Introduction
Mud volcanoes, also referred to as sedimentary volcanoes, form in regions where layers of fluidized sediments, such as silt and clay, undergo pressurization due to tectonic activity, such as in convergent plate boundary zones or through accumulation of hydrocarbon gases (Brown, Reference Brown1990; Bonini, Reference Bonini2012). At the surface, mud volcanoes release pressurized gas, of which >90% by volume is methane, with a lesser amount of carbon dioxide and nitrogen (Kopf, Reference Kopf2003), and saline water, occasionally enriched with traces of oil. These geological phenomena offer invaluable insights into the intricate processes governing the formation and migration of oil and gas, reaching depths up to 11 km (Mazzini and Etiope, Reference Mazzini and Etiope2017). The erupted water is rich in organic compounds, as well as in Br, B, Cl, I, and, eventually, Li (Mazzini and Etiope, Reference Mazzini and Etiope2017; Nikitenko and Ershov, Reference Nikitenko and Ershov2021). Volcanic edifices, reaching a height of a few hundred meters, can sprawl over several square kilometers. Notably, Azerbaijan (AZ) and the Caspian Sea have the highest concentration of mud volcanoes worldwide and are home to some of the largest of them (Mazzini and Etiope, Reference Mazzini and Etiope2017; Baloglanov et al., Reference Baloglanov, Abbasov and Akhundov2018).
Boron is generally enriched in clay minerals and saline water associated with sedimentary volcanism (Kopf and Deyhle, Reference Kopf and Deyhle2002). Based on quantitative clay mineralogy and the boron (B) content in mud, studies showed that with the advancing of smectite illitization in the B-rich fluid environment, fluid–mud interaction induces B to enter into the illite structure (Frederickson and Reynolds, Reference Frederickson, Reynolds and Swineford1960; Williams et al., Reference Williams, Hervig, Holloway and Hutcheon2001a).
In their study of mud products from Northern Apennines (NA) mud volcanoes, Battaglia and Pennisi (Reference Battaglia and Pennisi2016), suggested that B-rich water associated with mud volcanism can affect ‘illite crystallinity’, and support that hydro-chemical environment exerts an influence on the intensity of post-sedimentary transformations (Buryakovsky et al., Reference Buryakovsky, Djevanshir and Chilingar1995). The principal clay mineral sorbent is illite; however, other clay minerals such as smectite, chlorite, kaolinite, and mixed-layer illite-smectite can also contain significant amounts of B (Frederickson and Reynolds, Reference Frederickson, Reynolds and Swineford1960; Palmer et al., Reference Palmer, Spivack and Edmond1987; Ishikawa and Nakamura, Reference Ishikawa and Nakamura1993; Williams et al., Reference Williams, Hervig, Holloway and Hutcheon2001a).
Illite formed in a B-rich environment adsorbs B rapidly on its surface and B then diffuses slowly into the tetrahedral part of the structure where it replaces Si and/or Al (Fig. 1; Goldberg and Arrhenius, Reference Goldberg and Arrhenius1958; Couch and Grim, Reference Couch and Grim1968; Stubican and Roy, Reference Stubican and Roy1962; Fleet, Reference Fleet1965; Keren and O’Connor, Reference Keren and O’Connor1982; Williams and Hervig, Reference Williams and Hervig2002; Williams et al., Reference Williams, Hervig, Holloway and Hutcheon2001a; Williams et al., Reference Williams, Hervig and Hutcheon2001b). Direct evidence of the substitution of B into the tetrahedral sheet of the structure was obtained using infrared spectroscopy by Stubican and Roy (Reference Stubican and Roy1962). Molecular modeling of the illite structure using density functional theory revealed that B incorporation is energetically favored in tetrahedral sites, replacing Si atoms, rather than in the interlayer of expandable clays (Williams et al., Reference Williams, Turner and Hervig2007; Clauer et al., Reference Clauer, Williams, Lemarchand, Florian and Honty2018; Martos-Villa et al., Reference Martos-Villa, Mata, Williams, Nieto, Arroyo and Sainz-Díaz2020). Given that the B tetrahedral covalent radius is smaller (B: 88 pm; Al: 126 pm; Si: 117 pm; Shannon, Reference Shannon1976; Bailey, Reference Bailey2006), boron replacement of Si and/or Al in tetrahedral sites has the effect of changing the illite cell dimension, and the degree of ‘illite crystallinity,’ expressed in general as the Kübler index (KI), established as the full width at half-maximum height value (FWHM) of the (001) illite peak (Kübler, Reference Kübler1964; Kübler, Reference Kübler1967; Kübler, Reference Kübler1984).
Schematic representation of the illite structure (modified from Murray, Reference Murray2006).
The understanding gained on Northern Apennine mud volcanoes was expected to influence the KI in the sedimentary sequence at the illite–smectite transition. The conversion of smectite into illite depends on numerous factors such as availability of K+ in the system, reaction time, chemistry of the circulating fluids, and to a large extent on temperature and pressure (Colten-Bradley, Reference Colten-Bradley1987). Moreover, modification in smectite ordering and a progressive reduction in smectite content within mixed-layer illite-smectite (Ilt-Sme) occurring during burial have been utilized as mineral markers to define the thermal history of sedimentary sequences and trace diagenetic and low-grade metamorphic reactions (Velde et al., Reference Velde, Suzuki and Nicot1986; Velde and Vasseur, Reference Velde and Vasseur1992; Pollastro, Reference Pollastro1993; Hillier et al., Reference Hillier, Matyas, Matter and Vasseur1995; Varajao and Meunier, Reference Varajao and Meunier1996; Lanson et al., Reference Lanson, Velde and Meunier1998; Berger et al., Reference Berger, Velde and Aigouy1999). In this setting, Ilt-Sme clay minerals are commonly utilized as a mineral thermal indicator; as physical and chemical conditions change during burial, smectites become unstable and are converted to illite via transitional mixed-layer Ilt-Sme (abbreviations after Warr, Reference Warr2021).
In the present study, previous investigations on the effect of boron on ‘illite crystallinity’ (Battaglia and Pennisi, Reference Battaglia and Pennisi2016) were extended to three mud volcanoes in Eastern Azerbaijan (Perikushkul, Dashgil, and Shikhzahirli), and a comparison between the two case studies is discussed. The main objective is to explore the relationship between structurally fixed boron in illite and the KI, investigated here from a mineralogical and geochemical perspective. New insights are provided into the clay mineral phases, boron (and potassium) contents in the clay fraction (<2 μm), and physical properties, such as sediment grain size, of the AZ mud volcanoes, contributing data that are currently scarce in the literature.
Geological framework
Since the Plio-Pleistocene age, most of the convergence in the Eastern Caucasus region has been accommodated by the Kura Fold-and-Thrust Belt (FTB), located in the southeastern section of the Greater Caucasus. This domain has highly variable characteristics with isolated thrust folds propagating out of sequence in the western part and duplex structures with sequenced thrusts in the eastern part. The structures, which deform the Plio-Quaternary sedimentary deposits of the Kura basin, have a southward trend with shallow detachments (~5 km), located in the clayey formation of Maykop (Oligocene-Miocene) (Allen et al., Reference Allen, Vincent, Alsop, Ismail-zadeh and Flecker2003). Mud volcanoes in Azerbaijan are sited in the FTB area, mainly in the eastern part, and offshore in the South Caspian Basin (SCB). They generally develop on the crests of anticlinal folds that deform the fluvio-deltaic deposits of the Kura basin (up to 6 km thick and aged between the Upper Miocene and Lower Pliocene), and, generally, form edifices that have an elliptical base with a major axis parallel to the direction of maximum local horizontal stress (Bonini, Reference Bonini2012). The Kura is a foreland sedimentary basin of the Great Caucasus, consisting of a very thick synorogenic sedimentary sequence (up to 10 km onshore and 25 km in the SCB) of Paleocene-Quaternary age (Bonini et al., Reference Bonini, Tassi, Feyzullayev, Aliyev, Capecchiacci and Minissale2013). In the Oligocene-Lower Miocene rapid deposition of an anoxic level of fine sediments, rich in organic matter, with a variable thickness between 200 and 1200 m and continuous throughout the region, is recorded in the basin, which is known as the Maykop (Hudson et al., Reference Hudson, Johnson, Efendiyeva, Rowe, Feyzullayev and Aliyev2008). This formation, the top of which is attested at a depth between 4 and 6 km in eastern Azerbaijan and reaches the surface in the Great Caucasus area, represents the detachment level of the FTB and the primary source of hydrocarbons and sediments that generate mud volcanoes (Kopf et al., Reference Kopf, Deyhle, Lavrushin, Polyak, Gieskes, Buachidze, Wallmann and Eisenhauer2003). Above the Maykop Series is placed the formation of the Diatom Suite (Middle-Upper Miocene), which presents shale and marl with interlayers of sandstone and pelite. In the Lower Pliocene-Upper Miocene the sedimentation rate underwent a notable increase with the deposition of the Productive Series, formed by fluvio-deltaic deposits with interlayers of shale. This formation reaches a thickness of 5–7 km in the central part of the SCB and hosts the hydrocarbon and mud reservoirs and it is sealed at the top by the marine shales of the Akchagyl Series (Upper Pliocene). The sequence is completed by Quaternary deposits of the non-marine environment of the Absheron Series and recent sediments (Bonini et al., Reference Bonini, Tassi, Feyzullayev, Aliyev, Capecchiacci and Minissale2013). As a result of the compressive tectonics and the complex system of folds and thrusts, the depth of the Maykop Series (and consequently those above) varies locally within the FTB from 10 to 4 km. Furthermore, although the Maykop Series is inferred as the primary source of mud, clasts belonging to the less recent formations have also been found in the eruptive products, suggesting a mobilization of the material even from deeper reservoirs.
Sampling and experimental methods
Nine mud samples and associated waters extruded from the volcanoes Perikushkul, Dashgil, and Shikhzahirli, located in the Absheron region and from Daghlig–Shirvan in Gobustan (Azerbaijan) (Fig. 2), were collected in July 2017. Five samples were taken from different gryphons at Perikushkul (labeled as P1, P2, P3, P4, P5), where mud was gently pouring out at the time of sampling. Three samples were collected from gryphons at Dashgil (labeled as G1, G2, G3), and one sample was taken from the lake at the top of Shikhzahirli volcano (labeled SH), where intense gas bubbling was observed (Fig. 3). Temperature (°C) and electrical conductivity (EC; mS cm–1) were measured during sampling (Table 1). EC measurements were repeated after one day, once sediment had settled by gravity. Significant separation of the two phases was not observed after 1 day in samples P1, P2, G1, and SH.
Sampled mud volcano sites (in red).
Sampling of muds from active gryphons and mud pools at the Perikushkul and Shikhzahirli mud volcanoes.
Sample sites and field data
EC = electrical conductivity; n.d. = not determined; n.d.* = not determined, still turbid solution after 24 h.
The mud samples were stored in plastic bottles. In the laboratory, water was separated from sediments by decantation, after the suspension had settled. Sediments were subsequently dried at T<60°C to avoid any changes in the structure of the clay minerals (Moore and Reynolds, Reference Moore and Reynolds1997). The dried sludge was gently disaggregated and divided into quarters (Cavalcante and Belviso, Reference Cavalcante and Belviso2005) for particle size and mineralogical and chemical analysis. Separated water was filtered using 0.2 μm Sartorius filters, and analyzed for B using inductively coupled plasma-optical emission spectroscopy (Perkin Elmer Optima 2000DV instrument).
Grain size
The grain size of bulk samples was determined using a diffraction pattern analyzer (Malvern Mastersizer 2000, coupled to a Hydro 2000G wet sample dispersion unit: range of measurements 0.1 μm to 2 mm). Replicate analyses yielded 2SD = ±3% (n=5).
X-ray powder diffraction
The main mineral phases were identified through X-ray diffraction (XRD) analysis of powdered bulk-rock samples. For XRD analyses of oriented slides, the clay fraction (<2 μm grain size) was separated from the aqueous suspensions of whole-rock powders by differential settling, in accordance with Stokes’ law. The aqueous suspension of each fraction was pipetted, saturated with K⁺ and Mg²⁺ (KCl 1 M and MgCl2 0.1 M solutions), allowed to settle on the slides, and then dried at room temperature to produce highly oriented preparations. Care was taken to avoid excessively thin preparations; the amount of clay on each slide was within the range of 3–4 mg cm–² (Lezzerini et al., Reference Lezzerini, Sartori and Tamponi1995). No specific procedure was applied to remove the organic matter or carbonates to avoid deleterious effects on the phyllosilicates. A 1.82% mannitol solution was used to remove adsorbed B (Williams and Hervig, Reference Williams and Hervig2002). The main constituents of mud samples were identified as quartz, feldspars, and calcite, with associated dolomite in some samples (P1, P2, P4, P5, SH) (Fig. 4). Oriented slides indicated the occurrence of smectite, mixed-layer Ilt-Sme, illite, and chlorite (Fig. 5a–c). KI values are reported in Table 2, along with a quantitative analysis of clay-mineral phases in wt.% carried out according to the methodology described by Moore and Reynolds (Reference Moore and Reynolds1997). Instrumental setting and data processing methods were described by Battaglia and Pennisi (Reference Battaglia and Pennisi2016).
XRD spectra of bulk-sample powders. Qtz = quartz; Phyll = phyllosilicates; Kfs = K-feldspar; Pl = plagioclase; Cal = calcite. Wavelength = 0.15418 nm.
(a–c) XRD spectra of oriented slides from the clay fraction. Sme = smectite; Ilt-Sme = mixed-layer illite-smectite; Ilt = illite; Chl = chlorite; K-AD = samples saturated with KCl solution and air dried; Mg-AD = samples saturated with MgCl2 solution and air dried; Mg-EG = samples saturated with MgCl2 solution and ethylene glycol; wavelength = 0.15418 nm.
Mineralogy of clay minerals and associated KI
Phases are reported as wt.% of total <2 μm fraction (Sme = smectite; Ilt-Sme = illite-smectite mixed layers; Ilt = illite; Chl = chlorite; mineral abbreviation nomenclature according to Warr, Reference Warr2021).
The KI measurements were conducted on ethylene-glycol (EG) oriented slides on the 5 Å peak of the Ilt phase (002 basal peak) to avoid interference with the mixed-layer Ilt-Sme (Battaglia et al., Reference Battaglia, Leoni and Sartori2004), and were refined with an accurate fitting, as described by Battaglia and Pennisi (Reference Battaglia and Pennisi2016). To assess the reproducibility and accuracy of the results, multiple slides from a single batch of the sample suspension were measured several times. The relative standard deviation of KI measurements was 3–5% for samples with illite-chlorite clay mineral assemblages and 5–8% for those mixed-layer Ilt-Sme. The KI data were calibrated using the standard rock slab series established by Kübler, adhering to the procedures and calibration equation outlined by Leoni (Reference Leoni2001).
Quantitatively, the primary diagenetic clay reaction in shale is the gradual conversion of smectite into illite through the formation of mixed-layer Ilt-Sme, a process known as smectite illitization. Reynolds and Hower (Reference Reynolds and Hower1970) were the first to provide a detailed account of the relationship between Ilt-Sme and depth, identifying three interstratification forms of Ilt-Sme characterized by a different ‘Reichweite’ (R) order (Reynolds, Reference Reynolds, Brindley and Brown1980): random (R=0), short-range (R=1), and long-range (R=3). The method presented by Moore and Reynolds (Reference Moore and Reynolds1997) has been applied to determine the R ordering of the samples, by employing the d-spacing position of the 002/003 peak (which combines the Ilt 002 and EG smectite 003 reflections).
Thus the weight percentage of Ilt in the mixed-layer Ilt-Sme was evaluated by measuring the angular difference (°Δ2θ) between the 2θ position of peaks 001/002 and 002/003 of the mixed-layer Ilt-Sme and comparing it with tabulated values (Moore and Reynolds, Reference Moore and Reynolds1997) (Table 3).
Reichweite order obtained according to the Moore and Reynolds (Reference Moore and Reynolds1997) calculation method
Structural B content in clayey sediments
Sediments were processed using a 0.1 M MgCl2 solution with a Milli-Q B-free water (Moore and Reynolds, Reference Moore and Reynolds1997), stirred, allowed to settle overnight, and rinsed twice with B-free water to remove the interlayer B hosted in Ilt-Sme and Sme phases. Samples (sediment/water ratio of 1:100) were subsequently washed with a 1.82% mannitol solution (Williams and Hervig, Reference Williams and Hervig2002), ultrasonicated, and stirred gently for 30 min while maintaining the temperature below 40°C. Samples were then rinsed with B-free water. Precautions were taken to prevent any loss of the clay fraction by centrifuging the samples, allowing all suspended material to settle between steps. The entire process was repeated three times using plastic beakers and centrifuge tubes to avoid any B contamination from borosilicate glassware. The absence of clay loss was confirmed by grain-size distribution analyses performed before and after the process. In addition, XRD analyses permitted the exclusion of any clay assemblages in the dark-colored supernatant, hypothesized to be organic matter, which appeared in several samples throughout the washing treatment. After the procedures for removal of interlayer and adsorbed species, the remaining B in the sample was considered to be bound to the crystalline structure (i.e. the B atoms replacing Si and Al in tetrahedral groups). Structure-bound or ‘structural’ B, referred to hereafter as Bfix, was assessed via prompt gamma neutron activation analysis (Actlabs Laboratories, Canada). The K2O content was determined by inductively coupled plasma-optical emission spectroscopy (ICP-OES) of the clay fraction (<2 μm) after the mannitol and Mg2+ treatment.
The availability of K-rich fluids, associated primarily with the dissolution of K-feldspar and detrital mica, has traditionally been considered a key factor influencing the reduction of KI in Ilt during illitization (Offler and Prendergast, Reference Offler and Prendergast1985; Scotchman, Reference Scotchman1987; Kübler, Reference Kübler1990; Dellisanti et al., Reference Dellisanti, Pini, Tateo and Baudin2008), and this is evaluated in the Results section below.
Structural B (Bfix) in Ilt was calculated proportionally using the percentage of Ilt present in the clayey fraction of the samples (Bfix illite = Bfix clay × illite fraction; Table 4) because B is thought to enter the lattice of the Ilt phase preferentially with respect to the other clay phases (Środoń, Reference Środoń2010); however, Köster et al. (Reference Köster, Williams, Kudejova and Gilg2019) reported that B in solution substitutes into the tetrahedral sites of Ilt-Sme by forming new illitic layers, and potentially into pure authigenic smectites; a B range of 110–150 ppm was reported by Ishikawa and Nakamura (Reference Ishikawa and Nakamura1993) for B in marine smectite. Accordingly, and given the lack of target experimental data, the total B content was assumed for the clay-mineral phases from their percentage abundances.
Structural B concentration of total clay and Ilt are reported in mg kg–1; the Ilt fraction and K2O content are reported in wt.%
Results and Discussion
The grain-size distribution showed significant homogeneity (Fig. 6). For all samples the silt fraction prevailed (62–75%), followed by clay (17–24%) and sand (5–17%) (the International Soil Science Society (ISSS) classification was used here). The samples were classified within fields VI and XII according to the Pettijohn classification system (1975).
Pettijohn classification diagram for the various sample sites.
The results indicated that in all analyzed samples the Ilt content was inversely correlated to Ilt-Sme (Fig. 7), as observed for Northern Apennines mud samples (Battaglia and Pennisi, Reference Battaglia and Pennisi2016).
Inverse correlation between mixed-layer Ilt-Sme and illite.
This correlation suggested that for the samples studied, the mud source area had not reached the Ilt degradation zone (~4–6 km depth in the SCB; Buryakovsky et al., Reference Buryakovsky, Djevanshir and Chilingar1995). AZ mud samples showed a correlation between Ilt and mixed-layer Ilt-Sme, which was even greater than that observed in mud samples from NA. This is probably due to the smaller sampling area of the AZ campaign and, therefore, to a greater chemical-physical and mineralogical homogeneity. In AZ samples, the ‘illite crystallinity’ (expressed as KI) resulted in a positive correlation with structurally bound B in Ilt (Fig. 8). This may suggest that the samples belong to a common setting, with relatively small quantities of detrital Ilt compared with authigenic Ilt. The relation Bfix – ‘illite crystallinity’ can be explained by the incorporation of B into the tetrahedral structure of Ilt that can increase crystallochemical homogeneity and reduce the structural defects such as atomic vacancies and lattice distortions (Moore and Reynolds, Reference Moore and Reynolds1997). Despite the different geodynamic settings, KI vs Bfix data from AZ and NA sites aligned with good approximation along the same trend, confirming and implementing the findings from Battaglia and Pennisi (Reference Battaglia and Pennisi2016). A narrower range of structural B values is again observed in the AZ samples with respect to NA samples (53 mg kg–1 < Bfix illite < 182 mg kg–1, and 118 mg kg–1 < Bfix illite < 530 mg kg–1, respectively), which could also be related to the smaller sampling area investigated in AZ. The geochemical features observed could reasonably be related to the fluids associated with the extruded sediments. Relative to the seawater content (B = 5.0 mg L–1; Gonfiantini et al., Reference Gonfiantini, Tonarini, Gröning, Adorni-Braccesi, Al-Ammar, Astner, Bächler, Barnes, Bassett, Cocherie, Deyhle, Dini, Ferrara, Gaillardet, Grimm, Guerrot, Krähenbühl, Layne, Lemarchand, Meixner, Northington, Pennisi, Reitznerová, Rodushkin, Sugiura, Surberg, Tonn, Wiedenbeck, Wunderli, Xiao and Zack2003), waters associated with erupted mud showed an enrichment in B of 3× to 16× in the NA samples and from 15× to 30× in the AZ samples (Table 5). The δ11B signature ranged from +32‰ to +43‰ in NA samples (n. 6 data; M. Pennisi, unpublished data) and from +40‰ to +54‰ in AZ samples (n. 9 data; Salvadori, Reference Salvadori2019) (Table 5). The chemical and isotopic signature of B may reflect different processes of B desorption vs B incorporation in the Ilt structure controlled by mud/water interaction occurring at different stages of diagenesis at the two sites (You et al., Reference You, Spivack, Gieskes, Martin and Davisson1996; Kopf and Deyhle, Reference Kopf and Deyhle2002).
Relationship between KI and structural B content in AZ and NA (NA data from Battaglia and Pennisi, Reference Battaglia and Pennisi2016).
B content and isotope composition of waters from AZ mud volcanoes
The Ilt (wt.%) vs Bfix plot (Fig. 9) exhibits parallel regression lines, with that related to the AZ samples being shifted at higher Ilt values. Consequently, given a certain Bfix concentration, AZ samples were systematically enriched in Ilt compared with NA, as a result.
Positive correlation of Ilt content and Bfix for samples AZ (black circles) and NA (gray squares).
William and Hervig (2002) proposed potassium concentration as a factor controlling KI. However, in the present study, as shown for the NA samples (Battaglia and Pennisi, Reference Battaglia and Pennisi2016), a weak correlation between KI and K2O content in sediments was observed (Fig. 10). This study does not diminish or contradict the relationship between potassium and crystallinity, and no significant differences exist between the current study and previous research regarding the impact of K on ‘illite crystallinity’. Therefore, in both contexts, the variation of KI appeared to be influenced mainly by temperature and Bfix.
Weak correlation between KI and K content.
To find an explanation for this and for the greater slope of the AZ regression in the plot of KI vs Bfix with respect to NA samples, we note that in NA mud volcanoes the temperature range supports the development of cyclic compounds at a depth of 5–6 km, given an average geothermal gradient of 25°C km–1 (Tassi et al., Reference Tassi, Fiebig, Vaselli and Nocentini2012), while in AZ this depth can be extended to 7–8 km, with an average geothermal gradient of 18°C km–1. Furthermore, the gases emitted from AZ mud volcanoes are depleted in C5+ alkanes relative to those from NA mud volcanoes, indicating that the AZ samples have undergone more extensive secondary processes (Bonini et al., Reference Bonini, Tassi, Feyzullayev, Aliyev, Capecchiacci and Minissale2013). This finding agrees with the deeper source estimated for the AZ mud volcanoes, identified in the Maykop formation (Hudson et al., Reference Hudson, Johnson, Efendiyeva, Rowe, Feyzullayev and Aliyev2008), and is also confirmed by the R1 Reichweite order of most samples that reflect a temperature range above 120°C (Merriman and Frey, Reference Merriman, Frey, Frey and Robinson2009).
These results are consistent with previous conclusions about the maturity of hydrocarbon gases from SCB mud volcanoes, based on carbon isotopic composition and catagenetic maturity (Faber, Reference Faber1987; Guliyev et al., Reference Guliyev, Feyzullayev and Huseynov2001; Guliyev et al., Reference Guliyev, Huseynov and Feyzullayev2004). Based on the vitrinite reflectance measurements in the SCB, it is likely that the ethane in the mud volcanoes examined was produced at depths of 7–8 km.
Conclusions
The present study provides significant insights into the impact of B content on the crystal structure of Ilt during the illitization of smectite and Ilt-Sme phases. This process influences the ‘illite crystallinity’ as quantified by Kübler’s index. Observations were made in natural sedimentary basins characterized by the deposition of fine-grained sediments and the presence of B-enriched saline waters associated with mud volcanism, in Azerbaijan. The results agreed with the findings from a previous study on mud volcanoes in Italy, highlighting the correlation between structural B and KI across different geodynamic contexts and depositional environments. The AZ mud samples have shown a clayey silt composition, with the clay fraction ranging between 17 and 24%. The clay fraction consisted of smectite, illite, and mixed-layered minerals, indicating ongoing illitization, with Ilt content in the range of 35.1–60.3 wt.%. Boron-rich waters have been identified in association with mud extrusion. The comparison of the results obtained from the determination of the ‘Reichweite’ ordering parameter showed in AZ samples a greater degree of diagenesis (R1 order) compared with those from the NA (R1 for only 20% of the samples examined).
Physical, chemical, and mineralogical investigation of AZ mud samples allowed the authors to come to the following conclusions.
Investigation of Ilt in the mud volcanoes of Azerbaijan confirmed that the amount of structurally fixed B alters ‘illite crystallinity’ significantly.
Ilt crystallinity, expressed by the Kübler index (KI), and structurally fixed B (Bfix) were correlated by the equation KI = 0.771–0.00164(Bfix illite) + 1.41•10–6(Bfix illite)2 (R 2=0.93) in mud from Azerbaijan (this work) and Northern Apennines of Italy. Note that this equation resulted from independent chemical and mineralogical data on samples from the two natural systems investigated, and not from experimental laboratory tests.
The correlation between Bfix in illite and the KI confirmed the finding that in young sedimentary basins where interaction occurs between clays and high-salinity B-rich waters, the application of the sediment-depth estimation method based on the determination of the KI on Ilt can be misleading. The correlation between KI and B in Ilt is, however, independent of the Ilt content. The present study, based on the study of a natural system in the context of sedimentary volcanism, aimed to address the key question of Bfix–‘illite crystallinity’ index relationship using a mineralogical and geochemical perspective. The present results could lay the groundwork for a more specific, in-depth crystallographic analysis of B incorporation in the Ilt structure.
Data availability statement
All data used for the research described in the article have been shared as Tables within it.
Acknowledgements
Special thanks are extended to agronomist Elena Maserti for introducing M.P. to Azerbaijan; without her, this research would not have been possible. Gratitude is also expressed to Dr Dilzara Aghayeva and Dr Parvin Aghayeva for their friendship and for illuminating the challenges faced by pomegranates and chestnut trees. The authors sincerely acknowledge Manuele Scatena for conducting the granulometric analysis of the samples.
Author contribution
Matteo Salvadori: Conceptualization, Analyses, Data curation, Writing - Original Draft. Stefano Battaglia: Conceptualization, Analyses, Data curation, Writing - Original Draft. Marco Lezzerini: Data evalutation and processing, Writing - Review & Editing. Dadash Huseynov: Field work, Writing - Review & Editing. Maddalena Pennisi: Conceptualization, Field work, Data curation, Writing - Original Draft, Supervision.
Financial support
None.
Competing interests
The authors declare none.
Open access
