Mechanics and safety issues in tailing-based backfill：A review.pdf

Invited Review Mechanics and safety issues in tailing-based backfill A review Xu Zhao 1, Andy Fourie 1, and Chong-chong Qi 2 1 School of Civil, Environmental and Mining Engineering, University of Western Australia, Perth 6009, Australia 2 School of Resources and Safety Engineering, Central South University, Changsha 410083, China Received 17 October 2019; revised 2 February 2020; accepted 10 February 2020 Abstract Voids referred to as “stopes” are generally created during underground mining activities and can lead to both local and regional geotechnical instabilities. To assist in managing the stability of mining-related voids and improving the recovery of orebodies, tailing-based backfill technology has been widely used around the world. In the design of tailing-based backfill strategy, the specific function and engineer- ing requirements of the filling are intimately dependent on the stress distribution within the backfilled stope. In this paper, the main mechanics involved in tailing-based backfill in underground mines, which include arching, consolidation, hydration process, and movement of surround- ing rocks, are reviewed. Research on the safety of a barricade and stability of an exposed fill face are also presented. In conclusion, the back- filling process should be pered on the basis of a better understanding of the complicated interactions of the mechanisms of filling, consol- idation, and hydration process when cement is added. Keywords tailing-based backfill; arching; consolidation; barricade; safety 1. Introduction Extracting valuable minerals from the earth’s crust is the essence of mining [1]. Mining activities generally result in the creation of voids. Referred to as stopes, these voids vary in size but can be as large as 30 m 30 m in plan dimensions and more than 100 m tall, leading to various environmental impacts, such as ground surface settlement on both local and regional scales [2‒3]. To maintain geotechnical stability, sig- nificant pillars of valuable ore must be left between stopes, which can significantly reduce the quantity of ore recovered [4]. To assist in managing the stability of mining-related voids and improving the recovery of orebodies, backfilling has been widely used in many mined stopes [5‒6]. The types of fill applied and their specific functions and engineering re- quirements are intimately dependent on the mining s, mining strategies, and mining sequences. The main types of mine backfilling include hydraulic fill, in which high-density slurry is delivered through boreholes and pipelines to the underground workings; paste backfill, which is generated from full-stream tailings and is now a widely accepted alternative means of mine backfilling; and rock fill, which economically uses waste rock generated from mining operations as the main component of the fill material. This paper mainly focuses on tailing-based backfill, which includes both hydraulic fill and paste backfill. A key advant- age of tailing-based backfill is its capability to transport the material hydraulically. Given this characteristic, the material handling and delivery costs of tailing-based backfill are relat- ively low, whereas the establishment of the stress state in a tailing-based backfill is complicated because of the existence of water [7]. In cases where cement is added to the backfill material, attempts to characterize the behavior of the fill dur- ing deposition and subsequent hydration become even more complicated [8–10]. Fig. 1 schematically shows the backfilling system in a stope. At the bottom of the stope, an access tunnel also called “drawpoint” is connected to the stope for the recov- ery and removal of blasted ore. To enable filling of the stope, at a certain location along the access tunnel, a containment barricade is constructed to prevent the fill from flowing out of the stope during and after backfilling. In the design of the backfill strategy, the stress state in the backfilled stope should be controlled such that it does not present a risk to workers underground, e.g., if the full hy- draulic head within a stope is imposed on a barricade, it can result in rupture of the barricade, potentially causing an un- derground mud rush. This needs to be avoided at all costs, as it presents a serious risk to workers. However, limited know- Corresponding author Chong-chong Qi E-mail chongchong.qi University of Science and Technology Beijing and Springer-Verlag GmbH Germany, part of Springer Nature 2020 International Journal of Minerals , Metallurgy and Materials Volume 27, Number 9, September 2020, Page 1165 https//doi.org/10.1007/s12613-020-2004-5 ledge of the unexpected behavior of the fill can result in cata- strophic consequences, such as the failure of containment barricades or the failure of a fill mass during exposure. Therefore, more insights into the stress distribution within the backfilled stope are required to ensure the safety of tailing- based backfill systems. Plug fill Barricade Access tunnel Fill deposition point Stope 10−30 m 20−100 m Residual fill 10−30 m 5−6 m 5−6 m 5−6 m Fig. 1. Schematic of a typical layout in a backfilled stope. This paper presents an overview of the literature relevant to the mechanics and safety issues in the mine backfilling process. First, the non-geostatic stress state of “arching” in mine backfilling is discussed. Then, consolidation theory and cement hydration, the ologies of fully coupling these mechanisms, and the influence of the behavior of surround- ing rock mass are addressed. Finally, previous approaches used to analyze the stability of an exposed fill face and the safety of a barricade are presented. For a detailed discussion about the properties of the backfill body from the materials science point of view, interested readers can refer to the com- panion review paper from the same group [11]. 2. Arching theory In industry, the backfilling process is mainly divided into two steps, namely, plug filling and residual filling. To monit- or the stresses developed within the backfill or behind the backfill barricades, some in situ monitoring operations were conducted, where stress sensors were placed within the stope [12‒13]. In these monitoring programs, the initial stress mon- itored by the bottom sensor increases with the filling opera- tion, which is closely related to the self-weight stress or con- solidation loads. However, after plug filling, the measured stress slightly decreases. After residual filling, the measured stress at the bottom exhibits a gradual increase, and the in- crease rate is lower than the expected increase rate of the self- weight stress. This means that the total stress induced by the overlying backfill body was mainly transferred to the sur- rounding rock, and many scholars attribute this phenomenon to the mechanism known as “arching” [14–17]. The essence of arching within the backfilled stope is load transfer along the wall/fill interfaces induced by the shear forces. Notably, the “arching” in the backfilled stope is one of most important concepts adopted in this paper because it may have a signi- ficant impact on the stress state in the stope, thereby leading to safety issues in the exposed vertical face and barricade of the access tunnel. As a starting point, this section provides an overview of existing ologies to investigate the arch- ing phenomenon and highlights a number of areas where fur- ther research can improve the understanding of this phe- nomenon. The concept of stress arching has been discussed by many different researchers over the years, including the pioneering works of Janssen [18], Marston [19], and Terzaghi [20]. In recent years, this theory has been applied to the development of limit state solutions specific to the mine backfilling situ- ation [14,21–23]. Marston [19] provided a specific analytical solution for estimating the arching effect on a backfilled mine stope, which takes into account the shearing forces along the fill/rock interfaces. Fig. 2 shows the different force compon- ents of the backfilled mine stope. h H Stope Layer element L FcFc Fv dFv Fv Fs Fs dWdh σvH Fig. 2. Schematic of the forces acting on a typical vertical backfilled stope. h FvFvdFvFs FcdW dh σvH As shown in Fig. 2, L and H are defined as the width and height of the backfilled stope, respectively. A horizontal lay- er element at position is subjected to vertical compressive forces and , a shearing force , and a horizontal compressive force . is the self-weight of the layer and is the thickness of the layer. The average vertical stress at the bottom is expressed as follows 1166Int. J. Miner. Metall. Mater. , Vol. 27 , No. 9 , Sep. 2020 σvH γL−2c 1−exp−2KHtan δ/L 2Ktan δ 1 γ cδ where is the bulk unit weight of the fill, K is the earth pres- sure coefficient, is the cohesion of the backfill, and is the interfacial friction angle. Some useful modifications have also been made on the basis of this solution to adapt to wide geometric boundaries and external mechanical conditions. For example, Li and Au- bertin [24] extended the analytical solution for “fully drained” conditions to consider “hydrostatic” conditions, as follows σvH γ sat−γwL−2c 1−exp−2KHtan δ/L 2Ktan δ uw2 γsatγw uw where and are unit weights of the saturated fill and water, respectively, and is the water pressure under hydro- static equilibrium. Rajeev et al. [25] and Widisinghe [26] extended the ana- lytical expression of total stress to a three-dimensional scen- ario, such as that of a cylinder, as follows σHv γsat−γwL1−exp−4KHtan δ/L 4Ktan δ uw3 Lwhere is converted into the diameter of the cylinder. δ K K The calculated vertical stress of the analytical solution ex- pressed in Eq. 1 is highly dependent on the interfacial fric- tion angle and the earth pressure coefficient . For Mar- ston’s theory, three values of have been considered [27]. K01 Pressure at rest, K0 1−sin ϕ4 Ka2 Active pressure, Ka tan2 45◦− ϕ 2 5 Kp3 Passive pressure, Kp 1sin ϕ/1−sin ϕ6 ϕ K0Ka where is the friction angle of the fill material. Different val- ues of K or have been used in previous researches [14‒28]. The expressions of Marston’s theory [19] and the exten- ded theory are a useful and fast ology for the prelim- inary uation of stress distribution under some specific and simplified conditions. However, as with the limit equi- librium , this has the following internal limit- ations 1 It always assumes the full mobilization of shear- ing stress at the fill/rock interfaces [16]. 2 It ignores the fun- damental properties of cemented backfill that evolve with time, as well as the possibility of material yield at the fill/rock interfaces when large deation occurs. 3 It assumes a constant value for Rankine’s earth pressure coefficient, al- though investigations clearly show that this coefficient tends to vary with the location of a backfilled opening [15,29]. In conclusion, the previously mentioned investigations found that, even under “fully drained” conditions, the degree of arching is determined by the complicated interaction between the mechanical properties of the fill. The behavior of the fill is more complicated under “undrained” or “partially drained” conditions when the mechanism of consolidation is involved. As will be shown in the following section, neglect- ing pore pressure in the calculation of the stress distribution can lead to a gross oversimplification of the process. 3. Consolidation behavior of the fill This section presents the current research on the complic- ated interactions between the mechanism of stress arching and the consolidation process during filling placement and curing. Fourie et al. [30] explained the significance of effective stress to the analysis of the behavior of the fill. The adjective “effective” is used to indicate that it is this stress, and not the total stress, that dictates the strength and stiffness of unce- mented soils; furthermore, it is this effective stress that pro- duces the deation of the soil “skeleton.” Therefore, a thorough understanding of the mechanism of stress redistri- bution to the surrounding rock mass is constructed on the basis of the understanding of the effective stress. To under- stand effective stress, an understanding of the consolidation process is required. 3.1. Numerical consolidation analysis Many numerical investigations have been conducted to demonstrate the significance of consolidation on the arching phenomenon in the fill mass [15,17,26,31]. Fahey et al. [15] undertook a series of numerical simula- tions using a commercial code Plaxis 2D, where the Mohr–Coulomb constitutive model was chosen. In their study, a stope with the height of 50 m and width of 20 m was modeled. To determine the effect of consolidation on the stress state, as well as the arching mechanism, they divided these cases into either fully drained or saturated conditions 1 The fully drained case neglected the influence of pore wa- ter pressure; thus, it can be considered a valid representation of the “best case scenario.” Assuming that the water table had been drawn to the bottom of the stope quickly and atmo- spheric pressures existed throughout most of the fill mass, the importance of elastic parameters was investigated. 2 In a fully saturated backfill, the investigation was focused on the differences in stress state i.e., effective stress, total stress, and water pressure in the stope at the end of filling EOF, at the end of consolidation process EOC, and at the end of wa- ter table drawdown EOD. kN/m3 ν E The corresponding curves of stresses in three cases i.e., A, B, and C that assumed a fully drained situation are repro- duced in Fig. 3. The material properties adopted in this ana- lysis included the bulk unit weight of 20 , Poisson’s ratio of 0.2, Young’s modulus of 10 MPa, angle of X. Zhao et al., Mechanics and safety issues in tailing-based backfill A review1167 ψϕdilation of 10, and angle of friction of 45. Case A presented a plane strain stope with the width of 20 m, and case C showed an axisymmetric stope with the diameter of 20 m. The stress distribution within a stope with the width of 10 m case B, which is only half that of case A, was also in- vestigated. To demonstrate the influence of the arching the- ory, the overburden of fill without arching is shown in Fig. 3 denoted by a dashed line. 02004006008001000 0 10 20 30 40 50 Height / m Total stress / kPa Overburden no arching Total vertical stress, case A Total horizontal stress, case A Total vertical stress, case B Total horizontal stress, case B Total vertical stress, case C Total horizontal stress, case C Fig. 3. Vertical