Keywords

1 Introduction

Geopolymer material produced by the reaction of aluminosilicate sources with alkaline solutions has emerged as an alternative binder to ordinary Portland cement (OPC), essentially because of low energy consumption and low carbon dioxide emission when using industrial by-products in its production. The by-products used in the synthesis are mainly fly ash and ground granulated blast furnace slag (GGBS). Geopolymer produced from only fly ash is associated with a very long setting time and low compressive strength; therefore, the addition of slag is crucial to improve these properties [15]. The reaction mechanism of geopolymerization has been explained by some authors. The geopolymerization process starts with the dissolution of the solid aluminosilicate source by alkaline hydrolysis that releases [SiO4]− and [AlO4]− tetrahedral units. These tetrahedral units are linked with each other by sharing oxygen atoms to form polymeric bonds. The last process involves the polycondensation, where the geopolymer gel sets and forms a three-dimensional aluminosilicate network [4, 6, 18]. In fly ash/slag blends, the C-S-H phase and aluminosilicate gel co-exist in the system, which contributes to strength improvement [15].

On the other hand, designing buildings with good fire resistance is required in many sectors due to the experiences of fire disasters. Even though many buildings have been constructed with non-combustible materials such as brick, concrete, and steel, they cannot completely prevent the outbreak of fire and the building from collapsing. Some buildings are equipped or covered with heat-insulating materials, which can be very expensive. Geopolymer materials possess the potential as a fire-retardant due to their ceramic-like properties [3]. As shown by several previous studies, geopolymer material can maintain high compressive strength after exposure to high temperatures compared to OPC [8, 17, 20]. For instance, Kong and Sanjayan [8] compared the compressive strength of OPC paste with fly ash geopolymer paste after exposure to various temperatures. They found that OPC paste almost totally lost its compressive strength after being heated at 400 ℃, while geopolymer paste exhibited high compressive strength after exposure to a temperature of 800 ℃, which was 5.4% higher than its initial strength at room temperature. Another study by Zhang et al. [20] also compared the compressive strength of OPC and geopolymer pastes. They found that OPC paste completely lost its compressive strength after exposure to 800 ℃, while geopolymer still maintained its compressive strength after exposure to this temperature. The reason behind this is because of the re-crystallization of geopolymer at a very high temperature which leads to the formation of the analogies to natural minerals having extremely high fire resistance [5]. Moreover, some studies also claimed that geopolymer could maintain its strength at high temperatures due to the sintering effect [7, 10].

The fire resistance of geopolymer is affected by several factors, such as the type of raw materials [9, 14, 20], curing conditions [1], and alkali cation types [1, 11]. The combination of sodium silicate and sodium hydroxide is the most common activator solution used in geopolymer production. Although many studies have been conducted on the heat resistance of geopolymers, there is very little data on the heat resistance of geopolymer activated with potassium hydroxide (KOH). In addition, the molarity of alkali hydroxide solution can significantly affect the properties of geopolymer [2, 15]. Thus, this study aims to investigate the effect of the molarity of KOH solution on the heat resistance of fly ash-slag blended geopolymer paste. As a comparison, a heat resistance test on OPC paste was also conducted.

2 Materials and Method

2.1 Materials and Specimen Preparation

This study utilized fly ash (Blaine fineness = 3,550 cm2/g, density = 2.24 g/cm3) which conforms to JIS A 6201, GGBS (Blaine fineness = 4,170 cm2/g, density = 2.91 g/cm3) which conforms to JIS A 6206, and OPC (Blaine fineness = 3,340 cm2/g, density = 3.16 g/cm3). Their chemical compositions are shown in Table 1. Sodium silicate solution consisting of 13.4 mass% Na2O, 27.3 mass% SiO2, and 59.3 mass% H2O, and potassium hydroxide solution were utilized as alkaline solutions. The potassium hydroxide solutions were prepared in 10, 12, and 14 M by dissolving the potassium hydroxide pellets in water to form one liter solution. These solutions were prepared one day before mixing. The ratio of alkaline liquid to binder (l/b) and the ratio of sodium silicate to potassium hydroxide (SS/KH) were maintained constant at 0.52 and 1.35 by mass, respectively. The detail of the mix proportion is shown in Table 2.

Table 1 Chemical composition of raw materials. LOI is loss of ignition at 1000 ℃
Full size table
Table 2 Mix proportion
Full size table

The geopolymer paste was synthesized as follows: fly ash and GGBS were first dry-mixed together in a pan mixer for 3 min. Alkaline solutions were added to the dry materials and mixed rapidly and continuously for another 3 min. Subsequently, the fresh geopolymer paste was cast into cylindrical plastic molds (Φ50 × 100 mm). On the other hand, OPC paste was prepared with a water to cement ratio of 0.45 in accordance with the above mixing procedure. Just after casting, both OPC and geopolymer specimens were cured in an oven at 70 ℃ for 24 h under sealed condition. After oven curing, the specimens were stored at ambient temperature (approximately 20℃), demolded after aging for three days, and stored in a sealed plastic bag to prevent alkalinity loss or carbonation before the tests.

2.2 Heat Resistance Test

After curing for 28 days, specimens were exposed to high temperatures at 500, 750, and 950 ℃ by placing them into an electric furnace with a heating rate of 20 ℃/min. When the targeted temperature was reached, it was maintained for 1 h. Afterward, the furnace was turned off and let it cool down to room temperature. Some specimens were left unexposed for comparison.

2.3 Characterization and Measurement

The compressive strength of unexposed and exposed geopolymer paste and OPC paste specimens at the age of 28 days was evaluated following JIS R 5201. It was conducted using a universal testing machine with a compression capacity of 250 kN and a loading rate of 0.2 mm/min. The test was repeated for three specimens for each mixture, and an average was taken.

The phase of geopolymer and OPC product was characterized by X-ray diffraction (XRD). It was conducted using an X-ray diffractometer in the 2-theta range of 10–65° with a Cu Kα X-ray source.

The microstructure of unexposed and exposed geopolymer specimens was examined using scanning electron microscopy (SEM). It was conducted at 15 kV of accelerated voltage. The samples were coated with platinum.

3 Results and Discussion

3.1 Compressive Strength

The compressive strength of the specimens unexposed and exposed to high temperatures was measured and compared. Figure 1(a) shows that the increase of KOH molarity leads to the increase of initial compressive strength. The improvement is attributed to the fact that the increase in molarity increases the dissolution rate of the aluminosilicate precursors due to the high pH of high concentration of alkali solution [4, 12], resulting in more gel formation and denser microstructure. In addition, it can be seen that the compressive strength of all the geopolymer specimens is higher than that of OPC paste.

Fig. 1
An image with a line and a bar graph. Both have 4 lines and bars for 10, 12, 14, and O P C. All lines in the line graph begin from the top left and descend forming a slope till 750 and slightly ascending. In the bar graph, the highest bars are for 500 degrees Celsius, and among them, 14 M is the tallest and the lowest is for 750 degrees Celsius.

Compressive strength of geopolymer paste as a function of temperature a and retained compressive strength after heat exposure b

Full size image

After exposure to high temperatures, all the specimens exhibited a significant loss of compressive strength due to the decomposition of C-(A)-S-H gel. Similar to C-S-H in OPC paste, C-(A)-S-H in geopolymer is the main source of compressive strength development, and therefore, the decomposition of C-(A)-S-H in geopolymer and C-S-H in OPC paste will lead to a significant loss of compressive strength. Further heating at a higher temperature, all the specimens continuously lost their strength; however, when geopolymer specimens were heated up to 950 ℃, the compressive strength slightly increased and became higher than the retained compressive strength of specimens after exposure to 750 ℃.

Increasing the molarity improves the heat resistance of geopolymer. Figure 1(b) shows the percentage of retained compressive strength of geopolymer and OPC paste. Although the three series of geopolymer present a similar tendency of strength loss, the geopolymer with 14 M KOH shows the highest retained compressive strength corresponding to approximately 24% of its initial strength after heating at 950 ℃. In contrast, the OPC specimen retains only 4.5% of its initial compressive strength after heating at 950 ℃.

3.2 Phase Changes

Figure 2 represents the XRD patterns of unexposed and exposed OPC paste and geopolymer paste. Only the XRD pattern of geopolymer activated with 14 M KOH was shown since all the geopolymer specimens present similar phase changes after exposure to high temperatures. It can be seen in Fig. 2(a) that OPC paste mainly consists of portlandite, with some low-intensity peaks of C2S (belite), C3S (alite), ettringite, and calcite. At 500 ℃, the portlandite peak was still observable, and it started to decrease considerably at 750 ℃. At 750 ℃, a CaO peak was formed, and considerable amounts of C2S and C3S were observed. Calcite peak disappeared at 950 ℃, which indicates a complete decomposition of calcite at a temperature between 750 and 950 ℃. The peak intensity of CaO and C3S increased considerably with the increase of exposure temperature at 950 ℃. This is attributed to the further decomposition of portlandite, calcite, and C-S-H.

Fig. 2
An image with 2 line graphs and there are 4 lines with 5 shapes spread along with it indicating certain peaks for the graph on left and alphabets on right. All the lines run along horizontally placed one upon another forming some major and minor peaks. 2-Theta on the horizontal axis and intensity on the vertical axis .

XRD patterns of unexposed and exposed OPC pastes a, and geopolymer paste activated with 14 M KOH b. (Q = quartz, M = mullite, C = cristobalite, W = wollastonite, A = akermanite, L = leucite)

Full size image

Figure 2(b) represents the XRD pattern of geopolymer paste (14 M). In the unexposed specimen, the mullite (Al6Si2O13) and quartz (SiO2) peaks are originally from fly ash [16]. After exposure to 500 ℃, these peaks remain, while the C-(A)-S-H peak almost vanished due to dehydroxylation. Unlike the OPC specimen, further heating at 750 ℃, new crystalline peaks such as wollastonite (CaSiO3) and cristobalite (SiO2) were observed in geopolymer, proving the re-crystallization of this material into secondary mineral phases. Moreover, at 950 ℃, other new crystalline phases such as leucite (KAlSi2O6) and akermanite (Ca2Mg[Si2O7]) were observed. The presence of these two major phases is believed to contribute to the slight increase in the compressive strength after exposure to 950 ℃. The effect of molarity on phase change was insignificant because all the geopolymer specimens present similar phase changes at high temperatures; however, increasing the molarity increases the potassium content in the paste, and therefore, the amount of leucite formed after heating at 950 ℃ probably increased.

3.3 Microstructure

The geopolymer specimen showing the highest retained compressive strength after exposure to high temperatures was also characterized by SEM, and the result is shown in Fig. 3. In the unexposed specimen, it was evident that geopolymeric gel was found to co-exist with the calcium aluminate silicate gel (C-(A)-S-H), resulting in a compact microstructure.

Fig. 3
A grid of 4 images with a close-up view of a material that looks like small bubbles and pores on a block. As the temperature goes on increasing, the material looks ruptured with huge pores and openings and at 950 degrees Celsius, a bubble in middle is labeled sintering.

SEM images of unexposed and exposed geopolymer paste activated with 14 M KOH

Full size image

After heating at 500 ℃, the deterioration occurs due to the impact of dehydration and decomposition of the gels, resulting in a coarser microstructure compared to the unexposed sample. According to previous studies, the dehydration of the C-S-H in hardened OPC commences at approximately 110 ℃ and becomes completely decomposed at a temperature range of 600–800 ℃ [13, 19]. The dehydration of C-(A)-S-H in geopolymer is estimated to occur at a temperature similar to C-S-H in OPC. Further heating at 750 ℃, the microstructure becomes highly porous, resulting in very low compressive strength. However, after heating at 950 ℃, there was evidence of sintering. This could be the reason for slightly increasing the compressive strength from 750 to 950 ℃ of exposure temperature.

4 Conclusion

In this study, the effect of KOH molarity on the heat resistance of geopolymer exposed to 500, 750, and 950 ℃ was investigated. The following conclusions can be drawn from the experimental results.

  1. 1.

    The increase of molarity of KOH solution increases the initial compressive strength of geopolymer pastes.

    After exposure to high temperature, 14 M KOH solution performed the highest retained compressive strength, followed by 12 and 10 M. The compressive strength of all the geopolymer samples exposed to 950 ℃ was higher than that exposed to 750 ℃, which is attributed to the sintering effect and formation of new crystalline phases.

  2. 2.

    In comparison with the OPC paste, geopolymer pastes showed better heat resistance in terms of compressive strength loss. Unlike geopolymer which re-crystallized at high temperature, OPC paste presents an increase in the amount of C2S and C3S and the formation of CaO as the result of the thermal decomposition of C-S-H and calcium hydroxide.

Geopolymer made from fly ash and GGBS, activated with sodium silicate and potassium hydroxide solutions has been proven to have better performance than OPC paste when exposed to high temperatures,thus promising the use of geopolymer as an alternative material to OPC.