Atomic and Molecular Collisions at Liquid Surfaces - byui.edu · PDF fileAtomic and Molecular Collisions at Liquid Surfaces ... School of Engineering and Physical Sciences, Heriot-Watt

PC67CH22-Minton ARI 2 May 2016 21:34

Atomic and MolecularCollisions at Liquid SurfacesMaria A. Tesa-Serrate,1 Eric J. Smoll Jr.,2

Timothy K. Minton,2 and Kenneth G. McKendrick1

1Institute of Chemical Sciences, School of Engineering and Physical Sciences, Heriot-WattUniversity, Edinburgh EH14 4AS, United Kingdom; email: [email protected] of Chemistry and Biochemistry, Montana State University, Bozeman,Montana 59717; email: [email protected]

Annu. Rev. Phys. Chem. 2016. 67:515–40

First published online as a Review in Advance onApril 18, 2016

The Annual Review of Physical Chemistry is online atphyschem.annualreviews.org

This article’s doi:10.1146/annurev-physchem-040215-112355

Copyright c© 2016 by Annual Reviews.All rights reserved

Keywords

gas–surface scattering, gas–liquid interface, inelastic scattering, reactivescattering, surface composition, surface structure

Abstract

The gas–liquid interface remains one of the least explored, but neverthelessmost practically important, environments in which molecular collisions takeplace. These molecular-level processes underlie many bulk phenomena offundamental and applied interest, spanning evaporation, respiration, multi-phase catalysis, and atmospheric chemistry. We review here the research thathas, during the past decade or so, been unraveling the molecular-level mech-anisms of inelastic and reactive collisions at the gas–liquid interface. Armedwith the knowledge that such collisions with the outer layers of the interfacialregion can be unambiguously distinguished, we show that the scattering ofgas-phase projectiles is a promising new tool for the interrogation of liquidsurfaces with extreme surface sensitivity. Especially for reactive scattering,this method also offers absolute chemical selectivity for the groups that reactto produce a specific observed product.

515

Click here to view this article'sonline features:

• Download figures as PPT slides• Navigate linked references• Download citations• Explore related articles• Search keywords

ANNUAL REVIEWS Further

Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.

http://www.annualreviews.org/doi/full/10.1146/annurev-physchem-040215-112355


SAM: self-assembledmonolayer

1. INTRODUCTION

Although the gas–liquid interfacial region is ubiquitous in environmental (1–4) and industrial(5–7) chemical systems, it is still a challenge to study this region with current experimental andcomputational tools. Relative to bulk liquid and bulk gas, far less is known about the boundarybetween them. The investigation of collisions of atoms or molecules with liquid surfaces pro-vides unique insight into this important interface that complements spectroscopic and electron orneutron scattering studies.

Gas–liquid scattering dynamics are at the heart of the microscopic events that give rise tosolvation, phase transport, and heterogeneous catalysis. In this section, we summarize generaland important dynamical features of gas–liquid scattering. Many such features that have beenestablished for gas–solid scattering are also observed in gas–liquid scattering (8–10). For example,gas–liquid scattering can often productively be described as a binary combination of statistical(thermal) and nonstatistical (impulsive) scattering limiting distributions. Additionally, trends inthe dependence of energy transfer on deflection angle for impulsive scattering are consistent withhard-sphere kinematics for both liquids and solids. However, liquid interfaces are disordered,aperiodic, dynamic, and more prone to thermal roughening than solid surfaces (11, 12), resultingin broad energy and angular distributions for gas–liquid scattering.

The surface selectivity of gas–liquid collisions, which are restricted to or gated by the outermostregion of the surface, makes gas–liquid scattering a potentially very powerful analytical tool forthe study of liquid surface structure and composition. In Section 2, we highlight experiments thatdemonstrate this utility. Relative to solids, the liquid interface is less defined and more structurallycomplex (13, 14). The existence of surface-specific chemical and physical phenomena (15) suggeststhat the gas–liquid interface can function as a reaction medium distinct from vacuum and frombulk liquid. Although much remains to be learned about the surfaces of pure liquids, many currentgas–liquid scattering studies are building on the groundwork of previous studies and targetingcomplex liquid mixtures (16, 17).

This review is timely because of the substantial advances that have been made in the field ofgas–liquid interfaces in the past decade or so. These interfaces have been studied with a varietyof methods, including sum frequency generation, X-ray photoelectron spectroscopy, and X-ray,light, neutron, electron, and ion scattering. However, we focus here on atomic and molecular col-lisions with liquid surfaces because these interactions may provide a new way in which to obtaincomplementary information on surface composition and structure with a high degree of surfacespecificity and chemical selectivity. Owing to limitations of space, we restrict our review to the scat-tering of gaseous atoms, molecules, and radicals from hydrocarbons, polymers, ionic liquids, andliquid crystals. Experimental, theoretical, and computational results are interleaved and discussedthroughout. Lu et al. (18) have recently reviewed a large body of work on scattering dynamicsfrom self-assembled monolayers (SAMs). Here, we only include SAM studies that are relevant togas–liquid scattering dynamics. Ion–surface scattering, predominantly with much higher-energyprojectiles, as reviewed by Andersson & Ridings (19), is similarly omitted.

2. METHODOLOGIES

2.1. Experimental Approaches to Studying Gas–Liquid Collisions

Some features are common to all interfacial experiments on gas–liquid collision dynamics. Thebasic elements are a liquid surface, a means to generate gas-phase projectiles, and a method todetect the scattered products.

To obtain detailed dynamical information, the nascent products that scatter from the surfacemust be detected before they undergo any secondary encounters in the gas phase in order to

516 Tesa-Serrate et al.

Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


Wheel

Beam

Scraper

Motor

Scatteredproducts

Liquid

Liquidreservoir

Figure 1Liquid reservoir with a rotating wheel to create a continually refreshed liquid surface for molecular beam–surface scattering experiments.

avoid the exchange of energy and the loss of memory of the interfacial interaction. This needfor a reasonably large mean free path implies that experiments must be performed in a vacuumand therefore, without special measures, restricts the choice of liquids to those with low vaporpressures. Liquids that meet this requirement include glycerol, long-chain hydrocarbons suchas squalane (2,6,10,15,19,23-hexamethyltetracosane) or sufficiently long linear alkanes (n � 20),polymers such as PFPE (perfluoropolyether), ionic liquids, concentrated acids, and concentratedsalt solutions. Recently, some promising steps have been made toward the study of the gas–liquidinterfaces of aqueous solutions and other liquids with high vapor pressures (see Future Issues).The majority of work to date has involved low–vapor pressure liquids in static reservoirs or liquid-coated wheels in vacuum chambers. The use of a wetted wheel (Figure 1) is probably the mostpopular method for studies of gas–liquid interfacial dynamics in vacuo. In this approach, pioneeredin 1977 by Lednovich & Fenn (20), a disk of a suitable material is partially immersed in a bathcontaining the liquid of interest. As the wheel rotates, its surface becomes coated with the liquid,which generates an interface that is refreshed on each rotation. A scraper can be used to obtain anevenly covered, smooth surface.

Gas-phase projectiles are introduced into the vacuum chamber in most experiments as a molec-ular beam that is directed toward the surface. This technique has the advantage of giving a definedspeed and angle of incidence to the projectiles. The beam can be pulsed or chopped for temporalresolution. A supersonic expansion may be used to cool the incident internal-state distribution(often effectively into a single quantum state), providing for well-resolved information on internalenergy transfer. Radical species can be generated by laser photolysis or detonation or by electricdischarge of a suitable precursor in the source region of the molecular beam.

An alternative approach has also been adopted in which radical colliders are produced byphotolysis of a bulk pressure of the precursor close to the wheel. This pressure must be lowenough to ensure single-collision conditions. This approach generally produces radicals without

www.annualreviews.org • Collisions at Liquid Surfaces 517

Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


Pulsed valveand nozzle

CO2 laser

Sourcechamber

Liquidreservoir

Quadrupolemass filter

MirrorSkimmer

Chopper wheel

Ionizer

Collimating aperture

To ioncounting

system

Figure 2Schematic diagram of a hyperthermal molecular beam experiment with mass spectrometric detection (32). Shown are the laserdetonation source of high-velocity (hyperthermal) atomic oxygen, synchronized chopper wheel for velocity selection of the beam pulse,continually refreshed liquid surface, and rotatable mass spectrometer detector. Figure adapted with permission from Reference 32.Copyright 2015 American Chemical Society.

a well-defined initial velocity and with a broad distribution of speeds and internal states. Thisrange of initial states may be advantageous, however, as it allows investigation of scattering fromhigher-energy initial states, although the dynamical information will be convoluted over the initialdistributions.

The methods of detection of the scattered products can be broadly split into two categories:mass spectrometry (MS) and laser spectroscopy. MS yields well-resolved information on the trans-lational energy of the products. It is often paired with an incident molecular beam in a configurationwhere the angles of incidence, θ i (conventionally defined relative to the surface normal), and de-tection, θ f , are controlled and the velocity distributions of scattered products are obtained by thetime-of-flight method (Figure 2). Detection by laser spectroscopy, by contrast, provides productinternal-state distributions and, in some cases, translational energy information, although usuallywith more limited resolution. Spectroscopic techniques include laser-induced fluorescence (LIF)to detect I2 (21), OH (22), and NO (23); infrared (IR) absorption to detect HF (24) and CO2 (25);and resonance-enhanced multiphoton ionization (REMPI) to detect NO (26).

2.2. Simple Models and Theoretical Approaches

Experiments are unable to resolve the individual atomic interactions that sum to produce a gas–liquid scattering event. Thus, gas–liquid scattering experiments benefit greatly from efforts tounderstand and model these phenomena through theoretical simulations.

Although most applicable to gas-phase scattering, the hard-sphere binary-collision model(Equation 1) is often invoked to rationalize qualitative trends that are commonly observed ingas–liquid scattering experiments (27). The loss of kinetic energy, �E = Ef − Ei, is a single-valued function of the deflection angle, χ = 180◦ − (θ f + θ i), and is parametrically dependenton the ratio, μ, of the mass of the projectile to the effective mass of the surface, as well as thegas–surface interaction potential, V, and the liquid surface temperature, Tliq. For example, the


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


MD: moleculardynamics

QM/MM: quantummechanics/molecularmechanics

hard-sphere model predicts efficient energy transfer for heavy atoms or molecules that are incidenton low-density liquids, especially for low–impact parameter (head-on) collisions between the pro-jectile and a collection of atoms in the liquid–vacuum interfacial region (28). In many cases, a soft-sphere variant of the hard-sphere model (Equation 2), which accounts for the flow of translationalenergy into internal modes of the scattered species and the localized region of the surface where thecollision occurs, Eint, provides a better description of average fractional energy transfer data (29).This soft-sphere model has been used for high-energy projectiles, where the gas–surface interac-tion potential and thermal motion of the surface may be neglected. In particular, the soft-spheremodel gave improved agreement with experimental energy transfer for hyperthermal O(3P) and Arscattering on squalane, PFPE, alkylthiol SAMs (H-SAMs), and fully fluorinated alkylthiol SAMs(F-SAMs) (29). Equations 1 and 2 represent the hard-sphere and soft-sphere models, respectively:

(�EEi

)HS

= Ei − Ef

Ei≈ 2μ

(1 + μ)2

[1 + μ (sin χ )2 − cos χ

√1 − μ2(sin χ )2

] [1 + V − 2RT liq

Ei

],

(1)

(�EEi

)SS

= Ei − Ef

Ei

≈ 2μ

(1 + μ)2

[1 + μ(sin χ )2 − cos χ

√1 − μ2(sin χ )2 − Eint

Ei(μ + 1) + Eint

Ei

(μ + 1

2μ

)].

(2)

The washboard with moment of inertia (WBMI) model represents surface corrugation as aone-dimensional sinusoidal function and can easily be made to account for multiple collisions inthe interfacial region. Although WBMI calculations are not intended for quantitative comparisonwith experimental data, they do reproduce the bimodal scattering distributions that are oftenobserved in gas–surface scattering experiments (30). However, WBMI calculations also predictthat the effective surface mass of a more floppy liquid (or SAM) is lower than that of a stiffersurface, in contradiction to what has been derived using the soft-sphere model (29).

Both united-atom and all-atom classical molecular dynamics (MD) simulations can be used tocompute the structure and dynamics of liquid surfaces by equilibrating vacuum–liquid–vacuum slabstructures under periodic boundary conditions (Figure 3). The MD equilibrated vacuum–liquidinterface can then be used to estimate which atoms are geometrically accessible to a gas-phaseprojectile by applying accessible surface area or line of sight calculations (31, 32). Classical MDcan also be used to directly simulate nonreactive gas–liquid scattering trajectories (33–36) with anaccuracy that can be improved by fitting force field parameters to high-level quantum calculations(37–40).

The realistic simulation of gas–liquid scattering trajectories that allow for reactive events(bond forming, bond breaking, electron transfer, etc.) is much more challenging. Such simulationrequires the application of quantum mechanics (QM) in “on-the-fly” direct dynamics methods,where successive electronic structure calculations of the energies and forces steer the gas–liquidmotion (41). Computational limitations associated with these quantum Born–Oppenheimerdirect dynamics calculations often motivate decisions to approximate the liquid as a molecularcluster (42) or to adopt hybrid QM/MM schemes, where the gas–liquid system is partitioned intoregions treated classically or with QM (43–51).


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


y

z

xx

Figure 3The vacuum–liquid interface of squalane as predicted by classical molecular dynamics simulations. Theimage renders the unit cell of a vacuum–liquid–vacuum structure equilibrated for 4 ns at 323 K underperiodic boundary conditions. Blue lines trace the 10.0 × 9.3 × 18.3-nm unit cell boundary containing640 molecules of squalane (58,880 atoms). Optimized potentials for liquid simulations–all atom (OPLS-AA)force field parameters were used to describe all bonded and nonbonded interactions in the liquid. Carbonand hydrogen atoms near the upper vacuum–liquid interface are rendered as blue and white spheres,respectively. All other atoms in the liquid slab are rendered as gray points.

3. FUNDAMENTAL DYNAMICS OF INTERFACIAL COLLISIONS

3.1. Nonreactive Gas–Surface Interactions

Nonreactive systems have revealed a great deal about the fundamental scattering dynamics of gas–liquid interfacial collisions. In fact, early work using molecular beams of inert gases provided thebasic foundation for the development of the field (11, 12, 33, 52–56). These early studies have beenpreviously reviewed (28), so we summarize here only the key results that are relevant to all interfa-cial processes. The velocity distributions of the scattered products are found to be well describedempirically as the sum of two components. One component corresponds to fast-recoiling products


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


100 300 500 700 900 1,100

0

50

100

N(t

) / a

rbit

rary

uni

ts

Time of flight / μs

Direct orEley–Rideal

(IS)

Thermal orLangmuir–Hinshelwood

(TD)

Figure 4Representative time-of-flight distribution for reactively scattered HCl following reaction of an incidentCl-atom beam with an average incident energy of 〈Ei〉 = 118 kJ/mol on a squalane surface. The incident andfinal angles, with respect to the surface normal, were θ i = 60◦ and θ f = 45◦, respectively. Yellow circlesrepresent data points; the red curve is a MB fit to the slow (Langmuir–Hinshelwood) component; and theblue curve is the fast (Eley–Rideal) component, which comes from the difference between the data and theMB fit. Abbreviations: IS, impulsive scattering; MB, Maxwell–Boltzmann; N(t), number density; TD, thermaldesorption. Figure adapted with permission from Reference 80. Copyright 2000 AIP Publishing LLC.

IS: impulsivescattering

TD: thermaldesorption

that scatter in a narrow angular distribution at near-specular angles, with final translational energiesthat are superthermal and that depend on the incident energy. The other component is distributedin a cosine distribution around the surface normal and possesses much lower translational energies,resembling a Maxwell–Boltzmann distribution with a temperature close to that of the liquid sur-face. Figure 4 shows a typical time-of-flight distribution illustrating this bimodal behavior. Thisbehavior has been explained as the result of two distinct projectile–surface interactions: impulsivescattering (IS) and thermal desorption (TD). These concepts have been borrowed from studieson gas–solid dynamics (57). According to this distinction, atoms or molecules that scatter with IStrajectories collide with the surface once (or a few times at most) and scatter with relatively high,unaccommodated speeds. There is generally little additional energy transfer to the surface beyondthat resulting from a gas phase–like inelastic collision with a localized region of the surface. In con-trast, TD scattering involves more extended interactions with the surface, promoting exchange ofenergy and resulting in an apparently thermalized distribution of products. The IS and TD labelshave been successfully applied to categorize the bimodal behavior of most gas–liquid systems, in-cluding reactive scattering (of which an early example is shown in Figure 4), so they have becomepart of the common vocabulary of the field. Although the IS/TD dichotomy is undoubtedly avery useful phenomenological tool for characterizing experimental observations, care should beexercised not to overinterpret this simple picture of two limiting cases, as we discuss further below.

Further useful concepts have emerged from early molecular beam studies. Energy transfer tothe surface depends on the collision energy, the angle of incidence, and the masses of both thegas-phase projectile and the localized region of the liquid surface that participates in the collision.It is generally greatest for head-on collisions that lead to backward-scattered products and leastfor grazing collisions. Kinematic models of scattering (Equations 1 and 2) yield an effective surfacemass that decreases with incident energy because the timescale of the collision is shorter and fewersurface atoms move cooperatively. A fluorocarbon such as PFPE, with its heavy CFx groups and


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


dense structure, absorbs less energy than does a hydrocarbon such as squalane (54, 58–60), whosealiphatic chains form a soft surface (26) that promotes a large transfer of momentum. However,recent experiments and simulations using the soft-sphere model (Equation 2) of scattering ofoxygen and argon atoms from liquid and SAM surfaces have found that the effective surface massis greater for softer hydrocarbon surfaces (29). This apparently counterintuitive result is explainedby the fact that more atoms in the hydrocarbon participate in the interaction, so the projectile’senergy is dissipated more easily into the surface. In general, these collisions should be viewed notas local encounters with single CHx or CFx groups, but rather as interactions with many atomsthat are coupled through vibrational motion.

Another useful general concept that has emerged is that of surface roughness. Fluctuations ofliquid surfaces occur on a wide range of length scales, from the macroscopic, where they are mean-ingfully described as sinusoidal capillary waves, to the molecular (11). The shortest-wavelengthmotions, which are most relevant to the scattering of molecular projectiles, are rapidly dampedby viscosity and can be considered as transitory distortions of the local molecular arrangement.A rough surface promotes multiple-bounce collisions, facilitating energy transfer and resultingin a larger TD fraction and more out-of-plane scattering events. Thermal roughening of PFPEsurfaces was demonstrated by scattering different projectiles at multiple temperatures (11, 12). Asthe temperature increases, thermal motion generates a less densely packed surface that is micro-scopically rougher, as later MD simulations have shown (61).

A complete picture of the scattering dynamics requires knowledge of the distributions of boththe internal energy and the velocity of the scattered products. Thus, spectroscopic detectionmethods complement time-of-flight methods. Early studies of NO scattering from SAMs usingREMPI detection found significant translational-to-rotational energy transfer for IS collisions(26, 62). The first experiments with spectroscopic detection that involved a true liquid surfaceused LIF to investigate the scattering of I2 from polymers (PFPE and polydimethylsiloxane) andfrom squalane (21, 58, 63). In all cases, the TD fraction was rotationally and vibrationally warmerthan the incident jet-cooled I2, and there was evidence for weak translational-to-rotational energytransfer in the IS channel. More recently, IR spectroscopy was employed to study the inelasticscattering of a rotationally cold beam of CO2 from liquid surfaces in an extensive series of papersby Nesbitt and coworkers (25, 59, 64–69). The trends in energy transfer with different liquidsurfaces (59) and with different angles (64), collision energies (66), and liquid temperatures (66)were in qualitative agreement with other inelastic scattering experiments. Polarization-modulatedIR spectroscopy was used to investigate orientation and alignment effects on the collisions (67, 69),indicating a strong tendency for forward end-over-end tumbling of the scattered CO2 moleculesas a result of the corrugated structure of the surface, which promotes impulsive torques.

The CO2 rotational distributions were strongly bimodal (Figure 5) and were described by thetwo-temperature Boltzmann model in the following equation:

P (N )2J + 1

= C

[(α

T1

)e

− Erot (N)

kBT1 +(

1 − α

T2

)e

− Erot (N)

kBT2

]. (3)

In this model, the cold temperature T1 is assigned to the TD component and, in the work ofNesbitt and coworkers, constrained to 298 K, whereas the hot temperature T2 is assigned to the IScomponent (25). The TD fraction α has been interpreted by Nesbitt and coworkers as the fractionof incident molecules that become thermalized with the surface. In addition, the Doppler-resolvedabsorption profiles in the direction perpendicular to the surface normal for each rotational statewere fitted to a sum of IS and TD contributions. Although this approach lacks the high translationalresolution of a time-of-flight experiment, it indicates a correlation between the component of


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


7

6

5

4

3

2

1

00 200 400 600

Erot (cm–1)

ln{A

/[S J

(2J +

1)]

}

800 1,000 1,200

Experimental points

Two-temperature fit

Trot(IS) = 715 (36) K

Trot(TD) = 298 K

Figure 5Standard Boltzmann analysis of the J state populations of CO2 molecules that scatter inelastically fromPFPE at normal incidence with 〈Ei〉 = 10.6 kcal/mol. The number in parentheses is one standard deviationfrom the average of multiple measurements. Here Trot is the rotational temperature, A is the integratedabsorbance over a given spectral peak, and SJ is the Honl-London factor. Abbreviations: IS, impulsivescattering; PFPE, perfluoropolyether; TD, thermal desorption. Figure adapted with permission fromReference 25. Copyright 2005 American Chemical Society.

translational energy parallel to the surface and the rotational energy in the scattered CO2. Thevalues of the translational and rotational temperatures for the IS channel were similar, supportingthe IS/TD model. Although there was clear accommodation of translational and rotational energyon the surface, vibrational energy transfer was not so efficient. It was suggested that the timescaleof the interaction is too short to allow for efficient vibrational accommodation. Consistent withthese observations, simulations of CO2 scattering from perfluorinated SAMs found no significantvibrational excitation of CO2 (40, 70) and inefficient relaxation of its excited bending mode (71).

It should be emphasized that there is no a priori reason to expect the IS rotational distributionto be described by a single temperature when it arises from a nonequilibrium process involvingsingle or just a few collisions. Theoretical studies of CO2 collisions with F-SAMs classified di-rect trajectories as those that included both single and multiple collisions, and the TD fractionα obtained from two-temperature rotational fittings did not match the fraction of physisorptiontrajectories (40). Related studies of inert gases colliding with H-SAMs (37, 72–74) and squalane(36) further confounded the distinction between IS and TD, as they showed that single-collisionproducts could be characterized by Maxwell–Boltzmann distributions. It is clear that to fullydescribe the interfacial dynamics at the microscopic level, one needs to go beyond the simpletwo-channel IS/TD model. Inelastic scattering of OH radicals from hydrocarbon and polymersurfaces strengthened this argument (60, 75, 76). The translationally hot but rotationally near-thermal OH produced by HONO photolysis scattered with at least partial IS characteristics.Product rotational distributions could be fitted to a single rotational temperature that was higherthan the incident value, suggesting translational-to-rotational energy conversion in impulsive col-lisions at the surface. This result was supported by simulations of scattering from F-SAMs (77).Rotational temperatures and the component of translationally slower products that could be as-sociated with a TD fraction were consistent with trends in surface stiffness. Further experimentswere performed by changing the OH precursor to allyl alcohol. This generates OH radicals withmuch higher rotational energy, so the distinction between IS and TD channels should be clearer.Although the two-temperature model (Equation 3) could be fitted to the product rotational dis-tributions, it yielded values of α that were in strong disagreement with the TD fractions inferred


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


from translational energy distributions. These results cast doubt on the interpretation of α as asimple and direct measure of the TD fraction.

3.2. Reactive Scattering Systems

Interfacial gas–liquid reactions present added complexity relative to their gas-phase counterpartsbecause the nascent products can accommodate their energy (either partially or totally) at thesurface. The basic IS/TD characteristics of interfacial scattering are still present (24, 78, 79), butin addition, the reaction dynamics generally affect the energies of the products.

Many of the experimental studies to date have focused on hydrogen abstraction reactions fromliquid hydrocarbons, of which the interfacial O(3P) + squalane reaction is perhaps the most ex-tensively characterized example. Molecular beam experiments with MS detection by Minton andcoworkers (78, 80–82) demonstrated that the major product is OH, analogous to gas-phase reac-tions with small hydrocarbons (83). OH scatters in two distinct velocity distributions, consistentwith the IS/TD model for inelastic scattering systems. Borrowing terminology from conventionalsurface science, the direct gas–surface reaction (IS) may be referred to as an Eley–Rideal (ER)reaction, whereas the indirect reaction that leads to thermally desorbed OH would be referred toas a Langmuir–Hinshelwood (LH) reaction (27). TD (or LH) OH products might, in principle,arise from the reaction of thermally accommodated O with a C–H moiety on the surface (butnot in practice; see below) or through the thermal accommodation of a nascent OH after it hasbeen formed in an ER reaction. The OH product, whether produced via an ER or an LH reac-tion, may abstract a second H atom to form H2O. Depending on the sequence of collisions, theH2O product exits the surface with either thermal or superthermal translational energies; bothare found in practice (Figure 6). Given the complexity of the interactions that lead to H2O, itis not surprising that the H2O products scatter with a substantially larger contribution from TDscattering than do the OH products. The observation of a significant yield of nonthermal H2Oproduction is a rare example of a two-step ER reaction. A kinematic analysis of the IS componentof OH using the soft-sphere model (see Section 2.2) yielded an effective surface mass of approxi-mately 76 amu for squalane when the incident O-atom energy was 47 kJ/mol and approximately

0

100

200

300

400500

0°10° 20° 30° 40° 50°

60°70°

80°

Incident O beam

Translational energy / (kJ mol–1 )

Figure 6Plot of the distribution of scattered H2O flux as a function of final energy and angle, following the reactionof O(3P) with a squalane surface with 〈Ei〉 = 504 kJ/mol. The incident angle was θ i = 60◦, and the productswere detected over a range of final angles, θ f , from 0◦ to 80◦. All angles are in the same plane and are definedwith respect to the surface normal, with the incident and final angles being on opposite sides of the surfacenormal. Both thermal and nonthermal distributions are evident. The thermal distribution has a highintensity over a narrow range of low final energies, and it has been cropped. Figure adapted with permissionfrom Reference 81. Copyright 2002 AIP Publishing LLC.


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


43 amu when it was 504 kJ/mol (80, 81). In a similar study, interfacial H abstraction by Cl to formHCl was found to exhibit very similar reaction dynamics (80). Backward and sideways scatteringin the center-of-mass (c.m.) frame was favored for both H abstraction reactions, in contrast withthe analogous gas-phase reactions. This was explained as an unavoidable consequence of the ge-ometrical constraint imposed by the surface. At high c.m. collision energies (451–472 kJ/mol),additional reaction channels are open, and the C–C bond breakage product, CH3O, is observed toscatter from the surface (82) through ER and TD mechanisms. Related theoretical studies, whichcorroborate the experimental observations and conclusions (46, 47), suggest that an H-atom elim-ination reaction should also be significant, but the detection sensitivity for H was insufficient todetect the predicted H-atom products.

The information derived from the molecular beam experiments was complemented by LIFdetection of OH. The bulk photolytic method used to generate O(3P) produces lower (but stillsuperthermal) collision energies (22, 61, 84–86). The translational energy resolution is also morelimited; nevertheless, simulation of the OH appearance profiles (yield as a function of delay be-tween photolysis and probe laser pulses) showed conclusively that there must be a significant IScomponent (84). The presence of both IS and TD components was confirmed more incisively bythe rotational distributions. The overall distribution was described by a single Boltzmann tem-perature that was above the liquid temperature but lower than that from the analogous directgas-phase processes (87). This suggests that not all products are IS and that there was a significantTD contribution, although the IS and TD temperature values were too similar to yield a cleartwo-temperature fit. More refined measurements showed a clear correlation between translationaland rotational energy of the products, with different values of rotational temperature Trot at dif-ferent points of the appearance profile (Figure 7). Trot is superthermal and IS-like in the risingedge of the profiles, becoming close to the surface temperature at later delays (85). Moreover, thefact that the liquid temperature affects Trot only at later delays is additional strong evidence for aTD component contributing increasingly in the tail of the profile.

0 5 10 15 20 25

Rela

tive

fluo

resc

ence

inte

nsit

y

1.0

330 ± 5 K304 ± 3 K

373 ± 5 K381 ± 5 K

0.5

0

Photolysis–probe delay / μs

Figure 7Appearance profile of OH (v = 0) from the O(3P) + squalane reaction, measured on the Q1(1) transition ofthe A–X (1,0) band. Values of OH rotational temperature Trot are shown for the rising edge (6 µs) and peak(11 µs) of the appearance profiles, at surface temperatures of 273 K (blue) and 343 K (red ). Figure adaptedwith permission from Reference 85. Copyright 2006 American Chemical Society.


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


Survival probability(σ ): fraction ofunreactive gas-surfaceencounters

Reactive uptakecoefficient: fractionof gas-surfaceencounters that lead toreaction, defined asγ = 1−σ

The LIF experiments found a significant fraction of OH in its first vibrationally excited level(v = 1) that, by analogy with gas-phase reactions (87), must result from reaction at secondary andtertiary hydrocarbon sites on the surface. MD simulations showed that such sites are accessible,despite the slight preference for methyl groups to protrude toward the vacuum (31, 47). TheOH vibrational branching ratio was lower than had been observed in gas-phase experiments withsmaller hydrocarbons (84), although reproducibility of absolute vibrational branching ratios inindependent experiments is notoriously difficult even in the gas phase (83). Another observation,less dependent on external comparisons, was that the yield of OH (v = 1) was strongly sensitiveto surface temperature, as opposed to OH (v = 0), which was essentially independent (61). Itwas argued that vibrationally excited OH has a liquid temperature–dependent survival (ratherthan production) probability, and that the loss mechanism is collisional relaxation rather thanvibrationally enhanced secondary reaction. The increase in temperature causes the surface ofsqualane to be more open, as demonstrated in MD simulations, potentially enhancing the survivalprobability of OH (v = 1) (61). Importantly, the lack of liquid-temperature dependence in the yieldof the majority OH (v = 0) product precludes a significant contribution from the LH reaction ofthermally accommodated O atoms as the source of TD OH, which could not be ruled out fromthe earlier molecular beam scattering experiments.

Inelastic scattering of OH from different liquids offered further insights into its loss mech-anisms at the surface (60, 75, 76). The reactive uptake of OH by the liquid, derived from thesurvival probability of the incident OH, was larger for the unsaturated liquid squalene comparedto the fully saturated squalane. Experiments with HONO photolysis (75) found reactive uptakecoefficients, γ (squalane) = 0.30 ± 0.08 and γ (squalene) = 0.39 ± 0.07, in agreement with inde-pendent studies of uptake on squalane droplets (88) and other related measurements for thermalOH on hydrocarbon surfaces (89–92). Moreover, the fraction of scattered OH missing from squa-lene was the slower-moving component (75, 76). OH that thermalizes at the surface of squalaneis at least partially inhibited from abstracting a second hydrogen, owing to the 3–10-kJ/mol reac-tion barriers (93), and therefore has a reasonable probability of surviving and scattering with TDcharacteristics. On the squalene surface, in contrast, there is an additional mechanism for loss ofthermalized OH by addition to unsaturated sites, which is a barrierless reaction (94).

In O/OH + saturated hydrocarbon systems, the chemical energy released into the reactionproducts is lower than the collision energy, so the latter dominates the scattering dynamics. Theeffect of reaction energetics on the interfacial dynamics is best observed in the opposite situation,that is, highly exoergic reactions at low collision energies. F + squalane, O(1D) + squalane, andO(3P) + squalene are examples of such reactions. In the gas phase, F + hydrocarbon reactions giverise to HF products with high translational and internal energies. In the F + squalane reaction,both experiments and related simulations (24, 48, 95, 96) have found a fraction of hot products,in agreement with the gas-phase results, but also a component of rotationally and translationallycold products, indicating accommodation at the surface. However, the vibrational energy of thenascent HF is not accommodated efficiently. The situation is similar for the O(1D) + squalane(79) and O(3P) + squalene (97) reactions studied by McKendrick and coworkers: Rotationallyhot OH is produced in analogous gas-phase reactions, owing to reaction by a transient insertionmechanism [in the O(1D) + squalane reaction] or to higher available energy on breaking allylicC–H bonds [in the O(3P) + squalene reaction]. In both cases, LIF experiments found bimodalrotational distributions with a cold component resulting from secondary encounters with thesurface. In addition, the O(3P) + squalene reaction produces a lower yield of OH than the O(3P) +squalane reaction, despite the former’s greater exothermicity. This is further evidence of thepresence of unsaturated groups at the surface of squalene, as discussed above. Addition to doublebonds competes effectively with the direct abstraction reaction, hence diminishing the observable


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


yield of the OH-producing channel, a process consistent with related observations in reactionswith gas-phase alkenes (98–102).

Scattering dynamics have also been used to study proton exchange and solvation when reac-tive atoms and molecules impinge on polar and nonpolar liquids and solutions. In these studies,hydrogen/deuterium isotopic labeling is useful in resolving elementary acid–base events betweenthe projectile and the liquid. Hydrogen halides can scatter impulsively or undergo trapping at aliquid surface. Trapped HX or DX can desorb from the liquid interface without proton exchange,undergo rapid interfacial proton exchange followed by desorption, or succumb to solvation asdissociated ions in the bulk. On glycerol, the majority of incident HBr and HCl molecules, withincident energies of up to 100 kJ/mol, are trapped (at least momentarily) at the liquid–vacuum inter-face (103, 104). However, whereas all trapped DBr is observed to undergo bulk solvation, trappedDCl may desorb without exchange, desorb with exchange, or dissolve into the bulk (103, 104).Relative to pure glycerol, solutions containing NaI and CaI2 slightly increase the peak translationalenergy of IS, inhibit DCl trapping at high collision energies, and promote trapping-desorptionand trapping-exchange-desorption at the expense of trapping-solvation. These observations sug-gest that NaI, CaI2, and NaBr make the glycerol surface mechanically stiffer and smoother, andit is hypothesized that these salts modulate the post-trapping pathways of DCl by decreasing theavailability of the glycerol hydroxyl groups that promote DCl dissociation and H+/D+ transport(15, 105, 106). Like NaI, CaI2, and NaBr, glycerol solutions of the ionic surfactant tetrahexyl-ammonium bromide (THABr) also promote DCl trapping-desorption and trapping-exchange-desorption at the expense of trapping-solvation. It is observed that THABr-glycerol solutionsinfluence the post-trapping pathways of DCl in a manner similar to NaBr-glycerol solutions,only at much lower concentrations. This concentration difference can be rationalized by the sur-face enrichment of the THA+ cation and is supported by observations of reduced IS, indicatinga mechanically softer and rougher vacuum–liquid interface. Also, given that THABr-glyceroland NaBr-glycerol solutions are observed to impact DCl post-trapping pathways similarly withradically different cations, cation–liquid interactions may not be essential in enhancing trapping-desorption and trapping-exchange-desorption at the expense of trapping-solvation-desorptionpathways (107, 108). In contrast to the above salts, KF solutions in glycerol are observed toinhibit trapping-desorption and trapping-exchange-desorption by scavenging interfacial DCl toproduce HF and DF (109). Comparing cold LiBr-D2O solutions with NaBr-glycerol solutions,LiBr-D2O reduces DCl trapping-desorption, blocks trapping-exchange-desorption, and promotestrapping-solvation. All three observations imply that cold LiBr-D2O solutions are more effectiveat drawing interfacial DCl or [D+][Cl−] into the bulk. Brastad & Nathanson (110) suggest thatthis may be the result of superior H+/D+ transport and longer trapping times at the surface ofcold LiBr-D2O.

In an innovative application of atomic beam–liquid scattering, the ionization of Na atoms onthe protic liquid, glycerol, was used to form solvated electrons near the liquid–vacuum interface(42, 111). The solvated electrons induced a number of reactions that were identified throughthe use of selective deuteration of glycerol (Figure 8). Scattered Na atoms were observed inaddition to desorbed reaction products. The time-of-flight distributions of Na atoms showed onlyan IS component, indicating that all trapped Na atoms failed to desorb from the liquid (111).Supporting ab initio MD simulations (42) suggested that this irreversible trapping is because ofNa ionization, which would be expected to produce solvated electrons near the vacuum–liquidinterface. These simulations also identify glycerol hydroxyl group reorientation toward the Naatom as being important in promoting Na ionization (42). The volatile reactive products observedin the experiment (Figure 8) exited the glycerol surface with a Maxwell–Boltzmann velocitydistribution characterized by the surface temperature. A particularly surprising result from this


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


Na

Na+

~57%

OD–

Glycerol radical

Glycerolfragment

Within ~10 μsand ~50 Å

18 ± 3%D

D

D

24 ± 3%D2

58 ± 4%D2O

e–

~80%

~43%

C–Dless

O–Dmore

C–O

Figure 8Schematic diagram of reactive pathways that occur following the effusive molecular beam deposition ofsodium atoms at the vacuum–liquid interface of glycerol. These pathways were determined by the massspectrometric detection of reaction products using three isotopologues of glycerol. Although two pathwaysare believed to contribute to the formation and desorption of water, only the pathway involving OD− isshown. Figure adapted from Reference 111 with permission from AAAS.

study was the observation of H atoms, which demonstrated that electron-induced dissociation ofO–H bonds produced H atoms. Desorption of these H atoms competes with their reaction in thebulk because they are formed very close to the liquid–vacuum interface. In contrast to methodsthat generate solvated electrons with high-energy radiation or electrons, the gentle molecularbeam deposition of Na atoms prevents the creation of high-energy excitations, allowing the studyof a wide range of chemical events that are induced by solvated electrons formed with thermalenergies at the liquid–vacuum interface.

4. COLLISIONS AS AN ANALYTICAL PROBE OF SURFACES

It is undeniable that the field of interfacial collision dynamics has advanced greatly in the pasttwo decades, but its potential goes beyond the purely mechanistic studies discussed above. Nowthat the fundamental dynamics of a significant range of gas–liquid processes are well known andcharacterized, this knowledge can be used to interrogate the surfaces of liquids. Of particularinterest are typically heterogeneous surfaces where surface composition and structure might bearon real-world applications yet might not be easily probed by more conventional methods. Scat-tering of gas-phase species can yield information on the preferential orientation of molecules atthe surface and, in the more complex case of mixtures and solutions, can reveal and potentiallyquantify which components are exposed at the surface.

4.1. Studies of Surface Structure of Pure Molecular Liquids

The first observations of surface structural effects on scattering dynamics were made by Donaldsonand coworkers (112) on the inelastic scattering of I2 from the surface of 4-cyano-4′-pentylbiphenyl(5CB), a thermotropic liquid crystal. LIF detection showed substantial internal excitation in theTD fraction as a result of multiple encounters with the surface. Interestingly, there were importantdynamical differences between the isotropic and nematic phases of 5CB. In the isotropic phase, themolecules are disordered and the internal energy distribution of the scattered I2 resembled thatfrom other organic liquids (21, 58, 63). The nematic liquid, in contrast, induced a higher degree


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


of accommodation that depended on the direction of incidence with respect to the alignmentdirector. I2 molecules approaching perpendicular to the director became more internally excitedas a consequence of the surface being rougher at the microscopic level.

Another early example can be found in the study of the reactions between O(3P) and liquid linearalkanes carried out by McKendrick and coworkers (113). The barriers for abstraction of primary,secondary, and tertiary H atoms differ considerably (87). At the collision energies employed,this gives rise to different reaction probabilities for each C–H bond type, with the yield of OHincreasing in the order 1◦ < 2◦ < 3◦. The experiment found strong variations in the yield of OHwith the length of the alkane molecule. This was interpreted as being the result of different degreesof supramolecular ordering of the chains at temperatures just above the respective melting points.This surface freezing effect, known from earlier independent studies (114), aligns the moleculesperpendicular to the surface in a shallow interfacial region, preferentially exposing their primarygroups and therefore lowering their reactivity. Surface freezing was enhanced for shorter-chainalkanes when they were studied at lower temperatures.

Although the discovery of these structural effects on the dynamics was somewhat accidental,their potential for studying surface compositions was quickly realized. The next section discussesexamples of dynamical studies applied to the surface analysis of complex systems, such as solutionsand ionic liquids.

4.2. Studies of Surface Composition of Mixtures and Ionic Liquids

Atom– or molecule–surface scattering pathways at the liquid–vacuum interface typically involveonly the atoms in the first few atomic layers of the interface. IS is particularly surface specific.High translational energies, specular angular distributions, and reductions in energy transfer withincreasing incident angle imply that IS projectiles undergo one or few collisions with the outermostatoms of the interface. For example, hyperthermal Ar scattering experiments on glycerol solutionsof 2.6 M NaI, 2.6 M KI, 2.5 M LiI, 2.7 M NaBr, and 1.3 M CaI2 all show an increase in the fluxand average energy of IS Ar leaving the liquid surface relative to pure glycerol (15, 105, 106). Thisincreased flux is consistent with the formation of a smoother, stiffer interface, likely owing to OHinteractions with ions in the interfacial region. The increased average energy of IS Ar is consistentwith an increase in the effective surface mass produced by the presence of relatively massive ionsin the interface layer. Taken together, Ar scattering from alkali and alkaline halides in glycerolsuggests that these salts produce ions that occupy the vacuum–liquid interface.

Additional information on the surface composition of glycerol salt solutions can be providedby subthermal CO2 scattering experiments. These studies show that the high-energy componentof the scattered CO2 rotational distribution is enhanced for 2.5 M NaI in glycerol relative topure glycerol. Furthermore, these high-energy rotational states were observed to be sensitive toanion identity (Cl− versus I−) but not cation identity (K+ versus Na+) (115). If it is assumed thatcations and anions influence the structure of the interface equally and that the observed rotationaldistributions are equally sensitive to surface-accessible cations and anions, these results suggestthat the anions dominate the outermost region of the liquid surface.

In many cases, post-trapping pathways that do not involve solvation into the bulk have a verysmall penetration depth and are highly surface specific. By measuring the residence times ofproducts exiting the liquid surface and taking into account diffusion into and out of the liq-uid, Nathanson and coworkers (15, 109) estimated the maximum depth that products couldhave penetrated before scattering through trapping-desorption or trapping-exchange-desorptionpathways. For example, HCl that scatters through a trapping-exchange-desorption pathway (re-leasing DCl) from the surface of 0.5 M KF in glycerol diffuse on average 10 A or less from


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


SBS: sodium 1-butanesulfonate

Ionic liquids (ILs):salts with meltingpoints typically below100◦C, as opposed toinorganic salts withhigh melting points;the term RTIL(room-temperatureIL) is also in commonuse

[Cnmim]: 1-alkyl-3-methylimidazolium;n is the number ofcarbon atoms in thelinear alkyl chain

its trapping site before exiting the liquid. The detection of DF produced by the reaction ofF− with near-interfacial DCl provides evidence that F− can access this region of the interface(109).

Although trapping-solvation-desorption pathways are associated with long residence times inthe liquid bulk, they may be gated by events at the vacuum–liquid interface that are directlyrelated to surface composition. For example, butanol packs loosely on the surface of 60, 64,and 68 wt% D2SO4 at bulk concentrations below 0.18 M and promotes HCl to DCl exchangevia trapping-solvation-desorption pathways (116). In 60 wt% D2SO4, HCl to DCl exchange isobserved to increase with butanol concentration in a manner that correlates with surface tensionmeasurements of interfacial butanol concentration. This HCl to DCl exchange enhancement istoo large to originate from bulk dilution and was not observed for two different concentrationsof sodium 1-butane sulfonate (SBS) on 60 wt% D2SO4. Assuming that the surface structure andcompositions are similar for SBS-D2SO4 and butanol-D2SO4 solutions, these observations suggestthat trapping-solvation-desorption dynamics are sensitive to the concentration of alcohol hydroxylgroups at the liquid–vacuum interface (116). Interestingly, similar studies on hexanol (117) and2-ethyl butanol (118) show a transition from enhancement of HCl to DCl exchange to suppressionwith increasing alcohol concentration, as alkyl group packing more effectively blocks HCl fromreaching the acid surface (119).

Recently, both inelastic and reactive scattering have been employed to characterize the surfacesof ionic liquids (ILs). These are low-temperature molten salts containing large organic cationsand, usually, inorganic anions. Because of the growing interest in their industrial applications(120) and the potential to fine-tune their surface composition, a great deal of experimental efforthas been directed at understanding the principles that govern IL surfaces (see the sidebar, IonicLiquid Surface Analysis) (121, 122). The atomic and molecular scattering experiments reviewedbelow constitute only one of the multiple surface analysis techniques that have been employed tounderstand the surface composition and structure of ILs.

Chemical structures of representative ILs are shown in Figure 9. The majority of ILs studiedto date via dynamical scattering experiments are based on the 1-alkyl-3-methylimidazolium cation(abbreviated as [Cnmim], where n is the number of C atoms in the alkyl chain). Alkyl chains in thecation are the main nonpolar component; consequently, alkyl coverage has been the focus of many

IONIC LIQUID SURFACE ANALYSIS

In principle, the surface composition of an ionic liquid can be controlled by varying the choice of cation and anion,usually by changing the proportion of nonpolar groups with respect to the polar parts of the ion pair. A great deal ofexperimental effort has been directed toward characterizing the relationship between surface and bulk compositions,and each experimental technique has its own strengths and disadvantages. Nonlinear spectroscopies such as sumfrequency generation and second harmonic generation (143) are surface sensitive and provide information on theorientation of molecules, but they do not give quantitative information on surface composition. Being sensitive to allparts of the interfacial region that lack the isotropy of the bulk, they probe to some extent the lower layers of multi-layer structures and not only those exposed directly to vacuum. Scattering techniques such as neutron reflectometry(144), X-ray reflectivity (145), angle-resolved X-ray photoelectron spectroscopy (124), metastable atom–electronspectroscopy (123), and ion scattering [e.g., low-energy ion scattering (146), direct recoil spectrometry (147), andRutherford backscattering spectroscopy (148)] are sensitive to varying penetration depths, so the results of mea-surements with these techniques are typically convoluted with bulk information. In addition, the interpretation ofthe results is not unambiguous and relies on theoretical models of scattering.


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


NN

F

BF F

F

S

F

FF

O

O

O

SN

S

O O

O O

F

FF

F

FF

R

1-Alkyl-3-methylimidazolium

Tetrafluoroborate

Trifluoromethanesulfonate

Bis(trifluoromethylsulfonyl)imide

[BF4]–

[OTf]–

[Cnmim]+

[Tf2N]–

–

–

–

Figure 9Chemical structures of some common ions in room-temperature ionic liquids, shown with their chemicalnames and abbreviations.

surface analysis studies (123–127). The main goal has been to accurately characterize the packingand orientation of the chains and to determine whether it is possible to attain an IL surface thatis fully saturated in alkyl chains and therefore similar to a hydrocarbon surface.

One way to probe alkyl chain coverage is by energy transfer in an inelastic scattering experi-ment. This has been demonstrated by Nesbitt and coworkers by scattering NO (128) and CO2 (50,129) from [C4mim]+ ILs. The chemical identity of the anion [bis(trifluoromethylsulfonyl)imide([Tf2N]−), tetrafluoroborate ([BF4]−), or Cl−] was found to affect the values of the TD fraction andthe internal energies of the scattered products. This result suggests, although somewhat indirectly,the presence of anions at the surface. Cationic alkyl groups provide broadly similar inelastic scat-tering dynamics to well-studied hydrocarbon liquids such as squalane (130). In contrast, scatteringfrom charged ionic moieties has different inelastic dynamical features (usually they increase thestiffness of the surface). Different degrees of accommodation might arise from different fractionsof alkyl coverage, depending on anion identity. The TD fraction of inelastically scattered CO2

also depends on alkyl chain length, indicating a change in surface composition with n (131).The dependence of surface composition on chain length has been more incisively characterized

by reaction with a chemically specific atomic probe. The method, referred to as reactive atomscattering (RAS), provides the desired selectivity toward alkyl chains in addition to high surfacespecificity. The production of scattered OH when O(3P) atoms bombard a surface indicates theexclusive reaction of an incident O atom with an H atom at the liquid–vacuum interface (32, 130,132–134). Thus, the yield of OH in a RAS experiment is a quantitative measure of the hydrogencoverage of the surface. This chemical selectivity can be further tuned by varying the energy ofthe O(3P) atoms. As discussed above, O(3P) atoms resulting from NO2 photolysis give differentsensitivities to different C–H bond types. This means that O(3P) from bulk photolysis preferentiallyprobes secondary H atoms in the alkyl chains over primary H atoms from terminal methyl groups.The dependence of OH yield on C–H bond type has been experimentally corroborated by reactivescattering of O(3P) from deuterium-labeled SAMs (135). In contrast, O(3P) atoms with higherincidence energy, such as those in a hyperthermal beam source (130), have much more similarreactivity toward H atoms in any chemical environment.


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


RAS-MS: reactiveatom scattering withmass spectrometricdetection

RAS-LIF: reactiveatom scattering withlaser-inducedfluorescence detection

Both hyperthermal molecular beam experiments employing O(3P) atoms and MS detection(RAS-MS) (130) and bulk photolysis experiments that used O(3P) atoms paired with LIF (RAS-LIF) (132–134) have demonstrated a propensity for the alkyl chains to orient themselves towardthe vacuum phase. Despite previous conclusions to the contrary based on other surface analyticalmethods (123, 126, 136), the surface clearly does not become saturated with alkyl chains evenfor the IL with the [C12mim] cation. Further studies focusing on the effect of the anion on alkylcoverage (32) showed that anions with smaller ionic volume pack more efficiently at the surfaceand allow for more alkyl chains to cover the surface. MD simulations support all the experimentalfindings: Figure 10 demonstrates the predicted increase in alkyl surface coverage with either

a b c

d e f

g h i

Figure 10Molecular dynamics simulations of the vacuum–liquid interface of nine ionic liquids: (a) [C4mim][Tf2N],(b) [C6mim][Tf2N], (c) [C8mim][Tf2N], (d ) [C4mim][OTf], (e) [C6mim][OTf], ( f ) [C8mim][OTf],( g) [C4mim][BF4], (h) [C6mim][BF4], and (i ) [C8mim][BF4]. Each image is the snapshot of the equilibratedvacuum–liquid interface as one looks down along surface normal. Anions are shown in green. Carbon andnitrogen atoms of the cation imidazolium ring are shown in purple. Carbon atoms belonging to the methyland alkyl substituents on the cation imidazolium ring are shown in blue. Hydrogen atoms are shown inwhite. Figure adapted with permission from Reference 25. Copyright 2015 American Chemical Society.


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


increasing alkyl chain length or decreasing anionic volume, in semiquantitative agreement withRAS-LIF measurements. Thus, RAS has been shown to be an effective new tool to characterizeliquid surfaces.

SUMMARY POINTS

1. There are now well-established experimental methodologies to study inelastic and re-active scattering at the surfaces of low–vapor pressure liquids. Spectroscopic and massspectrometric (MS) methods of detection of nascent products provide complementaryinformation that deepens the understanding of underlying mechanisms.

2. Direct H-atom abstraction reactions from hydrocarbons and related liquids and protonexchange at the surfaces of protic liquids are the best-understood classes of reaction.

3. A pervasive observation is a combination of direct, gas phase–like scattering and morecomplex processes leading to extensive energy loss to the liquid surface. At least em-pirically, translational and rotational product-state distributions can be fitted to binarycombinations of impulsive scattering (IS) and thermal desorption (TD). However, bothexperiment and theory warn against treating these limiting cases too literally.

4. Both inelastic and reactive types of scattering are sensitive to the composition and struc-ture of the outer layers of liquids, with excellent surface specificity. This has been demon-strated for pure molecular liquids, for salty solutions, and especially for ionic liquids (ILs).Reactive atom scattering (RAS) is particularly promising as a surface analytical probe be-cause of its inherent chemical selectivity.

5. Surface structures of liquids predicted through molecular dynamics (MD) simulationsgenerally agree well with those inferred from experiments. It is much more demandingto simulate scattering dynamics successfully for nonreactive and reactive scattering.

FUTURE ISSUES

1. The dynamics of collisions on water and other liquids with high vapor pressures are yet tobe studied. Vacuum-based analytical techniques have successfully been applied to waterby using liquid microjets (137, 138) and microfluidic devices (139). These experimentalapproaches could potentially be adapted for gas–liquid scattering experiments. Indeed,scattering and evaporation of inert gases from liquid hydrocarbon microjets have alreadybeen demonstrated (140, 141), although scattering from water microjets still presentschallenges.

2. Further developments in the methods available to interrogate the mechanisms of dy-namical scattering can be anticipated. These are likely to include new and more sophisti-cated imaging methods for detecting the scattered products. For example, velocity-mapimaging (VMI) with resonance-enhanced multiphoton ionization (REMPI) detectionhas recently been demonstrated for the related scattering of HCl from self-assembledmonolayer (SAM) surfaces (142). This would both broaden the range of product speciesthat can be detected spectroscopically and provide additional mechanistically diagnosticinformation about state-specific velocity distributions.


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


3. In conjunction with improved detection methods, there is considerable scope for therange of inelastic and reactive scattering projectiles to be extended. These would bothopen up studies of distinct mechanistic classes of reaction and further the potential forreactive atom scattering (RAS) to be exploited as an analytical probe for the presence ofdifferent atoms and groups at a wider range of liquid surfaces.

4. Advances in computational power and/or the efficiency of mixed quantum mechanics/molecular mechanics (QM/MM) scattering methods are necessary to make the compar-ison of experimental scattering results with theoretical predictions feasible on a moreroutine basis, particularly for reactive scattering. Experiment currently leads theory sig-nificantly in this area.

DISCLOSURE STATEMENT

The authors are not aware of any affiliations, memberships, funding, or financial holdings thatmight be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

We appreciate the generous support of the US National Science Foundation (NSF-CHE-1266032) and of the UK Engineering and Physical Sciences Research Council (EP/K032062/1).We are grateful to Gilbert Nathanson, Duncan Bruce, John Slattery, George Schatz, BrooksMarshall, Simon Purcell, and Matthew Costen for many helpful discussions.

LITERATURE CITED

1. Pankow JF. 1987. Review and comparative analysis of the theories on partitioning between the gas andaerosol particulate phases in the atmosphere. Atmos. Environ. 21:2275–83

2. O’Dowd CD, Smith MH, Consterdine IE, Lowe JA. 1997. Marine aerosol, sea-salt, and the marinesulphur cycle: a short review. Atmos. Environ. 31:73–80

3. Keene WC, Sander R, Pszenny AAP, Vogt R, Crutzen PJ, Galloway JN. 1998. Aerosol pH in the marineboundary layer: a review and model evaluation. J. Aerosol Sci. 29:339–56

4. Kanakidou M, Seinfeld JH, Pandis SN, Barnes I, Dentener FJ, et al. 2005. Organic aerosol and globalclimate modelling: a review. Atmos. Chem. Phys. 5:1053–123

5. Hewick RM, Hunkapiller MW, Hood LE, Dreyer WJ. 1981. A gas–liquid solid phase peptide and proteinsequenator. J. Biol. Chem. 256:7990–97

6. Mehnert CP, Cook RA, Dispenziere NC, Afeworki M. 2002. Supported ionic liquid catalysis—a newconcept for homogeneous hydroformylation catalysis. J. Am. Chem. Soc. 124:12932–33

7. Kulkarni AA, Joshi JB. 2005. Bubble formation and bubble rise velocity in gas–liquid systems: a review.Ind. Eng. Chem. Res. 44:5873–931

8. Steinbruchel C. 1982. Gas–surface scattering distributions according to the hard-spheroid model. Surf.Sci. 115:247–58

9. Rettner CT, Barker JA, Bethune DS. 1991. Angular and velocity distributions characteristic of thetransition between the thermal and structure regimes of gas–surface scattering. Phys. Rev. Lett. 67:2183–86

10. Lahaye RJWE, Kleyn AW, Stolte S, Holloway S. 1995. The scattering of Ar from Ag(111): a moleculardynamics study. Surf. Sci. 338:169–82

11. King ME, Saecker ME, Nathanson GM. 1994. The thermal roughening of liquid surfaces and its effecton gas–liquid collisions. J. Chem. Phys. 101:2539–47


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


12. King ME, Fiehrer KM, Nathanson GM, Minton TK. 1997. Effects of thermal roughening on the angulardistributions of trapping and scattering in gas–liquid collisions. J. Phys. Chem. A 101:6556–61

13. Benjamin I. 1997. Molecular structure and dynamics at liquid–liquid interfaces. Annu. Rev. Phys. Chem.48:407–51

14. Benjamin I. 2015. Reaction dynamics at liquid interfaces. Annu. Rev. Phys. Chem. 66:165–8815. Muenter AH, DeZwaan JL, Nathanson GM. 2006. Collisions of DCl with pure and salty glycerol:

enhancement of interfacial D → H exchange by dissolved NaI. J. Phys. Chem. B 110:4881–9116. Lancaster DK, Johnson AM, Kappes K, Nathanson GM. 2015. Probing gas–liquid interfacial dynamics

by helium evaporation from hydrocarbon liquids and jet fuels. J. Phys. Chem. C 119:14613–2317. Ryder OS, Campbell NR, Shaloski M, Al-Mashat H, Nathanson GM, Bertram TH. 2015. Role of

organics in regulating ClNO2 production at the air–sea interface. J. Phys. Chem. A 119:8519–2618. Lu JW, Day BS, Fiegland LR, Davis ED, Alexander WA, et al. 2012. Interfacial energy exchange and

reaction dynamics in collisions of gases on model organic surfaces. Prog. Surf. Sci. 87:221–5219. Andersson G, Ridings C. 2014. Ion scattering studies of molecular structure at liquid surfaces with

applications in industrial and biological systems. Chem. Rev. 114:8361–8720. Lednovich SL, Fenn JB. 1977. Absolute evaporation rates for some polar and nonpolar liquids. AIChE

J. 23:454–5921. Kenyon AJ, McCaffery AJ, Quintella CM, Zidan MD. 1992. Liquid surface dynamics: a quantum-resolved

scattering study. Chem. Phys. Lett. 190:55–5822. Kelso H, Kohler SPK, Henderson DA, McKendrick KG. 2003. Dynamics of the gas–liquid interfacial

reaction of O(3P) atoms with hydrocarbons. J. Chem. Phys. 119:9985–8823. Ziemkiewicz MP, Roscioli JR, Nesbitt DJ. 2011. State-to-state dynamics at the gas–liquid metal interface:

rotationally and electronically inelastic scattering of NO [2�1/2(0.5)] from molten gallium. J. Chem. Phys.134:234703

24. Zolot AM, Harper WW, Perkins BG, Dagdigian PJ, Nesbitt DJ. 2006. Quantum-state resolved reac-tion dynamics at the gas-liquid interface: direct absorption detection of HF(v, J) product from F(2P) +squalane. J. Chem. Phys. 125:021101

25. Perkins BG, Haber T, Nesbitt DJ. 2005. Quantum state–resolved energy transfer dynamics at gas–liquidinterfaces: IR laser studies of CO2 scattering from perfluorinated liquids. J. Phys. Chem. B 109:16396–405

26. Cohen SR, Naaman R, Sagiv J. 1988. Rotational and state-resolved translational distributions of NOscattered from organized amphiphilic monolayers. J. Chem. Phys. 88:2757–63

27. Rettner CT, Ashfold MNR, eds. 1991. Dynamics of Gas–Surface Interactions. London: R. Soc. Chem.28. Nathanson GM. 2004. Molecular beam studies of gas–liquid interfaces. Annu. Rev. Phys. Chem. 55:231–5529. Alexander WA, Zhang J, Murray VJ, Nathanson GM, Minton TK. 2012. Kinematics and dynamics of

atomic-beam scattering on liquid and self-assembled monolayer surfaces. Faraday Discuss. 157:355–7430. Yan T, Hase WL, Tully JC. 2004. A washboard with moment of inertia model of gas–surface scattering.

J. Chem. Phys. 120:1031–4331. Kohler SPK, Reed SK, Westacott RE, McKendrick KG. 2006. Molecular dynamics study to identify the

reactive sites of a liquid squalane surface. J. Phys. Chem. B 110:11717–2432. Tesa-Serrate MA, Marshall BC, Smoll EJ, Purcell SM, Costen ML, et al. 2015. Ionic liquid–vacuum

interfaces probed by reactive atom scattering: influence of alkyl chain length and anion volume. J. Phys.Chem. C 119:5491–505

33. Lipkin N, Gerber RB, Moiseyev N, Nathanson GM. 1994. Atom scattering studies of liquid structureand dynamics: collisions of Xe with a model of squalane. J. Chem. Phys. 100:8408–17

34. Benjamin I, Wilson MA, Pohorille A, Nathanson GM. 1995. Scattering of water from the glycerolliquid–vacuum interface. Chem. Phys. Lett. 243:222–28

35. Szabo TJ, Siavosh-Haghighi A, Adams JE. 2006. Energy transfer at a gas–liquid interface: kinematics ina prototypical system. J. Phys. Chem. B 110:1319–25

36. Peng Y, Liu L, Cao Z, Li S, Mazyar OA, et al. 2008. Chemical dynamics simulation of Ne atom scatteringoff a squalane surface. J. Phys. Chem. C 112:20340–46

37. Bosio SBM, Hase WL. 1997. Energy transfer in rare gas collisions with self-assembled monolayers.J. Chem. Phys. 107:9677–86


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


38. Day BS, Morris JR, Troya D. 2005. Classical trajectory study of collisions of Ar with alkanethiolateself-assembled monolayers: potential-energy surface effects on dynamics. J. Chem. Phys. 122:214712

39. Alexander WA, Troya D. 2006. Theoretical study of the Ar–, Kr–, and Xe–CH4, –CF4 intermolecularpotential-energy surfaces. J. Phys. Chem. A 110:10834–43

40. Martınez-Nunez E, Rahaman A, Hase WL. 2007. Chemical dynamics simulations of CO2 scattering offa fluorinated self-assembled monolayer surface. J. Phys. Chem. C 111:354–64

41. Sun L, Hase WL. 2003. Born–Oppenheimer direct dynamics classical trajectory simulations. Rev. Comput.Chem. 19:79–146

42. Wiens JP, Nathanson GM, Alexander WA, Minton TK, Lakshmi S, Schatz GC. 2014. Collisions ofsodium atoms with liquid glycerol: insights into solvation and ionization. J. Am. Chem. Soc. 136:3065–74

43. Troya D, Pascual RZ, Garton DJ, Minton TK, Schatz GC. 2003. Theoretical studies of the O(3P) +ethane reaction. J. Phys. Chem. A 107:7161–69

44. Troya D, Pascual RZ, Schatz GC. 2003. Theoretical studies of the O(3P) + methane reaction. J. Phys.Chem. A 107:10497–506

45. Troya D, Schatz GC. 2004. Hyperthermal chemistry in the gas phase and on surfaces: theoretical studies.Int. Rev. Phys. Chem. 23:341–73

46. Troya D, Schatz GC. 2004. Theoretical studies of hyperthermal O(3P) collisions with hydrocarbonself-assembled monolayers. J. Chem. Phys. 120:7696–707

47. Kim D, Schatz GC. 2007. Theoretical investigation of hyperthermal reactions at the gas–liquid interface:O(3P) and squalane. J. Phys. Chem. A 111:5019–31

48. Radak BK, Yockel S, Kim D, Schatz GC. 2009. Modeling reactive scattering of F(2P) at a liquid squalaneinterface: a hybrid QM/MM molecular dynamics study. J. Phys. Chem. A 113:7218–26

49. Yockel S, Schatz GC. 2010. Modeling O(3P) and Ar scattering from the ionic liquid [emim][NO3] at5 eV with hybrid QM/MM molecular dynamics. J. Phys. Chem. B 114:14241–48

50. Li X, Schatz GC, Nesbitt DJ. 2012. Anion effects in the scattering of CO2 from the room-temperatureionic liquids [bmim][BF4] and [bmim][Tf2N]: insights from quantum mechanics/molecular mechanicstrajectories. J. Phys. Chem. B 116:3587–602

51. Yockel S, Schatz G. 2012. Dynamic QM/MM: a hybrid approach to simulating gas-liquid interactions. InMultiscale Molecular Methods in Applied Chemistry, ed. B Kirchner, J Vrabec, pp. 43–67. Berlin: Springer

52. Saecker ME, Govoni ST, Kowalski DV, King ME, Nathanson GM. 1991. Molecular beam scatteringfrom liquid surfaces. Science 252:1421–24

53. Saecker ME, Nathanson GM. 1993. Collisions of protic and aprotic gases with hydrogen-bonding andhydrocarbon liquids. J. Chem. Phys. 99:7056–75

54. King ME, Nathanson GM, Hanninglee MA, Minton TK. 1993. Probing the microscopic corrugationof liquid surfaces with gas–liquid collisions. Phys. Rev. Lett. 70:1026–29

55. Saecker ME, Nathanson GM. 1994. Collisions of protic and aprotic gases with a perfluorinated liquid.J. Chem. Phys. 100:3999–4005

56. Nathanson GM, Davidovits P, Worsnop DR, Kolb CE. 1996. Dynamics and kinetics at the gas–liquidinterface. J. Phys. Chem. 100:13007–20

57. Rettner CT, Auerbach DJ, Tully JC, Kleyn AW. 1996. Chemical dynamics at the gas–surface interface.J. Phys. Chem. 100:13021–33

58. Kenyon AJ, McCaffery AJ, Quintella CM, Zidan MD. 1993. Investigation of dynamical processes atliquid surfaces by molecular scattering. J. Chem. Soc. Faraday Trans. 89:3877–84

59. Perkins BG Jr., Nesbitt DJ. 2006. Quantum-state-resolved CO2 scattering dynamics at the gas–liquidinterface: incident collision energy and liquid dependence. J. Phys. Chem. B 110:17126–37

60. Bagot PAJ, Waring C, Costen ML, McKendrick KG. 2008. Dynamics of inelastic scattering of OHradicals from reactive and inert liquid surfaces. J. Phys. Chem. C 112:10868–77

61. Allan M, Bagot PAJ, Costen ML, McKendrick KG. 2007. Temperature dependence of OH yield, trans-lational energy, and vibrational branching in the reaction of O(3P) (g) with liquid squalane. J. Phys. Chem.C 111:14833–42

62. Cohen SR, Naaman R, Balint-Kurti GG. 1988. Energy distribution between spin-orbit states in NOscattered from organized amphiphilic monolayers. Chem. Phys. Lett. 152:269–73


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


63. Quintella CM, McCaffery AJ, Zidan MD. 1993. Direct observation of the kinetics and dynamics ofmolecular desorption from liquid surfaces. Chem. Phys. Lett. 214:563–68

64. Perkins BG Jr., Nesbitt DJ. 2007. Quantum-state-resolved CO2 scattering dynamics at the gas–liquidinterface: dependence on incident angle. J. Phys. Chem. A 111:7420–30

65. Perkins BG Jr., Nesbitt DJ. 2008. Correlated angular and quantum state–resolved CO2 scattering dy-namics at the gas–liquid interface. J. Phys. Chem. A 112:9324–35

66. Perkins BG Jr., Nesbitt DJ. 2008. Quantum state–resolved CO2 collisions at the gas–liquid interface:surface temperature–dependent scattering dynamics. J. Phys. Chem. B 112:507–19

67. Perkins BG Jr., Nesbitt DJ. 2008. Stereodynamics in state-resolved scattering at the gas–liquid interface.PNAS 105:12684–89

68. Perkins BG Jr., Nesbitt DJ. 2009. Toward three-dimensional quantum state–resolved collision dynamicsat the gas–liquid interface: theoretical investigation of incident angle. J. Phys. Chem. A 113:4613–25

69. Perkins BG Jr., Nesbitt DJ. 2010. Stereodynamics at the gas–liquid interface: orientation and alignmentof CO2 scattered from perfluorinated liquid surfaces. J. Phys. Chem. A 114:1398–410

70. Nogueira JJ, Vazquez SA, Mazyar OA, Hase WL, Perkins BG Jr., et al. 2009. Dynamics of CO2 scatteringoff a perfluorinated self-assembled monolayer. Influence of the incident collision energy, mass effects,and use of different surface models. J. Phys. Chem. A 113:3850–65

71. Nogueira JJ, Vazquez SA, Lourderaj U, Hase WL, Martınez-Nunez E. 2010. Chemical dynamics simu-lations of CO2 in the ground and first excited bend states colliding with a perfluorinated self-assembledmonolayer. J. Phys. Chem. C 114:18455–64

72. Yan T, Hase WL, Barker JR. 2000. Identifying trapping desorption in gas–surface scattering. Chem.Phys. Lett. 329:84–91

73. Yan T, Isa N, Gibson K, Sibener S, Hase WL. 2003. Role of surface intramolecular dynamics in theefficiency of energy transfer in Ne atom collisions with a n-hexylthiolate self-assembled monolayer.J. Phys. Chem. A 107:10600–7

74. Isa N, Gibson KD, Yan T, Hase W, Sibener SJ. 2004. Experimental and simulation study of neoncollision dynamics with a 1-decanethiol monolayer. J. Chem. Phys. 120:2417–33

75. Waring C, King KL, Bagot PAJ, Costen ML, McKendrick KG. 2011. Collision dynamics and reactiveuptake of OH radicals at liquid surfaces of atmospheric interest. Phys. Chem. Chem. Phys. 13:8457–69

76. King KL, Paterson G, Rossi GE, Iljina M, Westacott RE, et al. 2013. Inelastic scattering of OH radicalsfrom organic liquids: isolating the thermal desorption channel. Phys. Chem. Chem. Phys. 15:12852–63

77. Troya D. 2012. Dynamics of collisions of hydroxyl radicals with fluorinated self-assembled monolayers.Theor. Chem. Acc. 131:1072

78. Garton DJ, Minton TK, Alagia M, Balucani N, Casavecchia P, Volpi GG. 1997. Reactive scattering ofground-state and electronically excited oxygen atoms on a liquid hydrocarbon surface. Faraday Discuss.108:387–99

79. Waring C, King KL, Costen ML, McKendrick KG. 2011. Dynamics of the gas–liquid interfacial reactionof O(1D) with a liquid hydrocarbon. J. Phys. Chem. A 115:7210–19

80. Garton DJ, Minton TK, Alagia M, Balucani N, Casavecchia P, Volpi GG. 2000. Comparative dynamicsof Cl(2P) and O(3P) interactions with a hydrocarbon surface. J. Chem. Phys. 112:5975–84

81. Zhang JM, Garton DJ, Minton TK. 2002. Reactive and inelastic scattering dynamics of hyperthermaloxygen atoms on a saturated hydrocarbon surface. J. Chem. Phys. 117:6239–51

82. Zhang JM, Upadhyaya HP, Brunsvold AL, Minton TK. 2006. Hyperthermal reactions of O and O2 witha hydrocarbon surface: direct C–C bond breakage by O and H-atom abstraction by O2. J. Phys. Chem.B 110:12500–11

83. Ausfelder F, McKendrick KG. 2000. The dynamics of reactions of O(3P) atoms with saturated hydro-carbons and related compounds. Prog. React. Kinet. Mech. 25:299–370

84. Kohler SPK, Allan M, Kelso H, Henderson DA, McKendrick KG. 2005. The effects of surface temper-ature on the gas–liquid interfacial reaction dynamics of O(3P) plus squalane. J. Chem. Phys. 122:024712

85. Kohler SPK, Allan M, Costen ML, McKendrick KG. 2006. Direct gas–liquid interfacial dynamics: thereaction between O(3P) and a liquid hydrocarbon. J. Phys. Chem. B 110:2771–76

86. Allan M, Bagot PAJ, Koehler SPK, Reed SK, Westacott RE, et al. 2007. Dynamics of interfacial reactionsbetween O(3P) atoms and long-chain liquid hydrocarbons. Phys. Scr. 76:C42–47


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


87. Andresen P, Luntz AC. 1980. The chemical dynamics of the reactions of O(3P) with saturated hydro-carbons. I. Experiment. J. Chem. Phys. 72:5842–50

88. Smith JD, Kroll JH, Cappa CD, Che DL, Liu CL, et al. 2009. The heterogeneous reaction of hydroxylradicals with sub-micron squalane particles: a model system for understanding the oxidative aging ofambient aerosols. Atmos. Chem. Phys. 9:3209–22

89. Bertram AK, Ivanov AV, Hunter M, Molina LT, Molina MJ. 2001. The reaction probability of OH onorganic surfaces of tropospheric interest. J. Phys. Chem. A 105:9415–21

90. Rudich Y. 2003. Laboratory perspectives on the chemical transformations of organic matter in atmo-spheric particles. Chem. Rev. 103:5097–124

91. Molina MJ, Ivanov AV, Trakhtenberg S, Molina LT. 2004. Atmospheric evolution of organic aerosol.Geophys. Res. Lett. 31:L22104

92. Renbaum LH, Smith GD. 2011. Artifacts in measuring aerosol uptake kinetics: the roles of time, con-centration and adsorption. Atmos. Chem. Phys. 11:6881–93

93. Wilson EW, Hamilton WA, Kennington HR, Evans B, Scott NW, DeMore WB. 2006. Measurementand estimation of rate constants for the reactions of hydroxyl radical with several alkanes and cycloalkanes.J. Phys. Chem. A 110:3593–604

94. Atkinson R, Arey J. 2003. Atmospheric degradation of volatile organic compounds. Chem. Rev. 103:4605–38

95. Zolot AM, Dagdigian PJ, Nesbitt DJ. 2008. Quantum-state resolved reactive scattering at the gas–liquidinterface: F + squalane (C30H62) dynamics via high-resolution infrared absorption of nascent HF(v,J).J. Chem. Phys. 129:194705

96. Layfield JP, Troya D. 2010. Theoretical study of the dynamics of F + alkanethiol self-assembled mono-layer hydrogen-abstraction reactions. J. Chem. Phys. 132:134307

97. Tesa-Serrate MA, King KL, Paterson G, Costen ML, McKendrick KG. 2014. Site and bond-specificdynamics of reactions at the gas–liquid interface. Phys. Chem. Chem. Phys. 16:173–83

98. Kanofsky JR, Lucas D, Gutman D. 1973. Direct identification of free-radical products of O-atom reac-tions with olefins, using high-intensity molecular beams. Symp. Int. Combust. 14:285–94

99. Blumenberg B, Hoyermann K, Sievert R. 1977. Primary products in the reactions of oxygen atoms withsimple and substituted hydrocarbons. Symp. Int. Combust. 16:841–52

100. Adusei GY, Fontijn A. 1994. Comparison of the kinetics of O(3P) reactions with the four butenes overwide temperature ranges. J. Phys. Chem. 98:3732–39

101. Quandt R, Min Z, Wang X, Bersohn R. 1998. Reactions of O(3P) with alkenes: H, CH2CHO, CO, andOH channels. J. Phys. Chem. A 102:60–64

102. Min ZY, Wong TH, Su HM, Bersohn R. 2000. Reaction of O(3P) with alkenes: side chain versus doublebond attack. J. Phys. Chem. A 104:9941–43

103. Ringeisen BR, Muenter AH, Nathanson GM. 2002. Collisions of DCl with liquid glycerol: evidence forrapid, near-interfacial D → H exchange and desorption. J. Phys. Chem. B 106:4999–5010

104. Ringeisen BR, Muenter AH, Nathanson GM. 2002. Collisions of HCl, DCl, and HBr with liquid glycerol:gas uptake, D → H exchange, and solution thermodynamics. J. Phys. Chem. B 106:4988–98

105. Muenter AH, DeZwaan JL, Nathanson GM. 2007. Interfacial interactions of DO with salty glycerolsolutions of KI, NaI, LiI, and NaBr. J. Phys. Chem. C 111:15043–52

106. DeZwaan JL, Brastad SM, Nathanson GM. 2008. The roles of salt concentration and cation charge incollisions of Ar and DCl with salty glycerol solutions of NaI and CaI2. J. Phys. Chem. C 112:3008–17

107. Brastad SM, Albert DR, Huang M, Nathanson GM. 2009. Collisions of DCl with a solution coveredwith hydrophobic and hydrophilic ions: tetrahexylammonium bromide in glycerol. J. Phys. Chem. A113:7422–30

108. Faust JA, Dempsey LP, Nathanson GM. 2013. Surfactant-promoted reactions of Cl2 and Br2 with Br−

in glycerol. J. Phys. Chem. B 117:12602–12109. DeZwaan JL, Brastad SM, Nathanson GM. 2008. Evidence for interfacial [FDCl]− in collisions between

DCl and F− in KF–glycerol solutions. J. Phys. Chem. C 112:15449–57110. Brastad SM, Nathanson GM. 2011. Molecular beam studies of HCl dissolution and dissociation in cold

salty water. Phys. Chem. Chem. Phys. 13:8284–95


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


111. Alexander WA, Wiens JP, Minton TK, Nathanson GM. 2012. Reactions of solvated electrons initiatedby sodium atom ionization at the vacuum–liquid interface. Science 335:1072–75

112. Waclawik ER, Goh MC, Donaldson DJ. 1999. Inelastic scattering of atoms and molecules from liquidcrystal surfaces. J. Chem. Phys. 110:8098–103

113. Allan M, Bagot PAJ, Westacott RE, Costen ML, McKendrick KG. 2008. Influence of molecular andsupramolecular structure on the gas–liquid interfacial reactivity of hydrocarbon liquids with O(3P) atoms.J. Phys. Chem. C 112:1524–32

114. Ocko BM, Wu XZ, Sirota EB, Sinha SK, Gang O, Deutsch M. 1997. Surface freezing in chain molecules:normal alkanes. Phys. Rev. E 55:3164–82

115. Gisler AW, Nesbitt DJ. 2012. On probing ions at the gas–liquid interface by quantum state–resolvedmolecular beam scattering: the curious incident of the cation in the night time. Faraday Discuss. 157:297–305

116. Lawrence JR, Glass SV, Park S-C, Nathanson GM. 2005. Surfactant control of gas uptake: effect ofbutanol films on HCl and HBr entry into supercooled sulfuric acid. J. Phys. Chem. A 109:7458–65

117. Glass SV, Park S-C, Nathanson GM. 2006. Evaporation of water and uptake of HCl and HBr throughhexanol films at the surface of supercooled sulfuric acid. J. Phys. Chem. A 110:7593–601

118. Burden DK, Johnson AM, Krier JM, Nathanson GM. 2014. The entry of HCl through soluble surfactantson sulfuric acid: effects of chain branching. J. Phys. Chem. B 118:7993–8001

119. Park S-C, Burden DK, Nathanson GM. 2009. Surfactant control of gas transport and reactions at thesurface of sulfuric acid. Acc. Chem. Res. 42:379–87

120. Plechkova NV, Seddon KR. 2008. Applications of ionic liquids in the chemical industry. Chem. Soc. Rev.37:123–50

121. Lovelock KRJ. 2012. Influence of the ionic liquid/gas surface on ionic liquid chemistry. Phys. Chem.Chem. Phys. 14:5071–89

122. Grenoble Z, Baldelli S. 2014. Ionic liquids at the gas–liquid and solid–liquid interface—characterizationand properties. In Supported Ionic Liquids, ed. R Fehrmann, A Riisager, M Haumann, pp. 145–76.Weinheim, Ger.: Wiley-VCH Verlag

123. Iwahashi T, Nishi T, Yamane H, Miyamae T, Kanai K, et al. 2009. Surface structural study on ionicliquids using metastable atom electron spectroscopy. J. Phys. Chem. C 113:19237–43

124. Lovelock KRJ, Kolbeck C, Cremer T, Paape N, Schulz PS, et al. 2009. Influence of different substituentson the surface composition of ionic liquids studied using ARXPS. J. Phys. Chem. B 113:2854–64

125. Santos CS, Baldelli S. 2009. Alkyl chain interaction at the surface of room temperature ionic liq-uids: systematic variation of alkyl chain length (R = C1–C4, C8) in both cation and anion of [RMIM][R–OSO3] by sum frequency generation and surface tension. J. Phys. Chem. B 113:923–33

126. Hammer T, Reichelt M, Morgner H. 2010. Influence of the aliphatic chain length of imidazolium basedionic liquids on the surface structure. Phys. Chem. Chem. Phys. 12:11070–80

127. Maier F, Cremer T, Kolbeck C, Lovelock KRJ, Paape N, et al. 2010. Insights into the surface compositionand enrichment effects of ionic liquids and ionic liquid mixtures. Phys. Chem. Chem. Phys. 12:1905–15

128. Ziemkiewich MP, Zutz A, Nesbitt DJ. 2012. Inelastic scattering of radicals at the gas–ionic liquid inter-face: probing surface dynamics of BMIM–Cl, BMIM–BF4, and BMIM–Tf2N by rovibronic scatteringof NO [2�1/2(0.5)]. J. Phys. Chem. C 116:14284–94

129. Roscioli JR, Nesbitt DJ. 2010. State-resolved scattering at room-temperature ionic liquid–vacuum inter-faces: anion dependence and the role of dynamic versus equilibrium effects. J. Phys. Chem. Lett. 1:674–78

130. Wu B, Zhang J, Minton TK, McKendrick KG, Slattery JM, et al. 2010. Scattering dynamics of hyper-thermal oxygen atoms on ionic liquid surfaces: [emim][NTf2] and [C12mim][NTf2]. J. Phys. Chem. C114:4015–27

131. Roscioli JR, Nesbitt DJ. 2011. Quantum state resolved scattering from room-temperature ionic liquids:the role of cation versus anion structure at the interface. J. Phys. Chem. A 115:9764–73

132. Waring C, Bagot PAJ, Slattery JM, Costen ML, McKendrick KG. 2010. O(3P) atoms as a probe ofsurface ordering in 1-alkyl-3-methylimidazolium-based ionic liquids. J. Phys. Chem. Lett. 1:429–33

133. Waring C, Bagot PAJ, Slattery JM, Costen ML, McKendrick KG. 2010. O(3P) atoms as a chemicalprobe of surface ordering in ionic liquids. J. Phys. Chem. A 114:4896–904


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


134. Waring C, Bagot PAJ, Costen ML, McKendrick KG. 2011. Reactive scattering as a chemically specificanalytical probe of liquid surfaces. J. Phys. Chem. Lett. 2:12–18

135. Waring C, Bagot PAJ, Bebbington MWP, Raisanen MT, Buck M, et al. 2010. How penetrable arethioalkyl self-assembled monolayers? J. Phys. Chem. Lett. 1:1917–21

136. Ridings C, Lockett V, Andersson G. 2011. Effect of the aliphatic chain length on electrical double layerformation at the liquid/vacuum interface in the [Cnmim][BF4] ionic liquid series. Phys. Chem. Chem.Phys. 13:17177–84

137. Davidovits P, Kolb CE, Williams LR, Jayne JT, Worsnop DR. 2006. Mass accommodation and chemicalreactions at gas–liquid interfaces. Chem. Rev. 106:1323–54

138. Winter B, Faubel M. 2006. Photoemission from liquid aqueous solutions. Chem. Rev. 106:1176–211139. Yang L, Yu X-Y, Zhu Z, Thevuthasan T, Cowin JP. 2011. Making a hybrid microfluidic platform

compatible for in situ imaging by vacuum-based techniques. J. Vac. Sci. Technol. A 29:061101140. Lancaster DK, Johnson AM, Burden DK, Wiens JP, Nathanson GM. 2013. Inert gas scattering from

liquid hydrocarbon microjets. J. Phys. Chem. Lett. 4:3045–49141. Johnson AM, Lancaster DK, Faust JA, Hahn C, Reznickova A, Nathanson GM. 2014. Ballistic evapora-

tion and solvation of helium atoms at the surfaces of protic and hydrocarbon liquids. J. Phys. Chem. Lett.5:3914–18

142. Roscioli JR, Nesbitt DJ. 2011. Quantum state resolved velocity-map imaging spectroscopy: a new toolfor collision dynamics at gas/self-assembled monolayer interfaces. Faraday Discuss. 150:471–79

143. Martinez IS, Baldelli S. 2010. On the arrangement of ions in imidazolium-based room temperature ionicliquids at the gas–liquid interface, using sum frequency generation, surface potential, and surface tensionmeasurements. J. Phys. Chem. C 114:11564–75

144. Bowers J, Vergara-Gutierrez MC, Webster JRP. 2004. Surface ordering of amphiphilic ionic liquids.Langmuir 20:309–12

145. Jeon Y, Sung J, Bu W, Vaknin D, Ouchi Y, Kim D. 2008. Interfacial restructuring of ionic liquids deter-mined by sum-frequency generation spectroscopy and X-ray reflectivity. J. Phys. Chem. C 112:19649–54

146. Villar-Garcia IJ, Fearn S, De Gregorio GF, Ismail NL, Gschwend FJV, et al. 2014. The ionic liquid–vacuum outer atomic surface: a low-energy ion scattering study. Chem. Sci. 5:4404–18

147. Law G, Watson PR, Carmichael AJ, Seddon KR. 2001. Molecular composition and orientation at thesurface of room-temperature ionic liquids: effect of molecular structure. Phys. Chem. Chem. Phys. 3:2879–85

148. Nakajima K, Ohno A, Hashimoto H, Suzuki M, Kimura K. 2010. Observation of surface structureof 1-alkyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide using high-resolution Rutherfordbackscattering spectroscopy. J. Chem. Phys. 133:044702


Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.

PC67-FrontMatter ARI 9 May 2016 20:8

Annual Review ofPhysical Chemistry

Volume 67, 2016Contents

The Independence of the Junior Scientist’s Mind: At What Price?Giacinto Scoles � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1

Vacuum Ultraviolet Photoionization of Complex Chemical SystemsOleg Kostko, Biswajit Bandyopadhyay, and Musahid Ahmed � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �19

Real-Time Probing of Electron Dynamics Using AttosecondTime-Resolved SpectroscopyKrupa Ramasesha, Stephen R. Leone, and Daniel M. Neumark � � � � � � � � � � � � � � � � � � � � � � � � � � �41

Charge-Carrier Dynamics in Organic-Inorganic Metal HalidePerovskitesLaura M. Herz � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �65

Vibrational Control of Bimolecular Reactions with Methane by Mode,Bond, and Stereo SelectivityKopin Liu � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �91

Interfacial Charge Transfer States in Condensed Phase SystemsKoen Vandewal � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 113

Recent Advances in Quantum Dynamics of Bimolecular ReactionsDong H. Zhang and Hua Guo � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 135

Enhancing Important Fluctuations: Rare Events and Metadynamicsfrom a Conceptual ViewpointOmar Valsson, Pratyush Tiwary, and Michele Parrinello � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 159

Vibrational Heat Transport in Molecular JunctionsDvira Segal and Bijay Kumar Agarwalla � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 185

Gas-Phase Femtosecond Particle Spectroscopy: A Bottom-UpApproach to Nucleotide DynamicsVasilios G. Stavros and Jan R.R. Verlet � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 211

Geochemical Insight from Nonlinear Optical Studies ofMineral–Water InterfacesPaul A. Covert and Dennis K. Hore � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 233

v

Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


Charge Transfer Dynamics from Photoexcited SemiconductorQuantum DotsHaiming Zhu, Ye Yang, Kaifeng Wu, and Tianquan Lian � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 259

Valence Electronic Structure of Aqueous Solutions: Insights fromPhotoelectron SpectroscopyRobert Seidel, Bernd Winter, and Stephen E. Bradforth � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 283

Molecular Shape and the Hydrophobic EffectMatthew B. Hillyer and Bruce C. Gibb � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 307

Characterizing Localized Surface Plasmons Using ElectronEnergy-Loss SpectroscopyCharles Cherqui, Niket Thakkar, Guoliang Li, Jon P. Camden,

and David J. Masiello � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 331

Computational Amide I 2D IR Spectroscopy as a Probe of ProteinStructure and DynamicsMike Reppert and Andrei Tokmakoff � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 359

Understanding the Surface Hopping View of Electronic Transitionsand DecoherenceJoseph E. Subotnik, Amber Jain, Brian Landry, Andrew Petit, Wenjun Ouyang,

and Nicole Bellonzi � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 387

On the Nature of Bonding in Parallel Spins in Monovalent MetalClustersDavid Danovich and Sason Shaik � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 419

Biophysical Insights from Temperature-Dependent Single-MoleculeForster Resonance Energy TransferErik D. Holmstrom and David J. Nesbitt � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 441

Next-Generation Force Fields from Symmetry-Adapted PerturbationTheoryJesse G. McDaniel and J.R. Schmidt � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 467

Measuring the Hydrodynamic Size of Nanoparticles Using FluctuationCorrelation SpectroscopySergio Dominguez-Medina, Sishan Chen, Jan Blankenburg, Pattanawit Swanglap,

Christy F. Landes, and Stephan Link � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 489

Atomic and Molecular Collisions at Liquid SurfacesMaria A. Tesa-Serrate, Eric J. Smoll Jr., Timothy K. Minton,

and Kenneth G. McKendrick � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 515

Theory of Linear and Nonlinear Surface-Enhanced VibrationalSpectroscopiesDhabih V. Chulhai, Zhongwei Hu, Justin E. Moore, Xing Chen, and Lasse Jensen � � � � 541

vi Contents

Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.


Single-Molecule Studies in Live CellsJi Yu � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 565

Excited-State Properties of Molecular Solids from First PrinciplesLeeor Kronik and Jeffrey B. Neaton � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 587

Water-Mediated Hydrophobic InteractionsDor Ben-Amotz � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 617

Semiclassical Path Integral Dynamics: Photosynthetic Energy Transferwith Realistic Environment InteractionsMi Kyung Lee, Pengfei Huo, and David F. Coker � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 639

Reaction Coordinates and Mechanistic Hypothesis TestsBaron Peters � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 669

Fundamental Properties of One-Dimensional Zinc OxideNanomaterials and Implementations in Various Detection Modes ofEnhanced BiosensingJong-in Hahm � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 691

Liquid Cell Transmission Electron MicroscopyHong-Gang Liao and Haimei Zheng � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 719

Indexes

Cumulative Index of Contributing Authors, Volumes 63–67 � � � � � � � � � � � � � � � � � � � � � � � � � � � 749

Cumulative Index of Article Titles, Volumes 63–67 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 753

Errata

An online log of corrections to Annual Review of Physical Chemistry articles may befound at http://www.annualreviews.org/errata/physchem

Contents vii

Ann

u. R

ev. P

hys.

Che

m. 2

016.

67:5

15-5

40. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

Acc

ess

prov

ided

by

Mon

tana

Sta

te U

nive

rsity

- B

ozem

an o

n 06

/05/

16. F

or p

erso

nal u

se o

nly.

Documents

Atomic and Molecular Collisions at Liquid Surfaces - byui.edu · PDF fileAtomic and Molecular Collisions at Liquid Surfaces ... School of Engineering and Physical Sciences, Heriot-Watt