Durability of Repair Metakaolin Geopolymeric Cement under Different Factors

Abstract

Nowadays, energy saving, and green sustainability are influencing the development of all industries, including the construction industry. In recent years, geopolymeric cement and concrete have become hot topic materials as a replacement for traditional OPC; this work carried out orthogonal experiments to identify four potential factors affecting the basic properties of the metakaolin-geopolymeric cement specimens. The results showed that the metakaolin and activator contents were the two primary influencing factors. Given the importance of studying the durability of building restoration materials in harsh environments, this experiment focused on testing the bond strength, permeability resistance, sulphate corrosion resistance, and freeze-thaw resistance of metakaolin geopolymer pastes with different proportions of metakaolin dopant and alkali activator content. The findings are that durability of the formed specimens significantly improved when suitable metakaolin and activator contents were incorporated, and bond strength was also improved. Moreover, the microscopic tests, including SEM and FT-IR experiments, were used to better reflect the changing durability of pattern. The experiments showed that the best durability of the metakaolin geopolymeric cement was achieved when the ratio of metakaolin to cement was 1.5 and the ratio of activator to cementitious material was 0.3. It can be concluded that the appropriate content of metakaolin and activator can give the geopolymer excellent performance under harsh conditions, which will contribute to the wide application of geopolymer.

1. Introduction

As industrialisation accelerates, the variety of construction materials has been increasing. In 2019, China’s construction industry was worth nearly 3.6 trillion dollars and employed around 55 million people. The share of construction output in the country’s GDP has been rising year on year, hiring 7% of the total employed population; it is clear that the construction industry is a pillar industry in China. As a major component of civil structures, experimental research is essential, and it is linked to the improvement of people’s living standards.

Research into cement-based building materials has focused on the durability, high strength, and rapid setting properties of cement-based materials. However, the frequent use of Ordinary Portland Cement (OPC) has resulted in significant greenhouse gas emissions and growing environmental concerns [1]. OPC production consumes a large number of natural resources [2], with heat and electricity consumption exceeding 2.72 GJ/ton and 65 kWh/ton, respectively [2,3]. Furthermore, despite some preventive techniques, the degradation of civil infrastructure made of cement or concrete and the shortening of structural age are inevitable durability issues worldwide [4,5,6]. As a result, new materials are being sought that meet not only the strength requirements needed for basic building performance but also higher durability requirements. Geopolymer is a low carbon material first proposed by Davidovits in 1978 [7,8]. Geopolymeric cement can be produced by a reaction between industrial by-products (typically fly ash, slag, metakaolin, etc.) and alkaline solutions. The special structure of geopolymers makes them sustainable materials with many mechanical advantages [9] and durability [10]. Therefore, mineral geopolymers have been described as the most promising green cement material of the 21st century [11].

In recent years, the replacement of cement and silica fume with metakaolin in concrete has become a hot topic [12,13,14]. Metakaolin is classified as a reaction product of low calcium systems. Due to their highly cross-linked (mainly Q4) and zeolite-like structure, N-A-S-H gels have better mechanical properties than OPC [15]. Unlike cement as a cementitious material alone, the ratio of metakaolin to activator has an influence on performance [16], which means that the geopolymer content is highly important for this new system. The durability of the concrete is a crucial factor in determining the service life [12,17]. Recent studies on durability [5,18] have focused on the geopolymers of different materials and mechanisms of structural damage. Pasupathy et al. [19] investigated the durability of fly ash-based geopolymer concrete exposed to external environments to study resistance to carbonation. Davidovits et al., [20] highlighted that when fly ash geopolymers were placed in a 5% H₂SO₄ solution for a certain number of days, a mass loss of 7% was found for the metakaolin-based geopolymer. Hanrahan [21] found that the durability of fly ash-based geopolymer concrete is largely governed by the internal configuration of the aluminosilicate gel composition in extreme environments (5% Na₂SO₄ solution). Due to the outstanding durability characteristics of geopolymers, geopolymer concrete has potential applications in pavement rehabilitation. Keyu Chen et al., [22] mixed geopolymers with slag for pavement rehabilitation. Hawa et al. [23] used geopolymers for road rehabilitation. X.Q. Peng et al., and B.B. Jindal [24,25] searched for the preparation of geopolymeric materials and their application to the rapid repair of cement concrete pavement.

However, the proportion of metakaolin and alkaline activators as remediation materials has not been determined to obtain more information about the durability of the repair geopolymerical cement, including resistance to permeability, sulphate corrosion, and freeze-thaw resistance. Furthermore, there has been relatively little research on the effect of the dosage of alkali activator on the durability of the geopolymer system and the effect of geopolymer prepared with partial replacement of cement by metakaolin as a remediation material. The metakaolin/cement ratio and the amount of alkali activator still need to be explored to understand the durability of geopolymer systems as rapid repair pavement materials in adverse weather conditions. At the same time,

In this paper, four possible factors that have a large influence on the geopolymer system, including metakaolin, alkali activator, βs, and modulus, were first analysed by orthogonal experiments prior to the durability tests. Orthogonal experiments were conducted to determine the main influencing factors that modify the performance of the geopolymeric cement. Then, corresponding durability tests were carried out based on the two sets of variables. Combining bond strength and micro-morphology, it was found that the possible causes of the variation in durability were metakaolin and alkali activator contents. By forming a geopolymerical cement with a suitable formulation, a road repair material with good repair function and durability is formed; this experiment also lays the foundation for the combination of mechanised intelligence and engineering use. In the future, the durability of metakaolin polymers can be more widely applied in practical engineering. The flow chart is shown below in Figure 1.

2. Experimental Methods

2.1. Materials

2.1.1. Metakaolin and OPC

Metakaolin supplied by Datong Jinyuan Kaolin Co., Ltd. (Datong, Shanxi, China), was applied to part of solid mixtures. The parameters of metakaolin are in accordance with GB/T 14563-2008 and other relevant specifications. The relevant physical and chemical properties of metakaolin are shown in

Table 1. Chemical and Physical Composition of metakaolin.

Components.	SiO2	Al2O3	Fe2O3	TiO2	CaO	MgO	K2O	Na2O	Whiteness	PH Value	Water Content	Loss on Ignition	Average Grain Size	Fineness
Content	53.47%	44.02%	0.26%	0.47%	0.26%	0.21%	0.18%	0.08%	93.50%	6.8	0.20%	0.14%	13μm	10μm

. 42.5R ordinary Portland cement (OPC), which complies with Chinese standard GB175-2007 “General purpose portland cement”, was supplied by Hubei HuaXin cement Co., Ltd. (Wuhan, Hubei, China), and OPC was used as an alternative binder for solid materials in geotechnical systems. The properties of Normal Portland cement are shown in

Table 2. Chemical and Physical Compositions of OPC.

Compositions	CaO	SiO2	Al2O3	Fe2O3	MgO	SO3	Loss on Ignition	Initial Time	Final Time	Fineness
Content	55.30%	26.04%	6.61%	4.32%	2.44%	2.33%	3.12%	130 min	170 min	20 μm

.

2.1.2. Alkali Activator

A mixture of sodium silicate and sodium hydroxide at a ratio of SiO₂/Na₂O=3.20 was added as an activator. The NaOH particles were poured into the sodium silicate solution while stirring to adjust the modulus of the experimental solution. The composition of sodium silicate is shown in

Table 3. Composition of sodium silicate.

Components	SiO2	Na2O	°Bé	Modulus
Content	27.20%	8.75%	39	3.2

. NaOH was supplied by Tianjin Kaitong Chemical Reagent Co., Ltd. (Tianjin, China) )with an analytical purity over 96.0%. The solutions were sealed and stored in the environment for 24 h prior to synthesis.

2.2. Preparation

The experimental formulation was carried out according to

Table 4. Experiment compositions.

Test Label	Cement (g)	Water (g)	Metakaolin (g)	Modulus	Activator (g)	Standard Sand (g)
GS1	160	320	640	1.5	200	1600
GS2	200	320	600	1.5	200	1600
GS3	240	320	560	1.5	200	1600
GS4	280	320	520	1.5	200	1600
GS5	320	320	480	1.5	200	1600
GS6	360	320	440	1.5	200	1600
GS7	400	320	400	1.5	200	1600
GA1	320	320	480	1.5	80	1600
GA2	320	320	480	1.5	120	1600
GA3	320	320	480	1.5	160	1600
GA4	320	320	480	1.5	200	1600
GA5	320	320	480	1.5	240	1600
GA6	320	320	480	1.5	280	1600
GA7	320	320	480	1.5	320	1600

and Figure 2. Two different geopolymeric mortars (metakaolin content and activator content) were synthesised for the tests. The powdered metakaolin and the alkali activator were mixed in a planetary mixer. The specified materials were stirred slowly for 1 min and then at high speed for 1 min. The mixed specimens were then poured into standard moulds for the different experiments, shaken and smoothed. Apparatus with dimensions of 70.7 mm × 70.7 mm × 70.7 mm were used for both the freeze-thaw and sulphate corrosion resistance experiments. Six cubic test blocks were prepared for each set of experiments and the results were averaged. The curing time was adjusted according to different experiments, and the curing chamber was set at a temperature of 20 °C and relative humidity of 90–95%. Factor 1 was the amount of metakaolin, with a ratio of 0.8 to 0.5 metakaolin to cement. Factor 2 was the amount of activator, with a ratio of 0.1 to 0.4 activator to solid. Figure 1 shows the complete experimental procedure.

2.3. Characterisation

2.3.1. Permeability Resistance

Permeability resistance not only characterises the ability of cement and concrete to resist the passage of water, but also affects their performance against other durable properties such as chloride ion penetration. The amount of water penetration through the surface of the material after a period of time under constant water pressure was used as a measure of resistance to penetration. The higher the water pressure value, the better the impermeability of the geopolymeric slurry and the more compact the specimen. The SS-15 digital display mortar impermeability meter was used for the experiment. Before the experiment started, each group of 6 mortar specimens was numbered with the specimens evenly coated with glass cement on both sides and then were loaded into the mortar impermeability meter for fixing. The initial water pressure was set at 0.2 MPa and the maximum water pressure was set at 4.2 MPa. The test could be carried out after 24 h of complete setting of the glass cement. The mortar impermeability meter automatically rose by 0.1 MPa every 1 h and the specimen was observed every 30 min. The impermeability test values were determined using Equation (1):

P = H − 0.1

(1)

where:

P = average value of the resistance to permeation of the geopolymeric specimen paste, MPa;

H = pressure value of the specimen when permeation occurs in three of the six specimens, MPa.

2.3.2. Sulphate Corrosion Resistance

Most environments, such as the sea, salt lakes and groundwater, contain sulphate ions. Sulphate corrosion is a slow process. In order to accelerate the corrosion process in the test, a high concentration sulphate solution was applied to achieve the corrosion effect. The sulphate corrosion resistance test was measured in accordance with the National Standard GB/T 749-2008 [26]. The maintained GS and GA specimens were placed in plastic jars filled with sulphate solution and immersed for 28 and 56 days, respectively. After a certain number of days of immersion, the specimens were weighed and tested for compressive strength. Resistance to sulphate corrosion was measured by the rate of weight loss and corrosion resistance coefficient of the compressive strength. The coefficient of resistance to sulphate corrosion was determined using Equations (2) and (3):

$$▲Ms=\frac{M0-M1}{M0}\times 100\%$$

(2)

where:

▲Ms = weight loss rate of geopolymer slurry, %;

M0 = weight of the geopolymer slurry before immersion, g;

M1 = weight of the geopolymer slurry after 28/56 days of immersion, g.

$$Rf=\frac{Rc}{R}\times 100\%$$

(3)

where:

R_f = coefficient of corrosion resistance of the compressive strength of the geopolymer slurry, %;

R_c = average compressive strength of the geopolymer slurry after 28 d/56 d of immersion, MPa;

R = average compressive strength of the geopolymer slurry after 28 d/56 d without immersion, MPa.

2.3.3. Freeze-Thaw Resistance

When outdoor temperatures are low, building materials are often damaged by freeze-thaw cycles. Freezing damage alone does not cause large cracks in cement or concrete buildings. When building materials are repeatedly subjected to freeze-thaw damage, small cracks would gradually expand, and the stability of the structure is greatly threatened. The freeze-thaw resistance test was carried out in accordance with National Standard JGJ/T70-2009 [27]. The specimens are tested in a freeze-thaw testing machine for 50 and 100 times in a temperature range from −15 °C to 20 °C ± 2 °C. The freeze-thaw resistance test is reflected by the weight loss rate and the compressive strength loss rate, as shown in Equations (4) and (5):

$$▲M=\frac{Mfz0-Mfz1}{Mfz0}\times 100\%$$

(4)

where:

▲M = weight loss rate after 50/100 times freeze-thaw test, %;

M_fz0 = weight of specimens before the freeze-thaw test, g;

M_fz1 = weight of specimens after 50/100 times freeze-thaw test, g.

$$Cs=\frac{C0-C1}{C0}\times 100\%$$

(5)

where:

Cs = compressive strength loss rate, %;

C0 = average compressive strength of the control specimens, MPa;

C1 = 50/100 times the average compressive strength of the experimental specimen, MPa.

2.3.4. Bond Strength

The main indicator of the bond strength of a geopolymer mortar repair material is the flexural strength between the old and new materials. The 28 d flexural strength was tested for both variables according to a previous method [28].

2.3.5. Scanning Electron Microscopy

The S-3000N scanning electron microscope, manufactured by HITACHI, was used to obtain the SEM images of specimen paste. The resolution was up to 5 nm. At the same time, an experimental study was carried out using an E-1045 gold spraying instrument. After 28 days of curing of the specimen paste, small pieces of paste were taken at the centre of the specimen and placed in a constant temperature drying oven to be dried at 70 °C for 2 days. The dried specimens were put into the gold spraying instrument for surface gold spraying. After the specimens were processed, they were placed under a scanning electron microscope for microstructure observation.

2.3.6. Fourier Transform-Infrared Spectroscopy

The Nicolet Avatar 330 spectrometer, manufactured by Thermo Fisher Technology Co., was used to measure FT-IR. In the experiment, small pieces of specimen from the centre of the cement pastes were put into a laboratory bowl and ground to powder. A small amount of specimen was then mixed with KBr powder at a ratio of 1:100. The KBr powder should be stored in an IR drying oven to keep it dry. The FT-IR scanning range is 400 cm⁻¹ to 4000 cm⁻¹.

3. Results and Discussion

3.1. Permeability Resistance

Figure 3 shows the images of the cement mortar and the metakaolin geopolymer slurry in the anti-permeability experiments under two different significant influencing factors. The permeation pressure value for the control group was 3.3 MPa. The permeation pressure values of the moulded specimens increased and then decreased as the dose of metakaolin, and alkali activators was varied under the two different influencing factors. At GA5/GS5, the maximum value of permeation pressure was reached for both groups of specimens; it was also observed that the dose of the activator has a greater effect on the impermeability than the metakaolin doses before GA5/GS5. The change in activator content after GA/5GS5 also had a greater effect on the impermeability than the change in geopolymer contents. The water absorption of metakaolin was relatively high. For the same mass of metakaolin and cement, the density of metakaolin was smaller than that of cement. Thus, the volume of metakaolin was larger and the porosity was higher when more metakaolin powders were mixed to form specimens. Due to the high-water absorption of the metakaolin, there were more connected pores inside, and thus, during the experiment, water permeated into the interior first and the impermeability pressure values were small. When the mixture of metakaolin was gradually reduced and filled with cement, the formerly connected pores gradually changed to closed pores. Under such circumstances, water needed to pass through the closed pores to reach the interior of the specimen during the experiment, and the closed pores are more difficult to be penetrated by water than the connected pores, so the impermeability increases. The content of metakaolin became smaller as the GS7 specimen; the slurry formed by geopolymer cement mortar approximated the slurry formed by cement mortar; and the former’s permeability resistance pressure gradually approached to the latter.

On the other hand, an increase in the number of activators has a significant effect on impermeability. The amount of activator is directly related to the degree of denseness of the geopolymer mortar specimen. When the activator content is low, the geopolymer material does not have a sufficiently alkaline environment during the reaction of the metakaolin, only some of the material undergoes geotechnical polymerisation and the reaction produces substances that are silicate oligomers and the arrangement of the material in the mortar structure is not dense enough. There are more pores between these substances, which are interconnected. Water penetrates through the pores into the interior of the specimen, so that the permeation pressure of the geopolymer mortar is lower when the activator content is lower. Olugbenga Ayeni et al. [29] came to a similar conclusion experimentally, where an increase in alkaline concentration increased the solubility of Si and Al and these ions formed a denser structure through the reaction. The higher activator content will react sufficiently to form the structural units of the silicate aggregates. The highly dense structure makes it difficult for water to penetrate inside the structure [30]; this effectively enhanced anti-permeability. Excessive excitation caused the hardening of the retained Na₂SO₃ in the internal structure. At the junction of the Na₂SO₃ hardened part and geopolymer structure, the retained Na₂SO₃ also wraps around part of the geopolymer structure, and water enters the geopolymer mortar specimen, reducing the permeation pressure and leading to poor permeability resistance.

3.2. Resistance to Sulphate Corrosion

The resistance to sulphate corrosion was measured based on 28 days and 56 days weight loss rate and compressive strength corrosion resistance coefficients. The results are shown in Figure 4 and Figure 5. As can be seen from Figure 4, the cement paste is also highly resistant to sulphate corrosion. After 28 and 56 days of immersion in a 5% sodium sulphate solution, the weight loss rates of the cement mortar were 0.76% and 1.52%, respectively; it could be observed in Figure 4 that the weight loss rate showed a decreasing and then increasing trend after 28 and 56 days of solution immersion, regardless of whether the amount of metakaolin or activator was changed. At the same time, the weight loss rates for both influencing factors were similar or better than those for cement mortars at GA4/GS4 to GA6/GS6. The metakaolin content played a major role in the sulphate corrosion experiments. Reducing a certain amount of metakaolin and increasing a certain amount of activator content can enhance the resistance of geopolymer cement mortars to corrosion. Furthermore, reducing the amount of metakaolin had a greater effect on increasing the activator change in sulfate corrosion resistance. Figure 6 and Figure 7 shows the corrosion coefficient of compressive strength for the two influencing factors. Same as Figure 4, two factors had positive influences on GA/GS4 to GA/GS6 contents. When metakaolin is chosen as the main variable, the loss of strength is caused by the fracture of the molecular structure [22] and the formation of crystalline compounds [31]. In this process, the geopolymer reacted with Ca²⁺ generated by the hydration products of the cement [32,33]. Olugbenga Ayeni et al. and X. Chen et al. [29,33] have also shown experimentally that excess calcium ions negatively affect the structural compactness to a large extent, thus reducing strength. The high content of metakaolin makes a difference in enhancing geotechnical polymerisation reactions. In a sulphate environment, the erosion mechanism of the geopolymer varies depending on the calcium content [15]. Due to the low Ca²⁺ content, a small number of N-A-S-H gels were formed, causing some metakaolin to separate from the alkaline activator before the reaction occurred. The internal structure of the geopolymer was not dense and more internal pores allowed easy access to sulphate ions. The change in weight may be caused by the high porosity [34]. The generation of [SiO₄] and [AlO₄] combined with the cement hydration products increased the internal structural denseness of the specimens, and as the amount of cement increased, more Ca²⁺ was involved in the reaction. The amount of metakaolin gradually decreased and the amount of Ca(OH)₂ generated by the reaction increased, which could react with SO₄^2-, thus reducing the corrosion resistance.

When the alkali activator was chosen as the main variable, the lower activator content and less precursors produced by the geopolymerisation of the metakaolin in an alkaline environment resulted in less calcium silicate gel. The internal structure of the oligoaluminate formed by the inadequate reaction was not dense enough, causing sulphate ions to enter the inner space of the specimen easily. As the activator content continued to increase, the number of precursors gradually increased, as did the calcium silicate gel. Both [SiO₄] and [AlO₄] produced condensation polymerisation reactions with the silicate oligomers, which then produced aluminosilicate polymorphs [35]; this improved the denseness of the geopolymer mortar, preventing the sulphate ions from entering the interior of the mortar specimen easily. Due to its own alkalinity, excessive incorporation of activator affected the reaction process of the system, preventing sulphate ions from leaching into the internal structure while resulting in more pore space and a reduction in strength and weight.

3.3. Freezing-Thawing Resistance and Appearance

Figure 6 and Figure 7 illustrate the loss rates of weight and compressive strength after 50 and 100 freeze-thaw tests. Figure 8, Figure 9, Figure 10, Figure 11 and Figure 12 show the appearance of the specimens.

As can be seen in Figure 8, Figure 11 and Figure 12, the higher content of the metakaolin specimens showed more porosity and cracks. Figure 12 shows that some of the specimens show flaking at the edges and this has led to an increase in weight loss. With a 40% decrease in metakaolin content, as in GS5, only a small number of pores appeared on the surface of the specimens and the surface was relatively smooth. Specimens with different activator contents are depicted in Figure 9 and Figure 10. The surface of geopolymer mortar specimen with a small activator content was damaged and had more pores, while the surface of geopolymer mortar test blocks with an activator to solids ratio of 0.3 had no obvious damage and only a small amount of slurry was lost. The above comparative analysis showed that the freeze-thaw resistance of geopolymer mortar was influenced by the content of metakaolin and the content of the activator. The appropriate content of both factors helped improve the freeze-thaw resistance of the geopolymer system.

In Figure 6, as the number of freeze-thaw cycles increases, an increase in the weight loss of the specimens was observed, while the change in the contents of metakaolin and activator caused the weight loss of geopolymer mortar to first decrease and then increase. When the test ratio reached GA5/GS5, the freeze-thaw weight loss rate was lower than the cement specimen. The most important cause of weight loss was water expansion in the mortar, which was permeable while freezing occurred [36]. The onset of freezing led to an increase in water volume of approximately 9% and due to this effect, hydraulic pressure was generated [37]. The high content of metakaolin had a high-water absorption, which allowed more H₂O to fill the internal structure and the hydraulic pressure was greater than in specimens with less metakaolin. The force of the frozen ice exceeded the strength of the mortar; micro-cracks began to form; and weightless deterioration occurred. As a result, the compressive strength decreased, as shown in Figure 7. A. Allahverdi [38] and L. Basheer [39] also verified this interpretation and came to similar conclusions.

Other conclusions that can be drawn from Figure 7 are that the freeze-thaw resistance of the specimens improved as the content of metakaolin decreased. The presence of large amounts of calcium ions in the cement produced hydrated calcium silicate in the hydration reaction. Metakaolin is essentially an amorphous silicaalumina compound [19], which underwent progressive chain-end degradation catalysed by alkali-active cementitious materials, and the hydrated calcium silicate underwent condensation reactions with [SiO₄] and [AlO₄]. The hydrated calcium silicate produced by hydration reacted sufficiently with [SiO₄] and [AlO₄] in geopolymerisation and interaction, improving the internal structure of the geopolymer mortar. However, more cement influenced the geopolymerisation reaction.

Another factor in Figure 7 indicates that the magnitude of the activator content directly determines the degree of reaction of the metakaolin in the preparation of the geopolymer. Guanglong Yu et al. [16] also pointed out that an effective activator had a filling effect on the internal pore structure. When the activator content was low, the reaction of metakaolin with the alkali activator formed an oligoaluminate structure, which is less dense. On the other hand, the incomplete mixing of the metakaolin and the alkali activator due to the low activator content resulted in the presence of excess metakaolin in the specimen. Ahmadreza Mazaheri et al. [40] concluded that increasing the alkalinity of the pore solution by the appropriate addition of sodium oxide promoted the dissolution of the particles, thus facilitating the polymerisation reaction and the formation of the structure. However, further addition did not increase the optimisation of the pore solution. During freeze-thaw cycles, residual metakaolin tends to adsorb water molecules, resulting in poor freeze-thaw resistance of the geopolymer mortar. As the content increased, the metakaolin reacted sufficiently with the alkali activator and the linkages between these reaction products were mainly based on covalent and ionic bonds, forming a multi-polymer aluminosilicate structure that improved the denseness of geopolymer mortar and thus enhanced the freeze-thaw resistance of the geopolymer mortar. Excessive sodium silicate interfered with the reaction. During the test curing process, the sodium silicate hardened directly into a solid form, reducing the denseness of the geopolymer, and as the freeze-thaw cycle test progressed, water entered the specimen along the gap between the sodium silicate and the geopolymer mortar.

3.4. Bond Strength

Figure 13 describes the bond strength of metakaolin and activator content geopolymer of different doses. For a different dose of metakaolin (GS), with the decrease in metakaolin content, the bond strength increments of geopolymer paste were −16.00%, −7.43%, −1.14%, 3.43%, 12.00%, 6.86%, 4.00% compared to cement paste. Whereas for different doses of activator (GA), as the activator content increased, the bond strength of the geopolymer paste increased by −13.71%, −6.29%, 2.86%, 9.71%, 16.00%, 10.86%, and 6.85%, compared to the cement paste. The bond strength of the geopolymer mortar was mainly based on friction and chemical bonding forces [22]. At the macro level, the repair material wrapped the bonding surface with an uneven base, creating frictional forces on each other. Furthermore, the fineness of the source material has an effect [41]. At the microscopic level, the dense interfacial transition zone between the aggregate and the geopolymer paste gives the geopolymer a high bond strength [42].

3.5. Scanning Electron Microscopy

The two influencing factors topography of metakaolin geopolymer cement pastes durability and control cement paste topography are shown in Figure 14, Figure 15, Figure 16, Figure 17, Figure 18, Figure 19 and Figure 20. As shown in Figure 16, Ca(OH)₂, C-S-H and AFt were the main hydration products. From the extensive literature [43,44,45,46], it appears that the hexagonal ones are calcium hydroxide, while calcium alumina is elongated and acicular, and C-S-H is fluffy and fibrous. Figure 17, Figure 18 and Figure 19 images were different activator contents geopolymer pastes. At lower activator contents, the internal structure of the geopolymer was not homogeneous and a large amount of material was distributed in clusters. At the same time, a large number of metakaolin particles can be found encapsulated and not involved in the reaction, and the structural morphology in the tomographic section was different from the porous structure; this affected the performance of the metakaolin polymer in terms of durability. The porous structure made it easier for water to flow inwards and leak during the permeability test. As can be seen in Figure 5, the resistance to permeability was poorer at this point. Furthermore, for freeze-thaw cycles and resistance to sulphate corrosion, the water acted as a carrier due to the loose internal structure, carrying harmful ions into the interior and damaging the structure. The volume expanded when the frozen water cracked the test specimens. Figure 18 shows that the structure of the geopolymer was evenly distributed and densely arranged, with the metakaolin essentially completely reacting. The pores disappeared gradually, and the gel material wrapped around the particles, increasing the internal contact area and interconnecting the individual materials, thus improving the denseness of the structure. The polymetallic silicates are attached to the internal structure, giving the geopolymer better performance durability. As shown in Figure 19, the strength of the sodium silicate decreased as the activator content increased and the sodium silicate became a hardened body. At the same time, more pores appear in the internal structure, further undermining the durability of geopolymer pastes.

Figure 18, Figure 19 and Figure 20 showed another factor: the metakaolin content. Due to the high-water absorption of metakaolin, there were many flaky metakaolin particles on the surface of the gel material which were not involved in the reaction and were loosely packed. As with the low activator, the geopolymer performed poorly in terms of durability appearance. More porous and unreacted materials are presented in Figure 18. The metakaolin content was reduced, and the specimen images are shown in Figure 19 and Figure 20. The appropriate amount of cement was involved in the geopolymerisation reaction, improving the setting time of the geopolymer to achieve rapid hardening, and the hydrated Ca(OH)₂ intertwined with the aluminosilicate in the geopolymer to form a network structure, making the geopolymer denser. As can be seen in Figure 20, the cement and the metakaolin are mixed and the arrangement of some of the structures were tight.

3.6. Fourier Transform-Infrared Spectroscopy

The FT-IR images of two influencing factors of geopolymer are shown in Figure 21 and Figure 22. The FTIR absorbance peaks and their compositions are summarised in

Table 5. FTIR absorbance peaks for GA specimens.

Wavemunber Position [cm−1]			Bond Component
GA1	GA5	GA7
460.03	457.40	455.73	Bending of Si-O bonds [47]
975.70	977.70	979.70	Stretching of Si-OH bonds [47]
1423.59	1428.93	1427.86	Bending of O-C-O bonds ill CO32-ions [48]
1637.45	1640.96	1641.22	Bending of H-O-H bonds [47]
2925.74	2925.87	2927.43	Stretching of C-H bonds [49]
3446.46	3445.96	3453.60	Stretching of OH- groups [47]

and

Table 6. FTIR absorbance peaks for GS specimens.

Wavenumber Position [cm−1]			Bond Component
GS1	GS5	GS7
462.05	458.31	457.58	Bending of Si-O bonds [47]
559.31	554.71	540.01	Bending of Si-O-Al bonds [50]
973.54	977.97	978.00	Stretching of Si-OH bonds [47]
1097.97	1103.35	1096.00	Stretching of SiO42- units [49]
1431.75	1428.33	1428.53	Bending of O-C-O bonds in CO32- ions [48]
2927.14	2925.61	2927.01	Stretching of C-O bonds [49]
3449.44	3456.17	3455.19	Stretching of OH- groups [47]

. In Figure 22, with the change of metakaolin content, the peak area of 3449.44 cm⁻¹ band became progressively slower as the content of metakaolin changed. By contrast, the absorption peak area of O-H decreased and the Ca(OH)₂ produced by hydration was involved in the geopolymerisation during depolymerisation and condensation, consuming a large amount of Ca(OH)₂ and crystalline water. Figure 22 shows a peak near the 2927 cm⁻¹ band, which is the C-H stretching vibration on saturated carbon. At the 1097.97 cm⁻¹ band, the absorption peak gradually shifted from the high wave number band to the low one. The Si-O bond was replaced by the Al-O bond. The newly generated geopolymer structure gave rise to Si-O-Al bonds. Figure 21 shows the different activator contents. As the activator content increased, the shape of the 795.70 cm⁻¹ absorption peak gradually narrowed and became steeper. The formation of polyaluminosilicate structures by the geopolymer reaction steepened the peaks here. 1097 cm⁻¹ band absorption peaks are gradually shifted to the lower wavenumber band where different tetrahedral structures were combined to form new substance.

By comparing Figure 21 and Figure 22, the FT-IR showed a similar chemical bonding. This suggests that the two influences have a similar effect on the chemical bonds formed internally; these two factors altered the durability of the geopolymer through different actions. The microscopic experiments also provide further evidence for the macroscopic ones.

4. Conclusions

In this research, metakaolin and activator contents, the two main factors, were involved in experiments to study the durability of repair metakaolin geopolymer under different factors. Macroscopic experiments were carried out to determine the visual impact. More importantly, microscopic experiments, including SEM and FT-IR justified and explained the macroscopic phenomena. The following conclusions can be drawn:

The experiments tested durability, including bond strength, resistance to penetration, resistance to sulphate corrosion and resistance to freezing and thawing. Indicators including flexural strength, maximum pressure value, weight loss, compressive strength, and corrosion resistance coefficient were used to measure the performance of the repair material; it can be found that a suitable ratio of metakaolin and activator content significantly improves the durability and repairability of the geopolymeric cement. When the ratio of metakaolin to cement is 1.5 and the ratio of activator to solids is 0.3, the geopolymer are the most durable and can be used as repair materials and pavements.

From the SEM and FTIR, it was possible to see more clearly the effect of different factors changes on the internal structure of the forming geopolymer, so that the reasons affecting the durability changes could be reasonably analyzed. When the ratios of metakaolin to cement and activator to solids are appropriate, the geopolymer formed had a good internal structure with low porosity; this effectively prevented the entry of harmful ions and reduced the effects of crystallisation due to temperature reduction and volume expansion during freeze-thaw which destroys the internal structure and therefore provides better durability properties.