HYPOTHESIS:The surface-parallel depth-decaying layering at the room-temperature ionic liquid (RTIL)/air interface should exhibit a thermal behavior different from bulk's locally-layered structure. The absence of an (attractive) upper-half-space of matter above the surface should compress the layer spacing and provide for a larger thermal variation compared to the bulk. At the same time, the flat and sharp liquid surface is expected to align the layers macroscopically, and thus enhance the layering depth beyond the range of the bulk's local layering. EXPERIMENTS:Temperature (T)-dependent X-ray reflectivity (XRR) measurements off the free surfaces of a homologous series of model RTILs [Cnmim][NTf2] (Cn, even n=12-18, where n is the carbon number of the cationic hydrocarbon chain). Such combined (n,T) surface structure evolution is not currently available for any aprotic homologous RTIL series. FINDINGS:The surface layering exhibits an anomalous thermal contraction, as does the bulk, rather than a conventional expansion, with surface layer spacings falling below those of the bulk by 5-25%, and exhibiting larger contraction-increases with increasing T. A 10-35% increase in the spacing-normalized layering range was also found at the surface compared to the bulk for all n. Surprisingly, a hardening of the surface layers against T-variations is found for increasing n, shown by the decrease of the surface spacing's thermal slope |∂ds(T)/∂T| with increasing n, while the bulk softens, with its |∂db(T)/∂T|increasing with n. These bulk/surface differences originate in the effects hypothesized above.

This Comment raises several questions concerning the surface structure concluded in the paper referenced in the title. Specifically, that paper ignores previous experiments and simulations which demonstrate for the same ionic liquids depth-decaying, multilayered surface-normal density profiles rather than the claimed molecular mono- or bi-layers. We demonstrate that the claimed structure does not reproduce the measured X-ray reflectivity, which probes directly the surface-normal density profile. The measured reflectivities are found, however, to be well-reproduced by a multilayered density model. These results, and previous experimental and simulation results, cast severe doubt on the validity of the surface structure claimed in the paper referenced in the title.

Room temperature ionic liquids (RTILs), a novel class of liquid salts, are intensively studied for their basic science and numerous emerging applications. When undercooled, RTILs comprising long alkyl chains often exhibit liquid crystal (LC) bulk phases. However, only one molecular-resolution experimental structure study was published for their LC surface phases. We measured the temperature evolution of another LC surface phase, using surface specific angstrom-resolution X-ray methods. This phase's existence range, 90 degrees C, much exceeds the corresponding bulk phase's 3 degrees C. Its thickness, L, confirms the theory-predicted logarithmic temperature dependence, with an amplitude equaling the bulk correlation length. Surprisingly, at L's divergence temperature, a similar to 20 angstrom thick, hexagonally packed, crystalline monolayer forms at, and fully covers, the sample's surface. It is identified as a surface-frozen Langmuir-Gibbs film and fundamentally differs from the only reported RTIL surface crystal, a Coulomb-dominated, four-layer, island phase, covering only 5%-15% of the surface.

Significance This high-resolution X-ray study resolves the liquid–air interface structure for a long homologous series of room temperature ionic liquids (RTILs). RTILs are intensely studied for many potential “green” applications and for their intriguingly complex and rare combination of intermolecular interactions. Varying their cation’s alkyl chain length provides, therefore, an opportunity to tune the main interaction from mostly long-range electrostatic to mostly short-range van der Waals. This variation is found here to drive the interface structure from simple, to layered, to liquid crystalline. The quantitative results obtained constitute an accurate yardstick for testing simulations and theory, impact the bulk–surface structure relations in general, and provide currently scarce data for many RTIL applications, like batteries and supercapacitors.

Correction for ‘Self-segregated nanostructure in room temperature ionic liquids’ by Diego Pontoni et al., Soft Matter, 2017, DOI: 10.1039/c7sm01464c.

The nanosegregated bulk structure, and its evolution with the cation's alkyl length n, are studied by X-ray scattering for an unprecedentedly broad homologous series of a model room-temperature ionic liquid, [CnMIM][NTf2] (n = 4-22). A tri-periodic local structure is found, with the lateral periodicities, dII and dIII independent of n, and a longitudinal one, dI, linearly increasing with n. The results are consistent with a local structure comprising alternating layers of polar headgroups and apolar, interdigitated, partly overlapping, cations' alkyl tails, of an average macroscopic mass density close to that of liquid alkanes. A slope decrease in the linear dI(n) suggests a change from a lower to a higher rate of increase with n of chain overlap for n ≥ 12. The order decay lengths of the layering, and of the lateral chain packing, increase with n, as expected from the increasing van der Waals interaction's domination of the structure. The headgroups' lateral packing decay length decreases with n, due to increasing frustration between the longer lateral periodicity preferred by the headgroups, and the shorter lateral periodicity preferred by the chains. A comparison of the bulk and surface structures highlights the surface's ordering effect, which, however, does not induce here a surface phase different from the bulk, as it does in liquid crystals and liquid alkanes.

X-ray reflectivity (XR) and atomistic molecular dynamics (MD) simulations, carried out to determine the structure of the oil-water interface, provide new insight into the simplest liquid-liquid interface. For several oils (hexane, dodecane, and hexadecane) the XR shows very good agreement with a monotonic interface-normal electron density profile (EDP) broadened only by capillary waves. Similar agreement is also found for an EDP including a sub-Å thick electron depletion layer separating the oil and the water. The XR and MD derived depletions are much smaller than reported for the interface between solid-supported hydrophobic monolayers and water.

Molecular self-assembly is a key to wide-ranging nano- and microscale applications in numerous fields. Understanding its underlying molecular level science is therefore of prime importance. This study resolves the angstrom-scale structure of the earliest and simplest self-assembled monolayer (SAM), octadecanol on amorphous-SiO2-terminated Si(001) substrate, and determines the structures temperature evolution. At low temperatures lateral hexagonal order exists, with close-packed, surface-normal molecules. At similar to 12 degrees C above the alkanols bulk melting, a fully reversible disordering transition occurs to a novel stretched liquid phase, laterally disordered, but only similar to 15% thinner SAM than in the crystalline phase. The SAM persists to >= 100 degrees C. A thermodynamic model yields here a headgroup-substrate bond energy similar to 40% lower than on crystalline sapphire, highlighting the importance of the substrates order, and near-epitaxy, for the SAMs ordering and stability.

X-ray reflectivity measurements of increasingly more complex interfaces involving silicon (001) substrates reveal the existence of a thin low-density layer intruding between the single-crystalline silicon and the amorphous native SiO2 terminating it. The importance of accounting for this layer in modeling silicon/liquid interfaces and silicon-supported monolayers is demonstrated by comparing fits of the measured reflectivity curves by models including and excluding this layer. The inclusion of this layer, with 6-8 missing electrons per silicon unit cell area, consistent with one missing oxygen atom whose bonds remain hydrogen passivated, is found to be particularly important for an accurate and high-resolution determination of the surface normal density profile from reflectivities spanning extended momentum transfer ranges, now measurable at modern third-generation synchrotron sources.