Thermal effects on bond between FRP rebars and concrete
ARTICLE IN PRESS
DTD 5
1
57
2
58
3 4
Composites: Part A xx (xxxx) 1–8 www.elsevier.com/locate/compositesa
59 60
5
61
6
62
7
63
8 9
Thermal effects on bond between FRP rebars and concrete
64
Nestore Galatia,*, Antonio Nannia, Lokeswarappa R. Dharanib, Francesco Focaccic, Maria Antonietta Aiellod
66
10 11 12 13
a
17
Department of Civil, Architectural and Environmental Engineering, 221 Engineering Research Lab, University of Missouri-Rolla, Rolla, MO 65409, USA b Mechanical and Aerospace Engineering, University of Missouri-Rolla, Rolla, MO 65409, USA c Dipartimento di Costruzione dell’Architettura, Istituto University di Architettura di Venezia, Venice 30135, Italy d Department of Innovation Engineering, University of Lecce, Lecce 73100, Italy
18
Received 27 April 2005; accepted 13 May 2005
14 15 16
26 27 28 29 30 31 32 33 34
F O
1. Introduction
38
45 46 47 48 49 50 51 52 53 54 55 56
R
R
43 44
O
42
C
41
N
40
The use of FRP bars as reinforcement for concrete elements seems to be an effective solution for overcoming durability problems of traditional steel reinforced concrete structures due to the corrosion of metallic bars. For this reason, the replacement of steel with FRP bars is gaining popularity worldwide. The numerous experimental and theoretical studies carried out in the last years and many structural applications in addition to efforts in many countries to establish the guidelines of practical use of FRP reinforced concrete structures, confirm the increasing interest in this field [1]. However, in order to gain a deeper insight into the structural behavior of such new materials
U
39
72 73 75 76 77 78
90
35 36 37
71
Keywords: Fiber reinforced polymer (FRP); Thermal cracking; Bond degradation; Pull-out test; Shearing of the epoxy
O
25
70
79
PR
24
69
Bond between FRP bars and concrete depends on several parameters comprising environmental agents as the service temperature. As is known, FRP bars present high values of the transverse coefficient of the thermal expansion (CTE) with respect to concrete; as a consequence, when temperature increases, tensile stresses take place within the concrete that may produce splitting cracks affecting the bond performance. The present paper is devoted to the analysis of the bond between FRP bars and concrete under thermal loads taking into account already available data on bond–slip relationships and thermal behavior. An experimental investigation was carried out on concrete specimens reinforced with a FRP bar and subjected to thermal cycles with a maximum temperature value of 70 8C. After the thermal treatment, pull-out tests were performed at room temperature or higher. Untreated specimens were also tested for comparison. Results are reported and discussed in order to investigate the degradation of the concrete-reinforcement interface under thermal treatment and, as a consequence, the effects on bond–slip laws. Experimental results showed a significant degradation induced by exposure to relatively high temperatures. q 2005 Published by Elsevier Ltd.
D
23
Abstract
EC TE
22
67 68
74
19 20 21
65
* Corresponding author. Tel.: C1 573 341 6223; fax: C1 573 341 6215. E-mail address: [email protected] (N. Galati).
1359-835X/$ - see front matter q 2005 Published by Elsevier Ltd. doi:10.1016/j.compositesa.2005.05.043
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
and to guarantee their safe application, further researches are needed in specific areas that concern the structural performance and durability under service conditions [2] and among all the effect of the temperature increases. Unidirectional FRP bars, generally utilized in concrete, present an anisotropic behavior. As a consequence, the coefficient of thermal expansion (CTE) is different in the longitudinal and transverse directions. In particular, the longitudinal CTE, controlled by fibers, is low and even negative (as in the case of aramidic fibers, high strength carbon fibers and ultra-high modulus carbon fibers), while the transverse CTE, controlled by the resin, is up to 3–6 times that of the concrete [3]. As a result, an increase in temperature produces bursting stresses within the concrete that may cause splitting cracks. This fact involves a degradation of the bond between concrete and reinforcement [4,5], affecting the structural response. This paper aims to determine what effect elevated service temperatures have on the bond performance of FRP reinforced concrete members. A total of 36 specimens
80 81 82 83 84 85 86 87 88 89 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
ARTICLE IN PRESS
DTD 5
2
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
113
169
114
Notation
115 116
C
117
s
118 119 120
sr s
170
parameter of the BEP and mBEP bond–slip relationship slip, relative displacement between bar and concrete parameter of the CMR bond–slip relationship parameter of the mBEP bond–slip relationship
a
parameter of the BEP and mBEP bond–slip relationship parameter of the CMR bond–slip relationship bond shear stress at the bar/concrete interface parameter of the CMR bond–slip relationship.
b t tm
128 129
F
133
2. Experimental program 2.1. Test matrix
134 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
A total of 36 specimens for direct pull-out testing were constructed (Table 1). A 9.5 mm (3/8 in.) GFRP bar with different embedment lengths was placed inside a 152 mm3 (6 in.3) cube of concrete. Three different types of specimens were created based on the position of the bar, namely: centrally placed, and two and three times the diameter from one of the outer faces (Fig. 1). In addition, in order to create different embedment lengths, bond breakers were used. Consisting of soft polyethylene tubing placed over the bar to prevent direct contact with the concrete (Fig. 1). In order to have statistically relevant test results, three specimens for each configuration were tested (Fig. 2). The heat treated specimens were placed into an environmental chamber for 200 h at a temperature of 70 8C and at a humidity of 80%, corresponding to 70% of the glass transition temperature of the GFRP bars used. The testing of the specimens was undertaken at room temperature.
153
2.2. Material properties
160 161 162 163 164 165 166 167 168
O
C
159
N
158
Tests were performed to characterize the mechanical properties of the materials used in this investigation. The concrete used for the preparation of the specimens was designed to have a compressive strength of 27.6 MPa (4000 psi). Water to cement ratio for the concrete mixture was 0.45. Components in the concrete mixture were proportioned as follow by weight: 19% Portland cement, 40% crushed limestone, 33% sand, and 8% water. Once the specimens were cast, they were allowed to cure for 28 days at room temperature. The specimens were built from four different batches of concrete. Table 2 summarizes the average compressive strength [6] and the average tensile splitting strength [7] for each batch of concrete.
U
157
R
154 155 156
R
152
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
2.3. Test setup
199
The testing was completed using the guidelines listed by ACI 440 [9] for direct pull-out specimens. As already mentioned before, the specimens were designed as concrete cubes with the bar either embedded in the center or placed two or three times the diameter from one of the outer cube faces. The bar then had a long and a short end. The short end was used to measure the free end slip while load was applied to the longer end.
EC TE
135
O
131 132
O
130
The GFRP bars were manufactured by Hughes Brothers, Seward, Nebraska. These bars have a surface deformation called: ‘wrapped and sand coated design’. The ‘wrapped’ refers to the spiral glass fibers that are twisted around the bar. This decreases the amount of transverse thermal expansion and increases the mechanical interlock with the concrete. Bars are also sand coated to increase the friction and interlocking bond. The longitudinal and transverse coefficients of thermal expansion, as referred by the manufacturer, are 0.68O0.74!10K6/8C and 29!10K6/8C, per transverse and longitudinal direction, respectively. Tensile tests were performed on FRP bars to determine their engineering properties, which are related to fiber content. The average tensile strength, ultimate strain and modulus of elasticity obtained from the testing of the specimens (ASTM D3039) are presented in Table 3. Details of coupon fabrication and testing procedure are shown elsewhere [8].
PR
127
175
178
were tested under direct pull-out. The variables investigated were: bonded length, concrete cover and exposure to high temperature. Experimental results showed a significant degradation induced by exposure to relatively high temperatures.
D
126
174
177
122
125
173
176
121 123 124
171 172
200 201 202 203 204 205 206 207 208 209
Table 1 Test matrix
210
Bonded length mm (in.)
Thermal treatment
Concrete cover mm (in.)
Bar placement
76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6)
No No No No No No Yes Yes Yes Yes Yes Yes
148 (5.8) 148 (5.8) 19 (0.8) 19 (0.8) 29 (1.1) 29 (1.1) 148 (5.8) 148 (5.8) 19 (0.8) 19 (0.8) 29 (1.1) 29 (1.1)
Center Center 2f from 2f from 3f from 3f from Center Center 2f from 2f from 3f from 3f from
211 212 213 214 215
edge edge edge edge
216 217 218 219 220
edge edge edge edge
221 222 223 224
ARTICLE IN PRESS
DTD 5
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
281
2φ
225
3
226 227 228
282 283 284
230 231 232
Positions of the GFRP bar 3φ
229
233
285
152.4 mm (6 in)
152.4 mm (6 in)
Bond Breakers
286 287 288 289
234
290
235 236
291 292
152.4 mm (6 in) 152.4 mm (6 in)
237
293
Fig. 1. Thermal cracking specimens.
294
239
295
240
296
241
297
242
298
243 244
299 300
F
238
246 248 249 251 252 253 255 257
269 270 271
R
R
267 268
O
266
For direct pull-out testing, the concrete block is held in place while a load is applied to the long end of the bar. The measurements taken were free and loaded end slip of the bar and the load. Two LVDTs were used to measure the free end slip 1808 apart from one another. One LVDT was used to measure the loaded end slip because of the lack of room. This setup can be seen from Fig. 4. As the load was applied, readings were taken every second, until (1) rupture of the FRP bar occurred, (2) the enclosing concrete split, or (3) sufficient slippage of the free end of the bar occurred [10,11].
C
265
272 274
Average compressive strength MPa (psi)
Average splitting tensile strength MPa (psi)
277
1 2 3 4
33.9 (4920) 28.9 (4200) 28.6 (4150) 32.8 (4750)
3.39 (492) 3.05 (443) 3.03 (439) 3.31 (480)
U
Batch of concrete
279
309 310 311 312 313 314
The anchorage system required significant pre-test preparation. It was essential that a proper design was provided for the anchorage system to prevent damage to the rebars. Clearly, unprotected gripping crushes the FRP bar, and thus causes failure outside the gage length. The anchor system used consisted of a 200 mm (8 in.) long steel pipe, 25 mm (1 in.) in diameter. The pipe was placed on the end of the bar to be loaded and filled with a low viscosity epoxy resin. The low viscosity epoxy allowed for the epoxy to fill the pipe without leaving air pockets. The inside of the pipes were threaded to make sure the load was
280
315 316 317 318 319 320 321 322 323 324 325 326 327
Bar type
275 276 278
307 308
Table 3 Mechanical properties of FRP bars
Table 2 Mechanical properties of the concrete
N
273
306
Fig. 2. Test setup for pull-out test with central bar.
258
264
305
EC TE
256
263
304
D
254
262
303
PR
250
261
302
O
247
259 260
301
O
245
#3 GFRP bar
328 329
Dimensions of the bar
Average maximum strain (%)
Average maximum stress MPa (ksi)
Average elastic modulus MPa (ksi)
330
Nominal diameter 9.53 mm (0.375 in.)
1.85
760.0 (110.0)
40,800 (5920)
334
331 332 333 335 336
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
ARTICLE IN PRESS
DTD 5
4
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
transferred from the pipe to the resin. From Fig. 3 one can see that polymer caps were used to keep the pipe aligned with the bar. During testing, the load was applied to the pipe. For the specimens that had the off-center bar placement, a variation in the setup had to be introduced to prevent any bending on the bar. With reference to the sketch in Fig. 4, one can see that another six in block was tied to the one being tested. This kept the bar and the block straight, eliminating the bending moment that might have otherwise been created.
337 338 339 340 341 342 343
Fig. 3. Anchoring system.
344 345 346 347 348
(a) Straps Fastening Specimen and Concrete Block
349 350
152 mm3 (6 in3) Concrete Cube
352
A
A
353 354
357
O
359 360
Side View
361 363 364
(b) Specimen
152 mm (6 in)
PR
362
Concrete Block
152 mm (6 in)
365
369 370 371 372 373 374
Fixed End of the Testing Machine
Section A – A
376 377
Fig. 4. Test setup for the off-center specimens.
R
378
O
379 380 381
385 386 387 388 389 390 391
C
Bar placement
Bonded length mm (in.)
N
384
Center Center 2f from edge 2f from edge 3f from edge 3f from edge
76.2 (3) 152.4 (6) 76.2 (3) 152.4 (6) 76.2 (3) 152.4 (6)
397 398 399 400 401 402 403 404
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437
Table 4 Test results in terms of ultimate loads
U
383
R
375
382
D
Straps
EC TE
368
152 mm (6 in)
366 367
The specimens with 152.4 mm (6 in.) embedment lengths, 15 times the diameter of the bar, were too long for a 9.53 mm (0.375 in.) diameter bar. In some of the specimens, the bar broke before they would begin to slip. However, from the tensile tests on the bars, the breaking force was 54 kN (12 kip) while the maximum pullout force for the treated specimens was 42 kN (9 kip). Such inconsistency was explained as due to accidental imperfections such as, the eccentricity of the load generated by the non perfect alignment of the testing machine for the centered specimens or, the eccentricity generated at high load levels for eventual sliding between the two block. Test results in terms of average ultimate load are summarized in Table 4. By increasing the embedment length the maximum load that the specimens can hold also increases. From Table 4 it can be observed a 16% maximum reduction in the ultimate load for the thermally treated specimens consequent to bond degradation. The effects of thermal loads are more evident by looking at the load–slip curves. Typical load–slip curves for the pull-out tests performed are shown in Fig. 5. These curves show the difference between the behavior of the specimens with and without thermal treatment. For the untreated specimens there was almost no slip at the freeend until the load reached its ultimate value. For the thermal treated specimens instead, the same peak bond stress is attained but with higher slips values. The bond failure in this
O
Fixed End of the Testing Machine
Specimen
358
395 396
406
F
355 356
394
405
2.4. Test results and discussion
351
393
438 439 Ultimate load kN (kip)
Covariance (%)
Thermal treatment (PT)
No thermal treatment (PNT)
Thermal treatment
34.98 (7.86) 40.99 (9.21) 34.55 (7.77) 28.77 (6.47) 39.34 (8.84) 30.69 (6.90)
34.69 (7.80) 42.26 (9.50) 33.34 (7.50) 34.25 (7.70) 37.82 (8.50) 34.70 (7.80)
8.8 6.2 23.3 14.6 7.2 18.4
392
No thermal treatment 5.83 5.8 14.1 13.8 7.8 14.50
Strength loss (PNTKPT)/PNT! 100 (%) K0.84 3.01 K3.63 16.00 K4.02 11.56
440 441 442 443 444 445 446 447 448
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
ARTICLE IN PRESS
DTD 5
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
45 40 35 30 25 20 15 10 5 0
451 452 453 454 455 456
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
0
457
505
(b) 50 Load (kN)
(a) 50
450 Load (kN)
449
0.25
0.5
0.75
45 40 35 30 25 20 15 10 5 0
1
Bar Breaking
0
507 508
0.25
0.5
509 510 511
0.75
1
Displacement (mm)
Specimens having a 76.2 mm (3 in) Embedment Length, GFRP Bar Placed at the Center of the Specimen.
459 460
506 Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
Displacement (mm)
458
5
515 516 517
40 35 30 25 20 15 10 5 0
45 40 35 30 25 20 15 10 5 0
467 468 469
0
470
0.25
486
0.75
Displacement (mm)
483 484 485
0.5
495 496 497 498 499 500 501 502 503 504
531 532 Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
0.50
0.75
534 535 536 537 538 539 540
Specimens having a 152.4 mm (6 in) Embedment Length, GFRP bar placed 2 φ from edge.
541 542 543
R
544 545 546
R
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
1.00
533
Displacement (mm)
547 548
O
494
case is caused by the shearing of the matrix material from the fibers in the bar. The GFRP bar consists of a core of glass fibers with an outer layer of resin and sand. As a consequence of the thermal treatment there is a weakening of the matrix and thus it shears off as observed in Fig. 6. In Fig. 7, the pullout curves of treated specimens having different concrete cover are compared for the two analyzed embedment length. It can be observed that the weakening due to the thermal treatment is more evident when the concrete cover is small, especially for the six in embedment length specimens. Such behavior can be explained with the more extended micro cracking of the concrete when the bars are placed at a lower cover. Such micro-cracks are induced by the bursting stresses within the concrete due to the different CTE of GFRP bars and concrete; obviously they are more pronounced when lower covers are used.
529 530
549 550
C
493
0.25
526 528
Failure of the Bar Splitting of the Cover
551 552
N
491 492
45 40 35 30 25 20 15 10 5 0 0.00
553 554
U
490
1
527
Fig. 5. Comparison of free-end slips.
488 489
1
Specimens having a 76.2 mm (3 in) Embedment Length, GFRP bar placed 2 φ from edge.
487
525
0.75
Specimens having a 152.4 mm (6 in) Embedment Length, GFRP bar placed 3 φ from edge.
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
0.25
523 524
Displacement (mm)
(f) 50
0
0.5
522
O
45 40 35 30 25 20 15 10 5 0
482
0.25
D
475 476
481
0
Load (kN)
(e) 50
Load (kN)
474
480
1.25
Specimens having a 76.2 mm (3 in) Embedment Length, GFRP bar placed 3 φ from edge.
473
479
1
521
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
PR
472
478
0.75
520
Displacement (mm)
471
477
0.5
519
Concrete Splitting
EC TE
466
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
518 FRP Rupture
F
463
O
(d) 50 Load (kN)
(c) 45 Load (kN)
462
465
513 514
Specimens having a 152.4 mm (6 in) Embedment Length, GFRP Bar Placed at the Center of the Specimen.
461
464
512
555 556 557 558 559 Fig. 6. The shearing of the resin.
560
ARTICLE IN PRESS
DTD 5
6
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
561
Pullout force (kN)
562
617
Pullout force (kN)
45
45
618
563 564
40
40
35
35
619 620
565
30
30
621
566
25
25
567
20
568
Em. length: 3 in (76 mm) Center 3d cover 2d cover
15
569 570
10
571 572
5 0
573
15 10 5
0.2
574
0.4
0.6
0.8
1
1.2
0
623
Em. length: 6 in (152 mm) Center 3d cover 2d cover
20
Free end slip (mm)
0
622 624 625 626 627 628
Free end slip (mm)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
629 630
Fig. 7. Experimental free end slip-pullout force results for treated specimens.
575
631
576
632
Interfacial shear stress (N/mm2)
20
20 18
12
583
10
584
8
585
6
586
4
587 588
2 0
Em. length: 3 in (76 mm) Center 3d cover 2d cover mBEP CMR
0
0.5
1
1.5
14
Av. mBEP
12
Em. length: 6 in (152 mm) Center 3d cover 2d cover mBEP CMR
10 8 6 4 2
Slip (mm)
0 2
0
0.5
589
600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616
638 639
1
640 641 642 643 644
Slip (mm) 1.5
2
645
D
646
The dependence of the local t(s) by the cover is a consistent result, while the strong dependence on the embedment length is in contrast with the assumption of the existence of a local t(s) law.
EC TE
599
R
598
In Eqs. (1) and (2), t(s) is the interfacial shear stress corresponding to the bar/concrete slip, s. The unknown parameters a, C, s and b, tm, sr were found by minimizing the distance between the measured and calculated applied load N—free displacement, sf. The procedure takes into account the deformation of the bar within the embedment length. For all the sets of equal specimens the t(s) law was determined. These laws are plotted in Fig. 8. In this figure, the average t(s) curve, obtained by considering the average of the local t(s) curve fits to all the treated specimen results, is also plotted. It can be observed, in general, that better behaviors of the specimens result with shorter embedment lengths and large concrete cover.
R
597
O
595 596
The local bond–slip laws of the considered bars after thermal treatment have been determined according to [12], viz the modified Bertero Eligehhausen Popov (mBEP) Eq. (1) s tðsÞ Z Csa 1 K (1) s and the Cosenza, Manfredi Realfonzo [13] (CMR) Eq. (2): s b tðsÞ Z tm 1 K eKsr (2)
C
594
N
593
2.5. Analytical approach
U
592
637
Fig. 8. Local bond–slip relationships.
590 591
635 636
F
Av. mBEP
14
582
Av. CMR
16
16
581
634
18
Av. CMR
O
579 580
633
O
578
Interfacial shear stress (N/mm2)
PR
577
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
647 648 649 650 651 652 653
Interfacial shear stress (N/mm2)
654
20
655
18
656
Av. CMR
657
16
658
Av. mBEP
14
659 660
12
661
10
662
8
663 664
6
665
4
666
2
667 668
Slip (mm)
0 0
0.5
1
1.5
2
Fig. 9. Average local bond–slip relationships obtained considering all the treated specimens.
669 670 671 672
ARTICLE IN PRESS
DTD 5
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
731 732
Nestore Galati is a research engineer at the University of Missouri, Rolla. He obtained is PhD in Civil Engineering at the University of Lecce, Italy where he also received his BSc in Materials Engineering. He obtained his MSc degree in Engineering Mechanics at the University of Missouri-Rolla. His research has concentrated on the area of retrofitting of masonry and upgrade and in situ load testing of reinforced concrete structures (building and bridges). He is an EIT in the United States.
765
730
a
C
s
b
tm
sr
0.28
14.40
/CN
0.35
66.10
58.40
679 680 681 682 683 684 685 686 687
The final t(s) curves, an average of all treated specimens, computed from the mBEP and CMR equations are plotted in Fig. 9; the corresponding parameters are reported in Table 5. For the average mBEP law, it is found s /CN, which indicates that the same local bond–slip curve could also be obtained utilizing the Bertero Eligehhausen Popov bond–slip equation [14]. The result s /CN is an effect of averaging the results; for single specimens, s Z 1O 3 mm was found.
688 689 690 691 692 693
3. Conclusion On the basis of the experimental investigations, the following concluding remarks can be drawn:
694
701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
EC TE
699 700
R
698
R
697
O
696
(1) In most of the specimens the thermal treatment induced a slight degradation in the bond performance in terms of ultimate load. A more pronounced effect was observed for the bond stress–slip curves in terms of slip values due primarily to the degradation of the resin; (2) A number of specimens with an embedment length of 15 times the diameter of the rebar failed in the free length of the rebar before slip commenced; (3) Future research will be directed toward the direct pullout test at high temperature. This will enable any bond degradation to be determined when the structure is subject to high temperature, and thus verifying whether the current standards are adequate for the designing of internal FRP reinforcement; (4) The influence of the thermal treatment is more evident with the small values of the concrete cover. Such behavior was explained with the microcrackingoftheconcreteduetothe bursting stresses induced during the thermal treatment; (5) The local bond–slip laws utilized in the current investigation present a large scatter in the results, thus the average values of these laws can only be considered as representative of an order of magnitude. Due to the scattering of the results it was not possible to determine a correlation between the CMR/mod-BEP parameters o treated and un-treated specimens.
C
695
722
N
720 721
Acknowledgements
725
The support of the National Science Foundation Industry/ University Cooperative Research Center at the University of Missouri-Rolla. The authors would also like to acknowledge the support of the Rolla Technical Institute (RTI).
727 728
U
723 724 726
F
678
[1] Nanni A. Relevant applications of FRP composites in concrete structures. In: Figueiras J, Juvandes L, Furia R, editors. Proceedings, CCC 2001, Composites in Construction, Porto, Portugal, Oct. 10–12, 2001, 2001. p. 661–70. [2] Harries KA, Porter ML, Busel J. FRP materials and concrete— research needs. Concr Int 2003;25(10). [3] Gentry TR. Thermal compatibility of plastic composite reinforcement and concrete. In: El-Badry MM, editor. Advanced composite materials in bridges and structures. Montreal, Que.: Canadian Society for Civil Engineering; 1996. p. 149–56. [4] Tepfers R. Division of building technology. Go¨teborg: Chalmers University of Technology; 1998 [Publication no. 98:3, Work no. 22. p. 1–16]. [5] Tepfers R. Cracking of concrete cover along anchored deformed reinforcing bars. Mag Concr Res 1979;31(106):3–12. [6] ASTM C 39/C 39M. Standard test method for compressive strength of cylindrical concrete specimens. [7] ASTM C496-96. Standard test method for splitting tensile strength of cylindrical concrete specimens. [8] Secondin S. Masonry reinforced with FRP systems. Tesi di Laurea, Universita` degli Studi di Padova. Padova, Italy: Facolta` di Ingegneria; 2003. [9] ACI Committee 440. Recommended test methods for FRP rods and sheets. ACI 440-K.09-011. Farmington Hills, MI: American Concrete Institute International; 2001. [10] Boothby TE, Nanni A, Bakis CE. Accelerated test methods for frp/concrete systems in highway structures. In: Buckle IG, Friendland IM, editors. Fourth national workshop on bridge research in progress, 1996. p. 17–19 [Buffalo, NY]. [11] Conrad JO, Bakis CE, Bootby TE, Nanni A. Durability of bond of various FRP rods in concrete. Sherbrooke, Canada: CDCC-98; 1995 p. 299–310. [12] Focacci F, Nanni A, Bakis CE. Local bond–slip relationship for FRP reinforcement in concrete. J Compos Construct 2000;4(1):24–3. [13] Cosenza E, Manfredi G, Realfonzo R. Behavior and modeling of bond of FRP rebars to concrete. J Compos Construct 1997;1(2):40–5. [14] Eligehausen R, Popov EP, Bertero VV. Local bond stress–slip relationships of deformed bars under generalized excitations. Report No 83/23, EERC. Berkeley: University of California; 1983.
CMR
O
677
mBEP
729
O
675 676
References
PR
674
Table 5 Parameters of local bond–slip laws for treated bars (N, mm)
D
673
7
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
Antonio Nanni FACI, is the V and M Jones Professor of Civil Engineering at the University of Missouri, Rolla, Rolla, MO. He was the founding Chair of ACI Committee 440, Fiber Reinforced Polymer Reinforcement, and is the Chair of ACI Committee 437, Strength Evaluation of Existing Concrete Structures.
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 766 767 768 769 770 771 772 773 774 775 776 777
Lokeswarappa R. Dharani is Professor of Engineering Mechanics and Aerospace Engineering at the University of Missouri, Rolla. His research has concentrated in the fields aircraft structures, fracture mechanics, fatigue and failure analysis, micromechanics, composite materials and structures, process modeling of ceramic matrix composites, friction and wear of composites, fracture of laminated glazing under impact loads. He is a fellow of the American Society of Mechanical Engineers (ASME) and of the American Institute of Aeronautics and Astronautics (AIAA).
778 779 780 781 782 783 784
ARTICLE IN PRESS
DTD 5
8 785 786 787 788 789
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
Francesco Focacci is Associate Professor at the Dipartimento di Costruzione dell’Architettura (DCA), University of Venice, Italy. His research deals with the structural behavior of FRP reinforced concrete members, bond between FRP bars and concrete and structural rehabilitation using FRP.
790
Maria A. Aiello fib member, IABSE member, is Associate Professor at the Department of Innovation Engineering, University of Lecce, Italy: Her research interests include the structural analysis of concrete beams reinforced with FRP rebars, the structural behavior of RC beams strengthened with FRP laminates, the bond between concrete or masonry elements and FRP reinforcement, the stability analysis of composite laminates and sandwich panels.
841 842 843 844 845 846 847
792
848
793
849
794
850
795 796
851 852
797
853
798
854
799
855
800
856
801
857
802
858
803 804
859 860
F
791
O
805 806
O
807 808 809
PR
810 811 812 813
D
814 815
EC TE
816 817 818 819 820 821 822 823
R
824 825
R
826
O
827 828 829
C
830 831
834 835 836
U
833
N
832
861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892
837
893
838
894
839
895
840
896
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
DTD 5
1
57
2
58
3 4
Composites: Part A xx (xxxx) 1–8 www.elsevier.com/locate/compositesa
59 60
5
61
6
62
7
63
8 9
Thermal effects on bond between FRP rebars and concrete
64
Nestore Galatia,*, Antonio Nannia, Lokeswarappa R. Dharanib, Francesco Focaccic, Maria Antonietta Aiellod
66
10 11 12 13
a
17
Department of Civil, Architectural and Environmental Engineering, 221 Engineering Research Lab, University of Missouri-Rolla, Rolla, MO 65409, USA b Mechanical and Aerospace Engineering, University of Missouri-Rolla, Rolla, MO 65409, USA c Dipartimento di Costruzione dell’Architettura, Istituto University di Architettura di Venezia, Venice 30135, Italy d Department of Innovation Engineering, University of Lecce, Lecce 73100, Italy
18
Received 27 April 2005; accepted 13 May 2005
14 15 16
26 27 28 29 30 31 32 33 34
F O
1. Introduction
38
45 46 47 48 49 50 51 52 53 54 55 56
R
R
43 44
O
42
C
41
N
40
The use of FRP bars as reinforcement for concrete elements seems to be an effective solution for overcoming durability problems of traditional steel reinforced concrete structures due to the corrosion of metallic bars. For this reason, the replacement of steel with FRP bars is gaining popularity worldwide. The numerous experimental and theoretical studies carried out in the last years and many structural applications in addition to efforts in many countries to establish the guidelines of practical use of FRP reinforced concrete structures, confirm the increasing interest in this field [1]. However, in order to gain a deeper insight into the structural behavior of such new materials
U
39
72 73 75 76 77 78
90
35 36 37
71
Keywords: Fiber reinforced polymer (FRP); Thermal cracking; Bond degradation; Pull-out test; Shearing of the epoxy
O
25
70
79
PR
24
69
Bond between FRP bars and concrete depends on several parameters comprising environmental agents as the service temperature. As is known, FRP bars present high values of the transverse coefficient of the thermal expansion (CTE) with respect to concrete; as a consequence, when temperature increases, tensile stresses take place within the concrete that may produce splitting cracks affecting the bond performance. The present paper is devoted to the analysis of the bond between FRP bars and concrete under thermal loads taking into account already available data on bond–slip relationships and thermal behavior. An experimental investigation was carried out on concrete specimens reinforced with a FRP bar and subjected to thermal cycles with a maximum temperature value of 70 8C. After the thermal treatment, pull-out tests were performed at room temperature or higher. Untreated specimens were also tested for comparison. Results are reported and discussed in order to investigate the degradation of the concrete-reinforcement interface under thermal treatment and, as a consequence, the effects on bond–slip laws. Experimental results showed a significant degradation induced by exposure to relatively high temperatures. q 2005 Published by Elsevier Ltd.
D
23
Abstract
EC TE
22
67 68
74
19 20 21
65
* Corresponding author. Tel.: C1 573 341 6223; fax: C1 573 341 6215. E-mail address: [email protected] (N. Galati).
1359-835X/$ - see front matter q 2005 Published by Elsevier Ltd. doi:10.1016/j.compositesa.2005.05.043
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
and to guarantee their safe application, further researches are needed in specific areas that concern the structural performance and durability under service conditions [2] and among all the effect of the temperature increases. Unidirectional FRP bars, generally utilized in concrete, present an anisotropic behavior. As a consequence, the coefficient of thermal expansion (CTE) is different in the longitudinal and transverse directions. In particular, the longitudinal CTE, controlled by fibers, is low and even negative (as in the case of aramidic fibers, high strength carbon fibers and ultra-high modulus carbon fibers), while the transverse CTE, controlled by the resin, is up to 3–6 times that of the concrete [3]. As a result, an increase in temperature produces bursting stresses within the concrete that may cause splitting cracks. This fact involves a degradation of the bond between concrete and reinforcement [4,5], affecting the structural response. This paper aims to determine what effect elevated service temperatures have on the bond performance of FRP reinforced concrete members. A total of 36 specimens
80 81 82 83 84 85 86 87 88 89 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
ARTICLE IN PRESS
DTD 5
2
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
113
169
114
Notation
115 116
C
117
s
118 119 120
sr s
170
parameter of the BEP and mBEP bond–slip relationship slip, relative displacement between bar and concrete parameter of the CMR bond–slip relationship parameter of the mBEP bond–slip relationship
a
parameter of the BEP and mBEP bond–slip relationship parameter of the CMR bond–slip relationship bond shear stress at the bar/concrete interface parameter of the CMR bond–slip relationship.
b t tm
128 129
F
133
2. Experimental program 2.1. Test matrix
134 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
A total of 36 specimens for direct pull-out testing were constructed (Table 1). A 9.5 mm (3/8 in.) GFRP bar with different embedment lengths was placed inside a 152 mm3 (6 in.3) cube of concrete. Three different types of specimens were created based on the position of the bar, namely: centrally placed, and two and three times the diameter from one of the outer faces (Fig. 1). In addition, in order to create different embedment lengths, bond breakers were used. Consisting of soft polyethylene tubing placed over the bar to prevent direct contact with the concrete (Fig. 1). In order to have statistically relevant test results, three specimens for each configuration were tested (Fig. 2). The heat treated specimens were placed into an environmental chamber for 200 h at a temperature of 70 8C and at a humidity of 80%, corresponding to 70% of the glass transition temperature of the GFRP bars used. The testing of the specimens was undertaken at room temperature.
153
2.2. Material properties
160 161 162 163 164 165 166 167 168
O
C
159
N
158
Tests were performed to characterize the mechanical properties of the materials used in this investigation. The concrete used for the preparation of the specimens was designed to have a compressive strength of 27.6 MPa (4000 psi). Water to cement ratio for the concrete mixture was 0.45. Components in the concrete mixture were proportioned as follow by weight: 19% Portland cement, 40% crushed limestone, 33% sand, and 8% water. Once the specimens were cast, they were allowed to cure for 28 days at room temperature. The specimens were built from four different batches of concrete. Table 2 summarizes the average compressive strength [6] and the average tensile splitting strength [7] for each batch of concrete.
U
157
R
154 155 156
R
152
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
2.3. Test setup
199
The testing was completed using the guidelines listed by ACI 440 [9] for direct pull-out specimens. As already mentioned before, the specimens were designed as concrete cubes with the bar either embedded in the center or placed two or three times the diameter from one of the outer cube faces. The bar then had a long and a short end. The short end was used to measure the free end slip while load was applied to the longer end.
EC TE
135
O
131 132
O
130
The GFRP bars were manufactured by Hughes Brothers, Seward, Nebraska. These bars have a surface deformation called: ‘wrapped and sand coated design’. The ‘wrapped’ refers to the spiral glass fibers that are twisted around the bar. This decreases the amount of transverse thermal expansion and increases the mechanical interlock with the concrete. Bars are also sand coated to increase the friction and interlocking bond. The longitudinal and transverse coefficients of thermal expansion, as referred by the manufacturer, are 0.68O0.74!10K6/8C and 29!10K6/8C, per transverse and longitudinal direction, respectively. Tensile tests were performed on FRP bars to determine their engineering properties, which are related to fiber content. The average tensile strength, ultimate strain and modulus of elasticity obtained from the testing of the specimens (ASTM D3039) are presented in Table 3. Details of coupon fabrication and testing procedure are shown elsewhere [8].
PR
127
175
178
were tested under direct pull-out. The variables investigated were: bonded length, concrete cover and exposure to high temperature. Experimental results showed a significant degradation induced by exposure to relatively high temperatures.
D
126
174
177
122
125
173
176
121 123 124
171 172
200 201 202 203 204 205 206 207 208 209
Table 1 Test matrix
210
Bonded length mm (in.)
Thermal treatment
Concrete cover mm (in.)
Bar placement
76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6) 76 (3) 152 (6)
No No No No No No Yes Yes Yes Yes Yes Yes
148 (5.8) 148 (5.8) 19 (0.8) 19 (0.8) 29 (1.1) 29 (1.1) 148 (5.8) 148 (5.8) 19 (0.8) 19 (0.8) 29 (1.1) 29 (1.1)
Center Center 2f from 2f from 3f from 3f from Center Center 2f from 2f from 3f from 3f from
211 212 213 214 215
edge edge edge edge
216 217 218 219 220
edge edge edge edge
221 222 223 224
ARTICLE IN PRESS
DTD 5
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
281
2φ
225
3
226 227 228
282 283 284
230 231 232
Positions of the GFRP bar 3φ
229
233
285
152.4 mm (6 in)
152.4 mm (6 in)
Bond Breakers
286 287 288 289
234
290
235 236
291 292
152.4 mm (6 in) 152.4 mm (6 in)
237
293
Fig. 1. Thermal cracking specimens.
294
239
295
240
296
241
297
242
298
243 244
299 300
F
238
246 248 249 251 252 253 255 257
269 270 271
R
R
267 268
O
266
For direct pull-out testing, the concrete block is held in place while a load is applied to the long end of the bar. The measurements taken were free and loaded end slip of the bar and the load. Two LVDTs were used to measure the free end slip 1808 apart from one another. One LVDT was used to measure the loaded end slip because of the lack of room. This setup can be seen from Fig. 4. As the load was applied, readings were taken every second, until (1) rupture of the FRP bar occurred, (2) the enclosing concrete split, or (3) sufficient slippage of the free end of the bar occurred [10,11].
C
265
272 274
Average compressive strength MPa (psi)
Average splitting tensile strength MPa (psi)
277
1 2 3 4
33.9 (4920) 28.9 (4200) 28.6 (4150) 32.8 (4750)
3.39 (492) 3.05 (443) 3.03 (439) 3.31 (480)
U
Batch of concrete
279
309 310 311 312 313 314
The anchorage system required significant pre-test preparation. It was essential that a proper design was provided for the anchorage system to prevent damage to the rebars. Clearly, unprotected gripping crushes the FRP bar, and thus causes failure outside the gage length. The anchor system used consisted of a 200 mm (8 in.) long steel pipe, 25 mm (1 in.) in diameter. The pipe was placed on the end of the bar to be loaded and filled with a low viscosity epoxy resin. The low viscosity epoxy allowed for the epoxy to fill the pipe without leaving air pockets. The inside of the pipes were threaded to make sure the load was
280
315 316 317 318 319 320 321 322 323 324 325 326 327
Bar type
275 276 278
307 308
Table 3 Mechanical properties of FRP bars
Table 2 Mechanical properties of the concrete
N
273
306
Fig. 2. Test setup for pull-out test with central bar.
258
264
305
EC TE
256
263
304
D
254
262
303
PR
250
261
302
O
247
259 260
301
O
245
#3 GFRP bar
328 329
Dimensions of the bar
Average maximum strain (%)
Average maximum stress MPa (ksi)
Average elastic modulus MPa (ksi)
330
Nominal diameter 9.53 mm (0.375 in.)
1.85
760.0 (110.0)
40,800 (5920)
334
331 332 333 335 336
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
ARTICLE IN PRESS
DTD 5
4
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
transferred from the pipe to the resin. From Fig. 3 one can see that polymer caps were used to keep the pipe aligned with the bar. During testing, the load was applied to the pipe. For the specimens that had the off-center bar placement, a variation in the setup had to be introduced to prevent any bending on the bar. With reference to the sketch in Fig. 4, one can see that another six in block was tied to the one being tested. This kept the bar and the block straight, eliminating the bending moment that might have otherwise been created.
337 338 339 340 341 342 343
Fig. 3. Anchoring system.
344 345 346 347 348
(a) Straps Fastening Specimen and Concrete Block
349 350
152 mm3 (6 in3) Concrete Cube
352
A
A
353 354
357
O
359 360
Side View
361 363 364
(b) Specimen
152 mm (6 in)
PR
362
Concrete Block
152 mm (6 in)
365
369 370 371 372 373 374
Fixed End of the Testing Machine
Section A – A
376 377
Fig. 4. Test setup for the off-center specimens.
R
378
O
379 380 381
385 386 387 388 389 390 391
C
Bar placement
Bonded length mm (in.)
N
384
Center Center 2f from edge 2f from edge 3f from edge 3f from edge
76.2 (3) 152.4 (6) 76.2 (3) 152.4 (6) 76.2 (3) 152.4 (6)
397 398 399 400 401 402 403 404
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437
Table 4 Test results in terms of ultimate loads
U
383
R
375
382
D
Straps
EC TE
368
152 mm (6 in)
366 367
The specimens with 152.4 mm (6 in.) embedment lengths, 15 times the diameter of the bar, were too long for a 9.53 mm (0.375 in.) diameter bar. In some of the specimens, the bar broke before they would begin to slip. However, from the tensile tests on the bars, the breaking force was 54 kN (12 kip) while the maximum pullout force for the treated specimens was 42 kN (9 kip). Such inconsistency was explained as due to accidental imperfections such as, the eccentricity of the load generated by the non perfect alignment of the testing machine for the centered specimens or, the eccentricity generated at high load levels for eventual sliding between the two block. Test results in terms of average ultimate load are summarized in Table 4. By increasing the embedment length the maximum load that the specimens can hold also increases. From Table 4 it can be observed a 16% maximum reduction in the ultimate load for the thermally treated specimens consequent to bond degradation. The effects of thermal loads are more evident by looking at the load–slip curves. Typical load–slip curves for the pull-out tests performed are shown in Fig. 5. These curves show the difference between the behavior of the specimens with and without thermal treatment. For the untreated specimens there was almost no slip at the freeend until the load reached its ultimate value. For the thermal treated specimens instead, the same peak bond stress is attained but with higher slips values. The bond failure in this
O
Fixed End of the Testing Machine
Specimen
358
395 396
406
F
355 356
394
405
2.4. Test results and discussion
351
393
438 439 Ultimate load kN (kip)
Covariance (%)
Thermal treatment (PT)
No thermal treatment (PNT)
Thermal treatment
34.98 (7.86) 40.99 (9.21) 34.55 (7.77) 28.77 (6.47) 39.34 (8.84) 30.69 (6.90)
34.69 (7.80) 42.26 (9.50) 33.34 (7.50) 34.25 (7.70) 37.82 (8.50) 34.70 (7.80)
8.8 6.2 23.3 14.6 7.2 18.4
392
No thermal treatment 5.83 5.8 14.1 13.8 7.8 14.50
Strength loss (PNTKPT)/PNT! 100 (%) K0.84 3.01 K3.63 16.00 K4.02 11.56
440 441 442 443 444 445 446 447 448
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
ARTICLE IN PRESS
DTD 5
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
45 40 35 30 25 20 15 10 5 0
451 452 453 454 455 456
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
0
457
505
(b) 50 Load (kN)
(a) 50
450 Load (kN)
449
0.25
0.5
0.75
45 40 35 30 25 20 15 10 5 0
1
Bar Breaking
0
507 508
0.25
0.5
509 510 511
0.75
1
Displacement (mm)
Specimens having a 76.2 mm (3 in) Embedment Length, GFRP Bar Placed at the Center of the Specimen.
459 460
506 Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
Displacement (mm)
458
5
515 516 517
40 35 30 25 20 15 10 5 0
45 40 35 30 25 20 15 10 5 0
467 468 469
0
470
0.25
486
0.75
Displacement (mm)
483 484 485
0.5
495 496 497 498 499 500 501 502 503 504
531 532 Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
0.50
0.75
534 535 536 537 538 539 540
Specimens having a 152.4 mm (6 in) Embedment Length, GFRP bar placed 2 φ from edge.
541 542 543
R
544 545 546
R
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
1.00
533
Displacement (mm)
547 548
O
494
case is caused by the shearing of the matrix material from the fibers in the bar. The GFRP bar consists of a core of glass fibers with an outer layer of resin and sand. As a consequence of the thermal treatment there is a weakening of the matrix and thus it shears off as observed in Fig. 6. In Fig. 7, the pullout curves of treated specimens having different concrete cover are compared for the two analyzed embedment length. It can be observed that the weakening due to the thermal treatment is more evident when the concrete cover is small, especially for the six in embedment length specimens. Such behavior can be explained with the more extended micro cracking of the concrete when the bars are placed at a lower cover. Such micro-cracks are induced by the bursting stresses within the concrete due to the different CTE of GFRP bars and concrete; obviously they are more pronounced when lower covers are used.
529 530
549 550
C
493
0.25
526 528
Failure of the Bar Splitting of the Cover
551 552
N
491 492
45 40 35 30 25 20 15 10 5 0 0.00
553 554
U
490
1
527
Fig. 5. Comparison of free-end slips.
488 489
1
Specimens having a 76.2 mm (3 in) Embedment Length, GFRP bar placed 2 φ from edge.
487
525
0.75
Specimens having a 152.4 mm (6 in) Embedment Length, GFRP bar placed 3 φ from edge.
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
0.25
523 524
Displacement (mm)
(f) 50
0
0.5
522
O
45 40 35 30 25 20 15 10 5 0
482
0.25
D
475 476
481
0
Load (kN)
(e) 50
Load (kN)
474
480
1.25
Specimens having a 76.2 mm (3 in) Embedment Length, GFRP bar placed 3 φ from edge.
473
479
1
521
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
PR
472
478
0.75
520
Displacement (mm)
471
477
0.5
519
Concrete Splitting
EC TE
466
Specimen 1 (No Thermal) Specimen 1 (Thermal) Specimen 2 (No Thermal) Specimen 2 (Thermal) Specimen 3 (No Thermal) Specimen 3 (Thermal)
518 FRP Rupture
F
463
O
(d) 50 Load (kN)
(c) 45 Load (kN)
462
465
513 514
Specimens having a 152.4 mm (6 in) Embedment Length, GFRP Bar Placed at the Center of the Specimen.
461
464
512
555 556 557 558 559 Fig. 6. The shearing of the resin.
560
ARTICLE IN PRESS
DTD 5
6
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
561
Pullout force (kN)
562
617
Pullout force (kN)
45
45
618
563 564
40
40
35
35
619 620
565
30
30
621
566
25
25
567
20
568
Em. length: 3 in (76 mm) Center 3d cover 2d cover
15
569 570
10
571 572
5 0
573
15 10 5
0.2
574
0.4
0.6
0.8
1
1.2
0
623
Em. length: 6 in (152 mm) Center 3d cover 2d cover
20
Free end slip (mm)
0
622 624 625 626 627 628
Free end slip (mm)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
629 630
Fig. 7. Experimental free end slip-pullout force results for treated specimens.
575
631
576
632
Interfacial shear stress (N/mm2)
20
20 18
12
583
10
584
8
585
6
586
4
587 588
2 0
Em. length: 3 in (76 mm) Center 3d cover 2d cover mBEP CMR
0
0.5
1
1.5
14
Av. mBEP
12
Em. length: 6 in (152 mm) Center 3d cover 2d cover mBEP CMR
10 8 6 4 2
Slip (mm)
0 2
0
0.5
589
600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616
638 639
1
640 641 642 643 644
Slip (mm) 1.5
2
645
D
646
The dependence of the local t(s) by the cover is a consistent result, while the strong dependence on the embedment length is in contrast with the assumption of the existence of a local t(s) law.
EC TE
599
R
598
In Eqs. (1) and (2), t(s) is the interfacial shear stress corresponding to the bar/concrete slip, s. The unknown parameters a, C, s and b, tm, sr were found by minimizing the distance between the measured and calculated applied load N—free displacement, sf. The procedure takes into account the deformation of the bar within the embedment length. For all the sets of equal specimens the t(s) law was determined. These laws are plotted in Fig. 8. In this figure, the average t(s) curve, obtained by considering the average of the local t(s) curve fits to all the treated specimen results, is also plotted. It can be observed, in general, that better behaviors of the specimens result with shorter embedment lengths and large concrete cover.
R
597
O
595 596
The local bond–slip laws of the considered bars after thermal treatment have been determined according to [12], viz the modified Bertero Eligehhausen Popov (mBEP) Eq. (1) s tðsÞ Z Csa 1 K (1) s and the Cosenza, Manfredi Realfonzo [13] (CMR) Eq. (2): s b tðsÞ Z tm 1 K eKsr (2)
C
594
N
593
2.5. Analytical approach
U
592
637
Fig. 8. Local bond–slip relationships.
590 591
635 636
F
Av. mBEP
14
582
Av. CMR
16
16
581
634
18
Av. CMR
O
579 580
633
O
578
Interfacial shear stress (N/mm2)
PR
577
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
647 648 649 650 651 652 653
Interfacial shear stress (N/mm2)
654
20
655
18
656
Av. CMR
657
16
658
Av. mBEP
14
659 660
12
661
10
662
8
663 664
6
665
4
666
2
667 668
Slip (mm)
0 0
0.5
1
1.5
2
Fig. 9. Average local bond–slip relationships obtained considering all the treated specimens.
669 670 671 672
ARTICLE IN PRESS
DTD 5
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
731 732
Nestore Galati is a research engineer at the University of Missouri, Rolla. He obtained is PhD in Civil Engineering at the University of Lecce, Italy where he also received his BSc in Materials Engineering. He obtained his MSc degree in Engineering Mechanics at the University of Missouri-Rolla. His research has concentrated on the area of retrofitting of masonry and upgrade and in situ load testing of reinforced concrete structures (building and bridges). He is an EIT in the United States.
765
730
a
C
s
b
tm
sr
0.28
14.40
/CN
0.35
66.10
58.40
679 680 681 682 683 684 685 686 687
The final t(s) curves, an average of all treated specimens, computed from the mBEP and CMR equations are plotted in Fig. 9; the corresponding parameters are reported in Table 5. For the average mBEP law, it is found s /CN, which indicates that the same local bond–slip curve could also be obtained utilizing the Bertero Eligehhausen Popov bond–slip equation [14]. The result s /CN is an effect of averaging the results; for single specimens, s Z 1O 3 mm was found.
688 689 690 691 692 693
3. Conclusion On the basis of the experimental investigations, the following concluding remarks can be drawn:
694
701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
EC TE
699 700
R
698
R
697
O
696
(1) In most of the specimens the thermal treatment induced a slight degradation in the bond performance in terms of ultimate load. A more pronounced effect was observed for the bond stress–slip curves in terms of slip values due primarily to the degradation of the resin; (2) A number of specimens with an embedment length of 15 times the diameter of the rebar failed in the free length of the rebar before slip commenced; (3) Future research will be directed toward the direct pullout test at high temperature. This will enable any bond degradation to be determined when the structure is subject to high temperature, and thus verifying whether the current standards are adequate for the designing of internal FRP reinforcement; (4) The influence of the thermal treatment is more evident with the small values of the concrete cover. Such behavior was explained with the microcrackingoftheconcreteduetothe bursting stresses induced during the thermal treatment; (5) The local bond–slip laws utilized in the current investigation present a large scatter in the results, thus the average values of these laws can only be considered as representative of an order of magnitude. Due to the scattering of the results it was not possible to determine a correlation between the CMR/mod-BEP parameters o treated and un-treated specimens.
C
695
722
N
720 721
Acknowledgements
725
The support of the National Science Foundation Industry/ University Cooperative Research Center at the University of Missouri-Rolla. The authors would also like to acknowledge the support of the Rolla Technical Institute (RTI).
727 728
U
723 724 726
F
678
[1] Nanni A. Relevant applications of FRP composites in concrete structures. In: Figueiras J, Juvandes L, Furia R, editors. Proceedings, CCC 2001, Composites in Construction, Porto, Portugal, Oct. 10–12, 2001, 2001. p. 661–70. [2] Harries KA, Porter ML, Busel J. FRP materials and concrete— research needs. Concr Int 2003;25(10). [3] Gentry TR. Thermal compatibility of plastic composite reinforcement and concrete. In: El-Badry MM, editor. Advanced composite materials in bridges and structures. Montreal, Que.: Canadian Society for Civil Engineering; 1996. p. 149–56. [4] Tepfers R. Division of building technology. Go¨teborg: Chalmers University of Technology; 1998 [Publication no. 98:3, Work no. 22. p. 1–16]. [5] Tepfers R. Cracking of concrete cover along anchored deformed reinforcing bars. Mag Concr Res 1979;31(106):3–12. [6] ASTM C 39/C 39M. Standard test method for compressive strength of cylindrical concrete specimens. [7] ASTM C496-96. Standard test method for splitting tensile strength of cylindrical concrete specimens. [8] Secondin S. Masonry reinforced with FRP systems. Tesi di Laurea, Universita` degli Studi di Padova. Padova, Italy: Facolta` di Ingegneria; 2003. [9] ACI Committee 440. Recommended test methods for FRP rods and sheets. ACI 440-K.09-011. Farmington Hills, MI: American Concrete Institute International; 2001. [10] Boothby TE, Nanni A, Bakis CE. Accelerated test methods for frp/concrete systems in highway structures. In: Buckle IG, Friendland IM, editors. Fourth national workshop on bridge research in progress, 1996. p. 17–19 [Buffalo, NY]. [11] Conrad JO, Bakis CE, Bootby TE, Nanni A. Durability of bond of various FRP rods in concrete. Sherbrooke, Canada: CDCC-98; 1995 p. 299–310. [12] Focacci F, Nanni A, Bakis CE. Local bond–slip relationship for FRP reinforcement in concrete. J Compos Construct 2000;4(1):24–3. [13] Cosenza E, Manfredi G, Realfonzo R. Behavior and modeling of bond of FRP rebars to concrete. J Compos Construct 1997;1(2):40–5. [14] Eligehausen R, Popov EP, Bertero VV. Local bond stress–slip relationships of deformed bars under generalized excitations. Report No 83/23, EERC. Berkeley: University of California; 1983.
CMR
O
677
mBEP
729
O
675 676
References
PR
674
Table 5 Parameters of local bond–slip laws for treated bars (N, mm)
D
673
7
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
Antonio Nanni FACI, is the V and M Jones Professor of Civil Engineering at the University of Missouri, Rolla, Rolla, MO. He was the founding Chair of ACI Committee 440, Fiber Reinforced Polymer Reinforcement, and is the Chair of ACI Committee 437, Strength Evaluation of Existing Concrete Structures.
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 766 767 768 769 770 771 772 773 774 775 776 777
Lokeswarappa R. Dharani is Professor of Engineering Mechanics and Aerospace Engineering at the University of Missouri, Rolla. His research has concentrated in the fields aircraft structures, fracture mechanics, fatigue and failure analysis, micromechanics, composite materials and structures, process modeling of ceramic matrix composites, friction and wear of composites, fracture of laminated glazing under impact loads. He is a fellow of the American Society of Mechanical Engineers (ASME) and of the American Institute of Aeronautics and Astronautics (AIAA).
778 779 780 781 782 783 784
ARTICLE IN PRESS
DTD 5
8 785 786 787 788 789
N. Galati et al. / Composites: Part A xx (xxxx) 1–8
Francesco Focacci is Associate Professor at the Dipartimento di Costruzione dell’Architettura (DCA), University of Venice, Italy. His research deals with the structural behavior of FRP reinforced concrete members, bond between FRP bars and concrete and structural rehabilitation using FRP.
790
Maria A. Aiello fib member, IABSE member, is Associate Professor at the Department of Innovation Engineering, University of Lecce, Italy: Her research interests include the structural analysis of concrete beams reinforced with FRP rebars, the structural behavior of RC beams strengthened with FRP laminates, the bond between concrete or masonry elements and FRP reinforcement, the stability analysis of composite laminates and sandwich panels.
841 842 843 844 845 846 847
792
848
793
849
794
850
795 796
851 852
797
853
798
854
799
855
800
856
801
857
802
858
803 804
859 860
F
791
O
805 806
O
807 808 809
PR
810 811 812 813
D
814 815
EC TE
816 817 818 819 820 821 822 823
R
824 825
R
826
O
827 828 829
C
830 831
834 835 836
U
833
N
832
861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892
837
893
838
894
839
895
840
896
JCOMA 1618—1/7/2005—19:16—VEERA—154352—XML MODEL 5 – pp. 1–8
