Experimental Investigation of the Anchoring Effect of Two Different Types of Rock Bolts on Fractured Rock

A deep understanding of the anchoring effect of rock bolts on fractured rock is essential for support design in rock engineering. In this paper, cubic specimens containing a single preexisting flaw with different inclination angles were made by high-strength gypsum; uniaxial compression tests were conducted on bolted and unbolted specimens to study the anchoring effect of the fully bonded bolt and the prestressed bolt on fractured rock. +e mechanical parameters and failure characteristics of bolted and unbolted specimens were compared and analyzed in detail. +e results indicated that both the prestressed and fully bonded bolt had a significant influence on the mechanical behavior of fractured rock. +e average value of E, σi, σp, and σr of bolted specimens all increased due to the effect of the rock bolt. +e increase degree was the greatest for the specimens with flaw inclination angle of 45°. +e increase in residual strength, σr, was the most significant among all the mechanical parameters. +e mechanical parameters of specimens anchored with a prestressed bolt increased with an increase in pretension stress. Besides, the reinforcement effect of the two types of rock bolts was different for different mechanical parameters. +e bolted specimens displayed different failure characteristics compared to the unbolted specimens. Variation of tension stress in the prestressed bolt during the loading process was divided into three different stages. With an increase in the inclination angle from 30° to 60°, the peak tension stress value first increased and then decreased and obtained the maximum value at inclination angle of 45°. Besides, the peak tension stress value increased with pretension stress.


Introduction
Cracks usually initiate from the tips of preexisting flaws, joints, and other weak planes after rock excavation.e propagation and coalescence of these new cracks pose a great threat to the stability of the surrounding rock.Rock bolting is one of the most commonly used means of supporting in rock engineering to stabilize the ground and ensure safe working conditions [1].
e fully bonded bolt and the prestressed bolt are the most widely used rock bolts in civil and mining engineering due to their effectiveness and applicability.In recent years, new types of rock bolts, such as the cone bolt [2,3], the Durabar [4], the D-bolt [5][6][7], and other energy-absorbing bolts [8,9], have been developed to solve the issue of rock burst, which usually cause serious damage to on-site construction and pose a considerable threat to on-site workers in deep rock engineering because of the great amount of energy release [10][11][12][13][14][15][16].
A deep understanding of the working mechanism and anchoring effect of rock bolts is essential for support design in rock engineering.As such, extensive research studies have been conducted to address this problem in the past several decades using theoretical analysis [17][18][19], laboratory experiments [20][21][22][23][24], and numerical modeling [1,25,26].e support loads to the rock mass provided by a fully bonded bolt develop gradually when rock deforms [27].On the other hand, the prestressed bolts are installed with a certain pretension stress.Due to the effect of pretension stress, the prestressed bolts can provide support pressure to the surrounding rock immediately after installation.e anchoring effect of the two types of bolts is generally different due to the significant differences in their working mechanisms.Although the abovementioned studies have advanced our understanding of the working mechanisms and anchoring effects of rock bolts, the different anchoring effects of fully bonded and prestressed bolts on fractured rock have rarely been studied.e aim of this experimental research is to study the anchoring effect of the fully bonded bolt and the prestressed bolt on fractured rock.Cubic specimens containing a single preexisting flaw with different inclination angles were made by high-strength gypsum.Uniaxial compression tests were conducted on bolted and unbolted specimens.
e mechanical parameters and failure characteristics of bolted and unbolted specimens were compared and analyzed in detail.
ese research results are expected to provide useful guidance for support design in rock engineering.

Specimen Preparation.
e specimens were made from a mixture of high-strength gypsum, water, and quartz sand in a ratio of 3 : 1 : 0.5 by weight.Rectangular prismatic specimens, with dimensions of 120 × 60 × 40 mm (height × width × thickness), were prepared.
e physical and mechanical parameters of the intact specimen (with no preexisting flaws) are given in Table 1.e preexisting flaw with different inclination angles (30 °, 45 °, and 60 °) was made in the specimen by inserting a sheet of resin flake with a thickness of 0.2 mm and length of 20 mm during the specimen's solidification period (Figure 1(a)).All specimens were cured at room temperature condition for 28 days before the experimental tests.
A steel bar (70 mm in length and 4 mm in diameter) with elastic modulus of 200 GPa was used to simulate the rock bolt in this experiment (Figure 1(b)).A resistance strain gage was bonded to the steel bar to exactly measure the applied pretension stress and the variation of tension stress in the test.To apply the pretension stress, a spanner was used to tighten the nut which was set at the end of the bar.A bolt plate with dimensions of 35 × 35 × 2 mm (length × width × thickness) was also placed at each end of the steel bar to better simulate the real working conditions of the rock bolt.
A circular hole with a diameter of approximately 4.3 mm was drilled in the central position across the flaw to install the bolt into the sample.An anchoring agent which was a mixture of high-strength gypsum and water at a mass ratio of 3 : 1 was used to bond the steel bar and the borehole wall to simulate the fully bonded and untensioned rock bolt (Figure 1(c)).On the other hand, the steel bar was unbonded from the borehole and a certain pretension stress value was applied to simulate the prestressed rock bolt (Figure 1(d)).

Testing System and Program.
e uniaxial compression tests were conducted with the RMT-150C rock mechanics servo-controlled testing system in Institute of Rock and Soil Mechanics, Chinese Academy of Sciences.
e axial and lateral strains were measured by LVDT.e AEs during the compression tests were monitored with a 16-channel PAC-DISP system, and crack propagation process was recorded by a high-speed camera (Figure 2).e axial load was applied at a rate of 0.002 mm/s.Twelve groups of uniaxial compression tests under different test conditions were conducted, as listed in Table 2, and three samples were tested for each group.

Typical Stress-Strain Curves and Mechanical Parameters.
Typical stress-strain curves for the bolted and unbolted specimens under uniaxial compression are shown in Figure 3, which indicates that both the prestressed and fully bonded bolt had a significant influence on the mechanical behavior of fractured rock.e stress-strain curves of the bolted specimens possessed distinctive characteristics as compared with the unbolted specimens due to the effect of the bolt, which are summarized as follows: (1) Both the uniaxial compressive strength and residual strength were improved for the bolted specimens.Besides, the uniaxial compressive strength for the specimens anchored with the prestressed bolt increased with the pretension stress value.
(2) e drop rate of axial stress significantly reduced after peak strength for bolted specimens.While the axial stress dropped drastically to a low value (even to 0) after peak strength for the unbolted samples.(3) Lateral deformation was significantly inhibited after peak strength, especially for the samples containing flaws with inclination angles of 30 °(Figure 3(a)) and 60 °(Figure 3(c)).
e average values of the mechanical parameters of the specimens are listed in Table 2. e wing crack initiation strength, σ i , is difficult to be identified based on the stressstrain curves plotted in Figure 3.In this test, the AE hit rate curve was used to obtain the wing crack initiation strength, σ i .e axial stress and AE hit curves of sample 45-1 with flaw inclination angle of 45 °are plotted in Figure 4. e AE hit rate remained at a relatively low level during the initial loading stage.However, the AE hit rate increased sharply to 20 times per second when the axial stress reached about 18.66 MPa.Concurrently, the axial stress dropped rapidly from 18.66 MPa to 18.55 MPa.After that, a wing crack was observed to initiate from the tips of the prefabricate flaw.
erefore, the wing crack initiation strength, σ i , of sample 45-1 was identified to be 18.66 MPa. e e ect of rock bolts on the mechanical parameters E, σ i , σ p , and σ r of fractured specimens is shown in Figure 5, and the following conclusions can be obtained: (1) e mechanical parameters of E, σ i , and σ p decreased rst and then increased with an increase in the aw inclination angle from 30 °to 60 °for the   ese variations are consistent with previous research results [28][29][30][31].

Advances in Materials Science and Engineering
(2) e average E, σ i , σ p , and σ r of specimens with di erent aw inclination angles all increased due to the e ect of the rock bolt.e increase in residual strength, σ r , was the most signi cant among all the mechanical parameters.(3) For the fractured specimens anchored with a prestressed bolt or a fully bonded bolt, the increase degree of the mechanical parameters σ i and σ p was the greatest for the specimens with aw inclination angles of 45 °.Taking fractured specimens anchored with a fully bonded bolt as example, as α increased from 30 °to 60 °, the average value of σ i increased by 20.67%, 66.49%, and 14.17%, respectively, and σ p increased by 7.31%, 17.53%, and 16.37%, respectively.(4) e mechanical parameters E, σ i , σ p , and σ r of specimens anchored with a prestressed bolt increased with an increase in pretension stress.For example, the average value of E, σ i , σ p , and σ r increased by 4.03%, 14.13%, 4.47%, and 7.57%, respectively, with an increase in pretension stress from 36 MPa to 60 MPa for specimens with aw inclination angle of 45 °.
(5) e reinforcement e ect of the two types of rock bolts was di erent for di erent mechanical parameters.e fully bonded bolt was superior to the prestressed bolt with respect to increasing σ i of specimens.However, the prestressed bolt was more e cient than the fully bonded bolt with respect to increasing σ r of fractured specimens.

Typical Failure Modes
3.2.1.Unbolted Specimens.Typical failure modes of unbolted specimens are shown in Figure 6 (T stands for tensile crack, S stands for shear crack, and the grey area represents surface spalling).e cracking process of unbolted specimen can be described as follows.A tensile wing crack rst initiated from the tip of the aw and then propagated in a stable manner along the direction of axial stress.As the load increased, a secondary crack initiated from the tip of the aw and propagated along a coplanar plane along the aw in a stable manner.When the loads reached near peak strength, unstable failure occurred accompanied by an obvious shear movement, generally along the plane of the preexisting aw.After tests, as shown in Figure 7, the unbolted sample can be easily forced apart with hands into two parts along the macroscopic fracture surface of secondary shear crack.

Specimens Anchored with Prestressed Bolt.
Although the cracking process of the bolted specimens was similar to that of the unbolted specimens, a comparison of Figures 6 and 8 reveals some di erent failure characteristics between them.
(1) For the specimens anchored with the prestressed bolt, opening displacement between the two fracture surfaces of the tensile wing crack was greatly suppressed.e propagation path of the wing crack was di cult to clearly identify in the samples with aw inclination angles of 30 °(Figure 8(a)) and 60 °(Figure 8(c)).( 2) Macroscopic fractures propagating through the entire sample did not occur because of the e ect of the rock bolt.Furthermore, the fractured sample

6
Advances in Materials Science and Engineering was reinforced by the bolt and could not be easily forced apart by hand. is was the main reason why the residual strength of the specimens anchored with the prestressed bolt improved signi cantly.

Specimens Anchored with Fully Bonded Bolt.
Failure modes of specimens anchored with the fully bonded bolt are shown in Figure 9.When the sample is anchored with the fully bonded bolt, the secondary crack does not penetrate the whole sample.In addition, although the Advances in Materials Science and Engineering initiation and propagation of wing cracks occurred in the bolted sample, due to the restraint e ect of the bolt, the separation displacement between the two surfaces of tensile wing crack is greatly suppressed and the propagation path of wing crack is di cult to be identi ed.erefore, the bolted sample maintains a good integrity.
It is important to note that for the specimens anchored with the fully bonded bolt with aw inclination angle of 30 °, the cracking mechanism is di erent with unbolted samples.Anti-wing crack instead of wing crack initiated from the tips of preexisting aw is plotted in Figure 9(a).Unlike the wing crack, the formation mechanism of the anti-wing crack is more complicated.In this test, the fully bonded bolt directly interacts with the preexisting crack through the anchoring agent; this may be the main formation mechanism of the anti-wing crack.Under the action of the fully bonded bolt, the formation mechanism of anti-wing cracks needs to be further studied.

Tension Stress in the Prestressed
Bolt.Tension stress in the rock bolt is closely related to the deformation and failure process of the specimen, so an analysis of tension stress is of great signi cance to understand the working mechanism of the rock bolt.Variation of tension stress in the prestressed bolt during the loading process was divided into three di erent stages, as shown in Figure 10.For the specimens with aw inclination angles of 30 °and 60 °, the variation of tension stress in the rock bolts were very similar to that of specimen with aw inclination angle of 45 °.
However, the peak tension stress values were di erent, as shown in Figure 11.With an increase in the inclination angle from 30 °to 60 °, the peak tension stress value rst increased and then decreased and obtained the maximum value at inclination angle of 45 °.Besides, the peak tension stress value increased with pretension stress.For example, with an increase of pretension stress from 36 MPa to 60 MPa, the peak tension stress value increased from 82.14 MPa to 126.26 MPa for specimens with aw inclination angle of 30 °.

Discussion
In Section 3.1, we concluded that the fully bonded bolt was superior to the prestressed bolt with respect to increasing σ i , and the prestressed bolt was more e cient than the fully bonded bolt with respect to increasing σ r of fractured specimens.Due to the limited number of samples, only two di erent pretension stress values, i.e., 36 MPa and 60 MPa, were used for this test.If the pretension stress of the prestressed bolt was set to a value less than 36 MPa or a value greater than 60 MPa, the above conclusion may not be correct.Further experimental research is needed to validate this conclusion.
Based on the analysis of experimental results in Section 3.2, we can conclude that an increase in the pretension stress can improve the anchoring e ect of prestressed bolts on fractured rock.However, as analyzed in Section 3.3, a higher pretension stress will result in a larger peak tension stress value.In other words, tensile failure of the bolt can occur if the pretension stress is set too high.e experimental results provide some useful guidance for the design of bolt support in engineering.e pretension stress of rock bolts should be set to a reasonable value, which will not only improve the anchoring e ect on the surrounding rock but also avoid the tensile failure of the bolt.

Conclusions
Uniaxial compression tests were conducted to investigate the anchoring e ect of two di erent types of bolts on specimens containing a single aw with di erent inclination angles.Based on the analysis of experimental results, primary conclusions are summarized as follows.
Both the prestressed and fully bonded bolt had a signi cant in uence on the mechanical behavior of the fractured rock.e drop rate of axial stress signi cantly reduced and lateral deformation was signi cantly inhibited after peak strength for the bolted specimens.
e average value of E, σ i , σ p , and σ r of bolted specimens all increased due to the e ect of the rock bolt.e increase degree was the greatest for the specimens with aw inclination angle of 45 °. e increase in residual strength, σ r , was the most signi cant among all the mechanical parameters.Besides, the mechanical parameters of specimens anchored with a prestressed bolt increased with an increase in pretension stress.
e reinforcement e ect of the two types of rock bolts was di erent for di erent mechanical parameters.e fully bonded bolt was superior to the prestressed bolt with respect to increasing σ i of specimens.However, the prestressed bolt was more e cient than the fully bonded bolt with respect to increasing σ r of fractured specimens.
e bolted specimens displayed di erent failure characteristics compared to the unbolted specimens.Opening displacement was greatly suppressed between the two surfaces of the tensile wing crack.e fractured sample was reinforced by the bolt, and macroscopic fractures propagating through the entire sample did not occur.
Variation of tension stress in the prestressed bolt during the loading process was divided into three di erent stages: a slow growth stage, a stable growth stage, and a rapid growth stage.With an increase in the inclination angle from 30 °to 60 °, the peak tension stress value rst increased and then decreased and obtained the maximum value at inclination angle of 45 °.Besides, the peak tension stress value increased with pretension stress.

Figure 1 :Figure 2 :
Figure 1: Geometry of the specimens and two di erent types of rock bolt.(a) Specimens.(b) Steel bar.(c) Fully bonded bolt.(d) Prestressed bolt.

Figure 5 :
Figure 5: e e ect of the two types of bolts on the mechanical parameters of specimens.(a) Elastic modulus.(b) Wing crack initiation strength.(c) Uniaxial compressive strength.(d) Residual strength.

( 1 )
Stage A: slow growth stage.Elastic deformation occurred during the early loading period, and the lateral deformation of the sample increased at a low rate.As a result, the tension stress in the prestressed bolt increased slowly with time.(2)Stage B: stable growth stage.Wing crack initiated at the beginning of this stage, after which the tension stress of the bolt increased abruptly and then entered a stable growth stage.In this stage, the separation displacement between the two surfaces of the wing crack gradually increased, leading to a steady growth of the tension stress in the rock bolt.(3) Stage C: rapid growth stage.e tension stress in the bolt increased abruptly again after peak strength.During this stage, a macroscopic fracture had developed, and the separation deformation of the fractured sample further increased under the axial load.e tension stress increased at a relatively large rate due to the restriction e ect of the bolt on specimen deformation.Finally, the tension stress in the bolt reached the peak value when the bearing capacity of the sample no longer decreased.

Table 1 :
Physical and mechanical parameters of the intact sample.

Table 2 :
Test results of specimens under uniaxial compression.represents the elastic modulus.σ i is de ned as the wing crack initiation strength, and σ p is de ned as uniaxial compressive strength.σ r represents the residual strength. E