Diffraction-limited mid-infrared microspectroscopy to reveal a micron-thick interfacial water layer signature.

Mid-infrared microspectroscopy is a non-invasive tool for identifying the molecular structure and chemical composition at the scale of the probe, i.e. at the scale of the beam. Consequently, investigating small objects or domains (commensurable to the wavelength) requires high-resolution measurements, even down to the diffraction limit. Herein, different protocols and machines allowing high-resolution measurements in transmission mode (aperture size (i.e., beam size) from 15 × 15 µm to 3 × 3 µm) are tested using the same sample. The model sample is a closed cavity containing a water-air assemblage buried in a quartz fragment (fluid inclusion). The spectral range covers the water stretching band (3000-3800 cm-1), whose variations are followed as a function of the distance to the cavity wall. The experiments compare the performance of one focal plane array (FPA) detector associated with a Globar source with respect to a single-element mercury cadmium telluride (MCT) detector associated with a supercontinuum laser (SCL) or a synchrotron radiation source (SRS). This work also outlines the importance of post-experimental data processing, including interference fringe removal and Mie scattering correction, to ensure that the observed spectral signatures are not related to optical aberrations. We show that the SCL and the SRS-based setups detect specific spectral features along the quartz boundary (solid surface), invisible to the FPA imaging microscope. Additionally, the broadband SCL thus has the potential to substitute at the laboratory scale the SRS for conducting diffraction-limited high-resolution measurements.

Metastable phase equilibria in the ice II stability field. A Raman study of synthetic high-density water inclusions in quartz.

Microthermometric measurements of a synthetic high-density (984 kg m-3) water inclusion in quartz revealed that only part of the super-cooled liquid water (L) transforms to solid ice Ih upon ice nucleation (L → ice Ih + L). While ice nucleation occurs in the ice Ih stability field at -41 °C and 28 MPa the pressure increases instantaneously to 315 MPa into the ice II stability field. At this point, both phases, liquid water and ice Ih are metastable. The coexistence of these two phases was confirmed by Raman spectroscopy and could be traced down to -80 °C. The pressure along this low-temperature metastable extension of the ice Ih melting curve was determined by means of the frequency shift of the ice Ih peak position using both the O-H stretching band around 3100 cm-1 and the lattice translational band around 220 cm-1. At -80 °C and 466 MPa the super-cooled ice Ih melting curve encounters the homogeneous nucleation limit (TH) and the remaining liquid water transformed either to metastable ice IV (ice Ih + L → ice Ih + ice IV) or occasionally to metastable ice III (ice Ih + L → ice Ih + ice III). The nucleation of ice IV resulted in a pressure drop of about 180 MPa. Upon subsequent heating the pressure develops along a slightly negatively sloped ice Ih-ice IV equilibrium line terminating in a triple point at -32.7 °C and 273 MPa, where ice IV melts to liquid water (ice Ih + ice IV → ice Ih + L). Hitherto existing experimental data of the ice IV melting curve (ice IV → L) were found to be in line with the observed ice Ih-ice IV-liquid triple point. If, on the other hand, ice III nucleated at -80 °C (instead of ice IV) the associated pressure drop was about 260 MPa. The ice Ih-ice III-liquid triple point was determined at -22.0 °C and 207 MPa (ice Ih + ice III → ice Ih + L), which is in agreement with previous experimental data.

Pore scale modelling of DNAPL migration in a water-saturated porous medium.

A numerical simulator based on the discrete network model approach has been developed to simulate drainage processes in a water-saturated porous medium. To verify the predictive potential of the approach to simulate the unstable migration of a dense nonaqueous phase liquid (DNAPL) at the pore scale, the numerical model was applied to laboratory experiments conducted on a sand-filled column. The parameters relative to pore body size and pore throat size used in the construction of the equivalent network were derived from discrete grain-size distribution of the real porous medium. The observed water retention curve (WRC) was first simulated by desaturation of the network model. The good agreement of the modelled WRC with the experimental one highlights that the applied approach reproduces the main characteristics of the real pore space. The numerical model was then applied to rate controlled experiments performed on a homogenous sand-filled column to study the gravity-driven fingering phenomenon of immiscible two-phase flow of water and a DNAPL. The numerical results match within 10% based on the standard deviation with the experiments. They correctly reproduce the effect of several system parameters, such as flow mode (upward flow and downward flow) and the flow rate, on the stability of the water/DNAPL front in a saturated porous medium.

Oversolubility in the microvicinity of solid-solution interfaces.

Water-solid interactions at the macroscopic level (beyond tens of nanometers) are often viewed as the coexistence of two bulk phases with a sharp interface in many areas spanning from biology to (geo)chemistry and various technological fields (membranes, microfluidics, coatings, etc.). Here we present experimental evidence indicating that such a view may be a significant oversimplification. High-resolution infrared and Raman experiments were performed in a 60 × 20 µm(2) quartz cavity, synthetically created and initially filled with demineralized water. The IR mapping (3 × 3 µm(2) beam size) performed using the SOLEIL synchrotron radiation source displays two important features: (i) the presence of a dangling free-OH component, a signature of hydrophobic inner walls; (ii) a shift of the OH-stretching band which essentially makes the 3200 cm(-1) sub-band predominate over the usual main component at around 3400 cm(-1). Raman maps confirmed these signatures (though less marked than IR's) and afforded a refined spatial distribution of this interfacial signal. This spatial resolution, statistically treated, results in a puzzling image of a 1-3 µm thick marked-liquid layer along the entire liquid-solid interface. The common view is then challenged by this strong evidence that a µm-thick layer analogous to an interphase forms at the solid-liquid interface. The thermodynamic counterpart of the vibrational shifts amounts to around +1 kJ mol(-1) at the interface with a rapidly decreasing signature towards the cavity centre, meaning that vicinal water may form a reactive layer, of micrometer thickness, expected to have an elevated melting point, a depressed boiling temperature, and enhanced solvent properties.

Gibbs free energy of liquid water derived from infrared measurements.

Infrared spectra of pure liquid water were recorded from 20 cm(-1) to 4000 cm(-1) at temperatures ranging from 263 K to 363 K. The evolution of connectivity, libration, bending and OH stretching bands as a function of temperature follows the evolution of the inter-molecular dynamics, and so gives insight into the internal energy averaged over the measurement time and space. A partition function, which takes into account the inter-molecular and intra-molecular modes of vibration of water, all variable with the molecular networking, was developed to convert this vibrational absorption behavior of water into its macroscopic Gibbs free energy, assuming the vibrational energy to feature most of the water energy. Calculated Gibbs free energies along the thermal range are in close agreement with the literature values up to 318 K. Above this temperature, contributions specific to the non H-bonded molecules must be involved to closely fit the thermodynamics of water. We discussed this temperature threshold in relation to the well-known isosbestic point. Generally speaking, our approach is valuable to convert the IR molecular data into mean field properties, a quantitative basis to predict how water behaves in natural or industrial settings.

Influence of surface conductivity on the apparent zeta potential of TiO2 nanoparticles: application to the modeling of their aggregation kinetics.

Titanium dioxide nanoparticles (TiO2 NPs) are extensively used in consumer products. The release of these NPs into aquatic environments raises the question of their possible risks to the environment and human health. The magnitude of the threat may depend on whether TiO2 NPs are aggregated or dispersed. Currently, limited information is available on this subject. A new approach based on DLVO theory is proposed to describe aggregation kinetics of TiO2 NPs in aqueous dispersions. It has the advantage of using zeta potentials directly calculated by an electrostatic surface complexation model whose parameters are calibrated by ab initio calculations, crystallographic studies, potentiometric titration and electrophoretic mobility experiments. Indeed, the conversion of electrophoretic mobility measurements into zeta potentials is very complex for metal oxide nanoparticles. This is due to their very high surface electrical conductivity associated with the electromigration of counter and co-ions in their electrical double layer. Our model has only three adjustable parameters (the minimum separation distance between NPs, the Hamaker constant, and the effective interaction radius of the particle), and predicts very well the stability ratios of TiO2 NPs measured at different pH values and over a broad range of ionic strengths (KCl aqueous solution). We found an effective interaction radius that is significantly smaller than the radius of the aggregate and corresponds to the radius of surface crystallites or small clusters of surface crystallites formed during synthesis of primary particles. Our results confirm that DLVO theory is relevant to predict aggregation kinetics of TiO2 NPs if the double layer interaction energy is estimated accurately.

Infra-red imaging of bulk water and water-solid interfaces under stable and metastable conditions.

Superheated water has been studied by infrared spectroscopy to examine whether the special ability of liquid water to undergo such a metastable state corresponds to the development of peculiar inter-molecular networking under tension. As the best technique to superheat water is to trap the liquid inside micro-cavities in solids (the so-called "fluid inclusions"), the role of the water-solid interfaces to stabilize the adjoining liquid is also explored with the same infra-red micro-spectroscopy tool. The key signal is the intra-molecular OH stretching band, sensitive to the networking in the probed material. The sample of choice is liquid water occluded inside a quartz cavity of micrometric size, synthesized in laboratory from pure quartz and milli-Q water. The stretching band of the superheated water shows no significant spectral difference from that of a bulk "normal" water, which means that the molecular properties of the superheating liquid are quite similar to those of the stable bulk liquid. Liquid water is readily "superheatable" but retains its "normality" under these special conditions. Additionally, this result establishes a firm ground to justify that the properties of the former are predicted extrapolating the usual (though empirical) equation of state of the latter. The infra-red signals of the water-solid interfaces are more complex. The water-solid interfaces blue-shift the signal, affecting differently the three sub-bands of the OH-stretching. This effect was unexpected since the micro-IR spectroscopy probes volume beyond what is classically assigned for the interfacial properties. In addition, the interfacial signature is clearer under superheating than under the saturation conditions, which offers an interesting (and unexpected) way to interpret the special stability of the occluded metastable water. These encouraging results give confidence on the potentialities of the high-resolution micro-spectroscopy to get insights into the molecular basis of macroscopic properties.

Salt precipitation and trapped liquid cavitation in micrometric capillary tubes.

Laboratory evidence shows that the occurrence of solid salt in soil pores causes drastic changes in the topology of the porous spaces and possibly also in the properties of the occluded liquid. Observations were made on NaCl precipitation in micrometric cylindrical capillary tubes, filled with a 5.5 M NaCl aqueous solution and submitted to drying conditions. Solid plug-shaped NaCl (halite) commonly grows at the two liquid-air interfaces, isolating the inner liquid column. The initially homogeneous porosity of the capillary tube becomes heterogeneous because of these two NaCl plugs, apparently closing the micro-system on itself. After three months, we observed cavitation of a vapor bubble in the liquid behind the NaCl plugs. This event demonstrates that the occluded liquid underwent a metastable superheated state, controlled by the capillary state of thin capillary films persisting around the NaCl precipitates. These observations show, first, that salt precipitation can create a heterogeneous porous medium in an initially regular network, thus changing the transfer properties due to isolating significant micro-volumes of liquid. Second, our experiment illustrates that the secondary salt growth drastically modifies the thermo-chemical properties of the occluded liquid and thus its reactive behavior.

In-pore tensile stress by drying-induced capillary bridges inside porous materials.

We present here some evidences that capillary liquid bridges are able to deform micrometric cylindrical pores by tensile stress. Brine-soaked filter membranes are submitted to drying conditions leading to NaCl precipitation inside the 5-10 µm pores. A close examination demonstrated that two forms of NaCl crystallites are successively generated. First, primary cubic crystals grow, driven by the permanent evaporation. When this angular primary solid gets near the pore wall, while the evaporation makes the pore volume to be partly invaded by air, capillary liquid can bridge the now-small gap between the halite angles and the pore wall. In a second step, these small capillary bridges are frozen by a secondary precipitation event of concave-shaped NaCl. The proposed interpretation is that the liquid capillary bridges deform the host matrix of the membrane, and the situation is fossilized by the growth of solid capillary bridges. A quantitative interpretation is proposed and the consequences towards the natural media outlined.

Diffuse reflectance infrared Fourier transform spectroscopy as a tool to characterise water in adsorption/confinement situations.

We present experimental data acquired by diffuse reflectance infrared spectroscopy in the mid-IR (4000-400 cm(-1)), on micrometric-sized mineral grain powders. The spectral evolution of the OH-stretching band is followed when the adsorbed water film is thinned under dry conditions, from high to low hydration states. The IR bands are found to be characteristic of the degree of adsorption/confinement of the liquid water. The OH-stretching band is shifted toward shorter wavenumbers than in bulk water, showing that a significant portion of adsorbed water has a higher intermolecular bonding energy. Complementary treatment of the kinetics of water desorption, varying with the surface forces in the water film, confirms the relationships of these bands with the constrained water state. We distinguish different water types obeying liquid-liquid interactions (free and capillary water) or dominated by solid-water interactions (confined and adsorbed water). Part of this study is devoted to mesoporous silica MCM-41, of interest due to the restricted geometries of its mesopores (4.7 nm) favouring the confined water state. The methodology allows us to distinguish bulk and adsorbed/confined water, using spectral analysis coupled with an understanding of the dynamic behaviour of the desorption process.