Octahedral Vacancies Research Articles

Influence of Interface Structure on Magnetic Proximity Effect in Pt/Y3Fe5O12 Heterostructures

The influence of interface structure on the magnetic proximity effect (MPE) in Pt/Y3Fe5O12 (YIG) bilayered heterostructures is studied by first-principles calculations based on the density functional theory (DFT). When Pt atoms are in close proximity with Y or Fe ions at the interface, Pt-Y and Pt-Fe bonds are observed. The crystalline orientations and interface termination layers of the YIG strongly modify both the strength and the length of the Pt-Fe bonding and thereby influence the magnetic properties of the Pt. Point defects including tetrahedral Fe, octahedral Fe, and Y vacancies are introduced at the Pt/YIG interface to quantitatively evaluate the influence on the MPE from individual atoms. For the Pt(100)/YIG(100) structure, the interface tetrahedral Fe vacancies can significantly reduce or even completely diminish the magnetic moments in the Pt. In a stark contrast, the octahedral Fe vacancies slightly enhance the Pt magnetism, and the nonmagnetic Y vacancies cause little influences to the Pt magnetism. These results indicate that the strength of the MPE at the Pt/YIG interface strongly depends on the interface structure. This dependence originated from the direct exchange interaction between the Fe 3d and Pt 5d electrons via electronic state hybridization as well as the electron exchange coupling between the Pt atoms.

First-principles energetics of rare gases incorporation into uranium dioxide

First-principles density functional theory–generalized gradient approximation methods have been used to calculate the energetics (incorporation energy, formation energy and binding energy) of rare gases (He, Ne, Ar, Kr and Xe) at the three incorporation sites (octahedral interstitial, uranium and oxygen vacancies) of uranium dioxide. The Hubbard parameter U and van der Waals corrections have been used to describe the strongly correlated electronic behavior of uranium 5f electrons and the weak interactions of rare gases, respectively. The results indicate that the energetics of rare gases depend significantly on the incorporation sites and on the atomic properties such as atomic radius. All rare gases considered here are energetically unfavorable at the three incorporation sites. However, rare gases exhibit significant binding ability to both U and O vacancies. The main trends of relative stability of rare gases generally reflect a size effect: the rare gases become more unstable with increasing atomic number. Electronic structures of these systems containing rare gases also exhibit general trends in their relative stability and charge-transfer character.

Co on Fe3O4(001): Towards precise control of surface properties.

A novel approach to incorporate cobalt atoms into a magnetite single crystal is demonstrated by a combination of x-ray spectro-microscopy, low-energy electron diffraction, and density-functional theory calculations. Co is deposited at room temperature on the reconstructed magnetite (001) surface filling first the subsurface octahedral vacancies and then occupying adatom sites on the surface. Progressive annealing treatments at temperatures up to 733 K diffuse the Co atoms into deeper crystal positions, mainly into octahedral ones with a marked inversion level. The oxidation state, coordination, and magnetic moments of the cobalt atoms are followed from their adsorption to their final incorporation into the bulk, mostly as octahedral Co(2+). This precise control of the near-surface Co atoms location opens up the way to accurately tune the surface physical and magnetic properties of mixed spinel oxides.

First-Principles Analysis of Cation Diffusion in Mixed Metal Ferrite Spinels

Ferrite spinels are metal oxides used in a wide variety of applications, many of which are controlled by the diffusion of metal cations through the metal oxide lattice. In this work, we used density functional theory (DFT) to examine the diffusion of Fe, Co, and Ni cations through the Fe3O4, CoFe2O4, and NiFe2O4 ferrite spinels. We apply DFT and crystal field theory to uncover the principles that govern cation diffusion in ferrite spinels. We found that a migrating cation hops from its initial octahedral site to a neighboring octahedral vacancy via a tetrahedral metastable intermediate separated from octahedral sites by a trigonal planar transition state (TS). The cations hop with relative activation energies of Co ≈< Fe < Ni; the ordering of the diffusion barriers is controlled by the crystal field splitting of the diffusing cation. Specifically, the barriers depend on the orbital splitting and number of electrons which must be promoted into the higher energy t2g orbitals of the tetrahedral metastable in...

Understanding Hematite Doping with Group IV Elements: A DFT+U Study

Si, Ge, or Sn doped hematite (α-Fe2O3) photoanodes show significantly enhanced efficiency for photo-oxidization of water. We employed DFT+U to study the doping of α-Fe2O3 with group IV elements, i.e., Si, Ge, and Sn. From the calculated formation energies and chemical potentials, three key points are concluded. (1) Low oxygen pressure is favored for doping both substitutional and interstitial dopants. (2) Substitutional doping of the Fe atom at the lattice site is more stable than interstitial doping in the octahedral vacancies. (3) Most interestingly, Ge doping is found to be easiest among the three dopants. This result contradicts intuition based on atomic size and indicates that Ge doping should be more efficient than Si and Sn doping in increasing the charge carrier concentration. Incorporation of the dopants at the Fe site generates an electron polaron and the dopant with the +4 valence state by spontaneous transfer of one electron from the dopant atom to a surrounding Fe atom, according to the analy...

Phosphoran olivine overgrowths: Implications for multiple impacts to the Main Group pallasite parent body

Phosphoran olivine (1–7 wt% P2O5) is a metastable phase known from fewer than a dozen meteoritic or terrestrial occurrences. We have thoroughly examined phosphoran olivine in the Springwater pallasite to characterize its distribution, textural relationships, and geochemistry. Phosphoran olivine is abundant in Springwater as randomly distributed millimeter-scale partial overgrowths on the P-free olivine crystals. Geochemical analyses support the substitution mechanism of P into the tetrahedral Si site with octahedral site vacancies for charge balance; observed trace element variations, on the other hand, are not related to P substitution. Element mapping reveals fine-scale oscillatory P zoning in unusual serrate patterns, indicating rapid crystal nucleation from a melt as proposed by Boesenberg and Hewins (2010) and a subsequently variable rate of crystallization. The timing of phosphoran olivine formation in Springwater is constrained to after the period of macroscopic olivine rounding but before the cooling of the metal matrix; because the phosphoran overgrowths overprint specific host grain boundary modifications, we suggest that the episode of extremely rapid cooling necessary to crystallize and preserve this rare phase may have been triggered by an additional impact to the parent body.

Cobalt Spinel at Various Redox Conditions: DFT+U Investigations into the Structure and Surface Thermodynamics of the (100) Facet

Periodic spin unrestricted DFT-PW91+U calculations together with ab initio thermodynamic modeling were used to study the structure, defects, and stability of different terminations of the (100) surface of cobalt spinel under various redox conditions imposed by different oxygen partial pressure and temperature. Three terminations containing under-stoichiometric (100)-O, stoichiometric (100)-S, and overstoichiometric (100)-R amount of cobalt ions were analyzed, and their atomic and defect structure, reconstruction, and stability were elucidated. For the most stable (100)-S and (100)-O facets, formation of cationic and anionic vacancies was examined, and a surface redox state diagram of possible spinel (100) terminations in the stoichiometry range from Co2.75O4 to Co3O3.75 was constructed and discussed in detail. The results revealed that the bare (100)-S surface is the most stable at temperatures and pressures of typical catalytic processes (T ∼ 200 °C to ∼500 °C, pO2/p° ∼ 0.001 to ∼1). In more reducing conditions (T > 600 °C and pO2/p° < 0.0001), the (100)-S facet is readily reduced by formation of oxygen vacancies, whereas in the oxidizing conditions (T < 200 °C and pO2/p° > 10), coexistence of (100)-S and (100)-O terminations was revealed. Formation of the oxygen vacancies involves reduction of the octahedral trivalent cobalt and is accompanied by migration of the divalent tetrahedral cobalt into empty, interstitial octahedral positions. It was also found that the constituent octahedral Co cation proximal to the interstitial cobalt adopts a low spin configuration in contrast to the distal one that preserves its surface high spin state. In the case of the Co depleted surfaces, the octahedral vacancies are thermodynamically disfavored with respect to the tetrahedral ones in the whole range of the examined T and pO2 values. The obtained theoretical results, supported by TPD-O2 and TG experiments, show that the octahedral cobalt ions are directly involved in the redox processes of Co3O4.

Lithium isotope fractionation during uptake by gibbsite

The intercalation of lithium from solution into the six-membered μ2-oxo rings on the basal planes of gibbsite is well-constrained chemically. The product is a lithiated layered-double hydroxide solid that forms via in situ phase change. The reaction has well established kinetics and is associated with a distinct swelling of the gibbsite as counter ions enter the interlayer to balance the charge of lithiation. Lithium reacts to fill a fixed and well identifiable crystallographic site and has no solvation waters. Our lithium-isotope data shows that 6Li is favored during this intercalation and that the solid-solution fractionation depends on temperature, electrolyte concentration and counter ion identity (whether Cl−, NO3− or ClO4−). We find that the amount of isotopic fractionation between solid and solution (ΔLisolid-solution) varies with the amount of lithium taken up into the gibbsite structure, which itself depends upon the extent of conversion and also varies with electrolyte concentration and in the counter ion in the order: ClO4−<NO3−<Cl−. Higher electrolyte concentrations cause more rapid expansion of the gibbsite interlayer and some counter ions, such as Cl−, are more easily taken up than others, probably because they ease diffusion. The relationship between lithium loading and ΔLisolid-solution indicates two stages: (1) uptake into the crystallographic sites that favors light lithium, in parallel with adsorption of solvated cations, and (2) continued uptake of solvated cations after all available octahedral vacancies are filled; this second stage has no isotopic preference. The two-step reaction progress is supported by solid-state NMR spectra that clearly resolve a second reservoir of lithium in addition to the expected layered double-hydroxide phase.

Nature of the active sites of ferrospheres in the oxidative condensation of methane

The catalytic properties of ferrospheres containing 76–97 wt % Fe2O3 in the oxidative condensation of methane were compared with their phase composition and the distribution of iron cations over the crystallographic positions of iron-containing phases in a steady state. It was established that the reaction route of methane oxidation changed at a Fe2O3 content of 89 wt %. Deep oxidation was the main reaction route on ferrospheres with a Fe2O3 content of <88.8 wt %. At a Fe2O3 content of ≥89 wt %, the yield of C2 hydrocarbons sharply increased and the contribution of deep oxidation decreased. The yield of C2 hydrocarbons correlated with the amount of defects in the structure of iron spinel, which are iron ions with the tetrahedral cation of Ca2+ and the octahedral cation vacancy among the nearest neighbors.

Influence of Co doping on crystal structure and electrochemical performances of LiNi0.5Mn1.5O4

To investigate the influence of Co doping on spinel LiNi0.5Mn1.5O4 cathode, LiNi0.4Co0.1Mn1.5O4 and LiNi0.35Co0.1Mn1.5O4 were designed and successfully synthesised through a polymer assisted method. Inductively coupled plasma, X-ray diffraction, Fourier transform infrared spectra and X-ray photoelectron spectroscopy tests have been carried out. Results show that the compositions of the synthesised samples are consistent with the designed compositions, and Co doping could decrease the degree of cation ordering; Co doping together with octahedral vacancies could induce higher concentration of Mn3+ in the spinel structure. The result of cyclic voltammetry test suggests LiNi0.35Co0.1Mn1.5O4 has the best conductivity, and LiNi0.35Co0.1Mn1.5O4 exhibited the best rate capability among the three spinel cathodes.

Transmission electron microscopy and ab initio calculations to relate interfacial intermixing and the magnetism of core/shell nanoparticles

Significant efforts towards understanding bi-magnetic core-shell nanoparticles are underway currently as they provide a pathway towards properties unavailable with single-phased systems. Recently, we have demonstrated that the magnetism of γ-Fe2O3/CoO core-shell nanoparticles, in particular, at high temperatures, originates essentially from an interfacial doped iron-oxide layer that is formed by the migration of Co2+ from the CoO shell into the surface layers of the γ-Fe2O3 core [Skoropata et al., Phys. Rev. B 89, 024410 (2014)]. To examine directly the nature of the intermixed layer, we have used high-resolution transmission electron microscopy (HRTEM) and first-principles calculations to examine the impact of the core-shell intermixing at the atomic level. By analyzing the HRTEM images and energy dispersive spectra, the level and nature of intermixing was confirmed, mainly as doping of Co into the octahedral site vacancies of γ-Fe2O3. The average Co doping depths for different processing temperatures (150 °C and 235 °C) were 0.56 nm and 0.78 nm (determined to within 5% through simulation), respectively, establishing that the amount of core-shell intermixing can be altered purposefully with an appropriate change in synthesis conditions. Through first-principles calculations, we find that the intermixing phase of γ-Fe2O3 with Co doping is ferromagnetic, with even higher magnetization as compared to that of pure γ-Fe2O3. In addition, we show that Co doping into different octahedral sites can cause different magnetizations. This was reflected in a change in overall nanoparticle magnetization, where we observed a 25% reduction in magnetization for the 235 °C versus the 150 °C sample, despite a thicker intermixed layer.

Cation distribution in Zn–Cr spinel structure and its effects on synthesis of isobutanol from syngas: Structure–activity relationship

A series of Zn–Cr oxides nanoparticles with different degrees of cation disorder distribution were obtained by a coprecipitation and postcalcined method. Some qualitative analysis, such as Transmission Electron Microscope (TEM), Powder X-ray Diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and Raman spectroscopy were used to confirm the structure of Zn–Cr oxides and the existence of cation disorder distribution. Meanwhile, the Rietveld method was used to quantitatively investigate the changes of cation distribution between tetrahedral and octahedral vacancies of different catalysts. The results demonstrated that the degree of cation disorder distribution became higher with decreasing annealing temperature and increasing the molar ratio of Zn–Cr. Furthermore, the level of cation disorder distribution in Zn–Cr spinel structure affects tremendously both texture parameters and acid–base properties of catalysts which make isobutanol generation more effectively. Most importantly, we first discover that the isobutanol productivity presents approximately linear relationship to the level of cation disorder distribution of Zn–Cr spinel in quantitatively, unambiguously revealing the structure–activity relationship between the activity of isobutanol and the structure of Zn–Cr spinel.

Defect Formation Energy in Spinel LiNi0.5Mn1.5O4-δ Using Ab Initio DFT Calculations

Defect formation energies based on an oxygen vacancy model and a metal-excess model in Ni/Mn ordered P4332 and disordered Fd3m LiNi0.5Mn1.5O4 (LNMO) were evaluated by using ab initio density functional theory (DFT) calculations. The defect formation energy for the metal excess model was lower than that for the oxygen vacancy model in both P4332 and Fd3m. This indicates that oxygen vacancy formation reactions are unlikely, although interstitial cation occupation at the octahedral vacancies occurred in both P4332 and Fd3m LNMO spinel compounds. In addition, the corresponding defect formation energy in Fd3m was lower than that in P4332, indicating that the amount of defects is sensitive to the cation ordering/disordering in the spinel framework. This agrees with the experimental results that show that only Fd3m tends to possess oxygen defects.

Structure and properties of vanadium(V)-doped hexagonal turbostratic birnessite and its enhanced scavenging of Pb2+ from solutions

Vanadium(V)-doped hexagonal turbostratic birnessites were synthesized and characterized by multiple techniques and were used to remove Pb2+ from aqueous solutions. With increasing V content, the V(V)-doped birnessites have significantly decreased crystallinity, i.e., the thickness of crystals in the c axis decreases from 9.8nm to ∼0.7nm, and the amount of vacancies slightly increases from 0.063 to 0.089. The specific surface areas of these samples increase after doping while the Mn average oxidation sates are almost constant. V has a valence of +5 and tetrahedral symmetry, and exists as oxyanions, including V6O162−, and VO43− on birnessite edge sites by forming monodentate corning-sharing complexes. Pb LIII-edge extended X-ray absorption fine structure (EXAFS) spectra analysis shows that, at low V contents (V/Mn≤0.07) Pb2+ mainly binds with birnessite on octahedral vacancy and especially edge sites whereas at higher V contents (V/Mn>0.07) more Pb2+ associates with V oxyanions and form vanadinite [Pb5(VO4)3Cl]-like precipitates. With increasing V(V) content, the Pb2+ binding affinity on the V-doped birnessites significantly increases, ascribing to both the formation of the vanadinite precipitates and decreased particle sizes of birnessite. These results are useful to design environmentally benign materials for treatment of metal-polluted water.

Chemical controls on incipient Mg-silicate crystallization at 25°C: Implications for early and late diagenesis

AbstractMg-silicate minerals (e.g., stevensite, kerolite, talc, sepiolite) play an important role in the construction of facies models in lacustrine and peri-marine environments because they are sensitive to changes in solution chemistry. However, the response of Mg-silicate mineralogy to changing aqueous chemistry is only broadly understood because the mechanisms underpinning the coprecipitation of Mg2+and SiO2(aq) from surface water, and subsequent Mg-silicate crystallization, are unclear. Here we describe the results of experiments designed to systematically examine the effects of pH, Mg/Si and salinity of the parent solution on the nature of initially precipitated products. Structural interrogation of the products with X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR) and thermal analysis (TGA/DTA) allow comparison of synthetic products with naturally occurring crystalline counterparts. In general, Mg2+and SiO2(aq) co-precipitation and nucleation of Mg-silicate layer structures first involves the rapid formation of 2:1 layers with trioctahedral occupancy and a mean coherent X-ray scattering domain between 1–2 unit cells with respect to thecaxis. Well defined but diffusehkreflections indicate two-dimensional growth, turbostratic stacking and highly variable interlayer hydration. Diffuse reflectance FTIR shows numerous structural similarities with stevensite, kerolite and sepiolite. However, TGA/DTA analysis indicates the presence of variable kerolite/stevensite interstratification not readily detectable through XRD analyses, as well as a significant degree of surface and interlayer hydration (e.g. 15–20 wt.%).We observe a number of clear trends in the products with respect to solution chemistry. For example, at low salinity, kerolite-like products dominate at high Mg/Si and high pH, whereas sepiolite-like products are formed at lower pH and lower Mg/Si. At high salinity and high Mg/Si, stevensite-like products are favoured at high pH and kerolite-like products dominate at lower pH, whereas a decrease in Mg/Si of the solution leads to sepiolite-like products at low pH and only stevensite-like products at high pH. Higher pH leads to an increase in octahedral vacancies which favour stevensite-like products; this may result from a higher rate of two-dimensional tetrahedral sheet expansion relative to the octahedral sheet, as inferred from studies of silica oligomerization and brucite growth kinetics.Together, our results indicate that the neoformation of Mg-rich silicates from solution may often begin with the rapid nucleation of hydrated 2:1 layers. Subsequent dehydration leads to progressive layer stacking order and could occur in response to wetting/drying cycles, prolonged exposure to high salinity solutions, or burial and heating. The surface and interlayer water associated with these products is undoubtedly an important source of diagenetic water in Mg-silicate-bearing successions, and the chemistry of this water upon later diagenesis should be a focus of future investigation.

Processes of Zinc Attenuation by Biogenic Manganese Oxides Forming in the Hyporheic Zone of Pinal Creek, Arizona

The distribution and speciation of Zn sorbed to biogenic Mn oxides forming in the hyporheic zone of Pinal Creek, AZ, was investigated using extended X-ray absorption fine structure (EXAFS) and microfocused synchrotron X-ray fluorescence (μSXRF) mapping, and chemical extraction. μSXRF and chemical extractions show that contaminant Zn co-varied with Mn in streambed sediment grain coatings. Bulk and microfocused EXAFS spectra of Zn in the biogenic Mn oxide coating are indicative of Zn forming triple-corner-sharing inner-sphere complexes over octahedral vacancies in the Mn oxide sheet structure. Zn desorbed in response to the decrease in pH in batch experiments and resulted in near-equal dissolved Zn at each pH over a 10-fold range in the solid/solution ratio. The geometry of sorbed Zn was unchanged after 50% desorption at pH 5, indicating that desorption is not controlled by dissolution of secondary Zn phases. In summary, these findings support the idea that Zn attenuation in Pinal Creek is largely controlled by sorption to microbial Mn oxides forming in the streambed during hyporheic exchange. Sorption to biogenic Mn oxides is likely an important process of Zn attenuation in circum-neutral pH reaches of many acid-mine drainage contaminated streams when dissolved Mn is present.

Magnetism of iron oxide based core-shell nanoparticles from interface mixing with enhanced spin-orbit coupling

We show that the magnetism of core-shell nanoparticles (made of maghemite, $\ensuremath{\gamma}$-Fe${}_{2}$O${}_{3}$, cores and transition-metal and metal-oxide shells) is altered substantially by the interface, which is a doped iron-oxide layer formed naturally during the seed-mediated synthesis process, a route used typically to produce core-shell nanoparticles. Characteristics fundamental to useful applications, such as the anisotropy and superparamagnetic blocking temperature, were altered substantially with Cu, CoO, MnO, and NiO shells. To ascertain the origin of this behavior, the prototype $\ensuremath{\gamma}$-Fe${}_{2}$O${}_{3}$/CoO core-shell nanoparticles are described in detail. We show that the magnetism originates essentially from an interfacial doped iron-oxide layer formed via migration of shell ions, e.g., Co${}^{2+}$, into octahedral site vacancies in the surface layers of the $\ensuremath{\gamma}$-Fe${}_{2}$O${}_{3}$ core. For this system, an overall Fe ${m}_{\mathrm{orb}}/{m}_{\mathrm{spin}}=0.15\ifmmode\pm\else\textpm\fi{}0.03$ is measured (${m}_{\mathrm{orb}}\ensuremath{\sim}0$ for the Fe-oxides) and an enhanced Co ${m}_{\mathrm{orb}}/{m}_{\mathrm{spin}}=0.65\ifmmode\pm\else\textpm\fi{}0.03$ elucidates the origin of the unexpectedly high overall anisotropy of the nanoparticle. This interfacial layer is responsible for the overall (e.g., bulk) magnetism and provides a perspective on how the magnetism of core-shell nanoparticles manifests from the selected core and shell materials.

Coupling of the local defect and magnetic structure of wüstite Fe1−xO

The local nuclear and magnetic structure of wustite, Fe1-xO, and the coupling between them, has been examined using reverse Monte Carlo refinements of variable-temperature neutron total scattering data. The results from this analysis suggest that the individual units in a tetrahedral defect cluster are connected along <110> vectors into a Koch-Cohen-like arrangement, with the majority of octahedral vacancies concentrated near these defects. Bond valence calculations indicate a change in the charge distribution on the cations with the charge on the tetrahedral interstitials increasing on cooling. The magnetic structure is more complex than previously thought, corresponding to a non-collinear spin arrangement described by a superposition of a condensed spin wave on the established type-II antiferromagnetic ordering. This leads to an architecture with four groups of cations each with different spin directions. The cations within the interstitial clusters appear to be weakly ferromagnetically coupled and their spins are correlated to the spins of the octahedral cations closest to them. This work not only provides further insight into the local structure of wustite but also a better understanding of the coupling between defect structures and magnetic and charge-ordering in complex materials.

Positron Annihilation in Off-Stoichiometric and Ta-Doped Zn2 TiO4

While Zn vacancies are regarded as the charge compensators in Ti-rich Zn2TiO4, no evidence of cation vacancies in essentially single-phase samples containing up to 13 wt% excess TiO2 was evident from positron annihilation lifetime spectroscopy (PALS). Nor was there strong evidence of cation vacancies in Ta-doped samples targeted to contain either tetrahedral or octahedral vacancies, although there was a small increase in the lifetime as the doping level increased and the structure changed from cubic to tetragonal.

Extended planar defects and the rapid incorporation of Ti4+ into olivine

The formation of extended planar defects in minerals such as olivine is related to high point defect concentration and can be driven by large gradients in chemical potential, where the energy of the system is lowered by the ordering of defects along specific planes in the crystal. The presence of extended defects has the potential to create the (apparently) anomalous ionic diffusion in olivine as reported recently (Spandler and O’Neill in Contrib Mineral Petrol 159(6):791–818, 2010). High-resolution transmission electron microscopy and energy-filtered imaging were done using experimental samples designed to examine the impact of a TiO2 and f O2 on the potential to form such defects in ferromagnesian olivine. Doped basalt (5 wt% TiO2)–olivine reaction couple experiments were run at 1 atm and 1,310 and 1,410 °C for 50 h at various f O2, ranging from 102 below to 102 above the quartz–fayalite–magnetite buffer. Our results show that extended planar defects in olivine, parallel to {101}ol and occurring in ordered “clusters” with a prolate spheroid geometry ~5–25 nm across and extending up to 150 nm into the olivine, are present near the olivine–glass interfaces in all of our experimental high-TiO2 basalt–olivine samples. Increased Ti content in the olivine is associated with the defects; ordering of Ti4+ and octahedral site vacancies leads to a two- or three-layer superstructure in the olivine. Defect nucleation and growth is driven by the large TiO2 chemical potential gradient across the phase boundary at the start of the experiments, which provides access to microstructures not otherwise present.