The grottoes on the cliff were carved from the late Northern Wei Dynasty, flourished in the Tang Dynasty and ended in the Qing Dynasty. The total length is nearly 388 meters and the height is 45 meters. It forms a cave corridor composed of more than 950 Buddhist niches and more t

　　Qianfo Cliff is located in Guangyuan City, Sichuan Province. The grottoes on the cliff were chiseled in the late Northern Wei Dynasty. They flourished in the Tang Dynasty and ended in the Qing Dynasty. The total length is nearly 388 meters and the height is 45 meters. It forms a cave corridor composed of more than 950 Buddhist niches and more than 7,000 Buddha statues, so it is called "Thousand Buddha Cliffs". It is the largest Buddhist grotto in Sichuan Province, with a magnificent momentum. On the one hand, Guangyuan is located in a subtropical humid monsoon climate zone with heavy annual rainfall. On the other hand, the repeated dissolution-crystallization of soluble salts on the surface of sandstone is another important reason. Curing is necessary to restore intergranular cohesion. Inorganic reinforcement is the first choice for weathered sandstone restoration.

Magnesium Oxygen chloride Cement (MOC) is an air-hardened cement. This cement material is made of MgO, MgCl2 and H2O in a specific molar ratio and cured at room temperature based on water reaction. It has high mechanical strength, good thermal stability and excellent bonding ability. However, the process of using MOC mortar encountered some difficulties. First, due to its high viscosity, MOC mortar is difficult to penetrate into sandstone pores. Ethanol can disperse magnesium oxide particles, reduce agglomeration and increase penetration depth. Secondly, when using ordinary magnesium oxide as raw material to prepare the curing agent, magnesium oxide particles may clog the pores of the sandstone. Compared with ordinary magnesium oxide, nanomagnesium oxide has smaller particle size, higher specific surface area and stronger activity.

　This paper studies the feasibility of using ethanol as a dispersant and using nanomagnesium oxide to substitute ordinary magnesium oxide for consolidation of weathered sandstone samples, and explores the best treatment method (i.e., curing agent concentration and number of uses), and preliminarily evaluates the consolidation performance. The preparation process of the curing agent (MMH dispersion) is as follows: nanomagnesium oxide, MgCl2·6H2O and H2O are mixed in a glass beaker, and MMH mortar is prepared according to moles of n(MgO): n(MgCl2·6H2O): n(H2O)=5:1:16. Next, disperse with ethanol and sonicate. The application method is as follows: Use a disposable dropper to add MMH dispersion from the upper surface of AWS at a speed of 40 drops per minute. After treatment, the sample was kept at a temperature of 25°C and a relative humidity of 60% for 7 days. All samples were weighed before and after the consolidation treatment to record the absorbed solid phase.

Viscosity and XRD analysis showed that ethanol is a viable solvent that does not inhibit the formation of consolidation products. The formed 5Mg(OH)2·MgCl2·8H2O produces strong adhesion between the sand particles, which is the main source of mechanical strength. SEM-EDS characterization shows that consolidation treatment can improve the compactness of weathered samples and ultimately form a coherent structure. In addition, the consolidation performance is greatly affected by the concentration of curing agent and the number of times used. As the amount of consolidant permeates into the sample, the cured specimen exhibits significant improvements in mechanical strength and salt resistance, and the water absorption decreases as the consolidation product is partially filled in the sand voids. However, excessive accumulation of consolidates can lead to significant color changes on the surface of the sample and thus cause the risk of incompatibility. The results of this study show that in order to achieve optimal consolidation performance, the intermediate concentration of the MMH dispersion (30 wt.%) should be selected and the weathered samples should be repeated five times.