Search

The kinetics of electron attachment to pyruvic acid (CH

The kinetics of successive reactions of acetylene (C2H2) initiated on either vanadium or iron atomic cations have been investigated under thermal conditions using the variable-ion source and temperature-adjustable selected-ion flow tube apparatus. Consistent with the literature results, the reaction of Fe+ + C2H2 primarily yields Fe+(m/z = (C2H2)3); however, analysis via quantum chemical calculations and statistical modeling shows that the mechanism does not form benzene upon the third acetylene addition. The kinetics are more consistent with successive addition of three acetylene molecules, yielding Fe+(C2H2)3, followed by an addition of a fourth acetylene molecule, initiating cyclotrimerization, yielding either Fe+(C2H2) + neutral benzene or Fe+(Bz) + acetylene, where Bz is a benzene ligand. In contrast, the reaction of V+ + C2H2 yields products via successive associations V+(m/z = (C2H2)n) either with or without a bimolecular step involving loss of one H2 and V+C2(m/z = (C2H2)m), where n and m extend at least up to 11 under conditions of 0.32 Torr at 300 K. Stabilized V+(Bz) is not a significant intermediate in the association mechanism. We propose a plausible mechanism for the generation of neutral benzene in this reaction and compare with the Fe+ results. The reaction steps that produce benzene result in turnover of the single-atom catalyst, and the large hydrocarbons produced that remain associated to the catalyst are proposed to be polycyclic aromatic hydrocarbons.

We show how the powerful combination of temperature-dependent kinetics coupled with detailed statistical modeling can be used to derive dynamical information about transition state barrier heights, the importance of multiple entrance channel complexes, crossings between spin surfaces, energetics, product states, and other information for metal-ion reactions. The methods are not new, but with improved computers, ion sources, ion transport, and better detection techniques, the ability to derive such parameters from the combination of methods has improved greatly. Temperature-dependent kinetics is very sensitive to the above list of parameters because the energy is varied in a controlled way that can be easily modeled. The present measurements, performed in our variable-ion source temperature-adjustable selected-ion flow tube (VISTA-SIFT), have been enabled by advances in ion transport and injection improvements so that dim sources can be used. Replacing the quadrupole mass spectrometer detector with a time-of-flight mass spectrometer solved additional problems. Quantum chemical calculations have improved greatly and provide details about the surfaces, as well as frequencies, to use as starting points for the statistical modeling. For ion-molecule reactions, incorporation of both energy and angular momentum effects are important and we have developed an in-house computer program, based on the work of Juergen Troe, to rapidly compare statistical modeling predictions to the experimental data. As we show, modeling the kinetics data can often determine the most important parameters controlling the reactivity and deriving them is much simpler and usually more accurate than detailed ab initio calculations or dynamical modeling. Additionally, we show that even without statistical modeling, temperature-dependent rate constants as a function of metal anion cluster size can be used to show that such species react by the same mechanism as surfaces. In this review, we discuss reactions of metallic atomic ions, small metal oxide ions, mixed metal oxide ions, and a series of metallic anionic cluster reactions with small molecules such as CO, O2, CO2, N2O, CH4, and several other species. Particular attention was paid to reactions involving bond activation pertinent to catalysis.

The reactions of anionic metal clusters Mn– with O2 (M = V (n = 1–15), Cr (n = 1–15), Co (n = 1–12), and Ni (n = 1–14)) are investigated from 300 to 600 K using a selected-ion flow tube. All rate c...

The catalytic oxidation of CO by N2O promoted by Co+ was studied as a function of temperature in a variable-ion source temperature-adjustable selected-ion flow tube (VISTA-SIFT). Each step of the c...

The reactions of Al2O2+ + N2O and Al2O3+ + CO, forming a catalytic cycle oxidizing CO by N2O, have been investigated from 300 to 600 K in a variable ion source, temperature adjustable, selected-ion...

New insights into aluminum anion cluster reactivity with O2 were obtained through temperature-dependent kinetics measurements. Overall reactivity is controlled by a barrier at an avoided crossing w...

The kinetics of V+ + N2O and VO+ + N2O are studied using a selected-ion flow tube from 300-600 K at pressures of 0.25-0.70 Torr helium. V+ + N2O yields VO+ (k = 4.9 ± 1.0 (T/300 K)-0.3±0.2 × 10-10 cm3 s-1) in both ground and excited states. The secondary reaction VO+ + N2O → VO2+ + N2 proceeds near the collision rate at >10-10 cm3 s-1, whereas thermalized VO+ + N2O studied as a primary reaction proceeds more than 100× more slowly (k = 4.2 ± 1.0 (T/300 K)-1.4±0.2 × 10-12 cm3 s-1). The results are best explained by contributions of competing pathways in V+ + N2O: a spin crossing to the lower energy 3VO+ in the exit well and a spin-conserved reaction yielding an electronically excited 5VO+. The intersystem crossing occurs in 35 ± 20% and 37 ± 15% of reactive interactions at 300 and 600 K, respectively. Statistical modeling of relevant reaction coordinates supports the lack of a temperature dependence, indicates an intersystem crossing rate constant of 1011 s-1, and yields derived bond and transition state energies.

Mass-selected aluminum anion clusters, Aln–, were reacted with O2. Rate constants (300 K) for 2 < n < 30 and product branching fractions for 2 < n < 17 are reported. Reactivity is strongly anticorrelated to Aln– electron binding energy (EBE). Al13– reacts more slowly than predicted by EBE but notably is not inert, reacting at a measurable 0.05% efficiency (2.5 ± 1.5 × 10–13 cm3 s–1). Al6– is also an outlier, reacting more slowly than expected after accounting for other factors, suggesting that high symmetry increases stability. Implications of observed Al13– reactivity, contributions of both electronic shell-closing and geometric homogeneity to Aln– resistance to O2 etching, and future directions to more fully unravel the reaction mechanisms are discussed.

The kinetics of V nO m+ + CCl4 ( n, m = 2, 5; 3, 6-8; 4, 9-11; 5, 12-13) have been measured under thermal conditions using a selected-ion flow tube equipped with a laser vaporization ion source. All reactions proceed at approximately the capture rate limit, yielding three dominant categories of products: CCl3+ + V nO mCl (i.e., chloride transfer), COCl2 (phosgene) formation, and CO2 formation. Both CO2 and COCl2 are products of CCl4 reaction on a bulk vanadium oxide surface, while chloride (or chlorine) transfer is not observed. The product branching fraction of CCl3+ approaches 100% for small (V2) reactants and generally decreases with increasing cluster size down to

The temperature-dependent kinetics for reactions of V+, Fe+, and Co+ with OCS are measured using a selected ion flow tube apparatus heated to 300–600 K. All three reactions proceed solely by C–S activation at thermal energies, resulting in metal sulfide cation formation. Previously calculated reaction pathways were employed to inform statistical modeling of these reactions for comparison to the data. As surmised previously, all three reactions at thermal energies require spin crossing, with the Fe+ reaction crossing once circumventing a prohibitive transition state, before crossing again to form ground state products. The Fe+ and Co+ reaction efficiencies increase with energy. For the Co+ reaction, and to a lesser extent the Fe+ reaction, the apparent activation energies are less than the reaction endothermicities, possibly indicating increasing diabatic behavior of the spin crossings with energy. The V+ reaction was well modeled assuming an entirely adiabatic spin crossing, such that the resultant avoide...

Reactions of the cations Ar+, O2+, CO2+, and CF3+ with the C2 radicals C2H5, H2C2F3, C2F3, and C2F5 were investigated using the variable electron and neutral density attachment mass spectrometry technique in a flowing afterglow–Langmuir probe apparatus at room temperature. Rate coefficients for observed product channels for these 16 reactions are reported as well as rate coefficients and product branching fractions for the 16 reactions of the same cations with each of the stable neutrals used as radical precursors (the species RI, where R is the radical studied). Reactions with the stable neutrals proceed by charge transfer at or near the collisional rate coefficient where energetically allowed; where charge transfer is endothermic, bond-breaking/bond-making chemistry occurs. While also efficient, reactions with the radicals are more likely to occur at a smaller fraction of the collisional rate coefficient, and bond-breaking/bond-making chemistry occurs even in some cases where charge transfer is exotherm...

The pressure and temperature dependences of the reactions of Fe+ with methyl halides CH3X (X = Cl, Br, I) in He were measured in a selected ion flow tube over the ranges 0.4 to 1.2 Torr and 300-600 K. FeX+ was observed for all three halides and FeCH3+ was observed for the CH3I reaction. FeCH3X+ adducts (for all X) were detected in all reactions. The results were interpreted assuming two-state reactivity with spin-inversions between sextet and quartet potentials. Kinetic modeling allowed for a quantitative representation of the experiments and for extrapolation to conditions outside the experimentally accessible range. The modeling required quantum-chemical calculations of molecular parameters and detailed accounting of angular momentum effects. The results show that the FeX+ products come via an insertion mechanism, while the FeCH3+ can be produced from either insertion or SN2 mechanisms, but the latter we conclude is unlikely at thermal energies. A statistical modeling cannot reproduce the competition between the bimolecular pathways in the CH3I reaction, indicating that some more direct process must be important.

We present rate constants and product branching ratios for the reactions of FeOx+ (x = 0–4) with ozone at 500 K. Fe+ is observed to react with ozone at the collision rate to produce FeO+ + O2. The FeO+ in turn reacts with ozone at the collision rate to yield both Fe+ and FeO2+ product channels. Ions up to FeO4+ display similar reactivity patterns. Three-body clustering reactions with O2 prevent us from measuring accurate rate constants at 300 K although the data do suggest that the efficiency is also high. Therefore, it is probable that little to no temperature dependence exists over this range. Implications of our measurements to the regulation of atmospheric iron and ozone are discussed. Density functional calculations on the reaction of Fe+ with ozone show no substantial kinetic barriers to make the FeO+ + O2 product channel, which is consistent with the reaction’s efficiency. While a pathway to make FeO2+ + O is also found to be barrierless, our experiments indicate no primary FeO2+ formation for this...

The simplest Criegee intermediate, CH2OO, has been produced in a flowing afterglow using a novel technique. CH2I is produced by dissociative electron attachment to CH2I2, leading to the established reaction CH2I + O2 → CH2OO + I. The presence of CH2OO is established by observation of dissociative electron attachment to yield O(-) using the variable electron and neutral density attachment mass spectrometry (VENDAMS) technique. The measurements establish the electron attachment rate coefficient of thermal electrons at 300 K to CH2OO as 1.2 ± 0.3 × 10(-8) cm(3) s(-1). Thermal electron attachment is solely dissociative and is not a promising route to producing stable CH2OO(-). The results open the possibility of measuring ion-molecule chemistry involving Criegee intermediates, as well as the reactivity of other unstable radicals produced in an analogous manner.

Direct dynamics simulations and selected ion flow tube (SIFT) experiments were performed to study the kinetics and dynamics of the OH(-) + CH3I reaction versus temperature. This work complements previous direct dynamics simulation and molecular beam ion imaging experiments of this reaction versus reaction collision energy (Xie et al. J. Phys. Chem. A 2013, 117, 7162). The simulations and experiments are in quite good agreement. Both identify the SN2, OH(-) + CH3I → CH3OH + I(-), and proton transfer, OH(-) + CH3I → CH2I(-) + H2O, reactions as having nearly equal importance. In the experiments, the SN2 pathway constitutes 0.64 ± 0.05, 0.56 ± 0.05, 0.51 ± 0.05, and 0.46 ± 0.05 of the total reaction at 210, 300, 400, and 500 K, respectively. For the simulations this fraction is 0.56 ± 0.06, 0.55 ± 0.04, and 0.50 ± 0.05 at 300, 400, and 500 K, respectively. The experimental total reaction rate constant is (2.3 ± 0.6) × 10(-9), (1.7 ± 0.4) × 10(-9), (1.9 ± 0.5) × 10(-9), and (1.8 ± 0.5) × 10(-9) cm(3) s(-1) at 210, 300, 400, and 500 K, respectively, which is approximately 25% smaller than the collision capture value. The simulation values for this rate constant are (1.7 ± 0.2) × 10(-9), (1.8 ± 0.1) × 10(-9), and (1.6 ± 0.1) × 10(-9) cm(3)s(-1) at 300, 400, and 500 K. From the simulations, direct rebound and stripping mechanisms as well as multiple indirect mechanisms are identified as the atomic-level reaction mechanisms for both the SN2 and proton-transfer pathways. For the SN2 reaction the direct and indirect mechanisms have nearly equal probabilities; the direct mechanisms are slightly more probable, and direct rebound is more important than direct stripping. For the proton-transfer pathway the indirect mechanisms are more important than the direct mechanisms, and stripping is significantly more important than rebound for the latter. Calculations were performed with the OH(-) quantum number J equal to 0, 3, and 6 to investigate the effect of OH(-) rotational excitation on the OH(-) + CH3I reaction dynamics. The overall reaction probability and the probabilities for the SN2 and proton-transfer pathways have little dependence on J. Possible effects on the atomistic mechanisms were investigated for the SN2 pathway and the probability of the direct rebound mechanism increased with J. However, the other atomistic mechanisms were not appreciably affected by J.

We report for the first time temperature dependences (from 300 to 600 K) of the reactions of Fe(+) and FeO(+) with NO and NO(2). Both ions react quickly with NO(2), and their rate constants have weak negative temperature dependences. The former is consistent with the calculated energy profile along the Fe(+) + NO(2) reaction coordinate. Ground state Fe(+) reacts with NO(2) to produce only FeO(+), while FeO(+) reacts with NO(2) to produce NO(+) exclusively. Certain source conditions produce excited Fe(+), as evidenced by production of primary NO(+), which is endothermic with the ground state by 0.35 eV. The room temperature rate constants are in agreement with previous values. For the reactions of Fe(+) and FeO(+) with NO, we find an upper limit of

Recently, it was discovered that specific aluminum clusters (e.g., Al13–) that demonstrate enhanced resistance to reactivity with oxygen may do so not only because of a closed electronic jellium shell as originally supposed but also because of a forbidden spin-flip in the transition state of the reaction. Herein, we discuss an experiment using a multiple-species laminar flow reaction vessel coupled to a singlet oxygen generator. The present results suggest that all clusters react with singlet oxygen. Additionally, we observe Al9–, a cluster previously unidentified as having any notable stability, as being resistant to reaction with triplet oxygen. Furthermore, we discuss a means of estimating rate constants in a multiple-species flow tube where the products and reactants do not allow the use of traditional methods.

The dissociative photoionization of tetramethyltin (Me₄Sn) and hexamethylditin (Me₆Sn₂) has been investigated by threshold photoelectron-photoion coincidence (TPEPICO). Ions are energy-selected, and their 0 K dissociation onsets are measured by monitoring the mass spectra as a function of ion internal energy. Me₄Sn(+) dissociates rapidly by methyl loss, with a 0 K onset of E₀ = 9.382 ± 0.020 eV. The hexamethylditin ion dissociates slowly on the time scale of the experiment (i.e., during the 40 μs flight time to the detector) so that dissociation rate constants are measured as a function of the ion energy. RRKM and the simplified statistical adiabatic channel model (SSACM) are used to extrapolate the measured rate constants for methyl and Me₃Sn(•) loss to their 0 K dissociation onsets, which were found to be 8.986 ± 0.050 and 9.153 ± 0.075 eV, respectively. Updated values for the heats of formation of the neutral Me₄Sn and Me₆Sn₂ are used to derive the following 298.15 K gas-phase standard heats of formation, in kJ·mol⁻¹: Δ(f)H(m)(o)(Me₃Sn(+),g) = 746.3 ± 2.9; Δ(f)H(m)(o)(Me₅Sn₂(+),g) = 705.1 ± 7.5; Δ(f)H(m)(o)(Me₃Sn(•),g) = 116.6 ± 9.7; Δ(f)H(m)(o)(Me₂Sn,g) = 123.0 ± 16.5; Δ(f)H(m)(o)(MeSn(+),g) = 877.8 ± 16.4. These energetic values also lead to the following 298.15 K bond dissociation enthalpies, in kJ·mol⁻¹: BDE(Me₃Sn-Me) = 284.1 ± 9.9; BDE(Me₃Sn-SnMe₃) = 252.6 ± 14.8.

The dissociation dynamics of gas phase formic acid ions (HCOOH(+), DCOOD(+), HCOOD(+), DCOOH(+)) are investigated by threshold photoelectron-photoion coincidence (TPEPICO) spectroscopy and high level ab initio calculations. The slow rate constants for this seemingly simple H loss reaction and the large onset energy shifts due to isotopic substitution point to a substantial exit barrier through which the H or D atoms tunnel. Modeling of the HCOOH(+) experimental data using RRKM theory with tunneling through an Eckart potential are best fitted with a barrier of about 17 kJ mol(-1). High level ab initio calculations support the experimental findings with a computed barrier of 15.9 kJ mol(-1), which is associated with the substantial geometry change between the product HOCO(+) cation and the corresponding HCOOH(+) molecular ion. Because of this exit channel barrier, the formic acid ion dissociation does not provide a route for determination of the HOCO(+) heat of formation. Rather, the most accurate value comes from the calculations employing the high accuracy extrapolated ab initio thermochemistry (HEAT) scheme, which yields a Δ(f)H(o)(0K)[HOCO(+)] = 600.3 ± 1.0 kJ mol(-1) (Δ(f)H(o)(298K)[HOCO(+)] = 597.3 ± 1.0 kJ mol(-1)). The calculated proton affinity of CO(2) is thus 534.7 ± 1.0 kJ mol(-1) at 0 K and 539.3 ± 1.0 kJ mol(-1) at 298.15 K.

The dissociations of energy-selected di-t-butyl peroxide and di-t-butyl diazene ions have been studied by threshold photoelectron-photoion coincidence (TPEPICO) spectroscopy. Di-t-butyl peroxide ions dissociate via two parallel channels: (1) methyl loss at a 0 K onset (E0) of 9.58 +/- 0.04 eV followed by a sequential dissociation of the daughter ion to produce C4H9O+ and acetone; and (2) the dominant dissociation channel, producing t-butyl ion and t-butyl peroxy radical at an E0 of 9.758 +/- 0.020 eV. Di-t-butyl diazene ions dissociate through three parallel channels: (1) a rearrangement to form isobutene ion; (2) C-N bond cleavage with the charge staying on the t-butyl diazyl species (E0 = 8.069 +/- 0.050 eV); and (3) C-N bond cleavage with the charge instead on the t-butyl (E0 = 8.122 +/- 0.050 eV); the coproduct for this latter channel is a weakly, or possibly unbound, N2...t-butyl structure. Both the peroxide and diazene ion dissociations produce metastable daughters, and the dissociation rates are modeled with two rate theories: the Rice-Ramsperger-Kassel-Marcus (RRKM) theory and a simplified version of the statistical adiabatic channel model (SSACM). Due to a large kinetic shift, RRKM incorrectly models the peroxide ion rate curve. Using SSACM, the heat of formation of t-butyl peroxy radical is determined to be DeltaH0Kdegrees = - 81.1 +/- 3.9 kJ mol-1, and, using B3LYP/6-311++G(d,p) thermal energy, DeltaH298Kdegrees = - 109.7 +/- 3.9 kJ mol-1. Due to a competitive shift of the higher energy channel onsets, RRKM also incorrectly models the diazene rate curves. The 298 K heat of formation of the t-butyl diazyl ion, which is bound by 14 kJ mol-1, is determined to be 701.2 +/- 5.9 kJ mol-1.

Vibrational state resolved measurements of methane’s dissociation on Ni(111) show a strong surface temperature dependence near the translational energy threshold for reaction. The reactivity of molecules excited to v = 1 of the ν3 C−H stretching vibration and incident on the surface with a translational energy of 40 kJ mol−1 increased 8-fold as the surface temperature increased from 90 to 475 K. This enhancement is much larger than that reported for earlier studies at higher incident energies. These results support recent calculations that predict an important role for lattice deformation in transition state access. At higher surface temperatures, surface phonon excitation allows substrate atoms to sample lattice geometries with more favorable transition state energetics. We have also measured the coverage-dependent reactivity of these molecules at both surface temperatures and report a simple model that quantitatively predicts the observed coverage-dependent reactivity.

The 0 K onset of C(3)H(6) --C(3)H(5)(+) + H(*) is measured by threshold photoelectron-photoion coincidence (TPEPICO) spectroscopy. From the onset (11.898 +/- 0.025 eV) the heat of formation of the allyl ion (CH(2)CHCH(2)(+)) is determined to be DeltaH degrees (f,0K) = 967.2; DeltaH degrees (f,298K) = 955.4 +/- 2.5 kJ mol(-1). The value is significantly more positive than prior determinations, and resolves a discrepancy between measurements of the allyl radical and allyl ion heats of formation and recent highly precise measurements of the allyl radical adiabatic ionization energy. The new allyl ion heat of formation leads to a new proton affinity for propadiene (allene) of 765.0 +/- 2.6 kJ mol(-1). An attempt is made to determine the CH(3)CCH(2)(+) heat of formation by measuring the 0 K onset of 2-ClC(3)H(5) --C(3)H(5)(+) + Cl(*). However, C(3)H(5)(+) appears at too low an energy to be the higher energy CH(3)CCH(2)(+) structure. Rather, 2-ClC(3)H(5)(+) undergoes a concerted hydrogen transfer and Cl-loss via an intramolecular S(N)2 like mechanism to produce the allyl ion. The 0 K onset of 3-ClC(3)H(5) --C(3)H(5)(+) + Cl(*) (11.108 +/- 0.010 eV) is measured to determine the 3-ClC(3)H(5) heat of formation (DeltaH degrees (f,0K) = 14.9; DeltaH degrees (f,298K) = 1.1 +/- 2.7 kJ mol(-1)). 3-ClC(3)H(5)(+) is suggested to readily isomerize to trans 1-ClC(3)H(5)(+) prior to dissociation.

The 0 K onsets (E(0)) of a series of trichlorosilane derivatives SiCl(3)R --> SiCl(3)(+)+ R(*) (R = Cl, H, CH(3), C(2)H(5), C(2)H(3), CH(2)Cl, SiCl(3)) are measured by threshold photoelectron-photoion coincidence spectroscopy. The well-known heat of formation of SiCl(4) is used as an anchor to determine the heat of formation of SiCl(3)(+), which is, in turn, used as an anchor to determine the heats of formation of the other alkyltrichlorosilanes investigated. A series of isodesmic reactions at the G3 and CBS-QB3 levels are shown to accurately reproduce the experimental heats of formation, and this scheme is used to calculate the heat of formation of Si(2)Cl(6), from which the measured E(0) determines the SiCl(3)(*) heat of formation. The measured values then determine the IE of SiCl(3)(*) along with the Si-R bond dissociation enthalpies of the six neutral species investigated. The experimental heats of formation are also used in a series of isodesmic reaction calculations to determine the heats of formation of SiH(3)R (R = H, CH(3), C(2)H(5), C(2)H(3), CH(2)Cl, SiCl(3)).

Specific rate constants k(E) of the dissociation of the halobenzene ions C6H5X+ --C6H5+ + X* (X* = Cl, Br, and I) were measured over a range of 10(3)-10(7) s-1 by threshold photoelectron-photoion coincidence (TPEPICO) spectroscopy. The experimental data were analyzed by various statistical unimolecular rate theories in order to derive the threshold energies E0. Although rigid activated complex RRKM theory fits the data in the experimentally measured energy range, it significantly underestimates E0 for chloro- and bromobenzene. Phase space theory (PST) does not fit the experimentally measured rates. A parametrized version of the variational transition state theory (VTST) as well as a simplified version of the statistical adiabatic channel model (SSACM) incorporating an energy dependent rigidity factor provide excellent fits to the experimental data and predict the correct dissociation energies. Although both approaches have just two adjustable parameters, one of which is E0, SSACM is effective and particularly simple to apply.

The dissociative photoionization onsets for Cl and Br loss reactions were measured for HCCl3, HCCl2Br, HCClBr2, and HCBr3 by threshold photoelectron photoion coincidence (TPEPICO) in order to establish the heats of formation of the mixed halides as well as the following fragment ions: HCCl2(+), HCClBr(+), HCBr2(+). The first zero Kelvin onsets were measured with a precision of 10 meV. The second onsets, which are in competition with the lower energy onsets, were established with a precision of 60 meV. Because both the chloroform and bromoform have relatively well established heats of formation, these measurements provide a route for establishing the heats of formation of the mixed halomethanes within uncertainties of less than 5 kJ mol(-1).

A narrow bandwidth infrared (IR) laser selectively excites and activates one isotopic variant of a gas phase precursor to permit isotope-selective chemical vapor deposition. We show that selective vibrational excitation of the 12CH4 or 13CH4 isotopologue of methane controls the carbon isotope ratio of adsorbates deposited on a Ni substrate and permits significant isotope enrichment in a single reaction step. A model of the process based on known reaction probabilities and process variables predicts the enrichment we observe. While our current laser system permits a 9-fold enhancement of 12C or 13C deposition, optimization of deposition and detection strategies and use of a more powerful, commercially available laser source suggest isotope enrichment factors of 100-fold or more are readily attainable. Other gas−surface reactions activated by vibrational energy may also be candidates for this approach to isotope-selective deposition.

The product energy disposal in the abstraction reaction H + c-C 3 H 6 → H 2 (v', j') + C 3 H 5 at 1.6 eV collision energy has been characterized by quantum-state-selective, Doppler-resolved REMPI detection of the H 2 (v', j')product. The H 2 (v', j') product is observed with total energy, translational plus rotational plus vibrational, in excess of the total available energy if the C 3 H 5 product is a cyclopropyl radical, but the total H 2 (v', j') energy matches the total available energy if the C 3 H 5 product is allyl. This implies ring-opening of cyclopropyl to allyl concerted with abstraction and the deposition of a large fraction of the ring-opening energy into the H 2 (v', j') product. Such behavior is unprecedented and entirely unexpected. Calculations indicate that both direct H-by-H abstraction and H-addition/H 2 -elimination should occur too quickly to allow the isomerization to occur before product separation. Even if ring-opening did occur, the appearance of the total isomerization energy in the H 2 product would seem dynamically forbidden. The reaction must follow an unanticipated and surprising path.