Geopolymer ferrocement panels under flexural loading

2.1 Materials used

The following materials are employed in this work: ordinary portland cement (53 grade), fine aggregate (sand), pozzolan (low-calcium fly ash), alkaline activator (NaOH+sodium silicate), chicken mesh with hexagonal opening, and weld mesh with rectangular grid opening.

Ordinary Portland cement was used to produce a control mix mortar in this paper. The physical properties of the cement obtained from the tests conducted as per relevant IS 4031 (Part 4 and 5)-1988 [17] are shown in Table 1.

The fine aggregate used is ordinary river sand passed through sieve no. 8 (2.36 mm) with a specific gravity of 2.72, dry density of 1.6 g/cm³, and having a fineness modulus of 2.56.

The pozzolan used here is the low-calcium fly ash of class F obtained from the Tuticorin thermal power plant. The chemical composition of the fly ash, as determined by X-ray fluorescence analysis is given in Table 2.

Geopolymer is a combination of the following compounds: pozzolan (fly ash), activator solution (silicates of sodium), alkali powder (hydroxides of sodium), a high-range water-reducing lignosulfonated normal super plasticizer, and distilled water.

In this research, analytical grade NaOH in pellet form (NaOH with 98% purity) and sodium silicate solutions (14.7% Na₂O, 29.4% SiO_2, and 55.9% water by mass) were used as the alkaline activators. In order to avoid the effect of unknown contaminants in the mixing water, the NaOH pellets were dissolved in distilled water. The activator solution was prepared at least 1 day prior to its use. To improve the workability of fresh mortar, commercially available naphthalene-based super plasticizer was used. The constituents of geopolymer mortar are shown in Figure 1.

Figure 1

Composition of geopolymer mortar.

Wire woven chicken meshes (Figure 2) with a hexagonal opening of size 12 mm and wire thickness of 0.72 mm (20 gauge) are used. Machine-welded weld mesh (Figure 3) having a rectangular grid opening of size 76.2 mm×38.1 mm, with a thickness of 2.45 mm in the transverse direction and 3.45 mm in the longitudinal direction are used. Ultimate strength values of weld mesh and chicken mesh are 440 and 270 N/mm² as per ASTM A 185 [18].

Figure 2

Chicken mesh.

Figure 3

Weld mesh.

2.2 Volume fraction of reinforcement

When the same square or rectangular wire mesh is used throughout the depth of an FC element then the volume fraction of reinforcement can be calculated from Eq. (1) [19]:

where V_r is the volume fraction of reinforcement, N is the number of layers of mesh, π=3.14, d_w is the diameter of mesh wire, h is the thickness of FC element, D_l is the distance center to center between longitudinal wires, and D_t is the distance center to center between transverse wires.

For other types of mesh (hexagonal or chicken wire, expandable metal, fiber-reinforced plastic, or irregular meshes) Eq. (2) [19] is recommended to be used.

where V_r is the volume fraction of reinforcement, N is the number of layers of mesh, w_r is the unit weight of reinforcing mesh, h is the thickness of FC element, and δ_r is the density of reinforcement material.

2.3 Method of specimen preparation

Thirty FC panels (10 series, three in each series), each of size 1000 mm×200 mm×25 mm and 35-mm thickness (Figure 4), were cast in steel molds with cement mortar (8, 10, 12, and 14 m), geopolymer mortar reinforced with four layers of chicken mesh and one layer of weld mesh, which were tied firmly together for volume fractions of 1.8% and 1.3%.

Figure 4

Rectangular panel mold, 25/35-mm thickness.

The molds were made in a form such that each of the four side walls and the base of the formwork were detachable so that the mold could be easily separated from the cast element after its initial setting. The contact surfaces of the steel molds to the mortar were greased before casting the specimens to ease the demolding process. A minimum cover of 5 mm on all sides was maintained when placing the reinforcement inside the mold, and a cover block was used to maintain a 5-mm cover at the base. Indelible ink markings were made on the inside of the formwork and on all its sides to indicate levels of different volume fractions and different molar concentrations during casting.

Ordinary portland cement and river sand passing through No. 8 (2.38 mm) sieve and having a fineness modulus of 2.72 were used for casting. The geopolymer mortar used in this study is composed of low-calcium fly ash and alkaline solution composed of NaOH and sodium silicate combinations. NaOH is mixed with deionized water at a required concentration and kept for at least 24 h prior to casting. All geopolymer mortar specimens were made with sand-to-geopolymer ratio in equal proportion. The hydroxide-to-silicate ratio is kept constant as 1:1 Table 3. The fly ash and fine aggregate were dry mixed together in a mixer machine for 5 min, followed by the addition of activator solution containing hydroxide and silicate to the mixture, and mixed for another 10 min. The mixing was carried out at a room temperature of approximately 25–30°C (Figure 5).

Figure 5

Mixing of mortar.

Ordinary mesh obtained from the market was cut to obtain the required number of layers of desired size and orientation. The reinforcement was provided with a combination of chicken mesh and weld mesh. Chicken mesh with 12-mm hexagonal opening and gauge 20 were used. Weld mesh of rectangular opening of size 76.2 mm×38.1 mm was placed in such a way that the longer dimension was placed in the lengthwise direction of the panels. The reinforcements were tied firmly together in specified layers for various volume fractions. The geopolymer mortar was spread at the base of the mold, and on this base layer the mesh layers were laid. The molds were placed on the vibrating table and the mortar was spread evenly until the thickness of the panels was completely filled without voids (Figure 6). In each casting, three GF panels along with three mortar cubes of size 70.6 mm×70.6 mm×70.6 mm were cast to test the characteristic strength of the mortar mix (Figure 7). All these geopolymer mortars and GF specimens were heat cured for 48 h at 75°C in a heat curing chamber (Figure 8).

Figure 6

Specimen under table vibration.

Figure 7

Cube mold.

Figure 8

Heat-curing chamber under operation mode.

For all the FC specimens, the water-to-cement ratio of 0.3 and equal proportion of sand and cement (in order to compare with the geopolymer specimen) were taken by weight. The required amount of sand and cement was mixed dry in a mixer and then the required quantity of water and super plasticizer were added gradually while the mix was continuously stirred.

Similarly, FC panels were prepared by using cement mortar instead of geopolymer mortar in the mold. The specimens were air-dried for 24 h for initial setting and then immersed in water for curing. The specimens were removed from water after 28 days and were air-dried until the test was performed.

2.4 Testing of specimens

Two-point loading was used to conduct flexure tests. The bed of the testing machine is provided with two steel rollers on which the specimen is to be supported, and these rollers are mounted in such a way that the distance from center to center is 600 mm. The load is applied through two similar rollers mounted at the third point of the supporting span that is 200 mm center to center. The load is divided equally between the two loading rollers, and all rollers are mounted in such a manner that the load is applied axially and without subjecting the specimen to any torsional stress or restraints. For this experiment, the manual operating system was used and the readings were taken at an interval of 0.1 kN. The deflections were measured with the dial gauges having at least a count of 0.01 mm at the midsection of the element. Figure 9 shows the typical flexural test setup of GF panel. Before testing, all the elements were painted white so that the cracks could be easily observed and clearly photographed. The load was applied in a vertically upward direction by means of an electric jack. The load application was continued until the deflection became excessive (not exceeding 30 mm). The deflections were measured for each interval of load increment. On each day of testing, 70.6-mm mortar cubes of the respective mortar mix were also tested on the Universal testing machine to record the characteristic compressive strength of the specimen. The crack patterns were monitored and crack width was measured by using a microscope. The load at first crack, the maximum load, and corresponding deflections were also recorded. Table 4 shows the test results of all the tested specimens and Figure 10 shows a typical failure of a test specimen. After testing, the panels were removed from the test setup and both top and bottom sides were examined to investigate the sustained damage, such as yielding of reinforcement and cracking pattern at the bottom face.

Figure 9

Flexure test setup.

Figure 10

Typical failure of the specimen.

3.1 Compressive strength of different specimens

The testing of 70.6-mm mortar cubes of the respective mortar mix to record the characteristic compressive strength of the specimen is shown in Table 5 and Figure 11. Figure 11 illustrates the effect of NaOH concentration on the compressive strength of mortar and demonstrates that the compressive strength of geopolymer mortar increases with the increase in the concentration of NaOH. The compressive strength of geopolymer specimens increases as NaOH concentration in the aqueous phase increases from 8 m to 14 m. The NaOH concentration in the aqueous phase of the geopolymeric system acts on the dissolution process as well as on the bonding of solid particles in the final structure. The use of high concentration of NaOH leads to greater dissolution of the initial solid materials and increases geopolymerization reaction, and hence, higher compressive strength is achieved [20, 21]. The increment of sample strength is believed to be mainly due to the increase in Na ions and water molecules, which are the basic ingredients for geopolymerization that enhances the dissolution and reaction of the geopolymer mortars that eventually increases the final compressive strength. The percentage increase in compressive strength compared to the control specimen for 8, 10, 12, and 14 m geopolymer mortars is 9.015%, 36.09%, 58.95%, and 83.43%, respectively.

Figure 11

Compressive strength relationships.

3.2 Load-deflection characteristics

Deflection characteristics have been considered as essential criteria in the design of FC structures by practitioners. Deflection in FC is normally caused by the magnitude of loading it carries and also due to time-dependent factors such as shrinkage and creep [22]. One of the main advantages of having meshed reinforcement in FC is that the deflections due to shrinkage are significantly reduced by the presence of uniformly distributed mesh. From Figure 12A–E, the behavior of GF panels can be characterized by three distinct stages. Stage I can be designated as precracking of the specimen, wherein all the materials are elastic and the load-deflection curve is linear with the applied load. In this stage, the load is carried by the geopolymer matrix. Cracking load can be predicted by using the linear elastic theory. Stage II can be designated as multiple cracking of the specimen wherein more cracks occur after cracking of the geopolymer matrix. During this stage, a transfer of load from mortar to the wire mesh takes place. The wire mesh elongates under additional load, transferring the load back to the mortar matrix, forming new cracks. The final stage is named as failure of the specimen wherein there is widening of existing cracks with the increase of load. In this stage, the load is predominantly carried by the mesh. The specimen undergoes more deflection, and cracks can extend deep through the depth of the geopolymer matrix, reaching the top surface of the specimen.

Figure 12

Load vs. mid-span deflection curve of (A) FC panels, (B) 8 m GF panels, (C) 10 m GF panels, (D) 12 m GF panels, and (E) 14 m GF panels.

3.3 Cracking behavior

The failure of specimens results from the yielding of wire mesh reinforcement followed by the crushing of mortar. Initially, fine flexural cracks appeared at the bottom of the specimen. With further increase in the load, regularly spaced vertical cracks were observed, and they extended from the bottom of the specimen toward the top. The load was increased up to the ultimate stage and cracking pattern is observed. It was observed that the number of cracks formed in all specimens depended on the volume fraction of reinforcement. Also from the observations it is understood that a higher volume fraction of reinforcement provides a better crack control mechanism by the formation of a large number of well-distributed cracks. Table 6 presents the details of measured average crack width, number of cracks, and average crack spacing.

3.3.3 Crack pattern of GF panels and failure mode

Failures of tested GF panels were observed to occur when the extreme layer of steel mesh failed under tensile stress. No spalling of the mortar matrix was observed for any tested GF panels. A major continuous crack joining the bottom and top fibers was observed to be present on all tested panels. From Figure 13A, large numbers of closely spaced cracks were observed to be developed on test panels 12 m GF. Meanwhile, the relatively smaller number of cracks with larger intercrack spacings were detected on FC panels (Figure 13B). Only flexural cracks were observed on all tested GF and FC panels at ultimate failure. Hence, it can be concluded that all the tested panels failed in pure bending.

Figure 13

Cracking behavior of (A) 35-mm-thick 12 m GF panels and (B) 35-mm-thick FC panels with five layers of wire mesh.