Structure and Torsional Properties of Oxalyl ... - ACS Publications

ARTICLE pubs.acs.org/JPCA

Structure and Torsional Properties of Oxalyl Chloride Fluoride in the Gas Phase: An Electron-Diffraction Investigation Dwayne T. Friesen,† Robert J. G. Johnson, Lise Hedberg, and Kenneth Hedberg* Department of Chemistry, Oregon State University, Corvallis, Oregon 97331-4003, United States

bS Supporting Information ABSTRACT: The structure and torsional properties of oxalyl chloride fluoride in the gas phase have been measured by electron diffraction at temperatures of 22, 81, 158, and 310 C. The molecule may be regarded as a hybrid of oxalyl chloride and oxalyl fluoride. Since the former exists as a more stable periplanar anti form (φ = 180) in equilibrium with a less stable gauche form (φ = 60) and the latter as an equilibrium between two periplanar forms, anti and syn, the second form of oxalyl chloride fluoride is an interesting question. It was found to be gauche. The system was modeled as two rotational conformers related by a potential of the form 2V = V1(1 þ cos φ)
V2(1
cos 2φ) þ V3(1 þ cos 3φ). The anti/gauche bond distances and bond angles (rg/Angstroms, — R/degrees) with estimated 2σ uncertainties at 22 C are Ær(CdO)æ = 1.183(2)/1.182(2), Δr(CdO) = 0.0036/0.0026 (assumed from theory), r(C
F) = 1.329(3)/1.335(3), r(C
Cl) = 1.738(2)/1.753(2), — (C
C
Cl) = 112.0(3)/111.9(3), — (C
CdO3) = 123.0(4)/123.2(4), — (OdC
Cl) = 125.0(2)/ 1.249(2), — (OdC
F) = 123.0(3)/125.1(3), and — (Cl
C
C
F) = 180.0/59.8. The variation of composition with temperature afforded a determination of the standard enthalpy and entropy of the reaction anti f gauche. The results are ΔH = 2.5(12) kcal/mol and ΔS =
6.5(33) cal/(mol 3 K). The structures and equilibria are discussed.

’ INTRODUCTION The oxalyl halides (C2O2X2, X = F, Cl, Br) in the gas phase are composed of two easily interconvertible conformers generated by rotation around the carbon
carbon bond

The more stable form in each is periplanar anti ( — XCCX = 180, C2h symmetry), and the less stable form, after some uncertainties derived from early work, is now known to be gauche ( — XCCX ≈ 60
70, C2 symmetry) in C2O2Br21 and C2O2Cl22 (OxCl) and periplanar syn ( — XCCX = 0, C2v symmetry) in C2O2F23 (OxF). The reasons for the differing forms of the lower energy conformers are discussed in more detail at the end of this article, but in summary, the differences result from a competition between steric repulsion (O 3 3 3 O and X 3 3 3 X) and conjugation stabilization (or the anomeric effect), the latter of which operates more strongly in the difluoride than in the other molecules. In any case, these results suggest an immediate question about the structure of the hybrid molecule oxalyl chloride fluoride (OxClF), i.e., given that the likely lower energy conformer is periplanar anti, what is the form of the higher energy conformer where the above-mentioned effects are presumably intermediate between those operating in the difluoride and those operating in the dichloride? Related to this question is the matter of the nature of the torsional potential—are the barriers low enough to generate large amplitude torsional r 2011 American Chemical Society

modes for the two conformers? The first of these questions was answered4 by one of us (D.T.F.) some time ago (the second form is gauche, — XCCY ≈ 60, Figure 1), but the work was not published. We recently decided to reanalyze DTF’s data, making use of some theoretical results not accessible at that time but which in principle should increase the quality of the model. In the original work the framework structure of the system (bond lengths and bond angles) was assumed to be invariant with torsion around the C
C bond; reanalysis removes this constraint and brings the model into closer agreement with those used for the difluoride and dichloride.

’ EXPERIMENTAL SECTION The sample used in the diffraction experiments was prepared by a method similar to that outlined by Tullock and Coffman5 and modified by Hencher and King.6 It was further modified to increase the yield by slowly adding the sodium fluoride and tetrahydrothiophene-1,1-dioxide mixture to the oxalyl chloride, allowing the OxClF to distill directly into a cold trap as formed. The sample was then purified by fractional distillation. The appearance of the IR spectra showed no impurities in the final product. The diffraction experiments were made using the Oregon State apparatus with the following conditions: r3 sector; 8  10 in Kodak lantern slide, medium-contrast plates developed in D-19 diluted 1:1 for 10 min; exposure times, 40
240 s; beam currents, Received: April 1, 2011 Revised: April 21, 2011 Published: May 26, 2011 6702

dx.doi.org/10.1021/jp203060d | J. Phys. Chem. A 2011, 115, 6702–6708

The Journal of Physical Chemistry A

ARTICLE

Figure 1. Conformations and atom numbering of oxalyl chloride fluoride.

Figure 3. Radial distribution curve. The curves were calculated from composites of the experimental intensity data multiplied by (ZCZO/ FCFO)exp(
0.002s2) where Fi is an electron scattering amplitude. The vertical lines have lengths proportional to the weights of the scattering terms.

was found to be necessary to obtain good fits to the long camera data which contains more information on the torsion-sensitive distances. Averaged intensity and radial distribution curves for all temperatures are shown in Figures 2 and 3. The intensity data for all temperatures are found in the Supporting Information.

’ STRUCTURAL ANALYSIS

Figure 2. Curves of intensity data. Shorter curves are from the “long camera” and longer curves from the “middle camera”. Difference curves are experimental minus theoretical for the final model.

0.40
0.48 μA; nominal electron wavelength, 0.056
0.058 Å calibrated against CO2 in separate experiments assuming ra(CO) = 1.1646 Å and ra(OO) = 2.3244 Å; ambient apparatus pressure during sample run-in, 5.0
7.2  10
6 Torr. Experiments were carried out with nozzle temperatures of 22, 81, 158, and 310 C with nominal nozzle-to-plate distances of 75 cm (long camera, or LC) and 30 cm (middle camera, MC). Two to four plates were selected for analysis from the experiments at each temperature and each distance. Data reduction and generation of radial distribution curves were carried out in the same fashion as in the case of oxalyl fluoride3 except that the LC data were weighted twice as heavily as the MC data instead of the equal weighting used in oxalyl fluoride. The deviation from the normal weighting

Theoretical Calculations. Geometry optimizations of OxClF were carried out using the Gaussian 03 program package.7 Previous experience has shown a satisfactory combination for our work to be the B3LYP level of theory with the cc-pVTZ basis set. Structure optimizations with this combination were carried out and quadratic force fields evaluated for the two stable forms, anti (180) and gauche (ca. 60), as well as for a set of “pseudoconformers” specified by fixed values of the torsion angle — (F
C
C
Cl) over the range 0
180 in increments of 10. Since the optimized parameter values contained no representation of the thermal effects present in the experimental data, it was necessary to estimate these effects separately. This was done via normal coordinate analyses carried out with the program ASYM40,8 making use of the quadratic force fields for each of the 19 structures resulting from the optimizations but removing in each case the normal mode that consisted almost entirely of the torsional symmetry coordinate.9 These calculations provided estimates of the changes in the distances, angles, and vibrational amplitudes that occur as a consequence of changes in torsion 6703

dx.doi.org/10.1021/jp203060d |J. Phys. Chem. A 2011, 115, 6702–6708

The Journal of Physical Chemistry A

ARTICLE

substantially inferior to the latter two and in consequence was ignored. The 3CP gave slightly better fits to the experimental data than the 2S model, but the derived parameter values differed insignificantly. In view of the earlier results, we decided to base our reinvestigation of OxClF on the 3CP model. The form of the potential was identical to that above, but instead of D.T.F.’s choice of the trans form of the torsion angle Cl
C
C
F as equal to zero, we chose the zero of this angle as the cis form in keeping with current convention. Thus, φ þ 180 = φ(DTF) and our potential becomes 2V ðφÞ ¼ V 1 ½1
cosðφ þ 180Þ þ V 2 ½1
cos 2ðφ þ 180Þ þ V 3 ½1
cos 3ðφ þ 180Þ

ð1Þ

The model itself was similar to D.T.F.’s in that it consisted of 19 pseudoconformers defined by the torsion angle φ with values at 10 intervals over the range 0
180. The important difference between the present model and D.F.T.’s is the inclusion of the changes in the bond distances and bond angles of the pseudoconformers predicted from our molecular orbital calculations as a consequence of change in torsion angle. The periplanar anti form (φ = 180), which has the lowest energy, was chosen as the basic structure to which the structures of the others were tied by the calculated parameter differences. Each pseudoconformer was given a Boltzmann weight determined according to Wðφi Þ ¼ ½expð
V ðφi Þ=RT=

Figure 4. Calculated (B3LYP/cc-pVTZ) variation of bond lengths and bond angles with torsion angle.

angle and vibration of the molecular framework. The changes in the structural parameters are shown in Figure 4. The very many amplitude changes (not shown) are unexceptional. Models. The early work4 on OxClF tested three quite different models of the system. The first, designated the “two-conformer” (2C) model, consisted of a mixture of the anti and gauche forms each of which experienced only harmonic vibration, i.e., the torsional mode was of small amplitude. The second, the “double sigma” (2S) model, was an elaboration of the 2C model obtained by representing the torsional mode for each form as a set of pseudoconformers characterized by Gaussian distributions centered on the minima with shapes determined by refinable standard-deviation parameters. The third model, designated the “three-cosine” or 3C model, included a set of pseudoconformers distributed at regular intervals over the full range of torsion angle, with individual weights determined by a Gaussian distribution based on a potential energy assumed to be a sum of three cosine terms 2V ðφÞ ¼ V 1 ð1
cos φÞ þ V2 ð1
cos 2φÞ þ V3 ð1
cos 3φÞ The torsion angle φ = — (Cl
C
C
F) was taken as equal to zero for the periplanar anti conformation, and the Vi were treated as adjustable parameters.10 The 2C model was found to be

191

∑i¼ exp½
V ðφi Þ=RT

ð2Þ

The structural parameters of the periplanar anti form were the rR distances ÆCdOæ = (C1dO3 þ C2dO5)/2, Δ(CdO) = C1dO3
C2dO5, C
F, C
C, C
Cl and the — R bond angles C
C
Cl, C
CdO3, C
CdO5, and C
C
F. Additional parameters were the three Vi values that determined the torsional potential (eq 1) and several groups of vibrational amplitudes. As usual, the members of each amplitude group were tied together by the calculated differences as mentioned above. These may be seen by comparison of the values listed in Table 2. Refinement Results. Converged refinements of all structural parameters except Δ(CdO) were obtained for the experiments at each of the four temperatures; Δ(CdO) was fixed at the theoretical values. Refined values of the three potential constants were also obtained with the single exception of V3 for the 321 C experiment; it was given the refined value obtained for the 22 C experiment.11 The torsional potential reveals (by procedures discussed below) that the second form of OxClF is without doubt gauche, as previously found by D.T.F.4 The results for the four temperatures are shown in Table 1. Table 2 contains the distances and amplitudes for the anti and gauche forms obtained at 22 C; the values for the other temperatures are found in the Supporting Information. The correlation matrix for the model parameters obtained at 22 C is presented in Table 3. Correlation matrices for the other experiments are similar.

’ DISCUSSION Structure. The most interesting question about the structure of OxClF, a hybrid of OxCl and OxF, is the nature of the less stable of the two rotamers known to exist in the vapor—whether it is gauche as in OxCl or periplanar syn as in OxF—and why that form is preferred. Evidence that the less stable form is gauche is found in the distribution of pseudoconformers corresponding to 6704

dx.doi.org/10.1021/jp203060d |J. Phys. Chem. A 2011, 115, 6702–6708

The Journal of Physical Chemistry A

ARTICLE

Table 1. Temperature Dependency of Parameter Values (rR/Angstroms; — R/degrees) for Anti and Gauche Forms of Oxalyl Chloride Fluoridea 22 C parameter

anti

81 C gauche

anti

158 C gauche

anti

310 C

gauche

anti

gauche

theoryb anti

gauche

Ær(CdO)æ

1.183(2)

1.182(2)

1.180(2)

1.179(2)

1.178(2)

1.177(2)

1.178(2)

1.177(2)

1.178

1.178

Δr(CdO) r(C
F)

[0.0036] 1.329(3)

[0.0026] 1.335(3)

[0.0036] 1.331(5)

[0.0026] 1.337(5)

[0.0036] 1.329(5)

[0.0026] 1.335(5)

[0.0036] 1.325(4)

[0.0026] 1.331(4)

0.0036 1.333

0.0026 1.339

r(C
C)

1.543(4)

1.527(4)

1.546(6)

1.530(6)

1.551(6)

1.535(6)

1.546(6)

1.529(6)

1.544

1.527

r(C
Cl)

1.738(2)

1.753(2)

1.737(3)

1.752(3)

1.739(3)

1.754(3)

1.736(3)

1.751(3)

1.764

— (C
C
Cl)

1.780

112.0(3)

111.9(3)

112.1(4)

112.0(4)

111.4(6)

111.3(6)

111.1(9)

111.0(9)

112.1

111.8 124.1

— (C
CdO3)

123.0(4)

123.2(4)

123.0(7)

123.2(7)

123.3(7)

123.5(7)

123.2(7)

123.3(7)

123.8

— (OdC
Cl)

125.0(2)

124.9(4)

124.9(6)

124.4(6)

124.6(7)

124.6(5)

125.6(9)

125.6(9)

124.1

124.1

— (C
CdO5)

126.9(5)

124.9(5)

126.6(13)

124.7(13)

126.8(12)

124.9(12)

124.8(25)

122.9(25)

127.1

125.4

— (C
C
F) — (OdC
F)

110.1(3) 123.1(3)

109.9(3) 125.1(3)

111.1(12) 122.3(6)

111.0(12) 124.3(6)

110.6(11) 122.6(6)

110.5(11) 124.7(6)

112.7(23) 122.5(5)

112.6(23) 124.6(5)

109.3 123.5

110.9 123.7

— (Cl
C
C
F)c

[180.0]

— (OdC
C
F)

60.0 {59.8}

[0.0]

122.8

[180.0]

60.0 {57.6}

[0.0]

122.8

[180.0]

60.0 {58.5}

[0.0]

122.8

[180.0]

60.0 {66.2}

[0.0]

122.8

180.0

63.0

0.0

118.9

— (OdC
C
Cl)

[0.0]

120.9

[0.0]

120.9

[0.0]

120.9

[0.0]

120.9

0.0

117.9

— (OdC
CdO)

[180.0]

56.4

[180.0]

56.4

[180.0]

56.4

[180.0]

56.4

180.0

60.2

V1d

1.05(30)

0.32(30)

0.30(38)


0.089(31)

V2d

0.53(17)

0.21(14)

0.22(19)


0.32(21)

V3d Re

0.40(31) 0.056

0.29(50) 0.094

0.58(70) 0.089

[0.40] 0.073

a Reference 4. b B3LYP/cc-pVTZ. c Bracketed items for gauche are angles obtained from potential minima. d Units are kilocalories/mol. e R = [∑iwiΔi2/ ∑iwi(siIm,i(obsd))2]1/2, where Δi = siIm,i(obsd)
siIm,i(calcd).

Table 2. Distances and Amplitudes (r, l/Angstroms) for Oxalyl Chloride Fluoride at 22Ca gaucheb

anti rg

rR

ra

rg

/

1.185 (2)

1.187

1.188

0.046

(2)

1.183 (2)

1.186

1.187

0.045

1.181 (2)

1.183

1.185

0.045

1.181 (2)

1.183

1.185

0.045

C
F C
C

1.329 (3) 1.543 (4)

1.331 1.543

1.333 1.546

0.054 0.062

1.336 1.527

1.339 1.529

0.055 0.061

C
Cl

1.738 (2)

1.739

1.741

0.055

V (3)

1.335 (3) 1.527 (4) 1.753 (2)

1.755

1.756

0.057

V (3)

C1 3 O5

2.442 (7)

2.443

2.444

0.063

(4)

2.406 (7)

2.408

2.409

0.067

(4)

C2 3 O3

2.404 (6)

2.404

2.406

0.063

2.390 (6)

2.390

2.392

0.06

O5 3 F C1 3 F

2.208 (4)

2.209

2.210

0.054

2.233 (4)

2.235

2.236

0.054

2.357 (7)

2.358

2.360

0.067

2.347

2.349

0.070

O3 3 Cl

2.605 (5)

2.205

2.606

0.067

V (5)

2.346 (7) 2.616 (5)

2.617

2.618

0.068

C2 3 Cl O3 3 3 F

2.722 (6) 2.658 (12)

2.722 2.655

2.724 2.659

0.078 0.104

2.720 (6) 3.289 (8)

2.720 3.289

2.722 3.291

0.084 0.091

O3 3O O5 3 3 Cl

3.486 (6)

3.487

3.488

0.057

2.998 (8)

2.996

2.999

0.090

V (10)

2.978 (14)

2.976

2.980

0.101

(15)

3.654 (8)

3.654

3.656

0.085

(15)

F 3 3 Cl

3.899 (7)

3.899

3.900

0.069

(7)

3.020 (8)

3.019

3.023

0.114

(8)

s

V (5)

s

V (10)

(3) s

CdO3 CdO5

s

/

s

a

ra

s

rR

Dashes indicate bonded atoms; dots indicate atoms separated by one or two bond angles. Quantities in parentheses are estimated 2σ and include estimates of systematic error. The vertical arrows indicate amplitude groupings. b Data taken from the pseudoconformer with — Cl
C
C
F = 60. See Figure 5.

the potential functions. These distributions, calculated by weighting each pseudoconformer according to eq 2, are plotted in Figure 5 and show two maxima for each experimental temperature, one corresponding to the periplanar anti form at 180 and one at about 60 corresponding to a gauche form. The figure also shows, as expected, that the amount of gauche increases at the

expense of anti with an increase in temperature, as does Figure 3, where the decrease in the anti form is seen in the diminishing areas of the radial distribution peaks at about 3.9 Å. As mentioned in the Introduction, the torsional picture in the oxalyl halides may be understood in terms of competition between bond conjugation tending to force planarity and steric 6705

dx.doi.org/10.1021/jp203060d |J. Phys. Chem. A 2011, 115, 6702–6708

The Journal of Physical Chemistry A

ARTICLE

Table 3. Correlation Matrix (100) for Parameters of Oxalyl Chloride Fluoride at 22C σLSa

r1

Ær(CdO)æ

0.03

100

r(C
F)

0.08

59

r(C
C)

0.12

30

35

100

r(C
Cl)

0.05


22


19


16

100

— (C
C
Cl)

9.27

19

12

6


2

r2

r3

r4

—1

—2

—3

—4

V1

100

V2

V3


12


16


20

39


47

100

— (C
CdO5)

17.9

30

48

8


20

22


50

100

— (C
C
F) V1

12.4 15.2


2
39


10
50


7
42


4 38


19
18

30 40


74
73

100 42

27

38

8

6

48


9

41


44


28

100


24


31


22


20


23


26

14


12


23


59

100

V3

15.7

l3

l4

l5

l6

l7

100

14.4

8.58

l2

100

— (C
CdO3)

V2

l1

l(CdO)

0.05

12

29

20

4


8


4

19


1


25


1


2

100

l(C
Cl)

0.06


0.1

5


19

6


0.7

12

6

1

0.2

3


4

40

l(C 3 O5)b

0.09

1.1

8

16


10


29


16


22

51

4


35

7

21


9

100

0.17


40


41


57

2


31.2


24


3


2

41


39

44


7

3

22

100

0.34 0.51


20
36


20
45


34
47

15
5

13
30

4 2


14
12


1
0.3

21 18

16
55

12 79


12
13

1 1


14
2

19 56

100 24

100

0.24


22


21


41

20

21


2


7


14

28

31


2


13

5


20

27

41

14

l(O3 3 Cl)b

l(O 3 3 O)b l(O5 3 3 Cl)b l(F 3 3 Cl)b

100

100

Standard deviations (100) from least-squares refinement. Distances (r) and amplitudes (l) are in Angstroms, and angles ( — ) are in degrees. b Dots indicate the number of bond angles separating the atoms. a

Table 4. Composition (mole fraction) and Thermodynamic Properties of Oxalyl Chloride Fluoride in the Vapor Phase A modela anti

B

C

D (ref 4)

gauche

anti

gauche

anti

gauche

anti

gauche

22 C

0.80

0.20

0.76

0.24

0.73

0.27

0.65

0.35

81 C

0.47

0.53

0.45

0.55

0.45

0.55

0.55

0.45

158 C 0.42

0.58

0.41

0.59

0.40

0.60

0.54

0.45

310 C 0.28

0.72

0.27

0.73

0.20

0.80

0.40

0.60

ΔHb ΔSb

2544(1250)
6.55(330)

2266(1154)
5.98(304)

2652(1074)
7.28(230)

1129(436)
2.61(116)

See text in Discussion section. b For the reaction anti a gauche. ΔH in cal/mol, ΔS in cal/(mol 3 K). Uncertainties are two std’s. a

Figure 5. Calculated distribution of pseudoconformers at different temperatures.

repulsion tending to oppose it. In addition to the contribution to conjugation stabilization of the structure OdC
CdO present in all the molecules, there is contribution from the structure þ XdC

C
dYþ, where halogen atoms X and Y can be the same or different. In OxF such a structure contributes much more strongly than in OxCl where back π-bonding from chlorine atoms is less important. On the other hand, steric repulsion between chlorine atoms in a periplanar syn configuration of OxCl is greater than between fluorine atoms in OxF. The result is that the second form is gauche for the former and periplanar syn for the latter. Applied to OxClF, this argument suggests that the molecule behaves more like OxCl than OxF, i.e., the structure þ FdC

C
dClþ tending to force planarity is less effective than Cl 3 3 F steric repulsion is at preventing it. The quantitative aspect of the existence of the rotamers—the relative amounts of the two forms in each case—is subject to

uncertainty. These depend on the relative areas under a given curve that are assigned to each form. Since each pseudoconformer represents a part of this area, the areas assigned to anti and gauche depend on the number of pseudoconformers chosen to be associated with each maximum. The results of three different approaches to such a decision are given in Table 4, together with a fourth (model D) from our earlier work for comparison. For model A it was assumed that the total area of each conformer is represented by the height of its corresponding peak, for model B the areas were assumed to be determined by the sums of the heights of seven pseudoconformers symmetrically distributed about each peak maximum, and for model C the peak areas were obtained from the sum of all pseudoconformer heights divided into groups by a slightly more elaborate procedure.12 As Table 4 shows, the compositional agreement among models A, B, and C is very good for all experimental temperatures, but the values for model D indicate relatively less of the gauche form at the three highest temperatures. The variation in rotameric composition with temperature affords estimates of the enthalpy and entropy changes accompanying 6706

dx.doi.org/10.1021/jp203060d |J. Phys. Chem. A 2011, 115, 6702–6708

The Journal of Physical Chemistry A

ARTICLE

Figure 6. van’t Hoff plots of the compositional data for models A
D of oxalyl chloride fluoride.

the reaction anti h gauche according to the van’t Hoff equation ln K ¼
ΔGo =RT ¼
ΔH o =RT þ ΔSo =R

ð3Þ

where K = Ng/Na. van’t Hoff plots of the data in Table 4 are shown in Figure 6, from which least-squares fits yielded the tabulated values of the enthalpy and entropy changes for conversion of the anti form to gauche. The three models from the present work yield similar values for these changes, but the values from the earlier work differ from these by amounts at the margin of significance (2σ). Taking into account the imprecision of the tabulated enthalpy and entropy changes for conversion of the anti to the gauche form, the changes are consistent with results for other molecules that undergo similar hindered rotation. It should be noted that the entropy change includes the 2-fold degeneracy of the gauche form; if the entropy accompanying the conversion from the anti form to one of the two equivalent gauche forms is wanted, the quantity R ln 2 must be subtracted from the tabulated value. The 3C model of OxClF on which the present analysis is based provides unambiguous parameter values for the anti form of the molecule; however, the model does not yield similarly accurate values for the parameters of the gauche form. The reason is that calculations are carried out only for the individual pseudoconformers, one of which will correspond only by coincidence to the gauche torsion angle which is determined by the potential. Fortunately, this angle is quite close to 60 except for the experiment at 321 C. Inspection of the parameter changes predicted with torsion angle change (Figure 4) reveals that the differences between those at the actual gauche angle and those at 60 are estimated to be insignificant at the three lower experimental temperatures and less than 0.2 in the angles and 0.002 Å in the distances at the highest temperature. The gauche values listed in the tables may be accepted with confidence. Many of the parameter values for OxClF are nearly the same as the corresponding ones for OxCl2 and OxF.3 For example, the CdO bond distance (rR) in the low-temperature anti forms of the latter pair (respectively, 1.180(2) and 1.178(2) Å) differs very little from the average of 1.183(2) Å we find in OxClF. Certain bond angles in these anti forms are also similar: for example, in OxClF/ OxCl the value of — R(C
C
Cl) equals 112.0(3)/111.8(3)

and — R(C
C(Cl)dO) 123.4(4)/123.8(4). However, noteworthy differences also exist. The C
C distance at 1.533(3) Å in OxF is somewhat smaller than the nearly identical values of 1.543(4) and 1.545(8) Å in OxClF and OxCl, and in OxClF/OxF the angle — R(C
C
F) equals 110.1(3)/112.2(2) and — R(C
C(F)dO) equals 126.9(5)/124.2(2). These similarities and differences, like those discussed above for the torsion angles, may also be accounted for in terms of the effects of conjugation and steric repulsion. Thus, the shorter carbon
carbon bond length in OxF vs OxClF derives from a higher degree of conjugation in the former due to the influence of the structures OdC
CdO and þFdC

C
dFþ which operate more strongly in planar OxF. The angle differences mentioned above find their explanation in the distances they generate between the peripheral fluorine and oxygen atoms: in OxClF the distances F6 3 3 O3 and Cl4 3 3 O5 (Table 2) are each about 0.25 Å shorter than the sum of the van der Waals radii for these atoms, suggesting similar steric repulsion between the two ends of the OxClF molecule. In OxF the difference between the corresponding quantities for the F 3 3 O distance is similar at about 0.20 Å, and the peripheral F 3 O distances in OxClF/OxF are virtually identical at 2.208(4)/2.207(2) Å, suggesting very similar repulsions between the two atoms consistent with the similar F
CdO bond angles.

’ ASSOCIATED CONTENT

bS

Supporting Information. Tables of the molecular scattered intensities; tables of parameter values at the three highest experimental temperatures. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Present Addresses †

Bend Research, Incorporated, 64550 Research Road, Bend, Oregon 97701, United States.

’ ACKNOWLEDGMENT This work was supported by the National Science Foundation under grants CHE78-04258 and CHE-0613298. ’ REFERENCES (1) Hagen, K.; Hedberg, K. J. Am. Chem. Soc. 1973, 95, 4796. (2) (a) Danielson, D. D.; Hedberg, L.; Hedberg, K.; Hagen, K.; Trætteberg, M. J. Phys. Chem. 1995, 99, 9374. (b) Hagen, K.; Hedberg, K. J. Am. Chem. Soc. 1973, 95, 1003. (3) Friesen, D. T.; Borgers, T. R.; Hedberg, L.; Hedberg, K. J. Phys. Chem. A. 2006, 110, 12986. (4) Friesen, D. T. Ph.D. thesis, Oregon State University, 1981. (5) Tullock, C. W.; Coffman, D. D. J. Org. Chem. 1960, 25, 2016. (6) Hencher, J. L.; King, G. W. J. Mol. Spectrosc. 1965, 16, 168. (7) 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.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; 6707

dx.doi.org/10.1021/jp203060d |J. Phys. Chem. A 2011, 115, 6702–6708

The Journal of Physical Chemistry A

ARTICLE

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.; M. A. Al-Laham, 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 E.01; Gaussian, Inc.: Wallingford, CT, 2004. (8) (a) Hedberg, L.; Mills, I. M. J. Mol. Spectrosc. 2000, 203, 82. (b) Hedberg, L.; Mills, I. M. J. Mol. Spectrosc. 1993, 160, 117. (9) The vibrational averaging effects (the perpendicular amplitudes and amplitudes of vibration) of this low-frequency torsional mode cannot be realistically assessed by the usual harmonic approximation, which assumes linear atomic displacements. This problem is avoided by representing the actual curvilinear displacements of the atoms with a set of pseudoconformers distributed along the torsional coordinate and calculating the vibrational averaging effect for each with elimination of the contribution of the torsional mode: the sum of these contributions is (conceptionally) expressed in the distribution of the pseudoconformers. See Figure 5. (10) A fuller account of these and other models for large amplitude motion is given in the following: Gas-phase Electron Diffraction Applied to Molecules Undergoing Large Amplitude Motion. Hedberg, K. In Structures and Conformations of Non-Rigid Molecules; Laane, J., et al. , Eds.; Kluwer Academic Publishers: New York, 1993; pp 423
445. (11) The Vi’s have large uncertainties. It seemed that the lowest temperature value was among the most reliable, but in any case the assigned value is within a standard deviation of the others. (12) The area assigned to the gauche form was assumed to be proportional to the doubled sum of the heights of the pseudoconformers in the range 0
φ(gauche)/deg. The area of the periplanar anti form was taken to be the total area minus that of the gauche form.

6708

dx.doi.org/10.1021/jp203060d |J. Phys. Chem. A 2011, 115, 6702–6708