www.geologicacarpathica.sk
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
, JUNE 2013, 64, 3, 231—236 doi: 10.2478/geoca-2013-0017
Native selenium as a byproduct of microbial oxidation of
distorted pyrite crystals: the first occurrence in the
Carpathians
ELIGIUSZ SZEŁĘG, JANUSZ JANECZEK and PAWEŁ METELSKI
University of Silesia, Faculty of Earth Sciences, Będzińska ul. 60, 41-200 Sosnowiec, Poland;
eligiusz.szeleg@us.edu.pl; janusz.janeczek@us.edu.pl; pawel.metelski@o2.pl
(Manuscript received August 16, 2012; accepted in revised form March 14, 2013)
Abstract: Acicular crystals of native selenium up to 30 µm long occur together with barite on the surface of goethite
partial pseudomorphs after millimeter-sized pseudotetragonal-prismatic pyrite crystals in calcite veins that cross-cut
Senonian sandstones of the Silesian Nappe in the western Polish Outer Carpathians. Native selenium originated from
selenium apparently released during bacteria-induced oxidation of pyrite at neutral or near-neutral pH conditions. Oxi-
dizing bacteria preferentially colonized {100} faces of pyrite relative {111} faces.
Key words: Outer Carpathians, Poland, bacterial oxidation, goethite, pyrite, native selenium.
Introduction
Inspection of calcite veins that cross-cut thick-bedded sand-
stone in the Silesian Beskid Mountains of the western Polish
Outer Carpathians has revealed the abundance of pyrite dis-
playing various crystal habits including distorted (elongated)
pseudo-tetragonal prismatic forms and malformed crystals.
Some of those crystals are partially replaced by goethite with
microcrystals of native selenium and barite adhering to its
surface (Szełęg et al. 2012). To our knowledge this is the
first occurrence of native selenium in the Carpathians. In this
paper we discuss the origin of the native selenium. We also
provide morphological evidence of bacterial involvement in
the oxidation of pyrite.
Geological setting and samples
The native selenium-bearing samples were collected in a
sandstone quarry near the town of Wisła (Vistula) in the
Silesian Beskid Mts (Fig. 1). The sandstone that crops out in
the quarry belongs to the Senonian Lower Godula Beds of the
Silesian Nappe in the western part of the Polish Outer
Carpathians. The sandstone and associated conglomerate rep-
resent siliciclastic turbidites and fluxoturbidites (Cieszkowski
2004). Siliceous Godula sandstones have been classified as
quartz sandstones (quartz arenites) and oligomictic sandstones
(Kamieński et al. 1967). The abundance of glauconite grains
in the sandstone framework is evidence of moderately reduc-
ing conditions during glauconite formation – whereas the oc-
currence of framboidal pyrite and pyrite-encrusted Bryozoa
fossils in the sandstone framework suggest reducing condi-
tions of the sedimentary environment or of the diagenetic en-
vironment. The fine- to medium-grained (average grain size
0.2—0.3 mm) thick-bedded sandstone in the Wisła quarry is
cross-cut by numerous and rather randomly oriented calcite
veins up to 4 cm thick. Vugs and drusy-cavities within calcite
veins are lined by flattened rhombohedral {01
—
2} crystals of
calcite up to 5 mm in size (Fig. 2A). Double-terminated rock
crystals (variety known in the Carpathians as Marmaros dia-
mond) and tabular colourless crystals of barite occur occasion-
ally on calcite.
Numerous morphological varieties of pyrite crystals occur
in calcite veins. Cubo-octahedral and octahedral crystals are
the only isometric habits of pyrite observed in the veins. Pre-
dominant are distorted (axial) habits. The most common of
them are pseudo-tetragonal prismatic crystals with dominant
{100} faces and subordinate {210} faces, terminated by well-
developed octahedral {111} faces (Fig. 2B). Other varieties of
distorted crystals include pseudo-tetragonal prismatic {100}
forms terminated by pseudo-pinacoid {001} and poorly-de-
veloped {
111
} faces, whiskers, chains of autoepitaxially ag-
gregated cubo-octahedral crystals, and cylindrical forms
terminated by {111} faces. Longer axes of distorted crystals
range from a fraction of a millimeter up to a centimeter.
Some of the pyrite crystals are partially replaced by goethite
(Fig. 2A).
Mineral assemblages in the calcite veins represent low
temperature environment as suggested by calcite crystal habit
(Kostov & Kostov 1999) and the occurrence of the Marmaros
diamonds. According to Karwowski & Dorda (1986) the
Marmaros diamonds crystallized at 60—30 °C.
Analytical methods
Samples were examined by optical microscopy, analytical
scanning electron microscopy (ASEM), electron probe mi-
croanalysis (EPMA) and X-ray powder diffraction. Crystal
morphology and elemental composition of minerals were de-
termined using an environmental scanning electron micro-
scope Philips XL30 ESEM/EDAX (Faculty of Earth Sciences,
232
SZEŁEG, JANECZEK and METELSKI
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2013, 64, 3, 231—236
University of Silesia). The chemical composition of the inves-
tigated minerals was determined using a CAMECA SX100
electron microprobe at the Inter-institutional Laboratory of
Microanalysis of Minerals and Synthetic Materials, Univer-
sity of Warsaw. Electron-microprobe analyses of major ele-
ments in pyrite and goethite were performed at 15 kV and
40 nA; while the selenium in native selenium was deter-
mined at 15 kV and 20 nA. X-ray K lines of Fe, Cu, Mn,
Zn, Ni, Co, Si, Ti and S, and L of Se, Te, Ba, As, Ag, Cd,
Fig. 2. A – Elongated pyrite crystal encrusted by goethite in a calcite-lined vug in the Godula sandstone. B – SEM image of prismatic
crystal of pyrite terminated by octahedral faces.
Sb, and Sn were analysed using wavelength dispersive spec-
trometry (WDS). Minerals and synthetic materials (e.g.
Bi
2
Se
3
for Se) were used as standards appropriately selected
for each of the analysed minerals. X-ray powder diffraction
analyses of pyrite and weathered pyrite were performed us-
ing a Philips PW3710 diffractometer at the Faculty of Earth
Sciences, University of Silesia under the following operating
conditions: CoK radiation, acceleration voltage 45 kV, cur-
rent 30 mA, counting time 3 s per step and scan step 0.01° 2 .
Fig. 1. Simplified geological map of the Carpathians. The enlargement shows location of the sampling site in the sandstone quarry (modi-
fied after Ryłko & Paul 1992 and Szopa et al. 2012). 1 – Upper Cretaceous-Paleogene: sandstones and shales (Istebna beds); 2 – Upper
Cretaceous: sandstones and shales, Malinowskie conglomerates (upper Godulskie beds); 3 – Upper Cretaceous: sandstones and shales (midle
Godulskie beds); 4 – Upper Cretaceous: sandstones, conglomerates and shales (lower Godulskie beds); 5 – sandstone quarry – sampling
site; 6 – prominent mountain peaks; 7 – rivers and streams; 8 – state border.
233
NATIVE SELENIUM AS A BYPRODUCT OF MICROBIAL OXIDATION: FIRST OCCURRENCE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2013, 64, 3, 231—236
Fig. 3. A – Aggregates of platy barite crystals on the surface of goethite. Note the granular features on the surface of altered pyrite crystals
and needles of native selenium. B – A group of native selenium crystals. C – Enlargement of spindle-shaped native selenium crystals.
D– EDS spectrum of native selenium.
Results
Spindle-shaped acicular crystals of native selenium up to
30 µm long and 3 µm thick are adhered to the surface of goe-
thite replacing pyrite (Fig. 3). They are often associated with
platy barite in the form of barite “roses” (Fig. 3A) and occur
as single crystals or sub-parallel intergrowths of split crystals
(Fig. 3B—C). Native selenium has not been observed on the sur-
face of non-altered pyrite or anywhere else in the calcite veins.
The EDS spectra of native selenium (Fig. 3D) display
strong peaks of SeL and weak peaks of iron and oxygen,
most probably from the surrounding goethite. Due to the
small width of the crystals it was technically impossible to
obtain good quality electron microprobe data for the native
selenium. The best single-spot analysis gave 80.2 wt. % Se,
0.94 wt. % Fe, 0.46 wt. % Ca, and 0.21 wt. % S. Clearly, py-
rite, goethite and possibly calcite contributed to the analysis.
The low analytical total (81.79 wt. %) most probably resulted
from count loss. Detailed WDS scan revealed the weak peak
of oxygen which can be assigned to either the surrounding
material or surface oxidation of native selenium.
Goethite in the pseudomorphs after pyrite was identified
by X-ray powder diffraction. Its major peaks in the X-ray
pattern at 4.19 A, 2.45 A, 1.722 A, and 1.698 A are distinctly
broader than peaks of pyrite. This observation suggests a low
degree of crystallinity of goethite. The amount of goethite re-
placing pyrite in a few elongated crystals has been estimated
at ca. 31 wt. % based on the Rietveld refinement. Cross-sec-
tions of the affected pyrite crystals show replacement fronts
along crystal boundaries, crystal growth layers and sectors,
and along randomly oriented fractures (Fig. 4). In most cases
the pseudomorphic replacement of pyrite by goethite is lim-
ited to the outer portions of crystals leaving their interior
volumetrically dominated by pyrite.
The surface of altered pyrite is covered by aggregates of
elongated tubular structures (Fig. 5A—C). Individual tubes
are tens of micrometers long and consists of granular
goethite. The shape and morphology of the tubular structures
closely resemble products of bacterial (e.g. Leptothrix sp.)
iron oxidation (see, for instance figure 6 in Banfield &
Zhang 2001) including goethite-encrusted filaments from
oxidized pyrite ore (Hoffman & Farmer 2000). We infer
°
°
°
°
234
SZEŁEG, JANECZEK and METELSKI
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2013, 64, 3, 231—236
Fig. 4. Reflected light photomicrographs of pyrite crystals partially replaced by goethite. A – replacement of pyrite (bright) by goethite
(dark) in prismatic crystals occurring along crystal faces and fractures. Fractures perpendicular to the crystal longer axis are related to
growth layers in pyrite. B – replacement of prismatic pyrite overgrown by cubo-octahedral crystals. Note the complete removal of pyrite
from the octahedral growth sectors; whereas relics of pyrite are abundant in cubic growth sectors.
Fig. 5. SEM images of pyrite crystals covered by the goethite-encrusted filaments coalesced to forming mat-like fabrics. A – double-ter-
minated prismatic crystal of altered pyrite. A mat-like texture is confined to {100} form. B – detail of image (A) showing the difference
between {111} and {100} in the abundance of tubular sheaths of excretions of iron-oxidizing bacteria. There are only a few filaments on the
octahedral faces compared to the densely covered prismatic face. Bright needles are native selenium crystals. C – Detail of goethite-en-
crusted filaments with two spindle-shaped crystals of native selenium. D – Granular aggregates of goethite resembling coccoids on the
surface of altered pyrite. Bright platy crystals are barite.
235
NATIVE SELENIUM AS A BYPRODUCT OF MICROBIAL OXIDATION: FIRST OCCURRENCE
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2013, 64, 3, 231—236
from this striking similarity the bacterial-induced oxidation
of pyrite and its subsequent partial replacement by goethite.
Microbes preferentially inhabited {100} faces of prismatic
pyrite crystals covering them with a dense filamentous fab-
ric, whereas {111} were much less densely populated
(Fig. 5A and B). Granular features seen on the surface of
some other prismatic crystals can also be attributed to the
bacterial oxidation of pyrite because of their coccoid-like
morphology (Fig. 5D).
Discussion
The occurrence of native selenium in calcite veins from the
Godula sandstone is confined only to the surface of goethite
partial pseudomorphs after pyrite. This observation implies
liberation of Se from pyrite during its microbial oxidation and
subsequent precipitation of elemental selenium on the oxi-
dized crystals. Selenium can easily be incorporated into the
pyrite structure substituting sulphur due to similarities of
their ionic radii (RSe
2—
= 1.98 A, RS
2—
= 1.84 A) (Coleman &
Delevaux 1957; Chouinard et al. 2005). As a result, Se-bear-
ing pyrites have been found worldwide with Se content as
high as 6.68 wt. % (Zhu et al. 2004 and references therein).
Average concentration of Se in pyrite is 61 ppm (Paulo &
Strzelska-Smakowska 2003). Unfortunately, we were not able
to determine selenium concentration in an unaltered pyrite
from Wisła. However, Se concentration in that pyrite is cer-
tainly lower than its detection limit of about 200 ppm because
selenium L X-ray line was not observed during electron mi-
croprobe (wavelength dispersion) analysis. Moreover, the
small size of the native selenium crystals (Fig. 3) resulting in
their small mass (on the order of fractions of a nanogram) fur-
ther suggests that the investigated pyrite is not particularly
rich in Se compared to other Se-bearing pyrites. Release of Se
from Se-rich pyrites may lead to crystallization of millimeter-
or even centimeter-sized crystals of native selenium (Zhu et al.
2004 and references therein). The unit cell parameter of the in-
vestigated pyrite (5.4174 A) is also typical of “pure” pyrite.
Oxidation of pyrite in the calcite veins must have occurred
within the pH range constrained by the calcite stability,
namely in neutral or near-neutral pH conditions as suggested
by the lack of dissolution features in calcite. This is further
confirmed by the shape of goethite-encrusted filaments on
the surface of oxidized crystals closely resembling filaments
of neutrophilic Leptothrix sp. (Banfield & Zhu 2001). In a
neutral environment abiotic oxidation of iron is efficient and
rapid at high oxygen partial pressure – whereas neutrophilic
bacteria are capable of oxidizing Fe(II) at low oxygen partial
pressure, namely in microaerobic environments (Gault et al.
2011). That would explain why not all of the oxidized pyrite
crystals are covered by mat-like aggregates of filaments.
Possibly both processes, abiotic oxidation of pyrite and bac-
teria-induced oxidation of Fe(II) were either competing or
complementary. Relatively low oxygen partial pressure may
explain the partiality of pyrite oxidation.
Octahedral forms in prismatic crystals are volumetrically
more profoundly replaced by goethite than prismatic forms
(Fig. 4A). In cubo-octahedral crystals, the octahedral growth
sectors are entirely replaced by goethite; whereas there are
high amounts of pyrite relics in the cubic growth sectors
(Fig. 4B). These observations are in agreement with experi-
mental data that show higher oxidation rate of {111} growth
surfaces relative to {100} growth surfaces (Guevremont et
al. 1998). However, pyrite oxidizing bacteria seem to favour
the opposite trend. The preferential attachment of oxidizing
bacteria to the surface of {100} faces relative to {111} faces
seen in samples from Wisła (Fig. 5A,B) is not accidental.
Studies of surface colonization by pyrite oxidizing bacteria
showed that the orientation of bacterial cells to pyrite cubic
crystals was crystallographically controlled. The attached
cells were preferentially aligned along [100 ] and [110 ] direc-
tions (Edwards et al. 1998). Perhaps the atomic structure of
{100} faces occupied by disulphide molecules is favourable
for oxidizing bacteria.
Pseudomorphic replacement of pyrite by iron hydrooxides
caused release of sulphur and traces of selenium from the py-
rite. Selenium Se
2—
ions were oxidized to elemental selenium,
whereas S
2—
was oxidized to sulphate ions and bound to Ba
2+
to precipitate barite. The source of Ba ions is not obvious,
but the co-occurrence of tabular barite and calcite in the
veins suggests the increased activity of barium in solution.
Oxygen partial pressure during microbial oxidation of py-
rite was obviously not high enough to reach the stability
field of selenates. Elemental selenium is stable in aqueous
solutions over a wide range of pH and Eh values (Coleman
& Delevaux 1957; Howard III 1977; Zhu et al. 2004). The
stability fields of native selenium and FeOOH overlap in
standard conditions in the near neutral pH range.
Acknowledgments: We thank M. Markowicz and M. Dobczyń-
ski for their assistance in sampling, T. Krzykawski for his help
with X-ray powder diffraction, and M. Gardocki for sample
preparation. We are particularly grateful to Igor Broska,
Ruslan I. Kostov and an anonymous reviewer for their valu-
able comments and suggestions that substantially improved
the final version of our paper. This study was financially sup-
ported by the Statutory Fund of the Chair of Geochemistry,
Mineralogy and Petrography, University of Silesia.
References
Banfield J. & Zhu 2001: Nanoparticles in the environment. In: Ban-
field J.F. & Navrotsky A. (Eds.): MSA Reviews in Mineralogy
and Geochemistry 44, Washington D.C.
Chouinard A., Paquette J. & Williams-Jones A.E. 2005: Crystallo-
graphic controls on trace-element incorporation in Auriferous
pyrite from the Pascua epithermal high-sulfidation deposit,
Chile-Argentina. Canad. Mineralogist 43, 3, 951—963.
Cieszkowski M. 2004: The Outer Carpathian – general geology.
Miner. Soc. Pol., Spec. Pap. 24, 415—421.
Coleman R.G. & Delevaux M. 1957: Occurrence of selenium in sul-
fides from some sedimentary rocks of the western United
States. Econ. Geol. 52, 499—527.
Edwards K.J., Schrenk M.O., Hamers R. & Banfield J.F. 1998: Mi-
crobial oxidation of pyrite: Experiments using microorganisms
from anextreme acidic environment. Amer. Mineralogist 83,
1444—1453.
°
°
°
236
SZEŁEG, JANECZEK and METELSKI
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2013, 64, 3, 231—236
Gault A.G., Ibrahim A., Langley S., Renaud R., Takahashi Y.,
Boothman C., Lloyd J.R., Clark I.D., Ferris F.G. & Fortin D.
2011: Microbial and geochemical features suggest iron redox
cycling within bacteriogenic iron oxide-rich sediments. Chem.
Geol. 281, 1—2, 41—51.
Guevremont J.M., Elsetinow A.R., Strongin D.R., Bebie J. &
Schoonen M.A.A. 1998: Structure sensitivity of pyrite oxidation:
Comparison of the (100) and (111) planes. Amer. Mineralogist
83, 1353—1356.
Hofmann B.A. & Farmer J.D. 2000: Filamentous fabrics in low-
temperature mineral assemblages: are they fossil biomarkers?
Implications for the search for a subsurface fossil record on
the early Earth and Mars. Planet. and Space Sci. 48, 11,
1077—1086.
Howard III J.H. 1977: Geochemistry of selenium: formation of fer-
roselite and selenium behavior in the vicinity of oxidizing sul-
fide and uranium deposits. Geochim. Cosmochim. Acta 41, 11,
1665—1678.
Kamieński M., Peszat C. & Rutkowski J. 1967: Petrographic vari-
ability of Carpathian sandstones and the problem of sandstone
classification. Ann. Soc. Géol. Pol. 37, 4, 499—508.
Karwowski Ł. & Dorda J. 1986: The mineral-forming environment
of “Marmaros diamonds”. Mineral. Pol. 17, 3—16.
Kostov I. & Kostov R.I. 1999: Crystal habits of minerals. Bulgarian
Academic Monographs, 1. Prof. Marin Drinov Academic Pub-
lishing House & Pensoft Publishers, Sofia, 1—415.
Ryłko W. & Paul Z. 1992: Geological map of Poland without Qua-
ternary sediments. [Mapa geologiczna Polski bez utworów
czwartorzędowych.] Wydaw. kartograficzne Polskiej Agencji
Ekologicznej S.A.
Paulo A. & Strzelska-Smakowska B. 2003: Selenium at the turn of
XX and XXI Centuries. Przegl. Geol. 51, 6, 459—464 (in Polish).
Szełęg E., Metelski P. & Janeczek J. 2012: The first occurrence of
native selenium in the Carpathians. Acta Mineral. Petrogr.,
Abstract Ser., Szeged 7, 135.
Szopa K., Gawęda A., Müller A. & Sikorska M. 2012: The petrogen-
esis of granitoid rocks unusually rich in apatite in the Western
Tatra Mts. (S-Poland, Western Carpathians). Mineralogy and
Petrology, Doi: 10.1007/s00710-012-0262-2
Zhu J., Zuo W., Liang X., Li S. & Zheng B. 2004: Occurrence of
native selenium in Yutangba and its environmental implica-
tions. App. Geochem. 19, 3, 461—467.