Metastable decay of rare gas cluster ions: mechanism and kinetics

Metastable decay of cluster ions has been discovered only recently. It was noted that one has to take this metastable decay into account when using mass spectrometry to probe neutral clusters, because ion abundance anomalies in mass spectra of rare gas and molecular clusters are caused by delayed metastable evaporation of monomers following ion production. Moreover, it was found that(i) the individual metastable reaction rates k depend strongly on cluster size and cluster ion production pathways and that(ii) there exists experimental evidence (k=k(t)) and a theoretical prediction that a given mass selected cluster ion generated by electron impact ionization of a nozzle expansion beam will comprise a range of metastable decay rates. In addition, it was discovered that metastable Ar cluster ions which lose two monomers in the μs time regime decay via sequential decay series Ar _n ⁺ *→Ar _n?1 ⁺ *→Ar _n?2 ⁺ * with cluster sizes 7≤n≤10 andn=3 (similar results were obtained recently in case of N₂ cluster ions). Conversely, the dominant metastable decay channel of Ar ₄ ⁺ * into Ar ₂ ⁺ was found to proceed predominantly via a single step fissioning process.     

Photofragmentation of mass resolved carbon cluster ions

The results of a detailed study of the photodissociation of carbon cluster ions, C ₃ ⁺ to C ₂₀ ⁺ , are presented and discussed. The experiments were performed using internally cold cluster ions derived from pulsed laser evaporation of a graphite target rod in a helium buffer gas followed by supersonic expansion. The mass selected clusters were photodissociated using 248 nm and 351 nm light from an excimer laser. Photofragment branching ratios, photodissociation cross sections and data on the laser fluence dependence of photodissociation are reported. For almost all initial clusters, C _n ⁺ , the dominant photodissociation pathway was observed to be loss of a C₃ unit to give a C _n?3 ⁺ ion. This observation is interpreted as indicating that dissociation occurs by a statistical unimolecular process rather than by direct photodissociation. The photodissociation was found to be linear with laser fluence forn>5 with 248 nm and 351 nm light; quadratic forn=5 for 248 nm and 351 nm; and linear forn=4 at 248 nm. Dissociation energies for the carbon cluster ions implied by these results are discussed. The photodissociation cross sections were found to change dramatically with cluster size and with the wavelength of the photodissociating light.     

Multiphoton ionization and photodissociation of (NO)mArn clusters

Picosecond multiphoton ionization of (NO)_mAr_n clusters produced in a supersonic expansion of NO/Ar gas mixtures has been studied using time-of-flight mass spectrometry. Two-photon ionization with 266 nm photons show that dilute gas mixtures (1% NO/Ar) yield clusters limited to m≤7, but with as many as 37 argon atoms. Magic numbers are observed for NO⁺Ar₁₂, NO⁺Ar₁₈, (NO) ₂ ⁺ Ar₁₇, NO⁺Ar₂₂, and (NO) ₂ ⁺ Ar₂₁ and are understood in terms of solvation of the NO⁺ and (NO) ₂ ⁺ by argon in icosahedral arrangements. Four-photon ionization with 532 nm light produces dissociation of the clusters to yield only NO⁺Ar_n with n up to 54. This distribution exhibits an additional magic number at n=54, consistent with the completion of a second solvation sphere about the NO⁺. The known wavelength dependence for photodissociation of (NO) ₂ ⁺ and (NO) ₃ ⁺ and comparison of MPI spectra obtained with 266, 355, and 532 nm light indicate that the dissociation is occurring in the cluster ions.     

Guided ion beam studies of the reactions of O+, O 2 + , N+ and N 2 + with silane

Guided ion beam mass spectrometry is used to measure the cross sections as a function of kinetic energy for reaction of SiH₄ with O⁺(⁴S), O ₂ ⁺ (²Π_g,v=0), N⁺(³P), and N ₂ ⁺ (²Σ _g ⁺ ,v=0). All four ions react with silane by dissociative charge-transfer to form SiH _m ⁺ (m=0?3), and all but N ₂ ⁺ also form SiXH _m ⁺ products where (m=0?3) andX=O, O₂ or N. The overall reactivity of the O⁺, O ₂ ⁺ , and N⁺ systems show little dependence on kinetic energy, but for the case of N ₂ ⁺ , the reaction probability and product distribution relies heavily on the kinetic energy of the system. The present results are compared with those previously reported for reactions of the rare gas ions with silane [13] and are discussed in terms of vertical ionization from the 1t ₂ and 3a ₁ bands of SiH₄. Thermal reaction rates are also provided and dicussed.     

The chemistry and physics of molecular surfaces

Configuration interaction calculations are carried out to study the potential energy surface for the system Ar-Ar ₂ ⁺ . An all-electron as well as a pseudopotential treatment is employed. It is found that in the perpendicular Ar approach the Ar ₂ ⁺ partner remains essentially unchanged and the potential can be characterized by an electrostatic ion-induced dipole interaction. In the collinear mode of Ar approach the Ar ₂ ⁺ bond separation increases considerably, the charge is redistributed and the interaction can be characterized as chemical bonding. The minimum on the surface is found to be the linear symmetric molecule with bond lengths of 2.62 Å. The optimum structure in the perpendicular approach lies 0.13 eV above the minimum and is the T-shaped molecule in which the Ar is 3.65 Å away from the midpoint of the Ar ₂ ⁺ (r=2.46 Å) system; the best equilateral triangle structure has a bond length of 2.99 Å but is found to lie 0.64 eV above the Ar ₃ ⁺ minimum. The dissociation energy into Ar ₂ ⁺ + Ar is calculated to be 0.16 eV in reasonable agreement with experimental values of 0.21 eV. The potential curves for the four lowest states of Ar ₂ ⁺ are also treated.     

Ab initio potential energy surface for Ar 3 +

The potential energy surface (PES) of linear Ar ₃ ⁺ is calculated at the MP4/6-31G* level including all single, double, triple and quadruple excitations. The results show that the PES of the linear Ar ₃ ⁺ has a very flat valley along the asymmetric stretching vibration normal mode, ν₃. A higher level quadratic configuration interaction calculation including single, double and triple substitutions QCISD (T) along this flat valley suggests that an asymmetric geometry energy minimum reported earlier based on MP2 [1] is due to symmetry breaking in UHF. The global minimum of the PES is found to be for the symmetric geometry atR _ab =R _bc =2.66±0.01 Å, which is in good agreement with the MRD-CI calculation [2] and expectations from our earlier photodissociation experiments [3]. The calculational results are compared with other theoretical calculations, and are discussed in the context of the photodissociation and dynamics of dissociation experiments conducted on Ar ₃ ⁺ .     

On the role of dissociative ionization in the formation of argon dimer ions

The formation of Ar ₂ ⁺ ions has been investigated by means of the threshold photoelectron photoion coincidence (TPEPICO) technique. Two pathways for the formation of Ar ₂ ⁺ ions are important. One is a direct path via excitation of Rydberg states of Ar₂ with consecutive autoionization. The other path is dissociative ionization of larger argon clusters, in this case argon trimers. These two pathways lead to Ar ₂ ⁺ ions with different internal energy. The pathways are easily distinguished in the TPEPICO-TOF spectra by the kinetic energy released (KER) in the dissociative ionization. The KER for the reaction Ar ₃ ⁺ → Ar ₂ ⁺ + Ar was measured as a function of the photon energy and compared to the KER expected from statistical theory. The agreement is satisfying and confirms that Ar ₃ ⁺ ions do indeed dissociate at the thermochemical threshold. At higher photon energy the excited²Π(3/2)g state of Ar ₃ ⁺ is also detected from a second component in the KER. By applying a kinetic energy discrimination it is possible to measure cluster ion spectra in the presence of larger clusters but essentially without interference from the latter.     

Photofragment spectroscopy of N 2 + in the near UV

Photofragment spectroscopy of N ₂ ⁺ has been studied in the wavelength range 343–404 nm using an excimer-pumped dye laser with a spectral resolution of 0.2 cm^?1. The observed bands are assigned to transitions from thev″=23?26 levels of theX ²Σ _g ⁺ state to highlying rovibrational levels (v′≈46–48) of theB ²Σ _u ⁺ state, forming quasibound (predissociating) states above the dissociation limit N⁺(³ P)+N(⁴ S ⁰). Measurement of the photofragment kinetic energies allows to establish an absolute energy scale for the transitions with respect to the dissociation limit. Molecular constants for the lower and upper states of the observed transitions are determined. The measurements allow the first direct determination of the N ₂ ⁺ dissociation energyD ₀ ⁰ (N ₂ ⁺ ). Some high-resolution (0.04 cm^?1) measurements show the fine-structure splitting and lifetime broadening of the excitation lines.     

Helium cluster ions RgHe x + (Rg=Ne,Ar and Kr,x≦14) formed in a drift tube cooled by liquid helium

Rare gas ions Ne⁺, Ar⁺ and Kr⁺ are injected into a drift tube which is filled with helium gas and cooled by liquid helium. Helium cluster ions RgHe _x ⁺ (Rg=Ne, Ar and Kr,x≦14) are observed as products. Information regarding the stability of RgHe _x ⁺ is obtained from drift field dependence of the size distribution of the clusters, and magic numbers are determined. The magic numbers arex=11 and 13 for NeHe _x ⁺ andx=12 for ArHe _x ⁺ and KrHe _x ⁺ . NeHe _x ⁺ , Ar⁺ and Kr⁺ are proposed as the core ions for NeHe ₁₃ ⁺ , ArHe ₁₂ ⁺ and KrHe ₁₂ ⁺ , respectively.     

Theoretical study of the visible photodissociation spectrum of Ar 3 +

The photodissociation of Ar ₃ ⁺ is studied following a consistent theoretical approach from the Potential Energy Surfaces to the dynamics. Six P.E.S. are computed according to a D.I.M.-like model Hamiltonian. Transition dipole moments are determined using a similar method. The 4-D dynamics of this system is obtained with the H.W.D. method (Hemiquantal dynamic with the Whole DIM basis). All the 4 nuclear degrees of freedom and all the 6 electronic states are involved in the dynamical calculations, allowing for very general investigations. The main theoretical results are:

1. the spectrum essentially results from a Σ → Σg transition to the second excited electronic state along with a symmetric stretching motion
2. excited Ar ₃ ⁺ molecules almost all dissociate in Ar⁺ + 2 Ar
3. dissociation in Ar ₂ ⁺ + Ar requires special conditions such as low laser excitation and is predicted to increase with a specific excitation of the bending mode
4. the dominant symmetric stretching motion induces a bimodal kinetic energy distribution of the fragments.

All these points are in close agreement with experimental results.     

Reactivity of positively charged cobalt cluster ions with CH4, N2, H2, C2H4, and C2H2

Reactivity of positively charged cobalt cluster ions (Co _n ⁺ ,n=2?22), produce by laser vaporization, with various gas samples (CH₄, N₂, H₂, C₂H₄, and C₂H₂) were systematically investigated by using a fast-flow reactor. The reactivity of Co _n ⁺ with the various gas samples is qualitatively consistent with the adsorption rate of the gas to cobalt metal surfaces. Co _n ⁺ highly reacts with C₂H₂ as characterized by the adsorption rate to metal surfaces, and it indicates no size dependence. In contrast, the reactions of Co _n ⁺ with the other gas samples indicate a similar cluster size dependence; atn=4, 5, and 10?15, Co _n ⁺ highly reacts. The difference can be explained by the amount of the activation energy for chemisorption reaction. Compared with neutral cobalt clusters, the size dependence is almost similar except for Co ₄ ⁺ and Co ₅ ⁺ . The reactivity enhancement of Co ₄ ⁺ and Co ₅ ⁺ indicates that the cobalt cluster ions are presumed to have an active site for chemisorption atn=4 and 5, induced by the influence of positive charge.     

Spectroscopic properties of the Ar 2 * (5p) excimer states

The transitions between Ar ₂ ^* (5p) and Ar ₂ ^* (4sΣ _u) have been investigated by absorption spectrometry. The fine structure of the Ar ₂ ^* (5p ³ Π _g) was attributed to a predominantly Hund’s case a coupling. A spin orbit coupling constant of A = (9.8±0.3) cm^-1 results. Absorption by the singlet system allows one to determine the triplet/singlet splitting between the Ar ₂ ^* (4s Σ _u) states to be (540 ± 100) cm^-1. The transition probabilities of the Ar ₂ ^* (5p) and Ar ₂ ^* (6p) levels were determined by saturation spectrometry yielding values between (0.2–2.5) · 10⁶ s^-1.     

Formation of O 2 − in charge transfer collisions of fast molecular hydrogen ions with O2 molecules

Cross sections for the production of O ₂ ^? in charge transfer collisions of fast molecular hydrogen ions (H ₂ ⁺ , D ₂ ⁺ , H ₃ ⁺ , and D ₃ ⁺ of 10 to 140 keV kinetic energy) with O₂ molecules have been determined by means of a time-of-flight mass spectrometer analysing the slow negative product ions from the collisions. Within the measuring accuracy equivelocity H ₂ ⁺ and D ₂ ⁺ ions have the same cross sections for the generation of O ₂ ^? . The projectile velocity dependence curve of the cross section passes through a broad maximum with a peak value of about 6.5×10^?18 cm² around the Bohr velocity (25 keV/u) before showing an asymptotic decrease still within the limited energy range under investigation that is in inverse proportion to the square of velocity. Throughout the examined energy range H ₃ ⁺ ions yield a cross section which is about 1.4 times larger than that of H ₂ ⁺ ions of the same velocity. The fragment ion O^? has been found to appear with cross sections between 10^?19 and 10^?18 cm² upon collisional excitation in the energy range under investigation, with ever decreasing intensity when the energy of the positive hydrogen ions, the proton included, was increased.     

Electronic processes induced by high energy H+, H 2 + and H 3 + -ions: A scaling relation

A scaling relation is proposed which interrelates measurable quantities in the field of atomic collision physics performed with high velocity H⁺, H ₂ ⁺ and H ₃ ⁺ -ions. The relation may be written as $$Q(H^ + ) - 2*Q(H_2^ + ) + Q(H_3^ + ) = 0,$$ whereQ denotes an excitation or ionization cross section or a total or differential secondary particle yield evaluated at the same projectile velocity. The scaling relation will be tested by comparison with experimental data of yields and spectra from ion-induced secondary electron emission measurements and with cross section data for excitation and ionization of atoms and molecules. In general very good agreement is observed for high projectile velocities (v>2 a.u.).     

Analytical photoion spectroscopy applied to dissociative single and double photoionization of diatomic molecules (H2, N2, O2, CO,and NO)

This work reports the principle, advantage, and limitations of analytical photoion spectroscopy which has been applied to dissociative photoionization processes for diatomic molecules such as H₂, N₂, CO, and NO. Characteristic features observed in the differential photoion spectra are summarized with a focus on (pre)dissociation of(i) multielectron excitation states commonly observed in the inner valence regions,(ii) shape resonances, and(iii) doubly charged parent ions. Possible origins for negative peaks in the differential spectra are discussed. This spectroscopy is applied to the reported photoion branching ratios for D₂ (and H₂ at high energies). The main findings are as follows: (1) The direct dissociation of theX ²Σ _g ⁺ (1sσ _g ) state of D ₂ ⁺ , the two-electron excited state¹Σ _u ⁺ (2pσ _u 2sσ _g ) of D₂, and the²Σ _u ⁺ (2pσ _u ) state of D ₂ ⁺ appear clearly in the differential spectrum, as previously observed for H₂. (2) Decay of H ₂ ⁺ (D ₂ ⁺ ) to H⁺ (D⁺) above 38 eV is due to the direct dissociation of highly excited states of H ₂ ⁺ (D ₂ ⁺ ) such as the²Σ _g ⁺ (2sσ _g ) and high-lying Rydberg states converging on H ₂ ²⁺ (D ₂ ²⁺ ). (3) In the ionization continuum of H ₂ ²⁺ (D ₂ ²⁺ ) peculiar dissociation pathways are observed. The differential photoion spectra for O₂ derived from the reported photoion branching ratios are also presented. The (pre)dissociation of theb ⁴Σ _g ^? ,B ²Σ _g ^? , III²Π _u ,²Σ _u ^? , and^2,4Σ _g ^? states of O ₂ ⁺ appears as the corresponding positive values in the spectra in accord with previous observations. Some other dissociation pathways possibly contributing to the spectra are discussed including dissociative double ionization.     

Magic numbers in sims of Mn similar to those of charged Ar clusters

Manganese cluster ions Mn _k ⁺ (k?60) have been produced by 7 keV Xe ion bombardment and analyzed by a double-focusing mass spectrometer. Discontinuous variations of intensity are found atk=5, 14, 16, 29, 34, 45 and 54. Most of these magic numbers coincide with or differ by only one from those observed in Ar _k ⁺ . The similarity in magic numbers between Mn _k ⁺ and Ar _k ⁺ indicates that the bonding nature in the charged Mn clusters is similar to that in the charged Ar clusters; The polarization force between a positive ion in the center of a cluster and surrounding neutral atoms is dominant binding force.     

Direct measurement of the N 2 + dissociation energy

The accuracy for the direct measurement of the dissociation energy of the N ₂ ⁺ B²Σ _u ⁺-state was significantly improved by using frequency doubled laser light, which enables the authors to excite from lowerv″-levels and additionally to calibrate the fundamental laser wavelength with an iodine cell. The obtained value is:D ₈(N ₂ ⁺ )=70248±6 cm^?1.     

Vibrational population of H 2 + after electroionization of thermal H2

In an ion trap experiment we have determined the vibrational population of the lowest 9 vibrational levels of H ₂ ⁺ . We used photodissociation of the trapped molecules by 248 nm light from an excimer laser and the dependence of the photodissociation cross section from the vibrational state. Our results are in good agreement to calculations, which are based on the Franck-Condon principle, but include a variation of the internuclear distance in the transition matrix element.     

Reaction dynamics of Na n + in collision with molecular oxygen

Reaction dynamics of sodium cluster ions, Na _n ⁺ (n = 2–9), in collision with molecular oxygen, O₂ was investigated by measuring the absolute dissociation cross sections and the branching fractions by using a tandem mass spectrometer equipped with several octapole ion guides. The mass spectrum of the product ions show that the dominant reaction channels are production of oxide ions, Na_kO_i (i =1, 2), and intact ions, Na _p ⁺ (p < n). With increase in the collision energy, the cross section for the production of the oxide ions decreased, while that for the production of the intact ions increased. The collision-energy dependences of the cross section for the oxide formation reveals that electron harpooning from the molecule to Na _n ⁺ preludes the oxideion formation. On the other hand, the collision-energy dependences of the cross sections for the intact ion formation is explained by a hard-sphere-collision model similar to the collisional dissociation of Na _n ⁺ by rare-gas impact.     

Effect of solvation on the dynamics of H + CH3 association

The gas phase association of CH₃ with the HAr₂ cluster to form a vibrationally/rotationally excited CH ₄ ^* molecule is used as a model to study microscopic solvation dynamics. A potential energy surface for the reactive system is constructed from a previously fitted H + CH₃ ab initio potential and 12-6 Lennard-Jones Ar-Ar, Ar-C, and Ar-H potentials. Classical trajectory calculations performed with the chemical dynamics computer program VENUS are used to investigate the CH₃ + HAr₂ → CH ₄ ^* + Ar₂ reaction dynamics. Reaction is dominated by a mechanism in which the CH₃ “strips” the H-atom from HAr₂ during large impact parameter collisions. For a large initial relative translational energy the CH₃ + HAr₂ → CH ₄ ^* + Ar₂ cross section is the same as that for H + CH₃ association, so that HAr₂ acts like a “heavy” H-atom. However, at a low initial relative translational energy, the long-range Ar₂—CH₃ attractive potential apparently makes the CH₃ + HAr₂ association cross section larger than that for H + CH₃. Partitioning of energy to the CH ₄ ^* and Ar₂ products is consistent with a stripping mechanism. The initial and final relative translational energies are nearly identical and the CH ₄ ^* rotational energy is controlled by the initial CH₃ rotational energy. The velocity and orbital tilt scattering angles, θ(v _i ,v _f ) and θ(l _i ,l _f ), respectively, are consistent with the stripping mechanism. On average only a small amount of the product energy is partitioned to Ar₂ vibration/rotation and CH ₄ ^* + Ar₂ relative translation.