3.1 The compressive strength, elastic modulus, and surface Leeb hardness of the tabia samples
Fig. 2 and Table 2 show the compressive strength of the tabia samples. The samples NW, JY, and HY had higher compressive strength when compared with the other samples. The compressive strength of all these three samples surpassed 6.5 MPa. In this regard, a systematic investigation was performed to figure out why these three tabia samples had such high compressive strength.
Fig. 3 shows the strain-stress test results of the tabia samples, and the elastic modulus was calculated using the stress peak divided by the corresponding strain (listed in Table 3). The elastic moduli (over 350 Mpa) for the samples NW, JY, and HY were higher when compared with those of the other samples.
The surface Leeb hardness of the tabia samples was among 270 – 360 HLD (shown in Fig. 4 and Table 4). Compared with the result elastic moduli of the tabia samples, the surface Leeb hardness increased with the elastic modulus. Thus, the portable Leeb hardness tester could be used as a nondestructive method for approximating the elastic modulus of the tabia samples.
3.2 The density and porosity of the tabia samples
Fig. 5 and Table 5 show the density and porosity analysis results of the tabia samples. The results show that the density and porosity of the tabia samples were similar. The density of the tabia samples ranged from 1.53 to 1.76 g/cm3, and the porosity ranged from 41% to 49%. This might indicate that the craftsmanship of these coastal defense forts was alike.
3.3 Determination of the contents of calcite, sand and clay minerals in the tabia samples
The contents of calcite, sand and clay in the tabia samples are depicted in Fig. 6 and Table 6, which fluctuated markedly among different samples. Thermogravimetric analysis method was also used to determine the calcite content of tabia samples. Take sample HY as an example (as shown in Fig. 7), the calculated result of CaCO3 content was 29.8%, which was close to the acid dissolution method (31.7%). Thus, in this paper, we use the acid dissolution method result for further discussion. The content of sand ranged from 23.1% to 58.5%, the content of calcite ranged from 18.7% to 39.3%, and the content of clay minerals ranged from 22.3% to 53.5%. The differences among the tabia samples in the contents of calcite, sand, and clay indicated the variance in the craftsmanship. The ratio of calcite, sand, and clay in samples JY, NW, and HY (circled in red), was around 1 : 1 : 1. These results indicate that the appropriate ratio of sand, calcite, and clay could be an essential factor leading to the high compressive strength of tabia samples. in this regard, it is necessary to study the influence of the grain size distribution of sand on the high compressive strength of the tabia samples, because sand served as the aggregate and took up a large proportion of the tabia.
3.4 Determination of the grain size distribution of sand and clay minerals in the tabia samples.
Fig. 8 and Table 7 show the grain size distribution of sand in the tabia samples. The grain size for each tabia sample could be categorized as: (1) small < 0.18 mm; (2) middle = 0.18 – 0.85 mm; and (3) large > 0.85 mm. The ratios of small size mass, middle size mass, and large size mass of sand in the tabia samples were donated as Rs, Rm, and Rl, respectively. The Rs, Rm, and Rl of the tabia samples are listed in Table 8. It is noted that all the Rm values of samples NW, JY, and HY were below 50%. The Rm/l and Rm/s were calculated by dividing Rm by Rl and Rs, respectively. All the Rm/l and Rm/s values of samples NW, JY, and HY were below 2.0 (shown in Fig.9). These results indicate the relatively disperse distribution of sand would enhance the mechanical strength of the tabia.
The grain size distribution of clay minerals had similarity among the tabia samples, which was mainly concentrated in 20 – 40 μm (Fig. 10). This might indicate that the clay minerals used in these tabia coastal defense forts was processed by similar treatments.
3.5 The detection of organic additives in the tabia samples
The detection was performed for the organic additives including oil, protein, blood, sugar, and starch in the tabia samples. Among these organic additives, oil, protein, blood, and sugar were not detected in any of the tabia samples. For the starch test, the absorbance at 533.4 nm of the obtained solution of tabia samples (shown in Fig. 11) is shown in Table 9. The absorbance of samples JY, PY, and ZY was 0.128, 0.198, and 0.172, respectively, while the limit of detection (LOD) for this method was 0.096, as reported in [7]. This demonstrated that there was sticky rice in samples JY, PY, and ZY. The samples without starch might originally have it, but the starch was degraded. Herein, the detection of sticky rice in the tabia samples needs further clarification.
3.6 The morphology and mineral composition analysis
Fig. 12 shows the morphology of tabia samples. The clay minerals, sand, and calcite particles were found in samples TFG (Fig.12 a), LW (Fig.12 c), JY (Fig.12 d), PY (Fig.12 e), and ZY (Fig.12 f). While the hydraulic materials C-S-H fibers were found in samples NW (Fig.12 b) and HY (Fig.12 g). These C-S-H fibers might be generated by the chemical reaction between lime, quartz, and water: SiO2 + CaO + H2O → C-S-H, and these hydraulic materials C-S-H fibers could enhance the hardness of the tabia.
Fig. 13 shows the XRD test results of the tabia samples. The main minerals of tabia samples are quartz and calcite. There are also other minerals, for example albite, muscovite, katoite, vaterite, sanidine, and kaolinite observed in the tabia samples. The similar minerals composition of sample JY (Fig. 13 d), PY (Fig. 13 e), and ZY (Fig. 13 f) may indicate the same source of the soil of the three forts, which is different from the HY fort. The mineral difference between other tabia samples indicating the different sources of the soil.
3.7 The aging test of the tabia samples
The mass ratio of the tabia samples after each aging cycle is shown in Fig. 14. The cycles of 100% residual mass ratio were taken as the aging resistance capability, and the aging resistance capability of the samples increased with the increase of the calcite content. Sample NW with the highest calcite content of 39.3% had the highest aging resistance capability (87 cycles). However, sample HY with the calcite content of 31.7% had a lower aging resistance capability (34 cycles), which is lower than that of sample JY (72 cycles with 23.6% calcite), PY (68 cycles with 23.2% calcite) and ZY (56 cycles with 19.2% calcite). This observation was due to the inhomogeneous distribution of lime in sample HY. Comparing the aging test results with the mechanical properties results, composition test results, density and porosity results, and particle size distribution results, we found that sample NW, JY, and PY that with higher content of calcite, and lower content of sand (which could cause lower density) had higher aging resistance capability. Thus, tabia with well-mixed raw materials, high content of lime, low content of sand will have a high performance of aging resistance capability. The aging resistance test could be used as a method to evaluate the tabia’s craftsmanship of the tabia.
3.8 The capillary water absorption test of the tabia samples
Fig. 15 shows the capillary water absorption quantity Δ m of tabia samples at different absorption time, t. The capillary water absorption coefficient could be calculated from the equation Δ m = C * At0.5, where C is the water absorption coefficient and A is the bottom area of the tabia sample [23]. The results of capillary water absorption coefficient was divided into three parts. For sample TFG (with porosity 41.32%), the capillary water absorption coefficient is 0.024 g/cm2 min0.5. For samples JY, ZY, LW, and HY (with porosity from 43.72% to 46.34%), the corresponding capillary water absorption coefficients were from 0.044 to 0.059 g/cm2 min0.5, much higher than that of sample TFG. Samples NW and PY (with porosity from 48.72% to 48.95%) had a similar capillary water absorption coefficient 0.077 – 0.078 g/cm2 min0.5. These results show that the capillary water absorption coefficient of the tabia samples increase with the porosity.