Structural response of a fibre reinforced concrete pile-supported flat slab: full-scale test

1  Structural response of a fibre reinforced concrete pile-supported flat slab: full-scale test 1 Stanislav Aidarov a, b, *, Francisco Mena b, Albert de la Fuente b 2 a Smart Engineering Ltd., UPC Spin-Off, Jordi Girona 1-3, 08034 Barcelona, Spain 3 b Civil and Environmental Engineering Department, Universitat Politècnica de Catalunya (UPC), Jordi Girona 1-3, 08034 4 Barcelona, Spain 5 * Corresponding author. Tel.: +34 633 634 207; Full Postal address: 08242, 08034, Barcelona, Jordi Girona 1; Email 6 address: [email protected] ; [email protected] 7 Abstract 8 The total substitution of traditional reinforcement in the form of steel bars by fibres can be mainly found in 9 elements with favourable boundary conditions and subjected to low-moderate load levels. However, the 10 rigorous study of fibre reinforced concrete (FRC) and its potential fields of application over the last decades 11 permitted this material to face structural application with greater responsibility in terms of structural integrity 12 and mechanical capacity – construction of FRC flat slabs. This promising technology was used in a dozen 13 buildings with recognition of positive outcomes with respect to the optimization of resources, reduction of 14 execution time, and environmental impacts. Despite these achievements, the application of FRC in flat slabs 15 is still limited in the building sector because of certain concerns of the material capacities, and existence of 16 some aspects related both to service and ultimate limit states, which are still unclear. With this in mind, an 17 extensive experimental programme was carried out and focused on the construction of a full-scale FRC flat 18 slab and its loading protocol in order to analyse both crack and deflection patterns. Likewise, the structure 19 was led to failure, which also allowed assessing both the bearing and deformability capacities as well as the 20 fibre distribution and orientation. The results derived from this experimental program are expected to 21 increase the confidence of designers and practitioners on the use of FRC as structural material for column-22 supported flat slabs. 23 24 Keywords: elevated slab, two-way slab, ultimate behaviour, serviceability behaviour, ductility, full-scale test 25 26 1. Introduction 27 The incorporation of fibres in cement-based composites allows increasing the fracture energy of 28 the matrix [1] by providing post-cracking strength and, hence, improving the ductility [2], cracking 29 control [3–5], and fatigue [6,7]; fire and impact resistance can also be enhanced depending on 30 2  the type of fibres used [8–10]. These improvements enable the possibility of a partial or even total 31 substitution of the traditional steel bars by structural fibres [11]. 32 Over the last decades, a significant number of researchers has been investigating the FRC 33 technology and noticeable advances were achieved, even the acceptance of FRC as structural 34 material in the fib MC 2010 [11]. Nevertheless, so far, the FRC has been primarily used in 35 elements for which the applied bending moments during transient and service loading situations 36 are relatively low (usually below the cracking bending moment) and/or for those whose 37 consequences of failure are minor. For instance: slabs on grade [12–14], precast tunnel segments 38 [15–17], reinforced earth-retaining walls [18], and sewerage pipes [19]. The gained experience 39 along with ongoing research have expanded the knowledge base related to FRC and, 40 consequently, this is increasing its applicability in elements with higher structural responsibility. 41 One of the most complex challenges within this topic became the construction of FRC elevated 42 slabs for residential buildings with total substitution of traditional reinforcement because of the 43 occurrence of high tensile stresses during the service life of the structure. The first full-scale tests 44 proved that FRC with steel macrofiber content of 100 kg/m3 (volume fraction 1.3%) could provide 45 the required structural integrity even under the loads that considerably exceeded the standard 46 magnitudes for residential flat slabs at ultimate limit state (ULS) [20,21]. Other experimental 47 results pointed out: the remarkable redistribution capacity of FRC statically redundant systems 48 [22–24]; economically promising results of hybrid solutions (fibres + steel bars) [25,26]; 49 improvement of punching strength [27–29], and a sufficiently even distribution of fibres in the 50 concrete mix to consider the material as homogenous [30,31]. 51 Steel fibre reinforced concrete (SFRC) technology in construction of residential flat slabs was also 52 applied in real projects: the Ditton Nams shopping mall (Latvia), the Triangle office building 53 (Estonia), the Rocca Al-Mare office tower (Estonia), and the LKS office building (Spain) are 54 representative examples. Each of these projects highlighted technical advantages of using SFRC: 55 (1) the analysis of traditional reinforcement substitution by SFRC in Ditton Nams shopping mall 56 and Triangle office building acknowledged economic savings [32]; (2) the 9 week time-saving 57 effect along with reduction of the required machinery were mainly denoted during the construction 58 of 16-floor Rocca Al-Mare office tower [33]. 59 3  In case of the LKS office building, a more detailed comparative study was presented in [35]. This 60 research presented three different alternatives to be analysed: (1) steel bar reinforced concrete 61 (original design), (2) self-compacting SFRC with complementary use of steel bars in specific areas 62 of the slab plus anti-progressive collapse reinforcement (APC), (3) self-compacting SFRC with 63 complementary use of steel bars in specific areas of the slabs. Both SFRC solutions provided 64 economic benefits: 12% and 16% reduction of total costs in comparison with the original design 65 was reported for SFRC + APC and SFRC, respectively. Despite the fact that APC reinforcement 66 is, basically, a redundant reinforcement to prevent local failures [36] which could lead to the 67 collapse of the entire structure, the alternative with APC reinforcement was selected for the LKS 68 office building since this represented a pioneer experience in Spain and, as a consequence, an 69 additional safety margin was provided. 70 The achieved results in both experimental and real projects gathered in Table 1 could have 71 already turned SFRC technology into a competitive alternative for two-way slab construction. 72 However, a number of constraining factors that require further development are compromising 73 the widespread use of the SFRC for column-supported flat slabs. General comprehension of 74 SFRC in terms of fibre distribution and orientation [37,38], ULS capacity of SFRC in statically 75 redundant structural configurations [39,40], and the effect of long-term loads in terms of 76 deformation and cracking [41,42] are found to be those more relevant. Also, the potential 77 reduction of fibre content along with precise tools for thorough comparison with other alternatives 78 [43] and established criteria for quality control [44,45] demand further detailed investigation. 79 Table 1 Cases of SFRC application in two-way slabs 80 Author Type Cf [kg/m3] Lmax/h [m/m] Steel bars lf / Øf / Rm - type Di Prisco et al. [22] SSLT 35 13 Yes/None 60/0.9/1500 – DHE Blanco et al. [46] SSLT 40 15 None 50/0.62/1270 – SHE Fall et al. [47] SSLT 35 28 Yes/None 60/0.9/2300 – DHE Facconi et al. [25] SSLT 20/25 32 None 32/0.4/2200 – SHE Facconi et al. [25] SSLT 25 32 None 60/0.9/2200 – DHE Barros et al. [48] FSFT 90 16 Yes* 37/0.5/1100 – SHE Salehian et. al [35] FSFT 90 16 Yes* 37/0.5/1100 – SHE Destreé et al. [21] FSFT 45 19 No 50/1.3/850 – C Hedebratt et al. [26] FSFT 40/80 23 Yes/None 60/0.9/1160 – SHE 4  Døssland [31] FSFT 62 23 Yes/None 60/0.9/1000 – SHE Destreé et al. [21] FSFT 100 28 Yes* 50/1.3/850 – C Parmentier et al. [30] FSFT 70 30 Yes* 60/1.0/1450 – SHE Gossla [20] FSFT 100 30 Yes* 50/1.3/900 – C Ošlejs [49] RB 100 24 Yes* 50/1.3/900 – C Maturana [35] RB 100 27 Yes 50/1.3/900 – C Present study FSFT 70 30 Yes* 60/0.9/2300 - DHE NB: Cf – fibre content; Lmax – maximum span; h – depth of the slab; lf – fibre length in mm; Øf – fibre diameter in mm; Rm 81 – fibre tensile strength in MPa; SSLT – small scale laboratory test; FSFT – full scale field test; RB – real building; DHE – 82 double hooked-end; SHE – single hooked end; C – crimped; Yes* – only presence of APC steel bars. 83 To shed light on some aspects of abovementioned factors, an extensive experimental program 84 was carried out in this study by following these steps: (1) characterization of several self-85 compacting SFRC mixes with different fibre content and type; (2) construction of SFRC slabs of 86 different dimensions maintaining the same slenderness; (3) testing these slabs under various 87 boundary and load conditions. This paper focuses on the construction of a four-panel column-88 supported SFRC flat slab and its sequential loading protocol aimed at characterizing the structural 89 response of the structure for load levels representative of both service and ultimate limit states. 90 To this end, increasing uniformly distributed load (UDL) was applied onto the structure, in contrast 91 to previous experiences in which point loads were mainly used, especially for simulation of ULS 92 conditions (Table 2). After failure, 32 samples were cored from the tested SFRC slab to examine 93 fibre distribution both throughout the element and over the slab depth. Additionally, fibre 94 orientation was evaluated by means of an inductive test with the aim of assessing the contribution 95 of fibres in the direction of the three main axes, which could be valuable information for further 96 studies related to design procedures. 97 2. Experimental programme 98 2.1. Geometry of SFRC slab 99 A 12.0 × 10.0 × 0.2 m SFRC slab supported by nine columns with square cross section of 0.25 100 m was tested. These columns were supported on 1.85 × 1.85 × 0.7 m concrete footings and 101 formed four panels of 6.0 x 5.0 m2 each. Figure 1 presents the geometry of superstructure along 102 with the detailed description of the established reinforcement for concrete columns and footings. 103 The geometry was selected to reproduce common dimensions of slab panels that could be used 104 5  in office and residential buildings. Additionally, previous experiences were considered and the 105 achieved span-to-depth ratio (30) corresponded to the upper limit of those already constructed 106 (Table 1). 107 Table 2 Parameters studied within previous research programmes 108 Author Cf,real C i SLS ULS Cracking UDL G_UDL PL UDL W P E Di Prisco et al. [22] ● ● ● Blanco et al. [46] ● ● ● Fall et al. [47] ● ● Facconi et al. [25] ● ● ● Facconi et al. [25] ● ● ● Barros et al. [48] ● Salehian et. al [35] ● ● ● ● ● Destreé et al. [21] ● Hedebratt et al. [26] ● ● Døssland [31] ● ● ● Destreé et al. [21] ● Parmentier et al. [30] ● ● ● ● ● Gossla [20] ● ● ● Ošlejs [49] ● Maturana [35] ● ● Present study ● ● ● ● ● ● ● NB: Cf,real – real fibre distribution in the element; Ci – relative contribution of fibres in three main axes; G_UDL – gradual 109 uniformly distributed loading; PL – point loading; W – crack opening; P – crack pattern; E – crack evolution 110 111 Figure 1 Geometry of the SFRC prototype (units in [mm]) 112 2.2. Material properties 113 The experimental investigation presented herein was carried out by using a self-compacting 114 SFRC with a steel fibre content of 70 kg/m3 (volume fraction 0.9%) – the lower limit, to best 115 authors’ knowledge, reported in the scientific literature related to the erection and testing of full-116 scale SFRC two-way slabs of this slenderness. The used fibre was characterised by a high tensile 117 6  strength (2300 MPa), double hooked-end shape and an aspect ratio of 65. Despite the relatively 118 high content of fibres, the material was required to guarantee sufficient workability to avoid the 119 concrete vibration. The absence of vibrating minimized the external influence on the distribution 120 and orientation of fibres. Table 3 summarizes the details of the used concrete mix. 121 Table 3 Concrete mixture 122 Cement CEM II/A-L 45,2R (kg/m3) 425 Coarse aggregate 10/20 (kg/m3) 250 Coarse aggregate 4/10 (kg/m3) 150 Fine aggregate 0/4 (kg/m3) 725 Fine aggregate 0/2 (kg/m3) 600 Limestone filler (kg/m3) 25 Water-cement ratio 0.47 Additives (% on cement content) 2.77 Steel fibres (kg/m3) 70 123 Four concrete trucks were required for the construction of the SFRC prototype slab and each of 124 the batches was tested to evaluate the properties of the material. Mean values at 28 days for the 125 modulus of elasticity (𝐸), compressive strength (𝑓 ) and residual flexural strengths (𝑓,) are 126 gathered in Table 4. The presented properties at the hardened state were obtained in accordance 127 with standards EN 12390-13 [50], EN 12390-3 [51], and EN 14651 [52], respectively. Three 128 specimens per batch were tested to obtain 𝐸 and 𝑓 ; whereas four 150 × 150 × 600 mm 129 notched prisms for 𝑓,. 130 Table 4 Characterization of SFRC at 28 days 131 Batch 1 Batch 2 Batch 3 Batch 4 Overall Average (MPa) CV (%) Average (MPa) CV (%) Average (MPa) CV (%) Average (MPa) CV (%) Average (MPa) CV (%) 𝐸 25840 0.5 25030 1.4 26270 2.0 28080 3.1 26310 4.9 𝑓  42.2 0.9 39.5 1.2 42.3 1.5 47.5 6.3 42.9 7.8 𝑓  3.9 12.6 4.4 4.1 4.4 11.6 4.5 6.4 4.4 12.4 𝑓  5.8 30.0 7.8 9.1 5.3 51.5 10.0 10.2 7.2 35.3 𝑓  6.9 31.7 8.6 19.2 7.3 32.5 9.6 15.8 8.1 26.0 𝑓  6.7 25.7 8.4 21.6 6.9 33.4 9.0 13.4 7.7 24.7 𝑓  6.5 21.9 8.0 20.0 6.4 30.9 8.6 9.4 7.3 22.7 All tested batches had similar mechanical properties in terms of compressive strength and 132 modulus of elasticity. The relatively low values of the modulus of elasticity can be explained by 133 7  increased binder content that might have led to the reduction (<20%) of 𝐸 compared with 134 conventional concrete with equivalent 𝑓  [11]. The obtained results of 𝑓, had a higher scatter in 135 comparison with the expected values (especially for the given fibre content) [53,54]. This 136 phenomenon led to considerable reduction of the characteristic values of residual flexural 137 strengths 𝑓,, the SFRC resulting in a 3e strength class according to the fib MC-2010 138 classification, whereas the obtained mean values corresponded to the 7c strength class [11]. 139 However, taking into account that high density of cracks were expected, the mean 𝑓, was found 140 to be representative of the post-cracking response of the material in the structure as the intrinsic 141 scatter decreases due to the statistical scale effect reported in [55]. 142 2.3. Test configuration and procedure 143 The loads specified in the Spanish Building Code for residential buildings [56] were considered 144 as reference for designing the loading protocol. In this regard, in addition to the 4.8 kN/m2 due to 145 the self-weight (𝑞), a dead load (𝑞) and variable load (𝑞) of 2.0 and 3.0 kN/m2, respectively, 146 were considered. Load partial safety factors 𝛾1.35 and 𝛾1.50 were assumed for 147 computing the design load 𝑞 𝛾 ∙󰇛𝑞 𝑞 󰇜𝛾 ∙𝑞  13.7 ≈ 14.0 kN/m2 (for safe-side 148 purposes). 149 The load was applied gradually in different time-steps in order to evaluate the response of the 150 structure in terms of cracking and deformation. For this purpose, the load was provided by means 151 of concrete cubes of 0.5 × 0.5 × 0.6 m (≈ 350 kg each), dividing the process into four steps (Figure 152 2). Time-spans between loading phases (Table 5) depended on the structural response, i.e. the 153 next phase could only be started if the total deflection increment, measured in the centres of each 154 panel, during successive 10 days was less than 1 mm. The deflections of the structure were 155 monitored by means of eight prisms and topography (Figure 3). These were measured before 156 and after every load step to quantify the instantaneous component. The time-dependent 157 component of the deflections was measured three times per week. 158 8  Figure 2 Load phases: a) I b) II c) III and d) IV 159 160 Phases Applied Load (kN/m2) Duration (days) qSW – I 4.8 14 I – II 6.0 14 II – III 7.5 14 III – IV 8.7 22 IV – ULS_I 9.8 140 Figure 3 Position of installed prisms Table 5 Time-span between load phases Each loading step was complemented by a characterization of the bottom slab crack pattern 161 evolution. Multiple high-resolution pictures were taken from underneath the structure one day 162 before and two days after each load phase to evaluate the crack propagation. This procedure 163 permitted to distinguish the cracks produced by load increments from those caused by other time-164 dependant imposed deformations (i.e., shrinkage, creep and temperature gradients). 165 Since the fourth loading phase was applied, the preparation for load increment up ULS conditions 166 was started. Firstly, acknowledging the significant load magnitude of the fourth phase (9.8 kN/m2), 167 the duration of the stabilization period was increased (Table 5): the phase was considered finished 168 once the deflections measured by means of the prisms did not vary (with an accuracy of 1 mm) 169 9  during two months. Afterwards, water tanks were placed onto first and second panels of the SFRC 170 slab above the concrete cubes according to Figure 4a-b. The presented configuration provided 171 the possibility of increasing the load up to 16 kN/m2, i.e. a certain margin was left in comparison 172 with the 𝑞 of 14 kN/m2. 173 Each panel was loaded increasing the water level by steps of 300 mm (2 kN/m2 per step); the 174 rate of filling was controlled by the inspection of the structural response of the SFRC prototype. 175 The loading initiated by filling the tanks on the first and second panels, successively. The analysis 176 of the SFRC slab behaviour under the first phase of ULS loading (ULS_I) was followed by water 177 tank structure shifting from the already tested half of the element to the third and fourth panels 178 (Figure 4c-d). The loading procedure (ULS_II) was repeated on the other part of the structure in 179 a week. 180  Figure 4 SFRC flat slab under ULS conditions: a-b) ULS_I: loading of the first half of the slab (3D model 181 and real picture); c-d) ULS_II: loading of the rest of the slab (3D model and real picture) 182 After these two phases of loading, the analysis of both upper and lower surfaces (once the tanks 183 and cubes were removed from the structure) was carried out. The bottom crack pattern surface 184 was measured in accordance with the previously described method, whereas the top face of the 185 SFRC prototype was studied by means of a photogrammetry technique for better precision. 186 a) b) c) d) 16  factor 𝐾 = 1.40 could be assumed in accordance with MC 2010 [11]. As a result, an UDL of 9.6 316 kN/m2 is derived from considering these assumptions and Equation 1, which is 40% below the 317 experimental UDL (16 kN/m2). 318 Alternatively, considering 𝑓 = 7.7 MPa as representative for simulating the post-cracking 319 response of the FRC at the yielding line and considering 𝛾 = 1.00, the ultimate resisting moment 320 (𝑀) of the yielding line results to be 51.3 kNm/m, which leads to a UDL of 17.3 kN/m2 – 1.8 times 321 greater 𝑞. The test was stopped when a 16 kN/m2 was reached; however, further load increment 322 was possible since the prototype did not evidence signs of upcoming bending failure. Thus, the 323 analytical estimation of the load bearing capacity of the prototype by using mean values of the 324 mechanical parameters proved to provide a reasonable forecast. 325 3.3. Crack patterns 326 The first crack appeared at the vicinity of the central column after the formwork removal. This 327 crack was in line with the nonlinear simulations [72] that were carried out before the construction 328 of the SFRC prototype. The inspection of the lower surface presented no crack formation under 329 the self-weight of the structure; no visible cracks appeared after the second load phase (Figure 330 2b), which roughly corresponded to the quasi-permanent combination (𝑞, = 7.7 kN/m2) of load 331 as it was described in the Section 3.2. Cracking occurred once the third phase was carried out 332 (Figure 10a), i.e. once a UDL of 8.7 kN/m2 was applied (Figure 2c). This magnitude of UDL was 333 much closer to the characteristic load combination (𝑞,𝑞  𝑞 𝑞 ) of 9.8 kN/m2 rather 334 than the quasi-permanent. Nevertheless, the SFRC flat slab resulted to prove significant cracking 335 control under this load level: after the application of the load and stabilization process, none of 336 the observed cracks at this loading stage exceeded the width of 0.3 mm – the generally 337 established limit of crack widths (wmax) for several exposure classes under quasi-permanent load 338 combination of actions [11,66,73]. 339 17    340 Figure 10 Crack patterns of the lower surface: a) after the load phase III; b) after the load phase IV; c) 341 before ultimate limit loading 342 Further increment of the load during phase IV provoked the appearance of new cracks that are 343 depicted in red (Figure 10b). At this stage, several cracks did exceed the width of 0.3 mm, what 344 was expected because of the considerable UDL of 9.8 kN/m2. This magnitude of the sustained 345 load along with the thermo-hygral phenomena (i.e., shrinkage and creep) led to bending moment 346 and stresses redistribution – Figure 10c shows the evolution of the crack pattern under the same 347 load of 9.8 kN/m2 in comparison with the one presented by Figure 10b. This evolution of cracks 348 can be also observed in Table 7 which presents the number of observed cracks and the total 349 length of those for each load phase. 350 Table 7 Number/total lengths of produced cracks under different load phases 351 Applied Loads Number / Total length (m) of cracks wmax (mm) Panel 1 Panel 2 Panel 3 Panel 4 Average CV (%) Phases I-II 0 / 0 0 / 0 0 / 0 0 / 0 0 / 0 0 / 0 0 Phase III 14 / 5.9 19 / 12.0 13 / 6.8 12 / 13.3 14.5 / 9.5 21.4 / 38.9 < 0.3 Phase IVa 54 / 21.3 45 / 24.9 34 / 15.3 30 / 18.0 40.8 / 19.9 26.7 / 20.9 > 0.3 Phase IVb 96 / 43.1 74 / 41.1 71 / 34.3 69 / 37.9 77.5 / 39.1 16.1 / 9.8 > 0.3 ULS I-II 123 / 82.8 101 / 78.5 135 / 70.7 126 / 77.4 121.3 / 77.4 11.9 / 6.5 > 0.3 Achieving ULS conditions by filling the water tanks triggered a significant increase of both the 352 width and crack density (Figure 11a); the total crack lengths of all panels increased two times in 353 comparison with the values of phase IVb. Figure 11b confirms the presence of cracks on the top 354 surface of the prototype. The locations of these cracks are in total compliance with Johanssen´s 355 theory [62,63] – along the column lines where the negative yield lines were formed. In addition, 356 the effect of combined local flexural and shear stresses can be observed on Figure 11b by 357 a) b) c) PhaseIVa PhaseIVb 18  presence of cracking near corner and edge columns. Figure 12 presents the crack patterns of the 358 lateral surfaces of SFRC slab to provide detailed information related to these cracks. The lateral 359 surfaces are presented in accordance with corresponding panel for the sake of better scaling and, 360 as a consequence, easier comprehension. 361 a) b)  Figure 11 Crack patterns after the test 362 Panel 1  Panel 2  Panel 3  Panel 4  Figure 12 Lateral crack patterns after the test 363 Ly = 6000 mm Ly = 6000 mm L x = 5000 mm L x = 5000 mm Ly = 6000 mm L x = 5000 mm Ly = 6000 mm L x = 5000 mm 19  3.4. Fibre distribution 364 The structure maintained its integrity after the demolition process (Figure 8); hence, the cores 365 were drilled in accordance with the pattern established in Figure 13b. The height of extracted 366 specimens was measured to evaluate the real depth of the slab – an essential parameter for 367 further studies in terms of both analytical and nonlinear analyses, including uncertainties and 368 tolerances. The obtained mean thickness of 207.2 mm was slightly higher in comparison with the 369 design value (200 mm). This depth may serve as a reference value in following design procedures 370 due to similar measurements of slab thickness throughout the structure with a small coefficient of 371 variation (3.1%). 372  373 Figure13a)Extractionofcores;b)Numerationofextractedspecimens 374 Figure 14 depicts the distribution of fibres in the drilled cores. The average content of fibres along 375 the principal yield lines resulted to be 67.4 kg/m3 with a CV of 14.8%. This mean fibre content 376 and CV allows confirming that the fiber distribution is homogenous from the design point of view, 377 especially for this type of structures. However, the lack of homogeneity over the slab depth should 378 be pointed out: the average fibre content of the upper halves of drilled specimens (Figure 5) was 379 44.9 kg/m3, whereas the lower halves presented 89.5 kg/m3 in average – almost more than twice 380 the above value. 381 The tendency for an increase of fibre percentage from the top to the bottom of SFRC slab due to 382 presence of vibration was depicted in previous studies [48]. Nevertheless, in this experimental 383 programme no vibration was applied and a considerable variation of fibre content over depth was 384 also observed. This phenomenon may lead to development of insufficiently controlled cracks in 385 the zones with significant negative moments; therefore, the hybrid solutions with certain amount 386 a) b) 20  of conventional reinforcement and correspondently reduced fibre content could be an attractive 387 alternative for this type of structures in terms of SLS requirements. 388 389 Figure 14 Fibre distribution in extracted specimens 390 Evaluating the contribution of fibres along the direction of the main axes, a general trend for this 391 type of structural elements was acknowledged: the preferential fibre orientation was perpendicular 392 to the casting direction (Z axis) due to unrestricted radial flow of concrete along with the absence 393 of vibration during the casting process. The average fibre contribution of 76.8% in the horizontal 394 plane was detected by means of the inductive test; hence, this considerable percentage of fibres 395 enhanced the post-cracking flexural strength of SFRC prototype – the essential factor for elevated 396 flat slabs. 397 398 Figure 15 Fibre contribution in the horizontal plane 399 Figure 15 shows that both main directions represent at least 30% of fibre contribution in most of 400 the extracted cores (62.5%) and, therefore, no preferential orientation of the fibres along the slab 401 Theoretical Fibre Content 20 40 60 80 100 120 1...5 ..10..15..20..25..30.1...5 ..10..15..20..25..30. Number of Specimen Number of Specimen Fibre Content [kg/m3] Entire Specimen Upper Half Lower Half 20% 30% 40% 50% 60% 20% 30% 40% 50% 60% Fibre contribution along Y axis Fibre contribution along X axis Lower Part Upper Part 21  axis can be confirmed. A preferential orientation could be more noticeable nearby the formworks 402 due to geometrical boundary condition that those imposed to the concrete flow during the pouring. 403 4. Conclusions 404 In this paper, the structural response of a full-scale column-supported SFRC flat slab was 405 described and analysed in terms of structural integrity, cracking and deflections. Five different 406 uniformly-distributed load phases were applied in contrast to the previous experiences, in which 407 point loading was mainly used. Thereafter, 32 cores were drilled from the tested SFRC flat slab 408 in order to assess both fibre content and orientation. The following conclusions may be derived 409 from the obtained results: 410  A SFRC with 𝑓 = 7.2 N/mm2 (CoV = 35.3%) and 𝑓 = 7.7 N/mm2 (CoV = 24.7%) 411 proved to be sufficient to guarantee a suitable structural response for loads representative 412 of residential buildings in a column-supported slab with a span-to-depth ratio of 30. 413  A load increase from the characteristic load combination (𝑞 = 9.8 kN/m2) to 16.0 kN/m2 414 (14% superior to the design load combination, 𝑞 = 14.0 kN/m2) led to an increment of 415 deflections of 31 mm (𝐿250 ), without evidencing signs of failure; this proving both the 416 moment redistribution capacity and the ductility of the system. 417  The performance in terms of cracking and deflections resulted to be acceptable even 418 when the structural system was subjected to loads (q = 8.7 KN/m2) superior in 13% with 419 respect to the quasi-permanent load combination (𝑞, = 7.7 kN/m 2) for residential 420 buildings. 421  The results derived from performing the inductive test (non-destructive test oriented to 422 characterize distribution and amount of steel fibres) on drilled cores confirmed that a 423 76.8% were favourably oriented along the slab in-plane axis. 424  A higher amount of fibres (double in average) was detected in the middle-bottom part of 425 the slab height respect to the upper half (even without vibration). Although this issue did 426 not compromise the structural response of the slab, a steel mesh placed at the top side 427 of the slab at the vicinities of the columns is recommendable. 428 22  The results obtained in this experimental program represent a new evidence of the structural 429 suitability of the SFRC in elevated flat slabs, even with total substitution of the longitudinal steel 430 bars. Nevertheless, as any incipient technology (or considerable change in the design and safety 431 implications), certain aspects are still to be addressed for the acceptance of structural fibres as 432 unique concrete reinforcement in elevated flat slabs, and its effective implementation. In this 433 regard, monitoring time-dependent deformations for larger periods than that considered in this 434 experimental program, the structural effect of induced holes into the slab, the hybrid use of fibres 435 and steel mesh for optimization purposes are topics that should be further investigated. 436 Acknowledgements 437 The authors wish to express their gratitude to the Spanish Ministry of Economy, Industry and 438 Competitiveness (MINECO) for the financial support in the scope of the project eFIB (reference: 439 RTC-2016-5263-5) which was carried out along with SACYR Ingeniería e Infraestructura. The 440 authors also appreciate valuable input of Nikola Tošić into the manuscript. The first author, 441 personally, thanks the Department of Enterprise and Education of Catalan Government for 442 providing support through the PhD Industrial Fellowship (2018 DI 77) in collaboration with Smart 443 Engineering Ltd. (UPC’s Spin-Off). 444 REFERENCES 445 [1] Zollo RF. Fiber-reinforced Concrete : an Overview after 30 Years of Development. 446 Cem Concr Compos 1997;19:107–22. 447 [2] Liao L, de la Fuente A, Cavalaro S, Aguado A. Design of FRC tunnel segments 448 considering the ductility requirements of the Model Code 2010. Tunn Undergr Sp 449 Technol Inc Trenchless Technol Res 2015;47:200–10. 450 https://doi.org/10.1016/j.tust.2015.01.006. 451 [3] Pujadas P, Blanco A, de La Fuente A, Aguado A. Cracking behavior of FRC slabs 452 with traditional reinforcement. Mater Struct Constr 2012;45:707–25. 453 https://doi.org/10.1617/s11527-011-9791-0. 454 [4] Groli G, Pérez A, Marchetto F, Aríñez F. Serviceability performance of FRC 455 columns under imposed displacements : An experimental study. Eng Struct 456 2015;101:450–64. https://doi.org/10.1016/j.engstruct.2015.07.035. 457 [5] Vasanelli E, Micelli F, Aiello MA, Plizzari G. Crack width prediction of FRC beams 458 in short and long term bending condition. Mater Struct Constr 2014;47:39–54. 459 https://doi.org/10.1617/s11527-013-0043-3. 460 [6] Cachim PB, Figueiras JA, Pereira PAA. Fatigue behavior of fiber-reinforced 461 concrete in compression 2002;24:211–7. 462 23  [7] Carlesso DM, de la Fuente A, Cavalaro SHP. Fatigue of cracked high performance 463 fiber reinforced concrete subjected to bending. Constr Build Mater 2019;220:444–464 55. https://doi.org/10.1016/j.conbuildmat.2019.06.038. 465 [8] Nia AA, Hedayatian M, Nili M, Sabet VA. International Journal of Impact 466 Engineering An experimental and numerical study on how steel and polypropylene 467 fi bers affect the impact resistance in fi ber-reinforced concrete. Int J Impact Eng 468 2012;46:62–73. https://doi.org/10.1016/j.ijimpeng.2012.01.009. 469 [9] Teng T, Chu Y-A, Chang F-A, Shen B-C, Cheng D-S. Development and validation 470 of numerical model of steel fiber reinforced concrete for high-velocity impact. 471 Comput Mater Sci 2008;42:90–9. 472 https://doi.org/10.1016/j.commatsci.2007.06.013. 473 [10] Serrano R, Cobo A, Prieto MI, De M, González N. Analysis of fire resistance of 474 concrete with polypropylene or steel fibers 2016;122:302–9. 475 https://doi.org/10.1016/j.conbuildmat.2016.06.055. 476 [11] fib. fib Model Code for Concrete Structures 2010. 2010. 477 [12] Belletti B, Cerioni R, Meda A, Plizzari G. Design Aspects on Steel Fiber-478 Reinforced Concrete Pavements 2009;20:599–607. 479 https://doi.org/10.1061/(ASCE)0899-1561(2008)20. 480 [13] Alani AM, Beckett D. Mechanical properties of a large scale synthetic fibre 481 reinforced concrete ground slab. Constr Build Mater 2013;41:335–44. 482 https://doi.org/10.1016/j.conbuildmat.2012.11.043. 483 [14] Meda A, Plizzari GA, Riva P. Fracture behavior of SFRC slabs on grade. Mater 484 Struct Constr 2004;37:405–11. https://doi.org/10.1617/14093. 485 [15] Jamshidi M, Hoseini A, Vahdani S, Santos C De, De A. Seismic fragility curves for 486 vulnerability assessment of steel fiber reinforced concrete segmental tunnel 487 linings. Tunn Undergr Sp Technol 2018;78:259–74. 488 https://doi.org/10.1016/j.tust.2018.04.032. 489 [16] de la Fuente A, Pujadas P, Blanco A, Aguado A. Experiences in Barcelona with 490 the use of fibres in segmental linings. Tunn Undergr Sp Technol 2012;27:60–71. 491 https://doi.org/10.1016/j.tust.2011.07.001. 492 [17] Chiaia B, Fantilli AP, Vallini P. Combining fiber-reinforced concrete with traditional 493 reinforcement in tunnel linings. Eng Struct 2009;31:1600–6. 494 https://doi.org/10.1016/j.engstruct.2009.02.037. 495 [18] de la Fuente A, Aguado A, Molins C, Armengou J. Innovations on components 496 and testing for precast panels to be used in reinforced earth retaining walls. Constr 497 Build Mater 2011;25:2198–205. 498 https://doi.org/10.1016/j.conbuildmat.2010.11.003. 499 [19] de La Fuente A, Escariz RC, de Figueiredo AD, Aguado A. Design of macro-500 synthetic fibre reinforced concrete pipes. Constr Build Mater 2013;43:523–32. 501 https://doi.org/10.1016/j.conbuildmat.2013.02.036. 502 [20] Gossla U. Development of SFRC Free Suspended Elevated Flat Slabs 2005. 503 [21] Destrée X, Mandl J. Steel fibre only reinforced concrete in free suspended 504 elevated slabs: Case studies, design assisted by testing route, comparison to the 505 latest SFRC standard documents. Tailor Made Concr Struct 2008:437–43. 506 24  [22] di Prisco M, Colombo M, Pourzarabi A. Biaxial bending of SFRC slabs: Is 507 conventional reinforcement necessary? Mater Struct Constr 2019;52:1–15. 508 https://doi.org/10.1617/s11527-018-1302-0. 509 [23] Fall D, Shu J, Rempling R, Lundgren K. Two-way slabs : Experimental 510 investigation of load redistributions in steel fibre reinforced concrete 2014. 511 https://doi.org/10.1016/j.engstruct.2014.08.033. 512 [24] Pujadas P, Blanco A, Cavalaro S, Aguado A. Plastic fibres as the only 513 reinforcement for flat suspended slabs: Experimental investigation and numerical 514 simulation. Constr Build Mater 2014;57:92–104. 515 https://doi.org/10.1016/j.conbuildmat.2014.01.082. 516 [25] Facconi L, Minelli F, Plizzari G. Steel fiber reinforced self-compacting concrete 517 thin slabs – Experimental study and verification against Model Code 2010 518 provisions. Eng Struct 2016;122:226–37. 519 https://doi.org/10.1016/j.engstruct.2016.04.030. 520 [26] Hedebratt J, Silfwerbrand J. Full-scale test of a pile supported steel fibre concrete 521 slab. Mater Struct Constr 2014;47:647–66. https://doi.org/10.1617/s11527-013-522 0086-5. 523 [27] Tan KH, Venkateshwaran A. Punching shear in steel fiber-reinforced concrete 524 slabs with or without traditional reinforcement. ACI Struct J 2019;116:107–18. 525 https://doi.org/10.14359/51713291. 526 [28] Michels J, Waldmann D, Maas S, Zürbes A. Steel fibers as only reinforcement for 527 flat slab construction - Experimental investigation and design. Constr Build Mater 528 2012;26:145–55. https://doi.org/10.1016/j.conbuildmat.2011.06.004. 529 [29] Gouveia ND, Lapi M, Orlando M, Faria DM V. Experimental and theoretical 530 evaluation of punching strength of steel fiber reinforced concrete slabs 2017:13. 531 [30] Parmentier B, Itterbeeck P Van. The flexural behaviour of SFRC flat slabs : the 532 Limelette fullscale experiments for supporting design model codes 2015. 533 [31] Døssland ÅL. Fibre Reinforcement in Load Carrying Concrete Structures. 2008. 534 [32] Mandl J. Flat Slabs Made of Steel Fiber Reinforced Concrete (SFRC) 2008. 535 [33] Karv C. Shear and punching resistance of steel fibre reinforced concrete slabs. 536 Aalto University, 2017. 537 [34] Maturana A, Canales J, Orbe A, Cuadrado J, Cuadrado J. Análisis plástico y 538 Ensayos de Losas multidireccionales de HRFA. Inf La Construcción 539 2014;66:e031. https://doi.org/10.3989/ic.13.021. 540 [35] Maturana Orellana A. Estudio teórico-experimental de la aplicabilidad del 541 hormigón reforzado con fibras de acero a losas de forjado multidireccionales. 542 2013. https://doi.org/10.1174/021435502753511268. 543 [36] Mitchell D, Cook WD. Preventing Progressive Collapse of Slab Structures. J Struct 544 Eng 1984;110:1513–32. https://doi.org/10.1061/(ASCE)0733-545 9445(1984)110:7(1513). 546 [37] Blanco A, Pujadas P, Fuente A De, Cavalaro SHP, Aguado A. Assessment of the 547 fibre orientation factor in SFRC slabs. Compos PART B 2015;68:343–54. 548 https://doi.org/10.1016/j.compositesb.2014.09.001. 549 25  [38] Mudadu A, Tiberti G, Germano F, Plizzari GA, Morbi A. The effect of fiber 550 orientation on the post-cracking behavior of steel fiber reinforced concrete under 551 bending and uniaxial tensile tests. Cem Concr Compos 2018;93:274–88. 552 https://doi.org/10.1016/j.cemconcomp.2018.07.012. 553 [39] di Prisco M, Martinelli P, Dozio D. The structural redistribution coefficient KRd: A 554 numerical approach to its evaluation. 2016. 555 https://doi.org/10.1002/suco.201500118. 556 [40] Colombo M, Martinelli P, di Prisco M. On the evaluation of the structural 557 redistribution factor in FRC design: a yield line approach. Mater Struct 2017;50:1–558 18. https://doi.org/10.1617/s11527-016-0969-3. 559 [41] Tan H, Saha MK. Ten-Year Study on Steel Fiber-Reinforced Concrete Beams 560 Under Sustained Loads 2006. 561 [42] Monetti DH, Llano-torre A, Torrijos MC, Giaccio G, Zerbino R, Martí-vargas JR, et 562 al. Long-term behavior of cracked fiber reinforced concrete under service 563 conditions. Constr Build Mater 2019;196:649–58. 564 https://doi.org/10.1016/j.conbuildmat.2018.10.230. 565 [43] de la Fuente A, Casanovas-Rubio MDM, Pons O, Armengou J. Sustainability of 566 Column-Supported RC Slabs: Fiber Reinforcement as an Alternative. J Constr 567 Eng Manag 2019;145:1–12. https://doi.org/10.1061/(ASCE)CO.1943-568 7862.0001667. 569 [44] Galeote E, Blanco A, Cavalaro SHP, de la Fuente A. Correlation between the 570 Barcelona test and the bending test in fibre reinforced concrete. Constr Build 571 Mater 2017;152:529–38. https://doi.org/10.1016/j.conbuildmat.2017.07.028. 572 [45] Pujadas P, Blanco A, Cavalaro SHP, de la Fuente A, Aguado A. Multidirectional 573 double punch test to assess the post-cracking behaviour and fibre orientation of 574 FRC. Constr Build Mater 2014;58:214–24. 575 https://doi.org/10.1016/j.conbuildmat.2014.02.023. 576 [46] Blanco A, Pujadas P, de la Fuente A, Cavalaro SHP, Aguado A. Influence of the 577 type of fiber on the structural response and design of FRC slabs. J Struct Eng 578 (United States) 2016;142:1–11. https://doi.org/10.1061/(ASCE)ST.1943-579 541X.0001515. 580 [47] Fall D, Shu J, Rempling R, Lundgren K, Zandi K. Two-way slabs: Experimental 581 investigation of load redistributions in steel fibre reinforced concrete. 2014. 582 https://doi.org/10.1016/j.engstruct.2014.08.033. 583 [48] Barros J, Salehian H, Pires M, Gonçalves D. Design and testing elevated steel 584 fibre reinforced self-compacting concrete slabs. Fibre Reinf Concr 2012:1–12. 585 [49] Ošlejs J. New Frontiers for Steel Concrete. Concr Int 2008;3:45–50. 586 [50] CEN. EN 12390-13:2013. Testing hardened concrete. Determination of secant 587 modulus of elasticity in compression. 2013. 588 [51] CEN. EN 12390-3:2019. Testing hardened concrete. Compressive strength of test 589 specimens. 2019. 590 [52] CEN. EN 14651:2007. Test method for metallic fibre concrete. Measuring the 591 flexural tensile strength (limit of proportionality (LOP), residual). 2007. 592