Effect of elevated temperature environments on the

0 downloads 38 Views 6MB Size Report
Apr 22, 2016 - reinforced concrete (RC) circular columns strengthened with different ... strengthened with one layer of continuous glass fiber reinforced ...... concentrically and uniform loading is achieved, by providing gypsum capping at.

Construction and Building Materials 115 (2016) 345–361

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Effect of elevated temperature environments on the residual axial capacity of RC columns strengthened with different techniques Yousef A. Al-Salloum, Tarek H. Almusallam, Hussein M. Elsanadedy ⇑, Rizwan A. Iqbal MMB Chair for Research and Studies in Strengthening and Rehabilitation of Structures, Dept. of Civil Engineering, College of Engineering, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia

h i g h l i g h t s  Effect of high temperature on columns upgraded with different techniques is studied.  50 RC circular columns (14 unstrengthened and 36 strengthened) were prepared.  Columns were exposed to heating regimes from 100 °C to 800 °C for a period of 3 h.  Columns were left to cool down at room temperature and then axially loaded to failure.  Techniques in order of decreasing efficiency are: GF/NSM, TRM/NSM, CF and CFS/NSM.

a r t i c l e

i n f o

Article history: Received 7 October 2015 Received in revised form 18 February 2016 Accepted 8 April 2016 Available online 22 April 2016 Keywords: RC columns Elevated temperature Strengthening techniques FRP TRM

a b s t r a c t This paper examines the effect of different elevated temperature environments on performance of reinforced concrete (RC) circular columns strengthened with different techniques. For this purpose, 50 RC circular column specimens were prepared. The test matrix was divided into five groups, of which the first group was composed of 14 unstrengthened columns. The second group consisted of 14 columns strengthened with one layer of continuous carbon fiber reinforced polymer (CFRP) sheet. The third group was composed of 14 columns strengthened with one layer of 5 CFRP strips combined with near surface mounted (NSM) steel bars. Out of the 14 columns in the second or third groups, 4 specimens from each group were insulated using cement-based, fire protection mortar. The fourth group involved 4 columns strengthened with one layer of continuous glass fiber reinforced polymer (GFRP) sheet combined with NSM steel bars. The last group of columns consisted of 4 columns strengthened with continuous textile reinforced mortar (TRM) layers combined with NSM steel bars. It should be noted that strengthened columns of different groups were designed to have approximately the same axial load capacity at ambient temperature. Unstrengthened as well as strengthened columns were subjected to ambient temperature and heating regimes of 100 °C, 200 °C, 300 °C, 400 °C, 500 °C and 800 °C for a period of 3 h. Thereafter, they were left to cool down at ambient temperature and then loaded under uniaxial compression until failure. The behavior and failure modes of different column groups exposed to high degrees of temperature are presented. Comparisons between different strengthening techniques in terms of their resistance to elevated temperature are discussed. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Repair and/or strengthening of reinforced concrete (RC) columns is required for several reasons such as: extension of their lifetime, column degradation due to lack of maintenance, and the need to carry more loads than their designed values. Several strengthening measures have been developed by researchers and practicing engineers for RC columns. ⇑ Corresponding author. E-mail address: [email protected] (H.M. Elsanadedy). http://dx.doi.org/10.1016/j.conbuildmat.2016.04.041 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.

One of the most widely used methods to strengthen or repair RC columns is to construct a reinforced concrete jacket around the original column as studied by many researchers [1,2]. Another method of strengthening involves encasing the column with steel shells and filling the gap with a non-shrink grout in order to provide passive confinement to core concrete. Numerous studies investigated the use of steel jackets in strengthening of RC columns [3–5]. Another technique involves wrapping of RC columns by fiber reinforced polymer (FRP) composites for strength and ductility enhancement. FRP composites present an attractive option as an alternative and extremely efficient technique in strengthening of

346

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

Ø6mm @ 200 mm o.c 4Ø10 mm

Ø6 mm @ 200 mm o.c

H = 900 mm D = 242 mm

4Ø10 mm

30 mm

D = 242 mm

Elevation

Cross-section

Fig. 1. Details of unstrengthened column specimens (group C).

Table 1 Test matrix. Group ID

Strengthening system

Specimen ID

Temperaturea (°C)

Thermal insulation

C

Unstrengthened

C-R C-100 C-200 C-300 C-400 C-500 C-800

Room temp. (26 °C) 100 200 300 400 500 800

No No No No No No No

2 2 2 2 2 2 2

CF

One layer of full CFRP sheet

CF-R CF-100 CF-200 CF-300 CF-400 CF-500-IN CF-800-IN

Room temp. (26 °C) 100 200 300 400 500 800

No No No No No Yes Yes

2 2 2 2 2 2 2

CFS/NSM

One layer of 5 CFRP strips + 4Ø12 mm NSM steel bars

CFS/NSM-R CFS/NSM-100 CFS/NSM-200 CFS/NSM-300 CFS/NSM-400 CFS/NSM-500-IN CFS/NSM-800-IN

Room temp. (26 °C) 100 200 300 400 500 800

No No No No No Yes Yes

2 2 2 2 2 2 2

GF/NSM

One layer of full GFRP sheet + 4Ø16 mm NSM steel bars

GF/NSM-R GF/NSM-300

Room temp. (26 °C) 300

No No

2 2

TRM/NSM

3 layers of TRM composite + 4Ø16 mm NSM steel bars

TRM/NSM-R TRM/NSM-300

Room temp. (26 °C) 300

No No

2 2

Total No. of specimens a

No. of specimens

50

Elevated temperature exposure was for a duration of 3 h.

RC columns due to several reasons such as: their high strength-toweight and stiffness-to-weight ratios, large deformation capacity, minimal change in the geometry, corrosion resistance to environmental degradation and speed of application. Use of FRP jackets to strengthen RC columns has been studied by several investigators [6–17]. Another method of strengthening RC columns is to use near surface mounted (NSM) reinforcement, which involves cutting grooves into the concrete cover and bonding either steel or FRP bars inside the grooves through the use of proper filler (typically cement-based mortar or epoxy resin). This idea of NSM reinforcement started in Europe for steel bars, but recently when advanced materials, such as high quality epoxies and FRP composites, became available, the technique was given ample consideration by the research community and practicing engineers [18]. NSM FRP bars or strips were studied by some researchers for flexural capacity enhancement of RC columns subjected to bending and compression [19–21]. The viability of using the NSM reinforcement

in combination with external FRP confinement to enhance the flexural resistance and ductility of RC columns under axial load and bending was demonstrated in recent studies by Barros et al. [19] and by El-Maaddawy and El-Dieb [22]. As an alternative to FRP jacketing of columns, another technique is to use textilereinforced mortar (TRM) jacketing for column confinement as studied by Bournas et al. [23] and Ortlepp at al [24]. TRM composites can be combined with NSM reinforcement for flexural strengthening of RC columns as experimentally investigated by Bournas and Triantafillou [18]. Among all previous techniques used for strengthening of RC columns, FRP composites are mostly affected by exposure to fire or elevated temperature environments. FRP composites are vulnerable to ignition of their polymer matrix. Furthermore, strength and stiffness of polymer matrices get significantly reduced if heated above their glass transition temperature (Tg). Hence, if left uninsulated, FRP composites may ignite with increased flame

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

H = 900 mm

347

1 layer of CFRP sheet

D = 242 mm

Elevation

(a) Group CF

Elevation

C ro s s - s e c t i o n

C ro s s - s e c t i o n

(b) Group CFS/NSM

H = 900 mm

1 layer of GFRP sheet

D = 242 mm

Elevation

(c) Group GF/NSM

Cross-section

Fig. 2. Details of FRP-strengthened column specimens.

spread and toxic smoke evolution, and they may quickly lose bond and/or mechanical properties [25,26]. Therefore, the behavior of FRP-strengthened RC columns exposed to fire or elevated temperature environments is of big concern. As yet, research in this area is limited, and more work is needed. The objective of current research is to fill some of the gaps in learning the behavior of RC columns strengthened with different techniques, viz. externally bonded FRP composites and NSM reinforcement combined with either external FRP confinement or TRM composites, and exposed to high temperature.

2. Literature review Limited research exists on the performance of FRPstrengthened RC columns under high temperature exposure. The

effect of high temperature on GFRP-wrapped concrete cylinders was studied experimentally by Saafi and Romine [27]. In their study, the thermal resistance of the epoxy matrix was found to have a substantial effect on the performance of the wrapped cylinders. The GFRP composite laminates had a severe damage due to creep and melting of epoxy when the specimens were heated at a temperature equal to or above the glass transition temperature (Tg) of the epoxy matrix. This damage was more noticeable after 3 h of high temperature exposure. A significant loss in strength and ductility was noticed for specimens heated at temperature equal to 2Tg for a period of 3 h. Even though a 50% loss in compressive strength was evident after 3 h of exposure to high temperature, the residual capacity remained greater than the service strength. In a study by Cleary et al. [28], concrete cylinders were strengthened with externally bonded GFRP system, subjected to a range of high temperature between 120 °C and 180 °C and then

348

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

H = 900 mm

3 layers of TRM composite

D = 242 mm

Elevation

C r o s s - s e c ti o n

Fig. 3. Details of TRM-strengthened column specimens (group TRM/NSM).

(a) Formwork ready for casting

(b) Columns after casting & leveling

Fig. 4. Formwork & casting for RC columns.

(a) Carbon fibers used in this study

(b) CFRP sheet application

(c) Column after CFRP applied

Fig. 5. Strengthening of columns of group CF.

left to cool down to ambient temperature. Subsequently, GFRPwrapped cylinders were loaded in axial compression till failure.

It was found that there was no statistically substantial reduction of compressive strength until the elevated temperature was more

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

(a) Ø16 mm NSM bars (b) Column after all grooves filled placed in the with Sikadur-31 groove mortar

(c) Column after CFRP strips applied

349

(d) E-glass fibers used in this (e) Column after GFRP sheet study applied

Fig. 6. Strengthening of columns of groups CFS/NSM & GF/NSM.

(a) First layer of TRM applied with mortar

(b) Second layer of TRM applied

(c) Column after 3 layers of TRM applied

Fig. 7. Strengthening of columns of group TRM/NSM.

than 30 °C above the glass transition temperature of the epoxy. The loss of compressive strength due to high temperature exposure was reduced after the application of a fire protection material. El-Karmoty [29] studied experimentally the effect of high temperature on the behavior of RC circular columns strengthened with GFRP strips and thermally insulated with two types of fire protection systems. Seven RC circular columns were tested. Two control columns were loaded under monotonic axial compression till failure to evaluate the strength at room temperature; one specimen was unwrapped and the other was wrapped with strips of GFRP composite. The other five columns were exposed to high temperature of 600 °C, while being loaded, to simulate the actual situation in structures. Out of the five specimens, two columns were left unwrapped and were exposed to 600 °C for a duration of 1 and 2 h, respectively. Another specimen was wrapped with GFRP strips and exposed to 600 °C for 2 h. The last two specimens were confined with GFRP strips and then protected with two

different types of insulation materials and hence, they were heated at 600 °C for 2 h. Upon high temperature exposure for the specified duration, the five specimens were left to cool down for 24 h and then they were reloaded till failure to measure the residual strength. It was concluded that using insulation materials increased the ultimate load capacity of the GFRP-strengthened columns (with respect to the unprotected column) but the ultimate loads of the insulated columns were still smaller than that of the unheated GFRP-wrapped column. Cree et al. [30] investigated experimentally the behavior in fire of insulated FRP-strengthened RC columns. The test program comprised two fire tests: one circular and one square RC column wrapped with two and three layers of CFRP laminates, respectively. Both columns were then thermally insulated with a commercially available fire protection mortar with an average thickness of 44 mm for circular column, while the square column had an average insulation thickness on flat surfaces of 40 mm and average

350

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

Table 2 Material properties for test specimens. Material

a

Properties

Values

Notes On the day of test

Concrete

Specified compressive strength (MPa)

42

Steel bars

– Yield strength (MPa) Ultimate strength (MPa)

Ø6 Ø10 Ø12 Ø16 Based on tension tests 302 (280) 593 (520) 590 (520) 498 (420) (Based on manufacturer’s datasheet) 380 (350) 700 (650) 706 (650) 680 (600)

FRP material

– Thickness per layer (mm) Ultimate tensile strength (MPa) Ultimate tensile strain Tensile modulus of elasticity (GPa)

CFRP system 1.0 846 1.1% 77.3

Epoxy adhesive for FRP sheets

Tensile strength (MPa) Tensile modulus of elasticity (GPa) Tensile strain at break Glass transition temperature (°C) Thermal decomposition temperaturea (°C)

71.6 1.86 5.25% 85 345

Based on manufacturer’s datasheet

Epoxy adhesive mortar for NSM bars

Compressive strength (MPa) Flexural strength (MPa) Tensile strength (MPa) Young’s modulus (GPa) Elongation at break Bond strength to concrete (MPa) Glass transition temperature (°C)

65 30 20 8.5 1.5% 3.5 58

Based on manufacturer’s datasheet

Mortar for TRM composite

28-day tensile strength (MPa) 28-day compressive strength (MPa)

3.4 56.4

Based on standard test specimens

Bare carbon textile for TRM composite Smeared thickness per layer (mm) Tensile strength (MPa) Elastic modulus (GPa)

0.2 1500 150

Based on manufacturer’s datasheet

TRM composite

Thickness per layer (mm) Tensile strength (MPa) Elastic modulus (GPa)

5 4.25 0.5

Based on nonstandard test coupons

Insulation material (Sikacrete-213F)

Layer thickness (mm) Compressive strength (MPa) Thermal conductivity (W/mK)

40 2.0 0.23

Based on manufacturer’s datasheet

GFRP system 1.3 460 1.8% 20.9

Based on standard test coupons

Obtained from reference [39].

Fig. 8. Unstrengthened & strengthened specimens inside the oven ready for heating.

corner insulation thickness of 51 mm. The insulated FRPstrengthened columns were subjected to the CAN/ULC S101 [31] standard fire tests. The insulation system was efficient in protecting the strengthened columns such that they were able to reach 4 h fire endurance ratings in accordance with CAN/ULC S101 [31] and ASTM E119 [32]. However, the fire protection material couldn’t keep the temperature of the FRP below its glass transition temperature for the duration of the fire endurance test. Al-Salloum et al. [33] carried out a test program to study the effect of elevated temperature on the behavior of FRP-wrapped

concrete cylinders. In this program, 42 cylinder specimens were prepared. Fourteen specimens were left unwrapped; whereas the remaining cylinders were strengthened with one layer of CFRP (or GFRP) sheets. Some cylinders were subjected to ambient temperature; while other specimens were subjected to high temperatures of 100 °C and 200 °C for a period of 1, 2 or 3 h. Subsequently, they were cooled down to room temperature after which, they were tested under axial compression to failure. Results revealed that at a temperature of 100 °C (slightly more than the glass transition temperature (Tg) of the epoxy), both CFRP- and GFRPstrengthened cylinders had small loss in strength due to epoxy melting. The loss of strength was more evident for a temperature of 200 °C. Even though the behavior of FRP-upgraded concrete members at normal temperature is acceptable, information about their performance after being exposed to high temperatures is limited. The objective of this research is to examine the effect of different high temperature environments on residual axial capacity of RC circular columns strengthened with different techniques. These techniques involve: uninsulated and insulated continuous CFRP sheet, uninsulated and insulated CFRP strips combined with NSM steel bars, continuous GFRP sheet combined with NSM steel bars and TRM layers combined with NSM steel bars. Unstrengthened as well as strengthened column specimens were prepared and then subjected to ambient temperature and heating regimes of 100 °C, 200 °C, 300 °C, 400 °C, 500 °C and 800 °C for a period of 3 h. After being subjected to elevated temperatures, columns were left to cool down at room temperature and then tested under axial compression until failure.

351

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

Fig. 9. Individual time-temperature curves used in the study.

(a) Group C exposed to 100 °C

(b) Group CF exposed to 300 °C

(c) Groups C & CF exposed to 400 °C

(d) Insulated group CF exposed to 500 °C

(e) Insulated group CF exposed to 800 °C

Fig. 10. Post heated columns of groups C & CF.

mounted (NSM) steel reinforcement in addition to TRM layers with NSM steel bars. In this study, a total of 50 RC columns of diameter 242 mm and length 900 mm were cast and subjected to elevated temperatures ranging from 100 °C to 800 °C and then tested under uniaxial compression. Fig. 1 shows the details of the reinforcement arrangement for the unstrengthened column utilized in this study. As shown in the figure, the reinforcement comprised of 4Ø10 mm ribbed longitudinal bars tied by 6 mm diameter rebars at a center-to-center spacing of 200 mm. This provided a longitudinal steel ratio of 0.68% which is less than the minimum reinforcement (1%) for RC columns specified in the ACI 318-14 code [34]. The reinforcement area was selected in an effort to account for the loss in steel area as a result of steel corrosion in the aggressive environment of the coastal regions of Saudi Arabia, which would then necessitate the strengthening of columns using different measures. The design of strengthening systems was executed keeping in mind that the axial load carrying capacity of the strengthened specimen at room temperature for all systems shall be approximately same. The axial load capacity of strengthened RC columns at room temperature was calculated based on the recommendations provided in the ACI 440.2R-08 guidelines [35] (see Appendix A). As seen from the appendix, the maximum difference between the calculated nominal axial capacities at room temperature for all strengthened columns is 5.5%. Fig. 11. Post heated columns of group CFS/NSM. 3. Experimental program An experimental study was carried out in the current work to examine the effect of elevated temperature on reinforced concrete columns, axially strengthened with 4 different systems incorporating FRP sheets with and without near surface

3.1. Test matrix Table 1 shows the test matrix for the experimental program employed in this study. Out of the total 50 specimens cast, 14 columns belonged to group C which were unstrengthened, whereas the remaining columns were strengthened using four different systems. Group CF comprised of 14 columns strengthened using one full layer of CFRP sheet as shown in Fig. 2(a), and group CFS/NSM comprised

352

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

(a) Unstrengthened column

(b) Strengthened column

Fig. 13. Instrumented column specimen ready for testing.

(a) Group GF/NSM exposed to 300 °C

(b) Group TRM/NSM exposed to 300 °C

Fig. 12. Post heated columns of groups GF/NSM & TRM/NSM.

of another 14 columns which were strengthened using a combination system of NSM steel bars and CFRP strips. As shown in Fig. 2(b), the CFS/NSM strengthening system comprised of 4Ø12 mm NSM steel bars and 5 CFRP strips of 100 mm width. Group GF/NSM included 4 columns which were strengthened using a combination system of one full layer of GFRP sheet and 4Ø16 mm NSM steel bars as shown in Fig. 2(c). Finally, group TRM/NSM comprised of 4 columns strengthened with a combination system of 3 layers of textile reinforced mortar (TRM) and 4Ø16 mm NSM steel bars. The strengthening scheme for TRM/NSM is depicted in Fig. 3. Two columns from each group were tested at room temperature and were treated as control specimen for specific groups. All other columns were subjected to elevated temperatures for a period of 3 h as per the heating schedule depicted in Table 1 and then tested. Four column specimens each from group CF & CFS/ NSM were insulated using a locally available insulation material before subjecting them to the elevated temperature in order to investigate the effect of insulation.

3.2. Preparation of specimens Before placing concrete, the reinforcement cage was prepared as per the design and placed in formwork made of plastic with a constant cover of 30 mm on all sides. Special provisions were made within the formwork to achieve the grooves in the columns which would be used for placing the NSM bars. Ready-mix concrete was acquired from a local ready-mix company and all casting took place in the structural & concrete laboratories of King Saud University. All columns were cast using the same batch of concrete to avoid any material inconsistencies. Fig. 4(a) shows formwork ready for casting and Fig. 4(b) shows columns after casting is completed. Strengthening of RC columns was undertaken once the curing period of 28 days had expired. All columns to be strengthened were sand-blasted to clear any dirt, grime and loose materials from the surface. After applying acetone cleaner, the columns were ready for NSM rebar and FRP application. For the case of CFRPstrengthened specimens of group CF, standard manufacturer recommended procedures were adopted for FRP application, wherein the two-component epoxy-based resin was first applied on the column surface in a thin layer and then the CFRP sheet was completely saturated using the resin. The saturated sheet was then applied to the column surface taking care that all air voids in the sheet are removed. The CFRP sheet had the fibers oriented in the hoop direction with an overlap of 150 mm, and was applied in three circumferential strips over the length of a column owing to the limited width of the sheet as shown in Fig. 5. Same procedure was adopted for the GFRP sheet and CFRP strips wrapping of columns of groups GF/NSM & CFS/NSM, respectively. For the columns of groups CFS/NSM, GF/NSM & TRM/NSM, after sand blasting, steel rebars were placed in the center of the grooves as shown in Fig. 6(a) and a structural two part adhesive mortar (Sika 31) was applied all around the steel bar (see Fig. 6(b) and (c)) using a trowel. Care was taken to make sure the adhesive mortar covered the entire cross-sectional area within the groove. Once all the rebars were placed and the grooves filled with the adhesive material, the columns were then strengthened

using the specified FRP application as shown in Fig. 6(d) and (e). For the columns of group TRM/NSM, a 2-mm thick single layer of a three-component epoxy modified cementitious bonding agent was first applied on the column surface. A multipurpose polymer modified cementitious mortar was then mixed in water and used to bond the textile sheet on to the column surface. As shown in Fig. 7(a) and (b), a 5-mm thick layer of the mortar was first applied to the column over which the textile was slightly pressed until the mortar jutted out of the apertures between the textile rovings. A second layer of mortar was then applied so as to cover the textile sheet completely. The same procedure was then repeated for the second and third layers of TRM strengthening, with the third layer having an overlap of 300 mm as shown in Fig. 3. The insulation of the columns was carried out as per the manufacturer’s recommendation by using a minimum thickness of 40 mm and applying the insulation mortar material using a spray concrete system. The CFRP surface was prepared prior to application by cleaning and removing all carbon dust from the surface. The composite surface was then primed using epoxy resin, over which the fire-resistant mortar was then wet sprayed. 3.3. Material properties 3.3.1. Concrete Ready-mix concrete having a cement content of 400 kg/m3 was used for casting the RC column specimens. The maximum size of the aggregate used was 10 mm. The specified compressive strength measured as per the ASTM C39M [36] test for measuring compressive strength of cylinders at the time of the test, was 42 MPa. 3.3.2. Reinforcement bars Locally manufactured ribbed steel rebars were acquired for the longitudinal as well as tie reinforcement. Tensile tests were carried out in accordance with ASTM E8/E8M [37] on all diameters of rebars used in the study and the results are presented in Table 2. 3.3.3. FRP laminates Both CFRP and GFRP laminates used in this study were uni-directional. A twocomponent, resin-based epoxy was used as the adhesive and was to be mixed in a ratio of 100:34.5 by weight. Tensile tests in accordance with ASTM D3039/3039M [38] were carried out on standard coupons of both CFRP and GFRP materials. Table 2 enlists the properties for the composite gross laminate for both the CFRP and GFRP systems. The properties of the epoxy adhesive as provided by the manufacturer’s datasheet are also depicted in Table 2. 3.3.4. Textile reinforced mortar (TRM) Locally available patching and repair mortar (SikaRep) was used in combination with imported custom made bi-directional carbon fiber textile to form the textile reinforced mortar strengthening system. For determination of mortar compressive strength, 50-mm cubes were prepared and then tested according to ASTM C109/ C109M [40]. The tensile strength of mortar was measured using briquette specimens, which were prepared and tested according to ASTM C190 [41]. Fig. 7(a) shows the textile fiber used in this study. The carbon strands were 4 mm wide and spaced at 10 mm c/c in either directions. Properties of bare carbon textile as provided by the manufacturer’s datasheet are enlisted in Table 2. Material

353

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361 Table 3 Experimental results for column specimens. Group ID

Specimen ID

Specimen No.

Peak axial load (kN)

Secant stiffness at service load (kN/mm)

Peak average stress

Peak actual stress

C

C-R

1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average

1998 1966 1982 1880 1923 1901 1745 1786 1765 1507 1584 1545 1488 1454 1471 1354 1297 1325 780 883 832

1456 1358 1407 1230 1258 1244 1122 1191 1157 700 751 725 570 557 563 408 391 399 237 267 252

43.4 42.8 43.1 40.9 41.8 41.3 37.9 38.8 38.4 32.8 34.4 33.6 32 31.6 32.0 29.4 28.2 28.8 17.0 19.2 18.1

1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average

2738 2601 2669 2484 2379 2431 2189 2245 2217 2163 2130 2146 1897 1972 1934 2375 2284 2329 1452 1369 1410

1299 1235 1267 1258 1206 1232 1127 1155 1141 886 873 879 680 707 694 1027 988 1007 534 504 519

1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average 1 2 Average

2514 2609 2561 2474 2414 2444 2201 2174 2188 2026 2093 2060 1196 1256 1226 2321 2225 2273 1114 1205 1159

1 2 Average 1 2 Average

2602 2713 2658 2496 2387 2442

C-100

C-200

C-300

C-400

C-500

C-800

CF

CF-R

CF-100

CF-200

CF-300

CF-400

CF-500-IN

CF-800-IN

CFS/NSM

CFS/NSM-R

CFS/NSM-100

CFS/NSM-200

CFS/NSM-300

CFS/NSM-400

CFS/NSM-500-IN

CFS/NSM-800-IN

GF/NSM

GF/NSM-R

GF/NSM-300

Concrete compressive strength (MPa)

Axial strain at peak stress

Peak axial strain

Peak lateral strain

39.7 39.0 39.3 37.1 38.0 37.5 34.1 35.0 34.6 28.9 30.6 29.8 28 27.8 28.1 25.6 24.3 24.9 13.0 15.2 14.1

0.0019 0.0017 0.0018 0.0018 0.0020 0.0019 0.0027 0.0019 0.0023 0.0024 0.0030 0.0027 0.0036 0.0029 0.0033 0.0054 0.0044 0.0049 0.0040 0.0050 0.0045

0.0041 0.0037 0.0039 0.0028 0.0033 0.0030 0.0035 0.0034 0.0035 0.0032 0.0038 0.0035 0.0035 0.0030 0.0033 0.0067 0.0056 0.0062 0.0102 0.0121 0.0112

0.0011 0.0011 0.0011 0.0015 0.0019 0.0017 0.0024 0.0027 0.0025 0.0024 0.0027 0.0026 0.0032 0.0029 0.0030 0.0043 0.0038 0.0040 0.0081 0.0092 0.0087

59.5 56.5 58.0 54.0 51.7 52.9 47.6 48.8 48.2 47.0 46.3 46.7 41.2 42.9 42.1 51.6 49.7 50.6 31.6 29.8 30.7

55.9 52.9 54.4 50.3 48.0 49.1 43.9 45.1 44.5 43.3 42.6 42.9 37.4 39.1 38.3 47.9 45.9 46.9 27.7 25.9 26.8

0.0071 0.0075 0.0073 0.0073 0.0062 0.0068 0.0046 0.0052 0.0049 0.0049 0.0044 0.0047 0.0035 0.0039 0.0037 0.0065 0.0058 0.0062 0.0037 0.0033 0.0035

0.0076 0.0075 0.0076 0.0073 0.0062 0.0068 0.0049 0.0056 0.0052 0.0053 0.0047 0.0050 0.0034 0.0039 0.0037 0.0066 0.0058 0.0062 0.0049 0.0043 0.0046

0.0099 0.0085 0.0092 0.0086 0.0076 0.0081 0.0055 0.0063 0.0059 0.0058 0.0051 0.0055 0.0034 0.0039 0.0037 0.0081 0.0071 0.0076 0.0009 0.0008 0.0008

1578 1636 1607 1435 1400 1417 1120 1106 1113 827 854 840 504 529 517 930 892 911 418 453 435

54.7 56.7 55.7 53.8 52.5 53.1 47.9 47.3 47.6 44.1 45.5 44.8 26.0 27.3 26.7 50.5 48.4 49.4 24.2 26.2 25.2

47.7 45.6 46.6 44.7 43.4 44.0 38.7 38.1 38.4 34.8 36.3 35.5 16.4 17.7 17.1 41.3 39.2 40.2 14.6 16.6 15.6

0.0031 0.0035 0.0033 0.0049 0.0043 0.0046 0.0043 0.0038 0.0040 0.0053 0.0059 0.0056 0.0025 0.0029 0.0027 0.0069 0.0060 0.0065 0.0027 0.0032 0.0030

0.0092 0.0105 0.0099 0.0095 0.0084 0.0089 0.0082 0.0072 0.0077 0.0063 0.0072 0.0068 0.0039 0.0044 0.0042 0.0074 0.0066 0.0070 0.0058 0.0064 0.0061

0.0087 0.0099 0.0093 0.0090 0.0079 0.0085 0.0083 0.0073 0.0078 0.0067 0.0076 0.0071 0.0009 0.0011 0.0010 0.0041 0.0035 0.0038 0.0011 0.0014 0.0012

1274 1327 1300 1152 1102 1127

56.6 59.0 57.8 54.3 51.9 53.1

46.3 48.8 47.5 44.0 41.5 42.7

0.0032 0.0036 0.0034 0.0070 0.0059 0.0065

0.0099 0.0113 0.0106 0.0083 0.0076 0.0080

0.0137 0.0151 0.0144 0.0077 0.0064 0.0070

(continued on next page)

354

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

Table 3 (continued) Group ID

Specimen ID

Specimen No.

Peak axial load (kN)

Secant stiffness at service load (kN/mm)

TRM/NSM

TRM/NSM-R

1 2 Average 1 2 Average

2624 2777 2701 2178 2344 2261

1773 1872 1823 1252 1347 1299

TRM/NSM-300

properties of TRM composite are also presented in Table 2, which were obtained from non-standard TRM coupons prepared using the mortar and a single layer of carbon fiber textile, for testing under uni-axial tension until failure.

3.3.5. Insulation material The insulation material used in the study was locally available Sikacrete-213F, which is cement based dry-mix fire protection mortar. It is specifically designed to protect FRP-strengthened surfaces from elevated temperatures or fire. Manufacturer provided properties of the material have been presented in Table 2.

3.4. Heating of column specimens The heating of columns to the desired elevated temperatures was carried out using an electric oven of internal dimensions 1 m  1 m  1 m, which is shown in Fig. 8. The columns were heated after an age of two months after casting and one month after the strengthening systems were applied. At a time, 3–4 columns were placed in the oven and heated for a period of 3 h at the desired temperature as per the test matrix. Individual time-temperature curves used in the study for each temperature exposure are shown in Fig. 9. Type-K thermocouples provided by the manufacturer were used to measure the temperature inside the oven. The output of the sensors was directly fed to a read-out on the outside. After the desired elevated temperature was reached and the exposure time of 3 h expired, the oven would shut-off automatically. The columns were allowed to cool off naturally, first inside the oven and then after the temperature fell to 200 °C, by opening the oven door. For columns of group C, no visible damage was observed at temperatures of 100, 200 and 300 °C, whereas at 400, 500 and 800 °C concrete appeared to be darkened as a result of heating as shown in Fig. 10(c). Some thermal cracking was also observed on the surface as a result of removal of free water from within the concrete. For columns of group CF which were strengthened with one layer of CFRP sheet, at 100 & 200 °C, the epoxy appeared to melt while the columns were in the oven, however post-cooling the epoxy regained its original state. As the temperature was increased to 300 and 400 °C severe damage was noticed to the epoxy– fiber interface and charring of the fibers was also noticed. No de-bonding of CFRP sheet was noticed though. Insulated columns of this group subjected to 500 °C fared much better as a result of the insulation provided by the material. However, the insulated group CF columns subjected to 800 °C had visible charring of the composite strands but still no de-bonding was noticed as shown in Fig. 10. Post-heated condition of the FRP composite for columns of groups CFS/NSM and GF/NSM was similar to as described for group CF at the specified temperatures of exposure as shown in Figs. 11 and 12. Fig. 11 shows the post-heated condition of columns for group CFS/NSM after subjecting them to different elevated temperatures. For column subjected to 100 & 200 °C, the concrete and epoxy adhesive mortar of the NSM bars did not show any visible outward signs of material degradation. At temperatures of 300 and 400 °C, discoloration of epoxy mortar and some cracks on its surface were noticed especially at 400 °C. The bond between the epoxy mortar and concrete still appeared to be intact at 300 °C; whereas at 400 °C, bond failure was noticed between epoxy mortar and concrete at some locations. Insulation material did appear to be effective in reducing temperature induced damage at 500 °C, but severe charring of CFRP strips and discoloration of the exposed NSM epoxy mortar was noticed for the column subjected to 800 °C. Expansion of epoxy mortar was noticed at some locations indicating bond failure between concrete substrate and NSM epoxy at some locations. It was noticed that the columns of group TRM/NSM subjected to a temperature of 300 °C for a period of 3 h appeared to be undamaged as the mortar layer was effective in protecting the carbon fiber textile against the adverse effects of direct fiber exposure. However, discoloration of mortar was noticed and also some minor cracks were observed on the surface of the mortar as shown in Fig. 12.

Concrete compressive strength (MPa) Peak average stress

Peak actual stress

57.1 60.4 58.7 47.3 51.0 49.2

46.8 50.2 48.5 36.8 40.6 38.7

Axial strain at peak stress

Peak axial strain

Peak lateral strain

0.0027 0.0031 0.0029 0.0034 0.0040 0.0037

0.0159 0.0178 0.0168 0.0107 0.0125 0.0116

0.0056 0.0066 0.0061 0.0042 0.0052 0.0047

3.5. Test setup and procedure Testing of all RC columns was carried out using an AMSLER press with a compression capacity of 1000 tons. Instrumentation of the columns comprised of 2 linear voltage displacement transducers (LVDT) to measure the axial displacement in a gauge length of 300 mm within the middle-third of the column length. A built-in load cell of the machine was used to acquire the load data. Strain gauges were used to measure both concrete and FRP strains in the axial as well as lateral directions. A total of four strain gauges were attached at mid-height using cyanoacrylate adhesive, of which two were attached in the axial and two in lateral direction on opposite faces of the column. All columns were loaded until failure under uniaxial compression with a displacement controlled loading rate of 0.5 mm/min. Experimental data was recorded using a data acquisition system at intervals of 1 s. Extreme care was taken to make sure the columns were loaded concentrically and uniform loading is achieved, by providing gypsum capping at both the top and bottom ends. In order to avoid local failure at extreme ends of the columns due to stress concentration, the top and bottom 150 mm lengths of the columns were fortified using 3 layers of CFRP sheet as shown in Fig. 13.

4. Experimental results Table 3 shows a summary of experimental results for all column groups tested in this study. Results are presented in terms of peak axial load, secant stiffness at service load, peak average and actual axial stresses, and peak strains in axial and lateral directions. It should be noted that the individual test results of the replicate columns along with their average values are enlisted in Table 3. In this study, the secant stiffness at service load is defined as the ratio of service compressive axial load to the axial displacement at that load. The service load was considered to be 40% of the ultimate axial load. The peak average axial stress enlisted in Table 3 is obtained by dividing the peak axial load by the cross-sectional area of the column. However, the peak actual concrete stress presented in Table 3 was calculated from the following equation which is based on the ACI 318-14 code [34] and the ACI 440.2R08 guidelines [35]. 0

Peak actual stress ¼ f cc;T ¼

Pu  Ast f y Ag  Ast

ð1Þ

where Pu = peak axial load; Ast = area of longitudinal steel bars; fy = yield strength of steel bars and Ag = gross area of column section. Details of test results are described in the following. 4.1. Failure modes Fig. 14 shows the failure mode for columns of group C at room and after exposure to elevated temperatures. As seen from the figure, a typical failure mode was observed for the column at room temperature wherein the concrete at mid-height crushed as a result of pure axial compression. The crushing of concrete was followed by explosive spalling. Increase in loading resulted in complete spalling of concrete thereby exposing the steel rebars which were found to be bent at mid-height. Local failure at extreme ends was avoided as a result of restraining column ends using CFRP wraps. For the heated specimens, a less explosive failure was

355

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

noticed as the degraded concrete cover started spalling prematurely, ultimately resulting in the reduced axial capacity for columns damaged as a result of exposure to elevated temperature. Also as seen in Fig. 14, non-uniform loss of concrete cover was observed for the heat-damaged columns. Fig. 15 depicts the failure modes of group CF strengthened with a single layer of CFRP sheet. For the column at room temperature, failure was explosive with a loud noise as a result of sudden energy release and was characterized by complete rupture of the CFRP sheet at mid-height location as depicted in Fig. 15(a). However, the failure mode for heated specimens was far less explosive and uneven tearing of FRP sheet was noticed as failure initiated at regions of degraded FRP material. For the insulated column subjected to an elevated temperature of 800 °C, the CFRP sheet was completely damaged as shown in Fig. 15(e). Fig. 16 displays the failure modes for group CFS/NSM specimens. For the column at room temperature, failure was initiated with spalling of concrete cover, cracking of middle CFRP strip and the expansion of concrete core. As the test continued, final failure mode involved rupture of middle CFRP strip, cracking of NSM epoxy mortar and buckling of NSM bars. For the heat damaged columns, the failure mode was similar to the benchmark column except the damage to CFRP strips was more severe as the exposure temperature increased. For columns heated to 400 °C the NSM epoxy completely disintegrated towards the end of the test and the NSM bars were completely exposed. The effect of insulation was clearly visible for the 500 °C column (Fig. 16(d)) as compared with the 800 °C column (Fig. 16(e)). The failure mode for columns of groups GF/NSM and TRM/NSM tested at room and after an exposure to 300 °C is shown in Fig. 17. As seen from the figure, the failure mode of GF/NSM columns was similar to groups CF and CFS/NSM, identified by tearing of GFRP sheet which initiated the failure of NSM epoxy mortar. However, the complete wrapping of the GFRP jacket prevented severe damage to the NSM epoxy as the test progressed. The failure of these columns was also much less explosive as compared with the CFRP-wrapped columns. Fig. 17(c) and (d) shows the failure mode for columns of group TRM/NSM. Failure in these columns was initiated with spalling of the outermost layer of repair mortar, at midheight of the columns. Further expansion of the concrete core resulted in failure of mortar and textile interface and finally as a result of failure of TRM system, the confinement to the column

(a) Room

(b) 200 °C

(c) 300 °C

was lost resulting in the reduction of load-carrying capacity. The mode of failure for the heated specimen of this group was very similar to the benchmark column except as a result of degradation due to heating, some non-uniform spalling was observed in mortar material.

4.2. Stress-strain curves The stress-strain curves for the group C (unstrengthened) specimens at room and tested after exposure to elevated temperatures is shown in Fig. 18. In the figure, the axial stress is plotted against both the axial and lateral strains. In this group of columns, a general trend was observed where the axial stress gradually decreased with the increase in exposure temperature. Columns subjected to elevated temperatures behaved in a more ductile manner with an increase in the values of axial and lateral strains at peak axial stress with the increase in temperature. This is due to the fact that the exposure of concrete to elevated temperatures causes various physical (e.g., evaporation, condensation, water and vapor advection, vapor diffusion, heat conduction and advection, phase expansion), chemical (e.g., dehydration, thermo-chemical damage) and mechanical (e.g., thermo-mechanical damage, cracking, spalling) processes, resulting in concrete deterioration thus losing its strength and stiffness. Degradation in stiffness at high temperature exposure means higher axial/lateral strains at the same stress level which results in increased deformability (ductility) of columns, as mentioned previously. Both the decrease in the axial stress and the increase in axial and lateral strain values was more pronounced for specimens subjected to temperatures of 300 °C and more, compared with specimens subjected to temperatures of 100 and 200 °C. In general a reduction in column stiffness was noticed for all unstrengthened columns subjected to elevated temperatures. Fig. 19 shows the average axial stress versus axial and lateral strain curves for all groups of strengthened columns. Fig. 19(a) shows the same plot for columns of group CF and as seen from the figure a marked reduction in axial stress capacity was noticed for columns after exposure to elevated temperatures compared with the control column at room temperature. A reduction from peak axial stress of 58.0 MPa for CF-R specimen to 42.1 MPa for the CF-400 specimen was observed. In case of group CF specimens, the axial strains at peak load, peak axial strains and the peak lateral

(d) 400 °C

Fig. 14. Typical failure modes for group C specimens.

(e) 800 °C

356

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

(a) Room

(b) 300 °C

(c) 400 °C

(d) 500 °C Insulated (e) 800 °C Insulated

Fig. 15. Typical failure modes for group CF specimens.

(a) Room

(b) 300 °C

(c) 400 °C

(d) 500 °C Insulated

(e) 800 °C Insulated

Fig. 16. Typical failure modes for group CFS/NSM specimens.

strains, all decreased with the increase in the exposure temperature (100–400 °C). For the case of group CFS/NSM columns, a similar trend in axial stress was noticed as in groups C & CF where the peak axial stress decreased with the increase in exposure temperature as shown in Fig. 19(b). For uninsulated columns, peak axial stress decreased from 55.7 MPa to 26.7 MPa as the temperature increased from ambient to 400 °C. A general decrease in the peak axial and lateral strain values was also noticed for these columns. Fig. 19(c) and (d) shows the average axial stress versus axial and lateral strain plots for columns of groups GF/NSM and TRM/NSM, respectively. Exposing the columns to a temperature of 300 °C, resulted in a decrease in the peak axial stress from 57.8 MPa to 53.1 MPa and from 58.7 MPa to 49.2 MPa, respectively for the two mentioned column groups. The axial strain at peak load increased; yet, peak axial and lateral strains considerably decreased for the two groups due to elevated temperature exposure. A marked decrease in stiffness of the columns was also noticed in columns of all groups as the exposure temperature was increased.

5. Discussion of test results 5.1. Effect of elevated temperature exposure Fig. 20 shows the effect of exposure temperature on percentage loss in axial strength and stiffness for specimens of group C. As seen from Fig. 20 for unstrengthened specimens, as a result of exposure to elevated temperature, the percentage loss in the residual axial capacity and the secant stiffness of columns increases as the exposure temperature was increased from 100 °C to 800 °C. The effect of exposure temperature resulting in a decrease in both the axial strength and stiffness is much more pronounced at temperatures of 300 °C and above. A loss of 4% and 11% in the axial strength of the unstrengthened column was noticed for exposure temperatures of 100 & 200 °C whereas a maximum loss of 58% was noticed for the columns subjected to 800 °C. For exposure temperatures of 100, 200 and 800 °C, the losses in stiffness were 12%, 18% and 82%, respectively. Even though the unstrengthened columns subjected to 100 & 200 °C showed no real damage on a

357

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

(a) GF/NSM Room

(b) GF/NSM 300 °C

(c) TRM/NSM Room

(d) TRM/NSM 300 °C

Fig. 17. Typical failure modes for groups GF/NSM & TRM/NSM specimens.

50 45

Room Temp. 100 °C 200 °C 300 °C 400 °C 500 °C 800 °C

Average Stress (MPa)

40 35 30 25 20 15 10 5 0 -0.01

-0.008 -0.006 -0.004 -0.002

Lateral Strain

0

0.002

0.004

0.006

0.008

0.01

0.012

Axial Strain

Fig. 18. Stress-strain curves for unstrengthened specimens.

visual inspection, a reduction in residual load carrying capacity and stiffness was observed indicating degradation of the material at those temperatures. Fig. 21(a) and (b) shows the effect of exposure temperature on percentage loss of axial strength, stiffness and peak axial strain for groups CF and CFS/NSM, respectively. The same general trend wherein the losses in all measured strength parameters increased as the exposure temperature was increased, was noticed in specimens of both the groups. For the uninsulated columns of group CF, the percentage loss in strength increased from 9% to 28% as the temperature was increased from 100 °C to 400 °C. For the same columns the losses in stiffness were 3% & 45%, respectively. This steep decrease in the stiffness at 400 °C could be attributed to severe damage to epoxy matrix due to surpassing its decomposition temperature of 345 °C as enlisted in Table 2, which resulted in penetration of heat to concrete core making it porous and soft thereby increasing its ductile behavior. It is worth noting that even with severe visible damage to the CFRP interface, the axial strength of group CF column exposed to 400 °C was 31.5% greater than the unstrengthened column exposed to the same temperature which indicates the effectiveness of strengthening system. As shown in

Fig. 21(b), the axial strength and stiffness of group CFS/NSM subjected to 400 °C was reduced by 52% and 68%, respectively. 5.2. Effect of insulation As seen from Fig. 21(a) and (b) and Table 3, insulation did play a role in preserving the residual axial capacity for columns of both CF and CFS/NSM groups especially for an exposure temperature of 500 °C. The percentage loss in axial strength due to 500 °C exposure was 13% and 11% for the insulated groups CF and CFS/NSM, respectively. In terms of the axial strength performance, the insulated columns of both the groups exposed to a temperature of 500 °C fared even better than the uninsulated columns subjected to 200, 300 & 400 °C. This indicates excellent performance of the insulation material in preventing the heat induced damage to FRP-strengthened columns up to certain temperatures and exposure periods. The insulated column subjected to 800 °C, had a marked reduction in its axial stress capacity compared with specimens subjected to other temperatures. However, compared with the unstrengthened specimens subjected to the same temperature of 800 °C, the

358

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361 70 70 60

Average Stress (MPa)

Average Stress (MPa)

60 50 40 Room Temp. 100 °C

30

200 °C 300 °C

20

400 °C

50 40 30 20

500 °C - Insulated

10

800 °C - Insulated

10

Room Temp. 100 °C 200 °C 300 °C 400 °C 500 °C - Insulated 800 °C - Insulated

0

0 -0.01

-0.008

-0.006

-0.004

-0.002

0

0.002

Lateral Strain

0.004

0.006

-0.01

0.008

-0.008 -0.006 -0.004 -0.002

(a) Group CF

0.002

0.004

0.006

0.008

0.01

0.012

Axial Strain

(b) Group CFS/NSM

70

70

60

60

Average Stress (MPa)

Average Stress (MPa)

0

Lateral Strain

Axial Strain

50 40 30

Room Temp.

20

300 °C

10

Room Temp.

50

300 °C

40 30 20 10

0

0

-0.015

-0.012

-0.009

-0.006

-0.003

0

0.003

0.006

0.009

0.012

Axial Strain

Lateral Strain

-0.009

-0.006

-0.003

0

0.003

0.006

0.009

Lateral Strain

(c) Group GF/NSM

0.012

0.015

0.018

Axial Strain

(d) Group TRM/NSM

Fig. 19. Stress-strain curves for strengthened specimens.

% Loss due to Heating

90 80

Axial Strength

70

Secant Stiffness

60 50 40 30 20 10 0

100

200

300

400

500

800

Temperature (°C)

applied service load. As a result, for safety at fire or high temperature exposure, the ultimate capacity of the FRP-upgraded column should stay greater than its applied service load for the required duration of fire or high temperature [42]. Fig. 22 shows the variation of compressive strength enhancement ratio with exposure temperature. Compressive strength enhancement ratio is defined as the ratio of peak actual axial stress of confined concrete speci0 mens at an elevated temperature (f cc;T ) to the peak actual axial stress for unconfined concrete specimens at room temperature 0 (f co;R ). As seen in the figure as a result of insulation of both CF and CF/NSM groups, even exposing the specimen to a temperature of 800 °C for 3 h has kept the enhancement ratio above the service load level thereby demonstrating the effectiveness of insulation in preventing the column collapse at this level of temperature exposure.

Fig. 20. Effect of exposure temperature on percentage loss of axial strength and stiffness of unstrengthened specimens.

5.3. Effect of strengthening technique

axial stress capacity was 70% and 40% higher for insulated columns of groups CF and CFS/NSM, respectively. This indicates that the insulation was effective to some extent in preventing the damaging effects of the elevated temperatures from reaching the concrete core. When RC columns are strengthened using externally bonded FRP composites, their ultimate capacity increases, hence allowing for higher service loads to be applied. Under high temperature exposure, the ultimate capacity of the FRP-wrapped column gets reduced with increasing temperature and failure of the column would arise when its ultimate capacity becomes less than the

Performance comparison between different strengthening techniques used in this study at room temperature and after an exposure temperature of 300 °C is shown in Fig. 23(a) and (b). At room temperature, all techniques resulted in approximately same axial load carrying capacity owing to the design of the strengthening systems. However, in terms of strain energy per unit volume, which is defined as the area under the axial stress-strain curve, the strengthening technique incorporating textile reinforced mortar with near surface mounted steel bars (TRM/NSM) was found to be the most effective. This is also evident from Fig. 19(d) which shows the ductile behavior of the columns strengthened using this system. From Fig. 23(a), it is revealed that the strengthening

359

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

Uninsulated specimens

Insulated specimens

% Loss due to Heating

60 50 40

Axial Strength Secant Stiffness Peak Axial Strain

30 20 10 0 100

200

300

400

500

6

Enhancement Ratio due to Strengthening

70

5

Room Temperature Axial Strength Total Strain Energy per Unit Volume

4 3 2 1 0 C

800

CF

Temperature (°C)

CFS/NSM

GF/NSM

TRM/NSM

Group ID

(a) Group CF

(a) At room temperature

100 90

ca

60

Insulated specimens

Temperature = 300 °C

70 60 50

50

% Loss due to Heating

% Loss due to Heating

80 Axial Strength Secant Stiffness Peak Axial Strain

40 30 20 10

CF CFS/NSM

40

GF/NSM TRM/NSM

30 20 10

0

100

200

300

400

500

800

0 Axial Strength

Temperature (°C)

Secant Stiffness

Response Parameter

(b) Group CFS/NSM

(b) At 300 °C exposure

Fig. 21. Effect of exposure temperature on percentage loss of axial strength, stiffness and peak strain of CFRP-strengthened specimens.

techniques in order of decreasing ductility at ambient temperature are: TRM/NSM, GF/NSM, CFS/NSM and CF. Fig. 23(b) shows the performance of different strengthening systems in terms of axial strength, stiffness, peak axial strain and

Fig. 23. Performance comparison between column groups.

strain energy at an exposure temperature of 300 °C. Results are described in terms of percentage losses in parameters compared with the control specimen tested at room temperature for each group type. In terms of the loss in axial strength, the GF/NSM

1.4 Unstrengthened Uninsulated CF Insulated CF Uninsulated CFS/NSM Insulated CFS/NSM

1.2

f cc,T ’ /f co,R ’

1.0

0.8

0.6

0.4

Service load level for strengthened specimens Service load level for unstrengthened specimens

0.2

0.0 0

Peak Axial Strain Total Strain Energy per Unit Volume

200

400

600

800

1000

Temperature (°C) Fig. 22. Relationship between exposure temperature and compressive strength enhancement ratio.

360

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361

system was found to perform the best followed by the TRM/NSM system. Both the CFRP strengthening systems (CF & CFS/NSM) were found to perform similarly when exposed to elevated temperature of 300 °C. In terms of secant stiffness of the columns after exposure to an elevated temperature of 300 °C, the CFS/NSM system was found to perform the worst with a reduction in stiffness of almost 48%. Once again the GF/NSM system performed the best with only a 13% loss in stiffness as a result of exposure to elevated temperature. This surprising finding could be attributed to the fact that the voids within the strengthening fabric have an important role in the transmittal of the heat to the epoxy matrix. As seen earlier in Figs. 5 (a) and 6(d) which show, respectively, a close-up view of both the carbon and the E-glass fibers, the voids in the E-glass fibers are almost invisible and are covered-up by the custom fabric weave, whereas the voids in carbon fibers are visibly bigger. These voids may bring the epoxy matrix to be directly subjected to the adverse effects of the elevated temperatures resulting in deterioration of its strength. Nevertheless, in case of GFRP sheet, the lack of voids in between the glass fibers may safeguard the epoxy matrix from the hostile effects of the high temperatures. Same could be said about the TRM/NSM system in which the concrete as well as the textile fibers are well protected by a full layer of mortar. As seen in Fig. 22 and from Table 3, it was also noticed that the performance of group CFS/NSM subjected to a temperature of 400 °C was even worse than the unstrengthened group C exposed to the same temperature. This could be attributed to two reasons. Firstly, since the strengthening system incorporated separate CFRP strips and not a full wrap, some of the concrete and NSM epoxy surface was directly exposed to the heat thereby causing maximum degradation of materials, as pointed out earlier in Section 3.4, where at 400 °C, bond failure was noticed between epoxy mortar and concrete at some locations. Secondly, as a result of the grooves within the column cross-section for placement of NSM bars and with the epoxy being rendered ineffective as a result of exposure to temperature, the actual cross-sectional area of the column was now reduced, which resulted in a reduced residual axial load capacity.

5. The performance of RC columns strengthened using four systems and subjected to elevated temperatures was studied in terms of residual axial strength, stiffness and strain energy. It is revealed that the strengthening techniques in order of decreasing efficiency (level of performance) are: GF/NSM, TRM/NSM, CF and CFS/NSM.

Acknowledgements The project was supported by Deanship of Scientific Research Chairs at King Saud University, Saudi Arabia for MMB Chair of Research and Studies in Strengthening and Rehabilitation of Structures at Civil Engineering Department. Appendix A Calculation of nominal axial capacity of columns at room temperature For unstrengthened column C-R, the ACI 318-14 code [34] was followed in the calculation of its nominal axial capacity using the following equation. 0

Pn;u ¼ 0:85f c ðAg  Ast Þ þ Ast f yn

where = specified compressive strength of concrete; Ag = gross area of column section; Ast = total area of longitudinal steel bars; and fyn = nominal yield strength of longitudinal steel as provided by manufacturer (given in Table 2). Yet, for strengthened columns CF-R, CFS/NSM-R, GF/NSM-R & TRM/NSM-R, the nominal axial capacity was calculated as per the ACI 440.2R-08 guidelines [35] using the nominal yield strength of longitudinal steel, as given by the following equation. 0

Pn;s ¼ 0:85f cc ðAg  Ast Þ þ Ast f yn where 0 f cc

6. Conclusions The major conclusions derived from this study can be summarized as follows: 1. For both unstrengthened and strengthened RC columns, it was found that the reduction in their residual stiffness due to elevated temperature exposure was greater than the reduction in their residual axial load capacity. Therefore, more deliberation should be given to deformation and stress redistribution of post-fire RC columns. 2. The cement-based, fire protection mortar used in insulating test specimens was found to be effective in averting heat induced damage to CFRP-strengthened columns up to temperatures of 800 °C for an exposure period of 3 h. This is apparent from the fact that the compressive strength enhancement ratio did not reduce below the service load level for insulated specimens of both CF and CFS/NSM groups. 3. Results from this study indicate that it is better in terms of residual performance of columns subjected to elevated temperatures, for the strengthening techniques to be designed such that the entire column surface is covered, instead of having a system wherein a portion of material (concrete or other) is directly exposed to the heat. 4. A comparison of all strengthening techniques for RC columns used in this study revealed that at room temperature, the TRM/NSM system was most efficient in terms of total strain energy per unit volume, owing to its ductile behavior.

ðA1Þ

0 fc

¼

0 f cc

0 fc

ðA2Þ

= compressive strength of FRP-confined concrete, given by

þ 3:3wf ka f l

ðMPa unitsÞ

ðA3Þ

where = FRP strength reduction factor = 0.95; ka = efficiency factor accounting for section geometry = 1.0 for circular columns and fl = maximum confinement pressure given by

fl ¼

2Ef nt f efe D

ðA4Þ

where Ef = tensile modulus of elasticity of FRP material; n = No. of plies of FRP reinforcement; tf = thickness of one ply of FRP reinforcement; efe = effective strain level in the FRP at failure = 0.55  strain at rupture of the FRP reinforcement; and D = column diameter. Since we are not dealing with a design-related problem and the goal is to predict the axial capacity of test columns, a strength reduction factor of one (/ = 1) was utilized in Eqs. (A.1) & (A.2). It should be also noted that in case of TRM-strengthened column TRM/NSM-R, mortar contribution was ignored in axial capacity calculations and the strength reduction factor was assumed as 0.5. The above-mentioned detailed procedure was followed in the design of strengthened specimens such that they have nearly the same nominal axial capacity at room temperature. Table A1 enlists the nominal axial load capacity for unstrengthened as well as strengthened RC columns at room temperature. The experimental capacity (Pu-exp) of the same columns has also been presented in the table along with the ratio between the theoretical and experimental capacities. As seen from Table A1, the maximum difference between the axial capacities for all strengthened columns is 5.5% for both theoretical and experimental capacities. It is also noted that the above-mentioned procedure underestimated the axial capacity of columns at room temperature by about 9–15%.

361

Y.A. Al-Salloum et al. / Construction and Building Materials 115 (2016) 345–361 Table A1 Axial load capacity comparison for different columns at room temperature. Parameter

Unstrengthened specimen (C-R)

CF-R

CFS/NSM-R

GF/NSM-R

TRM/NSM-R

Min

Max

Max diff (%)

Ag (mm2) Ast (mm2)

45,996 314 (4Ø10) 0 – 0 42.0

45,996 314 (4Ø10) 1 0.95 3.9 54.1

45,996 766 (4Ø10 + 4Ø12) 0.5a 0.95 1.9 48.1

45,996 1118 (4Ø10 + 4Ø16) 1 0.95 2.2 49.0

45,996 1118 (4Ø10 + 4Ø16) 3 0.5b 4.1 48.8

– – – – – 1.9 48.1

– – – – – 4.1 54.1

– – – – – – –

1794 1982 0.91

2265 2669 0.85

2246 2561 0.88

2369 2658 0.89

2361 2701 0.87

2246 2561 0.85

2369 2701 0.89

5.5 5.5 –

n

wf f l (MPa) 0 f cc (MPa) P nth (kN) P uexp (kN) P nth =P uexp

Strengthened specimens

Italicized values are the calculated nominal axial capacity of strengthened columns at room temperature. a Equivalent number of layers based on assumed continuous FRP sheet. b Assumed.

References [1] B.J. Bett, R.E. Klingner, J.O. Jisra, Lateral load response of strengthened and repaired reinforced concrete columns, ACI Struct. J. 85 (5) (1988) 499–508. [2] M. Rodriguey, R. Park, Seismic load tests on reinforced concrete columns strengthened by jacketing, ACI Struct. J. 91 (2) (1994) 150–159. [3] M.F. Belal, H.M. Mohamed, S.A. Morad, Behavior of reinforced concrete columns strengthened by steel jacket, HBRC J. 11 (2) (2015) 201–212, http:// dx.doi.org/10.1016/j.hbrcj.2014.05.002. [4] M.J.N. Priestley, F. Seible, Y. Xiao, R. Verma, Steel jacket retrofitting of reinforced concrete bridge columns for enhanced shear strength. Part 1. Theoretical consideration and test design, ACI Struct. J. 91 (4) (1994) 394–405. [5] M.J.N. Priestley, F. Seible, Y. Xiao, R. Verma, Steel jacket retrofitting of reinforced concrete bridge columns for enhanced shear strength. Part 2. Test results and comparison with theory, ACI Struct. J. 91 (5) (1994) 537–551. [6] N. Moshiri, A. Hosseini, D. Mostofinejad, Strengthening of RC columns by longitudinal CFRP sheets: effect of strengthening technique, Constr. Build. Mater. 79 (2015) 318–325. [7] H.O. Al-Karaghool, Strength and ductility of axially loaded RC short columns confined with CFRP and GFRP wraps (MS thesis), The American University of Sharjah, United Arab Emirates, 2013. [8] N. Chikh, M. Gahmous, R. Benzaid, Structural performance of high strength concrete columns confined with CFRP sheets, Proc. of the World Congress on Eng. (WCE) vol. III, July 4–6, London, UK, 2012. [9] M. Quiertant, J.-L. Clement, Behavior of RC columns strengthened with different CFRP systems under eccentric loading, Constr. Build. Mater. 25 (2011) 452–460. [10] K. Gajdošová, J. Bilcˇík, Slender reinforced concrete columns strengthened with fibre reinforced polymers, Slovak J. Civ. Eng. XIX (2) (2011) 27–31. [11] K. Olivová, J. Bilcˇík, Strengthening of concrete columns with CFRP, Slovak J. Civ. Eng. 1 (2009) 1–9. [12] R. Benzaid, N. Chikh, H. Mesbah, Study of the compressive behavior of short concrete columns confined by fiber reinforced composite, Arab. J. Sci. Eng. 34 (1B) (2009) 15–26. [13] Y.A. Al-Salloum, Influence of edge sharpness on the strength of square concrete columns confined with composite laminates, Compos. Part B Eng. 38 (2007) 640–650. [14] H.M. Elsanadedy, T.H. Almusallam, H. Abbas, Y.A. Al-Salloum, S.H. Alsayed, Effect of blast loading on CFRP-Retrofitted RC columns – a numerical study, Lat. Am. J. Solids Struct. 8 (2011) 55–81. [15] H.M. Elsanadedy, Y.A. Al-Salloum, H. Abbas, S.H. Alsayed, Prediction of strength parameters of FRP confined concrete, Compos. Part B Eng. 43 (2) (2012) 228–239. [16] H.M. Elsanadedy, Y.A. Al-Salloum, S.H. Alsayed, R.A. Iqbal, Experimental and numerical investigation of size effects in FRP wrapped concrete columns, Constr. Build. Mater. 29 (2012) 56–72. [17] N.A. Siddiqui, S.H. Alsayed, Y.A. Al-Salloum, R.A. Iqbal, H. Abbas, Experimental investigation of slender circular RC columns strengthened with FRP composites, Constr. Build. Mater. 69 (2014) 323–334. [18] D. Bournas, T. Triantafillou, Strengthening of reinforced concrete columns with near-surface-mounted FRP or stainless steel, ACI Struct. J. 106 (4) (2009) 495–505. [19] J.A.O. Barros, R.K. Varma, J.M. Sena-Cruz, A.F.M. Azevedo, Near surface mounted CFRP strips for the flexural strengthening of RC columns: experimental and numerical research, Eng. Struct. 30 (2008) 3412–3425. [20] F. Danesh, B.B. Noveiri, Flexural behavior of concrete columns strengthened with near surface mounted FRP bars, CICE 2010 – The 5th Int Conf on FRP Composites in Civil Eng, September 27–29, 2010, Beijing, China, 2010, pp. 829–832. [21] M. Sarafraz, F. Danesh, Experimental study on flexural strengthening of RC columns with near surface mounted FRP bars, J. Seismol. Earthquake Eng. 12 (1 & 2) (2010) 39–50.

[22] T. El-Maaddawy, A.S. El-Dieb, Near-surface-mounted composite system for repair and strengthening of reinforced concrete columns subjected to axial load and biaxial bending, J. Compos. Constr. 15 (4) (2011) 602–614. [23] D. Bournas, P. Lontou, C. Papanicolaou, T. Triantafillou, Textile-reinforced mortar versus fiber-reinforced polymer confinement in reinforced concrete columns, ACI Struct. J. 104 (6) (2007) 740–748. [24] R. Ortlepp, A. Lorenz, M. Curbach, Column strengthening with TRC: influences of the column geometry onto the confinement effect, Adv. Mater. Sci. Eng. (2009) 5, http://dx.doi.org/10.1155/2009/493097 493097. [25] F. Apicella, M. Imbrogno, Fire performance of CFRP-composites used for repairing and strengthening concrete, Proc. of the 5th ASCE materials engineering congress, Cincinnati, Ohio, 1999, pp. 260–266. [26] L.A. Bisby, Fire behaviour of FRP reinforced or confined concrete (Ph.D. thesis), Department of Civil Engineering, Queen’s University, Kingston, Canada, 2003. [27] M Saafi, P. Romine, Effect of fire on concrete cylinders confined with GFRP, in: B. Benmokrane, EI-Salakawy (Eds.), Durability of Fibre Reinforced Polymer (FRP) Composites for Construction, 2002. [28] D.B. Cleary, C.D. Cassino, R. Tortorice, Effect of elevated temperatures on a fiber composite to strengthen concrete columns, J. Reinf. Plast. Compos. 22 (10) (2003) 881–895. [29] H.Z. El-Karmoty, Thermal protection of reinforced concrete columns strengthened by GFRP laminates (experimental and theoretical study), HBRC J. 8 (2012) 115–122, http://dx.doi.org/10.1016/j.hbrcj.2012.09.007. [30] D. Cree, E. Chowdhury, M. Green, L. Bisby, N. Benichou, Performance in fire of FRP-strengthened and insulated reinforced concrete columns, Fire Saf. J. 54 (2012) 86–95. [31] ULC, CAN/ULC-S101-M04, Standard Methods of Fire Endurance Tests of Building Construction and Materials, Underwriters’ Laboratories of Canada, Scarborough, Ont., 2004. [32] ASTM, ASTM E119-01, Standard Methods of Fire Test of Building Construction and Materials, American Society for Testing and Materials, West Conshohocken, PA, USA, 2001. [33] Y.A. Al-Salloum, H.M. Elsanadedy, A.A. Abadel, Behavior of FRP-confined concrete after high temperature exposure, Constr. Build. Mater. 25 (2) (2011) 838–850. [34] American Concrete Institute (ACI), Building code requirements for structural concrete and commentary, ACI 318-14, American Concrete Institute, Detroit, MI, USA, 2014. [35] American Concrete Institute (ACI), Guide for the design and construction of externally bonded FRP systems for strengthening concrete structures, ACI 440.2R-08, American Concrete Institute, Detroit, MI, USA, 2008. [36] ASTM, Standard test method for compressive strength of cylindrical concrete specimens, ASTM C39/C39M, American Society for Testing and Materials, West Conshohocken, PA, USA, 2010, http://dx.doi.org/10.1520/C0039_C0039M-10. [37] ASTM, Standard test methods for tension testing of metallic materials, ASTM E8/E8M, American Society for Testing and Materials, West Conshohocken, PA, USA, 2009, http://dx.doi.org/10.1520/E0008_E0008M-09. [38] ASTM, ASTM E119-01. ASTM D3039/D3039M – 08, Standard Test Method for Tensile Properties of Polymer Matrix Composite Materials, American Society for Testing and Materials, West Conshohocken, PA, 2008. [39] T. Khalifa, The effects of elevated temperatures on fibre reinforced polymers for strengthening concrete structures (MS thesis), Queen’s University, Kingston, Ontario, Canada, 2011. [40] ASTM, Standard test method for compressive strength of hydraulic cement mortars (using 2-in. or [50-mm] cube specimens), ASTM C109/C109M, American Society for Testing and Materials, West Conshohocken, PA, USA, 2008, http://dx.doi.org/10.1520/C0109_C0109M-08. [41] ASTM, Method of test for tensile strength of hydraulic cement mortars, ASTM C190, American Society for Testing and Materials, West Conshohocken, PA, USA, 1985. [42] V. Kodur, L.A. Bisby, M.F. Green, FRP retrofitted concrete under fire conditions, J. Concr. Int. 28 (12) (2006) 37–44.

Suggest Documents