Infrared and Raman spectra, conformations, quantum chemical calculations and spectral assignments of 1-methyl-1-silacyclohexane

Journal of Molecular Structure 1034 (2013) 207–215

Contents lists available at SciVerse ScienceDirect

Journal of Molecular Structure journal homepage: www.elsevier.com/locate/molstruc

Infrared and Raman spectra, conformations, quantum chemical calculations and spectral assignments of 1-methyl-1-silacyclohexane Peter Klaeboe a,⇑, Anne Horn a, Claus J. Nielsen a, Valdemaras Aleksa a,b, Gamil A. Guirgis c, Justin K. Wyatt c, Horace W. Dukes c a b c

CTCC, Department of Chemistry, University of Oslo, P.O. Box 1033, NO-0315 Oslo, Norway Department of General Physics and Spectroscopy, Vilnius University, Vilnius 2734, Lithuania Department of Chemistry & Biochemistry, College of Charleston, Charleston, SC 29403, USA

h i g h l i g h t s " Infrared and Raman spectra of 1-methyl-1-silacyclohexane. " Attribution of equatorial (e) and axial (a) conformers from spectra of the crystal. " Assignments of the A0 and A00 modes of the e and a conformers. " Determination of the enthalpy difference DH between the e and a conformers from variable temperature Raman spectra. " DFT calculations of vibrational modes in the anharmonic approximation.

a r t i c l e

i n f o

Article history: Received 17 September 2012 Received in revised form 22 October 2012 Accepted 22 October 2012 Available online 2 November 2012 Keywords: Infrared and Raman spectra Spectral assignments Enthalpy difference Conformations DFT calculations 1-Methyl-1-silacyclohexane

a b s t r a c t Raman spectra of of 1-methyl-1-silacyclohexane as a liquid were recorded at 293 K and depolarization data obtained. Additional Raman spectra were recorded at various temperatures between 293 and 143 K, and intensity changes with temperature of certain Raman bands were detected. A supercooled liquid appeared after slow cooling, but an amorphous phase was observed after shock freezing to 78 K. After annealing, first a plastic phase and subsequently a crystal were observed. In the crystalline phase spectral shifts and some vanishing bands were observed. The infrared spectra of the vapor and liquid were studied at ambient temperature and the solid sample investigated at 78 K. Negligible spectral changes ocurred at 78 K compared with the fluid state, but after annealing to ca. 170 K an apparent crystal was formed and a few bands vanished and others were shifted. The compound exists in two conformers, equatorial (e) and axial (a) in the liquid, amorphous and plastic phases, but only the e-conformer was present in the crystal. The experimental results suggest that the econformer has 0.6 kJ mol1 lower energy than a in the liquid. B3LYP calculations with various basis sets gave a conformational enthalpy difference DH (a–e) around 2.4 kJ mol1 while G3 model chemistry gave 0.6 kJ mol1. Infrared and Raman intensities, polarization ratios and vibrational frequencies for the e and a conformers were calculated. The fundamental wavenumbers were also derived in the anharmonic approximation in B3LYP/cc-pVTZ calculations. A relative deviation of 0.94% and 1.31% between the observed and calculated wave numbers for the 57 modes of the e-and a-conformers, respectively, was obtained. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction It has been established from numerous studies that the low energy form of the cyclohexane ring is the chair conformation whereas the additional boat and twist forms have much higher energies. They are therefore not observed under ordinary conditions, but the high temperature vapor has been trapped in argon matrices ⇑ Corresponding author. Tel.: +47 22855678; fax: +47 22855441. E-mail address: [email protected] (P. Klaeboe). 0022-2860/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.molstruc.2012.10.045

close to He temperature and the twist form identified. In most monosubstituted cyclohexanes [1,2] the equatorial (e) conformer has lower energies than the axial (a) conformer, leading to a larger abundance of the e-conformers in the vapor and in the liquid states. In the crystalline state obtained by cooling the liquids, these cyclohexanes generally exist only in the e-form since the a-conformers have higher energies and are not accommodated in the crystal lattice. When a carbon atom in cyclohexane is substituted with silicon, the steric and electronic effects are changed. However, much less experimental information is available, since the silacyclohexanes

208

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215

are more unstable than the corresponding cyclohexanes. The parent molecule silacyclohexane was studied by gaseous electron diffraction (GED) [3], microwave (MW) [4] and vibrational spectroscopy [5]. Silacyclohexane forms the backbone of an important group of nematic liquid cryctals. Many patents based on silacyclohexane have been reported on liquid crystals useful for displays. Some monosubstituted silacyclohexanes have been investigated [6] and quantum mechanical calculations [7–9] revealed that most of the substituted silacyclohexanes have the a-form as the preferred conformer. This conclusion has been verified by experimental methods such as MW [10], GED [11] and vibrational spectroscopy [12] on 1-fluoro-1-silacyclohexane, 1-chloro-1-silacyclohexane [13,14], 1iodo-1-silacyclohexane [15], 1-trifluoromethyl-1-silacyclohexane [16] and 1-bromo-1-silacyclohexane [17]. However, in 1-methyl1-silacyclohexane (MSC) (Fig. 1) GED and NMR spectroscopy gave an abundance of 68% in the vapor phase and 74% in solution of the e-conformer [18], while MW spectroscopy suggested a negligible energy difference between the conformers [19]. The apparent anomaly of the conformational abundance in MSC compared with the 1halo-1-silacyclohexanes, urged an infrared and Raman spectroscopic investigation of MSC. Quite recently [20] the Raman spectra of MSC were recorded at various temperatures, and coupled cluster calculations performed to obtain DH(a–e). The results obtained by vibrational spectroscopy and by DFT calculations for MSC are compared with the results for methylcyclohexane. 2. Experimental 2.1. Preparation The sample of 1-methyl-1-silacyclohexane was prepared by the formation of a double Grignard of 1,5-dibromopentane in dry diethylether. The Grignard was then transferred to a solution of freshly distilled methyl trichlorosilane in dry ether under nitrogen and refluxed overnight. The diethyl ether is distilled off under reduced pressure, dry pentane is added and the salt is filtered and washed twice with pentane under nitrogen. The pentane was distilled off at reduced pressure, the product was collected and the final purification was achieved by a trap to trap distillation. The sample is finally reduced by lithium aluminum hydride in dry dibutyl ether. The purity of the sample was checked by the presence of an intense SiAH stretching mode in the infrared spectrum of the gas and by 13C NMR spectroscopy. 2.2. Raman measurements The Raman spectra were recorded with a multichannel spectrometer from Horiba (Jobin Yvon) model T 64,000 employing both

H8 H9

H15

C11

Si4

H3 C1 H2

H16

C6 H10

2.3. Infrared measurements The middle infrared spectra (MIR) were recorded on two Fourier transform spectrometers: a Bruker spectrometer Vertex 80 and a Perkin-Elmer model 2000 (4000–400 cm1). Additional far infrared (FIR) spectra were obtained using the Vertex 80 spectrometer in the FIR set-up (600–100 cm1). Both spectrometers had DTGS detectors, beamsplitters of Ge coating on a KBr substrate were employed in the MIR region and a multilayer coating on Mylar in the FIR region. The spectrometer was flushed with dry nitrogen gas to diminish the absorption of water vapor in the FIR region. The vapor was studied in cells with CsI windows (10 cm path) in the MIR region and polyethylene windows (18 cm path) in the FIR range. MIR spectra of the liquid were recorded as a capillary be-

H17

H7

C5

a single and a triple monochromator. The spectrometer had a CCD detector, cooled to ca. 130 K. The spectra were excited by a Milennia Pro diode-pumped (Nd:YVO4 crystal) frequency-doubled laser from Spectra-Physics (Model J 40), employing 90° scattering geometry, adjusted to give approximately 50 mW of the 532 nm line on the sample. The spectrometer was applied with different optical set-ups: (1) a single monochromator using a notch filter, or a triple monochromator with (2) additive or (3) subtractive collection. The low wavenumber modes were recorded to 60 cm1 with the triple subtractive mode, compared to 120 cm1 with the single monochromator and the notch filter. The spectra were recorded at ca. 2 cm1 resolution, and depolarization measurements were recorded in the 90° mode, employing a polarizer and a scrambler. The polarization unit was calibrated with carbon tetrachloride, filled into a cylindrical tube identical to that employed for MSC. Except for very weak and/or overlapping bands, comprehensive depolarization data were obtained. Raman spectra of the liquid, including depolarization measurements were primarily recorded at room temperature, but additional spectra were obtained for each 10° between 293 and a supercooled liquid at 143 K in a tube of 2 mm inner diameter. The sample tube was surrounded by a glass Dewar and cooled with gaseous nitrogen evaporated from a tank [21] and controlled by an Eurotherm controller, using a calibrated Fe-constantan thermocouple touching the tube and giving approximately ±0.2 °C accuracy. From these spectra, the enthalpy difference DconfH between the conformers in the liquid was calculated. Various e/a band pairs were tested for determining the energy difference. The vapor of MSC was also condensed on a copper finger at 78 K and a glassy phase was formed. The sample was heated in steps to 160 K and the amorphous deposit converted to an apparently plastic solid which crystallized around 170 K.

H13

C12 H14

Equatorial

H19 H20 C18 H21

Axial

Fig. 1. The equatorial (e) and axial (a) conformers of 1-methyl-1-silacyclohexane (MSC), including numbering of atoms for the definition of symmetry coordinates.

209

tween two KBr windows, and FIR spectra with the sample confined in a closed cell of polyethylene of ca. 0.2 mm path. When the vapor was deposited on a CsI window and cooled with liquid nitrogen, an amorphous solid was observed on the window. The sample was annealed to various temperatures between 120 and 170 K and recooled to 78 K before recording of the spectra.

3. Results

Intensity

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215

3.1. Raman spectral results

Intensity

Raman spectra of MSC as a liquid at ambient temperature are presented in Figs. 2 and 3, whereas a depolarization spectrum is shown in Fig. S1 (Supplementary information). A region of the Raman spectra was recorded at 293 (liquid) and 143 K (supercooled liquid) and is presented in Fig. S2. These curves gave minor intensity changes, described to small variations in the conformational equilibria. The experimental results for the Raman and infrared spectra in various states of aggregation are collected in Table 1, and the vanishing bands which definitely disappear in the crystal are marked with asterisks. In more uncertain cases weak bands or shoulders in the fluid phases are merely left open if they are not observed in the crystal spectra. Raman spectra of the liquid were recorded for every 10° in two independent series of experiments between 293 and 143 K. Various intensity variations with temperature of Raman bands relative to neighboring bands were observed. The Raman bands which are enhanced at lower temperatures should belong to the e, and those diminishing upon cooling will be due to the a conformer. It is characteristic for MSC that a large majority of the Raman (and infrared) bands observed belong to both conformers. The intensities of each band pair were fitted to the van’t Hoff equation: ln{Ia(T)/Ie(T)} = DconfH RT + constant; where Ia/Ie is the ratio in peak heights or integrated areas and it is assumed that DconfH is constant with temperature. Both peak heights and integrated areas were employed for determining the band intensities. Three cases of e and a band pairs were suggested for DconfH (a–e) measurements: the 711/728 cm1 pair was employed in peak heights giving 0.42 kJ mol1 and in integrated areas giving 0.57 kJ mol1. Another pair 631/728 cm1 gave 0.89 kJ mol1 in peak heights (Fig. 4), but the intensity of the a-band at 631 cm1 was influenced by the very intense neighbor at 607 cm1. An average value of DconfH (e–a) = 0.6 kJ mol1 was obtained. Since the pure e and a bands are all weak and often close to strong neighbor

600

500

400

300

200

100

Wavenumber / cm-1 Fig. 3. Raman spectra of MSC in the (low) range 700–100 cm1 at 293 K, recorded with two ordinate values.

bands the calculated energy difference has a relatively high uncertainty in MSC (as discussed under spectral interpretation). In a very recent paper [20] the Raman spectra of MSC were recorded at various temperatures to obtain DconfH (e–a). It was observed that while MSC has intense overlapping bands for both conformers at 606 cm1 (see below) the deuterated molecule C5H10SiD(CH3) [20] had separate Raman bands at 635 (e) and 612 cm1 (a). This band pair, involving symmetric stretching of the heavy Si and C atoms, belong to the most intense Raman bands in the spectra of substituted silacyclohexanes. Provided that they are distinctly separated, these bands are well suitable for determination of the energy difference. The van’t Hoff plots gave the values 0.68 (peak heights) and 0.56 kJ mol1 (band areas) for C5H10SiD(CH3) with high precission in the temperature range 298– 210 K [20]. Ethalpy values between 0.53 and 0.72 kJ mol1 were reported for C5H10SiD(CH3) in n-pentane, dichloromethane and tetrahydrofurane solutions using peak heights and integrated areas [20]. Our B3LYP cc-pVTZ calculations gave a negligible difference of 0.01 kJ mol1 between DconfH (e–a) in MSC and in C5H10SiD(CH3). Additional Raman spectra of MSC were recorded in a cryostat at 78 K after the vapor was sprayed on a cold finger of Cu. The spectra suggested that the sample formed a supercooled, amorphous phase with spectra similar to those obtained at 143 K in the capillary (see above). The sample was subsequently annealed to six temperatures between 120 and 170 K and recooled to 78 K before the spectra were recorded. It was observed that the sample changed to a solid when annealed to 131–156 K, but the spectra were essentially unchanged, suggesting a possible plastic phase. Between 164 and 170 K the sample appeared crystalline, and after cooling to 78 K some Raman bands vanished and others were shifted relative to those of the unannealed sample as apparent from Fig. S3. The Raman bands at 1245, 879, 711, 631, 361 and 180 cm1 (having asterisks in Table 1) vanished and those at 216 and 207 cm1 were strongly reduced in intensities compared to the spectrum of the unannealed sample.

3.2. Infrared spectral results

3000

2500

2000

1500

1000

Wavenumber / cm-1 Fig. 2. Raman spectra of MSC as a liquid in the (high) range 3200–700 cm1 at 293 K, recorded with two ordinate values.

MIR vapor spectra of MSC in the CAH and SiAH stretching regions are presented in Fig. 5, and the range 1360–690 cm1 in Fig. 6. A few vapor contours with resolved P, Q and R peaks were observed in the spectra, but in other cases apparent fine structure is attributed to close-lying conformer peaks. An infrared spectrum of a liquid capillary in the MIR region is given in Fig. 7 and a FIR spectrum of the liquid in a polyethylene cell appears in Fig. 8.

210

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215

Table 1 Infrared and Raman spectral dataa for 1-methyl-1-silacyclohexane (MSC). Infrared

Interpretation

vapor 298 Kb

Liquid 298 Kb

Solid 78 Kb

Liquid 298 Kb

Solid 78 Kb

Equatorial (e)

Axial (a)

2966mc 2946m, sh 2924vvs 2910s, sh 2901s

2958s

2956m 2949m 2934s, br 2912vvs 2901s, sh

2958m, D

2958s

2931s, P 2913m, sh 2906vs, br 2883s, P 2877w, sh 2854vs, P

2926s, sh 2913s, sh 2904vvs, br 2878m 2847vs

m33 A00 d m 1 A0 m 2 A0 m 3 A0 m4 A0 , m5 A0 , m34 A00 m35 A00 m6 A0 , m36 A00 m7 A0 , m8 A0 , m37 A00

m33 A00 m 1 A0 m 2 A0 m 3 A0 m4 A0 , m5 A0 , m34 A00 m35 A00 m6 A0 , m36 A00 m7 A0 , m8 A0 , m37 A00 m 9 A0

2112s, br, P 1459w, sh 1449m, P 1410m, br, P 1344w 1331w 1291m 1264w, D 1252m 1245m 1198s, P 1180m, P

2105s, br 1460w, sh 1442s 1405m, br 1339w 1327w 1290m 1260w 1253w  1193w 1178m

m 9 A0 m10 A0 , m11 A0 , m38 A00 m12 A0 ,m39 A00 m13 A0 , m40 A00 m14 A0 , m41 A00 m42 A00 m15 A0 m16 A0 , m43 A00 m44 A00 m17 A0 m18 A0

m10 A0 , m38 A00 m11 A0 , m12 A0 ,m39 A00 m13 A0 , m40 A00 m14 A0 , m41 A00 m42 A00 m15 A0 m43 A00 m44 A00 m16 A0 m17 A0 m18 A0

1102w 1079m, D? 1051w 1045vw 1007m, P " 997w, sh. P 911vs, P

1096w 1074w

m45 A00

m45 A00

1008s 995w 906s

m46 A00 m47 A00 m19 A0 m20 A0 m21 A0

m46 A00 m47 A00 m19 A0 m20 A0 m21 A0

886m 882w

m48 A0 m22 A0 m49 A00

m48 A0

887w 879w 852 vw

 846w

m50 A00

795m, sh 792s

m23 A0 m24 A0

747m 727m

m51 A00 m25 A0

2870m, sh 2855m, sh 2122vs, br 2105s, br 1461w 1451w, br 1410w, br 1345w, br

1264m, br 1258s 1253m, br 1206w 9 1190m ; R = 1185s; Q ; 1178m; P 1100w 1052w 1045w 1003m, br 994m, br 9 918s; R = 913vs; Q ; 908s; P 891vs 887vs 882vvs 9 862s; R = 858vs; Q ; 851vs; P 798s 793s 787m 752vw 9 735m; R = 730s; Q ; 725m; P

2925vs 2914vs 2904vs 2882w 2879s 2852s 2118vs 2109vs 1460w 1446m 1411w 1342m 1330w 1291w 1262m 1252s 1246w,sh 1197m 1181s

1289w 1262m 1247vs  1198m 1180vs

1101w 1077w 1046m

1099m 1079w 1057w, br

1007s 996s 911vs

1007w 994m 913vs

887vs 878vs 874vw,sh 850vvs

2877m 2852s 2102vs 1458w 1447w, br 1410w 1340vw

890m, sh 890vs 880vs 854vs 847vs



798w,sh 791s 781vw 750w 727s

751w 728vs

799w,sh 794s, P 781vw, sh 752w, D 728m, P

606w, br 484w

710m 675m 664w 632vw 597s 482vs

 675s 668w  601m 483vw

711w, D 678s, P 666vw, sh 631w, D 607vvs, P 477m, P

367vw 357vw 349sh 303vw 246vw, br 225vw 209vw 172vw

362vs 358m, sh 345m 307s 244s 218m 209m 165vw

367m, P 361m 343m, P 306s, P 244w 216m, D 207w 166w, sh

151vw 122vw

114vw

160m, P 111w 93vw

711w, sh 675w, br

a

Raman

798w, sh 791vs

m22 A0 , m49 A00 m50 A00 m23 A0 m24 A0

 677s 668w  606vvs  476m 472m 366m  342m 307s 240m 220w () 213vw ()  180w 165w

m26 A0 m52 A00 m27 A0 , m53 A00 m28 A0 m29 A0 m54 A00 , m30 A0 m31 A0

m25 A0 , m51 A00 m52 A00 m26 A0 m53 A00 m27 A0 m28 A0 m54 A00 m30 A0 m29 A0 m55 A00 m31 A0

m55 A00

112w

m56 A00 m32 A0

81vw

m57 A00

m56 A00 m57 A00 m32 A0

Wavenumbers. Recording temperature. c Abbreviations; s, strong; m, medium; w, weak; v, very; sh, shoulder; br, broad; P, polarized; D, depolarized in the Raman spectra; P, Q and R signify rotational band contours in the infrared vapor spectra, asterisks denote IR and/or Raman bands of the fluid phases vanishing in the crystal. d The modes m1–m27 belong to species A0 , m28–m48 to species A00 . b

211

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215 -0.2 100

Transmittance

-0.3 -0.4

In Ia / Ie

-0.5 -0.6 -0.7

50

0

-0.8

3000

2500

2000

1500

1000

500

Wavenumber / cm-1

-0.9 0.003

0.004

0.005

0.006

0.007

0.008

1/T

Fig. 7. MIR spectrum of MSC as a liquid capillary between KBr plates at room temperature in the range 3300–500 cm with resolution 1 cm1.

Fig. 4. Van’t Hoff plots of the axial/equatorial band pairs 711/728 with peak heights (open circles) and with integrated areas (black squares) and 631/728 cm1 with peak heights (black triangles) of MSC in the temperature range 293–163 K. 1.0

Transmittance

0.3

Transmittance

0.2

0.5

0.1

0.0 0.0 600

500

400

300

200

Wavenumber / cm-1 -0.1 3000

2500

2200

Wavenumber /

2000

cm-1

Fig. 5. Middle infrared spectra (MIR) of MSC as a vapor at room temperature in the range 3300–2400 and 2300–2000 cm1 in a 10 cm cell with CsI windows, pressure 9 Torr (solid curve) and 3 Torr (dotted curve) with resolution 1 cm1.

Transmittance

0.4

Fig. 8. Far infrared (FIR) spectrum of MSC as a liquid in a polyethylene (throw away) cell of approximate thickness 0.2 mm at room temperature, in the range 600–150 cm1 with resolution 2 cm1.

Low temperature MIR spectra at 78 K, deposited on a CsI window are presented in Fig. S4. The sample was annealed to temperatures in the range 120–170 K, and certain small changes were observed, similar, but less prominent, than in the Raman recordings. Most significant, the IR bands around 710 cm1 vanished in the annealed sample. In earlier conformational studies we have frequently observed that the bulky samples used in Raman studies are more susceptible to crystallization and phase transitions than the thin capillary layer employed in infrared investigations. 3.3. Quantum chemical calculations

0.2

0.0 1300

1200

1100

1000

900

800

700

Wavenumber / cm-1 Fig. 6. MIR spectra of MSC as a vapor at room temperature in the range 1360– 690 cm1 in a 10 cm cell with CsI windows, pressure 9 Torr (solid curve) and 3 Torr (dotted curve) with resolution 1 cm1.

MP2 and DFT calculations were performed with Gaussian 09 [22] and the minima on the potential surface were found by relaxing the geometry. The conformational energy difference, DconfH, obtained from B3LYP calculations gave 2.1, 2.1 and 2.2 kJ mol1 employing the cc-pVDZ, cc-pVTZ and cc-pVQZ basis set, respectively, with equatorial being the low enthalpy conformer. A theoretical enthalpy difference value of 0.6 kJ mol1 was obtained in G3 calculations [23] whereas 0.8 kJ mol1 was reported [20] from coupled cluster calculations. In the present work B3LYP calculations was the standard method, and was previously employed for calculating the energy difference between the conformers of 1-fluoro-1-silacyclohexane

212

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215

Table 2 Observed and calculated vibrational modes of the e-conformer of 1-methyl-1-silacyclohexane (MSC). No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 a b c d e f

Sym. spec. 0

A A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00

Harma Waven.

Anharm. Waven.

Obs.b Waven.

IRc Int.

Rad Int.

Dep. ratio

PEDe

Description

3025 3094 3043 3053 3037 3004 3002 2993 2162 1492 1506 1467 1456 1380 1323 1292 1229 1211 1020 1017 928 898 810 796 722 677 600 475 334 304 236 96 3094 3053 3031 3003 2999 1464 1497 1453 1381 1369 1298 1285 1127 1089 1064 920 903 865 764 673 613 374 202 160 147

2947 2946 2905 2903 2897 2876 2849 2846 2094 1465 1464 1441 1419 1348 1292 1270 1201 1180 997 994 910 887 799 778 713 674 595 485 353 305 238 90 2946 2902 2889 2876 2847 1458 1449 1415 1342 1337 1268 1253 1107 1066 1035 898 883 846 749 659 606 366 183 141 84

2946 2924 2910 2901 2901 2870 2855 2855 2105 1461 1461 1442 1410 1345 1291f 1264 1206 1185 1003 994 913 887 798 793 730 675 606 484 367 303 246 122 2966 2901 2883f 2870 2855 1461 1451 1410 1345 1334f 1264 1258 1100 1052 1045 891 882 858 752 666f 606 367 172 151 81f

9.3 14.1 61.1 44.8 67.6 33.8 24.2 12.2 146.6 7.8 3.4 4.4 2.4 6.5 0.9 18.9 3.0 24.7 21.5 1.5 55.1 76.5 21.4 1.7 60.6 10.9 4.1 0.9 0.5 1.2 1.3 0.2 12.1 25.8 51.4 1.7 23.1 2.2 2.5 4.8 0.2 0.1 2.2 0.0 0.9 1.9 4.0 36.7 20.1 72.1 0.1 3.4 13.4 0.1 0.6 0.1 0.0

323.0 143.0 217.0 232.0 527.0 171.0 288.0 355.0 442.0 37.0 19.0 49.0 23.0 1.0 41.0 8.0 14.0 35.0 34.0 50.0 5.0 31.0 12.0 79.0 63.0 92.0 414.0 31.0 27.0 220.0 26.0 67.0 174.0 133.0 137.0 124.0 15.0 35.0 55.0 48.0 0.6 16.0 12.0 24.0 14.0 16.0 5.0 34.0 24.0 6.0 26.0 55.0 21.0 51.0 46.0 108.0 6.0

0.00 0.72 0.45 0.26 0.16 0.22 0.07 0.18 0.18 0.74 0.60 0.74 0.75 0.67 0.71 0.15 0.09 0.74 0.41 0.74 0.70 0.73 0.40 0.09 0.53 0.29 0.05 0.08 0.18 0.32 0.62 0.72 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75

100S29 100S28 73S9 + 12S8 80S13 76S11 + 11S8 77S12 + 10S13 77S10 + 14S8 61S8 + 20S9 + 12S11 100S14 89S16 92S18 91S32 87S22 77S20 69S21 + 19S24 94S30 28S24 + 21S19 + 14S25 + 12S21 30S25 + 28S24 + 14S2 31S25 + 18S2 + 17S19 + 13S17 27S2 + 19S1 + 15S24 + 14S19 19S17 + 14S5 + 14S19 + 11S2 49S26 + 36S31 56S23 + 23S17 61S1 + 17S2 62S15 + 11S3 26S26 + 22S31 + 21S3 43S3 + 16S15 + 15S31 + 10S26 26S5 + 24S4 + 13S19 + 12S23 35S4 + 29S5 + 19S7 35S6 + 26S7 + 15S27 34S27 + 29S7 39S27 + 30S6 + 18S7 100S54 84S41 + 12S40 86S39 57S40 + 35S38 57S38 + 20S40 + 11S39 95S56 90S44 91S48 64S42 + 12S33 + 12S47 54S46 + 16S43 + 12S34 27S50 + 19S47 + 14S36 20S43 + 19S49 + 19S51 + 19S53 34S50 + 16S47 + 10S34 31S34 + 26S33 + 11S51 33S53 + 25S51 + 14S34 + 11S33 32S36 + 30S34 + 23S49 43S49 + 28S45 + 12S34 17S55 + 15S37 + 14S34 + 13S49 42S49 + 29S55 + 14S45 26S35 + 25S49 + 19S37 + 13S53 54S49 + 38S53 52S36 + 27S53 + 14S37 72S53 + 16S36 51S37 + 40S52 63S57 + 13S37

CH3 sym. stretch CH3 antisym. stretch CH2 antisym. stretch CH2 antisym. stretch CH2 antisym. stretch CH2 sym. stretch CH2 sym. stretch CH2 sym. stretch Si–H stretch CH2 scissor CH2 scissor CH3 antisym. deform. CH2 scissor CH2 rock CH2 twist CH3 sym. deform. CH2 rock CH3 twist. CH2 twist C–C–C sym. stretch CH2 wag C–Si–H deform. CH2 wag CH3 antisym. deform. Si–C(H3) stretch CH3 antisym. def. C–Si–C sym. stretch CH2 wag C–C–C bend C–C–Si bend C–Si–C bend C–Si–C(H3) bend CH3 antisym. stretch CH2 antisym. stretch CH2 antisym. stretch CH2 sym. stretch CH2 sym. stretch CH3 antisym. deform. CH2 scissor CH2 scissor CH2 rock CH2 rock CH2 twist CH2 twist CH2 rock C–C–C antisym. stretch CH2 twist C–C–C antisym. stretch CH2 wag CH3 antisym. deform. CH3 antisym. def. C–Si–C antisym. stretch CH2 wag C–C–C bend Ring def. C–C–Si bend CH3 torsion

Calculated with B3LYP/cc-pVTZ. The observed wavenumbers are derived from IR vapor spectra.except when noted. Calculated infrared intensities in Km/mole. Calculated Raman scattering activities in arbitrary units, see text and Ref. [25]. For definition of symmetry coordinates, see Table S1. The observed wavenumbers are derived from Raman liquid spectra.

[12] and 1-chloro-1-silacyclohexane [14]. Vibrational wavenumbers were obtained for both conformers in B3LYP/cc-pVDZ, ccpVTZ and cc-pVQZ calculations. There were only minor differences between the results obtained with the triple and quadruple zeta basis sets, suggesting that the B3LYP electron density is described with sufficient accuracy employing the triple zeta basis set. Provided an adequate basis set is used in calcula-

tions of the /normal modes of vibration, the calculated wavenumbers are generally too high. A common procedure to remedy this discrepancy is to scale the calculated wavenumbers to compensate for, among others, mechanical anharmonicity. In the present study the cubic and quartic force constants were calculated allowing a prediction of the anharmonic vibrational wavenumbers directly.

213

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215 Table 3 Observed and calculated vibrational modes of the a-conformer of 1-methyl-1-silacyclohexane (MSC). No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 a b c d e

0

A A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00 A00

Harma Waven

Anharm. Waven.

Obs.b Waven.

IRc Int.

Rad Int.

Dep. ratio

PEDe

Description

3026 3096 3043 3054 3037 3010 2993 3002 2172 1493 1467 1508 1455 1380 1322 1297 1229 1212 1023 1017 928 906 811 797 707 673 593 484 332 368 212 104 3093 3053 3031 3009 2998 1499 1468 1454 1381 1368 1299 1285 1131 1088 1061 916 893 867 740 678 627 364 241 146 151

2940 2938 2905 2900 2897 2877 2846 2823 2119 1475 1446 1431 1418 1351 1294 1245 1202 1180 999 995 911 879 794 779 696 650 582 481 349 345 192 54 2949 2898 2890 2883 2829 1447 1437 1414 1342 1339 1268 1240 1111 1067 1036 893 871 849 711 671 615 351 215 97 96

2946 2924 2910 2901 2901 2870 2855 2855 2122 1461 1451 1451 1410 1345 1291f 1253 1206 1178f 1003 994 913 879 798 787 711 666f 606 484 357 349 209 81f 2966 2901 2883f 2870 2855 1461 1451 1410 1345 1334f 1264 1258 1100 1052 1045 891 879 858 711 675 631 367 225 122 93f

5.5 14.1 59.2 54.3 65.3 .21.6 9.8 34.5 183.3 7.4 3.6 4.0 2.9 5.1 0.6 22.5 1.4 19.1 22.4 3.1 29.2 106.0 42.4 6.5 23.9 4.4 1.1 2.7 0.2 5.5 0.9 0.0 12.8 20.5 60.9 0.6 20.3 2.4 1.6 7.8 0.1 0.3 1.4 0.1 0.0 0.0 1.6 7.7 0.1 117.6 20.1 0.3 0.9 1.9 0.6 0.0 0.1

325 113 228 204 514 291 399 126 585 41 10 24 55 1 43 5 16 37 15 50 3 45 3 85 74 81 518 6 20 35 114 130 167 130 122 106 44 45 31 57 5 12 14 19 18 14 3 37 16 16 39 11 91 16 53 137 18

0.01 0.74 0.36 0.40 0.13 0.18 0.17 0.08 0.21 0.72 0.71 0.62 0.75 0.75 0.67 0.10 0.13 0.72 0.28 0.74 0.61 0.74 0.17 0.05 0.71 0.75 0.02 0.48 0.45 0.46 0.50 0.64 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75

99S29 100S28 75S9 + 13S8 85S13 72S11 + 13S12 81S12 53S8 + 20S9 + 14S11 + 12S10 76S10 + 22S8 100S14 89S16 84S32 + 10S22 93S18 82S22 + 10S32 76S20 69S21 + 20S24 94S30 27S24 + 21S19 + 15S25 + 12S21 29S24 + 28S25 + 14S2 30S25 + 18S19 + 16S2 + 12S17 29S2 + 20S1 + 15S24 + 13S19 19S17 + 16S19 + 14S5 + 12S2 52S26 + 35S31 40S23 + 21S17 + 11S31 59S1 + 16S2 43S15 + 21S3 36S26 + 31S21 + 12S15 62S3 + 25S15 22S5 + 17S4 + 13S19 + 12S23 + 12S6 35S4 + 28S5 + 17S7 27S6 + 24S7 + 14S4 + 10S27 33S7 + 47S27 47S6 + 19S7 + 15S27 + 10S4 100S54 89S41 78S39 + 14S40 79S40 83S38 + 11S39 89S44 95S56 92S48 65S42 + 12S33 + 11S47 53S46 + 16S43 + 12S34 28S50 + 19S47 + 14S36 22S51 + 21S43 + 18S49 + 16S53 35S50 + 16S47 + 12S24 34S34 + 31S33 33S51 + 26S53 + 10S34 42S34 + 35S36 37S45 + 28S49 33S53 + 31S49 + 13S55 39S55 + 13S45 70S49 + 14S53 37S53 + 31S49 + 18S35 52S36 + 29S53 + 14S37 73S53 32S37 + 23S52 + 17S36 + 16S53 61S57 + 16S36 + 14S53

CH3 sym. stretch CH3 antisym. stretch CH3 antisym. stretch CH2 antisym. stretch CH2 antisym. stretch CH2 sym. stretch CH2 sym. stretch CH2 sym. stretch Si–H stretch CH2 scissor CH3 antisym. def. CH2 scissor CH2 scissor CH2 rock CH2 twist CH3 sym. def. CH2 rock CH2 rock CH2 twist C–C–C sym. stretch CH2 wag C–Si–H def. CH2 wag C–C–C sym stretch Si–C(H3) stretch C–Si–H def. C–Si–C sym. stretch. C–C–C bend. C–C–C bend. C–C–C bend. C–Si–C(H3) bend C–C–Si bend. CH3 antisym. stretch CH2 antisym. stretch CH2 antisym. stretch CH2 sym. stretch CH2 sym. stretch CH2 scissor CH3 antisym. def. CH2 scissor CH2 rock CH2 rock CH2 rock CH2 twist CH2 rock C–C–C antisym. stretch CH2 twist C–C–C antisym. stretch CH2 wag C–C–Si bend CH3 antisym. def. CH2 wag C–Si–C antisym. stretch C–C–C bend. Ring def. C–C–Si bend CH3 torsion

Calculated with B3LYP/cc-pVTZ. The observed wavenumbers are derived from IR vapor spectra, except when noted. Calculated infrared intensities in Km/mole. Calculated Raman scattering activities in arbitrary units. See text and Ref. [25]. For definition of symmetry coordinates, see Table S1.

3.4. Normal coordinate calculations Harmonic force constants were obtained for each of the two conformers of MSC in B3LYP/cc-pVTZ calculations. The harmonic, unscaled force constants were transformed from Cartesian to symmetry coordinates (Table S1), derived from a set of valence coordinates to obtain an approximate description of the normal modes; the numbering of the atoms appears in Fig. 1. These calculations were carried out both for the e-and the a-conformers,

employing the VIBROT program [24] and the results are listed in Tables 2 and 3. The infrared intensities obtained in the harmonic approximation, are included in these tables in combination with the experimental wavenumbers, relative intensities and Raman depolarization values. The theoretical Raman intensity, R, which mimics the measured Raman spectrum is related to the Raman scattering activity: Ri = C(mL–mi)4m1 B1 I Si, where mi is the frequency i and Si the corresponding Raman scattering activity of mode i, mL is the laser excitation frequency, Bi = 1 exp(hmi C/kT), and C a

214

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215

constant [25]. Finally, the wavenumbers derived from the anharmonic calculations are included in Tables 2 and 3 to be directly compared with the experimental values as observed in the infrared and Raman spectra. The PED (potential energy distribution) is expressed in terms of the symmetry coordinates (Table S1) and only PED terms larger than 10% have been included in Tables 2 and 3. 3.5. Conformations It can be seen from Tables 1–3 that the e and a bands have been tentatively identified as fundamentals while a few additional bands are presumably caused by combination bands or overtones. The enthalpy difference between the e and a-conformers in methylcyclohexane has been investigated by many different techniques including 13C NMR [26] and Raman spectroscopy [27,28] giving values as high as DconfH (a–e) = 8.5 kJ mol1 [28] in the liquid. Therefore, the equilibrium of methylcyclohexane is highly shifted towards the e-conformer compared to that in MSC, in which DconfH (a–e) is merely 0.6 kJ mol1. The potential energy surface and barrier to chair–chair conversion of silacyclohexane has been calculated [29–31] and compared with the results for cyclohexane [32]. The barrier in silacyclohexane between the chair (lowest point) and half chair (highest point on the path) was calculated to be 22.9 kJ mol1 [29–31]. In cyclohexane the barrier was found to be 53.5 kJ mol1 (MP2/ 6311+G(2df,2pd)) [32], about twice as large as in silacyclohexane. As is apparent from Tables 2 and 3 the vibrational modes derived from the anharmonic calculations are with a few exceptions at lower wavenumbers than those from the harmonic approximation both for the e and a conformers. Exceptions are the e modes m28, m29, m30, m31, and the a mode m30 which all are blue shifted in the anharmonic approximation. 3.6. Crystallization Generally, most simple compounds crystallize in a single conformer which is preferred in the crystal lattice. This leads to a simplification of the spectra compared with those in the fluid phases, which may contain two or more conformers. From these experimental results, quite reliable assignments of the vibrational modes to the proper conformers can be made, supplementing the information obtained from quantum chemical calculations. Many attempts were made to crystallize MSC using different cooling techniques, some of these attempts were unsuccessful and an amorphous solid remained in the low temperature domaine. 1. Slow cooling of the compound filled into a capillary tube from 293 to ca. 140 K, leads to a supercooled liquid as was demonstrated in the Raman procedures (see above). 2. Condensing the vapor on a CsI window at 78 K, to a thin layer suitable for IR studies, gives an amorphous phase. Annealing to temperatures in the range 100–130 K and subsequent cooling to 78 K leads to a possible plastic phase and further annealing until 170 K eventually to a crystal. The weak IR bands around 710 and possibly at 632 and 1246 cm1 in the liquid vanished in the crystal, but the evidence was less prominent than in the Raman spectra. 3. Condensing a sample on a Cu finger (previously roughened with sandpaper), suitable for Raman studies at 78 K, leads to a glassy pearl. Careful annealing to temperatures between 120 K and 150 K, and leaving the sample for ca. 10 min at the annealing temperatures brought visual changes around 140 K. Annealing to ca. 170 K, and recooling to 78 K resulted in some significant spectral shifts and a few vanishing bands (see Fig. S3), suggesting that an anisotropic solid (crystal) was formed.

It has been established by extensive Raman spectroscopic measurements at low temperatures [33–35] that methylcyclohexane forms normal liquid (293–150 K), undercooled liquid (150–80 K) and glass (below 85 K) and the various phase transitions have been thoroughly investigated. The Raman mode at 545.7 cm1 of the e-conformer was studied, attributed to a CH2 rocking mode [35], or as a ring puckering mode [28]. The glassy state depends on the history of the sample (e.g. different cooling rates) in contrast to states which are independent of history [35]. With the close structural similarity between MSC and methylcyclohexane it is not surprising that the complicated phase transitions observed for the latter compound [33–35] may also be relevant for MSC. As observed in Fig. S3 (Raman) and Fig. S4 (infrared) only a few Raman and IR bands of the amorphous solid vanished in the crystal spectra (see below). This is in good agreement with the results of the calculations, often suggesting identical or nearly identical wavenumbers for both conformers (Tables 2 and 3). Three examples of crystal splitting appeared in the spectra of MSC: a weak Raman band at 166 cm1 in the liquid was split into the doublet 180/ 165 cm1 and a medium intense band at 477 cm1 was split into the doublet 476/472 cm1. The IR spectrum of the liquid had a very intense band at 850 cm1 which appeared as a doublet at 854 and 847 cm1 in the crystal, interpreted as crystal splitting of the m50 A00 modes of coinciding e and a conformers.

4. Spectral interpretation With 21 atoms in MSC each conformer has 57 modes of vibration, and the e and a conformers both have a plane of symmetry (Cs symmetry). The fundamentals of both conformers will divide into 32 A0 and 25 A00 modes and give rise to polarized and depolarized Raman bands, respectively Since the order of the fundamentals sometimes change between the harmonic and the anharmonic calculations (Tables 2 and 3), the latter are made the basis for the numbering. Mostly, the assignments are in good agreement with the results of the B3LYP/cc-pVTZ anharmonic calculations. It is immediately apparent from Table 1 that among the 57 vibrational modes expected for each conformer, themajority (34) consists of overlapping e and a modes. This is supported by the fact that only 6 of the 57 a-modes vanish in the crystal spectra since they are mostly coinciding with e-modes. The results of the B3LYP/ccpVTZ calculations in the anharmonic approximation reveal that 28 of the corresponding e and a modes are situated closer than 10 cm1 apart, strongly suggesting coinciding fundamentals. Consequently, neither IR vapor contours nor Raman polarization measurements can provide definite aid for the assignments. Moreover, the calculated polarization data of Tables 2 and 3 show that in 5 cases for the e-conformer and 6 for the a-conformer the A0 modes have depolarization ratios equal to 0.75 or 0.74 and they will therefore easily be mistaken for A00 modes in the actual measurements. The 8 A0 and 5 A00 modes of the e and a conformers assigned to CH3 and CH2 antisymmetric and symmetric stretches are observed as highly overlapping modes between 3000 and 2850 cm1 in the vapor, and some of these Raman bands were among the most intense in the spectra. The Si-H stretch (m9 was attributed to the very intense IR vapor bands at 2122 (a) and 2105 cm1 (e) compared with the calculated values 2119 (a) and 2094 (e) cm1. In the liquid, two bands at 2118 and 2109 cm1 were observed in IR while only one broad, intense peak at 2112 cm1 was present in the Raman spectrum. In 1-fluoro-1-silacyclohexane [12] and 1-chloro-1-silacyclohexane [14] the a-conformer of Si-H stretch was also situated at slightly higher wavenumber than the e-confomer in the IR vapor spectra. With 13 hydrogens in MSC attached to carbon, we expect 20 vibrational modes involving CH2 scissor, wag, twist and rock and

P. Klaeboe et al. / Journal of Molecular Structure 1034 (2013) 207–215

three involving CH3 deformation. These vibrational modes are expected in the range 1500–700 cm1. A few skeletal stretches and bending modes m46 and m48 and m20 are intermingled with the CH3 and CH2 bending modes. As is apparent from Table 1 six bands (vanishing in the crystal) are described as separate a-fundamentals and a similar number was attributed to separate e-modes. Three cases of e and a band pairs were suggested for the determination of DconfH (a–e): 781 (a) 788 cm1 (e), 711 (a) 728 cm1 (e) and 631 (a) 628 cm1 (e). The first pair 781/788 cm1 was not reliable because of strong overlap between the components, and the two others resulted in DconfH (a–e) = 0.6 kJ mol1 (Fig. 4). Since the pure e and a bands are all weak and often close to strong neighbor bands the calculated energy difference has a relatively high uncertainty. It appears from Tables 2 and 3 that in the range below 700 cm1 we expect vibrational modes for the e and a conformers involving bending of C–C–C, C–Si–C, C–C–Si and ring deformation. These modes are mixed between 2, 3 or 4 symmetry coordinates and they give mostly rise to weak IR and Raman bands, except the C–Si–C symmetric stretch m27, which appear as a very intense Raman band. The average relative deviations between the observed and calculated (anharmonic) modes (Tables 2 and 3) give 0.94% for the e-conformer and 1.31% for the a-conformer. Acknowledgement The authors are grateful to Dr. Niels Højmark Andersen for his help with installing the low temperature Raman attachment. Valdemaras Aleksa is grateful to (CTCC) the Centre for Theoretical and Computational Chemistry, Department of Chemistry, University of Oslo for a working stipend in Norway. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.molstruc.2012.10. 045. References [1] T. Woldbaek, Acta Chem. Scand. A36 (1982) 641. [2] S.D. Christian, J. Grundnes, P. Klaeboe, E. Tørneng, T. Woldbaek, Acta Chem. Scand. A24 (1980) 391. [3] Q. Shen, R.L. Hildebrandt, V.S. Mastryukov, J. Mol. Struct. 54 (1979) 121. [4] L.B. Favero, W. Caminati, I. Arnason, A. Kvaran, J. Mol. Spectrosc. 229 (2005) 188. [5] G.A. Guirgis, C.J. Nielsen, A. Horn, V. Aleksa, P. Klaeboe, J. Mol. Struct. 1023 (2012) 189.

215

[6] E. Kleinpeter, Advances in Heterocyclic Chemistry, vol 86, Elsevier, Amsterdam, 2004. [7] R. Bjornsson, I. Arnason, Phys. Chem. Chem. Phys. 11 (2009) 8689. [8] A. Bodi, R. Bjornsson, I. Arnason, J. Mol. Struct. 978 (2010) 14. [9] A.J. Weldon, G.S. Tschumper, J. Quantum Chem. 107 (2007) 2261. [10] L.B. Favero, B. Velino, W. Caminati, I. Arnason, A. Kvaran, J. Phys. Chem. A110 (2006) 995. [11] A. Bodi, A. Kvaran, S. Jonsdottir, E. Antonsson, S.O. Wallevik, I. Arnason, A.V. Belyakov, A.A. Baskakov, M. Höbling, H. Oberhammer, Organometallics 26 (2007) 6544. [12] P. Klaeboe, V. Aleksa, C.J. Nielsen, A. Horn, G.A. Guirgis, M.D. Johnston, J. Mol. Struct. 1015 (2012) 120. [13] A.V. Belyakov, A.A. Baskakov, V.N. Naraev, A.N. Rykov, H. Oberhammer, I. Arnason, S.O. Wallevik, Russ. J. Gen. Chem. 81 (2011) 2257. [14] V. Aleksa, G.A. Guirgis, A. Horn, P. Klaeboe, R.J. Liberatore, C.J. Nielsen, Vib. Spectrosc. 61 (2012) 167. [15] A.V. Belyakov, A.A. Baskakov, R.J.F. Berger, N.W. Mitzel, H. Oberhammer, I. Arnason, S.O. Wallevik, J. Mol. Struct. 1012 (2012) 126. [16] G.V. Girichev, N.I. Giricheva, A. Bodi, P.I. Gudnason, S. Jonsdottir, A. Kvaran, I. Arnason, H. Oberhammer, Chem. Eur. J. 13 (2007) 1776. [17] A.V. Belyakov, A.A. Baskakov, V.N. Naraev, A.N. Rykov, H. Oberhammer, I. Arnason, S.O. Wallevik, Russ. J. Phys. Chem. A, vol. 86, N 10-C, in press. [18] I. Arnason, A. Kvaran, S. Jonsdottir, P.I. Gudnason, H. Oberhammer, J. Org. Chem. 67 (2002) 3827. [19] L.B. Favero, B. Velino, W. Caminati, I. Arnason, A. Kvaran, Organometallics 25 (2006) 3813. [20] T. Kern, M. Hölbling, A. Dzambaski, M. Flock, K. Hassler, S.Ó. Wallevik, I. Arnason, R. Bjornsson, J. Raman Spectrosc. 43 (2012) 1337. [21] F.A. Miller, B.M. Harney, Appl. Spectrosc. 24 (1970) 291. [22] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. Montgomery, J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision B.01, Gaussian, Inc., Wallingford CT, 2009. [23] L.A. Curtiss, K. Ragavachari, P.G. Redfern, V. Rasslov, J.A. Pople, J. Chem. Phys. 109 (1998) 7764. [24] G.O. Sørensen, VIBROT, Department of Chemistry, University of Copenhagen, Copenhagen, Denmark. [25] D. Michalska, R. Wysokinski, Chem. Phys. Lett. 403 (2005) 211. [26] R.J. Abraham, D.S. Ribiero, J. Chem. Soc. Perkin Trans. 2 (2001) 302. [27] D.J. Gardiner, G.S. Bassi, G.D. Galvin, A.C. Gorvin, B.P. Staughan, Appl. Spectrosc. 38 (1984) 313. [28] J.R. Durig, R.M. Ward, G.A. Guirgis, T.K. Gounev, J. Raman Spectrosc. 40 (2009) 1919. [29] I. Arnason, G.K. Thorarinsson, E. Matern, Z. Anorg, Allg. Chem. 626 (2000) 853. [30] I. Arnason, A. Kvaran, A. Bodi, Int. J. Quantum Chem. 106 (2006) 1975. [31] F. Freeman, C. Fang, B.A. Shainyan, Int. J. Quantum Chem. 100 (2004) 720. [32] J.R. Durig, C. Zheng, A.M. El Defrawy, R.M. Ward, T.K. Gounev, K. Ravindranath, N. Rajeswara Rao, J. Raman Spectrosc. 40 (2009) 197. [33] H. Abramczyk, G. Waliscewska, M. Lolodziejski, J. Phys. Chem. A 102 (1998) 7765. [34] H. Abramczyk, K. Paradowska-Moszkowska, J. Chem. Phys. 115 (2001) 11221. [35] H. Abramczyk, K. Paradowska-Moszkowska, J. Chem. Phys. 118 (2003) 4169.