In the third part of the experiment (samples 11-15), the same consolidant used in sample 6 on Meridian Gate tiles was applied to decorative floor tiles from the Mogao grottoes. Based on the first test results, this polyurethane consolidant, a Wang-brand thermoplastic resin, demonstrated a high degree of wear resistance. The Mogao tiles, produced during the Tang and Song dynasties, are not as strong and dense as those made for the Ming-period Meridian Gate; they contain more sand, less clay, and were fired at a lower temperature.
Abrasion Tests
Abrasion test results for each tile sample are shown in Table 1. Using a road-abrasion testing machine (from the former East Germany) with a rotating steel ball, 40 kg top load was applied and rotated eighty times on the Meridian Gate samples (samples 1-10). Because they are of poorer quality, the Mogao tiles (samples 11-15), were tested under 20 kg top load and rotated forty times. The polyurethane consolidants used in samples 5-8 gave satisfactory results, especially in samples 6 and 7. These consolidants remarkably increased the wear resistance of the tiles (Fig. 1).
Even better results were observed for the tiles treated with raw tung oil (samples 9 and 10). There is almost no abrasion loss for sample 10 (Fig. 2), which was treated by the soaking method. The tung-oil coated and soaked samples were also tested in a sandblasting machine. The samples were placed at a distance of 20 cm from the spray nozzle and blasted with 6 kg cm-2 of air pressure for fifteen seconds. These results demonstrated that treatment with raw tung oil can effectively increase the wear resistance of ancient tile.
The Mogao test pieces (samples 11-15) coated with the Wang polyurethane consolidant used in sample 6 also gave very satisfactory results (Fig. 3).
Figure 1
Sample tiles from Meridian Gate, Beijing Palace Museum, after abrasion-resistance testing; tile consolidated with polyurethane on the left and untreated tile on the right.
Figure 3
Samples of Mogao grottoes tiles after abrasion-resistance testing; tile consolidated with polyurethane on left and untreated tile on right.
Figure 2
Sample tiles from Meridian Gate after abrasion-resistance testing with raw tung oil; treated tile on left and untreated tile on right.
Freeze-Thaw Tests
Figure 4
Polyurethane treated and untreated tiles after 163 freeze-thaw cycles. The untreated tile is on the left. The tile on the right was treated with the polyurethane used in sample 6
(Table 1).
On-Site Experiments and Results
Freeze-thaw tests were conducted on the tile samples from the Meridian Gate (samples 1-10), using the Wang polyurethane tested in sample 6. The tiles were divided into three groups: (1) untreated samples; (2) samples coated with consolidant on 50% of the total surface area (one face and half of the four sides); and (3) samples coated on 100% of the surface area.
The dry weight of each tile was measured after the coating had been applied and allowed to dry. The samples were then soaked in water for five days. After soaking, each tile was weighed again to obtain its water-absorption rate.
To test the freeze-thaw rate, sample tiles were placed in boxes containing sticks at the bottom for support, with four tile pieces per test box. Water was gradually poured into each box until the top surface of the tile was submerged by 1-2 cm. Each box was then placed in a solution of calcium chloride at -20 °C to freeze for three hours. After this freeze period, the test boxes were transferred to a water bath at 16 °C for three hours to thaw. Temperature change was recorded with a resistance thermometer in each box. Freezing temperatures ranged from -15.95 °C to -1.35 °C, with an average of -6.9 °C. Thawing temperatures ranged from 12.5 °C to 17.7 °C, with an average of 16.6 °C. However, the volume of water in each box was too high for the samples to freeze completely in three hours.
After 40 freeze-thaw cycles, the experiment was modified to ensure effective freezing of the samples. The number of tiles in the box was reduced from four to three, and the water level in the boxes was measured at 1-2 cm above the base, rather than the top surface, of the tiles. The freeze-thaw cycle remained at an interval of three hours. Freezing temperatures measured with the new experimental procedure ranged from -11.3 °C to -3.3 °C, with an average of -7.5 °C. Thawing temperature was from 17.5 °C to 13.4 °C, with an average of 16.5 °C. The experiments were terminated after 163 cycles. Table 2 shows freeze-thaw test results for the seven samples tested. Figure 4 shows deterioration of two tile pieces after 163 freeze-thaw cycles.
The authors began on-site experiments at the Mogao grottoes in September 1980. Tests were conducted at three representative sites: Cave 45 (Tang dynasty), temple ruins in front of Cave 130, and Cave 328 (early Tang dynasty).
Two different consolidants were applied: raw tung oil mixed with kerosene in a ratio of 9:1; and polyurethane (the one tested in sample 6) mixed with a thinner (toluene or xylene and butyl acetate) in a ratio of 1:1.
First the tiles were cleaned and photographs were taken before coating with the consolidants. The coating solutions were then mixed on the basis of the number of tiles to be treated and the amount to be applied to each piece of tile. After treatment, photographs were taken again.
Both raw tung oil and polyurethane were applied to the tiles inside Cave 45 and outside Cave 130, and only polyurethane was used in Cave 328.
Several tiles in Caves 45 and 328 were left untreated and partially coated for comparison, and only four rows of tiles in Cave 130 were treated.
Raw tung oil penetrated rapidly and was readily absorbed; it took only twenty-four hours for the tiles at the site to absorb the same amount of oil that was used in the laboratory During application of the polyurethane, permeation gradually became more difficult because of the evaporation of the thinner, with only half the amount of consolidant penetrating the tiles at the site as compared with the laboratory test samples.
After more than ten years of observation of natural abrasion at the site, no further wear of the treated tiles has been observed, whereas obvious wear has been observed on the untreated tiles. Thus, results show that these two consolidants were ideal agents for the protection of floor tiles in situ.
Conclusions
Compounds formed on the surface of tiles treated with polyurethane con-solidants can improve the wear resistance of ancient tiles. Of these, the polyurethane used in samples 6 and 7 (Table 1) has produced the most outstanding results.
Results demonstrate that raw tung oil, a traditional material for the treatment of decorative tiles in China, is as effective as—and sometimes even better than—the best abrasion-resistant polyurethane consoli-dant studied in the testing project.
World History
