Mechanical Properties of Sandwich Panels Constructed from

Polystyrene/Cement Mixed Cores and Thin Cement Sheet Facings

Hamid Reza Tabatabaiefar 1*, †, Bita Mansoury 2, Mohammad Javad Khadivi Zand 2 and
Daniel Potter 1
1- Faculty of Science and Technology, Federation University Australia, Australia
2- Faculty of Design, Architecture and Building, University of Technology Sydney (UTS), Australia

ABSTRACT
Sandwich panels are made of two materials that are relatively weak in their separated state, but are
improved when they are constructed together in a sandwich panel. Sandwich panels can be used for
almost any section of a building including roofs, walls and floors. These building components are
regularly required to provide insulation properties, weatherproofing properties and durability in
addition to providing structural load bearing characteristics. Polystyrene/cement mixed cores and
thin cement sheet facings sandwich panels are Australian products made of cement-polystyrene
beaded mixture encapsulated between two thick cement board sheets. The structural properties of
sandwich panels constructed of polystyrene/cement cores and thin cement sheet facings are
relatively unknown. Therefore, in this study, in order to understand the mechanical behaviour and
properties of those sandwich panels, a series of experimental tests have been performed and the
outcomes have been explained and discussed. Based on the results of this study, values for modulus
of elasticity and ultimate strength of the sandwich panels in dry and saturated conditions have been
determined and proposed for practical applications.
Keywords: Sandwich Panels, Polystyrene/Cement Mixed Cores, Thin Cement Sheet Facings, Modulus of
Elasticity, Ultimate Strength

1. INTRODUCTION
The building and construction industry is ever increasing in size and the demand for
houses and industrial buildings all around Australia is currently greater than ever
(Australian Bureau Statistics, 2015a). The growing amount of houses and industrial
buildings being constructed is affecting the amount of building materials being
produced and is influencing the competitive prices needed to complete these
constructions (Australian Bureau Statistics, 2015b). The demand for building materials
has prompted the development of construction products and methods. These
construction products have been carefully thought out with careful considerations for
the future. The environment, the purpose and efficiency of materials are all
considerations that are continually thought of in the development of these materials.
The considerations for the future and the effects on the environment are the main
reasons why people produce construction materials. One of these construction materials
are sandwich panels. Sandwich panels are made of two materials that are relatively
weak in their separated state, but are improved when they are constructed together in a

* Correspondence to: Hamid Reza Tabatabaiefar, Lecturer in Structural Engineering, Federation University Australia,
Office Y231, Building Y, Mt Helen Campus, PO Box 663 Ballarat VIC 3353, Australia, Tel.: +61 3 5327 6718
† E-mail address: [email protected].
sandwich panel (Allan, 1969). Sandwich panels can be used for almost any section of a
building. The sections include the roof, the walls and the floor. These building sections
are regularly required to provide insulation properties, weatherproofing properties and
durability in addition to providing structural load bearing characteristics (Petras, 1998).
It is usually very common to find precast concrete panels as a part of a building’s
composition. Precast concrete panels are normally very strong, but are generally
extremely heavy and difficult to work with.Sandwich construction form has distinct
advantages over conventional structural sections because it promises high stiffness and
high strength-to-weight ratio (Araffa and Balaguru, 2006) as compared with a solid
member. Sandwich composite structure possesses excellent flexural and shear
properties. Their inherent lightweight characteristics make them ideal structural
components where weight reduction is desirable (Serrano et al., 2007). Thus structural
sandwich panels are becoming important elements in modern lightweight construction.
Among the other advantages, its good thermal insulation due to the cellular thick core
makes it an ideal external construction component (Bottcher and Lange, 2006). Some
recent investigations suggest their excellent energy-absorbing characteristics under
high-velocity impact loading conditions (Villanueva and Cantwell, 2004). Sandwich
structures have been considered as potential candidate to mitigate impulsive (short
duration) loads (Nemat-Nasser et al., 2007).
With the building requirements and the disadvantages of precast concrete panels,
sandwich panels provide an alternative to the precast concrete panels. There are many
different types of composite sandwich panels that can be constructed, but one of the
stronger types of sandwich panels is a panel made of a polystyrene/cement core that is
sandwiched between thin cement sheet facings (Davies, 1993). The structural properties
of sandwich panels constructed of polystyrene/cement cores and thin cement sheet
facings are relatively unknown (Mousa & Uddin, 2010). Therefore, there are a number
of questions that could be asked such as how the stress acts within the panel under
loads? How they are affected by moisture and what is the effect of a sustained load?
Besides, the stresses acting within the panel are the stresses acting in the different
materials within the panel. These stresses are unknown and an understanding of the
amount of stress carried by each of the core and the facings components would be
beneficial to provide a better awareness of the stresses acting and possibly provide
information that could be used to improve these panels. As results, in this study, in
order to understand the mechanical behaviour and properties of sandwich panels made
of polystyrene/cement mixed cores and thin cement sheet facings, a series of
experimental tests have been performed at the structures laboratory of Federation
University Australia and the outcomes have been explained and studied in this paper.

2. BACKGROUND
Composite sandwich panels have gradually become more popular due to their typical
benefits including strength, weight, ease of handling, durability, versatility, thermal and
acoustic properties (Gdoutos and Daniel, 2008). Many researchers are aware of these
benefits and have undertaken detailed research and publicised large amounts of scientific
papers on composite panels. Zenkert (1995) introduced sandwich constructions as materials
structured in a way that the bending stresses are supported by the face-sheets and the core
supports the shear loads and stabilises in order to influence wrinkling or buckling acting
against the face. Petrras (1998) investigated the failure modes for sandwich beams with a
range of varied thickness laminate skins and honeycomb cores under a number of loading
combinations to determine skin and core behaviour and comparing the results with
sandwich beam theory. Daienl et al. (2002) elucidated failure modes of composites
sandwich beams which revealed different failures of composite panels. These include
compressive face sheet failure, face sheet debonding, indentation failure and core failure
and face sheet wrinkling. In concrete construction, self-weight of structure itself
represents a very large proportion of the total load on the structures (Mouli and Khelafi,
2006; Tabatabaiefar et al., 2014a,b). Thus, reduction in the self-weight of the structures
by adopting an appropriate approach results in the reduction of element cross-section,
size of foundation and supporting elements thereby reduced overall cost of the project.
The lightweight structural elements can be applied for construction of the buildings on
soils with lower load-bearing capacity (Carmichael, 1986). Reduced self-weight of the
structures using lightweight concrete reduces the risk of earthquake damages to the
structures because the earth quake forces that will influence the civil engineering
structures and buildings are proportional to the mass of the structures and building.
Thus reducing the mass of the structure or building is of utmost importance to reduce
their risk due to earthquake acceleration (Ergul et al., 2004; Tabatabaiefar et al.,
2015a,b).
Gdoutos and Daniel (2008) detailed how composite sandwich panels are made of a core and
two skin layers that are fixed to either side of the core. The core material is generally a
lighter material where the thin skin layers are generally the stronger and stiffer. They
investigated failure modes of composite sandwich beams made of unidirectional
carbon/epoxy facings and PVC closed-cell foam cores which revealed different failures of
composite panels, which is described as dependent upon factors such as the materials
properties of the skin and the core, the dimensions of the test samples and the types of load
combinations applied. Followed by Gdoutos and Daniel (2008), Manalo (2010) compared
the failure mechanism and ultimate strength of structural fibre composite sandwich beams
loaded in an edgewise position compared to a flatwise position and acknowledges the
contribution of the strength of the core material on flexural and shear stiffness of the
composite sandwich while Plain (2009) studied the behaviour of fibre composite sandwich
panels for use in the civil infrastructure and building industry through a number of varied
point and uniformly distributed load testing conditions to gain material properties for
development of analytical design modelling. The above research outcomes haves
highlighted a number of differing properties of the derivatives of sandwich panels and
identified a number of differing failure modes to consider in conducting destructive testing
models for the thin faced reinforced concrete with polystyrene concrete core panels being
tested and analysed.
 3. POLYSTYRENE/CEMENT CORE & CEMENT FACINGS SANDWICH
PANELS
Polystyrene/Cement Core and Cement Facings Sandwich Panels (Figure 1a) are Australian
products made of cement-polystyrene beaded mixture encapsulated between two 6mm thick
cement board sheets. They can be purchased in thicknesses of 50mm thick or 75mm thick and
are constructed with a width of 600mm and a range of lengths, with the maximum length being
6.0m. The panels are typically produced in lengths of 2600mm and 3000mm. The panels are
normally orientated so that the short edge is at the bottom or top and are installed by sliding the
panel between top and bottom steel channels and are fixed together with a grooving fixture and
a tongue (Figure 1b). These types of panels are used by the industry as they are light-weight
and easier to handle/install than more traditional panel designs. The panels are easily
transported and handled on-site reducing the cost of construction and they are termite, pest and
rot resistant contributing to its long term performance in buildings. In addition, the panels have
better thermal, acoustic and fire resistant properties than traditional building materials used in
the building industry (Zenkert, 1995).
Reviewing the past research outcomes in Section 2, a greater understanding of the makeup
and considerations for design and use of sandwich panels has been developed encouraging
the scope of this study to be refined with the aim of determining three main components:

• Investigating the true stress distributions acting within the sandwich panels under
the applied loads;
• Determining the effects of moisture content of the mechanical properties of the
panels; and
• Elucidating the effects of sustained loads (creep effects).
The ultimate goal of this study is to investigate and analyse the structural properties of
composite sandwich panels comprising thin outer cement sheeting layers on either sides of
the core having cement paste and polystyrene beads. The investigation and analyses will
further develop an understanding of the behaviour and properties of Polystyrene/Cement
Core and Cement Facings Sandwich Panels.

4. MATERIAL CHARACTERISTICS
The flexural stiffness of a material is measured with the Modulus of Elasticity (E). The
modulus of elasticity for a rectangular beam that is simply supported with a mid-span
load (P) can be determined by using the equation for maximum deflection and the slope
of the load versus the displacement graph, where the slope is within the linear zone of
the graph as shown in Figure 2. The maximum deflection equation for a simply
supported rectangular beam is:
PL3
D= (1)
48 EI
where D is the maximum deflection of the beam; P is the mid-span load; E is the
modulus of elasticity and I is the moment of inertia of the section.
By rearranging Equation 1, Hilbbeler (1994) presented Equation 2 for determining the
modulus of elasticity:
P L3
E = ( )( ) (2)
D 4ab 3
ab 3
where, I = and a and b are the cross sectional dimensions, respectively.
12
The flexural strength of the panel can be calculated from simple bending theory. For a
simply supported beam with a midpoint load, the equation for stress with relation to
bending is:
My
σ= (3)
Ix
where, σ is the normal stress; M is the bending moment; y is the distance from neutral
axis and I x is the moment of inertia for a rectangular section.
Equation 3 is called the Bending Formula (Gross et al., 2011) which assumes a
homogeneous material with stress diagram shown in Figure 3a. As the panels are not
homogeneous, the stress distribution within the panel looks more like the stress diagram
in Figure 3b. To determine the stresses acting within a sandwich panel, the elasticity
modulus of the separate core component and the separate face sheeting component are
needed. As the bending formula assumes a uniform material, one of these different
material areas need to be changed into effective area of the other materials modulus.
This is done by using a ratio of the modulus of elasticity's to convert the area of one of
the materials. Figure 4a illustrates a normal sandwich panel with materials of different
moduli and Figure 4b shows a converted effective area with material of one elastic
modulus (Gross et al., 2011). The ratio used to convert the smaller elastic modulus into
the larger elastic modulus to create Figure 4b (Gross et al., 2011) is:
E Poly
× bPoly = bef (4)
EConc
where, E Poly is the elastic modulus of the polystyrene material; E Conc is the elastic
modulus of the concrete material; b poly is the width of the polystyrene material and b ef is
the effective width.
The moment of Inertia (I) can then be calculated for the new effective shape by using
the "parallel axis theorem" (Hibbeler, 1994). This can be done using Eq. 5 as follows:
I Total = ∑ ( I i + Ai d i2 ) (5)
where, I i is the moment of inertia of an individual shape about its own centroid axis, A i
is the area of an individual shape and d i is the distance of an individual shape to the
neutral axis.
This total moment of inertia (I Total ) can be used along with the calculated bending
moment at mid-span in the "Bending Formula" (Eqn. 3) to give the max stresses in the
top and bottom of the panel (Hibbeler, 1994). The stresses at the join between the
sheeting and core can be also calculated. These calculated stresses then may be
converted into strain using the effective shapes modulus by using:
Stress
E= (6)
Strain
These strains are then multiplied by the modulus of the existing core material to
produce stress diagrams similar to like Figure 3a.

5. MATERIAL TESTING PROGRAM

The testing of the samples in the testing program has been broken up into different
stages. These stages were the sample preparation, testing for true stresses, testing for
effect of moisture and the creep testing. These stages are outlined below.
5.1. Test Sample Preparation
The testing of samples required large 1.5m×0.6m panels to be cut to a size that could be
managed by the testing apparatus at the structures laboratory of Federation University
Australia. Several samples were required to be separated into their different components
(core and facing). The sample preparation was a large task and therefore was broken up
into three different sections. These were the cutting of the whole sample panels, cutting
of the core components and cutting of the facing components. The cutting of these
materials was conducted with the use of a circular saw with a diamond tip blade which
could cut through 75 mm think panels, a drop saw with a diamond tip blade which could
cut through 75mm thick panels and had a reach of 300mm minimum and, an angle
grinder with a stone disk. Protective equipment of ear muffs, glasses and a breathing
mask were also worn due to the amount of dust and debris. The panels were first
carefully measured with a tape and a square. They were then clamped and cut using the
circular saw (Figure 5) and then re-measured to make sure the sample was within a ±
1mm. All sandwich samples were cut and prepared using this method. The next stage in
cutting used the drop cut saw to either cut the sheeting off the core (Figure 6) if
requiring separate core component or cut the core off the sheeting if requiring the
sheeting component. This process has completed the separating of the core component
and the sheeting has been discarded. The final cutting of the samples required the use of
the angle grinder to finely grind off the excess core material to leave just the cement
sheeting (Figure 7). All of these samples were produced with millimetre accuracy to
provide accurate results as illustrated in Figure 8.
5.2. Stresses within the Panel Testing
The stresses acting within the sandwich panels were researched by conducting several
tests. These tests included testing of whole specimens, testing on the separated core
specimens and testing on the separated facing specimens. These tests determined the
flexural strength and the flexural stiffness of the materials. The three types of
specimens were all tested in bending in a horizontal flexural test. The horizontal
flexural tests for the whole specimens were carried out in the structures laboratory of
Federation University Australia. The test sample size and the arrangement were guided
by ASTM International (2012). The standards stated that the standard size of testing sample
of sandwich nature to be 600mm×75mm and have either a 4-point loading configuration
or a 3-point loading configuration.
5.2.1. Whole Sample Testing
Firstly, the 3-point loading configuration was used with a 600mm×75mm sample
spaned 560mm with a midpoint load. The equipment used to setup the first sample in
the mid-span loading included two support posts, a clamped steel beam between the two
supports to make the support posts ridged, one locked simple support roller, one simple
support roller, hanger that supports the load across the sample evenly (Figure 9a),
hanger to support the load (Figure 9b), three deflectometers and three small pieces of
flat steel. The three deflectometers were setup with one at each end of the sample over
the support and one at 15mm from mid-span due to the hanger preventing the
deflectometer from being at mid-span (Figure 10a). This small variance from mid-span
was estimated to give very little error and was ignored. The sample was also setup with
a small piece of 40×6×100mm flat steel under the supports to prevent the supports from
crushing into the sample and giving a wrong reading at the mid-span (Figure 10b). The
middle hangers also included a small 20×10×100mm flat steel to prevent the round
hanger pressing into the sample and causing a premature failure.
After the sample was setup in the apparatus, the sample was observed for deflection due
to self-weight, deflection due to load of hanger 1 (Figure 9a) and deflection due to load
of hanger 2 (Figure 9b). The increasing load on the sample was then applied by placing
8 kg weights on hanger 2 and recording the deflections on all of the deflectometers. The
recorded deflections on the deflectometers over the supports were generally very little
and have been averaged and subtracted from the total deflection at the supports.
As the second testing method, testing the whole samples was undertaken using the
Automax Smart Line control system with a flexure frame made by Controls Group
(Figure 11). This compression/flexure machine uses a hydraulic pump and ramp to
supply a load of between 0kN and 5000kN (Controls Group, 2014). According ASTM
International (2012) the test samples were arranged in a 2-point loading and 2 support load
configuration. The standards stated the 4-point loading configuration as a sample
spaned 560mm across a simply support and a roller support with the middle 2 loads
100mm apart. The 4-point loading apparatus was also setup (Figure 12) with three
deflectometer's, one at each end over the supports, and one at mid-span. As illustrated
in Figure 12, the mid-span deflectometer was unable to be put directly in the middle of
the sample due to the loading ram. This issue was overcome by clamping a light piece
of angled aluminium to the top of the sample that protruded to the left of the sample so
that the delectometer could be attached. This arrangement was considered as having
very minimal effects on the results. The apparatus also included two flat pieces of steel
(7mm thick) that were placed between the sample and the round support (Figure 13).
These flat pieces of steel were placed there to reduce the round supports from
compressing up into the sample and causing inaccurate results. The deflection due to
self-weight was also observed and recorded. The loads were then applied in 0.5kN
increments and deflections were recorded. It should be noted that it means each time
0.5kN has been added then the corresponding deflection was measured and recorded.
The two deflectometers over the supports were averaged and substituted from the mid-
span deflection. The deflection verse load was plotted.
The third and final method used to in testing of the whole samples was using the
Automax control system with the flexure frame with 3-point arrangement as shown in
Figure 14. The load was applied in 0.5kN increments and the deflections recorded until
the failure of the sample. The deflection versus load was plotted and the slope of the
elastic region of the graph was used in the rearranged deflection equation (Eqn. 2) to
determine the elastic modulus (E). As mentioned in Section 4, the employed panels are
not made of homogeneous materials. Therefore, the calculated moduli of elasticity for
the whole panels are not the true moduli. The true moduli of the materials within the
sandwich panels were calculated by testing the separated core component and testing of
the separated face sheeting component as will be explained in Section 5.2.2. Eventually,
the flexural strengths of the panels were determined by finding the stresses acting
within each material by means of the method explained in Section 4.
5.2.2. Core Sample Testing
After initially testing the whole samples, it was clear that the testing machine would not
be ideal for the testing of the separate sections due the increments of load that the
machine could be applied (only dealing with larger increments). Therefore a different
apparatus was required. The utilised apparatus was a 3-point loading apparatus with
manual weights connected to the hangers at the mid-span, which was the same setup as
the first method of testing the whole samples, explained in Section 5.2.1. As shown in
Figure 15, the 3-point apparatus used the three deflectometer's with two at the supports
and one at the mid-span located 15mm next to the hanger. The same flat steel plates
(7mm thick) were also utilised under the supports. Deflections due to self-weight, load
of the hangers and an increasing 1kg weights at mid-span were measured and recorded.
The loads versus deflections graphs were plotted afterwards and the moduli of elasticity
were determined according to Section 4.
5.2.3. Face Sheeting Sample Testing
The face sheeting samples testings have been performed using the same 3-point loading
configuration as the core component tests (Figure 16). Deflections due to self-weight,
load of the hangers and an increasing 0.4kg weights at mid-span were measured and
recorded. The loads versus deflections graphs were plotted afterwards and the moduli of
elasticity were determined according to Section 4.
5.3. Effects of Moisture Content
The tests on the effects of water were conducted in a similar apparatus as the tests of the
stresses within the panels. The tests were performed by soaking the whole samples over
a 6 month period, testing the bending strength of the panels and eventually comparing
the results with the results obtained from the dry samples. The samples were 600mm×
75mm × thickness of the panel (75mm). These were weighed and recorded before
submerging the samples in order to measure the amount of absorbed moisture. Figure
17 shows a saturated sample in 3-point loading using Automax control system and the
flexure frame apparatus. One interesting observation was that samples floated when
they were placed in the water and bricks were needed to fully submerge them (Figure
18).
5.4. Creep Effects
In order to determine the flexural creep of the material, tests were performed according to
ASTM International (2008). The effect of creep was planned to be observed by creating 6
creep apparatuses and testing the effect of a load over a 10 week period to determine the
creep rate of the panel. Preliminary destruction testing of the creep test panels was
undertaken by first testing four 900×75×75 mm simply supported samples (860mm span)
to obtain the average ultimate strength of the panels (1.4kN). It should be noted that the
main reason for testing four samples in the testing program was to show repeatability of the
test results and to achieve reliable average values. Other research was considered along
with the panel's modulus of elasticity to determine the appropriate percentage of the
ultimate strength loading to be placed on the creep test samples.

6. RESULTS AND DISCUSSION

The experimental measurements of each set of tests, explained in Section 5, are outlined
in this section. The load-deflection results for four different core samples and four
different face sheeting samples are measured and shown in Figures 19 and 20. From
those Figures, the slope of each load-deflection graph has been determined and the
modulus of elasticity was calculated. The calculated moduli of elasticity and ultimate
strengths for core and face sheeting samples are summarised in Tables 1 and 2. Based
on the summarised results, the average values for modulus of elasticity and ultimate
strength of the core samples are 400 MPa and 210 N, respectively while these values are
determined to be respectively 10890 MPa and 47.8 N for the face sheeting samples.
Figure 21 shows the whole sample tests results, were tested in a 4-point loading
arrangement with the Automax controls system and flexure flame. These results have
not been taken into consideration as several data points were predicted to be incorrect
due to the apparatus and movement of the samples. The issues involved with the 4-point
loading configuration were overcome with the test in the 3-point loading configuration.
The results from the 3-point configuration for the 600×75mm samples are illustrated in
Figure 22. From this Figure, the slope of each sample was determined and the modulus
of elasticity for each sample has been calculated accordingly. The calculated moduli of
elasticity and ultimate strengths are summarised in Table 3. It should be noted that it
was chosen to work out the elastic modulus for each sample instead of plotting one
trend line for all the data. This is due to the material being non-homogenise and a range
of moduli would be a better result for the other researchers to use in the future. As
mentioned in Section 4, the modulus of elasticity was determined by plotting all of the
samples tested and plotting a trend line (line of best fit) and determining the load-
deflection slope to sub into the equations. Based on the summarised results in Table 3,
the average values for modulus of elasticity and ultimate strength of the whole samples
are 4180 MPa and 2000 N, respectively. These values are recommended be adopted for
practical design applications.
According to Zhu and Chai (2013), three main failure modes for sandwich panels exist,
namely, facing fracture, core shear yield/fracture and overall core crushing. The mode
in which a sandwich panel structure fails on loading can be determined by establishing,
for a given panel geometry and materials properties, which mechanism has the lowest
failure load. This information can be plotted on a failure diagram where regions of the
diagram correspond to a given failure mechanism. In order to determine the failure
mode of the sandwich panels in this study, the stresses within a panel were determined
by using the method described in Section 4. Using the modulus of 400MPa for the core
material and 10,900MPa for the face sheeting, the true stress diagram has been plotted
for sample A10 at failure (Figure 23). As shown in Figure 24, the typical failure mode
of the 600×75mm sandwich panel occurred when the bottom cement facing failed
leading it to crack and protrude upwards. Therefore, the main failure mode of the
studied panels is facing fracture. As the thickness of facings of the examined panels are
thin compared to the other dimensions of the panels and considering that the cores are
of sufficient stiffness, the global deflections of the sandwich panels tend to be large and
as a result, there arise high in-plane tensile forces that will cause tensile cracking. As a
result, the cores have insignificant effect on the failure of the panels in bending until the
outer tension faces fail causing the cores to fail in a catastrophic mode.
The results from the fully saturated panel samples are shown in Figure 25. Similar to
the previous components, the slope of each load-deflection graph has been found and
the modulus of elasticity was determined. The calculated moduli of elasticity and
ultimate strengths are summarised in Table 4. Based on the summarised results, the
average values for modulus of elasticity and ultimate strengths of the fully saturated
panel samples are 4466 MPa and 1312 N, respectively. These results show that the fully
saturated panels possess larger moduli of elasticity. However, due to saturation, their
ultimate strengths reduce substantially.
Preliminary destruction testing of the creep test panels were undertaken by testing four
900×75×75mm simply supported samples that were to obtain the average ultimate
strength of the panels. The percentage of ultimate strength loading was determined to be
75% of the ultimate strength load, as it was appropriately within the range of previous
research conducted being 65% for wood and 85% for concrete. Three preliminary creep
tests were conducted by using a set of constructed apparatus that simply supported the
samples over a 2 week period. The samples were setup with the 75% of the ultimate
strength at the midpoint with the deflection recorded while the time increased. Based on
the summarised results in Table 5, the 75% of the ultimate strength at the midpoint is
1040N. It is understood that the creep testing was unable to provide conclusive results
as the deflection results from preliminary testing did not produce significant deflection
of the samples to measure enough to undertake the longer term tests.
 7. CONCLUSIONS AND RECOMMENDATIONS
In this study, the true stresses acting within typical cement sheet faced cemented
polystyrene core composite sandwich panels have been measured and studied. The main
finding that this investigation has highlighted is the strengths of the composite sandwich
panels in bending are highly dependent on the strength of the outer cement face
sheeting. Primarily, the ultimate strengths can be determined by determining the tensile
strengths of the tension faces of the panels. Therefore, the cores have insignificant
effect on the failure of the panels in bending until the outer tension faces fail causing
the cores to fail in a catastrophic mode. Based on the results of this study, the average
values for modulus of elasticity and ultimate strength of polystyrene/cement core and
cement facings sandwich panels are 4180 MPa and 2000 N, respectively. These values
can be adopted for practical applications. In addition, saturating the sandwich panels
reduce the strength of the outer facings resulting in 28% reduction in the total strength
compared to dry samples.
The creep testing in this study was unable to provide conclusive results as the deflection
results from preliminary testing did not produce significant deflection of the samples to
measure enough to undertake the longer term tests. As a result, further investigation
could be undertaken on the creep of the panels with consideration to the size of the
panel to get greater deflections and more discernible results.
Results of some recent research works on combination of cement, fibre and clay (Fatahi
et al., 2012; Fatahi et al., 2013) show that addition of fibre can increase the strength and
stiffness and reduce the shrinkage potential of the composite material. Therefore, it is
recommended that further studies and experimental works to be carried out on the
effects of addition of fibre on the strength, stiffness and shrinkage potential of
Polystyrene/Cement Core and Cement Facings Sandwich Panels.
References

Allan, H. (1969). Analysis and Design of Structural Sandwich Panels. London, United
Kingdom: Pergamon Press Ltd.
Araffa, M. and Balaguru, P.N. (2006). Flexural Behaviour of High Strength-High
Temperature Laminate Sandwich Beams. Proceedings of Eight International
Symposium and Workshop on Ferrocement and Thin Reinforced cement
Composites,6-8 February, Bangkok Thailand, IFS.189-201.
ASTM International. (2008). Standard Test Method for Flexure Creep of Sandwich
Constructions. West Conshohocken: American Society for Testing and Materials.
Retrieved March 5, 2014, from Sai Global.
ASTM International. (2012). Standard Test Method for Facing Properties of Sandwich
Constructions by Long Beam Flexure. West Conshohocken: American Society for
Testing and Materials. Retrieved March 31, 2014.
Australian Bureau Statistics. (2015a). Population Clock. Retrieved 2015, from Australian
Bureau Statistics:
http://www.abs.gov.au/ausstats/abs%40.nsf/94713ad445ff1425ca25682000192af2/1
647509ef7e25faaca2568a900154b63?OpenDocument.
Australian Bureau Statistics. (2015b). Producer Price Indexes. Retrieved 2015, from
Australian Bureau Statistics:
http://www.abs.gov.au/AUSSTATS/[email protected]/Lookup/6427.0Main+Features1Mar
%202015?OpenDocument.
Bottcher, M. and Lange, J. (2006). Sandwich Panels with Openings. Composite
Construction in Steel and Concrete, 186(14): 137-46.
Carmichael, J (1986). Pumice Concrete Panels. Concrete International. 8(11): 31– 32.
Controls Group. (2014). Compression Machines. Retrieved 2015, from Controls Group
Web site: http://www.controls-group.com/eng/concrete-testing-equipment/automax-
smart-line-fully-automatic-tester.php
Daniel, I., Gdoutos, E., Wang, K., & Abot, J. (2002). Failure Modes of Composites
Sandwich Beams. International Journal of Damage Mechanics, 16, 309-334.
Davies, J. (1993). Sandwich Panels. Thin Walled Structures, 16, 179-198.
Ergul, Y., Cengiz, D.A., Aleaettin, K. and Hassan, G. (2003). Strength and Properties of
Lightweight Concrete Made with Basaltic Pumice and Fly Ash. Materials Letters.
57(15): 2267-2270.
Fatahi, B, Khabbaz, H. and Fatahi, B. (2012). Mechanical Characteristics of Soft Clay
Treated with Fibre and Cement. Geosynthetics International. 19(3): 252–262.
Fatahi, B, Fatahi, B, Le, T.M. and Khabbaz, H. (2013). Small-Strain Properties of Soft Clay
Treated with Fibre and Cement. Geosynthetics International. 20(4), 286–300.
Gdoutos, E., & Daniel, I. (2008). Failure Modes of Composite Sandwich Beams.
Theoretical and Applied Mechanics, 35(1-3), 105-118.
Gross, D., Hauger, W., Schröder, J., A.Wall, W., & Bonet, J. (2011). Engineering
Mechanics 2. Springer Heidelberg Dordrecht London New York.
Hibbeler. (1994). Mechanics of Materials (2ed.). Macmillan.
Mouli, M. and Khelafi, H. (2006). Strength of Short Composite Rectangular Hollow
Section Columns Filled with Lightweight Aggregate Concrete. Engineering
Structures. 29 (8):1791–1797.
Mousa, M., & Uddin, N. (2010). Global Buckling of Composite Structural Insulated Wall
Panels. Materials and Design, 32(2), 766-772.
Nemat-Nasser, S., Kang, W.J, Mc Gee, J.D., Guo, E.G. and Isaacs (2007). Experimental
Investigation of Energy-Absorption Characteristics of Components of Sandwich
Structures. International Journal of Impact Engineering, 34(6): 1119-1146.
Petras, A. (1998). Design of Sandwich Structures. Cambridge University Engineering
Department, 1-100.
Plain, M. (2009). Behaviour of Innovative Fibre Composite Panels for Structural
Applications. University of Southern Queensland, 1-106.
Serrano, P. J.C, Uday, K.V. and Nasim, U. (2007). Low Velocity Impact Response of
Autoclaved Aerated Concrete/CFRP Sandwich Plates. Composite Structures. 80(4):
621-630.
Tabatabaiefar, H.R., Fatahi, B. & Samali, B. (2014a). An Empirical Relationship to
Determine Lateral Seismic Response of Mid-Rise Building Frames under Influence
of Soil-Structure Interaction. The Structural Design of Tall and Special Buildings,
23(7): 526-548.
Tabatabaiefar, H.R., Fatahi, B., & Samali, B. (2014b). Numerical and Experimental
Investigations on Seismic Response of Building Frames under Influence of Soil-
Structure Interaction, Advances in Structural Engineering, 17(1): 109-130.
Tabatabaiefar, H.R., Mansoury, B. & Khadivi Zand, M.J. (2015a). Experimental
Investigations on Behaviour of Steel Structure Buildings. Proceedings of the 10th
Pacific Conference on Earthquake Engineering, Australian Earthquake Engineering
Society, 6-8 November 2015, Sydney, Australia, Paper Number 160.
Tabatabaiefar, H.R. & Mansoury, B. (2015b). Detail Design, Building and Commissioning
of Tall Building Structural Models for Experimental Shaking Table Tests. The
Structural Design of Tall and Special Buildings, DOI: 10.1002/tal.1262, Wiley-
Blackwell.
Villanueva, G.R. and Cantwell, W.J. (2004). The High Velocity Impact Response of Composite and
FML-Reinforced Sandwich Structures. Composite Science and Technology. 64(1): 35-54.
Zenkert, D. (1995). An Introduction to Sandwich Construction. West Midlands, United
Kingdom: Engineering Materials Advisory Service Ltd.
Zhu, Z. & Chai, G.B. (2013). Damage and failure mode maps of composite sandwich panel
subjected to quasi-static indentation and low velocity impact. Composite Structures.
101:204-214.
 (a)

(b)

Figure 1: Cement/Polystyrene Core & Cement Sheet Facing Sandwich Panels

Figure 2: Typical Load-Displacement graph
 (a) (b)

Figure 3: (a) Stress diagram of homogeneous materials; (b) Estimated stress diagram of Sandwich Panels (Gross et al.,
2011)
 (a) (b)

Figure 4: Cross section of sandwich panels; (a) typical sandwich panels constructed from polystyrene & concrete with
different elastic moduli; (b) converted sandwich panels with one elastic modulus Panels (Gross et al., 2011)
Figure 5: Measuring and cutting of full panels
Figure 6: Cutting cement sheeting off core
Figure 7: Grinding off the excess core material
Figure 8: Prepared test samples
(a) (b)

Figure 9: (a) Hanger 1; (b) Hanger 2

 (a) (b)
Figure 10: (a) Support with flat steel; (b) Mid-span deflectometer
Figure 11: Automax Smart line control system with compression and flexure frames
Figure 12: Setup in horizontal flexure test machine
Figure 13: Flat steel between sample and the round support
Figure 14: 3-point test in Automax control flexure frame
Figure 15: 3-point testing apparatus of core
Figure 16: Face Sheeting tensile testing apparatus
Figure 17: Saturated sample in 3-point loading using Automax control system and the flexure frame apparatus
Figure 18: Submerged samples with bricks
Figure 19: The load-deflection results for four different core samples
Figure 20: The load-deflection results for four different sheeting samples
Figure 21: 4-Point Loading Results for three different whole samples
Figure 22: 3-Point loading results for four different whole samples
Stress Diagram at Failure (Sample A10)

Stress (MPa)

Figure 23: True Stress distribution within the whole panel

Figure 24: Typical failure of 600mm x 75mm sample in 4-point loading
Figure 25: 3-Point loading results for four different saturated samples
Figure 26: Results of Preliminary Creep Destruction Tests
Table 1: Moduli of elasticity and ultimate strengths for four different core samples
y/x
Sample E (MPa) Ultimate Strength (N)
(N/mm)

C1 157.92 369.7 228

C2 273.23 639.7 218
C3 113.07 264.7 188.5
C4 139.44 326.4 208
Average 400.1 210.6
Table 2: Moduli of elasticity and ultimate strengths for four different face sheeting samples
y/x
Sample E (MPa) Ultimate Strength (N)
(N/mm)

D5 3.8361 10396.3 48.6

D1 3.9397 10677.1 58.5
D3 4.033 10929.9 38.95
D4 4.2665 11562.7 48.6
Average 10891.5 47.8
Table 3: Moduli of elasticity and ultimate strengths for four different whole samples
y/x
Sample E (MPa) Ultimate Strength (N)
(N/mm)

A10 2796.4 3880.2 2200

A14 3379.1 4688.8 1900
A12 3103.6 4306.5 1950
A13 2772.7 3847.4 2000
Average 4180.7 2000
Table 4: Moduli of elasticity and ultimate strengths for four different saturated whole samples
y/x
Sample E (MPa) Ultimate Strength (N)
(N/mm)

B3 3237.6 4492.4 1300

B4 3375.2 4683.4 1250
B5 3098.6 4299.6 1350
B6 3163.8 4390.0 1350
Average 4466.4 1312.5
Table 5: Preliminary creep test results

Sample Ultimate Strength (N)

S1 1350
S2 1550
S3 1350
S4 1300
Average 1387.5
75% 1040.625