Flexural behavior of FRP strengthened concrete-wood composite beams

Abstract

An experimental investigation was conducted to study the structural performance of concrete-timber composite beams provided with systems of semi-rigid connectors of various configurations and stiffness. Eleven concrete-timber composite beams and one concrete-steel composite beam were tested in three-point flexure test. The construction of the beam web and the type of shear connector were the basic parameters of investigation. The feasibility of boosting the flexural performance of these beams when wrapping their webs with GFRP laminates was examined. Comparison of the flexural behavior of the proposed composite concrete-timber beams with that of both of post-stressed beams using external steel longitudinal rods and a concrete-steel composite beam was also evaluated. Test results showed that vertical glulam timber web is superior to both horizontal glulam and massive block timber webs in providing higher stiffness and load carrying capacity and larger toughness. Also, dovetail shear connector enhanced with insertion of a steel dowel at the center of the groove imparts a full composite action between concrete and timber web. Use of GFRP laminates wrapping along the full span of timber web is a highly effective method to take advantage of the use of enhanced dovetail shear connector. The behavior of the proposed concrete-timber composite beam was comparable to that of the concrete-steel web beam.

Full text

Flexural behavior of FRP strengthened concrete-wood composite beams
Yehia A. Zaher Ali
Faculty of Engg., Ain Shams University, Egypt
article info
Article history:
Received 13 July 2017
Revised 3 June 2018
Accepted 12 June 2018
Available online 26 November 2018
Keywords:
Shear connector
Timber
Glulam
Composite
Wrapping
Flexure
abstract
An experimental investigation was conducted to study the structural performance of concrete-timber
composite beams provided with systems of semi-rigid connectors of various configurations and stiffness.
Eleven concrete-timber composite beams and one concrete-steel composite beam were tested in three-
point flexure test. The construction of the beam web and the type of shear connector were the basic
parameters of investigation. The feasibility of boosting the flexural performance of these beams when
wrapping their webs with GFRP laminates was examined. Comparison of the flexural behavior of the pro-
posed composite concrete-timber beams with that of both of post-stressed beams using external steel
longitudinal rods and a concrete-steel composite beam was also evaluated. Test results showed that ver-
tical glulam timber web is superior to both horizontal glulam and massive block timber webs in provid-
ing higher stiffness and load carrying capacity and larger toughness. Also, dovetail shear connector
enhanced with insertion of a steel dowel at the center of the groove imparts a full composite action
between concrete and timber web. Use of GFRP laminates wrapping along the full span of timber web
is a highly effective method to take advantage of the use of enhanced dovetail shear connector. The
behavior of the proposed concrete-timber composite beam was comparable to that of the concrete-
steel web beam.
Ó2018 Ain Shams University. Production and hosting by Elsevier B.V. This is an open access article under
the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Reinforced concrete, steel or timber are those traditional mate-
rials used in the manufacture of structures as they verify the
requests of the serviceability limit state and the ultimate limit
state concepts. Since concrete is inherently a durable material, it
is a ubiquitous material, which has a long history in buildings,
highways, dams, sidewalks, and even artworks. The Romans
invented cement based concrete more than 2000 years ago and
used the material to build architectural masterpieces such as the
Pantheon [1]. Steel is the material of choice for commercial or
industrial truss construction because it is strong, readily available,
easily fabricated, and not excessively costly. Timber is one of the
earliest construction materials, and the structural use of wood
and wood-based materials continues to steadily increase. It is
becoming recognized as a highly attractive structural material for
large-scale building projects throughout the world. The orthotropic
nature of the timber imparts unique and independent mechanical
properties in the directions of three mutually perpendicular axes:
longitudinal, radial, and tangential.
The technique to associate these materials as a composite struc-
ture has great importance in the direction to search better
exploitation of the mechanical properties of each one of these
materials. The use of composites enhances some favorable proper-
ties such as stiffness, strength, toughness, and strain capacity. On
the other hand, it reduces some unfavorable properties such as
water absorption, permeability to gases and liquids, weight, and
manufacturing costs, installment and maintenance [2–4]. In this
constructive technique, the following associations are possible:
timber-concrete, steel-concrete or timber-steel, thus making possi-
ble, to exceed some inherent limitations of each one of these
materials.
Of these systems, timber-concrete composites are the easiest
and most prospective ones in the construction sector. The most
common shape of these composites is concrete deck associated
with timber web to form composite T-section beam. The timber
provides the tensile strength for the concrete working in compres-
sion. In building design such a system offers performance and con-
struction benefits. Technical committee report [5] stated that such
systems have been used in bridge constructions worldwide and
https://doi.org/10.1016/j.asej.2018.06.003
2090-4479/Ó2018 Ain Shams University. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer review under responsibility of Ain Shams University.
Production and hosting by Elsevier
E-mail addresses: yeh1612@yahoo.com,yeh1612@gmail.com
Ain Shams Engineering Journal 9 (2018) 3419–3424
Contents lists available at ScienceDirect
Ain Shams Engineering Journal
journal homepage: www.sciencedirect.com

that they proved success in service. The enhancement of the load
carrying capacity of timber-concrete composite floors for use in
traditional construction and bridge decks has been investigated
by Yttrup [6]. A comprehensive testing programme carried out
by Baldcock and McCullough [7] highlighted the technical and eco-
nomical feasibility of using composite timber and concrete bridge
construction. Different shear connectors such as steel dowels, steel
rings and notch type were investigated in their study. Since then,
various researches on the effectiveness and behavior of timber-
concrete composite section using different configurations of shear
connectors have been carried out [8–13].
Moreover, the incorporation of fiber-reinforced polymer (FRP)
composite in the strengthening techniques for wood glulam beams
is not new, and has been investigated by several researchers
[15,14]. However, relatively little work has been done on FRP-
reinforced beams with composite concrete slabs. Hence, the work
done in this research focuses on taking advantage of FRP reinforc-
ing applied to timber-concrete composite beams prepared using
different configurations of shear connectors.
2. Researches significance
The program has investigated the structural performance of
timber-concrete beams enhanced with systems of semi-rigid con-
nectors of various configurations and stiffness. The main objective
of the study was to examine the feasibility of enhancing the flexu-
ral performance of these beams when wrapped with GFRP lami-
nates. The evaluation depended on the values of the load
carrying capacity, the corresponding longitudinal strain, the mid-
span deflection and the shear deformations between the timber
web and the concrete flange.
The scope of the experimental program was planned to investi-
gate the continuity of the wrapping system along the span of the
beam. Comparison studies were included in this research. Post-
stressing of beams using external steel longitudinal rods was used
for the first comparison with FRP wrapped beams. The other com-
parison was carried out using steel-concrete composite beam. The
output of this study will open up new possibilities for the construc-
tion of highly efficient timber-concrete beams. Additional cost sav-
ings are also envisaged by reducing the cross-sectional area
required for the same design load.
3. Experimental study
As a scope of the experimental program to investigate the struc-
tural behavior of the tested specimens, 11 timber-concrete beams
and 1 steel-concrete beam were prepared for testing. The study
was accomplished through three stages in which stage I, consisted
of three beams, investigated different compositions of timber web.
Choosing the vertical glulam web (as an output of stage I), stage II,
consisted of six beams, handled the type of shear connector as well
as the use of FRP wraps to the timber web. Stage III, consisted of
three beams, was implemented to study the effect of external
stressing applied to the vertical glulam web and the massive tim-
ber web. Another objective of this stage was the monitoring of the
overall efficiency of using timber-concrete composite beams as
good alternatives for steel-concrete ones. Beyond the research pro-
gram proposed, the materials used in the construction of these
structural elements were characterized and described in this
section.
3.1. Materials
Cement:Portland cement complying with [17] was used in the
current research.
Aggregate:Crushed dolomite stone, 16 mm maximum nominal
size, was used as coarse aggregate. Natural silicate sand was
used as fine aggregate. The aggregates used complied with [18].
Water:Tap water was used as mixing water.
Timber: Pine timber commercially available was procured for
the entire experimental study from single source. The timber
was machined to the desired dimensions using electrical saws.
FRP System:The FRP system used in the current work consists
of bi-directional E-glass woven fabric laminates and polyester
resin as an impregnation resin. The system is provided by
Sika-Egypt Company. The properties of the glass fibers used
based on the product data sheet of the supplier are shown in
Table 1.
3.2. Test specimens
The timber-concrete ‘‘T” beams, of span 1.10 m between sup-
ports, were constituted of the web in pine timber (7.5 cm 15 cm)
and the cross-section of the flange in concrete (30 cm 6 cm). The
steel-concrete beam, of the same span and flange, was prepared by
replacing the timber web with standard steel I-beam web (IPE
100). The longitudinal reinforcement of the flange for all specimens
was five 8 mm diameter bars while the stirrups were 6 mm diam-
eter bars every 140 mm. Typical cross section of test beams is
shown in Fig. 1.
The designed slump of concrete used for the flange was 50 mm.
Test cubes were cured in water before testing at the ages of 7 and
28 days. The average compressive strength of the tested cubes after
28 days was 29 MPa. The study was fulfilled through three stages
as follows:
Stage I:Three beams were prepared for this stage to study the
efficiency of three different compositions of the timber web,
namely, vertical glulam, horizontal glulam and massive timber.
Hooked steel stud was implemented in this stage as a shear con-
nector at intervals of 10 cm along the span of the beam. The details
of beams used in this stage are shown in Fig. 2.
Stage II:As an output of stage I, vertical glulam web was chosen
for completing other stages of the study. The parameters of study
in this stage included the type of shear connector as well as encas-
ing the timber web with either discrete or continuous FRP lami-
nates. The shear connectors investigated were hooked steel stud,
steel mesh and interlocking dovetail. The timber web was wrapped
with three layers of U-shape FRP fabrics. Discrete and continuous
wrapping were adopted for the timber web for each shear connec-
tor. The details of beams used in this stage are shown in Fig. 3 that
includes a cutout view for the steel mesh shear connector.
Stage III:Post-stressing of the timber web in the longitudinal
direction by means of mechanical torque using 16 mm bars with
nuts was applied on two beams, one with vertical glulam web
and another with massive timber web. Hooked steel stud was
the shear connector used. Additional third beam was prepared to
evaluate the cost-saving aspect of using timber-concrete compos-
ite beams. In this beam, standard steel beam IPE 100 replaced
the timber web and hooked steel dowels of diameter 12 mm was
Table 1
Properties of glass fiber fabric.
Fiber orientation Bi-directional
Weight 600 g/m
2
Effective thickness (mm) 0.12
Fabric width (mm) 1000
Tensile strength of fibers (MPa) 1264
Tensile E-modules of fiber (MPa) 57,630
Elongation at break % 2.14
3420 Yehia A. Zaher Ali / Ain Shams Engineering Journal 9 (2018) 3419–3424

welded to the upper flange of the I-beam. The details of beams
used in this stage are shown in Fig. 4.
3.3. Testing
The beams are tested using the third point load over a simply
supported span with length of 1100 mm. The load was applied
gradually till failure. During the loading, observations and mea-
surements included: crack pattern; strain in the longitudinal direc-
tion on the surface of the concrete flange and the timber web;
central deflection, and the failure modes. Longitudinal deforma-
tions were measured using micro-measurement electrical resis-
tance gages designed for bonding to concrete while dial gages of
sensitivity 0.002 mm were used for recording the central deflec-
tion. Dial gages were also installed on the concrete top surface
directly over the supports to record any settlement of the supports
during testing. Fig. 5 shows the test set up featuring one of the test
specimens ready for attaching the instrumentation and applying
the load.
4. Results and discussion
The results of the experimental program are presented and dis-
cussed in this section. The load-strain curves for test columns are
shown in Figs. 6–8 and the ultimate carrying capacity is shown
in Fig. 9.
Fig. 6 illustrates the load-deflection curves for the three test
specimens investigated in stage I. It is shown from the graph that
the vertical glulam web imparts the test specimen the higher stiff-
ness and load carrying capacity and the larger toughness compared
to both horizontal glulam web and massive block web. The dis-
continuounity of natural defects of wood specially cracks and knots
in vertical glulam web minimized the weakness points in the web
and improved the overall performance of the beam compared to
that made of massive piece of wood. Moreover, the strong adhesive
used to bind vertical laminates of timber through two continuous
planes along the entire span of the beam provided additional rein-
forcing elements to resist tensile stresses and shear transfer. The
recorded failure loads for the test specimens were 7.2, 6.5 and
6.4 tons for vertical glulam, horizontal glulam and massive block
webs, respectively. The holes executed through the horizontal lay-
ers of adhesives in horizontal glulam web to fasten the shear con-
nectors limited the functional efficiency of these reinforcing layers.
Besides, the shear stresses transferred through horizontal planes of
adhesive contributed in limiting the load carrying capacity of hor-
izontal glulam.
Fig. 7 shows the load-deflection curves for the test specimens
for stage II proposed to study the effect of type of shear connector
as well as the role of FRP wrapping. The almost linear performance
of the two beams enhanced with steel mesh shear connector sup-
ported the observation of shear failure at the interface between the
concrete and the shear connector. The slip at the concrete-steel
mesh interface led to partial composite action between concrete
and timber and to increased bending deflections. Neither the FRP
layers wrapping the full section of the timber web nor the concrete
flange in compression zone boosted the load carrying capacity of
the two beams because of this shape of failure. Concisely, beams
manufactured using steel mesh shear connector showed marginal
decrease in load carrying capacity and stiffness compared to other
beams made with dovetail or hooked steel shear connectors.
On the other hand, beams provided with enhanced dovetail
shear connector exhibited superior behavior in flexure in view of
load carrying capacity, stiffness, toughness and ductility. This con-
clusion was verified in an experimental study on notched wood-
concrete composite beams [16]. The attribution of this enhanced
performance is a result of different reasons. First is the transfer
of the shear forces between the timber and concrete over as large
an area as possible. Other reasons include the swelling of timber
when the moist concrete is cast in addition to grasping the timber
with the shrinking concrete. The element responsible for the
enhanced ductility of these beams was the steel dowels provided
at the centers of the dovetail grooves.
The performance of beams made with hooked steel shear con-
nectors was intermediate between those of dovetail and steel mesh
shear connectors. The relative small area of timber acting to
restrain the shear dowel and the concentrated loads that result
limited the capability of hooked steel shear connectors of obtaining
full composite action and developing the shear transfer seen with
dovetail connectors.
Generally, wrapping the timber web with FRP laminates
increases the structural stiffness of the beam and thus results in
a smaller deflection at all load levels (complying with [19].It
was obvious that continuous wrapping along the whole span of
the beam increased the resistance to shear flow and minimized
the tendency to generate stress concentrations at transition zones
as in the case of dis-continuous FRP wraps. The continuous FRP
Fig. 1. Typical cross-section of test beams.
Fig. 2. Details of test beams in stage I.
Yehia A. Zaher Ali / Ain Shams Engineering Journal 9 (2018) 3419–3424 3421

Fig. 3. Details of test beams in stage II.
Fig. 4. Details of test beams in stage III.
3422 Yehia A. Zaher Ali / Ain Shams Engineering Journal 9 (2018) 3419–3424

wraps in beams with steel mesh shear connector limited the slip-
page between the shear connector and the timber web and hence
exhibited a marginal improvement in performance over dis-
continuous ones.
The ductile behavior of test beams is expressed by the ductility
index
l
, defined by Naaman and Jeong [20], as:
l
¼0:5ðE
tol
=E
el
þ1Þ
where E
tol
and E
el
are total energy and elastic energy, respectively,
computed under load-deflection curve. The curves were approxi-
mated to smaller engineering shapes for ease of calculating the
areas under curve. The computed values of
l
are shown in Table 2.
The bearing stresses between concrete and timber developed in
dovetail grooves and the deformability of timber with no apparent
failures in these locations strongly enhanced the ductile behavior
of beams with this shear connector compared with others. The
ductility capability shown by these beams provides very effective
composite action and also prevents total structural collapse.
To evaluate the success of the beam made with enhanced dove-
tail shear connector and wrapped with continuous FRP laminates,
Fig. 8 illustrates a the behavior of this beam along with three stiff-
ened web beams. The webs of these three beams were post-
stressed vertical glulam, post-stressed massive timber and stan-
dard steel beam IPE 100. It was clear that the behavior of dovetail
beam is comparable to the concrete-steel composite beam except
that the later one experienced large deformations near failure
due to the out of plane buckling of the steel web (Fig. 9). By consid-
ering the advantageous use of timber compared with steel in view
of the total weight and environmental impacts, this proposed
enhanced dovetail shear connector is highly recommended for
use in flexural composite elements.
Fig. 5. Test set up for beams.
0
10
20
30
40
50
60
70
80
0 5 10 15 20 25 3
0
Deflec tion mm
Load kN
Vl Glulam Hl Glulam Block
Fig. 6. Load-deflection curves for test beams in stage I.
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 3
0
Deflection mm
Load kN
Dovetail-continuous Dove tail-str ips Hook-continuous
Hook-strips Mesh-continuous Mesh-strips
Fig. 7. Load-deflection curves for test beams in stage II.
Table 2
Ductility index for test beams of stage II.
Dovetail-continuous 3.110 Dovetail-strips 2.830
Hook-continuous 1.715 Hook-strips 1.572
Mesh-continuous 1.384 Mesh-strips 1.275
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 3
5
Deflec tion mm
Load kN
Dovetail-continuous Ps Vl Glulam
Ps Block Steel web
Fig. 8. Load-deflection curves for test beams in stage III.
Fig. 9. Failure mode of concrete-steel composite beam.
Yehia A. Zaher Ali / Ain Shams Engineering Journal 9 (2018) 3419–3424 3423

Contrarily, the behavior of the two externally post-stressed
beams was inferior to other beams. However, using the external
longitudinal stressing bolts did increase the load carrying capacity
of the beams with only 15% compared with the corresponding
beams tested in stage I. The surplus strength of these beams may
be attributed to the increased resistance to shear flow through tim-
ber layers.
5. Conclusions
The following summarizes the findings of this investigation:
1. The present study shows that vertical glulam timber web can be
successfully used for the manufacture of concrete-timber com-
posite beams since it contributes in providing higher stiffness
and load carrying capacity and larger toughness compared to
both horizontal glulam web and massive block timber web.
2. Dovetail shear connector enhanced with insertion of a steel
dowel at the middle of the groove is the superior shear connec-
tor, among different types investigated in this study, to impart a
full composite action between concrete and timber web in
beams. It strongly enhances the load carrying capacity, stiffness,
toughness and ductility.
3. Use of GFRP laminates wrapping along the full span of timber
web in composite beams is a highly effective method to take
advantage of the use of enhanced dovetail shear connector.
4. The success of FRP laminates wrapping for timer webs con-
nected to concrete flange by enhanced dovetail shear connector
was emphasized by obtaining a comparable behavior of the
manufactured beam with that of concrete-steel web beams.
References
[1] Goho Alexandra. Concrete nation, bright future for ancient material. Sci News
2005;167(1):7.
[2] Baghi Hadi, Barros Joaquim AQ, Gouveia Antonio Ventura. Shear strengthening
of reinforced concrete beams with hybrid composite plates. Adv Struct Eng
2016;19(1):132.
[3] Wahby Wafeek S. Composite materials: yesterday, today and tomorrow. In:
The Second Middle East Symposium on Structural Composites for
Infrastructure Applications, Hurghada, Egypt; 1999.
[4] Mallick PK. Composite engineering handbook. New York: Marcel Dekker Inc;
1997.
[5] ASSE. ‘‘Classic Wood Structures”, Task committee on classic wood structures,
Committee of Wood Structural Division. New York: American Society of
Structural Engineers; 1989.
[6] Yttrup PJ, Nolan G. Performance of timber beam bridges in Tasmania, Australia.
In: International wood engineering conference, New Orleans, Louisiana, USA;
1996.
[7] Baldcock RH, McCullough CB. Loading tests on a new composite type short
span highway bridge combining concrete and timber in flexure. Oregon State
Highway Department, Technical Bulletin No. 1, Salem; 1941. p. 160.
[8] Al Ali M, Bajzecerova V, Kvocak V. Design methods of limber-concrete
composite ceiling structure. Mag Civil Eng 2017;5:88.
[9] Murthy CK. Timber concrete composites for low cost housing. Hous Sci 1984;8
(2):209–15.
[10] Ahmadi BH, Saka MP. Behaviour of composite concrete timber floors. J Struct
Eng 1993;119(10):3111–30.
[11] Meierhofer U. A timber/concrete composite system. Struct Eng Int 1993;2:104.
[12] Mantilla Carrasco EV, Oliveira SV. Behavior of composite timber-concrete
beams. In: Proceedings of the 1st RILEM symposium on timber engineering,
Stockholm, Sweden; 1999. p. 571–80.
[13] Rajèi
æ
M Rak. Continuous shear connecting-the best way to compose timber
and lightweight (EPS) concrete. In: The 7th world conference on timber
engineering, WCTE 2002, August 12–15, Shah Alam, Malaysia; 2002.
[14] Dagher HJ, Lindyberg RF, Abdel-Magid B, Shaler S. ‘‘Static Bending Test Results
Of Douglas-Fir And Western Hemlock FRP-Reinforced Glulam Beams”, AEWC
Report No. 98-4, University of Maine, Orono, ME; 1998.
[15] Kimball T. Feasibility of glulam beams reinforced with FRP sheets M.S.
Thesis. Orono, ME: Dept. of Civil and Environmental Engineering, University of
Maine; 1995.
[16] Yeoh D, Fragiacomo M, Abd Ghafar H, Buchanan A, Deam B, Crews K. LVL-
concrete composite floor systems: an effective solution for multi-storey
timber buildings. In: Australasian Structural Engineering Conference (2008),
Melbourne, Australia, Paper 236; 2008.
[17] Egyptian Standards 4756-1. Cement - Part 1: Composition, Specifications and
Conformity Criteria for Common Cements. Egyptian Organization for
Standardization and Quality, Cairo; 2007.
[18] Egyptian Standards 1109-2002. Concrete Aggregate from Natural Source.
Egyptian Organization for Standardization and Quality, Cairo; 2002.
[19] Mio Ho Jose Luiz, Dias Antonio Alves. Concrete composites: experimental
investigation into the connection system. Mater Res 2011;14(1):53.
[20] Naaman AE, Jeong SM. Structural ductility of concrete beams prestressed with
FRP tendons. In: Taerwe L, editor. Proc., 2nd Int. RILEM Symp. (FRPRCS-
2). Cachan, France: RILEM; 1995. p. 379–84.
Ali, Yehia Abdel Zaher is an Associate Professor at Ain
Shams University, Faculty of Engineering, Civil Engi-
neering Department. He was graduated and did his
M.Sc. from the same department. He did his Ph.D. at
IIT-Madras on 1998 in the field of building materials
and construction technology. He lives in Cairo, Egypt. He
loves general reading and football competitions.
3424 Yehia A. Zaher Ali / Ain Shams Engineering Journal 9 (2018) 3419–3424