Geologically Active – Williams et al. (eds)© 2010 Taylor & Francis Group, London, ISBN 978-0-415-60034-7

1973

Determining abrasivity of rock and soil in the laboratory

H. Käsling & K. ThuroEngineering Geology, Technische Universität München, Germany

ABSTRACT: The present work lights on a set of tests for the determination of abrasivity, whose results are used for the estimation of tool wear not only in TBM tunneling but also in rock drilling, in the use of road headers, in foundation construction by pilling etc. Besides the geologic and mineralogic way to estimate the abrasivity by thin section analysis, the Cer-char abrasivity test has to be highlighted as a widely used test especially during cost calcula-tion in TBM tunneling. Furthermore, the LCPC abrasivity test has become more and more important in rock and soil testing. Even as both laboratory tests have been partly regulated by standards, they are performed in a multitude of variations that result in highly differing results. Thus, an update of the test recommendations is needed urgently. As abrasivity assess-ment of soils becomes more and more important, the authors have set a capable testing pro-cedure and a suggestion of a unified classification scheme for both, rock and soil materials.

1 INTRODUCTION

The abrasivity of rock and even soil is a factor with considerable influence on the wear of tools. Hereby the wear is a question of material consumption and is an important indicator of rock excavation in tunneling, underground mining or quarrying in addition to the excavation speed (Fig. 1). On the one hand, the wear depends on the machinery being used for excavation; that are the devices and all tools which have contact to the rock or excavated material. On the other hand the rock and the geological conditions can be specified by geotechnical parameters.

The abrasivity of rocks can already be described by the petrografic composition, in par-ticular the contribution of hard minerals like quartz. This more geological way of determina-tion is used when the quartz or equivalent quartz content of rock is specified by microscopic examination of a thin section. Another, more technical way is to determine the abrasivity of rocks by laboratory tests where some kind of model or index test is used. In the following paper the Cerchar abrasivity test as well as the LCPC abrasivity test are explained, some tech-nical issues are commented and a unified classification system for both tests is presented.

2 MINERALOGICAL COMPOSITION OF ROCK MATERIAL

The most geological way to describe the abrasivity of rock, is to determine the mineralogical composition, in particular the contribution of hard minerals like quartz. This investigation by analyzing a thin section (Fig. 2) is used to determine the commonly used equivalent quartz content (Thuro 2002). The multiplication with the uniaxial compressive strength gives the Rock Abrasivity Index RAI (Plinninger 2008).

3 CERCHAR ABRASIVITY TEST

3.1 Testing principles

The Cerchar Abrasivity Test has been introduced in the 70s by the Centre d’Etudes et Recherches des Charbonages (CERCHAR) de France for abrasivtiy testing in coal

1974

bearing rocks. The test layout is described in Cerchar (1986) and in the French standard NF P94-430-1 in general.

The testing principle is based on a steel pin with defined geometry and hardness that scratches the surface of a rough rock sample over a distance of 10 mm under a static load of 70 N. The Cerchar-Abrasivity-Index (CAI) is then calculated from the measured diameter of the resulting wear flat on the pin:

CAId

c= ⋅10 (1)

where CAI = Cerchar-Abrasivity-Index (−); d = diameter of wear flat (mm); c = unit correc-tion factor (c = 1 mm).

As result of a worldwide survey it can be stated, that two testing devices with little modifications according to Cerchar (1986) and West (1989) are used in similar frequency (Figs. 3 and 4).

3.2 Variations and influencing factors

The Cerchar-Abrasivity-Index is used as a key parameter in prediction models for TBM tun-neling (Gehring 1995, Rostami et al. 2005) and for roadheader excavation. Therefore reliable

Figure 1. Parameters influencing tool wear and excavation performance.

Figure 2. Example of a thin section analysis (with counting grid) of a granite sample with determination of the Equivalent Quartz Content EQu under the microscope (Thuro & Käsling 2009).

1975

test results are needed to ensure the practicability of this index test as a quick and easy way to gain information about abrasivity of rocks worldwide.

Modifications of the test setup (Al-Ameen & Waller 1993, West 1989), who are partly not in familiar with the French standard headed to a multitude of testing variations and highly differing testing results all over the world. This leads to inadequate prediction of tool wear and often in unexpected cost over-runs. This inaccuracy could have been observed during several tunneling projects in Europe, North America and Australia in the last decade. Highly varying testing results from different laboratories have also been shown by Rostami (2005) and Rostami et al. (2005).

Numerous influences that have been evaluated during the last years, are described in detail in Käsling et al. (2007) and Käsling (in prep.).

Some of the influencing factors are shortly described in the following. At first the used testing equipment has to be stiff enough that the steel pin is accurately guided over the rock surface. Secondly and maybe the core point is the steel type and quality. Not only an adequate steel grade has to be used but also the required hardness of the pin has to be ensured. A world-wide survey at rock laboratories showed that steel pins of Rockwell hardness HRC54–56 like in the original literature and the French standard NF P94-430-1 are used as well as steel pins of a much lower hardness (HRC 40, West 1989). Figure 5 shows a correlation of testing results by both hard and soft steel pins. On the face of it, the results cannot be compared that easy, as stated by Michalakopoulos et al. (2005). In addition some laboratories carry out the test on plain, saw cut rock surfaces. As Figure 6 shows, the CAI derived on this smooth, saw-cut surface is a bit lower than the CAI derived on the rough, freshly broken rock surface recommended in the French standard. Again a reliable conversion from a saw-cut surface CAI in a rough-surface CAI and vice versa is difficult due to the high deviation especially at high CAI values. Furthermore the orientation of the test using anisotropic rocks and the precise reading of the wear flat of the steel pin are relevant for comprehensible results. In particular, the measurement of the wear flat diameter of the truncated cone has to be taken using a microscope (magnification about 50x) as an average of two, better four single read-ings perpendicular (normal) to each other using the cross section of the pin every 90°.

4 LCPC ABRASIVITY TEST

4.1 Testing principle

The LCPC abrasivity testing device (Fig. 7) is described in the French standard P18-579 and has been developed by the Laboratoire Central des Ponts et Chausées (LCPC) in France for

Figure 3. Setup of a modified Cerchar testing device according to Cerchar (1986). 1 – weight, 2 – pin chuck, 3 – steel pin, 4 – sample, 5 – vice, 6 – hand lever.

Figure 4. Setup of a testing device according to West (1989). 1 – weight, 2 – pin guide, 3 – steel pin, 4 – sample, 5 – vice sled, 6 – hand crank.

1976

testing rock and aggregates. The “abrasimeter” is built of a 750 W strong motor holding a metal impeller rotating in a cylindrical vessel which contains the granular sample. The rec-tangular impeller is made of standardized steel with a Rockwell hardness of HRB 60–75. As stated in the standard, the grain size of the rock sample has to be in a range between 4 to 6.3 mm; rock has to be crushed before the test accordingly.

With the aid of the LCPC abrasivity test, the breakability or brittleness of the sample material can be quantified too. A modified classification is given in Table 1. The LCPC-Breakability-Coefficient (LBC) is defined as the fraction below 1.6 mm of the sample mate-rial after the test:

LBCM

=( )M ⋅MM

(2)

where LBC = LCPC-Breakability-Coefficient (%); M1.6

= mass fraction <1.6 mm after LCPC test (g); M = mass of the sample material (= 0.0005 t).

Figure 5. Results of Cerchar abrasitity tests carried out with steel pins of different hardness (HRC 54–56 according to the French standard NF P94-430-1 and HRC 40 according to Al-Ameen & Waller (1993) or West (1989)).

Figure 6. Results of Cerchar abrasitity tests carried out on rough rock surfaces and smooth, saw cut surfaces in Käsling (in prep.).

Figure 7. LCPC abrasivity testing device according to the French standard P18-579 (1990). 1 – motor, 2 – funnel tube, 3 – steel impeller, 4 – sample container.

1977

4.2 Application for soils

For soils, the choice of a suitable testing procedure to determine the abrasivity is much more limited than for rocks. A determination of the equivalent quartz content or the Cerchar abra-sivity test is only practical for large components like coarse gravels and blocks. The only estab-lished way of determining abrasivity of soil is the LCPC abrasivity test, which was developed specially for “Granulate” by the Laboratoire Central des Ponts et Chausées (NF P18-579). As in Büchi et al. (1995) the test is only described for rock, a detailed procedure for soil testing has been proposed in Thuro et al. (2007), Thuro & Käsling (2009) and Käsling (in prep.).

The investigation method described in Nilsen et al. (2006), according to the statements given, is only suitable for material with grain size less than 1 mm and therefore is only feasible for silty sands and smaller grain sizes or comparable material.

When testing soil material, some considerations have to be done in order to agree with the technical recommendations: maximum grain size 6.3 mm due to the arrangement of the impeller and the capacity of the engine (Fig. 8).

• Testing the grain sizes between 4 and 6.3 mm of the soil sample. This fraction has to be obtained by sieving. This leads to low abrasivity values, which do not represent the real abrasivity of the entire soil sample

• Testing the grain sizes less than 6.3 mm of the sample accordingly, also leads to low values. Note, that originally the LCPC test was not intended to contain fine-grained materials less than 4 mm.

• Testing the entire soil sample and crushing the grains larger than 6.3 mm in a crusher. According to the original grain size distribution the crushed and therefore more angular material has to be added to the original material again. Depending on the scope of the abrasivity determination, the fines <4 mm have to be used for the test or excluded.

For a geotechnical interpretation of the obtained abrasivity values, a grain size distribu-tion analysis of the soil material before the test (and crushing) is essential. In addition, a mineralogical and petrological analysis of the components should be performed. The fines below 2 mm can be analyzed by X-ray diffractometer, whereas the larger components can be determined manually or optically.

Table 1. Classification of the LCPC-Breakability-Coefficient (LBC) according to Käsling (in prep.), modified from Büchi et al. (1995).

LBC [%] Breakability classification

0–25 Low

25–50 Medium

50–75 High

75–100 Very high

Figure 8. Example for sample preparation: Crystalline rich, sandy gravel before preparation (left), before (middle) and after (right) the LCPC abrasivity test. Scale: 1 black bar equals 1 cm.

1978

5 ABRASIVITY CLASSIFICATION

5.1 Abrasivity of rock

For common rock samples, the Cerchar-Abrasivity-Index varies between 0 and 6 and the LCPC-Abrasivity-Coefficient varies between 0 and 2000 g/t. As shown in Figure 9, there is a close linear correlation between the LAC and the CAI for the tested rock samples. Therefore the well-known Cerchar-Abrasivity-Index is used as a basis for a combined classification scheme as shown in Table 2 instead of the classification given in Büchi et al. (1995).

5.2 Classification of abrasivity of soils

There is generally little problem in categorization of rock into the proposed classification in Table 2; but for medium to coarse-grained soils (“loose rock”) additional factors have to be considered:

• Roundness of the grains: the abrasivity increases with a reduction in the degree of round-ness; sharp edged components cause higher wear.

• Mineral content of the components: crystalline pebbles possess a high abrasivity, carbon-ate pebbles (limestone or dolomite) are relatively insignificant.

Figure 9. Correlation between CAI and LCPC abrasivity testing results using data in Büchi et al. (1995) and results from own studies (modified from Thuro & Käsling 2009).

Table 2. Classification of the LCPC-Abrasivity-Coefficient (LAC) in connection with the CERCHAR-Abrasivity-Index (CAI) according to Thuro et al. (2007).

LAC [g/t] CAI [0.1] Abrasivity classification Examples

0–50 0.0–0.3 Not abrasive Organic material

50–100 0.3–0.5 Not very abrasive Mudstone, marl

100–250 0.5–1.0 Slightly abrasive Slate, limestone

250–500 1.0–2.0 (Medium) Abrasive Schist, sandstone

500–1250 2.0–4.0 Very abrasive Basalt, quartzitic sandstone

1250–2000 4.0–6.0 Extremely abrasive Amphibolite, quartzite

1979

Figure 10. Classification scheme for the LCPC-Abrasivity-Coefficient with allocation for different soil types (Thuro & Käsling 2009).

• Grading distribution: the abrasivity increases disproportional with the grain size. Stones and blocks can be decisive, because they have the same effect on the considerably smaller tools as sound rock.

In Figure 10 a synopsis of typical testing results of soil material is given. The diagram shows the LCPC-Abrasivity-Coefficient plotted against the medium grain diameter of the original sample at 50% (D

50) for different soil material. The absolute grain size is therefore

reflected in the “sharpness” of the grain size mixture prepared for the LCPC abrasivity test.

6 CONCLUSIONS

The Cerchar abrasivity test is in worldwide use for abrasivity assessment of rocks and wear prediction. The results are directly linked with the prediction model of the Colorado School of Mines for TBM cutter wear (Rostami et al. 2005, Frenzel in prep.) and is used for wear predictions of roadheaders too. Due to variations of the test occuring in the last decades, reliable and comparable testing results are occasionally non-existent. Revised testing recom-mendations for the Cerchar abrasivity test are in preparation by the DGGT working party 3.3 Versuchstechnik Fels (commission on rock testing techniques) and will include the main influencing factors that have been evaluated the last years.

The LCPC abrasivity test becomes more and more common for rock and soil testing in Europe. A French standard describes the testing facility in detail but lacks application on both, rock and soil samples. Ongoing work has to be done to implement testing of soil and aggregates satisfactory. Also for this test, testing recommendations are in preparation by the above mentioned DGGT working party.

In addition, a unified abrasivity classification for the Cerchar Abrasivity Index and the LCPC Abrasivity Coefficient has been presented in Table 2. It is based on the classification of Cerchar (1986) and has already been proven in construction practice successfully.

By the use of the presented laboratory tests, in some cases unusual test results have occurred. They can be caused by the specific test style and impact when testing a certain rock type; (e.g. very inhomogeneous or anisotropic rocks). Therefore it is helpful to combine these index tests with additional petrographic respectively thin section analysis. This reinsurance can help to avoid bad surprises and disputes during tunneling or foundation work in rock or soil.

1980

REFERENCES

Al-Ameen & Waller 1993. The influence of rock strength and abrasive mineral content on the Cerchar Abrasive Index. Engineering Geology 36: 293–301.

Büchi, E., Mathier J.-F. & Wyss, Ch. 1995. Gesteinsabrasivität – ein bedeutender Kostenfaktor beim mechanischen Abbau von Fest- und Lockergestein. Rock abrasivity – a significant cost factor for mechanical tunnelling in loose and hard rock. Tunnel 5: 38–44.

Cerchar - Centre d´Études et des Recherches des Charbonages de France. 1986. The Cerchar abrasive-ness index. Verneuil.

Frenzel, Ch. (in prep.). Kostenprognose für Schneidrollen bei maschinellen Tunnelvortrieben in Fest-gesteinen. Cost prediction for disc cutters in hard rock TBM tunnelling. Münchner Geowissenschaftli-che Abhandlungen, (B) 18, München: Verlag Dr. Friedrich Pfeil.

Gehring, K.-H. 2005. Leistungs- und Verschleißprognosen im maschinellen Tunnelbau. Felsbau 16 (6): 439–448.

Käsling, H. (in prep.). Bestimmung der Gesteinsabrasivität – Grundlagen, Anwendung und Einsatz-grenzen bei maschinellen Tunnelvortrieben. Determination of rock abrasiveness – pronciples, appli-cation and limitation of use in TBM tunelling. Münchner Geowissenschaftliche Abhandlungen, (B) 19, München: Verlag Dr. Friedrich Pfeil.

Käsling, H., Thiele, I. & Thuro, K. 2007. Abrasivitätsuntersuchungen mit dem Cerchar-Test – eine Evaluierung der Versuchsbedingungen. In Otto (ed.) Veröffentlichungen von der 16. Tagung für Ing-enieurgeologie, 7–10. März 2007. Bochum: Technische Fachhochschule Georg Acricola.

Michalakopoulos, T.N., Anagnostou, V.G., Bassanou, M.E. & Panagiotou, G.N. 2005. The influence of steel styli hardness on the Cerchar abrasiveness index value. Int. J. Rock Mech. Min Sci. & Geomech. Abstr. 43: 321–327.

Nilsen, B., Dahl, F., Holzhäuser, J. & Raleigh, P. 2006. SAT: NTNU’s new soil abrasion test. Tunnels & Tunnelling Int. 5, pp. 43–45.

Normalisation Française P18-579. 1990. Granulats: Essai d’abrasivité et de broyabilité. Paris: AFNOR Association française de normalisation.

Normalisation Française P94-430-1. 2000. Roches - Détérmination du pouvoir abrasive d’une roche. Partie 1: Essai de rayure avec une pointe. Paris: AFNOR Association française de normalisation.

Plinninger, R.J. 2008. Abrasiveness assessment for hard rock drilling, Geomechanik und Tunnelbau, 1: 39–46.

Rostami, J. 2005. CAI testing and its implications. Tunnels & Tunnelling International 37 (10): 43–46.Rostami, J., Ozdemir, L., Bruland, A. & Dahl, F. 2005. Review of Issues related to Cerchar Abrasivity

testing and Their Implications on Geotechnical Investigations and Cutter Cost Estimates. In J.D. Hutton, & W.D. Rogstad (eds) Proceedings of the 2005 Rapid Excavation and Tunnelling Conference (RETC) in Seatle. Littleton: Society for Mining, Metallurgy and Exploration.

Thuro, K. 2002. Geologisch-felsmechanische Grundlagen der Gebirgslösung im Tunnelbau. Münchner Geologische Hefte, B18. München: Eigenverlag.

Thuro, K. & Käsling, H. 2009. Classification of the abrasiveness of soil and rock. Geomechanics & Tun-nelling 2: 179–188.

Thuro, K., Singer, J., Käsling, H. & Bauer, M. 2007. Determining abrasivity with the LCPC Test. In E. Eberhardt, D. Stead & T. Morrison (eds.). Proceedings of the 1st Canada – U.S. Rock Mechanics Symposium, 27.-31.05.2007, Vancouver B.C., London: Taylor & Francis.

West. 1989. Rock abrasiveness testing for tunnelling. Int. J. Rock Mech. Min Sci. & Geomech. Abstr. 26: 321–327.