3.1.1. Load-Carrying Capacity and Failure Modes

The failure modes of the MPSs are similar. The development of cracks was analyzed based on the MP25H16 specimen as an example. Firstly, there were no changes in the surface of these specimens at the initial loading stage. As the loading continued, the fragments at the shear plane began to exfoliate. After that, the initial cracks appeared on the shear plane and several small cracks appeared instantaneously. Herein, it should be noted that the crack width of 0.05 mm is adopted as the criterion of visible initial cracking [28]. With the increase in loading, fine cracks further spread, connected and penetrated to form a crack zone along the shear failure surface, and the fibers between the crack zones were gradually pulled out or pulled off. Finally, with the further increase in the load, along with a huge sound, specimens were sheared and damaged. The condition of the crack development is shown in Figure 8. In order to verify whether the UHPC specimens still possess the bearing capacity after the main crack occurs, the test machine was restarted to continue loading. It turns out the load could still reach 1/2~2/3 of the ultimate load. On the basis of the testing results, it can be seen that there are two main crack modes in the failure modes for these specimens, namely the single main crack (Figure 9a) and multiple diagonal cracks (Figure 9b), respectively.

Since the failure characteristics of the specimens are similar, we took the MP25H13 specimen as an example to describe the shear load–slip relationship of the MPSs. The shear load–slip curve of MP25H13 specimen is shown in Figure 10. The results show that the direct shear cracking failure characteristics of UHPC can be divided into three domains. The first domain is a linear development stage in which the relationship between load and slip is basically linear (ideal status). The second domain is a crack growth stage. With the initial stiffness degenerating, the slope of the load–slip curve begins to decline. After the micro-crack occurs, the stiffness gradually decreases, and then the load enters a linear deviation domain. As a whole, because the bridging action of the steel fibers suppresses the crack expansion, the load increases in the alternation as the slip increases. The third domain is the shear failure stage. At this stage, large amounts of steel fibers are broken or pulled out from the matrix, and the specimens are rapidly destroyed. *Materials* **2017**, *10*, x FOR PEER REVIEW 9 of 17

**Figure 8.** Failure modes and the crack propagation process. **Figure 8.** Failure modes and the crack propagation process.

(**a**) (**b**)

**Figure 9.** Two typical crack forms: (**a**) single main crack and (**b**) multiple diagonal cracks.

from the matrix, and the specimens are rapidly destroyed.

Single main crack

Since the failure characteristics of the specimens are similar, we took the MP25H13 specimen as an example to describe the shear load–slip relationship of the MPSs. The shear load–slip curve of MP25H13 specimen is shown in Figure 10. The results show that the direct shear cracking failure characteristics of UHPC can be divided into three domains. The first domain is a linear development stage in which the relationship between load and slip is basically linear (ideal status). The second domain is a crack growth stage. With the initial stiffness degenerating, the slope of the load–slip curve begins to decline. After the micro-crack occurs, the stiffness gradually decreases, and then the load enters a linear deviation domain. As a whole, because the bridging action of the steel fibers suppresses the crack expansion, the load increases in the alternation as the slip increases. The third domain is the shear failure stage. At this stage, large amounts of steel fibers are broken or pulled out

**Figure 8.** Failure modes and the crack propagation process.

(**a**) (**b**) **Figure 9.** Two typical crack forms: (**a**) single main crack and (**b**) multiple diagonal cracks. **Figure 9.** Two typical crack forms: (**a**) single main crack and (**b**) multiple diagonal cracks. *Materials* **2017**, *10*, x FOR PEER REVIEW 10 of 17

**Figure 10.** UHPC shear load–slip curve. **Figure 10.** UHPC shear load–slip curve.

## 3.1.2. Shear Strength Results

MP25H16

MP30H16

MP25S13

MP25H13

286.4 515.6

291.0 469.8

284.2 454.3

3.1.2. Shear Strength Results For the MPSs, the baroclinic bar and the steel fibers passing through the shear plane act together to form a truss to resist the shear force along the shear plane. The effect of steel fibers on the shear For the MPSs, the baroclinic bar and the steel fibers passing through the shear plane act together to form a truss to resist the shear force along the shear plane. The effect of steel fibers on the shear properties of UHPC was analyzed, and the test results of fifteen MPSs are listed in Table 5.


properties of UHPC was analyzed, and the test results of fifteen MPSs are listed in Table 5.


254.1 420.8 21.04 1.656 1.604

296.4 494.8 24.74 1.669 1.669

261.3 481.5 24.08 1.843 1.630

21.53 0.059

24.72 0.035

23.13 0.028

22.98 0.035

1.800

1.614

1.599

430.6

494.5

462.7

459.8

273.8 465.5 23.28 1.700

25.78

307.3 473.0 23.65 1.539

291.5 444.6 22.23 1.525

22.72

306.2 443.5 22.18 1.447

According to Table 5, the maximum shear strength of UHPC can reach 24.72 MPa (MP30H16), which is 1.37 times the shear strength of MP20H16. It can be seen that the direct shear strength of UHPC increases with the increase of the fiber volume fractions. In addition, the shear strength of MP25H16 is 19.6% higher than that of MP20H16, while the shear strength of MP30H16 is only 14.8% higher than that of MP25H16. This shows that the shear strength of UHPC increases more obvious at the initial phase of the fiber volume fraction increment. According to the ratio of ultimate load to initial crack load of the specimens and shear slip relationship along the shear plane under different fiber volume fraction (Figure 11), when the fiber volume fraction is fewer, cracking failure characteristics of UHPC are close to brittle failure. Based on the above analysis, it can be considered


**Table 5.** *Cont.*

According to Table 5, the maximum shear strength of UHPC can reach 24.72 MPa (MP30H16), which is 1.37 times the shear strength of MP20H16. It can be seen that the direct shear strength of UHPC increases with the increase of the fiber volume fractions. In addition, the shear strength of MP25H16 is 19.6% higher than that of MP20H16, while the shear strength of MP30H16 is only 14.8% higher than that of MP25H16. This shows that the shear strength of UHPC increases more obvious at the initial phase of the fiber volume fraction increment. According to the ratio of ultimate load to initial crack load of the specimens and shear slip relationship along the shear plane under different fiber volume fraction (Figure 11), when the fiber volume fraction is fewer, cracking failure characteristics of UHPC are close to brittle failure. Based on the above analysis, it can be considered that the ductile characteristic and direct shear strength of the monolithic placement specimens with appropriate steel fibers can be improved in direct shear load. *Materials* **2017**, *10*, x FOR PEER REVIEW 11 of 17 that the ductile characteristic and direct shear strength of the monolithic placement specimens with appropriate steel fibers can be improved in direct shear load.

**Figure 11.** Shear load–slip curves of UHPC specimens with a different fiber volume fraction: (**a**) MP20H16, (**b**) MP25H16 and (**c**) MP30H16. **Figure 11.** Shear load–slip curves of UHPC specimens with a different fiber volume fraction: (**a**) MP20H16, (**b**) MP25H16 and (**c**) MP30H16.

As shown in Figure 11, the initial stiffness of the specimens (Stage 1) is reduced with the increase in steel fibers. This can be explained as follows. On the one hand, the uniformity of fiber mixing will

is predominantly dependent on the overall modulus of the matrix. Only after micro-cracks occur, the shear force will be balanced by the steel fibers and the matrix. Thus, with the increase in steel fiber content, the overall modulus of the matrix itself is reduced, leading to a reduced shear stiffness at Stage 1. With the further increase in load, micro-cracks occur, and the influence of the internal defects is gradually eliminated. The subsequent stiffness (Stage 2) mainly depends on the comprehensive modulus of the steel fiber and matrix. It should be noted that these phenomena are only reflected

Table 5 also shows the effect of different types of fibers on the UHPC shear strength. Under the same volume fraction of fibers (2.5%), the shear strengths of MP25S13, MP25H13 and MP25H16 are almost at the same level. It shows that the shear strength is not significantly influenced by the shape of the steel fibers. In the case of a certain fiber length, whether or not the hooked-type steel fiber is used has little effect on the shear strength of UHPC. However, when using the same shape fibers, increasing the fiber length will slightly reduce the shear strength of UHPC. This is because the UHPC specimens with short fibers possess more fibers at the same volume fraction. From the previous analysis, increasing the fiber numbers is an important way to enhance the compressive strength of UHPC. In addition, according to the ratio of ultimate load to the initial crack load of the specimens

when the fiber content is more than 2.5%.

As shown in Figure 11, the initial stiffness of the specimens (Stage 1) is reduced with the increase in steel fibers. This can be explained as follows. On the one hand, the uniformity of fiber mixing will be affected with the increase of steel fiber content, and the probability of internal defects (voids) of the UHPC will also be increased. On the other hand, the shear stiffness at the early stage of loading is predominantly dependent on the overall modulus of the matrix. Only after micro-cracks occur, the shear force will be balanced by the steel fibers and the matrix. Thus, with the increase in steel fiber content, the overall modulus of the matrix itself is reduced, leading to a reduced shear stiffness at Stage 1. With the further increase in load, micro-cracks occur, and the influence of the internal defects is gradually eliminated. The subsequent stiffness (Stage 2) mainly depends on the comprehensive modulus of the steel fiber and matrix. It should be noted that these phenomena are only reflected when the fiber content is more than 2.5%.

Table 5 also shows the effect of different types of fibers on the UHPC shear strength. Under the same volume fraction of fibers (2.5%), the shear strengths of MP25S13, MP25H13 and MP25H16 are almost at the same level. It shows that the shear strength is not significantly influenced by the shape of the steel fibers. In the case of a certain fiber length, whether or not the hooked-type steel fiber is used has little effect on the shear strength of UHPC. However, when using the same shape fibers, increasing the fiber length will slightly reduce the shear strength of UHPC. This is because the UHPC specimens with short fibers possess more fibers at the same volume fraction. From the previous analysis, increasing the fiber numbers is an important way to enhance the compressive strength of UHPC. In addition, according to the ratio of ultimate load to the initial crack load of the specimens and shear load–slip curves under different fiber types (Figure 12), the use of hooked-type fibers has little effect on the ductile failure characteristics of UHPC specimens. *Materials* **2017**, *10*, x FOR PEER REVIEW 12 of 17 and shear load–slip curves under different fiber types (Figure 12), the use of hooked-type fibers has little effect on the ductile failure characteristics of UHPC specimens.

**Figure 12.** Shear load–slip curves of UHPC specimens with different fiber types: (**a**) MP25S13 and (**b**) MP25H13. **Figure 12.** Shear load–slip curves of UHPC specimens with different fiber types: (**a**) MP25S13 and (**b**) MP25H13.

### *3.2. Test Results and Analysis of the WJTSs 3.2. Test Results and Analysis of the WJTSs*

steel fibers.

3.2.2. Shear Strength Results

### 3.2.1. Load-Carrying Capacity and Failure Modes

3.2.1. Load-Carrying Capacity and Failure Modes The shear failure characteristic of the WJTSs is a typical brittle failure, which occurs abruptly and without obvious symptoms. Similarly, the shear failure of the specimens is accompanied by a loud noise, especially for the WJ25H16, while the sound of the specimens without steel fibers will not be harsh. From the failure of the bond interface of the specimens, the failure surface of the specimens without fibers is relatively smooth (Figure 13a). But parts of the poured UHPC are embedded together successively, and the peeling phenomenon between the UHPC blocks interfaces is not obvious. The bonding effect of steel fiber reinforced specimens (especially for the WJ25H16) is The shear failure characteristic of the WJTSs is a typical brittle failure, which occurs abruptly and without obvious symptoms. Similarly, the shear failure of the specimens is accompanied by a loud noise, especially for the WJ25H16, while the sound of the specimens without steel fibers will not be harsh. From the failure of the bond interface of the specimens, the failure surface of the specimens without fibers is relatively smooth (Figure 13a). But parts of the poured UHPC are embedded together successively, and the peeling phenomenon between the UHPC blocks interfaces is not obvious. The bonding effect of steel fiber reinforced specimens (especially for the WJ25H16) is excellent due to the bridging effect of steel fibers (Figure 13b), which is similar to the MPSs. When the

steel fibers can play an important role in enhancing the shear plane strength for the WJTSs.

(**a**) (**b**)

**Figure 13.** Direct shear failure of cast-in-place specimens: (**a**) without steel fibers and (**b**) doped with

excellent due to the bridging effect of steel fibers (Figure 13b), which is similar to the MPSs. When the specimens are damaged, obvious fiber pullout marks and UHPC fragments can be seen. Thus,

MP25H13.

*3.2. Test Results and Analysis of the WJTSs* 

3.2.1. Load-Carrying Capacity and Failure Modes

specimens are damaged, obvious fiber pullout marks and UHPC fragments can be seen. Thus, steel fibers can play an important role in enhancing the shear plane strength for the WJTSs. excellent due to the bridging effect of steel fibers (Figure 13b), which is similar to the MPSs. When the specimens are damaged, obvious fiber pullout marks and UHPC fragments can be seen. Thus, steel fibers can play an important role in enhancing the shear plane strength for the WJTSs.

obvious. The bonding effect of steel fiber reinforced specimens (especially for the WJ25H16) is

*Materials* **2017**, *10*, x FOR PEER REVIEW 12 of 17

and shear load–slip curves under different fiber types (Figure 12), the use of hooked-type fibers has

(**a**) (**b**)

**Figure 12.** Shear load–slip curves of UHPC specimens with different fiber types: (**a**) MP25S13 and (**b**)

The shear failure characteristic of the WJTSs is a typical brittle failure, which occurs abruptly and without obvious symptoms. Similarly, the shear failure of the specimens is accompanied by a loud noise, especially for the WJ25H16, while the sound of the specimens without steel fibers will not be harsh. From the failure of the bond interface of the specimens, the failure surface of the specimens without fibers is relatively smooth (Figure 13a). But parts of the poured UHPC are embedded

little effect on the ductile failure characteristics of UHPC specimens.

**Figure 13.** Direct shear failure of cast-in-place specimens: (**a**) without steel fibers and (**b**) doped with steel fibers. **Figure 13.** Direct shear failure of cast-in-place specimens: (**a**) without steel fibers and (**b**) doped with steel fibers.
