Alkali feldspar is the most effective ice nucleating particle in airborne mineral dust and can initiate heterogeneous cloud ice formation at high temperatures [1]. It may thus influence precipitation formation and the Earth's radiation budget. The particularly high ice nucleation ability of microcline within the group of alkali feldspars was attributed to its complex perthitic microstructure in the form of Na-rich and K-rich exsolution lamellae [2], which naturally result from phase transformations during the cooling process after the magmatic and metamorphic crystallization, and defects like step edges, cracks, pores or cavities [2,3]. The lamellae and surface features are usually aligned with the non-rational Murchison plane with Miller indices between (-601) and (-801), subparallel to the direction where the elastic energy associated with exsolution is minimized [4]. Those features were hypothesized to expose small facets of particularly highly ice-nucleation active, but non-cleavable surfaces with the crystallographic (100) orientation [3]. This would explain the epitaxial relationship between feldspar (100) and the primary prismatic crystal planes of macroscopic ice crystals, observed in microscopic freezing experiments [3,5,6,7], but an understanding of this epitaxial relationship on the molecular level is still missing.We study ice formation from the vapor phase on (001) and (010) cleavage plates of gem quality (featureless reference), gem quality with chemically induced fractures along the Murchison plane, and natural perthitic alkali feldspar under atmospheric pressure using a newly developed in situ X-ray diffraction setup and synchrotron radiation. The high-resolution information allows us to quantify the average ice crystal orientation with respect to the crystallographic domains of feldspar and complement previous electron microscope experiments. For the first time, we confirm the epitaxial relationship between ice and feldspar on defect-rich samples under atmospheric-relevant conditions, as observed in our experiments through a narrow orientation distribution in reciprocal space. The highest fraction of oriented ice crystals is found on natural perthite surfaces of (010) orientation, while ice grows rather randomly on the gem quality reference. In addition, we always detect the XRD-signal of oriented ice well before the XRD-signal of the ice fraction growing with random orientation.[1] Atkinson et al., Nature (2013) 498(7454), 355-358, doi:10.1038/nature12278[2] Whale et al. Phys. Chem. Chem. Phys. (2017) 19, 31186—31193, doi:10.1039/c7cp04898j[3] Kiselev et al., Science (2017) 355, 367-371, doi:10.1126/science.aai8034[4] Petrishcheva et al., Contrib. Mineral. Petrol. (2023) 178, 77, doi:10.1007/s00410-023-02059-z[5] Pach and Verdaguer, J. Phys. Chem. C (2019) 123, 34, 20998–21004, doi:10.1021/acs.jpcc.9b05845[6] Kiselev et al., Atmos. Chem. Phys. (2021) 21, 11801-11814, doi:10.5194/acp-21-11801-2021[7] Keinert et al., Faraday Discussions (2022) 235, 148-161, doi:10.1039/d1fd00115a

Tracer diffusion of Na in natural alkali feldspars including sanidine, adularia and orthoclase with different Na:K ratios is measured using the radiotracer technique and applying the 22Na radioisotope. The tracer diffusion measurements along the crystallographic directions perpendicular to (001) and (010) in alularia feldspar reveled a slight (within a factor of two to three) anisotropy of Na diffusion with the faster diffusion rates along the perpendicular to (001) directions, while this difference was less prominent in orthoclase. The anisotropy of Na diffusion is specifically addressed for orthoclase and the full diffusion tensor is determined. The impact of temperature and composition on diffusion in natural alkali feldspars is discussed with respect to impurities and intrinsic defects. The state of Al/Si ordering is proposed to have a significant influence on the Na diffusion rates in alkali feldspars.

At high temperatures, alkali feldspar consists of a continuous solid-solution between the Na (albite) and K (K-feldspar) end members. Below about 600°C, a miscibility gap opens, the limits of which are determined by the so-called solvus. Alkali feldspar of intermediate composition tends to exsolve when cooled from temperatures of magmatic or metamorphic crystallization, forming an intergrowth of Na- and K-rich lamellae, a microstructure referred to as perthite. Initially, coherency is maintained across the lamellar interfaces and it may be preserved over geological times. Since the lattice parameters of alkali feldspars strongly depend on composition, the exsolution lamellae must be strained to maintain coherency at the interfaces. The elastic energy required to strain the lamellae counteracts exsolution and equilibrium compositions of coexisting coherent exsolution lamellae define a coherent solvus which lies below the solvus for strain-free phase equilibria. Segregation of Na and K and subsequent lamellar coarsening is achieved by thermally activated Na-K interdiffusion. Therefore, compositions and widths of the exsolution lamellae could be used to reconstruct cooling histories. Knowing the shape and position of the coherent solvus is key for the corresponding geo-speedometry applications. To determine the coherent solvus, initially homogeneous disordered, gem-quality alkali feldspar was annealed at temperatures between 440°C and 560°C and at atmospheric pressure, which caused it to exsolve into coherently intergrown 10 to 20 nm wide lamellae, the compositions of which were directly determined with atom probe tomography. The application of different annealing times showed that thermodynamic equilibrium was reached during the experiments. The obtained lamellar composition define points on the coherent solvus, which were for the first time measured directly for alkali feldspars. Additionally, equilibrium Na-K partitioning experiments between NaCl-KCl melt and the same alkali feldspars as used for the exsolution experiments were performed at atmospheric pressure and at temperatures between 800°C and 1000°C to calibrate a thermodynamic mixing model supplemented by a model for the elastic energy required for coherent exsolution. The coherent solvus calculated from the thermodynamic model and the directly measured coherent solvus are in excellent agreement. This indicates that the developed models provide an adequate description of phase equilibria in coherent lamellar intergrowth. The directly measured coherent solvus and the models form a solid base for potential applications of geo-speedometry in coherently exsolved alkali feldspars.

Alkali feldspar undergoes a variety of phase transformations during cooling from magmatic crystallization leading to increasing ordering of Al and Si on the tetrahedrally coordinated lattice sites and to grain-internal microstructures such as twins and exsolution lamellae and associated surface topography. Both, Al-Si ordering as well as the specific surface topography may contribute to the extraordinary ice nucleation activity of alkali feldspar. We studied seven natural alkali feldspars ranging from homogeneous and featureless gem-quality sanidine with disordered Al-Si to hydrothermally altered microcline with ordered Al-Si, several generations of exsolution lamellae and micropore-rich regions associated with domains of hydrothermal albitization. (010) and (001) cleavage plates were produced from each feldspar sample and mounted in a cooling stage. Then an array of 7nl droplets of ultra-pure water was applied and cooled at 2 K/min. Droplet freezing events were recorded with an infrared camera.The highest freezing temperatures are observed on (010) cleavage plates of K-rich (XK=0.94) microcline that exhibits 1-8 µm wide albite exsolution lamellae and 20-100 µm wide microporous regions along cracks related to hydrothermal albitization. In contrast, featureless (001) plates of gem-quality sanidines show freezing at over 10 K lower temperatures. The enhanced ice nucleation activity is tentatively ascribed to Si-Al ordering [1] and to heterogeneous ice nucleation on the surface features related to the grain-internal microstructures [2]. Which one of the two factors is more important is still unresolved. [1] Franceschi G., Conti A., Lezuo L., Abart R., Mittendorfer F., Schmid M., Diebold U. (2023) How Water Binds to Microcline Feldspar (001), J. Phys. Chem. Lett., Vol. 15, 1, 15–22, https://doi.org/10.1021/acs.jpclett.3c03235[2] Kiselev, A., Keinert, A., Gaedeke, T., Leisner, T., Sutter, C., Petrishcheva, E., Abart, R. (2021) Effect of chemically induced fracturing on the ice nucleation activity of alkali feldspar. Atmospheric Chemistry and Physics, 21 (15): 11801-11814, DOI 10.5194/acp-21-11801-202

The garnets in garnet pyroxenites from centimetre- to several-hundred-metre-sized mafic lenses embedded in felsic high-pressure granulites of the Gföhl Unit (Moldanubian Zone, Bohemian Massif) are relics of an early high-pressure–high-temperature metamorphic stage related to Variscan subduction and continental collision. Subsequent isothermal decompression to granulite-facies conditions led to the partial replacement of garnet by plagioclase-bearing assemblages. Associated with the partial replacement, a pronounced secondary compositional zoning developed in the relic garnets, which indicates relatively fast diffusion of Fe and Mg and comparatively slow diffusion of Ca. Based on inverse diffusion modelling, cooling rates in the range of 7– 1501 ^∘ C/Myr were estimated for the garnet pyroxenites, indicating rapid cooling and short-lived granulite-facies overprint after decompression. The petrological evidence is compatible with the extrusion of partially molten, buoyant felsic lithologies, which incorporated slivers of mafic lithologies en route. Through the heat they transported advectively, these lithologies produced perturbations of the thermal structure at mid-crustal levels, the decay times of which varied depending on the volumes of the hot material exhumed in different regions.

The equilibrium partitioning of Na and K between alkali feldspar and NaCl–KCl salt melt was determined at 800 ^∘ C, 850 ^∘ C, 900 ^∘ C, 950 ^∘ C and 1000 ^∘ C and close to ambient pressure. Four different natural gem-quality alkali feldspars with low degree of Al–Si ordering covering the range from orthoclase to high sanidine and with slightly different minor element concentrations were used as starting materials. The partitioning curves obtained for the four feldspars are indistinguishable indicating that Na–K partitioning independent of the differences of Al–Si ordering state and minor element concentrations existing amongst these feldspars. A sub-regular two parameter Margules type solution model was fitted to the partitioning data, and the excess Gibbs energy describing the thermodynamic non-ideality of the alkali feldspar solid-solution and the respective Margules parameters W_gK and W_gNa including their temperature dependence expressed as W_g=W_h-TW_s were determined: W_gK = 19754 ± 3140 J· mol ^-1 - T · 2.33 ± 2.67 J· mol ^-1· K^-1 W_gNa = 14916 ± 4272 J· mol ^-1 - T · 3.55 ± 3.64 J·mol ^-1· K^-1 The corresponding solvus has a critical temperature slightly above 650 ^∘ C and is well comparable with earlier direct experimental determinations of the low-sanidine-albite solvus curve. Comparison of the vibrational excess entropy determined from low-temperature heat capacity measurements with the total excess entropy derived from the temperature dependence of the excess Gibbs energy yields a negative configurational contribution to the excess entropy pointing towards short-range Na–K ordering on the alkali site.

At temperatures above about 600 °C, alkali feldspar forms a continuous solid solution between the Na and K end members. Towards lower temperatures a miscibility gap opens, and alkali feldspar of intermediate composition exsolves, forming an intergrowth of relatively more Na-rich and K-rich lamellae. During exsolution, the crystal structure usually remains coherent across the lamellar interfaces, a feature that may be preserved over geological times. Due to the compositional dependence of the lattice parameters, coherent intergrowth requires that the lamellae are elastically strained. The associated elastic strain energy counteracts exsolution, and the solvus delimiting the misciblity gap for coherent intergrowth lies below the solvus for strain free phase equilibria. To determine the coherent solvus, homogeneous gem quality alkali feldspar of intermediate composition was annealed at conditions falling into the two-phase region of the phase diagram. Thereby a coherent intergrowth of approximately 10–20 nanometers wide lamellae was produced. Lamellar compositions were determined with atom probe tomography defining points on the coherent solvus. In parallel, the coherent solvus was calculated using a thermodynamic mixing model calibrated on the same alkali feldspar as used for the exsolution experiments and accounting for the elastic strain energy associated with coherent lamellar intergrwoth. The experimentally determined and the calculated coherent solvus are in excellent agreement indicating that phase equilibria in coherent lamellar intergrowth of alkali feldspar are adequately described, providing a sound basis for the interpretation of phase relations in coherently exsolved alkali feldspar.

Mafic–ultramafic lenses embedded in felsic granulites of the Gföhl Unit, Moldanubian Zone, are considered to be mantle fragments incorporated into mid-crustal levels of the Variscan orogenic crust. We investigated a several 100 m sized mafic lens mainly formed by garnet pyroxenite. The primary mineral assemblage comprises calcium-rich garnet (XGrs = 0.4), kyanite, and sodium-rich clinopyroxene (XNa_M2 = 0.29) (± quartz), which indicates pressures above 1.8 GPa and temperatures around 1000 °C. Towards the margins of the mafic lens, the garnet pyroxenites were increasingly overprinted at lower pressures leading to the destabilization of kyanite, Na-rich clinopyroxene, and garnet. A first decompression phase is represented by garnet-hosted sapphirine–spinel–plagioclase symplectites supposedly replacing kyanite and clinopyroxene. A second stage is evident from the partial resorption of garnet by plagioclase and clinopyroxene in the form of a peculiar corrosion tubes penetrating the garnet in a worm-like fashion. Finally, the third stage decompression assemblage is represented by plagioclase–orthopyroxene–spinel symplectites partially replacing garnet. In all cases, garnet shows pronounced secondary compositional zoning towards the decompression products. The secondary zoning is qualitatively similar for the sapphirine–spinel–plagioclase symplectites and the plagioclase–clinopyroxene corrosion tubes and is characterized by a strong decrease of the Grs content accompanied by an increase of the Alm and Prp contents towards the decompression products. For the sapphirine–spinel–plagioclase symplectite, the garnet composition changes from Alm14Prp42Grs44 in the pristine garnet to Alm22Prp63Grs15 at the interface to the symplectite. The compositional change towards the corrosion tubes is from Alm19Prp40Grs41 to Alm30Prp54Grs16. The secondary zoning towards the plagioclase–orthopyroxene–spinel symplectites is characterized by an increase of XAlm from 0.19 to 0.27 and a concomitant decrease of XPrp from 0.55 to 0.49 at constant XGrs of 0.25. In all cases, the compositional changes are gradual suggesting diffusion-mediated re-equilibration of the garnet at decreasing pressures. Time scales for the duration of decompression were estimated by fitting a multicomponent diffusion model to the observed compositional patterns. Depending on the choice of the diffusion coefficients, the time scales vary from several hundreds to hundred thousands of years, whereby the earliest decompression features yield time scales that are five times longer than those obtained from the corrosion tubes and about ten times longer than those obtained from the plagioclase–orthopyroxene–spinel symplectites. These timescales reflect the duration from the onset of the different decompression-induced mineral reactions to the time when the rocks cooled below about 700 °C and the composition patterns of the garnet were effectively frozen. The longest timescales obtained from the early decompression reactions are on the order of 100,000 years and the shortest timescales obtained from the late-stage symplectites are on the order of 1,000 years. Considering the regional metamorphic setting of the Moldanubian Zone, such timescales are remarkably short and suggest rapid transport of the mafic–ultramafic lithologies from mantle depths to the mid-crustal level. Concomitant incorporation into a dominantly felsic environment led to immediate cooling.

A thermodynamic analysis of coherent lamellar intergrowth resulting from the exsolution of initially homogeneous alkali feldspar is presented. In contrast to earlier treatments, where the simplifying assumption of zero strain in the lamellar interfaces was used, our treatment is more general. The elastic stresses and strains associated with coherent lamellar intergrowth of Na-rich and K-rich alkali feldspar are calculated by minimising the overall elastic energy of the lamellar microstructure. At given pressure and temperature, the elastic energy depends on the volume proportions of the two lamellar types, and thus on the composition of the homogeneous precursor feldspar. As a consequence, there is no single coherent solvus for alkali feldspar, but coherent solvi are different for different compositions of the homogeneous precursor phase. Experimentally observed lamellar orientations agree with those predicted by minimising the strain energy on a set of all possible lamellar orientations.

Corona microstructures comprised of garnet (grt) and clinopyroxene (cpx) were observed at the contacts between plagioclase (pl) and Fe-rich orthopyroxene (opx) in meta-gabbroic rocks in a several 100 m sized (ultra-)mafic lens embedded in felsic granulite of the Gföhl unit (Moldanubian zone, Lower Austria). The corona microstructures are formed around monomineralic aggregates of opx and they are comprised of two layers, an inner about 100 μm thick layer of polycrystalline cpx and an outer, about 800 μm thick layer of polycrystalline garnet. The corona structures are surrounded by the pl-rich rock matrix. The cpx layer shows a weak but systematic chemical zoning characterized by increasing Mg and decreasing Na and Al contents from the contact with grt towards the contact with opx. The grt layer shows a pronounced and complex chemical zoning. There is a consistent trend of decreasing Mg and increasing Ca contents from the contact with the cpx layer, where the composition is Alm22 Prp67 Grs11 towards the contact with the rock matrix, where we observe Alm25 Prp48 Grs28. This pattern is interpreted as a primary growth zoning. Superimposed on the growth zoning there is a secondary zoning, which is evident from a decrease of the Ca content and a concomitant increase of the Mg content from the interior of the individual grains of the grt polycrystal forming the grt layer towards the grt grain boundaries. The secondary zoning is most pronounced in the outermost portions of the garnet layer, where the primary growth zoning shows the highest Ca and the lowest Mg contents. Locally the garnet grains contain abundant primary melt inclusions. In most segments of the corona, secondary opx and pl form layers along the contact between the primary cpx and grt layer, where the opx partially replaces the cpx layer and the pl partially replaces grt. The secondary opx has higher Mg and lower Na, Al, and Ca contents than the opx in the core of the corona structure. The secondary pl has the same composition as the matrix pl. At its outer edge, the garnet layer is locally replaced by spinel bearing cpx-pl symplectites. The primary compositional zoning of the garnet layer could be reproduced in equilibrium assemblage diagrams (pseudosections). Calculated equilibrium phase relations indicate that the grt-cpx corona formed at the contacts between opx and pl at supersolidus HP − HT conditions of P > 1.8 GPa and T > 900 °C and low H2O content. Growth of coronal grt and cpx requires the diffusive transport of Fe and Mg from the opx to the pl and concomitant transport of Ca and Al in the opposite direction. The secondary zoning of garnet, the back reaction forming secondary opx and pl at the contact between the primary grt and cpx layer and the spinel bearing pl-cpx symplectites locally replacing garnet at the outer edge of the grt layer are related to different decompression stages. Preservation of the secondary garnet zoning indicates relatively rapid cooling during late stages of or immediately after decompression.

Diagenetic carbonates in marine sediments contribute to the global burial of carbonates (Schrag et al., 2013; Sun & Turchyn, 2014). The carbonates often form in zones of enhanced anaerobic microbial activity, where the consumption and release of metabolites leads to supersaturation of the porewater with respect to carbonate minerals. Some diagenetic carbonates occur in zones of methanogenesis, where methane concentrations can be very high and reach gas hydrate stability. So far, it has not been clarified how carbonate formation is induced in methanogenic zones. The production of methane by both fermentation of acetate and reduction of carbonate by H2 is stoichiometrically linked to release of excess CO2 and, therefore, should lower carbonate supersaturation in the porewater. Nevertheless, porewater extracted from drill-cores across methanogenic zones, as at ODP Site 1230 in the Peru-Chile Trench, shows very high total alkalinity of 150 mmol/l, buffering the acidification imposed by the CO2. Based on full-speciation reaction-transport modelling (Meister et al., 2022), it is possible to reproduce alkalinity production as a result of the combined effects of dissimilatory release of ammonia and dissolution/alteration of clay minerals under high pCO2 conditions. Hence, acidification of the fluid is buffered by mineral reactions. In this way, silicate alteration in marine sediments may represent a significant CO2 buffer that contributes to the formation and burial of diagenetic carbonates. Schrag, D.P., Higgins, J.A., Macdonald, F.A., Johnston, D.T. (2013) Authigenic carbonate and the history of the global carbon cycle. Science 339, 540–3. Sun, X., Turchyn A.V. (2014) Significant contribution of authigenic carbonate to marine carbon burial. Nature Geoscience 7, 201. Meister, P., Herda, G., Petrishcheva, E., Gier, S., Dickens, G.R., Bauer, C., Liu, B. (2022) Microbial alkalinity production and silicate alteration in methane charged marine sediments: implications for porewater chemistry and diagenetic carbonate formation. Frontiers in Earth Science 9, 756591, 1-18. https://doi.org/10.3389/feart.2021.756591

A numerical reaction-transport model was developed to simulate the effects of microbial activity and mineral reactions on the composition of porewater in a 230-m-thick Pleistocene interval drilled in the Peru-Chile Trench (Ocean Drilling Program, Site 1230). This site has porewater profiles similar to those along many continental margins, where intense methanogenesis occurs and alkalinity surpasses 100 mmol/L. Simulations show that microbial sulphate reduction, anaerobic oxidation of methane, and ammonium release from organic matter degradation only account for parts of total alkalinity, and excess CO2 produced during methanogenesis leads to acidification of porewater. Additional alkalinity is produced by slow alteration of primary aluminosilicate minerals to kaolinite and SiO2. Overall, alkalinity production in the methanogenic zone is sufficient to prevent dissolution of carbonate minerals; indeed, it contributes to the formation of cemented carbonate layers at a supersaturation front near the sulphate-methane transition zone. Within the methanogenic zone, carbonate formation is largely inhibited by cation diffusion but occurs rapidly if cations are transported into the zone via fluid conduits, such as faults. The simulation presented here provides fundamental insight into the diagenetic effects of the deep biosphere and may also be applicable for the long-term prediction of the stability and safety of deep CO2 storage reservoirs.

Alkali feldspar is one of the most common rock forming minerals in magmatic and metamorphic rocks. It forms a solid-solution between the sodium and potassium end members. At temperatures above about 600°C alkali feldspar shows continuous miscibility. Towards lower temperatures, a miscibility gap exists. When cooled from super-solvus temperatures into the two phase region of the phase diagram, alkali feldspar of intermediate composition exsolves forming coherently intergrown lamellae of Na-rich and K-rich alkali feldspar, a microstructure referred to as perthite. The compositions and the characteristic widths of the exsolution lamellae reflect the cooling history. For a quantitative retrieval of cooling rates the thermodynamics of the solid solution including the effect of coherency strain and Na-K interdiffusion, which determines the coarsening kinetics, must be known. Four alkali feldspars with different degrees of Al-Si ordering were investigated, namely Madagascar Orthoclase, Volkesfeld Sanidine, Zillertal Adular and Zinggenstock Adular. For each feldspar a ther- modynamic mixing model describing the strain free solvus was derived from feldspar-NaCl-KCl salt Na-K partitioning experiments performed at 800°C, 900°C and 1000°C. The models show increasing non-ideality with increasing degree of Al-Si ordering. The corresponding coherent solvi and spinodes were calculated using the strain energy function of Robin (1974). The coarsening kinetics was obtained from exsolution experiments. To this end, each alkali feldspar was shifted to intermediate compositions by exchange with NaCl-KCl melt at 900°C for 35 days and subsequently tempered at 440°C, 480°C, 520°C and 560°C for 4, 8, 16, 32, 64, 128 or 256 days. Analyses of the run products by pXRD revealed splitting of reflections of the lattice planes that are subparallel to the lamellae subparallel to (-801), a feature that is diagnostic for coherent exsolution in feldspar. TEM investigation of foils extracted perpendicular to the crystallographic b-axis revealed fully coherent lamellae and lamellar widths between 8 and 30 nm. Lamellae growth rates were obtained from the time series experiments. For a given annealing time and temperature Madagascar Orthoclase shows relatively sharp and thick lamellae as compared to the other three feldspars. The coherency strain was derived from a comparison of the lattice parameters determined for the Na-rich and the K-rich lamellae by pXRD measurements of the experimental products with those of strain free feldspar as given by Kroll et al. (1986). The strain energy density calculated for the coherent intergrowth is by a factor of two smaller than the one given by Robin (1974). Kroll, H., Schmiemann, I., and Cölln, G. (1986). Feldspar solid solutions. American Mineralogist, 71:1–16. Robin, P.-Y. F. (1974). Stress and strain in cryptoperthite lamellae and coherent solvus of alkali feldspars. Am Mineral, 59:1299–1318.

A system of edge cracks was applied to polished (010) surfaces of K-rich gem-quality alkali feldspar by diffusion-mediated cation exchange between oriented feldspar plates and a Na-rich NaCl–KCl salt melt. The cation exchange produced a Na-rich layer at and beneath the specimen surface, and the associated strongly anisotropic lattice contraction lead to a tensile stress state at the specimen surface, which induced fracturing. Cation exchange along the newly formed crack flanks produced Na-enriched diffusion halos around the cracks, and the associated lattice contraction and tensile stress state caused continuous crack growth. The cracks nucleated with non-uniform spacing on the sample surface and quickly attained nearly uniform spacing below the surface by systematic turning along their early propagation paths. In places, conspicuous wavy cracks oscillating several times before attaining their final position between the neighboring cracks were produced. It is shown that the evolution of irregularly spaced towards regularly spaced cracks including the systematic turning and wavyness along the early propagation paths maximizes the rate of free energy dissipation in every evolutionary stage of the system. Maximization of the dissipation rate is suggested as a criterion for selection of the most probable evolution path for a system undergoing chemically induced diffusion mediated fracturing in an anisotropic homogeneous brittle material.

Feldspar is an important constituent of airborne mineral dust. Some alkali feldspars exhibit particularly high ice nucleation (IN) activity. This has been related to structural similarities of the ice (101‾0) prism planes and the (100) planes of alkali feldspar. Here the effect of generating feldspar surfaces with close to (100) orientation by means of chemically induced fracturing on the IN activity of alkali feldspar was investigated experimentally. To this end, gem-quality K-rich alkali feldspar was shifted towards more Na-rich compositions by cation exchange with an NaCl–KCl salt melt at 850 ∘C. By this procedure, a system of parallel cracks with an orientation close to the (100) plane of the feldspar was induced. Droplet-freezing assay experiments performed on grain mounts of the cation-exchanged alkali feldspars revealed an increase in the overall density of ice-nucleating active site (INAS) density with respect to the untreated feldspar. In addition, annealing at 550 ∘C subsequent to primary cation exchange further enhanced the INAS density and led to IN activity at exceptionally high temperatures. Although very efficient in experiment, fracturing by cation exchange with an alkali halide salt is unlikely to be of relevance in the conditioning of alkali feldspars in nature. However, parting planes with similar orientation as the chemically induced cracks may be generated in lamellar microstructures resulting from the exsolution of initially homogeneous alkali feldspar, a widespread phenomenon in natural alkali feldspar known as perthite formation. Perthitic alkali feldspars indeed show the highest IN activity. We tentatively ascribe this phenomenon to the preferential exposure of feldspar crystal surfaces oriented sub-parallel to (100).

We present a model for multicomponent diffusion in ionic crystals. The model accounts for vacancy-mediated diffusion on a sub-lattice and for diffusion due to binary exchange of different ionic species without involvement of vacancies on the same sub-lattice. The diffusive flux of a specific ionic species depends on the self-diffusion coefficients, on the diffusion coefficients related to the binary exchanges, and on the site fractions of all ionic species. The model delivers explicit expressions for these dependencies, which lead to a set of coupled non-linear diffusion equations. We applied the model to diffusion of $$^{23}$$Na, $$^{39}$$K, and $$^{41}$$K in alkali feldspar. To this end, gem-quality crystals of alkali feldspar were used together with $$^{41}$$K doped KCl salt as diffusion couples, which were annealed at temperatures between 800$$^\circ$$ and 950$$^\circ$$C. Concentration-distance data for $$^{23}$$Na, $$^{39}$$K, and $$^{41}$$K were obtained by Time of Flight Secondary Ion Mass Spectrometry. Over the entire investigated temperature range the Na self-diffusion coefficient is by a factor of $$\ge 500$$ higher than the K self-diffusion coefficient. Diffusion mediated by binary $$^{39}$$K–$$^{41}$$K exchange is required for obtaining satisfactory fits of the model curves to the experimental data, and the respective kinetic coefficient is well constrained.

A numerical reaction transport model was developed to simulate the effects of microbial activity and mineral reactions on the composition of the porewater in a 150-m-thick sedimentary interval drilled in the Peruvian deep-sea trench (Ocean Drilling Program, Site 1230). This site shows a zone of intense methanogenesis below 10 m sediment depth. The simulation shows that microbial activity accounts for most alkalinity production of up to 150 mmol/l, while the excess of CO2 produced during methanogenesis causes a strong acidification of the porewater. Ammonium production from organic matter degradation significantly contributes to alkalinity production, whereby ion exchange was simulated to compensate for hidden ammonium production not otherwise accounted for. Although clay minerals are reacting far too slowly to equilibrate with the porewater over millions of years, additional alkalinity is provided by alteration of chlorite, illite, and feldspar to kaolinite. Overall, alkalinity production in methanogenic zones is sufficient to prevent dissolution of carbonates and to induce carbonate formation either continuously as disseminated cryptic dolomite or episodically as hard lithified beds along a supersaturation front. The simulation presented here provides fundamental insight into the diagenetic effects of the deep biosphere and may also be applicable for the long-term prediction of the stability and safety of deep CO2 storage reservoirs.

We used atom probe tomography to complement electron microscopy for the investigation of spinodal decomposition in alkali feldspar. To this end, gem-quality alkali feldspar of intermediate composition with a mole fraction of $$a_{\text {K}}=0.43$$ of the K end-member was prepared from Madagascar orthoclase by ion-exchange with (NaK)Cl molten salt. During subsequent annealing at $$550\,^\circ \hbox {C}$$ and close to ambient pressure the ion-exchanged orthoclase unmixed producing a coherent lamellar intergrowth of Na-rich and K-rich lamellae. The chemical separation was completed, and equilibrium Na–K partitioning between the different lamellae was attained within four days, which was followed by microstructural coarsening. After annealing for 4 days, the wavelength of the lamellar microstructure was $$\approx 17\,\hbox {nm}$$ and it increased to $$\approx 30\,\hbox {nm}$$ after annealing for 16 days. The observed equilibrium compositions of the Na-rich and K-rich lamellae are in reasonable agreement with an earlier experimental determination of the coherent solvus. The excess energy associated with compositional gradients at the lamellar interfaces was quantified from the initial wavelength of the lamellar microstructure and the lamellar compositions as obtained from atom probe tomography using the Cahn–Hilliard theory. The capability of atom probe tomography to deliver quantitative chemical compositions at nm resolution opens new perspectives for studying the early stages of exsolution. In particular, it helps to shed light on the phase relations in nm scaled coherent intergrowth.

The chemically driven propagation of interacting parallel cracks in monoclinic alkali feldspar was studied experimentally. Single crystals of potassium-rich gem-quality sanidine were shifted towards more sodium-rich compositions by cation exchange with a NaCl–KCl salt melt at a temperature of \(850\,^{\circ }\hbox {C}\) and close to ambient pressure. Initially, a zone with elevated sodium content formed at the crystal surfaces due to the simultaneous in-diffusion of sodium and out-diffusion of potassium, where the rate of cation exchange was controlled by sodium–potassium interdiffusion within the feldspar. A chemical shift of potassium-rich alkali feldspar towards more sodium-rich compositions produces highly anisotropic contraction of the crystal lattice. This induced a tensile stress state in the sodium-rich surface layer of the crystals, which triggered the formation of a system of nearly equi-spaced parallel cracks oriented approximately perpendicular to the direction of maximum shortening. Crack propagation following their nucleation was driven by cation exchange occurring along the crack flanks and was controlled by the intimate coupling of the diffusion-mediated build-up of a tensile stress state around the crack tips and stress release by successive crack propagation. The critical energy release rate of fracturing was determined as 1.8–2.2 \(~ \hbox {J}\,\hbox {m}^{-2}\) from evaluation of the near-tip J-integral. The mechanism of diffusion-controlled crack propagation is discussed in the context of high-temperature feldspar alteration.