ArticlePDF Available

Abstract and Figures

A novel glass brick façade has been designed and engineered to reproduce the original brick façade of a former townhouse in Amsterdam. Based on the original design the resulting façade comprises more than 6500 solid glass bricks, reinterpreting the traditional brick pattern, and elaborated cast glass elements for the replication of the window and door frames. To achieve unhindered transparency, the 10 by 12 m glass block façade has to be self-supporting. Previous experimental work by Oikonomopoulou et al. (J Facade Design Eng 2(3–4):201–222, 2015b. doi:10.3233/fde-150021) concluded that it was necessary to use a clear, UV-curing adhesive of high stiffness as bonding material. Experimental work on prototype elements indicated that the desired monolithic structural performance of the glass masonry system, as well as a homogeneous visual result, are only achieved when the selected adhesive is applied in a 0.2–0.3 mm thick layer. The nearly zero thickness of the adhesive together with the request for unimpeded transparency introduced numerous engineering challenges. These include the production of highly accurate glass bricks and the homogeneous application of the adhesive to achieve the construction of the entire façade with remarkably tight allowable tolerances. This paper presents the main challenges confronted during the construction of the novel façade and records the innovative solutions implemented, from the casting of the glass units to the completion of the façade. Based on the conclusions of the research and the technical experience gained by the realization of the project, recommendations are made on the further improvement of the presented glass masonry system towards future applications.
This content is subject to copyright. Terms and conditions apply.
Glass Struct. Eng. (2018) 3:87–108
The construction of the Crystal Houses façade: challenges
and innovations
F. Oikonomopoulou ·T. Bristogianni ·
F. A. Veer ·R. Nijsse
Received: 28 September 2016 / Accepted: 9 March 2017 / Published online: 11 April 2017
© The Author(s) 2017. This article is an open access publication
Abstract A novel glass brick façade has been des-
igned and engineered to reproduce the original brick
façade of a former townhouse in Amsterdam. Based on
the original design the resulting façade comprises more
than 6500 solid glass bricks, reinterpreting the tradi-
tional brick pattern, and elaborated cast glass elements
for the replication of the window and door frames.
To achieve unhindered transparency, the 10 by 12 m
glass block façade has to be self-supporting. Previ-
ous experimental work by Oikonomopoulou et al. (J
Facade Design Eng 2(3–4):201–222, 2015b. doi:10.
3233/fde-150021) concluded that it was necessary to
use a clear, UV-curing adhesive of high stiffness as
bonding material. Experimental work on prototype ele-
ments indicated that the desired monolithic structural
performance of the glass masonry system, as well as
a homogeneous visual result, are only achieved when
the selected adhesive is applied in a 0.2–0.3 mm thick
layer. The nearly zero thickness of the adhesive together
with the request for unimpeded transparency intro-
duced numerous engineering challenges. These include
F. Oikonomopoulou (B)·F. A. Veer ·R. Nijsse
Department of Architectural Engineering and Technology,
Faculty of Architecture and the Built Environment,
Delft University of Technology, Delft 2628 BL,
The Netherlands
T. Bristogianni ·R. Nijsse
Department of Structural Engineering, Faculty of Civil
Engineering and Geosciences, Delft University of Technology,
Delft 2628 CN, The Netherlands
the production of highly accurate glass bricks and the
homogeneous application of the adhesive to achieve
the construction of the entire façade with remark-
ably tight allowable tolerances. This paper presents the
main challenges confronted during the construction of
the novel façade and records the innovative solutions
implemented, from the casting of the glass units to the
completion of the façade. Based on the conclusions
of the research and the technical experience gained
by the realization of the project, recommendations are
made on the further improvement of the presented glass
masonry system towards future applications.
Keywords Structural glass ·Solid glass bricks ·
Adhesive glass connections ·Glass masonry ·Crystal
Houses façade ·Cast glass
1 Introduction
A novel glass masonry façade has been designed and
engineered to replace the brick façade of a former
townhouse in Amsterdam, aiming to preserve the city’s
traditional architectural style and historical ensemble.
Designed by the MVRDV architectural studio (www., the innovative façade follows the original
nineteenth century elevation down to the layering of
the bricks and the details of the window frames, but
is stretched vertically to comply with updated zoning
laws and allow for increased interior space (MVRDV
Architects 2016). Based on the brick modules of the
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
88 F. Oikonomopoulou et al.
Fig. 1 Illustration by MVRDV of the concept behind the Crystal Houses façade
original masonry façade, the 10 by 12 m transparent
elevation employs more than 6500 solid glass bricks,
each 210(±1)mm thick by 65(±0.25)mm high, rein-
terpreting the traditional brickwork and the character-
istic architraves above the openings; while massive cast
glass elements reproduce the classic timber door and
window frames. As it ascends, terracotta bricks inter-
mingle with glass ones, gradually transforming the
glass elevation to the traditional brick façade of the
upper floor (see Figs. 1,2). The architects’ desire for
unimpeded transparency excluded the use of a metal
substructure, rendering the choice for an entirely self-
supporting glass brick system as a necessary and so far
unique solution.
2 Principles of the adhesively bonded, solid glass
block wall
An illustration of the structural scheme followed for
maximizing transparency is shown in Fig. 3. Based on
previous realized examples of solid glass block façades
Oikonomopoulou et al. (2015b), lists three intertwined
factors that define the structural performance and the
transparency level of a self-supporting glass block
façade: (1) the choice between hollow or solid glass
bricks, (2) the choice between structural adhesive bond-
ing or supporting substructure and (3) the façade’s
overall geometry. In principle, a bearing wall of the
aforementioned size employing exclusively solid glass
bricks is feasible owing to the compressive strength of
glass (stated between 400–600 MPa for uniaxial load-
ing by Fink (2000) and 300–420 MPa by Granta Design
Limited (2015) and the considerable cross-section of
the solid glass bricks (210 mm) that allow the façade to
carry its dead load and have an enhanced buckling resis-
tance. In comparison, a wall of the same dimensions
comprising hollow glass blocks would require a sup-
porting sub-structure. Their reduced thickness results
in internal buckling or stress concentrations that in turn
lead to a relatively low stated resistance in compressive
load [defined as low as 6 MPa in ISO 21690:2006 by
International Organization for Standardization (2006)].
The lateral stability of the façade is guaranteed by
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 89
Fig. 2 Left 3-D
visualization of the façade
by MVRDV. Right The
realized façade
Fig. 3 Principle of the proposed structural glass system
four buttresses erected towards the interior by inter-
laced glass bricks, resulting in a continuous envelope
of increased rigidity.
An entirely transparent structural system is obtained
by bonding the glass bricks together with a clear adhe-
sive. In the developed system, the mechanical prop-
erties of this adhesive are equally critical to the ones
of the glass blocks; it is their interaction as one struc-
tural unit that defines the system’s structural capacity
and behaviour. The most favourable structural perfor-
mance is when adhesive and glass bricks fully coop-
erate and the masonry wall behaves as a single rigid
unit under loading, resulting in a homogeneous load
distribution. Extended research and testing of various
adhesive types by Oikonomopoulou et al. (2015b) lead
to the eventual selection of Delo Photobond 4468; a col-
orless, UV-curing, one-component acrylate of the Delo
Photobond family, designed for high strength bond-
ing between glass components. Adhesives of the Delo
Photobond family have already been applied for the
bonding of all-glass structures, i.e. in the frames of the
glass shell of the Leibniz Institute for Solid State and
Materials Research (Delo Industrial Adhesives 2011).
The selected adhesive is optimized for high force
transduction in glass/glass and glass/metal bonds and
presents high shear stiffness, good short and long term
compressive behavior and high humidity resistance
(Delo Industrial Adhesives 2014). Visually, besides
being colorless, it has a similar refractive index to
glass (1.5) and does not discolor when exposed to sun-
light. Another important feature is its photo-catalytic
curing, allowing for fast construction: The adhesive
can be fully cured in a minimum of 40 s using
60mW/cm2UVA intensity (Delo Industrial Adhesives
2014). After curing, it obtains its full structural capacity
and becomes moisture- and water- resistant.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
90 F. Oikonomopoulou et al.
Fig. 4 Schematic illustration of the relation between a stiff adhe-
sive’s strength and thickness by Riewoldt (2014), den Ouden
(2009), Wurm (2007)
The medium viscosity of the selected acrylate
[7000 mPa s at 23 C, measured by Brookfield vis-
cosimeter (Delo Industrial Adhesives 2014)] suggested
that only the horizontal surfaces of the glass bricks are
bonded; the vertical ones are left dry, allowing as well
for thermal expansion.
There are no clear guidelines from the adhesive man-
ufacturer on the recommended application thickness of
Delo Photobond 4468. To the knowledge of the authors,
there is not yet a generally approved theory concerning
the effect of adhesive thickness in the strength of the
bond. Although the classical elastic analyses predict
that the strength increases with the adhesive thickness,
experimental results show the opposite (da Silva et al.
2006). Research by Grant et al. (2009), da Silva et al.
(2006), and Crocombe (1989) suggests different rea-
sons1why a thicker bondlayer provides a decreased
joint strength. Based on experimental work by den
Ouden (2009) and Riewoldt (2014), Fig. 4exhibits
how a comparatively thicker layer can negatively influ-
ence a rigid (i.e. epoxy or acrylate) adhesive’s bond
strength and subsequently the structural performance
of the entire system. In practice Wurm (2007) men-
tions that acrylates present their highest strength in an
application thickness between 0.1 and 0.5 mm, whereas
Puller and Sobek (2008) suggest an optimum thickness
of 0.2 mm for a glass to metal bond with Delo Photo-
bond 4468.
1Crocombe (1989) suggests that thicker single-lap joints have a
lower strength considering the plasticity of the adhesive, whereas
da Silva et al. (2006) found that interface stresses are higher for
thicker bondlines. Grantetal.(2009) suggests that as the bondline
thickness of a T joint increases, there is an increase in the bending
stress since the bending moment increases, reducing the strength
of the joint.
In our case, structural and visual experiments indi-
cated an optimum joint thickness for the selected adhe-
sive between 0.2 and 0.3 mm. Small wall prototypes
by Oikonomopoulou et al. (2015b), comprising glass
elements of different tolerance range, pointed out that
a homogeneous bond thicker than 0.3 mm cannot be
obtained due to the adhesive’s flow properties and
medium viscosity. Besides compromising the visual
result, inconsistent bonding introduces weaker struc-
tural zones. Especially voids against the glass substrate
in stiff adhesives can cause major stress concentra-
tions (O’ Regan 2014). Series of four-point bending
tests in a Zwick Z100 displacement controlled univer-
sal testing machine, with a speed of 2 mm/min and fail-
ure at a nominal flexural stress between 4.79 and 7.01
MPa demonstrated that the chosen adhesive enables the
glass brick wall to behave monolithically under load-
ing, when the adhesive is applied in a uniform layer of
the optimum thickness (Oikonomopoulou et al. 2015a).
The characteristic failure pattern can be seen in Fig. 5:
Failure occurs with a straight cut, parallel to the load-
ing path, splitting the specimen in two as if it was one
solid component. Specifically, the glass block of the
middle horizontal layer, spanning the vertical joint of
the top and bottom layers, is split in half. No significant
delamination is observed.
Based on the adhesive’s optimum application thick-
ness, it was determined that the glass blocks’ top and
bottom surfaces should be flat within 0.25 mm for guar-
anteeing an even adhesive layer of the highest strength.
The adhesive’s medium viscosity and ideal thickness
of a quarter of a millimeter together with the elastic
nature of glass require exceptionally strict tolerances
for a homogeneous application. The presented system
is fundamentally different from a conventional mor-
tar masonry, where the mortar can accommodate pos-
sible discrepancies in size and surface quality of the
bricks. In this case, any accumulated deviation larger
than the required 0.2–0.3 mm thickness of the adhesive
could lead to uneven and improper bonding. There-
fore, not only the size of each brick, but also the thick-
ness of each construction layer have to be confined
within a tight dimensional precision of a quarter of
a millimeter. The demand of this unprecedented high
level of accuracy and transparency, introduced vari-
ous challenges in the engineering and construction of
the Crystal Houses façade, calling for innovative solu-
tions. Such challenges and their solutions are presented
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 91
Fig. 5 Failure pattern of the glass beams
3 Manufacturing and quality control of the glass
The required ±0.25 mm tolerance influenced the
choice of glass recipe and mould used. Solid glass
bricks of comparable dimensions (200 mm ×300
mm ×70mm),usedintheAtocha Memorial, the
only other adhesively bonded glass block envelope, uti-
lized borosilicate glass and precision press moulds for
obtaining highly accurate units (Schober et al. 2007).
Borosilicate glass was favoured over soda-lime glass
owing to its comparably lower thermal expansion coef-
ficient [3.2–4 ×106/K]over soda-lime glass [9.1–
9.5 ×106/K](Granta Design Limited 2015). This
results in considerably less natural shrinkage during
cooling and accordingly to a cast element of higher
dimensional accuracy. A high precision press mould
further confines the cast element to the desired dimen-
sions, by pressing the molten glass during the ini-
tial, rapid cooling stage. The dimensional tolerance
achieved with this method for the aforementioned glass
bricks was ±1.0mm(Paech and Goppert 2008) with-
out any machine processing. However, in the case of
the Crystal Houses, the required ±0.25 mm tolerance
necessitates the mechanical post- processing of the
blocks’ horizontal (bonding) surfaces, even for borosil-
icate glass. Therefore, soda-lime glass and open pre-
cision moulds were opted for the final brick fabrica-
tion to avoid an unnecessary increase in manufacturing
costs. Soda-lime is the least expensive form of glass
(Corning Museum of Glass 2011) and requires a sig-
nificantly lower working temperature than borosilicate
[melting temperature is approximately 1200–1400 C
compared to 1400–1600 C(Shand 1968)]. As a draw-
back, the higher thermal expansion coefficient of soda-
lime requires a considerably longer annealing time and
thus manufacturing time of the components. For exam-
ple, the borosilicate glass blocks of 70 mm ×200 mm
×300 mm in dimensions and 8.4 kg weight (shown in
Fig. 6), used in the Atocha Memorial, required a total
annealing time of circa 20 h each (Paech and Goppert
2008). Whereas, the comparatively smaller soda-lime
glass bricks of 65 mm ×210 mm ×210 mm in dimen-
sions and 7.2 kg weight used in this project, required
36–38 h of annealing time respectively. High precision
open moulds were preferred over press moulds, since
the use of the latter was considered an expensive and
unnecessary solution in view of the inevitable post-
To ensure that the higher expansion coefficient of
soda-lime glass will not result in excessive thermal
stresses on the façade, a simulation of the expected ther-
mal loads in a yearly cycle was performed by an exter-
nal company specializing in building physics. Based
on the optical transmittance data provided by TU Delft
for the solar gain (see Fig. 7), the orientation of the
specific location, the height of the surrounding build-
ings and the assumption of a constant heating load in
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
92 F. Oikonomopoulou et al.
Fig. 6 Left: the 300 ×200 ×70 mm borosilicate glass block of
the Atocha Memorial made by press mould. Center: a 210 ×105
×65 mm soda-lime glass block of the Crystal Houses prior to
post-processing. Right: a 210 ×105 ×65 mm soda-lime glass
block of the Crystal Houses after post-processing
Fig. 7 Optical
transmittance data of a
standard Poe s ia brick by
Tijssen (2014)
winter and cooling load in summer from the indoors air-
conditioning, heat and light transmittance of the wall
were simulated. The results indicated acceptable ther-
mal strains (less than 14.3×103)for the soda-lime
cast glass even under the most extreme weather condi-
tions for Amsterdam.
The fabrication of the approximately 7500 solid
glass bricks was assigned to the Italian company Poe-
sia ( Each brick is man-
ually cast by pouring molten glass in high preci-
sion, open steel moulds with a removable bottom sur-
face (see Fig. 8). A low-iron glass recipe is used for
high optical quality. The final chemical composition
of the glass, as measured by a X-ray fluorescence
(XRF) spectrometer, is shown in Table 1. To attain
the desired smooth external texture the steel moulds
are preheated to a constant temperature of approxi-
mately 650–750 C. If the mould’s temperature falls
below this range, then the hot glass coming into con-
tact with the metal surface freezes instantly, creating a
rough, wavy surface. On the other hand, if the mould
is heated to a higher temperature, the glass tends to
adhere to the walls of the mould. A release coating on
the moulds further prevents the adhesion of the molten
glass to the working surface and the development of
After the glass is poured into the mould, it is left
at ambient temperature to rapidly cool until 700 C.
This rapid cooling through the critical crystallization
zone is essential to avoid the molecular arrangement
of the melt in crystals instead of an amorphous struc-
ture, which would result in a cloudy glass of reduced
transparency (Shelby 2005). During this initial cooling
phase, the glass has still low viscosity that can allow
any induced thermal stress to relax out to a negligi-
ble amount immediately (Shelby 2005). After the glass
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 93
Fig. 8 Left: The high precision, open steel moulds. Right: Molten glass bricks during the rapid cooling phase from 1200 to 700 C
Tabl e 1 Composition of the applied cast glass based on XRF chemical analysis
Compound Wt% Absolute
error (wt%)
Compound Wt% Absolute
error (wt%)
Compound Wt% Absolute
error (wt%)
SiO275.606 0.1Sb
2O30.821 0.03 ZrO20.043 0.006
Na2O15.833 0.1 CuO 0.402 0.02 SO30.034 0.006
CaO 5.142 0.07 Al2O30.165 0.01 TiO20.02 0.004
K2O1.836 0.04 MgO 0.072 0.008 Fe2O30.01 0.003
temperature drops to its softening point (720 C),2
the viscosity of the glass is sufficient for it to retain
its shape and not deform under its own weight (Shand
1968). At 700 C, the glass element is removed from
the mould by suction at the top surface, and moved into
the annealing oven.
There, a long and meticulously controlled annealing
process eliminates any possible differential strain built
up between casting and demoulding, as well as pre-
vents the generation of internal residual stresses during
further cooling. Upon this point, key reference tem-
peratures are the annealing point (545 C) and the
2The temperatures given here for the softening, annealing and
strain points are indicative for soda-lime glass and are based on
research by Napolitano and Hawkins (1964). The values may
differ according to the exact composition of the glass. The exact
temperatures referring to the soda-lime recipe of the glass blocks
have not been disclosed to the authors by Poe sia .
strain point (505 C) of soda-lime glass. The anneal-
ing point is defined as the temperature at which the vis-
cosity of glass will allow any induced stress to relax out
substantially in just a few minutes (Shelby 2005). The
strain point is the temperature where the same stress is
reduced to acceptable values in 4h (Shand and Armis-
tead 1958;Shand 1968). The cast glass should be main-
tained for adequate time at the annealing point to relieve
any existing strains and then cooled at a rate sufficiently
slow so that residual stresses will not reappear when the
glass temperature has reached equilibrium (Shand and
Armistead 1958). Effectively, below the strain point,
stress cannot relax in time and is considered perma-
nent (Watson 1999). When the temperature of the entire
glass component has dropped below the strain point, the
component can cool at a faster pace until ambient tem-
perature, yet still sufficiently slow to prevent breakage
due to thermal shock (Shand and Armistead 1958).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
94 F. Oikonomopoulou et al.
Fig. 9 Qualitative analysis of strain concentration by polariza-
tion test. Bricks such as the ones shown on the left image, with a
clear indication of residual stresses, were discarded. Specimens
such as the one on the right, with no visible considerable strain
concentration, were employed in the façade
Accordingly, during the annealing range, the mag-
nitude of the resulting internal stresses, is largely
determined by the temperature difference between the
warmest and coolest parts of the glass, its coefficient of
expansion and the thickness of the section (Shand and
Armistead 1958). However, the heat transfer needed
for accomplishing the desired temperature difference
in practice is influenced by various parameters that are
challenging to accurately simulate. These include the
element’s shape and mass distribution, its sides exposed
to cooling, the existence and amount of other thermal
masses in the furnace, as well as the geometry and
characteristics of the furnace itself. There are several
guides on the annealing cycle of cast objects in the
scientific and industrial literature, but they are often
tailored to very specific circumstances and include
unclear assumptions (Watson 1999).
Hence, even though the heat transfer needed for pre-
venting considerable stresses in a cast glass element
can be calculated, due to all the above reasons, in prac-
tice, the annealing schedule of large 3-dimensional cast
units is often based on practical experience. Based on
the experience and furnace facilities of Poesia, the com-
pany concluded that each Crystal Houses brick with
65 mm ×105 mm ×210 mm dimensions requires
approximately 8 h of annealing, whereas bricks of
double the volume (65 mm ×210 mm ×210 mm)
require 36–38 h respectively to prevent the genera-
tion of attributable permanent residual stresses. The 65
mm thickness of the components hinders an accurate
through-the-thickness stress measurement by a Scat-
tered Light Polariscope (SCALP) stress-meter (using
the current hardware/software). Instead a qualitative
analysis of strain concentration was performed using
a polarized white light source and a crossed polarized
film that blocks the transmission of light. If glass is
subjected to stress, it exhibits optical anisotropy. This
corresponds to two refractive indices, which result in
the presence of isochromatic fringes (coloured pat-
terns) when polarized light passes through the com-
ponent (see Fig. 9,left)(McKenzie and Hand 2011).
Glass without any stress will appear completely dark
(Schott AG 2004). If the specimen presents besides
black only grey-scale spectral composition,3it has low
3Shribak (2015) provides an extended analysis of the interfer-
ence colours seen through polarization. For small retardance
the brightness of the region increases, first with a white spec-
tral composition at 200 nm. As the retardance increases, colours
start to appear beginning with yellow, then red, blue and green.
The colour changes in this sequence three more times until the
retardance reaches 2000 nm. Then the interference colours turn
white again and the retardance can no longer be reliably deter-
mined using the region’s spectral composition. A continuous
presence of only black and white subsequently signifies low
residual stresses.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 95
Fig. 10 Glass bricks of 210
mm ×210 mm ×65 mm
coming out of the annealing
oven after circa 36–38 h.
The natural shrinkage is
visible on the top surface
Fig. 11 Glass bricks
rapidly cooled (prior to
annealing) in vertical
orientation. The natural
shrinkage is evident besides
the top surface, also along
the larger surfaces
residual stresses. When the colour spectrum appears
the amount of stress is higher but cannot be quanti-
fied. This method was used by Poesia to control all
produced bricks. Bricks such as the one on the left of
Fig. 9, with a clear indication of internal stresses by a
coloured spectrum, were discarded. Only bricks such
as the one on the right of Fig. 9, with dark and white
areas were used in the façade. The fracture pattern of
tested specimens also suggested low residual stresses—
there was no excessive fragmentation observed in the
During the initial rapid cooling, natural, inevitable
shrinkage occurs to the glass volume during the mate-
rial’s transition from liquid to solid state. The shrinkage
causes different dimensions between units, uneven sur-
faces and is larger on the top, open surface of the casting
component owing to the additional gravity force (see
Figs. 6,10).
Different casting orientations were tested for min-
imizing the resulting shrinkage in the larger, bonding
surfaces of the bricks. Figure 11 demonstrates that even
when molten glass is rapidly cooled in a vertical ori-
entation, there is still visible shrinkage on the longer,
bonding faces of the components.
As, regardless of orientation of the mould, the
bricks’ bonding surfaces required further processing,
the horizontal position was favoured in terms of aes-
thetics, where the non-bonding sides are not visibly
distorted. Thus, to achieve the desired ±0.25 mm pre-
cision, the blocks are cast slightly higher. After the
annealing process, a CNC machine mills the top layer
of each block to remove the natural convex and obtain
the precise height. Finally, both top and bottom faces
of each block, i.e. the bonding surfaces, are polished to
a smooth flat surface of the desired dimensional accu-
racy. The four vertical surfaces remain unprocessed as
they do not influence the structural system. Mechani-
cal testing on both CNC polished and unpolished bricks
showed no deterioration of the mechanical properties
of the former.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
96 F. Oikonomopoulou et al.
Fig. 12 a The jig used in the first dimensional control. bThe set-up of the second control for checking the height and flatness of the
The processed glass bricks are then subjected to two
separate dimensional controls to verify their confor-
mity. Both controls were performed first at Poesia and
then again at TU Delft for verification. The first control
is accomplished by a cut-out metal plate jig that con-
trols the total length and width of the brick in 1.00 mm
accuracy (Fig. 12a). The second control employs a cus-
tomized electromechanical measurement bench with
five Linear Variable Differential Transformer (LVDT)
sensors of 1 µm accuracy (Fig. 12b) attached to an alu-
minium frame. By taking point measurements close to
the four edges and at the center of each unit, the sen-
sors check if the bricks meet the required 0.25 mm
tolerance in both height and flatness from the nominal
height of 65.00 mm. It should be clarified that the range
of acceptable height varies between 64.75 and 65.25
mm but within each particular brick the height devi-
ation cannot exceed 0.25 mm. Accordingly, through
this measuring control, the acceptable bricks are sorted
in two groups based on the point with the maximum
height: Group A comprises bricks between 64.75 and
65.00 mm high and Group B comprises bricks between
65.00 and 65.25 mm high respectively. Only bricks of
the same group were used per row of construction to
maintain the 0.2–0.3 mm requirement for the adhesive
Besides the dimensional controls, a visual inspec-
tion of the bricks is performed as well at the construc-
tion site. Flaws at the bricks’ bonding surfaces, usually
in the form of minute cracks or scratches even less
than 1 mm deep, commonly caused during the han-
dling and transportation, can trigger the propagation
of visible cracks after the adhesive’s curing process.
In specific, during curing the adhesive shrinks by 9%
Fig. 13 Left: A minor crack on a brick’s bonding surface. Right:
The propagation of such a minor crack after the curing of the
vol at ambient temperature (Delo Industrial Adhesives
2014) because of polymerization triggered by UV-light
(Delo Industrial Adhesives 2007), introducing a con-
siderable amount of tension to the minute cracks that
can start to propagate, eventually resulting in visible
cracking (see Fig. 13). Only the glass components that
pass both the measuring and visual controls were used
in the construction of the façade.
4 Construction of the glass brick wall
4.1 Construction site set-up
The upper conventional masonry façade of the top
residential floor, based on a steel beam spanning its
length, was completed six months prior to the con-
struction of the glass elevation (see Fig. 14). The level
of complexity of the manual bonding process of the
glass façade called for a highly skilled building crew
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 97
Fig. 14 Left: The masonry wall was already constructed prior to the glass elevation. Centre: the installed aluminium place holders of
the opening. Right: The mast climbing working platform and one of the three mobile elevated platforms
and a strictly controlled construction. A 12 h work-
ing schedule was established, 5 days per week. Seven
to nine highly skilled workers bonded and sealed on
average 80–100 bricks per day under the supervision
of two quality control engineers and the construction
site supervisor.
The special characteristics of the adhesive required
the construction of the façade inside a UV-filtering tent
for protection against solar radiation, adverse weather
conditions and dust. To ensure a controlled level of tem-
perature and humidity, heating equipment was installed
inside the tent so that the bricks and the adhesive be
maintained within workable temperatures during win-
ter. During the summer, when the ambient tempera-
ture exceeded 30 C the construction would temporar-
ily stop. Due to limited space in the construction site,
the glass blocks were stored in pallets in a separate
warehouse and were gradually transported to the site
upon demand.
A scaffolding with a mast climbing working plat-
form was installed for the construction of the glass
brick wall (Fig. 14). Simultaneously, three mobile ele-
vated working platforms were placed at the inner side
of the wall for the construction of the buttresses. Bricks
for one full row of construction were loaded and lifted
each time on the mast climbing working platform, from
where they were distributed for bonding. An elabo-
rate network of horizontal aluminum guides was uti-
lized to prevent any misalignment during the erec-
tion of the wall. Customized vertical aluminium frames
were temporarily installed as place holders of the wall
4.2 Levelling the starting bonding surface
The erection of the glass masonry wall started on top
of a 0.60 m high by 0.20 m wide reinforced concrete
plinth, essential for the protection of the lower part of
the façade against hard body impact; it has been cal-
culated to resist a vehicle collision travelling with a
velocity of 50 km/h. To match the texture and color of
the glass wall, the vertical faces of the concrete base
are coated with a laminate of a stainless steel sheet
and annealed patterned glass, laminated together by
SentryGlas®foil. A 30 mm thick stainless steel plate
fixed by bolts on top of the plinth, forms the base for
the glass masonry wall (see Fig. 15).
The prerequisite for extreme accuracy of the devel-
oped glass block system necessitates a reference build-
ing surface of corresponding flatness. Accordingly, the
stainless steel plate had to be levelled to an accuracy
of 0.25 mm for the 12 m length of the façade. Such
high measuring accuracy called for the development of
an innovative measuring and levelling system. Specifi-
cally, the bolts, set 275 mm apart (see Fig. 15) allow for
the levelling of the stainless steel plate in consecutive
steps. By employing standard levelling equipment the
plate is initially levelled to an accuracy of 3 mm over
the 12 m length. Figure 16 demonstrates the principle
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
98 F. Oikonomopoulou et al.
Fig. 15 On the left the bolts used for levelling the stainless steel plate, seen on the right
Fig. 16 Principle of the developed levelling system
of the measuring system developed to further level the
stainless steel plate to the desired precision: A contin-
uous open metal conduit with both ends sealed, sup-
ported directly on the concrete surface, is filled with a
non-transparent liquid. When still, a liquid will achieve
nearly absolute horizontal flatness, establishing the ref-
erence level for calibrating the plate. A laser scanner
with a sensor of 1 µm precision, fixed on an aluminium
frame with three legs is then moved over a set of con-
secutive points on the stainless steel plate, taking mea-
surements in reference to the liquid’s surface, map-
ping the plate along its entire length. The use of an
opaque reflective liquid (e.g. full fat milk) is essential
for ensuring that the laser beam will take all measure-
ments exactly at the same reference level. After the
entire surface of the plate is mapped, by tightening or
releasing the nuts and counternuts of the bolts the plate
was successfully levelled with a maximum height devi-
ation of 0.24 mm per total 12 m length. The resulting
gap between the concrete base and the plate was filled
with non-shrinkage concrete and left to cure.
4.3 Bonding
The 0.2–0.3 mm optimum thickness of the adhesive
layer demanded extreme precision in each construc-
tion layer. In traditional terracotta brickwork the mortar
plays the dual role of bonding and accommodating tol-
erances in the size of the bricks. However, the selected
adhesive’s inability to compensate for any dimensional
discrepancies in the construction can result to an accu-
mulated offset of a few centimeters in the total height
of the façade, even when the allowable tolerance per
glass component is only ±0.25 mm. To eliminate the
development of fluctuations in the height of the con-
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 99
Fig. 17 Bricks of a new row laid down prior to bonding (left). Then a feeler gauge is used for checking the thickness of the resulting
seam (right). The blade of the feeler gauge is sufficiently flexible and round not to induce any surface damage on glass
Fig. 18 Common flaws occurring in the adhesive layer: air gaps, capillary action and dendroid patterns
struction, all the glass bricks of a new row are laid down
prior to bonding. The thickness of the resulting hori-
zontal joint between the laid bricks and the bonded ones
below is then checked by a feeler gauge (see Fig. 17).
When the seam is larger than the suggested 0.25 mm,
the corresponding brick is replaced with another one
that accomplishes better contact in the specific loca-
tion. The final selection of bricks is then numbered to
guarantee their correct bonding sequence.
Previous structural and visual tests by
Oikonomopoulou et al. (2015b) suggested the bonding
of the complete contact surface between blocks. The
uniform application of the adhesive besides ensuring
a homogeneous load distribution, is also essential for
maximizing transparency. Indeed, the façade’s visual
result is deeply affected by any form of air gaps and
bubbles in the adhesive layer, as well as from stains
caused by the adhesive’s overflow or capillary action
(see Fig. 18). To eliminate such defects a customized
bonding procedure was applied.
Initially, the bricks are visually inspected on site for
any defects, as explained in chapter 3. Then, the sur-
faces to be bonded are cleaned with 2-propanol. Spe-
cially designed self-reinforced polypropylene forms
out of PURE®(DIT 2016) are placed for the distri-
bution of the adhesive in an X pattern, controlling the
flow, spread and amount of the adhesive (see Fig. 19).
To prevent any capillary effect along the vertical faces
of the glass bricks, a special, UV beam light is used to
locally harden the liquid adhesive in case it arises on
the vertical seams.
Once the adhesive is evenly spread, it is initially
exposed to low intensity UV-light for 5 s while the brick
is kept in position and under pressure. This pre-curing
step was introduced for practical reasons4as this partial
curing stabilizes the glass brick while still allowing the
wiping-off of any adhesive overflow. After cleaning,
the adhesive is further cured by low and medium inten-
sity UV-radiation in the range of 20–60 mW/cm2and
for a period of 60–120 s, according to brick size. Once
4This pre-curing time was set experimentally. Testing of speci-
mens cured in this way did not reveal any differences with spec-
imens cured once off.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
100 F. Oikonomopoulou et al.
Fig. 19 Bonding steps from left to the right: 1. Application of
the adhesive with the aid of the PURE®form. 2. Resulting ×pat-
tern. 3. Local hardening of the adhesive by a UV beam light for
preventing capillary action. 4. UV-lamp used to cure the adhesive
for 60–120 s
a complete brick layer is bonded, all joints are sealed
in order to guarantee the dust, water- and moisture-
tightness of the façade. For the sealing, Delo Photo-
bond 4497 (Delo Industrial Adhesives 2016b), a more
flexible and viscous, clear UV-curing Delo-Photobond
acrylate, specially designed for outdoor applications,
is selected due to its good visual performance, com-
patibility with Delo-Photobond 4468 as well as for its
easy and quick application (see Fig. 20). This adhe-
sive requires only 10 s of UV-curing to be completely
The first row of glass blocks was directly bonded
onto the stainless steel base by Delo-Photobond 4468.
As previously mentioned, Delo Photobond 4468 is rec-
ommended by the manufacturer for glass to metal bond-
ing as well. Previous research on such a bond has
been conducted by Puller and Sobek (2008). The estab-
lished rigid connection was considered imperative by
the structural engineers in order to eliminate any hor-
izontal movements of the free-standing façade. Any
movements due to temperature strains in the structure
are compensated by the flexible connections at the sides
and top of the façade (see Chap. 4.6).
Every 2 m of elevation, the levelling along the total
length of the façade is recorded using a high accuracy
total station. Bricks with a 0.5 or 1.0 mm reduction in
height were specially manufactured for the levelling of
the wall in case of height deviations. Such bricks were
required to level the wall segments when reaching the
level of the architraves of the ground and first floor.
At the top of the elevation, the glass wall is connected
to a steel beam by a 22 mm thick structural modified
silane (MS-) polymer bond. This flexible connection
can accommodate displacements due to the different
thermal expansion and stiffness between the upper con-
struction and the glass wall. A flexible waterproof tube
filled the gap towards the interior of the wall, to fur-
ther prevent water leakage. As ceramic strips cover the
entire top row, the connection details are fully hidden
(Fig. 21).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 101
Fig. 20 Left: Sealing of the already bonded bricks. Right: the final, visual result achieved by the novel bonding method
Fig. 21 The top connection is completely masked by the ceramic strips
4.4 Construction and installation of the architraves
The architraves above the window and door openings
of the original nineteenth century elevation are also
reinterpreted into glass components by special tapered
glass bricks bonded together by the same adhesive
along their vertical surfaces. Due to the medium vis-
cosity of the Delo Photobond 4468 each architrave had
to be pre-assembled into one single component in a
custom made rotating steel fixture. The rotating fixture
ensures the horizontal application of the adhesive, as
well as the desired arch geometry, with a straight top
line in accordance to the maximum 0.25 mm devia-
tion rule. The finished components, fabricated in the
TU Delft Glass & Transparency Lab, were then trans-
ported on site and installed one by one in situ with
the aid of a jib fixture on the fork lift, as shown in
Fig. 22.
4.5 Transition layer between standard and glass
To obtain a smooth, gradual connection to the standard
brickwork of the final, residential floor of the building,
the initial intention of the architects was to realize a
transition zone of intermixing glass and normal terra-
cotta bricks towards the top of the façade. Nonetheless,
the structural blend of the two materials presented var-
ious practical implications, as can be seen in Fig. 23.
Besides having different mechanical properties, the two
types of bricks vary appreciably in acceptable toler-
ances. While in the glass bricks the required precision
in height is ±0.25 mm, for the terracotta bricks is at
least ±1.0 mm. Most importantly, the bonding between
the two brick types necessitates the application of dif-
ferent adhesives, involving the risk of their intermixing.
Lastly, the strongly alkaline character of most mortars
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
102 F. Oikonomopoulou et al.
Fig. 22 Left: The special rotating fixture for the preassembly of the architraves. Right: The installation of the bonded architraves on
Fig. 23 Practical implications encountered when combining ter-
racotta and glass blocks, such as differences in acceptable toler-
ances and in use of adhesives
used for the bonding of standard ceramic bricks attacks
the glass surface and must be avoided. It should be men-
tioned that, as the upper conventional masonry façade
was completed six months prior to the construction of
the glass elevation and the mortar was fully cured, there
was no hazard of alkaline reaction between the mortar
and glass.
Due to all the aforementioned reasons the option of
combining terracotta and glass was discarded. Instead
the following solution was applied: Glass bricks, 40
mm shorter in width, clad with an 18 mm thick ceramic
strip at each external side, replace the traditional bricks
in the intermixing zone. The ceramic strips are bonded
on the façade after all glass blocks have been bonded
in place, preventing the occurrence of adhesive stains
on their exterior surface. Tec 7 (Novatech 2016), a
brown coloured modified silane polymer is applied for
bonding the strips to the glass units. With an applica-
tion thickness of circa 3 mm the adhesive compensates
for any difference in thermal strains between the two
materials. Once all the ceramic strips are bonded to
the façade, the seams around the strips and the glass
are sealed by Zwaluw Joint Fix 310 ml lichtgrijs (Den
Braven 2017), an acrylic based mortar of similar texture
and color to the mortar used for building the wall above
(see Fig. 24). The selected mortar is less brittle than
normal mortar due to its acrylic content and features
considerably less volume shrinkage (5%) (Den Braven
2017) after hardening in comparison to standard mortar
types, preventing thus its delamination from the glass
blocks. The completed intermixing, gradient zone can
be seen in Fig. 25.
4.6 Boundary connections of the façade
The façade forms a free standing wall firmly connected
to the concrete plinth. To allow for displacements due to
the different thermal expansion and stiffness between
the glass wall and its boundaries, the façade is joined via
flexible connections to the top metal beam, supporting
the residential level above, and to the stainless steel
columns on the vertical sides. The top connection of the
two structures is realized by a modified silane (MS-)
polymer adhesive bond as analyzed in Chap. 4.3.
Regarding the connection along the vertical sides,
this varies between the right and left (as seen from the
street) side of the wall at the ground floor, since the left
side is self-supported by a buttress. On the right side at
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 103
Fig. 24 Left: Bonding of the ceramic strips to the shorter bricks. Right: The final visual result after the ceramic strips have been sealed
Fig. 25 End result of the intermixing, gradient zone
the ground floor, as well as on both sides at the first floor,
the glass masonry wall is connected by a 10 mm thick
layer of a clear silyl-terminated semi-elastic polymer to
the stainless steel L-shaped columns to compensate for
thermal displacements of the wall. Since for the curing
of the specific MS-polymer adhesive the contact with
atmospheric conditions is essential, the adhesive was
applied gradually with a glue-kit dispenser using com-
pressed air row by row, so that each glue layer can set
until the next row of bricks is completed (see Fig. 26).
The bricks at the right side of the ground floor are each
clad with two 1 mm thick stainless steel strips at their
adjacent to the L-shaped column sides, to mask the
rough detailing of the welded stainless steel structural
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
104 F. Oikonomopoulou et al.
Fig. 26 From left to right: Bonding of the steel plate to the corner brick. Positioning of the brick by a suction cup holder. Application
of the semi-elastic polymer
Fig. 27 The graphite
moulds used for the
fabrication of the frames
components (see Fig. 26). The cladding is applied to
the bricks prior their bonding to the façade. For such a
connection, DELO Photobond 4497 is used, to ensure
impact resistance.
4.7 Installation and bonding of the cast glass window
and door frames
The reproduction of the previous, historic elevation’s
wooden openings in cast glass was an extra challenge
added to the engineering and construction of the Crystal
Houses as it included the manufacturing and bonding
of massive cast glass elements. The glass frames were
cast by Poesia in open graphite moulds (see Fig. 27),
ground along their open surface to remove the mate-
rial shrinkage layer and polished with a rotational band
manually. As such pieces present larger tolerance prob-
lems, DELO Photobond 4494 (Delo Industrial Adhe-
sives 2016a) was chosen to bond the frame elements
together due to its higher viscosity and application
thickness that allow for easier tolerances, while main-
taining a clear optical result. This adhesive has a com-
paratively lower mechanical resistance to Delo Pho-
tobond 4468, yet sufficient for integrating the glass
frames into the construction.
The window and door frames were placed after the
completion of the glass wall. During the bonding pro-
cess, aluminum place holders were used to secure tem-
porarily the location of the openings. First each frame
was assembled in place by DELO Photobond 4494.
Based on the UV-measurements per m2done by Siko
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 105
Fig. 28 Left: The simultaneous application of the polymer from both sides. Centre and right: The final result of the bonded glass frames
b.v. the sill of each window frame, of 1145 mm ×
143 mm footprint, required 4 minutes of total curing
by two UV-lamps travelling back and forth along its
Once the frame was in place, the side and top con-
nection to the glass masonry wall was established. The
thickness of this connection was designed to be 8mm,
to compensate for horizontal and vertical deviations in
the glass masonry wall, and was achieved by the same
clear silyl-terminated semi-elastic polymer used also
at the top and side connections of the wall. To avoid
the entrapment of air, the polymer was injected at both
sides simultaneously from bottom to top (see Fig. 28).
After a few days, when the polymer had reached a satis-
factory strength, the aluminium frames were removed.
Then the cast glass mullions were bonded to the glass
layer. Finally, the glass panes were bonded to the mul-
lions by a standard transparent silicone, completing the
façade. The end result can be seen in Fig. 28.
5 Conclusions
A novel, completely transparent self-supporting glass
masonry wall system has been developed and real-
ized through pioneering research in the Crystal Houses
façade (Fig. 29). With the exclusive use of solid cast
glass elements bonded together by a clear, high stiff-
ness, adhesive and with the aid of geometry for enhanc-
ing the lateral stability, the 10m by 12m façade com-
bines the desired structural performance with pure
transparency. The façade is capable of carrying its own
weight and withstanding wind loads without any addi-
tional substructure when the adhesive-glass assembly
functions as one rigid unit under loading. Previous
experimental work by Oikonomopoulou et al. (2015b)
indicated Delo Photobond 4468, a one-component,
UV-curing acrylate for attaining both the desired mono-
lithic structural performance and high transparency
level. The experiments also demonstrated that the
desired structural and visual performance is only guar-
anteed when the adhesive is applied in a uniform layer
of a mere 0.2–0.3 mm thickness. This in turn leads to
an allowable dimensional tolerance of a quarter of a
millimetre in the height and flatness of the cast glass
components. This demand of extreme dimensional pre-
cision introduced new challenges in the engineering of
the façade from the manufacturing of the bricks to their
bonding method, calling for pioneering solutions.
Due to the inevitable natural shrinkage of molten
glass, such dimensional accuracy could only be attained
by CNC-cutting and polishing of the bricks’ horizontal
faces to the desired height. Soda-lime glass and open,
high precision moulds were preferred over borosilicate
glass and press moulds to reduce the manufacturing
cost. Special measuring equipment was developed for
controlling the dimensional accuracy of the compo-
Nonetheless, even blocks of such high dimen-
sional accuracy can still lead to a significant offset in
the façade’s total height. The fundamental difference
between a conventional brickwork and the developed
glass masonry system is that a standard mortar layer
compensates for deviations in the size of the bricks,
while the selected adhesive cannot. This manifests the
level of complexity deriving from the manual bonding
and the significance of constantly controlling the entire
construction with high precision methods.
A completely transparent façade is moreover linked
with the inability to hide any possible flaws in the con-
struction. The development of a novel bonding method
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
106 F. Oikonomopoulou et al.
Fig. 29 The completed Crystal Houses façade
for the homogeneous and flawless application of the
adhesive resulted in imperceptible connections in the
constructed façade.
6 Discussion and further research
Overall, the innovative glass masonry system devel-
oped for the Crystal Houses façade illustrates the great
potential of adhesively bonded cast glass blocks as an
answer to the quest of structural transparency and can
form the basis for novel architecture applications. The
system can be further engineered in order to simplify
and accelerate its application, minimize the interlinked
challenges and decrease the cost.
Most of the engineering puzzles elaborated in this
paper can be solved with the use of a thicker transparent
adhesive of equal structural performance that in turn
can allow for larger tolerances.
In this direction, different envelope geometries can
enhance the rigidity of the structure, allowing for
thicker, more elastic adhesives and correspondingly for
larger tolerances in the brick units.
Moreover, the development of a casting method of
glass units of higher accuracy without the need of post-
processing would significantly facilitate the entire pro-
duction and building process as well as enhance the
structural and architectural result.
Likewise, the choice of glass recipe plays a crucial
role in the total annealing time and in the scale of result-
ing natural shrinkage. Although a faster and more accu-
rate casting process can be achieved with borosilicate
glass instead of soda-lime, the total manufacturing cost
and dimensional precision prerequisites should be con-
sidered prior to the glass recipe choice.
Lastly, the casting of glass units can provide the
designer with a great freedom in the shapes and sizes
Fig. 30 Interlocking,
dry-assembly glass bricks
developed by
Bristogianni and Barou for
the 3TU.bouw Lighthouse
project: Restorative Glass
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
The construction of the Crystal Houses façade: challenges and innovations 107
of the masonry module. A promising solution for all
transparent, load-bearing glass structures is the devel-
opment of interlocking cast units. Currently, research
is being conducted by the authors on the development
of such a system that circumvents the use of adhesives
by employing dry connections instead (see Fig. 30). In
this case, the overall stability is attained by the total
weight of the construction in combination with the
unit’s interlocking geometry that provides the neces-
sary constraints against lateral movement. To prevent
stress concentrations due to the high contact pressure
between the glass elements, a transparent foil is placed
as an intermediate layer between the units, allowing as
well for dimensional tolerances in the cast components’
Acknowledgements The research work was conducted by
the TU Delft Glass & Transparency Lab for Ashendene-
Leeuwenstein BV whose permission to publish the results is
gratefully acknowledged. MVRDV and Gietermans & Van Dijk
are responsible for the architectural design, ABT b.v. for the
structural simulations, Wessels Zeist b.v. for the construction of
the Crystal House façade and Poesia Ltd. for the manufacturing
of the glass bricks. We thank Ashendene-Leeuwenstein BV and
MVRDV for the 3-D impressions of the case study. Weespecially
want to thank Rob Janssen from Siko BV for his valuable advice
and assistance. Ruud Hendrikx at the Department of Materials
Science and Engineering of the Delft University of Technology
is acknowledged for the X-Ray analysis.The authors gratefully
acknowledge Kees Baardolf and Kees Van Beek for their invalu-
able technical assistance and insight throughout the project.
Compliance with ethical standards
Conflict of interest On behalf of all authors, the corresponding
author states that there is no conflict of interest.
Open Access This article is distributed under the terms of
the Creative Commons Attribution 4.0 International License
(, which permits
unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons license, and
indicate if changes were made.
Corning Museum of Glass: Types of Glass. http://www.cmog.
org/article/types- glass (2011)
Crocombe, A.D.: Global yielding as a failure criterion for bonded
joints. Int. J. Adhes. Adhes. 9(3), 145–153 (1989)
da Silva, L.F.M., Rodrigues, T.N.S.S., Figueiredo, M.A.V., de
Moura, M.F.S.F., Chousal, J.A.G.: Effect of adhesive type
and thickness on the lap shear strength. J. Adhes. 82(11),
1091–1115 (2006)
Delo Industrial Adhesives: Bond it: Reference Book on Bonding
Technology, 2nd edn. Delo Industrial Adhesives, Munich
Delo Industrial Adhesives: Glass Bonding: Requirements. Prod-
uct Range and Design Examples. In. Delo Industrial Adhe-
sives, Munich (2011)
Delo Industrial Adhesives: Technical Information: Delo-
Photobond 4468. Delo Industrial Adhesives, Windach
Delo Industrial Adhesives: Delo Photobond 4494 Technis-
che Information. In. Delo Industrial Adhesives, Windach
Delo Industrial Adhesives: Delo Photobond 4497 Technical
Information. Delo Industrial Adhesives, Windach (2016b)
Den Braven: Zwaluw Joint Fix Technische datasheet. In: Den
Braven (ed.). Oosterhout (2017)
den Ouden, G.: Lastechnologie. VSSD, Delft (2009)
DIT b.v.: Pure-TechnicalData Sheet. In: DIT b.v. (ed.). Dinxperlo
Fink, A.: Ein Beitrag zum Einsatz von Floatglas als dauerhaft
tragender Konstruktionswerkstoff. Technische Universität
Darmstadt, Ph.D. (2000)
Grant, L.D.R., Adams, R.D., da Silva, L.F.M.: Experimental and
numerical analysis of single-lap joints for the automotive
industry. Int. J. Adhes. Adhes. 29(4), 405–413 (2009)
Granta Design Limited: CES EduPack 2015. Granta Design Lim-
ited, Cambridge (2015)
International Organization for Standardization: ISO 21690:2006
(en). In: Glass in Building—Glass Blocks—Specification
and test methods. ISO (2006)
McKenzie, H.W., Hand, R.J.: Basic Optical Stress Measurement
in Glass. Society of Glass Technology, Sheffield (2011)
MVRDV Architects: Crystal Houses. en/
projects/crystal-houses (2016). Accessed 21 Feb 2017
Napolitano, A., Hawkins, E.G.: Viscocity of a standard soda-
lime-silica glass. J. Res. Natl. Bur. Stand. A Phys. Chem.
68(5), 439–448 (1964)
Novatech N.V.: Tec7 Technical Data sheet. In: Novatech N.V.
(ed.). Olen (2016)
O’ Regan, C.: Structural Use of Glass in Buildings, 2nd ed. The
Institution of Structural Engineers, London (2014)
Oikonomopoulou, F., Bristogianni, T., Nijsse, R., Veer, F.A.:
Innovative structural applications of adhesively bonded
solid glass blocks. In: Vitkala, J. (ed.) Glass Performance
Days, pp. 256–261. Tampere (2015a)
Oikonomopoulou, F., Veer, F.A., Nijsse, R., Baardolf, K.: A
completely transparent, adhesively bonded soda-lime glass
block masonry system. J. Facade Design Eng. 2(3–4), 201–
222 (2015b). doi:10.3233/fde-150021
Paech, C., Goppert, K.: Innovative Glass Joints—The 11 March
Memorial in Madrid. In: Louter, C., Bos, F., Veer, F. (eds.)
Challenging Glass: Conference on Architectural and Struc-
tural Applications of Glass, pp. 111–118. IOS Press, Delft
Puller, K., Sobek, W.: Glass-Steel Connections Using Acrylate
Adhesives. In: Bos, F., Louter, C., Veer, F. (eds.) Challeng-
ing Glass, Delft 2008, pp. 273–278. IOS Press, Amsterdam
Riewoldt: Riewoldt—Adhesion—Cutting—Die Cutting/
knowledge Center.
center/fundamentals-of-adhesive-technology/ (2014)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
108 F. Oikonomopoulou et al.
Schober, H., Schneider, J., Justiz, S., Gugeler, J., Paech, C., Balz,
M.: Innovations with glass, steel and cables. In: Glass Per-
formance Days, pp. 198–201. Tampere (2007)
Schott AG: Technical Information: Stress in optical glass. In:
Schott (ed.) Advanced Optics. Schott AG, Mainz (2004)
Shand, E.B.: Engineering Glass, Modern Materials, vol. 6. Aca-
demic Press, New York (1968)
Shand, E.B., Armistead, W.H.: Glass Engineering Handbook.
McGraw-Hill, New York (1958)
Shelby, J.E.: Introduction to Glass Science and Technology: Edi-
tion 2 (2005)
Shribak, M.: Polychromatic polarization microscope: bringing
colors to a colorless world. Sci. Rep. 5, 17340 (2015).
Tijssen, M.: Transmissie glazen baksteen. In: DIMES. TU Delft,
Delft (2014)
Watson, D.M.: Practical Annealing In: 11th Biennial Ausglass
Conference, Wagga Wagga 1999. Mc Kinnon (1999)
Wurm, J.: Glass Structures: Design and Construction of Self-
supporting Skins. Birkhauser, Germany (2007)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”),
for small-scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are
maintained. By accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use
(“Terms”). For these purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or
a personal subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or
a personal subscription (to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the
Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data
internally within ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking,
analysis and reporting. We will not otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of
companies unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that
Users may not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to
circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil
liability, or is otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by
Springer Nature in writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer
Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates
revenue, royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain.
Springer Nature journal content cannot be used for inter-library loans and librarians may not upload Springer Nature journal
content on a large scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any
information or content on this website and may remove it or features or functionality at our sole discretion, at any time with or
without notice. Springer Nature may revoke this licence to you at any time and remove access to any copies of the Springer Nature
journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express
or implied with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or
warranties imposed by law, including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be
licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other
manner not expressly permitted by these Terms, please contact Springer Nature at
... On the contrary, each realized case-study opts for a different bonding media, based on the respective prioritized performance criteria. Application testing of the bonding media is crucial for verifying the constructability of the system, whereas structural tests are necessary to validate the overall performance of the assembly; as, similarly to conventional masonry structures, it is not the individual strength of each element, but rather the degree of collaboration between glass blocks and bonding media into one structural unit that defines the structural capacity and behaviour of the system (Oikonomopoulou et al. 2017a). ...
... They typically present low elongation at break, leading to a sudden, brittle failure. Failure is usually shown as plucking of the glass component as the bond is usually stronger than the glass; as demonstrated by 4point testing on adhesively-bonded glass brick beams by (Oikonomopoulou et al. 2015b;Oikonomopoulou 2019) Combining high strength with the possibility of maximized transparency, stiff adhesives have been traditionally preferred for adhesively-bonded cast glass applications, namely in the Atocha Memorial (Paech and Göppert, 2008;Schober et al. 2007), Crystal Houses (Oikonomopoulou et al. 2015b(Oikonomopoulou et al. , 2017a and LightVault (Parascho et al. 2020). Nonetheless, the high bond strength of such adhesives, which also ensures a good composite action and creep resistance (O' Regan 2014), can only be guaranteed when they are applied at their optimum bond thickness. ...
... In component level, mechanical post-processing of the glass bricks may be necessary to match the required accuracy in flatness and height of each component. In system level, a meticulous construction of extreme precision by a highly-specialized building crew is essential: the thickness of each construction layer has to be meticulously controlled, as any accumulated deviation larger than the required bonding thickness will lead to uneven and improper bonding (Oikonomopoulou et al. 2017a). This level of dimensional precision has a direct negative impact on both manufacturing and construction costs. ...
Full-text available
Cast glass is a promising, three-dimensional expression of the material for architectural and structural applications, particularly for the creation of all-transparent, self-supporting structures and envelopes. Typically applied in the form of solid blocks, cast glass components can be used as repetitive units to comprise fully-transparent, cast glass masonry walls. To maximize transparency and ensure an even load distribution, the glass blocks are bonded together by a colourless adhesive. Currently, there is a lack of standardized structural specifications, strength data and building guidelines for such adhesively-bonded cast glass-block systems. As a result, any new application is accompanied by experimental testing to select the adhesive and certify the adhesively bonded system. Since the choice of adhesive is highly dependent on the prerequisites set for each case-study -such as the structural and visual performance, available budget, the structure’s geometry and climate conditions- the preselection of the most prominent adhesive family at an early project stage can prevent an excessive budget and construction complications. This paper, therefore, aims to shed light on the selection process of adhesives for cast glass assemblies by first providing an overview of the most suitable bonding media families for such systems; these include stiff adhesives, flexible adhesives and cement-based mortars. Following, the paper reviews the research & development process of the adhesively-bonded glass-block systems in three distinct built projects, in which the TU Delft team has been involved: The Crystal Houses façade (NL), the LightVault, a robotically assembled glass vault (UK) and the Qaammat pavilion in the arctic circle (GL). The adhesive requirements for each of the three case studies are discussed in terms of structural and visual performance and ease-of-assembly (constructability). These criteria are decisive in pointing out the most promising bonding media family per case-study. The final shortlist of adhesive candidates within that bonding media family is subject to the full list of performance criteria, but also to market availability. The shortlist of adhesive candidates are typically experimentally evaluated, first via application testing and then via strength tests in order to choose the most suitable candidate. Based on the above, the review concludes in proposing guidelines for the effective selection, design and experimental verification of adhesively-bonded cast glass assemblies.
... Some notable examples include the Atocha Memorial (Paech and Göppert 2008), the Crown Fountain (Hannah 2009), the Optical House (Hiroshi 2013), the Crystal Houses (Oikonomopoulou et al. 2015(Oikonomopoulou et al. , 2018b (Fig. 1), the Qwalala Sculpture (Paech and Göppert 2018), the LightVault (Parascho et al. 2020) and the Qaammat Pavilion (Oikonomopoulou et al. 2022). What is common across all aforementioned projects is that the cast glass elements follow the shape of standardized bricks, mimicking the functionality, shape and size of ceramic masonry; a glass volume which can be annealed within a reasonable time length (Fig. 2). ...
... Both sharp corners and thin sections will cool faster than the remainder of the object resulting in uneven shrinkage, thus causing undesired levels of internal stresses. Based on the suggested fillet employed at the glass blocks of the Crystal Houses project (Oikonomopoulou et al. 2018b), a minimal fillet of 3 mm radius for sharp corners is assumed. It should be noted that within TO generated geometries, sharp angles are generally not found. ...
... An empirical estimation of the anticipated annealing times is therefore made through comparison to existing results. Three soda-lime glass brick elements from the Crystal Houses façade project have been selected for a quantitative comparison, using the recorded cooling times indicated in (Oikonomopoulou et al. 2018b). The main aspects that determine annealing times are the glass type, mass and thickness of each element. ...
Full-text available
Up to now, fabricating cast glass components of substantial mass and/or thickness involves a lengthy and perplex annealing process. This has limited the use of this glass manufacturing method in the built environment to simple objects up to the size of regular building bricks, which can be annealed within a few hours. For the first time, structural topological optimization (TO) is investigated as an approach to design monolithic loadbearing cast-glass elements of substantial mass and dimensions, with significantly reduced annealing times. The research is two-fold. First, a numerical exploration is performed. The potential of reducing mass while maintaining satisfactory stiffness of a structural component is done through a case-study, in which a cast-glass grid shell node is designed and optimised. To achieve this, several design criteria in respect to glass as a material, casting as the manufacturing process and TO as a design method, are formulated and applied in the optimisation. It is concluded that a TO approach fully suited for three-dimensional glass design is as of yet not available. For this research, strain- or compliance based TO is selected for the optimization of the three-dimensional, cast glass grid shell node; in our case, we consider that a strain based TO allows for a better exploration of the thickness reduction, which, in turn, has a major influence on the annealing time of cast glass. In comparison, in a stress-based optimization, the considerably lower tensile strength of glass would become the main restrain, leaving underutilized the higher compressive strength. Furthermore, it is determined that a single, unchanging and dominant load-case is most suited for TO optimisation. Using ANSYS Workbench, mass reductions of up to 69% compared to an initial, unoptimized geometry are achieved, reducing annealing times by an estimated 90%. Following this, the feasibility of manufacturing the resulting complex-shaped glass components is investigated though physical prototypes. Two manufacturing techniques are explored: lost-wax casting using 3D-printed wax geometries, and kiln-casting using 3D-printed disposable sand moulds. Several glass prototypes were successfully cast and annealed. From this, several conclusions are drawn regarding the applicability and limitations of TO for cast glass components and the potential of alternative manufacturing methods for making such complex-shaped glass components.
... The focus was placed primarily in finding a structural adhesive that functions similarly to a mortar in traditional brickwork in order to facilitate assembly: the adhesive should provide sufficient strength and at the same time absorb, within its thickness, the intolerances in size of the bricks and of the entire construction and allow for a fast and simple assembly. Subsequently, the prioritized performance criteria for the adhesive selection for the Qaammat pavilion, are fundamentally different to the ones followed by the previous realized examples of adhesively-bonded glass brick envelopes, namely the Crystal Houses façade [4] and the Atocha Memorial [10],and are more similar to the bonding solution followed at the Qwalala Sculpture (see Table 1), although in our case all joints should be sealed afterwards to prevent water/frost and dirt from entering. ...
... Moreover, the selected adhesive should allow for a fast fixing and curing time, which were set at < 30 min and < 24 h respectively. A quick fixing time was important for preventing the overflow of the adhesive and accidental movement (sliding) of the blocks but also for enabling a relatively quick construction, essential due to 1 3 the short Greenlandic summer: the pavilion should be built within a few weeks, thus, the adhesive should set quickly enough to allow for the built-up of several rows (3)(4) in one day. Lastly, due to the lack of electricity and of other common commodities in the specific location, it was essential that the construction could be realized without the need of strictly controlled environmental conditions (i.e. ...
... A good alternative was to order a battery-driven dispenser; however, after contacting multiple providers in EU and USA, we could not secure a battery-driven dispenser for such cartridge and mixing ratio in stock before September. Thus, given the limited timeframe for the building of the pavilion and the lack of alternatives, the construction started with the application of the adhesive via a manual dispenser for the first few rows (3)(4); in September, a battery-driven dispenser was obtained and the application of the adhesive was much easier and faster. In any case, due to the prolonged time needed for the DOWSIL Experimental Fast Curing adhesive to obtain its full strength, a maximum of 3 rows per wall was bonded each day. ...
Full-text available
An adhesively bonded, solid-glass brick pavilion has been designed by Konstantin Arkitekter as a landmark within the Aasivissuit – Nipisat UNESCO heritage in Greenland. The sculptural glass structure, measuring approximately 3.2 m in diameter × 2 m in height, faces a diverse set of engineering challenges compared to existing adhesively bonded glass brick structures. Placed in a remote location in the arctic circle, it has to withstand winter temperatures as low as -35 °C, and be built under a limited budget with the aid of the local population. Hence, key for the successful construction of the pavilion is finding an adhesive that satisfies the structural and aesthetic requirements of the project and simultaneously provides a simple and fast construction that spares the need for specialized building crew and sophisticated equipment, and is able to withstand the polar winter temperatures. Applicability and shear tests in (i) lab temperature conditions and (ii)) -5 °C lead to the final selection of: (a) 3M™ Scotch-Weld™ Polyurethane Adhesive DP610, which has a higher shear strength capacity, 1 mm gap filling capacity and is clear in colour, for bonding the bottom rows of the pavilion where higher strength is required due to the reduced overlapping of the bricks; and of (b) DOWSIL Experimental Fast Curing Adhesive developed by Dow Silicones Belgium particularly for this project, with a satisfactory shear strength, 3 mm gap filling capacity and white colour for the rest of the pavilion; its considerably larger gap filling capacity facilitates the ease of assembly as it can accommodate within the joint thickness the anticipated ± 1.5 mm standard size deviations of the soda-lime cast glass solid bricks and the possible accumulated deviations during construction. The paper further describes the application of the adhesive, first on a small-scale prototype, and then on site, and presents the encountered engineering and logistical challenges during the construction of the pavilion in Greenland.
... • Beams cut out in 20 * 30 * 350 mm size from Poesia cast glass frames. 5 5 The specific cast glass frames were produced as test pieces for the Crystal Houses façade (Oikonomopoulou et al. 2018a), in a 65 * 150 * 790 mm size. ...
... A lower internal friction of such borosilicate composition in comparison to SLS is also reported in bibliography (Duan et al. 2003). In the case of the "Borosilicate Tubes" glass, the low alkali to boron ratio, favours the association of the alkali to the boron, converting the oxygen coordination of boron from 3 (triangle) to 4 Footnote 16 continued 700 • C at atmospheric conditions, before the glass components are inserted in the annealing lehr (Oikonomopoulou et al. 2018a). 17 The internal friction Q −1 is calculated from the amplitude decay of the free vibration: ...
... Comparing the performance of the kiln-cast specimens to the reference samples, the "FT Float 1120 • C" samples have a lower average flexural strength (σ f = 48 MPa) than the single 10 mm float panes (56 MPa) or the adhesively bonded DELO 4468 samples, 25 in 25 DELO Photobond 4468 adhesive has been chosen due to the strong bonding to glass, which leads to monolithic behaviour of the bonded glass, as proven by Oikonomopoulou et al. (2018a). When the applied adhesive layer is thin enough, during the bending of the horizontally bonded glass sample, any shear gradient in the adhesive layer is eliminated, and thus the stress/strain is the case where the adhesive thickness was between 30-50 µm (53 MPa). ...
Full-text available
Cast glass has great potential for diverse load-bearing, architectural applications; through casting, volumetric glass components can be made that take full advantage of glass’s stated compressive strength. However, the lack of engineering, production and quality control standards for cast glass and the intertwined ambiguities over its mechanical properties-particularly due to the variety in chemical compositions and the lack of understanding of the influence of flaws occurring in the glass bulk-act as an impediment to its wide-spread application. Addressing the above uncertainties, this work studies a total of 64 silicate-based glass specimens, prepared in 20 * 30 * 350 mm beam size, either by kiln-casting at relatively low forming temperatures (970–1120 $$^{\circ }$$ ∘ C), or by modification of industrially produced glass. For the kiln-casting of the specimens, pure and contaminated recycled cullet are used, either individually or in combination (composite glasses). The defects introduced in the glass specimens during the casting process are identified with digital microscopy and qualitative stress analysis using cross polarized light, and are categorized as stress-inducing, strength-reducing or harmless. The Impulse Excitation Technique is employed to measure the Young’s modulus and internal friction of the different glasses. Differential Scanning Calorimetry is used on a selection of glasses, to investigate changes in the glass transition range and fictive temperature of the kiln-cast glasses due to the slower cooling and prolonged annealing. The four-point bending experiments are shedding light upon the flexural strength and stiffness of the different glasses, while the fractographic analysis pinpoints the most critical defects per glass category. The experiments show the flexural strength of cast glass ranging between 30–73 MPa, according to the level of contamination and the chemical composition. The measured E moduli by both methods are in close agreement, ranging between 60–79 GPa. The comparison of the flexural strength with prior testing of cast glass involving shorter span fixtures showed a decreasing strength with increasing size for the contaminated specimens, but similar strengths for pure compositions. The results highlight the versatile role of defects in determining the glass strength and the complexity that arises in creating statistical prediction models and performing quality control.
... The interlayer shall accommodate the size discrepancies of the individual blocks and achieve a homogeneous contact, to prevent the occurrence of localized peak stresses, which can lead to failure of the assembly (Oikonomopoulou, 2018). The ability of the interlayer to accommodate the size deviations of the individual blocks further contributes to decreasing the costs of a cast-glass structure, as it evades the post-processing of the blocks, such as in the Crystal Houses facade (Oikonomopoulou et al., 2017). Previous realized examples suggest anticipated size deviations in height and flatness of ±1 mm for (mould-pressed) borosilicate glass blocks (Paech, Göppert, 2008) and of ± 1.5 mm for (openmould) soda-lime cast glass blocks (Oikonomopoulou et al., 2022) of a size comparable to terracotta bricks. ...
... Based on previous realized examples, owing to the large thickness and thus contact area between the blocks, the anticipated permanent compressive stresses acting on a glass-block structure due to own weight are typically considerably less than 0.5 MPa. For example, at the lowest row of bricks, the permanent compressive stresses acting upon the Crystal Houses facade (12m tall) were < 0.2 MPa (Oikonomopoulou et al., 2017), <0.1 MPa for the Qwalala Structure (Paech, Goppert, 2018) and <0.15 MPa for the Qaammat Pavilion (Oikonomopoulou et al., 2022). Thus, the requirement of a >2MPa compressive strength should be able to satisfy most cast-glass block assemblies. ...
Full-text available
Interlocking cast glass assemblies are a promising solution for architectural cast-glass applications aiming for high transparency and a reversible structure that allows the reuse of the glass components (Oikonomopoulou et al.,2018; Oikonomopoulou,2019b). In such a system, an interlayer material between the glass elements is essential, to assist the homogenous stress distribution and account for the surface microasperities of the glass elements. Towards circularity, this material should be dry (and not an adhesive), allowing for the eventual disassembly of the system. Previous experimental work by (Aurik at al.,2018; Oikonomopoulou at al.,2019b) has focused on the use of PU and PVC interlayers as suitable candidates; the focus in those studies has been solely placed on the mechanical performance of the interlayer material. This research provides a review of potential material candidates suitable for interlayers of an interlocking cast glass assembly based on a set of revised design and performance criteria that are divided into primary and secondary. Furthermore, the impact their unique material properties have on the potential application of the interlocking system is examined. The whole process, from fabrication to construction of the entire assembly, based on an assumed building scenario, is presented in a chain reaction manner, whose starting point is the interlayer itself. After defining the design criteria the interlayer should adhere to, the proposed candidates are: PETG sheets (Vivak®), Neoprene, Aluminum, Laminated Polyurethane (PU) and a Soft-core aluminum interlayer. The unique properties and fabrication challenges of all five proposed interlayers are considered, as well as their properties in relation to assembly, which leads to the development of two distinct assembly sequences. The main distinction concerns the interlayers that risk creeping and those that do not. The research concludes with a comparison between the interlocking assembly and the other glass block assemblies currently applied.
... The use of cast glass as a structural material has been explored in several architectural projects such as the Atocha Memorial in Madrid (Paech and Goppert, 2008), the Optical Glass House in Hiroshima (Nakamura, 2012), the Crystal Houses in Amsterdam (Oikonomopoulou et al., 2018a;2018b) and the Robotic Glass Vault in London (Parascho et al., 2020). The success of these projects showcases the aesthetical and structural potential of cast glass, intriguing the architectural and engineering community to consider cast glass as a promising building material. ...
... Poesia is the producer of the cast glass bricks employed for the building of the Crystal Houses Façade (Oikonomopoulou et al., 2018a). ...
Full-text available
The emerging interest in the architectural applications of cast glass components reveals a knowledge gap on the mechanical properties of cast glass. Apart from its chemical composition, cast glass is characterized by its manufacturing history and thermal profile, often inheriting a set of defects that define its properties. The role that inhomogeneities in the bulk of voluminous glass components have on the strength of the final product is also uncertain. Systematic testing is therefore necessary for the safe structural application of cast glass. Towards this direction, the presented research aims to experimentally investigate the fracture resistance of cast glass under sharp contact loading, by means of a customized splitting test using a sharp linear indenter. Cubic specimens with 50 mm sides are kiln-cast at low forming temperatures, employing a variety of silicate-based cullet and firing schedules and their inherent defects are documented. The results of the splitting tests show that the borosilicate specimens fail at the highest splitting force, followed by the soda lime float specimens, while the fused or porous specimens have a significantly lower resistance to fracture. The strength order of the various glasses, as this results from the splitting tests-is opposite to that found earlier in four-point bending tests, due to the different fracture mechanisms activated. The fracture resistance of a glass specimen is governed, first by its ability to deform around the indenter to relief the developing stresses and then by its bond strength to resist crack propagation. Thus, a good balance between glass network flexibility and high bond dissociation energy is required, explaining why the tested homogeneous borosilicate and soda lime glasses are more resistant than the modified soda lime compositions with high alkali content. In addition, the fractographic analysis indicates that the non-stress inducing flaws in the bulk have a negligible contribution to the fracture resistance of the specimens.
... The construction of the Crystal House in Amsterdam, Netherlands, utilized an adhesive bonding material between individual glass bricks to create an all-glass façade. The base of this glass wall rested on stainless steel plates to separate it from the concrete foundation (Oikonomopoulou et al. 2018b). In larger structural applications, structural silicone is used to bond the glass at joints (de Krom et al. 2020a 2020b). ...
Full-text available
In this experimental research a transparent thermoplastic manufactured by the DOW Corporation and known as Surlyn is investigated for use as an interface material in fabrication of an all-glass pedestrian bridge. The bridge is modular in construction and fabricated from a series of interlocking hollow glass units (HGU) that are geometrically arranged to form a compression dominant structural system. Surlyn is used as a friction-based interface between neighbouring HGUs preventing direct glass-to-glass contact. An experimental program consisting of axial loading of short glass columns (SGC) sandwiched between Surlyn sheets is used to quantify the bearing capacity at which glass fracture occurs at the glass-Surlyn interface location. Applied load cases include 100,000 cycles of cyclic load followed by 12 hours of sustained load followed by monotonic load to cracking, and monotonic loading to cracking with no previous load history. Test results show that Surlyn functions as an effective interface material with glass fracture occurring at bearing stress levels in excess of the column-action capacity of an individual HGU. Furthermore, load cycling and creep loading had no effect on the glass fracture capacity. However, the load history had a nominal effect on Surlyn, increasing stiffness and reducing deformation.
... In view of the established efficiency of large-scale/high-volume glass production and lamination technologies, the advantages of upscaling AM to m²-scale glass devices or functional sheet remain unclear, although architectural-scale glass objects by AM have already been demonstrated (152). In this latter field, AM has the potential to replace specialized molding techniques used to fabricate certain kinds of structural glasses or cast glass bricks (153). Similarly, AM manufacturing of functional glass ceramic materials (e.g. ...
Among engineering materials, ceramics are indispensable in energy applications such as batteries, capacitors, solar cells, smart glass, fuel cells and electrolyzers, nuclear power plants, thermoelectrics, thermoionics, carbon capture and storage, control of harmful emission from combustion engines, piezoelectrics, turbines and heat exchangers, among others. Advances in additive manufacturing (AM) offer new opportunities to fabricate these devices in geometries unachievable previously and may provide higher efficiencies and performance, all at lower costs. This article reviews the state of the art in ceramic materials for various energy applications. The focus of the review is on material selections, processing, and opportunities for AM technologies in energy related ceramic materials manufacturing. The aim of the article is to provide a roadmap for stakeholders such as industry, academia and funding agencies on research and development in additive manufacturing of ceramic materials toward more efficient, cost-effective, and reliable energy systems.
Sunlight is essential for life on earth. Many biological processes are influenced by the light of the sun and humans also react sensitively to it. Therefore, this light and its incidence have a decisive influence on the comfort inside buildings. Hence, glass, as a transparent building material, is an elementary design element in modern architecture. To support the clear appearance, transparent adhesives represent an attractive alternative to opaque adhesives as well as to opaque structural sealants. But solar radiation affects such adhesives or sealants negatively through the absorption of UV-light that leads to unfavorable changes in material properties and has a significant effect on their durability. To estimate the service life of structural adhesive joints in glass constructions a large number of accelerated aging processes are carried out on a laboratory scale.[1] [2], In the case of UV aging, the accelerating effect is mainly achieved by continuously irradiating the samples. The radiation flux is mostly constant what only partially represent real solar radiation effects. A precise knowledge of the real light impact on the building envelope is needed to be able to evaluate the received material stresses after artificial aging correctly. In this study, transparent adhesives are conducted to standardized, time-shortened sunlight aging methods followed by the examination of the effects via spectroscopic and chemical analysis. These results are evaluated with regard to real amounts of radiation occurring on a building. Corresponding data are used from a simulation of the solar radiation exposure depending on season and weather influences as well as on the location and orientation of the building[3]. The aim is to identify zones with such high radiation amounts which cause damage of the adhesives and to investigate and recommend constructive protective measures.
Full-text available
Transparency, a highly sought aspect in architectural applications, has led to rising demand for glass in buildings and other structures. The changeover of the role of architectural glass makes it inevitable to characterize the mechanical properties of glass as individually as has been done for all other conventional materials employed in structural applications. Thus the awareness of glass strength is vital for the design of transparent structures made of structural glass. The present work reports displacement-controlled four-point bending tests performed on a set of annealed glass specimens to study the effect of pre-loading value and its duration on the flexural behaviour. Analysis of variance is carried out in Minitab 19 to study the effect of pre-loading and thickness of glass on the failure stress ratio. Analysis of variance results show that pre-load is the most significant factor for the failure stress ratio.
Full-text available
Interference of two combined white light beams produces Newton colors if one of the beams is retarded relative to the other by from 400 nm to 2000 nm. In this case the corresponding interfering spectral components are added as two scalars at the beam combination. If the retardance is below 400 nm the two-beam interference produces grey shades only. The interference colors are widely used for analyzing birefringent samples in mineralogy. However, many of biological structures have retardance <100 nm. Therefore, cells and tissues under a regular polarization microscope are seen as grey image, which contrast disappears at certain orientations. Here we are proposing for the first time using vector interference of polarized light in which the full spectrum colors are created at retardance of several nanometers, with the hue determined by orientation of the birefringent structure. The previously colorless birefringent images of organelles, cells, and tissues become vividly colored. This approach can open up new possibilities for the study of biological specimens with weak birefringent structures, diagnosing various diseases, imaging low birefringent crystals, and creating new methods for controlling colors of the light beam.
Full-text available
A pioneering, all transparent, self-supporting glass block facade is presented in this paper. Previously realized examples utilize embedded metal components in order to obtain the desired structural performance despite the fact that these elements greatly affect the facade’s overall transparency level. Undeniably, the oxymoron ‘transparency and strength’ remains the prime concern in such applications. In this paper, a new, innovative structural system for glass block facades is described, which demonstrably meets both criteria. The structure is exclusively constructed by monolithic glass blocks, bonded with a colourless, UV-curing adhesive, obtaining thus a maximum transparency. In addition, the desired structural performance is achieved solely through the masonry system, without any opaque substructure. Differing from previous realized projects, solid soda-lime glass blocks are used rather than borosilicate ones. This article provides an overview of the integrated architectural and structural design and discusses the choice of materials. The structural verification of the system is demonstrated. The results show that the adhesively bonded glass block structure has the required self-structural behaviour, but only if strict tolerances are met in the geometry of the glass blocks.
Full-text available
The effect of the adhesive thickness on the bond strength of single-lap adhesive joints is still not perfectly understood. The classical elastic analyses predict that the strength increases with the adhesive thickness, whereas experimental results show the opposite. Various theories have been proposed to explain this discrepancy, but more experimental tests are necessary to understand all the variables. The objective of the present study was to assess the effect of the adhesive thickness on the strength of single-lap joints for different kinds of adhesives. Three different adhesives were selected and tested in bulk. The strain to failure in tension ranged from 1.3% for the most brittle adhesive to 44% for the most ductile adhesive. The adherend selected was a high-strength steel to keep the adherends in the elastic range and simplify the analysis. Three thicknesses were studied for each adhesive: 0.2, 0.5, and 1 mm. A statistical analysis of the experimental results shows that the lap shear strength increases as the bondline gets thinner and the adhesive gets tougher.
The viscosity of a soda-lime-silica glass has been measured at the National Bureau of Standards and seven other laboratories. Determinations were made in the range of 102 to 1015 poises. The rotating cylinder was used at the higher temperatures (800 to 1450 °C) and the fiber elongation method at the lower temperatures (520 to 658 °C). The results have been critically evaluated and the glass has been issued as Standard Sample No. 710. This glass is available from the National Bureau of Standards.
This paper is concerned with predicting failure in adhesively bonded joints. After reviewing various methods that have been used for this purpose one technique is presented that in three independent studies has given reliable predictions of joint strength. This is based on a concept termed global yielding, which applies when a path of adhesive along the overlap region reaches the state in which it can sustain no further significant increase in applied load. Using an analysis technique such as the finite element method it is possible to ascertain the load at which this occurs fairly readily; this gives a good estimate of failure load. The three studies cited consist of experimental and analytical programmes on different types of joint. The first is concerned with the single lap joint, where this technique actually demonstrates why the strength of a joint increases with decreasing adhesive thickness. The second study is concerned with failure in double lap joints and the third with a new type of test termed the compressive shear test where it can be shown that loads considerably in excess of those expected can be sustained by eliminating the transverse tensile adhesive stresses.
Lap joints are used extensively in the manufacture of cars. In order to determine the effect of using a structural adhesive instead of spot-welding, a detailed series of tests and finite element analyses were conducted using a range of loadings. The adhesive was a toughened epoxy and the adherend was mild steel typical of that used in the manufacture of car bodyshells. The lap joints were tested in tension (which creates shear across the bondline), four-point loading (pure bending) and three-point loading (bending plus shear). Various parameters were investigated such as the overlap length, the bondline thickness and the spew fillet. The major finding is that three-point bending and tension loading are very similar in the way in which they affect the adhesive while the four-point bend test does not cause failure because the steel yields before the joint fails. A failure criterion has been proposed based on the tensile load and bending moment applied to the joint.
Delo Industrial Adhesives: Glass Bonding: Requirements. Product Range and Design Examples
Delo Industrial Adhesives: Glass Bonding: Requirements. Product Range and Design Examples. In. Delo Industrial Adhesives, Munich (2011)