You are on page 1of 6

International Journal of Mining Science and Technology xxx (2017) xxxxxx

Contents lists available at ScienceDirect

International Journal of Mining Science and Technology


journal homepage: www.elsevier.com/locate/ijmst

Numerical investigation into the effect of backfilling on coal pillar


strength in highwall mining
Mo S. a, Canbulat I. a, Zhang C. a, Oh J. a,, Shen B. b, Hagan P. a
a
School of Mining Engineering, The University of New South Wales, Sydney 2052, Australia
b
CSIRO Energy, QCAT, 1 Technology Court, Pullenvale 4069, Australia

a r t i c l e i n f o a b s t r a c t

Article history: This paper attempts to quantify the effect of backfilling on pillar strength in highwall mining using
Received 22 December 2016 numerical modelling. Calibration against the new empirical strength formula for highwall mining was
Received in revised form 15 March 2017 conducted to obtain the material parameters used in the numerical modelling. With the obtained coal
Accepted 1 May 2017
strength parameters, three sets of backfill properties were investigated. The results reveal that the behav-
Available online xxxx
ior of pillars varies with the type and amount of backfill as well as the pillar width to mining height ratio
(w/h). In case of cohesive backfill, generally 75% backfill shows a significant increase in peak strength, and
Keywords:
the increase in peak strength is more pronounced for the pillars having lower w/h ratios. In case of non-
Highwall mining
Backfill
cohesive backfill, the changes in both the peak and residual strengths with up to 92% backfill are negli-
Coal pillar gible while the residual strength constantly increases after reaching the peak strength only when 100%
Strength backfill is placed. Based on the modelling results, different backfilling strategies should be considered
Numerical modelling on a case by case basis depending on the type of backfill available and desired pillar dimension.
2017 Published by Elsevier B.V. on behalf of China University of Mining & Technology. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction [6]. In South Africa, backfilling was considered as a way of stabiliz-


ing old and/or under-designed bord and pillar workings as well as
Highwall mining is a mining method that produces coal from directly achieving higher recovery by increasing pillar height in
the base of highwall in surface mines as illustrated in Fig. 1 [1]. thick seam conditions [7,8]. In Australian coal operations, backfill-
A series of parallel entries are driven into the coal seam using ing has not been widely practised though a case of backfilling into
two types of highwall mining systems, namely, continuous high- pre-driven longwall recovery roadway in weak roof areas in a long-
wall mining (CHM) and auger mining. CHM system uses a contin- wall mine was reported [9]. In another application in Australia, old
uous miner generating typically 3.5 m wide rectangular entries coal mine workings were backfilled to prevent potential subsi-
while auger system excavates single or double holes typically from dence damage to a new road development on surface [10].
1.5 m to 1.8 m in diameter [2]. The penetration depths range from With the wide application of backfilling in the mining industry,
50 m to 500 m depending on the highwall mining systems and research was carried out in the late 1990s to investigate the possi-
mining conditions [3]. The pillars are then left between the mined bility of utilizing backfilling in highwall mining for the purpose of
entries to support the overburden. increasing coal recoveries [5,11]. As illustrated in Fig. 2, a typical
Backfilling technology has been widely used in the mining highwall mining layout incorporates the web pillars, generated
industry. In metalliferous mining, backfilling is typically used to by excavation of a series of entries using CHM system, and the bar-
provide a working platform, usually in cut and fill mining, or a sup- rier pillars that are left between the panels. If the entries are back-
port to the pillars and walls for higher productivity [4,5]. It has filled, the strength of web pillar will increase due to the
been widely applied in underground coal mining. In order to confinement. Therefore, with the use of backfill, potentially it
reduce surface subsidence, longwall goaf is backfilled to reduce may be possible to generate more slender pillars or omit the bar-
the mining void and subsequently the caving height in some Euro- rier pillars in highwall mining panel, which can lead an enhanced
pean mines. In Chinese mines backfill material is injected into part- recovery. In addition, the web pillars are known to fail suddenly
ing planes in the roof strata to prevent the closure of the partings accompanying catastrophic domino failures due to their low resid-
ual strengths [12,13]. Therefore, apart from the potential advan-
tage of an enhanced coal recovery by using backfilling in
Corresponding author.
E-mail address: joung.oh@unsw.edu.au (J. Oh).

http://dx.doi.org/10.1016/j.ijmst.2017.07.003
2095-2686/ 2017 Published by Elsevier B.V. on behalf of China University of Mining & Technology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Mo S et al. Numerical investigation into the effect of backfilling on coal pillar strength in highwall mining. Int J Min Sci
Technol (2017), http://dx.doi.org/10.1016/j.ijmst.2017.07.003
2 S. Mo et al. / International Journal of Mining Science and Technology xxx (2017) xxxxxx

a total of 29 cases with 14 collapsed cases and 15 uncollapsed


cases. Therefore, it is noted that the formula is only applicable to
the highwall mining pillars with width to height ratios (w/h) from
0.5 to 2.5.
It is considered that as the new formula has been developed
from an actual database of highwall mining, it represents the best
in-situ behavior of coal pillars in highwall mining. Due to the nat-
ure of long pillars in highwall mining conditions, plane strain anal-
ysis was conducted using the two-dimensional numerical software
FLAC [19]. The constitutive law for coal pillars was the strain-
Fig. 1. Highwall mining operation using CHM system [1].
softening model based on the Mohr-Coulomb failure criterion.
Therefore, the calibration process aimed to back calculate a set of
input parameters including the cohesion, angle of internal friction
and corresponding plastic strain range.

2.1. Model development

Fig. 3 shows the pillar model incorporating half of coal, roof and
floor along the symmetrical centerline of the pillar system which is
a repeating geometry in a highwall mining panel. The height of the
Fig. 2. Typical geometry of highwall mining panel with backfill.
roof and floor was 20 m and the mining height was fixed at 3 m
while the pillar widths varied in order to simulate pillars with w/
h ratios from 0.5 to 2.5. The uniform element size of 0.25 m was
highwall mining, pillar stability will be increased significantly if
applied to the coal and a smooth variation of zoning from the coal
there is an increase in residual strength with the use of backfill.
to the boundaries was used for roof and floor with appropriate
A number of forms of backfilling for highwall mining were pro-
aspect ratios to avoid numerical instability. Roller boundaries were
posed and relevant filling methods and indicative costs were pre-
applied along the side of the roof and floor, the bottom of the floor
sented in the past studies [5,11]. Even though backfilling is
and the vertical line.
thought to be effective in increasing coal pillar strength, it was
In this study, the roof and floor were considered to be an elastic
challenging to predict the behavior of the pillars with backfill
material with the stiffer material property [16,2022]. The inter-
due to the difficulties in modelling slender coal pillars and the
faces between coal and the roof and floor were taken from recent
interaction between pillars and different backfill materials.
coal pillar design studies [21,23]. The material properties used
This paper aims to quantify the effect of backfilling in highwall
for the calibration of model are provided in Table 1.
mining on pillar strength using numerical modelling. While few
After generating the grid and assigning the boundary conditions
numerical analyses of the effect of backfilling in highwall mining
and material properties, the model was stepped to an equilibrium
can be found, a limitation of the studies was the modelling of coal
state to develop in-situ stresses. The second equilibrium was then
pillars used in highwall mining [11,14]. In this paper, modelling of
achieved by excavating the entry. The entry width was fixed at
coal pillars is based on a recent study into a pillar design for high-
1.75 m which is a half width of a typical entry width of 3.5 m used
wall mining, which proposes a new empirical pillar strength for-
in the CHM method. Up until this point, very high cohesion and
mula for highwall mining [15]. The numerical investigations into
tensile strength values for coal were initially used to minimize
the interaction between the coal pillar and backfill in bord and pil-
the inertial effects so that more static solutions can be obtained.
lar mining were attempted by Ryder and Wang et al. [7,16], and
similar approach is used in this study. In Section 2, coal pillar prop-
erties are calibrated to represent the behavior of the slender pillars
in highwall mining. In Section 3, different types of backfill are
placed into the entry and the strength of coal pillar is monitored
with increasing backfill. The significance of the numerical results
is discussed in Section 4. It should be noted that only the consider-
ations for CHM system are included in this paper.

2. Calibration of model

Numerical models that are calibrated against empirical pillar


strength formulae can be a reliable way to explain the average
behavior of a pillar model [17]. In this regard, calibration of coal Fig. 3. Geometry of coal pillar model.
parameters was conducted against the proposed linear formula
for highwall mining recently developed by Canbulat et al. [15]:
Table 1
rp 4:660:56 0:44w=h 1 Material properties used for calibration of model.

where rp is the pillar strength, MPa; w/h is the pillar width to min- Material Youngs modulus (GPa) Poissons ratio
ing height ratio. Roof/floor 10 0.25
This new formula has been developed using the maximum like- Coal 2.5 0.30
lihood method, which is similar to how the UNSW pillar strength Interface Normal stiffness = 100 GPa/m; shear
formulae were originally derived [18]. The difference is that only stiffness = 50 GPa/m; cohesion = 0.5 MPa; Internal
friction angle = 30
a highwall mining pillar database has been used, which includes

Please cite this article in press as: Mo S et al. Numerical investigation into the effect of backfilling on coal pillar strength in highwall mining. Int J Min Sci
Technol (2017), http://dx.doi.org/10.1016/j.ijmst.2017.07.003
S. Mo et al. / International Journal of Mining Science and Technology xxx (2017) xxxxxx 3

Table 2
Coal properties calibrated against Eq. (1).

Material Youngs modulus (GPa) Poissons ratio Cohesion (MPa) Friction angle () Plastic strain
Peak Residual Peak Residual
Coal 2.5 0.3 0.9 0.1 35 12 0.012

Table 3
Peak strength results calibrated against Eq. (1).

Peak strength calculated by (MPa) w/h = 0.5 w/h = 1 w/h = 1.5 w/h = 2 w/h = 2.5
Eq. (1) 3.63 4.66 5.69 6.71 7.74
Model in Fig. 3 3.48 4.62 5.92 6.80 7.29

Then the cohesion and tensile values were set to their original val- strain-softening for the pillars with lower w/h ratios to strain-
ues and the model was stepped to equilibrium again. Finally, the hardening for the pillars with higher w/h ratios. The curves in this
increasing load was generated by applying constant velocity of figure describe the general behavior of in-situ coal pillars as agreed
106 m/s on the top of the roof in the model. by many researchers [6,12]. As field measurement data for high-
The coal pillar stress was estimated by taking the average of wall mining pillars are rarely available, the modelling of a series
vertical stresses generated in the elements located at the mid- of coal pillars, with w/h ratios from 0.5 to 10 in this study, can
point of the pillar. The vertical strain was calculated based on the be a practical way to validate the coal properties [24]. Overall,
ratio of the difference between the average vertical displacements the set of coal properties can be used for highwall mining pillars,
at the top and bottom of the coal pillar and the mining height. which follow the peak strengths calculated by Eq. (1) as well as
the general post-peak behaviors of coal pillars.
It is worthy of note that the obtained set of parameters may not
2.2. Determination of coal parameters
be the unique because different constitutive models, different
combinations of material parameters and different boundary con-
The input parameters for coal were calibrated against Eq. (1), by
ditions can produce the same calibration results [6,21]. It is also
monitoring the peak stresses with various sets of input parameters,
worth noting that the obtained values are dependent on mesh size,
which are shown in Table 2. The peak strengths of coal pillars with
therefore, those values can be changed if different mesh sizes are
w/h ratios of 0.5 and 1, as in this case, were mainly determined by
used [21,22].
the peak values of cohesion and internal friction angle, in contrast
to the pillars with w/h ratios of 1.5, 2 and 2.5 determined by both
the peak and residual values of them. The calibration results are 3. Numerical modelling of backfilling
given in Table 3 and Fig. 4.
While the peak strengths were modelled, the validation of the A few types of backfill, such as a dense fluid of unconsolidated
post-peak behaviors of the modelled pillar was attempted by con- rejects, a partially consolidated deposit of rejects and a saturated
structing coal pillar models with w/h ratios of up to 10 using the cement backfill paste were proposed for the application of backfill-
obtained set of coal properties. The stress-strain curves are ing in highwall mining [11]. In this paper, three sets of backfill
illustrated in Fig. 5, which shows the range of behaviors from properties were adopted to investigate the effect of backfilling on
coal pillar strength. These properties, as listed in Table 4, were
adapted from a study of mine remediation using fly ash backfilling
[25]. The mixture of fly ash and water was referred to as non-
cohesive backfill while the combination of fly ash, cement, water,
etc. was termed as cohesive backfill [25]. Two types of cohesive
backfill were considered with the unconfined compressive
strength (UCS) of 0.5 MPa and 1 MPa. The Mohr-Coulomb strength
criterion was used for the inelastic behavior of backfill. The input
parameters given in Table 4 are regarded as typical backfill proper-
ties [16].
For each type of backfill, the percentage fills from 8% to 100% of
Fig. 4. Stress-strain curves of modelled highwall pillars with w/h ratios from the pillar height were placed into the entry and the coal pillar
0.5 to 2.5. strengths with varying amounts of backfill were monitored. While
using the same pillar model and modelling process as in the pre-
ceding section, the entry was backfilled and the model was equili-
brated right before applying the increasing load in the model. The
geometry of the coal pillar model with backfill is provided in Fig. 6.

3.1. Cohesive backfill

The peak strengths of pillars with varying amounts of backfill


are listed in Table 5. In case of cohesive backfill, the increase in
peak strength is generally more pronounced as the w/h ratio of
the pillar decreases and the results can be grouped by the w/h ratio
Fig. 5. Stress-strain curves of coal pillar model with w/h ratios from 0.5 to 10. of the pillars. The peak strength of the pillars with w/h ratios of 0.5

Please cite this article in press as: Mo S et al. Numerical investigation into the effect of backfilling on coal pillar strength in highwall mining. Int J Min Sci
Technol (2017), http://dx.doi.org/10.1016/j.ijmst.2017.07.003
4 S. Mo et al. / International Journal of Mining Science and Technology xxx (2017) xxxxxx

Table 4
Backfill properties used for modelling [25].

Backfill type Youngs modulus (GPa) Poissons ratio Cohesion (MPa) Friction angle () Dilation angle () Tensile strength (MPa) UCS (MPa)
Cohesive 0.57 0.40 0.12 40 10 0.1 1
Cohesive 0.57 0.40 0.06 40 10 0.05 0.5
Non-cohesive 0.02 0.15 0 42 7.5 0 0

75% to 83% backfill. Meanwhile, the peak strength of the pillars


with w/h ratios of 2 and 2.5 increases less than 10% with backfill
up to 67%.
In almost all of the cases, the peak strength significantly
increases with 92% backfill height compared to 83% backfill. When
the entries are completely filled with backfill (i.e., 100%), the pillars
show an elastic behavior. The stress-strain curves of the pillars
with w/h ratios of 0.5, 1.5 and 2.5 with 1.0 MPa cohesive backfill
are illustrated in Fig. 7.
The residual strengths of pillars with varying amounts of back-
fill were also monitored, as shown in Table 6. As the values of post-
Fig. 6. Geometry of coal pillar model with backfill.
peak strength are not exactly converged over strain to the residual
strengths, those values at 2% of strain were taken, where the resid-
ual strengths are deemed to be reached.
and 1 generally increases with increasing backfill. Those pillars Considering the range of residual strengths without backfill, in
show an approximately 50% increase in peak strength when the order of 0.5 MPa, the residual strength increases considerably
backfill is placed up to 75%-83% of the pillar height. The pillar with with increasing backfill for the pillars with w/h ratios from 1 to
a w/h ratio of 1.5 also shows an increase in peak strength with 2.5. The rise in residual strength is more pronounced as the w/h
increasing backfill, but the percentage increase is nearly 20% with ratio of the pillar increases. The residual strength of the pillar with

Table 5
Peak strength with cohesive backfill (MPa).

Backfill Percentage of backfill (%) Peak strength (percentage increase compared to non-backfill value)
w/h = 0.5 w/h = 1.0 w/h = 1.5 w/h = 2.0 w/h = 2.5
Cohesive 8 3.50 (1%) 4.79 (4%) 6.07 (3%) 6.83 (0%) 7.27 (0%)
UCS 1 MPa 17 3.54 (2%) 5.04 (9%) 6.31 (7%) 6.98 (3%) 7.33 (1%)
25 3.63 (4%) 5.34 (16%) 6.55 (11%) 7.12 (5%) 7.36 (1%)
33 3.78 (9%) 5.67 (23%) 6.80 (15%) 7.23 (6%) 7.44 (2%)
42 4.00 (15%) 6.00 (30%) 6.98 (18%) 7.34 (8%) 7.52 (3%)
50 4.33 (24%) 6.30 (36%) 7.13 (20%) 7.44 (9%) 7.57 (4%)
58 4.52 (30%) 6.41 (39%) 7.14 (21%) 7.42 (9%) 7.64 (5%)
67 4.78 (37%) 6.57 (42%) 7.08 (20%) 7.41 (9%) 7.69 (5%)
75 5.04 (45%) 6.93 (50%) 7.41 (25%) 8.17 (20%) 8.30 (14%)
83 5.51 (58%) 7.39 (60%) 7.51 (27%) 8.05 (18%) 9.24 (27%)
92 5.98 (72%) 9.07 (96%) 9.76 (65%) 10.40 (53%) 11.69 (60%)
100 N/A N/A N/A N/A N/A
Cohesive 8 3.50 (1%) 4.75 (3%) 6.03 (2%) 6.76 (-1%) 7.17 (-2%)
UCS 0.5 MPa 17 3.53 (1%) 4.95 (7%) 6.26 (6%) 6.94 (2%) 7.31 (0%)
25 3.61 (4%) 5.24 (13%) 6.40 (8%) 7.22 (6%) 7.67 (5%)
33 3.75 (8%) 5.50 (19%) 6.67 (13%) 7.27 (7%) 7.62 (5%)
42 3.94 (13%) 5.74 (24%) 6.79 (15%) 7.24 (6%) 7.50 (3%)
50 4.20 (21%) 5.97 (29%) 6.85 (16%) 7.21 (6%) 7.38 (1%)
58 4.40 (26%) 6.13 (33%) 6.92 (17%) 7.22 (6%) 7.82 (7%)
67 4.62 (33%) 6.36 (38%) 7.01 (18%) 7.35 (8%) 7.83 (7%)
75 4.84 (39%) 6.69 (45%) 7.27 (23%) 7.87 (16%) 8.73 (20%)
83 5.14 (48%) 7.13 (54%) 7.59 (28%) 7.74 (14%) 8.54 (17%)
92 5.45 (57%) 7.77 (68%) 8.45 (43%) 8.98 (32%) 9.69 (33%)
100 N/A N/A N/A N/A N/A

Fig. 7. Stress-strain curves with 1 MPa cohesive backfill.

Please cite this article in press as: Mo S et al. Numerical investigation into the effect of backfilling on coal pillar strength in highwall mining. Int J Min Sci
Technol (2017), http://dx.doi.org/10.1016/j.ijmst.2017.07.003
S. Mo et al. / International Journal of Mining Science and Technology xxx (2017) xxxxxx 5

Table 6
Residual strength with cohesive backfill at 2% of strain (MPa).

Backfill Percentage of backfill (%) Residual strength


w/h = 0.5 w/h = 1.0 w/h = 1.5 w/h = 2.0 w/h = 2.5
Cohesive 8 0.35 0.37 0.42 0.55 0.69
UCS 1 MPa 17 0.35 0.37 0.46 0.58 0.77
25 0.35 0.40 0.59 0.77 0.91
33 0.45 0.60 0.74 1.05 1.24
42 0.41 0.53 0.67 1.02 1.36
50 0.43 0.80 1.07 1.31 1.35
58 0.53 0.86 1.13 1.23 1.44
67 0.90 0.99 1.35 1.42 1.69
75 1.18 1.51 1.56 1.79 1.96
83 0.98 1.47 2.10 2.30 2.84
92 1.77 2.63 3.23 3.51 3.81
100 N/A N/A N/A N/A N/A
Cohesive 8 0.35 0.37 0.43 0.55 0.74
UCS 0.5 MPa 17 0.35 0.36 0.45 0.58 0.76
25 0.38 0.43 0.68 0.81 0.98
33 0.39 0.63 0.80 1.00 1.24
42 0.41 0.73 0.89 1.08 1.29
50 0.68 0.72 1.00 1.12 1.32
58 0.77 0.81 1.05 1.18 1.43
67 0.88 1.11 1.09 1.40 1.49
75 0.95 1.47 1.35 1.66 1.90
83 1.04 1.35 1.72 1.98 2.09
92 1.19 1.65 2.04 2.39 2.56
100 N/A N/A N/A N/A N/A

Fig. 8. Stress-strain curves with non-cohesive backfill.

a w/h ratio of 0.5 is reached at around 1 MPa, only with the excep- Dw is the lateral displacement of pillar after backfill is placed;
tion when 92% of 1 MPa cohesive backfill is placed. In almost all of and B is the width of entry.
the cases, the residual strength significantly increases with 92% As can be seen from Eq. (2), the confining pressure is influenced
backfill compared to 83% backfill, which is similar to the results by the elastic modulus of backfill. With the non-cohesive backfill,
for peak strength. even 92% of backfill makes no substantial difference in terms of
both peak and residual strengths in this modelling study, possibly
3.2. Non-cohesive backfill due to the low elastic modulus of the non-cohesive backfill. How-
ever, non-cohesive backfill may be effective in other cases where
In case of non-cohesive backfill, only 100% backfill is effective the lateral displacement of pillars is sufficiently developed prior
while the changes in both the peak and residual strengths with to reaching the peak stresses. An example of non-cohesive backfill-
8% to 92% backfill are negligible. Moreover, the peak strength even ing in bord and pillar mining is found from Wang et al. [16].
hardly increases with 100% backfill while the residual strength It was also suggested by Galvin that the height of backfill should
constantly increases over strain as shown in Fig. 8. It is believed be more than two-thirds of the height of pillar as the lateral expan-
that confinement is not sufficiently generated before reaching sion is largest at mid-height of pillar [6]. The numerical results are
the peak stress, which will be discussed again in Section 4. generally consistent with the suggestion that the both peak and
residual strengths considerably increase with 75% cohesive backfill
4. Discussion for almost all of the modelled pillars.

4.1. Mechanism of confinement by backfilling 4.2. Consideration for backfilling in highwall mining

Galvin stated that the confinement by backfilling is primarily As predicted by numerical modelling in Section 3, the behavior
generated in response to lateral expansion of pillar [6]. The confin- of pillar varies with the type and amount of backfill as well as the
ing pressure developed by backfilling is given in Eq. (2) [8]: w/h ratio of the pillars. Therefore, different backfilling strategies
should be considered on a case by case basis depending on the type
rF EF eF EF Dw=B 2
of backfill available and desired pillar dimension. Some sugges-
where rF is the confining pressure developed by backfilling; EF is tions and considerations for backfilling in highwall mining are
the elastic modulus of backfill; eF is the lateral strain of backfill; summarized below:

Please cite this article in press as: Mo S et al. Numerical investigation into the effect of backfilling on coal pillar strength in highwall mining. Int J Min Sci
Technol (2017), http://dx.doi.org/10.1016/j.ijmst.2017.07.003
6 S. Mo et al. / International Journal of Mining Science and Technology xxx (2017) xxxxxx

(1) If a significant increase in peak strength is targeted, gener- [3] Verma CP, Porathur JL, Thote NR, Roy PP, Karekal S. Empirical approaches for
design of web pillars in highwall mining: review and analysis. Geotech Geol
ally 75% backfill using cohesive material may be enough to
Eng 2014;32(2):58799.
stabilize the pillars. The pillars with w/h ratios of 0.5 and 1 [4] Belem T, Benzaazoua M. An overview on the use of paste backfill technology as
show 50% increase in peak strength. Compared to that, even a ground support method in cut-and-fill mines. In: Proceedings of the 5th
though the increase in peak strength is not significant for the international symposium on ground support in mining and underground
construction. Perth (Western Australia): Taylor & Francis Group; 2004. p. 63750.
pillars with w/h ratios of 1.5 to 2.5, the percentage increase [5] BFP Consultants. Improved recovery in highwall mining using backfill.
in peak strength is approximately 20%. Furthermore, 75% Brisbane: Australian Coal Association Research Program (ACARP); 1997
backfill also brings improved long-term stability by the [ACARP C3052 Final Report].
[6] Galvin JM. Ground engineering- principles and practices for underground coal
increased residual strengths. mining. New York: Springer; 2016.
(2) The effect of the cohesive backfill with UCS of 0.5 MPa is not [7] Ryder JA. Determination of the quantity of ashfill required to stabilize and
much different from the one with UCS of 1 MPa, only with prevent undue subsidence of old workings. Johannesburg: Chamber of Mines
of South Africa Research Organisation; 1992 [Report No. B03/92].
the exception of 92% backfill. Unless more than 50% increase [8] Galvin JM. The significance, behavior and influence of ashfill on South African
in peak strength is required with 92% backfill, the 0.5 MPa thick seam mining operations. Johannesburg: Chamber of Mines of South
cohesive backfill may be as effective as the 1 MPa backfill. Africa Research Organisation; 1982 [Report No. 9/82].
[9] Rutty YM, Payne D, Mckenzie A. The evolution of pre-driven recovery
(3) When non-cohesive backfill is considered, full height of pil- roadways at Crinum mine. In: Proceedings of 16th Coal Oper Conf.
lar should be backfilled. Non-cohesive backfill may be pre- Wollongong: University of Wollongong; 2016. p. 17789.
ferred over cohesive backfill as cement doesnt need to be [10] Millar D, Holz B. Ipswich motorway upgrade-filling of abandoned coal mines.
Queensl Roads 2010;9:4661.
added and it can be injected properly into the entry due to
[11] Clark IH, Boyd GL. Highwall mining backfill. Brisbane: Australian Coal
flowability [25]. Association Research Program (ACARP); 1999 [ACARP C6014 Final Report].
(4) The property of backfill material typically changes with [12] Mark C. The evolution of intelligent coal pillar design: 19812006. In:
time. It is noted that this numerical study is not on a real Proceedings of 25th international conference on ground control in mining.
Morgantown: West Virginia University; 2006. p. 32534.
time basis and the modelling results assume that the prop- [13] Mark C. The state-of-the-art in coal pillar design. Trans-Soc Min Metall Explor
erty of backfill is settled over time. In reality, the strength Inc 1999;308:12328.
and the corresponding curing time of the chosen backfill [14] Sasaoka T, Karian T, Hamanaka A, Shimada H, Matsui K. Application of
highwall mining system in weak geological condition. Int J Coal Sci Technol
property should be checked. 2016;3:31121.
[15] Canbulat I, Mo S, Zhang C, Hagan PC, Oh J. Coal pillar design for highwall
5. Conclusions mining and geotechnical considerations in backfilling. Sydney: The University
of New South Wales; 2016.
The effect of backfilling in highwall mining was investigated by [16] Wang H, Poulsen BA, Shen B, Xue S, Jiang Y. The influence of roadway backfill
on the coal pillar strength by numerical investigation. Int J Rock Mech Min Sci
numerical analysis. Calibration of coal parameters was conducted 2011;48(3):44350.
against the new empirical pillar strength formula. Three sets of [17] Esterhuizen GS. Extending empirical evidence through numerical modelling in
backfill including cohesive backfill with UCS of 0.5 MPa and rock engineering design. J S Afr Inst Min Metall 2014;114(10):75564.
[18] Salamon MDG, Galvin JM, Hocking G, Anderson I. Coal pillar strength from
1 MPa and non-cohesive backfill were chosen. back-calculation. Sydney: The University of New South Wales; 1996 [Research
The modelling results indicate that the behavior of pillar varies Report 1/96].
with the type and amount of backfill as well as their w/h ratios. In [19] Itasca Consulting Group. FLAC (Fast Lagrangian Analysis of Continua), version
5.0. Minneapolis: Itasca Consulting Group Inc.; 2006.
case of cohesive backfill, the peak strength of pillars with w/h
[20] Duncan Fama ME, Trueman R, Craig MS. Two- and three- dimensional elasto-
ratios of 0.5, 1 and 1.5 increases with increasing backfill while plastic analysis for coal pillar design and its application to highwall mining. Int
the peak strength of pillars with w/h ratios of 2 and 2.5 consider- J Rock Mech & Min Sci Geomech Abstr 1995;32(3):21525.
ably increases with greater than 75% backfill. The residual strength [21] Esterhuizen E, Mark C, Murphy MM. Numerical model calibration for
simulating coal pillars, gob and overburden response. In: Proceedings of
of pillars generally increases with increasing cohesive backfill. 29th international conference on ground control in mining. Morgantown:
However, for non-cohesive backfill, only 100% backfill is effective. West Virginia University; 2010. p. 4657.
Based on the results, different backfilling strategies should be [22] Perry KA, Raffaldi MJ, Harris KW. Influence of highwall mining progression on
web and barrier pillar stability. Min Eng 2015;67:5967.
considered on a case by case basis depending on the type of backfill [23] Mohamed KM, Tulu IB, Klemetti T. Numerical simulation of deformation and
available and desired pillar dimension. failure process of coal-mass. In: 49th U.S. rock mechanics/geomechanics
symposium. San Francisco: American Rock Mechanics Association; 2015.
[24] Zipf RK. Ground control design for highwall mining. Trans-Soc Min Metall
References Explor Inc 2006;318:226.
[25] Shen B, Alehossein H, Poulsen B, Huddlestone-holmes C, Zhou B, Luo X.
[1] Shen B, Duncan Fama ME. Review of highwall mining experience in Australia Collingwood park mine remediation-subsidence control using fly ash
and a case study. Aust Geomech 2001;36(2):2532. backfilling. Brisbane: CSIRO; 2010 [CSIRO Earth Science and Resource
[2] Duncan Fama ME, Shen B, Maconochie P. Optimal design and monitoring of Engineering Report EP105068].
layout stability for highwall mining. Brisbane: Australian Coal Association
Research Program (ACARP); 2001 [ACARP C8033 Final Report].

Please cite this article in press as: Mo S et al. Numerical investigation into the effect of backfilling on coal pillar strength in highwall mining. Int J Min Sci
Technol (2017), http://dx.doi.org/10.1016/j.ijmst.2017.07.003