This action might not be possible to undo. Are you sure you want to continue?
Jianxin Ma1; Jörg Dietz1
Summary Experiences in the development of Ultra High Performance Concrete found a very high flowability in concrete with low water powder ratio. Because of the high viscosity of the cement paste, compaction was necessary. To improve the compaction characteristics of the concrete the idea of adding coarse aggregate was developed. The first tests showed a good workability of the fresh concrete and a good self compacting ability. The interest in reducing costs while improving workability, shrinkage tendency and the modulus of elasticity was leading to the results in this paper. 1 INTRODUCTION
Increasing the concrete strength is always one of the main desires of concrete technology. Since more than 20 years high strength concretes with compressive strength ranging from 50 N/mm² up to 130 N/mm² have been used worldwide in tall buildings and bridges with long spans or buildings in aggressive environments. Building elements made of high strength concrete are usually densely reinforced. The small distance between reinforcing bars may lead to defects in concrete. If high strength concrete is self-compacting, the production of densely reinforced building element from high strength concrete with high homogeneity would be an easy work. Self-compacting concrete is a concrete that flows and compacts only under gravity. It fills the whole mould completely without any defects. The usual self-compacting concretes have a compressive
Dipl.-Ing., Institut für Massivbau und Baustofftechnologie, Universität Leipzig
% . 20 Vol. βp is the volumetric ratio of water to powder material. According to the design method for conventional self-compacting concrete proposed by Okamura et al. In the case of ultra high strength self-compacting concrete with very low water-powder ratio. Concerning the filling effect. and the transition zone between these two phases.2 . silica fume and other fine components. Special attention must be paid to ensure the extremely high flowing ability required in self-compacting concrete.% . at which all voids among solid particles are just filled with water. fine aggregate and paste consisting of powder (<0. paste volume should be increased. An ideal powder mixture should be 34 . .% and 30 Vol. On the other hand. In ultrahigh performance concrete the water to cement ratio is even lower than 0. 18% silica fume (in cement weight) is theoretically enough for the total consumption of calcium hydroxide released from cement hydration . Reducing the water-cement ratio and the addition of pozzolanic admixtures like silica fume are often used to modify the microstructure of the matrix and to optimise the transition zone. 2 CONSIDERATION FOR THE SELF-COMPACTING HIGH STRENGTH CONCRETE It is well known that the properties of concrete are affected by cementitious matrix.LACER No. It can be determined according to the method described in section 4. Another way to increase concrete flowing ability is minimising the voids among particles of the powder mixture composed with cement. the volume of coarse aggregate. In the case of Portland cement. The lower βp is.125 mm) and water are approximate 50 Vol. decreasing water-cement ratio may negatively influence the flowing ability of the fresh concrete.28 and 0.1. aggregate. the lower is the voids volume in the powder mixture. In high performance concrete water to cement ratio ranges usually is between 0.38. 7. Voids among powder particles can be evaluated by the water demand βp. The Reduction of the water-cement ratio results in a decrease in porosity and refinement of capillary pores in matrix. 2002 strengths in the range of 60-100N/mm². In this paper the proportioning and properties of a ultra high performance self-compacting concrete (UHPSCC) with a strength about 150 N/mm² is presented. The effect of silica fume can be explained by its pozzolanic reaction with calcium hydroxide released from cement hydration and filling effect in the voids among cement or other powder material particles . more than 25% silica fume should be added to concrete to get the densest granular mixture .
The use of a superplasticizer on polyethercarboxylate basis ensured the flowing ability of the concrete. so that the voids among coarse particles can be filled with finer particles. A quartz powder with a particle diameter smaller than 10 µm was used as micro filler to optimise the grading curve of the granular mixture composed of 100 g cement. Its particles are in the range of 0.5 R and Portland cement with high sulphate resistance CEM I 42. 1.5R-HS were used in this experiment. As pozzolanic material a uncompressed white silica fume was added in concrete. 3 MATERIALS An ordinary Portland cement CEM I 42. The grading curves of row materials and the mixture are shown in Fig.3-0. 1 grading curves of raw materials and mixture 35 .Ultra High Performance Self Compacting Concrete composed of components with different fineness.8 g quartz powder.5R (z) z+MS+QM Fig. The aggregate mixture is composed of 30 % sand and 70% basalt in mass. 100% 90% 80% cumulated passing 70% 60% 50% 40% 30% 20% 10% 0% 0. 30 g silica fume and 42.1 to 1 µm.1 1 diameter (µm) 10 100 silica fume (MS) Quartz powder(QM) CEM I 42.8 mm and crushed basalt in the range of 2-5 mm were used as aggregates. Quartz sand with a size of 0. At high dosage it exhibits a strong retarding effect on the cement hydration.
1 EXPERIMENTS AND RESULTS Experiment on cementitious paste Flow tests were carried out on pastes containing different water to powder ratios or different superplasticizer dosages with a flow cone as in conventional selfcompacting concrete. For all latter experiments the superplasticizer dosage was determined about 2.6 0.0% in powder mass. micro silica and quartz powder.0 0 2 4 6 8 10 12 14 relative slump flow Vw/Vp ßp= 0. 0. 2 relationship between relative slump flow and Vw/Vp 36 . The intercept point with the ordinate is βp. 2002 4 4.2 0. The βp with a low value of 0. The relationship between relative slump flow and volumetric water to powder ratio Vw/Vp is shown in Fig.319 Fig.319 indicates a high packing density of the granular mixture composed from cement.1 0. The relative slump flow was calculated from measured slump flow as flowing: d Γm = d −1 0 2 (1) where d is the measured flow diameter and d0 is the flow cone diameter. The results showed that a superplasticizer dosage of more than 2% of the powder mass could not improve the flowing ability of the paste anymore. 7.LACER No.4 0.3 0.5 0.7 0. 2 (superplasticizer dosage was 2% in powder mass).
Before each measurement the concrete was mixed again for about 90 sec. with low water-binder ratio and the addition of silica fume often shows a rapid concrete setting . such as type and dosage of superplasticizer. and mineral admixtures. SCC must keep its high flowing ability for about 90 min to ensure casting at building place. fresh concrete temperature. This can be explained by the strong retarding effect of the superplasticizer on the cement hydration.2 show that the properties of the concrete have not changed significant during 2 hours after mixing. Results in Tab.1 the paste volume in concretes lies between 46% and 55%. During flowing neither segregation nor bleeding occurred. cement type and fineness. Experiences in HPC show that ordinary Portland cement.2. As shown in Tab. even though the density of crushed basalt is high. The distance between the steel bars of this block-ring was 16 mm.Ultra High Performance Self Compacting Concrete 4. The low water-powder ratio and the high fineness of silica fume and quartz powder formed a paste with high viscosity. water and superplasticizer have been added and the mixture was mixed about 6 min till a flowing and homogenous concrete was formed. 37 . water-powder ratio. It was observed that coarse aggregate was flowing together with viscous mortar through the obstacles of the block-ring without any difficulty. The evolution of fresh concrete properties was investigated on M6 with the obstacle-ring. after mixing. This purpose can’t be reached by the variation of the superplasticizer dosage. Blocking tendency of coarse aggregate was investigated with M6. The consistency loss of concrete can be influenced by some factors. During flowing coarse aggregate should not be blocked by reinforcing bars. to reduce the air content to a minimum level. but through increase of the paste volume in concrete. Slump flow and blocking tendency of coarse aggregate were measured every 30 min after mixing. A slight decrease in water to powder ratio leads to a significant increase in paste volume. 4.1 Properties of fresh concrete The flowing ability of fresh concrete is described with slump flow investigated with Abrams cone. Compared with conventional self-compacting concrete the slump flow of UHPSCC should be more than 700 mm instead of 650 mm. 2 min. Except high flowing ability a homogenous spreading of the concrete through narrow spaces between reinforcing bars is also of great importance.2 Experiment on concrete All concretes have been mixed in an intensive mixer with the following procedure: All dry materials were mixed till a homogenous mixture was built up. as in conventional self-compacting concrete.
LACER No.6 710 8 2590 160.7 155.543 46.65 Crushed basalt (2-5 mm) 3.431 54.8 0.21 0. Tab.1 Total water superplasticizer w/b=w/(z+MS) w/(z+MS+QM) Volumetric water to powder ratio Paste volume in concrete Slump flow (mm) t500 (sec.5% M2 609 183 261 398 936 163 21.05 CEM I 42.4% 162 231 440 1027 163 16.7 8.cyl.0% 540 97.21 0.155 0.1 0.63 Quartz sand (0. 38 . Elkem 2.100*300 Splitting tensile fsp.8 mm) 2.) Density of hardened concrete (kg/m³) Compressive fc.195 0.3 Grade 983 Quartz powder (0-10 µm) 2.9% M3 540 162 231 440 1027 163 16.7 710 9 2568 166.8 155.2 ----- 730 --2564 --150. 7.8 0.5R 3.2 296 443 1034 163 16. RH=78% in environment) Time after mixing Flowing value (mm) Flowing time t500 (sec.175 0.5R-HS Micro silica.7 °C.cube100*100 strength (N/mm²) fc. that the temperature in concrete did not increase during the first 24 hours after mixing.9 8.23 0.4 8. 2 Evolution of flowing ability of M6 (T=18.) Hight difference 1) 5 min 710 8 0 35 min 700 8 0 65 min 695 9 0 95 min 690 8 0 125 min 675 9 0 1) Height difference of concrete in or outside the block ring. 1 Composition and properties of UHPSCC Materials Density (g/cm³) z MS QM QS w SP M1 665 200 285 1019 178 23 0.4% M6 480 M4 M5 CEM I 42.cyl.175 0.100*300 540 144 205 475 1108 162 14.5 700 --2547 --151.9 0.2 151.23 0.6 148.487 50.431 61.175 0.0 ----- 48110 50118 52837 53944 Tab.155 0. Cement hydration was completely stopped during this time.8 0.100*200 strength (N/mm²) E-Modul (N/mm²) fc.487 50. 2002 Measurements of the hydration of a 150*150*700 mm beam showed.4% Vw/Vp 690 --2320 28d 28d 28d 28d 149.cyl.7 --- 700 11 2525 160.487 50.26 0.26 0.3-0.
M2. a plastic foil had been inserted in the form before casting. After casting the concrete was covered immediately to avoid any loss of water. The change in length of the concrete specimens was recorded every 15 min.1 are about 20% higher than experimental values. The value of k should be reduced to 1. 4. But deformation curves show a similar characteristic. M5 incommensurable. Then they were cured in water at 20°C till 3 days before testing.Ultra High Performance Self Compacting Concrete 4.1. the matrix and their ratio.2 for concrete containing basalt coarse aggregate Results calculated from this equation using cylinder compressive strength in Tab. After the dormant period during the first 24 hours. This can be explained with the lower coarse aggregate volume in UHPSCC than in conventional concrete.1 can be seen that the difference between compressive strength determined on cubes and cylinders is not significant.3 Autogenous shrinkage Autogenous shrinkage was measured with M1. splitting tensile strength and modulus of elasticity are shown in Tab. This phenomenon can be explained by the high powder content and the small size of the coarse aggregate . The specimens were demoulded 2 days after casting. and M5. which is 39 . M4. M3 and M4. CEB suggested an equation predicting the modulus of elasticity of HPC with the following equation: Ecm = E with c0 f ⋅ k ⋅ cm 10 13 (2) Modulus of elasticity of concrete Ecm: fcm: compressive strength of concrete Ec0: 20500 N/mm² k=1. The total deformation of all mixtures is shown in Fig. The modulus of elasticity of the concrete depends on the properties of coarse aggregate.2 Properties of hardened concrete The compressive strength. M2. M3. In order to prevent the friction between concrete and the mould.2. In Tab.0 for this UHPSCC. This indicates that the compressive strength of UHPSCC is not so strongly depending on the slenderness of the test specimen as conventional high strength concrete. This is the same case as in conventional self-compacting concrete. Concrete was cast into a metallic mould with the dimension of 50*100*1000 mm. Variations in the room temperature during the experiment make M1. 3.
At the same time the chemical shrinkage continued. Contraction caused by cooling and chemical shrinkage outbalanced. This contraction was the summation of two contrary deformations: chemical shrinkage caused by cement hydration and thermal expansion caused by temperature increasing in concrete. 2002 caused by the retarding effect of superplasticizer on cement hydration. 3 free length changes in sealed concrete beams 40 .LACER No. 0 -100 -200 shrinkage (µm/m) -300 -400 -500 -600 -700 -800 -900 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 concrete age (days) M1 M3 M2 M5 M4 dimension of specimen: 50*100*1000 mm Fig. concretes began to shrink. 7. Afterwards the temperature in the concrete decreased. A few hours later thermal expansion was dominant. The concrete shortened again. which resulted in a slight expansion.
As expected. all concretes show a similar development in autogenous shrinkage. but the higher paste volume in M1 leads to a much higher autogenous shrinkage than in M2. These concretes show a very 41 . M4 with higher silica fume content has a higher autogenous shrinkage than M5. It is noticeable that the great part of difference in autogenous shrinkage among these concretes occurred during the first 14 days after casting. 4 Autogenous shrinkage of concretes In Fig. 5 CONCLUSION The application of new superplastizicer and powders in high performance concrete give the opportunity to produce self-compacting concretes that easily reach a compressive strength of more than 150 MPa. The autogenous shrinkages in all tested concretes are much higher than in conventional high strength concrete. Afterwards. M3 with higher waterpowder ratio shows lower autogenous shrinkage values than M2. Except of these both factors paste volume in concrete is also a main factor influencing autogenous shrinkage.Ultra High Performance Self Compacting Concrete 0 -50 autogenous shrinkage (µm/m) -100 -150 -200 -250 -300 M4 -350 -400 -450 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 concrete age (days) M1 M2 M5 M3 dimension of specimen: 50*100*1000 mm Fig. Influences of water to binder ratio and silica fume content on autogenous shrinkage have the same tendency as in conventional high strength concrete. M1 and M2 had the same water-cement ratio and silica fume content. 4 the autogenous shrinkage was calculated form the time when concrete began to shrink again.
7.1  Gert König.1. E. Fehling und B.1501-1511. 1995  H. Bauhaus-Universität Weimar Schriftenreihe der Zementindustrie. Klaus Holschemacher und Frank Dehn. pp. KG  Vagelis G. Bornermann. Nguyen Viet Tue. Woermann GmbH & Co.LACER No.79-86  Thorsten Reschke Der Einfluss der Granulometrie der Feinstoffe auf die Gefügeentwicklung und die Festigkeit von Beton Dissertation. Herstellung und Anwendung Ernst & Sohn. Okamura. pp. 2001  Dirk Weiße Einfluss des Betonsalters auf die Festigkeits-. Heft 62/2000  Pierre Richard. Bauwerk Verlag GmbH. Future interests will try to reduce autogenous shrinkage to minimize the crack tendency of the young concrete. Vol. 2000. herausgegeben von Gert König. 2002 good workability in the fresh state. Papadakis: Experimental investigation and theoretical modeling of silica fume activity in concrete Cement and Concrete Research 29 (1999). Further investigations to optimize the mix proportions will aim to the application of this concrete on site. Schmidt.25. M. Ernst & Sohn  Ronald König. Middendorf Ultra-Hocheistungsbeton UHPC—Herstellung. 7. Formänderungs. K.und Verbundkennwerte Selbstverdichtender Beton. 2001 42 . 6 REFERENCE  R.und Stahlbetonbau 96. Marcel Cheyrezy: Composition of Reactive Powder Concrete Cement and Concrete Research. No. Jörg Peter Wagner Mikrosilika. Ouchi Self-compacting concrete Structural Concrete. Eigenschaften und Anwendungsmöglichkeiten Beton. also the hardened concrete shows excellent quality. No. Martin Zink Hochleistungsbeton---Bemessung. M. Heft 7. Ozawa. 2001.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.