Theuma slate (“Fruchtschiefer”)
Theumaer Fruchtschiefer is a highly valued building stone in the construction industry. This metamorphic slate is known for its good workability and excellent technical properties resulting in a wide range of applications, for example as façade panels, floor coverings, masonry, monuments or gravestones. In addition, a wide variety of processing techniques for the visible surface, such as bush-hammered, flamed, ground or brushed, can give the stone a wide range of different looks.
The church of Theuma, consecrated as early as 1456, was built with natural stones from the immediate vicinity and provides evidence of the centuries-old tradition of quarrying as well as the excellent weathering resistance of this rock. The mining of Theumaer Fruchtschiefer is documented to have begun in 1858.
The opencast mine currently covers an area of about 13 ha (Fig. 7a). On the terraced mining floors, natural fissures are drilled into with modern drilling equipment and blasted with black powder. The excavated blocks are finally cut to block sizes suitable for saw.
The size of the blocks varies between 1.8 m x 0.8 m x 0.4 m, i.e., a volume of just under 0.6 m³, and 2.8 m x 1.45 m x 1 m, i.e., about 4 m³. On average, a high price level of 800 €/m³ of rock is achieved for crevice blocks. The final production of the various natural stone products and their sale take place directly at the natural stone plant in Theuma (Fig. 7b).
The total monthly production is about 14,600 t of rock, of which, however, only about 1,200 t of ashlar are extracted. Of these, about 60 % are lost in industrial processing as cuttings, so that the net production at the end is only about 3 %. As a by-product of the quarrying of the natural stone blocks, about 2–3 % of the production can be used for gardening and landscaping purposes, e.g., paving stones. The remaining parts of the raw material that are not suitable for use as ashlar are processed into crushed grain mixtures, so that these components are also completely incorporated into saleable final products and 100% recycling of the material takes place.
Geological situation
The area of the Theumaer slate quarries lies in the contact zone of the Upper Carboniferous intruded granite massif of Bergen. The contact rocks (Fig. 7b), which originated from old Palaeozoic shales, are present as nodular or slate of variable petrographic composition. Essentially, two varieties of the Theumaer slate are distinguished:
-
a blue-grey, finely streaked slate (so called “Fruchtschiefer”, (Qgr) and
-
a mostly greenish, banded and striped to banded slate (Qp), often intercalated with lenses and bands of quartz (Peschel and Franz, 1968).
The finely striped, blue-grey coloured slate (Qgr) is restricted to some narrow zones in the contact zone of the Bergen granite massif. The formation strikes N-S or NNW-SSE and reaches a thickness of about 60–70 m. With increasing depth, increased quartz content can be observed both in the groundmass and in the form of increased bands, lenses and striations. From a depth of approximately 80 − 0 m only the banded slate Qp) occurs, which has a greater hardness than Qgr due to its high quartz content. In addition to both varieties of slate, amphibolite deposits ("black gravels") also occur. These strata-parallel diabase rocks were found in five areas and reach thicknesses varying from 0.2–0.7 m (Peschel and Franz, 1968). Typical for some areas at the Theuma quarries are mylonitic fault zones that strike approximately E-W. In the vicinity of this zones the slate is sometimes bleached and in deeper areas mostly hematite alteration occurs.
Lithological characterisation of the Theuma slate and its physical and technical properties
The schistose rock has a blue-grey, macroscopically homogeneous and dense ground mass, in which dark grey to black prismatic porphyroblasts are intercalated. The typical minerals of this stone are cordierite, which are also responsible for the naming of the slate, as their shape is reminiscent of cereal grains ("field fruits"). The spindle- to cigar-shaped cordierites are well oriented within the schistosity, while the long axes are randomly distributed. The foliation is defined due to an alternation of lighter and darker layers. The darker components are mainly ores (magnetite, ilmenite; Peschel and Franz 1968). According to Peschel and Franz (1968), the macroscopically homogeneous groundmass consists of a slaty to fine scale sericite-chlorite-biotite fabric with grain sizes of about 15 µm. Included are quartz grains of the same size. The cordierite is highly altered and shows in case also signs of pinitisation. Biotite and occasionally larger amounts of muscovite occur in the groundmass without a regular spatial arrangement. In addition, tiny streaky lenses of granoblastic quartz may occur in the schistosity plane (Peschel and Franz 1968).
Geological characterisation of the deposit and its ashlar potential
The area of the Theuma quarry (‘Plattenbruch') syncline is bounded on both sides by faults and divided into sub-areas by further parallel faults. Despite its large-scale tectonic development, however, a characteristic and relatively homogeneous cleavage surface structure is formed, which is defined by three cleavage groups (KS A, KS B and KS C, Fig. 8). The cleavage surfaces of cleavage layer A run parallel to the schistosity, generally dipping at approx. 30–40° to the WNW and consequently correlating with the strike direction of the fault zones laterally bounding the 'Plattenbruch'-syncline. Fissure B shows the same strike direction, but the fissures are much steeper with an average dip of 70–80° to the ESE. The spatial position of the third fissure layer is approximately perpendicular to the strike of the excavation walls and is also relatively steep at 70–80°. The interfaces of the two steeply standing fissure layers are usually not continuous. Frequently, a jumping can be observed, or they end at the fracture structures parallel to the schistosity.
From the schematic representation of the interface structure (Fig. 9) it is clear that, based on the natural interface system, a uniform spreading of well-dimensioned raw blocks is often not possible. Since the fracture structures of fracture set B are oriented at an acute angle (approx. 35°) to the strike of fracture set C, the spread of raw blocks deviate strongly from orthogonality and have to be formatted with high material losses (Fig. 9).
The distances of the interfaces are predominantly formed between 1 m and 2 m within the fracture set B and C, while the fissures developed parallel to the schistosity have smaller distances and are predominantly between 0 m and a maximum of 1 m in size (Fig. 9a). Based on the measured data, an average block size of 1.8 m x 1.4 m x 0.7 m is calculated, i.e., a raw block volume of approximately 1.6 m3 (Figs. 9).
Due to the strong deviation of the interface geometry from orthogonality, the determination of the dimensions of an average block must be considered critically. Therefore, in Fig. 9c, the maximum and minimum block dimensions documented by the Theuma company on the basis of many years of experience are shown as a supplement and comparison. It becomes clear that the average values determined on the basis of the fracture distance distribution correlate well with the average volume of the documented block dimensions, and can therefore be used as an initial estimate of emerging block volumes. Nevertheless, due to the unfavourable fracture geometry, it is very likely that blocks with lower volumes will be produced. This assumption is additionally supported by the comparative consideration of the arithmetic mean and the median (Fig. 9b). Thus, in the case of fracture groups A and C, the median is below the mean value, so that the frequency of small fracture distances predominates and consequently the emergence of smaller block sizes dominates.
The deposit is developed via five terraced levels (Fig. 7), which do not show any vertical deposit walls due to selective mining. The excavation walls are inclined according to the schisosity and are unevenly formed due to many offsets. Under these conditions, it was not possible to determine the raw block values, as this is based on the measured fracture distances along several profile lines. In addition, due to the general outcrop conditions of the deposit, modelling using the 3D-BlockExpert software proved to be difficult.
Concluding observations on the mining of the Theuma slate
Overall, the deposit is determined by unfavourable geological conditions, which are caused by the large tectonic structure as well as by the occurring interface structure. In view of the perspectives of a finite deposit as well as additional considerable restrictions in mining, a targeted and optimal utilisation of the resource is economically imperative in order to be able to extract the Theuma slate in its typical decor and its good mechanical and physical properties in the long term. However, an optimisation and thus an increase in production is difficult in the investigated deposit complex. Various potential changes in the mining process were simulated in order to achieve a possible increase in the block yield. However, a change in the mining method does not achieve any advantages. The main problem is that existing in-situ blocks can be cut and thus reduced in volume.
Building on the natural interface system, it therefore makes sense to consistently pursue selective mining that specifically targets singular large blocks. For each further mining step, a detailed recording and assessment of the interface structure would have to be carried out for individual rock sections in order to detect individual large blocks in the rock body.
Case study Seeberg – Rhaetian sandstone
For centuries, the sandstone of the Seeberg was an important raw material for building activity, especially for representative buildings in the city of Gotha, and was mined in numerous quarries, most of which are now abandoned. There is evidence of quarrying back to the 11th century when the stone was used in the construction of Wartburg Castle in Eisenach. It was also used, for example, in the construction of Sanssouci Palace in Potsdam and the Reichstag building in Berlin.
Today two quarries have been developed, one through the "Kammerbruch" and the other through the "Günthersleber Bruch". The “Kammerbruch” is a relatively small area of 50,000 m², where mining is done by small explosive charges detonated in boreholes. The resulting gangue blocks reach sizes of 1-1.2 m x 1.5 m x 2-2.5 m, i.e. block volumes of 3.8 m³ to 6 m³. The material that cannot be used as a building stone is processed to golden-yellow sand. Part of the "Kammerbruch" is designated as a geotope and should therefore be preserved for a long time.
Geological situation
The Triassic sediments of the Thuringian Basin form a widely stretched basin, which essentially formed during the younger Mesozoic. NW-SE striking fault zones dissect the basin into strip-shaped floes (Henningsen and Katzung, 2006). The Seeberg is part of the Hercynian-striking Eichenberg-Gotha-Saalfeld fault zone that runs through Thuringia and was formed during the Saxonian fracture tectonics. Part of this fault zone is also exposed in the quarry .
At present, the “Kammer”-quarry in production offers a good insight into the sandstone sequence of the Rhaetian (Upper Keuper) and the overlying strata of the Lias (Lower Jurassic). The deposition of the Rhaetian sandstone took place about 215 million years ago mainly under continental, fluviatile-estuarine, and to a lesser extent under brackish-lagoonal conditions (Henningsen and Katzung, 2006). The Upper Keuper formation reaches a thickness of about 40 m in the deposit area (Fig. 10), whereas the exploitable building stone beds of the Middle and Upper Keuper only reach a maximum thickness of 15 m in the entire area.
Lithological characterisation of the Seeberg sandstone and its physical and technical properties
The rocks of the Rhaetian Formation exposed in the Kammerbruch (Fig. 10) represent a yellow to yellow-grey coloured, very fine-grained to fine-grained as well as siliceously cemented quartz sandstone. Brownish limonite precipitates in the form of Liesegang’rings and are frequently observed (Fig. 11). In addition, washed-in plant remains, representing a structural inhomogeneity that is susceptible to weathering. Oblique stratification and flow ripples were observed in sediment structures. The Seeberg sandstone in its grain size-dependent varieties (fine-grained, very fine-grained) is a pure quartz sandstone (approximately 95–98 %). The very fine-grained sandstones also contain insignificant amounts of mica and glauconite (Schwate, 1994).
The physical and technical characteristics shows that the bulk density and porosity are in the middle range for sandstones (Mosch & Siegesmund 2007, Siegesmund & Duerrast, 2014), while the compressive and flexural strength of the rock are relatively high. This is due, among other things, to a very high quartz content especially in the form of the siliceous cement.
Geological characterisation of the deposit and its potential
Currently, four sandstone layers are being mined in the Kammerbruch (Fig. 10), with thicknesses varying between 2 m and 3 m (1st to 4th levels). Within these benches, clays and silts were observed, which can be several dm thick, but are often horizontally unstable and form larger, non-contiguous lenses or gullies.
On the slope of the sandstone layers viable for quarrying, there is an approximately 8 m thick cover of clay and silt layers, in which low-thickness layers of sandstones (dm-range) are intercalated. As a result, a thick layer of overburden must be removed before mining can begin, from which at most small masonry stones can be extracted for gardening and landscaping.
The interface system of the deposit is defined by two dominant fracture systems (KS A and KS B, Fig. 12), which are relatively steep with an average of 80°. Fracture A strikes approximately NNE-SSW, while fracture B is characterised by WNW-ESE striking structures, and thus has the same orientation as the fault zone. Orthogonal fracture geometry is common for sandstones. It causes a uniform dissection of the rock complex, so that well-dimensioned, approximately cuboidal rough blocks can be extracted here.
The steeply oriented planes are mostly rectilinear, but rarely run through the entire rock complex, instead they offset or end at the strata-parallel bearing fissures (Fig. 13). These, on the other hand, are horizontally stable, although their surfaces are increasingly wavy and irregularly developed. At first glance, it is obvious that the uppermost three beds are more fissured than the lowest sandstone layer. Furthermore, on the western side of the quarry near the fault zone, tectonic events in the form of staggered thrusting have been observed, but these are restricted to the uppermost floor and further dissect it.
In this context it is differentiated between interface distances within the upper three levels and those in the 4th level. In the x- and y-direction distances between 0 m and 1.5 m dominate, whereby the distances of the interfaces measured in the x-direction are generally smaller than in the y-direction.
Furthermore, the lowermost layer, compared to the upper layer, shows significantly less cleavage spacing between 0 m and 1 m and, especially in the y-direction relative to the x-direction, shows more cleavage spacing, ranging between 1.5 m and 2.5 m. While average blocks of 1.2 m x 1.4 m x 2.5 m in size and a resulting volume of about 4.2 m³ can be recovered from the main extraction layer, the raw blocks of the uppermost layers only reach an average size of 2 m³. In the entire rock complex, however, the determined medians of all profile directions are generally below the arithmetic mean, so that the frequency of smaller block sizes generally predominates.
Based on the dimensions of the minimum block, this results in an average raw-block potentiality of approximately 79 % for the uppermost levels, while the main extraction bench achieves an overall block potential of approximately 93 %. However, since it is very likely that extraction is usually based on larger volumes, the second calculation was carried out under the conditions of the average block sizes calculated on the basis of the distance distribution for the uppermost levels (0.8 m x 1 m x 2.5 m; 2 m³) as well as the main extraction bench (1.2 m x 1.4 m x 2.5 m; 4.2 m³).
As part of the geological characterisation of the deposit, a model was created for the excavation area illustrated in Fig. 14 using the 3D-BlockExpert software. Based on this illustration, it is clear that a large part of the fractures do not run through the entire rock complex, but are only developed in individual sandstone layers (Fig. 14). However, since the software describes all interfaces as continuous, this creates errors in the modelling that negatively affect the actual existing block volumes. Therefore, only interfaces that pass through at least two sandstone banks are included in the modelling (Fig. 14).
The frequency distribution of all block sizes occurring in the modelled rock complex shows that approximately 50 % of the in-situ blocks have volumes between 0 m³ and 2.5 m³ and a good 13 % are only formed up to a size of 0.5 m³. About 12 % of the in-situ blocks are extremely large, exceeding 10 m³. Based on the data obtained, an average block volume of about 3.8 m³ was calculated. The calculated median of 2.4 m³ is clearly below the mean value, so that the frequency of smaller blocks generally predominates. Overall, strong fluctuations occur within the block size distribution, which are caused by a high number of very small or very large blocks. These can be caused by the aforementioned limitations in the modelling. For example, fissures that are only formed in two sandstone layers are determined as a continuous interface, and thus possibly cut an already relatively small in-situ block a second time. On the other hand, the model does not consider fractures that only pass through one bed, so that large raw blocks may be additionally dissected by existing interfaces in reality. Considering these sources of error as well as the comparative consideration of mean, median and standard deviation, the corresponding values were additionally and restrictively determined for all occurring block sizes with a volume between 0.5 m³ and 10 m³. The average block calculated in this way is 3.2 m³, which is slightly below the calculated mean value of the total data set (3.8 m³). This value correlates with the average block volume determined on the basis of the fracture distance distribution (3.1 m³. In both cases, however, the median is below the mean value, so that the frequency of smaller block sizes generally predominates.
In principle, however, it is important to note that the model is still subject to errors due to the software-related limitations. In order to minimise the sources of error and to model the deposit in the best possible way, and not only to obtain knowledge about the average block sizes, but also to make possible statements regarding the amount of stock and the mineability of individual rock areas, separate modelling of all four sandstone benches would be necessary. As an example, such a model was created using the 4th level, i.e. the main extraction layer, which carries the largest potential of workable material, as well as the 2nd excavation level, which is probably the most dissected.
The modelled area of the 4th level represents a potential mining block with the dimensions 8 m x 15 m x 2.5 m. It is already clear here that the front part, due to stronger fracturing in the x-direction, leads to smaller in-situ blocks than in the rear areas. With the help of the function difference analysis the actually usable material can now be distinguished from the resulting overburden. While the sandstone bank in the front area carries a lot of overburden and only allows the extraction of small bricks or paving stones, the in-situ blocks located in the rear area show great potential for the production of building stones. Exemplarily, the quantified areas were exploited by fitting adapted rough blocks. These represent the average block with the dimensions 1.2 m x 1.4 m x 2.5 m determined on the basis of the gap distance distribution for the 4th extraction layer. Accordingly, 17 raw blocks with a volume of approximately 4.2 m³ could be extracted in the lowest level in the way shown.
Comparatively, the second invert was modelled, representing a potential mining block of 8 m x 15 m x 2.2 m. The more dissected sandstone bench consequently carries fewer large in-situ blocks than the fourth floor. Thus, six larger blocks with the dimensions of the average block from level 4 and 13 average blocks from the upper levels could be recovered from the schematically depicted mining series. For the average block dimensions, a size of only 2.2 m instead of 2.5 m was assumed here in accordance with the height of the excavation floor, so that the average volumes of the blocks must be corrected downwards accordingly compared to the modelling of the 4th excavation floor (1.8 m³ instead of 2 m³ and 3.7 m³ instead of 4.2 m³). Based on these representations, approximately 75 m³ of ashlar can be extracted from this layer (approximately 25 % of the total volume). The production in the second layer is significantly lower at approximately 38 m³ of ashlar (approximately 14.4 % of the total volume).
1.1. Case study Burgpreppach - Rhaetian sandstone
In the past, the ashlars from the Burgpreppach quarry area were mainly used as building sandstone. In addition to well-known buildings, such as Bamberg Cathedral, or modern facade cladding (Bamberg Concert and Congress Hall), this sandstone characterises the entire region around Burgpreppach. Currently, Burgpreppach sandstone is quarried and sold by the Schubert stone factory on an area of approximately 100,000 m² (oral communication from V. Deuster). The quarrying is done similarly to the Seeberg deposit complex with explosive charges that are detonated in boreholes. According to estimates, about half of the deposit in the mountain has been mined (oral communication from V. Deuster). In the trade, the sandstone is known as “Gelber Mainsandstein” (or Burgpreppacher Sandstone) and is used today partly for façade design and as flooring, but mainly as a wall stone for gardening and landscaping (Geyer and Schmidt-Kaler, 2006). In the case of larger raw blocks, the sandstone can also be used to make sculptures. According to current studies, the Rhaetian sandstone will also function as a hydraulic engineering stone in the future (Stein, 2009).
Geological situation
The Hassberge Mountains (Fig. 15) are built up by rocks of the Middle and Upper Keuper, a direct continuation of the Steigerwald, which lies S' of the Main (Geyer and Schmidt-Kaler, 2006). The Rhizosphere layers of the Upper Keuper are preserved in the Hassberge mainly as erosional relicts. According to the established lithostratigraphic division of the Keuper in Lower Franconia, the outcropping building stone beds are counted among the so-called Anoplophora sandstones, whose thickness is generally estimated at 10–12 m. The Anoplophora sandstone is a marine-influenced, with sedimentation occurring under estuarine conditions. These deposits are located in a marine-terrestrial transition zone and reflect the interaction between fluviatile processes and marine influences (Geyer and Schmidt-Kaler, 2006).
The Triassic plates of the South German stratigraphic plateau are hardly dissected by major faults (Henningsen & Katzung 2006). The significant faults as well as large saddle and trough structures all preferentially follow the Hercynian strike direction, i.e. are oriented roughly NW-SE (Geyer and Schmidt-Kaler, 2006).
A characteristic feature of the sandstone in the deposit is the strong siliceous cement, whereas the sandstones from the immediate vicinity contain significantly more clay, are not quartz-cemented and are therefore less suitable as an ornamental stone. The silicification could be due to post-diagenetic, hydrothermal influences, such as the supply of silicate solutions along tectonic fault paths. Such a phenomenon is described by v. Gehlen (1956) and Koch et al. (2003) for sandstones of the Keuper in the Nuremberg area. Here, the rocks are influenced by a fault system and were strongly hydrothermally silicified.
Lithological characterisation of the Burgpreppacher Rhätsandstein and its rock physical and technical properties
The Burgpreppach Rhaetian sandstone is characterised by a light grey to yellowish brown, sometimes also reddish colour. Overall it is fine-grained, with grain size changes occurring in part, which are formed as layer boundaries. In these areas there are coarse-grained sandstone layers, in which individual grains of predominantly grey quartz can be up to 5 mm in size. In individual sections of the deposit, cross-stratification is also formed over shorter distances of a few metres. The typical appearance of the sandstone is accentuated by a limonitic brown colouration (Fig. 16a).
The rock generally does not show any noticeable signs of intensive alteration. Only rarely are slightly higher contents of kaolinite occur. The rock consists of 98–99 % of quartz. The proportion of polycrystalline quartz, which may show lower strength under mechanical or chemical stress, is less than 5 %. The mineral grains usually show a secondary quartz accretion fringe (Fig. 16b). The frequently occurring grain to grain contacts causes a high mechanical strength of the rock. In some cases, clusters of quartz grains up to 5 mm in size have formed, which controls an additional mechanical stability.
The feldspar content is less than 1 %. The aggregates are fresh and show no clear signs of alteration. Sporadically, kaolinites as well as light mica and biotite are present, but they are so sparsely distributed that no slip planes can form under mechanical stress. The proportion of limonite and haematite has a colouring effect and is essentially encased as original grain adhesion to the primary quartz grains by the secondary quartz (Stein, 2009).
Geological characterisation of the deposit and its ashlar potential
Production takes place on the 2nd mining level, which forms the actual core rock of the
deposit. The natural stones are overlain by approximately 3 m thick, thin-banked and obliquely bedded sandstones, which are partly formed as platy sandstone clasts and contain grey clay lenses. The sandstones here are strongly limonitically interbedded and deconsolidated and contain individual benches that are more strongly silicified. Occasional lenses with a high kaolinite content also occur. The uppermost bed is considered overburden and must be removed before actual quarrying, similar to the case of the Seeberg sandstone, but here it reaches a significantly lower thickness of approximately 3 m. Under certain circumstances, these sandstones can still be used for the production of smaller wall stones or polygonal slabs for gardening and landscaping. If their clay content were lower, it would also be possible to grind up the rock.
The rock is strongly fractured. The horizontally layered fractures are usually boundaries of sedimentary sequences and usually have distinct layers with a higher proportion of clay minerals from cm to dm in thickness. The degradable bench heights here range from one to three metres. The spatial position of the developed interface structure was measured on the active excavation floor at three perpendicular deposit walls (Fig. 17, 18). The fracture surfaces determined in the y-direction (Fig. 17, 18) show a clear preferred direction with an almost N-S orientation (fracture layer A), perpendicular to the strike of the mining front. In addition, two further fracture sets are indicated (KS B and KS C), but these are rarely cut on this profile. These two sets, which were increasingly measured on the walls oriented in the x-direction, form an angle of approximately 50°, with an angle bisector perpendicular to the excavation front (Fig. 17b, 18). In addition, a fourth fracture set occurs here, which runs approximately perpendicular to the profile line (KS C).
Altogether, four preferred directions of the fracture surfaces can be determined, which run approximately N-S, E-W, WNW-ESE and WSW-ENE. The respective mining walls are oriented approximately N-S and E-W and thus well adapted to the strike of two fracture structures, which are consequently used as natural dividing surfaces in mining. However, despite the ideal orientation, more acute-angled in-situ blocks are formed during mining, which cause a high material loss when formatting to a net block (Fig. 17, 18).
The bedding surface system as a whole is heterogeneous and very changeable. While individual fracture sets are stable and run in a straight line, on the other hand it was often observed that fissures bend, jump, run out or end at horizontal interfaces. Bending and jumping of the fissures occurs especially at the horizontal interfaces as well as at the vertical fracture structures in the x-direction. Due to the heterogeneity of the interface, no reliable predictions can be made.
The average block size resulting from the joint sets and its distribution reaches a volume of about 1.5 m³, but must be considered critically because the calculation is not based on an orthogonal reference system. The determination of the raw blocks on the basis of the minimum block volume defined by Singewald (1992) results in a value of ~ 88 %. In contrast, the second calculation was made on the basis of the average values resulting from the fracture distance distribution, which yields an average value of about 74 %.
The termination behaviour of the fracture is also problematic, as the 3D-BlockExpert software in its current state of only determines discontinuties as straight and continuous surfaces. Figure 17 shows the course of the interfaces on the respective excavation walls as well as the extrapolated or simplified course of the fractures that were used in the modelling. This necessary but serious simplification means that the model can only represent reality to a limited extent.
The frequency distribution of all in-situ block sizes occurring in the modelled rock complex is dominated by volumes between 0 m³ and 2.5 m³ (~ 40 %). With a volume < 0.5 m³, a good 16 % of the blocks are not suitable as ashlars. A high number of extremely voluminous blocks (~ 50 m³) results in larger arithmetic mean of approx. 7 m³, which cannot reflect reality at all and probably results from the strong simplification of the model. Therefore, the mean value of all occurring in-situ blocks with a volume between 0.5 m³ and 10 m³, i.e. without considering the extreme block sizes, was calculated additionally and comparatively. This is by far lower with a value of 3.8 m³, but is still clearly above the average block volume (1.5 m³) determined on the basis of the joint distance distribution. In principle, it is to be expected that the average block sizes are below the calculated value from the modelling, as it is very likely, due to the heterogeneity in the interface system, that in addition to the fracture structures recorded at the outer walls, further fractures occur in the entire rock complex that are not visible at the excavation fronts.
1.2. Case study Treuchtlingen – “JURAMARMOR-limestone”
JURAMARMOR limestone is quarried (Fig. 19) in the Treuchtlingen-Solnhofen-Eichstätt triangle of towns and has been used on a large scale as a building stone since the beginning of the last century. The easy extraction of the stone was an important settlement criterion in earlier times for the construction of castles and houses in the upper classes. Thus, since the late Middle Ages, the ashlar was highly valued, especially in the episcopal town of Eichstätt, as a result of which the townscape is still characterised by buildings made of JURAMARMOR today. The first major building tasks were the façade of the cathedral, the staircase in the residence and the Willibaldsakristei at the cathedral, which were built from JURAMARMOR.
A few decades ago, a second main mining area was established in the Petersbuch-Erkertshofen-Kaldorf area (Hafner, 1989). Currently, the rock is extracted in about 30 quarries with mechanical use and blasting after removal of non-usable surface layers in open-cast mining. After the blocks have been split off, they are divided according to size and geological conditions and sawn into 1–4 cm thick slabs in the factory and then mostly ground or polished. The colours of JURAMARMOR range from white to golden yellow to grey-blue.
Geological situation
The Franconian Alb, one of the dominant mountain ranges of the South German stratified plain, is predominantly built up by rocks of the Jurassic formation. The Jurassic units in Franconia are altogether less than 400 m thick and are traditionally subdivided into the Black, Brown and White Jurassic (Lias, Dogger, Malm). For the present work, the strata of the White Jurassic, which are the most prominent in the Southern Germany and contain the deposits of the JURAMARMOR as well as the Solnhofer Plattenkalke, are particularly relevant. These are limestones which, due to their resistance to weathering, form a morphological escarpment, the so-called Albtrauf. The bedded, partly dolomitised limestones, especially of the younger Malm, are intercalated with larger unstratified blocks. Here are former reefs formed from algae and siliceous sponges in the then prevailing shelf sea environment (Henningsen and Katzung, 2006).
A special geological feature of the JURAMARMOR is its large-scale occurrence. Thus, the benches extend over an area of 22 km² across the southern Franconian Jura mountain range with more or less the same stratification and stratification sequence (Fig. 20). Therefore, a consistent quality can be guaranteed even with larger demands for building stones.
Lithological characterisation of the Jurassic limestone and its physical and technical properties
The trade name JURAMARMOR is misleading in the geological sense, since from a petrographic point of view it is a sedimentary rock, a fine-grained limestone. The rock found in the quarry under investigation is of the grey JURAMARMOR variety, although areas with a distinct yellow colouration have also been observed in places. The rock is overall predominantly light grey, micritic and shows dark grey to partly slightly reddish weathering features (Fig. 20). Furthermore, fossils like sponges, algae and ammonites can occur (Ritter-Höll, 2005). Washout and sinter formation occur especially on dominant fissure or stratified surfaces (Fig. 20b). In addition, coarse crystalline calcite veins penetrate the rock.
JURAMARMOR is comparatively soft, but in comparison to other softer rocks, it is characterised by the excellent polishing of its surface, which it owes to the extreme density of the material. This is why this natural stone is also called “marble”. The material density, the polishability, the numerous colour and structure variations and the good abrasion resistance result in the decisive properties for the versatile uses of the stone. The natural stone also has a particularly good heat storage effect, making it additionally suitable as an ideal floor covering for hot water or electric underfloor heating systems.
The JURAMARMOR has a high compressive and flexural strength, which is in the upper range of carbonate rocks.
Geological characterisation of the deposit and its ashlar potential
The JURAMAMOR deposit is developed as a kettle quarry. The thickness of the exploitable rocks is about 12 m with a current mining area of about 1500 m². The rock sequence is divided into individual benches up to 1.6 m thick, which are separated from each other by several cm-thick marl layers. On the slopes of the limestone beds there is a dolomite layer about 1–2 m thick, which has to be removed before the valuable rock can be quarried.
The stereographic projection of the fracture surfaces shows three preferred directions oriented approximately NNW-SSE, WNW-ESE and SW-NE. In addition, the strike directions of the deposit walls shown in Fig. 19 are illustrated, whereby the two mining fronts in the x-direction represent fracture surfaces or fracture zones. They run at an acute angle to each other and can give a first estimate of the geometry of the resulting in-situ blocks. The vertical interfaces mostly run in a straight line, but jump or end predominantly at the layer-parallel discontinuities. The spacing distribution of the interfaces is characterised by a large variability of the fractures occurring in the x- and y-direction, while in the z-direction fracture spacing between 0.5 m and 1 m clearly predominates. On average, blocks with a volume of about 1.6 m³ are to be expected according to the computationally determined characteristic values. Here, too, the determination of the average block must be viewed critically, as the calculation is not based on an orthogonal reference system. The determination of the raw block on the basis of the minimum block defined by Singewald (1992) results in a value of ~ 76 %. In contrast, based on the average blocks resulting from the fracture distance distribution, a value of ~ 66 % is calculated. Although this value is already lower, it must be assumed that the actual raw block average is still significantly lower due to the heterogeneities in the fracture system (Fig. 21). The first model describes the front part of the deposit, which is bounded by the two fracture surfaces oriented at an acute angle to each other. Figure 22 describes the course of the interfaces on the excavation walls (Fig. 22a), the simplification or extrapolation of these (Fig. 22b) and the model generated with 3D-BlockExpert. Well-dimensioned raw blocks can be extracted from the rock complex, as more or less orthogonal fracture geometry is given. Nevertheless, it must be noted that the probability of further interfaces in the y-direction, which have not been recorded here, is very high. However, these cannot under any circumstances be oriented at an acute angle to the other interfaces here, as they would otherwise have had to become apparent on the excavation walls examined.
The frequency distribution of the block sizes resulting from model 1 is dominated by volumes between 0 m³ and 2.5 m³. Extremely large blocks cause an arithmetic mean of 6.8 m³ that is clearly too high. Two things are responsible for this: part of the model represents an area of rock that has long been mined, so that some blocks must be reduced by at least half their volume. In addition, it is very likely that there are other discontinuities in the y-direction that are not covered by this model. Deducting the extreme block volumes (< 0.5 m³; > 10 m³) results in an average value of 3.4 m³. Even if this value is already lower and thus more realistic, the result must be viewed critically due to the lack of information about further potential fissure areas.
The second model describes a part of the current mining area. Since there is only information on fracture surfaces in the y-direction, a theoretical fracture distribution corresponding to the natural interface system is added at depth (Fig. 23). It describes fracture surfaces that run roughly parallel to the mining front and are characterised by distances of the mean fracture distance distribution in the x-direction. However, it must be noted that this model can only reflect reality to a limited extent and the deposit is most likely characterised by a much more complex fracture geometry.
The frequency distribution of all block sizes occurring in the modelled rock complex is determined to about 80 % by volumes between 0 m³ and 2.5 m³. In contrast to model 1, large blocks with > 10 m³ occur only occasionally, so that the arithmetic mean with 1.9 m³ and minus the extreme values with 2.3 m³ provides a realistic estimate.
Case study ROSA PORRIŇO
The ROSA PORRIŇO represents the best known granite in the Spanish mining region of Galicia (Fig. 24). On a total area of 1 km² here, monthly 1.300 m³ of rock was extracted by the company Canteiras de Granito (Blockgándara S.L.), which corresponds to a block yield of about 20 %. A further 30 % is used for the production of building or masonry stones, and the remaining 50 % is processed into crushed rock material.
The dimensions of the formatted raw blocks vary between 0.8–2.1 m x 1.4–1.85 m x 2.8–3.3 m, thus corresponding to volumes ranging from 3.1 m³ to maximum values of 12.8 m³. In general, however, the block sizes always depend on the machine equipment as well as on the requirements of the end product. Mining is done by blasting and diamond wire saws. The rock complex can be described by three spatial directions): "el norte", "el levante", "el andar". Due to a weak arrangement of biotites or veins, there is a slight preferred orientation in the granite complex, which is described by the surface "andar". Perpendicular to this is the surface "norte". Both are usually drilled and blasted. The third face ("levante") is usually sawn.
Lithological and rock characterisation of ROSA PORRIŇO
The stone ROSA PORRIŇO represents a slightly reddish, medium to coarsely crystalline, massive granite, which has a directionless granular structure in hand specimen (Fig. 25). In general, the average grain sizes in ROSA PORRIŇO 1 (Fig. 25a) are slightly larger than in ROSA PORRIŇO 2 (Fig. 25b). The reddish colour, which is already referred to in the trade name of the rock, is due to the high percentage of potassium feldspar. In addition, the rock is rich in quartz and plagioclase and contains biotite as dark components, which, however, have smaller grain sizes than the other minerals (Fig. 25). In places, fine-grained aplite veins while in a second deposit of the ROSA PORRIŇO granite, fine-grained mafic enclaves occur.
Geological characterisation of the deposit and its ashlar potential
ROSA PORRIŇO deposit complex 1
The ROSA PORRIŇO granite deposit 1 preserves the largest occurrence of this granite and is developed via numerous mining levels. The area can be divided into two sub-areas: the core areas, which mainly contain intact rock with the highest ashlar potential, and the peripheral areas, which are mostly highly fractured and mainly used for crushed stone mining.
The core area represents the current mining floor at the time of the site studies (Fig. 24) with vertical, on average up to 10 m high deposit walls, which are mostly arranged at right angles to each other. On a profile section of about 80 m, 14 parting planes, a fissure zone and an aplite vein of about 5 cm thickness were recorded on the excavation walls, so that this area is only slightly dissected overall. The fissures mostly run in a straight line, but only partially through the entire rock complex, while a curvature or branching of interfaces was observed very seldom.
The orientation of the interfaces shows three preferred directions, which strike about 080°, 100° and 170° and are predominantly steep with an average dip of about 80°. In addition, diagonal fracturing was recorded subordinately. The excavation walls were laid out according to the strike of the fracture layers oriented perpendicular to each other, which is to be evaluated positively with regard to the extraction of material suitable for quarrying. As a result of the natural interface system, irregularly formed to acute-angled in-situ blocks can be expected. A similar fracture geometry, characterised by three fracture layers oriented at an acute angle to each other, but partly with high variability in orientation, was determined over the entire deposit area.
Two models were created using the 3D-BlockExpert software (Model A and Model B). The position of the modelling was chosen in such a way that the majority of the fracture surfaces occurring on the excavation walls are captured. Due to the software related limitations, all fractures are represented as continuous interfaces, even if their course is not apparent on the excavation walls. However, since the fractures in a granite only rarely terminate at (sub-) horizontal interfaces and it is therefore very likely that they open further in the course of block excavation, the limitations mentioned only play a minor role in this case. In addition to recording the fracture structures, the course of an approximately 5 cm thick aplite vein was also modelled as a parting surface. Apart from an undesirable change in decor, which lowers the quality of the raw blocks, it is also possible that the rock breaks at this zone of weakness. The rock complex is clearly disturbed by four closely spaced partitioning planes in the central area, while to the left and right of these are well-dimensioned in-situ blocks. In accordance with the prevailing fissure geometry, these are not cuboidal, but in part strongly heterogeneous in shape.
The result of the second modelling shows an approximately 1 m wide fissure zone in the middle to rear area, which causes a strong dissection of the rock complex. In the front area, on the other hand, there are large in-situ blocks, which again show irregular shapes as a result of the natural interface geometry. About 40 % of the in-situ blocks resulting from model A are only between 0 m³ and 2.5 m³ in size, while in model B a similar percentage is given for block volumes larger than 35 m³. The proportion of these block sizes in model A is about 25 % (> 35 m³). The calculated mean values result in 36 m³ for model A and 50 m³ for model B, while the medians of 8 m³ and 24 m³ are both significantly below the arithmetic average.
Looking at the frequency distribution of the raw block sizes determined in the block storage (Fig. 26a), it becomes clear that a minimum volume of about 7 m³ is obvious (Fig. 26b). Consequently, this value can be regarded as an applicable requirement for the dimensions of the raw blocks to be extracted. This shows that about 48 % of the in-situ blocks located in deposit area 1 and about 23 % of the in-situ blocks located in deposit area 2 are highly unlikely to be processed into ashlars.
Since the developed models represent the rock complex realistically, it is possible to align the extraction with the course of the fracture surfaces in order to avoid any further processing. Finally, the usable material can be converted into raw blocks in the best possible way through targeted mining planning.
The singular large blocks deviate predominantly from the cuboid shape, so that their material can only be used in a limited way. For clarification, a comparison of two in-situ blocks: The first describes an irregular block located in the right front part of model A (brown, Fig. 27a), while the second represents a nearly rectangular block located in the left rear area (yellow, Fig. 27b). In each case, excavation slices with a thickness of 1.5 m, corresponding to the average block height (Fig. 26b), were created. In this way, both in-situ blocks show a comparable volume. However, while only three average blocks (3.2 x 1.9 x 1.5 m, Fig. 26b) can be extracted from the first one, it is possible to extract four blocks in the second case, so that the production is higher and the amount of overburden is lower.
Different mining approaches for the current extraction can be simulated through targeted planning steps (Fig. 28, 29). However, it must be noted here, especially from an economic point of view, that shifts in the arrangement of the raw blocks not only influence their possible number, but also the necessary processing work. Thus, although a higher number of blocks is realised in option B, their formatting is associated with an increased amount of work (relative cutting length) (Fig. 28, 29).
By assessing successive excavation volumes, prognostic statements can now be made about the number of recoverable raw blocks from the two rock complexes investigated. In relation to the total rock volume considered, a possible block yield of about 30 % can thus be predicted.
In order to explain the relevance of the interface structure with regard to the possibilities of quarry stone extraction in the investigated deposit in more detail, two characteristic marginal areas of the deposit are examined comparatively in the following. Area 1 shows a highly dissected rock complex characterised by two dominant fracture sets striking about 035° and 105° and dipping at an average of 70°. The formation of the interface system is homogeneous with respect to its orientation and thus provides good conditions for the extraction of ashlars. However, the low fracture spacing (< 0.5 m), which dominates the distribution within the approximately 035° striking fracture structures (> 75 %), has a limiting effect. Under these conditions, the investigated area can only be used for the extraction of fracture material. For this type of use, however, the given dissection is positive, as many natural separation areas already exist and thus less effort has to be made when extracting the materialArea 2 is also highly dissected Even though two preferred directions, sweeping about 045° and 150°, are recognisable, there is a high variability in terms of incidence angles and orientations in the whole. The separation distances, on the other hand, are predominantly between 0.5 m and 1.5 m and thus more widely spaced than in area 1. However, the aforementioned scattering with regard to the orientation of the interface is problematic, if selective extraction were possible, ongoing analyses of the interface system were necessary and material losses as well as time and machine input for the complex formatting of the extracted blocks had to be included. Under these cost-intensive conditions and the low price level of granite, a quarry stone extraction is consequently not economical even in the area 2. Profitable quarrying could only be realised if correspondingly high market prices could be achieved.
Concluding remarks on the ROSA PORRIŇO case study
On a total area of 1 km², 1,300 m³ of ROSA PORRIŇO granite are extracted every month, which corresponds to a block yield of about 20 %. The raw blocks stored on the mining floors have an average volume of about 9.5 m³ upwards. In order to achieve an appealing cost-performance ratio in the extraction of the ROSA PORRIŇO, extraction of ashlars can only take place in the weakly dissected core areas. Since the rock areas to be investigated have all the prerequisites for realistic modelling using 3D-BlockExpert, it makes sense to obtain such information in advance using models.
In this way, prognostic statements can be made about the resulting block sizes and numbers in progressive mining. By implementing potential mining plans, it is finally possible to achieve an increase in production in mining. In addition to supporting current mining planning, it is also possible to react directly to the needs of the market through targeted mining.
Case study GRIS ALBA
The GRIS ALBA granite deposit (Fig. 30) is located north of the town of Cañiza (Pontevedra province) and east of Porriño. The rock has been actively quarried at this site for about 19 years. Of the material moved, a block yield of about 70 % is recorded, while another 20 % is processed into small blocks. The average block sizes have a volume of about 7 m³. The dimensions vary between 2.5-3 m (length) x 1.5-2 m (width) x 1.3-2 m (height). But in general, the block sizes always depend on the mechanical equipment and the desired end product. Dismantling is done by blasting and by using the wire saw.
Lithological characterisation of the GRIS ALBA and its technical properties
The natural stone GRIS ALBA represents a grey to white, medium-grained massive granite with a homogeneous grain size and a slight foliation by intercalated biotite (Fig. 31a). The predominantly white to light grey colouration is due to a high percentage of quartz and plagioclase. The mineral grains occasionally reach sizes of 0.5-1 cm. In contrast to the ROSA PORRIŇO granite, potassium feldspar plays only a minor role. Compared to the other minerals, the biotites have smaller grain sizes.
The rock shows only occasional limonitic discolouration. A typical feature of the GRIS ALBA granite is the frequent occurrence of aplitic and pegmatitic dykes and mafic enclaves (Fig. 31b). The veins reach thicknesses of a few to a maximum of 25 cm, while the mafic enclaves can often be up to half a metre in size. Both elements are controlling the quality and décor of the rock.
Geological characterisation of the GRIS ALBA deposit
The GRIS ALBA granite deposit is accessible via seven excavation levels and probably represents the roof of a pluton (Fig. 30). The excavation walls of the 2nd to 7th levels are arranged perpendicularly and partly at right angles to each other. Area 1, an approximately vertical deposit wall with a maximum height of about 20 m is given (Fig. 32a). The orientation of the fractures shows three preferred directions, striking about 045°, 115° and 175° (Fig. 32b). The strike of the excavation wall is parallel to the orientation of the dominant fracture set (~ 115°), which is positive with regard to a building stone. Both horizontal and vertical partings are formed, which are only conditionally rectilinear. Pointed-angled outcrops are the result of the variable orientation of the interfaces.
Over short distances, there are clear differences in the distance distribution, with gap distances between 0.5 m and 2.5 m. On average, blocks with a volume of about 4 m³ are to be expected according to the calculated characteristic values. Considering the medians, which are in part significantly below the arithmetic mean, a general promotion of smaller block volumes must be expected here. Thus, the determined average volume of 4 m³ can hardly meet the currently applicable requirements for raw blocks (average 7 m³).
The calculation of the raw block potential based on the currently applicable requirements for raw blocks (average 7 m³) with edge dimensions of 2.9 m x 1.6 m x 1.5 m results in a value of 59 % (Fig. 33). However, this must be considered far too high, as the heterogeneity in the interface system often causes acute-angled in-situ blocks, which lead to material losses.
The core area of the deposit is hardly dissected by fracture planes. The fissure traces are only partly continuous and mostly show an irregular strike. They are often bent and strongly branched, show opening widths in the mm to cm range and partly show clear rock outcrops. Such irregular courses of fissure surfaces pose a problem in mining planning, as their course is difficult to predict.
The excavation walls are preferentially oriented in a N-S direction, while individual walls, mostly representing fissure surfaces, are oriented perpendicular to them (Fig. 34). These fissures show a high degree of continuity in their course, in contrast to the previously considered parting surfaces. They are mostly fissure zones that show a distinct regularity and occur approximately every 10–15 m. The main mining wall (N-S) was thus laid out perpendicular to the strike direction of the dominant fracture set, which is positive with regard to the extraction of material capable of working stone.
The schematic representation (Fig. 34) clearly shows that the fissures are irregular in their course, especially in the left area. By extrapolation of the fracture surfaces, the resulting models can only represent reality to a limited extent. In addition to the fracture surfaces, two 15 cm thick aplite veins were modelled as interfaces in subarea B, because these act as a zone of weakness on the one hand and are not desirable in the stone production on the other.
The modelled areas show a stronger dissection in subarea A, which is caused by a thick fissure zone within the right side (Fig. 35). As a result, the extraction of ashlars within this area is only possible to a limited extent. In the left half, on the other hand, there are large in-situ blocks with a maximum volume of 300 m³. Thus, on the basis of the model, selective mining that concentrates on extraction in the left-hand area of the rock body makes sense. However, even here there are limitations in the use of the material because the orientation of the interfaces causes an irregular shape of the in-situ blocks.
Area B is strongly dissected, especially in the central part, by the occurrence of a fissure zone and two aplite veins (Fig. 35). To the side of these, on the other hand, lie extremely large in-situ blocks with high ashlar potential. From the orientation of the fissure clusters to each other and the orientation of the excavation wall, approximately rectangular cuboids can be derived from the model for area B. The fissure zone and the aplite dikes are highly dissected. Thus, on the one hand, the extraction of blocks of very large volumes is possible, but on the other hand, the extraction of the material is made considerably more difficult, as no natural separation surfaces can be used in the extraction. This in turn leads to an increased demand for machinery and labour.
Projections of the number of resulting average blocks (7 m³ with 2.9 x 1.6 x 1.5 m) from the investigated rock complexes based on the consideration of successive excavation slices yield the following results: For subarea A, 85 blocks can be predicted, which means a block yield of about 39 % in relation to the total rock volume. In subarea B, on the other hand, this is about 64 % with a forecast of 289 blocks. The clear differences in the percentages are primarily due to the formation of the in-situ block shapes. The irregular in-situ blocks from sub-area A limit the use of the available material, so that an increased roughness is unavoidable.
As is clear from the foregoing, the models only reflect reality to a limited extent, which must be considered in Part A in particular. Nevertheless, on their basis it is possible to gain a first insight into the size of the resulting block sizes and thus into the potential of the respective mining beds. In areas of low fracture density or in those where the fractures are straight and can be followed on at least two surfaces, the implementation of a previously prepared mining plan can be carried out without further planning and production increased on the basis of this.
For an increase in production, the detection of highly disturbed areas within this deposit is of great importance, so that they can be identified in the mining process. It was observed that the extraction of raw blocks in the quarry area presented here is associated with an enormous amount of overburden, but the material is not used by 100 %. Bricks or building blocks are not produced and there is no processing into crushed grain mixtures.
If these unfavourable conditions of the interface orientation are associated with narrowly formed fracture spacing, a stone placement in this area can be excluded from the outset. Under these conditions, ongoing analyses of the interface system are necessary to identify suitable areas. Such conditions are given in the core area of the deposit, which is only insignificantly dissected by interfaces. Here, however, the orientation of the interfaces plays an important role and determines to what extent the material can be used or not. Therefore, it is important to determine the course of the fracture surfaces as far as possible and to record the shapes of the in-situ blocks. In this way, production can be increased on the basis of modelling with 3D-BlockExpert via prognostic statements and the implementation of suitable mining planning.