Experimental Work

2.4.2.1 Materials

The same type of specimens was used for all of the experimental investigations viz.

cubes with side length 100 mm, which were cut from beam specimens, cf. Figure 2.1.

Figure 2.1 Cutting of specimens from cast concrete beams.

2.4 Experimental and Analytical Procedure Chapter 2 Paper I

Cubes 1 - 6, cf. Figure 2.1 were used for the experimental investigations, whereas the end-pieces, ie the darker areas in Figure 2.1, were discarded as the mould ends affect the homogenous dispersion of the fibres.

The same type of steel fibres was used for all series, ie black steel fibres manufactured from cold drawn steel with hooked ends, d = 0.55 mm and l = 35 mm [Bekaert, 2011].

The aggregates used for all the series were excavated seabed materials, viz. naturally rounded, in accordance with [DS 2426, 2004] and tap water was used for the mixing-water. The materials and casting procedures for each series are given in separate sub-sections in the following.

2.4.2.2 Series A

The mixture design used for Series A is given in Table 2.4.

Table 2.4 Concrete compositions for Series A (Assuming s.s.d. conditions of ag-gregates and 1.0 vol.-% air).

Materials Quantity [kg/m³]

Plain concrete SFRC

0.0 vol.-% 0.5 vol.-% 1.0 vol.-% 1.5 vol.-%

Cement 500 500 500 500

Water 240 240 240 240

Sand (0-4 mm) 1563 1550 1536 1523

Steel fibres 0 39 78 117

The w/c ratio was 0.48. Though maximum aggregate size was 4 mm, Series A is in the following referred to as concrete. Aalborg Portland White cement, CEM I 52.5 R (HS/EA/≤2) was used for the casting and additional information about the cement type is available in [AAP, 2011a].

A standard pan-mixer was used for mixing and sand, cement, and fibres (if added) were dry-mixed for three min. Subsequently water was added and mixing was contin-ued for additional five min. Two beam specimens 100x100x840 mm per mix design were cast, and compacted by the use of a vibration table. After casting, the moulds were covered with plastic sheets to avoid evaporation from the fresh concrete surface, and left for curing at laboratory conditions, (~20ºC). The specimens were de-moulded after 24 h and stored in a water basin with lime rich water for further curing. After at least 28 days the specimens were cut into cubes with side length 100 mm, cf. Figure 2.1, and stored in lime rich water until time of testing.

Chapter 2 2.4 Experimental and Analytical Procedure Paper I

2.4.2.3 Series B

The concrete compositions used for Series B are given in Table 2.5.

Table 2.5 Concrete compositions for Series B (Assuming s.s.d. conditions of ag-gregates and 1.0 vol.-% air).

Materials Quantity [kg/m³]

0.0 wt.-% Cl^-/wt.-% cem. 4.0 wt.-% Cl^-/wt.-% cem. 6.0 wt.-% Cl^-/wt.-% cem.

Plain concrete SFRC Plain concrete SFRC Plain concrete SFRC

0.0 vol.-% 0.5 vol.-% 1.0 vol.-% 0.0 vol.-% 1.0 vol.-% 0.0 vol.-% 0.5 vol.-% 1.0 vol.-%

Cement 325 325 325 325 325 325 325 325

Water 163 163 163 163 163 163 163 163

Sand (0-4 mm) 814 814 814 814 814 814 814 814

Aggregates (4-8 mm) 1079 995 911 1079 911 1079 995 911

Steel fibres 0 39 78 0 78 0 39 78

Chlorides 0 0 0 20 20 31 31 31

The w/c ratio of Series B was 0.50, and the air content of the fresh concrete was as-sumed to be 1.0 vol.-%. The cement type used was Aalborg Portland Basis cement, CEM II/A-LL 52.5 N (IS/LA/≤2) [AAP, 2011b].

CaCl2, which was used as the source of chlorides, was dissolved in ¼ of the mixing water and added at the very end of the mixing and mixed for additional 30 seconds to distribute the chlorides in the fresh concrete. This procedure was adapted to facilitate mixing by limiting early acceleration of the hydration process caused by the chloride ions. One beam specimen, as shown in Figure 2.1 was cast per chloride content and fibre volume fraction.

A standard pan mixer was used for mixing and cement, sand, aggregates and ¾ of the mixing water, viz. original amount of water minus water with mixed in chlorides, were mixed for two minutes. Subsequently fibres (if added) were mixed in the con-crete for additional five min. The specimens were compacted by the use of a vibration table. Mixes containing chlorides needed more intense vibration, ie shorter vibration time and higher vibration frequency, than mixes without chlorides due to the acceler-ated setting caused by the addition of chlorides.

Finally the moulded specimens were covered with a plastic sheet to avoid evaporation from the fresh concrete surface and left for curing at laboratory conditions (~20ºC). 24 h after casting, the specimens were de-moulded and stored in a basin containing lime rich water for further curing. Specimens with mixed-in chlorides were stored in chlo-ride and lime containing water to minimize exchange of chlochlo-ride ions between speci-mens and water. Finally the specispeci-mens were cut in cubes, cf. Figure 2.1, after 28 days and stored saturated until time of testing.

2.4 Experimental and Analytical Procedure Chapter 2 Paper I

2.4.2.4 Series C

The concrete compositions used for Series C are given in Table 2.6.

Table 2.6 Concrete compositions for Series C (Assuming s.s.d. conditions of ag-gregates and 1.0 vol.-% air).

Materials Quantity [kg/m³]

Plain concrete SFRC

0.0 vol.-% 0.5 vol.-% 1.0 vol.-%

Cement 375 375 375

Water 163 163 163

Sand (0-4 mm) 761 755 750

Aggregates (4-8 mm) 56 55 55

Aggregates (8-16 mm) 1025 1018 1011

Steel fibres 0 39 78

The w/c ratio of the mixes given in Table 2.6 was 0.43. Aalborg Portland Basis ce-ment, CEM II/A-LL 52.5 N (IS/LA/≤2), was used as cement [AAP, 2011b]. The amount of aggregates was reduced for SFRC compared to plain concrete to account for the addition of fibres, cf. Table 2.6. The same grading of the aggregates was main-tained for the three different concrete compositions given in Table 2.6.

All concrete compositions given in Table 2.6 were mixed by the use of a standard pan mixer. Cement, water, sand, and aggregates were mixed for two min. Fibres (for SFRC mixes only) were added and the concrete was mixed for additionally two min.

For plain concrete, where fibres were not added, the mix was also mixed two times two minutes. Four beam specimens, as shown in Figure 2.1, were cast per fibre vol-ume fraction and the specimens were compacted by the use of a vibration table. The moulds were covered with a plastic sheet after casting to avoid evaporation from the fresh concrete surface. After storage for 24 h at laboratory conditions (~20ºC) the specimens were de-moulded. Subsequently to de-moulding, the specimens were stored in lime rich water for 28 days, at which age they were cut into cubes. The cubes were stored in lime rich water for another month. Four cubes per fibre volume fraction were conditioned to different levels of relative humidity (RH), viz. 45 and 75 % RH and additionally four specimens from each fibre volume fraction were stored in lime saturated water. Those four cubes subjected to the same RH, were selected from four different beam specimens to minimize the possible influence of the natural variation of the concrete material properties between the different beam specimens cast.

Cubes for RH = 45 % and RH = 75 % were conditioned using the rapid conditioning method described by [Rilem, 1999]: Initially the required mass loss was calculated based on sorption isotherms for the material. (The sorption isotherms were determined on smaller, crushed samples originating from companion cubes to those used for measurements of the electrical resistivity.) Subsequently the cubes were sealed on four sides to ensure 1-dimensional evaporation and oven-dried at 50 ^oC until the cal-culated mass loss was obtained. When the required mass-loss was reached, the cubes were sealed on all six sides and placed in the oven at 50 ^oC for at least 14 days to

ob-Chapter 2 2.4 Experimental and Analytical Procedure Paper I

tain a more homogenous moisture distribution in the cubes. Finally, the cubes were taken out of the oven and stored in small sealed containers to minimize the exchange of moisture with the ambient air until the experimental investigations were initiated.