INVESTIGATIONS ON STEEL FIBRE REINFORCED CONCRETE PAVEMENTS

Abstract

Fibre-reinforced concrete may be defined as concrete containing hydraulic cement, water, fine or fine and coarse aggregates and discontinuous discrete fibres. It may also contain pozzolans, superplasticisers and other admixtures commonly used with conventional concrete. Steel is the principal type of fibre and is most commonly used. Other types of fibres such as carbon, nylon, polypropylene, asbestos and alkali resistant glass are also used for specific applications. The incorporation of steel fibres in concrete increases its resistance to tensile stresses and modifies its brittle behaviour depending upon the fibre concentration orientation and distribution. The physical characteristics of the fibres affecting the concrete behaviour are the aspect ratio, shape and volume fraction. The incorporation of steel fibres in concrete is a relatively simple process which can be done manually or by fibre dispensers. The fibres can be hand cut or can be produced commercially with fibre cutting machines easily available in India. Plain concrete contains voids of various sizes and shapes in the matrix and numerous microcracks at the interface rone between the matrix and the coarse aggregates. At higher stress levels, the micro-cracks merge and propogate. Their number and size also increases with Increasing stress levels. The cracks in the matrix join up with the microcracks in the transition cone between the matrix and the coarse aggregate leading to initiation (ii) of the failure process. The initiation and growth of every new crack results in reduction of the available load carrying area which causes a concentration of stress at the crack tips. Once a few cracks are formed, the frequency of crack arrest is decreased. The plain concrete specimen breaks immediately after initiation of the first crack. Unlike plain concrete, fibre reinforced concrete does not break immediately after initiation of the first crack. The composite matrix can carry further load after the first cracking of the matrix. The failure of the composite matrix takes place due to fibre pullout or debonding. If the pullout resistance of the fibres at the cracked section is greater than the load at first crack the matrix does not resist any tension and the fibres carry the entire load taken by the composite. With an increasing load on the composite, the fibres will tend to transfer the additional stress to the matrix through bond stresses. If these bond stresses do not exceed the critical bond strength, there will be additional cracking in the matrix. This process of multiple cracking continues until either the fibres fail or the accumulated local debonding leads to fibre pullout. Fibre reinforced concrete, therefore, continues to sustain considerable loads even at deformations considerably in excess of the fracture deformation of plain concrete. The composite matrix is, therefore, characterised as a material having crack arrest mechanism due to presence of fibres and exhibits superior post (iii) cracking ductility and significantly higher load carrying capacity with substantially reduced cracking in terms of sise and number of cracks. The properties, characteristics and techniques of fibre reinforced concrete have been investigated and established by several researchers. Until now, the applications of concrete reinforced with steel fibres have been for pavements subjected to heavy traffic and for repair work in U.S.A., Canada, Australia and some European countries. The studies reported todate have mostly dealt with the material characteristics such as mix design parameters; workability, fibre loading, shape, size, amount of fibres and aspect ratio and the structural properties such as compressive strength, flexural strength, ultimate strength, ductility and shear strength. The advantages of SFRC* are now well recognised and traditional applications in pavements, where SFRC was initially used in heavy duty pavements, has now gained wide acceptance due to low cost and improved performance of thinner sections. Till recently, scanty information was available mostly dealing with the performance of certain experimental sections with respect to Physical appearance of cracks observed in test tracks under certain traffic conditions. Almost, no information is available on the structural _!^1:^!__°!__^!m!n^_^nder varyln« loading conditions. The *Steel Fibre Reinforced Concrete (iv) first crack and ultimate load carrying capacity of SFRC pavements and overlays needs to be investigated in detail. Also, there does not exist an appropriate method of thickness design for steel fibre reinforced concrete pavements having varying fibre content. The possibility of using the existing rigid pavement design methods for for SFRC pavements needs to be examined. Most of the laboratory studies on steel fibre reinforced concrete have reported that the composite matrix does not fail suddenly once the deflection corresponding to the ultimate flexural strength is exceeded but it continues to sustain considerable loads even at deflections considerably in excess of the fracture deflection of the plain concrete. The experimental investigation of load deflection characteristics is useful in the study of structural behaviour of SFRC pavements. In present experimental investigation, performance of SFRC Pavements of 1.8 x 1.8m and 1.8 x 3.6m size has been evaluated under the interior, edge and corner loading conditions. The experimental programme was planned to study the following: (a) Extent of stresses in steel fibre reinforced concrete pavements (volume fraction 0.5% to 2.0%, aspect ratio of 80) under different loading conditions. (b) Load deflection behaviour of SFRC pavements under interior, corner and edge loading conditions. (v) (c) Breaking load for SFRC pavements and the optimum percentage of fibres. (d) Structural behaviour of thin SFRC overlay over failed SFRC pavements. (e) Performance evaluation of ferro-fibro* overlay over SFRC pavement. (g) Extent of cracking, crack patterns and crack width in SFRC pavement slabs under varying loading conditions. The PCC pavement sections tested in this study were laid in a nominal thickness of 100 mm keeping in view the cracking/failure which could be observed under the available loading device. Thin SFRC pavement sections having 60 mm thickness were laid over subgrade having modulus of subgrade reaction 0.0302 N/cumm and 0.2414 N/cumm. The fibre volume in 60 mm thick sections were kept to be 0.5% to study the behaviour of the thinnest section having the least fibre volume which may be used. SFRC pavement sections having fibre volume between 0.5% and 2% were laid in 100 mm thickness and the fibre percentages were varied between 0.5% to 2% by volume of the mix. These pavements were laid over the compacted subgrade having modulus of subgrade reaction ks varying between 0.0414 N/cumm to 0.092 N/cumm. SFRC overlay was laid in 40 mm thickness over 60 mm SFRC pavement and the fibre volume was kept as 0.5% in both SFRC overlay and the pavement. The ferro-fibro overlay was laid in 20 mm thickness over 60 mm SFRC pavement. The fibre volume in *Ferro-cement with fibre reinforcement (vi) ferro-fibro overlay and the underlying SFRC pavement was kept as 0.5%. The experimental results obtained from test pavements and overlays have been compared with various theoretical approaches for rigid pavement design i.e.; Westergaard, IRC, Pickett and Meyerhof ultimate load theory. A finite element software incorporating numerically integrated finite and infinite elements has been employed to evaluate the stresses and deflections in pavements and overlays under central loading condition. The results of the experimental investigation show that suitably designed SFRC mixes having fibre percentage upto 2.0% by volume, can be conveniently mixed, placed and vibrated by conventional techniques normally used in pavement construction. Fibre balling which is normally a problem for SFRC mixes having low consistency may be prevented by using mixes with a slightly higher workability. The loss in strength due to a slightly higher water cement ratio is compensated by better compaction of such mixes. As is well known, a marked reduction in workability of SFRC mixed was noticed at fibre volumes between 1.5% and 2%. Almost negligible balling or nesting of fibres was observed at fibre concentration of 2% due to proper selection of mix ingredients and water cement ratio. The experience of laying SFRC pavements has indicated that inspite of the fact that SFRC mixes possess very low workability, the ability to place and compact these (vii) mixes under vibration is quite good. The laboratory investigation of SFRC mixes has indicated that addition of steel fibres to concrete does not significantly improve the compressive strength, Youngs' Modulus and Poisson's ratio while there is a considerable increase in the flexural strength with fibre addition upto 1.5% by volume. The maximum increase was observed at a fibre volume of 1.25%. The results of investigation of SFRC pavements have shown that the performance of concrete pavements can be substantially improved by introduction of steel fibres in percentages between 0.5-2% by volume of concrete. The composite matrices exhibit significantly lower deflection levels with a smaller pavement thickness. The pavement sections fail at loads much higher than those normally encountered in highway pavements. The composite matrix offers the possibility of laying the pavements directly over well compacted subgrade. Due to significant increase in flexural strength and the appreciable capacity of the matrix to bear stresses much beyond the flexural strength, the composite matrix can be advantageously employed in heavy duty pavements and as thin overlay sections where these pavements require strengthening due to increase in traffic and thinner overlay sections are required to be laid from the point of view of geometries. SFRC pavement having an optimum fibre volume of 1.25% improved the load carrying capacity by 17.60%, 60% and 29.2% (viii) under central, edge and corner loading conditions respectively at a stress level equal to the flexural strength of the composite, over that of plain cement concrete pavement having the same mix specifications. The maximum deflections were reduced to an extent of 85.4% under an interior load of 100 kN, 63% under an edge load of 90 kN and 81.5 % under 49 kN corner load. These deflections were appreciably lower than those permissible for rigid highway pavements. Full scale 100 mm thick SFRC pavement slab with optimum fibre content 1.25% by volume and laid over roller oompacted subgrade exhibited remarkable improvements in load carrying capacity. The pavement did not show any signs of cracking at 200 kN load under central loading whereas the corresponding plain cement concrete pavement failed at 55% of the load at first crack for the SFRC pavement. The SFRC pavement showed 150% and 300% increase in load carrying capacity under the edge and corner loading conditions respectively. The maximum deflections in SFRC pavements were found to be significantly lower at maximum load of 200 kN as compared to plain cement concrete pavements under all the three loading conditions. As a result of excellent behaviour of SFRC pavements under static plate load tests in various loading conditions, a thin ferro-fibro overlay having 20 mm thickness containing 0.5% fibre volume bonded with a thin SFRC pavement having 60 mm thickness (ix) and 0.5% fibre volume laid directly over subgrade has also been tested under static plate load test. The thin ferro-fibro overlay exhibited exceptional performance in load carrying capacity, resistance to crack formation and crack propagation and ability to keep the cracks tightly closed. 20 mm ferro-fibro overlay showed no signs of cracking at 200 kN load under central loading condition with appreciably lower level of maximum deflection. The overlay exhibited a unique performance exhibiting multiple cracking on increasing the load beyond first crack load. Ferro-fibro overlay is found to possess superior crack resistance, greater post cracking ductility and a unique cracking behaviour which is of signifigance from the considerations of damage to the pavement structure and its disintegration in that the composite matrix continues to take load beyond the first crack due to the transfer of additional stress by steel fibres to the matrix through bond stress. There may be additional cracking in the matrix if these bond stresses do not exceed the bond strength. The process of multiple cracking continues until either the fibres fail or the accumulated load debonding leads to fibre pullout. Due to the mechanism of multiple cracking, the crack widths are distributed over the larger surface area of the pavement surface and the crack widths are reduced to exceptionally narrow limits. A review of existing theoretical methods of rigid pavement design made in this study has indicated that the slab models are (x) limited to pavements consisting of two layers can not be employed for the analysis of multiple layered system and The layered system models have their limitations in evaluating the effect of pavement features such as edge and corner loading conditions. The Finite Element Method of analysis also, can be used to determine the stressed and deflections in rigid pavements of finite size and the solution is obtained by truncating the domain of analysis at large but finite distance from the point of load application. The infinite element method of analysis used in the present study has the major advantage that the element of uncertainity in respect of the extent of domain and the accuracy of results is reduced. The method has been found to be economical in terms of computational time due to reduction in the number of finite elements and the necessity of refining the mesh for want of accuracy. A review of various finite element models i.e. plain strain models, axisymmetric models, three dimensional models has been presented in the study. The coupling of finite element model with infinite and other models has been discussed and the steps in the infinite element analysis using coordinate ascent formulation have been presented. The analysis of stresses, strains and deflections of PCC pavements, SFRC pavements, SFRC and ferro-fibro overlays has been carried out using the material characteristics in the pavement layers. The results of experimental study have been compared with those obtained from existing theoretical methods and Infinite element method of analysis. SFRC pavements laid in a nominal (xi) thickness of 100 mm with fibre volumes varying between 0.5% - 2% have demonstrated a far superior load deformation characteristics, a remarkable improvement in load carrying capacity at first crack and ultimate stages. SFRC pavements have been found to possess a unique crack arrest behaviour such that a factor of safety more than 2 is available in 100 mm thick SFRC pavements having fibre volume 0.5% or more. The full scale SFRC pavement having optimum fibre volume 1.25% exhibited excellent performance. The pavement could take 200 kN load under central loading condition without any cracking. Under edge loading condition also, only a micro crack was observed at 200 kN load. The pavement could take upto 120 kN load under corner loading condition. The major findings of the investigation on steel fibre reinforced concrete pavements and overlay, thin SFRC pavements and ferro-fibro overlay are that, the steel fibre reinforced concrete offers itself as a new composite material which has a scope of its wide scale use in rigid pavement construction. The research results reveal the applicability of SFRC composite matrix in overlay construction in view of significantly higher load carrying capacity of SFRC overlays laid in a small thickness of 40 mm using a nominal fibre content of 0.5% by volume. The structural behaviour of thin ferro-fibro overlay having 20 mm thickness under central, edge and corner loading has revealed very interesting and useful results on the basis of (xii) which the composite material known as ferro-fibrocrete may find its place in overlay constructions where overlays are required to be provided for taking heavy loads. The superior crack arrest properties and unique crack formation in case of ferro-fibro overlay is of signifigance in view of extremely low crack widths observed at heavy loads of the order of 5 times the design load from the point of view of reduced maintenance requirements and longer life expectancy and at the same time better serviceability due to substantially low extent of cracking observed in such a composite overlay. The outcome of the study is as a result of laboratory investigation of pavements and overlays and the firm recommendations regarding use of SFRC and ferro-fibrocrete as pavement and overlay material can be made after the field performance of such constructions are observed under actual traffic, subgrade and environmental conditions for a considerable length of time. The considerably higher load carrying capacity and substantially lower crack widths and extent of cracking certainly indicates the longer life expectancy of these pavements. This however, has been estimated from the relative life expectancy analysed on the basis of flexural strength which has indicated a relative life expectancy between 2 to 7 times that of the PCC pavements for a constant thickness. The study has also indicated that for a constant life, the thickness may be reduced between 70 - 80% of that of a PCC pavement. (xiii) From the point of view of practical applications of the composite matrices in pavement and overlay constructions, an important conclusion has been derived as a result of this study that the composite matrix can be laid conveniently using conventional mixing, placing, compaction and finishing methods without a need for either special equipment or specialised skilled manpower for the construction. Also, the fibres can be either hand cut or machine cut at site and can be conveniently available and shall be economical if the same are produced commercially. The composite material using steel fibre and mesh reinforcement therefore offers a scope of large scale implementation in highway projects for pavement construction and in rehabilitation and strengthening of pavements. While stressses and deflections in concrete pavements due to wheel loads are determined using the solution of the plate bending problem, the solution cannot be applied to concrete pavements with discontinuities like joints and cracks. In that case finite element method can be applied to these problems. The finite element models for the analysis of concrete pavements involve two models of subgrade, a Winkler foundation and an elastic foundation. The former is assumed to act as a set of uniformly distributed springs supporting the concrete slabs. The latter is an elastic solid which is mathematically described by the well known Boussinesq equation. Under the present investigation, a finite element software incorporating numerically integrated finite and infinite elements has been (xiv) employed to evaluate stresses and deflections in plain cement concrete and steel fibre reinforced concrete pavements. The results computed by the finite element method were compared with the existing theories and the experimental values. It was found that the IFEM* results compared well with the Westergaard analysis. The results can be applied to predict the performance of plain cement concrete and SFRC pavements and overlays by adopting an appropriate load factor. The infinite element formulation is simple and requies nearly half the CPU time as compared to finite element analysis. The results of the present experimental investigation on SFRC pavements and overlays and ferro-fibro overlays as well as the theoretical analysis are expected to lead to a wide acceptance of the field applications of the composite matrix and the application of design methodology for the analysis of stresses and deflections in composite pavements and overlays using infinite element formulation.