Research of UHPC properties prepared with industrial mixer

Transcription

1 IOP Conference Series: Materials Science and Engineering PAPER OPEN ACCESS Research of UHPC properties prepared with industrial mixer To cite this article: E Šerelis et al 2017 IOP Conf. Ser.: Mater. Sci. Eng View the article online for updates and enhancements. Related content - The influence of nanoclay on the durability properties of asphalt mixtures for top and base layers Johan Blom, Bram De Kinder, Jannes Meeusen et al. - Durability Properties of Piezoelectric Stack Actuator Takenobu Sakai and Hiroshi Kawamoto - Post-treatment Effect of Particleboard on Dimensional Stability and Durability Properties of Particleboard Made From Sorghum Bagasse A H Iswanto, T Sucipto, S S D Nadeak et al. This content was downloaded from IP address on 08/04/2018 at 11:39

2 Research of UHPC properties prepared with industrial mixer E Šerelis 1, V Vaitkevičius 1 and V Kerševičius 1 1 Dept. of Building Materials, Kaunas University of Technology, Studentų 48, Kaunas, Lithuania evaldas.serelis@ktu.lt Abstract. Ultra-high performance concrete (UHPC) mixture with advanced mechanical and durability properties was created using decent Zyklos ZZ50HE mixer. Zyklos ZZ50HE rotating pan mixer is similar to mixer which has common concrete plants. In experiment UHPC was prepared with Zyklos ZZ50HE mixer and thereafter best composition was selected and prepared with industrial HPGM 1125 mixer. Experiment results revealed that UHPC with W/C=0.29 and advanced mechanical and durability properties can be prepared. In experiment tremendous amount of micro steel fibres (up to 147 kg/m 3 ) were incorporated in UHPC. Concrete with excellent salt scaling resistance and great mechanical properties was obtained. Compressive strength was increased about 30 % from 116 MPa to 150 MPa and flexural strength was increased about 5 times from 6.7 to 36.2 MPa. Salt-scaling resistance at 40 cycles in 3 % NaCl solution varied from kg/m 2 to kg/m 2. There were a few attempts to create UHPC and UHPFRC with decent technology, however, unsuccessfully till now. In the world practice this new material is currently used in the construction of bridges and viaducts. 1. Introduction Over a past few decades significant advances in research on cementitious materials have been made which led to develop a new class composition material with improved strength and durability properties. Probably the best durability and compressive strength properties could be attributed to ultra-high performance concrete (UHPC). In Europe UHPC is known as concrete which has compressive strength over 100 MPa. However by adding proper amount of necessary materials compressive strength could be increased up to 200 MPa [1-3]. Chemical admixtures are very important additives which helps reduce water to cement ratio below 0.30 and pozzolanic additives (fly ash, silica fume, ground-granulated blast-furnace slag and etc.) is very beneficial to gain additional mechanical strength during further hydration process. The most common pozzolanic material used in UHPC is silica fume [4-5]. Interaction of silica fume between reactive materials is very complicated and greatly affects workability and hydration process of cementitious materials. However silica fume is very expensive additive. Resent research showed, that silica fume better works with other pozzolanic materials such as glass powder. However deeper research is required to proper understand interaction of silica fume and glass powder on properties of ultra-high performance concrete. Glass is a solid material which generally consists of non-crystalline silica, sodium oxide, calcium oxide and other components [6-8]. The chemical composition mainly depends on the raw materials used and differs slightly for each glass type. The most common soda-lime glasses consist of 70 % amorphous SiO 2, 12 % Na 2O and 5 % CaO [9-11]. Being amorphous and containing relatively large quantities of silicon, glass should be an excellent pozzolanic material for concrete industry. Thus when Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by Ltd 1

3 glass is finely grounded to powder, in theory, it could be used as cement replacement. However glass powder has very high amounts of Na 2O which could initiate alkali-silica reaction. Therefore glass powder can be characterized by a few controversial facts: it has enough amorphous SiO 2 to be considered as pozzolanic material; also one of the main chemical components is lime so it can be considered as cementitious material and it contains a very high amount of Na 2O which is a source for alkali silica reaction. These truths are highly debated between various scientists; however it is not entirely clear how glass can affect certain properties of concrete. In experiment was revealed some controversial truth about glass powder utilisation in UHPC. The main aim of this research is to create UHPC composition which could be prepared with industrial mixer and without usage of advanced technology. New composition was prepared with different amount of micro steel fibres. Main properties of UHPC were determined by Chapelle test, mercury intrusion porosimetry (MIP), XRD analysis, compressive strength, flexural strength and saltscaling. Proposed composition could be used as innovative material for various elements made of concrete which could be prepared with industrial mixer. 2. Used materials Cement. Portland cement CEM I 52.5 R was used in experiment. Main properties: paste of normal consistency 28.5 %; specific surface (by Blaine) 4840 cm 2 /kg; soundness (by Le Chatielier) 1.0 mm; setting time (initial/final) 110/210 min; compressive strength (after 2/28 days) 32.3/63.1 MPa. Mineral composition: C 3S 68.70; C 2S 8.70; C 3A 0.20; C 4AF Particle size distribution of cement, silica fume, quarts powder and quartz sand are shown in Fig.1. Table 1. Chemical compositions of Portland cement, silica fume and glass powder Components Quantity, % CEM I 52.5 R Glass powder Silica fume SiO TiO Al2O Fe2O MnO MgO CaO SO Na2O K2O P2O Na2Oeq Loss of ignition Silica fume. Silica fume, also known as micro silica (MS) or condensed silica fume is a by-product of the production of silicon metal or ferrosilicon alloys. Main properties: density 2532 kg/m 3 ; bulk density 400 kg/m 3 ; ph 5.3. Quartz powder. Quartz powder was used in the experiment. Main properties: density 2671 kg/m 3 ; bulk density 900 kg/m 3 ; average particle size µm; specific surface (by Blaine) 3423 cm 2 /g. Quartz sand. Quartz sand was used in the experiment. Main properties: fraction: 0/0.5; density 2650 kg/m 3 ; specific surface (by Blaine) 91 cm 2 /g. Glass powder. Glass powder was used in the experiment. Main properties: density 2528 kg/m 3 ; average particle size µm; specific surface (by Blaine) 3350 cm 2 /g. Particle size distribution is shown in Fig. 2. Micro steel fibres. In experiment were used micro steel fibres with hooked ends. Main properties: length 30 mm, diameter 0.30 mm; tensile strength 1000 MPa. 2

4 Chemical admixture. Superplasticizer (SP), based on polycarbocylisc ether (PCE) polymers, was used in the experiment. Main properties: appearance: dark brown liquid, specific gravity (20 C) 1.08±0.02 g/cm 3 ; ph 7.0±1; viscosity 128±30 m Pa; alkali content 5.0 %, chloride content 0.1 %. Figure 1. Particle size distribution of Portland cement, silica fume, quartz powder and 0/0.5 mm fr. quarts sand Figure 2. Particle size distribution of glass powder 3. Methods Specific surface and particle size distribution. Specific surface was measured with Blaine instrument according to EN 196-6:2010 standard [12]. Particle size distribution was measured with Mastersize 2000 instrument produced by Malvern Instruments Ltd. Table 2. Compositions of ultra-high performance concrete Composition Micro filler, kg/m Water, Cement, Quarz Micro l kg/m 3 W/C Glass sand, SP, l fibres, Silica fume Quarz powder powder kg/m 3 kg/m 3 QP/GP ,25 99 QP/GP QP/GP100-F0 0 QP/GP100-F40 40 QP/GP100-F , QP/GP100-F QP/GP100-F QP/GP100-F Mixing, sample preparation and curing. Some fresh concrete mixes (Table 2) were prepared with Zyklos ZZ50HE mixer and some with HPGM 1125.Main properties of Zyklos ZZ50HE were: capacity 50 litres; high-speed whirler 1500 rpm (1.5/4 kw); power mixing star 2.2 kw. Main properties of HPGM 1125 mixer were: the volume is 1125 litres; the effective mixing volume is 750 litres, the engine power is 30 kw and the mass of the mixer is 3800 kg. It took about 12 min of time to prepare UHPC mixture with both mixers and thereafter 1 extra min for proper fiber distribution. Mixtures were prepared from dry aggregates. Cement, aggregates and chemical admixtures were dosed by weight. Quartz powder and silica fume to glass powder were substituted by volume. In experiment various sizes beams, cubes (10x10x10 cm) and prisms (10x10x40 cm) were created. Homogeneous mixes were cast in moulds (without compaction) and kept for 28 days at 20 C/95 RH. Mercury Intrusion Porosimetry (MIP) test. At age of 28 days cylinder (d=50 mm, h=50 mm) of each composition was broken into small fragments ant then placed in isopropanol. Later these fragments were dried in the oven at 40 C to remove all free water. All dried fragments were stored in sealed containers for mercury porosimetry tests. The pressure was applied from zero to 450 MPa. A 3

5 constant contact angle of 140 and a constant surface tension of mercury of 480 mn/m were assumed for the pore size calculation. X-ray diffraction (XRD) analysis. Hardened cement pastes were used for XRD analysis. The XRD measurements were performed with a XRD 3003 TT diffractometer of GE Sensing & Inspection Technologies GmbH with θ-θ configuration und CuKα radiation (λ=1.54 Å). The angular range was from 5 to 70 2 Theta with a step width of 0.02 and a measuring time of 6 sec / step. For XRD quantitative phase analysis the samples were mixed with 20 wt. % ZnO (a standard material widely used in XRD analysis) as an internal standard and stored in argon atmosphere until measurement using the Rietveld refinement. This permits the estimation of the amount of non-crystalline phases by the Rietveld fitting procedure. a) b) c) d) Figure 3. UHPC mixture preparation with Zyklos ZZ50HE (2-a) and with HPGM 1125 (2-c) mixer and the moulds filled with mixture (respectively 2-b and 2-d) The pozzolanic activity of glass powder. Pozzolanic activity is assessed by Chapelle test. One gram of glass powder is mixed with 1g of CaO and 200 ml of water. The suspension is boiled for 16 h and the free Ca(OH) 2 is determined by means of sucrose extraction and titration with a HCl solution. Compressive and flexural strength. Compressive and flexural strength were performed after 28 days according to EN and EN standard respectively [13, 14]. Compressive strength was obtained from 3 cubes (10x10x10 cm) and flexural strength was obtained from 3 prisms (10x10x40 cm) as an average value. Salt-scaling. Salt-scaling of concrete was performed according to CEN/TS :2006 standard [15]. 3 prisms (7x7x21 cm) were used for experiment. 4. Results and discussion The main aim of the research is to create UHPC composition under factory conditions. In experiment two different types of mixers were used (Zyklos ZZ50HE and HPGM 1125). Mixer Zyklos ZZ50HE is rotating pan mixer which has similar mixing efficiency as larger mixers used under factory conditions. In the research were created various UHPC compositions with different amount of glass powder by using Zyklos ZZ50HE mixer. Thereafter best composition was selected and prepared with various amount of micro steel fiber by using industrial HPGM 1125 mixer. The HPGM 1125 is planetary counter current concrete mixer. Main results of the research will be discussed further. 4

6 4.1. Optimal fineness of glass powder Optimal fineness of glass powder was determined by Chapelle test. According to Chapelle test (Fig. 3), the pozzolanic reactivity is evaluated on the basis of the amount of Ca(OH) 2 consumed during the pozzolanic reaction. Experiment results indicate, that pozzolanicity of glass powder after it oversteps fineness of 5240 cm 2 /g tends to increase insignificantly. When fineness of glass powder varies from 2119 cm 2 /g to 7399 cm 2 /g consumed amount of Ca(OH) 2 varies from 465 mg to 652 mg respectively. Optimal fineness of glass powder according to the consumed amount of Ca(OH) 2 is 5240 cm 2 /g. The same amount of silica fume consumes 1536 mg of Ca(OH) 2 and it is almost 3 times more reactive than glass powder. In Fig.4 specific surface of silica fume is not presented, because Blaine test method shows inaccurate results. Specific surface of silica fume usually varies from m 2 /kg to m 2 /kg by BET test method. Thereafter hardened cement pastes by hand were created and UHPC samples by Zyklos ZZ50HE mixer. Figure 4. Results of Chapelle test for glass powder and silica fume Figure 5. Relationship between water to cement ratio and density when heat treatment was not applied 4.2. XRD analysis Fig. 6 illustrates the XRD patterns of two hardened cement pastes with different amount of glass powder. CH phase was found at 17.9, 33.9, 46.8, 50.2 and Evidently the crystalline phase of CH was decreased with an increase of glass powder. C 2S and C 3S phases were found at 29.3, 29.7, 31.9, 38.3, 41.3, 45.3, 49.4, 51.2 and It was noticed, what with increased amount of glass powder also decreased intensities of C 2S and C 3S. Decreased C 2S and C 3S peaks probably related with better solubility of clinker phase. Decreased CH peaks probably related with the consumption by pozzolanic reaction of silica fume and glass powder. Figure 6. XRD patterns of hardened cement pastes with and without glass powder 5

7 4.3. Density of hardened UHPC As it was expected density in all compositions was very similar and it was about 2400 kg/m 3 (Fig.5). It was noticed that density intends more decrease due to W/C ratio than due to used pozzolanic materials. Slightly decrease of density probably could be attributed due to higher amount of evaporable water which tends to increase capillary porosity of concrete and thus decrease overall density. It seems that glass powders do not influence overall density of UHPC or the influence is insignificant Pore size distribution of hardened UHPC Experiment results revealed, that glass powder has positive effect on microstructure improvement of UHPC (Fig. 6 and Fig. 7). Significantly lowered porosity was obtained by composition (QP/GP100) with glass powder. Composition without glass powder (QP/GP0) had continuous pore distribution up to 700 µm, while pore distribution of composition with glass powder (QP/GP100) was smaller than 70 µm. Reduced macro porosity significantly affects mechanical properties of UHPC. Another interesting fact was observed, all compositions had higher concentration of porosity at Nano scale 0.1 µm. These types of pores do not influence mechanical properties, they mostly affect the shrinkage and creep of concrete. XRD analysis was applied in order to find out why composition with glass powder showed lowered pore size distribution. Figure 7. MIP pore diameter versus cumulative pore volume (W/C=0.25) Figure 8. MIP pore diameter versus log differential intrusion (W/C=0.25) 4.5. Compressive and flexural strength of hardened UHPC Interesting fact was noticed, when glass powder was incorporated in UHPC composition. UHPC compositions were prepared with Zyklos ZZ50HE mixer. With substitution up to 100 % of quartz powder by glass powder compressive strength increased about 40 MPa from 182 MPa (GP/GP0) to 221 MPa (GP/GP100), however, flexure strength remained almost the same - 20 MPa (Fig. 9). Such enormous compressive and flexure strength could be obtained in every concrete plant with advanced and sophisticated technology, however, most producers cannot afford such equipment. Water to cement ratio was increased up to Modified mixture was prepared with various amount of micro steel fibres by using HPGM 1125 mixer (Fig.10). Interesting fact was noticed (Fig. 10), that with increase up to 147 kg/m 3 of micro steel fibres, compressive strength increased about 30 % from 116 MPa (QP/GP100-F0) to 149 MPa (QP/GP100- F147), flexural strength increased more than 5 times from 6.7 MPa (QP/GP100-F0) to 36.2 MPa (QP/GP100-F147). According to the experiment results, micro steel fibres has positive effect on compressive ant flexural strength on UHPC. Despite the fact that optimal amount of micro steel fibres is about 40 kg/m 3, further increase up to 73.5 kg/m 3 has negligible effect on both flexural and compressive strength. Even further increase of micro steel fibres (up to 147 kg/m 3 ) has negligible 6

8 effect on compressive strength and has positive effect on flexural strength. However, with extreme amount of steel fibres could be drastically reduced workability and durability of concrete. In order to denote this postulate another test method was applied. Figure 9. Compressive and flexural strength of UHPC with different amount of glass powder (W/C=0.25) Figure 10. Compressive and flexural strength of UHPC with different amount of micro steel fibres (W/C=0.29) 4.6. Salt-scaling In cold climate zones the main durability problem is insufficient resistance to frost damage. In order to find out how different amount of micro steel fibres affects resistance to frost damage, salt-scaling test methods was applied (Fig. 11). It was noticed that with increase up to 147 kg/m 3 of micro steel fibres after 40 cycles in 3 % NaCl solution salt-scaling increased from kg/m 2 (QP/GP100-F0) to kg/m 2 (QP/GP100-F147). Critical losses of mass at 28 cycles are 1 kg/m 2 according to CEN/TS :2006 standard [15]. Although micro steel fibres has negative effect on salt-scaling, UHPC still has advanced resistance to frost damage. a) b) 100 µm Figure 11. Salt-scaling of UHPC with different amount of micro steel fibres at 40 cycles (W/C=0.29) Figure 12. Defects of structure identified by fluorescence test method: a) without magnification and b) with 50x magnification Visual inspection by fluorescence test method (Fig. 12) revealed that deterioration by surface scaling is not homogeneous. The greatest damage was observed on the frozen surface and gradually decreased in deeper layers. Surface scaling probably affected all cross section; however deterioration degree in each layer was different. Although concrete by nature is inhomogeneous material and in different circumstances applied methods should be perfect for structure analysis, however after cyclic 7

9 surface salt-scaling concrete should be considered as composite material, which has several layers with its own properties. 5. Conclusions Macro porosity of UHPC were eliminated. Largest porosity was observed at micro-scale ( 70 µm). All compositions of UHPC had highest concentration of pore at Nano-scale ( 0.1 µm). During experiment, it was founded that advanced mechanical properties of UHPC with industrial mixer can be obtained. Compressive strength can be increased by up to 30 % - from 116 MPa (without micro steel fibres) up to 149 MPa (with 147 kg/m 3 micro steel fibres). Flexural strength can be increased more than 5 times - from 6.7 MPa (without micro steel fibres) up to 36.2 MPa (with 147 kg/m 3 ). After 40 cycles, mass losses of researched compositions of salt-scaling were between kg/m 2 to kg/m 2. Highest mass losses were noticed in composition with 147 kg/m 3 of micro steel fibres. References [1] Singh L P, Karade S R, Bhattacharyya S K, Yousuf MM, Ahalawat S Beneficial role of nanosilica in cement based materials A review. Construction and Building Materials 47: [2] Dils J, Boel V, De Schutter G Influence of cement type and mixing pressure on air content, rheology and mechanical properties of UHPC. Construction and Building Materials 41: [3] Yu R, Spiesz P. Brouwers, H J H Mix design and properties assessment of Ultra-High Performance Fibre Reinforced Concrete (UHPFRC). Cement and Concrete Research 56: [4] Arroudj K, Zenati A, Oudjit M N, Bali A, Hamou A T Reactivity of Fine Quartz in Presence of Silica Fume and Slag, Engineering 3: [5] Maroliya M K Micro Structure Analysis of Reactive Powder Concrete. International Journal of Engineering Research and Development, 4(2): [6] Khatib J M, Negim E M, Sohl H S, Chileshe N Glass Powder Utilisation in Concrete Production. European Journal of Applied Sciences, 4(4): [7] Mirzahosseini M, Riding K A Effect of curing temperature and glass type on the pozzolanic reactivity of glass powder. Cement and Concrete Research, Volume 58, Pages [8] Zanotto E D Surface crystallization kinetics in soda-lime-silica glases. Journal of Non- Crystalline Solids, DOI: no [9] Pigeonneau F, Muller S The impact of iron content in oxidation front in soda-lime silicate glasses: An experimental and comparative study. Journal of Non-Crystalline Solids, 380: [10] Venkatanarayanan H K, Rangaraju P R Decoupling the effects of chemical composition and fineness of fly ash in mitigating alkali-silica reaction. Cement & Concrete Composites, 43: [11] Pignatelli R, Comi C, Monteiro P J M A coupled mechanical and chemical damage model for concrete affected by alkali silica reaction. Cement and Concrete Research, 53: [12] LST EN 196-6:2010. Methods of testing cement - Part 6: Determination of fineness. [13] EN :2000. Testing hardened concrete - Part 4: Compressive strength - Specification for testing machines. 8

10 [14] LST EN :2009/P:2011. Testing hardened concrete Part 5: Flexural strength of test specimens. [15] CEN/TS :2016. Testing hardened concrete - Part 9: Freeze-thaw resistance with deicing salts Scaling. 9