Rock Mech Rock Eng DOI 10.1007/s00603-015-0862-3
TECHNICAL NOTE
Study on the Correlations Between Abrasiveness and Mechanical Properties of Rocks Combining with the Microstructure Characteristic Jianming He1 • Shouding Li1 • Xiao Li1 • Xu Wang2 • Jingyun Guo1
Received: 12 April 2015 / Accepted: 30 September 2015 Springer-Verlag Wien 2015
Keywords Abrasiveness Cerchar abrasiveness index (CAI) Mechanical properties Microstructure characteristics
1 Introduction Rock abrasivity has become one of the necessary parameters for mechanical excavation of rock in the tunnelling industry (Wijk 1992; Hassanpour et al. 2011). The Cerchar abrasiveness test is widely used to assess the rock abrasivity. The test features a steel pin of specified shape and quality that is scratched over 10 mm of the specimen’s surface. The pin has a sharp conical point with cone angle of 90. A static force of 70 N is loaded on the pin and the test is carried out by moving the steel pin at a prescribed velocity on the testing surface. The Cerchar abrasiveness index (CAI) is calculated by multiplying the mean value of the wear flats stated in units of 0.01 mm by 10. The testing velocity is reduced from 10 mm/s of ‘‘Cerchar apparatus’’ to 1 mm/s of ‘‘West apparatus’’ to accommodate the controllability of the testing velocity (West 1981, 1989; Michael et al. 2014). It was concluded in the literature that CAI was not only related to the abrasive mineral content alone but also
& Jianming He
[email protected] 1
2
Key Laboratory of Shale Gas and Geoengineering, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, People’s Republic of China Geotechnical Engineering Company of Liaoning Provincial Building Design and Research Institute, Shenyang 110005, People’s Republic of China
influenced significantly by the rock strength (AL-Ameen and Waller 1994). The equivalent quartz content (EQC) alone was found to be not enough to interpret the abrasion values of the Cerchar scratch test (Plinninger et al. 2003). It was found that some technical factors had impact on the test result of the Cerchar scratch test. CAI was mainly influenced by the rock’s deformability and the content of abrasive minerals (Plinninger et al. 2003). Size of quartz grains was proved to be very important to the scratch tip wear. Smaller quartz grains gave low wear rate and larger grains gave higher rate (Beste et al. 2004). The relations between CAI value and petrographic properties of Coal Measures Rocks were examined and a good linear relationship between CAI and average quartz grain size parameters was revealed (Yarah et al. 2008). Many factors, including pin hardness, surface condition of specimens, and petrographical and geomechanical properties, were examined to study the influence of various parameters on Cerchar testing (Jamal et al. 2014). So far, there is little knowledge about the quantification of microstructure as well as the effect of microstructure characteristics on the abrasiveness of rocks. Microstructure coefficients are defined and integrated with the mechanical parameters in this study to account for the combined effects on the abrasiveness of rocks. The abrasiveness apparatus used in this study has been improved based on the previous test apparatus. The horizontal movement of the steel pin, which is driven by step-servo motor, can be controlled accurately through software (Fig. 1).
2 Cerchar Abrasiveness Testing The standard Cerchar tests were conducted on the fresh surfaces of different rocks. The steel pin used in the rock abrasiveness test had a tensile strength of 200 kg f/mm2
123
J. He et al.
Fig. 1 Improved testing apparatus to determine the CAI. 1 Sample vice, 2 rock sample, 3 steel pin, 4 pin chuck, 5 stress sensor, 6 displacement sensor, 7 steel wire, 8 vice sled, 9 step-servo motor, 10 displacement sensor, 11 steel wire
with HRC of C43 according to the technical note of rock abrasiveness testing for tunnelling (West 1989). Discshaped rock specimens were prepared and firmly held in the test apparatus. The testing surfaces were cleared from debris or loose grains. The steel pin was moved a total distance of 10.0 mm across the rock surface. A minimum of five test replications were made on each rock surface in order to obtain a more reasonable mean value of CAI. The tip of the steel pin was magnified using the electronic microscope with an accuracy of 0.01 mm for the measurement of the diameter of the wear flat (Michael et al.
2014). The CAI values of different rocks are listed in the Table 1. Pin penetration into the rock increases with reduced strength and abrasiveness of the rock (Mohammad et al. 2014). Some typical scratch damages of burnishing and ploughing can be observed in the tests, which can be classified into two different scratch types (Fig. 2). If the rock has sufficient strength to resist cracking, the surface may become burnished, such as the scratch on the tuff, diorite, syenogranite, monzonitic granite and feldspathic sandstone. The burnishing could include the removal of small particles. CAI values of the rocks with this type of scratch usually fall in the range of 2.0–4.0. If the loading on the tip causes extensive deformation, the tip will plough the rock surface, such as the scratch on the marble, argillaceous sandstone and mudstone. The ploughing can result in deep uneven tracks on the rock surface. CAI values of the rocks with this type of scratch are usually less than 2.0.
3 Microstructure Characteristics Three thin rock slices were made to show rock microstructure in different directions of horizontal, vertical and random angle for further study. Investigation and analysis of rock slices using polarizing microscope fitted
Table 1 CAI, EQC values, and main constituents of different rocks Lithology
CAI (0.1 mm)
Marble
1.84
3.61
0
0
0
95
0
5
Marble
1.96
1.97
0
0
0
85
0
15
Tuff Tuff
3.00 2.95
84.34 78.88
28 30
70 64
2 1
0 0
0 5
0 0
Diorite
3.65
49.16
70
20
8
0
0
2
Syenogranite
3.66
51.49
76
20
2
0
0
2
Argillaceous sandstone
1.62
50.44
0
48
16
0
36
0
Argillaceous sandstone
2.54
22.31
6
18
0.5
48
0
27.5
Mudstone
1.36
15.94
0
15
0
0
85
0
Mudstone
1.39
12.75
10
8
0
16
66
0
Monzonitic granite
3.47
59.66
60
34
2
0
0
4
Monzonitic granite
3.51
58.28
54
35
2
0
0
9
Monzonitic granite
3.08
59.65
60
34
2
0
0
4
Feldspathic sandstone
2.95
54.63
45
35
0
0
0
20
Quartz sandstone
3.08
91.86
0
88
0
1
5
6
Tuffaceous sandstone
2.43
28.62
5
25
35
0
6
29
Lithic sandstone
3.78
73.31
0
70
0
10
14
6
Lithic sandstone
3.16
78.35
0
75
0
1
21
3
Fine lithic sandstone Fine lithic sandstone
1.75 2.19
17.93 14.68
3 4
15 12
0 3
46 22
1 0
35 59
123
EQC (%)
Feldspar (%)
Quartz (%)
Mica (%)
Carbonate (%)
Clay (%)
Other (%)
Study on the Correlations Between Abrasiveness and Mechanical Properties of Rocks Combining…
Fig. 2 Typical scratch damages of burnishing or ploughing of the different rocks
with image analysis software were conducted to explore the microstructure characteristics of different rocks. 3.1 Petrographic Analysis Rocks with higher CAI usually have greater content of hard minerals. Diorite, syenogranite, monzonitic granite, quartz sandstone and lithic sandstone were distinguished in abrasivity with CAI values greater than 3.0. The slices of quartz sandstone and lithic sandstone contain approximately 64–88 % quartz, so rocks with higher content of quartz usually mean higher abrasivity. The slices of diorite, syenogranite and monzonitic granite, which contain medium contents of quartz with 20–35 %, also show relative higher abrasivity due to the presence of 54–76 % feldspar. The abrasivity of rocks with medium CAI less than or equal to 3.0, such as tuff and feldspathic sandstone, also have relative higher content of quartz and feldspar. Argillaceous sandstone and fine lithic sandstone show varied CAI, which mainly depend on the content of hard minerals. Rocks with lower content of quartz and feldspar, such as marble and mudstone, show poor abrasivity with CAI less than 2.0. So contents of the hard minerals in the rocks play an important role in the rock abrasivity. The
percentages of each mineral in the rock slices are listed in Table 1. Quartz represents the most common abrasive mineral. EQC in the rock slices can be determined by multiplying the percentage of minerals present in the rock by Rosiwal hardness values as Eq. (1) (Thuro 1997; Amirreza 2010): EQC ¼
n X
A i Ri
ð1Þ
i¼1
Therefore, each mineral amount Ai is multiplied by its relative Rosiwal hardness Ri to quartz (with quartz being 100 %). EQCs of the different rocks are listed in Table 1. 3.2 Definition of Microstructure Coefficient In the digital images of rock slices from the microscope, the boundaries of mineral grains in the micro-view to be measured were traced and defined using image analysis software. Size coefficient D and shape coefficient F defined as Eqs. (2) and (3) were used for the quantification of microstructure characteristics of individual mineral grain in each image of the rock slices. Mineral grain with greater
123
J. He et al.
size coefficient can lead to higher wear rate and greater CAI value. On the contrast, mineral grain with greater shape coefficient means more regular outline and can lead to smaller CAI value. rffiffiffiffiffiffiffi 4Ai D¼ ; ð2Þ p F¼
4pA ; L2
ð3Þ
where, Ai represents the area of the individual mineral grain and L represents the perimeter of the individual mineral grain. If a fraction of TC is defined as the numerator of size coefficient and denominator of shape coefficient as Eq. (4), it has obvious physical meaning and the greater TC means the bigger or more irregular mineral grains presented in the rock. So microstructure coefficient TC can be used as an indicator for the microstructure characteristic of the individual mineral grain: TC ¼
coefficient values of the three slices from one rock sample was much smaller, so average coefficient values of the three slices from one rock sample were used to represent its overall microstructure characteristic. The results of size coefficient, shape coefficient and microstructure coefficient of different rocks are listed in Table 2.
D : F
ð4Þ
Coefficients of the individual grain were calculated according to Eqs. (2–4). Corresponding average value of each coefficient of the mineral grains in one rock slice was obtained. It was found that the difference of the average
4 Correlations Between CAI and Microstructure Characteristics CAI against the average values of size and shape coefficient of the different rocks were plotted and the correlation was best fitted using linear regression. In general, rocks with greater size coefficient have higher wear rates, leading to relative greater CAI (Fig. 3a); but greater shape coefficient means more regular outline of the mineral grains, leading to relative smaller CAI contrarily (Fig. 3b). CAI against the average values of microstructure coefficient of the different rocks were plotted and the correlation was best fitted using logarithmic regression with a coefficient of determination (R2) value of 0.6750 (Fig. 3c), which was used to indicate how well the data fit a statistical model. CAI value shows increasing trend with the increase of microstructure coefficient, while the increment of CAI is reduced gradually.
Table 2 Values of CAI, microstructure coefficients and mechanical parameters Lithology
CAI (0.1 mm)
D (mm)
F
TC
H
UCS (MPa)
Marble Marble
1.84 1.96
0.4304 0.0646
0.6500 1.0285
0.6622 0.0628
46 48
98.20 112.50
5.90 6.80
25.10 24.30
Tuff
3.00
0.4703
0.5610
0.8383
60
313.20
16.00
29.10
Tuff
2.95
0.6460
0.5989
1.0787
60
273.30
13.84
26.40
Diorite
3.65
0.4813
0.5300
0.9081
56
171.20
5.30
46.40
Syenogranite
3.66
0.9250
0.5210
1.7754
50
138.10
6.30
22.10
Argillaceous sandstone
1.62
0.1008
0.608
0.1658
52
118.80
4.60
17.20
Argillaceous sandstone
2.54
0.9337
0.8608
1.0847
51
136.2
7.80
17.56
Mudstone
1.36
0.0756
0.7190
0.1051
30
9.50
0.79
10.00
Mudstone
1.39
0.0596
1.1614
0.0513
29
22.50
1.60
11.20
Monzonitic granite
3.47
1.0294
0.3488
2.9509
58
203.55
10.18
20.30
Monzonitic granite
3.51
1.4027
0.7285
1.9254
56
238.56
10.80
23.50
Monzonitic granite
3.08
0.9274
0.5534
1.6759
52
208.64
10.43
21.20
Feldspathic sandstone
2.95
0.2215
0.6340
0.3494
52
105.40
5.10
11.50
Quartz sandstone
3.08
0.2990
0.5749
0.5201
35
173.55
8.68
18.30
Tuffaceous sandstone Lithic sandstone
2.43 3.78
0.0942 0.7533
1.0075 0.6155
0.0935 1.2240
45 58
206.70 163.83
10.23 7.19
19.30 23.80
Lithic sandstone
3.16
0.3338
0.6478
0.5153
58
178.12
7.41
19.20
Fine lithic sandstone
1.75
0.1180
0.8743
0.1350
30
119.46
3.56
16.35
Fine lithic sandstone
2.19
0.1365
0.8105
0.1684
33
132.24
3.80
10.64
123
BTS (MPa)
E (GPa)
Study on the Correlations Between Abrasiveness and Mechanical Properties of Rocks Combining… Fig. 3 a CAI plotted against size coefficient D. b CAI plotted against shape coefficient F. c CAI plotted against microstructure coefficient TC. d CAI plotted against the product of microstructure coefficient TC and EQC
It is well known that the correlation of EQC and CAI shows positive trend in general (West 1986). Combination of the microstructure coefficient and EQC led to the finding that product of them could make the correlation to be improved with R2 value of 0.7176 using the same regression (Fig. 3d).
5 Correlations Between Mechanical Properties and CAI Some of the mechanical parameters including Scleroscope hardness H, unconfined compression strength UCS, Young’s Modulus E and Brazilian Tensile Strength BTS were investigated for the study on the correlations between CAI and mechanical properties of rocks. Values of CAI and mechanical parameters of the different rocks are listed in Table 2. CAI against the mechanical parameters of rocks were plotted and the correlation was best fitted using linear regression. The results indicate that all of the mechanical parameters show certain positive correlations with CAI, but the correlations are poor with the greatest R2 value of 0.4928 (Fig. 4). So single mechanical parameter cannot be
used to interpret the CAI, and some more intrinsic influencing factors should be combined.
6 Combination of Mechanical Properties and Microstructure Characteristics As is well known, CAI of rocks can be influenced by many intrinsic factors, including microstructure characteristics like mineral composition, hardness and grain size, and mechanical properties like rock strength and hardness (Beste et al. 2004; Yarah et al. 2008). So the better correlations can be obtained only if more relevant factors can be considered and combined. Combination of mechanical parameters, microstructure coefficient and EQC led to the finding that the product of them were best fitted using logarithmic regression with higher R2 values (Fig. 5), comparing with the cases only one or two aspects were considered. The fitted correlation curves show good and stable correlation trends and corresponding statistic equations are also shown (Fig. 5). The correlations can be improved to a certain degree and even the least R2 value is 0.7414.
123
J. He et al. Fig. 4 CAI plotted against the mechanical parameters of rocks
Fig. 5 CAI plotted against the products of mechanical parameters, microstructure coefficient and EQC
123
Study on the Correlations Between Abrasiveness and Mechanical Properties of Rocks Combining…
7 Conclusions The Cerchar abrasiveness test, microstructure measurement and mechanical test of different rocks have been carried out for study on the correlations between abrasiveness and mechanical properties combining with the microstructure characteristic. The apparatus used for the Cerchar abrasiveness test has been improved in the horizontal movement of the steel pin for better testing results. Size and shape coefficient can be used to indicate the grain size and outline irregularity of the mineral. The ratio of them is defined as microstructure coefficient and used to evaluate the microstructure of mineral grains in the rock. The mechanical parameters show certain positive correlations with CAI, but the correlations are poor. Combination of mechanical parameters, microstructure coefficient and EQC can lead to the finding that the product of them are best fitted using logarithmic regression with better correlations. Acknowledgments The authors are supported by the National Natural Science Foundation of China (Grant Nos. 41272351, 41272352 and 41227901). The authors are also supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant Nos. XDB10030301 and XDB10030304).
References AL-Ameen SL, Waller MD (1994) The influence of rock strength and abrasiveness mineral content on the Cerchar abrasiveness index. Eng Geol 36:293–301
Amirreza G (2010) Study of Cerchar abrasivity index and potential modifications for more consistent measurement of rock abrasion. Dissertation, The Pennsylvania State University Beste U, Lundvall A, Jacobson S (2004) Micro-scratch evaluation of rock types-a means to comprehend rock drill wear. Tribol Int 37:203–210 Hassanpour J, Rostami J, Zhao J (2011) A new hard rock TBM performance prediction model for project planning. Tunn Undergr Space Technol 26:595–603 Jamal R, Amireza G, Ehsan AG et al (2014) Study of dominant factors affecting Cerchar abrasivity index. Rock Mech Rock Eng 47:1905–1919 Michael A, Olgay Y, Filip D et al (2014) ISRM suggested method for determining the abrasivity of rock by the Cerchar abrasivity test. Rock Mech Rock Eng 47:261–266 Mohammad TH, Hossein M, Jamal R (2014) Continuous monitoring of pin tip wear and penetration into rock surface using a new Cerchar abrasivity testing device. Rock Mech Rock Eng 47:689–701 Plinninger R, Kasling H, Thuro K et al (2003) Testing conditions and geological properties influencing the Cerchar abrasiveness index (CAI) value. Int J Rock Mech Min Sci 40:259–263 Thuro K (1997) Prediction of drillability in hard rock tunneling by drilling and blasting. In: Golser J, Hinkel WJ, Schubert W (eds) Tunnels for people, 1st edn. CRC Press, Boca Raton, pp 103–108 West G (1981) A review of rock abrasiveness testing for tunneling. In: Proceedings of the international symposium on weak rock, Tokyo, pp 585–594 West G (1986) A relationship between abrasiveness and quartz content for some coal measures sediments. Int J Min Geol Eng 4:73–78 West G (1989) Technical note–rock abrasiveness testing for tunnelling. Int J Rock Mech Min Sci Geomech Abstr 26(2):151–160 Wijk G (1992) A model of tunnel boring machine performance. Geotech Geol Eng 10:19–40 Yarah O, Yasar E, Bacak G (2008) A study of rock abrasivity and tool wear in Coal Measures Rocks. Int J Coal Geol 74:53–66
123
