New data on the granite pedestal of the monument to Peter the Great “The Bronze Horseman” in Saint Petersburg

Introduction

It is known that the huge boulder Thunder-Stone, used as the material for the monument to Peter the Great (“The Bronze Horseman”), was discovered in Konnaya Lakhta near the Gulf of Finland in 1768. It is described in detail how the boulder was found, dug up, processed, installed on Senate Square and how the equestrian statue of the emperor was prepared and cast from bronze [4, 5]. It is indicated that the huge rock under the horseman is monolithic and made of Finnish granite. But both facts should be checked and clarified in order to expand knowledge about the stone decoration of Saint Petersburg and its preservation [3, 4, 15]. This is especially important in the time of 350th anniversary of Peter the Great.

The authors, with the participation of the staff of the State Museum of Urban Sculpture, had a unique opportunity to work with the pedestal of the monument, examine and photograph its condition, collect and study micro samples of the stone. For the first time, the granite was reliably diagnosed, its minerals were studied, the block structure and turns of the Thunder-Stone's blocks during the pedestal creation were established. It was revealed that the Thunder-Stone's granite clearly differs both from the vyborgites and piterlites, traditionally used in the stone decoration of Saint Petersburg, and from the facing slabs of the pavement around “The Bronze Horseman”.

The structure of the pedestal 

A careful examination of the monument reveals that the “rock” underneath is made up of four blocks of complex configuration [2, 8] (Fig.1). The largest block is marked as N 1, the smallest is N 4. One more block, marked as N 5, is 30 cm long and can be seen at the base of the pedestal at the border of blocks N 1 and N 2. Three types of rocks have been identified in these blocks. Firstly, the pedestal is composed mainly (about 90 %) of granite. Secondly, it contains three inclusions (xenoliths) of another rock – fine-grained granite (named so conventionally, according to the visual assessment). Inclusions are visible at the very top of block N 1, under the horseman's feet. Their shape is angular, visible dimensions are 50×30, 20×10 and 40×20 cm (Fig.2). Thirdly, in granite of the block N 1, one can see fragments of a pegmatite vein, about 50 cm thick. Not all of it is visible, since its main part is cleaved and covered by block N 2 (Fig.3). In addition, the granite contains thin pegmatite veinlets, 3-5 cm thick. They are visible in block N 1 on the left side of the pedestal, and in block N 2 at the front of the pedestal.

Different shades of color of granite blocks N 1-4 suggest that the color of the Thunder-Stone was not uniform. It is evident that, first of all, the most durable material was chosen and color shades were not taken into account. A detailed study of the orientation of granite trachytoid structure, the position of pegmatite veins and veinlets made it possible to determine from which part of the original boulder these blocks were made, how much they were processed and how exactly they were turned to give the pedestal the appearance of a “single rock” according to Falcone's idea [2]. Three-dimensional computer modeling showed that only 1/3 of the original Thunder-Stone was used for the pedestal, the total volume of which is estimated at 675 m³, which corresponds to ca. 1,755 tons of granite [2]. But the authors have not yet been able to identify location of the remaining parts of the original boulder.

Granite

The granite is light pink with different shades of gray. In the block N 1, granite is clearly pink with different color variations over the block volume. In the block N 2 it is pinkish-gray, in the block N 3 – light pink, in the block N 4 – grayish-pink. The main granite minerals are K-Na feldspar, plagioclase and quartz. Mica is presented by biotite and muscovite. According to [2], the composition of granite (in vol.%) is the following: feldspar – 68 %, quartz – 29 %, dark-colored minerals (mica) – 3 %. Among feldspars, the predominant one is K-Na feldspar (approximate ratio of K-Na feldspar to plagioclase is 3:1).

The granite is coarse-grained, with an average grain size of about 1.5-3 cm. Feldspar сrystals are subeuhedral, the average grain size is about 0.5-2 cm, in some places there are K-Na feldspar aggregates, 3-5 cm in size. The size of quartz grains varies from 0.3 to 0.5-1 cm. Mica crystals are euhedral, from 0.5 to 2 cm in size. The texture is trachytoid due to the parallel orientation of flattened feldspar crystals. Depending on the sections, the granite looks different in different blocks of the pedestal – in some cases it is almost massive, sometimes porphyric, sometimes clearly trachytoid (Fig.4). 

Pegmatite

In the front Neva-facing part of the granite block N 1, one can see two fragments of a pegmatite vein, about 50 cm thick. According to the mineral composition (K-Na feldspar, plagioclase, quartz, biotite and muscovite), this pegmatite is a granite one. The pegmatite vein is characterized by clearly expressed contacts with the host granite, graphic and pegmatoid structure, as well as zonal texture [8]. It is composed of a parallel-columnar aggregate of large (up to 15 cm long) feldspar crystals with small ingrowths of quartz, forming a graphic structure. In the axial part of the vein, feldspar crystals do not contain quartz and are euhedral with respect to the quartz core.

Pegmatite veinlets are mainly composed of feldspar (2-4 cm across) and quartz crystal aggregates. The contacts between the veinlets and granite are not sharp. The structure is partly zonal with xenomorphic segregations of quartz nests in the axial parts of veinlet bodies. In block N 1, the veinlets are parallel to each other and stretch along the block. In block N 2, the same veinlets stretch across its elongation.

Materials and methods of laboratory research

25 single-mineral samples, collected from the available micro-scaled shear fractures within the pedestal surface, were examined in this study. Topaz was diagnosed in two samples, collected in 2002 [8]. In 2016, five samples were taken:1) two pieces of pink K-Na feldspar from the granite chip, 2.3×3.9 mm in size; 2)one lamella of mica, 0.5×2 mm in size; 3) another mica fragment, 2×2 mm in size; 4) many small fragments of different minerals, 2×3 mm in size; 5) two grains measuring 1.2×2 mm and 0.02 mm. In 2018, 18 mineral samples were collected (Fig.5).

Samples taken in 2016 and 2018 were examined at the Research Park of St. Petersburg State University [8, 13]. The morphology of grains, their intergrowth and inclusions in them were studied in the Resource Center “Microscopy and Microanalysis” using a microscope microanalyzer Hitachi TM3000 (Japan), equipped with an energy dispersive microanalysis attachment OXFORD (UK), a system with focused electron and ion probes QUANTA 200 3D (FEI, Netherlands) and the analytical complex Pegasus 4000 (EDAX, USA). We obtained SE and BSE images of the studied samples. Electron probe microanalysis was carried out on a Pegasus 4000 energy dispersive diffractometer (SDD ApolloX, resolution at 5.9 KeV Mg Ka – 125.7 eV). The measured samples were rough and polished, covered with carbon in a high vacuum at an accelerating voltage of 15-20 kV. The degree of triclinicity of feldspar was determined by the powder method using an X-ray diffractometer UltimaIV (Rigaku), X-ray tube radiation CuKα₁₊₂; wavelengths l_CuKα1 = 1.54059 Å and l_CuKα2 = 1.54443 Å; tube operating mode 40 kV/30 mA; sample rotation speed 20 rpm; temperature 25 °C, atmosphere – air.

Results obtained

In total, more than 20 mineral phases have been identified, 15 of them have been reliably diagnosed (Table 1).

Table 1

Minerals diagnosed in the rocks of the Bronze Horseman pedestal

Note. In addition to the listed minerals, Fe-, Ti-, Nb-, Ta-bearing minerals, U phosphates, TRR were found in the monument pedestal, but not precisely diagnosed. Inclusions of iron and nickel were not indicated either, as their origin in the rocks of the pedestal has not been identified.

Rock-forming minerals – include feldspars and quartz. K-Na feldspar has a light pink color and an almost stoichiometric chemical composition KAlSi₃O₈ (Table 2). In three analyses, FeO impurity of up to 1-2 % was determined in the mineral. The degree of triclinicity at the two analyzed points was 87.6 % (in granite) and 86.1 % (in a pegmatite veinlet). The mineral is substantially homogeneous and contains rare perthites composed by albite N 5-10. Plagioclase in the granite has a lilac-gray color and is represented by oligoclase N 15-20. Albite was found in pegmatite. Quartz has not been studied in the laboratory. Here are examples of chemical formulas of the indicated minerals:

Sample N 1 – (K_0.84Na_0.08)_0.92(Si_3.00Al_1.02O₈) – microcline.

Sample N 9 – (Na_0.84Ca_0.15K_0.01)_1.00(Si_2.85Al_1.15O₈) – plagioclase N 15.

Table 2

Feldspars (element contents normalized to 100%)

Less abundant minerals. We have found mica (biotite and muscovite), magnetite, topaz, and fluorite. Biotite is represented by annite-siderophyllite with variable values of FeO to MgO ratio (Table 3). Muscovite is significantly enriched in Fe. Li was not detected in any micas. 

Table 3

Micas (element contents normalized to 100 %)

Examples of mineral formulas:

Sample N 16 – K_0.90(Fe²⁺_1.68Al_0.82Mg_0.07 Ti_0.04Fe³⁺_0.02)_2.63(Al_1.09Si_2.91O₁₀)(OH)₂ – biotite.

Sample N 18 – K_0.84(Fe²⁺_1.87Al_0.84Mg_0.07Ti_0.04Fe³⁺_0.02)_2.64(Al_1.10Si_2.90O_10.00)(OH_1,86) – biotite.

Sample N 21 – K_0.85(Al_1.53Fe²⁺_0.37Ti_0.03Mg_0.10)_2.03(Al_0.48Si_3.52O₁₀)(OH)₂ – muscovite.

Sample N 22 – K_0.88(Al_1.77Fe²⁺_0.25Mg_0.04Ti_0.01Fe³⁺_0.01)_2.08(Al_0.84Si_3.16O₁₀)(OH)₂ – muscovite.

Magnetite was found in two forms. It was identified in the form of flattened crystals with cleavage cracks. Based on a single analysis, it contains Cr (0.77 %) and Ti (1.37 %). Magnetite with a special morphology was found in the form of tiny spheres with a diameter of about 0.08 mm, containing impurities of Mg, Al and Si (about 3 % in total). Such magnetite spheres are encountered worldwide, and there are ongoing discussions about their origin.

Topaz was found in pegmatite veinlets in the form of single grains and in a pegmatite vein in the form of clearly visible clusters, where each individual grain varies in size from 0.3 to 2.5 mm, with an average value of about 1.5 mm. Topaz is transparent, colorless, bluish and greenish; it associates with quartz and feldspar (Fig.6). In two topaz samples we detected impurities of FeO (1.47 and 3.32 %). According to the analysis, the composition of topaz corresponds to formula Al_2.01Si_O.99O_4.00(F_1.86OH_0.14).  Fluorite was found in granite and pegmatite veinlets.

Accessory minerals. Seven minerals were identified in microsamples. In addition, 18 more mineral phases were identified but could not be diagnosed. Zircon, monazite, thorite and apatite were found in granite, xenotime – in pegmatite. Their chemical compositions were standard. Rutile (and equally possible anatase) was found in granite and pegmatite veinlets. Some of their grains had an ideal stoichiometric TiO₂ composition, but more often they were enriched with isomorphic impurities. The presence of Nb and Ta in undiagnosed phases is worth noting, which indicates either special varieties of rutile or such minerals as titanosilicates.

Uranium and copper phosphates. Five phosphates were identified, four of them were radioactive. They were found in pegmatite and pegmatite veinlets (Table 4) in the form of accumulations of very small (first microns) lamellar, tabular crystals with square or rounded outlines, as well as powder on the surface of feldspar, quartz and topaz grains. Analysis of samples N 59 and N 60 suggests that these can be torbernite and cheralite, respectively.  Unambiguous identification of the indicated minerals has not been achieved.

Table 4

Phosphates of Th, U and Cu (element contents normalized to 100 %)

Undiagnosed minerals. According to the analytical data, samples N 61, 62, 63 and 64 (Tables 5, 6) correspond to Fe silicates and aluminosilicates, whereas sample N 65 corresponds to a special Ti oxide. The presence of Nb and Ta titanosilicate was detected in samples N 66, 67 and 68.

Table 5

Silicates and aluminosilicates of Fe and Ti (element contents normalized to 100 %) 

Table 6

Titanosilicates of Nb and Ta (element contents normalized to 100 %)

Metals are represented by lamellar inclusions of nickel and iron in microcracks of feldspar crystals. The issue of the origin of such inclusions, whether they are natural or technogenic [13], remains controversial.

Discussion. First, let us pay attention to the ratio of the main rock-forming minerals, as well as structural and textural features of the Thunder-Stone. It is obvious that the volumetric ratios of quartz, feldspars and dark-colored minerals characterize the rock as biotite-muscovite K-feldspar granite. The uniform crystallinity, pronounced trachytoid texture and complete absence of rounded (ovoid) segregations of alkaline feldspar allow to exclude the similarity of this rock to the typical rapakivi granite – vyborgite (Baltic-Brown) and peterlite (Carmen-Red) – in accordance with the contemporary perception of the rocks from the rapakivi granite formation [17]. The idea of the Thunder-Stone as a “Finnish” rapakivi granite, common among historians and art experts, is apparently explained by its external resemblance to rapakivi, widely represented in the architecture of Saint Petersburg in the 18th – early 20th centuries [18].

In terms of structure and texture, the Thunder-Stone is not identical to the Olginsky boulder, located on the shore of the Gulf of Finland in the area of Konnaya Lakhta, where the former was once discovered. The Olginsky boulder is considered to be a part of the Thunder-Stone. However, this is a misperception – the boulder is composed of a completely different rock, ovoidal rapakivi granite. Three boulders on the banks of the Petrovsky Pond can neither be considered fragments of the Thunder-Stone. Observations suggest that this is also ovoidal rapakivi granite.

Secondly, the mineral composition, structure, texture, the presence of fluorite, monazite, apatite and other accessory minerals in the Thunder-Stone, as well as topaz and fluorite in pegmatites, allow us to draw attention to its similarity to coarse-grained trachytoid biotite-muscovite granites of the late magmatism phase, which occurred during the formation of the Vyborg rapakivi granite massif. This refers to biotite-muscovite granites, identified by D.A.Velikoslavinsky (1953) in the study of geological structure of the Vyborg massif and later attributed by A.M.Belyaev [1] to rare-metal biotite-muscovite topaz-bearing granites. The Kymi Massif (southeastern Finland) is given as a regional example of a geological object composed of such rocks. Trachytoid granites of this type, the so-called “even-grained rapakivi granites” of the Lappeenranta region, are described and studied in detail by Finnish researchers [14, 15, 16]. The age of the rocks in this massif is estimated as the Lower Proterozoic (1.5-1.6 Ga).

Genetic similarity of the Thunder-Stone to such rocks is also indicated by the presence of pegmatite veinlets. In terms of mineral composition and structure, they are similar to the so-called stockscheiders – topaz-bearing pegmatites, developed in the marginal zones of topaz-bearing granitoid bodies of the Vyborg massif [1]. Both in stockscheiders and in pegmatites, topaz is represented by intergrowths of tiny crystals.

In particular, the issue of topaz presence in the granite that composes the Thunder-Stone remains open. This mineral is found in biotite-muscovite granites of the Vyborg massif, and it would be important to check the similarity of the Thunder-Stone to such rock in this respect. For this purpose, it is necessary to study the stone pedestal of “The Bronze Horseman” in petrographic thin sections. For obvious reasons, it is not yet possible to carry out such study in practice.

The noted similarity of the rocks does not settle the questions of Thunder-Stone origin, but definitely directs on the way to further search for their solution. Of course, the natural outcrops of rapakivi granite in the Vyborg massif are located relatively close to the place of Thunder-Stone discovery, and the conclusion about their spatial proximity seems to suggest a conclusion about their genetic affinity. Nevertheless, it is wrong to exclude from consideration other rapakivi granite massifs, known in the mainland of Finland [20], on the Aland Islands and in other regions of Northwest Russia and neighboring countries. Moreover, the processes of destruction and movement of crystalline basement rock fragments on the Baltic shield during the last glaciation covered vast territories of Northern Europe [6, 10]. So, the noted mineralogical and petrographic similarity of rocks should be considered only as a precondition for determining the desired direction in further research. It can clearly be stated that in the future there is a need for a detailed comparative analysis between the rocks in the monument stone base, the rocks of the rapakivi granite formation and the granites of other formations in the northwestern region of Russia, Finland and Sweden [7, 9]. Examples of rapakivi granite studies [11, 12] show that for the correlation of this type of rocks it is especially important to take into account individual typomorphic features of accessory minerals.

For obvious reasons, petrochemical comparison of the rocks is impossible. Therefore, further study of “The Bronze Horseman” pedestal should focus on physiographic analysis of the compared rock structure and texture, as well as on electron-probe analysis of chemical (including isotopic) composition and age of accessory minerals, specifically, zircon. The authors recommend taking into account the following mineralogical features of the Thunder-Stone: the K content according to the results of radiometry is 6.4 %*, which in terms of K₂O corresponds to 7.4 %; the K/Na ratio is greater than 1, which is evident from the ratio of the main rock-forming minerals; the content of Fe and Mg in granite is low, which follows from a small amount of dark-colored minerals; the Fe/Mg ratio is high, which is evident from the chemical composition of biotite; the gamma activity of granite varies between 31and 38 mcR/h [8]; the content of radioactive elements is 10^-4 for Ra and 10^-4 for 65 Th. In granites and pegmatites of the Thunder-Stone, biotite is represented by annite-siderophyllite. At the same time, in mica from granite (Table 3) the Fe²⁺/(Fe²⁺ + Mg) ratio is 0.96, the Fe²⁺/(Fe³⁺ + Mg) ratio is 94.5/4.2/1.2, the annite/phlogopite/muscovite ratio is 65.4/2.7/31.9. The authors believe that these indicators are typical for micas from the late formation phases of the Vyborg rapakivi granite massif, the chemical composition of which is well studied [15, 19].

Conclusions

Thus, the famous Thunder-Stone, the four main parts of which make up the pedestal for “The Bronze Horseman”, is a boulder of biotite-muscovite K-feldspar granite. The structure of the granite boulder is complicated by the inclusions of xenoliths of other finer-grained granites. On one edge, it is composed of pegmatite lying in granites in the form of a vein, half a meter in thickness, which is only partially preserved. Numerous pegmatite veinlets are also developed in granites. The structural and textural features of granite, its mineral and chemical composition of the rock-forming and accessory minerals, together with the same characteristics of the pegmatites developed in it, indicate their similarity to trachytoid biotite-muscovite granites and topaz-bearing pegmatites (stockscheiders) of the late formation phase of the Vyborg rapakivi granite massif, hard rock outcrops of which are known in southern Finland. The presumed age of the Thunder-Stone formation of granite and pegmatites, as well as the rocks of the Vyborg massif, is estimated as 1.5-1.6 Ga.

The article explores Thunder-Stone origin, as well as other completely unexplored aspects of its history, in particular, how it was separated from the primary rocks, weathered and moved to the coast of the Gulf of Finland, where it was found in 1768. All these questions are undoubtedly interesting for both professional historians of Saint Petersburg and for a wide range of curious and inquisitive enthusiasts. After all, the Thunder-Stone and “The Bronze Horseman” monument to Peter the Great in Saint Petersburg are a part of the historical and cultural heritage of Russia.

Acknowledgments

Authors are grateful to Yu.B.Marin, E.G.Panova, A.M.Belyaev, G.V.Ivannikov, A.M.Larin, Finnish geologists P.Härmä, O.Selonen for the discussion of geological materials, to the staff of the State Museum of Urban Sculpture (St. Petersburg) – N.N.Efremova, G.V.Rytikova and V.V.Manurtdinova for active assistance in the study of the Thunder-Stone. Part of the study was financially supported by the European Union, Russia and Finland (KS1528).

Dedicated to the memory of the geologist, rapakivi granites researcher Alexei Dmitrievich Shebanov (1968-2017).

Reference

1. Belyaev A.M. Mineralogical and Geochemical Features of Vyborg Massif Granites. Vestnik Leningradskogo universiteta. 1983. Vol. 1. N 6, p. 17-26 (in Russian).
2. Bulakh A.G., Popov G.N., Ivanov M.A. Block Structure of the Granite Pedestal of the Bronze Horseman and Its Model. Muzei pod otkrytym nebom. Strategiya sokhraneniya skulptury v gorodskoi srede. St. Petersburg: Znak, 2018, p. 23-26 (in Russian).
3. State Museum of Urban Sculpture. Yesterday, Today, Tomorrow: Collection of Scientific Articles / Ed by N.N.Efremovoi. Nizhnii Novgorod: Kirillitsa, 2017, p. 116 (in Russian).
4. Ivanov G.I. Thunder-Stone. St. Petersburg: Stroiizdat, 1994, p. 112 (in Russian).
5. Kaganovich A.L. Bronze Horseman. The History of Monument Construction. Leningrad: Iskusstvo, 1982, p. 191 (in Russian).
6. Kuzin I.L. Erratic Boulders of Europe. Izvestiya Russkogo Geograficheskogo obshchestva. 2001. Vol. 133. Iss. 6, p. 45-60 (in Russian).
7. Larin A.M. Rapakivi Granites and Associated Rocks: Monograph. St. Petersburg: Nauka. 2011, p. 402 (in Russian).
8. Bulakh A.G., Popov G.N., Yanson S.Yu. Mineral composition and architectonics of the “Bronze Horseman” pedestal in Saint Petersburg. Proceedings of the Russian Mineralogical Society. 2017. Vol. 146. N 6, p. 111-125 (in Russian).
9. Sviridenko L.P. The Rapakivi Granites of the Fennoscandian Shield (The Case of Karelia). Transactions of the Karelian Research Centre of the Russian Academy of Sciences. 2014. N 1, p.17-21 (in Russian).
10. Chuvardinskii V.G. Quaternary Period. New Geological Concept. Apatity: Izd-vo Kolskogo nauchnogo tsentra RAN, 2012, p. 179 (in Russian).
11. Shebanov A.D. Typomorphic Features of Zircon from Trachytoid Rapakivi Granite (the Vyborg Massif). Zapiski Vserossiiskogo mineralogicheskogo obshchestva. 1992. N 4, p. 83-88 (in Russian).
12. Shebanov A.D., Poritskii M.S. Quartz Morphology, Properties and Generations in Rapakivi Granites. Zapiski Vserossiiskogo mineralogicheskogo obshchestva. 1993. N 4, p. 77-90 (in Russian).
13. Yanson S.Yu., Bulakh A.G., Popov G.N. Features of Thunder-Stone Granite Mineral Composition, Revisited. Muzei pod otkrytym nebom. Strategiya sokhraneniya skulptury v gorodskoi srede. St. Petersburg: Znak, 2018, p. 17-19 (in Russian).
14. Frank-Kamenetskaya O.V., Vlasov D.Yu., Rytikova V.V. The Effect of the Environment on Saint Petersburg’s Cultural Heritage. Springer, 2018, p.188. DOI: 10.1007/978-3-319-79072-5
15. Haapala I., Rämö O.T. Petrogenesis of the Proterozoic rapakivi granites of Finland. Geological Society of America Special Paper. 1990. Vol. 246, p. 279-286.
16. Härmä P., Selonen O. Natural stone production in the Wiborg rapakivi granite batholith in southeast. Geotechnical report 10/2018. Helsinki: The Finnish Natural Stone Association, 2018, p. 34.
17. Müller A. Rapakivi granites. Geology Today. 2007. Vol. 23. Iss. 3, p. 114-120. DOI: 10.1111/j.1365-2451.2007.00616.x
18. Bulakh A., Härmä P., Panova E., Selonen O. Rapakivi granite in architecture of St Petersburg as a potential global heritage stone from Finland and Russia. Geological Society, London, Special Publications. 2020. Vol. 486, p. 67-76.
    DOI: 10.1144/SP486-2018-5
19. Rieder M., Haapala I., Povondora P. Mineralogy of dark mica from Wiborg rapakivi batholite, southern Finland. European Journal of Mineralogy. 1996. Vol. 8, p. 597-605. DOI: 10.1127/ejm/8/3/0593
20. 20. Selonen O., Ehlers C., Luodes H. et al. The Vehmaa rapakivi granite batholith – production area for Balmoral Red granites in southwestern Finland. Geotechnical report 1/2016. Helsinki: Finnish Natural Stone Association, 2016, p. 47.