Ground Control for Underground Evaporite Mine in Turkey

Controlling the potential for hazardous ground movements and inundation of an underground mine to within acceptable limits is essential. Both hazards can result in serious harm or death of workers or persons that may inadvertently enter a mining area. The hazards are not always obvious. For example, the outcome of the hazard of a loose rock falling from a sidewall and striking someone can be fatal by either direct physical contact, or damaging the plant in which the worker is working. The presence of certain geological structure (natural planes of weakness in the rock) that override the effectiveness of any ground control measure, is not always obvious from within the mine.


Introduction
Ground control is the methodology applied to maintain all the risks associated with various forms of ground movement and inundation in underground mines within an acceptable level. It is applied to all stages of a mine -from feasibility through operation and finally abandonment.
Ground control methodology is largely determined as a function of the interaction of various qualities of the rock mass with various aspects of the mine planning and design methodologies. Depending on the nature of these interactions, rock support and reinforcement will required to achieve effective ground control.
Consequently, effective ground control may be considered to be a function of three main components:


Site ground characteristics,  Mine planning and design; and  Ground support and reinforcement Controlling the potential for hazardous ground movements and inundation of an underground mine to within acceptable limits is essential. Both hazards can result in serious harm or death of workers or persons that may inadvertently enter a mining area. The hazards are not always obvious. For example, the outcome of the hazard of a loose rock falling from a sidewall and striking someone can be fatal by either direct physical contact, or damaging the plant in which the worker is working. The presence of certain geological structure (natural planes of weakness in the rock) that override the effectiveness of any ground control measure, is not always obvious from within the mine.
In the case study given in this chapter, in situ and laboratory studies have been made to define the formations through which the drift has been driven. In situ engineering geology measurements were carried out consisting of field observation, mapping, boreholes and laboratory tests on samples collected from the trona field and the boreholes. After drift www.intechopen.com excavation and support installation, the deformations and the loads on the supports were measured. As a result of this study, the deformational characteristics of the trona ore beds and weak rocks and their effect on the main drift (T-2000) deformation have been determined. In this study, some parameters for understanding the deformational behavior of trona are investigated including in situ rock support loads, rock properties, and geology and excavation sequence.

Background for failure mechanism of underground evaporite mines
Geology is one of the most important parameters determining the overall deformational behavior of an underground opening. With careful investigation of the geological details, it is often possible to explain the unexpected deformational behavior of an excavation (Wang et all, 2000). Deformation of underground salt, trona and potash mines is generally time dependent, providing for gradual adjustment of strata to mining induced stresses. Time dependence can allow for higher extraction ratios provided eventual failure can be tolerated. However, this eventual failure can be violent if creep deformation can shift stress and potential energy to strong, brittle geologic units.
Generally, room and pillar methods, short-wall methods and solution mining methods have been applied as production methods for all evaporite mines in the world. The mine failure case studies reviewed here illustrate this process. Yield pillars and defects in bridging strata figure prominently in these cases. Yield pillars provide local and temporary support to the roof, temporarily delaying the cave; and allowing extraction ratios and overburden spans to increase beyond the long term capacity of overlying strata. Defects (faults, voids, thinning) of strong overburden strata reduce the critical span, sometimes to less than panel width. Analyses of many of these cases have focused on a cascading pillar failure mechanism, but recent work and this review point to failure of strong overburden strata as the essential element. The suddenness of failure and attendant seismic events pose hazards to miners and, in some cases, to those on the surface(Whyatt, J., 1998).
Trona is generally much stronger and stiffer than the immediate roof and floor rocks, with the degree of contrast varying with different beds and locations. Trona shows some timedependent behavior, but few laboratory creep tests have been performed in this study. Much of the time-dependency exhibited in the field may in fact represent creep of weaker roof and the floor strata rather than the trona bed itself. Many researchers suggest that, if the deformations around underground excavations are carefully measured in situ, the field measurements provide a better understanding of the deformational behavior of underground excavations (Onargan et al, 2006).

Location and geology
The study area is near Beypazari town that is 100km northwest of Ankara, and the Beypazari natural soda (trona) field is located in an area of 8 km 2 , which is 20 km northwest of Beypazari town center (

Laboratory studies
This study is divided into two sections: the first one includes identification of clay samples; the second one covers determination of the geotechnical parameters of the soft rocks and soda beds around the main drift. Rock types which occurred along the main drift line, based on thin sections and X-ray diffraction analyses, which were carried out in accordance with the methods described by the ISRM. Different block samples obtained from the tunnel face have been used in this study. The test results are summarized in Table 1.  www.intechopen.com Table 1. Geomechanical properties of hanging walls and soda beds (Onargan et all, 2004).
The geomechanical properties show great variety in the soda-bearing-bituminous shale formations due to the soda content. Two graphs illustrating the mechanical strength and deformability values obtained for the waste rocks and soda beds are given in Fig. 3. As seen in these diagrams, both the soda beds and waste rocks show a broad range of deformation properties, varying from ''very weak'' to ''medium'' characteristics.

Creep test results
The time-dependent deformation of trona, or creep, consists of four stages, namely: an elastic deformation stage, a transient or primary creep stage, a steady state or secondary creep stage, and a tertiary creep stage. The instantaneous elastic deformation occurs immediately after the creation of an excavation and is followed by the primary creep stage.

www.intechopen.com
In this primary creep stage, a high deformation rate decreases exponentially with time. The exponential decrease of the deformation rate under constant load in laboratory creep tests can be described as a strain hardening effect.
In this study, the creep tests have been carried out to determine the deformability properties of rocks versus time. The samples were tested under a load equal to 75% of the uniaxial compressive strength at 20 ºC room temperature. The deformability properties of clay Stone and trona beds (U4, U5, U6) versus elapsed time have been recorded. The creep properties of clay stone, U4 and U5 seem weaker when compared to U6. The clay stone, U4 and U5 ores complete the behavior of the 3rd creep region in 800-1000 min, while the U6 ore completes the same region in about 1400 min. (

Atterberg limits and swelling tests
The dominating mineral within the clayey formations is montmorillonite accompanied by a trace amount of illite. found to be about 10-40. According to the Unified Soil Classification Systems (USC); these clays, are classified as CH, and they are also under the group termed ''Inorganic Oily Clays with High Plasticity''. In the tests applied to un-altered samples, the consolidation index and swelling index were determined as 20% and 7%, respectively. These kinds of soils are classified as ''high swelling potential soils'' by O'Neil and Poormoayed. Hence, it is foreseen that additional support precautions will have to be taken at the necessary positions in the underground openings.

Field measurements at the T-2000 main drift and T-3000 roadway
The Main Drift, T-2000, and a roadway, T-3000, were excavated for Beypazari trona (natural soda) underground mine project between the years 1999 and 2001. The instruments were installed at several underground monitoring stations in order to obtain the necessary in situ data regarding displacement, deformation of supports and load using the load cell system. The magnitude and orientation of the loads were analyzed according to the results of the in situ measurements. Rock pressure was determined by the in situ measurements of the deformation of the support. Support load and convergence measurements were recorded daily for a period of about one year until the convergence rate decreased significantly. Six different stations were set up along the main entry drift for these measurements.
Load cells were placed at some of the points and the load increase over the support units was determined versus time. Three more stations (U4-a, U4-b and U4-c) in the roadway (T-3000) have been established within the soda bed (U4), again to determine the amount of load and convergence.
The stations in the main drift were coded as K-20, K-21, K-22, K-23, K-24 and K-25, which were all after the support installation. Vertical and lateral closures and narrowing were monitored daily for a period of 400 days.

Support loads
The support system load values were recorded by a digital gauge connected to the load cell system. The measured support loads at the stations versus time are plotted and shown in Fig. 6. According to these measurements, the support load at K-23 roof-rock station was found to be 0.09 MPa. At the U4-a station in the soda bed, the support load was 0.06 MPa. www.intechopen.com

Convergence monitoring
Convergence measurements at the main drift revealed variations in measured values over long periods of time. These values are independent of the face advance. Nine measuring stations were considered suitable for the analysis of the measured data. In these stations, horizontal and vertical convergences were measured between points of the rock wall.
This measurement system is illustrated in Figure 7 and 8. Between the tunnel length of 679 and 813 m, the main drift was driven in clay stone, bituminous shale, volcanic tuff and tuffitic claystone. In this part, 'rigid arches' were used as the supporting system by the management. For this section of the drift in the clayey formation, two kinds of problems were faced. Firstly, large deformations took place. Secondly, considerable swelling was observed, which was expected in the silty clay formations as mentioned before.
Floor heaving due to swelling capacity of the weak formation was also observed. The convergence measurements were performed using a tube-extensometer. The details of these measurements can be found in the research report (Onargan et all, 2001). All the resulting values indicate that deformations were below 1% on the convergence measurement lines Similar tests, for all the stations chosen, were carried out in different formations. Results can be seen in Figs. 9 and 10 in graphical form. It can be deduced from the foregoing studies that the total convergences at all the stations were found to be below 1% of the total gallery height and deformation rates were reduced significantly.

www.intechopen.com
In order to study the behavior of the support system, the convergence measuring stations were installed in the roadway (T-3000) driven in the soda bed. This roadway was driven with a 18 m2 cross section, using yielding steel sets of 29 kg/m TH profile with several spacings. These stations are termed U4-a, U4-b, and U4-c and convergence measurement and support loads were measured daily over a period of 400 days. As can be seen from Fig.  10, the maximum vertical closure was recorded as 198 mm, 5% of the original roadway height at the U4-a station. The values were obtained as 340mm (8.5%) at U4-b, and 198mm (5%) at U4-c for the other monitoring stations.  www.intechopen.com