University
stringclasses 19
values | Text
stringlengths 458
20.7k
|
|---|---|
Colorado School of Mines | Figure 2.1 Solution mining methods. [79]
Figure 2.2 shows a cross-sectional view of a solution mining system being utilized in an
impermeable salt deposit. As Figure 2.2 (A) illustrates, an initial cavity is formed at the bottom
of the drill hole through direct circulation of the leaching fluid. After cavity initiation, the
process of fluid injection is modified so the lixiviant (i.e. the fluid utilized to dissolve and
transport the valuable mineral) is introduced near the top of the cavity. In many applications,
heat may be applied to the lixiviant in order to assist in facilitating mineral dissolution. The
lixiviant is then horizontally distributed through jets designed to remove the target salt contained
in the periphery of the cavity. The saturated brine is then pumped from the bottom of the cavity
to the surface. Figure 2.2 (B & C) shows this reverse circulation leaching method and brine
production. The diameter of the typical cavities can extend from 80 to 150 m.
18 |
Colorado School of Mines | technique are controlling the direction of fracture propagation, intensity, and growth, and
predicting the fluid pressure required for successful fracturing. Mineral deposits suitable for in-
situ leaching are usually placed below the aquifer, where the technique has been successfully
utilized to economically extract copper, gold, uranium and soluble minerals, like halite, potash,
boron and magnesium. Leaching solutions commonly used for copper are sulfuric acid, while
uranium extraction relies on a variety of different acids, sodium bicarbonate, or treated water.
Conversely, salts, such as potash, sodium sulfate, and sodium chloride, can be effectively
extracted using plain water. [88]
While some fractures are naturally formed within a rock mass, hydraulic fracturing
(pressurization) is an applied technique that facilitates the propagation of fractures in a rock layer
through the introduction of pressurized fluid. Induced hydraulic fracturing, often referred to as
hydrofracturing, is routinely used to improve recovery of petroleum, natural gas (including shale
gas, and coalbed methane), and other energy fuels by artificially inducing permeability in the
rock mass surrounding the well. This type of well stimulation promotes the development of
fractures that radially extend from either horizontal or vertical wells (boreholes) drilled into
reservoir rock formations. Fracturing rocks at depth tends to be suppressed by the confining
pressure due to the load caused by the overlying rock strata. This is particularly true in the case
of "tensile" fractures, which require the walls of the fracture to move apart, working against the
confining pressure. Hydraulic fracturing occurs when the effective stress is sufficiently reduced
by an increase in the pressure of fluids within the rock such that the minimum principal stress
becomes tensile and exceeds the tensile strength of the material [16]. Fractures formed by
hydraulic pressurization are mainly oriented in the direction perpendicular to the minimum
21 |
Colorado School of Mines | principal stress and for this reason; induced hydraulic fractures in drill holes (wellbores) are
sometimes used to determine the orientation of stresses. [16]
A hydraulic fracture is formed by pumping the fracturing fluid into the wellbore at a rate
sufficient to increase pressure downhole to a magnitude that exceeds that of the fracture gradient
(pressure gradient) of the rock. The fracture gradient is defined as the pressure increase per unit
of depth due to its density and is usually measured in pounds per square inch per foot or bars per
meter. The rock cracks and the fracture fluid permeate further into the rock mass resulting in
crack propagation. Operators typically try to maintain "fracture width" and prevent the fracture
from closing (healing) by introducing a proppant, a material such as screened sand, ceramic, or
other particulates, into the injected fluid. This technique prevents the fractures from contracting
when the pressure of the injected fluid inside the fracture is reduced. [16]
2.2.3) Borehole Mining
Borehole mining utilizes a conventional drilling rig to place strategically oriented
boreholes through the overburden into the mineralized horizon. Once the mineralized zone is
reached, the borehole is cased and sealed and the drill rig is exchanged for a specialized mining
rig with customized tools designed to excavate the resource. The tool (cutting head assembly) is
lowered through the casing to the exposed face of the mineralized zone, and uses pressurized
water/drilling mud supplied by surface pumps to excavate the material from the orebody in a
360-degree arc around the borehole. An internal air line within the drill string provides a
continuous supply of air to depressurize the return pipe and create a vacuum to lift (bail) the
slurry containing the excavated material through the drill pipe to the surface. In most cases, this
22 |
Colorado School of Mines | system also utilizes a series of packers to control the pressure within the excavated cavity. Once
the region around a borehole is completely mined to the operating capacity of the excavation
system, the remaining cavity can be capped and left to collapse based on an approved
reclamation plan. If conditions warrant, the borehole can also be filled with a wide variety of
slurries, cement, or grouts. In this case, the borehole can be fully or partially backfilled, where
the top of the casing is generally plugged with bentonite and cement before final reclamation of
the site with a soil cap. In the latter case, issues related to subsidence, surface cracking, and
water quality must be considered. [18]
For some commodities, such as phosphate rock and coal, borehole excavation of the
entire rock mass without dissolving specific minerals may be accomplished through an alternate
approach. New technologies have demonstrated the capability to extend rock fracturing and the
cutting radius to tens of meters beyond the wellbores for many rock types, while maintaining
control of the cutting geometry and avoiding excessive dilution. Borehole mining, as with in-situ
leaching, tends to minimize the surface impact of the operation. The biggest challenge for
borehole mining is the development of tools that can break and remove rock a sufficient distance
from the well center-line at an acceptable production rate to make the process economically
viable. Various technologies can be envisioned for advancing this capability; such as flexible
mechanical and hydraulic cutters that can extend from the borehole in a pre-determined
geometry. In addition, these excavation systems will require the development of other
complementary tools, such as sensors and cavity monitor technologies, which are capable of
distinguishing ore from waste rock and controlling the excavation process. [19]
23 |
Colorado School of Mines | 2.2.4) Applications of In-situ Leaching
Mineral commodities exploited by in-situ mining (solution and leaching) methods include
salts, such as sodium chloride, potash, magnesia, soda ash, borates, sodium sulfate, and lithium,
as well as metals, including uranium, copper and gold. The mining processes associated with
soda ash and uranium are representative of solution and in-situ leaching methods respectively,
and are described below. The main difference between these two methods is the creation of a
cavity in the solution mining technique.
Soda ash
At the American Soda Project (ASP), solution mining was conducted by injecting
adequately pressurized, high temperature water (177 to 216˚C) in order to thermomechanically
fracture the nahcolitic oil shale and generate a bicarbonate liquor that is further processed on the
surface. ASP conducted their experiments and evaluations at the Yankee Gulch Sodium Minerals
Project (YGP) plant, located at Piceance Creek in Rio Blanco County, Colorado. The high-
temperature injection fluid was needed for two reasons. The first is to fracture the oil shale and
provide a means of access for the fluid to contact the disseminated nahcolite. The second is to
promote the solubility of nahcolite. The saturation of nahcolite in water is a temperature-
dependent parameter. The saturation concentration of nahcolite in water is over 35 percent
greater at temperatures exceeding 177°C (350°F), whereas the saturation concentration is less
than 20 percent at 93°C (200 F) compare to room temperature. The growth rate of the cavern,
which depends on the nahcolite grade and the extent of the thermally- induced cracking, is
accelerated at higher temperatures, where thermal cracking occurs as a consequence of the
induced stresses developed around the cavern. Testing conducted on core from the Yankee Gulch
24 |
Colorado School of Mines | lease confirmed the reduction in strength of the nahcolitic oil shale at elevated temperatures. Oil
shale also has a high thermal expansion coefficient in the range of 6 to 200 microstrains/F. The
combination of reduced strength and increased thermally-induced stress results in cracking,
fracturing, and/or yielding of the rock at the perimeter of the cavern and are major concerns that
directly impact well design and operations. Figure 2.4 shows a three-dimensional schematic of a
mature well field at YGP. The well field is empirically designed for long term stability by
spacing the wells so that the remaining undisturbed nahcolitic oil shale between “caverns” are
able to support the overlying strata. In this context, the term “cavern” is a misnomer as the
solution-mined volume is more correctly described as a leached zone. The well spacing is 90 m
(300 ft). This allows for growth of the caverns up to 60 m (200 ft) diameter, leaving sufficient,
unleached material between caverns for fluid containment and subsidence protection. Each of the
caverns is approximately 157 m (515 ft) high, with the roof at a depth of 550 m (1800 ft) from
the surface. [77]
Figure 2.4. Schematic, three-dimensional sketch of the mature well field at YGP. [77]
25 |
Colorado School of Mines | After solution mining, the caverns (or leached zones) are supported by residual undissolved
oil shale, which is residing in the walls and immediate roof of the cavern. At this site, the impact
to any overlying aquifers is expected to be negligible. The residual oil shale in the “cavern” is
enriched because of the removal of the nahcolite and the permeability of the surrounding host
rock. This enhanced permeability may later allow for the in-situ recovery of the oil shale or other
minerals, such as dawsonite. The processing plant produced 600,000 tons of soda ash in 2002.
Despite the apparent successes of the technology, there were some concerns related to the design
of the well field; these included:
The maximum extent of the cavern and the resource recovery per well (e.g., the
size and shape of the cavern), and the ability to predict failure,
The potential impact of subsidence on the overlying aquifers and on the ability to
economically extract the overlying oil shale resources,
The impact of solution mining on the oil shale resources within the zone of excavation,
The hydraulic conductivity of the host rock and its ability to contain the high fluid
pressures within the cavern without leakage,
Design efforts to minimize drilling and completion costs while satisfying the operational
requirements of the well,
The location of the injection and production strings to optimize recovery without
producing contaminants such as halite, and
Selection of injection temperatures and target flow rates to establish the number of wells
necessary to meet the production targets.
26 |
Colorado School of Mines | In this application, the oxidant used is hydrogen peroxide and the complexion agent is
sulfuric acid. In Kazakhstan, ISL mines generally do not employ an oxidant but use much higher
acid concentrations in the re-circulating solutions. ISL mines in the USA usually use an alkali
leach due to the presence of significant quantities of acid-consuming minerals in the host
aquifers, such as gypsum and limestone. The existence of more than just a few percent carbonate
minerals means that the alkali leach must be used in preference to the more efficient acid leach
due to PH neutralization. [17]
In 2011, a total of 24,180 tons of uranium were globally produced by ISL. Most of this
production originated from Kazakhstan at approximately 2,500 tons and lesser amounts from
Uzbekistan, USA, Australia, and Russia. This translates to approximately 45% of the total world
uranium production, a share which has risen steadily from 16% in 2000. ISL uranium mining
was first attempted on an experimental basis in Wyoming during the early 1960s, where the first
commercial operation began in 1974. Today, virtually all Kazakh and Uzbek, and most U.S.
uranium production come from ISL mining. Several projects are licensed to operate in
Wyoming, Nebraska, and Texas, where most of these U.S. operations are relatively small and
controlled by a number of international-owned mining and resource companies. [17]
Uranium deposits suitable for ISL occur in permeable sands or sandstones, confined
above and below by impermeable strata, and are located below the water table. They may either
be in tabular, or "roll front" type deposits, within a permeable sedimentary layer. Such deposits
were formed by the lateral movement of groundwater bearing oxidized uranium minerals through
the aquifer, with precipitation of the minerals occurring when the oxygen content decreased,
along oxidation-reduction interfaces [17]. The uranium minerals are usually uraninite (oxide) or
28 |
Colorado School of Mines | coffinite (silicate) coatings on individual sand grains. There are two dominant operating regimes
for ISL which are determined by geology and groundwater. If there is significant calcium in the
orebody (e.g., limestone or gypsum), alkaline (carbonate) leaching must be used. Otherwise, acid
(sulfate) leaching is generally preferred. In the latter case, the leach solution is normally at a pH
of 2.5 - 3.0, which is approximately equivalent to vinegar. [17]
The design of ISL well fields varies greatly depending on the local geologic conditions,
such as permeability, sand thickness, deposit type, and grade distribution. Regardless of the type
of pattern used, there is a mixture of injection wells to introduce the leach solution to the
orebody, and extraction wells with submersible pumps used to deliver the pregnant solution to a
processing plant.
Where large sheet-like deposits exist, such as in Kazakhstan, rows of injection wells
interleafed with rows of extraction wells and are generally configured in a pattern to maximize
production and cost efficiency (Figure 2.6).
Figure 2.6. Alternating lines of injection and extraction. [17] & [89]
29 |
Colorado School of Mines | This pattern has a relatively low installation cost and is simple to design. However, the
rate of extraction and recovery can be less than optimal due to the long distances between the
wells (typically 50 to 60 m).
In most western applications, as well as Kazakh operations, where depositional channels
are narrower than 60 m, closer spaced patterns are employed to achieve higher recoveries per
unit area than alternating line patterns. As shown in Figure 2.7, the most common types of
patterns employed are: 5-spot patterns (usually 20-30 m between the wells), and 7-Spot patterns
(usually about 17-18 m between the wells).
Figure 2.7. Five and seven spot patterns of injection and extraction. [17] & [89]
Regardless of the pattern type used, the well fields (usually a production unit that feeds to a
single collection point) are progressively established over the orebody as uranium is depleted. A
series of monitor wells are situated around each mineralized zone to detect any unintended
migration of leaching fluids outside the mining area. The wells are cased to ensure that liquors
only flow to and from the ore zone and do not impact any overlying strata or aquifers.
30 |
Colorado School of Mines | In the USA, the production life of an individual ISL well pattern is typically one to three
years [17], where most of the uranium is recovered during the first six months of the operation.
The most successful operations have achieved a total overall recovery of approximately 80% of
the resource, while the minimum rate of recovery can be as low as 60%. In Australia, individual
well patterns usually operate between 6 and 18 months and target recoveries of 70% in 12
months are typical. According to Ur-Energy Inc. [78], the production cost is estimated to range
from 18-25 $/lb uranium.
The disadvantages generally associated with uranium in-situ leaching include:
The risk of leaching liquid propagating outside of the targeted mining area, involving
subsequent groundwater contamination and mobilization,
Unpredictable impact of the leaching liquid on mineral constituencies within the rock
mass, and
The difficulties associated with restoring natural groundwater conditions after completion
of the leaching operations.
Other environmental concerns associated with in-situ leaching include the release of
considerable amounts of radon and the production of certain quantities of waste slurries and
water during recovery of the uranium from the liquid. After closure of an in-situ leaching
operation, the waste slurries produced must be safely disposed of and the aquifer restored, where
any chemical contamination caused by leaching activities is rectified. Groundwater restoration is
a very tedious and expensive process that is characterized through cyclic intervention activities
and testing.
31 |
Colorado School of Mines | There are several advantages associated with solution mining as compared to conventional
extraction methods (e.g. open pit mining), including lower capital and operating costs, improved
safety, and less detrimental environmental impacts. According to Rabb [20], the following are
the disadvantages that make solution mining non-feasible in many applications:
Not all ores or ore deposits are amenable to leaching,
Conditions may not be favorable for the degree of solution control that must be
maintained. In addition to subsurface hydrology, climate may impose additional
challenges such as freezing, permafrost, or heavy rainfalls,
Most leaching systems are not completely selective. Dissolution of undesirable elements
can cause problems in a leaching circuit (i.e. clogging of pipes; fluid contamination),
It sometimes takes longer time for heap dump or in-place leaching to reach a reasonable
rate of extraction. Extractions recoveries are generally lower, and
Such leaching solutions usually contain corrosive or hazardous chemicals and require
special safety measures and monitoring procedures. Special attention must be given to
solution control. Special precautions are required to account for fugitive efficient, system
leakage, waste spills, and fluid balances. [20]
2.2.5) Borehole Mining Definition
Borehole mining (BHM) is an excavation process that utilizes pressurized fluid at a
specified flow rate and a downhole slurry pumping system implemented through one or more
boreholes drilled from the surface into the mineralized resource. Despite the perceived benefits
32 |
Colorado School of Mines | of hydraulic excavation and waterjet technology, major concerns associated with their use are
energy efficiency per unit excavated and dilution (the unintended excavation of sub-economic
material). For a given horsepower, excavation efficiency generally increases as the flow rate
increases once the threshold pressure has been reached (i.e., when the fluid pressure measured
behind the nozzle orifice of the system is sufficient to fracture/break the material). As a nozzle
orifice diameter increases at a given pressure, the frictional losses associated with accelerating
the fluid through a conventional round orifice decreases. With greater flow, the throw (effective
standoff distance from the nozzle to the area of impingement) increases due to less frictional
nozzle losses and free-stream aerodynamic drag. The minimum and maximum flow rates may
also be influenced by material handling. Too small of flow rate may negatively impact
excavation efficiency, as well as be insufficient to bail (transport) the cuttings by slurry to a
central collection point or up through a borehole.
In most cases, hydraulic excavation systems are designed by determining the threshold
pressure for a given rock type at the desired standoff distance using a specific motion strategy.
Once the threshold pressure has been determined, empiric studies have shown that pump
horsepower should be devoted to increasing the flow rate rather than pressure to achieve greater
excavation efficiency. The effectiveness of a larger diameter stream to erode material is less
dependent upon motion than a smaller diameter nozzle that is travelling faster. In addition, a
larger jet has a greater probability to intersect and exploit the existing fractures and weaknesses
within the rock mass than smaller diameter jets. That said, too much flow rate can cause
problems associated with dilution and the control of excavation geometry and the transportation
of cuttings. It can also erode the piping network and down-hole equipment.
33 |
Colorado School of Mines | As the waterjet cuts the material, it results in the creation of slurry that flows into an
eductor pump near the base of the tool and is pumped to the surface. The eductor is a jet pump
that utilizes the kinetic energy of one liquid to cause another fluid to accelerate. Eductors operate
by taking a high pressure fluid stream and accelerating it through a tapered nozzle to increase the
velocity of the fluid (gas or liquid) in accordance with Bernoulli’s equation. This fluid then flows
through a secondary chamber where the friction between its molecules and those of a secondary
fluid (generally referred to as the suction fluid) causes this fluid to be pumped via a venturi.
These fluids are intimately mixed together and discharge from the jet pump.
Borehole mining (BHM) systems incorporate a wide range of excavation technologies
that are intended to extract economic mineral commodities through one or more strategically
positioned boreholes. While most of these systems utilize high volume/low pressure fluids to
erode soluble or semi-soluble minerals, the BHM method proposed in this research is designed to
selectively fragment and excavate targeted zones of mineral resources through a process that
utilizes a highly focused waterjets. The borehole mining systems described in this dissertation
utilize jet pressure greater than 4500 psi (31 MPa) and flow rates less than 100 gpm (6.3 L/s).
Such systems are considered as high pressure/low volume operations, where these parameters are
generally measured relative to the pump discharge and not at the nozzle assembly. As such,
frictional losses between the pump and the nozzle assembly are important and need to be
included in the system. Borehole mining under the proposed configuration is a selective method,
because only ore (economically viable material) is extracted and where dilution is minimized. In
order to achieve this advantage, either the rock must be strong enough to remain stable while the
ore is mined, or the excavated cavity must be kept small to prevent roof failure. The ramification
34 |
Colorado School of Mines | by Abramov et. al. [24] is illustrated in Figure 2.9. This system utilizes water pressure of 2000
psi (13.8 MPa) with flow rate of 1000 gpm (63 L/s) and is widely considered as a low
pressure/high volume system.
As shown in Figure 2.9, this borehole mining tool functions as follows: (1) a borehole is
drilled from the surface to the production interval and casing is installed through the caprock; (2)
a specified length of the borehole is drilled below the casing shoe to serve as a sump for the
collection of the water and slurry from the cavity; and (3) the tool is lowered into a borehole until
the hydromonitor is placed below the casing shoe, where actual mining can then take place. The
tool contains four hydraulic channels within the boreholes. Those are the inner duct, the gap
between the duct and the inner drill pipe column, the gap between pipe and outer pipe column,
and the gap between the pipe and the casing. These four channels are respectively connected to
an air compressor, a collecting tank, a water pump station, and a secondary agent source. The
inner drill pipe and outer pipe column are used for pumping down the working water and bailing
the slurry back to the surface. The inner duct is used to carry a secondary working agent such as
air. The high pressure water is pumped to the tool through one of two inner or outer pipes, where
it is split between the hydromonitor containing the nozzle assembly and the eductor. As fluid
passes through the nozzle, the flow accelerates to form a jet that is sufficiently powerful to break
and scale away the material being mined. This loosened rock/ore is then fluidized by way of
mixing with the spent water to create a productive slurry. The resulting slurry is drawn to the
surface via the educator, where the material removed produces a cavity of a predetermined
geometry. On the surface, the water contained in the slurry is separated from heavy particles in a
collecting pond. The clarified water is pumped to a holding tank and then reutilized in the mining
36 |
Colorado School of Mines | balance within the cavity was not maintained to assist with material removal or for achieving
geotechnical stability of the cavity.
According to Abramov et al. [24], the effectiveness of rock cutting and slurry recovery
are important factors in the overall performance of the mining system. One method of causing
the slurry to rise through the pipe is by pressurizing the entire system, including the cavity where
the mining operations are being performed. Thus, as a consequence of pressure differential, the
slurry is forced through the return pipe. An alternative method is to situate an eductor near the
lowest point on the tool in order to force the slurry into the return pipe. The eductor type pump
has been favored in borehole mining because of the simplicity of the device and its high
mechanical reliability. High reliability is deemed a critical parameter for these systems, since
failure of a component at depth can result in long down times and expensive procedures for
retrieving the tool from the borehole.
The bailing velocity of the slurry through the eductor can be varied to control the amount
of water in the cavity. If the eduction and injection (pumping) rates are balanced, the cavity
remains filled with water less any allowances for groundwater intrusion and fluid migration.
However, if the eduction rate greatly exceeds the injection rate and the water yield from the
formation being mined, air pockets might be created in the cavity. This is usually preferable
because the waterjets can reach a greater distance in air as compared to water. It results in giving
greater throw (standoff distance) to improve effective diameter of the excavation. Air flushing is
also used to clean the bottom of the hole and cavity of the cuttings and broken debris. The ability
of air to carry (bail) the extracted materials to the surface depends on the nature of the broken
38 |
Colorado School of Mines | material and depth of the excavation. To bail cuttings out of the hole, the air velocity must be
greater than the settling velocity of the material.
If the ore deposit is situated in or below an aquifer or groundwater, then water from the
aquifer will flow into the cavity. Pressurizing the cavity with a fluid having a pressure
approximately equal to the hydrostatic head applied by the aquifer will prevent the underground
water from flowing into the cavity. The cavity can be pressurized through a variety of means,
including the water injected into the borehole for mining. In this case, the pressure gradient in the
cavity can be maintained by adjusting the proportions of water being routed between the tool
(nozzle) and eductor. The proportioning of fluid between the nozzle and the eductor has a
corresponding equilibrium condition with respect to free surface level of the slurry (water and
rock debris) and the pressure within the cavity. According to Wenneborg et al [25], the
equilibrium condition depends upon a variety of factors such as the characteristics of the ore
deposit being mined, the density of the ore slurry, and the depth of the ore deposit. Both the
pressure and the quantity of the slurry in the cavity can be stabilized by regulating the
proportioning of fluid between the nozzle and the eductor and the use of packers in the borehole.
When the cavity is filled, its pressure will rise corresponding to the volumetric flow
through the nozzle into the void until it reaches the effective limits of the pump, packers, and
mechanical components of the system. For a fixed displacement, decreasing the portion of the
fluid flowing through the nozzle (lower volume flow rate) increases the remaining portion of
fluid flowing through the educator, thereby, increasing the rate of slurry removal. The pressure in
the cavity is usually measured at the depth of the eductor and therefore, is a function of the
39 |
Colorado School of Mines | height of the free surface level of the slurry until the cavity is filled. As such, the height of the
cavity fluid head within the borehole, and the potential energy of the jet are key parameters. [25]
Wenneborg et al [25] categorized three sets of conditions for describing pressure and the
free surface level of water slurry, referred to as high, medium and low. These ranges are not
intended to focus on numerical values for pressure or flow rate but rather conditions within the
cavity. The high conditions correspond to equilibrium where the cavity is full or near-full and the
pressure within the cavity as measured at the eductor is high. This situation is suitable when
mining occurs at great depths or when it is desirable to reduce the mining power requirements for
bailing cuttings. In order to change the conditions for mining from low or medium to high, the
proportioning of fluid between the nozzle and eductor is adjusted to cause the water/slurry to fill
and pressurized the cavity. The fluid pressure in the cavity assists the bailing of the slurry by
reducing the net pressure head required of the eductor. The net pressure head that must be
produced by the eductor is equal to the pressure head of the slurry in the return passage to the
surface minus the fluid pressure in the cavity and the friction of the flow in the pipe. Special care
must be taken in trying to elevate the cavity pressures beyond some predetermined level so as not
to fracture the cavity host rock and propagate fractures into the borehole and surrounding rock
mass. The medium category corresponds to equilibrium conditions that have a medium fluid
pressure in the cavity and are generally applied when an aquifer is in proximity to the
excavation. In order to have medium conditions, the fluid (waterjet) should cause a resulting
equilibrium slurry pressure in the cavity that is sufficient to essentially prevent the flow of water
from aquifer into the cavity. This pressure is often utilized as the preventive pressure. When
cavity pressure is lower than the preventive pressure, the amount of fluid pumped to the surface
40 |
Colorado School of Mines | is greater than the amount of fluid being pumped into the cavity as a consequence of fluid inflow.
The preventive pressure has been reached when this difference is zero (the inlet flow equals the
outlet flow). The low conditions correspond to a free surface level in the cavity drawn down to
below the mining element (nozzle) and correspond to a low fluid pressure measured at the
eductor. In essence, the cavity is not filled with fluid and is generally used when mining is
shallow and operating challenges associated with weak ground are minor. Mining under low
conditions has the advantage of maximizing the effective radial distance (standoff) for the jet,
where the fluid level in the cavity does not disrupt or interfere with the jet stream emanating
from the nozzle. As such, mining under low conditions maximize the mining rate and size of the
cavity excavation radius [25]. While the cavity condition in this dissertation is generally assumed
to be low and without an elevated fluid level, consideration is given to fracture formation caused
by the cyclical and repetitive pressurization of a cavity. In this case, it is possible for a collapse
to occur due to the absence of pressure being exerted against the cavity walls.
2.2.7) Waterjet Dynamics
Relative to the cost of mining, system productivity and excavation rate are important
parameters. In this context, the effective reach of the jet must be as large as possible and the
impinging jet stream must have sufficient volume and velocity to effectively fragment and
transport the ore. Fluid power delivered to the cavity to perform productive work (i.e. material
extraction) is a function of jet pressure and flow rate. For a fixed pump displacement and
horsepower, it is vital to find an appropriate relationship of pressure and flow rate for a given
application. The waterjet must have sufficient energy density and diameter in order to be able to
excavate the rock and stay coherent over a minimum standoff distance, as well as have enough
41 |
Colorado School of Mines | volume to bail the cuttings. For a fixed horsepower pump, high pressure waterjets are
characterized by smaller free-stream cross-sectional areas and higher velocities that have shorter
standoff distances due to nozzle friction and aerodynamic drag. Conversely, waterjets with lower
pressures can allocate more power to the generation of flow which improves energy efficiency
and often higher production rates. Choosing the particular pressure and flow rate of an operation
system is a complicated process of balancing these factors for the available power, operation
environment, intended material response, and economics [23].
In this research, the flow regimes furthest from the nozzle orifice are of greatest interest.
Immediately upon exiting the nozzle, and for some distance from it, the jet maintains its own
central nucleus and continues with a nearly constant velocity in free-stream. This portion of the
jet is named as the initial section (or coherent section), and within its limits, the axial dynamic
pressure of the jet remains unchanged [23]. Beyond the jet limits of the initial section, the axial
dynamic pressure gradually decreases according to a hyperbolic function. The reasons for this
are generally attributed to the gradual expansion and turbulence along the jet’s free stream
periphery (i.e., the boundary of the stream and surrounding air).
As the jet’s free-stream advances away from the nozzle, dispersion of the stream occurs.
Jet dispersion is characterized a non-linear function, where the stream’s cross-section area
increases and its unit density decreases due to air entrainment. Nikonov [26] concluded that an
increase in nozzle diameter (i.e. increasing flow rate) will cause an increase in the length of the
initial coherence section of the jet’s stream. An increase in the jet pressure, however, will cause a
decrease in its length. Fine, high pressure waterjets generally have shorter lengthens of
coherency, beyond which there is a sharp drop in velocity. This is one of the disadvantages of
42 |
Colorado School of Mines | applying high pressure/low volume waterjets in applications that require large standoff distances
[26].
According to Xuan et al. [27], it is important to establish the main technological
parameters of borehole mining prior to designing a system for a given application. Dimitrijevic
et al. [28] have derived the following formula for calculating the diameter (d ) of a nozzle for a
0
single orifice nozzle assembly:
4Q /3600
d 0 [mm] ( 2.1)
0
u 2gp102
Where d is the diameter of the nozzle; Q is the required flow rate needed to bail cuttings
0 0
from the cavity; u is the coefficient of useful effect of hydroelevator (bailing capacity); P is the
pumping pressure; and g is the acceleration of gravity. The required flow rate (Q ) depends on
0 ,
the requirements for the velocity of the return flow. Similarly, the pumping pressure (P) depends
on the shear strength and pressure loss in the fluid system as well as the required pressure needed
by the jet pump.
The main factors that affect coal fragmentation and slurry entrainment are the physical
and mechanical characteristics of coal and the energy delivered by the waterjet. The minimum
pressure required for rock breakage must be equal to or greater than the critical threshold
pressure necessary to facilitate rock damage. The jet velocity attributed to the threshold pressure,
as measured immediately behind the nozzle orifice where necessary to induce material failure
can be calculated by the following [28]:
2g
V 0 [m/s] (2.2)
r
0
43 |
Colorado School of Mines | Where V is velocity of high-pressure flow representing the threshold pressure; g is the
r
acceleration of gravity; τ is the shear strength of the rock; and γ is the specific weight of water.
0 0
According to Miller and Wang [1], the precise definition of pressure varies directly to
where in the system the measurement is being taken and the context of the application. For
continuous jets, a complex relationship between free-stream coherency, standoff distance, jet
motion, fluid mass flux, and impingement dynamics, relative to a desired material response
dictate the necessary fluid pressure that must be developed behind the nozzle orifice. Referred to
as threshold pressure, it is heavily influenced by the flow (frictional) coefficient of the nozzle
and the ability of the resulting free-stream to do some specific type of productive work (e.g.,
slotting, spalling, kerfing, etc.). In accordance with Bernoulli’s Equation, the potential energy
(nozzle pressure) developed by restricting flow through the nozzle orifice is converted to kinetic
energy represented by the velocity of the free-stream, less losses associated with fluid friction,
heat, and noise. From a practical perspective, the fluid pressure within the free-stream for the
velocity ranges being discussed can be viewed as being negligible. At the time of impingement,
elements of flow are trapped by the incoming fluid stream within the surface irregularities of the
rock mass (e.g., fractures, pore space, and kerfs) and create transient pressure gradients that aid
in material fragmentation and failure. While it’s possible to measure the force of impingement
and material deformation, it is extraordinarily difficult to establish a fluid pressure profile
associated with jet impingement against a rock mass. Rather, an empiric process is usually
employed that establishes the nozzle pressure required to produce a desired excavation rate
relative to jet motion, standoff, and other operating characteristics. [1]
Since there are many factors that dictate jet performance and efficiency, the threshold
44 |
Colorado School of Mines | fluid pressure will vary over time and as each of the interrelated parameters change. Frictional
fluid energy losses between the nozzle and the primary pump will determine the necessary pump
pressure required for a given system setup. The pump pressure is always greater than the nozzle
pressure. While it’s difficult to measure the effective nozzle pressure, it can be calculated by
determining the pressure drop between the pump and the nozzle and subtracting it from the
gauge pressure measured at the pump. [1]
Knoke [29] shows that the mining rate of a hydraulic borehole mining tool generally depends
on the following parameters:
(a) Cutting parameters: nozzle diameter (D ); nozzle pressure (P ); discharge coefficient of
c c
nozzle (c ); traverse rate (U) borehole tool rotational speed, standoff distance (distance
d t ;
between nozzle exit and coal face) (X), jet core region length (a measure of jet
coherency) (X ), and number of passes (N).
c
(b) Coal properties: Although there is some disagreement as to which material properties
most affect jet cutting, the following parameters are generally included in most
theoretical and experimental studies related to coal excavation: compressive strength (σ ),
c
direction of bedding planes (α), hydrodynamic coefficient of friction between solid and
liquid (C), and damping coefficient of coal ( η) (i.e. energy lost per cycle).
f
Figure 2.10 illustrates a typical configuration of a waterjet traversing across the surface of a
rock. The cutting strategies vary by system parameters and desired material response but are
usually categorized relative to slot/kerf formation or spalling. In terms of excavation efficiency,
spalling is generally preferred. [30].
45 |
Colorado School of Mines | consequence of the technical limitations of the time, the low mechanical availability of these
systems, their associated high costs, and the poor productivity of small diameter jets resulted in
faulty conclusions about the economic viability of the technology in many applications for
several decades. However, a wealth of research performed by numerous individuals in the early
1980s, disapproved the relationship between jet performance and a rock’s compressive strength.
For example, Summer [74] determined that it was easier for waterjets to penetrate granite than
marble, despite marble being significantly weaker. As tested, the uniaxial strength of the marble
was roughly 50-70% that of granite. It is believed that jets penetrate a rock by flowing into the
existing cracks, planes of weakness, and grains boundaries, where the fluid exploits weaknesses
in the material. This fluid is then pressurized by the arrival of the subsequent stream (fluid)
elements in the waterjet, creating transient zones of high pressure that increase the length of the
cracks to the point that they coalesce and result in the formation of rock fragments. Summer and
many others have concluded structural properties play a much more prominent role in rock
cutting by waterjets than do physical material properties.
Forman and Secor [75] advanced a theory in 1972 that introduced the effect of water
presence in cracks as part of a stress analysis model. They believed that when the water
impinged against the rock, it would penetrate into the surface and flow into the target under a
steady state condition resulting in the pressurized of the rock’s pore structure in accordance with
Darcy’s law. According to Forman and Secor, there are two different components of dynamic
loading on the rock surface. The first part is a result of the effective stress initiated under the
impacting waterjet. This effective stress is solely due to the total force of the water impinging on
the surface. The second part is generated within the rock mass due to the pore pressure. This
47 |
Colorado School of Mines | stress is generated by the migration of water into the accessible pore spaces in the rock’s exposed
surface. The initial pressure distribution under the jet impact is calculated by Equation 2.4.
P= P [1-3(r/a)2+2(r/a)3] ( 2.4)
max
Where P is the applied pressure, P is the peak pressure seen by the rock; r is the radial distance
max
from the nozzle axis (i.e. center of the jet impact), and “a” has been found experimentally to have
a value range from 2.5 to 3.0 times the nozzle radius. The value of P is equivalent to the
max
nozzle stagnation pressure 1/2ρv2 (ρ= fluid density, v= jet velocity) reduced by the pressure
losses due to nozzle friction and standoff distance. [75]
According to Forman and Secor [75], the pore pressure distribution couples with the
stress field due to surface loading caused by stream impingement to produce an effective stress
field, where tensile fracturing can result directly under the load. According to the Griffith failure
theory, a fracture is started when the net tensile stresses reach a certain critical level. These levels
are 2.5 T in Indiana limestone and 3.5 T in Barre granite, where T is the tensile strength of the
0 0 0
rock. In each instance, a shallow crater is first formed on the rock surface and the jet’s cross-
sectional area redistributes itself over this crater. This process continues as long as the applied
pressure is sufficient to force water to flow in the pore structure of the rock and create tensile
stresses larger than the critical value. Forman and Secor believed the ability of the jet to flow
through the interconnected pore space of the rock (permeability) plays an important role in the
degree of fracture initiation and extension.
Relative to coal, Jeramic [14] concluded that it was critical to know the differential coal
porosity along the seam profile and the direction of maximum permeability since these factors
48 |
Colorado School of Mines | govern water absorption into the rock mass and directly influence fragmentation efficiency. As
such, the technical and economic feasibility of borehole mining is predicted on understanding the
direction of coal cleats and depositional structure. As Figure 2.11 illustrates, the mechanism of
coal breaking by waterjets can be subdivided into three stages. These include; granular erosion
due to jet impact and fluid scouring on the coal face which depends on the coal hardness,
cleating, and fracture density; breakage due to induced shear stress by fluid penetration within
the rock’s fractures and structures which depends on coal cleating and permeability; and tensile
failure propagation along planes of weakness (Figure 2.12) which depends on interplanar coal
strength.
Figure 2.11. Mechanism of breaking coal with a waterjet. [14]
As Figure 2.12 illustrates, the excavation efficiency of a BHM system depends on the
orientation of the waterjets to the planes of weakness. If the jet is parallel to the planes of
weakness, the waterjets penetrate between them and break some portions of the coal effectively.
In this orientation, coal is commonly fragmented into relatively large slabs. If the jet is
49 |
Colorado School of Mines | perpendicular to planes of weakness, the waterjets have to penetrate into solid coal through the
top layer, so that further breaking can occur between the planes of weakness. If the planes of
weakness are relatively thin, this then becomes an efficient method of breaking coal. In the case
of jet that is oriented at some oblique angle to the planes of weakness, the waterjets penetrate at
an acute inclination. Coal is not broken as efficiently in the other two cases due to the limited
ability of fluid to access and exploit these structural planes. [14]
Figure 2.12. Coal breakage related to seam anisotropy: a. parallel to planes of weakness, b.
perpendicular to plane of weakness, c. at angle to planes of weakness. [14]
As Figure 2.13 shows, increasing the fluid pressure at a fixed volume results in greater
energy being applied to the rock and a linear increase in the depth of cutting over a fairly large
range. The specific energy, defined as the ratio of delivered power to the mining rate, is the real
measure of energy expenditure. As discussed, no material removal occurs below the threshold
cutting pressure. That said, this figure can be misleading because it measures the depth of cut
(kerf) rather than volumetric removal, which is the basis for determining excavation productivity
and efficiency. It does, however, demonstrate the influence of pressure for a fixed set of
variables, where it can be treated as linear function over specified range of values.
50 |
Colorado School of Mines | Figure 2.13. Pressure versus depth of cut for coal cutting. [29]
The traverse rate, U, is a function of the tool rotational rate and the cavity radius.
t
According to Knoke [29], it can be calculated as follow:
U = 1.257 Ωx (in. /sec) (2.5)
t
Where the rotational rate Ω is revolutions per minute, X is in inches, and U is in inches per
t
second.
In spite of the reduced depth of cut obtained at higher traverse rates, the mining (or
cutting) rate may increase. For spaced cutting, an optimum traverse rate and rotational rate exists
which results in a maximum mining rate. The cavity radius in BHM is equal to the cutting jet
standoff distance plus the tool radius. The latter, however, is generally considered negligible.
The effect of standoff distance is shown in Figure 2.14 for bituminous coal. Other pertinent
operating parameters are listed on the figure. At a cavity radii beyond the jet core length, the jet
quality decreases with distance from the nozzle. Jet core length is the distance from the nozzle
that jet preserves its own central nucleus of coherency and continues with nearly constant
velocity. In this region, both the effective pressure and the effective core diameter decrease.
51 |
Colorado School of Mines | (slope) from the coal face to the crusher inlet, percent of solids in the flow, vertical cutting
increment and its direction, traverse rate, capabilities of the crushing and pumping system, and
direct interaction of the cutting jet and the cavity flow. Figure 2.16 illustrates cross-section of a
BHM cavity that shows a hypothetical particle size and shape distribution. The larger, more
irregular pieces of coal tend to remain near the working face while the smaller pieces flow more
readily to the slurry inlet of the borehole tool. Figure 2.17 shows the effect of vertical increment
direction on the cavity flow. When incrementing downward, the particles are displaced
circumferentially behind the jet and thus take a spiral path to the slurry inlet. When
incrementing upward, the cuttings fall straight down into the previously formed cavity and flow
radially to the slurry inlet [29]. However, this schematic fails to show that the broken material
will continue to be regrinded and fragmented by the jetting process until it achieves a particle
size that lends itself for removal.
Figure 2.16. Hypothetical particle size distribution in the cavity. [29]
54 |
Colorado School of Mines | removed from the hole. In addition, borehole mining is often conducted in rock strata comprised
of friable, unconsolidated material, where the shape of the excavated cavity can be easily altered
by rock falls and/or collapse of the cavity. Thus, measurements made through intermittent
stoppages in production activities are not always advisable or effective. Optimally, down-hole
measurements would be made during active mining. As a consequence, current design tools
allow for attachment of monitoring devices which will enable observation of the borehole cavity
and measurement of system parameters in real-time. [24]
2.2.8) Cavity Stability Monitoring During BHM
Floyd [32] did a preliminary investigation on underground structures produced by
hydraulic borehole mining systems. The study was performed using the limited data available
from Wyoming Powder River Basin coal deposits. The analysis was conducted using the finite
element method computer code NASTRAN. In this model, a three layer rock strata was
considered, including an alluvium top layer, roof rock (sandstone) and coal. The panel diameter
(cavity), d, and the center-to center distance, s, determine the recovery factor. The stress
concentration factor at the borehole surface was calculated as a function of d/s ratio (Figure
2.18). After the stress distribution in a pillar was determined, the maximum d/s ratio was
calculated as a function of the overburden depth, h, whose density is assumed to be 121.92 lb/ft3
(Figure 2.19). Since the pillar was under compression, a safety factor of 4 was used.
According to Floyd [32], since the overburden depth was on the order of 200 ft (61m),
the roof would likely fail due to the tensile stress from bending. Floyd [32] assumed that the roof
was rigidly supported by the pillar. As discussed previously, a finite element method was used to
56 |
Colorado School of Mines | (iii) Recovery factor n = 0.907 × (d/s) 2 = 86.2%.
The sample case above indicates that 40-ft-diameter excavation diameter (radius equals 20 ft),
where center-to-center distances of 41 ft are feasible under the stated conditions [32].
While numerous parameters such as the effect of internal pressure around the cavity,
cavity geometry, and in-situ stress during BHM have not been considered in this approach, it can
be used to provide a simplistic prediction of cavity stability of a cavity after completion of BHM.
A major objective of this dissertation is to consider those neglected parameters and evaluate their
impacts on cavity stability.
One of the significant concerns in BHM is how to monitor the stability of the cavity roof
during mining. Savanic [10] utilized six monitoring wells during borehole mining of phosphate
in St. Johns County, Florida (Section 2.2.7.3) for monitoring roof failure. To monitor the effects
of the mining operation on the groundwater resources of the area, the U.S. Geological Survey
designed and implemented a hydrologic data collection network by six additional monitoring
wells that were constructed at different depths above and below the phosphate zone (Figure
2.21). Water-level measurements and water quality samples were collected before, at periodic
intervals during, and after the mining operation. Continuous pressure recorders were installed in
the wellheads of two artesian wells to measure the water levels in the aquifer below the
phosphate zone, as well as in aquifers immediately above the phosphate zone. The recorder in
the artesian well above the phosphate zone registered very large and sudden drops in fluid
pressure (Figure 2.22) when roof failures occurred in Boreholes 1 and 2 and mining in air was
attempted. No such pressure changes were observed in the aquifer below the deposit. The
59 |
Colorado School of Mines | 2.2.9) Borehole Mining (Low Pressure/High Volume Systems)
In 1932, Clayton received the first patent for a tool that utilized a waterjet nozzle
assembly to fragment rock adjacent to a borehole and a down-hole slurry pump to lift the broken
ore to the surface. Patents on similar borehole mining tools were issued in later years to
numerous researchers including Fly (1964), Wennenborg (1973), Archibald (1983), Wang
(1989) and Abramov (2002). Fly’s tool was utilized to excavate sandstones, limestones, and
shale to a maximum depth of 100 m. The achieved mining rate was 0.8 m3/min. The lateral
(radial) distance from the borehole for the excavated cavity was 9.1 m. The apparatus had two
sidewall nozzles that operated at 800 psi (5.52 MPa) and 200 gpm (12.6 L/s) per nozzle. The
slurry was made to flow into the intake of a down-hole jet pump that hoisted it to the surface
[33].
FMC Corp built an apparatus as described in the Wennenborg (1973) patent. This tool
was tested in several phosphate deposits in eastern North Carolina. A high-volume, low-pressure
waterjet was applied in this system in order to slurry the ore, where an eductor was used to lift
the slurry to the surface [33]. Marconaflo, Inc. built an apparatus as described in the Archibald
(1974) patent. This tool was applied to mine uraniferous sandstones and tar sands on an
experimental basis. The jet-cutting unit consisted of a nozzle and high-pressure piping that rode
on a vertical rail attached to the main body of the device. The role of the rail was to allow the
nozzle to move independently of the slurry pump. The cutting-jet was operated at 400 to 499 psi
(2.76 to 3.44 MPa) and 150 to 170 gpm (9.4 to 10.7 L/s) [33]. The main disadvantage of the
aforementioned tools were their limited area of application, where each of these tools was
designed for a specific borehole mining application and possessed very little flexibility.
61 |
Colorado School of Mines | 2.2.10) Borehole Mining and Coal Hydraulic Fragmentation (High Pressure/Low Volume
systems)
Cheung et al. [34] studied waterjet coal-cutting capability based on the results of a
literature survey of related empiric data. Their goal was to design a BHM tool for remote
extraction of coal. Based on their research, they established the pressure and the flow rate as
4,500 psi (31 MPa) and 100 gpm (6.31 L/s) respectively and found that a waterjet with an
diameter of 0.223 in. had the capability of cutting coal in a slot at least 7 ft (2.13 m) deep. They
achieved a mining rate of approximately 8 to 10 tons/hr for a 4-hour sustained period. Their
objective for the system was to reach 30 ton/hr. The results of the study showed that pumping 20
tons of coal per hour and 200 gpm (12.6 L/s) of water to a height of 200 ft requires a primary jet
of 670 psi (4.6 MPa) and 260 gpm (16.4 L/s). Figure 2.23 shows a top view of the jet cutting
pattern that was used during their testing. The small circle in the middle represents the top view
of the borehole, where L is the nozzle standoff distance to the face of the coal. The cutting
n
pattern employed in most subsequent BHM applications are similar to those in Figure 2.23.
Figure 2.23. Cutting pattern example. [34]
62 |
Colorado School of Mines | Several researchers have tested high pressure/low volume jets for coal slotting
applications. Due to the small volumes removed, applying a rock slotting excavation method in a
borehole mining is inherently not efficient. However, studying this method does provide insights
into cutting dynamics. The following tests/research did not include jet rotation or oscillation,
which would greatly improve cutting but rather a simple jet being traversed across a rock at a
given velocity.
Wallace, et al. [31] studied the waterjet cutting of bituminous coal and found that a
waterjet with a 4,000 psi (27.6 MPa) stagnation pressure and a 0.25-in. diameter nozzle, cut 2-in.
wide slots. The test was conducted in a slot cut by a previous pass, and thus, established the
condition of a confined jet. Olsen [26] conducted waterjet cutting experiments of Wilkeson
sandstone under both confined and unconfined conditions and found that the effective cutting
distance for an unconfined jet is at least twice that of a confined jet.
Fowkes and Wallace [31] conducted parametric studies on hydraulic coal cutting. They
found that the slot depth per pass was about 24 in. when the nozzle traverse speed was 5 in./sec.
In these laboratory tests, the waterjet stagnation pressure was approximately 4,000 psi (27.6
MPa), and the nozzle diameter was 0.37 in.
Nikonov and Goldin [35] studied the effectiveness of coal penetration by waterjets. They
found that, for a fixed nozzle standoff distance, the depth of slot per cut, h, can be approximately
predicted by the following empirical equation:
63 |
Colorado School of Mines | h p V
B( 0 0.2)( T )0.5
d V (2.6)
c 0
Where B is an experimentally determined constant (B = 0.5 for coal), P is the waterjet
0
pressure (psi), V is the traverse speed (in. /sec), V is the waterjet speed (in. /sec), σ is the
T 0 c
compressive strength of the rock (psi), and d is the diameter of the nozzle in inches. Their data
also showed that there is an optimum traverse velocity V , for a given V and d. The relationship
T 0
is:
V 1
T (2.7)
V 175584d
0
Malenka [22] conducted hydraulic mining tests on anthracite. Figure 2.24 shows the
cumulative depth of slot versus the number of cuts for the following parameters: P = 5,000 psi
o
(34.5 MPa), d = 0.29 in., and V 2.4 ips (in. /sec). In this equation, d and V are nozzle diameter
T = T
and traverse velocity, respectively. Malenka also found that for traverse velocities up to 44 in.
/sec, the excavation rate increased of traverse velocity.
Figure 2.24. Cumulative depth of cut versus number of passes on anthracite. [22]
64 |
Colorado School of Mines | Table 2.1 presents a summary of the coal fragmentation tests for high pressure/low
volume waterjets as prepared by Knoke [29].
Table 2.1. Data summary for high pressure/low volume waterjet. [29]
2.2.11) Applications of Borehole Mining For Different Minerals
2.2.11.1) Coal
Borehole mining operations were performed in 1975-76 by Flow Industries, Inc. under a
US Bureau of Mine (USBM) research contract. The site was located 4.8 km south of Wilkeson,
WA, and contained a 5.4 m thick seam of bituminous coal dipping at 42°. Two shallow
boreholes (7.6 m and 10.7 m) and one deeper borehole (26.8 m) were drilled through the dipping
coal seam and cased to the hanging wall. They used the shallow boreholes for preliminary tests
designed to optimize mining procedures to be followed during a 4-hr production test in the deep
borehole. Applying 4,500 psi (31 MPa) and 100 gpm (6.3 L/s) jets in the preliminary tests
induced a cutting radius of 3 m. The cutting radius attained in the production test was 4.5 m.
They applied 4,500 psi (31 MPa) and 200 gpm (12.6 L/s) jets in the production test. The
maximum production rate obtained was 7 tons/hour. Given the testing results and that no
mechanical failures of the mining tool occurred during these field trials; it supported the
conclusion that it was technically feasible to apply BHM to mine coal. However, the test
65 |
Colorado School of Mines | production rate was well below expectations. It was believed that even without the use of
parametric testing, the technology would require extensive work before it could be made
commercially viable [10].
2.2.11.2) Uranium
Marconaflo, Inc. and Rocky Mountain Energy (RME) conducted tests from 1973 to 1975
in the Powder River Basin and the Gas Hills of Wyoming using borehole mining. A total of
approximately 700 tons of uranium ore was mined, but problems were encountered with sand
plugging and severe wear of the slurry pump. The USBM cooperated with RME at its Nine-Mile
Lake site, Natrona County, WY, in a borehole-mining test in 1977 and 1978. RME prepared the
site, drilled a water supply well, constructed a pond and lined it with polyethylene, and drilled
three (406-mm) cased boreholes to a depth of 30.5 m into the Teapot sandstone orebody. Flow
Industries, Inc., under contract with the USBM, modified the tool used for coal at the Wilkeson,
WA, site and conducted tests at the sandstone mining operations. The modifications included
fitting the mining tool with a turning-vane nozzle ensemble designed to pass 300 gpm (18.9 L/s)
at 2,000 psi (13.8 MPa). During mining operations, approximately 940 tons of ore were mined
from depths of 22.9 to 30.5 m at an average rate of 7 tons/h. The standoff distances were as large
as 25 ft (7.6 m). Figure 2.25 shows the created cavity in one of the boreholes at Nine-Mile Lake.
The white, 2-in. (50.8-mm) diameter PVC pipes in the foreground are placed in monitor holes
drilled 3.05 and 6.10 m from the center of the borehole. This photographic survey showed that
roof failure was confined to a 7-ft (2.1-m) radius from the center of the borehole. This indicates
that rock within this radius was damaged during drilling. [10]
66 |
Colorado School of Mines | Figure 2.25. Cavityproduced during borehole mining. [10]
According to Savanic [33], uranium sands are considered to be a likely prospect for
borehole mining because: (1) the ore has a high unit value, (2) the sandstones can be cut by low-
pressure (1,000 to 3,000 psi or 6.90 to 20.69 MPa) waterjets, and (3) many deposits are shallow,
small, irregularly shaped, and isolated. Many of these deposits cannot be mined by conventional
methods, but are amenable to the selective capabilities of borehole systems.
2.2.11.3) Phosphate
Several borehole phosphate mining tests were conducted by the USBM and Flow
Industries, Inc., in cooperation with Agrico Mining Co. in St. Johns County, near St. Augustine,
FL, in 1980. An estimated 1,700 tons of phosphate ore were produced from three boreholes. The
borehole depths ranged from 70.7 to 77.1 m. Mining in the first hole was conducted below the
water table. The goal was to determine the feasibility of mining in a borehole and cavity filled
with water. By cutting with a submerged jet in a 360° arc, 860 tons of ore at an average rate of
36 ton/hr was obtained. After pumping water from the cavity, the immediate roof failed. This is
67 |
Colorado School of Mines | an indication that the water pressure had a significant influence in supporting the roof. Although
the roof in the cavity failed, no surface subsidence was observed. However, no further mining
could be performed in this hole. This experiment did indicate that borehole phosphate mining in
a submerged condition was technically feasible [10].
A second borehole was drilled and employed a system where the cavity was air-filled.
Mining was confined to a 30° arc and a 330° pillar supported the roof. Roof failure happened
after 300 tons of ore had been mined. It was concluded that the roof rock was not sufficiently
competent to permit mining in an air environment and that any future mining would require that
the cavity be filled with water. Finding a pillar with adequate size under different conditions to
keep the cavity stable was the ultimate goal of this project. The third borehole test was in
combination of flooded conditions and mining in air (air-shielding concept). Under this
condition, the waterjet stream was shrouded by compressed air, to provide it with greater
standoff distance. The air-shielded waterjet issued from a conventional borehole cutting nozzle
with an annulus for compressed air around it. Compressed air was supplied through conduits in
the borehole mining tool connected to an air compressor at the surface. About 430 tons of ore
was extracted without activating the air-shield. Air-shield was applied after the solids content of
the slurry began to decrease. This was an indication that the submerged jet had reached its
maximum radius in a submerged mode of operation. By activating the air-shield, an additional
137 tons at a rate of 25 tons/hr was obtained. The cavity radius was approximately 5.5 m. No
roof falls occurred while mining was taking place. The results indicated that phosphate can be
mined effectively in a flooded cavity and that air shielding substantially increased waterjet
effectiveness while operating submerged [10]. This research demonstrated that the nozzle could
68 |
Colorado School of Mines | be rotated 360 degree and translated along the axis of the borehole. By translating the jets,
extracting a larger vertical increment is possible while the slurry pump stays at the base of the
borehole where the slurry density is the highest. In Boreholes 1 and 3, the nozzle rotated 360
degree and the vertical increment ranged from 50.8 to 152.4 mm [23]. Tables 2.2 and 2.3 show
the waterjet’s specification for the first and third boreholes.
Table 2.2. Waterjet specifications for Borehole 1. [10]
Parameter Unit Specifications
Cutting jet pressure psi 500-2000
Cutting jet flow rate gpm 500-750
Cutting jet diameter in 0.475 and 0.96
Jet pump pressure psi 700-1500
Jet pump flow rate gpm 400-700
Jet pump nozzle diameter in 0.68 and 0.8
Jet pump throat diameter in 2 and 2.25
Turntable speed rpm 2-15
Mining arc degree 360
Mining depth ft 232-253
Vertical increment in 2-6
Table 2.3. Waterjet specifications for Borehole 3. [10]
Parameter Unit Specifications
Cutting jet pressure psi 1000-1900
Cutting jet flow rate gpm 423-499
Cutting jet diameter in 1
Air shield pressure psi 250
Air shield flow rate ft3/min 150
Air shield nozzle opening in 0.03
Jet pump pressure psi 490-1000
Jet pump flow rate gpm 432-491
Jet pump nozzle diameter in 0.7
Jet pump throat diameter in 2
Turntable speed rpm 1.8
Mining arc degree 360
Mining depth ft 235-249
Vertical increment in 2-6
69 |
Colorado School of Mines | According to Hrabik [36], borehole mining was found to be more economic if the
overburden thickness was 46 m (150 ft) or greater. A spacing of 21m (70 ft) between the
boreholes will yield a final barrier of 3 m (10 ft) between mined-out cavities. Barriers must be
maintained so that phosphate slurry is not washed into previously mined cavities and lost from
recovery. Mining recovery was estimated to be 67 percent. Future testing demonstrated that
mining recovery may be improved by modifying the shape of cavities or reducing the cavity
spacing.
2.2.11.4) Precious Metals
Placer deposits often contain a significant percentage of cobbles and boulders, associated
with permafrost and glacial debris. Because elasticity and heterogeneity of the ice-and-gravel
conglomerate causes erratic loading on direct contact tools, these deposits are extremely difficult
to excavate. Savanic et al. [37] did research on an approach that selectively mines the ore-
bearing portion of the deposit using a high-pressure waterjet to thaw the gravels and a downhole
hydraulic-lift-type pump to bring the excavated ore to the surface as slurry. The excavator
delivered a water flow of 890 liters/min. (235 gal/min.) through a 15.7 mm (0.62-in.) diameter
nozzle at 6.9 MPa (1,000 psi). Through empirical testing, the system proved to be capable of
2 2
excavating permafrost at a rate of approximately 9 tons/hr and drive a 0.1 m (1 ft ) horizontal
heading a distance of 4.6 meters (15 ft) in approximately 10 min. Figure 2.26 shows a schematic
of placer borehole miner. Two nozzle assemblies are located in drill holes approximately 7.5
2
meters (25 ft) apart. Each nozzle assembly cuts a long, horizontal passage approximately 0.1 m
2
(1-ft ) in cross section toward the other hole until the two passages intersect. The cutting jet of
one assembly forces slurry toward the inlet of the slurry pump located at the base of the second
70 |
Colorado School of Mines | cutting jet was placed into the auxiliary hole and directed toward the production hole. Production
during the experiment encountered two obstacles, clogging of the slurry inlet and large boulders.
Before developing this borehole system into a commercially viable technology, these two
problems need to be resolved.
Table 2.4 presents the jet pressure and flow rate that were used in BHM application for
different materials. According to Table 2.4, the maximum applied jet pressure and minimum
flow rate for coal were 4,500 psi (31 MPa) and 200 gpm (12.6 L/s), respectively. [23]
Table 2.4. Some jetting parameters that have been applied in BHM. [23]
Mineral Location Jet Pressure Flow rate
Coal Wilkeson, Wa 31 MPa (4500 psi) 200 gpm (12.6 L/s)
Sandstone Natrona county, Wy 10 MPa (1500 psi) 350 gpm (22.1 L/s)
Oil sands Taft, Ca 6.9 MPa ( 1000 psi) 300 gpm (18.9 L/s)
Phosphate St.Johns County, Fl 12 MPa (1800 psi) 420 gpm (26.5 L/s)
2.3) Influence of Geological Factors
According to Jeremic [14], the most important geological parameters for the successful
application of borehole mining are the shape, strike, dip, thickness, and structural irregularities of
the deposits. These factors influence the cavity geometry, coal mineability, strata stability, and
the economic feasibility of the method. The stability of any underground opening depends on the
lithology and the planes of weakness in the strata around the opening. In borehole mining,
however, the stability of the strata also depends on the impact of water saturation, rock strength,
72 |
Colorado School of Mines | and behavior (i.e. swelling) [14]. A premature roof collapse can lead to the loss of equipment and
mineral resource by the caved rock, where an effort to quantify a design protocol is the
motivated behind this research.
2.3.1) Shape of Coal Seams
The shape of the coal seam is a significant parameter in the design of a borehole mining
system. Favorable coal seam geometry will increase the productivity of the excavation system
and its recovery. The basic shape of coal seams is governed by their sedimentary origin which
produces bedded deposits. According to Jeramic [14], hydraulic transportation by slurry should
be in the direction of gravity, if at all possible. Coal seams with regular strikes will allow the
cavities to have a constant gradient that is critical for establishing a continuous gravity flow of
the coal slurry. Flat coal seams with inclination angles of 0-6° are insufficient to obtain the
required gradient that enables the coal slurry to flow by gravity. Under these conditions, the
extracted coal crushed by monitor jet has to be bailed by the eductor via fluid suction and
venturi.
The thickness of coal seams also affects the mode of extraction and it is important in
strata control. Borehole mining of thick seams can result in a build-up of excessive roof pressure
which could cause significant cavity displacement and deformation. It could ultimately result in
the roof caving before the completed extraction of coal from the cavity. Subsidence control is
generally more challenging due to high volume of the extracted cavity. Some geologic
conditions, like substantial variation in thickness and continuity, can cause significant
operational challenges for underground mining. However, due to the flexibility of borehole
73 |
Colorado School of Mines | mining, these challenges can be avoided by designing appropriate layouts and panels provided
that these structural irregularities are not densely spaced. [14]
2.3.2) Lithology of Rock Strata
According to Jeramic [14], borehole mining highly depends on lithology and the planes
of weakness in the strata around the excavated cavity. In borehole mining, the stability of the
strata depends on the influence of water saturation impacting rock strength and behavior (i.e.
swelling and slaking). The lithology of coal-bearing strata is dependent on the conditions that
were present during deposition of the vegetal matter from which the coal deposits were formed.
[14]
Williamson [39] has classified coal-bearing strata as the following:
1) Siliceous sediments with the following members
Conglomerate, highest strength in the series
Sandstone, often resistant to cave after coal extraction
Siltstone, usually cave after coal extraction
Shales and mudstones, susceptible to moisture, and may cave before coal
extraction is completed
Clay rocks, show low strength and higher moisture content with very low
bearing capacity [40].
2) Carbonate sediments are represented by the following two rock types
Limestone, moderate strength and good caving properties
Dolostone, similar properties to limestone
3) Integrated sediments are represented by hybrid rock types
74 |
Colorado School of Mines | Shaly limestone, lower strength
Silty limestone, lower strength
Lithology of the strata surrounding the coal deposit is critical in borehole mining design
due to the correlation between deposit structure and strength and the likelihood of roof failure.
Since borehole mining is a remote and unmanned mining method, roof sags and falls should not
affect coal production if the failure does not occur during the extraction procedure. One of the
important factors in the success of borehole mining designs is that the large cavity resulting from
the extraction of coal should collapse as soon as possible after completion of mining so that
adjacent cavities and pillars (segment between cavities) will be distressed. Sometimes when a
high rank coal seam is hosted in strong strata, such as sandstone, strain accumulates and is
released suddenly by violent fracturing [51]. Hydraulic fracturing can be utilized as an integral
part of the borehole mining operation. By water infusion in the coal seams, the property of the
coal can change from elasto-brittle to elasto-plastic. So, yielding deformations are induced
instead of sudden fracturing [51]. In some cases, coal seams lie in soft strata such as claystone.
These coal seams are usually represented by geologically younger coking-coal deposits. Rapid
extraction of the cavity may prevent development of accelerated deformation and collapse of
cavity roof before coal has been extracted [51].
One of the major concerns in designing borehole mining is the evaluation of cavity
stability based on stress concentration and relaxation around the extraction area. The stress
concentration generally depends mainly on two parameters: the volume of removed coal and
consequent load transfer on the cavity face, and the depth of cover which governs the strata
75 |
Colorado School of Mines | pressure. Cavity stability can be maintained by caving the mined-out area before the stress
concentration on the production face becomes critical [51]. Designing suitable cavity shape and
extraction orientation can be mitigation factors and are discussed relative to the objectives of this
dissertation.
2.3.3) Geological Structure Defects of Rocks
Besides lithology, planes of weakness are of greater concern compared to strength and
stability of coal bearing strata [41]. The quantification of geological structural defects of the rock
strata must be one of the key parameters in borehole design, since the extracted cavity is without
artificial support. In addition, the degree of deterioration of rock rigidity due to saturation of
planes of weakness should be evaluated [14]. The effects of different types of rock mass defects
on borehole mining are as follows:
1) Filled joints do not impact the stability of strata in borehole mining because of their wide
spacing.
2) Shear joints weaken the strength of the strata under influence by waterjets. However, rapid
extraction of coal does not permit development of their negative impacts on roof stability.
3) Shear slips are continuous defects at low angles or parallel to bedding with low residual
strength. They can have a damaging impact on borehole mining, since they can not be controlled
in unsupported openings. [14]
2.3.4) Strength Deterioration Due to Water Saturation
The presence of large quantities of water in borehole mining will result in the saturation
of strata within the cavity’s roof and floor. Water has a significant effect on intact rock and the
76 |
Colorado School of Mines | triaxial compressive strength of rocks decreases after the adsorption of water and the yielding
strength varies almost linearly with the water content [14]. This situation is similar for other
strength parameters such as stiffness, cohesive and internal friction angle. According to Jeramic
[14], based on several empiric experiments, the amount of moisture and its effect on rock
characteristics primarily depends on lithology and rock porosity. Tensile strength is an important
parameter in the evaluation of cavity stability. The magnitude of the tensile stresses is increased
by the pore water pressure reducing tensile strength. Brazilian tests of dry and saturated
sandstone discs at various loading angles to bedding planes showed reductions in strength
between three and eight times. Design of borehole mining based on the strength values of dry
rock can be misleading, since the saturated strength values in tension can be significantly lower.
[14]
2.4) Conclusions
By studying the literature, the following conclusions can be reached:
There are several in-situ mining methods (i.e solution mining and BHM) available. BHM
is more attractive for the mining some commodities including coal, since coal is not a
soluble mineral.
A majority of the applied BHM utilized low waterjet pressure (less than 4500 psi or 31
MPa) and high volume flow rates (more than 100 gpm or 6.3 L/s). Systems that use high
volume flow rates are not economically feasible for the extraction of thin seam deposits
because of dilution and the lack of geometry control governing cavity formation.
77 |
Colorado School of Mines | Choosing the particular pressure and flow rate depends on the available power and the
cuttability/cutting efficiency of the ore, the flow required to transport and bail the
cuttings, and the required cavity dimensions to achieve economic viability.
There are three preferred sets of conditions for the pressure and the free surface level of
ore slurry. Those conditions are referred to as high, medium and low conditions. Mining
under low conditions maximizes the mining rate and the size of the cavity that can be
mined. Mining under medium conditions can prevent interference with adjacent aquifers.
Mining under high conditions allow mining at great depths, assists the educting of the ore
slurry up to the surface and minimize the power required for educting.
Fine, high pressure waterjets have a short length of initial section, beyond which there is
a sharp drop in the stream velocity and impact force. In most applications, there is an
optimum distance (from nozzle) for a maximum mining rate between 200 and 400 nozzle
diameters.
In practice, the maximum borehole mining cavity radii have been measured as 26 ft (7.9
m) in sandstone.
If the jet is shrouded by compressed air, waterjets can be more effective underwater
(below water table). While borehole mining underwater has the advantage of providing
additional roof support during mining, the standoff distance that waterjet can reach
effectively will be reduced.
78 |
Colorado School of Mines | It is important to estimate some of the key operational parameters of the borehole mining
system such as standoff distance, nozzle diameter, pressure drop, cutting depth, to
determine project economics.
There are some approaches that can be used to predict the stability of the cavity after
completion of BHM. However, some parameters, such as the effect of internal pressure
around the cavity and in-situ stress during BHM have not been studied. The goal of this
dissertation is to study these confounding parameters and evaluate their impacts on the
cavity stability.
The most important geological parameters for successful applications of borehole mining
are the shape, strike, dip, thickness, and structural irregularities of the coal deposits. The
stability of any underground opening depends on the lithology and the planes of
weakness in the strata around the opening. In borehole mining, however, the stability of
the strata also depends on the impact of water saturation, rock strength and behavior (i.e.
swelling).
Tensile strength is an important parameter in the evaluation of cavity stability. The
magnitude of the tensile stresses is increased by pore water pressure reducing tensile
strength.
Large cavities resulting from the extraction of coal should collapse as soon as possible
after completion of mining so that adjacent cavities and pillars (segment between
cavities) will be distressed.
79 |
Colorado School of Mines | Sometimes when a high rank coal seam lie in strong strata such as sandstone, strain
accumulates and is released suddenly by violent fracturing. Hydraulic fracturing can be
utilized as an integral part of the borehole mining operation. By water infusion in the coal
seams, the property of the coal can be changed from elasto-brittle to elasto-plastic. So,
yielding deformations are induced instead of sudden fracturing.
In some cases, coal seams lie in soft strata such as claystone. These coal seams are
usually represented by geologically younger coking-coal deposits. Rapid extraction of the
cavity may prevent development of accelerated deformation and collapse of cavity roof
before coal has been extracted.
One of the major concerns in designing borehole mining is the evaluation of cavity
stability based on stress concentration and relaxation around the extraction area. The
stress concentration generally depends mainly on two parameters: the volume of removed
coal and the load transfer to the cavity face as a consequence of mining, and the depth of
cover which governs the strata pressure. Cavity stability can be maintained by caving the
mined-out area before the stress concentration on the face becomes critical. Designing
suitable cavity shape and extraction orientation can be mitigating factors.
80 |
Colorado School of Mines | the shape of an underground opening has an effect on the boundary stresses around the opening
[49].
Depending on the nature of the cavity/structure, some factors have more influence than
others. According to Bieniawski [48], the following are important for designing underground
working and cavities:
1) Modulus of elasticity (deformability),
2) Tensile strength,
3) Frictional properties (cohesion and friction angle) are important in yield zones, and
4) Post-failure modulus.
Cavity size and shape are two important factors that affect the stability of cavity-like
structures. Peng [50] identifies several factors that need to be considered for designing a panel
layout in longwall mining, which is applicable to this dissertation discussion. Those factors
include: panel dimensions, seam inclination (in this research, only a flat seam will be
investigated), roof and floor strata, geological anomalies, and in-situ stress, and multiple seam
mining.
3.3) Coal Geologic Factors and Rank
Although an extensive discussion of coal character and geologic composition is beyond
the scope of this dissertation, a few of the more important points are summarized to provide
82 |
Colorado School of Mines | some context for the subsequent discussion of mechanical properties and cavity (borehole)
stability.
3.3.1) Geologic Factors
According to Jeremic [51], the mechanical properties and performance of rock strata
depend on their lithology, sedimentary structure, and discontinuities. For mining, three groups of
sedimentary rocks are common.
1) Carbonate sediments: These sediments are usually deposited in delta or calcium rich
environments. Three rock types that may accompany those deposits are fine-grained limestone,
dolostone, and shaly limestone. Generally, the strength of limestone guarantees some degree of
stability; however, in the case of overloading conditions (high stress environments), the mine
floor will sometimes respond with rock bursts. Despite similar strength properties between
limestone and dolostone, the presence of joints and bedding in dolostone can increase the effects
of weathering which may cause a reduction in strength. This is particularly true for shallow
mines. This weathering or the presence of joints and bedding in dolostone can cause the rock
strata to lose structural integrity and cause additional mine stability problems. Shaly limestone is
associated with carbonaceous and clay laminas. This often results in a reduction of strength when
swelling occurs.
2) Siliceous sediments: These sediments are the most frequent of coal deposits mined
around the world. The history of coal basin deposition and cyclical climate change has had a
significant effect on the lateral and vertical continuity of siliceous rock. The main rock types are
conglomerates, sandstones, siltstones, and shales. Conglomerates don’t normally have the
continuity of lateral extension. If conglomerates are the primary rock type comprising the roof of
83 |
Colorado School of Mines | a cavity or underground working, stability can be dangerously affected since conglomerates are
not resistant to caving. The lack of caving tendency during extraction is a positive characteristic
of mining in sandstone. Although sandstones generally possess a high bearing capacity, it can
still host rock burst conditions by violently releasing its accumulated strain. Since siltstones
possess a transition characteristic somewhere between sandstones and shale, its interbedding can
extend to sandstone and shale beds with various compressive and tensile strengths.
3) Clay sediments: clay sediments are found in areas where lithification is incomplete.
Clay sediments are typically associated with the borders of flowing rivers. There are three types
of clay sediments: clay shales, mudstone-siltstones and clay sandstones. Clay shales are made of
clay with different mineralogy compositions. Depending on the composition, stability may be
affected. The results of laboratory research show that when the predominant deformation is
extrusion, the rock shows visco-plastic behavior. The strength of mudstone-siltstone depends on
the ratio of silt to clay, where sudden roof falls have been observed in mudstone or siltstone
strata. Both clay shales and mudstone-siltstones have fine bedded textures, where the failure
mode is usually elastic-plastic. [51]
The overlying roof strata above coal deposits vary widely, depending on regional and
local geology, as well as the depth of the coal seam being mined. The roof conditions can also be
affected by the presence of water and high horizontal stresses. The types of deposits immediately
above the coal can range from massive thick layers of sandstone or limestone, to strong
laminated layers of siltstone and sandstone, to thin weak layers of shale or mudstone. One of the
most frequently observed types of overlying strata in coal mines is a laminated shale roof that is
often prone to immediate skin failure. [51]
84 |
Colorado School of Mines | Peng, et al. [52] developed methodology for a roof classification with the following
categories: unstable, medium stable, and stable. Unstable immediate roofs consist of weak or soft
carbonaceous shale and fractured sandy shale. Medium stable immediate roofs consist of hard
shale, sandy shale, and sandstone. A stable immediate roof is thick and consists of strong sandy
shale or sandstone, conglomerate, and limestone. In many cases, these rock types can be left
unsupported for long periods of time (e.g., up to 8 hours).
One of the important coal strata characteristics is its bedded structure. Immediate roof
thickness, bedding planes, coal streaks, and soft rock bands have a significant impact on roof
stability. Increasing the thickness of the immediate roof will usually decrease the roof
convergence in unsupported areas. A stratified immediate roof has different stress and strain
features compared to those in a non-bedded immediate roof. Every layer in the stratified
immediate roof has its own tensile zone in the upper segment and compressive zone in the lower
part. While the tensile zone in the first layer is smaller; it has the largest tensile stresses [53].
3.3.2) Coal Rank
Coals are combustible sedimentary rocks formed primarily from plant debris. Different
types of coals exist, and they are defined by their constituent macerals. These macerals are the
elementary microscopic particles in coals, and are analogous to minerals in other types of
sedimentary rocks. Macerals consist of various kinds of lithified plant debris, including spores,
pollen, resins, waxes, and cuticle. The common groups of macerals are vitrinite, exinite, and
inertinite [54].
The physical and chemical properties of coal are established by two factors: (i) the nature
of the original plant debris, and (ii) the degree of chemical alteration (i.e., digenesis) of the
85 |
Colorado School of Mines | the Ruhr District of Germany, as compiled by Jones et al. [57] and Brauner [58], respectively.
Two significant points can be interpreted from these data: (i) carbon content alone does not
impact the absolute value of coal strength, as the U.S. coals are systematically stronger than the
German coals; and (ii) there is a consistent trend in the relative variation of coal strength across
the ranks represented by these two datasets. Specifically, the uniaxial compressive strength
(UCS) reaches a minimum value of roughly 5–7 MPa for ranks in the medium- to low-volatile
bituminous range (carbon content 75%–80%). For higher ranks, the UCS tends to increase with
increasing carbon content, where there is a trend of increasing UCS with decreasing carbon
content for the high-volatile bituminous ranks. The dry, mineral-matter free (dmmf) volatile
matter content is equal to 100 minus the dmmf carbon content [55].
Figure 3.1. Unconfined compressive strength of coals as a function of carbon content for
bituminous and anthracitic coals. [57-58]
87 |
Colorado School of Mines | 3.4) Stress Fields around the Coal Seams
Stress fields happen naturally in the ground prior to mining, and these effects are critical
when considering ground control issues. The horizontal and vertical stress fields can cause major
stability issues if they are not controlled through adequate barrier pillar design, cavity size, and
cavity orientation. The magnitude of vertical stress increases with depth and is normally the
result of the overburden weight above the excavation. Horizontal stress magnitudes and
directions vary primarily with geographic location and geologic structure, and can exceed the
vertical stress by a factor of 3 or more, particularly in shallower deposits. Research suggests that
tectonic forces are responsible for the magnitude and direction of horizontal stress fields, which
can cause roof buckling and bed separation, in addition to other structural issues [60]. Since the
1940’s, researchers have believed that large horizontal stresses were responsible for much of the
roof damage experienced in underground coal operations [59]. In addition to the magnitude of
horizontal and vertical stress, the ratio of horizontal stress to vertical stress (σ σ ) can
hmax/ v
influence roof stability as well. [60]
Horizontal in-situ stress is a major issue that impacts roof strata behavior, and orienting
the cavity in a particular direction can significantly improve its stability. Cavities are most stable
only in the direction of σ ; therefore, the most stable condition is to have cavities parallel to
hmax
maximum horizontal stress. The ratio of horizontal stress to vertical stress (σ /σ ) is known as
hmax v
a k value. A high k value can be an indicator of potential roof control problems. Additionally,
extremely low k values can cause as much potential instability as high values. Figure 3.2 depicts
the effect of low and high k values on safety factors for a coal mine. This plot was created by M.
Gadde [61] using computer modeling. [61]
88 |
Colorado School of Mines | Figure 3.2. Effect of low and high k values on safety factors. [61]
The roof layers (referred to as bed numbers) in Figure 3.2 begin with the weak immediate
roof and end with the most stable strata. This plot shows that an optimal k value for the highest
safety factor is between 0.5 and 1.0, with lower or higher ratios producing lower safety factors.
Orienting cavities (or extraction orientation) optimally is an excellent way to improve the
stability. Areas that have extremely low or high k values can potentially benefit from
appropriately directing their cavity’s orientation to optimize their roof stability. [60]
According to Braun [62], when underground openings are parallel to the in-situ principal
stress, there are no shear stresses in that direction. Instabilities commonly occur as breakouts
perpendicular to the opening or as fractures in the direction of the maximum principal stress.
3.5) Horizontal Borehole Stability Analyses for Coals
Horizontal directional drilling is used in some geological settings to produce methane
from a variety of shale deposits. In mining, this technology can also enhance the effectiveness of
89 |
Colorado School of Mines | coal degasification procedures and to aid in the delineation of coal reserves. Coal deposits tend to
be mechanically weak, where they are prone to borehole instability problems during drilling,
completion, and production operations. Due to similarities with borehole mining sequences, an
analogy can be established relative to the challenges and difficulties associated with this type of
drilling. The goal of the following sections of this chapter is to address some confounding
parameters that may affect the stability of the excavation process associated with borehole
mining.
Figure 3.3 illustrates different options for drilling horizontal (or near-horizontal)
boreholes into coal seams. According to Hawkes [55], for petroleum applications, drilling is
typically performed from a surface location. In the mining industry; however, it may be feasible
to drill from either surface or subsurface locations. The basic principles of directional drilling
methods in mining and petroleum applications are essentially the same, involving down-hole
drive motors mounted on adjustable, bent bottom hole assemblies. Hole sizes tend to be smaller
for mining boreholes (e.g., 96 mm or 3.78 in.), although for the purposes of degassing, options
for using larger boreholes up to 8 in (200 mm) have been applied. Based on the literature, the
vertical increment (hole diameter) for BHM ranged from 2 to 6 in (50.8 to 152.4 mm).
Another significant difference between these industries is the composition of the drilling
fluids commonly used. Pure fluids, such as water or air, are often used for mining, whereas a
wide spectrum of drilling “mud” base fluids (e.g., water, diesel oil, mineral oil) and additives
(e.g., bentonite) are common for petroleum applications. Usage of low-density drilling fluids
(e.g., nitrified water, foams, and mists) has increased in petroleum application involving
horizontal wells to prevent underbalanced drilling. Underbalanced drilling is performed with a
borehole fluid at a pressure less than the native formation pore pressure. This type of drilling
90 |
Colorado School of Mines | may cause formation damage. Near-well permeability reduction resulting from physical and/or
chemical interactions of drilling fluids with the producing rock formations are referred to, by
convention, as formation damage.
Figure 3.3. Methods for drilling horizontal boreholes into coal seams: (a) directional drilling
from the surface; (b) horizontal drilling from surface outcrop; (c) horizontal drilling from mine
shaft; and (d) tight-radius drilling from the surface. [55]
Borehole stability is primarily a function of how the rock surrounding a hole responds to
the induced stress concentration during drilling and material extraction. If the rock is stronger
than the induced stresses, then the borehole will remain stable. If not, rock yielding will occur,
and may cause detachment of yielded rock from the borehole wall. This may ultimately result in
the borehole wall collapsing or in convergence. Whether or not this will lead to serious drilling
problems is a function of many factors, including post-yield rock behavior, the volume of rock
that detaches, and the sensitivity of drilling operations to hole enlargement or convergence. The
91 |
Colorado School of Mines | critical factors affecting borehole instability include in-situ stresses, rock mechanical properties,
formation pore pressure, and fluid pressure within the borehole. [55]
According to Hawkes [55], a simple method for evaluating horizontal borehole stability
is to calculate the stresses near the borehole using a linear elastic model and to compare these
stresses with a rock strength criterion to determine if shear yielding will occur. By considering
the case of a borehole drilled parallel to the direction of principal in-situ stress where plane strain
conditions are in an isotropic, homogeneous rock mass, it is relatively simple to solve for the
stresses around the borehole. For a rock mass characterized by a linear Mohr–Coulomb failure
criterion, the critical wellbore pressure (p (critical)) at which shear yielding will initiate at a
w
point on the borehole wall is given by the following equation:
3 A UCS(N 1)P
P 1 3 P P a (3.1)
w N 1
Where:
σ and σ are the largest and smallest principal stresses, respectively, in the plane normal
to the borehole axis;
A is the poroelastic constant equal tov) / (1 v), where is Biot’s coefficient
p
and is usually equal to 1 for soft rocks, and v is the static Poisson’s ratio;
p is the change in pore pressure at the borehole wall equal to p - p , where p is the
a fm a
pore pressure just inside the borehole wall, and p is the in-situ formation (pore)
fm
pressure;
UCS is the unconfined compressive strength;
92 |
Colorado School of Mines | N is a frictional strength parameter equal to (sin) sin , where is the friction
angle); and p is the borehole pressure.
w
This model does not consider some material behaviors, such as nonlinear elasticity, strain
hardening, strain softening, and stress-dependent permeability. However, compared to more
complex models that account for these other factors, the linear elastic approach has the
significant advantages of being relatively easy to solve and requiring fewer input parameters
[55].
According to Hawkes [55], alterations in drilling-induced pore pressure may change the
in-situ saturation in coal. Often, there is no free gas phase in coals at in-situ conditions, since gas
molecules are adsorbed onto coal maceral surfaces. In overbalanced drilling, where there is no
reduction in pore pressure, this saturation condition will not change and the coal will remain
water-saturated. If the pressure in the borehole is kept lower than the pore pressure in the
formation being drilled (underbalanced), desorption of gas will be induced and a free gas phase
will develop. This might also produce capillary pressures which might decrease the rock
strength. The difference between the pressures of any two phases is referred to as capillary
pressure. In geologic terms, capillary pressure is the force required to squeeze a hydrocarbon
droplet through a pore opening.
According to Kang et al. [63], there are many factors that impact the stress redistribution
around the horizontal wellbore during drilling operations. The most influential factors are rock
mechanical properties, far field principal stresses, wellbore trajectory, pore pressure, drilling
fluid pressure, temperature, and time. [63]
93 |
Colorado School of Mines | 3.6) Orientation, Shape and Sequences of the Excavation
An essential aspect of this dissertation is the effect of excavation geometry on the overall
stability of the cavity produced. According to Pariseau [64], there are three major problems to
address in considering large, deep underground caverns: cavern orientation, cavity shape (i.e.,
dimension), and excavation sequence. Pariseau assumed a brick-shape cavern (Figure 3.4) in his
study. The data for his research was collected from the Yates formation at the Homestake Mine
in Lead, South Dakota. The first problem was addressed by alignment of the long axis of the
cavern with the major preexcavation principal stress (S ) and the short axis with the minor
1
preexcavation principal stress (S ). Figure 3.4 shows a block with dimensions of L , L , L ,
3 1 2 3
where L is the long dimension and L is the shortest dimension, and there are six orientations
1 3
possible. Based on the stress analysis, the most favorable orientation which minimizes stress
concentration and extent of yielding is present when L is parallel to S and L is parallel to S
1 1 3 3.
The second problem relates to the dimensions of a cuboid or block that minimize the stress
concentration for a fixed volume. In the favorable orientation, there is the possibility to optimize
edge lengths for the least stress concentration and minimal yielding, while meeting volume
requirements. The design guideline suggested here is to proportion the edge length in the same
ratios as the preexcavation principal stresses [64]. Thus, for the ratios M= S /S and M’= S / S ,
3 1 2 1
it should also have M= L / L and M’= L / L [64]. As previously discussed, the most favorable
3 1 2 1
orientation is when the long and short dimensions are parallel to the major and minor
preexcavation principal stresses.
Although numerical analysis shows that some yielding at the cavern walls is expected,
arching the walls into a spherical dome will further reduce the stress concentration. Spheroidal
shapes, as illustrated in Figure 3.5, may be advantageous from a rock mechanics point of view.
94 |
Colorado School of Mines | Through numerical modeling of these three traditional mining sequences, it is possible to
conclude that advancing parallel to the least preexcavation principal stress to reach the final
cavern dimensions is the most favorable method (Sequence 3). Utilizing Sequence 3 is
problematic, however, in BHM. The least favorable is advancing parallel to the greatest
preexcavation principal stress even though the working face is normal to the direction of advance
where the stress is smallest (Sequence 1). Yielding does not occur in the most favorable
sequence, but was presented surfaces normal to the advance direction as shown in the other two
sequences [64].
Based on the previous discussions, Pariseau [64] offered cavern design guidelines that
have some fundamental bearing relative to cavity design by borehole mining. The summaries of
Pariseau’s guidelines are as followings:
(1) the most favorable orientation is with the long edge (length) of the cavern parallel to
the major preexcavation in-situ stress and the short edge parallel to the least
preexcavation in-situ stress,
(2) the most favorable proportions of edge lengths are in the same ratio as the
preexcavation minor and intermediate to major principal stresses (compression positive),
and
(3) the most favorable excavation sequence is where production is advanced parallel to
the minor principal stress over the corresponding full cavern cross-sectional area [64].
97 |
Colorado School of Mines | 3.7) Hydraulic Fracturing
Although hydraulic fracturing is not part of this research effort and is assumed to play
little to no role in the proposed excavation process, there are conditions during BHM that may
exhibit similar characteristics to the fracture mechanics found during hydraulic pressurization.
Hydraulic fracturing is the propagation of fractures in a rock layer as a result of fluid penetration
and pressurization in the rock mass. A hydraulic fracture is formed by pumping fluid into a
borehole at a rate sufficient to increase the stagnation pressure down-hole until it exceeds the
fracture gradient (pressure gradient) of the rock. The fracture gradient is defined as the pressure
change per unit of depth due to fluid density and it is usually measured as an absolute pressure
per unit length (pounds per square inch per foot or bars per meter). The rock cracks as a
consequence of pressurized fluid exploiting weaknesses within the rock exposed by the borehole.
These resulting fractures are further extended by a process called water wedge. During the
process, the total fluid volume within the hole increases as a consequence of fracture density
extension and dilution. Coupled with fluid loss in the formation, the fluid pressure within the
borehole normally declines unless additional fluid is pumped into the system. [66]
There are three stress fields in the hydraulic-fracturing process as identified by Haimson
and Fairhurst [67]. They are induced by tectonic stress, pressurization of the borehole, and fluid
flow from the borehole into the surrounding rock. Each of these stresses will be addressed
separately.
The minimum tangential stress (S ) in the wall of a circular opening due to tectonic
1
stresses is:
98 |
Colorado School of Mines | S = 3(S - Po) - (S – P ) (3.2)
1 h H 0
Where; S = minimum in-situ principal stress, S = maximum in-situ principal stress, and P =
h H 0
preexisting pore pressure in the rock formation.
The minimum tangential stress in the wall of a borehole caused by fluid pressurization is
S = -P + 2P (3.3)
2 i 0
Where; P = borehole injection-fluid pressure.
i
The total tangential stress due to fluid flow at the borehole wall was obtained using an
analogy between the poroelastic and thermoelastic behaviors of material using Biot's constitutive
equation:
12
S = (P P ) (3.4)
3 1 i 0
Where; = 1 - (C/C ) = Biot's poroelastic parameter, C = compressibility of rock solid, C =
r b r b
compressibility of rock bulk, and v = Poisson's ratio
The total minimum tangential stress in the borehole is obtained by superimposing the
above three stress fields as the following:
S= S + S + S
t 1 2 3
12
S = 3S - S -P (P P ) (3.5)
t h H i+ 1 i 0
99 |
Colorado School of Mines | Where the total minimum effective tangential stress at the borehole wall is:
12
σ = 3S - S -P (P P )- P (3.6)
t h H i+ 1 i 0 p
Where; σ = total minimum effective tangential stress at the borehole wall and P = pore pressure
t p
in the rock at the borehole wall, where P <P <P
0 p i
Fracturing initiates when the total minimum effective tangential stress at the borehole
wall reaches the rock's tensile strength, that is:
σ = -T
t
P = P
i c
Where; T = tensile strength of rock in hydraulic fracturing and P = borehole fluid pressure at
c
breakdown (fracture initiation)
In Haimson and Fairhurst's derivation [67], Terzaghi's effective-stress law was applied at
failure:
P = P
p 0
Therefore, the poroelastic-breakdown equation of Haimson and Fairhurst is:
3S S T P
P h H 0 (3.7)
c 2
12
Where;
1
Based on these equations, it is possible to conclude that increasing the waterjet pressure
within an excavated cavity will increase the effective tangential stress at the cavity roof. Given
100 |
Colorado School of Mines | that this research effort will utilize much greater jet pressures than conventional borehole mining
systems, the resulting operating conditions could contribute to instability around the cavity. It is
also reasonable to conclude that more pore pressure at the cavity boundary (i.e. roof, floor and
rib) will decrease the effective tangential stress around the cavity.
3.8) Permeability Variation
Borehole mining of coal will likely cause disturbances of the surrounding rock mass by
increasing its permeability through a reduction of the stress, as well as the formation of new
fractures created during the excavation process. In addition, these conditions could also result in
the migration of methane gas contained in the disturbed rock into the low pressure cavity.
According to Palmer et al. [68], in naturally-fractured formations, such as coal, permeability is
sensitive to the changes in stress or pore pressure (i.e., changes in effective stress).
The following equation describes the importance of stress change in the permeability of
the coal:
ln (K) = K - 1/b (Effective stress) (3.8)
0
Where K= is the absolute permeability in a direction within the coal, K =is the permeability at
0
zero effective stress, and B= is the stress-permeability coefficient.
If a coal seam has a high permeability, any changes in permeability are unlikely to be
significant because the stress-permeability coefficient (b) is usually high. However, in the case of
low permeability coals, it is of extreme importance to be able to determine whether the
permeability will increase or decrease with degasification due to its implications regarding rock
stability. [69]
101 |
Colorado School of Mines | The consequence of changes in the formation pore pressure is significant. Induced pore
formation caused by excavation (i.e. borehole mining) will change the effective stress. As
previously discussed, effective stress variations will impact the permeability, and will increase
the gas flow as a function of increasing permeability.
According to Garber et al. [70], in low permeable porous formation, the excavation of
galleries leads to both mechanical and hydraulic disturbances in their surroundings (de-stressed
damage zone) and a significant reduction in pore water pressure. During and after excavation of
a gallery, fluid flow may occur, causing time-dependent changes in the redistribution of stresses
and displacement in the host rock.
Formation pore pressure plays an important role in the stability of a cavity. Changes in
the pore pressure lead to changes in the effective stresses. According to Settari et al. [71],
changes in pore pressure can also affect rock properties, such as porosity and permeability.
The effective stresses controlling failure are commonly neglected in the evaluation of an
underground opening Braun [62]. These effective stresses can also be influenced by changes in
the mechanical effectiveness of the existing pore pressure. The associate changes in the effective
stress will reduce the margin of safety. This phenomenon is more often observed in rocks with
low permeability and porosity.
3.9) Flow Mechanics
Gas flow mechanics and Darcy’s law governs the liberation of gas from crushed coal and
production faces into the excavation and working area. Based on Poiseuiille’s law, gas flow
depends on the permeability of the coal, internal gas pressure, distance from the face boundary,
and the coefficient of gas viscosity. Among other factors, the coefficient of coal strength
102 |
Colorado School of Mines | influences the permeability of the coal. The coefficient of coal strength depends on uniaxial
compressive strength of coal, width of the opening, depth of mining, and angle of seam
inclination. [51]
The following equation expresses the flow of gas by Poiseuiille’s law.
Qx= (-K /) + (dP/dx) (3.9)
x
Where K = permeability of the coal, P= Internal pressure of gas, x= Distance from the face
x
boundary, and = Coefficient of viscosity of the coal
Applying high pressure waterjets during the mining process may cause fractures and
variation of effective stress around the cavity. While the lengths of these fractures are likely to be
relatively small, this phenomena could translate to an increase in permeability which would in
potentially increase the rate of gas flow from the strata into the cavity
3.10) Subsidence
One of the key factors in any underground excavation, and especially in the case of cavity
failure during BHM, is surface subsidence. In this context, subsidence is the lowering of the
ground surface induced by the extraction of subsurface material. Besides the volume removed in
the extraction zone, there are many geological and mining parameters that may impact the
magnitude and extent of subsidence. Those parameters include extraction thickness (the vertical
height of the extracted material generally correlates to greater magnitude of subsidence), mining
depth, inclination of extraction horizon, degree of extraction (lowering the extraction ratios will
both reduce and delay the subsidence), the extent of the mined area, extraction rate (ground
103 |
Colorado School of Mines | surface subsidence follows the face as it advances, so a fairly rapid, constant advance rate should
be applied to minimize the effect of strain on surface), competence of surrounding materials, in-
situ stress state (high horizontal stresses may foster formation of an arch in the material above a
mined extracted void, thereby attenuating subsidence), geological discontinuities, near surface
geology and surface topography, hydrogeology, and elapsed time [24]. Subsidence is caused by
the change of equilibrium conditions after initiating the excavation of the cavity. Given the
objectives of this research and that only thin seam deposits are being considered, it is assumed
that the effect of subsidence is negligible. That said, a brief discussion of subsidence is warranted
due to the potential impacts should it occur.
Continuous subsidence will happen at the surface when thin, flat-dipping, tabular
orebodies are mined completely over large panels (Figure 3.7). If the fracture and discontinuous
movement of the rock are limited to the immediate vicinity of the cavity, then continuous
subsidence will occur. Caving-induced stress and depth of mining are two main factors that
control the movement of the fracture in and around the cavity. There are some standard
definitions for subsidence analysis. The vertical displacement of any point of the ground surface
is identified as subsidence (denoted as s). The maximum subsidence in a given profile is denoted
by S. For a specific area and width of extraction, known as the critical area and critical width, S
will take the maximum value possible for the particular seam, S . Areas or widths in which S<
max
S are described as subcritical. For supercritical areas or widths, the maximum subsidence is
max
reached over a finite width rather than at a single point that occurs in the critical case. The angle
of draw (ξ) is the angle made with the vertical by a line drawn from the base of the seam to the
point of zero surface subsidence. It depends on the mechanical properties of the rocks, being
lower for stronger rocks and higher for weaker rocks and soils. [72]
104 |
Colorado School of Mines | Among the exponential, hyperbolic, trigonometric, and error functions, the hyperbolic
appears to give more accurate results.
S(x) = 0.5S [1-tanh (bx/h)] (3.12)
max
Where; x is the horizontal distance from the point of inflection in the direction of decreasing
subsidence, h is the depth of the seam, and b is a constant controlling the slope at inflection
point. For UK mines, a value of b=5 is traditionally applied. [72]
The subsidence that occurs during the extraction process normally happens within a
critical area called active subsidence. Time-dependent subsidence due to consolidation or visco-
plastic behavior of the strata, which continues to develop after the point is no longer in the zone
of influence of the face, is called residual subsidence. In the case of multiple boreholes, rib
pillars are left between cavities that serve to limit the maximum subsidence. If the panels
(cavities) are narrow (with to mining height< 0.33) and the pillars are large (width to mining
height>0.2), the main strata will bridge across the pillars, where the subsidence over both the
pillars and panel centers will be minimal. [72]
Peng [45] categorizes the movement of the overburden zones based on their severity to
caved zone, fracture zone, deformation zone, and soil zone. The caved zone is the immediate
roof that falls into the cavity and fills it up. The fracture zone is immediately above the caved
zone. According to Peng, the important parameters that affect the height of the caved and
fracture zones are mining height, mining method, roof control method, rock properties, and
period of mining [45].
Peng [21] classifies the methods for predicting the subsidence into six categories:
theoretical, profile function, influence function, graphical, physical, and numerical modeling. He
106 |
Colorado School of Mines | introduces several equations for different cases. For finding the maximum possible subsidence
when the gob reaches the critical size, he has suggested the following equations:
S = aH cosα (3.13)
max
Where; a is the subsidence factor, H is the mining height, and α is the seam inclination.
Since only flat seams will be investigated in this dissertation, it is assumed that α=0.
The value of a is related to rock properties and, for many conditions, can be approximate by:
a= 0.35+ 0.5p (3.14)
Where p is the coefficient of combined strata properties and is obtained by the following:
N
(Qh )2
i i
P= i1 (3.15)
N
(h )2
i
i0
Where; N is the number of stratum layers in the overburden, Q is the rock factor of the ith layer
i
of the overburden strata, and h is the thickness of the ith layer of the overburden [21].
i
3.11) Conclusions
By studying the theoretical aspects of some confounding parameters, the following
conclusions were obtained. These conclusions will provide guidance in the development of a
borehole mining design protocol.
Increasing the waterjet pressure will increase the effective tangential stress at the cavity
roof.
107 |
Colorado School of Mines | The horizontal and vertical stress fields can cause major stability issues if they are not
controlled through adequate barrier pillar design (intact rock left between the cavities),
cavity size, and cavity orientation.
An optimal k value for highest safety factor is between 0.5 and 1.0, with lower or higher
ratios producing lower safety factors. Areas that have extremely low or high k values can
potentially benefit from orienting their cavities orientation appropriately to optimize their
roof stability.
Cavities are most stable only in the direction of σ , therefore the most stable condition
hmax
is to have cavities parallel to maximum horizontal stress. When underground openings
are parallel to the in-situ principal stress, there are no shear stresses in that direction.
There are three major problems to address in considering large, deep caverns
underground: cavern orientation, cavern shape (or dimension), and excavation sequence.
The most favorable orientation is with the long edge (length) of the cavern parallel to the
major preexcavation stress and the short edge parallel to the least pre-excavation stress.
The most favorable proportions of edge lengths are in the same ratio as the preexcavation
minor and intermediate to major principal stresses (compression positive). The most
favorable excavation sequence is to advance parallel to the minor principal stress over the
corresponding full cavity cross-sectional area.
108 |
Colorado School of Mines | CHAPTER 4
CASE STUDY AND NUMERICAL MODELING VERIFICATION
4.1) Introduction
The main objective of this chapter was to verify if a widely employed 2-dimensional
modeling software package (Flac2D) had sufficient accuracy to perform a stress analysis within
a borehole mining system as part of a predictive design protocol. To achieve this objective, data
derived from an empiric field study was modeled using Flac2D, and the results were compared to
those obtained from a case study that used a three dimensional model (Flac3D). This analysis
was deemed important because of the inherent advantages of using a less sophisticated numerical
tool to establish the design protocols.
Flac 2D (Fast Lagrangian Analysis of Continua) is a two-dimensional continuum code
developed for modelling soil, rock and structural behavior. FLAC is a general analysis and
design tool for geotechnical, civil, and mining engineers and can be applied to a broad range of
engineering problems. The explicit finite difference formulation of the code makes FLAC ideally
suited for modelling multi-stage geomechanical problems. The formulation can accommodate
large displacements and strains and non-linear material behavior, even if yield or failure occurs
over a large area or if total collapse occurs. [85]
FLAC accommodates a number of complex behaviors not readily suited to finite element
codes, including problems that consist of several stages, large displacements and strains, and
non-linear material behavior and unstable systems (even cases of yield/failure over large areas,
110 |
Colorado School of Mines | or total collapse). The Lagrangian formulation enables the grid to move and deform with the
material it represents since the incremental displacements are added to the coordinates. The
reported success of the use of the code and acceptance by the geotechnical engineering
community were also factors in the decision to use the software. [85]
It was observed that the results obtained by Flac2D (i.e. axisymmetry feature) were in
good agreement with those generated from Flac3D and utilizing Flac2D to simulate borehole
mining stress analysis will give a relatively accurate results. Some modifications were applied to
the model in order to study the impact of additional parameters, like cavity shape and internal
pressure. It was also concluded that the internal pressure and cavity shape effects on the induced
principal stresses was significant and needs more detailed study. The case study data used in this
chapter was obtained from a subsidence study prepared by Barr Engineering Company for
Cooperative Mineral Resources (CMR) [73]. CMR sought to collect samples from two enriched
manganese zones within an oxidized iron-formation at a site located near Emily, Minnesota.
Through bulk sampling, CMR intended to evaluate the potential of using a small scale Borehole
Mining (BHM) system that employed waterjet technology. CMR believed that the enriched
manganese zones in the resource were suitable to BHM excavation methods and as part of the
technical feasibility of the project; a subsidence study was performed by Barr Engineering. One
of the primary goals of this study was to determine if subsidence would occur under specific
operating conditions and, if so, to develop a range of potential depth and the radial extents of the
surface damage at the Project Area.
111 |
Colorado School of Mines | 4.2) Case Study
As described, the case study project area is located in the Emily District of the “Cuyuna
Iron-Range” in north-central Minnesota. The iron-formation, as measured near the collection
boreholes, has a thickness averaging approximately 249 ft (76 m). Two intervals within the iron
formation are enriched in manganese, with zones located at approximately 216 to 246 ft (66 to
75 m) and 321 to 446 ft (98 to 136 m) from the surface. Glacial material of sand-and-gravel
outwash overlies the manganese-bearing iron-formation at thicknesses ranging from 47 to 62 m
across the site. Surficial peat deposits are present in some locations between Anna Lake and Ruth
Lake, but the deposits in this area are predominantly highly permeable sand and gravel with little
to no silt and clay. Figure 4.1 illustrates the uppermost bedrock units in the vicinity of the bulk
sample project location. Figure 4.2 shows the detailed cross-sectional stratigraphy of iron
formation. [73]
Figure 4.1. Uppermost bedrock units in the vicinity of the bulk sample project location. [73]
112 |
Colorado School of Mines | typically contained gravel or larger rock pieces. The results of soils tests on the iron-formation
indicate that the soil matrix within the formation is weak. However, it does appear that where the
material is more soil-like or has experienced greater fracturing, the iron-formation will not be
able to maintain a cavity (e.g., maintain walls or a ceiling stability) for a significant amount of
time before failure. The glacial material is poorly-graded sand with small amounts of gravel. The
majority of the glacial outwash formation is below the water table, where material is likely to
flow toward a cavity in the presence of water.
The bulk sample collected from the cavity was modeled three-dimensionally using
Itasca’s FLAC3D. This software is a finite difference program that can model staged
excavations, large displacements and strains (mesh deformation), and unstable systems
(collapse). The FLAC3D computer code was used to simulate the response of the ground surface
to caving and deformation resulting from BHM. Theoretical mathematical models, empirical
data from mining-induced subsidence projects, and engineering judgment from past BHM
operations were used to evaluate the reliability of the modeling results. An isometric view of the
FLAC3D model’s numerical grid and parameterization are shown on Figure 4.3. Assuming that
the cavity is roughly ellipsoidal in shape, only one quadrant of the cavity needs to be modeled
due to symmetry. The model is comprised of a 200 ft (60 m) thick layer of glacial outwash
overburden underlain by the 200 ft (60 m) thick iron-formation and a 100 ft (30 m) layer of clay.
The maximum allowable bulk sample volume is 300,000 cubic feet (8495 cubic meters), where
two bulk sample sizes and locations were modeled. The lower enriched manganese zone is
approximately 115 feet (35 m) thick ranging from 295 to 410 ft (90-125 m) below surface. For
modeling purposes, the initial bulk sample (cavity) dimensions were approximately 100 ft (30 m)
114 |
Colorado School of Mines | model more conductive for evaluating three dimensions, the axisymmetry feature of Flac was
applied. Since Flac is a continuum modeling approach, it cannot accurately simulate the
disaggregation and volume expansion that will happen during borehole mining process. While
the model cannot simulate the actual mechanism of caving during the bulk mining, it can offer a
conservative estimate of angle of repose and horizontal extension of the subsidence at the ground
surface.
4.3.1) Model Parameters
A cavity induced by BHM is theoretically assumed to be ellipsoidal in shape, with
maximum dimensions of approximately 108 ft (33 m) in height and 105 ft (32 m) in diameter.
These dimensions were chosen to replicate the geometry of the models used in the case study.
The cavity extends upward and outward during the excavation (bulk sampling). While the
idealized shape of the cavity is approximately elliptical, a cylindrical cavity was easier to model.
The model boundary extends horizontally 492 ft (150 m) from the center of the collection
borehole in order to have sufficient distance to minimize the effect of the numerical boundary
conditions for the maximum excavation radius of 52 ft (16 m). Two additional models with
boundary horizontal extensions of 328 ft (100 m) and 590 ft (180 m) were tested and showed
similar results (Figure 4.5). The model accounts for 207 ft (63 m) of sandy glacial overburden,
6.6 ft (2 m) of a caprock underlaid by a 216 ft (66 m) thick layer of iron-formation, and 66 ft (20
m) clay at the bottom of the model. The caprock layer was added between the sandy glacial
outwash and the top of the soft iron-formation to represent the apparently discontinuous presence
of the hard rock. Material inputs were the same as those in Tables 4.1 and 4.2.
117 |
Colorado School of Mines | 35
30
la
c
it
r 25
e
v
g
Cylinder
n
it Spherical
s 20
ix
e
e
r
p
/
ss 15
ss
ee
rr
tt
ss
la
10
p
ic
n
i r
p
m 5
u
m
ix
a 0
M
0 5 10 15 20 25
Internal pressure/Cohesion
Figure 4.11. Influence of excavation shape and ratio internal pressure/cohesion on maximum
principal stresses at the cavity roof.
4.3.4) Third model setup and results
In this analysis phase, a cavity with a thickness of 1 m and radius of 5 m was modeled.
The modeling procedure utilized the axisymmetry feature, and the model inputs were the same as
those shown on Tables 4.1 and 4.2. Figure 4.12 shows the maximum principal stress in the model
at approximately 3 Mpa located at the side of the cavity. The effect of internal pressure on the
cavity walls was also studied. A cavity with 1 m thickness and radius of 5 m was modeled using
the axisymmetry feature. The modeling procedure was the same as the model setup illustrated in
Figure 4.5, where the depth of the cavity is 85 m. Internal pressures within the cavity of 5 and 30
Mpa were applied. Figures 4.13 and 4.14 illustrate the induced principal stress contours.
124 |
Colorado School of Mines | JOB TITLE : principal stress by applying internal pressure of 30 MPa (*10^2)
1.700
FLAC (Version 6.00)
LEGEND 1.500
25-Mar-13 21:19
1.300
step 37232
-5.000E+01 <x< 1.500E+02
-2.500E+01 <y< 1.750E+02
1.100
Maximum principal stress
-3.00E+07
-2.50E+07 0.900
-2.00E+07
-1.50E+07
-1.00E+07 0.700
-5.00E+06
0.00E+00
0.500
Contour interval= 5.00E+06
Extrap. by averaging
0.300
0.100
-0.100
-0.300 -0.100 0.100 0.300 0.500 0.700 0.900 1.100 1.300
(*10^2)
Figure 4.14. Maximum principal stress in the case of applying internal pressure of 30 Mpa.
By comparing the results shown in these figures with Figure 4.12, it was possible to
conclude that applying internal pressure on the cavity has significant impact on induced principal
stress around the cavity that can result in fractures initiation. As these two figures show, the
maximum principal stress is located at the top of the cavity.
Figures 4.15 and 4.16 show the plasticity indication status of the models by applying
internal pressures of 5 and 30 MPa, respectively. Applying 5 MPa internal pressure results in a
yield in tension on the sides of the cavity. However, applying 30 MPa internal pressure causes
shear failure at the cavity roof and floor. In these applications it is prudent to conclude that
applying higher pressure increases the risk of shear failure as compared to lower pressures. Shear
failure is not as simple as the tensile and compressive failure criteria, because it involves both
normal and shear stresses. The Coulomb criterion shows that shear stress tends to cause failure
126 |
Colorado School of Mines | Figure 4.17 shows the model geometry, where the objective was to observe if modeling a
system with a borehole and cavity will give relatively consistent and repeatable results. A cavity
with the same size as the one in Figure 4.12 was modeled in a 2-dimensional environment, where
the radius and thickness of the cavity were assumed to be 5 m and 1 m, respectively. The cavity
is connected to the surface by a borehole. The material inputs were held constant with those in
Tables 4.2 and 4.3. The axisymmetry feature is suitable to simulate the final cavity shape. It was
not used in this case. However, in this research, the effects of some dynamic parameters like
internal pressure are believed to be an important factor and will be discussed in the next chapter.
Figure 4.18 shows the maximum principal stress. The magnitudes (maximum of 3.5
MPa) suggest that the results are in relatively good agreement with Figure 4.12
(axisymmetry feature). As Figure 4.18 shows, the maximum induced principal stresses are in the
corners and sides of the cavity.
Figure 4.17 Model set up
128 |
Colorado School of Mines | CHAPTER 5
NUMERICAL RESULTS AND DESIGN PROTOCOLS
5.1) Introduction
In an effort to address one of the prevailing technical challenges impeding the
commercial applications of borehole mining in coal, the ultimate purpose of this research is to
establish a set of design protocols (guidelines) for estimating optimum cavity geometry and
orientation for several critical excavation and geological factors. The development of this design
methodology will contribute to a larger research effort which focuses on evaluating the economic
and technical development feasibility of in-situ borehole mining relative to thin seam deposits.
Towards this objective, a parametric study to investigate the effect of several parameters like
internal pressure, cavity size, and in-situ stress, on the stability has also been performed.
Based upon the positive results of the case study, a finite difference method (Flac2D) and
dimensional analysis have been applied to study a cavity developed through hydro-excavation of
a mineral resource during borehole mining. The structural impact of differential internal fluid
pressures on the cavity periphery has also been numerically modeled. The research was focused
on delineating the impact of critical factors associated with maintaining cavity stability during
the mining process and how it pertains to the development of a design protocol. A key element of
this goal was to observe the effect of cavity dimension, internal pressure (cavity pressurization),
K (horizontal stress/vertical stress), vacuum (negative pressure), depth and method of extraction
on the induced maximum principal stress. The largest principal stress is considered one of the
key parameters responsible for the initiation of fractures in rock. The maximum normal stress
criterion, also known as Coulomb’s criterion, is based on the Maximum normal stress theory.
According to this theory, failure occurs when the maximum principal stress reaches the ultimate
131 |
Colorado School of Mines | strength of the material for simple tension. Furthermore, the theory suggests that the largest
principal stress will initiate cracks in the rock and lead to failure of the rock mass.
In Section 5.2, cavities of different dimensions, under various internal pressures, have
been modeled. Different internal pressures represent different cavity pressurization values. The
effect of cavity dimension, internal pressure (extraction pressure), vacuum (negative pressure),
and depth on the induced maximum principal stress were studied. The influence of K (horizontal
stress/vertical stress) has been studied as well. The process and results of this analysis are
described in Section 5.2. In Section 5.3, the effect of two different mining sequences to reach the
final cavity geometry was studied. In the first case, several horizontal slices were used in order to
reach the ultimate cavity size. In the second method of extraction, the excavation process was
comprised of a series of vertical slices. The impacts of these two extraction methods on the
induced maximum principal stress were studied. Since the numerical models assumed that the
cavity was under pressurization, it is challenging to observe the failure and hence, the induced
maximum principal stress values were recorded as an indication of potential failure. In Section
5.4, the same numerical process used in Section 5.2 was replicated for a material with different
physical properties. A dimensional analysis was subsequently performed. In Section 5.5, the
effect of ultimate shape of a cavity on displacement was studied. Section 5.6 includes the
summary of the results achieved in Sections 5.2 to 5.5, design protocols are presented in Section
5.7. These design protocols were obtained based on the results of the numerical modeling in
Sections 5.2 to 5.5 and the conclusions reached in Chapter 2 and 3 as derived from the literature
and previous research.
132 |
Colorado School of Mines | 5.2) Effect of Cavity Dimension and Internal Pressure
This section describes the process of performing a stress analysis of a cavity created by
BHM, and includes consideration of fluid pressurization of the cavity periphery. To accomplish
this analysis, a finite difference method (Flac2D) was used. Figures 5.1 and 5.2 illustrate the
geometry of the models. As Figure 5.1 shows, depth of cover, cavity length (radius), and cavity
height are represented as H, L and D, respectively. To simplify the analysis, each side of the
cavity was modeled independently and the application of internal fluid pressure was applied to
only one side of the excavation. Figure 5.2 shows how internal pressure is applied inside the
cavity. Table 5.1 presents the mechanical rock properties that were applied in the models. These
properties belong to shale provided within the Flac2D database. For this study, the rock was
assumed to be a homogeneous, isotropic material with no internal structure (e.g., bedding or
jointing). The vertical in-situ stress was assumed to be gravitational, and equal to the unit weight
multiplied by depth. Depth stress gradients are assumed to be linear, with zero stress at the
ground surface. Gravity was specified in the analysis as 9.81 m/s2.
Figure 5.1 A view of model geometry. (Akbarzadeh & Miller [81])
133 |
Colorado School of Mines | the initial pore pressure distribution for this case. The geometry of the model was 60 × 120 m
which was made of 30 × 60 grids.
The modeling sequence consists of the following stages:
Stage I: Establish equilibrium conditions to initialize stresses.
Stage II: Excavation of the borehole and initiating a cavity in horizontal direction.
Stage III: Add internal pressure to the cavity roof, floor and side-walls, and cycle to
equilibrium.
Stage IV: Alteration of different cavity sizes, pressures and depth was
modeled.
The bottom of the mesh was fixed in the Y direction and the sides of the mesh were fixed
in x directions to prevent system instability. In addition, since the borehole will likely be cased in
most field applications, the walls of the borehole were also fixed.
Figure 5.3(a) – 5.3(d) show sequences of the model set-up and excavation process.
The height of the cavity was considered as one grid equal to 2 m. Four different values were
considered for cavity length ranging from 2 to 20 m (i.e., 2, 6, 10 and 20 m). The internal
pressure values were selected as 1.5, 7.5, 22.5 and 75 MPa. The reason for choosing this range
was to observe the sensitivity of different internal pressures on the maximum principal stress. In
this analysis, two different depths (i.e., 30 and 40 m) and the ratio of horizontal to vertical
effective stress (1.0) were also assumed. Figures 5.4 and 5.5 illustrate the maximum principal
stress around the cavity roof for different internal pressures, cavity length, and depth of cover.
135 |
Colorado School of Mines | 140
l 120
a
p
i c)
a
100
nP80 L=2m
i r PM
(60
L=6m
ms L=10m
s40
u
me L=20m
r20
t
iS
x
0
a
M
0 20 40 60 80
Internal Pressure (MPa)
Figure 5.5 Maximum principal stresses versus internal pressure for different cavity length for
depth of cover of 40 m. (Akbarzadeh & Miller [81])
As Figures 5.4 and 5.5 illustrate, increasing internal pressure and cavity length (radius) will
significantly increase the induced maximum principal stress. Depth of cover plays an important
role, but is not as significant as internal pressure and cavity size. As shown in the figures,
situations involving low applied internal pressures less than 5 MPa will generate similar
principal stresses for a variety of cavity sizes at a shallow depth. As the figures illustrate, cavities
with smaller dimensions are subject to fewer impacts derived from internal pressures as
compared to larger cavities.
Figures 5.6 to 5.10 show the effective principal stress before excavation (after stress is in
equilibrium) and borehole excavation (before applying pressure) for a variety of cavity lengths
(radii) (2 m, 10 m and 20 m). The internal pressure applied inside the cavity was 7.5 MPa. The
rationale was that 7.5 MPa will give a more meaningful result as compared to 1.5, 22.5 or 75
MPa because an internal pressure of 1.5 MPa does not have a significant effect on and that
internal pressures of 22.5 and 75 MPa cause some failure. Depth of cover was held constant at 30
137 |
Colorado School of Mines | In this analysis, pore pressures were generated as a result of mechanical volume change
during the borehole and cavity excavation. Figures 5.11 and 5.12 show the effective principal
stress distribution as a consequence of changing the water bulk modulus to 0.9 GPa. Varying the
water bulk modulus will result in changes in the pore pressure. Bulk modulus is sometimes
referred to as the incompressibility, and is a measure of the ability of a substance to withstand
changes in volume under compression on all sides. It is equal to the quotient of the applied
pressure divided by the relative deformation (change in volume divided by the original volume).
The cavity lengths (radii) in Figures 5.11 and 5.12 are 10 and 20 m, respectively. The internal
pressure for both cases was held at 7.5 MPa. Comparing Figures 5.9 and 5.10 with Figures 5.11
and 5.12 indicates that a change in pore pressure will influence the effective stress distribution,
and the extent of the tensile stress region. However, this influence appears to be relatively
insignificant.
Figure 5.11 Effective principal stress for cavity length of 10 m and internal pressure of 7.5 MPa,
and water bulk modulus of 0.9 GPa.
141 |
Colorado School of Mines | Figure 5.12 Effective principal stress for cavity length of 20 m and internal pressure of 7.5 MPa,
and water bulk modulus of 0.9 GPa.
In order to facilitate removal of the extracted material, the cavity will undergo an
intermittent condition of negative pressure during the extraction process. This condition is
modeled by applying negative pressure at the roof, floor, and walls of the cavity, where it was
assumed that cavity length (radius) was 10 m and the depth of cover equaled 30 m. All the other
properties were held constant as those outlined in Table 5.1. It was assumed that the cavity will
undergo cyclic pressurization (positive to negative and back to positive) beginning with a
positive internal pressure of 7.5 MPa and then dropping to a maximum negative value of 0.4
MPa. Figures 5.13 and 5.14 show the maximum principal stress contours for these assumptions.
According to Figure 5.14, applying negative pressure will cause stress concentrations to move
from the roof to the side-wall and corner of the cavity.
142 |
Colorado School of Mines | stress around the cavity (it can be the consequence of shallow depths and the internal pressure
being higher than horizontal and vertical stresses); the difference between the maximum and
minimum principal stresses changed significantly around the cavity periphery. Figure 5.15
illustrates the principal stress difference (σ - σ ) for different values of K. Two different cases
1 3
with internal pressure of 7.5 and 20 MPa were modeled, where cavity width was held constant as
10 m. The modeling procedure is same as those described in Chapter 5.2 and the material
properties were held constant to those introduced in Table 5.1. As Figure 5.15 shows, increasing
K will decrease (σ - σ ) around the cavity roof corner. This implies that increasing the horizontal
1 3
stress will result in a decrease of the induced shear stress at the cavity roof corners. According to
Figure 5.15, increasing the internal pressure inside the cavity will translate to an increase in the
principal stress difference (σ - σ ).
1 3
e 7
c
n
6
e
r
e
5
f
f
i
d
s)
a
4 P=7.5 MPa
s eP 3 P=20 MPa
M
r
t s( 2
l
a
p 1
i
c
n 0
i
r
P 0 1 2 3 4 5
K
Figure 5.15 Principal stresses difference (σ - σ ) versus different values of K.
1 3
(Akbarzadeh & Miller [81])
Cutter roof is a type of massive caving phenomenon and represents a failure that initiates
at cavity corner and extends approximately vertically. This can cause massive failure with little
144 |
Colorado School of Mines | warning. The rational for failure is that a cutter roof is caused by the shear stress at the cavity
corners exceeding the shear strength of the rock formation [86]. As such, the induced shear
stresses were measured at the cavity roof corner instead of cavity’s roof midpoint during
numerical modelling.
In order to observe the impact of depth in greater detail, a cavity with a length (radius) of
6 m was modeled for depths of 10, 20, 30 and 40 m. Three different internal fluid pressures of
25, 35 and 45 MPa were also applied to the analysis. As Figure 5.16 illustrates, the effects of
internal pressure are more significant than that of depth. This is in agreement with the results
obtained in Figures 5.4 and 5.5.
60
l
a
p 50
i )
c a
n P 40
i r M P=25MPa
P
(
30 P=35MPa
m s
s
u m e r 20 P=45MPa
t
i S
x
10
a
M
0
0 10 20 30 40 50
Depth of cover (m)
Figure 5.16. Maximum principal stresses versus depth of cover for different internal pressure for
cavity length of 6m. (Akbarzadeh & Miller [81])
5.3) Two Different Strategies for Extraction
The effects of different extraction strategies were also studied. The reason for this was to
observe if the different directions of the extraction would impact the stability around the ultimate
cavity shape. Two approaches were modeled by an axisymmetric feature within the software (it
145 |
Colorado School of Mines | The impacts of these two excavation approaches on the induced maximum principal stress
were studied. The record point for the maximum principal stress was located at the corner of the
cavity. Figures 5.19 and 5.20 illustrate the maximum principal stress as a function of K
(horizontal stress/vertical stress) for a depth of 100 and 1000 m, respectively. As Figure 5.19
shows, the maximum principal stress slightly increased in Case 2 (vertical slices). Thus, it was
possible to conclude that using the first excavation strategy (horizontal slices) would likely result
in a more stable excavation. In both extraction methods, increasing K after reaching a value of 1
will increase the induced maximum principal stress. This is an indication of possible stability
issue in the higher horizontal stress environment.
As Figure 5.20 illustrates, the induced principal stress at a depth of 1000 m is identical
between the two methods of extraction for K values more than 1. Increasing the K value will
increase the induced principal stress drastically. It can be concluded, that when horizontal stress
is significantly higher than internal pressure of 30 MPa, extraction sequence will not play an
important role in the stability of the cavity.
Figure 5.19 Maximum principal stress around the cavity as a function of K for depth of 100 m.
(Akbarzadeh & Miller [81])
147 |
Colorado School of Mines | The selected objective variable in this dimensional analysis is the largest principal stress in
the cavity roof, σ induced by applying internal pressure and extending the cavity. All the factors
1,
identified will affect the largest principal stress in varying degrees of magnitude. These factors
can be reduced by using the Buckingham П theorem. In this study, the values for ρ, ρ , B, B , S,
w w
C, σ φ , D, R and σ were kept constant. Therefore, the new dimensional function will be:
t, D h
σ σ = П (L/D, P/C, H/R, h )
1/ t
gH
Value of h (cited as K) was kept as one in this dimensional analysis.
gH
The numerical processes employed were the same as those used in Section 5.2. The material
properties inputs are presented in Table 5.2. These properties belong to a sandstone provided
within the Flac2D database. The reason for choosing a material with different properties was to
ensure that the results in Section 5.2 would be consistent for a variety of different materials. The
water properties were kept the same as those presented in Section 5.2.
Table 5.2 Input mechanical properties
Density 2700 kg/m3
Rock bulk modulus 4.93 GPa
Rock shear modulus 2.96 GPa
Friction angle (φ) 32˚
Cohesion (c) 3.7 MPa
Tensile strength (σ) 5 Mpa
t
Dilation angle 0.01˚
Figures 5.21 and 5.22 show the normalized maximum principal stress versus normalized
internal pressure for different normalized cavity geometries. Figures 5.21 and 5.22 utilize an H/R
of 15 and 20, respectively. As these figures indicate, the dimensionless principal stress will
149 |
Colorado School of Mines | 5.5) Stability of Different Cavity Ultimate Shapes
The ultimate cavity shape plays an important role in the stability of a borehole mining
design, where it will likely impact the stress distribution around the cavity. The most common
final cavity shape induced by borehole mining is cylindrical. In this section, a cylindrical volume
with three different front view sections was modeled. The purpose was to determine the most
suitable final cavity design for borehole mining projects. Figure 5.23 illustrates the three
different geometries that were modeled. Each cavity configuration had its own geometry with
different side shapes. An attempt was made to keep the cross sectional areas of the cavities as
close as possible so the areas were approximately the same. The Flac axisymmetry feature was
applied. The model input was from Section No.4103 Xinliu, an underground in central China
[82]. Table 5.3 presents the properties of the coal and main roof in this study. The depth of cover
was 240 m. The cavities were assumed to be subjected to in-situ stresses varying linearly with
depth. Unit weight of the overburden (sandstone) was assumed to be 0.026 MN/m3 and K was
equal to 1. The simulated cavities had an effective ratio of ½ radius/height, which translated to
approximately 4 m and 2 m radius and height, respectively. Figures 5.24, 5.25 and 5.26 illustrate
the geometric layouts of the three different cavity shapes in Flac.
Table 5.3 Input mechanical properties. [82]
Properies Shale (roof) Coal Limestone (main
roof)
Density 2106 kg/m3 1450 kg/m3 2612 kg/m3
Rock bulk modulus 12.2 GPa 13.8 GPa 18.1 GPa
Rock shear modulus 7 GPa 3 GPa 11.1 GPa
Friction angle (φ) 36˚ 30˚ 43˚
Cohesion (c) 7.25 MPa 2.7 MPa 11.46 MPa
Tensile strength (σ) 3.16 Mpa 1.33 Mpa 5.45 Mpa
t
Dilation angle 6 ˚ 0 13 ˚
151 |
Colorado School of Mines | Figure 5.26. Shape 3 in Flac model.
Figures 5.27 illustrates the maximum total displacement for Shapes 1, 2 and 3, at depths
of 100, 240, 500 and 1000 m. In all of the cases, the cavity roof near the axis of symmetry was
the most disturbed zone. According to Figure 5.27, although displacement values are close, it
suggested that a cavity with a final shape similar to Shape 2 was the most stable. As such, it can
be concluded that leaving the side walls of an excavated cavity in an arch shape will increase the
stability of the opening.
According to Boughrarou et.al. [80], rectangular shapes result in larger stress
concentration around the right-angled corners. By having a smooth geometry, the stresses are
evenly distributed around the opening. In hard massive rock, circular or oval shapes are generally
prefered. In stratified or jointed rock, choosing the final cavity geometry is complicated by the
presence of discontinuities. As one would expect, the selection of the cavity’s selected ultimate
shape should generates the smallest number of unstable rock blocks.
153 |
Colorado School of Mines | 16
t
n
e 14
m
e 12
c
a 10 Shape 1
l
p)
sm
8 Shape 2
i Dm
m( 6 Shape 3
u
4
m
i x 2
a
M
0
0 500 1000 1500
Depth of Cover (m)
Figure 5.27 Maximum displacement for different shapes.
Two cavities with different radii at the depth of 150 and 300 m were modeled by using
flac axisymmetry feature. The cavity height was assumed as 2 m. Table 5.4 presents the
properties of the coal and main roof in this study. The cavities were considered without any
internal pressure. The cavities were assumed to be subjected to in-situ stresses varying linearly
with depth. Unit weight of the overburden (sandstone) was assumed to be 0.023 MN/m3 and K
was equal to 1. Figure 5.28 shows the induced maximum principal stress at the roof of cavities
with different radii at the depth of 150 and 300 m.
As the Figure 5.28 illustrates, the maximum principal stress will increase as the depth of
cover and cavity radius increase. The maximum obtained stable cavity spans at the depths of 150
and 300 m were 8 and 6 m, respectively. After reaching these limits, shear failure was observed
at the cavity rib.
154 |
Colorado School of Mines | Depth of cover plays an important role, but not as significant as internal pressure and
cavity size in the case of shallow depths.
The influence of changing pore pressure on the distribution of effective stress and the
extent of the tensile stress region are minor.
Applying negative pressure will cause the stress concentration to move from the cavity
roof to the side-wall and corner of the cavity.
Increasing the horizontal stress will decrease the induced shear stress at the cavity roof
corner.
Increasing internal pressure inside the cavity will increase the induced shear stress at the
cavity roof corner.
Reaching the final cavity geometry through excavation processes that are comprised of
horizontal slices (cuts) is more stable than using vertical slices excavation
methods/techniques.
In a cavity with small dimensions, the impact of internal pressures is less than those with
larger geometries.
In a borehole mine design, leaving the side walls in an arched shape will increase the probability
for stability.
During a comprehensive literature search, no research was found that predict cavity stability
based upon geomechanical characteristics associated with borehole mining. In the future, these
models shall be verified and refined using different software, where the results will undergo a
156 |
Colorado School of Mines | comparative analysis. Based on the numeric results, cavity pressurization of maximum 30 MPa
at the depth of 40 m, will give an estimated maximum stable cavity radius of 10 m.
5.7) Design Protocol
Several models were analyzed using Flac2D to perform a parametric study of factors that
could potentially impact the borehole mining process, whereas information derived from the
literature was useful in identifying several parameters that could possibly affect cavity design
and the mining process. The results and observations of these studies (numerical modeling and
literature search) led to the following proposed protocol design for the creation of stable cavities
during borehole mining:
While mitigating premature roof collapse is the primary motivation behind this research,
it is also desirable for large cavities to collapse (fail) after the completion of mining so
that adjacent cavities and pillars (segment between cavities) will be distressed. In some
cases, coal seams lie in soft strata, such as claystone. These seams are usually
characterized by geologically younger coking-coal deposits. Rapid extraction of the
cavity may prevent development of accelerated deformation and collapse of cavity roof
before coal has been extracted. The following are the general precautions that need to be
considered regarding the immediate roof of a cavity as derived from literature search:
a) Conglomerate: highest strength in the series of rock types that can be associated
with a coal bearing strata,
b) Sandstone: often resistant to caving after coal extraction,
c) Siltstone: will usually cave after coal extraction,
157 |
Colorado School of Mines | d) Shales and mudstones: susceptible to moisture, and may cave before coal
extraction is completed,
e) Clay rocks: often show low strength and higher moisture content with very low
bearing capacity,
f) Limestone: usually of moderate strength and good caving properties,
g) Dolostone: similar properties to limestone,
h) Shaly limestone: characteristically possess lower strength. Silty limestone is
similar to shaly limestone and possesses lower strength.
According to literature survey, if the immediate roof is thick and consists of strong sandy
shale or sandstone, conglomerate and limestone, it can be left unsupported for extended period of
time (up to 8 hours).
Increasing internal pressure and cavity length will significantly increase the induced
maximum principal stress around the cavity. Increasing internal pressure inside the cavity
will also increase the induced shear stress at the cavity roof corner. In cavities with
smaller dimensions, the impact of internal pressures is less than those with larger
geometries. Depth of cover plays an important role, but in the case of shallow depths
(below 40 m) it is not as significant as internal pressure and cavity size.
a) During modelling, no plastic or shear yielding was observed for cavity radii of 2 m
or less, even in environments with higher internal pressures. By keeping internal
pressure up to approximately 10 MPa, no plastic and shear yielding were
observed in all the modeled cavities (i.e. radii of 2, 6, 10, and 20 m).
b) Applying internal pressure of 75 MPa caused plastic or shear yielding in all the
cavities.
158 |
Colorado School of Mines | c) Depending on the geomechanical properties of the rock formation (intact rock
with the geomechanical properties similar to those outlined in Table 5.1 and a
cavity height of 2 m), cavity radius of 10 m was determined to be the maximum
size that could be reached without increasing stability issues, provided the internal
pressure did not exceed a maximum of 30 MPa. It was observed that applying
internal pressure of more than 30 MPa would induce tension failure around the
cavity at a radius of 10 m. Applying internal pressure of 35 MPa will limit the
cavity radius up to 6 m. In practice, the maximum borehole mining cavity radii
has been empirically determined to be 26 ft (7.9 m) in sandstone.
d) Designing a cavity with a radius of more than 10 m without having stability
problems requires applying less internal pressure. In order to reach a cavity radius
of 20 m, the internal pressure shall not exceed 10 MPa. Applying small internal
pressure (extraction and pressurization) may have technical challenges associated
with shorter standoff distance and material removal.
The influence of changing pore pressure on the effective stress distribution, and the
extent of the tensile stress region, were determined to be minor. However, variation in
effective stress will impact permeability, and will increase the gas flow as a function of
increasing permeability. As discussed in Chapter 3, the absolute permeability in coal can
vary due to changes in the pressure of the formation. While borehole mining underwater
has the advantage of providing additional roof support during the excavation process, the
standoff distance that the waterjet can reach effectively will be reduced dramatically
influencing the economics. As such, borehole mining is preferred in deposits above the
water table in unsaturated environments.
159 |
Colorado School of Mines | Cavity cyclic pressurization (bailing) will cause the stress concentration to move from
cavity roof to the side-wall and corner. However, since negative pressure values are
relatively minor compared to the original applied internal pressure, it should not impact
cavity stability significantly.
Increasing the horizontal stress will decrease the induced shear stress at the cavity roof
corner.
One of the major concerns in the design of borehole mining systems is the evaluation of
cavity stability based on stress concentration and relaxation around the extraction area.
The stress concentration generally depends on two parameters: the volume of removed
coal and the load transfer to the cavity face as a consequence of mining. Cavity stability
can be maintained by caving the mined-out area before the stress concentration on the
face becomes critical. Designing suitable cavity shape and extraction orientation can be
mitigating factors. Two extraction strategies under different k values, and different cavity
shapes were studied.
a) Reaching final optimum cavity size by using horizontal slices extraction has
less stability concerns as compare to using vertical slices.
b) In a borehole mining design, leaving the side walls in an arch shape
(configuration), will increase the stability.
Figure 5.29 presents a suggested algorithm for applying two dimensional finite difference
method in order to predict the maximum cavity size that can be obtained without stability
concerns. The model set up process is similar to the procedure explained in Section 5.2.
160 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.