Glauconite and celadonite coexist at the nanometre scale in Early Jurassic
submarine volcanic rocks of the Betic Cordillera (southern Spain) as a
result of microbial activity. Samples from the limit between the two micas,
recognizable in scanning electron microscopy, have been extracted using the
focussed ion beam technique and studied by high-resolution analytical
electron microscopy. Both micas are present as randomly oriented
differentiated small crystals in the boundary area. They define clearly
distinct compositional fields with gaps affecting to Fe, Mg and K. At the
lattice scale, celadonite shows a high degree of order, with homogeneous
orientation of the visible lattice parameters being a difference from
glauconite, formed by packets no more than 10-layers thick. Smectite layers
were also detected alongside glauconite packets, in accordance with X-ray
diffractograms which indicate that glauconite is a mica–smectite
interstratification being more than 90 % mica layers. The compositional
gap indicates that celadonite is not the endmember of the glauconitic series
and the two micas represent two different structural tendencies of mica,
with glauconite having more distorted octahedral sheets, indicated by
systematically higher
Glauconites are green micas particularly significant in sedimentary
environments. Their restricted genetic conditions give them the status of
key minerals for palaeoenvironmental interpretations; they usually form in
shallow-marine settings (water depths
The equivalent non-interlayer-deficient mica, having all the crystal-chemical characteristics described in the previous paragraph, would be celadonite, according to the IMA classification. As the chemical composition of a given glauconite is dependent on its maturation degree (Odin and Matter, 1981; Baldermann et al., 2013), approaching that of celadonite with residence time, this latter mineral has been traditionally contemplated as a possible endmember for glauconites (e.g. Parron and Amouric, 1990). However, celadonite typically has a different origin, linked to the ocean floor hydrothermal alteration of intermediate and basic volcanic rocks (Odin et al., 1988).
Another ubiquitous characteristic of glauconites is the presence of
interstratified smectite layers. According to data from Baldermann et al. (2013) and
López-Quirós
et al. (2020), during maturation, the K content of glauconite evolves
with the proportion of mica layers in the mica–smectite mixed layer
following a logarithmic function, being asymptotic for 100 % mica layers
at a value of K
Reolid and Abad (2014) described glauconites and celadonites in close physical association, which were not linked to the usual sedimentary origin of glauconites but related to Early Jurassic submarine volcanism. The described case may be a clue to understanding the genetic and crystal-chemical relationships between glauconite and celadonite as (1) the two minerals coexist in contact, being differentiable at the scanning electron microscopy (SEM) scale, but their compositions are neatly different and separated by a compositional gap; (2) glauconite has a K content equivalent to the maximum one described in the compilation of data by Baldermann et al. (2013) and López-Quirós et al. (2020).
The aim of the work is the characterization of the textural-genetic relationships between glauconite and celadonite from Reolid and Abad (2014) at the nanoscale with the use of focused ion beam (FIB). This technique offers the possibility of extracting in the SEM nanometre-sized lamellas from the areas where the two minerals occur and to study them with high-resolution transmission electron microscopy (TEM) techniques, which allows for the nanoscale characterization of the textural-genetic relationships between the two minerals. As a complementary tool, Raman spectroscopy was also used to try to characterize structural features related to the OH stretching of both micas that could justify the existence of a compositional gap.
The studied green minerals are located in the Subbetic, Betic Cordillera, southern Spain (Fig. 1a). This mountain range is the westernmost European chain related to the Alpine Orogeny and is divided in external and internal zones. The Subbetic corresponds to the external zone of the Betic Cordillera and is constituted mainly of marine sedimentary rocks deposited in the Southern Iberian Palaeomargin from the Triassic to the Miocene. However, from the Early to Late Jurassic the palaeomargin was affected by intense submarine volcanic activity in wide sectors of the pelagic basin. The deposits related to the submarine volcanism are basaltic pillow-lava flows, interbedded with pyroclastic rocks and pelagic sediments. The analysed phyllosilicates of this study are associated with pillow-lava deposits. These volcanic rocks present a WSW–ENE orientation with an almost continuous outcrop around 85 km long and a maximum of 5 km wide. The maximum thickness of these volcanic rocks, described in the literature, is over 300 m (Vera et al., 1997). The Lower Jurassic submarine volcanic rocks of the Median Subbetic are considered as transitional alkaline, originated in an extensional process of crustal thinning (Portugal et al., 1995; Vera et al., 1997). The main minerals in the pillow lava are olivine, Ti-augite, plagioclase (variable from bytownite to oligoclase) and minor biotite, Ti-hornblende, apatite, and Fe–Ti oxides (Puga et al., 1989; Morata et al., 1996).
Location of the outcrop where samples were collected:
These rocks are well exposed close to the A-44 motorway at kilometre 79, between the cities of
Jaén and Granada, in the so-called Campotéjar outcrop
(Granada province, coordinates
The studied samples were retrieved from the mineralizations that infill the
voids among pillow-lava bodies. These voids are generally irregular spaces
with curved surfaces due to the surface of the pillow-lava bodies (Fig. 1c).
The veins with selected mineralizations developed in these spaces are
generally less than 15 cm long. The hand-scale observations were made on
13 polished slabs, and they are detailed in Reolid and Abad (2014). A total of
26 thin sections were used for petrographic observations. The X-ray
diffraction (XRD) data of the
Textural and chemical observations were made on carbon-coated polished thin
sections in a scanning electron microscope (SEM) using back-scattered
electrons (BSE) in atomic number contrast mode. BSE images were acquired at
15 kV with a working distance of 8 mm using an angle-selective backscatter (AsB) detector in a Merlin Carl
Zeiss electron microscope at the Centro de Instrumentación
Científico-Técnica (CICT, Universidad de Jaén). Microprobe
analyses of green micas on the polished thin sections were performed using
wavelength-dispersive spectroscopy (WDX) on a Cameca SX100 at the Centro de
Instrumentación Científica (CIC, Universidad de Granada). The
instrument was set at an accelerating voltage of 15 kV with a beam current
of 15 nA and an electron beam diameter of
For the textural and chemical characterization at the nanometre scale,
selected samples according to XRD analysis and SEM observations were
prepared for transmission electron microscopy (TEM) study. A dual beam
Helios 600 focused ion beam scanning electron microscope (FIB-SEM) was used
(LMA, Universidad de Zaragoza). The objective of using this technique is to
identify the most interesting areas for observation while keeping the
texture of the sample. Four lamellas were extracted from one of the thin
sections previously analysed by SEM. One lamella corresponds to the
glauconitic zone and another one to the celadonitic zone, and two more lamellas
were extracted including the contact between the glauconitic and celadonitic
areas. The selected areas were marked and trenched using a focused beam of
Ga
The TEM data were obtained using the high angle annular dark field (HAADF) FEI Titan G2 microscope, operated at 300 kV and with a point-to-point resolution of 0.08 nm in the TEM mode and 0.2 nm in the scanning transmission electron microscopy (STEM) mode (CIC, Universidad de Granada). Chemical analyses were obtained, using STEM mode, with a SuperX detector. Compositional maps were obtained from the whole interesting area, and the individual spectra of each pixel of homogeneous areas of the maps summed up to produce the average spectrum of all the area. In addition, scan windows including the entire analysed particle were used for the chemical analyses on the basis of high-angle annular dark field (HAADF) images.
Raman spectra were recorded on a Renishaw (inVia Reflex) spectrometer (CICT,
Universidad de Jaén) equipped with a Peltier-cooled (
The X-ray diffraction diagram of the
X-ray diffraction diagrams of the
A detailed textural description of the studied material using SEM can be
found in Reolid and Abad (2014). Here, we have focused on the transition
zone between glauconite and celadonite crystals inside the green laminated
crusts which are in turn composed of filaments. At the micrometre scale,
these are cylindrical filaments constituted of a coat surrounding a hollow.
The wall of these filaments is constituted of two parts: the inner one
without textural traits and an outer part consisting of a ring of
needle-like crystals disposed as a coating around the inner part (Fig. 3).
Actually, the inner part is composed of very small lath-like crystals
(
BSE images of the green micas surrounded by calcite:
The green micas' structural formulae based on microprobe analyses (Table 1)
easily allow the identification of the two dioctahedral micas: glauconite
and celadonite. In general, for these green micas the Si content is always
Structural formulae for green micas normalized to six cations
(IV
However, two different populations are identified as observed in the Fig. 4. The outer lath-like crystals corresponding to the celadonitic zone show
lower Fe contents (
Chemical plots showing the compositional ranges of the two types of green micas.
The characterization of the transition between the glauconitic zone and the celadonitic zone in lamellas extracted with FIB from thin sections in the TEM has allowed us to establish the main features of both minerals at the nanoscale.
In low-magnification images, both green micas show a similar aspect,
organized in clusters of platy crystals randomly oriented (Fig. 5), although
glauconite crystals appear to be thinner and less crystalline. In any case,
these subtle differences at that scale are very evident in the
lattice-fringe images (Figs. 6 and 7). At the lattice scale, celadonite
shows a high degree of order, with homogeneous orientation of the two
visible lattice parameters displayed in Fig. 6b. The 1
TEM low-magnification images showing the general aspect of
glauconitic zone
TEM images corresponding to celadonites:
TEM images corresponding to glauconite:
At the transition zone from glauconite to celadonite, the configuration of particles is chaotic, although the differentiation of the two green micas here was possible thanks to the differences in the chemistry, mainly the K and Fe contents, as shown by energy-dispersive X-ray (EDX) spectra (Fig. 8a). The domain which includes both micas shows clusters of crystals with different orientations and compositions without a clear organization among them (Fig. 8b, c). Nevertheless, the STEM-EDX maps show a net boundary between glauconite and celadonite (Fig. 8d), which confirms a compositional gap between them (Fig. 8a, d).
Figure 9 displays representative spectra corresponding to the 50–1200 and
3500–3700 cm
Representative Raman spectra of the two types of green micas:
glauconite
In consequence, and although spectra from both green micas are very similar, probably the celadonite features described at nanoscale (more crystalline, less defects) indicative of a more ordered structure imply better-defined bands in the Raman spectra.
Reolid and Abad (2014) pointed to the relation between the filamentous morphologies of the green micas and a possible microbial origin. Glauconite and celadonite crystallized forming a coating around organic filaments that have disappeared, leaving the void space within the elongated mineral structures (no more than a few tens of micrometres long) with rounded sections a few micrometres in diameter (Fig. 3). This makes sense with the microbial communities ubiquitously recorded on pillow lavas in recent ocean crust (e.g. Edwards et al., 2005; Santelli et al., 2008; Santelli, 2009). According to Reolid and Abad (2014) the microbes that lived in the spaces between the pillow lavas were extremophile, chemoorganotrophic organisms. In this frame, the chemical reactions related to the basalt alteration are capable of supplying enough energy for the chemoorganotrophic growth of microbes (Bach and Edwards, 2003; Santelli et al., 2008).
Textural observations of the mineral sequence coating the inner voids of the
filaments indicate that glauconite, which is closer to the inner voids of
the filaments, started to crystallize first. However, the transition from
glauconite to celadonite through a domain (4–5
As indicated before, there is a transitional zone where glauconite and celadonite coexist at the nanometre scale (Fig. 8); nevertheless, a continuous solid solution between the two phases does not exist. From a chemical point of view, a compositional gap between the two micas exists mainly in the K, Fe and Mg contents (Fig. 4). Assuming that the microorganisms got energy from redox reactions involving Fe, the metabolic Fe-rich byproducts could have been accumulated in the cellular walls, contributing to the glauconite genesis around them. Probably when the growth of the glauconitic coatings reached a critical thickness, the microorganisms were isolated, contributing to their death as there was no possible exchange with the environment. Then the input of Fe decreased, more Mg and K entered the structure, and, instead of glauconite, celadonite was formed. However, not only the chemical gradient from microbial filament that conditions the elements available in the environment but also the different structural characteristics of both micas suggested by the existence of a compositional gap controlled the crystallization of both phases.
According to Reolid and Abad (2014) the genesis of both micas was related to
a low-temperature hydrothermal activity with an open circulation regime,
oxidizing conditions and a high seawater
At a later stage, the environmental conditions in the spaces among
pillow lavas changed due to the deposition of marine sediments or new
pillow lava. Therefore, the isolation from the seawater favoured the
increase in
The peculiar origin of glauconite described above, which is not the typical one (Odin and Matter, 1981; Odin and Fullagar, 1988; Amorosi, 1995), allowed its coexistence with celadonite. Both micas are usually linked to different genetic environments: semi-confined microenvironments in the sediment pore water vs. ocean floor hydrothermal alteration of volcanic rocks. Therefore, the idea of celadonite as a possible endmember of the glauconitic series has been chiefly based in the chemical tendency of glaucony during its maturation, which roughly points to celadonite. Nevertheless, to our knowledge, this complete evolution has never been described in nature. It is for this that the described example from southern Spain offers a unique opportunity to describe the chemical and structural differences between the two micas at the nanometre scale.
Structural details of micas to show the differences between
celadonite and glauconite:
Normally, the existence of a compositional gap between two mineral phases
having similar structures is due to small structural differences motivated
for the necessity of the structure to accommodate different chemical
populations (Bloss, 1994). This includes the well-known cases of
albite vs. K-feldspars or paragonite vs. muscovite. In the two cases the
significant differences of ionic radii between Na
Relationship between the mean thickness of the octahedral sheet
and
The mica structure is composed of TOT (tetrahedral–octahedral–tetrahedral)
layers linked through electrostatic charge by cations (Fig. 10a). Hence, the
basal spacing (
Nevertheless, all these parameters refer to the
From the data in Fig. 11, glauconites are characterized by more distorted
octahedrons (longer in the
From the first description of celadonite structure (Zvyagin,
1957; Zhukhlistov et al. 1977,
Tsipurskii and Drits, 1986;
Zhukhlistov, 2005;
Dorset, 1992), it is known that the usual
difference in size between trans-octahedral (M1) and cis-octahedral (M2) positions,
generally found in all dioctahedral phyllosilicates, is minimum for
celadonite (Fig. 10c). Also
Brigatti
et al. (2005) described that M1 and M2 tend to become more similar with the
increase in celadonitic content of micas.
Schmidt et al. (2001) attribute the
change in tendency of
The
To sum up, celadonites and glauconites present enough differences in their structural details to justify the existence of a compositional gap. The knowledge of those details would need additional adequate methods of study, which are outside the scope of this paper. Nevertheless, the different atomic substitutions in the structure, which generates a different distribution of charges between the octahedral and tetrahedral positions, are at the origin of such differences. They produce for glauconites a higher size of tetrahedrons and possibly necessary accommodations in the linked octahedrons that, finally, determine an increase in the lateral dimensions of the TOT layer.
Spectroscopic methods and crystal-chemical models have been employed to
characterize the octahedral cationic distribution in celadonite and
glauconite by Drits et al. (1997),
Besson and Drits (1997), and
Dainyak et al. (2013). These studies have
described in glauconite a significant segregation of the octahedral cations,
with Fe
The study of glauconite and celadonite coexisting at the nanometre scale has
allowed us to acquire a deeper insight about their crystal-chemical
relationships. The two micas grew in confined spaces between pillow lavas as
a result of microbial activity. The chemistry of the local microenvironment
was conditioned at micrometre scale by its proximity to the microbial source
of Fe, producing two consecutive areas characterized by each of the micas:
glauconite closer to the Fe source and celadonite further away from it. In
the transitional zone, phyllosilicates with an intermediate composition
between celadonite and glauconite are lacking, and the two micas describe a
compositional gap characterized by net differences in Fe, Mg and K contents.
Therefore, celadonite is not the extreme term of the glauconitic series. The
chemical celadonite compositional field in the studied samples appears
limited by Mg
All relevant data are displayed in the article and additional ones are available upon request from the corresponding author.
Fieldwork was developed by MR and IA. FN, IA, BB and MR together worked on the conceptualization of the study, on the investigation and the interpretation of results, and on manuscript preparation (original draft and review).
The authors declare that they have no conflict of interest.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We acknowledge the use of the LMA-INA (Universidad de Zaragoza) for access to dual beam equipment and expertise, the technical and human support provided by CICT of Universidad de Jaén (UJA, MINECO, Junta de Andalucía, FEDER), and the use of the PANalytical X'Pert Pro diffractometer of the Departamento de Mineralogía y Petrología, Universidad de Granada, Spain. María del Mar Abad is especially recognized for helping with the HRTEM work. The access to the HAADF FEI TITAN G2 microscope and the Philips CM20 (STEM) microscope was facilitated by the Centro de Instrumentación Científica of the Universidad de Granada. Thanks are extended to Javier Cuadros and an anonymous reviewer for their critical reviews and very helpful comments and suggestions.
This research was carried out with the financial support of projects PGC2018-094573-B-I00, PID2019-104624RB-I00 and RTI2018-093419-B-I00 from the MCIU-AEIFEDER and the research groups RNM-325, RNM-200 and RNM-179 of the Junta de Andalucía and E18_17R of the Government of Aragon and European Fund.
This paper was edited by Martine Buatier and reviewed by Javier Cuadros and one anonymous referee.