1. INTRODUCTION
At the beginning of SCC technology extreme fine silica fume was firstly used as fine filler. A silica fume specific surface was 300 000 cm2.g-1 [1]. Usage of this material is economically justified only for SCC where high strength SCC is required. In SCC with lower strength demands it is possible to use also coarser materials. Therefore the current SCC are produced with usage of fly ash or ground granulated blastfurnace slag. Good results were obtained also with usage of ground limestone fines and stone powder, which originate during crush stone process.
As SCC fine filler it is possible to use different material types from various sources. These materials are different in already mentioned specific surface and a lot of other parameters. The question is arising concerning dependence of SCC features on kind and content of used fine filler. Hereafter we will describe mechanical and physical mortar and concrete tests where the main emphasis was done to verifying of dependence of fine filler choice on cement composite durability.
2. MONITORED PROPERTIES AND TESTING METHODS
One of the main properties of cement mortars is their durability. In general it is accepted precondition that higher amount of fine fillers cause higher mortar
Fig. 1: Hydration temperature measurement
To verify mechanical and physical properties of mortars standard procedures [3, 4, 5, 6 and 7] were mostly used. Temperature measurements during hydration process were realised by the GMH 3250 facility. Way of hydration temperature measurement shows Figure 1. All mortar tests were realised on beam samples of standard size 40*40*160 mm.
Modified slump test for mortar mixture. For mortar mixture moveability assessment was used modified slump test. A testing cone has followed dimension: inner lower diameter 116 mm, inner top diameter 58 mm and height 175 mm. It was made from cured silon - see Figure 2. French company Chryso S.A.S. uses this cone for moveability assessment of high flow cement mortars.
As mixing water was used ordinary drinking water from water supply system. In accordance with standard [8] and company standard [9] were used cement CEM I 42,5 R from Mokrá cement factory and silica sand PR 30 and PR 33 from Provodín region.
Fig. 2: Company Chryso S.A.S. testing cone
For verification of rheological properties of experimental concrete mixture, we carried out modified tests by slumping of the cone (Abrams), then by L-box, J-ring and Orimet tests. Slump test (Figure 3) for concrete mixture. When testing, Abrams cone must be placed with the smaller base on the smooth pad 750 x 750 mm and is filled up to the top edge with concrete mixture without compacting. Then by lifting of taper the slumping of concrete mixture on the pad can be carried out. We measure time needed for concrete mixture to be slumped into a cake of diameter 500 mm (T50) and final diameter of the cake (M).
Fig. 3: Diameter of slump flow test
L – box test for concrete mixture
Testing method simulates concrete mixture penetration through reinforcement. For the measurements we used the instrument whose diagram is on Figure 4. During the test vertical part of the instrument is filled up with concrete mixture and by lifting of sliding gate the mixture can freely leak out over inserted ribbed steel bars (in the present case 3 x profile 12 mm with axial distance 50 mm) into horizontal part of 4 L – box. We measure time T40 i.e. time when the face of concrete mixture in horizontal part of the instrument reaches the distance 400 mm from sliding door and time T60 i.e. time when the front of concrete mixture reaches the end of L box horizontal part. When the movement ends, we subtract further values H1 (the height of concrete mixture column in 2/3 of horizontal part of the instrument) and H2 (the height of concrete mixture column by the opening of vertical part of instrument). The ratio H1/H2 determines the movement locking of mixture through reinforcement.
Fig. 4: L-box Test
Orimet test for concrete mixture
We measure the flow through swaged opening of the plant. The test results give us the information, technically accurate enough, about mixture viscosity.
Fig. 5: J-ring Test
The combination of Orimet and J – ring tests for concrete mixture. Testing procedure described in the preceding paragraph is in this case completed with apparatus that enables to assess locking of mixture through
reinforcement – J – ring test (Figure 5) Basically it means the annulus inserted in constant distance of spiky steel bars of the same profile. The distance of bars as well as applied annulus profile depending on the max. used aggregate granule differs. J-ring test apparatus is placed centric under the opening of Orimet during the test.
We again measure the flow time of concrete mixture throw the opening of Orimet and visually assess the locking of concrete mixture movement through J-ring. Each concrete mixture, described in testing procedures, was observed for 120 minutes from the mixing (the interval between each sets of measurement was 30 minutes).
Fig. 6: Orimet Test
Fig. 7: Simple Orimet Test
3. COMPOSITION OF EXPERIMENTAL MORTARS AND CONCRETE MIXTURES
3.1 Composition of experimental mortars
In total 13 mortar mixtures were subjected to testing. First group of five mortars differed in kind of used fine filler. Its content in mortar mixture was constant (10 percent of mortar filler). Next six mortar mixtures were divided into two sets. For each set one kind of fine filler was chosen. Its ration in each mortar mixture set was modified to 6, 12 and 14 percent of mortar filler content. At the last two mortar mixtures was the 12 percent fine filler content divided into two parts. In the first case it was 135g of ground granulated blastfurnace slag (10 percent of mortar filler) and 27g of silica fume (2 percent of mortar filler). In the second case it was 135g ground limestone fines and 27g of silica fume.
Mortar mixture design was accomplished to obtain constant mortar rheological properties. Mortar mixture moveability adjustments were pursued by changes in mixing water content. For all 13 mortar mixtures was used the same ration of the same superplasticizer admixture called ViscoCrete 5 – 600 produced by Sika CZ Inc. Its ration was 1 percent of cement weight, which correspond to a maximal ration advised by the producer. The next design requirement was to bring mortar content and mortar moveability closer cement mortars of self-compaction concrete [10].
For the first group of mortars stone powder was selected as the fine filler was selected. Mortars of the second group were made with ground limestone fines admixture. Combination of fine fillers used at last two mortar mixtures are described in chapter 3.1.
3.2 Composition of experimental concrete mixture
While testing, it was especially observed, if the locking of mixture in L-box and J-ring doesn’t occur. In case of constant proportion of superplasticizer the segregation of aggregate granule was recorded, then water and filler proportion was corrected so as not to occur again. We similarly proceeded even if water separation mixture was recorded in spite of the fact that some authors mention in their works about the correctness of slight water separation when the mixture excessive water is absorbed back at the beginning of the hydrating process. Thanks to the water factor treatment and proportion of fine solids in the mixture we even eliminated unwanted behaviour of concrete mixture. The proportion of fillers then in individual concrete mixture ranged from 160 to 280/m3. On the basis of described rheologic test results, basic and combination compound formulas of concrete mixtures were specified for further experiments. (Tab. I a II)
These treated compound formulas were applied for sample production to test strength a frost resistance (beam: 100 x 100-x 400 mm). For verification of resistance against water and defrost elements (CHRL), testing cubes with the side of 150 mm were produced. Bending strength and tensile compression fraction tests were gradually carried out on the beam. Frost resistance of hardened concrete was observed by the decrease of sample resistance after 75 and 150 freezing cycles [11], and the measurement of concrete water resistance and CHRL in individual steps after 25 freezing cycles in some cases up to the total 250 cycles were carried out
3.2.1 Observed mortar and concrete mixtures fine fillers
To verify concrete mixture properties modified by various types and combinations following materials were used:
• silica fume (SF), Oravské ferozlitiarenské závody, a.s., Istebné (Slovensko, specific surface 230 568 cm2.g-1
• ground blastfurnace slag (GGBS), Kotouč Štramberk, specific surface 3629 cm2.g-1
• ground limestone (L), specific surface 4 857 cm2.g-1
• fly ash (FA), power plant Chvaletice, specific surface 2 426 cm2.g-1 stone powder (SP), from quarry Želešice, specific surface 4 345 cm2.g-1
4. TEST RESULTS
4.1 Mortar mixtures
To compare frost resistance of each mortar they were subjected to four periods each with 25 freezethaw cycles. Changes in flexure and compressive strengths were carefully monitored. At Figure 8 and 9 you can see that mortar strengths weren’t relevantly affected by the frost instrumentality.
Frost coefficients were calculated from mortar flexure strengths for each period. Their value range is between 0,84 and 1,09. These results represent very high mortar frost resistance particularly if they were determined for higher number of freeze-thaw cycles.
Fig. 8: Frost resistance – different fine filler
Fig. 9: Frost resistance – different fine filler volume
Only two mortar mixtures dissatisfied frost limit value based on [3]. The first one was mortar with 10 percent of stone powder and it exceeded the limit after 100 cycles. The second one was mortar modified by 10 percent of silica fume. In this case the frost limit value was exceeded sooner in 75 cycles and a drastic decreasing of this value arose between 50 and 75 cycles. Similar results we obtained for frost resistance value calculated from mortar compressive strength.
A mortar chemical resistance was verified by the instrumentality of 3 percent NaCl (sodium chloride) solution on a surface of testing samples. The samples were exposed to freeze-thaw cycling process. A stage of sample corrosion attack was measured every 25 freeze-thaw cycles. Mortar with 10 percent of scoria showed the highest sodium chloride resistance - see Figure 10. After 100 freeze-thaw cycles there was obtained 745,4 grams of outfall (convert to 1 square meter). With respect to a close-grained texture of an investigated mortar the result is really excellent. Conversely to this silica fume modified mortar samples proved a broad breakage and this process was finalised almost total disintegration of samples after 100 freeze-thaw cycles - see Figure 10.
Fig. 10: Chemical resistance – different fine filler
At Figure 11 you can see a clear connection between a sample degradation phase and a fine filler content of mortars with a various portion of stone powder. Inconsistent results were obtained by testing of mortars with ground limestone fine usage. In case of mortars with 6, 12 and 14 percent content of ground limestone fines it shows comparable resistance like mortars with scoria fine filler. While in case of mortar with 10 percent of scoria was exceeded a limit value of sample deterioration after 75 freeze-thaw cycles, which corresponds to a deterioration mode - ruined [4].
Fig. 11: Chemical resistance – different fine filler volume
By reason of a mortar sample different behaviour new testing samples were prepared from mortars with 10 and 12 percent of fine filler and the measurements were repeated. Difference of these measured values weren’t so broad - see Figure 12, but a significant difference in a resistance of both samples still remain. To clarify causation of this phenomenon it will be accomplish another tests and measurements.
At the Figure 12 there are also mentioned mortar resistance test results where a combination of silica fume fine filler and another fine filler like ground granulated blastfurnace slag or ground limestone fines were used. Resistance such mortars was sharply worsened with a growing number of freezethaw cycles. This note is primarily valid in case combination of silica fume and ground limestone fines. With reference to particularly positive strength test results of these mortars (vide infra) it is necessary to take an account their practical usage.
Fig. 12: Chemical resistance – another testes
At Figure 13 and 14 is mentioned compressive and flexural strength progress of mortars in phase from 1 to 60 days. Remarkable is the fact that silica fume modified mortar don’t reach a prospective higher strength, but mortars with combination of silica fume with another fine filler had much auspicious strength values.
Fig. 13: Mortar compressive strength test
Fig. 14: Mortar flexural strength test
At Figure 15 is an interesting comparison where you can see that the quickest start of setting and hardening process have cement mortars with silica fume fine filler. It can be caused by high grain fineness of this material and its sharply hydraulic features. On the other hand kind of used fine filler doesn’t affect mortar setting time.
Fig. 15: Setting process parameters – different fine filler
Dependence between fine filler mortar content and dynamic of hydration process start is evident from Figure 16. If you increase the mortar fine filler content the beginning of mortar setting process will be markedly accelerated. This dependence is obvious for both kinds of fine fillers.
Fig. 16: Setting process parameters – different volume of fine filler
Hydration heat progression in dependence on kind of used mortar fine filler is clear from Figure 17. Dependence between hydration heat progression and mortar fine filler volume you can see at Figure 18 (specifically for stone powder fine filler).
Fig. 17: Temperature of hydration - set I
Fig. 18: Temperature of hydration - set II
4.2 Concrete mixtures
The graphical representation of concrete compression strength test results of basic compound formulas set is shown in Figure 19. The strength of dispersion after 28 days of maturing (water placing) was approximately 40 % of max. measured value depending on the type of applied fine filler. The highest strength was achieved at samples produced from concrete mixture using silica fume as the fine filler (81 MPa after 90 days). For tensile strength under bending (Figure 20) the dispersion test results of basic compound formulas after 90 days were similar (approximately 35 % of max. measured value) and the highest strength was achieved at samples no. II and VI. (7.82 a 7.15 MPa). In graphs representing the comparison of concrete compound formulas show that entire dose of fine filler necessary for achievement of demanded rheological mixture properties is ensured by silica fume.
Fig. 19: Compressive strength gain with time
Fig. 20: Flexural strength gain with time
The measurement results in this case confirm preceding experiences, that so high silica fume doses (159 kg/m3 of fresh concrete) result in massive decrease of concrete durability (see graphs in Figure 21 and 22).
Fig. 21: Frost resistance – flexural strength
Fig. 22: Chemical resistance (3 % NaCl solution)
On the contrary very favourable results from durability point of view showed samples produced from concrete, that contained combination of stone dust removers and ground blastfurnace slag as fine fillers. The concrete produced only with slag admixture showed outstanding chemical resistance (3% NaCl solution), but determination results of its frost resistance at higher number of freezing cycles were distinctively worse. Equal results were obtained from frost and chemical resistance point of view from samples where admixture ground
limestone was used. The influence of applied combination of fine fillers on short-time concrete strength development is evident from graph in Figure 23.
Fig. 23: More types of fine filler in mix
The best results were achieved at formulas no. XII and XIII. The samples marked XIV up to XX were produced in particular to assess the influence of increasing content siliceous flies in SCC on its selected machine-physical properties. The required rheological properties of experimental mixtures were primarily achieved by adding admixture of ground blastfurnace slag. Volumes of silica fume dosed into concrete mixture are 2.5, 5.0, 7.5, 10.0, 15.0, and 20.0 % of concrete weight. Graphs of developments of tensile strength under bending and tensile compression at these formulas are shown in Figure 24 and 25.
Fig. 24: Compressive strength gain with time
Fig. 25: Flexural strength gain with time
From the course of individual curves, it is evident, that expected increase of strength SCC, depending on increasing silica fume dose, didn’t become apparent. The reason might be on relatively high volumes of ground blastfurnace slag applied in these mixtures for achievement of their rheological properties in fresh state. High doses of slag might at least partially suppress the effect of the strength increase of SCC by silica fume. From the graph in the Figure 26 on the contrary it is evident considerable effect of silica fume dose on SCC chemical resistance. Considerable increase of sample damage with 3% NaCl solution, occurred when we had content of silica fume in SCC over 10 % (referring to concrete weight) and higher number of freezing cycles. On the contrary, no effect of silica fume dose on frost resistance SCC (Figure 27) has been proved during the test.
Fig. 26: Chemical resistance (3 % NaCl solution)
Fig. 27: Frost resistance – flexural strength
CONCLUSION
Accomplished tests proved significant dependence between a kind of mortar, a fine filler volume and mechanical-physical features of these mortars. Knowledge of that and knowledge about affection extend of individual mortar technical parameters by increasing of fine filler content will enable a rational concrete structure proposal. By the help of their usage in practise there will be possibility to produce self-compacting concrete with controlled utility features, which will take an account of individual needs of concrete structures.
The results of the performed concrete tests show that there is a favourable effect of GGBS on strength values SCC as well as on chemical resistance of this concrete. However before concluding on unambiguous positive assessment, in that case it is necessary to wait for final durability (frost) test results. From frost resistance point of view, the most favourable results of already finished tests, were achieved from compound formulas no. I, III a VI and then all compound formulas observed within assessment of silica fume dose effect on SCC (concrete mixtures no. XIV up to XX).
Different frost resistance results at the samples produced from concrete mixtures no. II a XIV (both mixtures contained GGBS as a filler) might be caused by various volumes of blastfurnace slag in both compound formulas. This presumption will have to be verified by other tests. Comparing the compound formulas of concrete mixtures showing the highest frost resistance of produced SCC with the achieved chemical resistance is evident, that the highest durability was achieved at mixtures marked no. V, XIV, XV and XVI. The performed tests results showed that the usage of higher silica fume doses in concrete mixtures isn’t the best solution from the durability of self compacting point of view.
Foreign experiences and tests results performed by ourselves show that optimization of SCC proportion with consideration of demanded properties of fresh concrete mixture and resulting properties of hardened concrete require complete knowledge of individual characteristic of concrete mixture components in the course of all technological cycle of concrete production. E.g. according to American experience, durability of SCC using fly ash as filler is possible to be increased by concrete mixture aeration.
In that case it’s however necessary to pay high attention to their rheological properties because higher air content influences mobility of mixtures as well as its other properties. As a fine filler in the Czech Republic for utilization in SCC seems to be highly perspective combination GGBS with stone powder, resp. with fly ash, when taking in consideration the current prices.