Introduction

Background of the study

The consumption rate of mineral resources as well as increase in mining has increased tremendously in twenty-first century due to increase in urbanization, population and advancement in science and technology (Worldbank 2019). In some areas, mining activities have resulted in destruction of soil physiology, nutrient cycles, alteration of microbiological and macro-communities thereby destroying the aesthetics beauty of a productive ecosystem (Franks et al. 2011; Buta et al. 2019). The study areas, Gyel’A’ (Caha) and Kantoma in Plateau State, originally have numerous hillocks with gentle slopes emerging from the ground like mushrooms scattered with huge boulders creating unusual scenery of the Plateau. Unfortunately, mining activities has disrupted some part of the ecosystem and the aesthetic beauty of this Plateau with its characteristic mine tailings spreading over wide areas, deep pits, abandoned mine excavations (Fig. 1a–d), pilot ponds and failed earth dam/banks, etc.

Fig. 1
figure 1

Mine sites as seen in the study area

Full size image

During mining, very large voids are often created which are potential sources of subsidence, contaminants to groundwater, surface water, and the surrounding ecological system. Some of the abandoned mine pits in the study areas, Gyel’A’ and Kantoma mine sites have started caving/collapsing regardless of the underlying geology (Fig. 1b). To permit the continued mining of tin, protection and subsequent beneficial use of the mined land there is need for reclamation. Reclamation standards included backfilling, regrading, slope stability and recontouring, among other reclamation standards (Testa and James 2007), but this paper focused more on backfilling. In Canada, Australian mining industries and across the globe where safety is a prime consideration, mine backfilling is a technique that has been in use for decades (Thomas 1979; Hassani et al. 1989; Sivakugan et al. 2006).

Backfilling is the process of replacing or reusing the soil that has been removed during excavation back into an area that has been excavated (Arora 2004; Williams 2006; Rodriguez 2021). Materials such as cemented materials, high water content material and granular material like overburdens, waste rocks, tailings, quarried and crushed aggregate, and metallurgical processed tailings are often used in backfilling (Environment Canada 1987; Rhoades and Binkley 1996; Zhang et al. 2019). For the mine operator, pit backfill could mean shortened and potentially less costly downhill haulage routes, and a potential way to disposing mine tailings, and the overburdens (Parshley et al. 2006; Dowling et al. 2004). The backfilling of mine pits is loaded with many environmental benefits. Problems related to surface disposal of tailings can be reduced as substantial quantities of potential pollutants can be returned underground (Franks et al. 2011; Levesque et al. 2017). It controls the strata, reduces the mine site footprint, surface subsidence and also protects the environment by improving the local and regional stability, enabling safer and more efficient mining of the surrounding areas (Ercikdi et al. 2009; Fall et al. 2010; Villain 2011; Sivakugan et al. 2015; Sheoran et al. 2010; Sandor 2010).

However, despite all these benefits associated with backfilling materials, it is worth noting that the performance of a backfilling material depends on different geotechnical parameters, such as water holding capacity, permeability, porosity, specific gravity, bulk density, shear strength and most importantly grain size distribution. Owing that freshly placed backfill material tends to compress under its own weight and generates shear stress at the backfill–rock interface due to friction (Levesque et al. 2017). Thus, proper reclamation of mine pits involving backfilling materials requires accurate modeling. For instance, horizontal stress exerted by backfill material on excavated walls must be estimated accurately to avoid the failure of mine pillars (Gupta and Paul 2017). To reduce the susceptibility of the material to settlement, suitable compaction, optimum moisture content and maximum dry density are important when placing soils/tailings as backfill materials (Sullivan et al. 2011; Levesque et al. 2017). What may work for one type of mine pit may not work for another type, and may not work for the same type of mine pit at different geological sites (Water and Carroll 2007; Testa and James 2007).

Although there are few of research work on backfilling materials, but studies on modeling and characterization of backfilling/reclamation materials using integrated method is not quite extensive. In this study, possible backfilling materials such as rock mass and mine tailings were modeled by integrating geotechnical parameters (water holding capacity, permeability, porosity, specific gravity, bulk density, shear strength and grain size distribution) with statistical tool such as correlation and multivariate analysis to mitigate failure of backfilling materials.

The study area

Plateau state is underlain basically by hard rock such as basalts and phonolite trachyte, quartz porphyry, hormblende granite, coarse porphrytic biotite and biotite hornblende, migmatite and granitic gneiss (Fig. 2). It is one of the areas where older granites were intruded by younger granites. These younger granites are thought to be about 160 million years old. The phases of volcanic activities 50 million years ago involved in the formation of Plateau State have made it one of the mineral-rich states in the country (Hodder 1959). In Plateau State tin, cassiterite and columbite are present in veins and pegmatites associated with rhyolites and granites intrusion, and can also be found in metamorphic rocks (Hobert 2005). Tin is also present in placer deposit, when the rock containing minerals are weathered they remain intact and eventually concentrated in streams (Calvin and Rosann 2003). Most of the world’s total Sn production is derived from secondary alluvial deposits resulting from the disintegration of the primary deposits (Ndace and Danladi 2012). The study areas Gyel’A’ (Caha) and Kantoma areas of Plateau State lie within longitude 08° 48′ E to 09° 06′ E and latitude 090 24′ N to 10° 00′ N (Fig. 2).

Fig. 2
figure 2

Geology of Plateau State (NGSA 2006)

Full size image

Methodology

Sample collection

Four samples of possible backfilling materials; normal soil and mine tailing were systematically collected at varying depth of 10.2–19.4 ft around Gyel’A’ and Kantoma mine pits. Gyel’A’1 and Kantoma1 samples were basically the mine tailings while Gyel’A’C and KantomaC samples were from the normal soils. The samples were kept inside an air tight bag and then carefully labeled before sending them to the laboratory for geotechnical characterization.

Data analysis

In the laboratory, the samples were analysed for the following geotechnical parameters; sieve analysis/grain size distribution, quick undrained triaxial test, permeability test, oedometer consolidation test and modified proctor compaction test. All these tests were carried out in accordance to the specifications in the British Standards BS: 5930 (1981), 8004 (1986), 1337 (1990) and American Society for Testing and Materials ASTM; Designation 2487 (2011).

Sieve analysis/grain size distribution: Sieving was performed after 500 g of the samples were washed through the sieve size of 0.0075 mm and then oven dried for 105 °C for 24 h. The samples were analysed for grain size distribution in accordance with British Standard Part 2:1990 and then characterized using USCS classification indices; coefficient of uniformity (Cu), coefficient of curvature (Cc), sorting coefficient (C) or So). The particle sizes are D10, D25, D30, D50, D60 and D75 in mm and 10%, 25%, 30%, 60% and 75% by weight of soil, passing the respective sieve sizes (Casagrande 1948). Casagrande (1948) calculates Cu, Cc as follows:

$${\text{Cu}} = D60/D10$$
(1)
$${\text{Cc}} = D30^{2} / D60 \, \times \, D10.$$
(2)

Cu ≥ 4 indicates a well-graded soil. Cu < 2 indicates a uniform soil. Cc between 1 and 3 indicate a well-graded soil.

Cc < 0.1 indicates a possible gap-graded soil.

$${\text{While}}\;{\text{Folk}}\;(1968){\text{ calculates C or So}} = (D75/D25)^{1/2}$$
(3)

According to Folks’ classification scheme, (S0): < 0.35 is classified as very well sorted, 0.35–0.50 well sorted, 0.5–0.71 moderately well sorted, 0.71–1.00 moderately sorted, 1.00–2.00 poorly sorted, 2.00–4.00 very poorly sorted and > 4.00: extremely poorly sorted.

Quick undrained triaxial test: Consolidated undrained triaxial compression tests were performed to investigate the shear strength behavior of the backfilling material. The shear strength of a soil is the relative resistance of that soil to sliding when supporting a load (Sullivan et al. 2011). Shear also analyzes and solves stability problems. Here the samples from Gyel’A’ and Kantoma were placed inside a latex rubber sheath and sealed to a top and base cap by rubber O-rings with water pressure inside the confining cell to induce three equal principal stresses (triaxial) in the tailing samples. The stress was measured by means of pressure transducers as the cell pressure increased from 50 to 200 kN. Shear strength (\(\tau\)) of the soil is defined via Coulomb's law (Eq. 4) (Coulomb 1776) which is

$$\tau =C+\sigma \mathrm{tan}\varphi ,$$
(4)

where \(\complement\) = soil cohesion for total stresses, \(\sigma\) = total normal stress, \(\varphi\) = angle of internal friction for total stresses.

$$\sigma =\frac{\sigma 1+\sigma 3}{2}+\frac{\sigma 1-\sigma 3}{2}cos2\phi$$
(5)

\(\sigma 1=\) major principal plane, \(\sigma 3=\) minor principal plane.

Cohesion is a measure of the forces that cement particles of soils while internal friction is the measure of the shear strength of soil due to friction (Singh and Goel 2011).

Oedometer consolidation test: Four remolded samples measuring 71.4 mm in diameter and the height of 20 mm were prepared for the test. The test was conducted in the standard way, with load increment of 10–800 kN. At each loading stage, deformational readings 6, 15 and 30 s, 1, 2, 4, 8, 16, 30 min and at 1, 2, 4, 8 and 24 h, were taken systematically to develop a time–settlement curve. Effective stress and void ratio were also deduced at each loading stage. The final void ratio was plotted on the natural scale and the effective stress as abscissa. From the plot, the coefficient of compressibility (Mv) was deduced from the slope of final void ratio versus effective stress (Arora 2004). Mathematically;

$$Mv \, = \, \Delta e/\Delta p,$$
(6)

where ∆e is the rate of change in void ratio, ∆p is the change in applied effective pressure during compression.

Permeability test: The permeability test of the samples was measured with a permeameter by applying a variable head gradient under different consolidation stresses of 25, 50, 100, 200, 400, and 800 kN.

Modified proctor compaction test: Compaction test was conducted using the Modified proctor test (ASTM Designation: D1557) to determine the relationship between water content and dry unit weight of soils (compaction curve). Here, the backfilling samples were compacted in 5 layers and with a 4.54 kg rammer dropping from a height of 45 cm (18 in.). 25 blows were evenly distributed on each layer and the weight of the mould; samples and water content were also noted. The data were later plotted to represents a curvilinear relationship known as the compaction curve or moisture-dry unit weight curve from where the optimum water content and maximum dry unit weight were determined.

Statistical modeling

The correlation and multivariate analysis of the geotechnical data were statistically modeled using Statistical package for social science (SPSS) software version 23.0. In this research work, they were used to model relationships between the variables, the influence of one or more on the other as well as better characterization of the of backfilling materials (Abdella et al. 2017).

Correlation analysis modeled correlation coefficient of the geotechnical parameters using Pearson correlation coefficients at two-tailed test of 0.01 and 0.05 levels of significance. The higher the value the stronger the correlation of the parameters to each other is. Correlation coefficients greater than 0.7, 0.5 < r < 0.7 and less than 0.5 were considered as strong, moderate and weak correlation, respectively (Soltani et al. 2017; Ukah et al. 2019, 2020).

Multivariate data analysis used for high-quality experimental results is principal component analysis (PCA). PCA used in this research is Kaiser’s varimax normalization rotation method (Kaiser 1960). The principal factors which come from variables with eigenvalues > 1.0 were adopted in order to maintain maximum variances of factor loadings. Factor loadings > 0.71 are regarded as excellent and < 0.32 very poor. Large (either positive or negative) loadings indicate that a variable has a strong effect on that principal component.

Result and discussion

The results from the Gyel’A’ and Kantoma samples are displayed in both Tables 1, 2, 3, 4, 5, 6, 7, 8 and Figs. 3, 4, 5, 6, 7. They were geotechnically characterized into grain size distribution, consolidation, compaction, void ratio, porosity, water content, dry density, permeability, compressibility and shear strength.

Table 1 Result of particle size distribution analysis
Full size table
Table 2 Quick undrained triaxial compression test
Full size table
Table 3 Oedometer test
Full size table
Table 4 Compaction test
Full size table
Table 5 Permeability test result
Full size table
Table 6 Correlation of the geotechnical parameters
Full size table
Table 7 Total variance explained
Full size table
Table 8 Extraction method: principal component analysis
Full size table
Fig. 3
figure 3

Particle size distribution curve

Full size image
Fig. 4
figure 4

Quick undrained triaxial test

Full size image
Fig. 5
figure 5

Oedometer test

Full size image
Fig. 6
figure 6

Compaction curve

Full size image
Fig. 7
figure 7

Component plot in rotated space

Full size image

Geotechnical characterization of the backfilling/reclamation materials

Sieve analysis/grain size distribution

Grain size distribution result (Table 1) showed that percentage of gravel is in the range of 2.80–14.90%, sand 72.8–86.5% and fines 5.10–24.2%. The samples, Gyel’A’1 and Kantoma1 indicate higher percentage of coarse particles than fines when compared to Gyel’A’C and KantomaC. According to Campos (2008), Murinko (2010) and AASHTO (2012), coarse grained materials are preferable in backfilling as they provide better drainage capacity and less sensitivity to swell or shrinkage problems unlike fine grains like clay. Thus, the higher the percentage of fines in a material, the less permeable it becomes (Arora 2004). Many researchers such as Knapen et al. (2006), Yang et al. (2007), Ubido et al (2018) and Igwe and Chukwu (2018) have discovered that higher percentage of fines/clay on earth materials have been the cause of many structural collapse/failure. Moreover, materials with clay content greater than 20% exhibited expansion abilities (Mugagga et al. 2011). Results from Table 1 are modeled in Fig. 3 to characterized the geotechnical indices; coefficient of uniformity (Cu), coefficient of curvature (Cc), sorting coefficient (\(\complement\)) or SO using the percentage of soil passing against particle size diameter. Based on the indices, Gyel’A’C is the most well graded with Cu of 6.8 and SO of 2.934 while Cc (1.24) is higher in KantomaC than in the other samples (Table 1) (Arora 2004). According Folk (1968) classification index, the four samples calculated from the particle size model (Fig. 3) are well graded. They increase in the following order; Kantoma1 (1.900) poorly sorted, Gyel’A’1 (2.213), KantomaC (2.401) and Gyel’A’C (2.934) very poorly sorted. Poorly sorted soil in sedimentology is regarded as well-graded soil in geotechnics and the same goes to very poorly sorted soil (Tan 2013; USDA 2012; Jackson 2019). Result from the particle analysis showed that Gyel’A’C and KantomaC have more percentage of fines and are better graded than other samples.

Consolidation characteristics of the samples

Consolidation test is necessary on a backfilling material since it predicts the magnitude and the rate at which settlement occurs by determining the coefficient of compressibility and consolidation (Prakash and Jain 2002). If settlement is not kept to tolerable limit the erection of structures may be impaired and the design life of the Engineering structure may be reduced (Das 2014; Knappett and Craig 2012).

Quick undrained triaxial

The quick undrained triaxial test carried out on Gyel’A’ and Kantoma showed increase in compressive stress as the cell pressure increased from 50 to 200 KN (Table 2). Cohesion and angle of internal friction for the samples were deduced from the Mohr failure envelop plot of shear stress versus normal stress (Fig. 4). Gyel’A’1 (70°) and Kantoma1 (75°) have higher undrained cohesion as well as angle of internal friction (15° and 13°) respectively than Gyel’A’C and KantomaC (Table 2; Fig. 4). From Eqs. 4 and 5, the shear strength (Table 2) was calculated using the cohesion and angle of internal friction deduced from Fig. 4. Kantoma C (18.344) and Gyel’A’C (16.26) samples have higher shear strength than Kantoma1 (12.657) and Gyel’A’1 (0.9935). This suggests that, the normal soils (KantomaC and Gyel’A’C) will be less prone to rutting by shear compared to mine tailings. The shear strength depends indirectly on the soil permeability and directly on the effective stress, grain size and gradation of the individual particles (Langfelder and Nivargikar 2020; Roy and Bhalla 2017). Thus, soils are essentially frictional materials comprised of individual particles that can slide and roll relative to one another. The quick undrain triaxial result is in agreement with that from particle size analysis which showed KantamaC as sample with highest percent of fine followed by Gyel’A’C, thus shear strength of the soil is directly proportional to its grain size distribution (Alias et al 2014). Cohesion and angle of internal friction are the two major shear strength parameters that are strongly affected by the particle size distribution of soil mixtures, the larger the sand particles the higher the friction, and vice versa (Fei 2017; Oleg and Yefimov 2019).

Oedometer test

Oedometer test result (Table 3) showed how the void ratio decreases with increase in effective stress from 10 to 800 KN/m2 and vice versa. Hence, graphs of final void ratio (e) and their corresponding effective stress were plotted on a semi-log scale to determine the magnitude of settlement (Fig. 5). As the effective stress increased from 25 to 200 kN, the coefficient of compressibility in all the samples were higher but reduces drastically after the effective stress increases above 200 kN. However, mine tailings have lower coefficient of compressibility than the normal soils. Compressibility increases as the proportion of small particles increases and becomes highest in fine-grained soils as observed in normal soils samples, Gyel’A’C and KantomaC. Considering the high percentage of coarse grain, permeability, poor to fair gradation, mine tailings, Gyel’A’1 and Kantoma1 will undergo immediate settlement than the other samples (Das 1994; Bhandry et al. 2017; Ubido et al. 2018). Also, in terms of suitability as a backfilling material, Gyel’A’1 and Kantoma1 are better preferred since they have higher permeability and lower compressibility than other samples (Buta et al. 2019).

Compaction test

Result from compaction test carried out on the tailings and soil samples is displayed in Table 4 and modeled in Fig. 6. Optimum moisture content (OMC) is in the range of 12–18% and maximum dry density (MDD), 1.646–1.788 kg/m3. Gyel’A’1 and Kantoma1 mine tailings have the least values of OMC, 12 and 14% but highest MDD values, 1.788 and 1.736 kg/m3, respectively, compared to the normal soil samples, Gyel’A’C and KantomaC. This falls in the range of well graded sand (12–14%) while Gyel’A’C and KantomaC fall in low plasticity silt–clay (17–18%) (Terzaghi and Peck 1967). From the Zero air voids (ZAV) line (Fig. 6), it can be clearly observed that increase in maximum dry density results to decrease in optimum moisture content.

Permeability

Based on permeability tests results displayed on Table 5, coefficient of permeability (Kav) of the Sn tailings Gyrl’A’1 and Kantoma1 were higher (9.205 × 10–6 and 9.015 × 10–6 cm/s) than the normal soil 7.756 × 10–6 and 8.024 × 10–6 cm/s. This suggests that the mine tailings have higher permeability than normal soils. The result coincides with the findings of Hu et al. (2017), who stated that metal tailings generally present larger hydraulic conductivity. Buta et al. (2019) stated that the use of tailing as backfilling material requires that it has high permeability, low compressibility, and the ability to dewater rapidly for stability reasons. Judging from the permeability values observed in the samples, the Sn mine tailings are more suitable as a backfilling material than ordinary soil.

Statistical modeling of the backfilling materials

The statistical analysis of the geotechnical parameters were modeled to assist in the determination of the most important parameters affecting the backfilling material.

Correlation analysis

Correlation analysis of the samples were modeled using the following geotechnical parameters; moisture content (M/c), void ratio (e), porosity (η), undrained cohesion (\(C)\), angle of internal friction (\(\varphi\)), optimum moisture content (OMC %), maximum dry density (MDD) and shear strength using the test results from Tables 2, 4 and 5. Statistically at two-tailed significant level of 0.01, very strong positive correlation exist between moisture content and void ratio (0.999), > OMC (0.996), > porosity (0.994), between void ratio and porosity (0.997), > OMC (0.993) while between porosity and OMC (0.984). For frictional angle (\(\varphi )\) and shear strength (0.978) at two-tailed significant level of 0.05, strong correlation exist as well (Table 6). Strong positive correlation existing between the parameters suggest that increase in one parameter increases the other one while in strong negative correlation, increase in one parameter especially undrained cohesion results to decrease in other parameters (Li and Aubertin 2008; Su and Liu 2014; Ukah et al. 2018). However, non significant strong positive correlation exist between void ratio and \(\varphi\) (0.924), > MDD (0.908), > Shear strength (0.825), between M/c and MDD (0.924), > \(\varphi\) (0.908), > shear strength (0.803), between porosity and \(\varphi\) (0.947), > MDD (0.876), > shear strength (0.860), between angle of internal friction and OMC (0.890), > MDD (0.688), and non-significant strong negative correlation exists between undrained cohesion, \(\complement\) and the rest of the parameters (Table 6). OMC exhibit very strong non significant correlation with (MDD). Non-significant strong correlation exists between shear strength and all the parameters except angle of internal friction (\(\varphi )\). This correlation result suggests that shear strength is a function of frictional angle, and soil stability is based on soil strength (Watkins et al. 2010). Non-significant correlation shows how unimportant the relationships existing between some of the parameters are (Swan and Sandilands 1995).

Principal component analysis (PCA)

Principal component analysis (PCA) was modeled from Tables 2, 4 and 5 into a dimension reduction factor analysis extraction using the method of principal component analysis of Varimax with Kaiser Normalization rotation method converged in three iterations rotations (Kaiser 1960). In this study, factor loadings less than, < 0.5 are considered as low (insignificant), 0.5–0.75 as medium and above, > 0.75 as high. PCA replicate the correlation matrix shown in Table 6 using a set of components that are fewer in number and linear combinations of the original set of items (Tables 7, 8). PCA also transforms a number of (possibly) correlated variables into a (smaller) number of uncorrelated variables called principal components (Legendre and Legendre 1998; Swan and Sandilands 1995). The total variance explained table (Table 7), gives the total variance explained by each component. The first component PCA1 is 6.615, or (6.615/8) % = 82.691% and this accounts for 82.691% of the total variance, while the second component (PCA 2) account for 16.272% of the total variance. Because the same number of components was extracted as the number of items, the initial eigenvalues column is the same as the extraction sums of squared loadings column. Hence, the first component (PCA 1) explains the most variance, and the last component explains the least (Burstyn 2004). The PCA1 is dominated by moisture content, void ratio, porosity, angle of internal friction, optimum moisture content, maximum dry density and shear strength indicating very strong positive loadings, while undrained cohesion showed moderate negative loading (Table 8). This is graphically shown in component plot in rotated space (Fig. 7). Figure 7 gives a visual representation of the loadings plotted in a two-dimensional space. The plot shows how closely related the items are to each other and to the two components. PCA2 is dominated by moderately to strongly positive significant loading of MDD (0.533) and undrained cohesion (0.828). Positively significant loading component shows increase in one result is an increase in the other. Negative loadings indicate a negative correlation and shows increase in one result results to decrease in the other. Communality is total amount of variance an original variable shares with all other variables included in the analysis. According to Table 8, the total variance is 1, meaning that all variables contributed equally to that principal component. The statistical analysis modeled with PCA has so far revealed that apart from permeability and particle size distribution; moisture content, void ratio, porosity, angle of internal friction, optimum moisture content, maximum dry density and shear strength are the dominant parameters that determines the suitability of the choice of backfilling material unlike undrain cohesion (Roy and Bhalla 2017).

Conclusion

The modeling and characterization of mine pit backfilling/reclamation materials carried out around Gyel’A’ and Kantoma areas of Plateau state using integrated method of geotechnics and statistics revealed that;

  • Based on particle size analysis, normal soil samples (Gyel’A’C and KantomaC) have more percentage of fines and are more well graded than the mine tailings (Gyel’A’1 and Kantoma1).

  • Quick undrained triaxial test result from the normal soil samples (kantoma C (18.344) and Gyel’A’C (16.26)) have higher shear strength than the tailings Kantoma1 (12.657) and Gyel’A’1 (0.9935). This suggests that, the normal soils (KantomaC and Gyel’A’C) will be less prone to rutting by shear compared to mine tailings.

  • Mine tailings have lower coefficient of compressibility than the normal soils. Compressibility increases as the proportion of small particles increases and becomes highest in fine-grained soils as observed in normal soils samples, Gyel’A’C and KantomaC.

  • Permeability tests results of the Sn tailings Gyrl’A’1 and Kantoma1 were higher (9.205 × 10–6 and 9.015 × 10–6 cm/s) than the normal soil 7.756 × 10–6 and 8.024 × 10–6 cm/s. This suggests that the mine tailings are more permeable than normal soils.

  • Statistically, strong positive correlation exist between the geotechnical parameters moisture content (M/c), void ratio (e), porosity (η), undrained cohesion (\(C)\), angle of internal friction, optimum moisture content (OMC %), maximum dry density (MDD) and shear strength. This suggests that increase in one parameter increases the other and vice versa. Non significant strong correlation exists between shear strength and all the parameters except angle of internal friction, indicating that shear strength is a function of frictional angle. PCA revealed that apart from permeability and particle size distribution; moisture content, void ratio, porosity, angle of internal friction, optimum moisture content, maximum dry density and shear strength are the dominant parameters that determines the suitability of the choice of backfilling material unlike undrain cohesion

Furthermore, considering the high percentage of coarse grain, permeability, poor to fair gradation, mine tailings, Gyel’A’1 and Kantoma1 will undergo immediate settlement than normal soil. But in terms of suitability as a backfilling material, Gyel’A’1 and Kantoma1 are better preferred since they have higher permeability and lower compressibility than normal soil samples.