CC BY-NC-ND
0HBRC Journal (2012) 8, 26-35
Housing and Building National Research Center HBRC Journal
http://ees.elsevier.com/hbrcj
Behavior of thermally protected RC beams strengthened with CFRP under dual effect of elevated temperature and loading
A.E. Salama a, G.M. Ghanem a, S.F. Abd-Elnaby a'*, A.A. El-Hefnawy b, M. Abd-Elghaffar b
a Faculty of Engineering, Helwan University, Cairo, Egypt b Housing and Building National Research Center, Giza, Egypt
Received 5 July 2011; accepted 30 October 2011
KEYWORDS
Concrete;
Reinforcement;
Protection;
Strengthening;
Abstract Strengthening of reinforced concrete elements by externally bonded FRP is becoming increasingly popular in construction industry. Utilizing FRP offers several desirable attributes, such as resistance to corrosion, high strength-to-weight ratio, and electromagnetic neutrality. However, FRP materials have some disadvantages. In case of external strengthening with FRP which is directly exposed to the environment, efficiency of bond between FRP and concrete surface is affected by temperature. Current research work was carried out to investigate behavior of strengthened RC beams subjected to dual effect of elevated temperature and loading. The experimental program consists of two phases. First one comprises investigating thermal properties of six different cement-based mixes incorporating Perlite and Vermiculite in order to find out the most appropriate mix that possesses both low thermal conductivity and adequate strength. In the second phase, the obtained mix was applied as cement rendering to protect different reinforced concrete beams against elevated temperature. The beams were divided into four main groups in order to explore the flexural behavior of both unstrengthened and strengthened beams with CFRP subjected to dual effect of heating and loading. Results showed that a layer of 50 mm thickness of Perlite mortar can be used to protect CFRP strengthening system against 500 °C for three hours. Strengthened protected beams exhibited insignificant capacity loss when loaded under 500 °C for 3 h and cooled in ambient air, then loaded up to failure. When similar beam was subjected to the same loading
* Corresponding author.
E-mail address: hefnawyhbrc@hotmail.com (S.F. Abd-Elnaby). Peer review under responsibility of Housing and Building National Research Center
1687-4048 © 2012 Housing and Building National Research Center. Production and hosting by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.hbrcj.2012.08.005
and heating condition except that cooling was not allowed, a reduction in capacity by 22% was observed.
© 2012 Housing and Building National Research Center. Production and hosting by Elsevier B.V.
All rights reserved.
Introduction
Many structures that were built in the past decades suffered from significant deterioration. Many other structures were built without sufficient quality control on either design or construction. The resulting problems may take the form of cracks and excessive deflections. These cases need urgent and effective repair. There are many techniques which have been utilized to repair or strengthen damaged beams. The conventional methods cover imbedding steel reinforcement in the tension zone or using externally bonded steel plates or using steel sections such as angles. A new technique for external strengthening of reinforced concrete beams has been developed, using FRP (Fiber Reinforced Polymers). This technique is relatively easy and fast to apply. The most interesting advantages of these composite materials are their flexibility, lightweight, small thickness, non-corrosive nature, and ability to be applied to beams in any shape according to the strengthening requirement (flexure, shear...). One of the main problems that is usually associated with the use of FRP systems as well as some other.
Some traditional repair methods that use epoxy resins, is the possibility of losing structural integrity at temperatures exceeding the glass transition temperature (Tg) of the polymer (60-82 0C). As a result, in the case of fire, the strength of externally bonded FRP systems is assumed to be lost completely. For this reason the structural members without the FRP system should possess sufficient strength to resist all applicable loads during a fire without collapse. In order to guarantee that a structure can maintain its safety and integrity in case of fire, Egyptian FRP Code [1] imposes strengthening limits on design of any strengthening works. The code considers that an increase in ultimate strength that does not exceed 40% can be adequate enough to guarantee safety in case of fire, because combined ultimate load factor for a reinforced concrete structure is equal to 1.5 and strengths of both steel and concrete are reduced as a result of fire. It is mentioned that in case of using special class of polymers possessing better fire resistance and/ or using effective fire protection coating, only partial loss of FRP strengthening works will be expected to occur. As a result, Egyptian code allows strengthening limits to exceed 40% and reach higher values, the magnitudes of which should reflect the level of protection provided. However, it has been cited that higher values of strengthening limits are allowed only if it can be proved through testing and technical assessments that such fire protection systems can increase fire endurance of FRP systems to exceed the fire resistance rating of building [2].
The present research investigates the effect of properly designed protective cement based mortar on the performance of strengthened RC beams with externally bonded CFRP when exposed to dual effect of elevated temperature and loading.
Experimental program
The experimental program of current research consists of two phases [3]. The first one comprises investigating thermal properties of six different cement-based mixes incorporating Perlite and Vermiculite to find out the most appropriate mix that possesses both low thermal conductivity and adequate compres-sive strength. In the second phase, the most effective insulating mortar was used to protect different reinforced concrete beams against elevated temperature. Those beams were divided into four main categories in order to explore the effect of protective layer on flexural behavior of either unstrengthened or strengthened beams with CFRP subjected to dual effect of thermal and loading conditions.
Experimental program, phase 1: protective lightweight mortar
Lightweight aggregates like Perlite and Vermiculite were utilized throughout current research to produce an appropriate thermal protective cement based mortar layer that can be used to protect concrete elements against elevated temperature, as reported by other researchers [4].
Classification of lightweight aggregate
ASTM C332-07 [5] gives two general types of lightweight aggregate as follows:
Group (I): Aggregates prepared by expanding products such as Perlite or Vermiculite. These aggregates generally produce concrete with bulk density ranging from 240 to 800 kg/m3with thermal conductivity ranging from 0.065 to 0.22 W/m.k.
Group (II): Aggregates prepared by expanding, calcining, or sintering products such as basalt-furnace slag, clay, diato-mite, fly ash, shale, or slate; and aggregates prepared by processing natural materials , such as Pumice , Scoria, or Tuff. These aggregates generally produce concrete with bulk density ranging from 720 to 1440 kg/m3with thermal conductivity ranging from 0.15 to 0.43 W/m.k.
This means that group (I) is more efficient than group (II) in thermal insulation. Therefore, group (I) was considered through the current research.
Sampling and testing of the used materials were undertaken in accordance to ASTM C330-05 [6].
X-ray florescent test
X-ray test has been performed on Vermiculite and Perlite in order to determine its composition. Results are displayed in Table 1.
Mortars mix ingredients
ASTM C332-07[5] recommended that the density of mortar that used as a protective layer against temperature is about 240-800 kg/m3and using water cement ratio (w/c) just enough
Table 1 Composition of Vermiculite and Perlite by X-ray
analysis.
Substance Vermiculite composites (%) Perlite composites (%)
SiO2 43.38 76.9
AI2O3 7.77 9.8
Fe2O3 8.38 0.54
CaO 3.17 1.79
MgO 30.15 1.07
Na2O 0.98 2.55
K2O 2.44 2.41
SO3 0.57 0.19
TiO2 0.76 0.1
P2O5 0.09 0.03
L.O.I. 2.05 4.37
TOTAL 99.74 99.75
Table 2 Mix proportion of the protective materials.
Mix No. Cement content Perlite(kg) Vermiculite (kg) Water (kg)
P1 400 110 - 380
P2 450 427.5
P3 500 475
V1 400 - 412 480
V2 450 540
V3 500 600
to form the paste. However, to get an efficient protection, the protective layer must attain reasonable compressive strength. Therefore, several mixes with Perlite and Vermiculite aggregates and different cement contents were investigated. The studied mixes ingredients are given in Table 2.
Fig. 2 A mortar sample during conductivity test according to ASTM C177-07.
P/2 P/2
300 mm 250 mm 300 mm
8$ 8/m
2$ 8 2$ 8
750 mm
Fig. 3 Top head schematic.
170 mm
Table 3 Mechanical and thermal test results for Perlite and
Vermiculite mortar.
Mix No. Thermal conductivity Average compressive
(Watt/m °C) strength (kg/cm2)
P1 0.158 20.7
P2 0.26 22.67
P3 0.277 23.1
V1 0.396 19.05
V2 0.35 23.73
V3 0.329 30.1
Mortar specimens
As shown in Table 2 three mixes of Perlite and another three of Vermiculite were investigated. Three cubes 100 x 100 x 100 mm were prepared from each mix to experimentally determine its compressive strength. Additional plate 200 x 200 x 50 mm was prepared from each mix to experimentally determine its thermal properties.
Mixing and curing
For proper mixing, ASTM C332-07 [5] recommended to add the whole quantity of cement with only two-third of water and mixing them up to obtain a homogenous paste. Then, aggregate was added with the rest of water to the paste. The mixture was mixed until homogenous color was reached. It should be noticed that increasing mixing time causes crushing
of lightweight aggregate into smaller particles leading to density increase and thermal insulation decrease. Specimens were moist-cured for 7 days and then removed from the moist cabinet and stored at 22 0C and 55% relative humidity. At 28-day age, specimens were dried by a drying oven at 110 0C until the loss in weight was less than 1% during 24 h interval.
Compressive strength and thermal conductivity tests
The loose unit weight, q (bulk density) of Perlite is 120 kg/m3
while the loose density, q, of Vermiculite is 282.8 kg/m3.
Schematic of the device used to measure the thermal conductivity of different mixes is shown in Fig. 1. This device is complying with ASTM C177-07 [7]. Fig. 2 also shows the test device and equipments during testing of a mortar sample.
Top head
Water inlet
T3 Top-center thermo couple
T2 bottom-corner thermo couple
rner I
Test specimen
Water outlet
t Bottom head T bottom-center thermo couple
Fig. 1 Schematic of thermal conductivity test setup.
Table 4 Summary of tested specimens.
Category No. Category 1 Category 2 Category 3 Category 4
Strengthening No Yes No Yes
Thermal Protection No No Yes Yes
Loading condition A CBU1 CBS1 CBU2 CBS2
B LAU1 - LAU2 LAS
C LWU1 - LWU2 LWS
D HAU1 - HAU2 HAS
E HWU1 - - -
The device is basically composed of two heads. Top one contains path for cooling water as shown in Fig. 3. A thermo couple was attached to the top face of the sample to measure temperature (T3). The bottom head contains paths for heating coils. Two thermo couples were attached to the bottom face of the sample. One at the center to measure (T1) and the second at corner to measure (T2). The average of T1 and T2 were taken as the heating temperature. Special sensor was also used to determine heat flow (Q). All these thermo-couples and sensor were connected to a data acquisition device to collect different measurements.
Values of T1, T2, T3, heat flow (Q) and time were recorded until the difference between T3 and average of T1 & T2 becomes constant (thermal equilibrium state). Then, the thermal conductivity was calculated (K) in terms of W/m 0C as follows, [7]:K = (Q x t x 10)/(100 x (DT))Where:Q: heat flow in terms of W/m2t: sample thickness in cmDT: difference between T3 and average of T1 & T2
Results of compressive strength and thermal conductivity tests of Perlite & Vermiculite mortar samples are represented in Table 3.
From the previous results, it can be shown that Perlite mortar (P1) with cement content of 400 kg/m3 is more efficient than other mixes. It has the minimum thermal conductivity and accordingly maximum thermal protection. It also gained appropriate compressive strength.
Experimental program, phase 2: behavior of RC beams
This test phase included testing of (14) beams. All beams were tested under flexure test to investigate the behavior and load capacity of different beams category and to explore the effect of thermal protection on the first crack loads, failure loads, crack patterns, deflections and steel strains for the specimens. The beams were divided into four categories:
Category (1): Five unstrengthened and unprotected beams, control specimens.
Category (2): One strengthened and unprotected beam.
Table 5 Mechanical properties of CFRP wraps [8].
Property Test results
Fiber type High strength carbon fiber
Fiber weight (gm/m2) 225
Fiber orientation Unidirectional
Fabric design thickness (mm) 0.13
Fabric Width, mm 305
Tensile Strength of Fibers, kg/cm2 35700
Tensile E-Modulus of Fibers, kg/cm2 2346000
Elongation at Break, % 1.5
Category (3): Four unstrengthened and protected beams.
Category (4): Four strengthened and protected beams.
These categories are shown in Table 4.
Beams were exposed to loads and temperature according to following conditions:
Condition (A): Appling load on beams at normal room temperature up to failure. One beam of each category was tested under this condition as a control specimen.
Condition (B): Loading beam up to the working load that was obtained for all categories under condition (A), then the load was kept constant. While the working load was applied, temperature was increased up to 500 0C and remained constant for three hours. After that, the load and temperature were removed and beam was left to cool in air for one day. Then it was loaded up to failure. This condition was applied for all beams except category (2) because beams under this category were strengthened with CFRP and not protected.
Condition (C): Loading beam up to the working load and kept constant. Then temperature was increased up to 500 0C and remained constant for three hours. After that the load was increased up to failure. Again this condition was applied for all beams except category (2) as mentioned before.
Condition (D): Heating (unloaded) beam up to 500 0C and temperature was kept constant for three hours, then it was left to cool in air and loaded up to failure next day. This condition was applied for all categories except (2) as mentioned before.
Fig. 4 Concrete dimension and reinforcement details of test beams.
Fig. 5 Details of a strengthening beam.
Fig. 6 Bottom view of a strengthened beam.
Condition (E): Same as condition (D) except cooling method where beam was rapidly cooled by water. This condition was only applied on category (1).
Concrete dimensions and reinforcement of beams All beams had a rectangular section of 150 mm width, 170 mm depth and 850 mm length. The actual concrete strength based on150mm cube was 356 kg/cm2. Concrete clear cover was 20 mm. Each beam was reinforced in tension side with two 8 mm plain rounded bars of 2.91 t/cm2 yield strength. Two 8 mm plain rounded bars were used in compression side. Seven stirrups 8 mm plain bars were used to ensure flexure failure and prevent shear failure. Fig. 4 shows dimensions and reinforcement of a typical beam.
Strengthening by externally bonded CFRP
Carbon fiber reinforced polymer (CFRP) wraps were used to strengthen the tested beams. The CFRP wraps were Sika Wrap Hex-230C. The relevant mechanical properties, as stated in product data sheet are summarized in Table 5, [8].
Strengthening procedure of beams
The following are steps of beam strengthening, [9]:
1. Beam surface was prepared by grinding to get a smooth surface, followed by removal of loose particles and dust by compressed air jet.
2. Resin matrix was prepared by mixing the resin and hardener (Sika-dur 330).
3. An under-coat of resin was applied first to the beam surface using a paint brush to prepare the concrete surface and to fill voids.
,P/2 | P/2
300 mm 250 mm 300 mm
^ 250 mm ^
750 mm
170 mm
Fig. 8 Concrete beam after coating.
4. CFRP wrap was applied as one layer in the tension zone while pressing firmly down with a rag until resin was applied.
The strengthening geometry is shown in Fig. 5. To avoid debonding of CFRP and to ensure full utilization of CFRP, anchorage was secured in the manner shown in Fig. 6. It shows the bottom side of the strengthened beams.
Maximum capacity of strengthened section
capacity
: b x d2 = 0.214 x 356 x 15 x 16.5A2 =
31116.41 kg cm = 3.11 t.m.M = P/2 x L/3 ) P = 24.88 ton
If maximum reinforcement is used, the load which causes compression failure will be 24.88 tonTrying strip of CFRP with 10 cm breadthThickness = 0.013 cm, tensile strength = 35,000 kg/cm2, maximum tensile force in fibers Tf = 35000 x 0.013 x 10 = 4550 kg )Tf = 4.55 tonFig. 7 shows stress and strain distributions As = 1 cm2As' = 1 cm2, Fy = 2.91 t/ cm2(mild steel bars) ) Ts = 1 x 2.91 = 2.91 ton )T = Ts + Tf = 2.91 + 4.55 = 7.46 ton
C = Cc + Cs = 0.67 x Fcu x (0.8 x c): C = 0.67 x 356 x (0.8 x c) x 15 + Fs' = + Fs' = 7460 (1)
b + As': = 7460C =
Fs' = 7460, 2862.76 x c
ecc = es'/(c — d)
s' = Fs/Es, c/ s' = c/(c — d)c = 0.003, Es = 2100 t/cm2, 0.003/ (Fs/2100000) = c/(c — d) & d = 2 cm6300/ Fs = c/(c — 2), 6300 x c — 12600 = c x Fs Fs' = (6300 x c — 12600)/c (2) From equations 1 & 2 we get2862.24 x c + (6300 x c —
12600)/c = 7460 )2862.24 x c2 + 6300 x c — 12600 = 7460 x c2862.24 x c2 — 1160 x c — 12600 = 0 )c = 2.311 cmc/d = 2.311/16.5 = 0.14 > 0.1 ok )Fs' = (6300 x 2.311 — 12600) /2.311 = 847.8 kg/cm2Mcapacity = 0.67 x Fcu • 0.8 x c x b x (t — (0.8 x c)/2) + As' x Fs' (t — d) — As x Fy x d' = 0.67 x 356 x 0.8 x 2.311 x 15 x (17 — (0.8 x 2.311)/2) + 1 x 847.8 x (17 — 2) - 1 x 2910 x 0.5 = 117596 kg cmMcapacity = 1.18 t.m )M
P/2 x 0.25 )Pm
1.18 x 2/.25 = 9.44 ton
300 mm
Open oven with digital temperature control
250 mm
300 mm
550 mm
170 mm
750 mm
Fig. 9 Beam loading setup.
Fig. 10 Test setup for a beam exposed to elevated temperature during loading (elevation view).
Application of mortar layer
Beams of category three and four were thermally protected by 5 cm Perlite mortar overlay. Weak steel wire mesh was used to increase the ductility of the coating layer and to improve the bond strength between the coating and the concrete surface by bridging the areas covered by CFRP. Fig. 8 shows a beam mortar coating.
Beams test setup
In order to focus on the behavior of reinforced concrete beams under flexure loading, four points loading was used to avoid shear stresses in the studied zone as shown in Fig. 9. Figs. 10 and 11 illustrate test setup of beams that were exposed to dual effect of elevated temperature and loading.
Test results and discussion
Failure loads
The failure loads and the corresponding deflections for each beams category are shown in Table 6. Fig. 12 displays a bar chart for the failure loads of the beams arranged according to the loading conditions.
Strengthening by CFRP resulted in capacity increase by 78% for beam CBS1 over the capacity of beam CBU1. Results of CBU2 and CBS2 were very similar to those of CBU1 and CBS1, respectively. It indicated that no significant contribution of the protective layer to beam capacity under normal temperature condition.
Generally, all unstrengthened unprotected beams had almost same capacity under different conditions except beam LWU1 that was tested under condition C. Under this condition the beam was loaded up to failure while it was subjected to elevated temperature. Its capacity decreased by 23.8% compared with the control specimen under condition A. Loading condition C also resulted in the highest reduction in capacity of protected strengthened beam LWS where reduction of 22%, 19%, and 15% less than the CBS2, LAS, and HAS, respectively, was observed. Comparing the capacity of protected strengthened beams (LAS, LWS, and HAS) with the
Fig. 11 Test setup for a beam exposed to elevated temperature during loading (cross section).
capacity of protected unstrengthened beams (LAU2, LWU2, and HAU2), It can be noticed that minimum residual load in beam capacity due to strengthening happened under loading and temperature condition C. It means that this condition had the most adverse effect on CFRP strengthening system. Similar result was noted elsewhere [10].
Behavior of beams under different loading and heating conditions
The following sections illustrate differences in behavior between all beams categories subjected to the designated loading and heating conditions.
Loading of beams under normal temperature, condition A Effect of strengthening. Strengthened by CFRP resulted in beam capacity increase by 78% in case of normal temperature condition. Rupture of CFRF was noticed at beam failure. It indicated the adequacy of anchorage system to utilize full capacity of CFRP. Due to that rupture, sudden failure occurred and deflection corresponding to the failure load was lower than the unstrengthened beam. Fig. 13 shows the load-deflection curves for unstrengthened beam CBU1 and strengthened one CBS1. The failure loads for beams CBU1 and CBS1 were 5.6 and 9.97ton and the corresponding deflections 10.43 and 6.76 mm, respectively. It can also be noticed that CFRP resulted in beam stiffness increase. Fig. 14 shows insignificant contribution of protective layer on the flexural behavior of the protected beam.
Behavior of beams under dual effect of loading and heating, condition B
Protected and unstrengthened beams. Fig. 15 shows load-deflection curves for unprotected beam LAU1 and protected one LAU2. These two reinforced concrete beams were utilized to explore effect of the protective layer on behavior of reinforced concrete beams subjected to elevated temperature. Failure loads for these two beams were 5.38 and 5.61 ton and corresponding deflections 6.18 and 13.42 mm for LAU1 and LAU2, respectively. It can be observed that the failure load of the unprotected beam slightly decreased by less than 4%. However, deflection corresponding to failure load of the protected beam LAU2 was more than twice the deflection of
the uncovered beam LAU1. This may be attributed to adverse effect of elevated temperature on steel elongation in case of unprotected beam since deflection of beam LAU2 was near to the unprotected beam CBU1 tested under condition A. From previous results, it can be concluded that the use of thermal protective layer maintained beam ductility when subjected to dual effect of loading and elevated temperature provided that beam cooling is permitted.
Protected and strengthened beams. Fig. 16 demonstrates load-deflection curves for beams LAS and LAU2. It shows that failure loads were 5.61 and 9.4 ton and corresponding deflection 13.42 and 12.45 mm for LAU2 and LAS, respectively. The difference in failure load was due to CFRP used for beam LAS. CFRP increased the flexural capacity by 67.5% more than unstrengthened beam exposed to dual effect of loading and heating condition provided that beam cooling was permitted. The capacity of beam LAS decreased by only 3.4% less than the control beam CBS2. Similar result has been found by other researchers [11,12]. It should be noticed that the first steep red line represents load-deflection relationship of the beam under loading up to working load (without heating). The subsequent horizontal line represents load-deflection curve of the beam under constant working load simultaneous with elevated temperature. The dotted line represents the unloading phase. The second red line represents load-deflection curve of the beam after being cooled and reloaded up to failure. Comparing these two load-deflection curves it can be observed that beam ductility decreased as a result of relaxation that might have occurred to the CFRB matrix during beam heating.
Behavior of beams under dual effect of loading and heating, condition C
Protected and unstrengthened beams. Load-deflection curves for unprotected beam LWU1 and protected one LWU2 under loading condition C were plotted in Fig. 17. This loading condition was employed to explore the effect of continuous loading and heating up to failure without beam cooling. Failure loads of beam LWU1 and LWU2 were 4.27 and 5.75 ton and the corresponding deflection 17.6 and 39.4 mm, respectively. The effect of protective layer on beam capacity was remarkable since failure load of beam LWU2 increased by
Table 6 Failure loads and corresponding Deflection of different beams.
Category Beam Protective Strengthening Loading Failure load Deflection at failure Failure
No. designation layer with CFRP condition (ton) load (mm) mode
Category CBU1 - - Control (A) 5.6 10.43 Flexure
1 LAU1 - - B 5.38 6.18 Flexure
LWU1 - - C 4.27 17.6 Flexure
HAU1 - - D 6.08 13.88 Flexure
HWU1 - - E 5.72 14.27 Flexure
Category 2 CBS1 - Yes Control (A) 9.97 6.76 Flexure
Category CBU2 Yes - Control (A) 5.62 10.99 Flexure
3 LAU2 Yes - B 5.61 13.42 Flexure
LWU2 Yes - C 5.75 39.4 Flexure
HAU2 Yes - D 6.21 14.27 Flexure
Category CBS2 Yes Yes Control (A) 9.77 7.83 Flexure
4 LAS Yes Yes B 9.4 12.45 Flexure
LWS Yes Yes C 7.61 10.55 Flexure
HAS Yes Yes D 8.95 6.64 Shear
Deflection, Fig. 12 Failure loads for different beams.
Fig. 13 Load-deflection relationship of CBU1 and CBS1.
Fig. 14 Load-deflection relationship of CBU1 and CBU2.
/ LAS ^ LAU2
10 20 30 40 Deflection, mm
Fig.15 Load-deflection relationship of LAU1 and LAU2.
34.5% over the unprotected one LWU1 under loading condition C. The protected layer reduced the adverse effect of elevated temperature on properties of steel reinforcement. In addition, it showed the severe effect of elevated temperature
Def lectio
Fig. 16 Load-deflection relationship of LAS and LAU2.
on concrete compressive strength. Accordingly it delayed compression zone failure and prevented compression failure.
Protected and strengthened beams. Fig. 18 illustrates load-deflection curves of strengthened beam LWS and unstrength-ened one LWU2 subjected to lading condition C.
The failure loads of beams LWS and LWU2 were 7.61 and 5.75 ton and the corresponding deflections were 10.55 and 39.4 mm, respectively. It can be shown that capacity of protected strengthened beam LWS is higher than that of protected unstrengthened beam LWU2 by 32%. Rupture of CFRP was occurred during the test which means that strengthening by CFRP was fully utilized.
Behavior of beams, conditions D&E
Under condition D, beams HAU1, HAU2, and HAS were exposed to 500 0C for three hours then left to cool in ambient temperature and tested next day. Under condition E, beam HWU1was exposed to same temperature for the same duration but cooling was done rapidly by water then loaded up to failure.
Effect of cooling method on behavior of beams, condition D&E. The behavior of two unstrengthened and unprotected beams was compared. The difference was only the method of beam cooling. Beam HAU1 was cooled by ambient air while HWU1 cooled by water. Fig. 19 shows the load-deflection curve for these two beams.
It can be notice that beam HAU1 was slightly stiffer than beam HWU1. It was also noticed that crack width of beam HWU1 resulting from water-cooling was wider than beam HAU1. Failure loads for beams HAU1 and HWU1 were 6.08 and 5.72 ton, respectively, and the corresponding deflections were 13.88 and 14.27 mm, respectively. This indicates that beam cooling either by air or water and then loading up to failure gave insignificant difference in both capacity and
Fig. 17 Load-deflection relationship of LWU1 and LWU2.
Fig. 18 Load-deflection relationship of LWS and LWU2.
Deflection,
Fig. 21 Load-deflection relationship of HAS and HAU2.
HAU2. It means that HAU2 exhibited more ductile behavior due to thermal protection than HAU1. The failure loads of these two beams were 6.08 and 6.21 ton and the corresponding deflections were 13.88 and 14.27 mm, respectively.
Effect of strengthening and protection on behavior of beams, condition D. Fig. 21 shows load-deflection curve for protected strengthened beam HAS and protected unstrengthened beam HAU2. Failure load of HAS and HAU2 were 8.95 and 6.21 ton and the corresponding deflections were 6.64 and 14.27 mm, respectively. It was noticed for beam HAS, that shear failure was occurred followed by peeling of concrete resulted in anchorage rupture. Accordingly CFRP strengthening system was not able to sustain more loads.
Conclusions
Deflection,
Fig. 20 Load-deflection relationship of HAU1 and HAU2.
deflection. According to this observation, the effect of thermal protection and strengthening was studied in case of air-cooling only.
Effect of protective layer on behavior of unstrengthened beams, condition D. Load-deflection relationship for unprotected beam HAU1 and protected one HAU2 are displayed in Fig. 20. These two beams were utilized to investigate the effect of using protection on behavior of beams under this loading and heating condition. It can be shown that behavior of HAU1 and HAU2 were almost similar up to maximum load. Then resistance of HAU1 decreased more rapidly than
Based on results of current research work, the following conclusions can be drawn:
1. Perlite mortar exhibited 2.5 times less thermal conductivity than Vermiculite mortar and revealed adequate compres-sive strength.
2. Flexural capacity of strengthened beam with 100 mm width of CFRP bonded anchored sheet increased by 78% when tested under normal ambient temperature. Also, limited cracks propagation was observed.
3. Beam covered with thermal protective layer showed no capacity loss when it was loaded under 500 0C for three hours and cooled by air, then loaded up to failure.
4. Strengthened protected beam exhibited insignificant loss in its capacity, only 3.8%, when loaded under 500 0C for three hours and cooled by air, then it was loaded up to failure. It indicates that the protection layer provided successful protection against this heating condition.
5. Maximum reduction in capacity of strengthened protected beams was observed when the beam was loaded up to working load and heated up to 500 0C for three hours then load was increased up to failure. Under this condition, the beam suffered a loss of its capacity by 22%. The unstrengthened uncovered beam suffered a reduction of 23.75% under the same condition.
6. Strengthened and protected beam subjected to 500 0C for three hours (without application of loading at the same time) and kept to cool in air, then loaded up to failure showed capacity decrease by 8.3%.
References
[1] Egyptian Code of Practice for the Use of Fiber Reinforced Polymer in the Construction Field, ECP No.208, Egypt, 2005.
[2] H. Hosny, I. Mahfoz, Introduction to the Egyptian Code for the use of fiber reinforced polymers (FRP) in the construction fields, Proceeding of the Forth Middle East Symposium on Structural Composites for Infrastructure Applications, Alexandria, Egypt, 2005.
[3] M. Abd-Elghaffar, Effect of Fire on Reinforced Concrete Beams Strengthened Using Modern Fibrous Composite Materials, M.Sc., thesis, Faculty of Engineering, Cairo University, Egypt, 2007.
[4] M. Abdel-Razik, Coating protection of loaded R.C. columns against elevated temperature, International Conference on Future Vision and Challenges for Urban Development, Cairo, Egypt, 2004.
[5] ASTM C332-07, Standard Specification for Lightweight Aggregate for Insulating Concrete, American Society for Testing and Materials, Philadelphia, 2007.
[6] ASTM C330-05, Standard Specification for Lightweight Aggregate for Structural Concrete, American Society for Testing and Materials, Philadelphia, 2005.
[7] ASTM C177-7, Standard Test Method for Steady-State Heat Flux Measurements and Thermal Transmission Properties by Means of the Guarded-Hot-Plate Apparatus, American Society for Testing and Materials, Philadelphia, 2010.
[8] Sika Construction Chemicals, Product Data Sheet, Carbon Fiber fabric for Structural strengthening - Sikawrap Hex-230C, SikaSchweiz AG, Switzerland, 2005.
[9] A. Nanni, Fiber-Reinforced-Plastic (FRP) Reinforcement for Concrete Structures: Properties and Applications, Elsevier Science Publishers B.V., London, 1993.
[10] M. Adelzadeh, R. Eedson, L.A. Bisby, N. Benichou, X. Bao, W. Li, Performance in fire of fiber reinforced polymer strengthened concrete beams including embedded fiber optic sensors, Proceedings of the 6th International Conference, East Lansing, MI, USA, 2010, pp. 149-156.
[11] M.F. Green, N. Benichou, V.K.R. Kodur, L.A. Bisby, Design guidelines for fire resistance of FRP strengthened concrete structures, 8th International Conference on FRP in Reinforced Concrete Structures, Patras, Greece, 2007, pp. 1-10.
[12] B. Williams, L. Bisby, V. Kodur, J. Su, M. Green, An investigation on fire performance of FRP strengthened R/C Beams, 5th International Symposium on Fire Safety Science, Beijing, 2005, pp. 247-258.
