ARTICLE pubs.acs.org/JPCA
Adsorption of Water Molecules on Selected Charged SodiumChloride Clusters James A. Bradshaw,*,†,‡ Sidney L. Gordon,‡ Andrew J. Leavitt,§ and Robert L. Whetten‡ ‡
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia, United States Department of Chemistry, North Georgia College and State University, Dahlonega, Georgia, United States
§
ABSTRACT: The adsorption of water molecules (H2O) on sodium chloride cluster cations and anions was studied at 298 K over a mass range of 1001200 amu using a custom-built laser desorption ionization reactor and mass spectrometer. Under the conditions used, the cations Na3Cl2+ and Na4Cl3+ bind up to three water molecules, whereas the larger cations, Na5Cl4+ to Na19Cl18+, formed hydrates with one or two only. The overall trend is a decrease in hydration with increasing cluster size, with an abrupt drop occurring at the closed-shell Na14Cl13+. As compared to the cluster cations, the cluster anions showed almost no adsorption. Among smaller clusters, a weak adsorption of one water molecule was observed for the cluster anions Na6Cl7 and Na7Cl8. In the higher mass region, a substantial adsorption of one water molecule was observed for Na14Cl15. Density functional theory (DFT) computations were carried out for the adsorption of one molecule of H2O on the cations NanCln1+, for n = 28, and the anions NanCln+1, for n = 17. For each ion, the structure of the hydrate, the hydration energy, and the standardstate enthalpy, entropy, and Gibbs energy of hydration at 298 K were computed. In addition, it was useful to compute the distortion energy, defined as the electronic energy lost due to weakening of the NaCl bonds upon adsorption of H2O. The results show that strong adsorption of a H2O molecule occurs for the linear cations only at an end Na ion and for the nonlinear cations only at a corner Na ion bonded to two Cl ions. An unexpected result of the theoretical investigation for the anions is that certain low-energy isomers of Na6Cl7 and Na7Cl8 bind H2O strongly enough to produce the observed weak adsorption. The possible implications of these results for the initial hydration of extended NaCl surfaces are discussed.
I. INTRODUCTION Reversible adsorption of water (H2O) molecules to sodiumchloride (NaCl) nano- and microstructures is crucial to processes occurring in the marine and arctic aerosol environment, as well as to the processing of NaCl nanocrystals.1 These factors have motivated repeated theoretical investigations of this ubiquitous system,2 but to-date only limited experimental investigations have succeeded in addressing the initial stages of the adsorption processes.35 We have therefore developed a custom-built, variable-temperature cluster source and reactor, with mass-selective detection, to detect the full range of clusters and their adsorption-reaction products over a range of near-ambient conditions. The experimental results are complemented by electronic-structure computations, executed at the B3LYP functional level of densityfunctional theory, to provide structural interpretation of the adsorption phenomena. Oceanic sea-salt is one of the most abundant cluster (particulate) species in the atmosphere.6 Worldwide seawater has an average NaCl concentration7 of ∼0.6 M. Lunar and storm driven tide wave-action provides a constant and significant supply of aerosolized salt species to the atmosphere.6 The role of gas phase NaCl species is suspected in many atmospheric r 2011 American Chemical Society
chemical reactions though many, if not most, of the specific chemical contributions remain unknown. In fundamental chemistry, the hydration and dissolution of NaCl clusters is of significant interest both for their role in atmospheric processes as well as offering insight into the dissolution kinetics of alkali-halide salts. The full understanding of their chemical activity, such as possible catalytic sites for the destruction of tropospheric ozone or as deliquescent agents, remains deficient in the literature. The rudimentary understanding of the solvation of salts, with varying size and defect density, played a crucial role in the motivation for further investigation. An indication of their presence in interstellar dust grains8 also demands a complete understanding of all possible chemical processes. An exceptional collection of literature on dry NaCl clusters exists, both experimental911 and theoretical1214 treatments, describing predicted and confirmed15 structures. Investigations of water on bulk salt surfaces16,17 have also existed for some time and are relatively numerous. A small collection of studies3,4 exists for the hydration of smaller (NanClm+: n e 6) clusters species but are, at Received: July 7, 2011 Revised: November 14, 2011 Published: November 18, 2011 27
dx.doi.org/10.1021/jp206433r | J. Phys. Chem. A 2012, 116, 27–36
The Journal of Physical Chemistry A
ARTICLE
Figure 1. Mass spectra of NanCln1+ cations (a) without a reactant gas pulse over the mass range 100600 amu; (b) without a reactant gas pulse over the mass range 6001200 amu; (c) same as (a), but with a reactant gas pulse of the H2O/He mixture, where (n,m) is NanCln1+/(H2O)m; (d) same as (b), but with a reactant gas pulse of the H2O/He mixture, where (n,m) is NanCln1+/(H2O)m. Peak labeled 14* is the Na14Cl12 containing a free electron in place of a chlorine atom.
best, only tangentially related to the atmospheric analogue. More common in the literature are studies on NaCl parallels, such as NaI and other alkalihalide species.4 Prior to this investigation, adsorption on NaCl clusters was only vaguely understood and specific to very small unique and large bulk species. The molecular-to-bulk transition region, presumed to offer the most insight as to the driving kinetics that lead to aggregation in the atmosphere, was studied in detail for a select group of exclusive species under atmospherically relevant conditions.
minimize the likelihood or sticking to the die walls (making a sample largely unusable). Mass analysis of the sample targets with and without the use of cyclohexane where found to be identical and it therefore is assumed that the cyclohexane either evaporated or at most played a spectator role in the experiment. B. Apparatus. Salt clusters were produced in a laser desorption ionization (LDI) reactor, using the frequency-tripled output of a Nd:YAG laser (355 nm, 3.5 eV) as an ablation source as compared to the frequency doubled (532 nm, 2.3 eV) and fundamental (1064 nm, 1.2 eV). An inline iris was used to regulate roughly ∼15 mJ of energy per pulse. High purity helium carrier gas seeded with water vapor via an inline bubbler at a backing pressure of ∼7 bar was introduced into the reactor using a Series 9 pulsed valve (General Valve Company Inc.) A typical pulse width of ∼300 μs provided an estimated total reactor pressure of ∼0.5 bar. The calculated resident time for cluster aggregation and thermalization was significantly large, at 120 μs. For collection of dry NaCl cluster mass spectra, all carrier gas plumbing, starting at the He gas storage cylinder, was evacuated using an Edwards Inc. Stage-2 (#5) rotary vane pump and left overnight. An inline helium purification filter, UOP Inc. model P-1002, was placed at the output of the helium gas regulator. Regular maintenance and cleaning of this inline filter ensured that no unknown reactants entered the reactor assembly. The overall reactor assemblies cluster output was oriented orthogonal to a custom build time-of-flight mass spectrometer (TOF-MS) with a variable deflection voltage acting as a selective mass-range band-pass filter.
II. EXPERIMENTAL METHODS A. Sample Preparation. For experimental preparation, a 1/8” cylindrical LDI target rod was prepared in a custom hydraulic die using 99.999% (by % weight metal impurity) granular sodium chloride (Alfa Aeser #10862). Approximately 2 g of salt were placed inside the hardened high-carbon steel die and pressed, with a tungsten rod, at approximately 7.5kbar. The resulting salt sample was attached with cyano-acrylate glue to a 1/4” carrier sleeve and placed in the reactor assembly. A great deal of trial and error transpired in attempts to prepare the salt target. Previous work with salts and assorted varieties of organic components, such as para-aminobenzoic acid (PABA), showed that small additions of a select liquid maximized the yield and quality of the target. It was determined that the presence of a small quantity (
