In this study, the alkali activated binders were compared with an ordinary Portland cement concrete (OPCC). The compositions of the alkali activated binders are present in Table 1. For the reactive binders, the coal fly ash (bulk density: 2.42 g/cm³) “EFA-Füller” (class F) by the company Bauminerale GmbH and a granulated slag (bulk density: 2.85 g/cm³) by the company Holcim were used. The oxides compositions of the coal fly ashes and slag were determined by energy dispersive X-ray spectroscopy (EDX) and are shown in Table 2. The measured Blaine fineness of the coal fly ash and of the granulated slag was 7290 and 3740 cm²/g respectively. A quartz sand with a grain size of 0.1 to 1.0 mm was used as aggregate. The alkali resistant (AR) glass fibres, which lower the propensity to shrinkage cracking and can enhance first crack tensile strength in the concrete, were 12 mm long and had a length mass of 45 tex. The activator solutions consisted of a mixture of aqueous sodium silicate (SiO₂/Na₂O molar ratio: 2.1) and a solution of sodium hydroxide.

The sodium hydroxide solution was prepared by dissolving sodium hydroxide pellets (99% purity) in distilled water to a concentration of 2 - 10 mol/l and cooled to room temperature for 24 h. The water/binder ratio was 0.41.

The sodium hydroxide solution and the aqueous sodium silicate were stirred for 15 min by a magnetic stirrer. After that, the AAB specimens were manufactured by adding the activator solutions to the solid precursor (coal fly ash. granulated slag and aggregates). Then, all specimens were produced by mechanical mixing for 4 min. Subsequently, the AR-glass fibres were added to the fresh AAB and mixed together for 0.5 min. The detailed mixing proportion properties are shown in Table 3.

For the structural study of hardened AAB, the specimens were cast into corresponding molds and cured at a temperature of 20˚C and a relative humidity of 65% for 24 hours. Following this, the samples for the tests to be performed on the hardened concrete were stored dry, according to EN 12390-2.

The 3-point bending tensile strength was determined by means of the Toni Technik ToniNorm (load frame 20 kN) with three samples measuring 160 × 40 × 40 mm³ (length × width × height), based on EN12390-5 (Figure 1(a)). The compressive strength was determined by means of the Toni Technik ToniNorm (load frame 3000 kN) following EN 12390-3, with three cubes having an edge length of 150 mm (Figure 1(b)). The pre-load and the loading speed were 18 kN and 0.5 mm/min respectively.

To quantify the durability of the AAB-matrix, the freeze-thaw test, the water permeability and the acid resistance were investigated. The freeze-thaw test was measured with the CIF-Test (Capillary suction, Internal damage and Freeze-thaw test) by the Schleibinger Freeze-Thaw-Tester with standard agent solution according to the recommendations of RILEM TC 176 IDC (Figure 2(a)). The acid resistance was determined by the mass loss after storage of three specimens (28 days old) in 37% hydrochloric acid for 7 days at a temperature of 21˚C (Figure 2(b)). The determination of the water permeability was carried out according to DIN EN 1062-3 (Figure 2(c)).

Table 4 reveals the fresh and hardened AAB characteristics at the age of 14 days. Using an air content tester, a gross geometric density of 2.39 g/cm³ and air content of 2.4 volume percent and were determined in the fresh AAB. The 3-point bending tensile and compressive strengths of the hardened AAB were 5.5 and 53.4 MPa respectively. The pH-value of the fresh AAB was 14, whereby the pH-value decreased to 10 due to the reactions in the hardened AAB. A dynamic elastic modulus of 36 GPa was determined by the ultrasonic examination. The total shrinkage deformation was 0.45 mm/m.

Furthermore to the reactivity and the solids content of the alkali activated binders, the amount and concentration of the alkaline activator have an influence on the solubility and reactivity of the AAB [38] . For this influence, the 3-point bending tensile

strength (after seven days) is shown as a function of the NaOH concentration (1.2 to 10.0 mol/l) in Figure 3. Referring to Figure 3, an increase of the sodium hydroxide concentration leads to an increase of the 3-point bending tensile strength. The sodium hydroxide concentration of over 5.5 mol/l did not further increase the 3-point bending tensile strength, because the saturation point with respect to the solubility of the binder was reached, thus leading to a high level of degree of reaction. Sodium hydroxide concentrations over 5.5 mol/l did not result in an increasing of the 3-point bending tensile strength after seven days.

The thermogravimetry and the differential thermogravimetry of the AAB-matrix are illustrated in Figure 4. Regarding to Figure 4, it can be observed that the weight loss started at a temperature of 35˚C with maximum weight loss at 87˚C. This weight loss can be attributed to the removal of free and unbound absorbed water molecules. The further mass loss up to a temperature of about 300˚C corresponded to the physically bound water in the porous matrix and the zeolite-water from the nanocrystalline zeolite structures [39] [40] . The mass loss up to a temperature of around 520˚C resulted from the decomposition of the chemically bound water from the calcium silicate and the calcium aluminate phases. The cumulated weight loss was of 10.5% at a temperature of 900˚C.

The results of the durability of the alkali activated binder were investigated and the comparison with the ordinary Portland cement concrete is shown in Table 5. The results comprise the freeze-thaw test, the water permeability and the acid resistance. The AAB-matrix had a water absorption coefficient (w_t) of 0.095 kg/m²∙h^0.5 and reached the class 3 (low water absorption) according to DIN EN 1062-3 (Table 5). As opposed to this, the OPCC had a water absorption coefficient of 1.301 kg/m²∙h^0.5 and reached the class 1 (high water absorption).

In addition, the AAB-matrix exhibited a high frost resistance, which was verified through a successful CIF test with a relative dynamic modulus of elasticity of 93 percent (Ru, 28) and an average scaling of 30 g/m² (m₂₈) after 28 freeze-thaw cycles (Table 5 and Figure 5).

After seven days storage in 37% hydrochloric acid, the AAB-matrix had a mass loss of 15.9% (see Table 5 and Figure 6). In comparison, the mass loss of the OPCC was significantly higher with 75.9%. The high acid resistance of the AAB-matrix is mainly due to the portlandite-free phases.