FIB 95 FRC From Design To Structural Applications [PDF]

Zitiervorschau

ACI – fib – RILEM Joint Workshop

3rd FRC International Workshop Fibre Reinforced Concrete: from Design to Structural Applications 28-30 June 2018, Desenzano, Lake Garda, Italy Editors: B. Massicotte, F. Minelli, B. Mobasher, G. Plizzari

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

3rd FRC International Workshop Fibre Reinforced Concrete: from Design to Structural Applications

28-30 June 2018, Desenzano, Lake Garda, Italy

Editors: B. Massicotte, F. Minelli, B. Mobasher, G. Plizzari

i

Copyright 2018 Cartolibreria Snoopy s.n.c. ISBN 978-88-89252-44-4 Printed on June, 2018

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Preface The first international FRC workshop supported by RILEM and ACI was held in Bergamo (Italy) in 2004. At that time, lack of specific building codes and standards was identified as the main inhibitor to the application of this technology in engineering practice. The workshop aim was placed on identification of applications, guidelines, and research needs in order for this advanced technology to be transferred to professional practice. The second international FRC workshop, held in Montreal (Canada) in 2014, was the first ACI-fib joint technical event. Many of the objectives identified in 2004 had been achieved by various groups of researchers who shared a common interest in extending the application of FRC materials into the realm of structural engineering and design. The aim of the workshop was to provide the State-of-the-Art on the recent progress that had been made in term of specifications and actual applications for building and bridge projects in Europe and NorthAmerica. The rapid development of codes, the introduction of new materials and the growing interest of the construction industry suggest presenting this forum at closer intervals. In this context the third international FRC workshop is held in Desenzano (Italy), four years after Montreal. In this first ACI-fib-RILEM joint technical event, the maturity gained through the recent technological developments and large-scale applications will be used to show the acceptability of the concrete design using various fiber compositions. The growing interests of civil infrastructure owners in ultrahigh-performance fibre-reinforced concrete (UHPFR or UHPC?) and synthetic fibres in structural applications bring new challenges in terms of concrete technology and design recommendations. In such a short period of time we have witnessed the proliferation of the use of fibers as structural reinforcement in various applications such as industrial slabs, elevated slabs, precast tunnel lining sections, foundations, as well bridge decks. We are now moving towards addressing many durability based design requirements by the use of fibers, as well as the general serviceability based design. In that perspective, the aim of FRC2018 workshop is to provide the State-of-the-Art on the recent progress in term of specifications development, actual applications, and to expose users and researchers to tomorrow's challenges in the design and construction of a wide variety of structural applications. We appreciate the enormous support we have received from all three sponsoring organisations and look forward to paving the path for future collaborations in various areas of common interest so that the developmental work and implementation of new specifications and design procedures can be expedited internationally. June, 2018 Bruno Massicotte Fausto Minelli Barzin Mobasher Giovanni Plizzari

iii

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Dear Workshop Participants, On behalf of the ACI Board of Direction, Technical Activities Committee, and the entire ACI membership we welcome you to the ACI-fib-RILEM Workshop on Fibre Reinforced Concrete: from Design to Structural Applications. We are pleased that the American Concrete Institute worked together to co-sponsor this important technical workshop. The knowledge and expertise developed and disseminated at this workshop by experts such as yourselves is critical to advancing the concrete industry worldwide, to better utilizing resources, and to constructing a more sustainable environment. To capture information presented at the workshop, ACI-fib-RILEM are working collaboratively to review and publish all the presented papers. ACI will make the papers available in digital format. The leadership of ACI-fib-RILEM are confident that this workshop will set the groundwork for fruitful technical collaboration between our three groups and allow each group to accomplish more than it could working independently. During the workshop we hope each of you have the time to not only exchange technical information, but also to get to better know your peers from around the world. Wishing you all a most successful workshop.

David A. Lange, Ph.D., FACI President Executive

iv

Ronald G. Burg, P.E., FACI Vice President

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

It is with great pleasure that I welcome you, on behalf of fib, to the 3rd International Workshop on Fibre Reinforced Concrete, this time devoted to Design and Structural Applications, to be held for the second time at the University of Brescia. It will be for the second time a joint ACI-fib conference and, in addition, for the first time RILEM will join the organization of the conference. This kind of fruitful collaboration is the kind of actions fib likes to promote and to develop to the benefit of the global concrete community. FRC is a more mature field today. The fib has already produced, in its Model Code 2010, a good set of design guidelines. The fib is now in the process of preparing the new version of the Model Code, the Model Code 2020, and its goal is to produce a more coherent treatment of the different types of FRC’s and to try to adapt the models, we are currently using for normal concretes, to FRC. We sincerely hope that this Workshop can contribute to our goals and also can contribute in bringing together common approaches along the globe. I must thank the organizers for the initiative and to the hosts at the University of Brescia and wish all of you all a successful and enjoyable Workshop.

Hugo Corres Peiretti President of fib

v

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Dear Workshop Participants: Greetings from RILEM! It is a pleasure to see the series of International Workshops on Fibre Reinforced Concrete: From Design to Structural Applications continue with significant and continued interest from the scientific community and practising engineers. The first workshop in this series was held as a RILEM Workshop in Bergamo about 14 years back. This followed a landmark RILEM Symposium on Fibre Reinforced Concrete (BEFIB 2004) held in Varenna with a large international participation. This workshop in Brescia addresses the very important translational research that is essential for taking the developments made in the fundamental level to practice and implementation. It is of utmost need that technology implementation and design guidelines be undertaken so that the benefits of research performed over many decades can reach the society. RILEM is glad to support this and other activities in the area of fibre reinforced concretes, as in the past, and we hope to have many more such events in the future. I wish the event all success and thank the organizers for their exemplary effort.

Ravindra Gettu President-Elect, RILEM

vi

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Committees and local organiser Conference Chairman Bruno MASSICOTTE (Polytechinique Montreal, Canada) Fausto MINELLI (University of Brescia, Italy) Barzin MOBASHER (Arizona State University, USA) Giovanni PLIZZARI (University of Brescia, Italy) International Advisory Board György L. BALÁZS (Budapest University of Technology, Hungary) Arnon BENTUR (Technion – Israel institute of Technology, Israel) Andrzej BRANDT (Polish Academy of Sciences, Poland) Horst FALKNER (Technical University of Braunschweig, Germany) Hirozo MIHASHI (Tohoku University, Japan) Sidney MINDESS (University of British of Columbia, Canada) Antoine NAAMAN (University of Michigan, USA) Keitetsu ROKUGO (Gifu University, Japan) Surendra P. SHAH (Northwestern University, USA) Frank VECCHIO (University of Toronto, Canada) Joost WALRAVEN (Delf University of Technology, The Netherlands) Scientific Committee Antonio AGUADO (Spain) Maria Antonietta AIELLO (Italy) Mohammad AIHAMAYDEH (United Arab Emirates) Nemkumar BANTHIA (Canada) Joaquim BARROS (Portugal) Andrea BELLERI (Italy) Stefan BERNARD (Australia) Nicola BURATTI (Italy) Giuseppe CAMPIONE (Italy) Jean Philippe CHARRON (Canada) Matteo COLOMBO (Italy) Valeria CORINALDESI (Italy) Albert DE LA FUENTE (Spain) Frank DEHN (Germany) Marco DI PRISCO (Italy) vii

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Alessandro FANTILLI (Italy) Liberato FERRARA (Italy) Stephen FOSTER (Australia) Ravindra GETTU (India) Juan Navarro GREGORI (Spain) Steffen GRÜNEWALD (The Netherlands) Terje KANSTAD (Norway) Wolfgang KUSTERLE (Germany) Christopher LEUNG (Hong Kong) Karin LUNDGREN (Sweden) Peter MARK (Germany) Enzo MARTINELLI (Italy) Viktor MECHTCHERINE (Germany) Alberto MEDA (Italy) Günter MESCHKE (Germany) Aurelio MUTTONI (Switzerland) Benoit PARMENTIER (Belgium) Gustavo PARRA MONTESINOS (USA) Sergio PIALARISSI (Spain) Dario REDAELLI (Switzerland) Jacques RESPLENDINO (France) Zila RINALDI (Italy) Pierre ROSSI (France) Pedro SERNA (Spain) Luca SORELLI (Canada) Giuseppe TIBERTI (Italy) Romildo Dias TOLEDO FILHO (Brazil) Gideon P.A.G. VAN ZIJL (South Africa) Lucie VANDEWALLE (Belgium) Cristina ZANOTTI (Canada) Raul ZERBINO (Argentina) LI Zongjin (China) Local Organizing committee Antonio CONFORTI Luca FACCONI Fabiola IAVARONE

viii

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Sara LUCCHINI Giovanni METELLI Fausto MINELLI Antonio MUDADU Mónica Yolanda OÑA VERA Anthony PADERNO Giovanni PLIZZARI Marco PRETI Adriano REGGIA Giuseppe TIBERTI Ivan TRABUCCHI

ix

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Acknowledgement The chairs would like to thank the many people who contributed to organise the ACI-fibRILEM FRC2018 Workshop and the authors who contributed to share their knowledge. The Chairs would like to express their gratitude to the members of scientific community the reviewers whose help has ensured the high standards of the papers. The support of the sponsors has had a positive impact on the organisation. Finally the contribution of the local organising committee, graduate students and staff has been an essential component in the organisation of the workshop.

June, 2018 Bruno Massicotte Fausto Minelli Barzin Mobasher Giovanni Plizzari

x

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Contents Preface ...................................................................................................................................... iii Committees and local organiser............................................................................................ vii Acknowledgement .................................................................................................................... x Contents ................................................................................................................................... xi KEYNOTES ............................................................................................................................. 1 Twenty years experience on FRC structural precast elements: a gym for national and international standards ............................................................................................................... 2 Fiber-Reinforced Concrete - From Fresh Properties to Structural Design: New Tools, Guides, and Reports from ACI Technical Committee 544 ..................................................................... 4 High performance fibre reinforced concrete for structural applications .................................... 7 Material characterization and in-situ control of FRC ................................................................ 9 Recent developments in FRC tunnel linings ............................................................................ 12 Steel Fiber Reinforced Concrete For Structures Subjected To Severe Actions ....................... 14 SESSION: Mechanical characterisation .............................................................................. 17 Round-Robin Test on Various Test-Methods for Flexural Behavior of Steel Fiber Reinforced Sprayed Concretes.................................................................................................................... 18 Comparison of different methods of inverse analysis to assess the tensile law of UHPFRC from four points bending tests .......................................................................................................... 20 A new multi-scale hybrid fibre reinforced cement-based composites ..................................... 22 A sectional approach for the bending creep of FRC based on uniaxial tension creep tests..... 24 Optimized quality control procedure with the Barcelona test and the inductive method for FRC .................................................................................................................................................. 26 Influence of the beam size on the residual strength of fibre-reinforced concrete .................... 28 Mechanical properties of self-compacting concrete reinforced with hybrid mixes of hooked steel fibers of variable lengths ................................................................................................. 30 Influence of test methodology on the applicability of test results of fibre reinforced concrete for design ....................................................................................................................................... 32

xi

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Parameters affecting the properties of plain and fiber reinforced Self-Consolidating concrete .................................................................................................................................................. 34 Numerical Validation of a Simplified Inverse Analysis Method to Characterise the Tensile Behaviour of UHPFRC............................................................................................................. 36 Optimization of fibre combination using latest generation of steel and polyolefin fibres ....... 38 An application oriented state-of-art and research-need perspective on self-healing fibrereinforced cementitious composites ......................................................................................... 40 Early-age behaviour of High Performance Fiber Reinforced Concretes .................................. 42 Experimental and numerical analysis of fiber reinforced concrete beams in real scale with steel bars reinforcement .................................................................................................................... 44 An experimental investigation on the post-cracking behaviour of Recycled Steel Fibre Reinforced Concrete ................................................................................................................. 46 SESSION: Durability ............................................................................................................. 49 Influence of chloride corrosion on the surface aspect of steel fibre reinforced cementitous composites ................................................................................................................................ 50 The effect of fibres on corrosion of RC elements .................................................................... 52 Influence of the Post-Cracking Residual Strength Variability on the Partial Safety Factor .... 54 Comparability of bond tests for repair and retrofit of concrete structures with Fiber Reinforced Concrete .................................................................................................................................... 56 Self-monitoring of cracking development of fiber reinforced conductive concrete subjected to bending ..................................................................................................................................... 58 SESSION: Bridges/Elevated slabs ........................................................................................ 59 Hybrid Fiber Reinforced Concrete for the Application of Bridge Deck .................................. 60 Performance under fatigue loading of field-cast UHPFRC joints in positive moment regions between precast bridge deck panels.......................................................................................... 62 Design of precast prestressed SFRC T-girders for accelerated sustainable bridge construction .................................................................................................................................................. 64 SF+RC modelling using DIANA FEA ..................................................................................... 66 Steel Fibre Only Reinforced Concrete in Suspended Slabs: Design From Academical to Practical Methods ..................................................................................................................... 68 Effect of reinforcement configuration on the ductility requirements of real-scale slabs ......... 70

xii

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A new method for designing FRC elevated slabs according to Model Code 2010 ................. 72 Experimental and numerical comparative analysis of a SFRC slab on piles ........................... 74 FRC Hybrid slabs: reliability of Model Code approaches ....................................................... 76 Optimization of partially prefabricated HyFRC slabs ............................................................. 78 SESSION: Shear .................................................................................................................... 81 Experimental study of shear transfer in polypropylene fibre-reinforced concrete using precracked push-off specimens ..................................................................................................... 82 Shear crack behaviour and shear deformation of polypropylene fibre-reinforced concrete slender beams ........................................................................................................................... 84 Enhancements in Fracture Behavior and Shear capacity in Hybrid Steel and MacroPolypropylene Fiber reinforced concrete ................................................................................. 86 Experimental investigation of shear-critical prestressed steel fibre reinforced concrete beams .................................................................................................................................................. 88 Fibres as shear reinforcement in RC beams: an overview on assessment of material properties and design approaches ............................................................................................................. 90 SESSION: Seismic/Special loading conditions .................................................................... 93 Performance of Ductile FRCC under Cyclic Loads and Non-Linear FE Simulation .............. 94 Effect of synthetic fibers on the quasi-static and blast behaviour of reinforced concrete beams .................................................................................................................................................. 96 Strain-Based Fatigue Failure Criterion for Steel-Fiber Reinforced Concrete.......................... 98 Cyclic Damage on PVA Microfibre Embedded in Cementitious Matrix in Alternating TensionCompression Regime ............................................................................................................. 100 Seismic Performance of Fibre Reinforced Concrete in the Absence of Bars ........................ 102 Finite Element Analyses of Seismic Response of a 22-story RC Wall Building subjected to Drying Shrinkage Cracking and Application of SCRPCC .................................................... 104 Residual crack width in RC and R/FRC ties subjected to repeated loads .............................. 106 A method for strengthening and thermal-insulating cavity walls with Fiber Reinforced Mortar: the case of Groningen ............................................................................................................ 108 Flexural fatigue performance of plastic fiber and steel microfiber in reinforced concrete .... 110 SESSION: Structural rehabilitation .................................................................................. 111

xiii

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

The behavior of concrete columns confined by UHP-FRCC jacketing ................................. 112 Retrofitting a full-scale two-story hollow clay block masonry building by Steel Fiber Reinforced Mortar coating...................................................................................................... 114 Working Life Extension of RC Bridge Pier Through UHPFRC Jacketing ............................ 116 Improved Ductility Of SHCC Retrofitted Unreinforced Load Bearing Masonry Seismic Resistance ............................................................................................................................... 118 SESSION: Precast elements ................................................................................................ 121 New Concretes for Precast Panels in Uruguay ....................................................................... 122 Opportunities for synthetic fibre reinforcement in concrete tramlines................................... 124 Optimized reinforcement and performance of precast elements using polypropylene macro fibers ....................................................................................................................................... 126 Closed loop control in crushing test for fibre reinforced concrete pipes................................ 128 RC Beams with Steel Fibres - Towards Better Determination of their Minimum Conventional Reinforcement Ratio ............................................................................................................... 130 SESSION: Tunnel linings .................................................................................................... 133 Santoña–Laredo General Interceptor Collector – Challenges and Solutions ......................... 134 Segmental Lining Design using Macro Synthetic Fibre Reinforcement ................................ 136 An experimental study on the use of polypropylene fibers in precast segments for hydraulic and metro tunnel lining ................................................................................................................. 138 Design of SFRC Precast Tunnel Segments Supported by NLFEA ........................................ 140 Closed-form Solutions for Interaction Diagrams of Hybrid Fiber-Reinforced Tunnel Segments ................................................................................................................................................ 142 Steel and Polypropylene Fiber Reinforced Concrete for Secondary Tunnel Lining .............. 144 Experimental Behaviour of Precast Tunnel Segments in Steel Fiber Reinforcement with GFRP Rebars ..................................................................................................................................... 146 Structural behavior of precast tunnel segments with macro-synthetic fibers during TBM operations: a numerical study ................................................................................................. 148 SESSION: High Performance FRC .................................................................................... 151 Structural behavior of prestressed Ultra-High Performance Fibre-Reinforced Concrete beams with and without openings: comparison between experimental results and finite element modelling techniques .............................................................................................................. 152 xiv

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Use of a Probabilistic Explicit Cracking Model for Analyzing the mechanical behaviour of an Ultra-High Performance Fibre Reinforced Concrete (UHPFRC) Beam Subjected to Shear Loading .................................................................................................................................. 154 UHPFRC Precast slabs and field-cast joints of the Isabey-Darnley pedestrian bridge ......... 156 Fibre effect on shear behaviour of UHPC beams .................................................................. 158 Practical considerations on the application of the recent SIA 2052 guidelines on testing of Ultrahigh-performance fiber-reinforced concrete .......................................................................... 160 Effect of Fibers on the Flexural Behaviour of Beams Built with High-Strength Concrete and High-Strength Reinforcement ................................................................................................ 162 Mechanical characterization of fiber reinforced floor screeds: Influence of glass fibers on shrinkage and cracking mechanisms ...................................................................................... 164 SESSION: Service conditions ............................................................................................. 167 Evaluation of Effective Moment of Inertia for Calculation of Short-Term Deflections of Steel Fiber Reinforced Concrete Flexural Members....................................................................... 168 Reduction of Flexural Crack Widths Using Synthetic Fibres in Reinforced Concrete ......... 170 Serviceability Limit State Design of SFRC Members ........................................................... 172 Effect Of Fiber Dosage And Matrix Compressive Strength On MSFRC Performance ........ 174 Predictions of the Micro-crack Openings for Ultra High Performance Fiber Reinforced Concrete ................................................................................................................................. 176 Macrosynthetic fibers for end region crack control ............................................................... 178 Investigation of fiber effect on the geometrical property of crack and water permeability of cracked concrete ..................................................................................................................... 180 SESSION: Structural applications ..................................................................................... 183 Experimental Investigation of Hybrid Concrete Elements with Varying Fiber Reinforcement under Concentrated Load ....................................................................................................... 184 A holistic calculation and design tool for structural SFRC members .................................... 186 FRCcalc - Software for design of fiber reinforced concrete elements according to MC2010 recommendations ................................................................................................................... 188

xv

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

xvi

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

KEYNOTES

1

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Twenty years experience on FRC structural precast elements: a gym for national and international standards Marco Di Prisco1 , Claudio Failla2 , Giovanni Plizzari3 : University “Politecnico di Milano”, Milano, Italy: : Magnetti Building S.p.A., Bergamo, Italy; 3 : University of Brescia, Brescia, Italy. 1 2

Abstract The research activity concerning the structural application of fiber-reinforced concrete (FRC) in the precast industry began at Magnetti Building since the late 1990s, when preliminary investigations about the typologies of fibers used in concrete were carried out. Considering the results achieved at that time, the use of steel fibers was considered the most promising application; therefore, an experimentation program for the composition of steel fiber reinforced concrete (SFRC) was implemented. Basing on these first promising results, the patent “Use of fibers in concrete compositions for the production of structural elements in prestressed concrete and relevant reinforced elements” (1997), which was an input for the following indepth analysis of the material and its structural applications, was granted. The most relevant studies on application were carried out thanks to the research contract obtained by M.U.R.S.T. (Ministry of Universities and Scientific and Technological Research) concerning "Structural behavior, durability and fire behavior of prestressed fiber reinforced concrete elements" a 4-year program (1998-2002). This project allowed other institutions to be involved in the research, including the Polytechnic of Milan and the University of Brescia, that carried out an extensive experimental program on different structural precast elements. In parallel to the definition and optimization of different typologies of FRC, the first structural elements was realized with the goal of achieving the appropriate rheological and mechanical characteristics. In addition, experimental tests were carried out to gather information about the durability and fire resistance of SFRC. After the application on “standard” elements, a new roof profile was properly designed for this innovative material [Fig. 1]. The geometry, thicknesses and dimensions of this profile have been identified following several analyses and after having made more than 30 prototypes with different materials and construction details.

Figure 1:

structural roof elements in SFRC.

The “NGPL” element represented the first profile designed for structural applications of a SFRC prestressed element for roof components. It was realized placing transversal reinforcement only in the area closest to the extremities of the element, where prestressing diffuses and support reactions are located. The study of this roof element was carried out by

2

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

creating a formwork for casting a traditional reinforced concrete prototype [Fig. 2] and a SFRC element [Fig. 3]. In the FRC element, primary stresses (longitudinal) are controlled by the main reinforcement while the secondary ones (transverse) by SFRC.

Figure 2: Formwork for the construction of the NGPL element made by reinforced concrete.

Figure 3: Formwork for the construction of the NGPL element made by SFRC.

The research activity carried out was relevant to the development of two important standards: the UNI 11188 “Design, execution and control of structural elements in reinforced concrete with steel fibers” and the CNR-DT 204/2006 “Instructions for the Design, Execution and Control of Fiber Reinforced Concrete Structures”. Although good results were achieved in terms of bearing capacity of the structures, the lack of a national code restricted the use of this technology in the structural field; however, some buildings were built with FRC. Another application concerned the precast panels used for building façades. The SFRC panels were designed and tested up to failure [Fig. 4]. Nowadays, panels are still produced and used for the building façades.(only a few stirrups were placed at the beam ends) Other studies concerned the shear behavior of prestressed I beams made with conventional transverse reinforcement (stirrups) or, as an alternative, with FRC.

Figure 4:

Global view of the steel reaction frame of the panels.

With the latest developments of national and international standards other applications were developed. Magnetti is currently implementing the production of SFRC floors that have already been used for the construction of some buildings. The use of FRC in the field of structures was recently included in the Italian building code (NTC 2018); this could be the starting point for new research developments on the use of FRC in structural elements, taking into account the new typologies of products, different profiles and geometries.

Keywords Fiber Reinforced Concrete; Steel Fibers, Research Activity; Mechanical Behavior, Structural Applications

3

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Fiber-Reinforced Concrete - From Fresh Properties to Structural Design: New Tools, Guides, and Reports from ACI Technical Committee 544 Liberato Ferrara1and Barzin Mobasher2 1

: Department of Civil and Environmental Engineering, Politecnico di Milano, Italy; chair of ACI Technical Committee 544 – Fiber Reinforced Concrete. 2 : School of Sustainable Engineering and Built Environment, Ira A. Fulton Schools of Engineering, Arizona State University, Tempe, AZ, USA; past chair of ACI Technical Committee 544 – Fiber Reinforced Concrete

Abstract During the past 3 years, ACI Committee 544 has developed five new documents addressing the testing and fresh concrete properties, mechanical properties, back calculation of tensile properties, structural design with fiber-reinforced concrete (FRC), design of elevated slabs with FRC, and design of precast tunnel lining with FRC. These published reports (ACI 544.6R, 544.7R, 544.8R, 544.9R, and 544.3R) are currently available for the engineering community and offer a completely fresh way to use, design, and implement FRC in a variety of applications. The purpose of this paper is to provide an overview to the whole corpus of documents, illustrating each specific content and also addressing how the specific topics of documents interact with each other, paving the wat to implement and incorporate the knowledge in the design and specification.

Keywords Fiber reinforced cementitious composites; test methods; design approach; durability

Introduction After an about fifty year odyssey, starting with pioneer studies in the early 60s of last century, Fiber Reinforced Concrete (FRC) has now been fully recognized as a structural material. fib Model Code 2010 has provided a set of internationally recognized design rules for structural elements made of FRC, complemented with the necessary procedures for the experimental identification of material parameters which are used in the proposed structural design approaches (residual flexural strengths as per EN 14651). In this challenging international framework, and aiming at a much needed standard and code harmonization, ACI Technical Committee 544-Fiber Reinforced Concrete, which has accompanied technical development of FRC throughout the past decades since its early inception, has undertaken on the one hand a revision of current reports and guidelines on testing and design methods for FRC. On the other hand, in order to promote a more and more widespread use of FRC technologies and promptly update the technical community about the latest developments in the field, a series of Emerging Technology Reports addressing some recent massive applications of FRC dealing with, e.g.,

4

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

elevated slabs and precast tunnel segments. The presentation will provide an overview to the whole corpus of documents, illustrating each specific content and addressing how the specific topics of documents interact with each other, paving the wat to incorporate the knowledge in the design and specification.

The ACI 544 corpus of technical documents The set of ACI 544 technical documents, which has to be intended as an integrated corpus to be consistently used, covers the whole “value chain” of Fiber Reinforced Concrete materials and structures, encompassed by a “primer document” ACI 544.1R-96: Report on Fiber Reinforced Concrete, currently under revision, and by a document covering technological production issues, ACI 544.3R-08: Guide for Specifying, Proportioning, and Production of Fiber-Reinforced Concrete. The methods for the experimental identification of the material properties in the fresh and hardened state are reported in the detail in ACI 544.2R-17: Report on the Measurement of Fresh State Properties and Fiber Dispersion of Fiber Reinforced Concrete and ACI 544.9R-17: Report on Measuring Mechanical Properties of Hardened Fiber Reinforced Concrete, which also cover recently developed test methods for the assessment of fiber dispersion and orientation inside FRC structural elements, a topic which is becoming of the utmost relevance also in the sight of widespread use of fiber reinforced cementitious composites with adapted rheology, such as Fiber Reinforced Self-Compacting Concrete (FR-SCC). In this framework it is also work reminding the development of a joint document on FR-SCC together with ACI TC 237-Self consolidating concrete. The set of documents will be completed with a third one, currently under development, focusing on test methods for shrinkage, creep and durability related properties of FRC, with extended description on the effects of fibers on the aforementioned properties provided in ACI 544.5R-10: Report on the Physical Properties and Durability of Fiber-Reinforced Concrete. The document ACI 544.8R-16: Report on Indirect Method to Obtain Stress-Strain Response of Fiber-Reinforced Concrete (FRC), provides the necessary link between the experimental measurement of the material performance and the identification of those parameters, such as the “constitutive” tensile stress-strain and/or stress-crack opening relationship which have to be used in the design. Design methods are covered by ACI 544.4R: Guide to Design with Fiber Reinforced Concrete, whose completely revised version is going to be soon released. This document represents the corner stone of the whole ACI TC 544 activity. On the on hand it finalizes to structural applications the knowledge gathered in the other documents. On the other it consistently and comparatively addresses design methods based on the identification of design material parameters from different America, European and other international standards, in the context of code harmonization recalled above. Some examples of application to real structural concept, design and construction are provided in two Emerging Technology Reports: ACI 544.6R-15: Report on Design and Construction of Steel Fiber-Reinforced Concrete Elevated Slabs and ACI 544.7R-16: Report on Design and Construction of Fiber Reinforced Precast Concrete Tunnel Segments, which stand as a first example of consistent application of the whole set of tools and guidelines provided by ACI 544 to cutting edge engineering applications of FRCs.

5

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Material concept and production: ACI 544.1R-96 and ACI 544.3R-08

Material properties fresh and hardened state, early age and long term behaviour, durability ACI 544.2R-17; ACI 544.5R-10; ACI 544.9R:17; ACI 544.10R (in preparation)

Identification of design parameters: ACI 544.8R-16

Design methods: ACI 544.4R-18 (coming soon)

Applications (Emerging technology reports): tunnel segments, elevated slabs, FR-SCC … ACI 544.6R-15; ACI 544.7R-16; ACI 544.11R (in preparation)

ACI 237: SCC ACI 239: UHPC

ACI 544 FIBER REINFORCED CONCRETE

ACI 350: ENVIRONMENTAL ENGINEERING CONCRETE STRUCTURES

ACI 360: DESIGN OF SLABS ON GROUND

ACI 560 INSULATING CONCRETE FORMS ACI 506 SHOTCRETING

ACI 551: TILT-UP CONSTRUCTION

ACI … … to be continued

References ACI 544.1R-96: Report on Fiber Reinforced Concrete. ACI 544.2R-17: Report on the Measurement of Fresh State Properties and Fiber Dispersion of FRC. ACI 544.3R-08: Guide for Specifying, Proportioning, and Production of FRC. ACI 544.4R-18: Guide to Design with Fiber Reinforced Concrete (to appear). ACI 544.5R-10: Report on the Physical Properties and Durability of FRC. ACI 544.6R-15: Report on Design and Construction of Steel FRC Elevated Slabs. ACI 544.7R-16: Report on Design and Construction of Fiber Reinforced Precast Concrete Tunnel Segments. ACI 544.8R-16: Report on Indirect Method to Obtain Stress-Strain Response of FRC. ACI 544.9R-17: Report on Measuring Mechanical Properties of Hardened FRC. 6

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

High performance fibre reinforced concrete for structural applications Bruno Massicotte1 1

: Polytechnique Montreal, Montreal, Canada.

Abstract Fibres have been used by the construction industry for several decades, but their use in structural applications is still very modest if one considers the gigantic potential of concrete structures over the world and the benefits expected on their mechanical behaviour and durability. On the contrary, although their commercial availability is much more recent, UHPFRC are becoming attractive to structural engineers and stakeholders. The growing interest in UHPFRC, which occupy the upper end of a large family, might be the driving force that will bring the attention of designers to consider fibre reinforced concrete (FRC) as a whole. The exceptional mechanical and durability performances of UHPFRC make them very attractive materials for a wide range of applications. Stakeholders with long term vision are promoting the use of UHPFRC with promises of better service performances, higher longevity, and accelerated construction with enhanced quality. Despite the numerous advantages and the exceptional potential of these innovative construction materials, several challenges still need to be tackled. Which is fortunate: for maintaining the interest in research but also, and most importantly, for the innovation that will be generated and hopefully the long term benefits to the society. Fibre dispersion, mix design, casting procedures, material testing, design models, etc. are well known fibre concrete issues. These technical challenges are the easiest one to identify and, in most instances, the easiest ones to solve. Researchers and industrial actors are trained and dedicated to address such issues and one can confidently assume that most of today's known issues will be resolved in time. Economic challenges are getting under control. UHPFRC have shown to be economically advantageous at the construction time, in many instances due to the simplification of the construction procedure and the reduction of the construction time. When life cycle costs are included, as it should always be in a sustainable development strategy, using fibre concrete and very often UHPFRC generally, if not always, ends up with significant savings. This is true if owners are expecting to keep their assets for a long period or if infrastructure are planned to gain value with time. Two aspects however must also be considered: excessive quality should be avoided in infrastructure for which the purpose is expected to change in time, and the cost of deconstruction should be included, knowing that fibre concrete are tough materials. The success of recent projects, the increase in the number and magnitude of projects, and the contribution to sustainable development will definitely lead to an increase in the use of fibre concrete.

7

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

However, the most difficult challenge to tackle is to change the mentality of structural engineers. For the past 40 to 50 years, structural engineers have been trained and have gain experience in limit states design (or similar format), in which most of the attention is put on safety, with little, and often not appropriate, concern with serviceability. Structural engineering is probably the most codified area of engineering. Despite that, the deterioration of concrete structures due to environmental conditions clearly show that this aspect has not been addressed correctly in the past. Even today, with most of the attention being given to ultimate conditions, new concrete infrastructure show early signs of potential serviceability issues mostly attributed to cracking.

300

300

250

250

Benring moment (kN-m)

Benring moment (kN-m)

Figure 1 shows the flexural behaviour of a 0.25 m thick, 1.0 m wide panel for which the lower 0.1 m is made of either normal 30 MPa concrete, 80 MPa SFRC with 1% fibres and 120 MPa UHPFRC with 2% fibres, and reinforced with either 2000 mm2 normal reinforcement or 560 mm2 prestressing strands with an initial 1000 MPa stress. Although analytical results show the clear benefit of using fibre materials for limiting cracking along with the gain in strength, the softening flexural behaviour past the maximum strength (associated with crack localisation for both SFRC and UHPFRC and reinforcement yielding) are not naturally considered adequate from a structural engineer point of view. The general acceptance of such behaviour would require changing the usual engineering practice if one desires fibre concrete to be used efficiently. This would imply to less rely on prescriptive code requirements and to learn how to use numerical tools to understand the structural behaviour of components and structural systems that incorporate fibres and make the best use of FRC and UHPFRC.

200

150 NC / As 100

NC / Ap

FRC / As FRC / Ap

50

200

150 NC / As 100

NC / Ap

FRC / As FRC / Ap

50

UHPFRC / As

UHPFRC / As

UHPFRC / Ap

UHPFRC / Ap 0

0 0

20

Figure 1:

40

60

80

100

120

0

0.5

1

1.5

2

Curvature (km-1)

Crack opening (mm)

a) Moment-curvature

b) Crack opening

2.5

3

Precast panel behaviour.

UHPFRC, and fibre concrete in a broader viewpoint, unthoughtful have a promising future, because they contribute to mitigating concrete cracking and enhancing all concrete properties associated with durability, while improving several aspects of structural strength. For UHFRC to be correctly and increasingly used in structural applications, requires better educating structural engineers. If this objective is successful achieved, as said Pierre Rossi, fibre reinforced concrete may impact the construction industry practice in the XXIth Century as did prestressed concrete in the previous Century. It is hoped that in a near future one would have to justify why fibre concretes are not used.

Keywords UHPFRC; FRC; Structural performances; Sustainability; Durability; Accelerated construction

8

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Material characterization and in-situ control of FRC de la Fuente, Albert1 1

: Polytechnic University of Catalonia (UPC), Barcelona, Spain.

Abstract The quality control (QC) of fibre reinforced concrete (FRC) is performed in terms of the fibre content (Figure 1) and post-cracking residual strength (Figure 2) of the material. The former is usually assessed by means of the washing out test, which consist in weighting the fibres present in a certain volume of fresh concrete. This procedure takes around 45 minutes per test and requires the use of several litres of water. Consequently, this limit the number of tests that may be performed per day, and thus compromising both the statistical representativeness of the results and the effectiveness of the QC procedure. Alternatively, there exist other methods (non-destructive) based on the principles of the magnetic induction with high accuracy and that can be performed in less than two minutes. The specimens (cubic or cylindrical) are placed within a plastic container. Copper or aluminium wire coils are placed around the container, constituting the sensor element of the system. The inductance magnitude, properly correlated for each type of fibre (metallic), will provide the amount of fibres inside the specimen. The same specimen can be posteriorly tested to characterize a mechanical property.

Figure 1:

(a) Washing out test (destructive) and (b) inductive test (non-destructive).

The characterization of the post-cracking residual strength of FRC is normally performed throughout notched or unnotched prismatic beams tested considering one or two loading points. These tests are very useful to define the material constitutive equations for simulating the flexural tensile performance of the material. However, these test can be less appropriate for QC of massive concrete production since the beams have small specific failure surface and, together with the intrinsic variability of the fibre distribution in the cracked section, these tests use lead to a considerable scattering of the results (15-25% of variation coefficient) and, ultimately, to smaller characteristic residual strengths. Furthermore, the considerable weight of the specimen, together with the complex set up of the test and the need of the suitable closed-loop data 9

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

acquisition system, make it difficult and expensive the systematic characterisation of the material. Alternatively, and especially interesting for the QC of massive FRC production, there exist other test configurations derived from the double punching test that allow indirectly characterizing the FRC post-cracking tensile strength. Besides both the reduced weight of the specimens and the required time to perform these tests, the associated scatter is lower than that usually obtained with the beams test and, thus, the probability of having less uncomplaining batches is also lesser (Figure 3).

Figure 2:

Experimental setup for: (a) three edge point bending test on notched beam and (b) double punching test on cubic specimen.

% of uncomplayning batches

25

20

15

CVtest = 20%

10

5 CVtest = 13% 0 1,5

Figure 3:

1,7

1,9 2,1 Specified fRk (MPa)

2,3

2,5

Statistical simulation of the percentage of uncomplaining batches according to the specified value of fRk and the CV obtained in the quality control test (Cavalaro et al. 2015).

The main objective of this conference paper is to present the different tests configurations for dealing with the FRC quality control emphasizing those that can lead to improvements of its efficiency and robustness particularly for massive production.

10

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Acknowledgements The author wish to express their gratitude to the Spanish Ministry of Economy, Industry and Competitiveness for the economic support in the scope of the project SAES (BIA2016-78742C2-1-R).

References Cavalaro, S.H.P., Aguado, A., 2015. Intrinsic scatter of FRC: an alternative philosophy to estimate characteristic values. Mater. Struct. 48, 3537–3555. Cavalaro, S.H.P., López, R., Torrents, J.M., Aguado, A., 2014. Improved assessment of fibre content and orientation with inductive method in SFRC. Mater. Struct. 48(6), 1859–1873. de la Fuente A., Blanco A., Pujadas P., Aguado, A., 2012. Experiences in Barcelona with the use of fibres in segmental linings. Tunn. Undergr. Space Technol. 27(1), 60–71. di Prisco, Ferrara, L., Lamperti, M.G.L., 2013. Double-edge wedge splitting (DEWS): and indirect tension test to identify post-cracking behaviour of fibre reinforced cementicious composites. Mater. Struct. 46(11), 1893-1918. EN 14651:2007. Test method for metallic fibered concrete. Measuring the flexural tensile strength (limit of proportionality (LOP), residual). fib Bulletins 65-66 (2010), Model Code 2010. fédération internationale du béton (fib), Lausanne (Switzerland). Galeote, E., Blanco, A., Cavalaro, S.H.P., de la Fuente, A., 2017. Correlation between the Barcelona test and the bending test in fibre reinforced concrete. Constr. Build. Mater. 152, 529538. Pujadas, P., Blanco, A., Cavalaro, S., de la Fuente, A., Aguado, A., 2013. New analytical model to generalize the Barcelona test using axial displacement. J Civ Eng Manage. 19(2), 259– 271. Pujadas P., Blanco A., Cavalaro S.H.P., de la Fuente A., Aguado A., 2014. Multidirectional double punch test to assess the post–cracking behaviour and fibre orientation of FRC. Constr. Build. Mater. 58, 214–224. Monte, R., de la Fuente, A., de Figuereido, A.D., Aguado, A., 2016. Barcelona test as an alternative method to control and design fiber-reinforced concrete pipes. ACI Struct. J. 113(6), 1175-1184. Torrents, J.M., Blanco, A., Pujadas, P., Aguado, A., Juan-García, P., Sánchez-Moragues, M.A., 2012. Inductive method for assessing the amount and orientation of steel fibers in concrete. Mater. Struct. 45, 1577–1592. UNE 83515:2010. Fibre reinforced concrete. Determination of cracking strength, ductility and residual tensile strength. Barcelona test. Spanish Association for Standardization and Certification, AENOR. In English.

11

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Recent developments in FRC tunnel linings Alberto Meda1 1

: Tunnelling Engineering Research Centre, University of Rome “Tor Vergata”, Italy.

Abstract Tunnels, mechanical excavated with a TBM (Tunnel Boring Machine), are usually characterized by a precast segmental lining. The lining is installed by the TBM during the excavation process and it is used to push the machine during the boring phase. The use of Fiber Reinforced Concrete (FRC) in precast segments for tunnel lining is growing in the last years and this is becoming one of the principal application of this material. Typically, two solutions are proposed: a hybrid solution, where fiber can partially substitute the steel cage, and a FRC only solution, where the ordinary reinforcement is totally substituted by the fiber reinforcement. The main advantages linked to the use of FRC in tunnel segments are related not only to the cost, but also to the possibility of enhancing structural and durability performances. The application of FRC in segmental tunnel lining require a proper design in order to obtaining a successful application. In the first tunnels made with FRC, design was mainly based on a test proofing procedure. It was immediately onset the need of guidelines and codes for design FRC structures in general, but also with particular regard to this tunnel application. FRC & RC/FRC precast tunnel segments: case studies over the years 2011-2017

2006-2010

2000-2005

'90s

'80s

Hybrid RC/FRC

0

Figure 1:

5

10

15

20

25

30

35

FRC

40

FRC tunnel application in the recent years.

After the publication of a series of National Codes, the publication of key documents as Model Code 2010, certainly boosted the use of FRC application in tunnels. Already several tunnels have been designed adopting Model Code 2010 as references document. The documents related the use of FRC, as Model Code 2010, refer to general structures and they mainly consider the solution with a partially substitution of the ordinary reinforcement. Furthermore, a series of guidelines specific for the FRC tunnel application were issued, giving a valid support to designer, clients and construction companies. In particular, Fib

12

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Bulletin 83 Precast tunnel segments in fibre-reinforced concrete was recently published with the aim of suggest a design procedure for FRC precast segments, consider also aspects related to quality control, durability and sustainability. This document highlighted the necessity of a conceptual design approach in order to clarify the requests in terms of material performance and structural performance, that should be balanced with considerations related to cost. As a consequence, it appears important to make correct choices from the beginning of the design process in order to obtain the best performances. This is particular important when an FRC only solution is proposed since aspects related to structural ductility (important for Ultimate Limit State design) and cracking control (at Serviceability Limit State) should be ensured. As an example, it is important to consider the possibility to have segments with strain hardening behavior in bending, that can be achieved with a correct design of the FRC material.

Figure 2:

Behavior of a segment with a 250 mm thickness adopting a C40 4.0c FRC.

Another important aspect that should be considered in the design is related to the lining durability. Requests in term of service life for tunneling are becoming more severe (from 120 to 200 years). If correctly designed against durability, FRC solution can be an optimal solution, also considering the effect of stray current. Finally, it has to consider the advantages offered by the fiber reinforcement in the future technologies related to tunneling. For example, great interest is related to the possibility of having a precast production that can real time follow the TBM excavation, avoiding segment stocking and transportation. These industrialized processes can take great advantages by the adoption of FRC instead to a traditional reinforcement.

Keywords FRC Precast Tunnel Segments, Mechanical excavated tunnels, Tunnelling design

13

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Steel Fiber Reinforced Concrete For Structures Subjected To Severe Actions Gustavo J. Parra-Montesinos1 1 : University of Wisconsin-Madison, Madison, WI 53706, USA

Abstract The use of deformed steel fibers as a means to improve the behavior or simplify the construction of reinforced concrete structures has been investigated for over five decades. Among applications investigated in the past few decades is the use of steel fiber reinforced concrete (SFRC) in structures prone to be subjected to severe loading conditions such as earthquakes, impact and blasts, given the significant tension and compression ductility exhibited by some types of steel fiber reinforced concrete. Also, as SFRC is increasingly used in a wide variety of structures, the effects other extreme conditions such as severe corrosive environments and elevated temperatures have on material mechanical performance and structural behavior need to be well understood. Strong ground motions often induce large inelastic displacement reversals in structural members/systems. In order for concrete structures to withstand such displacements while maintaining gravity load carrying capacity, careful design of transverse and longitudinal reinforcement is required. The need for closely spaced transverse reinforcement to confine the concrete, provide shear resistance and laterally support longitudinal reinforcement in critical regions of earthquake-resistant structures often translates into significant reinforcement congestion with the associated construction difficulties. This has led several investigators to evaluate the potential of using deformed steel fibers in the concrete as a means to reduce reliance on transverse reinforcement in critical regions of earthquake-resistant structures without compromising performance. These regions include coupling beams and the base of structural walls, beam-column joints, and beam and column plastic hinge regions. Research on the use of SFRC in coupling beams has shown that elimination of diagonal reinforcement is possible in coupling beams with span-to-depth ratio of at least 2.0 and subjected to average shear stresses as high as the maximum limit in ACI 318-14. While in most cases tensile strain-hardening fiber reinforced concretes are required to achieve adequate deformation capacity, it has recently been shown that strain-softening materials are sufficiently effective in coupling beams with span-to-depth ratios of approximately 3.0 or greater and moderate levels of shear stress. The elimination of diagonal reinforcement through the use of deformed steel fibers is accompanied by important reductions in transverse reinforcement. The significant simplifications achieved in the construction of coupling beams, particularly by eliminating diagonal reinforcement, has led to the adoption of this concept in several high-rise structures on the west coast of the USA (Figure 1), also demonstrating that SFRCs with fiber volume fractions as high as 1.5% can successfully be produced in large volumes and cast in normal field conditions through a crane and bucket operation (Figure 1b). This and other investigations on the use of SFRC in earthquake-resistant structures has led to two consistent findings: steel fibers allow a reduction in shear and confinement reinforcement; and steel fibers

14

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

lead to a higher deformation tolerance by enhancing cracking distribution and concrete deformation capacity.

(a) A core wall at Lincoln Square Expansion, Bellevue, WA, USA (Courtesy of Cary Kopczynski & Co.) Figure 1:

(b) Casting of steel fiber reinforced concrete in The Martin building, Seattle, WA, USA.

Examples of use of steel fiber reinforced concrete in coupling beams of earthquakeresistant wall structures.

The significant toughness exhibited by strain-hardening fiber reinforced concretes, particularly ultra-high performance fiber reinforced concretes (UHPFRC), has led to the evaluation of these materials for use in structures prone to impact and blast loading. Recent investigations at various institutions have shown that UHPFRC structural members exhibit superior impact and blast resistance compared to members constructed with regular concrete. In general, the use of UHPFRC leads to a reduction in maximum and residual displacements, as well as to less damage compared to regular concrete members. The use of UHPFRC as overlays to regular concrete structural members has also been shown to provide excellent resistant against impact and blast loading. Although there is much to be done in terms of laboratory and analytical research on the behavior of SFRCs under extreme loading conditions, there is ample evidence indicating their great potential to enhance the performance of structures subjected to strong earthquakes, impact and blasts, while in some cases allowing for significant simplifications in construction.

15

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

16

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Mechanical characterisation

17

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Round-Robin Test on Various Test-Methods for Flexural Behavior of Steel Fiber Reinforced Sprayed Concretes Sven Plückelmann1, Rolf Breitenbücher1 1

: Ruhr University Bochum, Bochum, Germany.

Keywords Sprayed concrete, steel fibers, round-robin test, flexural behavior

Abstract According to the actual European standard EN 14487-1 the potential of steel fiber reinforced sprayed concrete is characterized by flexural strength tests (first peak, ultimate and residual). In most cases, this is performed by a four-point bending test on beam specimens, specified in EN 14488-3. As an alternative test method, a three-point bending test on square panels with notch is recommended by EFNARC. It is argued as main benefit of the latter test method, that the geometry and dimensions of the panels are equal to those of specimens used for measuring the energy absorption capacity according to EN 14488-5. Hence, the specification of the ductility of fiber reinforced concretes according to EN 14487-1 in terms of residual strength and energy absorption capacity can be achieved preparing only one type of specimen. Furthermore, the EFNARC guideline points out a smaller scatter of test results, compared to the beam tests according to EN 14488-3. Before the EFNARC method will be considered in EN 14487-1, the relevant CEN TC 104/WG10 requests for adequate proofs. These were performed within a round-robin test (RRT) on testing the flexural behavior of steel fiber reinforced sprayed concrete by the standardized EN 14488-3 method as well as by the proposed EFNARC method. The aim of this RRT was to investigate the comparability and correlation between the two test methods. Furthermore, the scatter of both methods was assessed. The RRT has been organized by Ruhr University Bochum, in whose labs the steel fiber reinforced sprayed concrete specimens (beams according to EN 14488-3 and square panels according to EFNARC guideline) were produced using a robotic spraying machine. The specimens were then tested in five independent laboratories Europe-wide. In Figure 1 the stress-deflection and stress-CMOD curves of all specimens (beams according to EN 14488-3: B1-15; panels according to EFNARC: P1-20) tested within the RRT and the respective laboratories are illustrated. The flexural tensile strengths (first peak/LOP, ultimate and residual) were calculated from the presented stress-displacement curves of the beams and panels. The total mean values MV, standard deviations s and variation coefficients v were determined and are listed in Table 1.

18

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 2:

(a) (b) Test results of the five labs for the flexural tests according the EN 14488-3 method (a) and according to the EFNARC method (b).

Table 1:

Flexural strengths according to EN 14488-3 / EFNARC.

Flexural strength

Ultimate strength fult First peak strength ffp Residual strength fR1 Residual strength fR2 Residual strength fR4 Ultimate strength fult LOP fct,L Residual strength fR1 Residual strength fR2 Residual strength fR3 Residual strength fR4

Mean value MV Standard deviation s [N/mm²] [N/mm²] Beams according to EN 14488-3 4.68 0.33 4.65 0.34 3.45 0.62 3.26 0.52 2.75 0.61 Panels according to EFNARC 4.70 0.48 4.68 0.49 3.93 0.49 4.07 0.55 3.73 0.49 3.26 0.43

Variation coefficient v [%] 7.1 7.4 17.8 16.0 22.2 10.2 10.4 12.5 13.4 13.2 13.0

With regard to the total mean values of ultimate strengths as well as first peak strengths and LOPs, the test results showed an almost exact correlation between the test methods. The total scatter of ultimate strengths as well as first peak strengths and LOPs was slightly larger for the EFNARC test method compared to the EN 14488-3 method. In contrast, the total scatter of residual strengths was somewhat larger for the EN 14488-3 method compared to the EFNARC method. However, a direct comparison between the absolute as well as the statistical values of the residual strengths of the beams and panels is unfeasible due to their different definitions. All in all, it can be stated that there are no significant differences concerning the scatter of results. Hence, it is quite conceivable that the EFNARC method would be considered in EN 14487-1. However, the classification of residual strengths according to EN 14487-1 needs to be adapted.

19

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Comparison of different methods of inverse analysis to assess the tensile law of UHPFRC from four points bending tests David Bouchard1, Luca Sorelli1, David Conciatori1, Svatopluk Dobrusky2 : University Laval – Department of Civil Engineering, Quebec City, Canada. 2 : LafargeHolcim – San Quentin Fallavier, France 1

Introduction Ultra-high Performance Fiber Reinforced Concrete (UHPFRC) is a relatively new cement based composite which is emerging in several structural applications thanks to its outstanding compressive strength and remarkable toughness. Moreover, new design recommendations have been developed which account the tensile law of UHPFRC in design. Tensile tests have already been developed to characterize UHPFRC tensile law, using either dumbbell-shaped specimens or straight, constant cross-section specimens. However, several factors may affect the tensile behavior of UHPFRC such as misalignment of the specimen, parasite moment due to the effect of boundary conditions, or the effect of grip ends on the stress distribution. Therefore, inverse analysis methods have been developed to obtain the UHPFRC tensile law from bending tests. The flexural response obtained is then analyzed by a model based on continuum mechanics in order to assess the tensile law. Different methods has been developed to model the flexural behavior of UHPFRC. For instance, AFGC employs the method of Chanvillard (2000) to analyze the flexural behavior of UHPFRC beam by considering the crack height evolution. More recently, the Swiss code SIA has proposed a simplified method of inverse analysis based on the explicit equations of the work of Chanvillard (2013). The present study focuses on a comparison between the tensile laws of the material obtained from the methods developed by Dobrusky and Chanvillard (2017) and Lopez ((2015) and (2016)). The present results compare the capacity of these methods to predict the tensile law of UHPFRC with 1% and 2% of steel fibers.

Experimental program Specimen geometry The UHPFRC beams tested in the present research had a cross-section of 100 mm (height) x 40 mm (width), with a length of 600 mm (span-to-depth L/h ratio of 4 and shear span-todepth a/h ratio of 1). The specimens were separated in two groups of four beams, with a fiber content by volume of 1 or 2%. Fibers used had a length of 14 mm and a diameter of 0.185 mm, for an aspect ratio (L/) of 75 and were provided by Bekeart.

Test set-up The beams were tested using 4PBT set-up, accordingly to Figure 1. The span was set to 400 mm (L/h = 4). The load was applied by means of a closed-loop control at a rate of 0.3 mm/min 20

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

(measured on the mid-span deflection) up until the maximal load. The rate of loading was then further increased to 0.6 mm/min until the end of the test. The specimens were heat treated for 24h (90°C and 100% RH) and were tested, un-notched, about 1 year after casting to prevent the effect of ageing on the results.

Figure 1.

Four points bending test set-up.

Methods Lopez and al. (2015) proposed an iterative inverse analysis method based on the minimization of the error between two curves: experimental flexural strength vs experimental curvature and analytical flexural strength vs experimental curvature. The relationship between the experimental deflection and the curvature is also given in his work. The minimization of the error is done by iteration on the tensile parameters, which in turn influences the analytical curve. Lopez and al. (2016) proposed a simplified inverse analysis method based on five key points identified on the experimental equivalent bending strength vs deflection at mid-span of unnotched 4PBT. 6 different parameters are calculated with these key points, and the tensile law of the material can be plotted accordingly to these parameters. Dobrusky and Chanvillard (2017) also developed an inverse analysis procedure for the determination of the tensile law of UHPFRC. The model assume the beam specimen as a nonlinear continuum with a nonlinear hinge at the crack location. Shear contribution is also included in the determination of the deflection of the specimen. The method is also valid for specimen with rebar reinforcement.

21

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A new multi-scale hybrid fibre reinforced cement-based composites Li Li1, Mingli Cao1 1

: Dalian University of Technology, Dalian, China.

Abstract Conventional cement-based composites are very brittle materials and the failure of cementbased composites is a progressive and multilevel process, pre-existing or newborn microncracks grow to meso-cracks and macro-cracks. Eventually, the macro-cracks lead to the fracture. In this sense, combination of fibres with different sizes and constitutions is generally drawn into cement-based composites to restrict the growth of cracks at different stages in the failure process and improve its mechanical properties. Unfortunately, traditional hybrid fibres can hardly delay micro crack initiation and propagate because of the relatively large length and diameter. A new hybrid fibre reinforced cement-based composites (HFRCC), including steel, polyvinyl alcohol (PVA) fibre and micron calcium carbonate (CaCO3) whisker (20-30 μm length and 0.5-2 μm diameter), has been developed. Six mix groups with different type and dosage of steel fibre, PVA fibre and whisker are illustrated in Table 1. Table 1:

Content of different fibre in HFRCC.

Group Pure S2 S1.5P0.5 S1.5P0.4W1 S1.25P0.75 S1.25P0.55W2

Fibre type Plain 35mm hooked end steel fibre and 6mm PVA fibre

Volume fraction（%） Steel fibre PVA fibre Whisker

Fiber and whisker cost ($/m3)

0

0

0

0

2

0

0

437

1.5

0.5

0

366

1.5

0.4

1

365

1.25

0.75

0

331

1.25

0.55

2

329

Three prismatic and three cube specimens of each group were casted. Prismatic specimens with dimensions of 100 (width) × 100(thickness) ×400(length) mm3 were used to test flexural behaviour. The cylinder specimens with dimensions of Φ100 (diameter) × 200(height) mm3 were employed to test compressive behaviours. For the flexural testing, a computer-controlled electro-hydraulic servo universal equipment of 300kN capacity was used according to ASTM C1609/C1609M. For the compressive strength testing, a computer-controlled electro-hydraulic servo universal equipment of 2000kN capacity at a displacement rate of 1 mm/min was used according to ASTM C469/C469M. From test results it turns out that, compared to the conventional hybrid fibers (steel fibre and PVA fibre), the addition of multi-scale fibres (steel fibre, PVA fibre and CaCO3 whisker) can

22

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

improve the compressive strength and toughness of cement-based composite, as shown in Figure 1. Multi-scale fibre usage leads to notable increase to both the flexural strength and flexural toughness values of cement-based composite, as shown in Figure 2. 50

70

Pure S2 S1.5P0.5 S1.5P0.4W1 S1.25P0.75 S1.25P0.55W2

Stress (MPa)

50

40

Strain energy (N/mm2)

60

40 30 20

30

20

10

10 0 0.0

0.5

1.0

1.5

Strain (%)

2.0

2.5

0

3.0

Pure

(a) Stress–strain curve Figure 1:

S2

S1.5P0.5

S1.5P0.4W1 S1.25P0.75 S1.25P0.55W2

(b) Strain energy

Compressive behaviors of various mixtures.

55

60

50 45

Flexural toughness (N*m)

S1.25P0.55W2

40

Load(kN)

35 S1.5P0.4W1

30 25 20

S1.5P0.5

S2.0

15 10 Pure

5

S1.25P0.75

0 0

1

2

3

4

50

40

30

20

10

0

Deflection(mm)

(a) Load-deflection curve Figure 2: Flexural behaviors of various mixtures.

Pure

S2

S1.5P0.5

S1.5P0.4W1 S1.25P0.75 S1.25P0.55W2

(b) Flexural toughness

Based on these test results, we can conclude that there is a remarkable fiber synergy in multiscale hybrid fibre reinforced cement-based composites with 35mm hooked end steel fibre, 6mm PVA fibre and CaCO3 whisker. It seems possible that the steel fibres and PVA fibres can be partly replaced by CaCO3 whiskers, which is very beneficial in decreasing the production cost (shown as the sixth column in Table 1) of fiber reinforced concrete for potential structural applications.

Keywords Multi-scale hybrid fibre, CaCO3 whiskers, compressive behavior, flexural behavior

23

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A sectional approach for the bending creep of FRC based on uniaxial tension creep tests Rutger Vrijdaghs1, Marco di Prisco2, Lucie Vandewalle1 1

: Department of civil engineering, KU Leuven, Leuven, Belgium. 2 : Department of structural engineering, Politecnico di Milano, Milan, Italy.

Abstract The creep behavior of FRC elements remains an important obstacle to use FRC in structural applications. Owing to the residual post-cracking strength properties of FRC, creep deformations play an important role in the cracked sections and influence durability and SLS requirements of structural elements. Therefore, it is of high importance to take creep deformations into account in the design phase. In this paper, two topics are discussed. Firstly, the results of an experimental campaign into the creep behavior of cracked polypropylene FRC under sustained uniaxial tension loading is presented. Furthermore, the results of the creep tests were modelled in the dedicated finiteelement program DIANA. Secondly, a sectional approach is presented in which the results of the uniaxial tensile creep tests can be incorporated. By using inverse analysis to determine stresses and crack widths in the cracked section of the beam, creep functions are assigned to the discretized beam section. The experimental program focusses on a normal strength concrete (fcm = 43 MPa) with 1 V% of polypropylene fibers and consists of a series of short-term characterization and long-term creep tests. The FRC is firstly characterized in short-term tests, both in bending as well as in uniaxial tension. Furthermore, 6 beams were tested in a EN14651 three-point bending configuration at cyclic varying loads. Five LVDTs were placed over the cracked section to measure crack opening and closing upon un- and reloading in bending. Secondly, 14 specimens were cored from the FRC beams and subjected to uniaxial tensile creep loading after precracking to an initial crack width of 0.2 mm in the tensile frame. Two different load ratios were considered, i.e. 30 % and 45 % of the residual strength at the precracking level. Based on the results of the bending tests, the FRC is classified according to the Model Code 2010 as a 2d concrete. The results of the bending test can also be used to predict the uniaxial behavior, i.e. the stress-crack width (σ-w) behavior based on equation found in the Model Code and as proposed by di Prisco et al. [1]. This combined MC10-[1] model is able to accurately describe the uniaxial behavior based on bending results when compared to experimental data. The uniaxial tension creep behavior is modelled in a finite element model with the aim of describing and predicting the creep behavior in uniaxial tension based on the results of the single fiber and pull-out behavior, using a discrete treatment of the fibers. The general crack width growth is quite accurately captured, but the initial deformations upon loading are overestimated by the numerical model. Nevertheless, the creep deformations as obtained from the FEA are retained. Even though FEA is a very powerful tool, in most real design problems, it is quite often prohibitively expensive to construct, run and analyze the results. Therefore, most design approaches consider a sectional approach which can be easily understood and adopted. A 24

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

proposal for such a sectional approach is given to take creep deformation into account. The basis of this approach is a discretized cracked section subjected to bending moments. Uniaxial creep deformations are then assigned to the discretized elements in the cracked section. Before creep deformations can be taken into account, the first step is to calculate the stresses in the cracked section, both upon loading (i.e. following the EN 14651 test) and unloading (i.e. at the beginning of the creep test). In the loading phase, the stresses and deformations can be calculated using the uniaxial constitutive relationship, assuming a plane cracked section. The approach is able to quite accurately capture the test results, albeit it overestimates the bending tensile strength of the concrete somewhat. Upon unloading however, the plane section approach cannot be assumed correct anymore, as unloading under such an assumption would entail complete recovery of crack widths, which is not observed experimentally. Therefore, a bilinear deformation distribution over the section height is proposed with the inflexion point at the height of the undamaged zone. In an inverse analysis, the algorithm tries to find the scalar damage evolution D(w) to describe the complete unloading path of the cyclic FRC bending test. Using the deformation and stress profiles calculated from this bilinear distribution as a starting point, it is possible to predict the bending creep based on the uniaxial creep compliance functions. It is assumed that the undamaged tensile creep behavior is identical to the undamaged compressive creep, and that the damaged tensile creep compliance is independent of the initial crack width. While the results provide an indication of the expected increase in deformations, it should be noted that in the real beam, a stress redistribution will take place owing to the increase of the crack height and higher associated compressive stresses, further increasing the creep deformations. This time-dependent cracking needs further investigation, both on an experimental level as well as in the sectional approach. As such, the predicted creep deformations should be regarded as the lower limit of the real deformations. Nevertheless, in the presented model, with creep coefficients exceeding 7 after 120 days, the predicted creep deformations are significant and should be taken into account where SLS requirements are of importance.

Keywords Sectional analysis, creep of FRC, polymeric FRC, tensile and bending creep [1]di Prisco, M., M. Colombo, and D. Dozio, Fibre-reinforced concrete in fib Model Code 2010: principles, models and test validation. Structural Concrete, 2013. 14(4): p. 342-361.

25

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Optimized quality control procedure with the Barcelona test and the inductive method for FRC Eduardo Galeote1, Ana Blanco2, Sergio H.P. Cavalaro2, Albert de la Fuente1 1

: Universitat Politècnica de Catalunya, Barcelona, Spain. 2 : Loughborough University, Loughborough, United Kingdom.

Abstract The three-point bending test is one of the most extended methods to determine the postcracking strength of FRC. However, as suggested by the Belgian standard NBN B 15-238, the three-point bending test is not a suitable method for the systematic quality control of FRC. In this regard, the Barcelona test arises as an alternative to streamline the quality control given its simpler setup and lower scatter. Additionally, the influence of the content of fibres on the residual strength of FRC is an outstanding issue. For this, the inductive method is here presented as a complementary test for quality control to determine and verify the content of fibres in a concrete mix. The simplicity of the Barcelona test represents a clear asset with respect to the three-point bending test. The smaller dimensions of the specimens in the Barcelona test compared to those required for the bending test or the lack of complex initial preparation allows reducing the time of testing and simplifies the testing procedure. Another important advantage of the Barcelona test is that it does not require additional devices and the test can be performed in a conventional compressive equipment, thus reducing the costs. In this regard, the main objective is to describe the methodology to correlate the results of the Barcelona test and the bending test for quality control. Moreover, the procedure to determine the content of fibres using the inductive test and the method for its calibration is also shown. Three batches of the same concrete mix were produced to be tested under the three-point bending test and the Barcelona test. For this, prismatic beams with dimensions of 150x150x600mm and cubic specimens of 150x150 mm were casted for the flexural test and the Barcelona test, respectively. To calibrate the inductive test with the type of fibre used in this experimental program, the cubic specimens were crushed after the Barcelona test. The fibres of these specimens were separated from concrete, weighted to determine the content and correlated with the inductance measurements of the inductive test. To achieve a correlation between the mechanical parameters of the three-point bending test and the Barcelona test, it was necessary to determine an equation using the experimental results obtained through both methods. A parametric analysis yielded that if using the load and the energy of the Barcelona test at displacements of 0.5, 1.5, 2.5 and 3.5 mm it is possible to obtain a correlated value of the load of the bending test at crack openings of 0.5, 1.5, 2.5 and 3.5 mm. To take into account the intrinsic variability of FRC, a confidence interval of 99% is calculated and presented according to the results of the calibration. The results show similar trends of both the experimental results and the correlated values, being the experimental results of the threepoint bending test slightly higher than the average results obtained by the correlation. According to the results of the correlation, the differences between the average experimental-correlated

26

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

were even lower than the scatter between batches when conducting the three-point bending test. The confidence intervals reduce the correlated value to remain on the conservative side. The calibration between the inductance and the content of fibres separated manually from the concrete and the inductance measured revealed an average content of fibres of 27.95 kg/m3 with a variability of 10.5%. This scatter is mainly attributed to the heterogeneity of the material and the random distribution of fibres within the concrete. The relation between the inductance and the content of fibres presents an adjustment of R2=0.99.

Keywords FRC, quality control, bending test, Barcelona test, inductive method, content of fibres

27

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Influence of the beam size on the residual strength of fibrereinforced concrete Eduardo Galeote1, Ana Blanco2, Sergio H.P. Cavalaro2, Albert de la Fuente1 1

: Universitat Politècnica de Catalunya, Barcelona, Spain. 2 : Loughborough University, Loughborough, United Kingdom.

Abstract The constitutive law for fibre reinforced concrete (FRC) of the fib Model Code 2010 (MC2010) defines the post-cracking behaviour in terms of Strength-Strain with parameters obtained from beams tested under a three-point bending configuration. However, the response of concrete elements of different dimensions is generally affected by the size-effect. In this regard, the MC2010 does not specify clearly how to take into account the influence of the dimensions of the elements to determine their structural response. In this line, using specimens with smaller dimensions than the standardized beams to characterize FRC arises as an attractive possibility. For this, the influence of using small beams to conduct bending tests and determine the parameters of the constitutive law for FRC needs to be assessed. Accordingly, an experimental program involving four mixes of high performance fibre reinforced concrete (HPFRC) with microfibres in contents of 90 and 190 kg/m3 and two water to cement ratios was designed. The mechanical characterization of the four concrete mixes involved the compression strength, the modulus of elasticity and three-point bending tests. To analyse the size effect on the post-cracking strength of HPFRC, prismatic samples with dimensions of 150x150x600 mm, 100x100x400 mm and 40x40x160 mm were casted. To conduct the bending tests with smaller specimens the ratios depth-span and depth-notch were kept constant to maintain the proportions of the beams. To compare the average flexural strength between the different sized beams, the results were addressed in terms of StrengthRotation, exhibiting a rotation-hardening behaviour and a clear influence of the specimen size on the residual strength. The parameters of these tests were used to calculate the constitutive laws for FRC and subsequently compute the Strength-Rotation curve by means of a backanalysis. A back-calculation based on an analysis of evolutionary sections (AES) with a multi-layer approach was used to determine analytically the flexural behaviour of FRC. The constitutive model of the MC2010 describing the behaviour of FRC at the post-cracking stage was implemented in the AES using the experimental data of the three-point bending tests conducted on the different sized beams. Given the lack of specific indications to calculate the parameters related to the constitutive law of FRC when non-standard specimen dimensions are used, two approaches were considered to determine these parameters. The first approach consists of considering the full crack opening (FCO) regardless of the dimension of the specimen. The second approach suggests using an equivalent crack opening (ECO) depending on the depth of the specimen. The approach affects both the strains 𝜀𝑆𝐿𝑆 and 𝜀𝑈𝐿𝑆 since these parameters are calculated through the value of the crack opening.

28

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Consequently, the mechanism of the FCO exhibits a CMOD in smaller specimens comparatively higher than in bigger samples. Conversely, the ECO leads to identical deflections and strains for different sized beams. A comparison between the analytical curves obtained using the FCO and the ECO approach reveals that calculating the constitutive law using the ECO approach, the analytical StrengthRotation curves present a more accurate fitting with regard to the experimental results. These differences lay either on the strains and the strengths associated to each approach. Also the content of fibres seems to affect the data fitting. The differences between these two approaches become more pronounced while decreasing the size of the specimens and also depend on the content of fibres of the mixes.

Keywords FRC, size effect, constitutive law, back-calculation, specimen dimension

29

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Mechanical properties of self-compacting concrete reinforced with hybrid mixes of hooked steel fibers of variable lengths Małgorzata Pająk1 1

: Silesian University of Technology, Faculty of Civil Engineering, Department of Structural Engineering, Gliwice, Poland.

Introduction Randomly distributed short steel fibers enhance the tensile mechanical properties, fracture energy and behaviour under impact of brittle concrete matrix. The effectiveness of fibers in the concrete matrix depends mainly on their geometrical parameters, material and the matrix (Brandt (2008)). The fibers delay the micro-crack formation in the pre-peak stage and improve the ductility of concrete in the post-peak regime. However, fracture in concrete is a multiscale process, which comes gradually. One type of fibers can provide reinforcement at one limited range of strains. Combining fibers of different lengths could be more effective in comparison with only one type of fiber. In recent years, in the field of interest of scientists were hybrid mixtures with combination of different types of fibers (Banthia et al. (2014), Pająk (2016)). In the present research the attempt to join the advantages of the self-compacting concrete with the hooked fibers of different lengths were made. The present research were focused on analyzing different proportions between two types of fibers to determine the most effective mix, which could produce a synergy.

Experimental program The aim of this study was to experimentally examine the influence of hybrid mixes of steel hooked fibers on the compressive and flexural properties of the SCC as well as its rheological parameters. The combinations of steel hooked fibers of two lengths: 30 mm and 50 mm, with comparable aspect ratio were applied to SCC. The proportions between the types of fibers were changing while the total volume fraction of fibers was constant and equal to 0.76%, what was a fiber content equal to 60kg/m3. In summary, five types of mixes were analyzed. The compression tests and four-point bending tests on notched beams with the dimensions of 150×150×600 mm3 were determined (Fig. 1).

Test results, discussion and conclusions The results from the rheological tests indicate that the fibers decreased the workability of the SCC, what was tested in slump flow test. Further, the mixes were not able to appropriately pass the reinforcement bars in the L-box test, thus the requirements for FRSCC were not fullfiled. However, this mixes could be still placed in the strucural elements with no congested reinforcement. The hybrid fibers slighlty decreased the compressive strength of the SCC matrix, which was equal to 63.1 MPa. The reduction of the comressive strength was in a range of 3% to 9% depending of the hybrid mix.

30

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

The average load-CMOD curves obtained in flexural tests of HFR-SCC were presented in the Fig. 2. The collation of the results indicated that replacing the 30 mm long hooked fibers with 50 mm ones enhanced the post-peak parameters of the matrix. This conclusion confirmed the findings that the shorter fibers were more effective in delaying the appearance of the microcracks, what affects mainly the hardening phase. Meanwhile, the longer fibers bridged to prevent from rapid opening of localized cracks (Stähli et al. (2007)). Generally, the increase of the amount of long fibers in hybrid mix resulted in the improvement of the flexural parameters. However, the combination of 15 kg/m3 of the 30 mm long fibers with 45 kg/m3 of fibers with the length equal to 50 mm (mix H_15-45) was the most efficient from all tested mixes. Considering the number of fibers in the mix, it appeared that the number of shorter and longer fibers in this mix was almost equal. It can be concluded, that proportions between the fibers equal to 75-25% resulted in synergy effect. Exactly the same proportions between the fibers were indicated by other scientists as the most effective.

Figure 13: Test setup for flexural three-point bending tests on notched beams.

Figure 2: The comparison of the load-CMOD curves of HFR-SCC.

The flexural tensile strengths derived from three-point flexural tests on notched beams could be further used in structural calculations of FRC (di Prisco et al. (2013)). The analysis of the flexural parameters indicated that the HFR-SCC could be used not only to decrease the crack width and distance between the cracks but also as a substitute of the conventional reinforcement, however, not congested one taking into account blocking of the fibers observed in rheological tests.

References Banthia N., Majdzadeh F., Wu J., Bindiganavile V. (2014). Fiber synergy in Hybrid Fiber Reinforced Concrete (HyFRC) in flexure and direct shear. Cement & Concrete Composites 48, 91–97. Brandt A.M. (2008). Fibre reinforced cement-based (FRC) composites after over 40 years of development in building and civil engineering. Composite Structures 86, 3–9. di Prisco M., Colombo M., Dozio D. (2013). Fibre-reinforced concrete in fib Model Code 2010: principles, models and test validation. Structural Concrete 14(4), 342-361.

Pająk M. (2016). The investigation on flexural properties of hybrid fiber reinforced selfcompacting concrete. Procedia Engineering 161, 121-126. Stähli P., van Mier J.G.M. (2007). Manufacturing, fibre anisotropy and fracture of hybrid fibre concrete. Engineering Fracture Mechanics 74, 223–242.

31

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Influence of test methodology on the applicability of test results of fibre reinforced concrete for design Karoly Peter Juhasz1, Peter Schaul2, Dr. Ralf Winterberg3 1

: Budapest University of Technology and Economics, Department of Mechanics, Materials & Structures, Műegyetem rakpart 1-3, Budapest, Hungary 2 : Budapest University of Technology and Economics, Department of Construction Materials and Technologies, Műegyetem rakpart 1-3, Budapest, Hungary 3 : EPC Holdings Pty Ltd, Singapore

Abstract The design of fibre reinforced shotcrete (FRS) linings is commonly based on the Q-System or Barton charts. This performance based design approach accesses the results of experimental tests, carried out on panel specimens according to existing standards or guidelines. This is different to the general methodology to access and determine the performance of fibre reinforced concrete (FRC) using standardized beam tests. Panel and beam test results yield significantly different information on the performance of FRC and it is problematic to correlate them. The beam test yields a stress-strain relationship for a small displacement range only. Based on the significantly different working and failure mechanisms, structural tests to evaluate the post-crack performance and the ductility of FRS linings are typically conducted on different types of panels rather than on traditional beams. As a consequence, test results based on beam tests may lead to an overestimation of FRC performance in panels and vice versa. In order to avoid uneconomic designs the most appropriate material must be found using the most appropriate test methodology. To design tunnels with Finite Element Analysis (FEA) software the measured energy absorption values are not sufficient; it is required to determine different concrete-specific and fibre reinforced concrete-specific parameters as well. One of the key parameters is the residual flexural strength, which describes and quantifies the fibre effect in the concrete. This parameter can be determined by using the harmonized European beam test, measuring the load vs. the crack mouth opening displacement (CMOD). This is a three-point bending test with a notch in the centre of the beam span. The process of a panel test is relatively quicker compared to a beam test, with a lower variability of results. The beam tests need more preparation and the results could be misleading because of limited crack propagation and the variability is usually high with COV’s up to over 30%. Laboratory research is presented in this paper, where panels and beams were tested. Results were compared and a numerical FEA model was made to estimate the results of both types of test with the same material model parameters. The possibility of estimating the material parameters from panel tests was then examined. The test matrix can be seen in Table 1. The beam and the panel test specimens were all stored under water. The testing date of the specimens was at the age of 28 days.

32

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Table 1:

Test matrix.

Fibre dosage Square panels Beams

Plain concrete SQ PC 1-3 cast B PC 1-4 cast

2.5 kg/m3 BC48 SQ 2.5 1-3 cast B 2.5 1-4 cast

5.0 kg/m3 BC48 SQ 5.0 1-3 cast B 5.0 1-4 cast

7.5 kg/m3 BC48 SQ 7.5 1-3 cast B 7.5 1-4 cast

3.0 kg/m3 BC54 SQ 3.0 1-3 sprayed B 3.0 1-3 cast

Due to their different working mechanism, the correlation between panel and beam tests cannot be formulated directly, but the test could be modelled using advanced Finite Element Analysis (FEA). The numerical verification was made using ATENA. To be able to model the curling rise of the panel’s corners from the steel formwork, a non-linear interface material was applied between the concrete panel and the centrally located steel loading plate. The material parameters used were determined by inverse analysis.

Figure 1:

Results of the panel (a) and beam tests (b) with their modelling by FEA.

The mean values of the tests and the numerical results of the panels can be seen in Figure 1a. The behaviour and value of the numerical model closely matches the test results for each dosage of fibre. The peak load is almost the same, and the slope of the curve from FEA is close to the test curve for each fibre dosage. Note that the maximum difference in the areas under the curves was only 7%. Using the same material parameters the beam tests were modelled with ATENA and the results can be seen in Figure 1b. The differences in the areas under the curves range from 4% to 18% using modified mean values, and 4% to 10% using mean values. Material parameters can be derived from panel tests, where there is much lower variability of the results than for beam tests. The residual strength parameter is a function of the fibre dosage, which produced a nearly linear function in the test series. Using FEA a correlation can be made between the dosage of fibres and their performance. Even using a linear residual strength model the correlation is acceptable, leading to a proper material model.

33

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Parameters affecting the properties of plain and fiber reinforced Self-Consolidating concrete Armin Javadian 1, Amir Mahdavi 1, Othman Benamrane 1, Muslim Majeed Aoude 1 1

1

and Hassan

: University of Ottawa, Ottawa, Canada

Introduction The addition of fibers to concrete improves its tensile capacity, post-cracking resistance, and toughness. However, the addition of fibers in traditional concrete reduces workability and can result in difficulties in concrete placement, particularly at the higher fiber contents required for structural applications. The combined use of self-consolidating concrete (SCC) and steel fibers has been proposed as a solution to this problem. This research study examines the effect of fiber properties on the fresh-state and hardened-state properties of selfconsolidating fiber-reinforced concrete (SCFRC). As part of the study the effect of fiber content, length, aspect ratio, and hybrid use of fibers is investigated.

Experimental program A total of 16 SCFRC were examined as part of this study. The various SCFRC mixes were cast using the same base mix (see Table 1). The properties of the five fibers included in this study are summarized in Table 2. Fiber content was varied between 0% and 1.6% by volume of concrete. Mix nomenclature reflects the fiber type (ZP, BP, 5D, ML, TUF (see Figure) and fiber content (0%, 0.4%, 0.8%, 1.2% or 1.6% by volume of concrete) in each SCFRC mix. For example, SCC0% represents the control mix, SCC0.8%ZP indicates a mix having 0.8% of ZP steel fibers, while mix SCC0.8%ZPML indicates the combined use of 0.4% ZP and 0.4% ML hybrid fibers. The fresh-state properties of the mixtures were studied using the slump flow test and V-funnel test. A modified fiber funnel, which had larger dimensions when compared to the standard V-funnel was also used (see Figure 2). The consistency and segregation resistance of the mixes was also assessed qualitatively using the visual stability index (VSI) test. The mechanical properties of the SCFRC mixtures were studied by testing standard cylinders in compression and standard prisms in flexure in accordance with the ASTM C1609 standard. Using the flexural test results the toughness T150 was calculated. Table 1:

SCC mix design (base mix).

GUB-8SF (kg/m3)

Slag (kg/m3)

Crushed limestone (kg/m3)

Natural sand (kg/m3)

W/C ratio

Air entrainment (ml/m3)

Super plasticizer (L/100kg/cm)

400

100

810

780

0.44

590

0.48

Table 2:

Fiber type and dimension.

Fiber ID

Type

Shape

Diameter (mm)

Length (mm)

Aspect ratio

Tensile strength (MPa)

ZP

Steel

Hooked-end

0.55

30

55

1100

BP

Steel

Hooked-end

0.3

30

80

2300

5D

Steel

Double hooked-end

0.9

60

65

2300

ML

Steel

Straight

0.3

13

43

2500

TUF

Synthetic

Fibrillating

0.37 x 1.1

51

74

625

34

Figure 1: Steel fibers (from left to right): ML, ZP, BP and 5D fiber.

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Discussion of Results Addition of fibers decreases the workability of SCC, with reduction in mean slump spread and increase in V-funnel time. For example, the slump for the ZP series reduced from 755 mm for the base SCC mix to 690, 640 and 595 mm after addition of 0.4%, 0.8% and 1.2%ZP fibers (see Figure 2). Increasing fiber content in SCC beyond certain limits can lead to loss of mix uniformity (e.g. VSI = 2 at 1.6% ZP fibers and VSI =3 at 1.6% ML). Using fibers with higher aspect-ratio (e.g. BP fibers) and longer length (e.g. 5D fibers) reduces workability and increases the potential for segregation. In contrast, the use of micro fibers (e.g. ML fibers) has little effect on fresh-state characteristics of SCFRC as long as the limiting fiber content is not exceeded. The hybrid mixes followed the same trend observed in the ZP series with a higher level of segregation at 1.6% of fiber content. Addition of low amount of fibers can increase compressive strength of SCC, although this effect is more significant in mixes with micro fibers (e.g. 20% strength gain when using 1.2% ML fibers). At higher fiber contents strength reduction was observed, with a considerable loss in strength when the limiting fiber contents for each fiber type is exceeded. Addition of fibers significantly improves the flexural behaviour of SCC and the enhancement becomes more prominent as the fiber content is increased. Furthermore, SCFRC with higher fiber contents can show deflection-hardening behaviour (with peak load being higher than the first cracking load), leading to further improvements in SCFRC toughness. SCFRC mixes with macro-fibers show higher bending toughness in comparison with mixes with micro fibers (with the latter having greater effect on first-cracking strength). Use of fibers with longer length or increased aspect ratio can result in further improvement in toughness of SCFRC. The use of hybrid fibers was also shown to improve overall toughness of SCFRC, with a synergy effect between the short and longer fibers.

(b) Modified V-funnel dimensions V-funnel time (sec) (Standard)

10

700 600

ZP BP 5D ML MLZP (hybrid) TUF

500

(c) Slump flow test and funnel test

ZP BP ML MLZP (hybrid) TUF

8 6

V-funnel time (sec) (Modified)

800

4 2 0

400 0

0.4 0.8 1.2 1.6 Fiber content (%)

ZP 5D MLZP (hybrid)

60

20

ZP BP 5D ML MLZP (hybrid) TUF

6 4 2

2

0

(e) Stand. V-Funnel results BP ML

40

(f)

0.4 0.8 1.2 1.6 Fiber content (%)

2

Modif. V-funnel results

60

40 MLZ BP TUF

20

ZP 5D ML

0

0 0

0.4 0.8 1.2 1.6 Fiber content (%)

(g) Toughness T150 results

Figure 2:

0.4 0.8 1.2 1.6 Fiber content (%)

Compressive Strength (MPa)

80

8

0 0

2

(d) Slump Flow results Toughness, T150 (J)

Avg. slump flow (mm)

(a) Standard V-funnel dimensions

2

0

0.4

0.8 1.2 Fiber content (%)

1.6

2

(h) Compressive strength results

Fresh state tests and research program results as a function of fiber content/type.

35

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Numerical Validation of a Simplified Inverse Analysis Method to Characterise the Tensile Behaviour of UHPFRC Eduardo J. Mezquida-Alcaraz 1, Juan Navarro-Gregori 1, Juan Ángel López Martínez 1, Pedro Serna Ros 1 1

: Institute of Science and Concrete Technology, ICITECH, Universitat Politècnica de València, València, 46022, Spain.

Abstract Nowadays the characterisation of Ultra-High-Performance Fibre-Reinforced Concrete (UHPFRC) tensile behaviour still remains a challenge for researchers. For this purpose, a simplified closed-form non-linear hinge model based on the Third Point Bending Test (TPBT) was developed by the authors. This model has been used as the basis of a simplified inverse analysis methodology to derive the tensile material properties from load-deflection response obtained from TPBT experimental tests. The aim of this work is the numerical validation of the simplified inverse analysis method to characterise the tensile properties of UHPFRC. To get this objective a Finite Element Model (FEM) is carried out. The parameters to characterize the concrete properties from the simplified inverse analysis method by means of TPBT are used in the numerical modelling. The constitutive model for UHPFRC is modelled using two assumptions. One is based on the smeared cracking approach where a fixed total strain crack model, expressed as function of a crack opening fibre-reinforced concrete fib curve, is used. The other is based on a discrete cracking model for the macrocrack position. Numerical validation accuracy is reasonable for the smeared crack case and excellent for the discrete crack approach.

Keywords UHPFRC, hinge model, TPBT, numerical validation, FEM

Introduction A 2D finite element numerical model is presented in this paper with the objective of validating the non-linear hinge model (López et al. 2016) to derive the tensile UHPFRC’s properties from load-deflection response obtained from TPBT experimental tests. The Simplified Five-Point Inverse Analysis Method (5P-IA) is fully described in (López et al. 2016). This new simplified methodology is based on the closed-form non-linear hinge model developed in (López 2017; López et al. 2016).

Numerical model and experimental validation The non-linear closed-form hinge model and its derived simplified 5P-IA method have been validated resorting to a robust non-linear finite element modelling and a set of TPBT with variable depth, slenderness and hinge length (López 2017).

36

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

In this work, a numerical modelling is developed using the FE software DIANA. To model the tensile UHPFRC constitutive behaviour two different approaches are used: a smeared cracking approach and a discrete cracking approach. The discrete cracking approach of the numerical model is more adequate for modelling the tensile behaviour of UHPFRC in a TPBT. When using the tensile parameters resulting from the iterative inverse analysis method (I-IA) using the closed-form non-linear hinge model presented and its derived simplified 5P-IA method, the discrete cracking approach describes accurately: (a) the load deflection response, (b) the bending curvatures, and (c) the average longitudinal strains measured in the bending tests (Figure 1). As a result, inverse analysis methodologies based on the closed-form non-linear hinge model and the derived simplified 5P-IA method proposed can be recommended to obtain UHPFRC’s tensile properties in Third-Point Bending Tests.

Figure 1:

Normal stresses (a); Equivalent bending stress -displacement at mid-span (b).

References López, J. Á. (2017). “CHARACTERISATION OF THE TENSILE BEHAVIOUR OF UHPFRC BY MEANS OF FOUR-POINT BENDING TESTS.” Universitat Politècnica de València, Valencia (Spain). López, J. Á., Serna, P., Navarro-Gregori, J., and Coll, H. (2016). “A simplified five-point inverse analysis method to determine the tensile properties of UHPFRC from unnotched four-point bending tests.” Composites Part B: Engineering, 91(Supplement C), 189–204.

37

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Optimization of fibre combination using latest generation of steel and polyolefin fibres M. G. Alberti1. J.C. Gálvez1. A. Enfedaque1. A. Cortez1 1

Departamento de Ingeniería Civil: Construcción. E.T.S de Ingenieros de Caminos. Canales y Puertos. Universidad Politécnica de Madrid. C / Profesor Aranguren. s/n. 28040. Madrid.

Abstract Concrete has certain limitations like its low ratio tensile-compressive strength and its low fracture energy, behaving as a quasi-brittle material. Hence, in the last decades a significant number of researchers have carried out a wide range of campaigns, models as well as modifications and combinations with other materials with the only aim of improving concrete tensile characteristics. Among the various alternatives proposed in literature, the incorporation of randomly distributed fibres into the concrete matrix has enabled to reach residual load-bearing capacities that can be considered in the structural design. These structural fibres can be synthetic or made of steel. The presence of fibres improves the response of concrete under tensile and flexural stresses and can replace the use of rebars to reinforce concrete. Fibres suppose not only an increase of toughness but also they can control the development and propagation of cracks and therefore enhance its durability. This study aimed to characterize a high-performance self-compacting concrete reinforced with the latest generation steel fibres and the combination of these with polyolefin fibres through a series of fracture tests. Three-point bending tests followed EN-14651 and permitted assessing the post-cracking behaviour and, thus, the structural contributions of these types of fibres. The significance of this research relies in the optimization of several fibre cocktails that combined three steel fibres types and polyolefin fibres. The results showed that it is possible to achieve a fibre combination that can take advantage of the best performance of each fibre type.

Keywords Self-compacting, fibre reinforced concrete, fracture, residual strengths

Introduction From several variations of the shapes, sizes and materials it is possible to achieve enhancements in the behaviour of fibre reinforced concrete and to meet a wide range of requirements. In addition, it is possible to find a combination of fibres that can be optimized as a function of the requirements and final usage of fibre reinforced concrete. Given the new advances in steel and polymer fibres, their possibilities as well as new possible combinations were still pending in literature. This is the case of polyolefin fibres or steel fibres with improved mechanical properties (OL) or new shapes (5D). In this study, 3D, 5D, OL and polyolefin fibres have been used as a reinforcement of a self-compacting concrete matrix. Additionally, 5D and OL fibres were combined keeping constant the fibre dosage in 70 kg/m³. With the same aim, another mixture combining 5D, OL and polyolefin fibres in the same volume proportions was manufactured keeping constant the total volume fraction of fibres. The significance of this 38

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

research relies on the assessment of the fresh and hardened state properties of the mixtures with the most advanced steel fibres and the two mixtures reinforced with the fibre cocktails. This is of special relevance in what regards to residual tensile strengths that are the values used for the structural design of concrete elements in Ultimate Limit State (ULS).

Experimental program, results and main conclusion In this study, three types of steel fibres and one of polyolefin were used: 5D steel fibres (considered latest generation technology because they combine an optimized shape of the pins, a great wire ductility and high fibres tensile strength), OL fibres (which are characterized by high performance and by their high resistance to cracking that make them ideal for optimal ductility) and 3D fibres (the most common fibres). The mix proportioning was carefully studied in order to reach self-compacting properties even with dosages of fibres of 0.89% in volume or 70 kg/m³ in weight of steel fibres. The fresh properties and the main mechanical properties compressive strength, indirect tensile strength and elasticity modulus were assessed with the standard tests. In order to assess the flexural tensile strength, fracture tests following EN 14651 were carried out on three prismatic specimens of each type of concrete. Figure 1 and Figure 2 show the curves load vs. crack mouth opening displacement (CMOD) of the two concrete mixtures with a combination of fibres.

Figure 1: LOAD-CMOD curves of 5D-OL specimens.

Figure 2: LOAD-CMOD curves of 5D-OLPOLY specimens.

Concluding notes • • • •

At crack openings of 5 mm, the combination 5D-OL fibres showed the highest residual performance. The maximum value of peak load was obtained with the concrete reinforced with 3D fibres (around 42 kN) while the concretes reinforced with OL fibres and with the combination of 5D-OL-POLY were the ones that reached peak loads around 25 kN. If we compare the use of 5D and OL fibres separately or combined in equal proportions, it was observed that together they optimize their behaviour as concrete reinforcement. The combination of 5D and OL steel fibres with polyolefin showed an outstanding stable residual branch with almost bilinear behaviour.

39

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

An application oriented state-of-art and research-need perspective on self-healing fibre-reinforced cementitious composites Liberato Ferrara1, Estefania Cuenca Asensio1, Francesco Lo Monte1, Didier Snoeck2 and Nele De Belie2 1

: Department of Civil and Environmental Engineering, Politecnico di Milano, Italy. : Magnel Laboratory for Concrete Research, Faculty of Engineering and Architecture, Ghent University, Belgium. 2

Abstract The synergy between fibre-reinforced cementitious composites and self-healing techniques may result into promising solutions. Fibres improve the self-healing process due to their capacity to restrict crack widths and enhance multiple crack formation. In particular, cracks smaller than 30-50 µm are able to heal completely. Moreover, in the case of HPFRCCs, high content of cementitious/pozzolanic materials and low water-binder ratios are likely to make the composites naturally conducive to self-healing. A state-of-the-art survey on self-healing of fibre reinforced cementitious composites will be provided, which analyses the current knowledge with the goal of providing a “healable crack opening based” design concept which could pave the way for the incorporation of healing concepts into design approaches for FRC and also conventional R/C structures. On the other hand, the same state-of-the-art will be instrumental at identifying research needs, which still have to be addressed to carry out a proper use of self-healing fibre-reinforced cementitious composites in the construction field.

Keywords Self-healing; fibre reinforced cementitious composites; test methods; healable crack width

Introduction In current design codes, it is appropriately recognized that the achievement of the required durability is the complex outcome of the suitable choice of structure concept and shape, material selection, as well as of the enforcement of “operational” design criteria, which limit the crack width under the anticipated actions to suitable scenario-based threshold values. In order to provide effective control of the crack width, Fibre Reinforced Concrete (FRC) has been developed over the past fifty years, pushing ahead the boundaries of this concept up to the formulation of the so-called High Performance Fibre Reinforced Cementitious Composites (HPFRCCs), whose composition is designed through micro-mechanical concepts based on fibre pull-out and crack tip toughness balance. Thanks to this, after the formation of a first crack, fibres effectively provide a through crack stress redistribution, which enables new multiple cracks to be formed while controlling the opening of the previously formed ones, which are basically stopped from further widening, up to the unstable localization of one major

40

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

crack. This makes the material able to “spread” the entity of a single damage (crack) into a series of tightly spaces and narrowly opened multiple cracks, whose single width is hence possibly much less detrimental to the structural durability. The composition of HPFRCCs features high contents of cement and supplementary binders, with either pozzolanic (fly ash, silica fume) or delayed cementitious activity (slag) and low water content. The resulting high amount of reactive material, which remains un-hydrated and entrapped inside the bulk volume of a structural element may be, upon cracking exposed to outdoor environment, in case featuring presence of liquid or vapour water. These can both activate delayed hydration reactions as well as carbonation ones, whose products precipitate onto the crack surfaces sealing it. The reconstruction of the through crack matrix continuity and, in case, matrix densification at the interface with the fibres, which also yields improved fibre-matrix bond, may also result into “proprie dictum” material healing, i.e. a recovery of the post-cracking mechanical performance, in terms of load bearing capacity, stiffness, toughness and ductility etc.. The presence of a homogenously dispersed fibre reinforcement, effectively controlling and limiting the crack width, is helpful to self-healing, since narrower cracks are definitely much easier to be healed. Different techniques have been proposed and validated to stimulate and enhance the aforementioned autogenous healing capacity, making its effectiveness less scattered and more reliable. The most consolidated ones consist in the use of crystalline admixtures and superabsorbent polymers (SAPs).

The “healable crack” concept In a real structural design framework, the “crack width” is the calculated output of a structure service scenario, which encompasses the anticipated type and level of actions, the exposure conditions, the structure dimensions and reinforcement details as well as the material characteristics and performance. All the aforementioned variables contribute to define an “accepted” serviceability and durability scenario, one of whose distinctive parameters is the crack width itself. The synergy between FRC and self-healing concrete leads to rethink the same concept of serviceability crack opening-based design through the formulation of a “healable crack width” concept. To substantiate from a design-wise operational point of view the aforementioned concept, actually depending on a multi-fold set of variables, the following needs hold: - A sound experimental methodology aimed at quantifying the healable crack as a function of the material, the anticipated service scenario and the required performance as a function of the intended structural use. Such a quantification has to rely upon a “holistic” characterization of the healing, encompassing crack sealing, recovery of durability related and mechanical performance and nano- to micro-scale characterization of healed cracks and healing products. - A multi-physics based model of the evolution of the material performance, including degradation, aging and healing, coupled with a model of the uncertainty which enables to predict with “deep enough” confidence the time evolution of the structural performance and provide output for LCA/LCC/SLCA evaluation of the benefits of using self-healing fibre reinforced cementitious composites in building new and/or retrofitting existing structures.

41

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Early-age behaviour of High Performance Fiber Reinforced Concretes Valeria Corinaldesi1, Jacopo Donnini1, Glauco Merlonetti1 1

: Department of Materials, Environmental Sciences and Urban Planning, Università Politecnica delle Marche, Ancona, Italy

Abstract The aim of this experimental activity was to study the early-age behavior of several High Performance Fiber-Reinforced Concretes (HPFRCs) containing expansive agent. In particular, the investigation concerned the evaluation of the influence of different amounts of fibers (dosages of 2.0%, 1.75% and 1.5% by volume of HPFRC) on the mechanical performance of HPFRCs. The attention was focused on the strength development at early ages. Mechanical tests were carried out at 0.25 (i.e. 6 hours, that is time of demolding), 1, 2, 7 up to 28 days of curing. The properties of HPFRCs were characterized at the fresh state, by measuring flow ability and consistency as well as at hardened state by measuring compressive and flexural strength up to 28 days. The different dosage of fibers did not influence the values of compressive strength, while there is a significant difference in terms of 28-day flexural strength between the several mixtures depending on the different amount of fibers. In all cases at least 20 MPa of 28-day peak flexural strength were achieved.

Keywords Advanced Materials, Building, Early age, Fiber Reinforced Concrete, High Performance Concrete, HPFRC

Introduction This work was aimed at evaluating the influence of different amounts of steel fibers on the mechanical properties of HPFRCs prepared with CaO-based expansive agent. Mixtures were prepared by varying the dosage of hooked brass-coated fibers: ranging from 1.5% to 1.75% up to 2% by volume of concrete, respectively. In particular, the attention was focused on the mechanical performance at early ages: 6 hours (time of demoulding), 1, 2, 7 and 28 days of wet curing. The experimental data collected after only 6 hours are important for likely application in precast concrete plant.

Experimental program All the mixtures were prepared with the same w/c ratio of 0.35 and inert/cement ratio of 2.2, as well as the same dosage (2.5% by weight of the cement) of a 26.8±1.3% aqueous solution of a polycarboxylate-based water-reducing admixture. Their mixture proportions are shown in Table 1.

42

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Table 1:

Mixture proportions (kg/m3). Ingredients Cement

Br-2% 600

Mixtures Br-1.75% 600

Br-1.5% 600

Water

210

210

210

Sand (0-4 mm)

1320

1320

1320

Superplasticizer

15

15

15

Silica Fume

110

110

110

Brass fibers

155

135

115

Limestone filler

70

70

70

Expansive Agent

40

40

40

The dosages of fibers did not influence the compressive strength; in all cases a 1-day compressive strength value of at least 50 MPa was achieved, while the 28-day compressive strength was in the range of 115-120 MPa, independently of the amount of fibers. In terms of flexural strength at very early ages (up to 24 hours) no difference was detected, while after 7 days of curing a difference of about 3 MPa between the mixtures with minimum (1.5%) and maximum dosage (2%) of fibers was observed. However, at least 13 MPa of 1-day flexural strength was reached in all cases, as well as a minimum 28-day flexural strength of 20 MPa. Residual flexural strength at a displacement of 3.5 mm shows to be independent of the amount of fibers up to 7 days of curing, then the higher dosage of 2% by volume seems to promote a better post-cracking behaviour.

43

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental and numerical analysis of fiber reinforced concrete beams in real scale with steel bars reinforcement Marianela Ripani1, Paula Folino1, Hernán Xargay1,2, Nicolás Rocca1 1

: Universidad de Buenos Aires, Facultad de Ingeniería, Laboratorio de Métodos Numéricos en Ingeniería (LMNI), Instituto INTECIN (UBA-CONICET), Buenos Aires, Argentina

2

: Comisión Nacional de Energía Atómica (CNEA), Departamento ICES, Buenos Aires, Argentina

Abstract The aim of this work is to present some results of an extensive experimental campaign involving reinforced concrete beams in real scale. Both steel reinforcing bars and industrial steel fibers are considered. This research was motivated by the fact that although a very important number of scientific papers related with fiber reinforced concrete can be found in the literature, scarce is the number of works considering real scale fiber reinforced concrete specimens. Regarding available design recommendations, the contribution of new experimental data could contribute to improve actual design equations. In this experimental campaign, a reference concrete mixture designed for a target 28 days mean cylindrical compressive strength of 30 MPa was used. Moreover, two different fiber contents of 40 kg/m3 and 60 kg/m3 were considered. Wirand FF3 end hooked steel fibers were selected. A total of twenty-four reinforced concrete beams of dimensions 120 mm (width), 300 mm (height) and 2400 mm (length) were casted for the experimental campaign. These beams were tested in a four-point loading configuration with a span of 2100 mm. Besides the variable fiber contents, three different reinforcing bar layouts were designed and adopted, corresponding to three expected cases of failure that will be detailed below. The experimental campaign also included compressive and splitting tensile tests on cylindrical samples and three-point bending tests on small notched beams. Case (1) Ductile bending failure tension-controlled: two 12 mm diameter reinforcing bars and 6 mm diameter closed shear stirrups, at a spacing of 150 mm (6@15). Nine beams were poured, corresponding three specimens for each fiber content. Case (2) Shear failure: four 12 mm diameter reinforcing bars and only three stirrups (6@80). Nine beams were casted, corresponding three specimens for each fiber content. Case (3) Brittle bending failure by tension: no reinforcing bars. Six beams were poured, including three with a fiber content of 40 kg/m3and three with 60 kg/m3. The experimental results of bending tests performed on full-scale beams are depicted in Figure 1. Figure 1a shows the Load-Deflection curves obtained for all the analyzed beams. It can be seen that fiber reinforced beams presented less dispersion on test results than plain concrete beams. On the other hand, in Figure 2b only mean curves corresponding to Case 1 are shown, it can be observed that yielding deflection appeared to decrease as fiber content increased and only a small increment in peak load was observed for higher fiber contents. Regarding experimental results of Case 3, it is concluded that no bar reinforced beams seems to be not suitable for real structural purposes regarding the extremely localized failure mode observed. Therefore, from the experimental results, it can be concluded that the benefits of the 44

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

addition of fibers were particularly related with a reduction of the dispersion of results and with structural integrity in the post peak behavior. Last, a comparison between experimental and numerical ultimate loads was performed, being the latter based on RILEM recommendations for obtaining the design parameters from the three-point beam tests (see Fig. 2). From comparison between peak loads, it can be highlighted that in Case 1 beams (failure caused by bending) theoretical and experimental values resulted almost the same for fiber reinforced concrete, while in Case 2 beams (shear failure) numerical predictions were more conservative and a higher residual strength was evidenced. On the other hand, in both cases, it was observed that curvature decreases when fiber content increases. However, this reduction in curvature is not as large as estimated in available design guidelines.

(a) Figure 1:

Load–Deflection curves corresponding to full-scale beam bending tests: (a) All the considered specimens and (b) only mean results for beams Case 1.

100 90

(b)

P u [kN]

100

Theoretical

90

Experimental

80

80

70

70

60

60

50

50

40

40

30

30

20

20

10

Fibers [kg/m3]

0 0

Figure 2:

40

60

P u [kN]

Theoretical Experimental

10

Fibers [kg/m3]

0

0

40

60

(a) (b) Comparison of experimental and numerical predictions of ultimate loads for (a) Case 1 and (b) Case 2 of reinforcement, respectively.

Keywords Fiber reinforced concrete, structural design, flexural strength, shear strength

45

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

An experimental investigation on the post-cracking behaviour of Recycled Steel Fibre Reinforced Concrete Cristina Frazão1, Joaquim Barros1, J. Alexandre Bogas2, Kypros Pilakoutas3 1

: ISISE, University of Minho, Guimarães, Portugal. : CERIS-ICIST, Instituto Superior Técnico, Technical University of Lisbon, Portugal. 3 : The University of Sheffield, United Kingdom. 2

Abstract For assessing the potentialities of recycled steel fibres (RSF) as concrete reinforcement, an experimental program was performed in the present work by comparing the following properties of concrete reinforced with industrial steel fibres (ISF) and with RSF: compressive strength, modulus of elasticity, flexural strength, flexural toughness and indirect tensile strength. To evaluate the corrosion effects on the post-cracking response of Recycled Steel Fibre Reinforced Concrete (RSFRC), double edge wedge splitting tensile tests were conducted in RSFRC specimens previously exposed to aggressive chloride environment.

Keywords Recycled steel fibres, industrial steel fibres, RSFRC, post-cracking behaviour, corrosion.

Introduction Steel fibres resulting from the industry of tyre recycling have high potential as an effective reinforcement of concrete, especially in terms of improving its post-cracking tensile behaviour, and increase its flexural, shear and impact strength (Micelli et al. 2014). In this context, this study reports the results of an experimental research carried out at the University of Minho (Portugal) that aims to investigate the post-cracking behaviour of RSFRC and perform its comparison to the one registered in Industrial Steel Fibre Reinforced Concrete (ISFRC), by performing 3-point bending tests (3PNBBT), round panel tests (RPT-3ps) and double edge wedge splitting tests (DEWST). The corrosion effects on the post-cracking response of RSFRC previously subjected to chloride attack were also analysed from DEWST.

Experimental program The post-cracking behaviour of RSFRC1% and ISFRC1% (equal composition with 1% of fibres by volume of concrete) was analysed using different test methods, namely, 3PNBBT according to the CEB-FIP Model Code recommendations (2013), RPT-3ps according to the recommendations of ASTM C1550 (2005) and DEWST according with a new test method (Lameiras et al., 2015). Figure 1a represents the average force-deflection response registered in the plain concrete (PC), RSFRC1% and ISFRC1% by executing 3PNBBT. In Figure 1b is depicted the average force-central deflection relationship of RSFRC1% and ISFRC1% panels obtained by performing RPT-3ps. Figure 1c shows the average splitting tensile stress  t ,split versus crack

46

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

9

27

40

18

30

PC RSFRC ISFRC

21

7

18

6

15

5

12

4

9

3

6

2

3

1

0 0.00

0.03

0.06

0.09

0.12

0.15

0.18

16 14 12 10

0 0.21

8 6 4

30

1.0

1.5

2.0

2.5

3.0

3.5

0

10

15

20

25

30

35

40

45

50

Central deflection (mm) 5.0

RSFRC ISFRC

10 days Cl

4.5 4.0

3.5

3.5

(MPa)

4.0

3.0

t,split

2.5 2.0

-

Class 1 (reference) - REF Class 1 (reference) - COR Class 2 (350ºC) - REF Class 2 (350ºC) - COR Class 3 (polished) - REF Class 3 (polished) - COR

3.0 2.5 2.0

1.5

1.5

1.0

1.0

0.5

0.5

0.0 0.0

5

(b)

4.5

(MPa)

15

0

5.0

t,split

20

5

0 4.0

Deflection (mm)

(a)

25

10

2 0.5

RSFRC ISFRC

35

8

24

Force (kN)

60 55 50 45 40 35 30 25 20 15 10 5 0 0.0

Flexural stress (MPa)

Force (kN)

width curves for RSFRC1% and ISFRC1% obtained in DWEST. Figure 1d shows the average  t ,split versus crack width curves for specimens of 3 different classes of RSF considered to analyse the influence of the small rubber debris attached to RSF surface (Class 1 – Reference RSF, as were received; Class 2 – RSF pre-treated at 350ºC; Class 3 – Polished RSF).

0.0 0.5

1.0

(c)

1.5

2.0

CMOD (mm)

Figure 1:

2.5

3.0

0.0

3.5

(d)

0.5

1.0

1.5

2.0

2.5

3.0

3.5

CMOD (mm)

(a) Average force/flexural stress versus deflection (3PNBBT), and (b) Average force versus central deflection (RPT-3ps), (c) Average splitting tensile stress versus crack width for specimens of RSFRC1% and ISFRC1%, and (d) Average splitting tensile stress versus crack width for specimens with RSF of class 1, 2 and 3, not exposed to NaCl solution (REF) or under corrosion action (COR).

From 3PNBBT and RPT-3ps, RSFRC exhibited higher flexural strength and energy absorption capacity than ISFRC of equal composition. No significant differences were observed between the post-cracking tensile behaviour of RSFRC and ISFRC in DEWST. The corrosion action caused a slight decrease of the average post-cracking tensile strength of the RSFRC with reference RSF. The small percentage of rubber attached to RSF surface had a negligible effect in the corrosion resistance of RSFRC.

References Micelli F., Leone M., Centonze G., Aiello M. (2014). Chapter: Go Green: Using Waste and Recycling Materials, Infrastructure Corrosion and Durability Sustainability Study, Edited by Yang Lu, OMICS Group eBooks, USA. CEB-FIP Model Code 2010 – Volume 1. Tomas Telford, Lausanne, Switzerland, 2013. ASTM C1550-05 (2005). Standard test method for flexural toughness of fiber reinforced concrete (using centrally loaded round panel). ASTM International. Lameiras R., Barros J., Azenha M. (2015). Influence of casting condition on the anisotropy of the fracture properties of SFRSCC. Cement & Concrete Composites, 59, 60-76.

47

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

48

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Durability

49

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Influence of chloride corrosion on the surface aspect of steel fibre reinforced cementitous composites Mylene M. Vieira1, Sergio H. P. Cavalaro2, Antonio Aguado.1 1

: Polytechnic University of Catalonia, Barcelona, Spain. 2 : Loughborough University, Loughborough , United Kingdom.

Abstract The corrosion of steel fibre at surface of a cement composite brings a damaging visual effect regarding the aesthetic of the structure. For high performance steel fibre reinforced cementitious composites (HPSFRCCs), higher fibre content is added into the mixes with the aim of improving characteristics such as, crack control, post-peak structural behaviour and ductility. The increment of fibre content into the mix may increase the amount of fibres close to the surface of the structures. In spite of the dense matrix of HPSFRCCs, a level of porosity at surface and also the presence of cracks may allow the ingress of deleterious substances. For the enhanced properties and durability, these materials have mainly applications in infrastructures subjected to aggressive environments. Marine structures, particularly in the splash and tidal zone is one of the preferable engineering applications. When exposed to cyclic conditions with the presence of chlorides, a level of the damage regarding steel fibre corrosion may occur throughout the life-cycle of the structure. It has been previously proved that steel fibres present and excellent durability against corrosion in uncracked cementitious composites. In aggressive conditions where conventional steel bars show a high rate of corrosion, the steel fibres still with no sign of damage. However, the fibres randomly distributed into the cementitious matrix lie on the surface of the structure with almost no protective cover from the environment. So the chlorides, the oxygen and humidity availability at surface may penetrate through the pores and react with the iron of the fibres localized at the superficial layer of the structure. Moreover, the wetting and drying cyclic condition may increase the corrosive process for the increment of oxygen and humidity over cycles. Firstly, during the wetting the ingress of humidity and chlorides into the pores at surface may occur due to the capillary suction. Then, during the drying the water tends to evaporate from the pores and the chlorides precipitate out of the deeper pores to the pores at surface increasing its concentration in this layer. Also, the oxygen diffuses fast for the dryness of the surface. Finally, all these mechanisms may result in a level of corrosion of the steel fibres over time. Few studies in the literature focus on establishing relationships between the fibre content and the surface corrosion of specimens when subjected to aggressive conditions. So, further research is required to fully understand the behaviour of high content of steel fibre in HPSFRCCs structures subjected to cyclic process of chloride exposure. The objective of this study is to evaluate the correlation between the fibre content and the level of surface corrosion from a quantitative standpoint. Two mixes of high performance cementitious composites with different steel fibre content (90 and 190 Kg/m3) and the same water/cement ratio were designed. Prismatic specimens were 50

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

cast and put through a cyclic process of wetting (48 hours submerged in salt water) and drying (48 hours at ambient temperature). To ensure the same condition of the specimens in the beginning of tests and over cycles, firstly the specimens were subjected to two cycles of wetting (48 hours submerged in distilled water) and drying (48 hours at ambient temperature). At each 10 cycles, pictures were taken of the surface of each specimen and the area of corrosion was measured by means an algorithm developed for the analysis. Results showed that the increase of fibre content is correlated with the increment of surface area of corrosion. Besides that, it was observed that in HPSFRCCs subjected to chlorides, the corrosion occurs in the steel fibres localized in a very thin layer of the specimens close to the surface.

Keywords: Cement composite, Steel fibre, Corrosion, Aesthetic aspect

51

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

The effect of fibres on corrosion of RC elements Berrocal, Carlos G.1,2, Lundgren, Karin1, Löfgren, Ingemar1,2 1

: Department of Architecture and Civil Engineering, Division of Structural Engineering, Chalmers University of Technology, Gothenburg, Sweden. 2 : Thomas Concrete Group, Gothenburg, Sweden.

Abstract In the present paper, long-term experiments involving natural corrosion of RC beams subjected to chloride solution cyclic exposure were carried out to investigate the effect of fibres on different aspects of the corrosion process as well as their contribution to the structural behaviour of RC elements damaged by corrosion. The long-term experiments were complemented with short-term accelerated corrosion experiments and mechanical tests to investigate the influence that low fibre contents may have on individual mechanisms that play an important role in the corrosion process of steel in concrete. These showed that fibres promote crack branching which results in a change of the internal crack pattern towards multiple thinner cracks, particularly near the reinforcement. This agrees with the long-term experiment results, which exhibited longer times to corrosion initiation for FRC beams with bending cracks and revealed a more distributed corrosion with more pits but less cross-sectional loss compared to bars in plain concrete. Fibres also proved beneficial in delaying corrosion-induced cracks and preventing cover spalling, which greatly enhanced the bond-behaviour of corroded bars. Furthermore, a positive effect of the fibres was also observed on the residual flexural capacity of corroded beams, which generally increased the load-carrying capacity and rotation capacity.

Keywords Chloride-induced corrosion, durability, cracking, reinforcement bond, residual flexural capacity

Introduction Despite the increasing interest in using FRC in a broader range of structural applications, due to the casting-dependent distribution and orientation of the fibres throughout the concrete matrix, in many cases fibre reinforcement may be only used in combination with conventional reinforcement bars. One of the main advantages of combining FRC and conventional rebars is an improved crack control leading to narrower and more closely spaced cracks, which could be beneficial to delay the initiation of reinforcement corrosion. However, studies addressing the potential impact of fibres on the corrosion process of conventional reinforcement and on the structural behaviour of RC elements with corroding reinforcement are very scarce. A sound understanding of these aspects, which are experimentally investigated in the present paper, is essential to achieve a generalized deployment of FRC into a broader range of structures, including those susceptible to suffer corrosion damage.

52

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental programme The experimental programme presented in this work is comprised of three different studies, the main study consisting of long-term experiments, referred to as experimental study A, and two short-term complementary experiments, referred to as experimental study B and C, respectively (see Figure 1). Experimental study A was intended to encompass the different stages of the corrosion process while reproducing, as much as possible, realistic conditions. These conditions included the combination of naturally induced accelerated corrosion, flexural cracks formed under mechanical loading, varying crack widths and different loading conditions. The short-term experiments, on the other hand, were designed to isolate the effect of fibres on specific parameters relevant to the corrosion process, namely the effect on the internal crack morphology (study B) and on the initiation of corrosion-induced cracks as well as on the bond behaviour between concrete and corroding reinforcement (study C).

Figure 1.

A summary of the experimental programme including experimental studies A, B and C.

53

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Influence of the Post-Cracking Residual Strength Variability on the Partial Safety Factor de la Fuente, Albert1; Cavalaro, Sergio Henrique2; Bairán, Jesús Miguel1 1

: Polytechnic University of Catalonia (UPC), Barcelona, Spain. 2 : Loughborough University, Leicester, United Kingdom.

Abstract The use of fibre reinforced concrete (FRC) in segmental linings built with tunnel boring machines (TBM) has increased in past years. So far, more than 50 tunnels have been successfully constructed with FRC. As, in service, the tunnel segments are mainly loaded in compression, their reinforcement is designed to avoid cracking during transient and handling situations. Design formulations for FRC exist in current design codes considering the semiprobabilistic approach and Ultimate-Limit-State (ULS) verifications. Usually, the strength reduction factor for fFtu (ultimate residual strength of FRC) is assumed the same as for concrete in compression (fck); although, the coefficient of variation of fFtu can double that of fck. In this paper, we propose a calibrated material safety factor for fFtu. that takes into account the influence of the element size. We assess the error of the design model through comparison with test on real-scale segments. Further, we calibrate the model safety factor for varying target reliability indexes using the First Order Reliability Method (FORM). The findings provide a basis to support the revision of safety considerations found in codes in guidelines and may serve as a reference for future studies for other types of structures made with FRC. Intrinsic scatter of fiber reinforced concrete (FRC) strength In previous studies in the literature, it was highlighted that the FRC strength measured on small-scale specimens may not be representative on real-scale elements. The sources of material variability has been divided into those related to production process, testing process and intrinsic variability of the material. In the paper, the influence on the size of the segment on the intrinsic variability was investigated with a numerical model that considered nonuniform distribution of the material properties in the element through a Monte Carlo simulation. Different element’s depth and width were considered and stable results of the intrinsic C.V. after 300.000 simulations for each specimen size. It was found that the intrinsic C.V. in the real-scale can reduce more than 50% with respect to the reference value of the small-scale test. The real-scale C.V. tends to stabilize for b=1500mm. A model for the size effect on the intrinsic variability was calibrated and the total variability of the structure was accounted for considering the influence of the other effects (production and testing) as an independent process. To calibrate the material safety factor, the model bias was assessed through a database of 23 real-scale experimental tests on precast segments from the literature. All elements were fiber reinforced without ordinary steel bars and tested in flexure. Their size were representative of actual conditions in tunnel segments, with b varying between 1000mm to 1800 mm and thickness from 235 to 400 mm. The ratio of the experimental observation to model prediction was found to have an average value of 1.04 and a C.V. of 33%.

54

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Further, the reliability analysis was conducted through a FORM analysis. The random variables considered were the model bias, the concrete flexural strength, the geometric error and the post-cracking FRC strength for CMO=2.5 mm (fR3). Considering the design equation (1), the material safety factor (γfR3k ) was optimized for a range of target reliability indexes and representative design set were concrete class varied from C25 to C60 and the element’s thickness varied between 200 and 400 mm. 𝑀𝑢 = 𝛾

𝑓𝑅3 𝑏ℎ2

𝑓𝑅3𝑘

6

≥ 𝑀𝑐𝑟 =

𝑏ℎ2 6

𝑓𝑐𝑡,𝑓𝑙

(1)

A relationship between the calibrated safety factor, the target reliability and the C.V. of the FRC was found, Fig. 1. Equation (2) provides a suitable model to define the safety factor in terms of the target reliability index and the C.V. The effect of the size of the specimen is taken into account through its influence on the value of the C.V. as found in the proposed model. 𝛾𝑓𝑅3𝑘 = 0.80𝑒

Figure. 1.

0.56𝛽∙𝐶𝑉𝑓0.27 𝑅3

(2)

Relationship between β – γfR3k for FRC tunnel segments subjected to flexure in ULS.

It is highlighted that the currently used safety factor (1.5) will only justify a nominal target reliability () between 1.5 to 2.3. For the typical target index in ULS, γfR3k may exceed 2.5. However, it should also be noticed that no load-safety factor has been used in the load part of Eq. (1). Nevertheless, it is evident that the current design approach should be revised.

55

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Comparability of bond tests for repair and retrofit of concrete structures with Fiber Reinforced Concrete Cristina Zanotti1, Norbert Randl2, Parisa Setayesh Gar1, Bardia Kabiri Far1, Martin Steiner2, 1

: University of British Columbia, Vancouver, Canada. 2 : Carinthia University of Applied Sciences, Villach, Austria.

Introduction Repair and retrofit of concrete structures requires appropriate and durable bond between old and new concretes. Fiber reinforcement enhances substrate-repair bond strength and durability, beyond improving the overall repair/retrofit performance. Nevertheless, such benefits are not fully utilized mostly due to complex interrelations among different parameters and to uncertainties on the appropriateness of common bond tests. As part of a larger research collaboration aimed at addressing those issues, an experimental study on interfacial bond strength between concrete substrates and different types of repair materials was conducted in two independent labs in Canada and Austria, focussing on the effect of fibers, different repair properties and various testing methods.

Experimental program A concrete substrate with compressive strength fc = 60-65 MPa, a normal strength (NS) repair mortar with similar strength and a high strength (HS) repair concrete (fc ≈ 90 MPa) were adopted. The repairs were applied plain (NS-0%, HS-0%) and with 0.5% volume fraction (Vf) of 13 mm steel fibers (NS-0.5%, HS-0.5%). Repairs were cast against substrates aged at least 28 days and were cured for another 28 days before testing. The interface was sandblasted (Mean Texture Depth = 1.3-1.5 mm) and prepared in Saturated Surface Dry (SSD) condition before casting the repair. The different test set-ups adopted for tensile and shear bond are shown in Fig. 1.

Results Bond failures were observed with all test set-ups except for slant shear test (Fig. 1e-f) with HS concrete (material compressive failure) and overlay-type pull-off test (Fig. 1c, substrate tensile failure). Highest tensile bond was obtained from splitting tests as expected (Fig. 1b). In direct tension and pull-off tests, data can be more sensitive to localized stress concentrations and drilling-induced damage. Among those 3 set-ups (Fig. 1a,c,d), the one shown in Fig. 1d offered the highest bond strength values. In shear tests, bond strength decreased for decreasing levels of compression at the interface. As compression and associated frictional/interlocking effects decreased, the cohesive component of bond played a more important role and steel fibers were more effective in enhancing bond strength. When comparing same slant angles, slant shear tests exhibited sensitivity to cylinder size and proportions (distance between loading edges and interface), affecting both bond strength and failure mode. This is due to specimen size effect on the mechanical response of materials at and nearby the interface and to the effect of same roughness 56

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

on shear bond of interfaces with different areas, but is also associated with local confinement due to friction in the loading areas and ensuing stress distributions though the cylinder and the interface. Interfacial shear and normal stresses at failure from slant tests with different bond angles (different shear-normal stress ratios) are plotted in Fig. 2. Overall, bond was higher in HS repair concretes and in concretes with fibers. By comparing cohesion and friction coefficients extrapolated with the Mohr-Coulomb approach, one can appreciate the beneficial effect of repair strength and fiber reinforcement on cohesive shear bond. Similarly, fibers were beneficial to the direct shear bond strength assessed with push-out test (Fig. 1h), but the values obtained with this test were lower than those form slant shear test (even when compared to extrapolated cohesion), mostly due to pronounced frictional/interlocking effects in the slant test, and slight bending due to small eccentricities in the push-out test.

Figure 1:

Bond tests performed (all dimensions in mm).

Figure 2:

Average interfacial shear and normal stresses at failure from slant shear tests, and extrapolated values of cohesion, c, and friction angle, ϕ.

57

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Self-monitoring of cracking development of fiber reinforced conductive concrete subjected to bending Yining DING 1, Genjin LIU 1, Zhongyue DU 1 1

:State Key Laboratory of Coastal and Offshore Engineering, Dalian University of

Technology, Dalian 116024, China

Abstract Currently, the piezoresistivity of cementitious materials and its applications in SHM is widely investigated (Chung 2001, Han, Guan et al. 2007). But, the self-monitoring ability of conductive concrete with coarse aggregate is still less studied. Also the electrical property of concrete with deflection hardening behavior is very few investigated. The addition of SF, CB and CF does not show significant trend of improving the compressive strength. The advantage of deflection hardening fiber composite is its superior energy absorption capacity than that exhibits deflection softening behavior (Shaikh 2013). The peak load can be improved clearly by addition of 60kg/m3 SF. Compared with PC beam, the additions of SF enhance the post crack behavior of concrete significantly. A great increment of FCR is observed during the first cracking. For series 40 specimens, a monotone increasing relationship between FCR and COD is illustrated. However, for series 60 specimens, bilinearity increasing relationships between FCR and COD are observed.

Conclusion (1) The addition of conductive admixture can improve the flexural capacity and toughness. (2) As the load bearing capacity declines at the first cracking point, the FCR of all beams shows a significant increment.

References Chung, D. (1995). "Strain sensors based on the electrical resistance change accompanying the reversible pull-out of conducting short fibers in a less conducting matrix." Smart materials and Structures 4(1): 59. Han, B., X. Guan and J. Ou (2007). "Electrode design, measuring method and data acquisition system of carbon fiber cement paste piezoresistive sensors." Sensors and Actuators A: Physical 135(2): 360-369. Shaikh, F. (2013). "Deflection hardening behaviour of short fibre reinforced fly ash based geopolymer composites." Materials & Design 50: 674-682.

58

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Bridges/Elevated slabs

59

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Hybrid Fiber Reinforced Concrete for the Application of Bridge Deck Mehran Khan1, Mingli Cao1, Majid Ali2 1:

Department of Civil Engineering, Dalian University of Technology, Dalian, China. 2: Department of Civil Engineering, Capital University of Sciences and Technology, Islamabad, Pakistan.

Abstract There are many flaws in bridge decks like early age micro cracking, surface erosion, spalling and crazing. The main flaw in concrete bridge deck is early age micro cracking (EAMC). The cement content, tensile strength and compressive strength are the factors affecting EAMC in bridge decks. The less resistance to early age cracking is due to the low tensile strength of concrete in bridge decks. Therefore, improving tensile strength of concrete with material other than cement will results in more resistance against EAMC. The lateral load carrying capacity is reduced due to penetration of moisture through micro cracks in concrete cover which results in cracks on surface of embedded rebar in concrete matrix. The structural durability can be greatly increased by producing a more crack-resistant concrete. At the preliminary phase of micro-cracking, the micro-fibers arrest the micro-cracks by bridging. When the stress increases, these micro-cracks spread and convert into meso-cracks where meso-fibers are more operational to resists these meso-cracks. The meso-cracks converts into macro-cracks when stresses are further increased. Hybrid fiber reinforced concrete (HFRC) is used as an alternative to plain concrete (PC) due to its several structural benefits. The hybridization of different scale fibers can control cracking at different levels and also improves the durability. Nowadays the use of different fibers in concrete for the application of bridge deck has showed better performance against early age micro cracking. The steel fibers arrest macro-cracks ultimately resulting in improved toughness. On the other hand, use of calcium carbonate (CaCO3) whisker also results in less cracking. CaCO3 whisker is a new type of inorganic micro-fiber which is used in cement mortar to improve its mechanical properties. Also, nowadays the basalt fiber has gained the popularity due to its environmental friendly manufacturing process and excellent mechanical properties in concrete. The tensile strength of basalt fibers is greater than that of Eglass fibers. A new kind of fiber hybridization with the combination of CaCO3 whisker, basalt fibers and steel fibers is considered. Therefore, EAMC can be controlled with the use of CaCO3 whisker, steel and basalt fibers. In this work, the effect of different basalt fiber content on splitting-tensile properties of hybrid fiber reinforced concrete (HFRC) will be investigated. The CaCO3 whisker, basalt fiber, steel fiber, coarse aggregate, silica sand, cement, water and super plasticizer are the raw materials used. The maximum size of aggregates was 18 mm. The XRD pattern and micromorphology of CaCO3 whisker is also shown. The mix design ratio of HFRC is 1:2:1.5:0.05 (cement: sand: aggregate: CaCO3 whisker) with a water cement ratio of 0.42. The steel and basalt fiber length is 35 mm and 12 mm, respectively. The steel fiber and CaCO3 whisker both content are added 5%, by cement mass. Different basalt fiber contents of 2%, 4% and 6%, by cement mass, are added to prepare HFRC2, HFRC4 and HFRC6, respectively. The super plasticizer content of 1%, by cement mass, is added to HFRC2, HFRC4 and HFRC6. For production of HFRC mix, the whole material was added into the mixer using one-third layers 60

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

approach to avoid balling effect. A one-third layer of aggregate was added first, followed by layers of one-third sand, steel fibers, cement, CaCO3 whisker and basalt fibers. The other layers were also added using the same sequence until all the material was finished. For each batch, cylinders are cast and test under splitting-tensile load as per ASTM standard. The slump values of HFRC are reduced up to 74% as compared to that of PC. The load-time curves are recorded under split-tension load. Also, the strengths, pre/post and total energy absorbed and toughness for split-tension load are determined. The SEM analysis is also performed to study bridging and cracking of hybrid fibers in the matrix. The fiber-matrix bond shows the proper bonding between the matrix, CaCO3 whisker and basalt fiber. Also, the basalt fiber surrounded by the matrix will results in improved split-tension properties. The split-tension strength, pre-crack energy absorbed, total energy absorbed and toughness index of HFRC are increased up to 37%, 11%, 95% and 75%, respectively, than that of PC. It is found that, with increasing content of basalt fiber up to 4%, there is an increase in tensile strength of hybrid fiber reinforced concrete. This may ensure the reduction of EAMC in bridge deck. Also, the crack arresting power of hybrid fibers at micro-, meso- and macro-level does not allow the moisture to penetrate in to the concrete which will ultimately reduce the deterioration of concrete which leads towards the durable bridge deck. Future recommendation is to optimize the basalt fiber content in HFRC for the application of bridge deck.

Keywords Early age micro cracking; CaCO3 whisker; basalt fiber; hybrid fiber reinforced concrete; bridge deck

61

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Performance under fatigue loading of field-cast UHPFRC joints in positive moment regions between precast bridge deck panels Benoît Marleau1, Bruno Massicotte1, Jean-Philippe Charron1 1

: Polytechnique Montreal, Montreal, Canada.

Abstract In the past decades, many research projects aimed at studying the use of ultrahigh performance fibre-reinforced concrete (UHPFRC) in field-cast joints between precast bridge deck elements have been carried out in North-America and led to the construction of several bridges using this new technology. However validation is still needed before introducing this innovative concept in bridge design codes and generalised its application. Past tests led to design specifications for UHPFRC field-cast joints positioned above girders whereas very limited experimental results were available on the fatigue behaviour in positive moment regions of longitudinal joints between main bridge girders or transverse joints perpendicular to the main girders. An extensive experimental campaign on 17 UHPFRC jointed slab specimens and 3 control specimens of normal concrete was carried out to investigate the contribution of parameters such as the joint performance under high cycle fatigue loading, the effect of fibre volume in UHPFRC, the lapped bar layout and the advantage of using fibres in precast panels. Several specimens were first subjected to over 5 million fatigue cycles and ultimately loaded to rupture. Others specimens were only tested to rupture with a monotonic loading for comparison purposes. The cyclic test protocol included 5 cycles at 60% of the yield strength followed by fatigue loading generating a stress variation of 125 MPa in the bars. For longitudinal joints in the deck positive moment regions, the Quebec Ministry of Transportation (QMT) design uses 15 mm U-shape bars alternately lapped over a length equal to 10 db whereas straight alternate lapped bars are specified for transverse joints as shown in Figure 1. The project included the validation of the usual QMT joint design with 2% (by volume) 13x0.2 mm steel fibres and the consideration of joints made with 3% fibre content.

a) Longitudinal joint

Figure 1:

b) Transverse joint

Longitudinal and transverse joint configurations.

For specimens subjected to fatigue loading deflection and crack width progressed with the number of cycles. The growth rate was variable, ranging from 0.05 mm/million cycles to 0.15 mm/million cycles. Despite this observation, the ultimate strength was not affected as 62

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

illustrated in Figure 2, where similar load deflection response were obtained for specimens subjected or not to fatigue loading. It is worth noting that the behaviour of the control specimen made of normal concrete with a conventional 600 mm long lap splice (pink and purple curves with BO in Figure 2) subjected to fatigue loading was comparable to the specimens with short UHPFRC joints.

Average deflection (mm)

Figure 2:

Comparison of the behaviour to failure of joints previously subjected to fatigue to control specimens only loaded monotically to failure.

Results from this project confirmed the adequacy of QMT design using UHPFRC containing 2% of 13×0.2 mm fibres for longitudinal and transverse joints in positive moment regions with a minimum lap slice length equal to 10db. Results also indicated that using UHPFRC with fibre volume more than 2% might enable reducing the lap splice length below 10db. Finally experimental evidence indicated that using precast slabs made of steel fibre reinforced concrete increases the bridge deck performance under fatigue loading and at failure.

Keywords Bridge deck; UHPFRC; Fatigue; Accelerated Bridge Construction; Sustainability

63

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Design of precast prestressed SFRC T-girders for accelerated sustainable bridge construction Bruno Massicotte1, Nicola Cordoni2 1

: Polytechnique Montreal, Montreal, Canada. 2 : WSP, Vancouver, Canada.

Abstract Accelerated bridge construction is becoming a subject of major importance. Combining steel fibre reinforced concrete (SFRC) and prestressing offers a unique opportunity to fulfill the demand of more sustainable infrastructure with enhanced durability and life-cycle cost reduction. A project on precast T-girders was initiated with the aim of developing a new set of prestressed girders for new bridges in the 10 to 30 m span range. The proposed concept main innovation consists in the combination of steel fibres, optimised prestressing layout and reduced conventional reinforcement, for flexure and shear, and reduced top flange transverse flexural reinforcement connected through UHPFRC field-cast joints between adjacent girders. The cross-section dimensions are shown on Figure 1 for the 20 m span girders.

175

1500 Additionnal T15 prestressing strands if required

250 Single-leg stirrups

1000

Transverse reinforcement

300

6@50

T15 prestressing strands

70

50 200

Figure 1: Transverse cross-section for the 20 m span girders.

Figure 2: Moment–deflection behaviour obtained using NLFE.

The geometrical characteristics were chosen to standardise the cross-section profile for reducing the costs related with the production process, mainly associated to the formworks. The cable configuration and the geometrical limitations for the beam web were chosen to minimise any issues related with fibre flow and optimised fibre orientation. The lateral profile of the web is inclined with top flange and bottom dimensions kept constant for all beam sizes. The use of minimum shear reinforcement is envisaged with single 15 mm in diameter stirrups located at the center of the web between the two rows of strands (see Fig. 1). The selected SFRC mechanical characteristics were based on the self-compacting mix successfully used in past projects using 80 kg/m3 of hooked-end 35 mm long and 0.55 mm in diameter fibres (De Brouker 2013). This mix has shown its efficiency in service conditions for controlling flexural crack opening both longitudinally in the girder web and transversely in the girder flange, and also for its performance at ultimate limit states in shear. 64

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

The cross section design was done using a sectional analysis software using material nonlinear behaviour at all loading stages. The use of three types of longitudinal reinforcement for the lowermost layer was investigated: bonded T15 tendons, un-bonded T15 tendons and 30 mm diameter reinforcing bars. The preliminary study tends to confirm that there are no substantial advantages in terms of carrying capacity, cracking pattern and stress range to privilege one type of longitudinal reinforcement for the bottom layer, as confirm experimentally in a parallel project. The choice of unbonded strands for the lower layer could be recommended to limit the stress range associated with fatigue if cracking was allowed. Live load deflection limitations govern the girder depth to L/20. The adopted transverse reinforcement consists of 15 mm diameter single-leg stirrups spaced at s=h/2 (see Fig. 1). The shear resistance is calculated following the Canadian Bridge Code (CSA, 2014) for the concrete (Vc) and reinforcement (Vs) contributions. The fibre contribution (VF) was determined according to Casanova and Rossi (1996) model. The fibre contribution was taken as the post-cracking strength corresponding to a crack width of 1.5 mm as suggested in the Australian Code (2014), equal to 1.25 MPa for the material properties used in this design. Figure 2 presents the moment deflection response of the 20 m girder analysed using nonlinear finite element analysis. In all cases analytical results showed closely spaced cracks in service whereas a few cracks governed the behaviour at failure which was attributed to the rupture of presetressing strands. The governing shear strength was obtained for a loading point located at a distance equal twice the member depth from support. In this case the combined contribution of the fibres was determined equal to 1.45 MPa, which is slightly larger than the value of 1.25 MPa selected in the preliminary design.

References DR AS 5100.5 (2014), Draft for Public Comment Australian Standard, Bridge Design Part 5:Concrete, Standards Australia, Sydney, Australia. Casanova, P., Rossi, P. (1996), Analysis of metallic-fibre reinforced concrete beams submitted to bending. Materials and Structures, 29, 354-361. CSA (2014), Canadian Highway Bridge Design Code - CSA-S6-14. Canadian Standard Association, Toronto, Ontario, Canada. De Broucker, W. (2013). Flexural and shear behaviour of prestressed SFRC girders. M.A.Sc. Thesis, Polytechnique de Montréal, Montreal, Canada. (In French).

Keywords Precast; Prestressing; SFRC; Bridges; Accelerated Bridge Construction; Sustainability

65

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SF+RC modelling using DIANA FEA Ab van den Bos1, Saurabh Dhanmeher1, Yue Dai2 1

: DIANA FEA BV, Delft, Netherlands. 2 : Delft University of Technology, Delft, Netherlands.

Abstract The design of structural members utilising steel fibre + reinforced concrete (SF+RC) is gaining popularity. SF+RC implies enhancing normal designed reinforced concrete by adding fibre content, thereby making it more robust. Usually, this enhancement enables reduction in size of concrete sections and/or the amount of traditional reinforcement. The modelling of the material is often based on a smeared crack approach. In DIANA this is already possible for a long time using a total strain crack model. Since a few years, an official tensile curve named FRCCON has been available. The softening behaviour of the concrete combined with the steel fibres can be input as a stress-CMOD or stress-strain relation. Several papers on this material have already been published, Bos, A.A. van den et.Al. (2015), K. Younis et.Al. (2014), Bos, A.A. van den, et.Al. (2016) The parameters of this tensile curve are obtained either from the Model Code 2010 and/or via tests done on small bending prisms. This could be a 3-point notched beam test or a 4-point bending test. Via inverse modelling, one can obtain the parameters for material input. Many different types of fibres exist, which can be added in different dosages and combined with concrete to produce SF+RC. In order to standardise design procedures, it is of interest to have an idea about the combined behaviour of a particular fibre-type added to concrete in a particular dosage. However, since the number of possible combinations are very large and test results are valid only for a certain combination of volumes, the inputs for a new combination in structural design can be difficult to identify. This paper aims at developing a strategy to numerically calculate the response of SFRC such that the effect of different fibre-dosages and fibre-types can be considered within a single model. For the application it is important to describe the influence of the different parameters on the ultimate strength of the structural member. The main influencing parameters are investigated and described. This will be the concrete strength, the shear along the fibre, the stiffness and strength of the end anchors and the orientation and distribution factors. In the first approach, the effect of steel fibres is considered in the constitutive law for the concrete material. A Stress vs. Crack Mouth Opening Displacement relationship, as suggested by the fib Model Code for Concrete Structures 2010, is used as input. In the second approach, the steel fibres are modelled as discrete beams/trusses embedded in continuum elements of concrete. The fibres follow a separate constitutive law. A linear elastic stress-strain relationship is assumed for this analysis. In order to account for the friction stress and slip of the fibres, interface elements are introduced along the fibre-length, which are governed by a Shear traction-Relative displacement diagram.

66

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A combination of the two approaches discussed in the previous section is conceived, which can allow numerical evaluation of FRC structures with any dosage, length and orientation of steel fibres. Based on experimental data, max-min stress-CMOD curves are assumed for concrete class C30, reinforced with 20kg of reused steel fibre from car tyres. The response of this combination is calibrated using the FRCCON material model available in DIANA.

(a) (b) Figure 1:

Stress-CMOD (Experimental vs. numerical) (a) and discrete fibre input (b).

Keywords SFRC, DIANA FEA, Hybrid (SF+RC), multiple fibre dosage

67

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Steel Fibre Only Reinforced Concrete in Suspended Slabs: Design From Academical to Practical Methods X. Destrée, ir. 1 1

: ARCELORMITTAL, Luxembourg

Abstract

Introduction Over the last 20 years (Destrée, X) the total replacement of all traditional rebar by steel fibers has become completely routine in applications such as industrial and commercial suspended slabs on piles as well as elevated suspended slabs. The typical pile spacing is in the range 2.5m to 5 m in each direction, with span to depth ratios from 12 to 20. The Model Code 2010 provisions includes an experimental structural indeterminacy redistribution factor applicable to the resisting moment: K Rd = (P max,k / P max,m) / (f Ftum / F tuk) < 1.4. The more recent Swedish Standard SS812310 (2013) introduces also a structural indeterminacy factor Ƞ det = 2 and is indeed much more practical and closer to reality value. The ultimate loading intensity of the real suspended slab is however 4 to 7 times larger than its most onerous service loading. Fifteen million square meters have been completed to date and no known case of moment failure, no known case of punching-out!

Design example Using KRd, at a typical FIB Model Code value derived from a number of real full scale tests (Kleinman), a 210 mm thick slab SS 812310 conform for 40 kN/m² loading intensity, shall need to be thickened up to 250 mm according to the Model Code provisions. The 210 mm thickness being kept, the loading intensity should have been reduced down to 28 kN/m² thus a 40 % loss of service ability following the Model Code 2010 provisions. The FIB Model Code 2010 results into overdesigned uncompetitive solutions that are not in line with the 25 year experience.

Full scale test verification The design example piled slab, has been test loaded during 7 days in full scale up to 40 kN/m² in Tingstat (Sweden) while there was no contact with the ground underneath. It did not show more than 0,5mm deflection and no visible cracking! The same observations were made in more full scale loading tests of piled slabs in Nieuw Vennep (Holland), in Klaipeda (Lithuania), in Ternat (Belgium) and Townsville (Australia).

68

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Shear/Punching-out SFRC suspended slabs are totally insensitive to punching-out. This is confirmed by the total lack of any known punching-out event of the pile head into the slab in almost 2 million cases over the last 25 years and also according to real scale punching-out testing. a T.U. Eindhoven-Holland testing of a 160 mm thick suspended steel fibre reinforced slab with 30 kg/m³ dosage rate has been tested under punching-out: 250 kN loading intensity at first crack and 700 kN point loading were needed to cause the punching-out rupture of a punchedout cone of 700 mm diameter thus showing a 20 ° angle punching-out cone. In Bissen-Luxembourg, a 210 mm thick full-scale slab weakened by 60% coring-out of the section along the critical perimeter, failed in punching-out along the free edge at 550 kN. The full slab far from the edge should indeed show a multiple of that loading intensity and far from the free edge to 2750 kN at punching-out rupture. Flexion prevails, by far, over punching-out. Thus in the last ten years most SFRC piled slabs were designed without pile heads to enjoy an easily installed slab and so, a much more economical flat bottom slab.

Concluding remarks The KRd redistribution factor as defined in the Model Code 2010 in case of suspended slab is unpractical as impossible to be known at the design stage and is of a too low value, at most 1,4, so that it results into a gross overdesign of the S.F.R.C solution or into the necessary introduction of supplemental reinforcing bars. The numerous SFRC suspended elevated full-scale tests achieved in the last 25 years and the multimillions square metres of completed experience gained in the real activity together with the Swedish Standards SS 812 310 provisions, demonstrate that KRd = 2 is a more realistic minimum and safe value.

69

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Effect of reinforcement configuration on the ductility requirements of real-scale slabs Eduardo Galeote1, Ana Blanco2, Albert de la Fuente1 1

: Universitat Politècnica de Catalunya, Barcelona, Spain. 2 : Loughborough University, Loughborough, United Kingdom.

Abstract According to design specifications, structural elements require a certain degree of ductility in failure regime. This ductility is provided by a minimum amount of reinforcement that guarantees that the ultimate failure moment Mu of the structure can resist the cracking moment Mcr. Structural elements such as segments for tunnel linings behave as isostatic slabs during transitory stages such as demoulding, stocking, transport or manipulation. During transient stages segments are subjected to their self-weight and the weight of additional segments stacked on top. These loads, even though being more restrictive, produce low bending moments that are not high enough to achieve Mcr. Hence, a reinforcement below the minimum necessary to achieve a ductility associated to Mcr would suffice to provide enough ductility to the structure. To analyse the influence of different reinforcement configurations based on fibres and traditional reinforcement, an experimental program was conducted. The characterization of the material was performed on one mix of plain concrete (PC) and two mixes of fibre reinforced concrete (FRC) with contents of 25 and 30 kg/m3 of fibres (FRC-25 and FRC-30). Additionally, five slabs with a length of 3.0 m, a width of 1.0 m and a height of 0.4 m were produced. The reinforcements consisted of one slab with steel corrugated bars (RCS), two slabs with fibre contents of 25 and 30 kg/m3 (FRCS-25 and FRCS-30) and two additional slabs with a hybrid combination of traditional reinforcement and fibres in contents of 25 and 30 kg/m3 (HRCS-25 and HRCS-30). The properties assessed of the material were the compressive strength, the modulus of elasticity and the flexural test through three-point bending tests. The slabs were tested on a fourpoint isostatic configuration with a span of 2.7 m and a distance between loading points of 0.9 m. The deflection was measured using a laser placed at the mid-span over the surface of the slab. The crack opening was also measured at both sides of the slab through two displacement transducers encompassing a central distance of 90 cm. Additionally, one LVDT measured the vertical displacement of each support at the short edges of the slabs. Load-Displacement curves were obtained during the test and Moment-Curvature diagrams (M-χ) were determined through a sectional analysis using the material characterization results as data input. In RCS, the load bearing capacity was activated after cracking with rebars absorbing the tensile stresses while increasing the strength and the deflections. Slabs FRCS-25 and FRCS-30 presented a brittle behaviour exhibiting a strength drop after cracking. Hybrid reinforced slabs presented a ductile response achieving the maximum loads at low deflections. The only elements with crack localization were slabs FRCS-25 and FRCS-30 due to the brittle failure. The rest of the slabs presented two cracks in at least one of the faces. Ramifications appeared especially at the main cracks, these branching off at the half top of the slab following the direction of the longitudinal reinforcement placed at the top surface.

70

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Increasing the amount of fibres also produced a greater number of cracks due to the reduction of the bond transfer length. These results evinced that there are different solutions and reinforcement combinations that verify the ductility requirements.

Keywords FRC, real-scale tests, ductility, design optimization, crack width

71

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A new method for designing FRC elevated slabs according to Model Code 2010 Luca Facconi1, Fausto Minelli1, Giovanni Plizzari1 1

: University of Brescia, Brescia, Italy.

Abstract After several years of discussion within the research community, Fiber Reinforced Concrete (FRC) is nowadays recognized as a structural material considered by international (fib Model Code 2010) and national (DAfStb, NTC 2008) structural codes. Based on the design requirements of fib Model Code 2010 (MC2010), a simplified procedure for designing the flexural reinforcement in Hybrid Reinforced Concrete (HRC) elevated slabs is proposed herein. The method consists of two main stages named as “proportioning stage” and “verification stage”, respectively. The former provides some guidelines to preliminary design conventional reinforcement and to determine the main geometrical properties of the slab. The latter focuses on the verification of the HRC slab by performing Non-Linear Finite element analyses (NLFEAs). The main features of the two design stages are described in the following. Proportioning stage The proportioning stage consists of the following main steps: 1. 2. 3. 4. 5.

Choice of the slab thickness (t) so that 1/35≤t/L≤1/25 (t/L=thickness/span length). Choice of mechanical properties of materials (FRC and reinforcing steel). Determination of the design loads (Ed) according to the typical load combinations suggested by EN 1990 (2006). Determination of the internal actions through Linear Elastic Finite Element Analysis (LEFEA). Design of conventional reinforcement combined with fibers. Based on the results of the LEFEA, the maximum design bending moments (mEd,x, mEd,y) acting in the two orthogonal directions (x and y) can be evaluated as follows: mEd,x =md,x ±|md,xy | ; mEd,y =md,y ±|md,xy |

1)

where md,x, md,y are the design bending moments whereas md,xy is torsional moment. The design resisting moment provided by fibers only can be estimated as follows: 1

mRd,FRC = 2 f

Ftu,d

∙t∙(t-x)=0.45∙fFtu,d ∙t2

2)

where fFtu,d=fR3k/(3×c); fR3k is the residual flexural strength at a Crack Mouth Opening Displacement (CMOD) of 2.5 mm according to EN14651; c=1.5 is the partial safety factor for FRC according MC2010. Conventional reinforcement has to be placed in the areas of the slab where the design internal bending moment (mEd) is higher than the resisting moment provided by fibers only (mRd,FRC). The intersection between the resisting moment provided by fibers (i.e. mRd,FRC) and the envelope curve of design bending moments provides the length (Lint) along which bending moments have to be integrated to determine the area of conventional reinforcement. The latter results from the following relation:

72

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

L

As,x =

∫0 int (mEd,x -mRd,FRC )dx 0.9∙fyd ∙d

L

; As,y =

∫0 int (mEd,y -mRd,FRC )dy 0.9∙fyd ∙d

3)

where As,x and As,y are the total reinforcement areas in x and y-direction, respectively; d is the effective depth of the slab; fyd=fyk/s, fyk and s=1.15 are the design yield strength, the characteristic yield strength and the material safety factor of conventional reinforcing steel. A typical conventional reinforcement layout is depicted in Figure 1.

Figure 1:

(a) (b) Typical rebars layout for HRC elevated slabs: top view (a); bottom view (b).

Verification stage The following procedure is proposed for structure verification of the slab: 1.

Determine the global resistance (Rd) of the slab by performing NLFEAs including the tensile constitutive laws of FRC suggested by the MC2010 (clause 5.6.4). Knowing Rd, the following condition has to be fulfilled (MC2010 - clause 7.11.3): Ed ≤Rd =Rm /(γ*R ∙γRd )

2. 3.

4)

where Rd and Rm are respectively the design and mean global resistance of the structure; *R and Rd are the global resistance and the model uncertainty factors, respectively. Check the safety and serviceability minimum requirements for FRC structures according MC2010 (see clause 7.7). If needed, check the punching resistance of the slab (see MC2010 - clause 7.7.3.5.3).

Keywords Elevated slabs; Fiber Reinforced Concrete; Design; Hybrid Reinforcement; fib Model Code 2010; Finite element analysis; Non-linear analysis

73

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental and numerical comparative analysis of a SFRC slab on piles Lukáš Dlouhý 1, Rémy Roland 2 1

: Product Development Engineer, SCIA, Brno, Czech Republic. 2 : Technical Manager South Europe & Africa, Bekaert, Zwevegem, Belgium.

Abstract In order to investigate the behaviour of SFRC slabs on piles, a campaign of tests has been conducted at the Tongji University of Shanghai, China. All the specimen are reinforced with the 5D 65/60BG high performance fibre from Bekaert, at dosages of 30 and 35 kg/m³. Tests were conducted on four plates of dimensions 4.3x4.3m, with a thickness of 18cm, loaded by one central point and supported on four 4 points. To gain a fundamental understanding of the structural behaviour, we have built non-linear FEM models using the SCIA Engineer 17 software, which has a built-in module for SFRC structures. Numerical simulation results are compared to experimental performance in terms of ultimate load and crack pattern. As the steel reinforced fibre concrete is very often used mainly for foundation the real behaviour and investigation of a structure such as a slab on piles is essential. In this scope, this study aims to predict the ultimate capacity of the SFRC slab on piles of the Nanhui project in Shanghai. This design has been done both by testing and by using a finite element model using SCIA engineer. Concerning the test campaign: The projected structure is a square slab on piles with dimensions 14x14m supported on 49 piles having a central spacing of 3m and a peripheral spacing of 2m. This is of course too big to be tested in a laboratory so in a first step we will derive a suitable testing model with a size reduced to 4.3x4.3m. Then we describe the test setup and give the results. Concerning the finite element model: The model is made of plate elements with dimensions close to 5cm. Due to the symmetry of the test specimen only one quarter of the slab was modelled (i.e. 2.15x2.15m). The nonlinear calculation uses a very efficient damage material model called the Mazars model. In a first step, we did a complete finite element model of the slab to see the distribution of the bending moments. Then, we have derived an equivalent test structure with one central hydraulic ram and four bearings at the corners, the rotation of the model’s edge will be constrained to simulate the behaviour of the full plate. The concrete is a C40/50 and the post-crack flexural tensile strength have been measured according to EN 14651 for two dosages 30 kg/m³ and 35kg/m³ on 6 beams for each dosage.

74

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

EN14651 mean values. Dosage [kg/m³] 30 35

Delta FL (mm) 0,04 0,04

fℓ [MPa] 5,20 5,27

fR,1 [MPa] 4,52 5,00

fR,2 [MPa] 5,36 6,00

fR,3 [MPa] 5,34 6,08

fR,4 [MPa] 5,15 5,46

The force/displacement curves show an extended ductility far beyond the ultimate limit state, up to a deflection of 10cm. The presented nonlinear model provides good match with the experimental test. The comparison of the results with the load-displacement diagram coming from the test is visible on figure [Fig. 1] for Dramix fibres 5D65/60BG with dosage 30kg/m3 and 35kg/m3. This capacity is at the load level 283,9kN (299,0kN for 35kg/m3 respectively) where the ultimate tensile strain in steel fibre reinforced concrete (25‰) is reached. In fact graphs Calc-30 correspond to P-5D-30-1-LS-1 and Calc-35 to P-5D-35-1-LS-2.

Figure 1: Load displacement distribution.

The experimental and numerical findings presented in the present paper allows to draw the following concluding remarks: 1. From test results it turns out that the slabs reinforced only with 30 to 35 kg/m³ of 5D 65/60BG fibers exhibited a good structural behaviour with an extended duxtility 2. The non-linear calculation performed in software Scia Engineer provides a very good

comparison with the test measurement. The study proves that this tool can be used for the analysis of 2D structures and modelling of the real behaviour of the structure with respect to cracks.

Keywords Steel Fibre Reinforced Concrete, Slab on piles, Finite Element Modelling

75

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

FRC Hybrid slabs: reliability of Model Code approaches Ali Pourzarabi, Matteo Colombo, Paolo Martinelli, Marco di Prisco Politecnico di Milano, Department of Civil and Environmental Engineering, Milan, Italy.

Introduction Fibre reinforced concrete (FRC) material is characterized by a high intrinsic scatter in the results when tested according to standard notched beams. However, it is observed that structures affected by a high stress redistribution show a significantly reduced scatter in their structural behaviour. Therefore, the use of the characteristic material constitutive parameters from a standard test leads to overly conservative design. The Model Code 2010 has introduced a coefficient, named structural redistribution factor, that is able to take into account a reduced variability of the structural response, when compared to that identified from a standard material test. The high scatter in the results is due to the small fracture area, low number, and random distribution of fibres. The coefficient introduced in the Model Code 2010 counteracts the reduced characteristic values of material properties for structures that are able to redistribute stresses, and in which a large fracture volume is involved. The larger fracture volume is taken into account through the random assignment of the material properties to different cells in the structure. The coefficient is directly multiplied to the design load to get the ultimate design load as follows: P Rd = K Rd P( f Fd ) ;

K Rd =

max Prand , k f Ftu , m

K Rd 3 =

u Prand , k f Ftu , m

u Prand ,k

 1.4 or or K Rd 4= u u max Prand Phom, k Prand , m f Ftu , k , m f Ftu , k where PRd is the ultimate design load, KRd the redistribution factor, and P(fFd) is the design load computed based on material characteristic properties. Moreover, the new revision of Eurocode 2 has introduced a κG factor which is equal to KRd4 with the difference that it is only applied to the design value of FRC constitutive tensile strength, f Ftu,k /γF, instead of being multiplied directly to the ultimate design load. In this work, reference is made to the four definitions of the redistribution factor already mentioned. Pmaxrand,m and Pmaxrand,k, Purand,m and Purand,k are the mean and characteristic values of the maximum and ultimate load obtained starting from a normal distribution assumption for the fR1 and fR3 which are randomly assigned to the structure, Puhom,k is the ultimate load obtained by assuming the characteristic values of material properties in a homogenous manner for the whole structure. MC

Experiments and Discussion To study the redistribution coefficient, 9 standard notched specimens were tested in a threepoint bending test set-up; 6 concrete slabs of 2×2×0.15 m were tested under a concentrated load applied in the center and supported at the mid-length of each side. Two SFRC slabs, two R/C slabs, and two slabs with both reinforcement (named Hybrid slabs) were tested. The redistribution factor has been evaluated through a yield line approach with random assignment of material properties assuming a normal distribution for the post-peak residual tensile 76

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

properties of fR1 and fR3. The testing setup is shown in Figure 1a and the discretised yield line mechanism in Figure 1b.

(a) Figure 1:

(b) (a) slab test setup (b) discretized yield line mechanism adopted.

The results obtained from the notched beams and slabs are shown in Figure 2 in terms of nominal stress–CMOD and load–deflection, respectively. Looking at the results of the SFRC slabs (SF1, SF2), it can be observed that the structural behaviour of the FRC material is very closely repeated in the slab test while depicting a high scatter in the standard beam tests.

(a) Figure 2:

(b) (a) stress-CMOD obtained from three-point bending test (b) slab load-deflection.

Table 1 gives the values of the ultimate design load and the safety coefficients based on different definitions of the redistribution factor. It is observed that when the redistribution factor is applied to the whole resistant load (the case for KRd3, KRd4, KMCRd) in the Hybrid slab, no distinction is made between the fibre contribution and the contribution of the conventional reinforcement. However, applying the G factor to the design constitutive tensile strength values of the FRC material gives a safety factor comparable to that of the R/C slab. Table1: KRd definition

KRd3 KRd4

K G

MCRd

The redistribution factor obtained for the SFRC and Hybrid series with safety factors obtained. SFRC Hybrid RC KRd

1.66 1.48 1.4 1.48

PRd

γ=Puexp,m/ PRd

KRd

PRd

γ= Puexp,m / PRd

172 153.3 145 119.5

1.38 1.55 1.64 1.99

1.67 1.17 1.4 1.17

471.6 330.4 395.4 275.5

1.04 1.50 1.25 1.80

γ= Puexp,m / P(fFd)

1.88

77

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Optimization of partially prefabricated HyFRC slabs Giulio Zani1, Matteo Colombo1, Claudio Failla2, Marco di Prisco1 1

: Politecnico di Milano, Dept. of Civil and Environmental Eng., Milano, Italy. 2 : Magnetti Building, Carvico (BG), Italy.

Abstract A new partially prefabricated elevated slab has been recently introduced in two different industrial buildings, to propose a viable alternative to the classical double tee deck with the addition of an in situ RC topping. The solution is characterized by an adjustable spacing in the orthogonal direction, 40 mm thick FRC plates used as predalles and a cast-in-place FRC finishing, designed according to a continuous slab resting on the simply-supported beams. The proposed deck is a structural solution that tries to fit different issues like construction speed, transport and cost reduction, structural optimization, high fire resistance (R120) and quality performance. All elements are made from SFRC, with varying amounts of steel fibers. This paper presents a design investigation on this kind of floor element, aimed at optimizing the global structural solution by minimizing the whole floor weight. Longitudinal and transverse bending, as well as vibration limit state, were considered in the design. The optimization strategy will be here presented, through the discussion of the parameters considered in the design, the variables taken into account and the constraints adopted within the procedure. A Model Code 2010 design approach was followed.

Keywords Elevated slab; HyFRC; SFRC; fiber-reinforced concrete; precast concrete; prestressed concrete; structural optimization

Introduction Steel Fiber Reinforced Concrete elevated slabs are an interesting design solution, because they can simultaneously guarantee robustness, high construction speed, flexibility and economy, especially if combined with conventional reinforcement. In the paper, an industrial deck solution aimed at combining both the advantages of prefabrication and cast-in-place solutions is investigated. Classical TT elements are substituted by prestressed prefabricated beams, designed to reach a high fire resistance, and thin prefabricated slabs with integrated lattice girders, resting on the top chords of the beams; these latter elements allow to increase and adjust the spacing of the beams, reducing the transport costs without penalizing the construction speed. The final cast-in-place layer guarantees a two-way stiffness, not only in its own plane, but also in bending, thus profiting of the slab redundancy. A scheme of the structural solution is presented in Figure 1, where the different parts of the system are highlighted. Starting from two existing case studies, the paper describes a design optimization aimed at minimizing the global weight of the structural solution. This investigation follows the Model Code provisions and represents a preliminary approach, since further developments are needed to guarantee a comprehensive structural design.

78

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

General scheme of the proposed structural solution.

Optimization procedure and results An optimization procedure was implemented to define the optimal solution for a given longitudinal span and for a given characteristic live load acting over the floor. The transverse bending of the floor was studied considering both transient and final conditions, leading to the identification of the electro-welded lattice girder solution able to guarantee the maximum beam spacing. Following the selection of the electro-welded lattice girder, therefore the definition of the maximum distance between the beams related to each span and load case, the depth of the beam and the tendons distribution within the beam cross-section were defined, focusing on the longitudinal bending behaviour. To this aim, a general constrained minimization procedure, solved adopting a Sequential Quadratic Programming (SQP) algorithm and able to minimize the weight of the structural element, was implemented. The results of the optimization procedure are presented in the paper, where the global permanent loads g0 of the considered structural solution are represented as a function of the beam spacing, respectively for longitudinal beam spans equal to 10, 15 and 20 m and for live loads of 10, 20 and 30 kN/m2. Among the global set of solutions, the ones guaranteeing the smallest permanent load g0 were finally selected, leading to the identification of the optimum structural layouts.

Conclusions The paper presents a preliminary optimization procedure applied to a new FRC solution for partially prefabricated slabs. The main goal is to maximize the beam spacing and to reduce the global permanent load of the structure. The preliminary procedure here presented simply focus on longitudinal and transverse bending behaviours. All the other checks, such as shear and fire resistance, need to be carried out in the aftermath of the solution choice. Further investigation will also comprise these parameters in the optimization procedure and will include alternative cross-sections, such as the beam ends.

79

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

80

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Shear

81

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental study of shear transfer in polypropylene fibrereinforced concrete using pre-cracked push-off specimens Francisco Ortiz-Navas1, Luca Scaroni2, Juan Navarro-Gregori1, Pedro Serna-Ros1 1

: Polytechnic University of Valencia, Valencia, Spain. 2 : University of Brescia, Brescia, Italy.

Abstract Shear transfer mechanisms in fibre-reinforced concrete (FRC) have been studied by many authors based on pre-cracked and non-cracked push-off specimens. However in many of these research works, the study of crack behaviour related to normal stress and crack width-slip remains unclear. The experimental results of the shear behaviour of 14 pre-cracked push-off specimens (40 MPa of concrete compressive strength), made with plain concrete (PC) and polypropylene fibre-reinforced concrete (PFRC), are presented. Fibres were dosed in PFRC specimens in Vf = 1.1% (10kg/m3). Five types of pre-crack width were used (w0 = 0, 0.25, 0.50, 0.75 and 1.0 mm) and controlled by a rigid steel frames system that confined specimens. Afterwards, pre-crack specimens were tested under direct shear load. Shear behaviour was analysed by means of crack width and slip versus the applied load transformed to normal and shear stresses. Finally, a comparison of behaviour between PFRC and PC was made. The results obtained with fibres showed significantly improved shear strength and a noticeable increase in shear stiffness compared to PC, despite normal stresses remaining similar in both concretes.

Keywords Push-off, shear crack, polypropylene fibre-reinforced concrete, shear

Introduction The objective of the present study was to determine the contribution of polypropylene fibres to shear transfer. For this purpose, the “Z”-shaped specimens were tested under a shear concentrated load. Five pre-crack opening were used (w0 = 0, 0.25, 0.50, 0.75 and 1.0 mm) and controlled by a rigid steel frame system that confines specimens. Finally, a brief comparison with steel fibres was made.

Experimental programme Fourteen “Z”-shaped specimens were tested under direct shear load. The studied parameters were: concrete type and pre-crack opening. Thus five specimens corresponding to PC, three to steel fibre-reinforced concrete (SFRC) and seven to PFRC were tested. For the PFRC specimens, polypropylene fibres were used at a dose of 10 kg/m3, while in the SFRC specimens, hooked-end steel cold-drawn fibres were used at a dose of 30 kg/m3. The mean concrete compressive strength was 41.3 MPa. The entire test was divided into two phases: the precracking process and the push-off test. Five pre-crack opening types were performed: w0 = 0, 0.25, 0.50, 0.75 and 1.0 mm. 82

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

a) Figure 1:

b)

Building the shear interaction diagram for the a) PC, b) PFRC.

As it can be seen in Figure 1, the behaviour of polypropylene fibres showed improvement, especially for the shear stress with narrow crack opening values, as well as, fibres increased the shear stiffness in PFRC compared to PC. No significant differences were found in the normal stress behaviour in both concrete types.

References Echegaray-Oviedo, J., Navarro-Gregori, J., Cuenca, E., and Serna, P. (2017). “Modified push-off test for analysing the shear behaviour of concrete cracks.” Strain, 53(6), 1–17.

83

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Shear crack behaviour and shear deformation of polypropylene fibre-reinforced concrete slender beams Francisco Ortiz Navas1, Juan Navarro-Gregori1, Pedro Serna Ros1, Gabriel Leiva Herdocia1 1

: Institute of Science and Concrete Technology, ICITECH, Universitat Politècnica de València, Valencia, 46022, Spain.

Abstract In recent years, much research has shown improvements in shear behaviour and deformation capability when an adequate amount of macro fibres is provided in concrete. However, very few experiments have used macro synthetic fibres. In this paper, the shear capability of deformation in slender beams was studied by analysing the shear crack path, the crack openingslip relationship and shear deformation of polypropylene fibre-reinforced concrete (PFRC) beams. Shear cracks and deformations were measured by non-contact image measurement techniques. The results are compared with those of plain concrete (PC), steel fibre-reinforced concrete (SFRC) and reinforced concrete (RC) beams. Both types of fibres were dosed so that similar average residual tensile strengths would remain similar to one another. The crack path analysis results showed that synthetic fibres delayed the formation of shear cracks and their propagation into compression zone, and improved the behaviour of secondary cracks due to loss of bond with longitudinal reinforcement. Finally, the crack opening-slip relationship varied widely along the crack and location in beams.

Keywords Polypropylene fibre, shear strength, shear deformation, image-measurement

Introduction The main objective of this paper is to study the cracked behaviour of shear-critical PFRC beams. It particularly intends to study (a) the crack pattern evolution, (a) shear deformation on beams, and (c) the kinematic behaviour of critical diagonal cracks. For this propose, an experimental programme that considered eight PFRC beams was carried out.

Experimental programme Eight full scale beams with a shear span to effective depth ratios of 3.80 and 4.80, and fibre contents of 10 kg/m3 (PFRC) and 30 kg/m3 (SFRC) were tested under a concentrated load at mid-span. Moreover, to analyse crack openings and shear deformation, targets were painted on the beam surface at 100-mm intervals in the vertical and horizontal directions by a non-contact image-based displacement technique, the relative displacements between the selected targets were measured, and hence deformations and strains could be calculated as it is shown in Figure 1.

84

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

Shear stress - strain response.

When comparing the shear strains between beams PC and PFRC (Figure 4), we found that the shear strain at peak stress on the PC beam was in the order of 0.8 mm/m, while the strain was approximately 6.5 mm/m for the PFRC beams. This increase in shear strain represented around 8-fold the shear strain provided by the PC beams, and was almost the same shear stressstrain provided by the A2 beam, including transverse reinforcement.

References Conforti, A., Minelli, F., and Plizzari, G. A. (2017). “Shear behaviour of prestressed double tees in self-compacting polypropylene fibre reinforced concrete.” Engineering Structures, Elsevier Ltd, 146, 93–104.. Navarro-Gregori, J., Ortiz Navas, F., Leiva Herdoncia, G. E., Serna, P., and Cuenca E. (2016). “Experimental reexamination of classic shear-critical concrete beams tests including fibers.” Proceedings of 9th RILEM International Symposium on Fibre-Reinforced Concrete, Vancouver.

85

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Enhancements in Fracture Behavior and Shear capacity in Hybrid Steel and Macro-Polypropylene Fiber reinforced concrete Sahith Gali1, Chiranjeevi Reddy Kamasani1, Subramaniam Kolluru. V. L2 1 2

: Research Assistant, Indian Institute of Technology Hyderabad, Hyderabad, India. : Professor, Indian Institute of Technology Hyderabad, Hyderabad, India.

Abstract The load response from the fracture tests on notched beams of control (SFN00), the steel fiber (SFN05) and the hybrid blend (HYN05) have an initial peak load, which is followed by a softening indicated by a decrease in the load with increasing crack opening displacement. The rate of load decrease with increasing CTOD in the softening part of the load response is significantly smaller in the HYN05 specimens when compared with SFN05 specimens. Following the softening, the residual load carrying capacity is sustained at a high value with increasing CTOD in HYN05. With increasing CTOD, there is a strain hardening in the load response of the SFN05 specimens. Following UNI 11039-2, the first crack flexural strength (fIf) and the equivalent flexural strengths (feq(0-0.6), feq(0.6-1.5)) were determined and are listed in Table 1. The hybrid fiber blend provides better early crack control due to higher early resistance to crack opening when compared to steel fibers. Fiber blends at identical volume fraction as steel fibers produce significant improvement in the early post-peak fracture energy. Table 1

Fracture Parameters per UNI 11039-2.

Specimens SFN00 SFN05 HYN05

FL (kN) 13.5 15.48 15.9

fIf (MPa) 3.89 4.45 4.57

feq(0-0.6) (MPa) 3.37 3.52

feq(0.6-1.5) (MPa) 3.99 3.67

The load responses of reinforced concrete beams tested in shear are shown in Figure 1. In the control specimens, abrupt failure was produced by a dominant shear crack. Failure in concrete without fibers is produced by the loss of stress transfer across the primary shear crack. The failure could not be controlled even in displacement control. With the addition of 0.5% volume fraction of steel fibers, there is an increase in the peak load carrying capacity. Following peak load, there is progressive softening in the load response and a gradual decrease in the load carrying capacity with increasing deflection. A significant improvement in the shear capacity is obtained from the hybrid blend when compared with the steel fibers at identical volume fractions. In HYS05 series, one beam labelled HYS05-1 exhibited an almost constant load carrying capacity with increasing deflection after peak load. A softening type load response with a decrease in the load carrying capacity after peak was observed from HYS05-2. Crack patterns in control, steel fiber and hybrid fiber reinforced beams at the peak loads identified from strain contours obtained from Digital Image Correlation (DIC) technique are shown in Figures 2(a), (b) and (c), respectively. In the control beam (SFS00-1), the abrupt

86

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

opening of the primary shear crack produced failure. In SFS05-1 beam, the primary shear crack was fully established between the flexural reinforcement and the load point at the peak load (Point III in Figure 1(b)). In HYS05-2, a secondary shear crack was formed at the peak load (Point V in Figure 1(c)). DIC measurements indicate a continuous increase in the opening displacement across the primary crack up to peak load. A plot of the load as a function of the crack opening across the primary shear crack is shown in Figure 2(d). The continuous increase in the crack opening across the primary shear crack during the load response indicates a continuous dilatant response in all specimens. The high early resistance provided by hybrid fibers results in a delay in crack formation and a significant increase in the load with an increase in crack opening. A higher load is sustained for the same crack opening in the hybrid blend when compared with steel fibers.

(a) Figure 1:

(a) Figure 2:

(b)

(c)

Load-deflection response of shear beams: (a) SFS00, (b) SFS05 and (c) HYS05.

(b)

(c)

(d)

Crack patterns at the peak load of shear beams: (a) SFS00, (b) SFS05, (c) HYS05 and (d) crack opening across the primary shear crack during the load response.

The variation in the continuous dilatant response across the primary shear crack measured from the control, steel fiber only and hybrid blend indicates different levels of crack opening control in all three. Even at the large crack opening, the fibers provide crack-bridging stresses, which ensures contact stress across the crack plane. The cohesive stress produced by the fibers therefore increases the shear capacity of the frictional interface across the crack faces. The increase in the cohesive traction with crack opening is reflected in the equivalent flexural strength, feq(0-0.6), which provides a measure of the energy dissipated in the crack opening up to 0.6 mm. There is an increase in the feq(0-0.6) obtained from the hybrid fibers when compared with the steel fibers. Improvements in shear behavior are related to the increase in the post-peak fracture energy. The fracture energy using UNI 11039-2 procedure for crack opening up to 0.6 mm provides a good indication of shear capacity enhancement obtained with fibers.

87

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental investigation of shear-critical prestressed steel fibre reinforced concrete beams Maure De Smedt1, Kristof De Wilder1, Dimitrios Anastasopoulos2, Edwin Reynders2, Guido De Roeck2, Lucie Vandewalle1 1

: Building Materials and Building Technology Section, Department of Civil Engineering, KU Leuven, Belgium 2 : Structural Mechanics Section, Department of Civil Engineering, KU Leuven, Belgium

Abstract This paper presents an experimental research on the shear behavior and capacity of prestressed steel fibre reinforced concrete (SFRC) beams. It aims to contribute to the overall understanding of shear in structural SFRC elements and to investigate existing analytical design models. The very brittle shear failure mode could specifically benefit from the enhanced postcracking behavior of adding steel fibres to concrete. However, the use of SFRC is still limited with respect to its potentials due to the lack of international building codes for the shear behavior of SFRC. This is mainly caused by the complex behavior and many interrelated parameters, resulting in an unclear analytical and numerical modelling of shear in SFRC beams. Six prestressed I-shaped beams (numbered from B401 to B406) with a length of 7000 mm and a height of 630 mm were subjected to a force-controlled four-point bending test until failure. Table 1 gives an overview of the beams. The three investigated parameters were the amount of prestressing force p0, the presence of shear reinforcement w, and the fibre dosage Vf (Dramix RC-80/30-CP). The concrete mixtures have a mean cylindrical compressive strength of 74.3 MPa (standard deviation of 6.7 MPa), according to EN 12390-3. The concrete mixture with 20 kg/m³ steel fibres shows a softening behavior while 40 kg/m³ of fibres shows a hardening behavior under bending loads, according to EN 14651. During the progressive damage loading test, failure mode and load, as well as deformations, displacements and cracking pattern properties were observed by means of conventional measurement devices and advanced optical techniques. The latter include Bragg grated optical fibres (FBG), to accurately monitor the horizontal strains at top and bottom flange, and digital image correlation technique (DIC), to analyse the full-field displacement and deformation field at left and right side of the beam. Table 1:

Overview of experimental program and investigated parameters. w = 2.693 ·10-3

w = 0

w = 0

Vf = 0 kg/m³

Vf = 20 kg/m³

Vf = 40 kg/m³

p0 = 1488 MPa

B401

B402

B403

p0 = 750 MPa

B404

B405

B406

The experimentally observed load-displacement curves are shown in Figure 1(a). All beams displayed a post-cracking behavior and failed in a shear failure mode due to diagonal tension. A decreased prestressing force leads to a limited elastic region; the occurrence of bending 88

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

cracks at first (detected by the horizontal strains of the FBG measurements); an increased inclination of the shear cracks; smaller crack widths (measured by the DIC results); a lower amount of released energy and a less brittle failure; and a decreased shear capacity but an unchanged deflection at midspan. An increased fibre dosage leads to a larger post-cracking behavior and multiple cracks at failure; a less brittle failure due to a redistribution of the internal forces and a gradual energy dissipation; and an increased shear capacity. Beams without steel fibres but with shear reinforcement have the largest post-cracking behavior but the shortest elastic region and crack load; and one major crack and very brittle failure occurs. The experimentally observed failure loads of the beams (Vu,exp) are compared to analytical predictions (Vu,pred), based on different models and by omitting safety factors and using the measured average material properties. Eurocode 2 is used for B401 and B404, and Model Code 2010 for B402, B403, B405 and B406. For the six beams, an average experimental-to-predicted failure load ratio of 1.43 was found with a coefficient of variation of 7.2%. Furthermore, four other analytical models for shear design of SFRC are investigated, namely DRAMIX Guideline, RILEM TC 162-TDF sigma-epsilon method, CNR-DT 204/2006 model and a model proposed by Soetens. The results in Figure 1(b) show an underestimation of the predicted shear capacities of prestressed SFRC beams in all the models. The underestimation increases for a higher prestress level, whereas the correlation with the fibre dosage varies within the models. Furthermore, both the experimental and analytical failure loads show that replacing the amount of shear reinforcement w = 2.693 ·10-3 by 40 kg/m³ of steel fibres results in a similar shear capacity for the specimens tested in this research.

(a) Load-displacement curves as experimentally observed at location of the loading point. Figure 1:

(b) Comparison of failure loads and analytical predictions of the shear capacity of SFRC beams.

Results of the progressive four-point bending tests of the six (SFRC) beams.

Keywords Shear, prestressed steel fibre reinforced concrete, experimental mechanics, analytical models, advanced optical measurement techniques

89

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Fibres as shear reinforcement in RC beams: an overview on assessment of material properties and design approaches Joaquim A.O. Barros1, Stephen J. Foster 2 1 : Full Prof., ISISE, University of Minho, Guimarães, Portugal. 2 : Professor and Head, School of Civil and Environmental Engineering, UNSW Sydney.

Objectives The concept of a physical-mechanical model, referred to as the integrated shear model (ISM), is described for determining the shear strength of beams where Mode II fracture is considered. The results are compared to fib Model Code 2010 (MC2010) model results for determining one-way shear capacity of beams.

Integrated shear model (ISM): formulation and predictive performance The ISM is based on coupling comprehensibly the following main modules (Table 1 and Fig. 1a): i) a fibre orientation profile for determining the number of fibres crossing the shear crack in discrete intervals of fibre orientation; ii) the fibre pull-out constitutive law determined according to the unified variable engagement model (UVEM); iii) and the modified compression field theory (MCFT) for the evaluation of the crack width at the shear failure stage (wu). Table 1 - Relevant equations of the ISM.  ( w) = kG min ( 0.4 f R 2 + 1.2 ( f R 4 − f R 2 )  ( w), fctk ,min ) ;  ( w) =  w − 0.25

(

we = 1 3.5 d f tan 3   max  2

)

(7)

(

Lcru ,i = d f  fu

 =  −  for 0    

(8)  =   fu fu

 max =  +  2 for  2   max  

(9)  bu ,i = kb

PFPCL ( wu ) = ku ,i d f  bu ,i Lbf ,o

(10)

i

(

) ( 2 bu,i )

( 2 umax )

fcm + f 1 − cos( u ,i 2)   

)2 k

f Ri = k1 V f l f d f ... i = 1, 2, 3, 4 k1 = 10.5, 9.2, 8.0. 7.0 and k2 = 0.80, 0.75, 0.70, 0.65 for fR1, fR2, fR3 and fR4, respectively (the values for fR2 are interpolated from those for fR1 and fR3)

(6) (11) (12) (13) (16)

  wu  we,i   ... for  wu  Lbf ,o  0  ku ,i =    Lbf ,o − wu  Lcru ,i   ... for we,i  wu  Lbf ,o  2 Lbf ,o − wu l f  =5/12 for EN 14651 and RILEM TC 162-TDF;  =1/3 for ASTM C1609;  =43/84 for the UNI 11039; kG=0.7 for ASTM 1609; kG=0.6 for EN 14651, RILEM TC 162-TDF and UNI 11039; fctk ,min = 0.7 ( 0.3 fck 2/3 ) ; we =fibre’s

((

))

engagement crack displacement; df=fibre’s diameter; lf=fibre’s length; Lbf,o=fibre average bonded length (=lf /4); Lcru ,i =critical fibre embedment length;  fu =fibre’s effective tensile strength;  fu =fibre’s uniaxial tensile strength;  bu,i =fibre’s average fibre bond strength; f = 4.5 MPa; kb=bond coefficient dependent steel fibre type.

90

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Testing the ISM using a database of 122 reinforced SFRC beams, a predictive performance better than the MC2010 approaches was obtained (Figure 1b-1d). The MC2010_EEN and MC2010_MCFT approaches represent the MC2010 models based on the concept of fFtuk, and MCFT with  k ( wu ) evaluated according to Eq. (6), respectively. In all analyses unit values were used for the safety factors, and average values were used for the material properties. The one-way shear models are evaluated and their performance compared. The average and test-tomodel prediction ratio () and coefficient of variation (CoV) for the ISM approach was 1.12 and 16.6%, respectively, whereas for the MC2010_EEN and MC2010_MCFT models the averages and CoVs were 1.32 and 23.4% and 1.32 and 24.2%, respectively.

(a)

(b)

(c)

(d)

Figure 1: (a) Fibre orientation and loading direction in fibre engagement concept; (b) to (d): Test to model comparisons for shear strength: (b) ISM; (c) MC2010_EEN; (d) MC2010_MCFT.

Acknowledgements The authors wish to acknowledge the grant SFRH/BSAB/114302/2016 provided by FCT and the Australian Research Council grant DP150104107, as well as the support provided by the UNSW for the research activities carried out under the status of Visiting Professorial Fellow for the first author. The support of the FCT through the project PTDC/ECM-EST/2635/2014 is also acknowledged.

91

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

92

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Seismic/Special loading conditions

93

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Performance of Ductile FRCC under Cyclic Loads and NonLinear FE Simulation Georgiou Antroula1, Pantazopoulou Stavroula2 1

: University of Cyprus, Nicosia, Cyprus. 2 : York University, Toronto, Canada.

Abstract Earthquake ground motion imposes reversed cyclic displacement demands on structures and where Capacity Design concepts are enforced, the “strong column-weak beam” requirement for frames is implemented in the form of plastic hinge formation in the beam-ends and at the base of ground-floor columns. Beam and column hinging regions must be detailed specifically in order to sustain excessive yielding in longitudinal reinforcement without shear failure (Eurocode 8-III, 2005). Whereas concrete tensile strength is neglected in estimating flexural strength, yet, spalling of the concrete cover in the compression zone is a usual failure that limits member resistance. In light of these practical limitations the use of a new type of composite material reinforced with short discontinuous fibers that exhibits strain hardening properties (SHFRCC) under tension and improved performance in compression is considered an ideal alternative. The improved performance in tension benefits the composite’s behavior in shear as seen in the short span beams subjected to cyclic loading in previous studies (Kanta, Watanabe, & Li, (1998), Fischer & Li (2007), Yuan and Pan (2013)). Limited experimental research concerning the reversed cyclic loading of steel reinforced SHFRCC members has been published todate [Lequesne, Parra-Montesinos, & Wight, (2016), Fischer & Li, (2003), Yuan, Pan, Dong, & Leung, (2014), Parra-Montesinos & Chompreda, (2006)] to support generalized conclusions; however, it is evident that – owing to its unique properties, SHFRCC can work effectively together with longitudinal steel reinforcement and improve the ductility of reinforced concrete structural members without the requirement of significant additional transverse reinforcement. The present paper investigates the behavior of SHFRCC-RC Structural members under cyclic loads, simulating seismic effects. To this end, the experimental responses of two half-scale interior beam column connections subjected to reversed cyclic loading are compared; one of the connections was constructed with a cementitious matrix without fibers, and was detailed according with the Eurocode provisions for ductility class M (moderate, μ=3.5). The other connection was constructed with a SHFRCC mix (2% by volume of PVA fibers was used to reinforce the matrix) and non-seismic detailing amounting to minimum code-specified transverse reinforcement;. Specimens were tested under flexure shear with no simultaneous axial load. The purpose of the tests was to investigate the beneficial action of the fibers on the shear/flexural capacities of members under reversed cyclic loading. The experimental results (Fig. 1) indicate that the SHFRCC specimen showed improved behavior under reversed cyclic loading due to the presence of the fibers that contributed to the transfer of loads, the confinement of the regions under tension and compression, the increase of shear capacity of the member and the sustained load capacity up to large levels of deflection (enhanced deformation capacity). After vigorous cyclic displacement reversals the test of the SHFRCC specimen was terminated, attaining the limit of stroke of the actuator. There was no 94

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

evidence of strength loss in the load-resistance curve up to a ductility of six in terms of the relative drift ratio (μθ=6). By comparison, the Control specimen (reinforced as per EC8-I DCM) attained a drift ductility of μθ=4/1=4 at a strength loss of 20% of peak load. The experimental findings suggest the potential benefit from a shift of construction towards High Performance Structures. (a)

(b)

Figure 1:

Shear Load-Right Beam deflection for the (a) CONTROL and (b) SHFRCC.

Additionally, the experimentally determined properties in uniaxial tension, split and compression of two mix designs, with and without fibers, were used to calibrate analytical models of the behavior of those composites. Results obtained from FE modeling analysis were compared with the experimental load-deflection curves (Fig. 2). As tensile strength is a very important analysis parameter, procedures for its proper determination for the simulation of SHFRCC members must be incorporated into Modern Codes for these types of materials. Results obtained from direct tensile tests may underestimate the resistance and strain capacity of the composite when this is placed in a 3-D state of stress. Flaws and size effects (in terms of possible voids during compaction for the estimation of stress from the specimens’ area, and the gauge length used in estimation of average strain from crack widths) may result in lower values than what may be supported in the presence of a more complex stress state and therefore the split cylinder results must be used for determining the materials’ tensile parameter for simulation. Experimental results Finite Element Modeling results

yield

max load

δ max

Figure 2:

Simulation of the SHFRCC member under monotonic push-off analysis and comparison to the cyclic behavior, stresses and corresponding strains.

95

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Effect of synthetic fibers on the quasi-static and blast behaviour of reinforced concrete beams Hassan Aoude1, Yang Li1, Roukaya Bastami1, Steve Castonguay1, Corey GuertinNormoyle1 and Fawwaz Munir Ibrahim1 1

: University of Ottawa, Ottawa, ON, Canada.

Research description This paper presents the results of a study examining the effect of macro-synthetic fibers on the flexural and shear behavior of beams tested under quasi-static and blast loads. In total, ten beams built with normal-strength concrete and synthetic fibers are studied, with five specimens tested under quasi-static four-point bending and a companion set of five beams tested under simulated blast loads using the shock-tube at the University of Ottawa. The beams in this study had a cross-section of 125 mm x 250 mm, a total length of 2440 mm and were simply supported over a span of 2232 mm, with a constant moment region of 750 mm and two equal shear spans of 741 mm. Longitudinal reinforcement consisted of 2-15M bars. Control beams were built with plain SCC, while fiber-reinforced concrete specimens (FRC) where built with SCC and macro-synthetic fibers (type S1 or S2) at volume fractions of 0.75% or 1% (6.75 or 9 kg/m 3). The SCC used in all specimens had a specified strength of 40 MPa (average strength for the concrete in all beams is reported in Table 1). The S1 fibers had a length of 50 mm, aspect-ratio of 74 and tensile strength of 625 MPa. The S2 fibers had a length of 50 mm, aspect ratio of 64, with a tensile strength of 400 MPa. One set of SCC and FRC specimens was built without transverse reinforcement to observe shear response, with the remaining beams built with Ushaped stirrups made from 6.3 mm wire spaced at 100 mm in the shear spans. Specimen nomenclature in Table 1 reflects indicates concrete type (plain SCC or FRC), fiber type (S1 or S2), fiber content (0%, 0.75% or 1%) and shear reinforcement details (“100” for beams built with stirrups). Static testing was conducted under slowly-applied four-point bending (see Figure 1a). Dynamic testing was conducted under gradually increasing blast pressures using the setup shown in Figure 1b. Sample shockwaves corresponding to Blasts 1 to 4 are shown in Figure 1c. Results from the static and blast tests are shown in Figure 1d and Figure 1e-g, respectively. Table 1: Series

Control

FRC

96

Beam test matrix. Specimen Designation

Concrete Strength [MPa]

Fiber Type

Fiber content [%]

Stirrup spacing [mm]

Long. steel

Test type

SCC-0% -0

48

-

-

-

2-15M

S/D

SCC-0% -100

48

-

-

100

2-15M

S/D

FRC-0.75%(S1)-0

52

S1

0.75

-

2-15M

S/D

FRC-0.75%(S1)-100

52

S1

0.75

100

2-15M

S/D

FRC-1%(S2)-0

44

S2

1.0

-

2-15M

S/D

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Results 1.1 Effect of synthetic fibers on shear response Beams SCC-0%-0, FRC-0.75%(S1)-0, and FRC-1%(S2)-0 can be studied to examine the effect of synthetic fibers on shear behaviour under static and blast loads. The beams in this set were detailed without stirrups. Results from static testing show that the use of synthetic fibers did not lead to an increase in shear capacity, however it improved ductility (see Figure 1d). In the control beam, the failure is brittle and occurs suddenly after the formation of the first diagonal shear crack, which leads to beam collapse. The provision of synthetic fibers allows the FRC beams to carry load after the initiation of shear cracking, leading to a more gradual failure. In terms of the effect of fiber type, both FRC beams show a similar response. Under blast loading, all three beams failed in shear, but at different blast intensities (see Figure 1e). The control beam failed at Blast 2 (reflected impulse, Ir = 330 kPa-ms), while beams reinforced with 0.75% of fiber S1 and 1% of fiber S2 failed at Blast 3 (Ir = 460 kPa-ms).

1.2 Effect of synthetic fibers on flexural response The effect of synthetic fibers on flexural response under static and dynamic loading can be examined by comparing the behaviour of beams SCC-0%-100 and FRC-0.75%(S1)-100 which were detailed with plain SCC and FRC having 0.75% of S1 fibers, respectively. The beams in this comparison had transverse reinforcement spaced at 100 mm in the shear spans. The results from static testing indicates that use of synthetic fibers does not lead to a significant effect on beam strength and stiffness. However, the use of fibers delays crushing of concrete in compression and therefore leads to an improvement in beam ductility and toughness. Under blast loading the beams show a similar response, failing in flexure at Blast 4. At Blast 3 & 4, the use of synthetic fibers had a more important effect on reducing residual displacements (see Figure 1f and 1g). The use of fibers also led to improved damage tolerance, and eliminated spalling and secondary fragments which can pose a danger for building occupants.

(a) static setup

(b) Blast setup Blast 2

200

10

Shear failure

0 -10

0 0

10

20

30

40

50

Displacement (mm)

(d) Static results

Figure 1:

60

70

-20 250

300

350

400

450

500

Time (ms)

(e) Blast result (Blast 2)

30 20 10

60 40 20

0

0

-10 300

-20 300

340

380

420

460

500

Time (ms)

(f) Blast result (Blast 3)

SCC-0%-100 FRC-0.75%(S1)-100

80

Displacement (mm)

50

20

100

SCC-0%-100 FRC-0.75%(S1)-100

40

Displacement (mm)

100

Blast 4

50

SCC-0%-0 FRC-0.75%(S1)-0

30

Displacement (mm)

150

Load P (KN)

40

SCC-0%-0 SCC-0%-100 FRC-0.75%(S1)-0 FRC-0.75%(S1)-100 FRC-1%(S2)-0

(c) Blast loads

Blast 3

340

380

420

460

500

Time (ms)

(g) Blast result (Blast 4)

Blast setup, blast load sequence, static results and dynamic test results.

97

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Strain-Based Fatigue Failure Criterion for Steel-Fiber Reinforced Concrete Elisa Poveda 1, Gonzalo Ruiz 1, Héctor Cifuentes 2, Rena C. Yu 1 and Xiaoxin Zhang 1 1

: ETSI Caminos, C. y P., Universidad de Castilla-La Mancha, Ciudad Real, Spain. 2 : ETS de Ingeniería, Universidad de Sevilla, Seville, Spain.

This work proposes a new strain-based fatigue failure criterion for steel-fiber reinforced concrete (SFRC). It is based on the well-known relation between the secondary strain rate per cycle in compressive fatigue, ∂ε/∂n, and the life expressed as the number of cycles at failure, N, of a definite specimen [1], which is log (∂ε/∂n) = m + s log N (1) where m and s are constants. Figure 1a shows the logarithm of the secondary strain rate per cycle with respect to the logarithm of the number of cycles resisted for five SFRCs that share the same concrete matrix. As can be observed, all these points adjust perfectly to a single straight line that responds to Equation 1. Then this relationship is independent of the amount of fiber and therefore it is a characteristic of the base concrete [2], i. e. the matrix of the SFRC. A basic integration of this equation allows for proposing a strain-based failure criterion as the sum of the initial instantaneous strain in the specimen due to the maximum compressive stress, ε1, plus the increment of strain due to the cyclic loading and companion creep, that can be estimated m s+1 as 10 N . So, the equation m

s+1

εc = ε1 + 10 N (2) where εc is the critical strain at fatigue failure, states that a specimen fails when it is strained to εc during the cyclic loading. Contrariwise, it keeps resisting while its accumulated strain is smaller than the critical one, ε ≤ εc. Figure 1b shows the failure criteria and three different cyclic curves in which the failure occurs when the strain reaches the thickest curve. Figure 2 shows

Figure 1: a) Adjustment of regression in the experimental tests of the logarithm of the secondary strain rate with respect to the logarithm of the number of cycles, b) fatigue failure criterion with three cyclic curves drawn.

98

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 2: a) Steel fiber used, b) number of cycles to failure in the compressive fatigue tests per concrete and c) influence of the fiber dosage.

the fibers used in the SFRC as well as the fatigue behavior of the experimental tests for each concrete: Figure 2b shows the cycles resisted and the Weibull fitting in solid lines, whereas Fig. 2c plots the mean value of the log N for each fiber content. Our results also prove that it happens to exist an optimum in fiber content that results in the longest fatigue life. By contrast, fiber ratios beyond the optimum destructure the matrix and lead to shorter fatigue lives.

References [1] P.R. Sparks, J.B. Menzies (1973). The effect of rate of loading upon the static and fatigue strengths of plain concrete in compression. Magazine of Concrete Research 25:83, 73–80. [2] E. Poveda, G. Ruiz, H. Cifuentes, R.C. Yu, X.X. Zhang (2017). Influence of the fiber content on the compressive low-cycle fatigue behavior of self-compacting SFRC. International Journal of Fatigue 101, 9–17.

99

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Cyclic Damage on PVA Microfibre Embedded in Cementitious Matrix in Alternating Tension-Compression Regime Majid Ranjbarian1, Viktor Mechtcherine1 1

: Technische Universität Dresden, Dresden, Germany

Introduction Strain-hardening cement-based composites (SHCCs) are a suitable type of material for structures which must exhibit high resistance against seismic, impact and cyclic loadings (Kesner & Billington, 2004). However, mechanical performance of SHCC pronouncedly depends on the loading regime. Specifically, under alternating tension-compression loading deterioration of fibres squashed between the crack faces and resulting degradation of their bridging capacity are responsible for early failure necessitating further investigations on microlevel (Müller & Mechtcherine, 2017). Ranjbarian and Mechtcherine recently developed a novel setup (Ranjbarian & Mechtcherine, 2018), which is used in this work for investigating the influence of cyclic pre-loading in tension-compression mode on PVA microfibre bridging behaviour. The results are compared with the crack bridging behaviour of PVA microfibre tested under quasi-static monotonic regime with the same setup.

Experimental program The test setup is shown in Figure 1a. PVA microfibre Kuralon K-II REC 15, Kuraray (Japan) is used with the matrix of the SHCC investigated also in other research projects at the TU Dresden, e.g. Müller & Mechtcherine (2017). The force-displacement curve progression during entire loading regime is schematically shown in Figure 2b. After applying monotonic tensile load for breaking the matrix on the notches and reaching a crack width of 100 µm, cyclic loading is applied with 200 and 2000 number of cycles, cases “C200” and “C2000”, respectively, followed by pulling the fibre out of the matrix. The displacement controlled tests conducted with displacement rates of 0.01, 1 and 0.01 mm/s for these three stages, respectively. Figure 1c shows the bridging behaviour of PVA fibre embedded in cementitious matrix under monotonic loading, see Ranjbarian & Mechtcherine (2018) for more details.

(a) Figure 1:

(b)

(c)

(a) Double-sided specimen geometry (thickness 4 mm) and setup configuration, (b) applied loading regime, (c) fibre bridging behaviour under monotonic loading.

Results and conclusions Figure 2a shows the force-displacement curves in the final pull-out stage of double-sided fibre pull-out tests under different loading regimes. The cyclic pre-loading leads to different 100

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

pull-out behaviour. For “C200”, the fibre is in final pull-out stage completely pulled out from the matrix in “softening” regime, however, PVA fibre usually tends to rupture in postdebonding stage. The repeated fibre pull-out and push-in during cyclic loading regime causes superficial deteriorations of the embedded fibre, i.e. bidirectional scratching and abrasion, which changes the damage development, leading to complete fibre pull-out. In the specimen “C2000”, the fibre eventually ruptured in the final pull-out stage. However, the rupture force and the corresponding displacement are in this case much lower than those of the control specimen. Frictional forces between the fibre and edges of matrix tunnel close to the fracture surface lead to abrasion on the fibre surface in cyclic damage zone between the crack faces. Also defibrillation is involved in degradation process of the microfibre, see Figure 2b,c.

Figure 2:

(a) (b) (c) Fibre bridging behaviour influenced by alternating tension-compression preloading, (b) and (c) the damage zone belongs to the part of fibre between crack faces for (“C200”) and (“C2000”), respectively.

It can be concluded that there is a possibility of a change in pull-out behaviour of PVA microfibre embedded in cementitious matrix from “fibre rupture” to “fibre pull-out”, as well as from “hardening” to “softening” in post-debonding stage due to the damage induced under cyclic pre-loading in tension-compression mode. The findings support the damage mechanisms suggested for SHCC under cyclic tension-compression regime by Müller & Mechtcherine (2017). Local deterioration of the fibre due to alternating tension-compression loading compromises the bridging capacity of the fibre and can prevent multiple cracking, leading to lower ductility on composite level.

References Kesner, K., & Billington, S. L. (2004). “Tension, Compression and Cyclic Testing of Engineered Cementitious Composite Materials.” Mceer, 120. Müller, S., & Mechtcherine, V. (2017). Fatigue behaviour of strain-hardening cement-based composites (SHCC). Cement and Concrete Research, 92, 75–83. http://doi.org/10.1016/j.cemconres.2016.11.003 Ranjbarian, M., & Mechtcherine, V. (2018). A novel test setup for the characterization of bridging behaviour of single microfibres embedded in a mineral-based matrix. Cement and Concrete Composites, under Review.

101

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Seismic Performance of Fibre Reinforced Concrete in the Absence of Bars Erik Stefan Bernard1 1

: Technologies in Structural Engineering Pty Ltd, Sydney, Australia.

Keywords Reverse-cycle loading, post-crack performance, seismic loading, robustness, reinforcement

Abstract Numerous investigations of the effect of fibre addition on the seismic performance of conventionally reinforced concrete members have been published. These generally show that fibres can improve robustness and survivability during reverse-cycle loading, but the dosage rate of fibre required to achieve significant improvements in performance is substantial. Recently, pure FRC members have increasingly been used in structures such as tunnel linings, including both fibre reinforced shotcrete and pre-cast FRC segments. Concerns have been raised about the absence of data on the seismic resistance of such members given that all previous research on seismic performance has essentially involved hybrid members incorporating both steel reinforcing bars and fibres. The present investigation has focused on the reverse-cycle flexural performance of FRC members in the absence of conventional steel reinforcing bars. Laboratory testing was performed on plain, bar-reinforced, and steel fibre reinforced concrete members, and their performance was compared. The tests indicate that steel fibres provide a small improvement in flexural capacity under reverse-cycle loading compared to plain concrete, but that the robustness of pure FRC members is relatively poor compared to steel bar-reinforced members incorporating steel stirrups. The data suggest that, when used at practical dosage rates, large hooked-end steel fibres cannot be relied upon to provide seismic performance in flexure comparable to steel bar reinforced concrete members. Ductility in reinforced concrete members is an important property to satisfy strength requirements in response to many forms of structural loading. However, there is some concern over how fibres perform across cracks that repeatedly open and close. Moreover, cracks arise in many orientations during reverse-cycle loading, so loss of lateral confinement may potentially degrade fibre performance. Reverse-cycle loading appears to produce different responses in FRC compared to monotonic flexural loading, so good performance in pure bending tests such as ASTM C1609/C1609M and EN14651 does not necessarily predict good performance in a seismic event. When fibres are used in the absence of conventional steel bars, flexural cracks are widely spaced and strain localisation is likely to be concentrated, especially if the axial compressive stress is low. The result is a limited number of wide cracks.

102

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

Relationship between drift and peak cycle moment for plain concrete and SFRC beams compared to hybrid beams reinforced with 20 and 40 kg/m3 Dramix RC65/60 3D fibres.

The experimental program consisted of a series of reverse-cycle tests on beam-columns made of plain concrete and steel FRC, with some specimens also reinforced with conventional steel bars and stirrups. Each set of four nominally identical beam-columns was cast using a 32 MPa concrete mix. The steel FRC mixtures included either 20 or 40 kg/m 3 Dramix RC65/60 3D fibres both with and without conventional reinforcing bars. The specimens were cast and consolidated in a horizontal position, stripped after two days, and cured in lime-saturated water for four months before testing. The beams-columns all measured 2000 mm in length and 200 × 200 mm in cross-section. The conventionally reinforced specimens had one 12 mm longitudinal reinforcing bar in each corner. The RC beam-columns included stirrups consisting of 6 mm plain round bar bent into a fully enclosed square with 40 mm long end hooks spaced at 250 mm centres along the beam-column. Each series of four beam-columns was tested at the same age, using the same ACI 374-based testing procedure, leading to destructive failures in the high moment region near the centre of each specimen. Peak moment capacity has been plotted as a function of drift in Figure 1. The plain concrete specimens survived remarkably well, with the cracks initially opening and closing in a rocking manner without much damage accumulating in the adjoining concrete. However, beyond about 8-10% drift, longitudinal cracks started to appear perpendicular to the flexural cracks and finally most of the specimens suffered a shearing failure. The central part of each specimen generally broke into large pieces of concrete and collapsed by the end of the test. The inclusion of steel fibres in the mix at dosage rates of 20 and 40 kg/m 3 increased the robustness of the beam-columns at high levels of drift, especially when conventional steel reinforcing bars were also included, but had a negligible effect on performance in the absence of conventional steel reinforcing bars (Figure 1).

103

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Finite Element Analyses of Seismic Response of a 22-story RC Wall Building subjected to Drying Shrinkage Cracking and Application of SCRPCC Keisuke Kitazawa1, Yuichi Sato1, Kazuhiro Naganuma2, Yoshio Kaneko1 1 2

: Kyoto University, Kyoto, Japan. : Nihon University, Tokyo, Japan.

Abstract Reducing of CO2 emission became important tasks in the entire industry. The reuse of steel materials contributes to a reduction in the environmental load. The authors attempted to develop and implement steel chip reinforced cementitious composite (SCRCC), which is reinforced with steel chips instead of conventional steel fibers. The steel chips are produced when steel plates are precisely machined on numerically controlled lathes. To enhance strength, adhesion, bond, waterproofness and durability of building structures, the polymer cement composite (PCC) is one of the promising solutions. On the other hand, it has been thought that the shrinkage cracks may reduce the stiffness of the RC building and affect the vibration characteristics, but few attempt of quantitative estimation has been conducted. The analyses in this paper employ a 3D finite element model of the entire building to estimate the long period (8400 days, or approximately 23 years) drying shrinkage behavior of. The building model is then subjected to the dynamic time history analyses by applying 3D acceleration record. It is supposed that the shrinkage crack increase the response drift of the building during the vibration while the response of a building made of the SCRPCC will be reduced because the SCRPCC reduce the opening of shrinkage cracks. These phenomena are quantitatively discussed in the following sections. The analysis method proposed by the authors’ previous study adopts the bond model based on fib Model Code 2012, models of shrinkage strain and creep based on CEB-FIP Model Code 1990 and a finite element program incorporated with the discrete-like crack simulation method. The discrete-like crack simulation method was developed by aiming at an explicit computation of crack widths, which used to be impossible by using the conventional smeared crack-based finite element method. The summed strain of the shrinkage strain and the creep strain is input for the shrinkage cracking analysis. The analyzed building is 22-story-high RC wall building (Fig. 7a) located in Nonoalco Tlatelolco housing complex in Mexico City and severely damaged in the 1985 Mexico Earthquake (AIJ 1986, Petrovski et al. 1988). The structure is 68.81 m-high, composed of 200 mm-thick walls and 280 mm-thick waffle slabs. The reinforcement ratios of the walls range from 0.23% up to 1.13%, and the bar diameter from 9.4 mm up to 31.8 mm. The building was designed based on old guidelines in 1950’s and it is not probable that a new kind of FRC like the SCRPCC is used for such an old structure. However, the building was supposed to attain large drifts by relatively long-period vibration so as the concrete material was subjected to critical stresses and strains. This may be one of ideal conditions to examine the performance of the SCRPCC.

104

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

The analyses conduct the drying shrinkage cracking computations before the time history analyses. The analysis parameters are the material (ordinary concrete and the SCRPCC) and consideration of drying shrinkage (with/without shrinkage crack analyses). Therefore, four cases are analyzed; (1) ordinary concrete without shrinkage, (2) SCRPCC without shrinkage, (3) ordinary concrete with shrinkage, and (4) SCRPCC with shrinkage. The same constitutive models are adopted for the ordinary concrete and SCRPCC except the tensile softening characteristic (Fig. 2b). The drying shrinkage strain is calculated based on the CEB-FIP Model Code 1990 and impose the strain corresponding to 8400 days to the building by the initial 250 steps. For the time history seismic response analyses, an acceleration data SCT1 (AIJ 1986) recorded is input to the north-south, east-west and up-down direction. The building model is made of 18,228 four-node quadrilateral shell elements and 226 two-node truss elements. The former are used for the wall while the latter for the braces. Each quadrilateral shell element is divided into ten layers along the thickness direction so that the element can consider the out-ofplane bending. The total degrees of freedom of the model are 111,865. The gravity acceleration is input at the initial two steps. Then the time history analyses are conducted by 0.01-second time increment up to 6000 steps (i.e., 60 seconds). The coefficients β=0.25 and γ=0.5 of time increment are adopted for Newmark β method. Damping is primary stiffness proportional type, and we add 1% damping in a rate of primary natural period 1.22 seconds, which is evaluated by the eigenvalue analyses using the Subspace iteration method. Each case attained the maximum drift around 1% by the time history analysis. The vibration patterns look similar, but considerable differences are found with respect to the maximum drift. For the NS direction, the maximum drift of the ordinary concrete model is increased by 10.8% (i.e. from 1.02% to 1.13%) by considering the shrinkage. The increasing rate is reduced to 6.9% (i.e. from 1.02% to 1.09%) by the SCRPCC. For the EW direction, the maximum drift of the ordinary concrete model is decreased by 5.6% (i.e. from 1.07% to 1.01%). It is supposed that the shrinkage cracks and the cracks induced during the vibration excessively elongate the natural period of the building and reduce the resonance effect with the ground motion. The maximum drift of the SCRPCC model is reduced by 13.2% (i.e. from 1.06% to 0.92%), which is considerably smaller than the ordinary concrete model. The average crack widths of the models without the shrinkage are 0.35 mm for the both model. If the shrinkage is considered, the width is increased up to 0.45 mm for the ordinary concrete model. The width is reduced to 0.40 mm for the SCRPCC model.

Keywords Drying shrinkage crack, Steel chip reinforced polymer cementitious composite, 3D FEM

105

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Residual crack width in RC and R/FRC ties subjected to repeated loads Alessandro P. Fantilli1, Francesco Tondolo1 : Politecnico di Torino, DISEG – Department of Structural, Building and Geotechnical Engineering. 1

Abstract In this research project, crack width is measured on a series of specimens by using traditional mechanical strain gauges and a new device based on the optical conoscopic holography. The latter allows the non-contact measure of crack profile, at the end of each loading cycle, both in plain and fiber-reinforced ties subjected to sets of repeated loads. The specimens consist of RC and R/FRC ties with circular cross section as shown in Fig.1a, where a scheme of the test setup is illustrated. On each specimen, three couples of bases for the mechanical extensometer (positioned at 120 degrees around the specimens in Fig.1a) were installed. Two potentiometer transducers were positioned on the ends of the specimens to measure the relative slip between the reinforcing bar and the concrete, whereas two LVDTs where used to obtain the whole elongation of specimen measured on the reinforcing bar (see Fig.1a). The whole experimental campaign included 20 RC and 20 R/FRC specimens grouped in the 8 series reported in Fig.1b, which differ only by the presence of fibers and of geometrical properties.

Series

Figure 1:

Bar diameter

Concrete radius

ρ=As /Ac

Fiber volume

[mm]

[mm]

[%]

[%]

D10-LP

10

87

1.34

-

D10-HP

10

60.4

2.82

-

D20-LP

20

174

1.34

-

D20-HP

20

121

2.81

-

D10-LF

10

87

1.34

0.5

D10-HF

10

60.4

2.82

0.5

D20-LF

20

174

1.34

0.5

D20-HF

20

121

2.81

0.5

The specimens investigated in this research project: a) the arrangement of the instrumentation, b) the specimens tested in uniaxial tension.

The results of the tests in terms of crack width vs. number of cumulative cycles are illustrated in Fig.2. In particular, in each diagram the crack widths of two similar series (with and without fibers) are compared both at the peak of load and at the end of each cycle.

106

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 2:

a)

b)

c)

d)

Average values of crack width (peak – continuous lines, and residual – dashed lines) for the same type of specimens (RC – orange lines, and R/FRC – blue lines): a) D10-LP and D10-LF, b) D10-HP and D10-HF, c) D20-LP and D20-LF, d) D20HP and D20-HF.

In the case of low reinforcement ratio (Fig.2a and Fig.2c), at the peak of load, crack widths in RC specimens (continuous orange lines) are always larger than that of FRC specimens (continuous blue lines). Whereas, there is not a great difference between plain concrete and fiber-reinforced ties when the reinforcement ratio is higher (Fig.2b and Fig.2d). Accordingly, the fiber capability of bridging the crack surfaces is more evident when the bond stresses at the interface between rebar and concrete is low, as happens herein in the specimens with =1.34. As the load increases with the number of cycles, crack width increases as well, also denoting a progressive reduction of the bond between steel and concrete during the cycles. In addition, larger crack widths were also measured in the larger specimens, according to the well-known size-effect phenomena Also residual cracks show different widths depending on the presence of fibers, especially in the case of low reinforcement ratio (Fig.2a and Fig.2c). Nevertheless, crack width in R/FRC samples (dashed blue lines) is generally larger than that of the corresponding RC ties crack width in RC specimens (dashed orange lines). This result can be ascribed to the well-known mismatch of the crack surfaces that occurs at unloading, which seems furtherly amplified by the presence of fibers. In all the cases, residual crack widths are more or less constant regardless of the number of cycles, and, similarly to the crack width at peak, they are wider at larger scale.

107

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A method for strengthening and thermal-insulating cavity walls with Fiber Reinforced Mortar: the case of Groningen Fabiola Iavarone1, Sara Lucchini1, Luca Facconi1, Antonio Conforti1, Stefano Sacrato2, Giovanni Plizzari1 1 : University of Brescia, Brescia, Italy. 2 : Tecnologia e Ricerca Italiana srl, Brescia, Italy.

Abstract After many years of micro-earthquakes in the Groningen area (the Netherlands) resulting from the production of shale gas by hydraulic fracturing (fracking), a new retrofitting method has been developed and experimentally assessed to increase the seismic resistance of houses built with masonry cavity walls. The three-years research aimed at the following main targets: 1. Development of a proper mix design able to provide Fiber Reinforced Mortar (FRM) compressive and tensile mechanical properties suitable for its employment as a retrofitting material in seismic design. Particular attention has been also devoted to the insulating properties of FRM in order to reduce thermal energy consumption of existing masonry cavity wall buildings; 2. Study of the long-term behavior to verify humidity transfer with severe environmental parameters (thermal shock, heavy rains, icing-thaw) and mechanical-parameters stability; 3. Definition of correct method statement for the full-scale implementation of the proposed retrofitting technique. Material properties The ARMOX CWG material was prepared by using non-conventional binder based on Sulphoallumante cement, inorganic nano-binders and Portland Cement. The aggregate of FRM is mainly thermal-insulating cellular glass and acicular Aluminum Silicate grains. Material mix design has been completed with extruded polypropylene macro synthetic fibers formed into a crimped profile, 40 mm long and with a diameter of 0.75 mm. Flexural and Physical tests on masonry wallets infilled with FRM The low-density FRM has a hardening post-cracking behavior under flexure, as proved by three-point-bending tests performed according to EN 14651. The adhesion between mortar and masonry is so significant that no delamination has been observed at the masonry-to-FRM interface during the flexural tests on masonry wallets infilled with FRM (Figure 1). Therefore, no additional devices are required to provide connection between FRM and masonry leaves. In addition, thermal insulation resulted 34% higher as compared to a not-infilled wall. 40

Load [kN]

Front Back

35 30 25

20 15 10 5

Crack opening [mm] 0

0

2

4

6

(a) Figure 1:

108

8

10

12

14

(b)

Flexural test: Load – Crack-opening response (a), failure pattern (b).

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Results mechanical properties tests Test ID

Test description

Units

2b

2c

UNI EN 1015-11 2

N/mm

Test results Min

Max

10,01 2,94 4570

7,87 2,74 4527

11,43 3,19 4618

UNI EN 12390-13

0,87 1,17 1,35 1,19 0,98

0,78 0,84 1,01 0,89 0,68

0,98 1,77 1,98 1,65 1,21

2

N/mm

UNI EN 14651

Test ID

Test description

Units

Test protocol

3a

Coefficient of thermal conducibility

W/mK

EN ISO 12664

2 1/2

3b

Capillarity water absorption

3c 3d 3e 3f

Water absorption at atmospheric pressure Vapour permeability Vapour transfer and influence on building envelope Freezing-Thaw resistance with de-icing salts - spalling Determination of adhesive strength of hardened rendering and plastering mortars on substrates

3g

Average

N/mm2

Compressive and flexural test Compressive strength Flexural strength Modulus of Elasticity Post cracking flexural test fL fR1 fR2 fR3 fR4 Results physical properties tests

2a

Test protocol

UNI EN 15801

kg/m s

kg/m

UNI 7699 UNI EN 1015‐19 UNI EN 16322 UNI CEN/TS 12390-9:2017

MPa

UNI EN 1015-12

% Kg/msPa 2

Test results Average Min Max 0,146 0,144 0,148 0,0102 0,0228 16 1,66E-12 1,48E-12 9,06E-12 0,18845 0,1448 0,2553 1,08 0,45 1,59 0,83

0,9

1,94

Long-term behavior No mechanical and physical properties deteriorations under severe environmental parameters (thermal shock, heavy rains, icing-thaw) were observed. FEM analyses 1000

V [kN]

800 600 400 200

δ [mm]

Name

0 -200 without

-400

ARMOX (plaster)

-600

ARMOX with 20% less energy

-800

ARMOX CWG (Theoretical 1) ARMOX CWG (Theoretical)

-1000

ARMOX CWG (P13)

-1200 -15

-10

Figure 2:

-5

0

5

10

15

without ARMOX (plaster) ARMOX WITH 20% LESS ENERGY ARMOX CWG (THEORETICAL 1) ARMOX CWG (THEORETICAL) ARMOX CWG (P13)

(ag/g)+ 0,228 0,445

Increment 95%

(ag/g)0,250 0,525

Increment 110%

0,440

93%

0,515

106%

0,420

84%

0,460

84%

0,405

78%

0,455

82%

0,370

62%

0,425

70%

Pushover analyses: Load-Displacement response of a typical terraced house.

Method Statement A method statement has been proposed and tested on full-scale mock up models. The latter has shown that a real house can be strengthened in a couple of days by working only from the outside of the building and without forcing the inhabitants to move out. Furthermore, the original aspect of the structure was totally preserved. Keywords Groningen; Cavity walls; Infill material; Strengthening; Thermal-insulating; Fiber Reinforced Mortar; Hybrid reinforcement; Non-linear finite element analysis

109

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Flexural fatigue performance of plastic fiber and steel microfiber in reinforced concrete Débora Martinello Carlesso1, Albert de la Fuente Antequera1, Sergio Henrique Pialarissi Cavalaro2, Wellington Longuini Repette3 1

: Polytechnic University of Catalonia - BarcelonaTECH, Barcelona, Spain. : Loughborough University, Loughborough, United Kingdom 3 : Federal University of Santa Catarina, Florianópolis, Brazil. 2

Abstract The present paper deals with an experimental study on the fatigue behavior under bending of fiber reinforced concrete pre-cracked beams. The post-crack fatigue performance of two different plastic fibers (polypropylene copolymer based and modified olefin based) and the influence of specimen size were compared by means of three point bending tests, considering an initial crack width accepted in the service limit state. Apart, an investigation with ultra-high performance fiber reinforced concrete (UHPC) was conducted considering four different values for the upper limit of applied load. The mechanical response of these three concretes was evaluated through compressive strength, elastic modulus and static monotonic bending test. The results of the fatigue tests were analyzed as a function of the crack mouth opening displacement (CMOD) and total number of cycles. In addition, the analysis was performed regarding the crack opening range, toughness and the number of fibers in the cross section. Results suggest that fatigue in plastic fiber reinforced concrete (PFRC) depends on the fiber stress level varying from cementitious matrix failure to a secondary fatigue of the fiber, depending of the number of cycles. In contrast, fatigue in UHPC suggests the occurrence of fiber pull-out. From the experimental intrinsic scatter of the fatigue phenomenon observed, in particular for high levels of applied fatigue load, the amount of fibers in the cracked cross section seems to play an important role in withstanding the fatigue load.

Keywords Cracked section; Fiber reinforced concrete; Fatigue; Plastic fibers; Steel microfibers

110

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Structural rehabilitation

111

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

The behavior of concrete columns confined by UHP-FRCC jacketing Alessandro P. Fantilli 1, Tomoya Nishiwaki 2, Valerio Lisi 1, 2, Monica Longo 1, 2 : Politecnico di Torino, DISEG – Department of Structural, Building and Geotechnical Engineering, Torino, Italy. 2 : Tohoku University, LCEL – Life Cycle Engineering Laboratory, Sendai, Japan. 1

Abstract Ultra-High Performance – Fiber Reinforced Cementitious Composites (UHP-FRCC) show excellent mechanical performances, and therefore can be effectively used to retrofit concrete structures. Similarly to the traditional Reinforced Concrete (RC) jacketing, also when a layer of UHP-FRCC is applied on an existing column a sort of confinement can be obtained. Accordingly, the purpose of this study is to investigate the performances of plain concrete cylinders, confined by UHP-FRCC and subjected to uniaxial compression. In some of the layers, high volume fly ash has also been used to replace part of the cement and reduce the environmental impact. An experimental campaign on concrete cylinders confined by UHPFRCC jackets is described in this work. The cores (Fig.1 and Table 1), with a radius R = 50 mm and a height L = 200 mm, were made with normal concrete (NC) and reinforced by a UHPFRCC jacket, whose length Lj was lower than L. In this way, the jacket only confined the NC cores without offering a direct resisting contribution to compression.

Table 1:

Figure 1:

Geometrical properties of the specimens.

Core radius, R [mm]

Core length, L [m]

Jacket thickness, tj [mm]

Jacket length, Lj [mm]

Total diameter, D [mm]

50

200

25

176

150

NC core confined by UHP-FRCC jacket.

Three series of UHP-FRCC were tailored for the jacket layer shown in Fig.1. With respect to the absence of fly ash (FA0 series), in FA20 and FA70 series, the replacement rates of cement with fly ash were 20 and 70% by weight, respectively. The mix proportions used in this research project are shown in Table 2.

112

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Table 2:

Mix proportions (wt.%) of the UHP-FRCC jackets. Binder (B)

Series

LH Cement / B (%)

FA / B (%)

FA0

82

0

FA20

65.6

16.4

FA70

24.6

57.4

SF / B (%)

S/B (%)

Wo / B (%)

W/B (%)

SP / B (%)

DA / B (%)

18

35

13

13.8

2.2

0.02

Note: B: binder; LH Cement: Low Heat Cement; FA: fly ash; SF: silica fume; S: aggregate; Wo: wollastonite; W: water; SP: Superplasticizer; D: defoaming agent; Ol: steel micro-fibers; HDR: steel macro-fibers.

As illustrated in Fig.2, with respect to the unconfined NC cores (i.e., NO JACKET), the strength of the specimens confined with FA0 jackets increases of about 40% (see Table 3). In all the other cases, the higher the amount of cement in the UHP-FRCC mixtures, the higher the confinement effect on the NC core. Specifically, the jackets made with FA20 and FA70 mixtures, improve the strength of the plain core of only 31% and 12%, respectively. Thus, replacing cement with fly ash leads to a reduced compressive strength of the NC cores, even if both the ductility and the environmental benefits remain. Table 3: Improvement due to jacketing. Compressive Max stress / fccore strength increase [-] [%] FA0

1.37

36.85

FA20

1.31

31.05

FA70

1.12

11.89

Figure 2: Stress-strain curves referred to fc, core.

113

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Retrofitting a full-scale two-story hollow clay block masonry building by Steel Fiber Reinforced Mortar coating Sara Lucchini1, Luca Facconi1, Fausto Minelli1, Giovanni Plizzari1 1

: University of Brescia, Brescia, Italy.

Abstract The use of mortar coating reinforced only with randomly diffused steel fibers represents an effective technique for seismic retrofitting of existing masonry buildings. The present work aims at proving the effectiveness of that technique by testing a full-scale two-story hollow clay block masonry building subjected to a quasi-static cyclic lateral loading. The test prototype has been built by using a quite weak masonry material consisting of clay bricks with vertical holes having a percentage of voids of 62%. The structure was designed to represent the behavior of an existing building in which floors have a high in-plane stiffness and an effective connection with the bearing walls. Such an assumption aims both at preventing any out-of-plane mechanism and at promoting a global-type response of the building when subjected to earthquake actions. The masonry prototype building has a total length of 5750 mm, a width of 4250 mm and a maximum height of 6700 mm. A 1500 kN electromechanic jack fixed to a reaction wall was used to apply the lateral load to the test prototype. In order to ensure a load distribution proportional to the building masses, a distributor beam was designed so that the 60% of the total lateral load was applied to the first floor and the residual 40% to the second floor. A total of 28 potentiometric transducers were installed on the East and West façade of the building to monitor crack formation. Horizontal displacements were detected by LVDTs placed in correspondence of the floors and of the slab foundation. The quasi-static test was performed under displacement control, by progressively increasing the average top displacement of the building till the achievement of its maximum capacity. The first test performed on the building without coating is carried out to pre-damage masonry in order to simulate the effects of a seismic action significant for Ultimate Limit State conditions. The not-retrofitted building behavior was governed by the in-plane walls response, particularly by a shear failure of masonry piers. The first part of the response was linear until diagonal cracks started to grow in the central area of the piers of longitudinal walls at ground floor, causing a gradual loss of stiffness that led to the attainment of the maximum capacity, equal to 180 kN (Figure 1(a)). The building response presented a progressive reduction of the lateral stiffness in both loading directions, due to the formation of new shear cracks and cracks starting from the corner of the openings due to the rotation of piers. After having performed the first test for pre-damaging the building, a 25 mm thick Steel Fiber Reinforced Mortar (SFRM) coating will be applied on the external surface of the four façades to improve both the in-plane and the out-of-plane resistance of the walls. Coating-tomasonry connectors consisting of a nylon plug with steel screws are drilled into the masonry walls and provided with a 50x50x1.8 mm3 steel anchor plate placed within the coating thickness. Furthermore, conventional reinforcing bars are placed vertically along the base perimeter of the walls to connect coating to the RC foundation and bent rebars are installed to ensure connection between the RC chords to coating. Since the test on the retrofitted building is still ongoing, the related results will be described and discussed elsewhere. 114

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A series of 3D Non-Linear Finite Element Analyses (NLFEAs) were carried out to predict the response of the full-scale building before and after retrofitting, by means of the Finite Element (FE) program DIANA 10.1. The simulations include one analysis on a not-retrofitted building (MB) and two analyses on buildings (MB-R and MB-R+SR) retrofitted with coating. Unlike MB-R, MB-R+SR was provided with steel reinforcing bars to connect coating to the RC foundation. The total base shear-lateral deflection response of Figure 1(a) highlights the strength and stiffness improvement provided by the proposed technique. The steel rebars used to prevent rocking mechanism of the masonry walls, allowed to effectively exploit the tensile toughness of SFRM. Therefore, the lateral capacity of MB-R+SR is approximately two times higher than the one obtained from the building without coating-to-foundation connections (MBR). Figure 1(b) shows the comparison between the failure patterns of the East façade before (b1) and after (b2) retrofitting with coating connected to the foundation. The not-retrofitted building showed a diagonal shear failure, whereas MB-R+SR behavior is mainly governed by flexure without any significant shear mechanism. Figure 1(a) reports also the comparison between experimental and numerical lateral load-displacement curves related to not-retrofitted building. The numerical initial stiffness was in good agreement with the experimental one, while the peak load is 18% underestimated. Finally, the failure mechanism of the not-retrofitted building was properly predicted by the FE model. East (+)

800

V [kN]

East (-)

600

400

200

δ [mm]

(b1): MB

0

-20

-15

-10

-5

0

5

10

15

20

East (+)

East (-)

-200

-400 MB, num -600

MB-R, num

MB-R+SR, num MB, exp -800

(a)

Figure 1:

(b2): MB-R+SR

Base shear (V) - lateral displacement () curves: comparison between numerical and experimental response (a); contours of principal tensile post-cracking strains of East façades (b1, b2).

Keywords Steel Fiber Reinforced Mortar; Mortar coating; Retrofitting; Hollow block masonry; Cyclic test; Full-scale masonry building

115

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Working Life Extension of RC Bridge Pier Through UHPFRC Jacketing Adriano Reggia1, Alessandro Morbi2, Giovanni A. Plizzari1 1

: DICATAM, University of Brescia, Brescia, Italy. 2 : Global Product Innovation, Heidelberg Cement Group, Bergamo, Italy.

Abstract Prior the introduction of modern codes, the design working life of reinforced concrete bridges was never explicitly prescribed. After more than 50 years from the construction of large reinforced concrete infrastructures in many western countries, their condition is now raising the problem of the evaluation of the remaining working life. In this long period, reinforced concrete structures may have experienced structural damages (due to lower-than-design earthquakes or exceptional actions) and materials’ degradation (due to harsh environmental conditions or to the effects of human actions) so large to require a major repair. In this paper, an innovative strengthening technique involving ultra-high performance fiber reinforced concrete (UHPFRC) and another based on traditional reinforced concrete are compared as possible solutions for seismic retrofitting of a bridge pier. The resulting extended working life is discussed considering the structural response at the ultimate limit state, by the evaluation of the lateral load bearing capacity of the structures. In the end, the results of a 1:4 scaled experimental test on the bridge pier are briefly presented as a proof of concept of the UHPFRC retrofitting technique. The case study is a 23,32 m tall highway bridge pier supporting three 50,00 m single-span box girders. It has a multicellular cross-section with external dimensions of 600×250 cm and two internal cells of 210×130 cm. At the base, it is reinforced by 314 Ø26 longitudinal bars and Ø16 transverse stirrups spaced 100 mm with a concrete cover of 40 mm. Two seismic retrofitting techniques are considered in the study: an UHPFRC jacketing with a thickness equal to 12 cm and a RC jacketing with a thickness of 30 cm. Additional reinforcement is provided in both cases to restore the tensile capacity of the jacketing at the base section. The main properties of the elements are listed in Table 1. The period of the un-strengthened pier is 1,95 s, while those of the UHPFRC and RC strengthened piers are 1,55 s (-20%) and 1,46 s (-25%) respectively. The maximum seismic acceleration sustainable by the un-strengthened pier is 0,231·g, while those of UHPFRC and RC strengthened piers are both 0,324·g (+40%). The working life of each element has been calculated by means of the adoption of performance limit demand spectra for which the maximum acceleration is reached for the different periods of the structures, as shown in Figure 1. The estimated working life (in a high seismicity zone in Italy) of the un-strengthened pier is 67 years, while those of UHPFRC and RC strengthened piers are 107 years (+60%) and 81 years (+22%) respectively. In order to validate the UHPFRC jacketing as a reliable retrofitting technique, the results of the test performed on the scaled laboratory specimen (scale 1:4) of the highway bridge pier are presented. The specimen was tested under static, uni-directional, cyclic loading, as shown in Figure 2. The specimen was cyclically loaded in the horizontal direction by means of an electromechanic actuator with a capacity of 1500 kN under a constant vertical load of 1000 kN. Seismic response of the strengthened specimen was enhanced both in the maximum load and 116

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

in the maximum displacement reached at the end of the test with respect to the un-strengthened specimen calculated analytically (see Figure 2). The ultimate load increased up to +80% and the ultimate displacement increased up to +57%. The load determining first cracking and the first yielding of steel reinforcements increased up to +42% and +90% respectively. Table 1: Element Designation Unstrengthened UHPFRC RC

Specimen designation and main properties considered in the analytical study. Axial force [kN] 22˙985

Bending moment [kNm] 105˙567

Effective stiffness [kNm2] 87˙235˙072

Perio d [s] 1,95

Maximum seismic acceleration [m/s2] 0,231·g

Working life [years]

24˙101 25˙891

151˙924 158˙760

141˙545˙489 168˙720˙490

1,55 1,46

0,324·g 0,324·g

107 81

67

Figure 1:

Fictitious location of the structure and performance limit demand spectra.

Figure 2:

Test setup (measures in centimeters) and experimental response of the UHPFRC strengthened laboratory specimen (envelope curves).

117

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Improved Ductility Of SHCC Retrofitted Unreinforced Load Bearing Masonry Seismic Resistance Dirk J.A. de Jager 1, Gideon P.A.G. van Zijl 1 1

: Civil Engineering, Stellenbosch University, South Africa

Abstract

Introduction Retrofitting of unreinforced load bearing masonry (ULM) in seismic regions has been proposed by the research group in Stellenbosch, South Africa. This was motivated by the significant stock of three and four storey ULM buildings in the region that were constructed before seismic requirements were included in the South African National standard. Sprayed application of thin layers of strain-hardening cement-based composite (SHCC) was developed. Characterising triplet tests were developed for the interfacial bond between the masonry and SHCC, and the multiple-cracking shear response of the SHCC. The characterised behaviour enabled the design of the overlay thickness, and validating tests on walls subjected to in-plane shear. Here, nonlinear finite element (FE) analysis of the overlay strategy is reported. The calibrated data for the masonry and SHCC overlay are used for double-leaf masonry walls, retrofitted with 15 mm SHCC overlays and subjected to pull-over tests while maintaining vertical restraint simulating upper storeys. Eight walls were tested in the structures laboratory in Stellenbosch, varying wall thickness (single leaf, double leaf masonry) and overlay thickness (15 mm, 30 mm). The test results enable validation of the computational modelling strategy and characterised model parameters. Masonry is modelled as an anisotropic continuum with multisurface plasticity-based limit functions, allowing compressive hardening and subsequent softening, as well as tensile softening of the masonry. The SHCC overlay is modelled with a Rankine-Rankine limit function, with both compressive and tensile strain-hardening and subsequent softening. Plane interfaces are used to model the interface between the masonry and the overlay, and allow Coulomb-friction shearing interaction and tensile de-bonding. The validated computational model is used to extrapolate the physical laboratory shear wall test boundary conditions, in order to evaluate alternate retrofitting strategies to improve the ductility in seismic resistance.

Retrofitting towards improved ductility of shearing response Enhanced shear resistance is obtained when a shear wall is retrofitted with a 15 mm SHCC overlay. However, an improved ductility performance was yet to be observed. Analyses were performed on the surface roughness of the bonding interface of SHCC. When only considering a mechanical load with SHCC applied to concrete beams with different surface profiles, a smooth surface led to an increased crack distribution over a larger debonding area. The enlarged crack distribution holds advantages for the ductility of the structure as the finer additional cracks provides high shear resistance at higher deformations. Cracks distributed in a smaller area in the case of a rough surface, leading to more localised cracks in the concrete beam. Brittle

118

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

localised failure is more likely to occur under these conditions. However, the effects of drying shrinkage were not considered in the previously discussed surface roughness analyses. Uncontrolled debonding of the SHCC overlay due to drying shrinkage is prone to occur in cases where the bonding strength of the surface profile is too low. Thus, a balance must be obtained in the interface bond for maximum shear resistance and ductility. Here, the influence of diagonal debonded strips is investigated computationally, after calibrating the finite element model against laboratory experiments on retrofitted walls with fully bonded SHCC overlays. Figure 1 shows the increased ductility computed for a retrofitted wall with three unbonded, diagonal strips, each 75 mm wide, and spaced 150 mm centre to centre diagonally. (a) 300

Shear force (kN)

250

(b)

200 150 100

SW220-15 SW220-15, 75 mm Strip

50 0 0

(c) Figure 1:

5

10

15

20

Displacement (mm)

Shear displacement response for SHCC retrofitted masonry walls, showing maximum principal strain (E1) in the SHCC overlay at (a) 5.8 mm displacement for the case without debonding strips, (b) 9.6 mm with debonding strips, and (c) the shear force-displacement comparison.

Conclusions The computational model was calibrated by validation against experimental test results on bare masonry wall shearing response, as well as masonry walls retrofitted with 15 mm thick, fully bonded SHCC overlays. Subsequent computational analysis of a similarly retrofitted masonry wall, but with three 75 mm wide unbonded diagonal strips resulted in a significant increase in the diagonal cracked region in the overlay, leading to significantly enhanced shear ductility.

119

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

120

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Precast elements

121

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

New Concretes for Precast Panels in Uruguay Gemma Rodriguez de Sensale1, Iliana Rodriguez Viacava1, Rosana Rolfi1, María Esther Fernández1 1

: University of the Republic (UdelaR), Montevideo, Uruguay.

Abstract International experience in precast has shown advantages of use new concretes instead of conventional concretes. In Uruguay, the largest experiences in precast products date back to the 1960s and are based on heavy precast systems, whose performance has been variable. The material used in all cases is conventional concrete (C), which determines technical characteristics of the final products, manufacturing and assembly. The versatility of the applications of fiber reinforced concretes (FRC) and self-compacting concretes(SCC), make these special concretes an alternative of maximum interest for our country. In this paper the results obtained at the Project “Application of new concrete for precast” corresponding to experimental stage in prototypes are presented. Considering uruguayan typical precast panel, a rectangular panel of conventional concrete (P1) was taken as reference. The variables analyzed were the reinforcement used (conventional, and replacement of metal mesh by metal and synthetic fibers) and the type of concrete (C and SCC). Therefore, real-scale panels made of fiber reinforced concrete (FRC), self-compacting concrete (SCC) without fibers and with fibers (FRSCC) were studied, which were compared with the typical precast panel with C. In a Uruguayan precast company the different mix concretes where made in a mixer of 1m3 capacity, different series of real-scale panels and test specimens of the concretes were made. The six series of precast panels tested in the present research were rectangular with dimensions of 2400 (length) × 1200 (width) × 200 (thickness) mm. All panels were made with expanded polystyrene inner core 100mm thick (900mm wide x 2100mm high), a frame of 150mm wide around its entire contour and two outer layers of 50mm thick each of C, SCC, FRC or FRSCC according to the panel type; this frame include a triangular steel reinforcement formed by three longitudinal bars of 8 mm diameter joined together by a cross-link of 4.2mm diameter. Panels with C and SCC, named P1 and P2 respectively, have the steel reinforcement typically used at uruguayan panels, with an upper and lower electrowelded steel mesh (of 4.2mm diameter separated every 15cm in both directions) included in both outer layers; in the remaining panels the mesh was eliminated using structural fibers: 20 kg/m3 of metallic fibers (MFP1 and MFP2) or 6 kg/m3 of synthetic fibers (SFP1 and SFP2). After 12 hours of execution of each series of panels, they were subjected to steam cure during 8 hours. Once reached 24 hours of panel execution, they were exposed to environmental cure after wetting their surfaces to avoid shrinkage cracks by drying. After 28 days, the panels and specimens of each type of concrete were transported from the factory to the Laboratory of the Institute of Building Construction to be tested. The structural strength of the panels was evaluated using the ASTM E72-15 standard. The test was carried out in an instrumented loading portic. Each panel was placed in a horizontal position, simply supported, with light between supports of 2.10m. In the test, two uniformly distributed linear loads located one quarter of the light between supports were applied. In this way, the central section is submitted to pure bending moment. Deflections were measured throughout test.

122

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

In fresh state, the incorporation of fibers has a significant influence on the results obtained; for SCC with and without fibers the slump diameter are higher than 60 cm and dj ≥ d-5 cm, fulfilling the requirements of Schedule 17 EHE-08. In hardened state, the use of fibers in different concretes carried out influenced the properties related to mechanical resistance; moderately increases the compressive strength and improves the results of resistance to bending in relation to concretes without fibers. Table 1 shows the average values of the maximum load (Fmax), maximum instantaneous deflection (Definst), maximum deflection at 5 minutes of application of the maximum load (Def5min) and the residual deflection (Defresidual) reached for each series of 3 panels tested following the ASTM E72-15 standard. Variations of these values (Def) are also presented in relation to the P1 panels that are commercialized by the company in our country. Table1:

Panel test results following ASTM E72-15.

Panels

Fmáx (kN)

P1

105.86

MFP1 SFP1

Def 5min (mm)

Def residual(mm)

10.19

4.87

2.95

90

2.19

0.62

0.16

80

2.48

0.87

0.23

P2

127.10

9.0

4.3

2.50

MFP2

97.25

3.9

2.36

1.43

SFP2

86.10

4.5

2.50

1.90

Def 5min (%)

Def resid.(%)

Fmáx (%)

Def inst.(mm)

Def inst.(%)

MFP1

-14,98

-78,51

-87,27

-94.58

SFP1

-24.43

-75.66

-82,13

-92.20

P2

+20.06

-11.68

-11.70

-15.25

MFP2

-8.13

-61.73

-51.54

-51.52

SFP2

-18.66

-55.83

-48.66

-35.60

At the studied panels, it was demonstrated that it is possible to replace the upper and lower reinforcement mesh for fibers; better results are obtained with metallic fibers than with synthetic fibers. The use of SCC improves the behavior of the panels subjected to ASTM E7215 test, obtaining higher final load and lower deflections than with the use of conventional concrete commonly used in them. The results obtained show that the SCC, FRC and FRSCC are an alternative of great interest for their application in precast in our country.

Keywords Fiber reinforced concrete; Self-compacting concrete; Structural fibers; Precast panels

123

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Opportunities for synthetic fibre reinforcement in concrete tramlines Karoly Peter Juhasz1, Peter Schaul2, 1

: Budapest University of Technology and Economics, Department of Mechanics, Materials & Structures, Műegyetem rakpart 1-3, Budapest, Hungary 2 : Budapest University of Technology and Economics, Department of Construction Materials and Technologies, Műegyetem rakpart 1-3, Budapest, Hungary Today the construction of modern concrete slab track plays a prominent role in the construction industry. Besides keeping in mind having an economic solution, more emphasis is placed on its durability and its resistance to environmental factors such as moisture, de-icing salts etc. The economic solution can be achieved primarily by decreasing the thickness of the slab and shortening the construction time. Durability can be significantly increased by designing for fatigue and by using materials that are resistant to these environmental factors. Because of this macro synthetic fibres are being used more often for the reinforcement of the concrete for both cast in place or precast structures. Corrosion resistance is the greatest benefit of macro synthetic fibre where durability can be assured but also synthetic fibres behave better with dynamic loads than steel fibres, therefore their use for tramline or railway track slab is very favourable. Added to this are the economic advantages such as a reduction in labour who would traditionally set and tie the steel reinforcement into place. The first macro synthetic fibre reinforced track slab was constructed in Japan in 2002: Elasto Ballast track railway. The goal of using macro synthetic fibre was, beside from the reduction of the vibration and noise, to increase the speed of the construction process. The first synthetic fibre reinforced track slab in Europe was the Docklands Light Railway near London in 2004. The first only macro synthetic fibre concrete cast in place tramline in Europe was built in Szeged, Hungary. Macro synthetic fibre was considered over steel reinforcement as at a certain point in the track it was possible to use any steel reinforcement due to the operation of a special switch that collected stray current and so this part of the track was reinforced with macro synthetic fibre reinforced concrete. Because of the positive experiences and cost saving in this part of the track the contractor changed to this solution for the entire track and thus replaced all the steel reinforcement with macro synthetic fibres. The Szeged tramway project was a huge success. After the system proved to be fully functional several other tram tracks were constructed using very similar solutions and using macro synthetic fibres. These tram tracks were constructed in St. Petersburg, Russia, and in Tallinn, Estonia. Beside the cast in place solution the use of precast concrete tramline elements started to spread, mainly because of the same benefits of shortening construction time. These elements also needed to be designed for temporary situations, such as demoulding, lifting, transporting and placing on site. The first and only macro synthetic fibre reinforced concrete slab track to date is the PreCast Advanced Truck (PCAT) system. This precast element is highly optimised both by the dosage of the fibre and its geometrical shape. The numerical modelling of the PCAT slabs were done with ATENA FEA software. 124

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

PCAT calculation with ATENA.

Real scale model was carried out to determine the deflection of the slab and to be able to verify the finite element results to real scale test. The results in every loading case were close to each other. The finite element analysis closely mirrored what happened in reality and the differences between the measured deflections in the model and in the test was less than 0.1 mm.

Figure 2:

Results of RTST and FEA.

Design and optimizing of the track slab means the determining of the required thickness of the track slab and the macro synthetic fibre dosage for the varying load cases, such as ultimate and serviceability limit state, and fatigue. For these special tasks advanced finite element software is needed. The influence of the fibres could be taken into account by the modification of the fracture energy of the plain concrete.

125

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Optimized reinforcement and performance of precast elements using polypropylene macro fibers Àlex Calvo Costa1, Jose Maria Vaquero1, Carles Cots Corominas1; Jürgen Bokern2, Abel Medel Ramírez3 1 : BASF Construction Chemicals Spain, L’Hospitalet, Spain 2 : BASF Construction Solutions GmbH, Mannheim, Germany 3 : Uniblok Velatia, Seseña

Abstract There is a growing interest in the precast industry in the application of polypropylene (PP) fibers in combination with conventional reinforcement (hybrid) or as full substitution. In this paper two examples are discussed for the use of PP macro fibers. In the first case, a precast panel with a 3 x 6 m² surface area and a thickness of 15 cm was produced considering synthetic fiber reinforced self-compacting concrete (SFRSCC). Characteristic concrete compressive strength was 40 MPa (fck,cube). The original reinforcement foreseen consisted of perimeter lattice girders 8ϕ8 mm (As=402 mm2) and a standard mesh 200x200ø5mm (As=98 mm2). Since an automatic table was used to lift the panel during demolding phase and panels were transported and installed in vertical position, the design stresses in the panel were calculated considering exclusively wind loads (1,5 kN/m² wind load; Md=6.9 KNm/m, assuming the panel was part of a building located in wind zone C, grade III, according to the Spanish Standard SEAE).

a) Figure 1:

b)

Lifting tables: (a) original steel reinforcement of the panel (b) remaining steel reinforcement considering PP fiber reinforced concrete for the panel.

Anticipating pre-determined post crack stresses (fR1,k= 1,06 MPa; fR3,k=1,40 MPa), it could be proven that a combination of polypropylene fibers (4 kg/m3) and the conventional lattice in the perimeter allow to get the required structural capacity for the design bending moment. In addition, according to the Spanish Code EH-08, the design had to fulfill the ductility requirements as a guarantee to prevent fragile fracture of the concrete (As f yd + 0,4 Ac fctR,d ≥ 0,04 Ac fcd).

126

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Compared to the initial solution, 45 % of the previous steel reinforcement (fig. 1) was saved. Moreover, reinforcement related CO2 emissions were reduced by 25 %. Finally, durability of the elements was improved, because cracks showed lower crack width and led to lower permeable concrete. In the second case the target was to find a more efficient design for the different elements that constitute a precast electrical equipment shelter. Again, FRSCC had to be considered. In this paper we focus on the wall panels. If standard meshes are to be used in production the actual reinforcement degree is often significantly higher than what is needed based on the design calculation. Thus, the objective of the study was, to partially replace the steel reinforcement by PP macro fibers in a way that utilization of available standard meshes is maximized. The residual tensile performance of FRSCC and MC 2010 design rules provide the tools. Three different FRSCC according to Spanish Codes varying in fiber dosage as indicated below were considered in the study: 4.5kg/m3 - HAF-40/P-1.12-1.07/AC/12-4.8/IIa 6.0kg/m3 - HAF-40/P-1.47-1.59/AC/12-4.8/IIa 8.0kg/m3 - HAF-40/P-2.05-2.55/AC/12-4.8/IIa Results of the analysis are compiled in the following table. Table 1:

Concrete strength class

C40 C40 C40 C40

Steel reinforcement in dependence of selected synthetic fiber reinforced concrete.

Fiber (kg/m3)

Required Standard steel Utilization steel mesh degree mesh section Material Dosage Asr(mm2) ϕ n Asd(mm2) % polypropylene 436 10 8 628 69 polypropylene 4,5 401 8 8 402 100 polypropylene 6,0 393 8 8 402 94 polypropylene 8,0 369 8 8 402 92

It shows that by using PP macro fibers the theoretically required reinforcement of 436 mm² could be reduced by between 8% and 15% at fiber dosages between 4,5 kg/m³ and 8,0 kg/m³, respectively. Comparison of the mesh reinforcements put in place indicates, that actually a reduction by 36 % from 628 mm² to 402 mm² could be achieved. Moreover, the utilization of the actually placed steel was optimized in this way. From originally 69 % it went up to > 90 %. At maximum 100 % utilization could be achieved at a fiber dosage of 4,5 kg/m³. Additional benefits that have been realized are an increased impact resistance and a significant reduction in crack width, both contributing normally to substantial finishing activities before delivery of precast elements to the end-user.

Keywords Fiber reinforced concrete; Polypropylene fibers; Precast panel; Precast electrical equipment shelter, CO2 emissions, sustainability.

127

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Closed loop control in crushing test for fibre reinforced concrete pipes Renata Monte1, Albert de la Fuente Antequera2, Antonio Domingues de Figueiredo1 1

: University of São Paulo. São Paulo, Brazil. 2 : Barcelona Tech (UPC). Barcelona, Spain

Abstract The mechanical behaviour of fibre reinforced concrete pipes (FRCP) is usually verified through three-edge bearing test (TEBT). More recent studies propose a design approach according to the fib Model Code 2010, involving the evaluation of mechanical behaviour in service (SLS) and ultimate (ULS) limit states. The challenge consists in assessing with the proper accuracy and reliability the mechanical parameters by means of the TEBT method. It is generally performed using a hydraulic equipment with load control and without using devices to measure the diametric displacement. In this research program, TEBT was carried out using three different control systems. Two closed-loop controls, using pipe diametric displacement or actuator displacement, and an open-loop control by loading rate. The crack width development was also measured simultaneously with the diametric displacement, only for closed-loop control. An increase in the post-crack strength of the FRCP were observed when the closed-loop system was used. The results show significant difference between the mechanical performances of pipes tested in open or closed-loop control. Closed-loop tests performed with actuator displacement control or pipe diametric displacement control show no significant differences. Crack opening measurement allows establishing a linear correlation with diametric displacement, making possible to establish the values corresponding to the SLS and ULS deformation and respective loads. Consequently, the connection with the Model Code approach become possible and likewise the design optimization of the FRCP.

Keywords Fibre reinforced concrete. Concrete pipes. Crushing test. Closed-loop control

Research program The experimental campaign involved the real scale production of pipes with internal diameter Di = 1000 mm, thickness h = 90 mm and length l = 1190 mm, manufactured with steel fibres of the type DRAMIX® RC80/60BN in contents of 10, 20 and 30 kg/m³. The pipes were tested in TEBT using closed control system, with feedback control from actuator or diametric displacements, and compared to open-loop tests from literature. The Model for the Analysis of Pipes (MAP) is used to simulate the global response of the pipe during the TEBT and the tensile behaviour of FRC was simulate using parameters derived from the Barcelona test.

128

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Results and discussion

(a) Figure 1:

(b) (c) F–v average experimental and numerical curves for the SFRCPs with: 10 kg/m³ (a); 20 kg/m³ (b); 30-35 kg/m³ (c).

Table 1:

Comparison between experimental and numerical results for different load levels.

10 kg/m³ Exp MAP [kN/m] [kN/m]

 [%]

20 kg/m³ Exp MAP [kN/m] [kN/m]

 [%]

30 kg/m³ Exp MAP [kN/m] [kN/m]

 [%]

Fcr

39.4

52.2

-33

47.5

50.5

-6

46.4

47.1

-1

F2mm

65.3

52.3

20

71.3

56.4

21

76.5

52.7

31

F6mm

45.4

38.8

14

59.1

47.6

19

80.4

45.2

44

F10mm

31.4

27.0

14

49.1

35.3

28

70.8

33.4

53

The control provided by the actuator or pipe displacement can be considered equivalents. But, when tests are executed with a load control (LC) the results are significantly lower. The MAP is capable to reproduce, from the safe side, the post-cracking tendency. Nonetheless, the numerical simulation for the pipe with 30 kg/m3 should be reconsidered taking into account the differences obtained (which can lead to unnecessary conservative designs). This lower performance can be directly attributed to the dispersion of the fibre amount measured in BCN specimens.

129

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

RC Beams with Steel Fibres - Towards Better Determination of their Minimum Conventional Reinforcement Ratio Avraham N. Dancygier1, Yuri S. Karinski1 1

Technion – Israel Institute of Technology, Haifa, Israel

Abstract Minimum longitudinal reinforcement ratio in reinforced concrete (RC) beams is determined by criteria that refer to flexural capacity. However, relatively recent experiments have shown that RC beams with steel fibres (RFRC beams) with the above-determined minimum reinforcement, exhibit a pronounced reduction in their flexural ductility, compared with similar plain RC beams. This phenomenon was characterized by cracking localization, when only one or two cracks widened significantly more than the other cracks did. This led to increase of the steel plastic strains, even up to its rupture, in the rebars that crossed these cracks. The ductility reduction in this case impairs the structural response of RFRC beams compared with RC beams without fibres. Therefore, this effect of the fibres must also be considered when determining the minimum longitudinal reinforcement ratio in RFRC beams. The paper overviews results from two experimental programs of high and normal strength RFRC beams. The programs included four-point bending tests of plain (control) and fibrereinforced beam specimens, where the latter included two amounts of hooked-end steel fibres and conventional reinforcement ratios,  that varied from 0.15% to ~3.5%. These results show that the ductility ratios of all specimens in these experiments with low reinforcement ratios were significantly smaller than those of similar plain RC beams. When the reinforcement ratio increased, the flexural ductility of the RFRC beams increased as well and for a larger reinforcement ratio, the ductility ratios of the RFRC beams were higher than those of the control RC specimens. Hence, to guarantee sufficient ductility of RFRC beams, their minimum reinforcement ratio should be increased. In addition, processing of the experimental data shows that as the reinforcement ratio decreases below a certain value (~0.6% for normal strength concrete), which is considerably larger than the minimum required by the codes, the structural ductility also decreases. For larger amounts of conventional reinforcement, the dependence of ductility on the reinforcement ratio in RFRC and plain RC beams is similar. As stated above, this flexural ductility reduction is caused by cracking localization, which is characterized by significant widening of one or only few cracks, compared with the other ones. The level of cracking localization can be quantified by the ratio m/n, where m is the number of significantly wide cracks and n is the total number of cracks. Smaller m/n ratios indicate higher localization levels and vice versa. Processing of the experimental data included counting of the total number of cracks, n, within the constant moment zone of the beam specimens, up to yielding of the conventional reinforcement and the number of wide cracks, m, at the end of the test. In specimens with low values of  only a single wide crack was observed. When the reinforcement ratio increased, the number of wide cracks increased as well, where for the largest value of  all cracks widened almost uniformly (which is typical of plain RC beams). Moreover,

130

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

it is shown that the ratio m/n is more affected by  at its lower range and that lower level of localization is observed at lower amounts of fibres, which conforms to the diminishing of the phenomenon at plain RC beams. Considering the dependencies of both structural ductility and cracking localization level (m/n) on  in RFRC beam, shows that within a range of small reinforcement ratios, these dependences are similar. This means that there is an influence of the cracking localization on the flexural ductility ratio. For larger reinforcement ratios, the localization level diminishes and its influence on ductility no longer exists. These dependencies are combined in Figure 1 to show the relation between the ductility ratio and the ratio m/n, where the ascending branch indicates the range within which m/n affects ductility while within the descending branch there is no localization effect.

Figure 1:

Relation between flexural ductility and m/n.

This finding indicates that the minimum reinforcement ratio in RFRC beams, discussed above, may be expressed in terms of cracking localization level. This level may be obtained theoretically, based on the fibres non-uniform distribution along the beam (such model for tensile RFRC elements is presented elsewhere).

Keywords Cracking localization; Fibre-reinforced concrete; Flexural ductility; Minimum reinforcement ratio

131

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

132

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Tunnel linings

133

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Santoña–Laredo General Interceptor Collector – Challenges and Solutions Ralf Winterberg1, Rolando Justa Cámara2, David Sualdea Abad2 1

Elasto Plastic Concrete, Singapore. 2 Acciona Infraestructuras SA, Madrid, Spain.

Abstract The Santoña–Laredo General Interceptor Collector is a 1.5 km subsea tunnel under construction in Northern Spain. The tunnel is part of the Santoña Marshlands Sanitation Project. It is constructed with a dia. 4.30 m Mixshield TBM across the Santoña bay area using fibre reinforced concrete segments. The tunnel excavation starts in a deep shaft and passes through two different and consistent geological units withstanding pressures of up to four bars. Firstly, a karstified limestone formation in the first section and silty sands of fluvial, lacustrine or marine depositions with pockets of muddy silt in the rest of the alignment. This tunnel lining with exposure to corrosive media, as a subsea tunnel, has high requirements on robustness and durability. The project presents numerous technical challenges: • Umbilical assembly of the TBM at the bottom of a deep shaft in order to avoid excavation of an assembly cavern; •

Space constraints in the bottom of the 40 m deep shaft and on the surface due to the worksite location in an urban area;

•

Expected pressure up to 4 bars and permanent ground water pumping in order to minimize ground pressure against the shaft walls;

•

Employment of macro synthetic fibre as primary reinforcement of the segments due to marine environment and associated corrosion issues;

•

Design of the fibre reinforced concrete segments.

View into Santoña launch shaft (left) and storage of the rings at the precast plant (right).

This paper addresses the solutions to the technical challenges of the project, the design of the macro synthetic fibre reinforced segments and the benefits in production and construction associated with the replacement of conventional rebar cages. 134

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Due to the location of the Santoña shaft, the lack of available space on the surface for installations and stockyard is one of the constraints of the project. The design of the Santoña Shaft requires the ground water level to be maintained under the -17 m elevation. Considering the location of the shaft, at a distance of 50 m from the sea coast, a drainage system through the shaft floor slab with a ground water pumping system are working continuously. Geotechnical research conducted on the job site indicated the possibility of finding mixed terrain in the excavation face on a long section, because the project kept the alignment in the area of contact between the bedrock and the fluvial, lacustrine or marine deposits. Although the choice of the Mixshield TBM was considered to resolve the scenario of excavation fronts with mixed ground, to reduce associated geotechnical risks, a modification of the layout was suggested, avoiding this contact surface as a preventive measure. Thus the problem of finding mixed fronts rock-soil in an important length is avoided or reduced. Another constraint was that an assembly gallery shall not be excavated. Considering that requirement, the Mixshield had to be assembled in an umbilical mode until the length of the excavated tunnel enables the assembly of all gantries in their normal position. Thus, a special umbilical assembly process has been designed, using the space in both shaft levels to place the essential parts of the back-up in order to excavate in this umbilical mode. During all bore and assembly stages, the electrical cables and water, air, bentonite and slurry hoses were disposed over the shaft and tunnel in a temporary condition and have been extended in a manual way. One of the most difficult operations during all the stages until the end of the assembly was the manipulation and extension of the provisional slurry feed and extraction hoses, used to connect the pipes of the tunnel and the TBM. Lack of space inside the shaft has been a problem during all assembly stages, as well as during the normal operation of the machine. The design of the auxiliary installations inside the shaft needed to meet the requirements of these limited dimensions. Also, a dewatering system at groundwater level behind the shaft walls had to be effective during all stages. Due to humidity and salinity of groundwater inflow, dealing with TBM assembly operations led to more difficulties with the tunnel progress. The initial segment design yielded a conventional steel reinforcement cage of 95 kg/m3. A review of the design and the manufacture of the tunnel segments to meet the aggressive environment led to the decision to replace conventional rebar cages with EPC’s BarChip Macro Synthetic Fibre. A cost assessment including segment manufacture and reduced repair or reject rate, due to significantly improved robustness of the segments, revealed a total cost saving of nearly 40%, compared to the traditional rebar cage design. Aside from the direct cost advantages, the switch to the macro synthetic fibre eliminated the rebar cage and its inherent labour and reduced production cycle times by nearly 50%. The experience shows us that launching TBMs in shafts with lack of space is a possible practice, which requires reduced advance ratios, but provides a reduction of shaft structure costs. A reduction in the launching shaft bill of quantities can be reached using a complex vertical assembly of the TBM, but it can be a high skill manpower demanding process.

135

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Segmental Lining Design using Macro Synthetic Fibre Reinforcement Ralf Winterberg1, Luis Mey Rodríguez2, Rolando Justa Cámara3, David Sualdea Abad3 1

Elasto Plastic Concrete, Singapore. 2 Ingemey Consultores, Madrid, Spain. 3 Acciona Infraestructuras SA, Madrid, Spain.

Abstract Fibre reinforced concrete (FRC) is becoming widely utilized in segmental linings due to the improved mechanical performance, robustness and durability of the segments. Further, significant cost savings can be achieved in segment production and by reduced repair rates during temporary loading conditions. Structural fibres can replace or reduce ordinary rebar cages, acting as the primary structural reinforcement. Significant cost savings are often achieved by the use of fibres; mainly by the partial or entire replacement of ordinary reinforcement in the production, but also by improving the robustness, serviceability and durability and hence, reducing maintenance costs. Segments reinforced with traditional steel rebar cages require tight serviceability limits in order to protect the steel reinforcement from corrosion. Crack width control given by the fibre reinforcement can change a crack control governed design (SLS) into a pure structurally required design. This not only yields economic benefits but creates significant freedom in design and detailing. FRC without conventional reinforcement is able to limit the developing crack width where load-redistribution is possible as given in statically indeterminate elements (rotation capacity), such as tunnel segments under axial thrust. Macro synthetic fibres (MSF) are non-corrosive and thus ideal for segmental linings in critical environments. Although fibre reinforcement for segments is relatively new, recent publications such as the ITAtech report No. 7 or the British PAS 8810 have now given more credibility to this reinforcement type and the basis for design.

Large tunnels or tunnels in soft grounds might still require traditional steel reinforcement due to high bending moments in temporary or permanent condition. In these cases hybrid reinforcement, i.e. a combination of fibre and steel rebar cage, is often the most durable and cost-efficient solution due to the partial replacement of rebar by fibres.

136

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Recent research has shown that the addition of high performance macro synthetic fibres to steel rebar reinforced concrete reduces crack width and crack spacing by 30% in bending of simply supported beams. Thus, MSF can add a significant gain of design life to steel reinforced concrete structures, such as hybrid reinforced segments. This paper presents and discusses the design methodology for precast tunnel segments and in particular the tasks associated with the use of MSF reinforcement. Temporary loadings as well as long term load behaviour will be addressed. A case history from the Santoña–Laredo General Interceptor Collector, currently under construction in northern Spain, will illustrate the specific benefits of MSF reinforcement for segmental linings. The Santoña–Laredo General Interceptor Collector is a 1.5 km subsea tunnel currently under construction in northern Spain. The tunnel is part of the Santoña Marshlands Sanitation Project. It is constructed with a 4.30 m Mixshield TBM across the Santoña bay using macro synthetic fibre reinforced concrete segments. The review of the manufacturing process and especially of the design of the tunnel segments led to the decision to replace conventional rebar cages with macro synthetic fibre to cope with the aggressive marine environment. The initial segment design yielded a conventional steel reinforcement cage of 95 kg/m3. Regarding improvements in the precast operations and the related cost savings the main contractor Acciona Infraestructuras reviewed this design with Ingemey Consultores, the final design consultant. Switching to EPC’s BarChip fibre reinforcement eliminated more than 80% of the steel reinforcement. The remaining bursting ladders (16 kg/m3) are solely for jacking forces where the synthetic fibre is the primary segment reinforcement. Aside from the direct cost advantages, the switch to macro synthetic fibre eliminated the rebar cage and its inherent labour and reduced production cycle times by nearly 50%. A cost assessment, including segment manufacture and reduced repair or reject rate due to significantly improved robustness of the FRC segments, revealed a total cost saving of nearly 40% for the rings, compared to the traditional rebar cage design. The experience gained in the Santoña-Laredo project shows that macro synthetic fibre reinforced segments perform very robust and satisfactory even under difficult conditions. The successful completion of this project will build further confidence in macro synthetic fibre reinforced segmental linings. The success and gained experience of this project will surely lead to the implementation of this technology in other tunnel projects. These types of utility tunnelling projects (e.g. sewage, power cables, irrigation, gas transfer or hydropower) are widely existing in the world market and present a huge opportunity for macro synthetic fibre reinforced concrete segmental linings, benefiting from the given advantages.

Keywords Macro synthetic fibre, fibre reinforced concrete, segmental lining, TBM, segment design

137

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

An experimental study on the use of polypropylene fibers in precast segments for hydraulic and metro tunnel lining Antonio Conforti1, Ivan Trabucchi1, Giuseppe Tiberti1, Giovanni Plizzari1, Angelo Caratelli2, Alberto Meda2, Sandro Moro3, Martin Hunger4 1

: University of Brescia, Brescia, Italy. : University of Rome “Tor Vergata”, Rome, Italy. 3 : BASF Construction Chemicals Italia, Treviso, Italy. 4 : BASF Construction Solutions GmbH, Trostberg, Germany. 2

Abstract There is a growing interest in the scientific community on the structural applicability of polypropylene fibers (in combination or not with conventional reinforcement) as spread reinforcement of precast tunnel segmental linings. Polypropylene fibers could be used in precast tunnel segments as shear reinforcement and for withstanding splitting and spalling tensile stresses which occur under and between TBM rams, respectively. In addition, fibers could also simplify and boost the tunnel element production process. In this context, the present study investigates the possibility of using polypropylene macro fiber reinforcement in precast tunnel segments for both hydraulic and metro tunnel linings. Four full-scale segments (two counter-key segments of a hydraulic tunnel and two trapezoidal shaped segments of a metro tunnel) were experimentally evaluated under point load tests (Figure 1). The latter simulates the TBM actions on segments during the excavation process (TBM thrust jack phase), which is generally the most critical temporary loading condition for segments. For both Hydraulic (specimen designation: H) and Metro (specimen designation: M) tunnel segments, the following two reinforcement solutions were considered: 1. conventional reinforcement solution (hydraulic tunnel segment H-RC and metro tunnel segment M-RC), in which the typical amount of conventional reinforcement for practice was adopted. These segments (H-RC and M-RC) are also considered as reference samples; 2. hybrid solution (hydraulic tunnel segment H-RCO+PFRC and metro tunnel segment MRCO+PFRC), in which a combination of reinforcing bars and polypropylene fibers was properly designed to obtain an optimized reinforcement solution. In case of hydraulic tunnel, both reinforcement solutions allowed segments to reach the maximum applied load of 3000 kN (much higher than both design and maximum thrust loads) showing a very similar stiffness and final crack pattern. At both design and maximum thrust loads only a spalling crack were observed and the maximum crack opening was similar for HRC and H-RCO+PFRC segments, i.e. 0.05 mm. Therefore, H-RCO+PFRC segment showed a global structural response similar to the H-RC one with a good control of spalling cracks. In fact, in the hybrid segment smaller and more closely spaced spalling cracks were observed as compared to H-RC segment. Noteworthy is also that polypropylene fibers in the hybrid solution were the only reinforcement used to withstand tangential splitting cracks and they were able to control them in a similar way of rebars in the RC sample.

138

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Concerning metro tunnel segments (Figure 1a), it can be observed that also in this case both reinforcement solutions allowed segments to reach the maximum capacity of the loading system (Figure 1b), i.e. 4000 kN. The latter is significantly higher than the design and the maximum thrust loads and it was reached by both segments showing a similar stiffness. The spalling crack width was comparable in M-RC and M-RCO+PFRC segments, even if it resulted slightly higher in the hybrid reinforcement solution. At the design load, this crack was not present in RC segment, while it was very small in M-RCO+PFRC one. Splitting cracks were well controlled at any load levels by polypropylene fibers in the M-RCO+PFRC segment by a transverse stress redistribution that continuously provide equilibrium with the external applied load. Therefore, both reinforcement solutions guaranteed the force equilibrium and a good control of splitting cracks up to maximum capacity of the loading system (which is significantly greater than the maximum thrust load). These results on metro tunnel segments, extend and confirm the previous outcomes obtained on hydraulic samples.

(a) Figure 1:

(b)

Point load tests on metro tunnel segments: test set-up (a) and experimental response in terms of single shoe load vs. V#1 displacement curves (b).

The experimental results proved that a combination of polypropylene fibers and a low amount of conventional reinforcement is an attractive reinforcement solution for both hydraulic and metro tunnel segments. In fact, polypropylene fibers can be used in precast tunnel segments to withstand shear stresses and tensile forces in the splitting zones under TBM loading shoes, while spalling crack control and flexural resistance is better guaranteed by an optimized combination of fibers and a low amount of curved rebars.

Keywords Fiber reinforced concrete; Polypropylene fibers; Point load test; Precast tunnel segments; TBM thrust jack phase

139

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Design of SFRC Precast Tunnel Segments Supported by NLFEA Radomír Pukl 1 1:

Červenka Consulting s.r.o., Prague, Czech Republic.

Abstract The paper deals with design of precast tunnel segments made from steel fibre reinforced concrete (SFRC) used for Ejpovice railway tunnels in the Czech Republic. Recent possibilities of the non-linear finite element analysis (NLFEA) offer a versatile tool for design of SFRC structures. For the modeling of the SFRC material special numerical models are available accounting for the SFRC specifics such as shape of tensile softening branch, high toughness and ductility. Appropriate input material parameters can be identified from the measured response of material bending and compressive material tests using inverse analysis procedure. Design of the segmental tunnel lining has been performed based on a combination of the NLFEA and laboratory tests of the precast SFRC segments. The numerical investigation and design of the structure - precast tunnel segments for TBM technology - was performed for various construction stages and various loading situations. During project preparation and design procedure the results from numerical simulation of SFRC tunnel lining segments were compared with the experimental evidence. Response of the structural members under service loads and their damage under limit loads were evaluated in order to verify and support NLFEAbased design of the SFRC tunnel segments.

Keywords Steel fibre reinforced concrete, segmental tunnel lining, finite element analysis, non-linear material models.

Introduction Utilization of the steel fibre reinforced concrete (SFRC) for segmental tunnel lining promises potential advantages in comparison to the traditionally reinforced concrete (RC) structures efficient manufacturing, lower risk of corrosion, less damage during transport, etc.

Design of SFRC segments by nonlinear analysis As part of modernization of the railway line Rokycany – Pilsen two single track railway tunnels are constructed. Lengths of southern and northern tunnel were 4150 m and 4176 m respectively. The circular segmental lining is used for the tunnels. The lining consists of precast SFRC segments and reinforced concrete segments with inner radius 4.35 m in both cases. The segment thickness is 0.40 m and the width of load carrying ring is 2.0 m. Due to the lack of standards for SFRC structures design the nonlinear finite element analysis was utilized.

140

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Recent possibilities of the non-linear finite element analysis (NLFEA) offer a versatile tool for design of SFRC structures. Special numerical material models were developed for modeling SFRC taking into account SFRC specifics such as shape of tensile softening branch, high toughness and ductility. Bending tests of SFRC precast segment performed by Klokner Institute in Prague were used for identification of material parameters which are necessary as an input for the nonlinear material model in ATENA software. The appropriate material parameters were determined by inverse analysis of load displacement curve and failure mode of test and model. Comparison of results and failure modes between the laboratory test and the model confirmed that the identified model is well suitable for modeling SFRC structures. ATENA software was used for design and assessment of SFRC tunnel lining as well as for identification of material parameters. Crack width, strain, limitation of compressive stress and ovalization of the tunnel lining were assessed within the serviceability limit state. In the ultimate limit state the structure was subjected to various load cases and combinations. The cross section subjected to critical combination was assessed in compression and bending including check of the shear forces.

Concluding remarks Nonlinear computer simulation analysis was utilized for modeling of SFRC precast tunnel segments and design and assessment of the segmental tunnel lining. Appropriate material characteristics for SFRC numerical model were obtained by inverse analysis from laboratory tests. The approved models were then used in design of the structure in serviceability as well as ultimate limit states. The suitability of the nonlinear finite element analysis for design and assessment of SFRC structures was confirmed on a practical project.

Acknowledgements The presented results are based on the research performed within Eurostars project E!10316 "Virtual Lab for Fibre Reinforced Concrete Design by Simulation Prototyping". Authors gratefully acknowledge this support.

141

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Closed-form Solutions for Interaction Diagrams of Hybrid Fiber-Reinforced Tunnel Segments Yiming Yao1, Mehdi Bakhshi2, Verya Nasri2, Barzin Mobasher3 Precast concrete segments reinforced by steel bars are the predominant support method used in tunnels dug by Tunnel Boring Machines (TBM), where a significant amount of time and labor are needed to assemble the cages and place the reinforcing bar. As alternative reinforcement, fibers may considerably improve the concrete post-cracking behaviour, crack control and enhance handling and placement of precast concrete segments. A hybrid solution of combined rebar and fiber reinforcement has been adopted when the use of FRC is sometimes not adequate as the sole reinforcing mechanism. Tunnel segmental linings are mainly subjected to combined axial force and bending moment due to multiple loading cases that occur during manufacturing, transportation, installation, and service conditions. Axial force-bending moment interaction (P-M) diagram is therefore, an important tool that tunnel engineers use for the structural design. This paper presents material models, derivations and for the first time, closed-form solutions to construct a full range P–M interaction diagram of HRC segments considering the contributions of fibers. The model is verified by comparing with experimental and numerical results from published data. The proposed P–M diagram can be used as a tool for design of HRC tunnel segments especially in large diameter tunnels. Three distinct material models are used to derive parametric responses of HRC beams which address concrete and steel’s tensile and compressive piecewise linear constitutive laws. The simplified bilinear tension model is capable of covering a wide range of tension softening behavior exhibiting both deflection-hardening and deflection-softening responses in flexure. The effect of lateral ties on the compressive behavior is characterized by an improved compressive strength and ultimate strain that can be specified with design codes. In derivation of axial force (P) and bending moment (M) for a rectangular cross section, the assumption of plane section remaining plane is used. Taking the normalized strain at the bottom extreme fiber as the independent variable to incrementally impose axial and flexural deformations for three modes of failure. The strain and stress distributions are subsequently constructed using the linearized material models. The next step is to integrate the stresses over the area in each zone to obtain the parametric solutions of force terms, which are expressed in quadratic expressions. Internal moment is obtained by integrating the force components using the distance to the center line as the moment arm. Strain limits of the materials are required by design codes to ensure sufficient ductility of structural members subjected to axial compression and bending moment. The proposed model is used on a case study of precast tunnel lining segments reinforced by longitudinal rebars and steel fibers. Tiberti (2009) performed numerical analysis on different case studies of the tunneling segments with internal diameters ranging from 7.25 m to 14.9 m. The effect of steel fiber was considered in the study by using a similar concept of residual tensile strength “χ” that ranges from 0 to 0.75. Each case was performed in two configurations: plain SFRC and HRC sections with a reinforcement ratio of ρ=0.4%. Other assumptions include the design compressive strength of the concrete fcd = 22.7 MPa, tensile strength fcld = 1.6 MPa, concrete modulus E = 35 GPa, design yielding strength of steel fyd = 391 MPa and steel modulus Es = 200 GPa. The case study of “Barcelona” lining with a tunnel diameter of 10.9 m is presented in this work by comparing the original analysis with the proposed model. Fig. 1 compares the interaction diagrams for the sections made of SFRC and HRC for the Barcelona 142

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

lining configuration. The cross sectional geometries and model parameters are indicated in the figures. In order to address the residual strength used in the original study, same values are assigned to the normalized residual strength parameter μ, which are 0.25, 0.50 and 0.75, respectively. Similarly, increments in the ultimate moment capacity are observed with increasing μ value.

(a) Figure 1.

(b) P-M diagram for the Barcelona lining configuration: (a) SFRC section, and (b) HRC section (Tiberti, 2009).

This paper presents, for the first time, analytical closed-form solutions to construct full range P-M interaction diagram for HRC segments based on parametric multi-linear material models for tension and compression of FRC matrix and steel model for rebar. All failure modes including all compression, compression controlled and tension controlled were covered and the closed-form equations for each case were explicitly derived. The accuracy of the model was verified by simulating experimental results from the literature for cases of structural columns, and comparison with model-predicted results of other established numerical and finite-element analyses on precast concrete segments and cast-in-place concrete tunnel linings. Results show that using appropriate material models for fiber and reinforcing bar, engineers can use the proposed methodology to obtain P-M interaction diagrams for HRC tunnel segments. The proposed methodology can be used by tunnel engineers as one of their main tools for design of HRC segments that are subjected to combined axial force and bending moment.

143

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Steel and Polypropylene Fiber Reinforced Concrete for Secondary Tunnel Lining Jakob Sustersic1, Marko Lutman2, Andrej Likar3, Melanija Huis3, Andrej Zajc1 1

: IRMA Institute for Research in Materials and Applications, Ljubljana, Slovenia. 2 : DRI Investment Management Ltd, Ljubljana, Slovenia. 3 : Geoportal, Ljubljana, Slovenia.

Introduction In the paper, properties of Steel and Polypropylene Fiber Reinforced Concrete (SPFRC) are shown and discussed. It was used for construction of secondary lining of the tunnel Sten on the road bypass around the town of Škofja Loka in Slovenia. Due to the uneven excavation, thickness of the secondary lining varies between 30 and 80 cm. Therefore, lining made of concrete with higher tensile strength and resistant to crack propagation has to be applied. There was one of the basic requirements of the project, no cracks on the surface of the secondary lining. Based on a preliminary study, SPFRC was selected as the most suitable material that could be used for the construction of secondary lining because it has properties that would greatly contribute to reducing the occurrence and propagation of cracks in the lining. In addition to the achievement of the required properties in the harden state, SPFRC mixture had to be suitable for pumping and filling the space between framework and waterproofing layer on the primary lining.

Materials and Mix Proportions Two mixtures of SPFRC were prepared at the concrete plant which differed in particular from the maximum grain of the aggregate: with Dmax = 16 mm (composite designation SPFRC16) and with Dmax = 32 mm (composite designation SPFRC-32). 360 kg/m3 Blended Portland cement with addition of granulated blast-furnace slag was added to both SPFRCs. Effective water – cement ratio (w/c)eff of SPRFC-16 was 0.43 and (w/c)eff of SPFRC-32 was 0.44. Air entraining admixture with super-plasticizing effect was added to SPFRC-32, while hyperplasticizer and air entraining admixture were added to SPFRC-16. Hooked steel fibers with length of 16 mm and diameter of 0.50 mm in the amount of 30 kg/m3 and fibrillated polypropylene fibers with length of 10 mm in the amount of 1 kg/m3 were added to both SPFRCs.

Results and discussion Workability of SPFRC-16 was better compared with that of SPFRC-32 after mixing in the concrete plant. Decrease of workability during transport from concrete plant to site was minimal in the case of SPFRC-16 while, decrease of workability of SPERC-32 was significant. Transport of the SPFRC lasted between 20 and 30 minutes depending on the traffic on the road. However, both SPFRC-16 and SPFRC-32 were easy pumped and placed, without any break and blockage of pump tube. Fibers were not obstructing SPFRC pumping because they are short, and also their quantity was not large. Obtained average porosity of SPFRC-16 is a little 144

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

bit large than the calculated value. While, obtained and calculated values of SPFRC-32 porosity is the same. Practically, porosities of SPFRC-16 and SPFRC-32 did not change during the transport. Measurement of heat of hydration was carried out by procedure given in the paper by Ng P.L. et al. Adiabatic thermal characteristics of SPFRC-32 are measured on a site in semiadiabatic conditions. From the obtained results it can be concluded that T = 45.7°C at the initial temperature of fresh SPFRC-32 - T0 = 23.0°C. Such result was expected and has been taken into account in preliminary calculations. Test results of resistance to water penetration under pressure and resistance to-freeze/thaw without a de-icing agent show that the SPFRC meets the requirements of the project. There was also a slight shrinkage of SPFRC to the age of 180 days - 0.40 mm/m. The compressive strength of SPFRC-16 is slightly higher compared to SPFRC-32 at all ages. The difference is increasing with the age of the SPFRC. In any case, the initial strengths of both SPFRC allow relatively early removal of the formwork and thus the rapid progress of the work in the tunnel. In addition, both meet the conformity criteria for the required strength class C25/30. Ultimate split strength fct, and equivalent strength up to the crack width of 0.2 mm f0.2 were calculated by taking into account parameters of load - CMOD diagrams obtained by WST method. The areas under diagrams of SPFRC-16 are larger compared with those of SPFRC-32. This means that a greater amount of energy is absorbed during the loading of the SPFRC-16, which leads to a more tough behavior. Therefore, SPFRC-16 exhibits greater resistance to crack propagation. The resistance of SPFRC to crack propagation can also be estimated with a ratio of f0.2/fct. This ratio is almost the same for all ages of SPFRC (Figure 1). However, it should be taken into account that the SPFRC-16 has a higher fct compared to the SPFRC-32. This difference increases with increasing into the age of both SPFRCs. The most important finding is that the young SPFRC-16 shows the highest ratio f0,2/fct. This property proved to be very suitable for preventing cracks in earlier times after the placement of SPFRC-16 in the secondary lining of the Sten tunnel. The behavior of the SPFRC-16 in the tunnel lining for a period of two years is appropriate. No significant vertical cracks have been observed.

Figure 1:

The ratio f0.2/fct versus time.

145

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental Behaviour of Precast Tunnel Segments in Steel Fiber Reinforcement with GFRP Rebars Alberto Meda1, Zila Rinaldi1, Simone Spagnuolo1, Benoit De Rivaz2, Nello Giamundo3 1

: University of Rome Tor Vergata, Rome, Italy. 2 : BMUS, Aalst, Belgium. 3 : ATP, Angri, Italy.

Abstract The interest in using fiber reinforced concrete (FRC) for the production of precast segments in tunnel lining, installed with Tunnel Boring Machines (TBMs), is continuously growing, as witnessed by the studies available in literature and by the actual applications. The possibility of adopting a hybrid solution of FRC tunnel segments with GFRP reinforcement is investigated herein. Full-scale tests were carried out on FRC segments with and without GFRP cage, with a typical geometry of metro tunnels In particular, both flexural and point load full-scale tests were carried out, for the evaluation of the structural performances (both in terms of structural capacity and crack pattern evolution) under bending, and under the TBM thrust. Finally, the obtained results are compared, in order to judge the effectiveness of the proposed technical solution.

Segment geometry and materials Four full-scale fiber reinforced concrete segments were cast in moulds available at the Laboratory of the University of Rome Tor Vergata. The specimen geometry is shown in Figure 1. Two of the four segments were further reinforced with a perimetric GFRP cage.

Figure 1:

Segment geometry

Bending tests The bending tests were performed with the loading set-up illustrated in Figure 1a, The obtained results expressed in terms of load displacement are shown in Figure 1b

146

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

a)

Figure 2.

b)

Bending test:a) set up; b) results.

Point load tests The point load test was performed by applying three-point loads at the segment, and adopting the same steel plates used by the TBM machine (Fig. 3).

Figure 3.

Point load tests: test set-up and crack patterns.

Concluding remarks 1. The results of bending tests, clearly show the synergic effects of the two materials (fibers and GFRP reinforcement), by increasing the peak load and reducing the crack width. 2. The results of the point load test confirm the effectiveness of the solution.

147

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Structural behavior of precast tunnel segments with macrosynthetic fibers during TBM operations: a numerical study Ivan Trabucchi1, Antonio Conforti1, Giuseppe Tiberti1, Giovanni Plizzari1 and Ralf Winterberg2 1 2

: University of Brescia, Brescia, Italy. : Elasto Plastic Concrete, Singapore.

Abstract The use of fiber reinforced concrete in tunnel linings, with or without conventional rebars, has increased in the two last decades, especially in segmental linings. In the meanwhile, in the scientific community there was a growing interest on macro-synthetic fibers for use in underground structures. Within this framework, the present study investigates the possibility of using macro-synthetic fiber reinforcement in precast tunnel segments by means of a numerical study. Firstly, an experimental program was carried out for the characterization of polypropylene fiber reinforced concretes (PFRCs) adequate for tunnel elements. Polypropylene (PP) fibers were added to a base concrete with a target mean cube compressive strength of about 6065 MPa in two different amounts: 6 kg/m3 (PFRC6 series) and 10 kg/m3 (PFRC10 series). For each matrix, twelve small beams 150x150x550 mm according to EN 14651 standard for determining post-cracking behavior were produced. Both mixtures showed a stable postcracking response. PFRC matrixes were characterized by a softening behavior with a load drop after the peak load, followed by an increment of the residual flexural tensile strength. For larger values of crack width, a soft and progressive decrease of the residual flexural tensile strength was observed. By using higher fiber contents, the load drop (after the peak) decreases and the increment of the residual flexural tensile strength becomes more pronounced. These good postcracking performances given by the adopted PP fibers are mainly due to their embossed shape, which significantly increases the bond between fiber and matrix. The obtained fracture properties of PFRC6 and PFRC10 fulfil the requirements of fib Model Code 2010 for use in structural elements, as well as satisfy the requirement of Equation 7.7-14 of Model Code 2010 allowing to use only PP fibers as minimum shear reinforcement. Secondly, the corresponding stress vs. crack opening laws, representative of the PFRCs investigated, were evaluated through inverse analysis procedure. Then, a segment of a typical tunnel lining having small diameter was adopted as reference to optimize the reinforcement solution and to study its structural behavior by numerical analyses. Particular attention was devoted to the Tunnel Boring Machine (TBM) thrust jack phase, in which the TBM moves forward by pushing the thrust jacks on the bearing pads of the latest assembled ring, introducing high-concentrated forces in the lining. The following three reinforcement solutions were studied: 1. PFRC6 and PFRC10 segments: PP fiber reinforcement only; 2. RCO+PFRC6 and RCO+PFRC10 segments: hybrid solution based on a combination of PP fibers and conventional reinforcement; 3. RC segment: typical conventional reinforcement generally adopted in practice.

148

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

3.0

2500

2.5

2000

2.0 Maximum thrust load

1500

Service thrust load

1000

PFRC6 PFRC10 RCO + PFRC6 RCO + PFRC10 RC

500 0 0.00

0.20

0.40

(a) Figure 1:

3000

Eccentric loading condition

3.0

2500

2.5

2000

2.0

1.5

1500

1.0

1000

0.5

500

0.0 0.60 0.80 1.00 Spalling crack opening [mm]

0

Maximum thrust load

Service thrust load

PFRC6 PFRC10 RCO + PFRC6 RCO + PFRC10 RC 0.0

0.2

Load/Service load [-]

Normal loading condition

Load [kN]

3000

Load/service load [-]

Load [kN]

The numerical analyses were performed by means of the finite element code DIANA 10.1, using a three dimensional numerical model of one single tunnel segment with two pairs of actuators. Four loading configurations were considered: 1) normal loading condition: perfect placement of TBM thrust shoes and perfect positioning of the segments on the previous assembled ring; 2) eccentric placement of TBM shoes: an outward eccentricity of about 20 mm was considered; 3) un-even supports: the partial loss of the contact (due to a possible gap) in one support was considered; 4) un-even supports combined with outward eccentricity as a combination of previous mentioned irregularities. In case of normal loading condition (Figure 1a), PFRC and RCO+PFRC segments were able to guarantee the required structural performance at both service and maximum TBM thrust loads in a similar way than RC samples. Concerning unfavorable conditions (Figure 1b), which may occur during the construction of the lining, it was observed that the bearing capacity of precast tunnel segments can significantly decrease. For example, the partial loss of the supports behind the segment (un-even support) can lead to an average reduction of the maximum applied load of about 40% with respect to the normal loading condition. In addition, the reinforcement solution with fiber reinforcement only (PFRC6 and PFRC10) was observed to be more vulnerable in terms of expected spalling crack widths against unfavorable conditions (especially for loads higher than 1.5 times the service load), while the combination of a low amount of conventional rebars and macro-synthetic fibers was able to guarantee a local and global behavior of segments similar to the one observed in the RC solution, even when a critical loading condition occurs.

1.5

1.0

0.5

0.0

0.4 0.6 0.8 1.0 Spalling relative displacement [mm]

(b)

Spalling crack opening between TBM thrust shoes at different load levels: normal loading condition (a) and eccentric placement of TBM shoes (b).

Keywords Fiber reinforced concrete; Macro-synthetic fibers, Numerical analyses; Splitting phenomena; Thrust jack

149

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

150

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: High Performance FRC

151

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Structural behavior of prestressed Ultra-High Performance Fibre-Reinforced Concrete beams with and without openings: comparison between experimental results and finite element modelling techniques Dario Redaelli1, Joanna Yared Nseir2 1 2

: University of Applied Sciences Western Switzerland, Fribourg, Switzerland. : Saint Joseph University, Beirut, Lebanon.

Abstract This paper presents the results of a numerical study carried out by the authors to better understand the structural behavior of prestressed beams with a complex geometry and to identify numerical modelling techniques that allow to adequately predict such behavior. UltraHigh Performance Fibre Reinfored Concrete (UHPC) beams with and without openings are considered, with a focus on shear controlled failure modes. For all the beams considered in this study, prestressing is used to resist the main bending moment. However, no other reinforcement is added to the beams in order to emphasize the structural contribution of the fibers and to focus on solutions that could be economically competitive for the precast industry. The results of non-linear simulations performed with existing finite elements codes are compared and validated against experimental results of tests carried out at the University of Applied Sciences of Western Switzerland. The main assumptions of the numerical simulations are discussed, as well as the results and the limits of the analysis.

Keywords Ultra-High Performance Concrete, Finite Elements, Prestressed Beams, Cellular Beams

Introduction UHPC structural elements usually present complex geometries. The prediction of the mechanical behavior of irregular elements of this kind requires appropriate analytical or numerical modelling techniques. In order to acquire a better knowledge on the structural behavior of UHPC beams with irregular geometry, a test series of prestressed UHPC cellular beams was analyzed with nonlinear finite element models.

UHPC material properties for modelling Tensile laws for modelling very derived from bending tests carried out on sawn prisms.

152

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

Constitutive curves: left) exponential model; right) linear model.

Finite element modelling based on actual material laws A first nonlinear modelling approach was based on the actual material laws derived above.

Figure 2:

Comparison between experimental and numerical results obtained with ATENA.

Finite element modelling based on elasto-plastic stress fields Secondly, modelling was carried out based on elasto-plastic stress fields.

Figure 3:

Comparison between experimental and numerical results obtained with j-Conc.

153

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Use of a Probabilistic Explicit Cracking Model for Analyzing the mechanical behaviour of an Ultra-High Performance Fibre Reinforced Concrete (UHPFRC) Beam Subjected to Shear Loading Pierre Rossi1, Dominic Daviau-Desnoyers2, Jean-Louis Tailhan1 1 2

: Université Paris-Est, IFSTTAR, MAT, F-75732, Paris, France : CIMA+, Montreal, Canada.

Introduction Existing national and international recommendations do not provide sufficient relevant information regarding cracking at the serviceability limit state (crack opening and spacing). In this way, the best approach for designing structures with respect to both safety and sustainable development is the use of finite element analysis. The objective of this paper is to use the probabilistic explicit cracking model, developed by IFSTTAR, to simulate the behaviour of a reinforced UHPFRC beam subjected to a bending load leading to shear failure.

Probabilistic Explicit Cracking Model of Steel Fibre Reinforced Concrete (SFRC)

Principles of the explicit probabilistic cracking model of SFRC.

154

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Numerical Modelling of a reinforced UHPFRC beam submitted to shear loading The probabilistic explicit cracking model developed for SFRC has been used to simulate the mechanical behaviour and cracking process of a longitudinally reinforced UHPFRC beam without transverse reinforcement bars.

1.1 Experimental Details The beam (bulb-T section - 38 cm high) containing longitudinal reinforcement was designed to avoid bending failure. The reinforcement consisted of 5-HA20 and 1-HA25, the latter being placed in the middle of the bottom layer of rebars. The test realized on this reinforced beam was a 4-point bending test. The spans between the loading points and the supports were respectively 480 and 2000 mm. The UHPFRC mix design used was a self-compacting concrete made of fine sand with a maximum aggregate size of 0.8mm and straight steel fibres with a length of 13 mm and a diameter of 0.2 mm at 2.5 % by volume. The UHPFR had a 28-day mean compressive strength of 212 MPa, a Young’s modulus of 56 GPa and a Poisson’s ratio of 0.185.

1.2

Numerical Simulation of the Shear Behaviour of the Beam

The model parameters obtained through the inverse analysis approach are summarized in following Table. Specimen designation and main properties. ft [MPa]

σ(ft) [MPa]

W [MPa*mm]

σ(W) [MPa*mm]

ζ0 [mm]

ζc [mm]

8.43

1.31

60

72

0.035

5

The following Figure presents a comparison between the experimental load-deflection curve and the three numerical simulations.

Experimental and numerical load-deflection curves of the reinforced UHPFRC beam submitted to shear loading.

155

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

UHPFRC Precast slabs and field-cast joints of the IsabeyDarnley pedestrian bridge Charron Jean-Philippe1, Desmettre Clélia1, Cantin Bellemare Étienne2 1

: Department of Civil, Geological and Mining Engineering, École Polytechnique Montréal P.O. Box 6079, Station Centre-ville, Montréal, QC, H3C 3A7, Canada 2 : Bridges and Tunnels Division - City of Montreal, Quebec, Canada e-mail: [email protected], web page: www.polymtl.ca/structures.

Project overview The Isabey-Darnley pedestrian bridge is located in Montreal (Quebec, Canada) over Highway 520 (average annual daily traffic of 30 000) close to the International Airport. This 61 m long footbridge built in 1967 suffered from extensive concrete spalling and rebar corrosion after 40 years of service and needed to be replaced. Among the replacement requirements, traffic delays and redirection below the structures had to be minimized and the new structure had to ensure an excellent durability under severe exposure conditions. In this context, precast UHPFRC slabs linked by UHPFRC field-cast joints were selected for the slab of the new structure.

Design of precast UHPFRC slabs and field-cast UHPFRC joints 1.3

Concept of the slab

The slab of the Isabey-Darnley pedestrian bridge consists in 11 precast UHPFRC slabs of 75 mm-thickness, 2.25 m-width and around 6 m-long (except for one shorter slab of 2.1 m) linked by 125 mm-wide UHPFRC field-cast joints (100 mm rebar splices for 10M rebar). The precast slab includes curbs which serve as a security cladding for pedestrians, favour water drainage and strengthen the 75 mm-thick slab. The cross-section of this slab (Figure 1) illustrates the limited reinforcement in positive and negative bending as well as for crack control requirements.

Figure 1:

1.4

The 75mm-thick precast UHPFRC slab.

Methodology

The design of the precast slabs and field-cast joints were made considering the Canadian Highway Bridge Design Code (2006) for loads and load combinations. Pending the introduction of design equations for UHPFRC structures in the code 2019 edition, the recommendations of the French Association of Civil Engineering (AFGC, 2013) for UHPFRC structures were used in this pilot project for determining mechanical constitutive laws, durability requirements and quality

156

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

control recommendations. Finite-element analysis with Atena 3D software were performed to evaluate the structural behavior of the precast slabs and joints.

1.5

Design highlights

The exceptional durability of uncracked and cracked UHPFRC in comparison to conventional concrete allowed utilization of a concrete cover of 25 mm. Moreover, the outstanding mechanical properties of UHPFRC permitted significant reductions of 64% in concrete volume and 91% in rebar volume in comparison to a conventional reinforced concrete slab. Furthermore, the improved bond of UHPFRC to rebar allowed an important reduction of the bar lap splice (100 mm that represents 10 db, instead of 40 db for reinforced concrete) in the joints connecting the slabs, and thus a reduction of the joint width (125 mm).

Feed-back on the construction A proof specimen of a precast UHPFRC slab (one third of its length) was tested under positive bending moment at Polytechnique Montreal Structural laboratory. The ULS design moment requirement and SLS crack opening limit were met easily. The mechanical behavior of the proof specimen of the slab was adequate and fulfilled all design specifications. The installation of the UHPFRC precast slabs was completed in 2016 in two phases of 8 hours during the night (complete closing of the highway), while 3 periods of 8 hours during the night (partial closing of the highway) were required for the casting of the UHPFRC field cast joints under winter conditions. With the experience gained, only one period of 8 hours could have been required for the installation of the precast slabs and only 2 periods of 8 hours for the casting of the joints, limiting further the impact on the road traffic.

Conclusion The Isabey-Darnley pedestrian bridge was a pilot application including precast UHPFRC slabs and field-cast UHPFRC joints. The project objectives, which were to design an innovative, durable and aesthetic footbridge minimizing impact on traffic during construction, were achieved. The main conclusions regarding the bridge slab are summarized below. •

The design was completed with nonlinear finite element calculations. The mechanical behavior of the precast slab and field-cast joints models under the critical conditions demonstrated that the selected design met easily the ULS design moment requirement and SLS crack opening limit.

•

A mechanical test performed on a full size precast UHPFRC slab under positive bending moment confirmed an adequate mechanical behavior and fulfilled all design specifications.

•

The utilization of UHPFRC precast slabs and field-cast joints allowed a significant reduction of on-site construction time, as well as a decrease of the traffic management and traffic perturbation around the bridge.

157

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Fibre effect on shear behaviour of UHPC beams Norbert Randl1, Tamás Mészöly1 1

: Carinthia University of Applied Sciences, Austria.

Abstract The load bearing behaviour of fibre reinforced UHPC members in shear has not yet been investigated sufficiently. The difficulty is that on one side shear behaviour itself is a complex phenomenon, while on the other side UHPC is a rather new material that still needs to be fully explored. Above that, it is known that fibres have a significant impact on a concrete beam’s shear load bearing capacity in general. Therefore an experimental campaign was conducted, investigating the behaviour of I-shaped beams made of Ultra-High Performance Concrete (UHPC) subjected to shear loading. The height of the beams was 350 mm, and the single span length 3.00 m. The main parameters were the fibre content (up to 2% by volume) and the degree of stirrup reinforcement (no stirrups or diameter 10 mm bars, spacing 125 to 300 mm). A total of 20 tests were performed, all of them ending up in typical shear failure: Pronounced shear cracks formed along the web between support and load introduction at a level of 60-70% of the ultimate load. Close to ultimate load, the formation of a critical shear crack was observed. The cracking process was monitored by a digital image correlation (DIC) measurement system. A decisive effect of the fibres was observed: The addition of 1% to 2% fibres by volume led to a more distributed crack pattern as well as a significant increase of the ultimate load. The enhancement of the shear load capacity was not proportional to the fibre content, but corresponding quite well to the UHPFRC’s maximum tensile strength in the post-cracking stage. The interaction with varying degrees of stirrup reinforcement was assessed as well. The contribution of fibres to the overall shear resistance was not very sensitive to variations in the degree of stirrup reinforcement. The effect of stirrups in turn reduced somewhat when adding fibres.

Keywords UHPC, UHPFRC, fibre reinforcement, shear load

Experimental program A number of 20 beams made of UHPC were subjected to monotonic shear loading until failure. The test setup was a three-point bending configuration with the exception of the first two beams (B1 and B2) where a four-point setup was chosen. In order to ensure the intended shear failure mode, an I-shaped cross section was designed. The single span of the 3.50 m long beams was 3.00 m and the height of the I-shaped cross section was 350 mm, the thickness of the web 58 mm and the width of the flanges 200 mm. The shear-span-to-depth ratio a/d was larger than 3 (3.5 and 3.2) in all tests in order to prevent possible direct load transfer to the support closest to the load introduction. The following Table 1 summarizes the different tested configurations. The steel fibre content was varied from 0 vol.-% to 2 vol.-%. Several beams were provided with stirrups ø10, the different spacings are listed in Table 1. The average value

158

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

of the compressive strength measured on cubes with a side length of 100 mm was around 166 MPa on the 28th day and 171 MPa on the day of testing. Table 1:

Overview of tested beam configurations.

Fibre content [vol.-%]

Stirrups

No stirrups

Ø10/125mm

Ø10/200 mm

Ø10/300 mm

0

B21, B23, B27

B22, B35

B16

B18

1

B20, B24, B29

–

B36

B28

2

B19, B25, B30

B1, B2

B15, B26

B17

Main observations from tests All beam tests resulted in a typical shear failure (Fig. 1) and confirmed the significant contribution of steel fibres to the ultimate shear bearing capacity. The test results were evaluated on the basis of the provisions given in the French “Ultra High Performance Fibre-Reinforced Recommendations” (revised ed. 2013). As a conclusion it can be stated that the approach predicts the test results quite well, however being about 15 to 20% on the conservative side. The mean ratio of tested versus calculated shear strength is 1.18 and the coefficient of variation 0.19. The French recommendation in the present study slightly underestimates the contribution of the fibres especially without stirrup reinforcement and in turn overestimates the contribution of the stirrups (Fig. 1).

Figure 1:

Shear failure (left) and relationship between Vu,test / Vu,calc and fibre content (right).

159

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Practical considerations on the application of the recent SIA 2052 guidelines on testing of Ultra-high-performance fiberreinforced concrete Roman Loser1, Janis Justs1, Pietro Lura1,2 1

: Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland. 2 : Institute for Building Materials (IfB), ETH Zurich, Switzerland.

Abstract Within the last few years, Ultra-high-performance fiber-reinforced concrete (UHPFRC) has progressed from the research lab and from niche applications to becoming a viable and in specific cases the preferred solution for increasingly large construction projects. Due to the recent increase in the use of UHPFRC, standards and guidelines were published recently or they are in preparation in different countries. In Switzerland, the guidelines SIA MB 2052: Ultrahigh performance fiber reinforced concrete (UHPFRC) - Materials, design and execution was published in 2016. According to SIA MB 2052, UHPFRC is divided into three different classes. The classes are defined by the following material properties: fUtek characteristic value of elastic limit tensile strength; fUtuk characteristic value of tensile strength; Utu strain-hardening (strain at tensile strength); fUck characteristic value of compressive strength. The compressive strength is obtained from compression tests on either cubes or cylinders. No strength classes are defined, but only a minimum value for the strength that needs to be met. With 120 MPa, it is the same for all classes defined above. fUtek, fUtuk and Utu are derived from direct tensile tests and bending tensile tests defined in the appendix of the guidelines. These measurements have to be performed by manufacturers at defined intervals as a quality-control assessment. The average of a series of 6 specimens is used as the characteristic value to assign the tested UHPFRC to one of the three classes. The results of the direct tensile test and the bending tensile test are crucial for the classification of a UHPFRC. Therefore, the results have to represent the materials behavior and they have to be reproducible. However, as the stress in the tensile tests is oriented in one direction, the fiber orientation has a large impact on both the tensile strength fUtu and the strain-hardening Utu. The highest and most reproducible values with the lowest scatter result when all fibers are perfectly oriented in the direction of the tensile stress induced by the test setup. As soon as there are some small local inhomogeneities, mainly resulting from local differences in fiber distribution, considerably lower values are consistently found. In addition, the scatter in tensile strength fUtu and strain-hardening Utu between nominally identical specimens increases substantially in these conditions. The experience gathered at Empa after developing and testing several UHPFRC recipes shows that the manner of formwork filling is the critical parameter. An example from practice is illustrated in Figure 1, which shows results of direct tensile tests according to the SIA 2052 guidelines. The black curves (A) correspond to the first trial, in which the specimens were produced in the manufacturer’s laboratory and brought to the testing lab. In the experimental results, a large scatter after having reached the elastic limit strength is apparent, while the curves do not show any clear strain-hardening behavior for all specimens. Based on these results, the 160

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

composition of the mixtures was optimized in different steps, resulting in the green curves (B). The scatter was still high, with some specimens showing a strain-hardening behavior and others not, even if the manufacturer paid attention when filling the formworks. The large scatter and the lack of strain hardening behavior of most specimens in series (B) of Figure 1 are caused by random local inhomogeneities, which arise due to the production process of the tested specimens. In addition, the reproducibility of the results is clearly limited, which additionally adversely affects material development.

Figure 1:

Example curves from direct tensile tests according to the SIA 2052 guidelines (see text for details).

To investigate further the above-discussed influence of specimen manufacturing on the material properties, a UHPFRC with identical mixture composition as for Series (B) was mixed again at the Concrete and Construction Chemistry Laboratory at Empa. The formwork was filled in several layers with a long half pipe, allowing the UHPFRC to flow very homogenously into the formwork from one side to another. The resulting stress-strain curves (blue curves (C) in Figure 1) are completely different from the curves (B), with much higher values for fUte, fUtu and Utu and with very small scatter. These results have a very good reproducibility within one series, which makes it possible to optimize the mixture in regard to the mechanical properties. As a last step, the UHPFRC was produced again with the same mixture composition and the same manner of specimen production but at the construction site. The obtained values (red curves (D) in Figure 1) were somewhat lower, with scatter very limited up to quite large strains. These results are much better and more homogeneous compared to series (B), which emphasizes the importance of the manner of formwork filling. However, even in case of perfect formwork filling it can be concluded that a certain safety margin must be provided to fulfill the requirements for a certain class, even in cases where the specimens are produced under construction site conditions.

Keywords UHPFRC, tensile test, flexural test, strain hardening, fiber orientation, reducing factor

161

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Effect of Fibers on the Flexural Behaviour of Beams Built with High-Strength Concrete and High-Strength Reinforcement Roukaya Bastami1, Yang Li1, Omar Algassem1 and Hassan Aoude1

1

: University of Ottawa, Ottawa, Canada.

Research description This paper summarizes the findings from a research project examining the performance enhancements that can be achieved by using fibers in high-strength concrete (HSC) beams built with normal-strength and high-strength reinforcement. As part of the study, a total of six HSC beams, built with and without fibers, and reinforced with normal-strength and high-strength bars were tested under four-point bending until failure; the properties of the specimens are summarized in Table 1. Beams had 125 mm x 250 mm cross-sections, a length of 2440 mm and were reinforced with either 2-15M Canadian size normal-strength (Grade 400 MPa) bars or 2-No.5 American size high-strength ASTM A1035 (Grade 690 MPa). Transverse reinforcement in four of the six beams consisted of U-shaped stirrups made from 6.3 mm wire, spaced at 100 mm in the shear spans (HSFRC beams in series 1 were detailed without stirrups). HSFRC beams were made from HSC reinforced with either 30 mm hooked-end steel fibers (ZP) at a volumetric ratio (V f) of 1% (78 kg/m3), or 50 mm long macro-synthetic fibers (S1) at Vf = 0.75% (6.75 kg/m3). The compressive strengths of the various concretes are reported in Table 1. Sample properties of concrete in compression and steel in tension are shown in Figure 1a-b. Beam load-deflection responses are shown in Figure 1c.

Results 1.1 Effects of fibers in beams with Grade 400 MPa reinforcement Comparisons between the responses of beams HSC-0%-15M, HSC-1%ZP-15M and HSC0.75%S1-15M, constructed with plain HSC and HSFRC containing steel and synthetic fibers, respectively, highlights the effects of fibers on the flexural response of high-strength concrete beams built with conventional longitudinal steel reinforcement. The results show that use of 1% steel fibres substitutes for transverse reinforcement and improves flexural response of the HSC beams by increasing strength, stiffness and ductility. The use of macro-synthetic fibers allows the beam to reach a larger maximum displacement when compared to the control beam, though it was not sufficient to completely substitute for transverse reinforcement.

1.2 Effects of fibers in beams with Grade 690 MPa reinforcement Comparison between the responses of beams HSC-0%-No.5(HS), HSC-1%ZP- No.5(HS) and HSC-0.75%S1- No.5(HS) allows for a study on the effect of fibers on the behaviour of beams detailed with high-strength bars. As observed in Figure 1c the use of high-strength steel in the plain HSC concrete beams leads to an increase in strength. However, the increase in strength comes at the compromise of ductility due to sudden failure of compression concrete soon after steel yielding. The provision of steel fibers solves this problem and allows the beam to carry loads past yielding, with a more ductile and gradual failure. The improved response 162

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

results from the enhanced compressive strain capacity and toughness of HSFRC when compared to plain high-strength concrete. The use of synthetic fibers in the beams with highstrength bars allows for an increase in load–carrying capacity and stiffness when compared to the plain HSC companion, although the fibers are not able to fully mitigate against the highstrain demands imposed by the high-strength bars on the concrete in the compression zone.

1.3 Summary In summary, the results show clear benefits associated with the use of fibers in high-strength concrete (HSC) beams. In the case of HSC beams detailed with conventional reinforcement the use of fibers improves flexural ductility (and in the case of steel fibers, can substitute for transverse reinforcement). In the case of beams detailed with high-strength bars, the enhanced toughness and strain capacity of HSFRC allows for a synergy effect which allows for a better utilisation of the capacity of the high-strength bars.

Concrete

Stirrups provided?

Mix

f’c (MPa)

Type

Length (mm)

Vf (%)

Type

Grade (MPa)

Amount & Size

HSC

107

-

-

-

NS

400

2-15M

Yes

HSC-1%ZP-15M

HSFRC

106

ZP

30

1.0

NS

400

2-15M

No

HSC-0.75%S1-15M

HSFRC

71

S1

50

0.75

NS

400

2-15M

No

HSC-0%-No.5(HS)

HSC

95

-

-

-

HS

400

2-No.5

Yes

HSC-1%ZP-No.5(HS)

HSFRC

103

ZP

30

1.0

HS

690

2-No.5

Yes

HSC-0.75%S1-No.5(HS)

HSFRC

93

S1

50

0.75

HS

400

2-No.5

Yes

Beam I.D. HSC-0%-15M

120

400

1400 HSC - 1%ZP HSC - 0.75%S1 HSC-0%

80

40

No.5 - HS steel 15M - NS steel

1200

Stress (MPa)

Stress (MPa)

Longitudinal steel

Fibers

1000 800 600

HSC-0%-15M HSC-1%ZP-15M HSC-0.75%S1-15M HSC-0%-No.5(HS) HSC-1%ZP-No.5(HS) HSC-0.75%S1-No.5(HS)

300

Load (kN)

Series 2

Series 1

Series

Table 1: Properties of specimens tested in the experimental program.

400

200

100

200 0

0 0

0.004

0.008

0.012

0.016

Strain (mm/mm)

a) Concrete (compression)

Figure 1:

0

0

0.03 0.06 0.09 0.12 0.15

0

b) Long. Steel in tension

20

40

60

80

Displacement (mm)

Strain (mm/mm)

b) Beam load-deflection curves

Material properties and Load-deflection beam results.

163

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Mechanical characterization of fiber reinforced floor screeds: Influence of glass fibers on shrinkage and cracking mechanisms Fella Chelha1; Syed Yasir Alam1; Ahmed Zakarya Bendimerad1; Ahmed Loukili1 1

: Ecole Centrale de Nantes, Civil engineering and Mechanics Research Institute (GeM) UMR CNRS 6183, 1 rue de la Noé, 44321 Nantes, France

Abstract Self-compacting mortars and concretes for horizontal structures are cementitious mixtures that are both fluid and homogeneous, with the particularity of flowing under the effect of their own weight. Thanks to their homogeneous texture they offer the possibility of achieving good quality of finishing and many such advantages become the reason for their applications especially in slabs and floors. However, self-compacting mortars or concretes show considerable shrinkage and cracking problems when used in floors and slabs. Because of their large moisture exchange surfaces, the floor screeds are subjected to significant drying effects and in particular plastic shrinkage. If the movements are restrained, the risk of cracking is high. In this respect the use of fibers is a good alternative to using reinforcement bars and welded mesh. Indeed on site a clear decrease in cracking caused mainly by the shrinkage can be observed as soon as the fibers were incorporated in the screed, especially during the heating of the heated floors. This study is conducted with the aim to demonstrate the effectiveness and limitation of glass fibers on the closure of shrinkage cracks by determining their mechanisms of action, both at young age and in the long term. The study is carried out in two parts: Firstly, free shrinkage behavior is analyzed in the fiber reinforced floor screed. Secondly, the restrained behavior at young ages using recently developed uni-axial tensile testing machine and at long term using ring specimens are investigated. The impact of temperature increase (during heating of floors) is also inspected on the effectiveness of fibers to reduce the shrinkage cracks and strains in hardened screed.

Keywords shrinkage; glass fibers; self-compacting mortar; floor screed; cracking

Introduction The aim of this research is to study the influence of small microfibers on the early age behavior of fiber reinforced cementitious floor screeds. The purpose of the experimental program is to show the mechanism by which the glass fibers reduce the microcrack and their effect before the initiation of the macrocrack. To do this the tests were based on the measurement of free plastic shrinkage and tensile strength at young age. In fact, considering the decoupled approach of cracking at young age, these two tests make it possible to highlight the effect of the fibers on the mechanical action (free plastic shrinkage) and the strength (tensile strength and strain capacity before rupture). These tests are then correlated with other

164

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

measurements. A variation of the fiber dosage was made to determine the optimum fiber dosage.

Experimental program Four mixes of self-compacting mortars were tested in the present research with and without fibers. The mortar mix composition remains the same however; the fiber dosage was varied in order to determine the optimal dosage which allows the reduction of the microcracking and thus the elimination of cracking at an early age. The young cracking test was conducted with wind on cracking benches. At 24 hours casting time, we photographed the cracking benches. Reference mix without fibers cracks and the length of the crack corresponds to the length of the notch on the mold. However, the addition of fibers in the mix shows their ability to close the cracks at young age. The results of plastic shrinkage shows that the reference mix has a greater shrinkage compared to mixes with different fiber dosages, indeed we note that the addition of fibers decreases shrinkage. The extent of shrinkage is reduced by about 46% with a dosage of 0.6 kg/m3, however a dosage of 1.8 kg/m3 leads to a decrease of about 60%, then a stability with a dosage of 3 kg/m3 is noted. It is also observed that there is a decrease in the mass loss compared to the reference in particular after the first hours. A dosage of 3kg/m3 fibers causes 34% decrease in mass loss, and therefore drying, compared to the reference. This reduction in mass loss, even if minimal, contributes to the reduction of the plastic shrinkage, indeed, the influence of the fibers on the evaporation was observed from the first hours and which correspond to the compaction of the particles, this puts us assuming that the fibers can trap the flow of water by hindering the consolidation of the solid particles. At young age, crack opening can be related to the strain capacity of the material before failure. A high strain capacity indicates a reduced risk of cracking. In this study, the strain capacity corresponds to the tensile strain before failure and is determined from the direct tensile test at 24 hours. It is found that the strain capacity increases according to the dosage, at a dosage of 3kg/m3 the strain capacity reached 78μm/m, an increase of about 34%, compared to an increase of 8% and 5% for dosages 1.8 and 0.6 kg/m3 respectively. The same observation is made for the tensile strength, little variation compared to the reference was observed for the dosage 0.6 kg/m3 and a slight decrease for a dosage of 1.8 kg/m3. An increase of 0.3 MPa was noted for the dosage of 3 kg/m3 compared to the reference. The effect of fibers on tensile strength is more noticeable at high dosages.

165

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

166

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Service conditions

167

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Evaluation of Effective Moment of Inertia for Calculation of Short-Term Deflections of Steel Fiber Reinforced Concrete Flexural Members Luis B. Fargier-Gabaldon1, Mohamed Al-Tameemi2, Gustavo J. Parra-Montesinos3 1

: Visiting Professor, University of Wisconsin-Madison, USA : Research Assistant, University of Wisconsin-Madison, USA 3 : C.K. Wang Professor of Structural Engineering, University of Wisconsin-Madison, USA. 2

Abstract The effect of discontinued randomly distributed steel fibers on the effective moment of inertia (𝐼𝑒 ) of lightly reinforced flexural members is evaluated through the testing of three pairs of specimens under four-point bending. The specimens consisted of a simply supported, 3660 mm long, 254 mm deep, and 610 mm wide one-way slab strip (Figure 1). All slab specimens contained minimum flexural reinforcement according to the ACI 318-14 Building Code (0.18%). The first pair featured regular concrete (no fibers), while the second and third pairs included steel fibers in a volume fraction (𝑉𝑓 ) of 0.26% and 0.38%, respectively. The fibers were Dramix® 5D 65/60 BG (60 mm long and 0.90 mm in diameter, 𝐿𝑓 /𝐷𝑓 ≈ 65) with a multi hooked-end design and a nominal tensile strength of 2300 MPa. Beyond cracking, a substantial drop in the flexural stiffness was noticed in all specimens. The behavior of the test slabs with and without fibers was similar in terms of crack distribution and crack width up to yielding. The post-cracking and effective flexural rigidity (𝐸𝐼𝑒 , where E is the Young’s Modulus) of the slab specimens with fibers, however, increased by nearly 70 and 20%, respectively, when compared to their counterparts without fibers. These increments can be primarily attributed to the ability of fibers to transfer tension across cracks. The responses of the specimens can be idealized as a trilinear relationship representing three stages of behavior; stage one up to cracking, stage two between cracking and first yield, and stage three after yielding of the longitudinal reinforcement (Figure 2). These idealized relationships were normalized in terms of the applied moment to cracking moment ratio (𝑀𝑎 /𝑀𝑐𝑟 ) versus the corresponding normalized displacement ratio (∆𝑎 /∆𝑐𝑟 ), and compared to the calculated mid-span deflection with the effective moment of inertia given by the equation proposed by Bischoff (2005), 𝐼𝑒 =

𝐼𝑐𝑟 1 − 𝛽𝑐 [1 − 𝐼𝑐𝑟 ⁄𝐼𝑔 ] (

𝑀𝑐𝑟 ) 𝑀𝑎

≤ 𝐼𝑔

where 𝐼𝑐𝑟 and 𝐼𝑔 are the cracked and gross moments of inertia, respectively, and 𝛽𝑐 is the tension stiffening factor. For reinforced concrete and fiber-reinforced polymer flexural members, Bischoff and Scanlon (2007) recommended setting the stiffening factor (𝛽𝑐 ) equal to the ratio 𝑀𝑐𝑟 /𝑀𝑎 (see Figure 2a, for a comparison with slabs with no fibers), while results from this investigation suggest that a stiffening factor equal to one is appropriate when estimating shortterm deflection of slabs made of steel fiber-reinforced concrete in a volume fraction of 0.26% and 0.38% and minimum longitudinal reinforcement, regardless of the magnitude of the applied loads, as can be observed in Figures 2b and 2c, respectively.

168

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Section A-A

610

Reaction Beam

2

3

54

8

Actuator

4  10 mm P

Beam

P

Load A Aspreader

Test Specimen S uppo rt

2 54 1 A A 1220

50

a

50 P

Figure 1. Test set-up and specimen cross section.

P

M /M

1

1220

1220

M a/M cr

cr

2.5

2.5

I

I

g

g

FRC, V =0.26%

2

(Pair 2, both specimens)

Yielding

1.5

1.5

1

1

Yielding

f

2

RC (Pair 1, both specimens)

 =1.0 c

 =M /M c

cr

a

Cracking

Cracking

I

cr

I

cr

0.5

0.5

0

0 0

2

4

6

 / a

8

10

12

0

14

2

4

6

cr

(a) Pair 1 (regular concrete, no fibers, both

8

 a/ cr

10

12

14

(b) Pair 2 (SFRC, 𝑉𝑓 = 0.26%, both specimens).

specimens). M a/M cr 2.5 I

g

FRC, V =0.38%

Yielding

f

2

(Pair 3, both specimens)

1.5  =1.0 c

1 I

Cracking

cr

0.5

0 0

2

4

6

8

 a/ cr

10

12

14

(c) Pair 3 (SFRC, 𝑉𝑓 = 0.38%, both specimens). Figure 2. Idealized moment vs. displacement envelope responses (normalized).

169

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Reduction of Flexural Crack Widths Using Synthetic Fibres in Reinforced Concrete Erik Stefan Bernard1 1

: Technologies in Structural Engineering Pty Ltd, Sydney, Australia.

Keywords Crack width, post-crack performance, flexure, tension stiffening, reinforcement

Abstract The use of steel fibres to reduce the width of flexural and tensile cracks in reinforced concrete members is now well recognised and aspects of this phenomenon have been incorporated into several codes internationally. Almost all the research work upon which this has been based has involved hooked-end steel fibres, with very little work undertaken using other types of fibre. However, the theories underlying how fibres assist in reducing crack widths are not specific to hooked-end steel fibres, so alternative types of fibre could work in a similar way to reduce crack widths in flexural and tensile members. The current paper outlines work recently completed on the effect of several types of synthetic fibre on flexural crack widths in reinforced concrete members. The fibres have predominantly been manufactured using polypropylene, but other materials can also be used for this purpose and have been included in this investigation. Laboratory testing has demonstrated that synthetic fibres are fully capable of reducing flexural crack widths, but their efficacy is not consistent across all fibre types and designs. Testing appears to demonstrate that for a given dosage rate of fibre, the capacity of a FRC mix to limit flexural crack widths is related to the post-crack residual strength. However, some fibres can limit crack widths more effectively than is predicted by existing expressions for design width that are based primarily on post-crack residual strength. This suggests that post-crack toughness is not the only parameter contributing to crack width reduction in Reinforced Concrete members. Flexural crack width estimates have presently been limited to the Serviceability Limit State at which the stress in the reinforcing bars is relatively moderate. This has nominally been taken to be 50 percent of the ultimate load capacity of a member. For ease of comparison, all the fibres examined in this investigation have been tested using an identical RC member configuration and the same nominal mix design for the concrete. While this has facilitated comparison of fibre performance, use of a single conventional reinforcement configuration has meant that identifying the full range of performance characteristics for each type of fibre in this application is not possible. Performance has been compared to predictions according to Model Code 2010 made using post-crack performance measured in ASTM C1609/C1609M beam tests.

170

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

a) Relationship between mean crack spacing and width for all beam specimen sets at 70 kN service load (50% of ultimate load), and b) comparison of measured mean crack width at 70 kN load and predicted design crack width (based on ASTM C1609 beam results).

The experimental component of the investigation consisted of a series of trials in which 3600 mm long beams of 300×300 mm cross-section were cast using either plain concrete (as the control) or the same mixture reinforced with various types of fibre. Conventional steel reinforcement comprising 3×16 mm deformed steel bars of 500 MPa nominal tensile strength and normal ductility were placed with 30 mm clearance from the base. Four replicate RC beams were cast for each trial together with seven ASTM C1609/C1609M beams for post-crack performance assessment. The fibres examined in the investigation included Dramix RC65/60 3D hooked-end steel fibres acting as a SFRC control. Three macro-synthetic fibre types were included, BC48 as an embossed polypropylene (PP) fibre, MQ58 as an embossed polymer blend fibre, and Fibermesh 650 as a smooth tape PP fibre. Fibermesh 150 is a smooth circular microsynthetic PP fibre. Of the remaining fibres, all were circular in cross-section and smooth except for the aramide fibre which had a rectangular cross-section and was partially embossed. Both the glass and basalt fibres were initially bundled but fragmented during mixing to produce dispersed micro fibres that were finely distributed throughout the mixture. Beams were progressively loaded until yielding of the steel bars and crack widths were manually measured after each 1.5 mm increase in deflection using a field microscope The Fibremesh 150 micro-synthetic fibres not only proved very difficult to consolidate but ineffective for reducing crack widths. The PA (Nylon) fibres were less onerous in terms of mixability, but also proved ineffective for crack width reduction. The glass and basalt-based micro fibres were similarly unable to reduce crack widths compared to the plain concrete control. All the macro-synthetic fibres produced a similar reduction in crack width and appeared comparable to the Dramix RC65/60 3D steel fibres at 20 kg/m3 in terms of effectiveness at a moderate dosage rate. The stand-out performer in terms of both crack spacing and crack width reduction was the PVA micro-fibre (Fig. 1), the mean crack width was reduced consistently across all four specimens and was equivalent to a 48% reduction in crack width compared to the plain concrete control for the addition of only 4 kg/m3 of fibre.

171

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Serviceability Limit State Design of SFRC Members Ali Amin1, R. Ian Gilbert2 1

: The University of Sydney, Sydney, Australia. 2 : The University of New South Wales, Sydney, Australia.

Abstract A significant body of research is available on the strength of steel fibre reinforced concrete (SFRC) members subjected to shear and flexure. The behaviour of SFRC under service loads has received less research attention. As the fibres are capable of transmitting tensile stress across a crack, the average tensile strain at a crack in a reinforced concrete member containing fibres is less than that in a similar member without fibres. As a result, the cracking and deformation characteristics of reinforced concrete structures can be significantly improved by adding fibres to the concrete mix. Despite the increased awareness of the benefits of steel fibres by both practitioners and researchers, the use of SFRC in Australia has largely been limited to non-critical members, even though significant potential exists for full or partial replacement of costly, manually placed, steel bar reinforcement in a wide range of applications. SFRC systems have the potential to improve architectural freedom, allowing for non-orthogonal shapes and floor plans, which may otherwise be difficult with conventional reinforcing bar layouts and formwork systems. The relatively slow take up of this material is possibly because standardized procedures for the design of SFRC have not been developed and, as a result, practicing engineers have little design guidance. In addition, some practitioners may be uncomfortable with the idea of relying on randomly placed fibres to provide post-cracking strength in bending or direct tension. Test methods that can correctly, and accurately, establish the constitutive behaviour of the material are required, particularly the post-cracking response in tension. Only when the material constitutive relationships for SFRC is determined with confidence can the material be reliably incorporated into structural elements. A catalyst in promoting this material to structural practitioners, however, is the establishment of physical mechanical models that can accurately explain and predict the material’s response to a variety of load cases. This has been addressed recently for SFRC in the publication of the fib Model Code 2010 and the Australian Bridge Code for Concrete Structures AS5100.5-2017. In order to develop a complete understanding of the performance of SFRC under in-service loads, it is prudent to understand the interaction of SFRC with conventional steel reinforcing bars and its effect on structural behaviour. Fibres not only transmit tensile stress across cracks, they also enhance the interaction and stress transfer from the reinforcing bar to the matrix by inhibiting the propagation of splitting cracks that occur as a result of localized stress points around the ribs of a deformed bar. Consider the instantaneous tensile load-displacement response of a plain concrete tension tie, square in cross section and reinforced with a single concentric reinforcing bar (see Figure 1). Neglecting any effects due to shrinkage, the response of the member prior to cracking is linearelastic. The initial slope of the load-displacement relationship is directly proportional to the area of the transformed section. On first cracking, the stiffness of the tie drops significantly, however, with only one crack, much of the tie is uncracked. With an increase in applied load, more cracks 172

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

develop along the length of the tie and the average stiffness of the member further decreases. If the concrete between the cracks carried no tensile stress, then the response would follow that of the bare bar. The difference between the actual behaviour of the tie shown in Figure 1 and that of the bare bar is referred to as ‘tension stiffening’ and is attributed to the ability of concrete to carry tension between cracks as a result of the bond which is developed between the reinforcing steel and concrete matrix. Tension stiffening increases the member’s rigidity/stiffness prior to the yielding of the steel bar and hence, has the capacity to affect the deformation and cracking characteristics of the member. The tension stiffening behaviour of SFRC is different to that of plain concrete. In plain concrete ties, the concrete carries no stress at the cracks, and can only carry tension between cracks. In SFRC tension ties, the fibres in the concrete matrix can carry tension across the crack and the fibre-concrete also carries tension between the cracks due to the bond between the steel fibres and the concrete and the reinforcing bar and the concrete. This leads to a greater resistance in tension, not just after cracking, but also after yielding of the reinforcing bar (as the fibres can still be effective at this stage).

P Py

SFRC

Pcr

Plain RC

Bare Bar

 Figure 1:

Tension stiffening in SFRC and plain concrete tension ties.

This paper first describes a physically rationale model of the tension stiffening behaviour of SFRC as qualitatively explained above. With this behaviour quantified, expressions suitable for the design of SFRC members are derived for the control of instantaneous deflections and crack widths, followed by a short example.

Keywords steel fibre, concrete, serviceability, deflections, crack widths

173

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Effect Of Fiber Dosage And Matrix Compressive Strength On MSFRC Performance C. Del Prete1, A. Tilocca1, N. Buratti1, C. Mazzotti1 1

DICAM – Structural Engineering, University of Bologna, Italy.

Objectives and experimental programme The addition of fibers to concrete increases considerably its toughness. On the other hand, compressive strength is not improved significantly. Many researches have shown that the overall performance of the material is accountable to a variety of factors concerning fibers and concrete matrix, e.g. the shape and material of fibers, their dosage and the concrete mix design. The fiber dosage is a parameter that sometime is expressed also trough the number of fibers across the crack section because they are strictly involved in the cracking process of material. The present research is aimed at evaluating the influence of fiber dosage and concrete compressive strength on the nominal flexural residual tensile strength. The FRC analyzed contained macro polypropylene fibers. The fibers used have a crippled shape, a length of 39 mm and a diameter of 0.79 mm. Four different dosages were adopted 2, 4, 6 or 8 kg/m 3. To investigate the concrete compressive strength influence, three different concrete mixes were used. They had similar aggregate grading but featured different contents of cement and had different water/cement ratios. The experimental campaign consisted of compressive and flexural tests performed on cubic, 150x150x150 mm3, and notched prismatic specimens, 150x150x600 mm3, respectively. Flexural tests were carried out according to EN 14651. The effects concrete compressive strength and fiber dosage were evaluated by means of a statistical elaboration of all experimental results.

Relevant results / case study The compression tests performed have revealed values of strength ranging from 33 MPa to 55 MPa. Figure 1a shows a comparison of the mean nominal residual flexural tensile strength versus CMOD curves for the specimens of all the batches containing 6 kg/m3 o fibers. They are grouped by concrete mix. Each batch contained 8 prisms. In order to better understand the effect of fibers and avoid systematic errors related to their uneven distribution, in Figure 1a fR3 values are plotted, for all the specimens tested, against the number of fibers crossing the crack, as counted in the portion of the cross section closer to the notch (2/3 of the cross section). A very strong correlation can be observed in this case, furthermore no significant batch-to-batch variation can be noticed.

174

5 4,5 4 3,5 3 2,5 2 1,5 1 0,5 0

B8_L_6 (34 MPa)

B9_M_6 (43 MPa) B10_H_6 (55 MPa) fR3 (M Pa)

σN [M Pa]

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

0

0,5

1

1,5 2 2,5 CM OD [mm]

3

3,5

(a) Figure 1

4

5,0 4,5 4,0 3,5 3,0 2,5 2,0 1,5 1,0 0,5 0,0 0

50 100 N Fiber (A1+A2)

150

(b)

Mean nominal residual flexural tensile strength versus CMOD curves for the specimens containing 6 kg/m3 of fibers (a) and correlation between the number of fibers crossing the crack and fR3 (b). Different colours and symbols correspond to different batches which feature different compressive strengths.

The relationship between the FRC post peak strength and fiber dosage and concrete compressive strength, has been described by means of a linear regression model. The form of the regression model is:

f r ,i = bi + m1,i  ( N fibers , A1+ A2 ) + m2,i  Rcm

[1]

where bi , m1,i , m2,i are regression coefficients and i = 1, … 4. Significance tests were then performed on the regression parameters by computing their p-values. Using a threshold value of 0.05, the test suggests that concrete compressive strength, i.e. the m2 parameter, has no statistical significance for the FRC under consideration.

Conclusion The results of the experimental campaign carried out, allow to draw the following conclusions on the behavior of the MSFRC under consideration: 1. The FRC nominal flexural residual tensile strength is strongly dependent on the number fibers bridging the crack surfaces, and therefore on fiber dosage. 2. The compressive strength of the cementitious matrix has no significant influence on nominal flexural residual tensile strength. It is worth noticing that this conclusion is specific to the fibers and concrete admixtures tested. 3. Increasing fibers dosage, from 2 kg/m3 to 8 kg/m3, the scattering of results rises: this is most likely due to a less uniform distribution of fiber in the specimen (also dependent on specimen size). Furthermore, entrapped air in concrete might increase as the fiber dosage increases, and therefore might affect the mechanical performances of the composite.

175

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Predictions of the Micro-crack Openings for Ultra High Performance Fiber Reinforced Concrete Vicky Turgeon-Mallette1, Luca Sorelli1, David Conciatori1, Vanessa Durand2, Julien Réthoré3 1

Université Laval, Québec, Canada. 2 Ministère des Transports, de la Mobilité durable et de l’Électrification des transports, Québec, Canada. 3 Université de Lyon, Lyon, France.

Abstract Ultra-High Performance Fiber Reinforced Concretes (UHPFRC) are characterized with an outstanding durability compared to normal concrete, allows the formation of multiple microcrack instead of localized large cracks under tensile stress. Microcrack width and spacing can strongly reduce the water flow within a cracked concrete. It is known that permeability is related to the cube of the crack width. Therefore, the permeability of several microcracks is lower than the permeability of a single crack where the cumulated crack width is the same for both cases. A reliable estimation of crack width is then critical for assessing the water transport in damaged concrete. Several methods are available in open literature for estimating the crack widths of fiber reinforced concrete, but their application to UHPFRC is not yet validated. Among existing methods for predicting maximum concrete crack width, we consider the following ones: 1\ Eurocode method for normal concrete. The method considered that when cracking occurs, the tensile stress is completely transfers to the reinforced steel bars. The Eurocode method is considered in this study despite the fact that its specified for normal concrete as a comparison for the other methods; 2\ RILEM TC 162-TDF extends the Eurocode method for SFRC. Unlike the Eurocode’s method, it is considered that part of the tensile stress is taken by the steel fiber reinforced concrete even after cracking therefore reducing the stress in the steel bars; 3\ Moffatt et al. proposed a modified method of the Eurocode to take fiber influence into account after cracking, which is done by considering a residual tensile stress in concrete after cracking; 4\ Deluce et al. proposed a method based on the CEB-FIP 1978 crack spacing formulation and modified to include influence of steel fiber in concrete. The maximum crack width is obtained from the average crack width with the application of a coefficient. The beams tested were made of Ductal® UHPFRC with 2% by volume of steel fibers and steel rebar which correspond to 1.5% steel reinforcement. This UHPFRC exhibits tension hardening behavior in tension with multiple microcrack formation. The experimental 4PBT were carried on un-notched UHPFRC beams. The mid-span deflection was measured by a LVDT, the bottom surface was instrumented with 3 LVDTs in order to measure Crack Mouth Opening Displacement (CMOD) and a strain gauge was located on the steel bar to obtain its deformation. For one beam, the steel stress was also monitored with a strain gauge. The forces 176

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

were applied with a closed loop control system. An inverse analysis of the bending tests allowed estimating the tensile law. The cracks spacing was also visually measured by eye. The DIC pictures of the beams are taken every 10 seconds throughout the bending tests with a total of 250-300 pictures. The crack width was also calculated according to the methods by DIC analysis by assessing the distance from 2 points aside of observed microcracks. The crack number was estimated from DIC analysis by plotting the strain distribution on the bottom surface of the beam as explained in the following. Four different load levels were considered, such as: cracking onset load (M0), the maximum load applied (Mmax) and two intermediary points in between M1 and M2. The scope of this work is to make a preliminary study of the suitability of existing methods to predict the crack width of UHPFRC beam under flexion at different damaged levels. Digital Image Correlation (DIC) technique is employed to observe and measure microcracks in concrete under tensile stress. By comparing the experimental results with considered methods in terms of crack width at different damage, the present work provides a preliminary consideration on the suitability of existing methods to predict the crack widths of steel reinforced UHPFRC under bending at different load level. Figure 1 shows the experimental and theoretical values of maximum and average crack width obtain from all the calculation method and DIC analysis. Theses results shows that at M 0, all methods underestimate the crack width and at Mmax only Deluce’s method gives an accurate prediction of crack width with a slight overestimation. At intermediate load levels M1 and M2, the considered methods from Eurocode, Moffatt and Deluce overestimate the actual crack, while RILEM’s method underestimates the crack width. The average crack width from Deluce’s method is accurate except at M0 where it is greatly underestimated. 1,E-01

1,E-01 Crack opening (mm)

Crack opening (mm)

1,E+00

1,E-02

1,E-03

1,E-04

M0

M1

(a) Figure 1.

M2

Mmax

1,E-02

1,E-03

1,E-04 M0

M1

M2

Mmax

(b)

Comparison of the experimental and theoretical values for (a) maximum crack size and (b) average crack size.

177

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Macrosynthetic fibers for end region crack control Glenda Diaz-Acosta1, G. R. Consolazio1, H. R. Hamilton1 1

: University of Florida, Gainesville, United States.

Abstract In the U.S., precast bridge girders are typically prestressed with bonded pretensioned strands. In many cases, strands are straight and are not deviated, which can result is significant tensile stresses in the concrete in the end region of the section. In the State of Florida, partial strand debonding and added mild steel reinforcement is utilized to control the resulting cracking. At times, however, this reinforcement does not effectively control cracking, resulting in construction delays, potential repairs, additional costs and potential compromise of long-term durability. In addition, Florida utilizes self-consolidating concrete (SCC) in producing precast bridge girders for their highway system. Fiber reinforced concrete (FRC) has been found to significantly improve concrete tensile capacity and help in crack control. Since end region cracking is a serviceability state and not an ultimate strength state, crack control using fibers may provide better results than conventional reinforcement. The objective of this research is to evaluate the effectiveness of steel or macrosynthetic fiber reinforcement at controlling end region cracking while still maintaining SCC fresh properties. To better understand the behavior of FRC girders under service conditions, this investigation contains both analytical and experimental work. FRC mixtures containing macrosynthetic and steel fiber reinforcement at volume fractions ranging between 0.1-0.5% were evaluated in laboratory scale and results are to be used to predict full-scale behavior during and immediately following prestress transfer. This paper covers both completed laboratory work and analytical work currently being performed. The main focus of this paper is the material model calibration and methodology of the analytical work. The results of these analyses will help evaluate the effectiveness of fibers to control end-region cracking and aid in developing guidelines for implementing fiber reinforcement into precast girder construction. The experimental work involved mixture development, evaluation of fresh and residual strength, and full-scale production. Residual strength of FRC was evaluated following test procedures of ASTM C1399 and a modified EN 14651. Experimental results suggest that the EN 14651 provides more repeatable and stable results, especially when working with low volume FRC. This is due to the use of crack width opening to control the loading. This ensures stable and repeatable loading of the beam. In addition this test was selected for material calibration in the analytical work because it provides continuous data collection through the entire duration of the testing. Calibration of the material model that predicts FRC behavior under service conditions was focused on the post-cracking residual strength between first cracking and fR,1 (CMOD=0.5mm) (Figure 1). The analytical study covered in this paper is divided into two phases. The first phase is focused on material calibration. The experimental results from laboratory testing were used to calibrate a material model that accurately predicts FRC behavior under service conditions for fiber volumes in the range of 0.1-0.5%. In the second phase, this calibrated material model is incorporated into the full-scale prestressed concrete girder model. The effectiveness of the

178

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

various fibers at redistributing stresses and controlling end region cracking will be evaluated by comparing stresses between control (no fiber reinforcement) and FRC beams.

LOAD

Area of interest

0.5

Figure 1:

1.5

2.5 CMOD (mm)

3.5

EN 14651 typical curve.

Preliminary results from the material calibration shows good agreement in the pre-cracking behavior, with initial stiffness and cracking load been accurately predicted (Figure 2). Additional calibration work is still necessary in the post-cracking stage. The work conducted to date, however, indicates that the smeared reinforcement technique can be used to simulate service condition behavior of FRC elements. Macrosynthetic fibers CMOD (mm) 0.02

0.04

CMOD (mm)

0.06 3,000

8,000

1,000

0

4,000

0

0.001

0.002

CMOD (in)

(a) Figure 2:

0 0.003

0.02

0.04

0.06 Analytical 12 Experimental

Load (lb)

2,000

Load (kN)

Load (lb)

Analytical 12,000 Experimental

0

2,000

8

1,000

0

4

0

0.001

0.002

Load (kN)

3,000

0

0 0.003

CMOD (in)

(b)

Experimental vs analytical work macrosynthetic (a) and steel (b) fiber.

Although additional work is necessary to improve material model calibration, preliminary work with the full-scale analytical model with conventional concrete properties has been performed for verification. Analysis of vertical displacements, longitudinal stresses and forces generated along the steel reinforcement show good agreement with theoretical values. Visual inspection of crack pattern and crack width measurements agree well with the results from the analytical model for a full-scale FIB cross-section with crack widths in the range of 0.003in. (0.15 mm) to 0.013in. (0.30 mm).

179

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Investigation of fiber effect on the geometrical property of crack and water permeability of cracked concrete Yining DING 1, Dong LI 1, Wei ZENG 1 1

:Dalian University of Technology, Dalian, China.

In real RC structures, the tensile cracks due to drying shrinkage and mechanical loading are almost unavoidable (Nilson et al. 2009). The existence of cracks can significantly increase the penetration of moisture and salts into concrete. Rapoport et al. (2002) and Picadent et al. (2009) studied the influence of steel fiber reinforcement on permeability of HSC, and the results showed that the steel fibers demonstrated positive influence on the impermeability of concrete with crack. However, from the paper review, there is still a lack of a systematic investigation of the fiber effect on crack geometry and permeability of concrete. In this work, the widely used macro steel fibers have been added into the concrete samples. The feedback controlled splitting test was introduced. The crack widths between 50μm and 250μm have been selected for water permeability test. A new developed equipment was introduced to the investigation of permeability. For each FRC mix, ten specimens are tested, and the statistical analyses of the crack tortuosity factor τ and the surface roughness factor Rs are shown in Table 1. Table 1:

Comparison of the crack geometry. Crack tortuosity, τ

Types Average

Standard deviation, σ

NC

0.8995

0.0249

SF25

0.8271

SF35 SF55

Surface roughness, Rs

Coefficient variation, CV(%)

of Average

Standard deviation, σ

Coefficient variation, CV(%)

2.8

1.1534

0.0596

5.2

0.0214

2.6

1.2257

0.0285

2.3

0.8164

0.0543

6.6

1.2776

0.0662

5.2

0.7424

0.0724

9.7

1.5161

0.1632

10.8

of

From Table 1, it can be seen that: •

The standard deviation and the variation coefficient of τ and Rs are relatively small. It means that the τ and Rs can be used to verify the crack geometry.

•

The value of τ decreases obviously with the increasing of fiber dosage. Compared to the NC specimen, the value of τ of the SF55 decreases by about 17%.

•

The value of Rs increases with the increasing of fiber dosage. Compared to the NC specimen, the value of Rs of the SF55increases by about 31%. The surface roughness Rs is introduced to modify the Poiseuille law, see Eq.(1).

180

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

 beff2 k= 12 ( −51 + 45Rs1.5 )

(1) The correlation coefficient R of the test results and the values predicted by the Eq.(1) of all the specimens are illustrated in Table 2. 2

Table 2:

Results predicted by the recommended method. τ

Rs

R2

NC

0.8995

1.1534

0.974

SF25

0.8271

1.2257

0.962

SF35

0.8164

1.2776

0.917

SF55

0.7424

1.5161

0.837

Types

From Table 2, it can be seen that all the values of the correlation coefficient R2 of all the specimen are larger than 0.82. It means that the modified Poiseuille law can be used to predict the permeability of cracked concrete. Based on the results of the investigation, the following conclusions can be drawn: 1. The crack tortuosity factor τ and surface roughness factor Rs can be adopted to evaluate the crack geometry. 2. The steel fibers can change the crack geometry (tortuosity and roughness) obviously. 3. The relative surface roughness Rr can be replaced by the surface roughness factor Rs to modify the Poiseuille law. 4. The modified Poiseuille law can be used to predict the permeability of the cracked concrete.

References Nilson A, Darwin D, Dolan C (2009) Design of concrete structures. USA: McGraw-Hill. Picandet V, Khelidj A, Bellegou H. (2009) Crack effects on gas and water permeability of concretes. Cement and Concrete Research, 39(6): 537-547. Rapoport J, Aldea C M, Shah S P, et al. (2002) Permeability of cracked steel fiber-reinforced concrete. Journal of Materials in Civil Engineering, 14(4): 355-358.

181

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

182

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

SESSION: Structural applications

183

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Experimental Investigation of Hybrid Concrete Elements with Varying Fiber Reinforcement under Concentrated Load Sven Plückelmann1, Rolf Breitenbücher1 1

: Ruhr University Bochum, Bochum, Germany.

Keywords Steel fiber reinforcement, hybrid reinforcement, splitting stresses, bearing behavior

Abstract In special cases, concrete members are exposed to high locally concentrated loadings. Precast tunnel lining segments represent a typical examples for such loadings as they are subjected to high and concentrated jacking forces of the boring machine during the construction stage. Such concentrated loadings lead to a multi-dimensional stress state beneath the loaded area. Due to the load diffusion, large splitting tensile stresses are generated in the upper regions of the concrete member (i.e. St. Venant disturbance zone) and spread along directions perpendicular to the load. In order to resist these splitting tensile stresses, the state of the art is to reinforce concrete members with transverse steel reinforcement. An alternative approach is to add steel fibers to the concrete matrix. However, regarding economic concerns it is not appropriate to reinforce the entire concrete member with an adequate high amount of steel fibers, rather only those zones where high splitting stresses are expected. Based upon this fact, the main objective of an experimental study conducted at the Ruhr University Bochum was to investigate the load-bearing and fracture behavior of hybrid concrete elements with splitting fiber reinforcement under concentrated load. For this purpose, in a first step, hybrid specimens (15 × 15 × 30 cm³) were produced containing both plain and fiber concretes. The reference specimens consisted exclusively of plain concrete, while the hybrid specimens were partially strengthened with various types of steel fibers only in the St. Venant disturbance zone (i.e. in the upper half of the specimens for the cases investigated here), instead of a full range fiber reinforcement. As the incorporation of fiber reinforcement can alter the distribution of tensile splitting stresses and further the position of the maximum tensile splitting stresses, the thickness of the reinforcement layer was varied in order to determine the optimal configuration of fiber reinforcement. Taking into account the influence of the casting direction on the fiber orientation and consequently on the bearing and fracture behavior, the hybrid specimens were cast either in standing or in lying molds by means of a “wet-on-wet” casting technique. In the latter case, both concretes were simultaneously placed in the mold, however, temporarily separated by a movable interior formwork (Figure 1). These hybrid elements were loaded either concentrically and eccentrically at constant area ratio (i.e. ratio of total area to loaded area) under identical testing conditions as applied for the plain and fully reinforced specimens.

184

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

Photograph and schematic description of a special formwork used for the “wet-onwet” production of hybrid concrete specimens.

The test results showed that under concentric loads the maximum bearing stress of the hybrid specimens produced in standing molds increased progressively with growing thickness of the fiber reinforced concrete layer. In contrast to the plain concrete specimens, the fiber reinforcement led to a remarkable improvement in the post-cracking ductility. Compared to the fully reinforced specimens, the hybrid specimens with a reinforcement thickness of 150 mm exhibited - besides an almost identical bearing capacity - a similar stress-displacement behavior in the post-cracking zone (Figure 2a). This implied that in this case, a full range reinforcement is not necessary and thus not very efficient. For hybrid specimens produced in lying molds, the improvement in terms of bearable concentrated loads as well as post-cracking behavior caused by the incorporation of fiber reinforcement was comparatively inconsiderable due to the unfavorable orientation of fibers (Figure 2b).

Figure 2:

(a) (b) Stress-displacement curves for specimens produced in standing molds under (a) and lying molds (b).

185

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

A holistic calculation and design tool for structural SFRC members Katharina Look 1, Peter Mark 1 1

: Institute of Concrete Structures, Ruhr-University Bochum, Germany

Keywords Design tool, spreadsheet analysis, SFRC, optimisation, yield line theory, cross section design

Abstract An easy adaptable design tool for holistic calculation of structural members made of steel fibre reinforced concrete (SFRC) is developed employing spreadsheet analyses, optimisation methods and iterative analytical routines (Mark, 2003). The tool evaluates sectional forces based on the non-linear yield line theory (Figure 1 (a)) and supports users to design arbitrary cross sections in ultimate and serviceability limit states (Meyerhof, 1962). Additionally to a forward computation of sectional forces for slabs, where fibre classes are commonly preselected by the user, a reversal kind of optimisation that yields the best fibre class regarding applied loads is optional. In order to benefit from the additional bearing capacities of SFRC in the post-cracking domain, non-linear methods are necessary. The routines for bending and shear design as well as for a limitation of crack widths automatically pick the best steel fibre class regarding a chosen crack width by optimisation (Heek et al., 2017). Especially the combination and interconnection of these three aims of design makes this tool practical and enables users to adapt the input to their specific needs. Comparative calculations of SFRC and reinforced concrete (RC) members highlight the advantages of SFRC and quantify potential savings of rebar. The tool is free of any specific code regulations, just bases on the assumptions of plane strains and perfect bond and requires pre-defined uniaxial stress-strain laws, strain boundaries and fundamental design formulas only (TC 162-TDF, 2003), (Gödde et al., 2010). The optimisation for bending design employs reduced gradients. A division of the solution space into subspaces, which are explored for an extremum by iteration, ensures to find the global minimum of the objective function. By variation of initial values the stability of the result is assured. Boundary conditions, material parameters and sectional properties as well as common results like stress and strain distributions for both, RC and SFRC, performance ratios and potentials of improvement are visualised in commented figures (Figure 1 (b)). Contrasted to the calculated reinforcement ratios for conventional RC (0% utilisation) and SFRC with the highest possible fibre class (100% utilisation) they give feedback about potential capacity losses or gains regarding the chosen fibre class. The program is ready to use and user-friendly implemented in spreadsheets which can be downloaded for free. Computation lasts few seconds only.

186

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

Figure 1:

(a) (b) Input data sheet of slab calculation (a); iterative solution of strain and stress distributions and results of bending design (b).

References Gödde, L.; Mark, P. (2015): Numerical simulation of the structural behaviour of SFRC slabs with or without rebar and prestressing, Materials & Structures 48(6), pp. 1689-1701. (DOI 10.1617/s11527-014-0265-z) Gödde, L.; Strack, M.; Mark, P. (2010): Bauteile aus Stahlfaserbeton und stahlfaserverstärktem Stahlbeton – Hilfsmittel für Bemessung und Verformungsab-schätzung nach DAfStb-Richtlinie „Stahlfaserbeton“, Beton- und Stahlbetonbau 105(2), Ernst & Sohn Verlag, Berlin, pp. 78-91. Heek, P.; Ahrens, M. A.; Mark, P. (2017-1): Incremental-iterative model for time-variant analysis of SFRC subjected to flexural fatigue, Materials & Structures 50: 62, (https://doi:10.1617/s11527-016-0928-z). Heek, P.; Look, K.; Minelli, F.; Mark, P.; Plizzari, G. (2017-2): Datenbank für querkraftbeanspruchte Stahlfaserbauteile – Bewertung der Bemessungsansätze nach DAfStbRichtlinie und fib Model Code 2010, Beton- und Stahlbetonbau 112(3), Ernst & Sohn Verlag, Berlin, pp. 144-154. (DOI: 10.1002/best.201600075) Mark, P. (2003): Optimierungsmethoden zur Biegebemessung von Stahlbetonquerschnitten, Beton- und Stahlbetonbau 98(9), Ernst & Sohn Verlag, Berlin, pp. 511-519. Meyerhof, G. G. (1962): Load-carrying capacity of concrete pavements, Journal of the Soil Mechanics and Foundations Division 88(3), pp. 89-116. Rilem Technical Committees: Rilem TC 162-TDF (2003): Test and design methods for steel fibre-reinforced concrete, final recommendation: σ–ε design method. Materials and Structures, Rilem, 36(8), pp. 560–567. (https://doi.org/10.1007/BF02480834) 187

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

FRCcalc - Software for design of fiber reinforced concrete elements according to MC2010 recommendations Tiago Valente1, Joaquim Barros2 and Lúcio Lourenço1 1

: CiviTest, Pesquisa de Novos Materiais para a Engenharia Civil, Lda., V. N. de Famalicão, Portugal. 2 : University of Minho, Guimarães, Portugal.

Objectives The manual application of the design recommendations and guidelines presented in fib Model Code 2010 (MC2010) for the design of fiber reinforced concrete (FRC) structural members can represent a significative effort in the design of complex structures, as FRC design introduces concepts that can be new for some structural designers. A software capable of automating the safety verifications of FRC members according to the design recommendations of MC2010 can represent a very useful tool in the design process of structures with this material that can present economic and technical advantages in several applications.

Software for design of FRC members A new software was developed to assist and automate the safety verifications of FRC members for serviceability (SLS) and ultimate limit state (ULS) conditions. The software, denominated FRCcalc, is guided to the analysis of rectangular cross-sections with or without conventional passive steel reinforcements, submitted to axial, bending and shear forces. A main feature of the software is the possibility to run a comparative analysis between FRC and reinforced concrete (RC) cross-sections, which was implemented to assess the technical and economic attributes provided using fiber reinforcement as a total or partial replacement of conventional reinforcement. FRCcalc allows to perform the following analysis: (i) ultimate flexural capacity of FRC and RC members; (ii) evaluation of the moment vs. curvature relationship of FRC and RC members at ULS; (iii) ultimate shear capacity of FRC and RC members; (iv) evaluation of design crack width of FRC and RC members; (v) determination of moment vs. design crack width relationship of FRC and RC members SLS; (vi) evaluation of stress limitation criteria of FRC and RC members at SLS.

Design examples Two design examples are analyzed regarding the cross-sections of a FRC and RC (i) slab (Figure 1) and (ii) of a beam (Figure 3). The design examples explored the possibility of partial replacement of conventional steel reinforcement by the addition of fibers to concrete, the demonstration of the increased structural performance provided by the fiber reinforcements and where also considered to compare the predictive performance of FRCcalc with the software DOCROS.

188

FRC2018: Fibre Reinforced Concrete: from Design to Structural Applications Joint ACI-fib-RILEM International Workshop

FRC:

RC: C25/30 | d g = 16mm |

h=200

c=25 1

C25/30 | 3b | ( f R1k = 3.0 MPa |

As1

f R 3 k = 2.1MPa ) | d g = 16mm |

c = 0 |

c = 0 (creep factor) |

As1 = 392.699mm

(dimensions in mm)

Figure 1.

Cross-section data of example no. 1. 50

35

45

30

40

25 20 15 10

FRC

5

RC

Moment [kN.m]

Moment [kN.m]

A500NR

2 As1 = 251.327 mm A500NR

b=1000

2

35 30

25 20

15

0

10

FRC

5

RC

0 0

0.1

0.2

0.3

0.4

0

0.1

0.2

Curvature [1/m]

Figure 2.

0.4

0.5

0.6

0.7

0.8

0.9

1

Crack opening [mm]

a) b) FRC and RC cross-section results of example no.1 slab determined by FRCcalc: a) Bending moment vs. curvature relationship at ULS conditions; b) Design crack width vs. resisting bending moment relationship at SLS conditions. As2

FRC/RC:

c=30 2

C25/30 | FRC: f R1k = 2.678MPa | f R 2 k = 2.508MPa

f R 4 k = 2.950 MPa (Toughness class 2.5c) | d g

Asw

| f R 3k = 2.941MPa | = 16mm |  c = 0

2 2 As1 = 314.16mm (4Ø10) A500NR | As 2 = 157.08mm (2Ø10) A500NR

Asw

c=30 1

h=500

0.3

/ s = 502.66mm2 / m(Ø8 / /200mm) A500NR

As1 b=200

(dimensions in mm) 80

90

70

80

60

70

50 40 30

FRC (FRCcalc) FRC (DOCROS) RC (FRCcalc) RC (DOCROS)

20 10 0 0

0.02

0.04

0.06

Curvature [1/m]

Figure 4.

0.08

0.1

Moment [kN.m]

Moment [kN.m]

Figure 34. Cross-section data of example no. 2.

60 50 40 30 20

FRC

10

RC

0 0

0.1

0.2

0.3

0.4

0.5

Crack opening [mm]

a) b) FRC and RC cross-section results of example no.2 beam: a) Bending moment vs. curvature relationship determined by FRCcalc and DOCROS at ULS conditions; b) Design crack width vs. resisting bending moment relationship determined by FRCcalc at SLS conditions

189

Platinum sponsors

Gold sponsors

Silver Sponsors