State-Specific Reactions of Cu+(1S,3D) with SF6 and SF5Cl - The

Mar 25, 2016 - (30) An unintended consequence of these applications has been the increase in the atmospheric concentration of this anthropogenic gas t...
2 downloads 5 Views 2MB Size
Download PDF
Article pubs.acs.org/JPCA

State-Specific Reactions of Cu+(1S,3D) with SF6 and SF5Cl William S. Taylor,* Xavier S. Redmon,† and Benjamin A. Scheuter Department of Chemistry, University of Central Arkansas, Conway, Arkansas 72035, United States ABSTRACT: State-specific reactions of Cu+(1S,3D) were carried out in a selected ion drift cell apparatus with SF6 and SF5Cl. Copper ions were prepared in a glow discharge utilizing Ne as the working gas. Analysis of Cu+ states using ion mobility mass spectrometry (IMS) indicated the presence of both Cu+(3d10) and Cu+(3d94s1) configurations attributable to the 1S ground and 3 D first excited states of this metal ion, respectively. State-specific product formation in reactions of these ions with the two neutral substrates of interest here was determined using IMS along with both known and calculated energetic requirements for product formation. These experiments indicate that Cu+(1S) associates with both SF6 and SF5Cl; however, the process is approximately four times as efficient with the latter neutral under these conditions. Association is also observed as a minor product between Cu+(3D) and both neutral reactants. Inefficient formation of SF3+ occurs as the sole bimolecular product from SF6 via Cu+(3D). In contrast, Cu+(3D) reacts with SF5Cl in rapid parallel bimolecular processes yielding SF3+ and CuCl+. These results also indicate that CuCl+ initiates additional higher-order processes which result in SF5+ and SF4Cl+. The energetics associated with the formation of SF3+ suggest that a copper halide neutral byproduct must also be formed, requiring a more complex mechanism than simple dissociative charge-transfer.

■

INTRODUCTION A number of surveys have illustrated that the ability of first-row transition metal ions to induce rearrangements in small substrate molecules under thermal conditions tends to diminish as d-orbital occupancy increases.1−10 Often, this trend toward inertness can be explained on the basis of unfavorable thermochemistry, but other factors, (e.g.; the availability of an orbital of suitable symmetry, conservation of electron spin) may also play a role.11,12 Thus, as a consequence of its closed-shell [Ar]3d10 configuration, the exothermic chemistry of gas phase Cu+(1S) with many small neutral reactants is often largely restricted to simple association. However, exceptions to this generalization have been reported for reactions in which the neutral reactant contains a sufficiently polar functional group.13−22 In contrast, the 3D first excited state of Cu+ has been shown by us23−26 and others27 to be capable of exothermic bimolecular chemistry with a number of neutral partners at room-temperature. Specifically, we have reported that in reactions with several halogenated methane analogues, Cu+(3D) readily abstracts a labile halogen atom to form CuX+ (X = Cl, Br, I).23−25 A subsequent study has also revealed that Cu+(3D) reacts with SF5CF3 in three competing processes to yield SF3+, SF2+, and CF3+, in which CuF2 is the likely neutral byproduct containing the metal.26 In all cases, reaction outcomes appear to be well-described in terms of the available thermochemistry and overall conservation of spin. In the work reported here, we extend our examination of the state-specific reactions of Cu+ to include SF6 and SF5Cl. SF6 is used widely as a dielectric in high voltage applications,28 as an etchant in reactive plasmas,29 and as a cover gas in magnesium smelting.30 An unintended consequence of these applications © 2016 American Chemical Society

has been the increase in the atmospheric concentration of this anthropogenic gas to its current level of 8 ppt.31 With a global warming potential of ∼24 000 and an atmospheric lifetime of 3200 years, SF6 is a powerful greenhouse gas.32 Although its contribution to the overall atmospheric greenhouse gas concentration is less than 1%, global levels continue to increase as a result of its commercial use.33 Several previous studies have examined the chemistry of SF6 with a number of neutral and ionic species relevant to atmospheric processes.32,34−37 In addition, previous studies of gas phase transition metal ions with SF6 have been carried out which show that the early metals participate in a range of exothermic fluorine- and fluoridetransfer processes, while the late metals (including Cu+(1S)) are largely unreactive with this molecule under these energetic conditions.2,4 Studies describing the behavior and uses of SF5Cl are more limited. Having said this, SF5Cl has been shown to be a useful reagent for adding the − SF5 functional group to a variety of organic substrates resulting in modifications to their physical and chemical properties.38−44 The first reported examinations of the gas phase ion−molecule chemistry of SF5Cl appeared in 2002, where the reactions of this molecule with 22 different cations were described and compared with the behavior of SF6.45 That work revealed that these ions exhibit several processes with SF5Cl, including abstraction of both fluoride and chloride. To the best of our knowledge, the work described herein represents the first examination of the gas phase Received: February 14, 2016 Revised: March 19, 2016 Published: March 25, 2016 2295

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A

configurations which differ by either the presence or absence of a 4s electron. Because of its larger size, the s orbital experiences a less attractive interaction between the ion and the He bath gas, thus reducing the number of capture-collisions. For Cu+, this means that the 3d94s1 configuration has a higher mobility (22 cm2/V·s) in He than the 3d10 configuration (15.8 cm2/V· s). As a consequence, a pulse of ions containing both configurations will be separated within the drift cell such that they appear as different peaks in an arrival time distribution (ATD). Under all discharge conditions employed here, both Cu+ configurations were formed. A recent analysis of the energetics of excited state production within the discharge provides evidence that it is capable of producing excited metal ion states with energies up to approximately 11.5 eV above the atom ground state.51 For Cu+, this energy range includes the 1 S(3d10) ground state, as well as the 3D(3d94s1) and 1D(3d94s1) excited states which lie 2.808 eV (averaged over J-levels) and 3.257 eV above the 1S state. The low-mobility feature in our Cu+ ATDs must be comprised exclusively of the 1S ground state, but because the first and second excited states are indistinguishable on the basis of their mobilities, we cannot rule out the presence of both within the high mobility feature on the basis of IMS alone. We have argued previously that while some amount of Cu+(1D) may be present in the 3d94s1 (high mobility) feature, it is likely that the major contributors to this configuration are the energetically more accessible 3D3,2,1 states.25 Charge-transfer bracketing tests utilizing C3F6 do in fact result in the production of small amounts of C3F6+, consistent with the presence of Cu+(1D) within the 3d94s1 population extracted from the discharge. On the basis of this diagnostic, we estimate that Cu+(1D) represents less than 2% of the 3d94s1 population. Indeed, in the reactions described here, we observed no definitive evidence of product formation that could be attributed exclusively to Cu+(1D). Reagents. Copper cathodes used as sputter targets were fashioned from used oxygen-free copper gaskets into 5.0 mm diameter rods. Research grade neon (the discharge gas) was obtained from Matheson Tri-Gas Inc. Helium used as the buffer gas in the drift cell was obtained from Air Products Inc. with a purity of 99.9999%. SF6 was obtained from Nexair with a purity of 99.9%. SF5Cl was obtained from Synquest Gases Inc. and certified as 94% pure. Significant contaminants in this gas include SF4 at 3.4%, and ClF at 1.7%. Other trace contaminants were not quantified by the supplier. Possible implications of the presence of these contaminants are discussed below. Both reagent gases were obtained with their natural isotopic abundances. Given that Cu, S, and Cl all possess more than one naturally occurring isotope, mass spectra containing species with more than one of these elements exhibit the characteristic patterns resulting from the possible combinations of the individual elemental isotopes. Computational Methods. Where published thermochemical data were unavailable, reaction thermochemistry was calculated with the Gaussian 09 suite of programs using density functional methods.52 All calculated energies were corrected for zero-point energy contributions and include a thermal correction to 298 K. These calculations were carried out using the B3LYP functional in conjunction with the aug-ccpVTZ basis set.53−55 This is a correlation-consistent basis set that includes polarization functions with triple-ζ split-valence and diffuse functions in the augmented version. In those cases where sufficient experimental thermochemistry is available, calculated reaction thermochemistry was compared to the

chemistry of SF5Cl with any metal ion, and provides useful comparisons to the reactions of both SF6 and SF5CF3 with Cu+(1S) and Cu+(3D).

■

EXPERIMENTAL METHODS Instrumental Description. Experiments were carried out using a selected ion drift cell apparatus which has been described previously; therefore only a brief overview will be presented here.46,24 Cu+ ions used in this work were generated with a dc glow discharge ion source which has also been described in detail.47 These Cu+ ions were subsequently directed through a quadrupole deflector (turning lens) and then to a quadrupole mass filter operated in low resolution mode for mass selection such that both the 63Cu+ (69.17% natural abundance) and 65Cu+ (30.83% natural abundance) isotopes were transmitted. The reactant ion beam was then focused onto the entrance aperture of a 4.0 cm drift cell. The drift cell was charged to a total pressure of 3.5 Torr with a mixture of the desired neutral reactant gas in He at a mole fraction of approximately 10 −4 . This concentration is sufficiently greater than the ion number density that pseudo first-order conditions exist with respect to depletion of Cu+. Reactant ions were drawn through the drift cell by means of a small electric field maintained by a set of seven guard rings. Experiments described here were carried out at an E/N of 8.5 Td (1 Td = 1 × 10−17 cm2·V), which equates to a center-ofmass kinetic energy of approximately 0.07 eV for these reacting systems. Residence times for reactant ions were on the order of 100 μs. These reaction conditions are such that little translational heating occurs and only exothermic or thermoneutral reactions are typically observed. Temperature control of the drift cell can be accomplished via a copper shroud through which heated or cooled gases can be circulated. The reactions described in this work were carried out exclusively at room temperature. Temperatures within the drift cell were monitored using a Pt-RTD (resistance temperature device). Ions exiting the drift cell were mass-analyzed with the use of a second quadrupole, and detected using an electron multiplier operated in pulse-counting mode. Metal Ion Source. When operated as in the experiments described here, the glow discharge ion source produces metal ions via a sputter bombardment process in which ions of a working gas (Ne in this case) are accelerated to a cathode made from the desired metal and sputter atoms from its surface. These sputtered atoms diffuse into the discharge plasma and are subsequently ionized by either Penning ionization via metastables of the working gas, or by electron impact ionization via fast electrons being accelerated from the cathode. Metal ions are sampled directly from the discharge plasma. We have previously demonstrated that this ion source is capable of producing metal ions in excited states as well as in their ground states. Further, excited state populations can be controlled to some extent by manipulation of both the working gas pressure and the distance between the cathode and the sampling aperture.48 Both methods were utilized in this work to alter the relative amounts of excited and ground states of Cu+ extracted from the discharge. Determination of Cu+ State Distribution. Specific configurations of Cu+ ions produced in the glow discharge were differentiated within the drift cell using ion mobility spectrometry (IMS), which characterizes them on the basis of their mobilities in He.49,50 Applied to first-row transition metal ions, IMS is effective in distinguishing between electronic 2296

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A Table 1. Reaction Thermochemistry for Observed Bimolecular Product Formation from Cu+(1S) and Cu+(3D) thermochemistry (kJ/mol) 1

3

S

neutral reactant

observed product ion

SF6

SF3+

SF5Cld

SF3+ [S]

[S]

SF5+ [S] CuCl+ [D]

a

possible neutral products

a

CuF2 [D] + F CuF3 [T] CuF2 [D] + Cl CuClF [D] + F CuF [S] + ClF CuClF2 [T] CuCl [S] SF5 [D]

expt 236 ± − 43 ± − 126 ± − 11 ± 107 ±

36 36 38 16 18c

DFT

Δ∑

expt

221 72 39 115 123 −40 57 49

±1,0 ±1 ±1,0 ±1,0 0 ±1 0 ±1,0

−35 ± − −228 ± − −145 ± − −260 ± −164 ±

36 36 38 16 18c

D

DFTb

Δ∑

−51 −200 −233 −156 −149 −312 −214 −223

±2,±1,0 ±2,0 ±2,±1,0 ±2,±1,0 ±1 ±2,0 ±1 ±2,±1,0

a

Molecular singlet, doublet, and triplet states indicated by S, D, and T respectively. bIncorporates calculated Cu+(1S)/Cu+(3D) splitting of 271.5 kJ/ mol. Individual J-levels not explicitly treated at this level of theory. cIncorporates Cu+-Cl binding energy of 91 ± 10 kJ/mol obtained from ref 71. d Experimental values incorporate ΔfH°0(SF5Cl) lower limit of −985 kJ/mol from ref 45 (see text).

Scheme 1. Generalized Model Illustrating the Competition between Three-Body Association and Bimolecular Product Formationa

a

X = F, Cl.

experimental values in order to evaluate the accuracy of this level of theory. These comparisons are given in Table 1, and show that, in most cases, the calculated thermochemistry is within experimental error of the measured values.

■

RESULTS AND DISCUSSION Under the multicollisional conditions present in the drift cell, association competes with bimolecular product formation according to the model in Scheme 1: Here, Cu+ forms an initial energized complex with the neutral via the pathway described by ka. This initial encounter complex must subsequently redistribute the initial interaction energy in one of three ways. Decomposition back to reactants via kd results in an unproductive encounter; however, the energized complex may also access any available bimolecular product channels (kp) provided the overall energetics are favorable and there are no substantial kinetic barriers to reaction. Alternatively, if the lifetime of the complex is sufficiently long, collisions with a stabilizing third body such as He can yield an association product (ks[He]). The reacting systems described here display evidence of both association and a number of bimolecular product scenarios. In addition, evidence for secondary association was observed with both neutrals, as well as higher-order bimolecular product formation with SF5Cl. A more detailed analysis of these products follows. Primary bimolecular product formation with SF6 was limited exclusively to SF3+; however, formation of this ion is minor when compared with association. As we discuss below, this is largely due to unfavorable kinetics. Conversely, SF3+ appears prominently in product spectra for the Cu+/SF5Cl system. In this case, however, product growth curves for this reaction given in Figure 1 indicate that primary production of SF3+ competes with a second product channel yielding CuCl+. We

Figure 1. Product intensities for the reaction of Cu+(3D) with SF5Cl as a function of SF5Cl concentration. Dwell time per mass = 4 ms with signal intensity at each [SF5Cl] summed over 1 min acquisition time. T = 307 K, P = 3.5 Torr, E/N = 8.4 Td.

note that both of these species are also thermochemically accessible from SF4 and ClF respectively (the two contaminants in our SF5Cl) via Cu+(3D). However, the abundance of SF3+ and CuCl+ observed to be formed points to SF5Cl as their predominant source as they are both major products. SF5+ is also observed in this reaction; however, it is unclear from the data in Figure 1 whether this species is formed in a primary or secondary step. This said, other observations suggest that the most significant pathway for this species is one in which it is formed in a secondary step. This is discussed in more detail 2297

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A below. In addition, SF4Cl+ is formed in a higher-order step in the Cu+/SF5Cl system. State-Specificity and Bimolecular Reaction Sequencing. Since both Cu+(1S) and Cu+(3D) are present simultaneously in the drift cell, products arising from both states can be observed. State-specificity and reaction sequencing (primary vs secondary product formation) were deduced using a combination of criteria, foremost of these being the aforementioned requirement of exothermicity in order for a reaction to be observed. Possible primary bimolecular product channels are summarized in Table 1 for which at least one exothermic process exists from one of the Cu+ states. Unless otherwise noted, thermochemical changes for each reaction were calculated using 298 K enthalpy of formation data available from the work of Afeefy et al.,56 Kramida et al.,57 and Chase.58 Ionization energies were obtained from compilations by Lias59 (Cu and SF5) and Lias et al.60 (SF3). For SF5Cl, we have used ΔHf ≥ −985 kJ/mol as reported by Atterbury et al.45 This lower limit is in reasonable agreement with a later computational result using the G3 composite method,61 and yields better agreement (see Table 1) between our DFT results and the available experimental thermochemistry than the −1038.9 kJ/mol value currently reported for this parameter.58 Since the ΔH f (SF 5 Cl) is a lower bound, the SF 5 Cl thermochemical values listed in Table 1 therefore represent upper limits. Where the necessary thermochemistry is unavailable, reaction energetics were calculated using the density functional methods described above. As seen in Table 1, the exothermicity criterion is satisfied only via Cu+(3D) for all but one of the reactions resulting in fragmentation of either of the substrate neutrals. The sole exception to this observation is the possibility of exothermic formation of CuClF2 and SF3+ via Cu+(1S) (discussed below). Conservation of electron spin may also play a role, although these effects cannot be confirmed without detailed knowledge of the specific reaction surfaces involved. In addition, numerous examples of transition metal ion reactions have been reported in which spin is not conserved.12,62 This being noted, previous results with Cu+/small molecule systems are consistent with overall conservation of electron spin.23−26 Further, although all thermochemically accessible product channels are considered, none of the observations reported here provide definitive evidence requiring us to conclude that this is not the case in these systems since exothermic product channels for which spin is conserved can be proposed for all observed product ions. With respect to SF3+ formation for both neutrals, reaction energetics provides indirect evidence that concomitant formation of a neutral copper halide must occur. Although this ion is known to be formed via dissociative ionization of both SF6 and SF5Cl, dissociative charge-transfer to either Cu+(1S) or Cu+(3D) is energetically unfavorable for both molecules.63,64 In the Cu+(3D)/SF6 reaction, exothermic processes resulting in SF3+ include parallel formation of CuF2 and F, which is exothermic by 35 kJ/mol. Another intriguing possibility is that CuF3 is formed as the sole neutral product accompanying SF3+ production. This species has been examined in several independent computational studies with regard to its behavior as a superhalogen, i.e., a species with an electron affinity higher than any element in the periodic table.65−67 This previous work has yielded contradictory results regarding the ground state of CuF3. A singlet ground state is predicted by coupled-cluster methods,65 whereas DFT results using the B3LYP functional and a variety of triple-ζ basis sets

(including those incorporating effective core potentials and relativistic effects) suggest that the ground state is triplet.66,67 Our own DFT calculations also indicate that the triplet state lies lower than the singlet by approximately 15 kJ/mol. We have therefore based our thermochemical evaluations on the assumption that this is the ground state. Utilizing this computational result, production of SF3+ and CuF3 from Cu+(3D) is exothermic by 200 kJ/mol in addition to conserving spin overall. Exothermic SF3+ production from SF5Cl presents a number of possibilities with respect to possible neutral byproducts. Here also, CuF2 formation is thermochemically accessible, as is production of CuClF. However, no distinction can be made between these two solely on the basis of energetics. We are unaware of any experimental or theoretical determinations of the ground spin state of the mixed copper dihalide, but given that this species is likely to be structurally similar to CuCl2, the ground state of which is doublet, it is likely that spin is conserved in this product channel as well, and there is no a priori reason to exclude formation of either CuF2 or CuClF.68−70 It is also thermochemically possible to form CuF and ClF as byproducts of SF3+ in the Cu+(3D)/SF5Cl reaction in a process which does not conserve spin. Here, as with SF6, formation of the trivalent copper species CuClF2 can also be envisioned as a possible neutral product arising along with SF3+. In this case, however, DFT calculations indicate that this process is exothermic from both Cu+(1S) and Cu+(3D), making it impossible to assign state-specificity on the basis of energetics alone. However, as we demonstrate below, other evidence indicates it is unlikely that Cu+(1S) contributes to SF3+ formation in this reaction. Additional indications of state-specificity were obtained by correlating reactant and product ATDs as well as by comparing product spectra in the presence of different Cu+ state populations.23−26 Comparison spectra are shown in Figure 2 for the Cu+/SF5Cl reaction where, consistent with the exothermicity criterion, production of both CuCl+ and SF5+ can be seen to be enhanced when the proportion of Cu+(3D) is increased relative to the overall Cu+ population. SF3+ formation

Figure 2. Comparison spectra for Cu+/SF5Cl reaction in the presence of different Cu+(3D) populations. E/N = 8.4 Td; [SF5Cl] = 1.7 × 1013 mol/cm3; T = 306 K; Ptot = 3.5 Torr. Spectra have been normalized to the Cu+ intensity (not shown). 2298

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A is likewise enhanced by increased Cu+(3D). However, the previously discussed thermochemical ambiguity with respect to the possible formation of CuClF2 from either Cu+ state makes it difficult to draw conclusions regarding state-specificity in this way, since it is not possible to completely eliminate Cu+(3D) from the reaction in our instrument. Having said this, product correlation ATDs for SF3+ given in Figure 3 correlate well with

formation of this ion despite its apparent thermochemical favorability with respect to formation of CuClF2 as the neutral byproduct. Thus, we conclude that SF3+ is formed exclusively via Cu+(3D). Correlation of the SF3+ ATD in the Cu+/SF6 system is inhibited by the low abundance of this product in that reaction and is therefore less informative. However, peak SF3+ arrival times are delayed from that of the Cu+(3D) feature despite the fact that the energetic requirements for production of this ion dictate that it must come from the Cu+(3D) state. The reasons for this poor correlation are not clear, but could be related to the inefficiency of its formation. It is possible that SF3+ proceeds via the formation of a weakly bound metastable, Cu+(3D)·SF6 association complex which subsequently decomposes to yield SF3+. Such an association complex would have a lower mobility (as discussed below), and therefore any product arising indirectly from it would exhibit a delayed arrival time. Further, the weakly bound nature of such a metastable complex would be consistent with inefficient production of any species arising from it. Presumably, a portion of any population of such metastable association species would be collisionally stabilized within the drift cell and therefore observed. Indeed, clear evidence for the production of an association structure arising from Cu+(3D) is indicated in the Cu+·SF6 ATD that is discussed below. On the basis of the assessments of state-specificity given above, bimolecular product formation in the Cu+/SF5Cl system can be summarized by the possible reaction sequences emanating from Cu+(3D) shown in Scheme 2. Primary formation of CuCl+, SF3+, and SF5+ are described by steps 1a−1e, where all thermochemically accessible processes are shown. Having said this, we note that primary production of SF5+ remains uncertain, despite the fact that it appears at low extents of reaction. SF5+ has been previously demonstrated to result from dissociative ionization of SF5Cl;64 however, as in the case of SF3+, production of SF5+ via dissociative chargetransfer to Cu+(3D) is thermochemically inaccessible in our instrument. This suggests that, if occurring, production of SF5+ directly from Cu+(3D) must be accompanied by formation of CuCl in order to satisfy the exothermicity criterion. Direct

Figure 3. Correlation of SF3+ (dotted lines) and Cu+ ATDs (solid lines) at different relative populations of Cu+(3D). All ATDs are acquired at 306 K, E/N = 8.4 Td. Cu+ ATDs are normalized to the Cu+(1S) intensity. SF3+ ATDs acquired with XSF5Cl = 1.41 × 10−4 and are scaled to be proportional to the acquisition time for the associated Cu+ ATD.

Cu+(3D), indicating definitively that this Cu+ state contributes to its production. Furthermore, when ionizing parameters are adjusted to enhance the Cu+(3D) population present in the drift cell, SF3+ ATDs representative of acquisition times proportional to that of the related Cu+ ATD trend consistently with the amount of Cu+(3D) in both intensity and shape. This suggests that Cu+(1S) has little or no contribution to the

Scheme 2. Possible Exothermic Bimolecular Reaction Sequences for the Cu+(3D)/SF5Cl Systema

a

Dotted lines indicate processes in which spin is not conserved. 2299

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A production of SF5+ via Cl− abstraction by Cu+(3D) (process 1b in Scheme 2) is exothermic by 260 kJ/mol; consequently, SF5+ is included as a possible primary product in Scheme 2, step 1b. If all products are formed in their ground states, this reaction does not conserve spin. Alternatively, SF5+ could arise in a secondary step. SF3+ has been shown previously to be unreactive with SF5Cl.45 This is consistent with its behavior in Figure 1, which indicates that SF3+ is not depleted at higher extents of reaction. Thus, SF3+ can be ruled out as a possible SF5+ precursor. In contrast, CuCl+ (produced in parallel to SF3+) is clearly consumed at higher extents of reaction by SF5Cl, and could therefore be the source of SF5+ via Cl− abstraction (step 2a). This scenario is further supported by product analysis carried out in which the fractions that individual product ions represent relative to total product intensity are examined as a function of extent of reaction as defined by SF5Cl concentration and shown in Figure 4. Here, it

similarly spin-allowed. Thus, production of SF4Cl+ via CuCl+ must be given consideration as a possible process. This being said, SF5+ has been shown previously to abstract fluoride slowly from SF5Cl to yield SF4Cl+ (step 3), and could be the sole source of this product ion.45 Indeed, the delayed onset of the SF4Cl+ product fraction shown in Figure 4 is more consistent with tertiary production via SF5+ than with secondary production via CuCl+, thus arguing against step 2b as a significant means of production for this species. In light of the observations described above, we conclude that the sequence of steps: 1a → 2a → 3 is the most likely pathway leading to the production of CuCl+, SF5+, and SF4Cl+, and that this process occurs in parallel with SF3+ production. We also note that the sequence of reactions initiated by Cl abstraction is analogous to one we have postulated for the reaction of Cu+(3D) with several halogenated methane analogues.24,25 Finally, we point out that formation of all observed product ions can be explained by invoking reaction pathways consisting only of steps in which overall spin is conserved. Association Products. Association was also observed to occur with both reactant neutrals and arises almost entirely from Cu+(1S) as shown by the correlation ATDs given in Figure 5. State-specific production of the Cu+·SF6 adduct, via Cu+(1S) is consistent with previously reported observations for this reaction at low interaction energy.2,4 In addition, the correlation ATD for association with both neutral reactants suggests some contribution from Cu+(3D) as evidenced by the small portion of the adduct ATD in each case that precedes the Cu+(1S) feature. Geometry optimizations for these singlet association products as well as the unassociated neutral reactants were carried out using the DFT methods outlined above and are given in Figure 6. For the neutral substrates, experimental and previous calculated geometric parameters are available for comparison and are also provided. Our calculated bond lengths and angles for both SF6 and SF5Cl are in good agreement with experimental values and with those determined in previous calculations.58,61 Stable singlet association geometries for adducts of Cu+ with both SF6 and SF5Cl were located, all of which exhibited η2 coordination to the substrate molecule. For the Cu+(1S)/SF6 system, one association structure was identified. The calculated 298 K binding energy for this structure is 65 kJ/mol. In the case of SF5Cl, a somewhat more robust interaction with Cu+(1S) occurs in which the metal interacts directly with Cl and F atoms. This “frontal” geometry has a predicted binding energy of 117 kJ/mol. A search for additional stationary points for the Cu+(1S)·SF5Cl complex revealed two more association geometries in which Cu+ interacts exclusively with the fluorines. If we arbitrarily stipulate that Cl occupies an axial position, we can distinguish these two alternate geometries as axial and equatorial with respect to the fluorines interacting with Cu+. Although representative of stable stationary points, these two additional geometries are nonoptimal in that the binding in both is somewhat weaker than that of the frontal interaction (84 and 80 kJ/mol for the axial and equatorial structures respectively). All Cu+(1S)·SF5Cl association geometries are of Cs symmetry. For all three, the presence of the Cu+ subtly distorts the geometry of the unassociated neutral in that the bond lengths to sulfur for the atoms interacting directly with Cu+ are somewhat lengthened. In addition, the X-S-X (X = F,Cl) bond angles for the atoms interacting directly with Cu+ are slightly compressed. The behavior Cu+(1S)·SF5Cl system appears to be similar to that of

Figure 4. Product fractions for species originating from Cu+(3D) as a function of SF5Cl number density. SF3+ fractions not included for clarity. Trendlines added for visualization purposes. T = 307 K, P = 3.5 Torr, E/N = 8.4 Td.

is apparent that the depletion of CuCl+ occurs simultaneously with the growth of SF5+ at early extents of reaction, consistent with a process in which the former produces the latter as shown in step 2a. Using an enthalpy of formation of −5.9 kJ/mol for CuCl2(g) calculated as the sum of the ΔHf of the solid and the enthalpy of sublimation of 200 kJ/mol,72 the thermochemistry of step 2a can be shown to be ≤ −126 kJ/mol. Although differing in magnitude, this indication of exothermicity is in general agreement with thermochemistry obtained via DFT methods predicting that production of SF5+ via step 2a is exothermic by 31 kJ/mol. Thus, it seems likely that secondary production of SF5+ via this pathway is energetically possible in addition to conserving spin overall if CuCl2 is formed in its doublet ground state.68−70 In contrast to SF5+, the growth curve for SF4Cl+ shown in Figure 1 occurs at much higher extents of reaction, arguing against primary production. However, formation of this ion can be envisioned as arising in parallel with SF5+ in a secondary step via F− abstraction by CuCl+ (step 2b). This process is predicted to be essentially thermoneutral using DFT methods (−0.3 kJ/ mol), and presuming that ground state CuClF is doublet, is 2300

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A

Figure 5. Correlation ATDs for primary association products in the (a) Cu+(1S,3D)/SF6 and (b) Cu+(1S,3D)/SF5Cl reactions. E/N = 8.4 Td for both reactions. (a) T = 307 K, XSF6 = 2.3 × 10−4 (b) T = 308 K, XSF5Cl = 3.5 × 10−5.

Cu+(1S) association complexes incorporating the fluorinated methane analogues CF3Br and CF3I, where both frontal and backside association geometries are likewise predicted.25 As in those systems, we anticipate that the initial encounter complex for both Cu+(1S)·SF6 and Cu+(1S)·SF5Cl will sample multiple geometric possibilities at room temperature prior to being relaxed via collisions with He. In the case of the chlorinated neutral, however, at least three geometric options are possible− two with binding energies that are comparable to that of Cu+(1S)·SF6, and one which is stronger. Reaction Kinetics and Possible Mechanisms. Statespecific rate constants for depletion of Cu+ were determined for both reactant neutrals by observing the decay of the Cu+(3D) and Cu+(1S) ATD features in the presence of varying neutral concentrations at fixed reaction time. This is shown in Figure 7 for SF6. Analysis of the peak areas in these ATDs yields pseudo first-order kinetic plots such as those shown in Figure 8. Rate constants obtained from these measurements are given in Table 2. Note that in the case of the Cu+/SF5Cl system, the slope of the Cu+(1S) decay in Figure 8 is slightly higher than that of the Cu+(3D) decay, while the ordering of the rate constants given in Table 2 is reversed (the triplet reaction is faster). Notwithstanding the fact that these two rate constants are indistinguishable within the reproducibility of our experiment, this apparent inconsistency arises because the mobility of Cu+(3D) is higher than that of Cu+(1S), resulting a slightly shorter residence time in the drift cell for the triplet (8.1 × 10−5 s vs 1.1 × 10−4 s for the singlet under the conditions employed here). Since the rate constant is calculated by dividing the slope of the decay by the residence time, this leads to a higher rate constant for the triplet in this case. There is little loss in ATD resolution in Figure 7 in the presence of SF6. This is significant in that evidence of bridging between the arrival time features would indicate copper ions with drift times intermediate between the two states extracted from the discharge, pointing to an SF6-induced quenching process: Cu+(3D) → Cu+(1S). The absence of such bridging suggests that the triplet and singlet reaction surfaces remain isolated from one another during the initial ion/neutral interaction where dissociation back to the reactants is still

possible. Similar behavior was observed in the SF5Cl reactions. We note however that the Cu+(3D) feature in Figure 7 does exhibit a slight displacement to longer arrival times as the SF6 concentration is increased. This behavior is consistent with the formation of a weakly bound triplet association complex as evidenced by the small ATD feature preceding the major association peak in Figure 5. Cu+(3D) ions resulting from the dissociation of a “failed” Cu+(3D)·SF6 adduct exhibit slightly longer arrival times as a result of the portion of time spent in the drift cell as the lower mobility association species. Overall observed rate constants for all reactions were compared to their theoretical collision-limited values. The collision rate constant (kc) for SF6 represents the Langevin limit,73 whereas kc for SF5Cl was calculated using the parametrized average dipole orientation (ADO) model.74 We are unaware of any published experimental values for the polarizability (α) and dipole moment (μ) for SF5Cl (needed to calculate kc). We have therefore used α = 9.55 Å3 and μ = 0.538 D as calculated by Van Doren, et al. via density functional methods, yielding a limiting rate constant of 1.2 × 10−9cm3/ mol·s.61 Our measured values for depletion of both Cu+ states are higher than this, but agree within our estimated uncertainty of ±30%. Thus, we conclude that SF5Cl reacts with both Cu+ states at the collision rate. Since association is the sole process to occur in the Cu+(1S)/SF5Cl reaction, this indicates that the Cu+(1S)/SF5Cl binding in the initial encounter complex is strong enough such that little or no dissociation occurs before it is stabilized by sufficient collisions with He to yield the observed association product. Determination of branching ratios for the primary bimolecular products in the Cu+(3D)/ SF5Cl system is complicated by the possible contribution of primary SF5+ formation. However, the growth curves for SF3+ and CuCl+ in Figure 1 indicate that these two products are formed at similar rates. In light of the high efficiency of the triplet reaction, this suggests that production of both ions is rapid. With respect to SF3+ formation, this also indicates that if more than one of the four possible pathways in Table 1 leading to it is occurring, the sum of their individual efficiencies must approach unity. Production of SF3+ with concomitant formation of CuClF2 (step 1e in Scheme 2) is likely to proceed through a 2301

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A

Figure 6. Optimized structures for singlet association geometries and reactant substrate molecules. Bond lengths in angstroms, bond angles in degrees. All structures optimized at B3LYP/aug-cc-pVTZ level of theory. Experimental values are given in parentheses for comparison.

tight transition state requiring interaction between the Cu+ and three halogen atoms simultaneously, thereby lowering the efficiency of this product channel. Likewise, if occurring, step 1c

requires a spin change which may inhibit access to this product channel as well. It therefore seems reasonable to expect that the two possible processes described by step 1d represent the most 2302

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A

significant amount in this reaction. DFT calculations predict two exothermic product channels that could yield this ion via Cu+(3D). The first is accompanied by CuF3 production and the second yields CuF2 and F. In the latter case, the DFT result is consistent with an experimentally derived upper limit of −103 kJ/mol which incorporates a previous estimate of ΔHf(SF2Cl+) of 453 kJ/mol.45 For both pathways resulting in SF2Cl+, Cu+(3D) must interact solely with the fluorines. Since SF2Cl+ is not observed despite the likelihood of thermochemical favorability, this implies that chlorine is the preferred reactive site on SF5Cl. Similarly, SF2+ is notably absent from the product spectrum. For reaction channels resulting in parallel formation of either CuF or CuCl, this can be ascribed to unfavorable thermochemistry. However, we have previously observed SF2+ as a product in the reaction of Cu+(3D) with SF5CF3, where CuF2 and CF4 were hypothesized as the neutral byproducts via heterolytic cleavage of the C−S bond and followed (or accompanied) by fluoride transfer to CF3.26 In the Cu+(3D)/ SF5Cl reaction, the analogous process yielding SF2+ could result in concomitant formation of either CuF2 (with ClF) or CuClF (with F2). It is tentatively possible to distinguish between these two possibilities using the required energetics. Thermochemistry based on DFT calculations predicts that SF2+ production accompanied by CuClF and F2 is endothermic via Cu+(3D) by 61 kJ/mol. Alternatively, experimental thermochemical data suggest that SF2+ formation yielding CuF2 and ClF, is accompanied by an energy change of ≤ −10 ± 23 kJ/mol which includes the aforementioned −985 kJ/mol lower limit for ΔHf (SF5Cl).45 Thus, it is possible that SF2+ production via this product channel will not be prohibited on the basis of energetics. Thus, if the mechanism for SF2+ production does in fact require parallel formation of a neutral copper dihalide, the only energetically possible means for accomplishing this again requires that the Cu+(3D) interact only with the SF5Cl fluorines. In light of the fact that we do not observe SF2+, this again suggests that interaction with the chlorine is preferred. Since SF2+ production is thermochemically prohibited to proceed from an initial interaction geometry in which Cu+(3D) associates with the Cl, bimolecular product formation resulting from this interaction is restricted to those processes which are energetically possible, i.e., SF 3 + and CuCl + production. With respect to the former, this suggests that CuClF and F are the likely neutral byproducts. The rate constant for the Cu+(1S)/SF6 reaction has been reported previously to be 8.5 × 10−11 cm3/mol·sec at 295 K and 0.35 Torr.4 Even considering the uncertainty in both values, this is somewhat lower than the value of 3.2 × 10−10 cm3/mol·sec that we observe at 3.5 Torr. However, since association is the only product channel available to this Cu+ state, we would expect this parameter to exhibit a pressure dependence. Thus, our higher value does not seem unreasonable and reflects the enhanced collisional stabilization frequency at higher pressure. Nonetheless, SF6 association with Cu+(1S) is only 36% as efficient as the same process with SF5Cl. This is most likely the result of stronger binding between Cu+(1S) and the latter due to the presence of the larger, more polarizable Cl atom. The Cu+(3D)/SF6 rate constant is slower than that of the ground state by nearly an order of magnitude. Since we would expect the binding between Cu+(3D) and SF6 to be weaker than that of the ground state, it is perhaps not surprising that association via the excited state does not represent an efficient mechanism for its depletion. However,

Figure 7. Cu+ ATDs in the presence of varying SF6 concentrations at fixed reaction time. T = 305 K; E/N = 8.4 Td.

Figure 8. State-specific kinetic decay at T = 306 K, P = 3.5 Torr for Cu+(1S) (squares) and Cu+(3D) (circles) with SF6 (open symbols) and SF5Cl (solid symbols).

Table 2. State-Specific Second-Order Rate Constants for the Depletion of Cu+(1S) and Cu+(3D)a,b Cu+(1S) ‑9

reactant neutral

k/10

SF6 SF5Cl

0.32 1.4

Cu+(3D) ‑9

k/kc

k/10

k/kc

0.36 1

0.04 1.5

0.043 1

Rate constants are in units of cm3•molec−1•s−1. uncertainty estimated at ±30%. a

b

Experimental

likely means of SF3+ formation. Of these two, there is reason to believe that formation of CuClF is preferred. This is supported by two observations. First, the van der Waals radius of Cl (1.75 Å) is 19% larger than that of F (1.47 Å), and thus it presents a larger target for Cu+(3D).75 Further, Cl is substantially more polarizable (2.18 Å3 vs 0.557 Å3).76 These two factors make Cl a more favorable initial interaction site for the metal ion than F. Indirect evidence in support of this idea is given by the preferred frontal geometry in the Cu+(1S)/SF5Cl association product discussed above. Furthermore, we note that the chlorinated SF3+ analogue, SF2Cl+, is not observed in any 2303

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A Cu+(3D) is consumed inefficiently despite the presence of the parallel bimolecular process yielding SF3+, suggesting the possible presence of a kinetic barrier restricting access to this product channel. Alternatively, we note that the uncertainty in the available thermochemistry for SF3+ production via Cu+(3D) is substantial, and includes that possibility that this process is essentially thermoneutral. If occurring, this scenario would be kinetically indistinguishable from an exothermic process with a barrier of similar magnitude in the exit channel.

represents a minor contribution to overall SF5+ formation. In general, the behavior of the Cu+(3D)/SF5Cl reaction is in many ways reminiscent of that of CF3X (X = Cl, Br, I). Clearly, more products are thermochemically possible with SF5Cl than in the halocarbon systems, and are in fact observed. Nonetheless, we can generalize to say that, as with the halogenated methanes, the reactive options with Cu+ for these sulfur-containing substrate reactants are enhanced by the incorporation of a larger, more labile halogen.

■

■

SUMMARY AND CONCLUSIONS The state-specific results reported here for the reactions of Cu+(1S,3D) with SF6 and SF5Cl present contrasts between not only the reactivity of the two copper ion states, but between that of the two neutrals as well. Under the near-thermal conditions occurring in our instrument, SF6 exhibits a limited capability to react with either Cu+ state, resulting in association proceeding primarily from Cu+(1S) with minor contributions to adduct formation also exhibited by Cu+(3D). In addition, inefficient formation of SF3+ via Cu+(3D) is also observed. By comparison, SF5Cl presents a much more reactive substrate. As with SF6, association is the only product formed via Cu+(1S); however, the presence of the Cl provides an enhanced binding site for the metal which results in efficient adduct formation between Cu+(1S) and SF5Cl. Here also, Cu+(3D) contributes inefficiently to association. Whereas DFT calculations reveal a single association structure for Cu+(1S)/SF6, three distinct structures were located for Cu+(1S)/SF5Cl association. One of these structures represents the global minimum and can be described as a frontal geometry in which Cu+(1S) interacts with the chlorine and one of the fluorines. The other two represent local minima in which the metal center interacts exclusively with two of the SF5Cl fluorines. Significantly, the frontal geometry exhibits a binding energy predicted to be nearly twice that of the Cu+(SF6) adduct, indicating the likely reason for the improved association efficiency in the Cu+(1S)/SF5Cl system. In addition, the chlorinated neutral readily consumes Cu+(3D) in at least two bimolecular processes resulting in production of SF3+ and CuCl+. SF5+ may also occur as a primary product in this reaction, but this cannot be definitively concluded from the evidence discussed here. For both neutral substrates, dissociative charge-transfer resulting in bimolecular product formation can ruled out on the basis of unfavorable energetics. This means primary production of SF3+ (from both neutrals) and SF5+ (from SF5Cl) must be accompanied by the formation of a molecular copper halide species in order to satisfy the energetic requirements for reaction. For SF3+ via SF6, two possibilities exist resulting in production of either CuF2 or CuF3, which cannot be distinguished from one another based on these results. Production of SF3+ via Cu+(3D)/SF5Cl can likewise be envisioned to proceed with concomitant formation of a number of possible copper halide species including CuF, CuF2, CuClF, and CuClF2. In this case however, the observed product formation combined with the kinetics of Cu+(3D) depletion suggest that the most likely process results in production of CuClF and F. Similarly, primary SF5+ formation via Cu+(3D)/SF5Cl must occur by Cl− abstraction to yield CuCl as the neutral byproduct in a process in which spin is not conserved. However, this cannot be confirmed since the appearance of this ion can also be rationalized to proceed via a secondary step initiated by CuCl+. Indeed, the results presented here suggest that secondary production of SF5+ via CuCl+ is more likely, and that if occurring, the primary pathway

AUTHOR INFORMATION

Corresponding Author

*(W.S.T.) E-mail: [email protected]. Telephone: 501-852-2529. Present Address †

Ralph E. Martin Department of Chemical Engineering, 3202 Bell Engineering Center, University of Arkansas, Fayetteville, AR 72701−1201 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Support for this research was provided by the National Science Foundation under Grant No. CHE-0956393.

■

REFERENCES

(1) Seemeyer, K.; Schröder, D.; Kempf, M.; Lettau, O.; Müller, J.; Schwarz, H. Face Selectivity of the C-H Bond Activation of Cyclohexane by the “Bare” First-Row Transition Metal Cations Sc+Zn+. Organometallics 1995, 14, 4465−4470. (2) Jiao, C. Q.; Freiser, B. S. Reactions of Transition-Metal Ions with Sulfur Hexafluoride in the Gas Phase. J. Am. Chem. Soc. 1993, 115, 6268−6276. (3) Shayesteh, A.; Lavrov, V. V.; Koyanagi, G. K.; Bohme, D. K. Reactions of Atomic Cations with Methane: Gas Phase RoomTemperature Kinetics and Periodicities in Reactivity. J. Phys. Chem. A 2009, 113, 5602−5611. (4) Cheng, P.; Shayesteh, A.; Bohme, D. K. Gas-Phase Reactions of Sulfur Hexafluoride with Transition Metal and Main Group Atomic Cations: Room-Temperature Kinetics and Periodicities in Reactivity. Inorg. Chem. 2009, 48, 1018−1029. (5) Koyanagi, G. K.; Bohme, D. K. Gas-Phase Reactions of Carbon Dioxide with Atomic Transition-Metal and Main-Group Cations: Room-Temperature Kinetics and Periodicities in Reactivity. J. Phys. Chem. A 2006, 110, 1232−1241. (6) Lavrov, V. V.; Blagojevic, V.; Koyanagi, G. K.; Orlova, G.; Bohme, D. K. Gas-Phase Oxidation and Nitration of First-, Second-, and ThirdRow Atomic Cations in Reactions with Nitrous Oxide. J. Phys. Chem. A 2004, 108, 5610−5624. (7) Cheng, P.; Koyanagi, G. K.; Bohme, D. K. Heavy Water Reactions with Atomic Transition-Metal and Main-Group Cations: Gas Phase Room-Temperature Kinetics and Periodicities in Reactivity. J. Phys. Chem. A 2007, 111, 8561−8573. (8) Blagojevic, V.; Flaim, E.; Jarvis, M. J. Y.; Koyanagi, G. K.; Bohme, D. K. Nitric Oxide as an Electron Donor, an Atom Acceptor, and a Ligand in Reactions with Atomic Transition-Metal and Main-Group Cations in the Gas Phase. J. Phys. Chem. A 2005, 109, 11224−11235. (9) Koyanagi, G. K.; Caraiman, D.; Blagojevic, V.; Bohme, D. K. GasPhase Reactions of Transition-Metal Ions with Molecular Oxygen: Room-Temperature Kinetics and Periodicities in Reactivity. J. Phys. Chem. A 2002, 106, 4581−4590. (10) Caraiman, D.; Koyanagi, G. K.; Bohme, D. K. Gas-Phase Reactions of Transition-Metal Ions with Hexafluorobenzene: RoomTemperature Kinetics and Periodicities in Reactivity. J. Phys. Chem. A 2004, 108, 978−986. 2304

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A

(32) Totterdill, A.; Kovács, T.; Gómez Martín, J. C.; Feng, W.; Plane, J. M. C. Mesospheric Removal of Very Long-Lived Greenhouse Gases SF6 and CFC-115 by Metal Reactions, Lyman-α Photolysis, and Electron Attachment. J. Phys. Chem. A 2015, 119, 2016−2025. (33) IPCC. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, U.K., 2013. (34) Fehsenfeld, F. C. Ion Chemistry of SF6. J. Chem. Phys. 1971, 54, 438−439. (35) Jarvis, G. K.; Kennedy, R. A.; Mayhew, C. A.; Tuckett, R. P. A Selected Ion Flow Tube Study of the Reactions of Several Cations with the Group 6B Hexafluorides SF6, SeF6, and TeF6. J. Phys. Chem. A 2000, 104, 10766−10776. (36) Nichols, C. M.; Yang, Z.; Worker, B. B.; Hager, D. R.; Nibbering, N. M. M.; Bierbaum, V. M. Gas-Phase Reactions of the Atomic Oxygen Radical Cation with Halogenated Compounds. Phys. Chem. Chem. Phys. 2013, 15, 561−567. (37) Morris, R. A.; Miller, T. M.; Viggiano, A. A.; Paulson, J. F.; Solomon, S.; Reid, G. Effects of Electron and Ion Reactions on Atmospheric Lifetimes of Fully Fluorinated Compounds. J. Geophys. Res. 1995, 100, 1287−1294. (38) Case, J. R.; Ray, N. H.; Roberts, H. L. Sulphur Chloride Pentafluoride: Reaction with Unsaturated Hydrocarbons. J. Chem. Soc. 1961, 2066−2070. (39) Winter, R.; Gard, G. L. New Pentafluorothio (SF5) Esters. J. Fluorine Chem. 1994, 66, 109−116. (40) Aït-Mohand, S.; Dolbier, W. R., Jr. New and Convenient Method for Incorporation of Pentafluorosulfanyl (SF5) Substituents into Aliphatic Organic Compounds. Org. Lett. 2002, 4, 3013−3015. (41) Brel, V. K. Synthesis and Diels-Alder Reactions of Dienophiles with Pentafluoro-λ6-sulfanyl (SF5) Moiety. Synthesis 2006, 2, 339−343. (42) Dolbier, W. R., Jr.; Aït-Mohand, S.; Schertz, T. D.; Sergeeva, T. A.; Cradlebaugh, J. A.; Mitani, A.; Gard, G. L.; Winter, R. W.; Thrasher, J. S. A Convenient and Efficient Method for Incorporation of Pentafluorosulfanyl (SF5) Substituents into Aliphatic Compounds. J. Fluorine Chem. 2006, 127, 1302−1310. (43) Ponomarenko, M. V.; Serguchev, Y. A.; Röschenthaler, G. The Reactions of 3,7-Dimethylenebicyclo[3.3.1]nonane, Norbornadiene and cis,cis-1,5-Cyclooctadiene with Pentafluoro-λ6-sulfanyl Chloride. J. Fluorine Chem. 2010, 131, 270−273. (44) Ponomarenko, M. V.; Kalinovich, N.; Serguchev, Y. A.; Bremer, M.; Röschenthaler, G. Synthesis of Pentafluoro-λ6-sulfanyl Substituted Acetylenes for Novel Liquid Crystals. J. Fluorine Chem. 2012, 135, 68− 74. (45) Atterbury, C.; Critchley, A. D. J.; Kennedy, R. A.; Mayhew, C. A.; Tuckett, R. P. A Study of the Gas Phase Reactions of Various Cations with Two Derivatives of SF6; SF5CF3 and SF5Cl. Phys. Chem. Chem. Phys. 2002, 4, 2206−2223. (46) Taylor, W. S.; Campbell, A. S.; Barnas, D. F.; Babcock, L. M.; Linder, C. B. Thermal and Near-Thermal Reactions of Pt+, and Au+ with Small Alkenes. J. Phys. Chem. A 1997, 101, 2654−2661. (47) Taylor, W. S.; Everett, W. R.; Babcock, L. M.; McNeal, T. L. Application of a Glow Discharge Ion Source in a Flowing Afterglow Study of Transition Metal Ion Chemistry. Int. J. Mass Spectrom. Ion Processes 1993, 125, 45−54. (48) Taylor, W. S.; Spicer, E. M.; Barnas, D. F. Metastable Metal Ion Production in Sputtering DC Glow Discharge Plasmas: Characterization by Electronic State Chromatography. J. Phys. Chem. A 1999, 103, 643−650. (49) Kemper, P. R.; Bowers, M. T. Electronic-State Chromatography: Application to First-Row Transition Metal Ions. J. Phys. Chem. 1991, 95, 5134−5146. (50) Armentrout, P. B. Electronic State Chromatography: Ion Mobility of Atomic Cations and Their Electronic States. In Ion Mobility Mass Spectrometry: Theory and Applications; Wilkins, C. L., Trimpin, S., Eds.; CRC Press: Boca Raton, FL, 2011. (51) Taylor, W. S.; Fancher, K. G.; Chen, L. C.; Ward, B. K.; Cameron, C. A. Probing Ni+ Metastable State Production in

(11) Schwarz, H. How and Why do Cluster Size, Charge State, and Ligands Affect the Course of Metal-Mediated Gas-Phase Activation of Methane? Isr. J. Chem. 2014, 54, 1413−1431. (12) Roithová, J.; Schröder, D. Selective Activation of Alkanes by Gas-Phase Metal Ions. Chem. Rev. 2010, 110, 1170−1211. (13) Schalley, C. A.; Schröder, D.; Schwarz, H. Activation of C-C and C-H Bonds of cis- and trans-1-Acetyl-2-methylcyclopropane by Bare Metal(I) Cations in the Gas Phase: Comparative Study of the FirstRow Transition-Metal Ions Cr+-Cu+. Organometallics 1995, 14, 317− 326. (14) Eller, K.; Schwarz, H. Gas-Phase Chemistry of Isolated Metal Ions: Are Predictions on Catalytic Properties Possible Based on the study of Simple Model Reactions? Ber. Bunsenges. Phys. Chem. 1990, 94, 1339−1343. (15) Lebrilla, C. B.; Drewello, T.; Schwarz, H. Only the Cu+ Coordinated to the π-System of the CN Group Activates Upon Collision C-H and C-C Bonds of Nitriles. Organometallics 1987, 6, 2450−2451. (16) Eller, K.; Zummack, W.; Schwarz, H. Gas-Phase Chemistry of Bare Transition Metal Ions in Comparison. J. Am. Chem. Soc. 1990, 112, 621−627. (17) Eller, K.; Karrass, S.; Schwarz, H. Gas-Phase Chemistry of Bare Transition Metal Ions in Comparison 2. Reactions of Ti+ to Zn+ with 2-Ethylbutanenitrile. Organometallics 1992, 11, 1637−1646. (18) Eller, K.; Schröder, D.; Schwarz, H. Gas-Phase Reactions of Secondary and Tertiary Isocyanates with Ti+ and Zn+: Model Studies for Catalytic Processes. J. Am. Chem. Soc. 1992, 114, 6173−6182. (19) Burnier, R. C.; Byrd, G. D.; Freiser, B. S. Copper(I) Chemical Ionization-Mass Spectrometric Analysis of Esters and Ketones. Anal. Chem. 1980, 52, 1641−1650. (20) Cassady, C. J.; Freiser, B. S. Gas-Phase Reactions of Transition Metal Ions with Methyl Nitrite and Nitromethane. J. Am. Chem. Soc. 1985, 107, 1566−1573. (21) Jones, R. W.; Staley, R. H. Gas-Phase Chemistry of Cu+ with Alkyl Chlorides. J. Am. Chem. Soc. 1980, 102, 3794−3798. (22) Jones, R. W.; Staley, R. H. Gas-Phase Chemistry of Ag+ with Alkyl Chlorides and of Cu+ and Ag+ with Alcohols. J. Phys. Chem. 1982, 86, 1669−1674. (23) Taylor, W. S.; May, J. C.; Lasater, A. S. Reactions of Cu+ (1S, 3 D) and Au+(1S, 3D) with CH3Br. J. Phys. Chem. A 2003, 107, 2209− 2215. (24) Taylor, W. S.; Matthews, C. C.; Parkhill, K. S. Reactions of Cu+(1S, 3D) with CH3Cl, CH2ClF, CHClF2, and CClF3. J. Phys. Chem. A 2005, 109, 356−365. (25) Taylor, W. S.; Abrams, M. L.; Matthews, C. C.; Byers, S.; Musial, S.; Nichols, C. M. State-Specific Reactions of Cu+(1S, 3D) with CH3X and CF3X (X = Cl, Br, I): Exploring the Influence of Dipole Orientation on Association and C−X Bond Activation. J. Phys. Chem. A 2012, 116, 3979−3988. (26) Taylor, W. S.; Manion, J. M.; Church, C. M.; Redmon, X. S.; Scheuter, B. A. State-Specific Reactions of Cu+(1S,3D,1D) with the Super Greenhouse Gas SF5CF3. J. Phys. Chem. A 2014, 118, 10919− 10926. (27) Rodgers, M. T.; Walker, B.; Armentrout, P. B. Reactions of Cu+(1S and 3D) with O2, CO, CO2, N2, NO, N2O and NO2 Studied by Guided Ion Beam Mass Spectrometry. Int. J. Mass Spectrom. 1999, 182183, 99−120. (28) Yanabu, S.; Murayama, Y.; Matsumoto, S. SF6 Insulation and its Application to HV Equipment. IEEE Trans. Electr. Insul. 1991, 26, 358−366. (29) Donnelly, V. M.; Kornblit, A. Plasma Etching: Yesterday, Today, and Tomorrow. J. Vac. Sci. Technol., A 2013, 31, 050825. (30) Luo, A. A. Magnesium Casting Technology for Structural Applications. J. Magnesium Alloys 2013, 1, 2−22. (31) Stiller, G. P.; von Clarmann, T.; Hoepfner, M.; Glatthor, N.; Grabowski, U.; Kellmann, S.; Kleinert, A.; Linden, A.; Milz, M.; Reddmann, T.; et al. Global Distribution of Mean Age of Stratospheric Air from MIPAS SF6 Measurements. Atmos. Chem. Phys. 2008, 8, 677− 695. 2305

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306

Article

The Journal of Physical Chemistry A Sputtering Glow Discharges with the Use of Diagnostic Reactions. Int. J. Mass Spectrom. 2013, 352, 29−36. (52) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford, CT, 2009. (53) Balabanov, N. B.; Peterson, K. A. Systematically Convergent Basis Sets for Transition Metals. I. All-Electron Correlation Consistent Basis Sets for the 3d Elements Sc-Zn. J. Chem. Phys. 2005, 123, 064107. (54) Dunning, T. H., Jr. Gaussian Basis Sets for use in Correlated Molecular Calculations. I. The Atoms Boron Through Neon and Hydrogen. J. Chem. Phys. 1989, 90, 1007−1023. (55) Woon, D. E.; Dunning, T. H., Jr. Gaussian Basis Sets for use in Correlated Molecular Calculations. III. The Atoms Aluminum Through Argon. J. Chem. Phys. 1993, 98, 1358−1371. (56) Afeefy, H. Y.; Liebman, J. F., Stein, S. E. In Neutral Thermochemical Data, NIST Chemistry WebBook, NIST Standard Reference Database Number 69; Linstrom, P. J., Mallard, W. G., Eds.; National Institute of Standards and Technology: Gaithersburg, MD, 2015. (57) Kramida, A.; Ralchenko, Yu.; Reader, J.; NIST ASD Team in NIST Atomic Spectra Database (version 5.2), National Institute of Standards and Technology: Gaithersburg, MD, 2014. (58) Chase, M. W. NIST-JANAF Thermochemical Tables, 4th ed., Journal of Physical and Chemical Reference, Monograph 9; AIP: New York, 1998; pp 1−195. (59) Lias, S. G. In Ion Energy Evaluation, NIST Chemistry WebBook, NIST Standard Reference Database Number 69; Linstrom, P. J., Mallard, W. G., Eds.; National Institute of Standards and Technology: Gaithersburg, MD, 2015. (60) Lias, S. G., Liebman, J. F. in Ion Energetics Data, NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Eds. Linstrom, P. J., Mallard, W. G., National Institute of Standards and Technology: Gaithersburg, 2015. (61) Van Doren, J. M.; Miller, T. M.; Viggiano, A. A. G3 and Density Functional Theory Investigations of the Structures and Energies of SFnCl (n = 0−5) and Their Anions. J. Chem. Phys. 2008, 128, 094310. (62) Schröder, D.; Shaik, S.; Schwarz, H. Two-State Reactivity as a New Concept in Organometallic Chemistry. Acc. Chem. Res. 2000, 33, 139−145. (63) Pullen, B. P.; Stockdale, J. A. D. Dissociative Ionization of SF6 by Electron Impact. Int. J. Mass Spectrom. Ion Phys. 1976, 21, 35−42. (64) Chim, R. Y. L.; Cicman, P.; Märk, T. D.; Mayhew, C. A.; Scheier, P.; Tuckett, R. P. Vacuum-UV and Dissociative Ionisation Studies of SF5Cl. Int. J. Mass Spectrom. 2007, 261, 208−217. (65) Solomonik, V. G.; Boggs, J. E.; Stanton, J. F. The Ground Singlet and Low-Lying Triplet Electronic States of CuF3. J. Mol. Struct.: THEOCHEM 2000, 496, 213−216. (66) Wang, Q.; Sun, Q.; Jena, P. Superhalogen Properties of CuFn Clusters. J. Chem. Phys. 2009, 131, 124301. (67) Koirala, P.; Willis, M.; Kiran, B.; Kandalam, A. K.; Jena, P. Superhalogen Properties of Fluorinated Coinage Metal Clusters. J. Phys. Chem. C 2010, 114, 16018−16024. (68) Wang, S. G.; Schwarz, W. H. E. Density Functional Study of First Row Transition Metal Dihalides. J. Chem. Phys. 1998, 109, 7252− 7262. (69) Ross, A. J.; Bacis, R.; Bouvier, A. J.; Churassy, S.; Coste, J. C.; Crozet, P.; Russier, I. The Ground State of the Cupric Chloride Molecule from Laser-Induced Fluorescence. J. Mol. Spectrosc. 1993, 158, 27−39. (70) Barnes, M. P.; Carter, R. T.; Lakin, N. M.; Brown, J. M. Observation and Analysis of Rotational and Nuclear Hyperfine Structure in Bands of the Red System of Copper Dichloride. J. Chem. Soc., Faraday Trans. 1993, 89, 3205−3215. (71) Luo, Y. R. Comprehensive Handbook of Chemical Bond Energies; CRC Press: Boca Raton, FL, 2007.

(72) Feber, R. C. Heats of Dissociation of Gaseous Halides; Los Alamos Report 3164, Los Alamos Scientific Laboratory of the University of California: 1965. (73) Gioumousis, G.; Stevenson, D. P. Reactions of Gaseous Molecule Ions with Gaseous Molecules. V. Theory. J. Chem. Phys. 1958, 29, 294−299. (74) Su, T.; Chesnavich, W. J. Parameterization of the Ion-Polar Molecule Collision Rate Constant by Trajectory Calculations. J. Chem. Phys. 1982, 76, 5183−5185. (75) Bondi, A. van der Waals Volumes and Radii. J. Phys. Chem. 1964, 68, 441−451. (76) Miller, T. M.; Bederson, B. Atomic and Molecular Polarizabilities-A Review of Recent Advances. Adv. At. Mol. Phys. 1978, 13, 1−55.

2306

DOI: 10.1021/acs.jpca.6b01534 J. Phys. Chem. A 2016, 120, 2295−2306