- Email: [email protected]
Abrasion wear resistance of wall lining materials in bins and chutes during iron ore mining
International Journal of Mineral Processing 167 (2017) 42–48
Contents lists available at ScienceDirect
International Journal of Mineral Processing journal homepage: www.elsevier.com/locate/ijminpro
Abrasion wear resistance of wall lining materials in bins and chutes during iron ore mining Wei Chen ⁎, Subhankar Biswas, Alan Roberts, Jayne O'Shea, Kenneth Williams Centre for Bulk Solids and Particulate Technologies, The University of Newcastle, Callaghan, Australia
a r t i c l e
i n f o
Article history: Received 31 January 2017 Received in revised form 2 August 2017 Accepted 14 August 2017 Available online 15 August 2017 Keywords: Abrasion wear Iron ore Material handling equipment Lining materials
a b s t r a c t The wear problem on the internal lining of bins and chutes needs to be addressed before any significant efficiency gains during iron ore mining operations. This study aims to investigate the factors determining the wear resistance of common lining materials used in iron ore mining operations. A purposely designed experimental system was utilised to quantitatively assess the wear resistance of a suite of wall lining materials against an iron ore abrading medium, from which a wear rate for each liner is determined. Results suggested that the hardness of a lining material can be utilised to indicate its abrasion wear resistance. From experimental results, prediction models of the service life of selected lining materials in bins and chutes are also obtained. © 2017 Elsevier B.V. All rights reserved.
1. Introduction The increasing global demand for iron ore is placing greater emphasis on the need for higher mining production capacity coupled with the objectives of more efficient, low cost operations. Directly associated with these objectives, is the need to increase the service life, combined with reducing maintenance costs linked to the various components in the mining, conveying and handling chain. Of the various parameters affecting service life, there is none more significant than the need for equipment wear reduction and extended life performance. These remarks refer to such items as conveyors, feeders, bins and chutes, for which the correct choice of wear lining materials is of paramount importance (Cortie and McEwan, 1996). This is the motivation for work presented in this paper. There are two principal types of wear mechanisms associated with bins and chutes in iron ore mining, namely, the impact wear and abrasion wear (Roberts et al., 1989; Tuckey, 2003). Impact wear occurs when bulk material exhibits oblique to normal angular contacts with the wall lining materials at relatively high speeds (Scott and Choules, 1993). Such impacts often lead to localised fracture or chipping on the lining surface, particularly when the impact angles are high and near 90°. At lower impact angles, gouging type impact wear is more evident. Wear is especially pronounced when the iron ore particles are highly angular in shape (Biswas et al., 2014; Cenna et al., 2011). Abrasive wear is caused by the prolonged frictional rubbing on the surface of the lining
material (Hutchings, 1993; Wiche et al., 2005). Whereas impact wear is localised, abrasive wear is more uniformly distributed and, often, is less severe than impact wear. Appropriate design of the material flow pattern within plants is able to minimise or even eliminate the undesirable localised impact wear by creating the more desirable and efficient streamline flow. This is illustrated in the transfer chute shown in Fig. 1 (a), in which the curved hood component effectively guides the flow. Previous studies have reported that the normal pressure and the frictional velocity of an abrading medium applied on the surface of a liner are linearly proportional to its wear rate (Roberts and Wiche, 1993). For a given lining material, varying the normal pressure and/or the flow velocity can improve the liner's wear performance. More importantly, appropriate choice of lining material can significantly enhance the wear resistance of a plant. Various types of wear lining materials, such as ceramics and metals, have been developed with different physical, chemical and metallurgical properties aimed at minimising the abrasion wear. However, it is difficult to predict the wear performance based on the fundamental properties of a particular lining material, and experimental testing is often required. The purpose of this study is to experimentally investigate the abrasive wear performance of a suite of wear lining materials against iron ore, from which wear performance can be correlated to the fundamental liner characteristics. By this means, the abrasive wear resistance of a lining material can be simply indicated by its material properties. 2. Lining material selection and characterisation
⁎ Corresponding author at: Centre for Bulk Solids and Particulate Technologies, Newcastle Institute for Energy and Resources, The University of Newcastle, Callaghan 2308, Australia. E-mail address: [email protected] (W. Chen).
http://dx.doi.org/10.1016/j.minpro.2017.08.002 0301-7516/© 2017 Elsevier B.V. All rights reserved.
A total of nine lining materials (shown in Table 1) which are commonly used as internal wear liners during iron ore operations were selected in this study, including four ceramic materials and five metal
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
43
Fig. 1. Abrasive wear in typical chutes and bins during iron ore mining operation.
linings. Out of the five metal lining materials, one mild steel liner was used for comparative purposes. For each lining material, the additional material properties are characterised below.
curve. This portion is presented as a new graph with the X-axis extending in the same direction as the average line and the Y-axis representing the magnitude, as shown in Fig. 2. The roughness curve is represented by y= f(x).
2.1. Energy dispersive X-Ray spectroscopy (EDS) The energy dispersive X-Ray spectroscopy (EDS) is a quantitative elemental analysis technique. The measurement principle is based on the interactions between an X-Ray and the atom structure of a particular material (Shindo and Oikawa, 2002). Results are presented in Table 2 below for all liners. Among the four ceramic liners, C is a silica carbide (SiC) type of ceramic lining. A, B and D are typical technical ceramics composed of alumina (Al2O3) and carbon (C) for reinforcement. There is also additional platinum (Pt) added into B and D lining materials for consistent mechanical strength at high temperatures. The metal type lining materials E–H contained chromium (Cr) for additional strength. There is small percent of niobium (Nb) formulated into lining material E to enhanced the hardness. Liner X is a typical carbon steel. 2.2. Surface roughness Surface roughness is calculated as an arithmetical average roughness (Ra) using a surface roughness tester. Arithmetical average roughness (Ra) is determined from a portion stretching over a reference length in the direction in which an average line is cut out of the roughness
Ra ¼
1 L
Z 0
L
j f ðxÞjdx
Ten measurements were carried out randomly on each lining surface in its as-manufactured condition. The test was performed in accordance to AS2382 (Australian Standard, 1981). Test results for all liners are shown in Table 3.
2.3. Knoop hardness Thirdly, the hardness of the lining materials is characterised using a Knoop hardness test following the ASTM Standard (ASTM, 2012). The Knoop hardness test is suitable for both the metals and ceramic materials. During a test, a pyramidal diamond with a pre-determined geometry was pressed into the surface of a liner sample with a known load for a specific dwell time, the resulting indentation area left on the surface indicated the hardness of lining material. Table 3 summarises the Knoop hardness results of all lining materials selected.
Table 2 Chemical elemental analysis results for all liners using EDS.
Table 1 Selected lining material label, material type and density. Liner label
Material type
Density – kg/m3
A B C D E F G H X
Ceramic Ceramic Ceramic Ceramic Metal Metal Metal Metal Mild Steel
3640 3610 3050 4020 7600 7400 7400 7240 7850
ð1Þ
44
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
Fig. 2. Schematic of the measurement principle for the surface roughness. Table 3 Surface roughness and Knoop hardness test results for all selected lining materials. Liner label
Surface roughness (Ra) - μm
Knoop hardness - HK
A B C D E F G H X
1.3 1.5 1.6 1.6 4.5 6.1 1.1 5.8 3.9
1092 1247 2039 1364 953 749 763 728 132
3. Experimental abrasion wear testing 3.1. Test equipment and procedure The purposely designed experimental system employed in this study is shown in Fig. 3. The test facility incorporates the following components: a) A rotating annular bed with the following features: i. A 200 mm wide by 200 mm deep annulus. ii. Continuous transportation of the material bed underneath four ‘stationary’ wall lining test samples. By inclining the wear sample with a small angle θ (0.5–2.5°) with respect to the constrained bed, the sample is able to plane over the surface of the wear media, as shown in Fig. 4. With careful adjustment of the plane angle (θ), even wear over the entire surface of the wear sample is achieved. iii. Significant quantity of the material storage to distribute degradation and heat generated as the annular bed passes under the sample.
iv. A bulk solids moisture monitoring system to ensure the consistency of the sample moisture during testing by dripping waters into the abrading agent. b) A counterweighted load-arm allowing moderate loads to be applied to the test sample. This is dependent on the normal stress of the wear media in relation to the sample/media friction. Other features include: i. The capability to pivot at the shear plane. Only one degree of freedom. ii. Weights placed on the sample holder to provide the total normal load. iii. A wall sample holding clamp that allows easy placement and removal of a standard sample size. iv. Plane angle adjustment for the wall sample. v. Each sample holder can accommodate a wear liner with a surface area of 100 mm by 100 mm, and up to 50 mm in thickness c) A plough and grading mixing system to ensure that the wear media is completely remixed. The plough's vertical position is adjustable to suit the wear media selected. The aim is to present a fresh surface of bulk solids to the lining material for each contact cycle. d) A variable speed hydraulic drive to allow for testing at different velocities. This unit provides all the mechanical power to the system. e) A passive consolidator to increase the wear media's bulk density before its presentation to the wall sample material. This reduces the plane angle required. After placement of a sufficient quantity of the bulk material as the wear media in the circular annular channel, the turntable is set to run at the required speed with the plough set at least 3 particle sizes above the bottom of the trench. It is noted that the grader blade should only level the bulk solid. When clamping each sample in place, the nosepiece of each sample holder is adjusted according to the sample thickness to ensure the full abrasion wear contact on the liner surface. The pivoting load arm assembly is then counterbalanced. The nominated normal test load is then applied. It needs to be noted that the sample plane angle should be adjusted to allow the radiused nose-piece to be in line with the approaching bed surface. For instance, if the test sample plane is below the bed surface, the angle is too fine. If only the rear half of the sample planes on the bed surface, the angle is too coarse. Correct adjustment is vital so that the sample wears evenly. 3.2. Abrasion wear test program At the beginning of each test, the fresh iron ore sample was deposited in the annular channel. Four wear liner samples were then installed
Fig. 3. The abrasion wear testing system.
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
45
Fig. 4. Schematic of the liner setup for rubbing against the abrading agent.
on four load arms under a nominated normal load, after which the load arms were placed in the centre of the annulus in contact with the iron ore material. The tester was set in motion with the platform rotating at a nominated velocity. The tests were conducted with the moisture monitoring system activated to ensure a relative consistent moisture throughout the test. Every 5 h, tests were interrupted to allow liner samples to be removed and cleaned. The corresponding sample weight loss was measured. The samples were then re-installed and the tests continued. A total of 40 h of abrading time was performed for each sample to complete a test. As discussed above, a total of 9 different lining materials were selected for this study. Each test material was prepared to have a 100 mm by 100 mm surface area. The thickness of the samples ranged from 20 mm–40 mm. Noting that each suite of tests can only accommodate a maximum of 4 liner test samples. Therefore, three separate wear test suites were performed to complete the test program. Each suite of testing was repeated three times with new liner samples. The mild steel liner was used in each test suit as a benchmarking lining. Fig. 5 shows the lining sample arrangements in all three suites. The selection of the normal pressure applied to the liners and the velocity of the abrading medium in the experiment are deemed to reflect the operational conditions in material handling plants during iron ore mining. In material storage bins, the abrasive wear tends to concentrate at the hopper discharge section; whereas in chutes, abrasive wear occurs along the material flow path (Corder and Thorpe, 1987; Corder and Thorpe, 1995; Roberts, 1988). In both cases, the iron ore abrading medium exhibited relatively fast flow velocity under small normal consolidation pressure. Therefore, a linear speed of 1 m/s for the iron ore material and a normal load of 2 kPa was applied to the lining samples in all tests. A typical iron ore fines product was utilised as the abrading agent in this study. The bulk material properties of the sample, including the particle size distribution, the particle density and the moisture content are shown in Fig. 6.
Fig. 6. Particle size distribution of the used iron ore abrading agent.
4. Results and discussion 4.1. Abrasive wear resistance ranking After the completion of three suites of testing, the weight loss in grams of each liner due to abrasive wear for a total testing duration of 40 h was obtained. However, it is more useful to compare the abrasive wear resistance by normalising the weight loss to liner thickness. Although the liner thickness may be directly measured using techniques such as a vernier, the measurement may not be able to account for the deep surficial cut on the ductile metal linings and the localised chipping on the brittle ceramic liners, which leads to underestimation of the abrasion wear. Based on the forgoing comments, the weight loss results are subsequently converted to thickness loss, in microns, with the following expression: Thickness Loss ¼
M 103 Aρ
ð2Þ
where M is mass loss in grams, A is the contact surface area and ρ is the test liner sample density. The contact surface area was fixed to be 0.01 m2 for each sample. Results of the thickness loss for all liners are shown in Fig. 7. Fig. 7 (a) shows that the thickness loss results of mild steel liners are consistent for the three separate testing suites, indicating the abrasive wear resistance can be directly compared across different testing suites. The thickness loss for the mild steel liners was observed to be generally higher than other lining materials, indicating significantly lower abrasive wear resistance. Among linings A–H, metal type of lining materials generally exhibited higher thickness loss, thus lower abrasion wear
Fig. 5. Wall linings arrangements in three separate test suites.
46
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48 Table 4 Abrasive wear rate for each lining material through linear regression of the wear testing results. Ranking
Liner label
Abrasive wear rate - Ψ [μm/h]
1 2 3 4 5 6 7 8 9
C D A E B G F H X1 X2 X3 X4 Average of X
0.099 ± 0.02 0.352 ± 0.05 0.458 ± 0.04 0.527 ± 0.05 0.603 ± 0.04 0.942 ± 0.06 1.090 ± 0.05 1.281 ± 0.07 11.17 ± 1.3 9.83 ± 1.4 9.99 ± 1.2 10.63 ± 1.6 10.40 ± 1.4
the lining to enhance its mechanical strength, which is also indicated from the Knoop hardness results. Among the three other metal linings, no distinct abrasive wear resistance is observed between Linings F and G. However, Lining H exhibits the lowest wear resistance among all metal linings, which may be due to its lower chromium (Cr) contents. Furthermore, as shown in Fig. 8, the Knoop hardness value is observed to be proportional to the logarithmic abrasion wear rate. This is suggesting that the abrasive wear resistance of a specific lining material can be simply indicated by its Knoop hardness value. Practically, when selecting lining materials in bins and chutes to handle iron ore against abrasive wear, the Knoop hardness test can be used to predict its abrasion wear performance. Lastly, the surface roughness exhibits no obvious correlation with the ranking of the abrasive wear rate. Fig. 7. Thickness loss results for each liner based on Eq. (2).
resistance compared to the ceramic linings. Nevertheless, metal lining sample E showed higher abrasive wear resistance compared to ceramic lining sample B. Additionally, all results exhibited a quasi-linear relationship between the thickness loss and the testing duration, from which an abrasive wear rate - Ψ (μm/h) for each lining material can be subsequently defined. The abrasive wear rate represents a simple indication of the abrasive wear resistance of a lining material. Table 4 showed the abrasive wear rate for each lining material and its corresponding ranking based on a linear regression approach. 4.2. Abrasive wear resistance correlation with material properties Ranking of the abrasive wear resistance is predominately determined by the lining material properties. The lining material properties, including the surface roughness and hardness, vary as the abrasive wear propagates into the material from frictional rubbing and localised heat generation. Therefore, it is more useful to predict the abrasive wear resistance of a particulate lining material based on its material properties at its as-fabricated condition, which is investigated below. Initially, based on the chemical compositions of the lining material, ceramic types of linings generally outperformed the metal liners in terms of abrasive wear resistance. Lining C, the silica carbide (SiC) type of ceramic is suggested to be more abrasive wear resistant comparing to the alumina (Al2O3) type of ceramics (A, B and D). Within the three alumina type ceramics, Lining D, with the additional zirconium dioxide (ZrO2) trace, exhibits higher abrasive wear resistance when comparing to Linings A and B. Lining B shows the lowest abrasion resistance among ceramics, which may be due to lower alumina content. Additionally, no enhancement on abrasive wear resistance is observed by adding the platinum (Pt) into the linings. It is also interesting to observe that the complex metal lining material E shows comparable abrasive wear resistance to the ceramic lining material B. This may be due to the small percent of niobium (Nb) formulated into
4.3. Service life predictions in chutes and bins Abrasive wear is assumed to be a function of the normal pressure, the rubbing or sliding velocity at the boundary and the friction coefficient. From research detailed in (Roberts and Wiche, 1993), it is reasonable to assume that abrasive wear is a linear function of normal pressure and rubbing velocity. Abrasive wear, Wa, expressed in units of thickness loss per unit time (μm/s) may be defined as follows: Wa ¼
σ W VS tanϕ ½μm=s σ Wp
ð3Þ
where • σW is the normal pressure at the boundary (kPa) • VS is the velocity of bulk solid adjacent to the boundary (m/s)
Fig. 8. Correlation between the abrasive wear rate and the Knoop hardness across all liners.
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48 Table 5 Service life prediction results for the example chute based on the wear testing results. Lining label
Wear rate in example chute [μm/h]
Service life – [h]
A B C D E F G H X
3.18 4.19 0.69 2.44 3.66 7.57 6.54 8.90 72.2
9432 7164 43,636 12,273 8197 3963 4586 3372 415
the service life of each lining material at other normal pressures and rubbing velocities in chute and bin applications (Roberts and Wiche, 1993). Examples are presented below. Firstly, for a straight, parallel transfer chute processing the iron ore product described in this study, the following typical geometrical and operational conditions are assumed: • Chute width, w • Tonnage, M • Sliding velocity, v • Wall friction angle (all linings), ϕ • Lining thickness (all linings), d • Bulk density, ρ
σW ¼ The abrasive wear parameter, σWp, can be established for the wear samples using Eq. (3) and the results obtained from the abrasion wear testing. Essentially, Eq. (3) may be transformed to the following form: σ W VS tanϕ Wa
½kPa
ð4Þ
In the abrasion wear tests, the normal pressure σW was set at 2 kPa, and the relative velocity of the material to the liner VS was 1 m/s. Providing a fixed wall friction angle ϕ, the abrasion wear parameter σWp is obtained below for a particular liner by utilising the abrasion Ψ ) as shown in Table 4. wear rate (Wa ¼ 3600 • σWp ~15,720 tan ϕ (106 kPa) • σWp ~11,940 tan ϕ (106 kPa) • σWp ~72,727 tan ϕ (106 kPa) • σWp ~20,454 tan ϕ (106 kPa) • σWp ~13,662 tan ϕ (106 kPa) • σWp ~6605 tan ϕ (106 kPa) • σWp ~7643 tan ϕ (106 kPa) • σWp ~5620 tan ϕ (106 kPa) • σWp ~692 tan ϕ (106 kPa)
=2 m =10,000 t/h =5 m/s =25° =30 mm =2500 kg/m3
The normal pressure acting on the lining surface was estimated as,
• ϕ is the wall friction angle • σWp is the abrasive wear parameter (106 kPa)
σ Wp ¼
47
Liner A Liner B Liner C Liner D Liner E Liner F Liner G Liner H Liner X
Having determined the abrasive wear factor for a particular lining material against this iron ore product, it is now possible to estimate
5 Mg ¼ 2:8 ½kPa 18 wv
ð5Þ
The wear rate and life span of the different lining material can be estimated using Eq. (2) and abrasive wear parameter derived from the wear testing. Results are shown in Table 5. Secondly, for an axisymmetric mass flow bin (shown in Fig. 9 (a)) discharging the iron ore product, the follow typical geometrical and operational conditions are assumed: • Diameter, D • Outlet diameter, B • Bin height, Hb • Hopper height, Hh • Hopper half angle, α • Effective internal frictional angle, δ • Bulk density, ρ • Discharge rate, M • Wall friction angle (all linings), ϕ • Lining thickness (all linings), d
=12 m =1.5 m =24 m =12.37 m =23° =50° =2500 kg/m3 =2500 t/h =25° =30 mm
The normal wall pressures and the sliding velocity of the material during symmetrical discharge are calculated according to the Australian Standard – loads on bulk solids containers (Australian Standard, 1996). Results are shown in Fig. 9. The normal pressure on the bin wall increases from the top towards the transition from the vertical section to the hopper section, at which the highest normal wall pressure is indicated. From the hopper transition section to the discharge opening, the normal wall pressure continuously decreases. In the meantime, the
Fig. 9. Normal wall pressure and sliding velocity during discharging in the example bin.
48
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
Fig. 10. Predicted service life along the height of the example bin for each lining material selected in this study.
material sliding velocity continuously increases from the top to the discharge opening of the bin. Based upon the normal wall pressure and the sliding velocity of the material, the wear rate and the service life of wall linings within the example bin are estimated. Results of the predicted service life are shown in Fig. 10. From the results, it is indicated that the transition from vertical section to the hopper section and the outlet region exhibits higher abrasive wear within the bin. Therefore, the abrasive wear lining design and installation strategy should reflect the localised abrasion wear rate in order to extend the service duration of the bin. 5. Conclusion A comprehensive study was performed on the abrasive wear resistance of wall lining materials in bins and chutes during iron ore mining operations. A suite of experimental investigations and modelling on predicting the service life of these wall linings were conducted. This study yielded the following major findings: • A comparative abrasive wear testing system was developed and was capable of obtaining the abrasive wear rate for a particular lining from a selected abrading agent. • The Knoop hardness test can be utilised to indicate the ranking of abrasive wear resistance for various lining materials. • The lining material service life prediction model developed in this study can be applied to bins and chutes for wear pattern predictions under various geometrical and operational conditions.
Consequently, the outcome of this study provided a technical guide for selecting lining materials to be installed in chutes and bins during
iron ore mining for reduced abrasive wear and more accurate maintenance planning. References ASTM, A., 2012. E384: Standard Test Method for Knoop and Vickers Hardness of Materials. ASTM Stand, pp. 1–43. Australian Standard, 1981. AS 2382-1981 Surface Roughness Comparison Specimens. Australian Standard, 1996. AS 3774-1996 Loads on Bulk Solids Containers. Biswas, S., Cenna, A., Williams, K., Jones, M., 2014. Subsurface behavior of ductile material by particle impacts and its influence on wear mechanism. Proc. Eng. 90, 160–165. Cenna, A.A., Williams, K.C., Jones, M.G., 2011. Analysis of impact energy factors in ductile materials using single particle impact tests on gas gun. Tribol. Int. 44 (12), 1920–1925. Corder, G.D., Thorpe, R.B., 1987. Wear of Hopper Walls. Tribology in Particulate Technology. 1987. Adam Hilger. Corder, G.D., Thorpe, R.B., 1995. An experimental study of the wear at hopper walls. KONA Powder Part. J. 13 (0), 105–112. Cortie, D.P., McEwan, J.J., 1996. Materials selection in the mining industry: old issues and new challenges. J. South. Afr. Inst. Min. Metall. 96 (4), 145–157. Hutchings, I.M., 1993. Mechanisms of wear in powder technology: a review. Powder Technol. 76 (1), 3–13. Roberts, A.W., 1988. Friction adhesion and wear in bulk materials handling. Proc. Antiwear 88. The Royal Society London. Roberts, A.W., Wiche, S.J., 1993. Prediction of lining wear life of bins and chutes in bulk solids handling operations. Tribol. Int. 26 (5), 345–351. Roberts, A.W., Ooms, M., Wiche, S.J., 1989. Concepts of boundary friction, adhesion and wear in bulk solids handling operations. Third International Conference on Bulk Materials, Storage, Handling and Transportation: Preprints of Papers. Institution of Engineers, Australia, p. 349. Scott, O.J., Choules, P.R., 1993. The use of impact plates in conveyor transfers. Tribol. Int. 26 (5), 353–359. Shindo, D., Oikawa, T., 2002. Energy dispersive X-ray spectroscopy. Analytical Electron Microscopy for Materials Science. Springer, pp. 81–102. Tuckey, K.R.G., 2003. Intelligent selection of engineered wear linings in iron ore plant. Miner. Process. Ext. Metall. 112 (1), 33–38. Wiche, S.J., Keys, S., Roberts, A.W., 2005. Abrasion wear tester for bulk solids handling applications. Wear 258 (1), 251–257.
Contents lists available at ScienceDirect
International Journal of Mineral Processing journal homepage: www.elsevier.com/locate/ijminpro
Abrasion wear resistance of wall lining materials in bins and chutes during iron ore mining Wei Chen ⁎, Subhankar Biswas, Alan Roberts, Jayne O'Shea, Kenneth Williams Centre for Bulk Solids and Particulate Technologies, The University of Newcastle, Callaghan, Australia
a r t i c l e
i n f o
Article history: Received 31 January 2017 Received in revised form 2 August 2017 Accepted 14 August 2017 Available online 15 August 2017 Keywords: Abrasion wear Iron ore Material handling equipment Lining materials
a b s t r a c t The wear problem on the internal lining of bins and chutes needs to be addressed before any significant efficiency gains during iron ore mining operations. This study aims to investigate the factors determining the wear resistance of common lining materials used in iron ore mining operations. A purposely designed experimental system was utilised to quantitatively assess the wear resistance of a suite of wall lining materials against an iron ore abrading medium, from which a wear rate for each liner is determined. Results suggested that the hardness of a lining material can be utilised to indicate its abrasion wear resistance. From experimental results, prediction models of the service life of selected lining materials in bins and chutes are also obtained. © 2017 Elsevier B.V. All rights reserved.
1. Introduction The increasing global demand for iron ore is placing greater emphasis on the need for higher mining production capacity coupled with the objectives of more efficient, low cost operations. Directly associated with these objectives, is the need to increase the service life, combined with reducing maintenance costs linked to the various components in the mining, conveying and handling chain. Of the various parameters affecting service life, there is none more significant than the need for equipment wear reduction and extended life performance. These remarks refer to such items as conveyors, feeders, bins and chutes, for which the correct choice of wear lining materials is of paramount importance (Cortie and McEwan, 1996). This is the motivation for work presented in this paper. There are two principal types of wear mechanisms associated with bins and chutes in iron ore mining, namely, the impact wear and abrasion wear (Roberts et al., 1989; Tuckey, 2003). Impact wear occurs when bulk material exhibits oblique to normal angular contacts with the wall lining materials at relatively high speeds (Scott and Choules, 1993). Such impacts often lead to localised fracture or chipping on the lining surface, particularly when the impact angles are high and near 90°. At lower impact angles, gouging type impact wear is more evident. Wear is especially pronounced when the iron ore particles are highly angular in shape (Biswas et al., 2014; Cenna et al., 2011). Abrasive wear is caused by the prolonged frictional rubbing on the surface of the lining
material (Hutchings, 1993; Wiche et al., 2005). Whereas impact wear is localised, abrasive wear is more uniformly distributed and, often, is less severe than impact wear. Appropriate design of the material flow pattern within plants is able to minimise or even eliminate the undesirable localised impact wear by creating the more desirable and efficient streamline flow. This is illustrated in the transfer chute shown in Fig. 1 (a), in which the curved hood component effectively guides the flow. Previous studies have reported that the normal pressure and the frictional velocity of an abrading medium applied on the surface of a liner are linearly proportional to its wear rate (Roberts and Wiche, 1993). For a given lining material, varying the normal pressure and/or the flow velocity can improve the liner's wear performance. More importantly, appropriate choice of lining material can significantly enhance the wear resistance of a plant. Various types of wear lining materials, such as ceramics and metals, have been developed with different physical, chemical and metallurgical properties aimed at minimising the abrasion wear. However, it is difficult to predict the wear performance based on the fundamental properties of a particular lining material, and experimental testing is often required. The purpose of this study is to experimentally investigate the abrasive wear performance of a suite of wear lining materials against iron ore, from which wear performance can be correlated to the fundamental liner characteristics. By this means, the abrasive wear resistance of a lining material can be simply indicated by its material properties. 2. Lining material selection and characterisation
⁎ Corresponding author at: Centre for Bulk Solids and Particulate Technologies, Newcastle Institute for Energy and Resources, The University of Newcastle, Callaghan 2308, Australia. E-mail address: [email protected] (W. Chen).
http://dx.doi.org/10.1016/j.minpro.2017.08.002 0301-7516/© 2017 Elsevier B.V. All rights reserved.
A total of nine lining materials (shown in Table 1) which are commonly used as internal wear liners during iron ore operations were selected in this study, including four ceramic materials and five metal
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
43
Fig. 1. Abrasive wear in typical chutes and bins during iron ore mining operation.
linings. Out of the five metal lining materials, one mild steel liner was used for comparative purposes. For each lining material, the additional material properties are characterised below.
curve. This portion is presented as a new graph with the X-axis extending in the same direction as the average line and the Y-axis representing the magnitude, as shown in Fig. 2. The roughness curve is represented by y= f(x).
2.1. Energy dispersive X-Ray spectroscopy (EDS) The energy dispersive X-Ray spectroscopy (EDS) is a quantitative elemental analysis technique. The measurement principle is based on the interactions between an X-Ray and the atom structure of a particular material (Shindo and Oikawa, 2002). Results are presented in Table 2 below for all liners. Among the four ceramic liners, C is a silica carbide (SiC) type of ceramic lining. A, B and D are typical technical ceramics composed of alumina (Al2O3) and carbon (C) for reinforcement. There is also additional platinum (Pt) added into B and D lining materials for consistent mechanical strength at high temperatures. The metal type lining materials E–H contained chromium (Cr) for additional strength. There is small percent of niobium (Nb) formulated into lining material E to enhanced the hardness. Liner X is a typical carbon steel. 2.2. Surface roughness Surface roughness is calculated as an arithmetical average roughness (Ra) using a surface roughness tester. Arithmetical average roughness (Ra) is determined from a portion stretching over a reference length in the direction in which an average line is cut out of the roughness
Ra ¼
1 L
Z 0
L
j f ðxÞjdx
Ten measurements were carried out randomly on each lining surface in its as-manufactured condition. The test was performed in accordance to AS2382 (Australian Standard, 1981). Test results for all liners are shown in Table 3.
2.3. Knoop hardness Thirdly, the hardness of the lining materials is characterised using a Knoop hardness test following the ASTM Standard (ASTM, 2012). The Knoop hardness test is suitable for both the metals and ceramic materials. During a test, a pyramidal diamond with a pre-determined geometry was pressed into the surface of a liner sample with a known load for a specific dwell time, the resulting indentation area left on the surface indicated the hardness of lining material. Table 3 summarises the Knoop hardness results of all lining materials selected.
Table 2 Chemical elemental analysis results for all liners using EDS.
Table 1 Selected lining material label, material type and density. Liner label
Material type
Density – kg/m3
A B C D E F G H X
Ceramic Ceramic Ceramic Ceramic Metal Metal Metal Metal Mild Steel
3640 3610 3050 4020 7600 7400 7400 7240 7850
ð1Þ
44
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
Fig. 2. Schematic of the measurement principle for the surface roughness. Table 3 Surface roughness and Knoop hardness test results for all selected lining materials. Liner label
Surface roughness (Ra) - μm
Knoop hardness - HK
A B C D E F G H X
1.3 1.5 1.6 1.6 4.5 6.1 1.1 5.8 3.9
1092 1247 2039 1364 953 749 763 728 132
3. Experimental abrasion wear testing 3.1. Test equipment and procedure The purposely designed experimental system employed in this study is shown in Fig. 3. The test facility incorporates the following components: a) A rotating annular bed with the following features: i. A 200 mm wide by 200 mm deep annulus. ii. Continuous transportation of the material bed underneath four ‘stationary’ wall lining test samples. By inclining the wear sample with a small angle θ (0.5–2.5°) with respect to the constrained bed, the sample is able to plane over the surface of the wear media, as shown in Fig. 4. With careful adjustment of the plane angle (θ), even wear over the entire surface of the wear sample is achieved. iii. Significant quantity of the material storage to distribute degradation and heat generated as the annular bed passes under the sample.
iv. A bulk solids moisture monitoring system to ensure the consistency of the sample moisture during testing by dripping waters into the abrading agent. b) A counterweighted load-arm allowing moderate loads to be applied to the test sample. This is dependent on the normal stress of the wear media in relation to the sample/media friction. Other features include: i. The capability to pivot at the shear plane. Only one degree of freedom. ii. Weights placed on the sample holder to provide the total normal load. iii. A wall sample holding clamp that allows easy placement and removal of a standard sample size. iv. Plane angle adjustment for the wall sample. v. Each sample holder can accommodate a wear liner with a surface area of 100 mm by 100 mm, and up to 50 mm in thickness c) A plough and grading mixing system to ensure that the wear media is completely remixed. The plough's vertical position is adjustable to suit the wear media selected. The aim is to present a fresh surface of bulk solids to the lining material for each contact cycle. d) A variable speed hydraulic drive to allow for testing at different velocities. This unit provides all the mechanical power to the system. e) A passive consolidator to increase the wear media's bulk density before its presentation to the wall sample material. This reduces the plane angle required. After placement of a sufficient quantity of the bulk material as the wear media in the circular annular channel, the turntable is set to run at the required speed with the plough set at least 3 particle sizes above the bottom of the trench. It is noted that the grader blade should only level the bulk solid. When clamping each sample in place, the nosepiece of each sample holder is adjusted according to the sample thickness to ensure the full abrasion wear contact on the liner surface. The pivoting load arm assembly is then counterbalanced. The nominated normal test load is then applied. It needs to be noted that the sample plane angle should be adjusted to allow the radiused nose-piece to be in line with the approaching bed surface. For instance, if the test sample plane is below the bed surface, the angle is too fine. If only the rear half of the sample planes on the bed surface, the angle is too coarse. Correct adjustment is vital so that the sample wears evenly. 3.2. Abrasion wear test program At the beginning of each test, the fresh iron ore sample was deposited in the annular channel. Four wear liner samples were then installed
Fig. 3. The abrasion wear testing system.
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
45
Fig. 4. Schematic of the liner setup for rubbing against the abrading agent.
on four load arms under a nominated normal load, after which the load arms were placed in the centre of the annulus in contact with the iron ore material. The tester was set in motion with the platform rotating at a nominated velocity. The tests were conducted with the moisture monitoring system activated to ensure a relative consistent moisture throughout the test. Every 5 h, tests were interrupted to allow liner samples to be removed and cleaned. The corresponding sample weight loss was measured. The samples were then re-installed and the tests continued. A total of 40 h of abrading time was performed for each sample to complete a test. As discussed above, a total of 9 different lining materials were selected for this study. Each test material was prepared to have a 100 mm by 100 mm surface area. The thickness of the samples ranged from 20 mm–40 mm. Noting that each suite of tests can only accommodate a maximum of 4 liner test samples. Therefore, three separate wear test suites were performed to complete the test program. Each suite of testing was repeated three times with new liner samples. The mild steel liner was used in each test suit as a benchmarking lining. Fig. 5 shows the lining sample arrangements in all three suites. The selection of the normal pressure applied to the liners and the velocity of the abrading medium in the experiment are deemed to reflect the operational conditions in material handling plants during iron ore mining. In material storage bins, the abrasive wear tends to concentrate at the hopper discharge section; whereas in chutes, abrasive wear occurs along the material flow path (Corder and Thorpe, 1987; Corder and Thorpe, 1995; Roberts, 1988). In both cases, the iron ore abrading medium exhibited relatively fast flow velocity under small normal consolidation pressure. Therefore, a linear speed of 1 m/s for the iron ore material and a normal load of 2 kPa was applied to the lining samples in all tests. A typical iron ore fines product was utilised as the abrading agent in this study. The bulk material properties of the sample, including the particle size distribution, the particle density and the moisture content are shown in Fig. 6.
Fig. 6. Particle size distribution of the used iron ore abrading agent.
4. Results and discussion 4.1. Abrasive wear resistance ranking After the completion of three suites of testing, the weight loss in grams of each liner due to abrasive wear for a total testing duration of 40 h was obtained. However, it is more useful to compare the abrasive wear resistance by normalising the weight loss to liner thickness. Although the liner thickness may be directly measured using techniques such as a vernier, the measurement may not be able to account for the deep surficial cut on the ductile metal linings and the localised chipping on the brittle ceramic liners, which leads to underestimation of the abrasion wear. Based on the forgoing comments, the weight loss results are subsequently converted to thickness loss, in microns, with the following expression: Thickness Loss ¼
M 103 Aρ
ð2Þ
where M is mass loss in grams, A is the contact surface area and ρ is the test liner sample density. The contact surface area was fixed to be 0.01 m2 for each sample. Results of the thickness loss for all liners are shown in Fig. 7. Fig. 7 (a) shows that the thickness loss results of mild steel liners are consistent for the three separate testing suites, indicating the abrasive wear resistance can be directly compared across different testing suites. The thickness loss for the mild steel liners was observed to be generally higher than other lining materials, indicating significantly lower abrasive wear resistance. Among linings A–H, metal type of lining materials generally exhibited higher thickness loss, thus lower abrasion wear
Fig. 5. Wall linings arrangements in three separate test suites.
46
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48 Table 4 Abrasive wear rate for each lining material through linear regression of the wear testing results. Ranking
Liner label
Abrasive wear rate - Ψ [μm/h]
1 2 3 4 5 6 7 8 9
C D A E B G F H X1 X2 X3 X4 Average of X
0.099 ± 0.02 0.352 ± 0.05 0.458 ± 0.04 0.527 ± 0.05 0.603 ± 0.04 0.942 ± 0.06 1.090 ± 0.05 1.281 ± 0.07 11.17 ± 1.3 9.83 ± 1.4 9.99 ± 1.2 10.63 ± 1.6 10.40 ± 1.4
the lining to enhance its mechanical strength, which is also indicated from the Knoop hardness results. Among the three other metal linings, no distinct abrasive wear resistance is observed between Linings F and G. However, Lining H exhibits the lowest wear resistance among all metal linings, which may be due to its lower chromium (Cr) contents. Furthermore, as shown in Fig. 8, the Knoop hardness value is observed to be proportional to the logarithmic abrasion wear rate. This is suggesting that the abrasive wear resistance of a specific lining material can be simply indicated by its Knoop hardness value. Practically, when selecting lining materials in bins and chutes to handle iron ore against abrasive wear, the Knoop hardness test can be used to predict its abrasion wear performance. Lastly, the surface roughness exhibits no obvious correlation with the ranking of the abrasive wear rate. Fig. 7. Thickness loss results for each liner based on Eq. (2).
resistance compared to the ceramic linings. Nevertheless, metal lining sample E showed higher abrasive wear resistance compared to ceramic lining sample B. Additionally, all results exhibited a quasi-linear relationship between the thickness loss and the testing duration, from which an abrasive wear rate - Ψ (μm/h) for each lining material can be subsequently defined. The abrasive wear rate represents a simple indication of the abrasive wear resistance of a lining material. Table 4 showed the abrasive wear rate for each lining material and its corresponding ranking based on a linear regression approach. 4.2. Abrasive wear resistance correlation with material properties Ranking of the abrasive wear resistance is predominately determined by the lining material properties. The lining material properties, including the surface roughness and hardness, vary as the abrasive wear propagates into the material from frictional rubbing and localised heat generation. Therefore, it is more useful to predict the abrasive wear resistance of a particulate lining material based on its material properties at its as-fabricated condition, which is investigated below. Initially, based on the chemical compositions of the lining material, ceramic types of linings generally outperformed the metal liners in terms of abrasive wear resistance. Lining C, the silica carbide (SiC) type of ceramic is suggested to be more abrasive wear resistant comparing to the alumina (Al2O3) type of ceramics (A, B and D). Within the three alumina type ceramics, Lining D, with the additional zirconium dioxide (ZrO2) trace, exhibits higher abrasive wear resistance when comparing to Linings A and B. Lining B shows the lowest abrasion resistance among ceramics, which may be due to lower alumina content. Additionally, no enhancement on abrasive wear resistance is observed by adding the platinum (Pt) into the linings. It is also interesting to observe that the complex metal lining material E shows comparable abrasive wear resistance to the ceramic lining material B. This may be due to the small percent of niobium (Nb) formulated into
4.3. Service life predictions in chutes and bins Abrasive wear is assumed to be a function of the normal pressure, the rubbing or sliding velocity at the boundary and the friction coefficient. From research detailed in (Roberts and Wiche, 1993), it is reasonable to assume that abrasive wear is a linear function of normal pressure and rubbing velocity. Abrasive wear, Wa, expressed in units of thickness loss per unit time (μm/s) may be defined as follows: Wa ¼
σ W VS tanϕ ½μm=s σ Wp
ð3Þ
where • σW is the normal pressure at the boundary (kPa) • VS is the velocity of bulk solid adjacent to the boundary (m/s)
Fig. 8. Correlation between the abrasive wear rate and the Knoop hardness across all liners.
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48 Table 5 Service life prediction results for the example chute based on the wear testing results. Lining label
Wear rate in example chute [μm/h]
Service life – [h]
A B C D E F G H X
3.18 4.19 0.69 2.44 3.66 7.57 6.54 8.90 72.2
9432 7164 43,636 12,273 8197 3963 4586 3372 415
the service life of each lining material at other normal pressures and rubbing velocities in chute and bin applications (Roberts and Wiche, 1993). Examples are presented below. Firstly, for a straight, parallel transfer chute processing the iron ore product described in this study, the following typical geometrical and operational conditions are assumed: • Chute width, w • Tonnage, M • Sliding velocity, v • Wall friction angle (all linings), ϕ • Lining thickness (all linings), d • Bulk density, ρ
σW ¼ The abrasive wear parameter, σWp, can be established for the wear samples using Eq. (3) and the results obtained from the abrasion wear testing. Essentially, Eq. (3) may be transformed to the following form: σ W VS tanϕ Wa
½kPa
ð4Þ
In the abrasion wear tests, the normal pressure σW was set at 2 kPa, and the relative velocity of the material to the liner VS was 1 m/s. Providing a fixed wall friction angle ϕ, the abrasion wear parameter σWp is obtained below for a particular liner by utilising the abrasion Ψ ) as shown in Table 4. wear rate (Wa ¼ 3600 • σWp ~15,720 tan ϕ (106 kPa) • σWp ~11,940 tan ϕ (106 kPa) • σWp ~72,727 tan ϕ (106 kPa) • σWp ~20,454 tan ϕ (106 kPa) • σWp ~13,662 tan ϕ (106 kPa) • σWp ~6605 tan ϕ (106 kPa) • σWp ~7643 tan ϕ (106 kPa) • σWp ~5620 tan ϕ (106 kPa) • σWp ~692 tan ϕ (106 kPa)
=2 m =10,000 t/h =5 m/s =25° =30 mm =2500 kg/m3
The normal pressure acting on the lining surface was estimated as,
• ϕ is the wall friction angle • σWp is the abrasive wear parameter (106 kPa)
σ Wp ¼
47
Liner A Liner B Liner C Liner D Liner E Liner F Liner G Liner H Liner X
Having determined the abrasive wear factor for a particular lining material against this iron ore product, it is now possible to estimate
5 Mg ¼ 2:8 ½kPa 18 wv
ð5Þ
The wear rate and life span of the different lining material can be estimated using Eq. (2) and abrasive wear parameter derived from the wear testing. Results are shown in Table 5. Secondly, for an axisymmetric mass flow bin (shown in Fig. 9 (a)) discharging the iron ore product, the follow typical geometrical and operational conditions are assumed: • Diameter, D • Outlet diameter, B • Bin height, Hb • Hopper height, Hh • Hopper half angle, α • Effective internal frictional angle, δ • Bulk density, ρ • Discharge rate, M • Wall friction angle (all linings), ϕ • Lining thickness (all linings), d
=12 m =1.5 m =24 m =12.37 m =23° =50° =2500 kg/m3 =2500 t/h =25° =30 mm
The normal wall pressures and the sliding velocity of the material during symmetrical discharge are calculated according to the Australian Standard – loads on bulk solids containers (Australian Standard, 1996). Results are shown in Fig. 9. The normal pressure on the bin wall increases from the top towards the transition from the vertical section to the hopper section, at which the highest normal wall pressure is indicated. From the hopper transition section to the discharge opening, the normal wall pressure continuously decreases. In the meantime, the
Fig. 9. Normal wall pressure and sliding velocity during discharging in the example bin.
48
W. Chen et al. / International Journal of Mineral Processing 167 (2017) 42–48
Fig. 10. Predicted service life along the height of the example bin for each lining material selected in this study.
material sliding velocity continuously increases from the top to the discharge opening of the bin. Based upon the normal wall pressure and the sliding velocity of the material, the wear rate and the service life of wall linings within the example bin are estimated. Results of the predicted service life are shown in Fig. 10. From the results, it is indicated that the transition from vertical section to the hopper section and the outlet region exhibits higher abrasive wear within the bin. Therefore, the abrasive wear lining design and installation strategy should reflect the localised abrasion wear rate in order to extend the service duration of the bin. 5. Conclusion A comprehensive study was performed on the abrasive wear resistance of wall lining materials in bins and chutes during iron ore mining operations. A suite of experimental investigations and modelling on predicting the service life of these wall linings were conducted. This study yielded the following major findings: • A comparative abrasive wear testing system was developed and was capable of obtaining the abrasive wear rate for a particular lining from a selected abrading agent. • The Knoop hardness test can be utilised to indicate the ranking of abrasive wear resistance for various lining materials. • The lining material service life prediction model developed in this study can be applied to bins and chutes for wear pattern predictions under various geometrical and operational conditions.
Consequently, the outcome of this study provided a technical guide for selecting lining materials to be installed in chutes and bins during
iron ore mining for reduced abrasive wear and more accurate maintenance planning. References ASTM, A., 2012. E384: Standard Test Method for Knoop and Vickers Hardness of Materials. ASTM Stand, pp. 1–43. Australian Standard, 1981. AS 2382-1981 Surface Roughness Comparison Specimens. Australian Standard, 1996. AS 3774-1996 Loads on Bulk Solids Containers. Biswas, S., Cenna, A., Williams, K., Jones, M., 2014. Subsurface behavior of ductile material by particle impacts and its influence on wear mechanism. Proc. Eng. 90, 160–165. Cenna, A.A., Williams, K.C., Jones, M.G., 2011. Analysis of impact energy factors in ductile materials using single particle impact tests on gas gun. Tribol. Int. 44 (12), 1920–1925. Corder, G.D., Thorpe, R.B., 1987. Wear of Hopper Walls. Tribology in Particulate Technology. 1987. Adam Hilger. Corder, G.D., Thorpe, R.B., 1995. An experimental study of the wear at hopper walls. KONA Powder Part. J. 13 (0), 105–112. Cortie, D.P., McEwan, J.J., 1996. Materials selection in the mining industry: old issues and new challenges. J. South. Afr. Inst. Min. Metall. 96 (4), 145–157. Hutchings, I.M., 1993. Mechanisms of wear in powder technology: a review. Powder Technol. 76 (1), 3–13. Roberts, A.W., 1988. Friction adhesion and wear in bulk materials handling. Proc. Antiwear 88. The Royal Society London. Roberts, A.W., Wiche, S.J., 1993. Prediction of lining wear life of bins and chutes in bulk solids handling operations. Tribol. Int. 26 (5), 345–351. Roberts, A.W., Ooms, M., Wiche, S.J., 1989. Concepts of boundary friction, adhesion and wear in bulk solids handling operations. Third International Conference on Bulk Materials, Storage, Handling and Transportation: Preprints of Papers. Institution of Engineers, Australia, p. 349. Scott, O.J., Choules, P.R., 1993. The use of impact plates in conveyor transfers. Tribol. Int. 26 (5), 353–359. Shindo, D., Oikawa, T., 2002. Energy dispersive X-ray spectroscopy. Analytical Electron Microscopy for Materials Science. Springer, pp. 81–102. Tuckey, K.R.G., 2003. Intelligent selection of engineered wear linings in iron ore plant. Miner. Process. Ext. Metall. 112 (1), 33–38. Wiche, S.J., Keys, S., Roberts, A.W., 2005. Abrasion wear tester for bulk solids handling applications. Wear 258 (1), 251–257.