Description:

_Journal of Structural Geology 32 (2010) 507-522_ Contents lists available at ScienceDirect Journal of Structural Geology journal homepage: www.elsevier.com/locate/jsg Impact of ephemeral cataclastic fabrics on laser diffraction particle size distribution analysis in loose carbonate fault breccia Fabrizio Storti*, Fabrizio Balsamo Dipartimento di Scienze Geologiche, Università “Roma Tre”, Largo S.L. Murialdo 1, I-00146 Roma, Italy Article info Article history: Received 12 December 2008 Received in revised form 19 January 2010 Accepted 25 February 2010 Available online 4 March 2010 Keywords: Fault breccia Particle size Particle shape Laser diffraction granulometry Particle disintegration Cataclasis Abstract Modern laser diffraction particle size analysers provide the possibility of fast particle size data acquisition over a wide size range by using a variety of analytical methods, named standard operating procedures. We performed specific tests on poorly coherent carbonate platform cataclastic rocks from a fault zone in the Central Apennines, Italy, by combining laser diffraction granulometry, thin section analysis, and optical morphometry. During laser diffraction granulometry tests, we used several wet and dry operating procedures that included different pump speeds, analyses with and without sample ultrasonication, and different dispersant liquids. The variability of particle size distributions from a given sample, as a function of the adopted operating procedure, has the same magnitude as that theoretically predicted in natural cataclastic rocks, from low-to high-deformation shear zones. Thin section image analysis and optical morphometry support mechanical disintegration of internally microfractured coarser particles in ephemeral cataclastic fabrics as the major cause of such a size variability. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction Cataclastic rocks exert a primary control on the frictional strength, stability, seismic velocity, and permeability properties of fault zones (Tullis and Weeks, 1986; Sammis et al., 1987; Marone and Scholz, 1989; Marone and Kilgore, 1993; Hadizadeh, 1994; Antonellini and Aydin, 1995; Caine et al., 1996; Evans et al., 1997; Ben-Zion and Sammis, 2003; Anthony and Marone, 2005; Reches and Dewers, 2005; Billi and Di Toro, 2008). Particle size distributions are central to these studies and have been widely used to investigate the evolution of cataclasis (Borg et al., 1960; Engelder, 1974; Sammis et al., 1986, 1987; Blenkinsop, 1991; Morgan, 1999; Mair et al., 2002; Monzawa and Otsuki, 2003; Storti et al., 2003; Wilson et al., 2005; Heilbronner and Keulen, 2006; Fossen et al., 2007; Sammis and King, 2007; Sammis and Ben Zion, 2008; Torabi et al., 2008). Despite the large amount of available data, interpretation of particle size distributions from poorly cohesive cataclastic rocks is still controversial, particularly because results from microscopic and sieve analyses (e.g. Sammis et al., 1986; Billi and Storti, 2004), and from laser diffraction granulometry (e.g. Wilson et al., 2005; Reches and Dewers, 2005) have been used to support either fractal or non-fractal behaviors, respectively. A possible investigation strategy relies on the interplay between grain weakness in poorly cohesive fault rocks (e.g. Mair and Abe, 2008), and sample biasing associated with different particle size measurement techniques. Careful thin sectioning of epoxy-impregnated samples ensures negligible alteration and reliable grain size data in two dimensions (e.g. Panozzo, 1982). Both sieving and laser diffraction provide indirect measurements of spherically equivalent particle size distributions. Laser diffraction particle size analysers, however, ensure much more effective sampling strategies because they cover a wide size range, need short analysis time, and require very small amounts of material (e.g. Beuselinck et al., 1998), thus facilitating very detailed studies of particle size distributions in fault zones. Laser diffraction particle size analysers provide a wide variety of operating procedures that include the use of wet and dry dispersion units, different dispersant agents, and ultrasonication to aid sample disgregation and dispersion. The selection of the pump speed, the length of the measurement time, the number of measurement runs etc., raises the question of their influence on measured results. Sample ultrasonication can aid particle disgregation by collision (e.g. Blott et al., 2004) and consequently it can produce a possible bias on particle size distributions, particularly on fault core rocks where particles are mechanically weakened by fracturing. To further investigate this, we adopted a twofold working strategy that includes the study of poorly cohesive carbonate cataclastic rocks in terms of their internal fabric, and of the response of such a fabric to different operating procedures in laser diffraction particle size analysers. We sampled cataclastic rocks in the core of the active Assergi extensional fault system (e.g. D'Agostino et al., 1998), which bounds to the south the Gran Sasso Massif in the Central Apennines, Italy (Fig. 1). The complex structural architecture in the cataclastic fault core of this fault system is characterized by incipient calcite cementation at the fault core-footwall damage zone transition, which allowed sampling for thin section analysis, and by loose breccias bounding intensely fractured shear lenses, which were sampled for laser diffraction and morphometric analyses. Thin section analysis was used to provide the 2D particle size distribution of the cataclastic fabric, aimed at investigating the role that abundant intragranular fracturing may play on data obtained by laser diffraction particle size analysers. Particle shape analysis was also used for morphometric grain characterization before and after laser diffraction analyses. Our results indicate a significant dependence of particle size data on the adopted analytical procedures in laser diffraction granulometry. 2. Thin section analysis A representative thin section image of the cataclastic fabric in the Assergi extensional fault system is shown in Fig. 2a. A typical feature of the cataclastic fabric is intense microfracturing characterizing coarser clasts embedded in a finer matrix where strongly comminuted shear bands are abundant (Fig. 3). Gray scale digital microphotographs of four selected sub-areas were acquired for automated particle size distribution analysis by the Optimas 6.51? image analysis software. Both actual and manually modified particle size distributions were computed. In the first case, manual retouching of the automatically detected fabric was used only to separate touching particles (Rawling and Goodwin, 2003; Storti et al., 2007). In the second case, retouching consisted of manually splitting larger clasts into constitutive fragments by exploiting all intragranular fractures clearly resolvable in the parent clasts. Plotting particle numbers versus size in bilogarithmic graphs allowed us to calculate the corresponding two-dimensional fractal dimensions of both distributions as the slope of the best fit lines (e.g. Blenkinsop, 1991). Cumulative particle size distributions obtained by merging data from the four sub-images are shown in the bilogarithmic graph of Fig. 2b. Although producing an exhaustive scale-independent analysis was not our purpose in this work, data are well aligned along a best fit line in the 10 mm to 2000 mm size range, having a slope D ? 1.52, which is quite similar to what obtained from wider observation ranges (e.g. Sammis et al., 1986). When transformed into a 3D value by adding 1 (Turcotte, 1986; Sammis et al., 1987), D ? 2.52 fits very well in the range of fractal dimensions obtained from natural bulk cataclastic bodies (e.g. Sammis et al., 1986; Billi and Storti, 2004). Results after manual splitting coarser particles show a still satisfactory fit in a bilogarithmic graph, with a best fit line slope D ? 1.87 (Fig. 2c). The corresponding three-dimensional value D ? 2.87 falls in the typical range of localized cataclastic shear bands (e.g. Marone and Scholz, 1989; Storti et al., 2003; Billi and Storti, 2004). Ключевые слова: laser diffraction, geological, irregular object, wet, great number, equivalent diameter, probe displacement, ner size, text, wilson, journal, performed, dispersant, granulometric curve, air pressure, dispersion, long test, numerical simulation, curve, size range, particle number, merging data, probe, signicant, tullis, total, sub-sample aliquot, recirculation, roma tre, long, diameter, mechanical, particle size, analysis, cataclasis, dispersant liquid, geological society, fractal, peak, blenkinsop, modal peak, progressive subtraction, ultrasonication intensity, speed, bilogarithmic graph, operating procedure, laser obscuration, wet dispersion, dimension, sample aliquot, structural, applied, decrease, ?rst, averaged, coarser fraction, aspect ratio, size distribution, journal structural, sample, comparison, displacement, journal structural geology, mair, laser, balsamo journal, pump, fault rock, doi, diffraction particle, dry, ri, sammis, surface tension, convexity, data collected, smaller, tap, aliquot, parent clast, sample ultrasonication, tectonophysics, performed measurement, number, particle size distribution, geophysical, cataclastic rock, fractal dimension, procedure, dispersion unit, sub-sample, graph, data averaged, coarse particle, analytical procedure, marone, fabric, acquired, clast, image, sieve analysis, granulometric, ?ner, breccia, fossen, bulletin, society, rock, analysing aliquot, result, cataclastic, shear, balsamo, elsevier, measurement, deformation band, fragmentation, malvern instruments, analysis performed, nature, pump speed, fast failure, compound clast, data, data acquired, operating, zone, circularity, ultrasonication probe, distribution, equivalent, curves averaged, wa, structural geology, test, water, storti, sample biasing, fault zone, material, nova, fault, geology, unit, dry dispersion, carbonate, gouge, ratio, volume, panozzo, evans, environmental condition, rpm, deformation, journal geophysical, signicant increase, particle, liquid, storti balsamo, shape, rst measurement, volume weighted, average curve, dry measurement, time, performing measurement, mm, coarsest particle, decalci?ed tap, recirculation time, theoretically predicted, close similarity, sample recirculation, decalci?ed, trend, volume percentage, size, ultrasonication, close, measurement time, subsidiary peak, modal, cataclastic fabric, optical property, diffraction, sperazza, billi, ner particle, inset graph, boer, tap water, range