You are on page 1of 11

 

ECOLOGICAL CONCRETE AND WORKABILITY: A MARRIAGE WITH FUTURE?
S.A.A.M. Fennis*, Delft University of Technology, the Netherlands J.C. Walraven, Delft University of Technology, the Netherlands
36th Conference on OUR WORLD IN CONCRETE & STRUCTURES: 14 - 16 August 2011, Singapore

Article Online Id: 100036007 The online version of this article can be found at: http://cipremier.com/100036007    

This article is brought to you with the support of  Singapore Concrete Institute  www.scinst.org.sg    All Rights reserved for CI‐Premier PTE LTD  You are not Allowed to re‐distribute or re‐sale the article in any format without written approval of  CI‐Premier PTE LTD  Visit Our Website for more information  www.cipremier.com  

which reduces the water demand and increases the strength of concrete. Abstract. the cement content in concrete can be reduced from the prescribed minimum of 260 kg/m3 in ordinary concrete to 110 kg/m3 in ecological concrete. Based on the experience with the cement pastes and mortars. Fennis* and J.a. Walraven† Civil Engineering and Geosciences Delft University of Technology Stevinweg 1. The reduced amount of water and low water/cement ratio can be used to save a certain amount of cement when concrete is designed for a fixed strength class. August 14-16. are used in this design procedure to predict the strength of ecological concrete. in the future it will be very important to use particle packing models to control the water demand and workability of ecological concrete. it is possible to lower the cement content in concrete without changing concrete properties in a negative way.C. the amount of water in concrete mixtures can be reduced. Delft. The high density of the particle structure leaves less space for voids to be filled with water. 2628 CN. particle packing. Twenty cement pastes and more than hundred mortar mixtures were tested to relate the packing density to the workability of mixtures. energy consumption and CO2 emissions of concrete are reduced. A slight increase in water demand in mixtures with a low cement content has a larger effect on the water/cement ratio than in mixtures with a high cement content. it is important to control the water demand of these mixtures.A. Thus. rheology. Optimizing concrete mixtures with regard to cement content is one of the most important factors in the design of ecological concrete.m. webpage: www.tudelft. the particle packing density of the materials should be known. With this method.nl * Keywords: Ecological concrete. To be able to predict how much water is necessary to fill up the voids in between the particles of a concrete mixture and predict the workability.36 Conference on Our World in Concrete & Structures Singapore. This was done following the cyclic design procedure presented in this paper.fennis@tudelft. the Netherlands e-mail: s. In this way.concrete. By minimizing the cement content. The relations between packing density and water demand and between particle structure and the cement spacing. However. 2011 th ECOLOGICAL CONCRETE AND WORKABILITY: A MARRIAGE WITH FUTURE? S.a. _______________________________________ † Delft University of Technology . Experiments have shown that with this 57% reduction of Portland cement. and residual products from other industries are reused. material properties such as strength and modulus of elasticity can remain constant or even improve.M. By making use of particle packing optimization techniques.nl. three ecological concrete mixtures with low cement contents were designed.citg.A. Particle packing density of powders and mixtures was tested by the mixing energy method.

Flowability increases with higher amounts of excess water in the mixture.M. This is done by optimizing the concrete composition in such a way that the highest packing density is achieved. which reduces the water demand and increases the strength of concrete. The optimization results in a stiff and strong particle structure. where the same amount of particles in a stable particle structure (left hand side) is packed closer than in a real mixture (right hand side). In this paper it is presented in what way particle packing density and water demand are related and how this can be used to design concrete with lower cement contents. 2 PARTICLE PACKING DENSITY AND WATER DEMAND The particle packing density of an ecological concrete mixture has an important influence on its water demand. By making use of particle packing optimization techniques. By minimizing the cement content. it is possible to lower the cement content in 1 concrete without changing concrete properties in a negative way . the solid content of the mixture φmix decreases. The definition of particle packing density is the solid volume of particles in a unit volume. are reused. This is shown in Figure 1. In that case. Furthermore. In such a stable structure. a distinction should be made between a stable particle structure and the solid content φmix of a real concrete mixture. less material is dumped as landfill and more natural resources are saved. not only energy consumption and CO 2 emissions of concrete are reduced. In a real concrete mixture part of the water is used to fill the voids between the particles. In a stable particle structure all particles are in contact with each other. Fennis and J.A. about 50% of the total CO2 emitted during the construction of a concrete structure comes from the use of Portland cement. by minimizing material transports or by using recycled materials. Therefore. for instance by optimizing the concrete mixture. the key to design ecological concrete is to control the water demand and workability.S. This excess water provides the flowability of the mixture.A. the particles are packed with a certain packing density αt. The water demand of the particle structure is very important since a slight increase in water demand in mixtures with low cement content results in higher water/cement ratios. optimizing concrete mixtures with regard to cement content is one of the most important factors in the design of ecological concrete.C. which has a positive influence on the mechanical properties such as shrinkage and creep. Walraven 1 INTRODUCTION Ecological concrete can be created in several ways. This packing density is higher than the solid content φmix of a real concrete mixture containing that particle structure. However. like fly ash or incinerator bottom ash. Excess water Stable parƟcle structure Mixture Figure 1: The volume occupied by a stable particle structure (left hand side) compared to the volume occupied by a flowable mixture consisting of that particles (right hand side) . the high density of the particle structure leaves less space for voids to be filled with water. but also residual products from other industries. Therefore. while the rest of the water is regarded as excess water. A higher water/cement ratio decreases strength and durability. Therefore. However.

calculated in four directions.) of the particle structure. with fine particles (< 1 mm). less void water is necessary. for instance. However. Therefore. which is very useful in the design of ecological concrete mixtures. The average amount of water in the mixture at maximum power consumption was taken as the void volume between the particles to calculate the packing density. mixing was continued with a constant water supply of ± 0. The slump flow was taken as the average spread diameter. PAAR Physica MC1. The maximum packing density of dry particles can be determined according to NEN-EN 10973:1998 for loose bulk density. The pastes were mixed in a three-liter Hobart mixer.mixing energy • Proctor test (NEN-EN 13286-2) 5 • Centrifugal consolidation 6 • Water demand – Japan 7 • Rheology – Krieger and Dougherty 7 • (Gas) pressure filtration 8 • Vicat test 9.10 • Shear compactor • … For the experiments presented in this paper. wet.3 ml/s until maximum mixing power consumption was reached in about 2 minutes.002 [-].S. agglomeration of particles. Walraven Increasing the packing density improves the workability of a mixture. the water that first filled up the voids between the particles now becomes available as excess water and provides more flowability. to determine the maximum packing density of wet particles. it is important to measure the maximum packing density of the particles under the same conditions as under which the particles would be used in concrete. The method can be extended to determine the maximum packing density at a certain compaction level. If in this optimized mixture the water dosage would be kept constant.A. To relate the packing density to the workability of mixtures. Since the inter-particle forces depend on the conditions (dry. First. The mortars were mixed in a three-litre Hobart mixer following the mixing procedure of NEN-EN 1961:2005.M. The mixing procedure started by mixing the dry powders with (an estimated) 95% of the water and the superplasticizer (1. no single method is generally accepted and therefore different countries use their own test methods to determine packing density and/or water demand of fine particles. The slump flow was determined by a mini cone test (upper/lower diameter 20/37 mm and height 57 mm) on a flow table (Tonindustrie) with a 300 mm diameter glass plate. These inter-particle forces can cause. the particle packing density of the materials should be known.C. also packing density is influenced by this. If the particles of a concrete mixture are optimized in such a way that the particle packing density increases. . Tests on each powder type or size fraction were performed at least twice to ensure a packing density precision of 0. tests were performed on cement pastes and mortar mixtures. use of superplasticizer etc. The slump flow was determined by a mini cone test (upper/lower diameter 38/90 mm and height 75 mm) on a flow table (Tonindustrie) with a 300 mm diameter glass plate. Fennis and J. 3 TESTING METHODS To be able to predict how much water is required to fill up the voids in between the particles of a concrete mixture and predict the workability. by applying external loads such as vibration or top-weight. the mixing energy method was used to determine the packing density of the powder.2% kg/kg Glenium 51) in a three-liter Hobart mixer for one minute. After two minutes resting and scraping. the dry powders were mixed for ten seconds after which the water and superplasticizer were added. was used to determine the viscosity of the paste. then rested and scraped for one minute and subsequently mixed for another minute at low speed. thus lowering the packing density. The paste was mixed for 1 minute at low speed. However. A coaxial cylinder viscometer.A. Each mixture consisted of 1500 grams of powder. The following list gives a short summary of the most frequently used techniques: 2 • Water demand France 3 • Water demand Germany 4 • Water demand . the inter-particle forces become increasingly important.

3 378.0 18. which were placed in longitudinal direction in the center of the long sides with 136 mm measuring length.33 0.3 421.26 0.43 WPR [-] 0.3 378. M600 Sibelco) or a combination of both.29 0. 4 MATERIALS AND MIXTURE DESIGN The maximum packing density of seventeen pastes was tested by the mixing energy method. one measurement only ** below measuring range .37 0.2% per gram cement) and with an adjusted water content to ensure stability of the mixtures.28 1.A.0 18.29 0.602 0.29 0.0 18. the same powder compositions were tested according to NEN-EN 1961:2005.0 18.5 N ENCI Maastricht).26 0.43 αe αt CIPM Viscosity [Ns/m ] 1. Cube compressive strength and tensile splitting strength were measured according to NEN-EN 12390:2009 on 150 mm cubes after 2. 20.639 0.637 0.22 1.598 0. The mixtures were designed for strength class Table 1: Mixture compositions and test results of cement pastes series A. all mixtures consisted of 1500 grams of powder to which the water and superplasticizer (Glenium 51 BASF con. Quartz powder from three different size classes was used (M6: d50=33µm.3 496.26 0.33 1. 10.0 18. Fennis and J.3 378.43 0. M300.33 0.45 0.M.33 0.61 0.4 637. The longitudinal displacements were recorded continuously by means of four LVDT’s per specimen.29 0.605 0.33 0.603 0.542 0. Based on the experience with the cement pastes and mortars. concrete mixtures were tested on strength development in time.26 0.C.19 1.0 18.0 18.0 18.3 378.602 0.666 0.604 0.2 [-] [-] 0.55 0.656 0.0 18.604 0. For the mortar tests.37 0.4 Glenium 51 [g] 18.3 378.35%) were added according to the procedure described in the testing methods.50 1.8 475.596 0. Additionally.0 18.0 WCR [-] 0.26 0.589 0.604 0.604 0.542 0.3 378. The experiment was executed by loading three 100×100×400 mm prisms with a -3 constant deformation speed of 10 mm/s.0 18.43 0.3 378.604 0.37 0.604 0.527 - * Due to required high mixing energy.649 0. 28.606 0.67 0.0 18.A.583 0. The mixtures consisted of cement (CEM I 42.35 1.5 N [g] A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14 A15 A16 A17 A18 A19 A20 CP100_W026 CP90-M6 CP80-M6 CP70-M6 CP60-M6 CP90-M300 CP80-M300 CP70-M300 CP60-M300 CP90-M600 CP80-M600 CP70-M600 CP60-M600 CP100_W029 CP100_W033 CP100_W037 CP100_W043 M6 M300 M600 1500 1350 1200 1050 900 1350 1200 1050 900 1350 1200 1050 900 1500 1500 1500 1500 Quartz powder M6 [g] 0 150 300 450 600 0 0 0 0 0 0 0 0 0 0 0 0 1500 M300 [g] 0 0 0 0 0 150 300 450 600 0 0 0 0 0 0 0 0 1500 M600 [g] 0 0 0 0 0 0 0 0 0 150 300 450 600 0 0 0 0 1500 Water [g] 378.3 378. 7.0 18. prism compressive strength and modulus of elasticity were determined after 28 days.575 0.33 0. three ecological concrete mixtures with low cement contents were designed.0 18.600 0.26 0.37 0. M300: d50=14µm. including the same amount of superplasticizer (1. The mortar mixtures were tested on flowability and cube compressive strength at 7 and 28 days. CEM I 42. Walraven Furthermore.601 0.3 378.582 0.S.26 0.27 1.2 466.26 0.583 0.3 378.604 0.8 393.604 0.669 0.26 0. At every point in time three specimens for compressive strength and three specimens for splitting tensile strength were tested.26 0.8 545.604 0. For mixture compositions see Table 1. To minimize the differences in mixing energy between the mixtures.26 0.33 0.29 0.95 0.27 ** ** 2 Kt=12.3 378. 30 or 40% [kg/kg] of the total cement volume was replaced by quartz powder.0 18.26 0.614 0. quartz powder (M6. M600: d50=3µm).660 0.43 0.37 1.604 0.616 0.23 1.0 18. The mixture compositions are presented in Table 2.597 0.26 0.26 0.527* 0. 56 and 90 days.669 0.37 0.3 378.

6 60. the experimentally determined packing density αe is decreased a little.798 0.9 57.0 34.37 0.46 0.6 37.793 0.37 0.8 10.30 0. Table 2: Mixture compositions and test results of mortars series B.5 59.813 0.4 52. the cement content was minimized to 110 kg/m in order to reach 2 the estimated design strength of 33 N/mm .8 10.37 0.795 0.37 0.A. a relatively high amount of water had to be dosed.37 αt Slump Slump Kt=9 [-] 0.8 33.41 0. The flow measurements of the mortar mixtures show the same trends as the paste mixtures. Fennis and J. The addition of M600 shows a significant improvement of the packing density.8 10. and ground incinerator bottom ash as residual product with low environmental impact.46 0.9 54.5 N (ENCI Maastricht).8 10.37 0.0 68.6 38. strength flow value 7-day 28 day 2 2 [mm] [mm] [N/mm ] [N/mm ] 83 91 83 84 89 86 87 92 93 90 92 134 159 86 85 95 95 95 91 102 98 97 124 113 97 140 177 202 96 94 49.37 0.6 46.53 0.8 10.8 10. see Table 1.5 N [g] B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13 B14 B15 C100 C90-M6 C80-M6 C70-M6 C60-M6 C90-M300 C80-M300 C70-M300 C60-M300 C90-M600 C80-M600 C70-M600 C60-M600 900 810 720 630 540 810 720 630 540 810 720 630 540 Quartz Sand Water powder [g] 90 180 270 360 90 180 270 360 90 180 270 360 [g] 2700 2700 2700 2700 2700 2700 2700 2700 2700 2700 2700 2700 2700 [g] 326 326 326 326 326 326 326 326 326 326 326 326 326 SP 51 [g] 10.8 10.62 0. the required amount of water at this consistency level could be reduced by more than 20%. The reference mixture C1 was designed with 260 kg cement per m concrete. When cement is partly replaced by M300. The increased packing density and increased flowability that are the result of replacing cement with M600.37 0. Replacing cement by the coarser quartz powders M6 and M300 has less influence on the packing density and flowability of the mortar mixtures than using the fine quartz powder M600.37 0. The mixture compositions of the three ecological mixtures and their reference mixture are presented in Table 3 3 as series C.801 0.792 0.5 N was used to increase the packing density even further and create a more ecological mixture containing less Portland clinker. Mixtures C2 and C3 were optimized on packing density. Three types of cement replacing materials were used: fly ash because of its ball bearing effect and high packing density.6 65.829 [mm] 2 5 5 2 12 3 3 13 16 15 27 39 48 4 2 Flow Comp. Replacing 20% or 30% of cement by M600 can increase the packing density by more than 10%.8 10.8 36.7 61. which is the minimum cement content according to the Dutch standards. Therefore.821 0.8 10.37 0. the number of measurements was too low to demonstrate a direct relation between packing density and viscosity.8 WCR [-] 0.46 0.8 10. Because of the increased packing density these mixtures require less water to fill the voids.37 0. 5 RESULTS Replacing part of the cement by the coarse filler M6 slightly improves the packing density of the pastes.41 0. CEM I 42.8 259 10.4 C90-M600 810 C80-M600 720 90 2700 180 2700 293 10. Since the mixture was 2 designed for strength class C20/25 with design strength 33 N/mm . Using the design 3 cycle as presented in section 6.37 0.37 WPR [-] 0.62 0.6 48.C. quartz powder because of its fine and narrow particle size distribution.6 24.6 49.801 0.37 0.796 0.8 0.7 35.6 42. In mixture C3 blast furnace slag cement CEM III/B 42.A.37 0.813 0.821 .41 0.62 0.S. can be used to optimize mixtures.827 0. Walraven C20/25 with minimum amounts of cement CEM I 42.0 74. The relatively small differences in the particle packing density of the pastes containing M6 and M300 do not show a clear relation with the viscosity measurements.33 0.9 35. Mixture C2 contained a combination of fly ash and quartz powder M600.798 0.M.53 0.8 47.8 10. Adding M600 to the cement paste clearly shows an increase in the packing density as well as a decrease of the viscosity. This was done for mixtures B14 and B15 which were designed to have the same flowability as the reference mixture B1.8 10.53 0.5 62.8 47 38.1 47.803 0. However. Mixture C4 was designed with wet ground municipal solid waste incinerator bottom ash from INASHCO.6 27.6 50.0 42.37 0.

5 N/mm for mixture C3 is explained by the high particle packing density of the powders and possible additional pozzolanic effects of the fly ash and quartz powder. Also the measured moduli of elasticity of the ecological mixtures comply with the relation for compressive strength versus modulus of elasticity as described in Eurocode 2 for normal concrete.C. Slump measurements (NEN-EN 12350-2:2009). Mixture C4 contained ground IBA as cement 2 replacing material and reached a 28-day cube compressive strength of 37. . The contribution of the fly ash and quartz powder to decrease the water demand is substantial.A.6 N/mm for mixture C2 and up to 33. air content (NEN-EN 12350-7:2009) and density (NEN-EN 12350-6:2009) are presented in Table 3. Increasing the packing density by using the right type of cement replacing material decreases the water demand. All mixtures 2 reached the mean strength of 33 N/mm for strength class C20/25 within 28 days. mixtures 3 could be designed for the same strength class C20/25 with only 110-125 kg/m cement. To increase the ability to cast in practice. The results are presented in Figure 2 as a function of the solid volume in a real mixture φmix over the maximum calculated packing density αt.M. 250 200 11 Flow value [mm] 150 100 R² = 0. Fennis and J. Based on the water/binder ratio the 2 28-day strength of these mixtures should have been between 20 and 25 N/mm . Mixture C2 has a water/cement ratio of 0. showing a relation between flow value and the packing density of a mixture. leading to easy collapse of the slump. In combination with the binding effect of the fly ash and bottom ash.89. which was especially noted when filling the moulds. All the mixtures had sufficient workability for casting and vibration. These two mixtures had similar workability as mixture C2.9 0. Walraven Further investigations were performed on more than hundred mortar mixtures . The cube compressive strength development of series C is presented in Table 3. However.9 N/mm .A. Mixture C3 was judged as harsh compared to the other mixtures.94 and a water/binder ratio of 0. The relationship between the average cube compressive strength and the tensile splitting strength of the ecological mixtures is the same as for normal concrete. This is also confirmed by electrical resistance measurements.95 1 11 φmix/αt [m3] Figure 2: The flow value as a function of φmix/αt for mortar mixtures The ecological concrete mixtures were designed based on the results from the mortar tests. This will also improve the cohesion of the mixtures which was evaluated as low for mixtures C3 and C4. it is recommended to increase the total powder content. by virtue of the dense particle structure. Very low segregation and little bleeding of the mixtures was observed. neither the conventional water/cement ratio nor the water/binder ratio of mixtures C2 and C3 can explain the strengths attained by these mixtures.83 (NEN-EN 206-1:2001). namely consistency class C1 earth moist.S.87 50 Mortar mixtures 0 0. Mixture C3 has a water/cement ratio of 0.85 0. Evaluation of the fresh state of these ecological concrete mixtures showed that all mixtures in series C were homogeneous and stable. The additional 2 2 strength gain up to 39. both improving the microstructure of the concrete.94 and a water/binder ratio of 0.

94 0. increasing the packing density influences the concrete properties in different ways.0 29. Since the amount of water in the mixtures was kept constant. which affects concrete properties such as strength and shrinkage.3 39. Increasing the packing density reduces the water demand and affects the flowability of concrete.2 39.5 30500 C2 Mixtures C3 44 66 65 85 1160 866 1.1 2.0 2.2 112 0.8 2406 15. with an optimized packing density the particles in the structure are closer to each other.1 2.1 2.2 32.1 36.6 32500 6 DESIGN METHOD FOR ECOLOGICAL CONCRETE In the mixture design of ecological concrete.7 3.9 2424 7.6 2.A.5 2.898 12 0. the water demand of a mixture depends on the required amount of void and excess water.9 2366 13.0 3.0 3.62 0.5 N (ENCI) Fly ash (SMZ Maasvlakte) Quartz powder M600 (SIBELCO) Ground IBA (INASHCO) Aggregates 4-16 (Filcom) Sand 0-4 (Filcom) Glenium 51 [% kg/kg of powders] Effective amount of water Water/cement ratio Water/powder ratio Estimated density (1% air) Packing density CIPM Rheological properties Slump Air content Density Mechanical properties 2-day cube compressive strength 7-day cube compressive strength 28-day cube compressive strength 56-day cube compressive strength 90-day cube compressive strength 7-day tensile splitting strength 28-day tensile splitting strength 56-day tensile splitting strength 90-day tensile splitting strength 28-day prism compressive strength 28-day modulus of elasticity [kg/m ] 3 [kg/m ] 3 [kg/m ] 3 [kg/m ] 3 [kg/m ] 3 [kg/m ] 3 [kg/m ] [%] 3 [kg/m ] [-] [-] 3 [kg/m ] [-] [cm] [%] 3 [kg/m ] [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 [N/mm ] 2 3 3 C1 260 1193 718 0.9 2456 12.886 20 0. Increasing the packing density lowers the required amount of void water.1 30500 110 88 62 1162 867 0.897 3 1.7 29. they require less (void) water for this flowability. the mixtures could be designed with less water and the possibility was created to maintain a constant water/cement ratio while replacing cement.6 33.5 3. These two effects can be used in the design of ecological concrete. mixtures B14 and B15 were designed to have the same flowability as the reference mixture B1.5 35.9 48.890 14 0. The other way around.62 2351 0.S.8 103 0. Furthermore.C.94 0. Because B14 and B15 have a higher packing density.7 26. Walraven Table 3: Mixture compositions [kg/m ] and the measured material properties of series C. Fennis and J.1 1.2 25.9 22.A.90 0.2 103 0.40 2408 0. Composition CEM I 42.3 4.0 2.9 24. The relation between packing density and water demand can be used to predict the required amount of water of a mixture.0 30500 C4 125 75 50 1157 864 1.40 2409 0. For instance.9 53.4 36.2 17.8 162 0.7 2.45 2402 0.4 2. As described in section 2.0 55. the water became available as excess water and the flowability of these mixtures increased. in mixtures B1 and B10-B13 packing density increased and less void water was required.6 20.9 37.6 39.M.4 2. .5 N (ENCI) CEM III/B 42. Therefore.

the space between the cement particles will vary. the strength of the mixture can be predicted . the packing density increases. This value is directly related to the flowability of a mixture. By using 11 this concept of cement spacing factor.A. In that case. but it can also decrease the space between the cement particles. When water absorption of the materials and predicted air content are available.S. In the following step the strength is predicted based on this mixture composition.C. the cement particles are further apart. the packing density remains constant. Therefore. Fennis and J. With a high packing density in the mixture. increases the space between the cement particles. This leads to higher strengths. as is shown in Figure 3. This can be explained by considering a container filled by cement particles. However. however. It 2. With higher amounts of water and higher water/cement ratios.A. The cyclic design procedure starts by determining the packing density αt of the particles in a concrete mixture. C C C F C F C C Figure 3: The volume occupied by a stable particle structure with coarse filler (F) compared to the volume occupied by a stable particle structure with fine filler (C=Cement). its packing density and its particle structure. when cement is replaced by finer particles. In that case. Depending on the particle size of the used filler. In the next step the water demand of the mixture is determined by using of φmix/αt.M. such as M600. cement particles and other particles are close to each other. The CIPM was developed and tested to include wet packings of microparticles in combination with superplasticizer Glenium 51. the hydration products of the cement particles need to bridge a larger distance. such as M6 and M300. the amount of water to be dosed during the mixing process is known and thus the mixture composition is available. This means that the cement particles are relatively further away from each other (see also Figure 3 left hand side). In total the same volume percentage of the container is still occupied by the particles. the 80% cement particles can be present in a smaller container (see Figure 3 right hand side) and therefore the space between the cement particles is smaller than with the coarse filler. so φmix and thus the volume of water can be adjusted to the required consistency. now only 80% of the original cement particles are present in that container. In ecological concrete this factor is important because the cement is replaced by fillers (or binders). the cement glues the aggregates together and the strength depends on the space to be bridged to connect all aggregates.12-14 can be calculated by analytical particle packing models . reducing the space that needs to be filled by hydration products. Walraven Optimizing the packing density is not only beneficial for the water demand. The relations between packing density and water demand and between particle structure and the cement spacing. Replacing cement by coarse fillers. Figure 4 presents the cyclic design procedure for ecological concrete based on these two relations. For this research project on 2 ecological concrete the Compressible Packing Model was extended to the Compaction 11 Interaction Packing Model (CIPM) to improve its performance for fine fillers . The result of this first design step is a known particle size distribution with its corresponding packing density. If 20% of the cement is replaced by a filler of similar size. . This space between the cement particles depends on the particle structure of the concrete. see also Figure 2. during the hydration process. eventually leading to lower strengths. can be used in the design of ecological concrete. The strength prediction is based on the assumption that in low strength ecological concrete all aggregates are stronger than the produced concrete.

Experiments have shown that with this 57% reduction of Portland cement. Walraven packing xtu Mi mi / x α t re ju ad φ en stm t water CSF strength Figure 4: Cyclic design procedure for ecological concrete. 7 CONCLUSIONS AND FUTURE PROSPECTS The workability of ecological concrete is very important when designing these types of concrete mixtures. This cyclic design is repeated until the mixture meets all requirements set by the user. For this project on ecological concrete the only controlling parameter in the design procedure was set to be the strength requirement. when the strength is still higher than required. the amount of water in concrete mixtures can be reduced. In that case. it will be very important to use particle packing models to control the water demand and workability of ecological concrete.A. a higher water/cement ratio and lower solid volume lead to larger distances between the cement particles and thus. determination of the water demand and prediction of the strength. This can be done by using the cyclic design method. the cement content in 3 concrete can be reduced from the prescribed minimum of 260 kg/m in ordinary concrete to 110 3 kg/m in ecological concrete. while the rest is used to lubricate particles and allow flowability. adjustments can be made to the mixture composition to include requirements of the user. Higher packing densities decrease the void volume between the particles and therefore decrease the amount of water necessary to fill this void volume. material properties such as strength and modulus of elasticity can remain constant or even improve. cement can be replaced by a filler in order to create a more ecological mixture.S. this effect is even more important than in ordinary concrete. However. Fennis and J. In this way. Thus.A. This is because a slight increase in water demand of mixtures with a low cement content has a larger effect on the water/cement ratio than in mixtures with a high cement content. particle size distribution changes and the cyclic design procedure should be followed again: calculation of the new packing density. With this method. thus increasing the volumetric distance between the cement particles. part of the added amount of water is used to fill voids. The reduced amount of water and low water/cement ratio can be used to save a certain amount of cement when concrete is designed for a fixed strength class. After the strength prediction. to a lower strength due to the fact that the hydration products of the cement particles need to bridge larger distances. or mixture design based on defined performance requirements. It is important to control this amount of water because the excess water pushes the cement particles further away from each other. in the future. The flow value is related to the packing density and the amount of water in the mixture. . After changing the mixture composition. In a real mixture. An increased water/cement ratio will decrease strength and durability.C. the set-up of the cyclic procedure allows also for concrete mixture design based on mixture composition restrictions. In ecological concrete with high amounts of cement replacing materials.M. For instance. eventually.

(1996) 2545-2556. 10. Fennis. Tattersall and P. Okamura. Computer Modelling of Concrete Mixtures. Miller. Toufar. Cement and Concrete Research Vol. Schwanda. VEB Deutscher Verlag für Grundstoffindustrie (1976) 29-44. London: E & FN Spon (1999). Water retention and granular rheological behavior of fresh C3S paste as a function of concentration. Colombet and H. Wasseranspruch von feinen Kornhauf-werken.A. K. Rostock. McCarthy and K. which is very much appreciated. P.M. Beitrag zur Optimierung der Packungsdichte Polydisperser körniger Systeme. Centrifugation. Zukoski. The experiments were performed by the technicians of the Stevin Laboratory of Delft University of Technology. Banfill. Born and E. de Larrard. EF & Spon. [5] K. [4] I. Mix design for self-compacting concrete. Melant and C.A. [2] F. Klose. (1999) 1441-1453.5 (2002) 242-248. The Rheology of Fresh Concrete. Walraven ACKNOWLEDGEMENTS This research was supported by the Dutch Technology Foundation STW. Journal of the American Ceramic Society. M. 79. Boston (1983). No. Comparison of the compressive yield response of aggregated suspensions: Pressure filtration. Wolter.. Design of Ecological Concrete by Particle Packing Optimization.J. Concrete mixture proportioning. Cement international. London (1999). Mansoutre.H. Dhir. [6] H. Fennis and J. PhD Thesis. [14] J.. Brouwers.K.C. [11] S. Ein Misschungskonzept für selbstverdichtenden Beton auf der Basis der Volumenkenngrössen und Wasseransprüche der Ausgangsstoffe. Freiberger Forschungsheft A 558. Dewar. A scientific approach. M.8). [10] G. Vol. Of ISCE.A.1-1320.T. 7 (1/2009) 1-8.F.F. Ozawa.G. (1995) 107-120. Zement und Beton 37 (1966) 8-17.M. Materials and Structures 38 (275) (2005) 1-9. No 25. applied science division of NWO and the Technology Program of the Ministry of Economic Affairs.A. Das rechnerische Verfahren zur Bestimmung des Hohlraumes und Zementleimanspruches von Zuschlägen und seine Bedeutung für Spannbetonbau. Concrete library of the JSCE. Pitman. (2006) 1-1313 . Germany. Fachgebiet Baustoffe. . 29. [3] W. and osmotic consolidation. 16. Puntke. Fak. Weimar.J. Germany (2002). Vol. Univ. Ibausil. [7] S.A. Bauhaus Universität. Paine. R. [8] M.D. für Ingenieurswissenschaften. Marquardt. Delft: Delft University of Technology (2011).H. Palm and A. [9] S. no 496/v-24. Development of Self-Compacting Eco-Concrete. [13] W. [12] F. Fachbereich Bauingenieurswesen. H. Determining and optimizing the void filling of dry particle systems. (Translation from Proc. REFERENCES [1] R.M. Hunger and H. Engineering property and structural design relationships for new and developing concretes. 1994. In proceedings of the internationale Baustofftagung. beton 52.S. Van Damme.