Experiments and theory both show that chloride, bromide, and iodide ions are readily available at the interface.  In fact, iodide ions prefer the interface.  Calculations indicate that nitrate prefers the interface while sulfate does not. 

Experiments include aerosol chamber experiments and diffuse reflectance infrared Fourier transform spectrometry (DRIFTS).  The aerosol chamber is a 561 L chamber equipped with a variety of particle generating and measurement systems, including:

• an atomizer and a nebulizer

• a differential mobility analyzer with a condensation particle counter

• an aerosol particle spectrometer

In situ spectroscopic analytical techniques include long path FTIR and UV-visible.  A tandem quadrupole mass spectrometer with atmospheric pressure ionization provides high sensitivity for measurement of a variety of gases such as the molecular halogens.

Photo of our aerosol chamber.  On the right are the optics for long path FTIR and UV-visible and on the left is the atmospheric pressure ionization mass spectrometer.  The inset shows an end-on view of the chamber when the photolysis lamps, mounted on top of the chamber, are on. 
See enlarged view

In a collaborative effort with Alex Laskin, Dan Gaspar, and Jim Cowin at PNNL, particles are collected from the chamber using the PNNL TRAC system.  They are then analyzed using SEM-EDS and TOF-SIMS.

 Photo of the TRAC system at PNNL

Quantitative interpretation of the aerosol chamber data requires modeling all of the physical (e.g. diffusion, mass accommodation, etc.) and chemical processes accurately.  The MAGIC model was developed for this purpose.

The MAGIC (Model of Aerosol and Gas Interfacial Chemistry) model, which was developed by the Dabdub group, includes detailed halogen chemistry in the gas phase, in the bulk liquid of particles, and at the interface.  It is currently being expanded to include oxide of nitrogen and sulfur chemistry for application to the atmosphere. 

DRIFTS (Diffuse Reflectance Infrared Fourier Transfer Spectrometry) provides a sensitive, real-time method for monitoring the formation of infrared-absorbing products of salt reactions.  The cell also has a window that permits photolysis to be carried out  while monitoring the infrared spectrum of the solid.

Photo of our DRIFTS apparatus.  The left photo shows the elliptical mirrors swung back out of the optical path and the sampling base with a salt (white) sample.  The right photo shows the mirrors in position and the base holding the sample covered by the vacuum chamber.

Typical data for the formation of nitrate on NaCl by its reaction with gas phase NO₂ are shown below.  Also shown are changes that occur upon photolysis of this surface nitrate film, where nitrate is observed to decrease but other products on the salt surface such as nitrite are not formed.  This illustrates that the photochemistry of thin surface films of nitrate is significantly different than that of bulk crystalline nitrate salts which are well known to form nitrite on photolysis.  

DRIFTS spectrum of NaCl (a) after reaction with NO₂.  The bands observed are due to the nitrate ion.  (b) After photolysis of the nitrate ion formed as in (a).  Unlike photolysis of NaNO₃, no nitrite is formed which suggests that the photochemistry of nitrate at the interface is different than bulk  crystalline nitrate.  From Vogt and Finlayson-Pitts, J. Phys. Chem. 99 17269 (1995).

The segregation of halide ions both on the surface of solid salts and in thin water films on the salts is being studied using X-ray photoelectron spectroscopy (XPS).

On solid salts, reactions with species such as gas phase nitric acid or ozone exposure are followed using conventional high vacuum XPS after exposure of the salts to the relevant gases in a fast-entry lock chamber or using a doser.  The apparatus is shown in the photograph below.

Photograph of XPS system in Hemminger laboratory

On solid salts, reactions with species such as gas phase nitric acid or ozone exposure are followed using conventional high vacuum XPS after exposure of the salts to the relevant gases in a fast-entry lock chamber or using a doser. 

In collaborative studies with Miquel Salmeron at the advanced light source (ALS) at Lawrence Berkeley National Laboratory, the segregation of bromide and chloride ions at the surface of deliquescing particles has been studied using high pressure XPS.  In these experiments, XPS on salt surfaces can be carried out in the presence of several Torr of water vapor which, if the sample is maintained at low temperature, is above the deliquescence point.  (See High Pressure Photoelectron Spectroscopy (HPPES) system schematic.)

Molecular dynamics calculations provide guidance and support for the experiments.  These theoretical studies are carried out by the Tobias, Jungwirth, and Gerber groups.  For example, snapshots of the distribution of sodium, fluoride, chloride, bromide, and iodide ions from the MD simulations are shown below.  These simulations show that fluoride ions shy away from the interface, chloride ions are distributed about equally over the bulk and interface but that bromide and iodide ions are enhanced at the interface.

Results of MD simulations of sodium halides dissolved in water, showing the distribution of ions in the water layer.  The left side shows snapshots of the distribution of the ions in the water and the right shows the relative concentration profiles; blue is the oxygen (water) distribution, green is the sodium ion, while black, yellow, orange, and magenta represent the halide ions.  From Jungwirth and Tobias, J. Phys. Chem. B. 105 10468 (2001).

Molecular dynamics simulations are also being integrated with the photoelectron spectroscopy (PES) experiments of Lai-Sheng Wang at PNNL/Washington State University.  For example, the PES experiments have shown that the sulfate anion is embedded inside water clusters [X. B. Wang, J. B. Nicholas and L.-S. Wang, J. Chem. Phys. 113, 10837 (2000)], which is supported by MD calculations [P. Jungwirth, J. E. Curtis, and D. J. Tobias, Chem. Phys. Lett. 367; 704 (2003)].

In collaborations with Bruce Garrett and Liem Dang at PNNL and Chris Mundy at LLNL, the propensity of the interface to hold gases such as O₃, OH and H₂O₂  and the electronic structure of the interface are being explored.