No abstract

We report a new experimental method to study reactive ion-molecule collisions at very low temperatures. A source of laser-cooled ions in a linear Paul trap has been combined with a quadrupole-guide velocity selector to investigate the reaction of Ca+ with CH3F at collision energies E[over](coll)/k(B)> or =1 K with single-particle sensitivity. The technique represents a general approach to study reactive collisions between ions and polar molecules over a wide temperature range down to the cold regime.

Ensembles of cold atomic and molecular ions in ion traps prepared at millikelvin temperatures by laser and sympathetic cooling have recently found considerable interest in both physics and chemistry. At very low temperatures the ions form ordered structures in the trap also known as "Coulomb crystals". Ion Coulomb crystals exhibit a range of intriguing properties which render them attractive systems for novel experiments in chemical dynamics, ultrahigh-resolution spectroscopy and quantum-information processing. In this article we review the methods used to prepare atomic and molecular ion Coulomb crystals and discuss some recent studies in mass spectrometry, low-temperature chemistry and precision spectroscopy to illustrate their scientific potential for chemical applications. Finally, we conclude with an outlook on outstanding challenges and prospective further developments in the field.

The recent development of a range of techniques for producing cold atoms and molecules at very low translational temperatures T < or = 1 K has provided the opportunity to investigate collisional processes in a new physical regime. We have recently presented a new experimental method to study low-temperature reactive collisions between translationally cold ions and neutral molecules (S. Willitsch et al., Phys. Rev. Lett. 2008, 100, 043203). Our technique relies on the combination of a quadrupole-guide velocity selector for the generation of translationally cold neutral molecules with a facility to produce ordered structures of cold ions (Coulomb crystals) by laser cooling in a linear quadrupole ion trap. The strong localisation of the ions in the trap in combination with the high sensitivity of laser-induced-fluorescence detection enabled us to study chemical reactions on the single-particle level, down to temperatures of T approximately 1 K. In the current paper, we present a detailed characterisation of the scope and limitations of this method based on our study of the reaction between laser-cooled Ca+ ions and velocity-selected CH3F molecules. The properties of our cold-neutrals source and the dependence of the measured rate constant on the shape of the Coulomb crystals, trapping and laser-cooling parameters are discussed. An extension of our technique for the study of low-temperature reactions with sympathetically cooled molecular ions (translational temperature T > 10 mK) is presented and first results on the charge-transfer reaction between OCS+ and ND3 are discussed. Finally, perspectives for further developments of our method are explored.

Rate constants for chemical reactions of laser-cooled Ca(+) ions and neutral polar molecules (CH(3)F, CH(2)F(2), or CH(3)Cl) have been measured at low collision energies (/k(B)=5-243 K). Low kinetic energy ensembles of (40)Ca(+) ions are prepared through Doppler laser cooling to form "Coulomb crystals" in which the ions form a latticelike arrangement in the trapping potential. The trapped ions react with translationally cold beams of polar molecules produced by a quadrupole guide velocity selector or with room-temperature gas admitted into the vacuum chamber. Imaging of the Ca(+) ion fluorescence allows the progress of the reaction to be monitored. Product ions are sympathetically cooled into the crystal structure and are unambiguously identified through resonance-excitation mass spectrometry using just two trapped ions. Variations of the laser-cooling parameters are shown to result in different steady-state populations of the electronic states of (40)Ca(+) involved in the laser-cooling cycle, and these are modeled by solving the optical Bloch equations for the eight-level system. Systematic variation of the steady-state populations over a series of reaction experiments allows the extraction of bimolecular rate constants for reactions of the ground state ((2)S(1/2)) and the combined excited states ((2)D(3/2) and (2)P(1/2)) of (40)Ca(+). These results are analyzed in the context of capture theories and ab initio electronic structure calculations of the reaction profiles. In each case, suppression of the ground state rate constant is explained by the presence of a submerged or real barrier on the ground state potential surface. Rate constants for the excited states are generally found to be in line with capture theories.

A wave-packet propagation study is presented of the ionization dynamics of xenon and hydrogen Rydberg atoms interacting with a metal surface in the presence of an external field. The calculations are performed using a Coulomb-wave discrete variable representation, which allows an efficient extension of previous calculations to a higher principal quantum number. The wave-packet calculations include nonadiabatic effects at avoided energy level crossings. Ionization probabilities as a function of distance from the surface are compared with complex-scaling calculations, which assume purely adiabatic traversal of the avoided crossings. A comparison is made between the dynamics calculated for the "normal" experimental situation, where the applied field is oriented so as to repel positive ions away from the surface, versus the dynamics for the reversed field situation, in which electrons are repelled from the surface. Overall it is clear that reversing the field direction has a pronounced effect on the ionization dynamics for any given starting level and that the nonadiabatic effects are most pronounced in the reversed field case. For certain field ranges, electron flux is found to be "backscattered" away from the surface in the reversed field configuration. Preliminary mean-field calculations are also presented to evaluate the effect of the acceleration of the atom on the ionization dynamics.

The measurement of the rotational state distribution of a velocity-selected, buffer-gas-cooled beam of ND3 is described. In an apparatus recently constructed to study cold ion-molecule collisions, the ND3 beam is extracted from a cryogenically cooled buffer-gas cell using a 2.15 m long electrostatic quadrupole guide with three 90° bends. (2+1) resonance enhanced multiphoton ionization spectra of molecules exiting the guide show that beams of ND3 can be produced with rotational state populations corresponding to approximately T(rot) = 9-18 K, achieved through manipulation of the temperature of the buffer-gas cell (operated at 6 K or 17 K), the identity of the buffer gas (He or Ne), or the relative densities of the buffer gas and ND3. The translational temperature of the guided ND3 is found to be similar in a 6 K helium and 17 K neon buffer-gas cell (peak kinetic energies of 6.92(0.13) K and 5.90(0.01) K, respectively). The characterization of this cold-molecule source provides an opportunity for the first experimental investigations into the rotational dependence of reaction cross sections in low temperature collisions.

The production of a translationally cold (T o 1 K) sample of bromine atoms with estimated densities of up to 10 8 cm À3 using photodissociation is presented. A molecular beam of Br 2 seeded in Kr is photodissociated into Br + Br* fragments, and the velocity distribution of the atomic fragments is determined using (2 + 1) REMPI and velocity map ion imaging. By recording images with varying delay times between the dissociation and probe lasers, we investigate the length of time after dissociation for which atoms remain in the laser focus, and determine the velocity spread of those atoms. By careful selection of the photolysis energy, it is found that a fraction of the atoms can be detected for delay times in excess of 100 ms. These are atoms for which the fragment recoil velocity vector is directly opposed and equal in magnitude to the parent beam velocity leading to a resultant lab frame velocity of approximately zero. The FWHM velocity spreads of detected atoms along the beam axis after 100 ms are less than 5 ms À1 , corresponding to temperatures in the milliKelvin range, opening the possibility that this technique could be utilized as a slow Br atom source.