Search

Conspectus: Photoelectron spectroscopy (PES) is a powerful tool for the investigation of liquid–vapor interfaces, with applications in many fields from environmental chemistry to fundamental physics. Among the aspects that have been addressed with PES is the question of how molecules and ions arrange and distribute themselves within the interface, that is, the first few nanometers into solution. This information is of crucial importance, for instance, for atmospheric chemistry, to determine which species are exposed in what concentration to the gas-phase environment. Other topics of interest include the surface propensity of surfactants, their tendency for orientation and self-assembly, as well as ion double layers beneath the liquid–vapor interface. The chemical specificity and surface sensitivity of PES make it in principle well suited for this endeavor. Ideally, one would want to access complete atomic-density distributions along the surface normal, which, however, is difficult to achieve experimentally for reasons to be outlined in this Account. A major complication is the lack of accurate information on electron transport and scattering properties, especially in the kinetic-energy regime below 100 eV, a pre-requisite to retrieving the depth information contained in photoelectron signals.In this Account, we discuss the measurement of the photoelectron angular distributions (PADs) as a way to obtain depth information. Photoelectrons scatter with a certain probability when moving through the bulk liquid before being expelled into a vacuum. Elastic scattering changes the electron direction without a change in the electron kinetic energy, in contrast to inelastic scattering. Random elastic-scattering events usually lead to a reduction of the measured anisotropy as compared to the initial, that is, nascent PAD. This effect that would be considered parasitic when attempting to retrieve information on photoionization dynamics from nascent liquid-phase PADs can be turned into a powerful tool to access information on elastic scattering, and hence probing depth, by measuring core-level PADs. Core-level PADs are relatively unaffected by effects other than elastic scattering, such as orbital character changes due to solvation. By comparing a molecule’s gas-phase angular anisotropy, assumed to represent the nascent PAD, with its liquid-phase anisotropy, one can estimate the magnitude of elastic versus inelastic scattering experienced by photoelectrons on their way to the surface from the site at which they were generated. Scattering events increase with increasing depth into solution, and thus it is possible to correlate the observed reduction in angular anisotropy with the depth below the surface along the surface normal. We will showcase this approach for a few examples. In particular, our recent works on surfactant molecules demonstrated that one can indeed probe atomic distances within these molecules with a high sensitivity of ∼1 Å resolution along the surface normal. We were also able to show that the anisotropy reduction scales linearly with the distance along the surface normal within certain limits. The limits and prospects of this technique are discussed at the end, with a focus on possible future applications, including depth profiling at solid–vapor interfaces.

We report valence band photoelectron spectroscopy measurements of gas-phase and liquid-phase benzene as well as those of benzene dissolved in liquid ammonia, complemented by electronic structure calculations. The origins of the sizable gas-to-liquid-phase shifts in electron binding energies deduced from the benzene valence band spectral features are quantitatively characterized in terms of the Born-Haber solvation model. This model also allows to rationalize the observation of almost identical shifts in liquid ammonia and benzene despite the fact that the former solvent is polar while the latter is not. For neutral solutes like benzene, it is the electronic polarization response determined by the high frequency dielectric constant of the solvent, which is practically the same in the two liquids, that primarily determines the observed gas-to-liquid shifts.

Liquid-jet photoelectron spectroscopy was applied to determine the first acid dissociation constant (pKa) of aqueous-phase glucose while simultaneously identifying the spectroscopic signature of the respective deprotonation site. Valence spectra from solutions at pH values below and above the first pKa reveal a change in glucose’s lowest ionization energy upon the deprotonation of neutral glucose and the subsequent emergence of its anionic counterpart. Site-specific insights into the solution-pH-dependent molecular structure changes are also shown to be accessible via C 1s photoelectron spectroscopy. The spectra reveal a considerably lower C 1s binding energy of the carbon site associated with the deprotonated hydroxyl group. The occurrence of photoelectron spectral fingerprints of cyclic and linear glucose prior to and upon deprotonation are also discussed. The experimental data are interpreted with the aid of electronic structure calculations. Our findings highlight the potential of liquid-jet photoelectron spectroscopy to act as a site-selective probe of the molecular structures that underpin the acid–base chemistry of polyprotic systems with relevance to environmental chemistry and biochemistry.

Nature 595(7869), 673 - 676 (2021). doi:10.1038/s41586-021-03646-5, Insulating materials can in principle be made metallic by applying pressure. In the case of pure water, this is estimated1 to require a pressure of 48 megabar, which is beyond current experimental capabilities and may only exist in the interior of large planets or stars2,3,4. Indeed, recent estimates and experiments indicate that water at pressures accessible in the laboratory will at best be superionic with high protonic conductivity5, but not metallic with conductive electrons1. Here we show that a metallic water solution can be prepared by massive doping with electrons upon reacting water with alkali metals. Although analogous metallic solutions of liquid ammonia with high concentrations of solvated electrons have long been known and characterized6,7,8,9, the explosive interaction between alkali metals and water10,11 has so far only permitted the preparation of aqueous solutions with low, submetallic electron concentrations12,13,14. We found that the explosive behaviour of the water–alkali metal reaction can be suppressed by adsorbing water vapour at a low pressure of about 10$^{−4}$ millibar onto liquid sodium–potassium alloy drops ejected into a vacuum chamber. This set-up leads to the formation of a transient gold-coloured layer of a metallic water solution covering the metal alloy drops. The metallic character of this layer, doped with around 5 × 10$^{21}$ electrons per cubic centimetre, is confirmed using optical reflection and synchrotron X-ray photoelectron spectroscopies., Published by Nature Publ. Group, London [u.a.]

A versatile, temperature controlled apparatus is presented, which generates deeply cooled liquid microjets of condensed gases, expelling them via a small aperture into vacuum for use in photoelectron spectroscopy (PES). The functionality of the design is demonstrated by temperature- and concentration-dependent PES measurements of liquid ammonia and solutions of KI and NH4I in liquid ammonia. The experimental setup is not limited to the usage of liquid ammonia solutions solely.

For deeply supercooled liquids the transition from a two-stage freezing process to complete solidification in just one freezing step occurs at the hypercooling temperature, a term that seems to be almost unknown in water research; to our knowledge, it has only been mentioned by Dolan et al. for high-pressure ice. The reason for the absence of this expression may be that the best estimate to be found in the literature for the hypercooling temperature of water is about −160 °C (113 K). This temperature is far below the limit of experimentally realizable degrees of supercooling near −40 °C (233 K), which marks the homogeneous nucleation temperature TH of common pure water; in fact, it is even below the glass-transition temperature (133 K). Here we show that, surprisingly, a more thorough analysis taking into account the temperature dependence of the heat capacities of water and ice as well as of the enthalpy of freezing shows that the hypercooling temperature of water is about −64 °C or 209 K, almost 100 K h...

Photoelectron spectroscopy of microjets expanded into vacuum allows access to orbital energies for solute or solvent molecules in the liquid phase. Microjets of water, acetonitrile and alcohols have previously been studied; however, it has been unclear whether jets of low temperature molecular solvents could be realized. Here we demonstrate a stable 20 μm jet of liquid ammonia (-60 °C) in a vacuum, which we use to record both valence and core-level band photoelectron spectra using soft X-ray synchrotron radiation. Significant shifts from isolated ammonia in the gas-phase are observed, as is the liquid-phase photoelectron angular anisotropy. Comparisons with spectra of ammonia in clusters and the solid phase, as well as spectra for water in various phases potentially reveal how hydrogen bonding is reflected in the condensed phase electronic structure.

The object of the study is the C2D4 molecule, as it is important to know its properties to address numerous problems of molecular physics. The analysis of high-resolution spectra of the deuterated ethylene molecule was made in the range of 600–1200 cm–1, specifically bands ν7 and ν12. The results obtained were used to determine high-accurate values of the vibrational-rotational levels of the ground vibrational state of the C2D4 molecule.

The freezing of freely suspended supercooled water droplets with a diameter of bigger than a few micrometers splits into two rather different freezing stages. Within the first very fast dendritic freezing stage a spongy network ice with an ice portion of less than one-third forms and more than two-thirds of liquid water remain. In the present work the distribution of the ice portion in the droplet directly after the dendritic freezing phase as well as the evolution of the ice and temperature distribution has been investigated in dependence of the most relevant parameters as droplet diameter, dendritic freezing velocity (which correlates with the supercooling) and heat transfer coefficient to the surroundings (which correlates with the relative droplet velocity compared to the ambient air and with the droplet size). For this purpose on the experimental side acoustically levitated droplets in climate chambers have been investigated in combination with high-speed cameras. The obtained results have been used for finite element method (FEM) simulations of the dendritic freezing phase under consideration of the beginning second, much slower heat-transfer dominated freezing phase. A theoretical model covering 30 layers and 5 shells of the droplet has been developed which allows one to describe the evolution of both freezing phases at the same time. The simulated results are in good agreement with experimental as well as with calculated results exploiting the heat balance equation. The most striking result of this work is the critical radius of the droplet which describes the transition of one-stage freezing of the supercooled water droplet toward the thermodynamically forced dendritical two-stage freezing in which the droplet cannot sufficiently get rid of the formation heat anymore. Depending on the parameters named above this critical radius was found to be in the range of 0.1 to 1 μm by FEM simulation.

The high resolution infrared spectrum of the 34 S 16 O 2 molecule was re-investigated in the region of 1600–1900 cm −1 . The earlier studied by Lafferty et al., the ν 2 + ν 3 band was re-analyzed under consideration of transitions with higher values of quantum numbers J and K a . The weak ν 1 + ν 2 cold band and the 2 ν 2 + ν 3 - ν 2 hot band were analyzed for the first time. The obtained experimental data were used in the weighted fit procedure based on the effective Hamiltonian which takes into account resonance interactions between a variety of vibrational states. The obtained set of spectroscopic parameters reproduces the initial experimental data with accuracies close to experimental uncertainties.

The high resolution infrared spectrum of the 32S16O18O molecule was recorded for the first time with a Bruker IFS 120 HR Fourier transform interferometer in the region of 930–1580 cm−1 where the bands ν1 and ν3 are located. More than 3000 and about 2400 transitions were assigned in the experimental spectrum with the maximum values of quantum numbers J max . / K a max . equal to 58/23 and 68/23 to the bands ν1 and ν3, respectively. The further weighted fit of experimentally assigned transitions was made with the Hamiltonian model which takes into account Coriolis resonance interaction between the vibrational states (100) and (001). The 81 microwave transitions of the states (100) and (001) known from the literature also were taken into account. As the result, a set of 26 fitted parameters was obtained which reproduces the experiment-based 2690 ro-vibrational energy values of the two bands with the d rms = 1.8 × 10 − 4 cm −1. Microwave transitions are also reproduced with the accuracy close to experimental uncertainty.

The high resolution infrared spectrum of the 34S16O2 molecule was recorded in the region of 2600–2810 cm−1 where the strong 2 ν 3 band is located. About 2500 transitions with maximum values of quantum numbers J max . = 76 and K a max . = 26 were assigned to the 2 ν 3 band. The very weak 2 ν 1 + ν 2 band was recorded and analyzed for the first time. This gave us the possibility to find and analyze, for the first time, the hot band 2 ν 1 + ν 2 - ν 2 , which is located in the 2200–2380 cm−1 region and totally covered by the strong 2 ν 1 band. In general, about 1370 transitions were assigned to the bands 2 ν 1 + ν 2 and 2 ν 1 + ν 2 - ν 2 , and 302 ro-vibrational energies of the (2 1 0) state were determined. Ro-vibrational energies obtained from experimental data for both the (0 0 2), and (2 1 0) vibrational states were used in the weighted list square fit of the parameters of the effective Hamiltonian, which takes into account additional interactions with the (1 3 0) and (0 5 0) vibrational states. The set of parameters obtained from the fit reproduces the initial experimental data with accuracies close to experimental uncertainties.

Highly accurate, ~ ( 1 − 2 ) × 10 − 4 cm − 1 , ro-vibrational spectrum of S18O2 was recorded with Bruker IFS 120 HR Fourier transform interferometer in the region of 1050–1400 cm−1 where the bands ν1 and ν3 are located. About 1560 and 1840 transitions were assigned in the experimental spectrum with the maximum values of quantum numbers J max . / K a max . equal to 65/22 and 58/16 to the bands ν3 and ν1, respectively. The further weighted fit of experimentally assigned transitions was made with the Hamiltonian model which takes into account Coriolis resonance interaction between the vibrational states (100) and (001). To make the ro-vibrational analysis physically more suitable, the initial values of main spectroscopic parameters have been estimated from the values of corresponding parameters of the S16O2 species on the basis of the results of the Isotopic Substitution theory. Finally, the set of 23 spectroscopic parameters obtained from the fit reproduces values of 1292 initial “experimental” ro-vibrational energy levels (about 3400 transitions assigned in the experimental spectrum) with the d rms = 0.00015 cm − 1 . Also, the ground state parameters of the S18O2 molecule were improved.

Alkali metals can react explosively with water and it is textbook knowledge that this vigorous behaviour results from heat release, steam formation and ignition of the hydrogen gas that is produced. Here we suggest that the initial process enabling the alkali metal explosion in water is, however, of a completely different nature. High-speed camera imaging of liquid drops of a sodium/potassium alloy in water reveals submillisecond formation of metal spikes that protrude from the surface of the drop. Molecular dynamics simulations demonstrate that on immersion in water there is an almost immediate release of electrons from the metal surface. The system thus quickly reaches the Rayleigh instability limit, which leads to a ‘coulomb explosion’ of the alkali metal drop. Consequently, a new metal surface in contact with water is formed, which explains why the reaction does not become self-quenched by its products, but can rather lead to explosive behaviour. The explosion of alkali metals in water is a typical high-school chemistry experiment, but its mechanism is not fully understood. Using high-speed cameras and molecular simulations it is now shown that a key early step in this reaction is the migration of electrons from the alkali metal into water, leading to a charging of the metal's surface and subsequent Coulomb explosion.

An electrical phenomenon consisting of two alternating voltage peaks of up to 6 V amplitude was observed during the rapid dendritic freezing phase of supercooled water droplets in the millimeter size range with supercoolings ΔT in the range of 5 to 20 K. For correlation of the dendritic freezing front with the electric potential, a fast recording oscilloscope was combined with a high-speed camera operating at up to 5000 frames per second. The strength of the effect is roughly proportional to the supercooling and dendritic freezing speed. Furthermore, during the subsequent second freezing phase, which is much slower than the dendritic, a qualitatively different electric potential evolution of similar magnitude has been found which resembles the well-investigated Workman-Reynolds freezing potential (WRFP). The experiments show clear evidence that the first rapid dendritic freezing stage significantly influences direction and amount of the electric potential during the second slow freezing stage. Compared to the WRFP, which takes place for much smaller supercoolings of ΔT ≪ 5 K, the evolution of the presented dendritic freezing potential occurs about 10(4) times faster, is about 10 times smaller in view of the maximum voltage, and shows similar break off concentrations but remarkably does not vanish at low foreign ion concentrations. This phenomenon has direct relevance to atmospheric freezing processes of the Earth, other planets, and satellites.

Alkali metals in water are always at the brink of explosion. Herein, we show that this vigorous reaction can be kept in a non-exploding regime, revealing a fascinating richness of hitherto unexplored chemical processes. A combination of high-speed camera imaging and visible/near-infrared/infrared spectroscopy allowed us to catch and characterize the system at each stage of the reaction. After gently placing a drop of a sodium/potassium alloy on water under an inert atmosphere, the production of solvated electrons became so strong that their characteristic blue color could be observed with the naked eye. The exoergic reaction leading to the formation of hydrogen and hydroxide eventually heated the alkali metal drop such that it became glowing red, and part of the metal evaporated. As a result of the reaction, a perfectly transparent drop consisting of molten hydroxide was temporarily stabilized on water through the Leidenfrost effect, bursting spectacularly after it had cooled sufficiently.

The characterization of liquid–vapor interfaces at the molecular level is an important underpinning for a basic understanding of fundamental heterogeneous processes in many areas, such as atmospheric science. Here we use X-ray photoelectron spectroscopy to study the adsorption of a model surfactant, octanoic acid, at the water–gas interface. In particular, we examine the information contained in photoelectron angular distributions and show that information about the relative depth of molecules and functional groups within molecules can be obtained from these measurements. Focusing on the relative location of carboxylate (COO-) and carboxylic acid (COOH) groups at different solution pH, the former is found to be immersed deeper into the liquid–vapor interface, which is confirmed by classical molecular dynamics simulations. These results help establish photoelectron angular distributions as a sensitive tool for the characterization of molecules at the liquid–vapor interface.