Abstract

Despite reports on previous research associated with the dynamic strength of mudded intercalations during cyclic loading, a systematic investigation of the impact factors of this strength is still valuable. This work aimed at experimentally revealing the impact factors of the strength along with their impacts. The potential impact factors considered in this work include (i) water content, (ii) clay mineral composition, (iii) clay content, (iv) confining pressure, and (v) cyclic failure time. Specimens of mudded intercalations were collected from China and were remolded and prepared for a dynamic triaxial test under cyclic loads. The test results showed that the dynamic strength is impacted by water content (strongly), clay mineral composition (moderately), confining pressure (moderately), and cyclic failure time (weakly); no significant impact of clay content was detected. Moreover, the dynamic cohesion is correlated with clay mineral composition (strongly), water content (moderately), and cyclic failure time (weakly); no significant correlation with clay content or confining pressure was detected. Finally, the dynamic friction angle is correlated with water content (strongly), clay content (moderately), and cyclic failure time (weakly); no significant correlation with clay mineral composition or confining pressure was detected.

1. Introduction

The presence of mudded intercalations is known to deteriorate rock engineering. There exists a limited body of research on the dynamic strength associated with mudded intercalations during cyclic loading. Significant advances in this field include the report by Xue and Wang [1], who determined the dynamic strength indexes of mudded intercalations collected from the Xiaolangdi hydroproject using a cyclic simple shear test and a dynamic triaxial test. However, the potential impact of various factors (e.g., clay mineral composition, water content, grain gradation, and confining pressure) has not yet been reported. Despite numerous previous works investigating the impact factors of the dynamic strength of common fine-grain soils [2–11], little is known as to whether the same impacts occur in mudded intercalations, which are a special soil type whose main mineral composition is clay with breccia and rock fragments.

This work aims at experimentally identifying the impact factors of the dynamic strength of this special soil type and specifying their impacts. The potential impact factors considered in this work include (i) water content, ω, (ii) clay mineral composition, M, (iii) clay content, C, (iv) confining pressure, σ3, and (v) cyclic failure time, Nf. Test specimens were collected from geological audits that service the large hydroproject near the Hukou waterfall of the Yellow River. Eighty-one groups of specimens with different values of the considered potential impact factors were remolded and prepared for dynamic triaxial testing under cyclic loads. The potential impacts of the factors were investigated and compared with those of related soils.

2. Materials and Methods

2.1. Testing Instrument and Specimens

Dynamic triaxial tests were performed on a DDS-70 electromagnetic vibration triaxial apparatus, as shown in Figure 1. The maximum allowable axial displacement and maximum allowable axial force are 20 mm and 1370 N, respectively. The allowable frequency range is between 0 Hz and 10 Hz.

(a)
(a)
(b)
(b)
Figure 1 
DDS-70 dynamic triaxial test apparatus: (a) view; (b) schematic diagram.

The specimens were sampled from a geological audit of a large hydroproject (location shown in Figure 2) and remolded into a standard size (shown in Figure 3). To investigate the potential impacts of (i) water content (ω), (ii) clay mineral composition (M), and (iii) clay content (C) on the dynamic strength and its indexes, specimens with various physical properties, as listed in Table 1, were collected. The grading results of the particle sizes are shown in Figure 4.


Figure 2 
Specimen sampling sites: (a) location; (b) view.
(a)
(a)
(b)
(b)
(c)
(c)
Figure 3 
Specimens after remolding: (a) Groups 1∼3; (b) Groups 4∼6; (c) Groups 7∼9.
Table 1 
Physical properties of remolded specimens.

Figure 4 
Grading of the particle sizes.
2.2. Test Procedure

The test was conducted a total of 81 times. As listed in Table 1, there are 9 groups of remolded specimens for each type of main clay mineral composition. For each group of remolded specimens, 3 levels of confining pressure (σ3) are applied: 100 kPa, 200 kPa, and 300 kPa. For each level of confining pressure, it is necessary to apply different axial dynamic loads (σde) to the 3 specimens.

The loading process is shown in Figure 5. An equivalent sinusoidal wave with a frequency (f) of 1 Hz is used as the axial dynamic load, and the consolidation stress ratio (Kc) is set at 1. The test procedure is shown in Figure 6.


Figure 5 
Loading process.

Figure 6 
Test procedure.

In general, under isobaric pressure consolidation, typical soils can be expected to fail once the strain reaches 5% [3, 4]. However, the strength of the mudded intercalations tends to be smaller in comparison with common soils. As a result, under cyclic loading, the initial strain values are larger and the development of damage is slower. This 5% strain failure criterion is therefore not quite suitable for mudded intercalations. Instead, the 10% strain failure criterion [12] is more reasonable and was therefore adopted in the test.

3. Results

3.1. Properties of Dynamic Strength, τdf

The cyclic time corresponding to the time point at which the cumulative strain (εd) meets the prescriptive strain failure criterion is defined as the cyclic failure time (Nf). The dynamic shear stress occurring when the cycle number (N) meets Nf is defined as the dynamic strength (τdf).

Figure 7 shows the relationship between εd and N. εd initially increases slowly with N, followed by a more rapid increase. This trend is not affected by the confining pressure, σ3. Moreover, the increase of σ3 and the cyclic stress ratio (rd; rd = σde/σ3) contribute to an increased initial cumulative strain value (ε0), an increased rate of εd, and a decreased Nf.


Figure 7 
Cumulative strain, εd, versus cycle number, N (no. 8 remolded specimens).

The relationship between τdf and Nf is shown in Figure 8.


Figure 8 
Relationship between τdf and Nf.

As shown in Figure 8, τdf decreases with Nf. Their relationship is fitted aswhere and are fitted coefficients.

The obtained values of A and B are listed in Table 2. More than 63% of the R2 values are greater than 0.7, indicating a good fit.

Table 2 
Obtained values of A and B.

An orthogonal test [3] (Table 3) shows that A is impacted by M (strongly), σ3 (moderately, positive), and ω (weakly, initially negative and then positive). The B is impacted by ω (strongly, positive), M (slightly), and σ3 (weakly, negative).

Table 3 
Orthogonal test results. Kjm is the sum of the parameters for the row level m repeated tests for the j line factor. kjm is the average value of the parameters. Rj is the range (i.e., the difference between the maximum and minimum) of kjm. , such that ηj represents the contribution of factor j to A or B.
3.2. Impact Factors of Dynamic Strength, τdf

The orthogonal test results show that τdf is correlated with ω (strongly), M, σ3 (moderately), and Nf (weakly). Figure 9 shows τdf∼Nf under different conditions, while Figure 10 shows the impact of factors on τdf, where the ordinate is kjm/kjmmax and the abscissa is the impact of the factors.

(a)
(a)
(b)
(b)
(c)
(c)
Figure 9 
τdf versus Nf for different cases of (a) ω, (b) σ3, and (c) M.
(a)
(a)
(b)
(b)
(c)
(c)
(d)
(d)
Figure 10 
Factor effects on τdf. kjm/kjmmax versus (a) ω, (b) M, (c) σ3, and (d) Nf.
3.2.1. Impact Factor I: Water Content, ω

Figure 9(a) indicates that τdf decreases with ω when the main clay mineral composition is kaolinite and σ3 is 200 kPa. In comparison, τdf initially increases slightly with ω but then decreases (Figure 10(a)). This phenomenon may be attributed to the critical water content, ω0. That is, for an ω value smaller than ω0, τdf increases with ω, whereas it decreases for an ω value greater than ω0. This behavior agrees with findings for common soil reported in [13, 14].

3.2.2. Impact Factor II: Confining Pressure, σ3

Figure 9(b) shows that τdf increases with σ3, where ω is 15.1% and the main clay mineral composition is mixed montmorillonite/illite. A similar result is exhibited in Figure 10(c). These results are consistent with those of ordinary soils, as demonstrated in [4, 15, 16].

3.2.3. Impact Factor III: Clay Mineral Composition, M

Figure 9(c) shows τdf when the main clay mineral composition is primarily illite. A similar result is exhibited in Figure 10(b). This result differs from that revealed in [17–19].

3.2.4. Impact Factor IV: Cyclic Failure Time, Nf

Figure 8 shows that τdf decreases with Nf. A different result is presented in Figure 10(d), which shows that τdf does not always decrease with Nf. The difference is likely caused by a variation in the compactness. That is, the compactness rises with Nf, resulting in an increase in τdf [5].

3.3. Impact Factors of the Dynamic Strength Indexes, Cd and φd

The dynamic strength indexes, Cd and φd, are determined as follows: first, τdf is evaluated based on Seed’s [20] equivalent cyclic failure time. Then, using τdf, Mohr’s stress circle can be obtained, and Cd and φd are consequently determined (see results in Table 4). As shown in Table 4, Cd is closely related to the clay mineral composition, which has a contribution rate of 67.23%. φd is closely related to ω. Figures 11 and 12 show the impacts of various factors on Cd and φd.

Table 4 
Values of Cd and φd determined for the case of Nf = 20.
(a)
(a)
(b)
(b)
(c)
(c)
Figure 11 
Factor effects on Cd. kjm/kjmmax versus (a) ω, (b) M, and (c) Nf.
(a)
(a)
(b)
(b)
(c)
(c)
Figure 12 
Factor effects on φd. kjm/kjmmax versus (a) ω, (b) M, and (c) Nf.
3.3.1. Impact Factor I: Water Content, ω

As shown in Figure 11(a), Cd increases with ω when ω is less than ω0, and vice versa for ω greater than ω0. φd decreases with ω (Figure 12(a)).

3.3.2. Impact Factor II: Clay Mineral Composition, M, and Clay Content, C

Figure 11(b) shows that Cd is affected by the clay mineral composition. Specifically, the Cd values for illite as the main clay mineral composition are greater than those with mixed montmorillonite/illite but lower than those with kaolinite. Moreover, φd decreases with C (Figure 12(b)).

3.3.3. Impact Factor III: Cyclic Failure Time, Nf

Figure 11(c) shows that Cd initially increases slightly with Nf and then decreases. A similar variation in φd is observed (Figure 12(c)).

3.4. Comparison with Other Soils

The dynamic strength indexes of the mudded intercalations are compared to those of soils from other sites, as documented in [1, 21–23]. The test results are shown in Figure 13.


Figure 13 
Dynamic strength limit of mudded intercalations.

Figure 13 compares the upper and lower limit values of the dynamic strength for samples from various sites. The comparison shows that the dynamic strength of mudded intercalations is generally smaller than that of loess, silt, clay, and other common soils. Possible reasons for this phenomenon are as follows: (a) a structural disturbance due to remolding leads to a significant decrease in the dynamic strength; (b) the mudded intercalations tested in this work are partially composed of mixed montmorillonite/illite, which have relatively low strengths; and (c) the clay content varies depending upon the site, resulting in differences in the dynamic strength.

4. Discussion

Previous literature has concluded that the τdf of mudded intercalations with kaolinite is higher than that for illite [17–19]. However, a different result was observed here; namely, the former is lower than the latter (Figure 9(c)). As a factor that accounts for the disparity between the two results, the clay content of the mudded intercalations in this study corresponding to illite is only 21.0%. This is much lower than the clay content of kaolinite, which reaches 48.8%. The greater the clay content, the smaller the dynamic strength [24–26]. A similar relationship is observed for the dynamic strength indexes of the mudded intercalations, as shown in Figure 11(b). As the mechanical properties of the main mineral composition improve, Cd of the mudded intercalations increases. Moreover, the impact of the clay content on φd is more significant, as shown in Figure 12(b), where φd decreases with clay content. The clay mineral composition and clay content constitute two primary impact factors for the dynamic strength indexes of mudded intercalations, which are different from common soils.

5. Conclusion

The impacts of various factors on the mudded intercalation dynamic strength, including (i) clay mineral composition, (ii) water content, (iii) clay content, (iv) confining pressure, and (v) cyclic failure time, were investigated. The following conclusions were drawn:(1)A greater confining pressure and cyclic stress ratio contribute to lower cyclic failure times.(2)The dynamic strength is strongly impacted by the water content. When the water content exceeds a critical value, the dynamic strength decreases with increasing water content. An opposite variation in dynamic strength with increasing water content is observed when the water content is less than the critical value.(3)The dynamic strength is impacted by the clay content and clay mineral composition. The strength is correlated with the main clay mineral composition and decreases with the clay content.(4)The dynamic cohesion is impacted by the clay mineral composition, water content, and cyclic failure time. Specifically, the dynamic cohesion of the mudded intercalations with illite is greater than that for mixed montmorillonite/illite but lower than that for kaolinite. Such cohesions initially increase slightly with water content and cyclic failure time but then decrease.(5)The dynamic friction angle is strongly impacted by the water content. Specifically, as the water content rises, the dynamic friction angle decreases.(6)The dynamic strength, cohesion, and friction angle of the mudded intercalations are smaller than those of loess, silt, clay, and common soils.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this article.

Acknowledgments

This research was financially supported by the National Natural Science Foundation of China (U1504523), the Key Research and Development Plan in Henan Province (182102210014), and the Open Foundation of Jiangxi Engineering Research Centre of Water Engineering Safety and Resources Efficient Utilization (OF201602).