Rotational Spectroscopy and Molecular Structure of the 1,1,2

DOI: 10.1021/jp107494t. Publication Date (Web): September 28, 2010. Copyright © 2010 American Chemical Society. * To whom correspondence should be ad...
1 downloads 10 Views 196KB Size
J. Phys. Chem. A 2010, 114, 10975–10980

10975

Rotational Spectroscopy and Molecular Structure of the 1,1,2-Trifluoroethylene-Hydrogen Chloride Complex Helen O. Leung,* Mark D. Marshall,* Margaret R. Ray, and Justin T. Kang Department of Chemistry, Amherst College, P.O. Box 5000, Amherst, Massachusetts 01002-5000, United States ReceiVed: August 9, 2010; ReVised Manuscript ReceiVed: September 14, 2010

Fourier transform microwave spectra in the 6-20 GHz region are obtained for the complex formed between 1,1,2-trifluoroethylene and hydrogen chloride, including both 35Cl and 37Cl isotopomers. Analysis of the spectra provides rotational constants and additionally, the complete quadrupole hyperfine coupling tensor in both the inertial and principal electric field gradient axis systems. The inertial information contained in the rotational constants combined with the results of the hyperfine analysis provides the structure for CF2CHF-HCl. A primary, hydrogen bonding interaction exists between the HCl donor and the F atom geminal to the H atom on the substituted ethylene. The hydrogen bond is bent from linearity to allow a secondary interaction to form between this H atom and the Cl atom. Comparisons made to similar complexes involving both other protic acids (HF and HCCH) and fluoroethylenes (vinyl fluoride and 1,1-difluoroethylene) reveal the effects of varying gas phase hydrogen bond donor strength, of increasing fluorine substitution on fluorine atom nucleophilicity, and on the relative importance of steric versus electrostatic effects in determining the structures of these species. Introduction In an effort to understand the effects that changes in electronic distribution have on the character of a molecule’s participation in intermolecular interactions, we have been focusing on weakly bound complexes formed between members of a class of molecules with multiple functionalities, halogen substituted ethylenes, and one of three protic acids (HF, HCl, and HCCH, collectively represented by HX in the following). These acids contain both an electropositive proton and an electron-rich portion (F, Cl, and the acetylenic bond, respectively). By changing the number and type of halogen substituents on the ethylenic portion of the complex, we are able to tune the electronic properties of the functionalities of this subunit and observe changes in its interactions with acids of differing hydrogen bond donor ability. Thus far, with the exception of the one complex, 1,1,2-trifluoroethylene-HCl, our group and the Legon group have completed studies on three sets of these complexes, namely, vinyl fluoride-HX,1-3 1,1-difluoroethylene-HX,4-6 and 1,1,2-trifluoroethylene-HX.7,8 The findings for these complexes are summarized below. In each case, two interactions are observed for these complexes: a primary hydrogen bond from HX to an F atom in the substituted ethylene, and a secondary interaction between an H atom in the substituted ethylene and the nucleophilic portion of HX. For complexes containing vinyl fluoride and 1,1difluoroethylene, HX interacts with the F and H atoms located cis to each other, that is, HX binds to these substituted ethylenes across the double bond. On the other hand, for the complexes containing 1,1,2-trifluoroethylene, HX interacts with the F and H atoms connected to the same C atom, that is, HX is positioned at one end of the double bond. The mode of bonding and the structural parameters of these complexes provide details of the * To whom correspondence should be addressed. Phone: +01 413-542-2006 (M. D. M.), +01 413-542-2660 (H. O. L.). E-mail: [email protected] (M. D. M.), [email protected] (H. O. L.); Fax: +01 413-542-2735.

nature of the interactions and of the electronic distribution in the substituted ethylenes. For the fluorine substituted ethylenes, as the number of fluorine substituents increases, there is a corresponding decrease in the number of H atoms. The F atoms must share the polarizable electron density of the double bond and of the smaller number of hydrogen atoms. As a result, each F atom should be less nucleophilic while each H atom should become more electropositive. Indeed, it is observed that with the same acid partner, the primary hydrogen bond between the acid and the fluorine substituted ethylene becomes longer, and thus weaker, as one proceeds from vinyl fluoride to 1,1-difluoroethylene to 1,1,2-trifluoroethylene. Even more interestingly, HX interacts with a different pair of H and F atoms in 1,1,2-trifluoroethylene than it does in the vinyl fluoride and 1,1-difluoroethylene complexes, suggesting that, since it is this geminal F atom that participates in the hydrogen bond, the F atom in 1,1,2trifluoroethylene geminal to the H atom is more nucleophilic than the one located cis to the H atom. The secondary interaction is made possible by a bend in the primary hydrogen bond, and therefore, the extent that this bond deviates from linearity provides information concerning the delicate balance between the attractive secondary interaction and the unfavorable distortion. For complexes with the same HX partner, the atoms that participate in the secondary interactions remain the same. Therefore, the value of the angle of deviation from linearity is an indication of the strength of the primary interaction. Indeed, experimentally, this value increases from vinyl fluoride to 1,1-difluoroethylene to 1,1,2-trifluoroethylene, paralleling the trends of the increasing length (and the decreasing strength) of the primary hydrogen bond. Given the location of the primary hydrogen bond for each substituted ethylene, there is only one possible site for the secondary interaction for the complexes containing 1,1-difluoroethylene and 1,1,2-trifluoroethylene but there are two possible sites for vinyl fluoride: the H atoms located cis and geminal to the F atom. Although the geminal H atom should be more

10.1021/jp107494t  2010 American Chemical Society Published on Web 09/28/2010

10976

J. Phys. Chem. A, Vol. 114, No. 41, 2010

electropositive, given that it is only two bonds away (instead of three bonds as is the cis H atom) from the F atom, the observed secondary interaction in vinyl fluoride complexes involves the H atom cis to the F atom, indicating that effects other than electrostatics considerations are at play. In fact, our work on the trans-1,2-difluoroethylene-HF complex9 shows that when electrostatic factors are the same for both possible secondary interaction sites, HF will bind across the double bond. This configuration is therefore sterically more favorable. Applying this finding to the vinyl fluoride-HX complexes, it can be concluded that steric effects are more important than electrostatics factors in determining the secondary binding site on vinyl fluoride. Although comparisons of the structures between fluorine substituted ethylene-HX complexes with the same acid subunit yield information concerning the substituted ethylene subunits, comparisons of the structures of these complexes with the same substituted ethylene subunit reveal the behavior of the three different acids in intermolecular interactions. In the series HF to HCl to HCCH, both the length of the primary bond and its deviation from linearity increase, signifying a weakening of the hydrogen bond. (There appears to be one exception: the deviation of the hydrogen bond from linearity for vinyl fluoride-HF and vinyl fluoride-HCl are the same within experimental uncertainty.10) It is further interesting to note that the values of the CF · · · H angle in the vinyl fluoride-HX and 1,1-difluoroethylene-HX complexes are all approximately 120° whereas those in the two 1,1,2-trifluoroethylene-HX complexes studied so far are substantially smaller, between 104 and 109°. We report here the structure of the 1,1,2-trifluoroethyleneHCl complex. This work completes our series of trifluoroethylene-containing complexes and is therefore intended to further our understanding of the behavior of halogen-substituted ethylenes in intermolecular interactions. Comparisons both among complexes with a common protic acid and among those with a common fluoroethylene demonstrate that although steric factors play the primary role in determining the structure of complexes with vinyl fluoride, electrostatic interactions predominate in the complexes with 1,1,2-trifluoroethylene. Experimental Section The rotational spectra of two isotopomers, CF2CHF-H35Cl and CF2CHF-H37Cl, each in natural abundance, are studied in the 6-20 GHz region with a Balle-Flygare pulsed molecular beam, Fourier transform microwave spectrometer.11 The complex is formed by expanding a mixture of 1% CF2CHF and 1% HCl in argon through a 0.8 mm diameter nozzle at a stagnation pressure of approximately 2 atm into the cavity. After a suitable delay to allow for the supersonic beam expansion, the complexes are excited by a microwave pulse. Following its dissipation, if a rotational transition lies within the bandwidth of this excitation pulse, the free induction decay time-domain signal is coupled out of the cavity. The signal, after being down converted in two steps by a two stage heterodyne detection system to a center frequency of 2.5 MHz, is typically digitized for either 1024 or 2048 data points at a sampling frequency of 10 MHz, and corrected for background. The background corrected signal is averaged and then zero filled to a record length twice as long as the number of data points collected (that is, to a 2048- or 4096-point record length) before Fourier transformation to give a frequency domain signal, which has a resolution element of 4.8 or 2.4 kHz, respectively. Because the molecular beam is parallel to the mirror axis, each transition appears as a Doppler doublet; the rest frequency of the transition is the mean frequency of the two Doppler components.

Leung et al.

Figure 1. The 818-717 rotational transition for the CF2CHF-H35Cl complex showing four chlorine quadrupole hyperfine components labeled by F′-F′′.

Unlike the two trifluoroethylene-containing species we studied previously, a quadrupolar nucleus (Cl) is present in CF2CHF-HCl. Thus, nuclear quadrupole coupling interactions split each rotational level into several sublevels, contributing to significantly weaker intensities for the spectral lines than those that were observed for CF2CHF-HF and CF2CHF-HCCH. Consequently, we are not able to observe any 13C-containing isotopomers in natural abundance as we did for the other two complexes. Additionally, we did not attempt to record the spectrum for DCl-containing isotopomers because the presence of a second quadrupolar nucleus, D, would further dilute the intensity of each spectral line. Moreover, because HCl and DCl are likely to have different zero point motions in their respective complexes, analysis of spectra for the CF2CHF-DCl isotopomers may not, in fact, provide useful assistance in determining the structural parameters of the complex. Results Spectral Analysis. We have observed 63 and 37 a and b type transitions for the CF2CHF-H35Cl and CF2CHF-H37Cl isotopomers, respectively. Tables of observed and calculated transition frequencies for all isotopomers studied are available as Supporting Information. A typical rotational transition showing the chlorine nuclear quadrupole hyperfine structure is shown in Figure 1. The rotational spectrum for each isotopomer is analyzed using the Watson S-reduced Hamiltonian in the Ir representation12 with the inclusion of nuclear quadrupole coupling interaction for the Cl nucleus. The angular momentum coupling scheme employed couples the rotational angular momentum (J) to the spin angular momentum of Cl (I) to give the total angular momentum (F). The spectroscopic constants for each isotopomersincluding 3 rotational constants, 5 quartic centrifugal distortion constants, and chlorine quadrupole coupling constantssare fitted using Pickett’s nonlinear least-squares program,13 SPFIT, and the results are listed in Table 1. The highest J quantum number accessed in the fits is 11 for the 35Cl-containing isotopomer and 10 for the 37Cl species. The highest value for Ka is 2 in each fit. The rms deviations are 1.47 kHz (35Cl) and 1.24 kHz (37Cl), which are commensurate with the measurement uncertainty. Structure Determination. The values of the asymmetry parameter are, respectively, -0.975 and -0.976 for the most abundant and 37Cl-containing isotopomers; thus, the CF2CHFHCl complex is a near prolate asymmetric top. The values of the inertial defect are slightly negative (-0.667 and -0.665

1,1,2-Trifluoroethylene-Hydrogen Chloride

J. Phys. Chem. A, Vol. 114, No. 41, 2010 10977

TABLE 1: Spectroscopic Constantsa,b (in MHz, Except As Otherwise Noted) for Two Isotopomers of CF2CHF-HCl A B C DJ/10-3 DJK/10-3 DK/10-3 d1/10-6 d2/10-6 χaac χbbc χccc χabd No. of rot. transitions No. of a type No. of b type No. of hyperfine components highest J highest Ka rms/kHz

CF2CHF-H35Cl

CF2CHF-H37Cl

6869.44797(21) 778.052627(32) 699.539068(23) 0.67312(17) -9.8890(21) 185.842(42) -98.463(77) -7.249(58) 0.1206(16) -26.7947(13) 26.6740(11) 37.967(39) 63 26 37 263 11 2 1.47

6857.91549(30) 754.052785(52) 679.962719(29) 0.63971(30) -9.6570(42) 185.383(55) -90.88(16) -6.53(13) -0.0508(23) -20.9812(14) 21.0320(17) 29.951(64) 37 15 22 141 10 2 1.24

a S-reduction, Ir-representation.12 b 1σ standard deviations in the parameters are given in parentheses. c The values of these nuclear quadrupole coupling constants are fitted as 1.5χaa and (χbb - χcc)/4 and the values of the individual constants are then calculated using the Laplace condition. d Even though only the magnitude of χab can be determined, the structure of the complex allows the sign to be established.

amu Å2, respectively, for the two isotopomers), suggesting that the complex is planar and exhibits the effects of out-of-plane vibration. Assuming the geometries of CF2CHF7 and HCl14 remain unchanged upon complexation, three parameters are required to describe the geometry of the complex: a distance between the two subunits and the angular orientation of each subunit. For a planar complex, only two of three moments of inertia are independent for each isotopomer. Thus, we have four moments of inertia from the two isotopomers at our disposal but these proved not to be sufficient to determine the structure of the complex. We therefore extract the orientation of the HCl subunit from the nuclear quadrupole coupling constants. The complete chlorine nuclear quadrupole coupling tensor in the inertial axis system of the complex has been determined experimentally: the three diagonal components and χab from spectroscopic data, while the remaining nondiagonal components are 0 because the complex is planar. Although the absence of a first-order effect and entering into energy level expressions at the second-order often makes it impossible to determine χab (and other off-diagonal elements) from spectroscopic data alone, there are several near degeneracies between rotational levels connected by the appropriate matrix elements to have a significant effect on the observed spectrum for this complex. For example, in the absence of quadrupole splitting, the 313 and 404 levels in CF2CHF-H35Cl differ only by 4.287 MHz. Given the typically small values of rotational constants appropriate to a weakly bound complex, the relatively large value of χab, and the precision of microwave spectroscopy, many transitions show a small, but significant, dependence on the value of this constant, and fits that do not include this constant show a global increase in residuals. Diagonlization of the tensor so determined in the inertial axis system of the complex gives the tensor in its principal electric field gradient axis system, x, y, z, with the values of χxx, χyy, and χzz determined to be 26.9445, 26.6740, and -53.6185 MHz, respectively, for the 35Cl-containing species and 21.2107, 21.0320, and -42.2427 MHz for 37Cl. The x- and

Figure 2. The structure of 1,1,2-trifluoroethylene-HCl determined from spectroscopic constants.

Figure 3. The structures of the two isomers of 1,1,2-trifluoroethylene-HCl determined from ab initio calculation.

z-axes are in the molecular plane, with the z-axis taken to be that of the quadrupole coupling constant of the greatest magnitude. The values so obtained for these quadrupole coupling constants represent averages over the zero-point motion of the HCl in the complex about its equilibrium position, which is the z-axis in the absence of electric field gradient perturbation.15 As a result, in principle they depend both on the extent of electric field gradient perturbation and on the zero-point motion of the complex. It is argued below that the electric field gradient perturbation effects, if any, are negligible for this complex. The diagonalization of the quadrupole coupling tensor for CF2CHF-H35Cl yields an angle of 54.76° between the a inertial axis and the z electric field gradient axis. Fixing the angle between HCl and the a-axis to this value, we determine the remaining two structural parameters by a fit to Ia and Ic of each isotopomer using Schwendeman’s STRFTQ program.16 Although we find that different choices for the specific structural parameters used in the fits give the same geometry for the complex, some choices lead to correlation of the parameters. A choice that leads to no correlation is the distance of the line connecting Cl and the center of the CdC bond and the angle this line makes with the CdC bond. The rms deviation of the fit is 0.0448 amu Å2. Given values for these two quantities, it is possible to determine more chemically relevant geometric parameters, which are shown in Figure 2. Ab Initio Calculations. Ab initio calculations using Gaussian 0317 are carried out at the MP2/6-311++G(2d,2p) level to examine two possible configurations of the CF2CHF-HCl complex: HCl binding across versus at one end of the double bond of CF2CHF. As with the experimental analysis, the structures of the monomers are held fixed at their literature values.7,14 The two optimized structures are shown in Figure 3. It is interesting to note that the primary hydrogen bond length is 0.18 Å shorter, and therefore stronger, when HCl is bonded at the C-2 end of CF2CHF than when it is bonded across the double bond. The hydrogen bond in the side bonding configuration deviates slightly less from linearity (30 vs 34°), and a longer (by 0.11 Å), and hence weaker, secondary interaction is formed. This is consistent with the stronger hydrogen bonding interaction in this “side-bonding” configuration compared to the

10978

J. Phys. Chem. A, Vol. 114, No. 41, 2010

Leung et al.

TABLE 2: Rotational Constants of Two Isomers of CF2CHF-H35Cl Calculated Using Gaussian 03 at the MP2/6-311++G(2d,2p) Level HCl at side of double bond

HCl across the double bond

7253 770 696

4076 1012 811

A/MHz B/MHz C/MHz

“top-bonding.” Additionally, the CF · · · H angle is 14° smaller in the side bonding configuration. The side bonding configuration, with a counterpoise corrected18 binding energy of 670 cm-1 (8.02 kJ mol-1), is 201 cm-1 lower in energy than that with HCl across the double bond. Thus, the energies and structures of these two configurations suggest that the stronger primary interaction is the most significant factor in determining the stability of the complex. The rotational constants for these two isomers are listed in Table 2 and indeed, the ab initio value for the A constant agrees with the experimental value to 5.5%, and those for the B and C constants are within 1% of the corresponding experimental values. Discussion The validity of the use of the chlorine nuclear quadrupole coupling tensor to supply structural information depends on the assumption that there is no measurable electric field gradient perturbation in the HCl subunit upon complexation. Here we justify this assumption. The absolute value of the asymmetry parameter for the nuclear quadrupole coupling tensor expressed in its principal axis system, |η| ) |(χxx - χyy)/χzz|, is 0.0050 for CF2CHF-H35Cl and 0.0042 for CF2CHF-H37Cl, indicating that there is only a very small deviation from a cylindrical charge distribution about the chlorine nucleus. This is likely to have arisen from the zero point motions of HCl and indicates the absence of significant electric field gradient perturbation. The values of χxx, χyy, and χzz determined in the complex are averaged over the motion of the HCl subunit, making them smaller than those of the respective components for the free HCl monomer, -χmon/2, -χmon/2, and χmon. Because the z-axis corresponds to the equilibrium position of the HCl subunit, the value determined for χzz is a tensor projection of the monomer value,

χzz )





3 cos2 θz - 1 χmon 2

where θz is the angle between the instantaneous position of the HCl molecular axis and its equilibrium location. Thus, a comparison of the values for χzz and χmon provides a measure of the extent of the HCl motion about its equilibrium position in the complex. In CF2CHF-H35Cl and using the value (and 1σ uncertainty limits) of -67.60916(10) MHz for χmon,19 the value of θz is calculated to be 21.80°. Similarly, a comparison between the values of χyy and χmon gives 74.62° as the average value for the angle θy between the electric field gradient y-axis and the H35Cl molecular axis. Since the y-axis is normal to the plane of the molecule, the average out-of-plane angle for HCl in the complex is about 15°. Given the structure determined for the CF2CHF-HCl complex, a counterclockwise rotation of 35.24° would be required to bring the b inertial axis of the complex into coincidence with the HCl axis for the 35Cl-containing isotopomer. The direction of rotation and the values for χxx, χyy, and χzz are consistent with a positive χab value for the isotopomer. Similar analysis shows

Figure 4. A comparison of the structures of analogous complexes with HCl: (a) vinyl fluoride-HCl (refs 2 and 22); (b) 1,1-difluoroethylene-HCl, redetermined in ref 4 using spectroscopic constants from ref 5.

that the sign of the χab value for the 37Cl-containing isotopomer is also positive. We have obtained useful ab initio results for quadrupole coupling constants even at this modest level of theory in the case of monomeric species.20 Because of the small to vanishing effect of complexation on electron distributions and the strong dependence of the measured values for these parameters on complex geometry via the projection formulas, our experience is that ab initio predictions of quadrupole coupling constants for weakly bound species are equivalent to the much easier procedure of rotating the tensor calculated for the monomer into the inertial axis system of the complex. Conversely, we find that such constants predicted in an ab initio calculation for a complex need to be corrected for differences between the theoretically predicted geometry and that observed experimentally. Of course, vibrational averaging over the large amplitude motions associated with the intermolecular modes will also have an effect. An example of the effects of inertial axis rotation can be seen in the values of χaa for CHFCF2-H35Cl and CHFCF2-H37Cl reported in Table 1. The HCl molecular axis is very nearly at the “magic” angle (54.74°) to the a inertial axis of the complex so that the P2 projection is very nearly zero and the derivative ∂χaa/∂θa is near its maximum value. As a consequence of the inertial axis rotation caused by the isotopic substitution, the values for χaa in the two isotopomers are nearly zero and have different signs. Thus, the ratios of corresponding tensor components in the inertial axis systems for the two isotopomers should not be expected to reflect that of the nuclear quadrupole moments of 35Cl and 37Cl [1.26889(3)].21 Even when expressed in the principal axis system the effects of vibrational averaging discussed above may be expected to cause small deviations between the ratios of the quadrupole coupling constants and the nuclear quadrupole moments for the two chlorine isotopes. Ab initio calculation of the structure of the lowest energy isomer agrees well with the experimental structure, with the experimental and theoretical values of the CF · · · H angle being within 1.2° of each other, and those for the secondary interaction within 0.1 Å. The length of the primary bond in the theoretical structure is 0.23 Å shorter than that determined for the experimental structure. This is not unexpected since not only do MP2 calculations typically overestimate the strength of intermolecular interactions, but we are also comparing a theoretical, equilibrium bond length with an experimental, average bond length, which is typically longer. The shorter hydrogen bond in the theoretical structure combined with the absence of the effects of zero-point motion, particularly the angular motion of the HCl subunit, also contributes to a smaller deviation from

1,1,2-Trifluoroethylene-Hydrogen Chloride

J. Phys. Chem. A, Vol. 114, No. 41, 2010 10979

Figure 5. A comparison of the structures of analogous complexes of 1,1,2-trifluoroethylene: (a) 1,1,2-trifluoroethylene-HF (ref 7); (b) 1,1,2trifluoroethylene-HCCH (ref 8).

linearity, or alternatively, a larger F · · · HCl angle as compared with the experimental result. The effect of an increase in fluorine substitution on intermolecular interactions is revealed through a comparison of the structure of the 1,1,2-trifluoroethylene-HCl complex in Figure 2 with those of its vinyl fluoride2,22 and 1,1-difluoroethylene5 (reanalyzed in ref 4) counterparts, as shown in Figure 4. As the number of fluorine substitutents increases from 1 to 3 in vinyl fluoride to 1,1-difluoroethylene to 1,1,2-trifluoroethylene, the primary hydrogen bond becomes longer, 2.123(1), 2.33094(36), and 2.3416(7) Å, respectively. Additionally, the longer hydrogen bond bends more from linearity, specifically, 18.3(1), 34.22, and 47.729(13)°, respectively, for these complexes. This trend is consistent with a decrease in nucleophilicity of the hydrogen bonded F atom as the number of F substituents increases. Note that the very slight increase in bond length between the 1,1difluoroethylene-HCl and 1,1,2-trifluoroethylene-HCl complexes is also observed in the analogous HF complexes. The primary bond lengths for these complexes are 1.98833(44) Å 4 and 2.020(41) Å,7 respectively. As observed in the analogous complexes with HF,4 there is a remarkable consistency in the length of the three secondary interactions of these three fluoroethlyenes with HCl. The individual values differ from their mean by 1.8%, suggesting that 3.08-3.16 Å is the optimal distance for this secondary interaction and that the HCl will bend sufficiently to achieve this distance. The correlation between primary bond strength and hydrogen bond donor ability becomes apparent when the structures of 1,1,2-trifluoroethylene complexed with HF, HCl, and HCCH are compared (Figures 2 and 5). In progressing from HF to HCl to HCCH, the hydrogen bond length increases from 2.020(41) to 2.3416(7) to 2.748(15) Å and additionally bends more from linearity: 41.6(15) to 47.729(13) to 69.24(67)°. It is notable that the values of the CF · · · H angle for all three 1,1,2-trifluoroethylene-containing complexes are similar and lie between 104.5 and 109.7°, suggesting that indeed, for a particular fluorine substituted ethylene, the electron density that interacts with an acid depends only on the ethylene species and not on the acid. Conclusions The rotational spectra of two isotopomers of 1,1,2-trifluoroethylene-hydrogen chloride complex have been observed and analyzed. The complete nuclear quadrupole hyperfine coupling tensor in the inertial axis system of the complex is determined and transformed into the principal electric field gradient system, making possible the determination of the orientation of the HCl subunit in the complex. This information combined with a fit to moments of inertia allows a determination of chemically

relevant structural parameters for this species. The complex is formed via a primary interaction in which the hydrogen chloride acts as a hydrogen bond donor and the fluorine atom on carbon 2 of 1,1,2-trifluoroethylene serves as the acceptor. The hydrogen bond deviates from linearity to allow a secondary interaction between the chlorine atom of HCl and the hydrogen atom geminal to the hydrogen bond accepting fluorine atom. This structure with the HCl at one end of the ethylenic moiety contrasts with the “top binding” complexes formed with vinyl fluoride and 1,1-difluoroethylene where the primary and secondary interactions involve atoms cis to each other, but is in accord with other protic acid (HF, HCCH) complexes of 1,1,2-trifluoroethylene. For the 1,1-difluoroethylene complexes, the top binding arrangement is the only possibility that allows the formation of a secondary interaction and the additional stability that provides. Both top binding and side binding are possible for complexes of protic acids with vinyl fluoride and 1,1,2difluoroethylene. For the vinyl fluoride complexes, steric effects render the top binding configuration more stable in spite of more favorable electrostatic interactions for the primary and secondary interactions individually with side binding. In complexes with 1,1,2-trifluoroethylene, unfavorable steric factors on the side of the ethylene are outweighed by the increased electrostatic stability afforded by hydrogen bond formation with the fluorine atom at carbon 2. The length of the hydrogen bond and the accompanying deviation from linearity in 1,1,2-trifluoroethylene-protic acid complexes show a clear correlation with protic acid identity, with both structural features increasing along the series HX, X ) F, Cl, CCH. As observed in the complexes of hydrogen fluoride and acetylene with fluorine-substituted ethylenes,1,3,4,6-8 complexes with hydrogen chloride show the same increase in both hydrogen bond length and deviation from linearity as the number of fluorine atoms on the ethylene subunit increases, a result attributed to an accompanying decrease in the nucleophilicity of each F atom. Additionally, as also observed in the complexes with hydrogen fluoride, the secondary interaction length in CH2CHF-HCl,2 CH2CF2-HCl,4,5 and CF2CHF-HCl (this work) displays a remarkable consistency among the complexes despite the two different modes of bonding observed, with the individual values varying less than 2% from their mean. Acknowledgment. This material is based on work supported by the National Science Foundation under Grant No. CHE0809542. MRR thanks the Arnold and Mabel Beckman Foundation for support. Supporting Information Available: Tables of observed and calculated transition frequencies for all isotopomers of 1,1,2-

10980

J. Phys. Chem. A, Vol. 114, No. 41, 2010

trifluoroethylene-hydrogen chloride in this study are available as Supporting Information. Cartesian coordinates in the respective principal inertial axis systems predicted by ab initio calculation for the two structures displayed in Figure 3 are also provided. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Cole, G. C.; Legon, A. C. Chem. Phys. Lett. 2004, 400, 419–424. (2) Kisiel, Z.; Fowler, P. W.; Legon, A. C. J. Chem. Phys. 1990, 93, 3054–3062. (3) Cole, G. C.; Legon, A. C. Chem. Phys. Lett. 2003, 369, 31–40. (4) Leung, H. O.; Marshall, M. D.; Drake, T. L.; Pudlik, T.; Savji, N.; McCune, D. W. J. Chem. Phys. 2009, 131, 204301. (5) Kisiel, Z.; Fowler, P. W.; Legon, A. C. J. Chem. Soc. Faraday Trans. 1992, 88, 3385–3391. (6) Leung, H. O.; Marshall, M. D. J. Chem. Phys. 2006, 125, 154301. (7) Leung, H. O.; Marshall, M. D. J. Chem. Phys. 2007, 126, 114310. (8) Leung, H. O.; Marshall, M. D.; Cashion, W. T.; Chen, V. L. J. Chem. Phys. 2007, 128, 064315. (9) Leung, H. O.; Marshall, M. D.; Amberger, B. K. J. Chem. Phys. 2009, 131, 204302. (10) Cole, G. C.; Hughes, R. A.; Legon, A. C. J. Chem. Phys. 2005, 122, 134311. (11) Leung, H. O.; Gangwani, D.; Grabow, J. U. J. Mol. Spectrosc. 1997, 184, 106–112. (12) Watson, J. K. G. Aspects of Quartic and Sextic Centrifugal Effects on Rotational Energy Levels. In: Vibrational Spectra and Structure; Durig, J. R., Ed.; Elsevier Scientific Publishing: Amsterdam, 1977; pp 1-89. (13) Pickett, H. M. J. Mol. Spectrosc. 1991, 148, 371–377.

Leung et al. (14) DeLucia, F. C.; Helminger, P.; Gordy, W. Phys. ReV. A. 1971, 3, 1849–1857. (15) Legon, A. C.; Thorn, J. C. Chem. Phys. Lett. 1994, 227, 472–479. (16) Schwendeman, R. H. In: Critical EValuation of Chemical and Physical Structural Information; Lide, D. R., Paul, M. A. , Eds.; National Academy of Science: Washington, DC, 1974. (17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (18) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553–566. (19) Klaus, T.; Belov, S. P.; Winnewisser, G. J. Mol. Spectrosc. 1998, 187, 109–117. (20) Leung, H. O.; Marshall, M. D.; Vasta, A. L.; Craig, N. C. J. Mol. Spectrosc. 2009, 253, 116–121. (21) Legon, A. C.; Thorn, J. C. Chem. Phys. Lett. 1993, 215, 554–560. (22) Legon, A. C.; Ottaviani, P. Phys. Chem. Chem. Phys. 2002, 4, 4103–4108.

JP107494T