Rapakivi granites

Abstract

Rapakivi granites, especially the famous rapakivis from southern Finland, are some of the most attractive building and decorative stones, used world-wide because of their striking textures and beautiful colours. But rapakivi granites are more than just building material, they have also been studied scientifically for over 100 years to try to explain the formation of their plagioclase-mantled alkali feldspar ovoids, the rapakivi feldspars which are responsible for the unique texture of the rock. Despite all the work which has been done on them, the formation of this distinctive texture is still not fully understood. Rapakivi granite complexes are not just a ‘pretty face’, they are also associated with important tin deposits in Rondônia and Amazonas in Brazil.

Most people interested in geology have seen rapakivi granite used as a decorative stone and wondered how the amazing rings of feldspar formed. This is a good question; the answer still puzzles geologists more than 100 years after rapakivi granite was first studied. If you have not heard the term ‘rapakivi’ before, you may have come across the names ‘Baltic Brown’ or ‘Carmen Red’, which are trade names for rapakivi granite, or wiborgite and pyterlite, rapakivi granite varieties named after their original ‘type’ locality in southern Finland. You might be surprised to learn that the name rapakivi does not relate to the feldspar rings but is Finnish for disintegrated or crumbly rock and refers to the distinctive weathering behaviour of rapakivi granites.

The term was probably first used as long ago as 1694 by the Swedish doctor, chemist and mineralogist Urban Hjärne for the saprophytic (insitu chemically rotten rock) outcrops of granite that occur in southern Finland. The terms rapakivi granite and rapakivi texture were introduced into the international geological literature later on in 1891 by Jakob J. Sederholm in his classic work Ueber die finnlänndischen Rapakiwigesteine. Sederholm transformed the meaning of the word rapakivi from a weathering state to a mineral texture. He described rapakivis as granites containing ovoid phenocrysts (these are 1 to 5 cm diameter crystals set in a finer groundmass) of pink or brown alkali feldspar (potassium-rich feldspar) that are entirely mantled by grey plagioclase (calcium-rich feldspar, commonly oligoclase) forming the rapakivi feldspar, or rapakivi texture (Figs 1, 2).

Figure 1. Polished rapakivi granite. A. Pinkish variety of Wiborgite (trade name Baltic Brown) from the Wiborg batholith in SE Finland, showing seven alkali feldspar ovoids (1-7) within a medium-grained matrix consisting of quartz, mica, hornblende, plagioclase and alkali feldspar. Four of the ovoids (1, 3, 5 and 6) are entirely mantled by plagioclase. Photo: author. B. Beige variety of Wiborgite from SE Finland with large rapakivi feldspar ovoids. Photo: with permission from the Natural History Museum, London. C.Åland rapakivi granite from SW Finland containing only a few rapakivi feldspars (white arrows). Photo: Leif Bergman. D. Pyterlite (trade name Carmen Red) from the Wiborg batholith in SE Finland. The pink alkali feldspar phenocrysts are mainly surrounded by drop-like quartz crystals. Plagioclase-mantled alkali feldspars are very rare in this variety. Photo: with permission from http://www.finstone.com.

Figure 2. Photographs of rapakivi feldspars. A. Beige alkali feldspar mantled by dark greenish grey plagioclase, Wiborgite, SE Finland. Photo: with permission from the Natural History Museum, London. B. White/brown alkali feldspar with carmine plagioclase mantle, Altenberg-Frauenstein microgranite, E Germany. Photo: author. C. Broken pink alkali feldspar mantled by white plagioclase, Hammarudda porphyry, SW Finland. Photo: author. D. white alkali feldspar with greenish yellow plagioclase mantle, Land's End granite, SW England, photo: Author.

The colour of rapakivi feldspar is rather variable, as shown in Fig. 2. The rapakivi feldspars are embedded in finer-grained groundmass, usually consisting of quartz, alkali feldspar, plagioclase, amphibole and biotite. Mantled feldspars occur side-by-side with non-mantled alkali feldspar phenocrysts. In continuation of this work, rapakivi texture was characterized sensu stricto by Atso Vorma in 1976 as:

• 1
  an ovoidal shape of the alkali feldspar phenocrysts;
• 2
  mantling of part of the ovoids by plagioclase shells; and
• 3
  the presence of two generations of alkali feldspar and quartz.

This is not quite the last word on the definition of rapakivi granite, as you will see below.

The age of rapakivi granites varies from Late Archaean to the Tertiary. However, most of the rapakivi-forming magmas are very old, having crystallized between 1800 and 1000 million years ago (Mesoproterozoic). Probably the most famous example of ‘non-Proterozoic’ rapakivi granite is the Lower Permian (c. 290 million years) Lindum rapakivi granite, which belongs to the Drammen granite complex in the Norwegian Oslo region.

Rapakivi granites are most famous from SE and SW Finland because this is where they were first explored in connection with building and decorative stone production. At the end of the nineteenth and the beginning of the twentieth centuries, rapakivi granites were also described from the Ukraine and Sweden, and after the Second World War, a new period of activity commenced with description of numerous rapakivi granite occurrences from the Baltic countries, South Greenland, Labrador, mid-continental and western USA, Venezuela, Brazil, Botswana and several other Precambrian shield areas (Fig. 3). Since then the formation of Proterozoic rapakivi granites and associated anorthosites has been considered as an important tectonomagmatic Proterozoic event in the construction of the continental crust.

Figure 3. Global map showing the distribution of rapakivi granites. Simplified from Fig. 1 in Rämö and Haapala (1995).

Definition and classification

Rapakivi granite is a member of the granite family, which comprises medium- to coarse-grained igneous rocks containing quartz, feldspar and some darker minerals such as mica and amphibole.

Igneous rocks are named by plotting the proportions of the rock-forming minerals in the quartz-alkali-feldspar-plagioclase-foidite (QAPF) ‘Streckeisen’ diagram or by plotting the concentration of Na₂O + K₂O versus SiO₂ (in weight per cent) in the total alkali-silica (TAS) diagram if the rock has been chemically analysed. Both classifications are used to assign names to most of the common types of igneous rocks. Rapakivi granite is not an official classification of granite and does not have its own field in these diagrams. Chemical analyses of rapakivi granites plot in the granite field of the QAP diagram and some in the granodiorite and quartz monzonite field of the TAS diagram (Fig. 4).

Figure 4. A. Nomenclature of plutonic rocks plotted on the QAP diagram. The Finnish rapakivi granites (names of the batholiths are listed in the legend) plot in the granite field. Names for the numbered fields are: 1—alkali feldspar granite, 2—alkali feldspar syenite, 3—tonalite, and 4—diorite, gabbro, anorthosite. B. Nomenclature of plutonic rocks plotted on the total alkali-silica (TAS) diagram. The Finnish rapakivi granites plot mostly in the granite field and some of them in the granodiorite and quartz monzonite field. Names for the numbered fields are: 1—foid monzosyenite, 2—foid monzodiorite, 3—quartz monzonite, 4—monzonite, 5—monzdiorite, 6—monzogabbro, 7—peridotgabbro, and 8—gabbroic diorite. Data are from Tapani Rämö's PhD ‘Petrogenesis of the Proterozoic rapakivi granites and related basic rocks of southeastern Fennoscandia: Nd and Pb isotopic and general geochemical constraints’Geological Survey of Finland Bulletin 1991, 355.

The International Union of Geosciences subcommission on the classification of igneous rocks gave the following definition: ‘Rapakivi granite is a term used both as an adjective and noun, for a variety of granite usually containing amphibole and biotite and characterized by the presence of large oval grains (ovoids) of orthoclase which are usually mantled by plagioclase. Two generations of quartz are often present’.

This definition is very close to the original literature definition given in the introduction and the one recommended for general use. However, geologists often subdivide rocks of the granite family according to their geochemistry and tectonic association, into those generated during the evolution of large fold belts (orogenic; M-, I- and S-types) and those associated with uplift and major strike-slip faulting (anorogenic; A-type). Rapakivis are classified as A-type granites because they have high silica, sodium and potassium, iron to magnesium ratio, and zirconium and low calcium and strontium. The A-type chemistry of the rapakivi granites is particularly characteristic and so there is another current definition of rapakivi granite used by geologists and defined by Haapala and Rämö in 1992: ‘Rapakivi granite is A-type granite characterized by the presence, at least in the larger batholiths, of granite varieties showing the rapakivi texture.’

How to recognize Rapakivi granite

Ovoid, pale red or pinkish alkali feldspar phenocrysts entirely rimmed by grey or pale green plagioclase are the most characteristic macroscopic textural feature of rapakivi granites. The phenocrysts are usually between 1 and 5 cm in diameter embedded in a finer, medium- to coarse-grained groundmass and form up to 40 per cent of the rock. The characteristic colour of the alkali feldspars gives the rock a ‘meat-like’ colour and some rapakivi varieties even look a bit like black pudding. The ovoids are often almost perfectly spherical, and frequently contain concentric zones of inclusions, which are generally finer grain size than the corresponding groundmass phases.

Many of the large ovoids do not develop plagioclase mantles and in some rapakivi granites this most distinctive feature may be very rare (Fig. 1d). The other minerals in the rock are euhedral plagioclase and ovoid quartz phenocrysts usually between 0.3 and 1 cm, and groundmass minerals, readily identified by their anhedral (poorly formed) crystal shape and small grain size relative to the large phenocrysts, comprising alkali feldspar, quartz, biotite, and amphibole, with very small accessory amounts of fluorite, ilmenite, zircon, and apatite.

Formation

Rapakivi granites are plutonic rocks that cooled and crystallised slowly from molten rock in magma chambers situated 2 to 12 km beneath the Earth's surface. They form sheet-like bodies called batholiths that are sometimes more than 100 km across (e.g., the Wiborg rapakivi granite in southern Finland) and have a maximum thickness of about 10 km. These granite magmas originate much deeper down in the crust, as deep as 2 5 km, by partial melting of other rocks such as slates, greywackes and gneisses or by differentiation from mantle magmas with higher proportions of iron and magnesium minerals (as illustrated in Fig. 5). Most rapakivi granite complexes are linked with basaltic dykes, gabbros and anorthosites that are considered to derive from partial melting in the Earth's upper mantle. However, the majority of the rapakivi-forming magmas derive from lower crustal rocks.

Figure 5. Scheme of the magmatic differentiation from gabbro to granite magma in connection with Bowen's reaction series of minerals. The lower row (pale green) represents the series of Fe- and Mg-rich minerals and the upper row that of alkali- and Si-rich minerals. Plagioclase crystallises prior to alkali feldspar (yellow ellipse) if the granite magma cools continuously but the rapakivi plagioclase-mantled alkali feldspars reflect a reverse process.

Crystal size depends largely upon the cooling history of the granite but also on the presence of volatiles such as water, fluorine and carbon dioxide. Rapakivi granites contain two generations of crystals: first, phenocrysts, representing the early stage of magma crystallization in deep magma chambers and reservoirs and, second, much smaller, anhedral groundmass minerals that crystallized at the final intrusion level.

The ascent rate of granite magmas is slow in comparison with basaltic magmas because of their high viscosity, and during the long journey from the original source of partial melting, the granite magma may be stored at various different crustal levels forming magma reservoirs or chambers. Within these ‘route stops’, crystals begin to nucleate and grow. These early crystals have enough space within the magma and enough time to form ideal (euhedral) crystal shapes. Alkali feldspar phenocrysts in granites grew over a long period of more than 10 000 years in deep magma chambers. The phenocrysts record changes of magma temperature, pressure and composition as they grow resulting in textural and chemical zoning on a micrometre-scale (1 micrometre = 1/1000 of a millimetre). The complexity of zoning in many phenocrysts indicates that they have moved around chemically stratified magma chambers and alternatively experienced different temperatures, pressures and chemistries during growth. The application of modern analytical techniques, such as electron probe micro analysis (EPMA) and laser ablation inductively coupled mass spectrometry (LA-ICP-MS), enables geologists to decode the chemical zoning of phenocrysts, reconstruct the crystallization and ascent history of the magma.

Formation of the rapakivi plagioclase-mantled alkali feldspar phenocrysts that are the key component of the rapakivi granites required an extreme change in the physicochemical growth conditions in the magma chamber because the crystallisation of potassium alkali feldspar prior to the calcium-bearing plagioclase feldspar contradicts the usual progressive magma reaction series, the so-called Bowen's reaction series (Fig. 5). The reaction series of mafic (Fe- and Mg-rich) minerals is shown in the lower row and that of felsic minerals (feldspars, quartz) in the upper row of Fig. 5. With decreasing magma temperature the anorthite component (An; CaAl₂Si₂O₈) in feldspar decreases. During continuous magma crystallization plagioclase crystallizes before alkali feldspar (marked by the yellow ellipse in Fig. 5) but the rapakivi texture requires the opposite crystallization trend. Thus the formation of rapakivi texture is still under debate in the scientific community; read on to find out more.

Scientific importance—controversy over the formation of rapakivi texture

As mentioned above, the plagioclase crystal rims on alkali feldspar phenocrysts require the alkali feldspar to have crystallized before the plagioclase, which is contrary to the established physiochemical behaviour of such magmas. So the formation of mantled ovoidal alkali feldspar phenocrysts requires a distinct change in the physicochemical crystallization conditions, such as increasing temperature, pressure loss or mixing with more mafic magma that would stabilise plagioclase in favour of alkali feldspar, allowing plagioclase to nucleate on the alkali feldspar crystals and form the distinctive mantles. The granite journey from source to magma chamber is not a simple process. There are three proposed mechanisms for the formation of rapakivi texture.

• 1
  Mixing of two magmas of different composition. A catalyst for inducing the changes in the physiochemical conditions of the magma chamber may be the influx of ‘fresh’ mafic magma from below, upsetting the status quo of the felsic magma and triggering magma convection and possibly magma ascent. This is supported by the fact that most of the rapakivi granite complexes are associated with mafic igneous rocks. There seems to be a correlation between the frequency of rapakivi feldspars in the rock and the portion of mafic magma involved in the mixing process.
• 2
  Crystallisation of granite melt under conditions involving marked pressure release combined with just a small change in temperature. Some scientists propose that rapakivi textures are formed when hot and dry A-type granite magma (∼780 °C) containing quartz and potassium-feldspar phenocrysts is transported sub-adiabatically (without temperature change) from intermediate depths (15-20 km), where potassium-feldspar and quartz are stable, to upper crustal levels. During this ascent there is no change in magma chemistry and very little change in temperature but a large decrease in pressure, so that conditions become such that potassium-feldspar and quartz are resorbed while plagioclase feldspar becomes stable and precipitates. This would explain the ovoid shape of the alkali feldspars. Multiple plagioclase mantling including anti-rapakivi textures which are commonly observed in Proterozoic rapakivi granites would hint to step-wise adiabatic ascent. However, adiabatic magma ascent by dyke formation seems to be a common process and so the formation of rapakivi texture should be common in many places and during all geological ages. The concentration of rapakivi textures in Proterozoic A-type granites and rareity in younger granites is a problem for this proposal.
• 3
  Synneusis—the drifting together and systematic attachment of crystals in the melt—of plagioclase crystals at the surface of potassium-feldspar phenocryst during magma flow and/or crystal sinking. This process requires the potassium-feldspar to move into plagioclase-rich (mafic) melt batches in order for the plagioclase to form pure and complete mantles around the potassium-feldspar. This process has mostly been excluded but the passive attachment of previously crystallized plagioclase crystals at the K-feldspar surface might have contributed to the plagioclase rim formation in some cases.

In conclusion, rapakivi can be produced by multiple processes and the texture and chemistry of each granite must be considered carefully and interpreted individually.

Economic importance

Finnish dimension stone Quarrying of rapakivi granite began with the exploitation of erratics in southern Finland and the rocks were used for centuries in southern Finland and southern Russian Karelia. In particular, many churches were built with rapakivi from the early eleventh century to the late thirteenth century after Christianity was introduced into Scandinavia.

Rapakivi granite became internationally famous when it was used in the Isaac Cathedral and then the Alexander Column in St. Petersburg, Russia. The Alexander Column is the most impressive monument made of rapakivi granite (Fig. 6). It was deliberately designed to compete with the pyramids and obelisques of the ancient Egyptians, and at 47.5m high (similar height to Nelson's column which is 51.7m but not made of a single stone) the column is a terrific piece of architecture and engineering. Situated at the focal point of Palace Square in front of the Winter Palace, it honours Tsar Alexander I and his victory over Napoleon in 1812. It was designed by Montferrand and built between 1830 and 1834. The various phases of quarrying, working, transportation and raising of the monolith were described by Montferrand in the lavishly illustrated folio volume Plans et détails du monument consacréà la mémoire de l'Empereur Alexandre, 1836. The red rapakivi granite monolith (today's trade name ‘Carmen Red’) was quarried at Pyterlahti, Finland. Hundreds of stonecutters cut a granite block by hand which was over 30 m long and more than 4 m thick. The finished monolith that had to be transported weighed some 600 tonnes. It almost sank when being loaded onto a special vessel for the journey across the Gulf of Finland, forcing workers and soldiers to struggle for 48 hours to save it. On 1 July 1832 the vessel arrived in St Petersburg and 12 days later the monolith was brought ashore. The column rests on granite blocks in the pedestal, underneath which are 1250 pine piles, each 6.4 m long, driven into the soft marshy earth. The column is topped by a sculpture of an angel holding a cross and treading on a snake.

Figure 6. The 47.5 m high Alexander Column in St. Petersburg in front of the Winter Palace. The column is made of Pyterlite—a Finnish rapakivi granite variety which today has the trade name Carmen Red. Photo: Lyubov Neyevina.

With the industrial revolution and the development of quarrying tools, the exploitation of rapakivi granites became easier and more rational. The granite was used for railway construction, bridges, paving and buildings. However, after 1910 the construction of monumental stone buildings declined. A turning point was reached in the 1970s when Finnish rapakivi appeared on the global market and was recognized as an attractive decoration stone world-wide. Since then major buildings, monuments and fountains using it have been installed all over the world. Examples include the Core Pacific City building in Taipei, Taiwan (granite ‘Red Eagle’) by the architect David E. Rodgers in 2001 (Fig. 7) and the fountain ‘Horses of Helios’ near Piccadilly Circus in London by the artist Rudy Weller in 1992 (Fig. 8).

Figure 7. The Core Pacific City shopping centre in Taipei, Taiwan, decorated with Finnish Red Eagle rapakivi granite. Photo: Palin Granit Oy.

Figure 8. Fountain ‘Horses of Helios’ near Piccadilly Circus in London by the artist Rudy Weller in 1992. The basin and edging of the fountain is made of a greenish dark variety of the Wiborgite rapakivi granite with fascinating rapakivi textures (foreground). Photo: author.

At the present time, Finland is a world-leader in granite production with 4 per cent of the world market share. The turnover of the granite industry was about 100 million Euro in 2004. Most of the Finnish granites produced are rapakivi granites and Finland is probably the only country in the world which exports rapakivi granite products. Italy is the largest importer of Finnish granites with c. 100 000 tons in 2004 followed by China with 65 000 tons. Finish rapakivi granite products are known as Baltic Brown (quarried at Ylämaa, Husu and Parkkola), Carmen Red (quarried at Virolahti, Lypsyniemi and Pyterlathi) and Eagle Red (quarried at Pyhtää and Kotka).

Ore deposits related to rapakivi granites Rapakivi granites have traditionally been regarded as barren rocks, meaning they do not contain concentrations of any minerals of economic interest. However, as long ago as 1907 Otto Trüstedt interpreted the skarn-type tin-beryllium-copper-zinc deposits of Pitkäranta to be genetically related to the nearby Salmi rapakivi granite batholith in Russian Karelia. In the 1970s rapakivi granite complexes were recognized as showing traits of the typical tin-bearing granites, such as the Cornish granites, and discoveries of tin mineralisation were made in Fennoscandia, Brazil and Missouri. It is now evident that the Proterozoic rapakivi granites represent a world-wide magmatic association with remarkable ore-generating capacity, with economically viable rapakivi-related tinpolymetallic deposits known from Amazonas and Rondônia, Brazil; Missouri, USA; Karelia, Russia and India. Ilmari Haapala (1995) distinguished two major types of ore deposits occurring within Proterozoic rapakivi granite plutons: (i) greisen-, vein-, and skarn-type tin (-tungsten-beryllium-zinc-copper-lead) deposits associated with specialized late-stage granites; and 2, iron oxide-copper (-uranium-gold-silver) deposits. One of the largest mineral deposits in the World at Olympic Dam in South Australia is an iron-copper-uranium-gold-silver deposit in a hydrothermally mineralized hematite breccia complex situated in a 1590 million year old rapakivi granite pluton.

Weathering of rapakivi granite

All rocks weather with time if they are exposed to the atmosphere but rapakivi granites have a very distinctive weathering behaviour. Despite being generally a very hard and strong rock, a fresh rapakivi exposure can suddenly change to coarsegrained, sharp-edged saprolitic crumble or grus (locally called moro). This disintegration is confined to certain coarse-grained and biotite/amphibole-rich granite varieties and to certain areas. In vertical quarry walls horizontal banks of saprolitic rapakivi can occur under the solid and fresh rock. Remarkably, the saprolite and the adjacent rock show no differences in their macroscopic appearance. In 1930 Pentti Eskola in Finland wrote a classical work ‘On the disintegration of rapakivi’ and concluded that the disposition to crumble is related to the rock texture, especially the smooth boundary surfaces between the coarse mineral grains. Mineral content is of less significance than grain size although a high content of efflorescent micas speeds up the rock destruction. The sharp contact of weathered and fresh rock may be explained in many ways: density of rock joints and microfractures, differences in rock matrix and texture, or differences in the cutting erosional planes.