Effect of an eccentric decoupled charge on rock mass blasting Effect of an eccentric decoupled charge on rock mass blasting

Abstract Experimental and numerical analyses were conducted to investigate the effect of an eccentric loaded contour hole on a rock mass. In the concrete blocks used for the analyses, detonating cords were placed at the centre of the blast hole and eccentrically against the wall of the blast hole. PFC 2D and AUTODYN were used for the numerical analyses, and the results of these software showed that an eccentric decoupled charge can result in the directional development of fractures, thereby enabling the control of cracks in the opposite direction. Even though both types of blasting have identical decoupling indexes, the crack and fracture patterns were affected by the location of the explosive, tamping, and other conditions. The results showed that an eccentric charge holder can be applicable to control the fracture direction and the damaged zone. For an eccentric charge, the initial crack was generated at 0.01 ms and expanded in the intended direction. For the eccentric charge, the maximum pressure at the area in contact with the blast hole wall exceeded that for the central decoupled charge by a factor of 5.5. Furthermore, the pressure in the intended direction was twice of that in the opposite direction. Abstract Experimental and numerical analyses were conducted to investigate the effect of an eccentric loaded contour hole on a rock mass. In the concrete blocks used for the analyses, detonating cords were placed at the centre of the blast hole and eccentrically against the wall of the blast hole. PFC 2D and AUTODYN were used for the numerical analyses, and the results of these software showed that an eccentric decoupled charge can result in the directional development of fractures, thereby enabling the control of cracks in the opposite direction. Even though both types of blasting have identical decoupling indexes, the crack and fracture patterns were affected by the location of the explosive, tamping, and other conditions. The results showed that an eccentric charge holder can be applicable to control the fracture direction and the damaged zone. For an eccentric charge, the initial crack was generated at 0.01 ms and expanded in the intended direction. For the eccentric charge, the maximum pressure at the area in contact with the blast hole wall exceeded that for the central decoupled charge by a factor of 5.5. Furthermore, the pressure in the intended direction was twice of that in the opposite direction.


Introduction
S everal experimental studies have been conducted to analyze the dynamic mechanical behavior of materials and structures. A semi-disc dynamic three-point bending test involving impact loading was conducted in a previous study; the results showed that the dynamic loading rate has a significant influence on the fracture toughness and arrest toughness of materials [1,2]. Smooth blasting, which is the controlled blasting typically used for tunnel excavations, is based on decoupling effects. However, the blasting pressure radiates in all directions, including the wall and the free surface. The use of eccentric charges is an ideal method to reduce the damage caused to the final wall and to concentrate the energy toward the free face, for a well fracture [3e5]. Our experiment is the initial work aimed at controlling the charge based on an analysis of the geological setting, which elucidates the direction of the charge, based on whether the drilling is deep or semi-deep. A majority of these studies have examined the column removal scenario, wherein one of the columns is removed, and the load of this column is applied vertically on the upper face to assess the behavior of the frame until failure [6,7]. Decoupled charges are generally adopted to minimise the damage zone on the final wall. The decoupling index is the ratio of the blast-hole diameter and the explosive diameter. The space between the wall and the explosive acts as a cushion and prevents the dynamic force from acting directly on the wall. Explosives are loaded at the centre of the hole and are not in contact with the wall [8]. Researcher indicted the relationship between burden and borehole diameter which is linear. Also, he carried out burden in underground mine were lower than in surface mine which related to higher ore densities, greater confinement in blasting and finally greater demand for wall fragmented rock [9]. Cost and saving time are the main target for using Measurement while drilling (MWD) techniques. This technique provides a useful tool to manage drill and blasting in surface mining also, improve the reliability of the blast design by providing the drill and blast engineer with the information specially tailored for use. From this, decisions can be made on the most appropriate type and amount of explosive charge to place in a per blasthole or to optimize the interhole timing detonation time of different decks and blastholes [10,11]. An improved blasting operation depending on blast pattern to modified fragmentation related to drilling logs. Many factors recorded during drilling operations with analyzed. Author mentioned the rock blastablity and determined the rock quality index as index for rock characterization. Two methods for the powder factor determination are suggested and their uses illustrated. Better knowledge of bench geology and structure will allow a better evaluation of zones' blastability and a timely blast design modification, and will therefore facilitate production and processing processes and reduce the operation cost [12]. They discussed on blast fragmented with assume the rock properties. They used stochastic modeling approach to quality the variation. This model used thousand iteration, generates a statistical representation of the expected fragmentation resulting from a production blast. The various blast-related parameters influence different parts of the fragmentation distribution, e.g., rock strength and explosive velocity of detonation have most impact on the fines. The technique is used to identify the parameters that have the greatest influence on various size fractions. Such an analysis will be useful to direct resources to efficiently minimise the variation [13].
In this study, an eccentric charge was used to maximize the decoupling effect to the final wall and to concentrate the blasting energy toward the free face. By loading the charges toward the free face, the side wall blasting energy is concentrated in the direction of the free face. The final wall on the opposite side has a greater stand-off than that under normal decoupling and consequently should be protected properly. The damages to the wall and the free surface were compared for two different types of charges: centrally decoupled charges and eccentrically decoupled charges. Cylindrical concrete models were blasted, and the corresponding numerical models were simulated using PFC(Particle Flow Code) 2D and AUTODYN software to investigate pressure concentration and fracture propagation [14,15]. The damaged or disturbed rock zone (DRZ) is defined as the alternating zone induced by mechanical, thermal, hydraulic, or chemical processes. The resulting modulus, hydraulic conductivity, and seismic velocity differ from the original values. The DRZ includes the failure zone, damaged zone, and disturbed zone. If the DRZ is caused due to excavation, it is termed as the excavation disturbed or damaged zone (EDZ). Even though the disturbance and damage are not clearly defined, their features and areas are mainly dependent on the excavation method, size of the structure, rock strength, and initial stress. The EDZ is the rock zone around an underground excavation where irreversible changes occur. Such zones are inevitable because of the changing conditions resulting from the removal of rock during the construction process. As EDZ zone exists deeper in the rock; in this zone the changes associated with construction are reversible. Previous studies have focused on crack/ fracture evolution and the distinction between the mechanical and flow EDZ to establish the extent to which a coupled model is required [16e19].

Experimental work
This section describes the preparation of the materials, different blast-hole designs, models, and investigation of the experimental results. Five cases A, B, C, D, and E were tested, and the results were evaluated to determine if fracture is generated in the intended direction by eccentric decoupling [20]. Normal Portland cement was used for a concrete cylinder model, according to concrete quality standards (KS L 5201). To enhance the durability and strength of the concrete cylinder model, an air entraining water reducing admixture was added, according to the chemical admixture standards (KS F 2560). Furthermore, detonating cords of 10 g/m and 20 g/m were used. Five concrete cylinders, with a diameter and length of 20 cm and 40 cm, respectively, and a ∅45-mm blast hole, were cured for 28 days and set in the ground. Styrofoam and circular plastic films were used to fix the detonation cords to the centre or on the sidewalls. Fig. 1 depicts the fracture patterns obtained when a 20 g/m detonating cord was eccentrically loaded on the sidewall of the hole and blasted (Case A). The hole was stemmed, and it completely fractured, as shown in Fig. 1. In Case B, the concrete cylinder was eccentrically loaded with a 10 g/m detonating cord and blasted. Fine cracks were generated at the point of contact of the detonating cord. Additional cracks were not observed, except at the blast point. Case C involved eccentrically loading the concrete cylinder with a 20 g/m cord and blasting it. Several evident cracks were generated from the hole in the radial direction, and some portion of the collar was crushed around the contact point. In Case D, the concrete cylinder was blasted with a central-loaded 10 g/m cord. No cracks were generated on the cylinder wall, and only small parts around the collar of the hole were evenly crushed. In Case E, the concrete cylinder was blasted with a central-loaded 20 g/m detonating cord. Three crack lines were observed on the surface, and some parts of the collar were fractured [21,22]. Thereafter, the results of these five cases that were blasted under different conditions were compared. Cases A and C had the same amount of explosive; however, stemming was only performed in Case A. Thus, the results indicate additional fracturing in Case A, as compared to Case C. The tests on Cases B and E were conducted using the same amount of explosives, and both cases illustrated that the additional cracks and fractures were generated in the case wherein the detonating cord was located on the sidewall of the hole, as compared to the cracks generated when the detonating cord was located at the centre. Table 1 presents the results of each experiment.

Numerical analyses
The damaged zones can be analyzed via numerical analyses, using PFC 2D , which employs the DEM (Distinct Element Model), and AUTODYN 2D , which employs the FEM (Finite Element Model). Both these software were used to verify the effect of eccentric charges based on a comparison of the damaged zones [23], so numerical analyses were carried out using these software in this study.

PFC 2D
PFC 2D is the distinct-element modeling software provided by Itasca. It was used to verify the effect of eccentric charges by comparing the damaged zones. The material characteristics of concrete and the detonating cord were determined using the microparameters listed in Table 2. The fracture patterns and stress for the centrally loaded and eccentrically loaded cases were compared.
SF is the scale factor. SFn is the value multiplied by the average vertical strength (F), and SFs is the value multiplied by the average shear strength (G). According to the change in the SF, Young's modulus, Poisson's ratio, uniaxial compressive strength, and the crack initiation strength of cement mortar were calculated and applied at SF 0.2. s c is the average vertical strength of the contact bond, t c is the average shear strength of the contact bond, and H is the ratio of the standard deviations of the average vertical and shear strengths of the contact bond. Eight factors, except the scale factors (SFn and SFs), were the variables that define the constituent particles and contact characteristics of the analytical PFC model.
The blasting source was simulated by applying pressure control to the surrounding wall, according to the change in the size of the particles in the blasting hole. The blasting source was fixed, and the vibration history was monitored. The porosity, disconnection of chains, and stress components were investigated for two loading cases, for the circular investigation areas with a diameter of 75 mm around the blast hole (Fig. 2).

Crack propagation
The crack propagation patterns for the five cases are depicted in Fig. 3. The cracks were generated until 5 ms after detonation. The blasting pressures are fully decayed, and no cracks were generated after 5 ms. The lower point of the hole was the eccentric loading point.

Blasting-induced stress
The blasting-induced stress (s xx , s yy) were determined, as shown in Fig. 4. The eccentrically loaded cases exhibited a significant difference between the stress components, whereas the centre-loaded holes have approximately identical components. This phenomenon can be attributed to the asymmetrical loading conditions, which can be a favorable condition for shear failure.
"Up" refers to the portion away from the detonating cord when the sidewall is loaded, and "Down" refers to the portion where the detonating cord is attached. The X axis represents time, and the Y axis represents stress and porosity. A twodimensional stress field is assumed for the PFC model. The two-dimensional analysis did not consider ignition in the vertical direction and ignored the velocity of detonation of the detonating cord. The model is a simple radial load model.

Porosity change
The porosity is changed due to crack development. Even though the cracks were not developed, and the pores were not well expanded, the difference between the two loading methods is evident because of the low pressures observed in Cases B, D and E (Fig. 5). In the figure, 0.156 indicates 15.6% and not 0.156%. The general porosity of concrete is 15e20%; hence, the PFC model is adequate. In general, it is known that the porosity and permeability are linear [24], and the crack propagation is performed in a direction in which the porosity of the concrete material is large. Therefore, it can be estimated that the greater the permeability of the material, the better the crack propagation. The numerical analysis of the eccentrically decoupled holes indicates greater fracture growth, higher porosity between the particles, and greater pressure propagation in the intended direction, as compared to that in central decoupling.

AUTODYN 2D
The specific features and capabilities of AUTO-DYN 2D are described in this section. In particular, the software incorporates all the function required  for the generation of a numerical model, analysis, and display of results in a single graphical menudriven package. The codes can be compiled at the same functionality, albeit at varying speeds, on personal computers and engineering workstations as well as mainframes and supercomputers.
The central and eccentric charges in the cylinder holes were simulated using the hydro code of AUTODYN. PETN was selected as the charge equivalent of the detonating cord. The corresponding material properties for concrete are listed in Table 3. The boundary condition was set as "transmit" for infinite radiation.

Crack propagation
The damage level until 0.05 ms after detonation is depicted in Fig. 6. In the central detonation case, radial and even propagation of cracks was observed, according to the stress wave radiation. Conversely, the eccentric charge yielded crack generation at the contact point, 0.01 ms after detonation. Moreover, crack expansion was prevalent toward the intended direction in the right side. 1. The boundary condition between the concrete cylinder model and the external atmosphere is not clearly stated in the paper. However, in the analysis for evaluating the impact of failure, the concrete cylinder and the external atmosphere are generally set to free surfaces.
When the blast pressure propagates through the blast hole in the cylinder, the compression wave proceeds along the concrete cylinder, which is a solid material, and the free surface reaches the free surface, tensile stress waves are formed due to the difference between the impeller and the solid material. These tensile stress waves induce the tensile failure of the concrete cylinder, thereby extending the crack in the concrete cylinder between the blast source and the artificial joint.

Pressure change
Gauge points were assigned along the intended direction at 10 mm-intervals to observe the changes in the stress during the first 0.1 ms (Fig. 7). The maximum pressure of gauge #2 for the central charge was 31 MPa at 0.012 ms, whereas it was 170 MPa at 0.004 ms in the case of the eccentric charge, at the same position. Additionally, the other    Fig. 8 presents the crack development, pressure contours, and pressure distribution toward the intended side and the opposite side, 0.05 ms after detonation, to compare the failure states. Several cracks are induced at the charged (intended) side by the initial stress wave. The pressure distribution at the intended side also exhibits a wider high-pressure zone than that in the opposite direction. The maximum pressure in the intended direction attains a value that is twice of that in the opposite direction.

Stress distribution
The AUTODYN 2D analysis demonstrates that the stress wave was directly applied on the wall at the contact point of the perimeter, and the initial cracks were generated more easily and developed well in the intended direction for the eccentric case, than that in the central charge case. This indicates that, by using an eccentric charge, the initial crack generation and fracture development can be controlled in the intended direction and consequently fragmented to the free surface, while ensuring that the opposite final wall is protected, similar to the central decoupled holes.

Conclusions
Experiments and numerical analyses of eccentric charges for contour holes were conducted with an aim to reduce the damage caused to the rock wall as well as to control the direction of crack generation toward the free surface. In this study, centrally and eccentrically charged concrete cylinder models were blasted. Furthermore, PFC 2D and AUTODYN 2D analyses were conducted for these models to verify the controlled decoupling effects. The results of this study are summarized as follows: 1) The experimental results show that eccentric charges produce more cracks and fractures than central charges, with more cracks being generated in the intended direction than in the opposite direction.
2) The results of the PFC 2D analysis agree with the experimental results. Additional cracks were generated and developed in the intended direction, resulting in fracture expansion and increased porosity. The differential stress in the eccentric case was higher than that in the central charge case, which indicates that shear failure occurs easily.
3) The results of the AUTODYN analysis show that decoupling in the central charge causes a radial generation of cracks. Conversely, in the eccentric charge case, it indicates an initial crack generation at 0.01 ms and expansion in the intended direction. The maximum pressure in the eccentric case exceeded that in the central charged case by a factor of 5.5. The pressure in the intended direction was twice of that in the opposite direction. 4) The results of the experiments and numerical analyses show that high stress and initial fractures can be generated in the intended direction by using eccentric decoupling. Although the same decoupling index was applied in both cases, the shock wave and initial fracture could be concentrated toward the free face via eccentric loading, while ensuring that the damage caused to the bed rock was minimized.

Conflicts of interest
None declared.

Ethical statement
Authors state that the research was conducted according to ethical standards.

Funding body
None.