ANALYSIS OF LONG-TERM INFLUENCE OF CHLORIDE AGGRESSIVE ENVIRONMENT ON THE UHPC

Transcription

1 Proceedings of the 6th International Conference on Mechanics and Materials in Design, Editors: J.F. Silva Gomes & S.A. Meguid, P.Delgada/Azores, July 2015 PAPER REF: 5573 ANALYSIS OF LONG-TERM INFLUENCE OF CHLORIDE AGGRESSIVE ENVIRONMENT ON THE UHPC Radka Pernicová 1(*), Daniel Dobiáš 1 1 Klokner Institute, Czech Technical University in Prague, Czech Republic (*) radka.pernicova@klok.cvut.cz ABSTRACT Depth of Chloride penetration of Ultra High Performance Concrete (UHPC) with dispersed reinforcement in long term horizon is determined in this paper. This work deals with the one dimensional chloride diffusion, which can be potentially dangerous particularly for the durability of concrete structures. For the simulation of aggressive environment there was used chloride solution, where the samples had been stored for one year. Measured data were examined in relation to the depth of penetration of chloride ions into the concrete structure. Comparative measurements with normal strength concrete have been done as well. The experimental results showed that UHCP have more improved resistance of chlorides penetration than NSC and also chloride diffusion depth is significantly lower in UHCP. Keywords: UHPC, one dimensional diffusion, long-term, dispersed reinforcement. INTRODUCTION The deterioration of building materials is caused in many cases by water-soluble salts (Dhir, 1990). The most common salts are of chloride, sulphate and nitrate origin. They can come into building structures from several sources. The largest amounts of chloride come from using of de-icing salts on roads and pavements during winter season. Chlorides are potentially very dangerous for most porous building materials due to their high crystallization pressures. Chloride capacity belongs to the most important parameters which can indicate the extent of this danger (Luping, 2007). High strength concretes have been developed since the 80s of 20th century, however bigger advancement appeared only during last decade with nanotechnology approach. Ultra high performance concrete denoted as UHPC differs from the common concrete marked as NSC (Normal Strength Concrete) mainly by size of the grain and quantity of aggregate, type of used components and production technology. Designing recipe of the Ultra High Performance Concrete with dispersed reinforcement is a complex problem. The UHPC is a new cementitious composite material which is used for reducing of structure weight and eliminated, partly or entirely, conventional steel reinforcement bars and reinforcing cage (Reda, 1999). Due to the high tensile strength and possibility of applications in large quantities, fibers can effectively bridge the cracks and help with cracked concrete structure to withstand greater loads and ensure that the design will have elastic properties. The fibers provide a non-corrosive solution that is thinner and lighter than traditional approaches and is less demanding on equipment, thereby reducing the cost of labor and materials. The strongest influence on ability of material to transport chloride ions has its specific composition and therefore not only the impact of individual basic materials but also the -775-

2 Track_I Composite and Advanced Materials impact of additives and admixtures (Pernicova, 2007). For those reasons the diffusion of chlorides is considered a tool for estimating durability of UHPC (Graybeal, 2007). MATERIALS AND SAMPLES In this paper two concretes were studied. The first one was ultra-high performance concrete with dispersed reinforcement (UHPC) and the second one was normal strength concrete (NSC) for comparative measurement. Both concretes were prepared in mass of ingredients: a binder (cement, micro silica, including additives) and standardized sand. Composition of individual concrete mixtures is described in Table 1. Water coefficient of individual mixtures was modified according to required workability. Table 1 - Concrete Composition Composition of Mixture [kg] UHPC NSC Cement + micro silica + fly ash Cement + micro silica Aggregate Plastificator Water In designing of UHPC recipe as dispersed reinforcement the fibers with high tensile strength (more than 2200 MPa), made of high carbon steel coated brass, have been used. The fibers have a regular cylindrical shape with a length of 13 mm and a radius of 0.2 mm. Their shape and size allowed the use of large amounts in m 3 without compromising the ability of mixing concrete. Fig.1 Fibers (carbon steel coated brass) Samples were mixed in a blender according to ČSN EN standard. Specimens (150x150x150 mm in dimension) were cut from bigger concrete element removed from the mould after 24 hours and then cured at temperature 20 ± 1 C and kept in an environment with high relative humidity for the period of 28 days

3 Proceedings of the 6th International Conference on Mechanics and Materials in Design, Editors: J.F. Silva Gomes & S.A. Meguid, P.Delgada/Azores, July 2015 MEASUREMENT METHODS AND RESULTS Basic measurements were performed in air-conditioned lab under the temperature 23±1ºC and 30±5% of relative humidity. Bulk density ρ b [kg.m -3 ] and compressive strengths f c [MPa] were determined as fundamental physical material characteristics. The bulk density was determined using water saturation of samples under reduced compression and subsequent weighing of maximum saturated samples by their immersion under water, where the so-called Archimedes weight is determined. Results are shown in Table 2. These results are the average of three and more measured values. Table 2 - Basic Properties of Hardened Concrete UHPC NSC Bulk density [kg/m 3 ] Compression strength [MPa] Investigation of chloride resistance parameters was carried out according to CEN/TS standard. It is a method for determining the unidirectional non-steady state chloride penetration parameters. The testing method allows determination of penetration of chlorides in certain time. The specimen is divided into two sub-specimens, a "profile specimen" that is used to determine the chloride profile after exposure to chloride solution during one year, and an initial chloride sub-specimen that is used to determine the initial chloride level, C i. The obtained C i value is accepted as a zero level of chlorides. The steady-state chloride diffusion coefficient is measured on water saturated samples. The profile specimen is first vacuum saturated with distilled water for 24 hours, then coated on all sides but one, and then the uncoated face is exposed to a chloride exposure solution. After sealing the surfaces, the specimens were placed in saturated calcium hydroxide solution for at least 18 hrs. After storage in saturated calcium hydroxide solution, the specimens were transferred directly to 3% solution of sodium chloride without surface drying. Cl - solution was changed three times during the period (every 90 days) to be sure that concentrations are still 3%. After one year of exposure, several parallel layers of the chloride exposed surface were ground from the dry profile specimen. First six layers were approximately one millimeter thick, second set of six layers has the thickness of 2 mm and all the others were 10 mm. Amount of chloride ions was determined by capillary electrophoresis system. Method is based on leaching ground samples in distilled water. Final values of Cl - were calculated as content area under curve of chloride peak. Fig. 2 - Curve of Cl - Content in Dependence on Depth of Penetrationn -777-

4 Track_I Composite and Advanced Materials Figure 2 shows the decreasing curve of the Cl- content depending on the distance from the exposed surface. Diversity of slope of the line which points the depth of diffusion ions can be observed. Initial increase of the UHPC curve between the first and second layers is probably caused by backward leaching of Cl- ions from the surface layer of the sample to solution of sodium chloride. Final content of Cl - in the sample was converted to weight of binder (cement + admixtures) in 1 m 3 of concrete according to ČSN EN 206 standard. The depth where chloride content first reaches a value between C 1 and C 1 + 0,015 %, is important. This is the last point used in the regression analysis and it is called the "zero point". Table 3 - Content of Cl - on the Weight of Binder in 1m 3 of Concrete Depth [mm] UHPC [%] NSC [%] 0 (Initial) , Depth of 0 mm means initial values measured prior to immersing the sample in the sodium chloride solution. Zero point means depth where values of the content of chlorides reach initial values. This distance should not exceed thickness of the cover layer in order to prevent degradation. The dispersed reinforcement in UHPC would have to be examined in the future from the aspect of degradation of actual threats caused by chlorides

5 Proceedings of the 6th International Conference on Mechanics and Materials in Design, Editors: J.F. Silva Gomes & S.A. Meguid, P.Delgada/Azores, July 2015 Fig. 3 - Content of Cl- depending on the Weight of Binder in 1 m 3 of Concrete in Dependence on Depth from the Surface Chloride content in the first few layers is very similar for both concretes. Values vary within several percent. However, with increasing distance a huge difference begins to appear in the decline of chloride curves and in the place of zero point of UHPC; NSC values are several times higher than its zero point. In the depth of about 8-10 mm m values already approach the zero point, reaching it finally at 12 mm depth. If we compare measured values with NSC concrete where the drop rate is only moderate, we find out that the resistance to penetration of chloride salts is therefore multiplying well. In the place of zero point of UHPC, the NSC values are approximately thirty times higher than its zero point. NSC concrete reached its zero point even in 100 mm of depth. CONCLUSION Chloride resistance of Ultra High Performance Concrete with dispersed reinforcement in long term horizon was determined in this paper. Comparative measurements with normal strength concrete have been done as well. Main attention was focused to determine the penetration of the chloride ions to the structure of materials. Results obtained by capillary electrophoresis method proved that UHPC has a higher resistance to penetration of chloride ions than ordinary concretes and even the depth affected from the surface of exposed area is remarkably lower in UHPC. A distance to achieve the zero point of UHCP (12 mm) is approximately ten times shorter than the distance of NSC (100 mm). Penetration of chloride into the concrete structure varies significantly depending on time of exposure to aggressive environments. In previous work the sample has been exposed to sodium chloride only for 90 days. After one year, the final depth of "zero point" and also the slope of the curve are different from the data measured one year ago (Dobias, 2015)

6 Track_I Composite and Advanced Materials This study shows that there are substantial differences in chloride penetration depending on type of concrete mixture, its additives and admixtures. In the future it is necessary to focus on resistance to another aggressive environment (how does sulphate or carbon dioxide act ) and on dislocation of their zero points. ACKNOWLEDGMENTS This research has been supported by GACR Grant Agency of the Czech Republic No. GACR S. REFERENCES [1]-Dhir R.K, Jones M.R, Ahmed H.E.H, Determination of Total and Soluble Chlorides in Concrete. Cement and Concrete Research, 1990, p [2]-Dobiáš D, Pernicová R. Diffusion of Chloride Ions in Ultra High Performance Concrete. Advanced Materials Research, 2015, Vol. 1106, p [3]-Graybeal, J. Tanesi. Durability of an Ultrahigh-performance Concrete, Journal of Materials in Civil Engineering, Volume 19, Issue 10, October 2007, p [4]-Luping T. Chloride Transport in Concrete Measurement and Prediction, Chalmers University of Technolofy, Goteborg [5]-Pernicová R, Pavlíková M, Černý R. Effect of Metakaolin on Chloride Binding in Limebased Composites, WIT Transactions on Modelling and Simulation, 2007, Volume 46, 2007, p [6]-Reda M.M, Shrive N.G, Gillott J.E. Microstructural Investigation of Innovative UHPC. Cement and Concrete Research, Volume 29, Issue 3, March 1999, p