Effect of Metakaolin on the Water Resistance of Magnesium Phosphate Cement Mortar

Abstract

In order to study the effect of metakaolin on the water resistance of magnesium phosphate cement mortar, we took four factors as research objects: the setting time, fluidity, compressive strength, and water resistance of magnesium phosphate cement mortar. These were studied after adding equal gradations of metakaolin, and we then carried out SEM microcosmic experiments on the best group. The results showed that with the increase in metakaolin content, the setting time of magnesium phosphate cement decreased gradually, and the fluidity of the mortar decreased as a whole. The effect of metakaolin on the fluidity of the magnesium phosphate cement mortar was not significant when the content of metakaolin was less than 8%. Metakaolin was able to improve the pore structure of magnesium phosphate cement mortar; for example, it improved its compressive strength and water resistance due to both chemical and physical interactions. The water resistance of mortar samples in the same timeframe increased first and then decreased with the increase in metakaolin content. When the content of metakaolin was 12%, the water resistance of magnesium phosphate cement mortar achieved the optimal results. A 50.32% improvement of the 56 d strength retention rate and a 54.89% improvement of the 90 d strength retention rate proves its effective long-term water resistance.

1. Introduction

Magnesium phosphate cement (MPC) is a new-type inorganic hydraulic cementitious material, which is produced via chemical interactions between a mixture of MgO, NH₄H₂PO₄, Na₅P₃O₁₀, Na₂B₄O₇·10H₂O and other raw materials in certain proportions, and water. It has advantages such as rapid setting, early high strength, strong adhesion, good wear resistance, good heat resistance, obvious frost resistance, etc. [1,2,3,4,5,6]. It has been widely used in biological materials, waste treatment, construction materials, etc. [7,8,9,10,11,12,13]. However, in the process of its use, it has been found that magnesium phosphate cement has poor water resistance [9,14,15,16,17,18]; this greatly restricts its application in many scenarios [9,14,15,16,17,18]. Some scholars have performed research on improving the water resistance of magnesium phosphate cement mortar. A simple and effective way is to add admixtures and change the crystal structure of magnesium phosphate cement. One of the more common methods is to use fly ash to improve the water resistance [19,20,21,22,23,24,25]. It has been found that with an increase in fly ash content, the setting time of magnesium phosphate cement first decreases and then increases, and the fluidity first increases and then decreases, whereas the compressive strength decreases. In addition, some scholars have used other admixtures to improve the performance of magnesium phosphate cement or have used new materials [26]. Li Dongxu et al. [27] analyzed the influence of different raw material proportions on the water resistance of MPC and the composition and microstructure of hydration products of MPC under different curing conditions. They found that the effect of raw material proportions on the water resistance of MPC was obvious, and the effect of phosphate content on the water resistance of MPC was significant. The main reason for the poor water resistance of MPC is the dissolution of a small amount of unreacted phosphate under water curing, which changes the pH value of the aqueous solution. The main hydration product, MgKPO₄·6H₂O, dissolves in acidic conditions, which causes an increase in the void ratio and a decrease in strength. Gao Ming et al. [28] found that the water resistance of magnesium phosphate cement mortar could be greatly improved by adding micro-silica powder to the magnesium phosphate cement mortar, mainly because the density of the hydration products was improved. Shi Yaven et al. [29] and LU X et al. [30] improved the performance of magnesium phosphate cement by partially replacing magnesium oxide with metakaolin. It was believed that this method would significantly prolong the setting time of magnesium phosphate cement mortar and greatly improve its 1 h strength and bond strength. The main reason for this was that the active Al₂O₃ in metakaolin was thought to take part in the reaction and form amorphous gels, such as AlH₃(PO₄)₂·H₂O and AlPO₄, which would effectively improve the mechanical properties. Wu Qing et al. [31] studied the effects of alumina–silica mineral admixtures (metakaolin and illite) on the physical and mechanical properties of MKPC by using heavy burned magnesium oxide, potassium dihydrogen phosphate, composite retarder, air-entraining type water reducer, and quartz sand as the main raw materials to produce MKPC. The results showed that the active substances SiO₂ and Al₂O₃ in metakaolin and illite could promote hydration and form AlPO₄ and Al(PO₃)₃ gels, which caused the hydration product and whole structure to become more compact and led to an obvious filling effect. Fu Xinyu et al. [32] compared the effects of metakaolin and fly ash on the bonding performance between potassium magnesium phosphate cement (MKPC) slurry and concrete through a four-point bending test and oblique shear test. The results indicated that the addition of metakaolin and fly ash could effectively improve the bonding strength and compressive strength of MKPC slurry. Fu Cuihong [33] studied the performance of mineral powders (fly ash, silica fume, metakaolin) on magnesium phosphate cement mortar (MKPC). The results showed that the setting time and fluidity of fly ash magnesium phosphate cement mortar were slightly improved. The addition of 15% silica fume could significantly improve the setting time, fluidity, and compressive strength of magnesium phosphate cement mortar. The addition of 10% metakaolin shortened the setting time of MPC mortar and reduced the fluidity and compressive strength, but was beneficial for improving the flexural strength. Hao Kuo et al. [34] studied the effects of admixtures on the microstructure and water resistance of magnesium phosphate cement in both static and dynamic water environments. The results showed that the modification of fly ash, Portland cement, and silica fume could improve the water resistance of magnesium phosphate cement, with silica fume working best. Bai Weiliang et al. [35] conducted an experimental study on the modification of magnesium phosphate cement using metakaolin and hollow glass microspheres and found that the optimal dosage of metakaolin and hollow glass microspheres was 0.5–0.8 and 8%–12%, respectively. Lin Ludan [36] conducted an experimental study on the water stability of MPC by regulating water consumption; the results showed that the degree of hydration reaction and the water maintenance system jointly affect the water stability of MPC. Fatheali A. Shilar et al. [37] used marble as a binder and MK to produce geological polymer concrete (GPC) and evaluated the structural performance of polymer concrete in metakaolin soil. The results showed that soluble silicate can accelerate the dissolution of metakaolin soil. T. P. P. Coelho et al. [38] used metakaolin and calcined diatomaceous earth to prepare silicate-free geopolymers with enhanced low-temperature performance. The research results showed superior physical and mechanical properties were achieved by using metakaolin in the preparation of geopolymers. Liu Runqing et al. [39] utilized metakaolin to improve the processability and mechanical properties of MPC and studied the static and dynamic mechanical properties after high-temperature treatment, as well as the hydration products of MPC. The results indicated that adding metakaolin could prolong the setting time of MPC and significantly improve its high-temperature impact resistance. The method of adopting appropriate additives to modify MPC not only improves the performance of MPC but also reduces the production cost of MPC. The modification effect of admixtures on MPC has been attributed to the physical and chemical effects caused by admixtures reacting with MPC, such as the adsorption effect, filling effect, activity effect, etc. These effects influence the hydration degree and final strength of MPC. Metakaolin, as a mineral admixture with finer particles and stronger activity, has attracted significant attention.

Metakaolin (MK) is a highly active volcanic ash and is produced from kaolinite by high-temperature calcination (at temperatures ranging from 700 to 900 °C) and is used as a cement substitute for traditional concrete [37,40]. Compared with Portland cement, MK has a smaller particle size of about 1–2 μm, which is larger than that of silicon powder [41] and a larger specific surface. Compared with ordinary Portland cement, MK has a smaller particle size, larger surface area, and faster reaction speed [42]. Many scholars have conducted research on modifying magnesium phosphate cement mortar with metakaolin (MK), but there are relatively fewer studies on the long-term water resistance of the resulting product. With regard to the lifespan of building materials, long-term water resistance is very significant. The authors of this paper considered MK as a mineral admixture and added it into magnesium phosphate cement mortar. Then, we studied the effects of MK content on the setting time, fluidity, compressive strength, and water resistance of magnesium phosphate cement in order to provide a reference for the use and further research of magnesium phosphate cement.

2. Test Content

2.1. Raw Materials and Proportions

2.1.1. Reburned Magnesium Oxide Powder

Reburned magnesium oxide powder (MgO, hereinafter referred to as M), with a light-yellow color, was produced in Yingkou City, Liaoning province, China. It was directly calcined from magnesite (at 1800 °C for the calcination temperature, CO₂ escape point), with a purity of 97% and a particle size range of 200 mesh. The specific chemical composition is shown in

Table 1. Chemical composition of magnesium oxide.

Materials	MgO	SiO2	Al2O3	Fe2O3	CaO	Others
M	95.0	2.1	0.4	0.4	1.7	0.4

.

2.1.2. Ammonium Dihydrogen Phosphate

Ammonium dihydrogen phosphate (NH₄H₂PO₄, hereinafter referred to as P), a material mainly involved in hydration reactions in magnesium phosphate cement components, was produced by Chongqing Sanjiang Chemical Co., Ltd. in Chongqing City, China. It was industrial grade, with a density of 1.80 g/cm³, a purity of 98%, and a particle size range of 70–300 mesh.

2.1.3. Retarder

The retarder consisted of borax and sodium tripolyphosphate. Borax (Na₂B₄O₇·10H₂O), abbreviated as B, was produced by Dashiqiao Xinsheng Boron Industry Co., Ltd. in Yingkou City, Liaoning province, China. It was of industrial grade, with a density of 1.70 g/cm³ and a purity of over 95%. Sodium tripolyphosphate (Na₅P₃O₁₀), abbreviated as STP, was produced by Suzhou Xunlaili Environmental Protection Technology Co., Ltd. in Suzhou City, Jiangsu province, China. It was of industrial grade, with a density of 2.51 g/cm³ and a purity of over 95%.

2.1.4. Metakaolin

Metakaolin (MK), a white powder with a particle size of 1250 mesh, was produced by Shanxi Xingle Kaolin Co., Ltd. in Shuozhou City, Shanxi province, China. Its chemical composition is shown in

Table 2. Chemical composition of metakaolin.

Materials	SiO2	Al2O3	Fe2O3	CaO	Others
MK	51.3	45.9	0.5	0.3	2.0

.

2.1.5. Aggregate

ISO standard sand, with a particle size of 0.08–2.0 mm, was produced by Xiamen Aisiou Standard Sand Co., Ltd. in Xiamen City, Fujian province, China.

2.1.6. Water

Tap water is abbreviated as W.

2.1.7. Mix Proportion

The mix proportion of magnesium phosphate cement mortar is shown in

Table 3. Mix ratio of magnesium phosphate cement mortar.

Group Number	Mix Ratio (%)		Mass Ratio		STP/M (%)	B/M (%)
	(M + P + B + STP)	MK	P/M	S/M
MK0	100	0	1/3.3	1/1.1	4.7	2.2
MK4	100	4	1/3.3	1/1.1	4.7	2.2
MK8	100	8	1/3.3	1/1.1	4.7	2.2
MK12	100	12	1/3.3	1/1.1	4.7	2.2
MK16	100	16	1/3.3	1/1.1	4.7	2.2

Note: Water-solid ratio is 0.15.

.

2.2. Manufacture of Test Samples

First, magnesium oxide, ammonium dihydrogen phosphate, borax, sodium tripolyphosphate, standard sand, and metakaolin were added into a mixing pot in proportion and stirred evenly. Then, water was added into the mixing pot and mixed for 90 s, and the fluidity and setting time of the mixed mortar were tested. Next, we loaded the mixture into a 40 mm × 40 mm × 40 mm mold, placed it on a vibration table, and vibrated it 120 times to give it a compact form. After 1 h, we demolded it and transferred it for natural curing and water curing, respectively.

2.3. Test Methods

2.3.1. Setting Time Test

The setting time test was carried out in accordance with GB/T 1346-2011 [43] “Test methods for water requirement of normal consistency, setting time, and soundness of portland cements”. The setting time was measured with a Vicat apparatus, as shown in Figure 1. The testing process was as follows: record the time when all cement is added into the water as the starting time t₁ of the setting time; place the test mold containing the specimen under the test needle, lower the height of the test needle in order contact it with the cement slurry; tighten the screws for 1–2 s, and then relax. Observe the reading of the test needle when it stops sinking or releasing for 30 s. When the test needle is (4 ± 1) mm away from the bottom plate, the cement has reached the initial setting state. Record the time t₂ at this time. The initial setting time is (t₂ − t₁). The setting time is measured every 30 s, and the results are timed by the minute. Due to the short interval between the initial and final setting times of magnesium phosphate cement, the initial setting time is used as the final setting time. Test each mix ratio three times, and take the average of the three times as the setting time.

2.3.2. Fluidity Test

The fluidity test was conducted according to GB/T 2419-2005 [44] “Test method for fluidity of cement mortar”, and the fluidity of magnesium phosphate mortar was tested with the NLD-3 cement mortar fluidity tester. The drop distance was 9.8 mm to 10.2 mm, and the vibration frequency was 1 Hz. The testing process was as follows: first, place the mortar in a copper mold in the center of the cement mortar fluidity tester, remove the copper mold, and continuously vibrate 25 times. Then, use a caliper to measure the diameter of the mortar test cake and its vertical direction. Next, calculate the average value and round it to obtain the cement mortar fluidity of the water amount. Finally, test each mix ratio three times and take the average of the three times as the fluidity of the cement mortar.

2.3.3. Curing Conditions

The specimen was cured under two different conditions: (1) natural curing conditions, temperature of 19 °C to 21 °C; (2) water curing conditions, temperature of 19 °C to 21 °C.

2.3.4. Compressive Strength Test

The strength test was conducted according to GB/T 17671-1999 [45] “Method of testing cements-determination of strength”. The compressive strength of magnesium phosphate mortar was tested with a TYE-300 press, and the peak compressive strength was obtained. The size of the test piece was 40 mm × 40 mm × 40 mm. The maximum test force was 300 kN, the accuracy was ±1%, and the loading speed was 2.4 kN/s. We found the compressive strength values at test ages of 1 hour (1 h), 1 day (1 d), 7 days (7 d), 28 days (28 d), 56 days (56 d), and 90 days (90 d). We tested one set of six blocks per mix ratio and age and took the average of the six results as the compressive strength value.

2.3.5. Water Resistance Test

The mortars were molded in a 40 mm × 40 mm × 40 mm mold and demolded after 1 h after they were first cured at room temperature for 1 day. We placed each of them in indoor natural environments and water environments for 56 and 90 days and then began the compressive strength tests. Specimens cured in water were taken out 3 h before testing and placed in a natural environment for drying. In an indoor natural environment, we tested one group of six blocks per mix ratio and age and took the average of the six results as the compressive strength value. After water curing, we tested one group of six blocks per mix ratio and age and took the average of the six results as the compressive strength value.

2.3.6. Microcosmic Testing

After the specimens were naturally cured and cured in water to the specified age, they were crushed. A specimen with a length, width, and height of about 8 mm was selected, soaked in anhydrous alcohol to terminate hydration, and dried for 48 h. Then, the specimen was sanded with sandpaper, and its surface was treated with gold spray, followed by SEM testing.

3. Results and Discussion

3.1. Analysis of Setting Time and Fluidity

Figure 2 shows the setting time and mortar fluidity curves of magnesium phosphate cement with different amounts of metakaolin.

From Figure 2, it can be seen that in terms of setting time, the setting time of the MK0 group without metakaolin was 21.2 min. When the metakaolin content was 4%, 8%, 12%, and 16%, the corresponding setting times were, respectively, 19.5 min, 17.0 min, 16.0 min, and 14.2 min. Compared with MPC mortar without metakaolin, the setting times were, respectively, shortened by 8.0%, 19.8%, 24.5%, and 33.0%, which indicated that with the increase in metakaolin content, the setting time was gradually shortened. Related studies have also shown that metakaolin has a large specific surface area and strong water absorption. With the increase in the dosage, the amount of water absorbed by metakaolin increased, and the water–cement ratio decreased. In addition, a large amount of active Al₂O₃ in metakaolin was able to react with P to form cementitious substances such as AlH₃(PO₄)₂·H₂O and AlPO₄. Within a certain range of dosage, the setting time of cement can be shortened [46]. Fu Cuihong [33] conducted an experimental study on the effect of metakaolin on the setting time of MPC and found that metakaolin is a highly active substance that can accelerate the hydration of MPC mortar and reduce its setting time. In summary, within a certain range, with the increase in metakaolin content, the setting time of MPC mortar will gradually decrease.

In terms of fluidity, the fluidity of the MK0 group was 204 mm, the fluidity of the MK4 group was 206 mm, and the fluidity of the MK8 group was 203 mm. When the content of metakaolin was less than or equal to 8%, the fluidity remained basically unchanged, which indicates that a small amount of metakaolin has little effect on the fluidity of MPC. When the content of metakaolin was 12% and 16%, the corresponding fluidity was 190 mm and 163 mm, respectively. The fluidity decreased rapidly with the increase in metakaolin content. Compared with the basic group MK0, the fluidity of the MK12 and MK16 groups decreased by 6.9% and 20.1%, respectively. In the MK16 group, the decrease in fluidity was particularly significant. Therefore, when the content of metakaolin was small (≤8%), the effect on the flow performance was not significant; when the dosage was large (>8%), it led to a decrease in the fluidity of MPC. There are two reasons for this: firstly, the specific surface area of metakaolin is larger than that of reburned magnesium oxide powder, which has a strong adsorption effect on water molecules. The increased water demand, coupled with the generated cementitious substances, leads to a significant decrease in fluidity. Secondly, due to the irregular sheet-like structure of MK, it cannot produce ball-shaped effects like FA [1,47].

3.2. Compressive Strength Analysis

Figure 3 shows the effect of different amounts of metakaolin on the compressive strength of magnesium phosphate mortar at different ages.

From Figure 3, it can be seen that:

(1) The compressive strength of equigradient metakaolin mixed with magnesium phosphate cement mortar gradually increases with age. (2) In terms of the 1 h strength, the 1 h strength of the MK0 group was 45.04 MPa, while the 1 h strength of the MK4, MK8, MK12, and MK16 groups was 45.15 MPa, 46.55 MPa, 49.07 MPa, and 44.45 MPa, respectively. The strength of the MK16 group reached over 40 MPa after 1 day, and with the increase in the metakaolin content, the strength showed a trend of first increasing and then decreasing after 1 h. The strength of the MK16 group showed a reverse contraction, which was lower than that of the MK0 group during the same period. (3) In terms of 1 d strength, the 1 d strength of the MK0 group was 60.03 MPa, while the 1 d strength of the MK4, MK8, MK12, and MK16 groups was 60.79 MPa, 66.27 MPa, 64.04 MPa, and 58.45 MPa, respectively. With the increase in metakaolin content, the 1 d strength showed a trend of first increasing and then decreasing. The strength of the MK12 and MK16 groups showed an inverted trend, with the strength of the MK16 group being lower than that of the MK0 group during the same period. (4) The 1 h strength of the MK0, MK4, MK8, MK12, and MK16 groups reached 75.0%, 74.3%, 70.2%, 76.6%, and 76.0% of the 1 d strength, respectively. The 1 h strength difference in the five groups of materials was not very significant, and the strength and development speed were basically equivalent. (5) In terms of the 56 d strength, the 56 d strength of the MK0 group was 80.99 MPa, while the 56 d strength of the MK4, MK8, MK12, and MK16 groups was 87.68 MPa, 92.15 MPa, 94.97 MPa, and 97.13 MPa, respectively. With the increase in metakaolin content, the 56 d strength showed an increasing trend. (6) In terms of the 90 d strength, the 90 d strength of the MK0 group was 92.56 MPa, while the 90 d strength of the MK4, MK8, MK12, and MK16 groups was 101.06 MPa, 104.41 MPa, 107.12 MPa, and 110.34 MPa, respectively. With the increase in metakaolin content, the 56 d strength showed an increasing trend. (7) Compared to the long-term strength (90 d), the 1 d strength of the MK0, MK4, MK8, MK12, and MK16 groups reached 64.9%, 60.2%, 63.5%, 59.8%, and 53.0% of the 90 d strength, respectively. When the content of metakaolin was less than or equal to 12%, the 1 d strength reached about 60% of the 90 d strength. When the content of metakaolin was 16%, the 1 d strength reached about 53.0% of the 90 d strength. This indicates that an increase in metakaolin content is unfavorable for early strength development and beneficial for long-term strength growth. (8) At the same age, with the increase in metakaolin content, the overall strength of magnesium phosphate cement mortar showed an increasing trend. In terms of the 1 d strength, the mortar strength of the MK12 and MK16 groups showed a decreasing trend compared to the other groups. After 7 days of aging, with the increase in metakaolin content, the strength of the MPC mortar gradually increased, indicating that metakaolin is beneficial for the long-term strength of MPC mortar. (9) Compared to the other groups, the MK12 group had the highest 1 h strength, with a balanced growth rate with age and a higher long-term strength. Compared with the basic group MK0, the MK12 mortar group showed an 8.9% increase in 1 h strength, 6.7% increase in 1 d strength, 17.3% increase in 56 d strength, and 15.7% increase in 90 d strength. Overall, a dosage of 12% was found to be optimal.

The reasons for the above phenomenon are as follows: the particle size of metakaolin is smaller than that of magnesium oxide. During the reaction process, metakaolin particles can fill cement gaps, improve the pore structure, and increase compactness. At the same time, when MgO phosphate undergoes an acid–base reaction, exothermic heat will stimulate active Al₂O₃ in metakaolin, which reacts with phosphate ions to form a cementitious material and fill the pores. In addition, it can also enhance the strength and durability of MPC by improving the microstructure of pores and increasing compactness [1]. However, a higher metakaolin content does not continue to lead to improvements. At a content of 16%, the strength in the early stage was lower than that of the MK0 group at the same age. Excessive MK will weaken the compressive strength due to the replacement of MgO with MK, reducing hydration products in the early stage. The heat released by early acid–base reactions is not enough to stimulate most of the Al₂O₃ reactions, resulting in low early strength [1]. Moreover, an excessive reaction between Al₂O₃ and P in metakaolin results in the early expansion of aluminum phosphate cementitious material. This causes increased porosity, lowering the compactness of the mortar, and lower compressive strength. The sustained and stable growth of strength in the later stage can be attributed to an increase in hydration products over time, with a stronger filling effect and low porosity. As the acid–base reaction progresses, in the middle and later stages of the reaction, the heat released stimulates the active Al₂O₃ in the metakaolin, which converts more reactants and intermediate products into hydration products. This can increase the compactness of the microstructure and is generally beneficial to the improvement of compressive strength. LIU N [48] believed that with the increase in Al₂O₃ content, the overall compressive strength would be improved. Due to the high content of Al₂O₃ in metakaolin, it can be inferred that within a certain dosage range, the addition of metakaolin is beneficial for the increase in MPC strength.

3.3. Water Resistance Analysis

Water resistance is the ability of magnesium phosphate cement mortar to resist water damage. The water resistance can be reflected directly and accurately by two indexes: mechanical properties and strength retention.

3.3.1. Effect of Metakaolin on the Mechanical Strength of Magnesium Phosphate Cement Mortar

Figure 4 shows the effect of different amounts of metakaolin on the compressive strength of magnesium phosphate cement mortar under natural and water curing conditions.

From Figure 4, it can be seen that:

(1) Under natural curing conditions, the compressive strength of magnesium phosphate cement mortar was higher than that under water curing conditions, for both the 56 d compressive strength and 90 d compressive strength. This indicates that the compressive strength of magnesium phosphate cement mortar is reduced in an aqueous environment. This may be attributed to the dissolution of unreacted ammonium dihydrogen phosphate in the magnesium phosphate cement mortar, which can result in an acidic pore water solution inside the magnesium phosphate cement mortar and the precipitation of hydration products, and then cause a loss of compressive strength. (2) Under natural curing conditions, the compressive strength of magnesium phosphate cement mortar showed an increasing trend with the increase in metakaolin content. When the metakaolin content was 16%, the compressive strength was the highest. The compressive strength of 97.13 MPa at 56 d showed an increase of 19.9% compared to the basic group MK0 group, at 80.99 MPa. The value of 110.34 MPa at 90 d was 19.2% higher than the basic group, at 92.56 MPa. (3) Under water curing conditions, the compressive strength of magnesium phosphate cement mortar specimens was significantly improved when they were mixed with metakaolin. Their compressive strength showed a trend of first increasing and then decreasing with the increase in metakaolin content. When the content of metakaolin was 12%, the compressive strength was the highest. The compressive strength of 47.79 MPa at 56 d showed an increase of 18.9% compared to the basic group MK0 group at 40.19 MPa. The compressive strength of 58.80 MPa at 90 d showed an increase of 63.1% compared to the basic MK0 group, at 36.06 MPa. This may be attributed to the following: (1) In a water curing environment, the unreacted ammonium dihydrogen phosphate in magnesium phosphate cement mortar dissolves with water, resulting in an increase in MPC pores. At the same time, it can cause an acidic environment inside the mortar and precipitation of hydration products. Due to the weak bonding force in acidic environments, these products result in the appearance of pores and microcracks, which loosen the internal structure and further damage the internal structure of the mortar. (2) When metakaolin is added, oxides such as SiO₂, Al₂O₃, and Fe₂O₃ in metakaolin will consume some phosphates and react with MPC to generate products such as Al(H₂PO₄)₃, NaAl₃H₁₄(PO₄)₈·14H₂O, and NaFeH₁₄(PO₄)₈·14H₂O. These products can fill pores to improve material density and water resistance and enhance bonding strength, which can improve the performance of MPC [49,50,51]. The addition of metakaolin can, to some extent, compensate for the strength loss of magnesium phosphate mortar under water curing conditions and improve its water resistance. (3) Qin [52] pointed out that under water curing conditions, metakaolin can provide a uniform location for the crystallization of MPC hydration products. It can promote crystal crystallization and growth, improve the compactness of the internal structure, and reduce the generation of cracks. (4) When excessive metakaolin is added, the compressive strength of the water-cured mortar specimen showed a decreasing trend. This was possibly because the large specific surface area of metakaolin can adsorb a large amount of water. Excessive metakaolin leads to a decrease in the fluidity of the specimen, an increase in internal pores, and a decrease in water resistance.

Therefore, metakaolin has a significant impact on improving the water resistance of magnesium phosphate mortar. It can have both physical and chemical effects on MPC. Under water-based conditions, as the age increases, the strength of magnesium phosphate mortar mixed with an appropriate amount of metakaolin increases more significantly. However, when the content of metakaolin is 16%, its compressive strength decreases. This indicates that metakaolin can improve the water resistance of magnesium phosphate mortar, but when the dosage is too high, the strength decreases. Therefore, the optimal dosage of metakaolin is 12%.

3.3.2. Effect of Metakaolin on the Strength Retention of Magnesium Phosphate Mortar

Compared to OPC, MPC shows severe strength loss in aquatic environments, i.e., poor water resistance. Improving water resistance is of great significance in MPC. The authors adopted the strength retention rate to reflect the influence of metakaolin on the water resistance of magnesium phosphate mortar. The larger the strength retention rate value, the better the water resistance of magnesium phosphate mortar, and vice versa. The formula for calculating the strength retention rate [53] is shown in Equation (1):

$$W_N=\frac{R_{CN}}{R_C}$$

(1)

W_N is the strength retention rate of a specimen cured in water for N days, R_CN is the compressive strength of a specimen cured in water for N days, and R_C is the compressive strength of a specimen cured at room temperature for N days.

Figure 5 shows the strength retention rate of metakaolin magnesium phosphate mortar at different dosages under two different curing conditions.

From Figure 5, it can be seen that:

(1) As the age increased, the strength retention rate of magnesium phosphate cement mortar decreased. (2) The water resistance of mortar specimens at the same age showed a trend of first increasing and then decreasing with the increase in metakaolin content. When the content was 12%, the water resistance reached the highest value, with a 56 d strength retention rate of 50.32%, which was 1.4% higher than the basic MK0 group at 49.62%. The 90 d strength retention rate of 54.89% was 38.96% higher than the basic MK0 group, an increase of 40.9%. The effect was significant. (3) By comparing the water resistance of specimens at different ages (56 d, 90 d), it was observed that the strength retention rate of specimens with a low content of metakaolin (<8%) decreased, while the strength retention rates of the MK0 and MK4 groups decreased by 21.5% and 1.9%, respectively. When the content of metakaolin soil was less than 8%, the long-term water resistance of the mortar was low, and as the content of metakaolin soil increased, the strength retention rate gradually decreased with age. Specimens with a high content of metakaolin (≥8%) showed a significant improvement in water resistance after 90 days compared to 56 days. As the content of metakaolin increased, the strength retention rate gradually increased. The 90 d strength retention rate of the MK8, MK12, and MK16 groups was greater than the 56 d strength retention rate, and the strength retention rate increased by 8.5%, 9.1%, and 16.2%, respectively. From the above results, it can be concluded that as the amount of metakaolin increases, it becomes more effective in improving the long-term water resistance of mortar. As in Section 3.3.1, the optimal amount of metakaolin was determined to be 12%. The reasons for strength loss in water curing environments were as follows: on the one hand, magnesium phosphate mortar is alkaline when it is cured in water, which can accelerate the erosion of magnesium phosphate crystals. On the other hand, the hydration products of magnesium phosphate mortar have lower stability in water, which leads to an increase in porosity and a decrease in the structural compactness of magnesium phosphate mortar. After adding metakaolin, oxides such as SiO₂, Al₂O₃, and Fe₂O₃ in metakaolin will consume some phosphate and react with MPC. This can enhance the stability of the final hydration products and improve the impermeability of magnesium phosphate mortar and its water resistance. MO L [19] and XU B [54] believed that the addition of MK could improve the pore structure and make the microstructure more dense by increasing the number of micropores, which could enhance its water resistance. In addition, because the density of metakaolin itself is smaller than that of reburning magnesium oxide, it may also reduce the density by replacing reburning magnesium oxide with equal amounts of metakaolin. However, excessive metakaolin will weaken the strengthening effect. The strength retention rate of the MK16 group was lower than that of other dosage groups, with a downward trend. This was possibly because the large specific surface area of metakaolin can adsorb a large amount of water. Excessive metakaolin led to a decrease in the fluidity of the test block, an increase in internal pores, and a decrease in water resistance.

3.4. Microanalysis

According to the above analysis, it can be seen that the optimal dosage of metakaolin is 12%. To fully elucidate the mechanism for improving the water resistance of mortar with metakaolin, microscopic analysis was conducted on the MK12 group. Figure 6a shows an electron microscope photo of MK12 cured in a natural environment for 90 days, and Figure 6b shows an electron microscope photo of MK12 cured in water for 90 days.

From Figure 6, it can be seen that there are a large number of polygonal phosphates in the pores of magnesium phosphate mortar, with a diameter of approximately 5 μm. These phosphates fill the pores of magnesium phosphate cement, giving magnesium phosphate cement mortar the advantage of rapid strength growth. Figure 6b shows an electron microscope photo of MK12 cured in water for 90 days. Compared to Figure 6a, most of the polygonal-shaped phosphates in the pores had dissolved, and a large number of cracks appeared on the surface, resulting in a decrease in strength. After magnesium phosphate cement is soaked in water for a long time, some fillers in the pores dissolve, which increases porosity. Some substances generated via hydration also decompose in alkaline water, further expanding cracks and increasing microcracks. Metakaolin has a significant effect on improving the water resistance of MPC. The increase in the strength retention rate is a result of both chemical and physical actions. The chemical action is mainly due to the active ingredients in metakaolin consuming unreacted phosphates and reacting with the MPC mortar structure, further filling the pores with the products and enhancing solid-phase bonding. The physical effects are mainly due to the finer particles and larger specific surface area of metakaolin, which can fill the internal pores of the mortar, increasing the compactness and improving the water resistance. In the water curing environment, phosphate is dissolved, and the concentration of the solution decreases, which causes a decrease in the production of the main product struvite (MgNH₄PO₄·6H₂O) and other hydration products. Therefore, a large number of cracks appear on the surface, which lowers its effects.

3.5. Mechanism Analysis of MK Modified MPC

3.5.1. Retarding Mechanism

Borax (Na₂B₄O₇·10H₂O) will ionize and decompose into B₄O₇²⁻ and Na⁺ in solution. B₄O₇²⁻ forms a precipitation film layer of MgB₄O₇ by combining with Mg²⁺ in the solution. MgB₄O₇ inhibits the dissolution of MgO and reduces the reaction rate of MPC by adsorbing on the surface of MgO. Meanwhile, the increasing effects of heat absorption and dissolution lower the hydration temperature of the mixture and delay the hydration reaction [55]. At the same time, borax, which is ionized and hydrolyzed in water, makes the solution alkaline, which, to some extent, inhibits the dissolution of MgO in acidic environments. The retarding mechanism of sodium tripolyphosphate (Na₅P₃O₁₀) is the same as that of borax. By analyzing the setting time and fluidity, the MK retarding effect was found to not be significant.

3.5.2. Hydration and Hardening Mechanism

Rouzic M L [56] pointed out that the hardening of MPC can be divided into three stages, and the related literature [16,57] suggests that Al₂O₃, Fe₂O₃, and SiO₂ in the admixture participate in the reaction of the MPC system.

By analyzing the existing literature, we found that the hydration mechanism of MPC is essentially an acid–base neutralization reaction. The entire solution reaction process is as follows: due to excessive MgO, the MPC matrix may be composed of unreacted MgO and hydration products, which are often used as cementitious materials to harden MgO in chemically bonded ceramic bodies. MPC hydration and hardening are usually divided into three stages.

Stage I: In the dissolution of phosphate (NH₄H₂PO₄), after adding water, the phosphate will immediately ionize and decompose into H⁺, NH₄⁺, H₂PO₄⁻, HPO₄²⁻ and PO₄³⁻, because it is easily dissolved in water. The solution is weakly acidic, and the ionization equation is shown in Equations (2)–(4):

$${\mathrm{NH}}_4{\mathrm{H}}_2{\mathrm{PO}}_4\rightleftharpoons {\mathrm{NH}}_4^{+}{+\mathrm{H}}_2{\mathrm{PO}}_4^{-}$$

(2)

$${\mathrm{H}}_2{\mathrm{PO}}_4^{-}\rightleftharpoons {\mathrm{H}}^{+}{+\mathrm{HPO}}_4^{2-}$$

(3)

$${\mathrm{HPO}}_4^{2-}\rightleftharpoons {\mathrm{H}}^{+}{+\mathrm{PO}}_4^{3-}$$

(4)

Stage II: MgO is dissolved in a weakly acidic solution environment. MgO itself is difficult to dissolve in water, but when it meets H⁺ in the solution, it will react to generate Mg(OH)₂ (as shown in Equation (5)), which will ionize in water and decompose into Mg²⁺ and OH⁻ (as shown in Equation (6)), causing the pH value of the solution to increase. The H⁺ generated by decomposing in (2)–(4) undergoes a neutralization reaction with the OH⁻ generated by decomposing in (6), promoting the reaction in the forward direction, as shown in Equation (7).

$${\mathrm{MgO}+\mathrm{H}}_2\mathrm{O}\rightleftharpoons {\mathrm{Mg}\left(\mathrm{OH}\right)}_2$$

(5)

$${\mathrm{Mg}\left(\mathrm{OH}\right)}_2\rightleftharpoons {\mathrm{Mg}}^{2+}{+2\mathrm{OH}}^{-}$$

(6)

$${\mathrm{OH}}^{-}{+\mathrm{H}}^{+}\rightleftharpoons {\mathrm{H}}_2\mathrm{O}$$

(7)

The ionization equilibrium constant of H₂PO₄⁻ is much smaller than that of HPO₄²⁻. Therefore, it is believed that the beginning of the reaction mainly occurs in the processes of (2) and (3), with higher concentrations of H₂PO₄⁻ and HPO₄²⁻ in the solution, while the concentration of PO₄³⁻ produced by step ionization is lower.

In addition, because MK contains highly active SiO₂ and Al₂O₃, which account for over 95% of the total composition, these active oxides will also undergo dissolution at this stage, as shown in Equations (8)–(10). During the reaction, a large amount of OH⁻ will be generated, causing an increase in the pH value of the solution. The H⁺ generated by decomposition in (2)–(4) undergoes a neutralization reaction with the OH⁻ generated by decomposition in (8)–(10), promoting the reaction in the forward direction, as shown in Equation (7). At the same time, heat release during the solution reaction process will also promote the reaction between SiO₂ and MgO to generate MgSiO₃. The reaction is shown in Equation (11):

$${\mathrm{Al}}_2{\mathrm{O}}_3{+6\mathrm{H}}_2{\mathrm{PO}}_4^{-}+3{\mathrm{H}}_2\mathrm{O}\rightleftharpoons 2{\mathrm{Al}(\mathrm{H}}_2{\mathrm{PO}}_4{)}_3{+6\mathrm{OH}}^{-}$$

(8)

$${3\mathrm{Al}}_2{\mathrm{O}}_3{+16\mathrm{H}}_2{\mathrm{PO}}_4^{-}+2{\mathrm{Na}}^{+}{+33\mathrm{H}}_2\mathrm{O}\rightleftharpoons 2{\mathrm{NaAl}}_3{\mathrm{H}}_{14}{{(\mathrm{PO}}_4)}_8\cdot {14\mathrm{H}}_2{\mathrm{O}+14\mathrm{OH}}^{-}$$

(9)

$${3\mathrm{Fe}}_2{\mathrm{O}}_3{+16\mathrm{H}}_2{\mathrm{PO}}_4^{-}+2{\mathrm{Na}}^{+}+33{\mathrm{H}}_2\mathrm{O}\rightleftharpoons 2{\mathrm{NaFe}}_3{\mathrm{H}}_{14}{{(\mathrm{PO}}_4)}_8\cdot {14\mathrm{H}}_2{\mathrm{O}+14\mathrm{OH}}^{-}$$

(10)

$${\mathrm{SiO}}_2+\mathrm{MgO}\rightleftharpoons {\mathrm{MgSiO}}_3$$

(11)

It should be emphasized that compared to MgO, it is better to replace reburned magnesium oxide powder with partial MK. Al₂O₃ in early MK will react with H₂PO₄⁻, resulting in an increase in hydration heat. Some studies have also suggested that temperature is more sensitive to the setting time of MPC. The higher the temperature, the shorter the setting time. This also partly explains why the setting time of MPC gradually decreases with the increase in metakaolin content.

Stage III: The MPC hardened body is formed. As Mg²⁺ continues to dissolve, H⁺ is consumed in large quantities. The pH value of the solution continues to rise, and the reaction continues to develop in a positive direction. PO₄³⁻ will continue to appear, and PO₄³⁻, Mg²⁺, and NH₄⁺ will interact with H₂O to form guanosine (MgNH₄PO₄·6H₂O). The reaction formula is shown in Equation (12). PO₄³⁻ reacts with Al³⁺ to generate AlPO₄ [57], as shown in Equation (13). H₂PO₄⁻ and HPO₄²⁻ react with Al³⁺ to generate the intermediate product Al₃H₃(PO₄)₂·H₂O. Al₃H₃(PO₄)₂·H₂O further reacts with Al₂O₃ in metakaolin to generate AlPO₄ [16,33], with the reaction formula shown in Equations (14) and (15). At the same time, reactions among NH₄⁺, H₂PO₄⁻, and HPO₄²⁻ gradually begin to produce products. These products continuously crystallize, interlace, overlap, grow, and bond with unreacted MgO, ultimately forming high-strength MPC. The hardened body is ultimately based on MgO as the framework, and the phosphate hydration product is the crystalline structure grid, giving good mechanical properties to the hardened body. The mechanism of action of MPC and MK is shown in Figure 7:

$${\mathrm{Mg}}^{2+}{+\mathrm{PO}}_4^{3-}+{\mathrm{NH}}_4^{+}{+6\mathrm{H}}_2\mathrm{O}\rightleftharpoons {\mathrm{MgNH}}_4{\mathrm{PO}}_4\cdot {6\mathrm{H}}_2\mathrm{O}$$

(12)

$${\mathrm{Al}}^{3+}{+\mathrm{PO}}_4^{3-}\rightleftharpoons {\mathrm{AlPO}}_4$$

(13)

$${\mathrm{Al}}^{3+}{+\mathrm{H}}_2{\mathrm{PO}}_4^{-}+{\mathrm{HPO}}_4^{2-}{+6\mathrm{H}}_2\mathrm{O}\rightleftharpoons {\mathrm{Al}}_3{\mathrm{H}}_3{{(\mathrm{PO}}_4)}_2\cdot {\mathrm{H}}_2\mathrm{O}$$

(14)

$${\mathrm{Al}}_2{\mathrm{O}}_3{+2\mathrm{AlH}}_3{{(\mathrm{PO}}_4)}_2\cdot {\mathrm{H}}_2\mathrm{O}\rightleftharpoons 4{\mathrm{AlPO}}_4+5{\mathrm{H}}_2\mathrm{O}$$

(15)

In addition to the above hydration products, many scholars have also found that magnesium phosphate hydration products contain unhydrated MgO, ammonium dihydrogen phosphate, and other hydration products such as MgSiO₃, MgNH₄PO₄·H₂O, Mg₃(PO₄)₂·4H₂O, (NH₄)₂Mg₃(HPO₄)₄·8H₂O, etc. [28,33,49,50,58]. Although there are many types of hydration products, the main hydration product is struvite (MgNH₄PO₄·6H₂O).

In summary, MK, as an admixture added into MPC, will play an important role in the process of hydration. The active components in MK will participate in the hydration reaction, during which it will consume some phosphate ions, change the pH value of the solution, promote the forward progress of the reaction, and generate some hydration products containing Si, Al, and Fe. The mechanism of action is shown in Figure 7. Meanwhile, due to the fine particle size and large specific surface area of MK, the micro aggregate effect can be stimulated by filling the internal pores of the mortar and increasing compactness. Therefore, the MPC performance can be improved by adding metakaolin.

Under water curing conditions, the dissolution rate of MgO and NH₄H₂PO₄ will be quickened with a lower solution concentration. The generated amounts of the main product struvite (MgNH₄PO₄·6H₂O) and other hydration products will be decreased, with a lower strength phenomenon. Secondly, according to existing research results, the main hydration products of magnesium phosphate cement also have a certain degree of solubility. Adding a certain amount of metakaolin not only causes chemical reactions but also exerts a micro aggregate effect, improving the impermeability of MPC and its water resistance. However, when the amount of metakaolin exceeds a certain level, the water resistance shows a decreasing trend. There may be two reasons for this: first, when the internal pores of the hardened body are filled, the surplus metakaolin finds it difficult to form a whole gel, making the strength lower. Secondly, kaolin, with a larger specific surface area and smaller particle size, can adsorb a large amount of water, leading to a decrease in water resistance. This may explain why there is an optimal dosage of metakaolin for its impact on water resistance.

4. Conclusions

In order to study the effect of metakaolin on the water resistance of magnesium phosphate cement mortar, experiments were conducted by adding equigradient metakaolin and assessing the setting time, fluidity, compressive strength, and water resistance of the magnesium phosphate cement mortar. The following conclusions were drawn:

(1)

Setting time: within a certain range, with an increase in metakaolin content, the setting time of MPC mortar gradually decreases.

(2)

Fluidity: with an increase in metakaolin content, the effect on fluidity is not significant when the metakaolin content is low (≤8%); when the dosage is larger (>8%), it will lead to a decrease in the fluidity of MPC. When the content of metakaolin is 12% and 16%, the corresponding fluidity is 190 mm and 163 mm, respectively. The fluidity decreases rapidly with the increase in metakaolin content. Compared with the basic group MK0, the fluidity of the MK12 and MK16 groups, respectively, decreased by 6.9% and 20.1%. In the MK16 group, the decrease in fluidity was particularly significant.

(3)

Compressive strength: (a) Metakaolin can effectively improve the compressive strength of magnesium phosphate mortar, and its compressive strength gradually increases with age. (b) At the same age, with the increase in metakaolin content, the overall strength of magnesium phosphate cement mortar shows an increasing trend. In terms of 1d strength, the mortar strength of the MK12 and MK16 groups showed a decreasing trend compared to the other groups. After 7 days of aging, with the increase in metakaolin content, the mortar strength of MPC gradually increased. This indicates that metakaolin is beneficial for the long-term strength of MPC mortar. (c) Compared to the other groups, the MK12 group had the highest strength after 1 h, with a regular growth rate with age and high long-term strength. Overall, a 12% metakaolin content was found to be optimal.

(4)

Water resistance: (a) With an increase in age, the strength retention rate of magnesium phosphate cement mortar showed a reverse shrinkage. (b) The water resistance of mortar specimens at the same age showed a trend of first increasing and then decreasing with the increase in metakaolin content and reached the highest value when the content was 12%. The retention rate of 56 d strength was 50.32%, and the retention rate of 90 d strength was 54.89%. The long-term water resistance effect was obvious. (c) Metakaolin can improve the pore structure of magnesium phosphate cement mortar and has a significant effect on improving the water resistance of MPC. The increase in the strength retention rate is a result of both chemical and physical actions.