X-ray Crystal Structure, Raman Spectroscopy, and Ab Initio Density

The salt 1,1,3,3-tetramethylguanidinium bromide, [((CH3)2N)2C═NH2]+Br− or [tmgH]Br, was found to melt at 135(5) °C, forming what may be referred ...
0 downloads 0 Views 2MB Size
J. Phys. Chem. A 2010, 114, 13175–13181

13175

X-ray Crystal Structure, Raman Spectroscopy, and Ab Initio Density Functional Theory Calculations on 1,1,3,3-Tetramethylguanidinium Bromide Rolf W. Berg,*,† Anders Riisager,†,‡ Olivier N. Van Buu,†,‡ Steffen Buus Kristensen,†,‡ Rasmus Fehrmann,†,‡ Pernille Harris,† and Anna C. Brunetti§ Department of Chemistry and Centre for Catalysis and Sustainable Chemistry, Technical UniVersity of Denmark, KemitorVet, Building 207, DK-2800 Kgs. Lyngby, Denmark, and Department of Photonics Engineering, Technical UniVersity of Denmark, Ørsteds Plads, Building 345V, DK-2800 Kgs. Lyngby, Denmark ReceiVed: July 30, 2010; ReVised Manuscript ReceiVed: NoVember 8, 2010

The salt 1,1,3,3-tetramethylguanidinium bromide, [((CH3)2N)2CdNH2]+Br- or [tmgH]Br, was found to melt at 135(5) °C, forming what may be referred to as a moderate temperature ionic liquid. The chemistry was studied and compared with the corresponding chloride compound. We present X-ray diffraction and Raman evidence to show that also the bromide salt contains dimeric ion pair “molecules” in the crystalline state and probably also in the liquid state. The structure of [tmgH]Br determined at 120(2) K was found to be monoclinic, space group P21/n, with a ) 7.2072(14), b ) 13.335(3), c ) 9.378(2) Å, β )104.31(3)°, Z ) 2, based on 11769 reflections, measured from θ ) 2.71-28.00° on a small colorless needle crystal. Raman and IR spectra are presented and assigned. When heated, both the chloride and the bromide salts form vapor phases. The Raman spectra of the vapors are surprisingly alike, showing, for example, a characteristic strong band at 2229 cm-1. This band was interpreted by some of us to show that the [tmgH]Cl gas phase should consist of monomeric ion pair “molecules” held together by a single N-H+ · · · Cl- hydrogen bond, the stretching vibration of which should be causing the band, based on ab initio molecular orbital density functional theory type calculations. It is not likely that both the bromide and chloride should have identical spectra. As explanation, the formation of 1,1-dimethylcyanamide gas is proposed, by decomposition of [tmgH]X leaving dimethylammonium halogenide (X ) Cl, Br). The Raman spectra of all gas phases were quite identical and fitted the calculated spectrum of dimethylcyanamide. It is concluded that monomeric ion pair “molecules” held together by single N-H+ · · · X- hydrogen bonds probably do not exist in the vapor phase over the solids at about 200-230 °C. Introduction and Previous Work Ionic compounds being molten at or near room temperature exhibit interesting properties and have potential applications.1 Since the discovery that certain ionic compounds could be vaporized and even distilled at moderate temperatures,2 the properties of the gases over the ionic liquids have been intensively studied, see for example refs 3-9. Thus, Armstrong et al.6 have by mass spectrometry found that eight common imidazolium-based ionic liquids evaporate as ionic pairs. Photoelectron spectroscopy on the vapor over 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide has provided similar results.7 Rebolo et al.8,10 have found by ion cyclotron resonance mass spectrometry (ICR-MS) that aprotic ionic liquids A+X- form a gas phase composed by discrete anion-cation pairs [AX], with no free ions or higher aggregates. However, for the gas phase above the 1-methyl-imidazolium acetate, a protic ionic liquid, the vapor was found to consist of mostly isolated neutral molecules of 1-methyl-imidazole and acetic acid.8,11 In this work, we continue studying ionic salts containing the 1,1,3,3-tetramethylguanidinium cation, that is, (C5H14N3)+ or [((CH3)2N)2CdNH2]+ or [tmgH]+. Many of such compounds * To whom correspondence should be addressed. Phone: +45 45 25 24 12. Fax: +45 45 88 31 36. E-mail: [email protected]. † Department of Chemistry. ‡ Center for Catalysis and Sustainable Chemistry. § Department of Photonics Engineering.

are liquid at temperatures not far above room temperature. The compounds are of interest, because of potential use as extraction media for SO2 and CO2 in stack gases.12,13 Thus, the bromide compound, [tmgH]Br, readily absorbs SO2 and CO2 forming a room temperature ionic liquid. This detailed study of [tmgH]Br was carried out to elucidate the affinity of the compound to absorb the gases. Also, we investigated if [tmgH]Br would form a gas complex with a single NH+ · · · Br- hydrogen bond of the kind presumed for [tmgH]Cl.14 In that work, Raman scattering and IR absorption spectra were studied, supplemented by ab initio molecular orbital calculations to bring information on the chemical properties. Such comparison of experimental structures and spectra with model calculations has been reviewed in detail.15 The [tmgH]Cl crystal structure contains discrete centrosymmetric dimers linked by four classical N-H+ · · · Cl- hydrogen bonds according to previous X-ray diffraction results on single crystals.16 In the present work, a similar study of [tmgH]Br is carried out. Experimental Methods and Computational Details Materials. 1,1,3,3-tetramethylguanidine (TMG, 99%), hydrobromic acid (47%), ethanol (99%), and diethylether (99%) were all purchased from Aldrich and used as received. (Caution: Use safety goggles when working with TMG because of eye damage risks17). The 1,1,3,3-tetramethylguanidinium bromide salt, [tmgH]Br, was prepared by a modification of the previously

10.1021/jp107152x  2010 American Chemical Society Published on Web 12/02/2010

13176

J. Phys. Chem. A, Vol. 114, No. 50, 2010

TABLE 1: Crystal Data of [tmgH]Br formula Mw/g mol-1 crystal size/mm crystal system space group

C5H14BrN3 196.10 0.3 × 0.05 × 0.04 monoclinic P21/n (no. 14, C2h5) unique axis b, cell choice 2 7.2072(14) 13.335(3) 9.378(2) 104.31(3) 873.3(3) 1.492 not measured 120(2) 4 400 4.638 0.71073 empirical 2.71-28.00 11769 -9 e h e 9; -17 e k e 17; -12 e l e 12 0.0617 2107 and 1617

a/Å b/Å c/Å β/deg V/Å3 Dc/g cm-3 Dm/g cm-3 temperature/K Z F(000) µ (Mo KR)/mm-1 wavelength/Å absorption correction θ range for data collection/deg no. of reflections reflections collected R(int) reflections, total and with I > 2σ(I) no. of parameters R1 ) ∑|F0| - | Fc|/∑|F0| for all reflections for those with F02 > 2σ(F02)

86 0.0524 0.0346

wR2 ) [∑w(F02 - Fc2)2/∑wF04] /2 for all reflections for those with F02 > 2σ(F02)

0.1007 0.0871

weight function w, where P ) (F02 + 2Fc2)/3 goodness of fit residual charge density/e- Å-3

w-1 ) σ2(F02) + (0.0759P)2 + 1.5513P 0.766 -0.3421 < F < 1 0.015

1

reported method17 involving direct neutralization of an ethanolic TMG solution with an equivalent amount of hydrobromic acid. In a typical procedure, 2.90 mL of 47% HBr (25 mmol) was cautiously added to a stirred solution of 2.94 g of TMG (25 mmol) in 50 mL of ethanol while maintaining the solution at room temperature (Caution: The neutralization is highly exothermic). After continuous stirring for 1 h, the solvent was removed under reduced pressure, leaving a quantitative yield. The raw product was recrystallized from methanol/diethylether (1:3), followed by drying under vacuum (50 °C, 0.1 mbar) to produce a white solid (yield ) 4.54 g, 91%). N,N-dimethylcyanamid (DMCA, >98%) was obtained from Fluka AG, Switzerland. X-ray Diffraction. Profile data from ω scans were collected on a Bruker/Siemens SMART CCD platform diffractometer4 using graphite monochromated Mo KR radiation (λ ) 0.71073 Å; µ ) 4.898 mm-1). Unit cell dimensions were refined and intensity data reduced (corrected for Lorentz and polarization effects) by use of the Siemens SMART and SAINT program systems.4 SHELXTL software5 was used to solve the structure by direct methods and refine it by weighted full-matrix leastsquares fitting of positional and isotropic or anisotropic thermal parameters to F2 using all data to obtain nearly flat difference maps. Hydrogen atoms were observed and modeled at ideal positions (at distances 0.96 Å for CH and 0.86 Å for NH, respectively). Data collection and refinement details are given in Table 1. Positional and equivalent isotropic and anisotropic thermal parameters are listed in Tables S1 and S2 of the

Berg et al. TABLE 2: Selected Bond Distances and Angles in the [tmgH]Br Structure and Comparable Dataa geometry, distance/angle

[tmgH]Br X-ray

[tmgH]Cl X-ray16

[tmgH][NTf2] X-ray14

N1-C1/Å N2-C1/Å N3-C1/Å N2-C2/Å N2-C3/Å N3-C4/Å N3-C5/Å C-H/Å N1-C1-N3/deg N1-C1-N2/deg N3-C1-N2/deg N1-H1A/Å N1-H1B/Å

1.332(4) 1.345(4) 1.348(4) 1.467(4) 1.466(4) 1.449(4) 1.462(4) 0.9800 119.8(3) 121.1(3) 119.1(1) 0.8800 0.8800

1.3304(15) 1.3370(14) 1.3417(15) 1.4630(17) 1.4543(15) 1.4580(15) 1.4621(14) 0.98(1) 119.7(1) 120.91(10) 119.36(10) 0.918(17) 0.886(17)

1.3408(15) 1.3393(15) 1.3433(15) 1.4686(16) 1.4684(16) 1.4635(16) 1.4693(15) 0.9600 119.28(11) 119.46(11) 121.25(11) 0.8600 0.8600

a

E.s.d.s are given in parentheses.

Supporting Information. Some selected bond lengths and angles are included in Table 2, and the structure is illustrated in Figure 1. Raman Spectroscopy. Samples were studied in sealed cylindrical tubes. Visible laser light (green 532 nm with nominal power up to 4W) was used to excite spectra of the samples, directly through the silica glass. A dispersive DILOR-XY spectrometer was used as described in refs 18-20. Rayleigh scattered light was removed through a notch filter. Raman light was detected with a liquid-N2 cooled CCD to give unpolarized spectra with a spectral resolution of approximately 4 cm-1. The spectra were repeatedly collected in several overlaid sections that were combined after removal of cosmic spikes. In a few cases where a broad fluorescent background was observed, a polynomial was fitted to the background and subtracted, allowing for the essential spectrum to be shown more clearly. The wavelength scales were calibrated with the use of cyclohexane and benzonitrile spectra to a precision of about 1 cm-1.21 Samples were sealed inside evacuated silica vials and heated by means of a four-window furnace, designed and built in our laboratory.19,22 Infrared Absorption. FT-IR spectra were obtained at approximately 25 °C from a pressed disk of 1.5 mg of polycrystalline sample powder in 250 mg of KBr, measured against a similar empty reference. A Perkin-Elmer 1710 FT instrument with a diamond ATR attachment was used. The IR spectral resolution was ∼4 cm-1 (100 scans). Mass Spectrometry. A heated Pfeiffer Vacuum Inc. Thermostar GSD 301 T quadrupole mass spectrometer was used. Ab Initio Molecular Orbital (MO) Quantum Chemical Calculations. The calculations were performed with the GAUSSIAN 03W23 program on a Pentium R4 3 GHz personal computer operated under Windows XP. The total geometric/conformational energies of the guessed molecular species were minimized by use of Hartree-Fock/Kohn-Sham density dunctional theory (DFT) procedures at a level of approximation limited by use of restricted-spin Becke’s three parameter hybrid exchange functional (B3), Lee-Yang-Parr correlation and exchange functionals (LYP) and with Pople’s polarization split valence Gaussian basis set functions, augmented with d- and p-type polarization functions and diffuse orbitals on non-hydrogen orbitals (RB3LYP, 6-311+G(d,p)). This level of modeling has proved to be satisfactory to describe, for example, methanol clusters24 and N-containing compounds.25 The Gaussian 03W software was used as implemented with the modified GDIIS algorithm and with tight optimization convergence criteria.23

[((CH3)2N)2CdNH2]+Br- Structure and Raman Spectra

J. Phys. Chem. A, Vol. 114, No. 50, 2010 13177

Figure 1. (A) Structure of the [tmgH]Br crystal, showing two nearly planar [tmgH]+ cations, hydrogen bonded to two bromide anions and (B) the crystal packing showing how these nearly planar dimeric “molecules” are arranged in the unit cell.

The molecules and ions were taken to be in assumed gaseous free state and without any preassumed symmetry. The vibrational frequencies and eigenvectors for each normal mode were calculated without adjusting the force constants. These results are used as a basis for assigning the observed bands according to dominating group frequency motions. Results and Discussion Crystal Structure. The structure of a small needle crystal of [tmgH]Br showed [tmgH]+ cations with a nearly planar CN3 core and a reasonable geometry (see Figure 1). Most notably, the data revealed that the monoclinic bromide salt contains discrete centrosymmetric dimers linked by four classical N-H+ · · · Br- hydrogen bonds, quite similar to those found in the chloride salt.16 The found bond lengths and angles are given in Table 2, which also shows previously determined X-ray structure data for other tetramethylguanidinium compounds: for example,thechloride16 andthebis{(trifluoromethyl)sulfonyl}amide (NTf2-)26 salts. The geometry data compare well to each other. We refer to the Cambridge Structural Database27 for further similar structure determinations, and we conclude that the [tmgH]Br compound is isostructural with the chloride. Ab Initio Calculations. Extended molecular orbital (MO) model calculations are quite efficient to predict chemical

Figure 2. Calculated optimized geometries of the [tmgH]Br ion pair molecule (A) and the dimeric [tmgH]2Br2 ion pair molecule (B).

structures and Raman scattering and IR emission vibrational spectra. In previous work,14 ab initio calculations on the isolated [tmgH]+ ion and [tmgH]Cl and [tmgH]2Cl2 ion pair molecules were reported. It was concluded that these species could be modeled with a reasonable accuracy with the Gaussian 03W software.23 The calculations are performed here to see how similar bromide results compare with the chloride ones. The optimized equilibrium geometries of the isolated [tmgH]Br and [tmgH]2Br2 ion pair molecules were calculated at the B3LYP/ DFT/6-311+G(d,p) level, neglecting further interactions. The optimized structure results are shown in Figure 2 and Table 3. In Table 3, also experimental structure data for the [tmgH]2Br2

13178

J. Phys. Chem. A, Vol. 114, No. 50, 2010

Berg et al.

TABLE 3: Ab Initio Optimized Geometric Structures as Determined by Calculationa for an Isolated Tetramethylguanidinium Ion, [tmgH]+, and for Isolated [tmgH]Br and [tmgH]2Br2 Ion Pairs, Compared to Experimental Data from Crystalline [tmgH]2Br2 geometric quantity, distance/angle

calculated [tmgH]+ ion14

calculated [tmgH]+Br- ion pair

calculated [tmgH]2Br2 dimer ion pair

X-ray structure [tmgH]2Br2

N1-C1/Å N2-C1/Å N3-C1/Å C-H/Å N1-C1-N3/deg N1-C1-N2/deg N3-C1-N2/deg C4-N3-C1-N1/deg C5-N3-C1-N1/deg C2-N2-C1-N1/deg C3-N2-C1-N1/deg N1-H1/Å H1-Br/Å N1-H1 · · · Br/Å N1-H1-Br/deg dipole moment/Debye RB + HF - LYP minimum energy/au

1.34753 1.34400 1.34400 1.09(1) 119.13084 119.13084 121.73833 -21.44330 146.76858 -21.44328 -146.76875 1.00735 1.4001 -363.107356

1.3288 1.3496 1.3617 1.09(1) 119.185 121.356 119.456 -18.553 140.498 -22.348 143.957 1.0734 2.070 3.117 164.4 12.8866 -2937.485883

1.32004 1.35841 1.35841 1.09(1) 120.537 120.537 118.926 -20.702 141.631 -20.702 140.631 1.0336 2.3198 3.3454 171.4 0.0001 -5875.018748

1.332(4) Å 1.345(4) 1.348(4) 0.980(1) 119.9(3) 121.1(3) 119.1(1) 22.7(5)b -145.2(3)b 23.1(5)b -149.2(3)b 0.880 2.628 3.377 146.2 -

a

Gaussian 03W DFT/B3LYP/6-311+G(d,p).23 b Sign of angle depends on definition.

compound found here have been included to facilitate comparison. Although small deviations between calculated and experimental quantities can be seen, it appears that quite reasonable parameter values are obtained for the calculated isolated ion pairs, showing that ab initio DFT B3LYP/6311+G(d,p) modeling is able to predict reasonably accurate structures (compare Table 3 columns). Like previously found for the chloride compound,14 the dimer [tmgH]2Br2 ion pair molecule gives the best fit to the known values in the solid. It should be noted that the total minimum energy of the [tmgH]2Br2 ion pair molecule was found as -5875.018748 au, that is, lower than the sum (-5874.906211 au) of total minimum energies from two TMG and two HBr molecules (TMG: -362.700012 au and HBr: -2574.753093 au at similar sophistication levels of modeling), predicting that the dimer ion pair molecule would be 0.112538 au more stable than two separate TMG and two HBr molecules. For the monomer [tmgH]Br ion pair molecule, the energy came out as -2937.485883 au, also lower than the sum (-2937.453105 au) of energies from TMG and HBr molecules, predicting that the ion pair molecule would be 0.032778 au more stable than separate TMG and HBr molecules. The size of these energies shows that the isolated [tmgH]+ · · · Br-/ Br-. · · · [tmgH]+ dimer ion pair molecule came out with nearly two times better stabilization per unit formula than the monomer ion pair. The Gaussian software allows for calculation of vibrational spectra. The IR and Raman spectra calculated for the ab initio optimized isolated [tmgH]+ · · · Br-/Br- · · · [tmgH]+ dimer ion pair molecule are shown in Figure 3. New numerical results for [tmgH]2Br2 and [tmgH]Br ion pair molecules are summarized in Tables S3 and S4 of the Supporting Information, and calculated IR and Raman spectra of the [tmgH]+ ion can be found in our previous work.14 Table S3 of the Supporting Information shows experimental bands assigned to approximate group vibrations based on the calculated normal modes (no imaginary frequencies). It is seen that the NH2 vibrations in the two [tmgH]+ ions are much influenced by the presence of the bromide ions: before bonding to the Br- anions, NH2 symmetric and asymmetric stretchings were predicted at 3609.6 and 3731.1 cm-1,14 whereas intense bands were predicted at 3130.2 (IR), 3150.4 (Raman), 3194.1 (Raman), and 3250.2 cm-1 (IR) after dimer ion pair formation (see Table S3 of the

Figure 3. IR absorption (A) and Raman scattering spectra (B) of the [tmgH]2Br2 dimer ion pair molecule, calculated by DFT/6-311G+(d,p)/ B3LYP Gaussian modeling. Curves were arbitrarily scaled and lifted.

Supporting Information, mode nos. 121, 130-132). These NH2 symmetric and asymmetric stretching bands discussed above were observed at 3110 vs (IR), 3120 s (Raman), 3260 m (Raman), and 3310 s cm-1 (IR) for crystals, according to our interpretation. Spectra. Experimental spectra of crystalline [tmgH]2Br2 were obtained in this work as shown in Figure 4. IR absorption (top) and Raman scattering spectra (bottom curve) were arbitrarily scaled and lifted. The spectra of the chloride and bromide compounds are much alike, as may be seen by comparing Figure 4 to Figure 3 in ref 14. For the Raman spectra in aqueous solution (the solubility in water is not high), a satisfactory spectrum could be obtained. The low solubility is attributed to the formation of the dimeric ion pairs. When heated to temperatures higher than 135 °C in an evacuated ampule, the compound melted, and like for the chloride,14 the melt gave strong red fluorescence when trying to record the Raman spectrum, even though not much sample

[((CH3)2N)2CdNH2]+Br- Structure and Raman Spectra

Figure 4. Experimental IR absorption (A) and Raman scattering spectra (B) of crystalline [tmgH]2Br2 obtained in this work. Curves were arbitrarily scaled and lifted. For position of bands see Table S3 of the Supporting Information.

Figure 5. Comparison of experimental Raman scattering spectra of (A) crystalline [tmgH]2Br2, (B) a drop of distillate over the melt, (C) a nearly saturated solution in water with water subtracted, and (D) the calculated spectrum. Curves were arbitrarily scaled and lifted. Raman signals from the silica are seen at around 500 cm-1 in (B). Noise due to water band subtraction is seen at around 3000-3600 cm-1 in (C).

degradation was seen for the liquid. However, it was possible to record a good spectrum from a condensed drop at 180 °C inside the ampule in the furnace. Raman spectra are shown in Figure 5. When comparing the measured Raman spectra of different states (A-C in Figure 5), the accordance between experiments is quite good. The Raman signals from silica glass are seen at around 500 cm-1 in (B), and noise due to water band subtraction is seen in (C) at around 3000-3600 cm-1. The accordance with the calculated spectrum (D in Figure 5) is not perfect, probably due to the simplicity of the model. We conclude that Figure 5 can be taken to indicate that the compound retains its structure in the different solutions (with four N-H+ · · · Br- hydrogen bonds).

J. Phys. Chem. A, Vol. 114, No. 50, 2010 13179

Figure 6. Raman spectra of (A) the gas phase at 225 °C over about 200 mg of liquid [tmgH]Br in evacuated sealed 10 mm diameter ampule and (B) the same for [tmgH]Cl. Also shown are spectra of (C) dimethylcyanamide gas saturated at approximately 200 °C and (D) the liquid at 25 °C. Gas phase spectra were obtained by long time exposures (2 h/frame at 2 W of 532 nm excitation power). Curves were arbitrarily scaled and lifted. The ab initio calculated geometry of the dimethylcyanamide molecule and its spectrum are shown (E). Asterisks indicate bands due to H2O and N2, and also indications of CO2 and O2 can be seen in (C) at high zoom conditions. For position of bands, see Table 4.

Vapor Phase. Raman spectra of gases are often very weak.28 It was attempted to acquire the Raman spectrum of the gas phase by heating sealed silica vials containing the melted crystals. Accordingly, the gas-phase spectrum of the vapor over the [tmgH]Br melt at 225 °C was recorded over a long time at high excitation intensities (2-4 W of laser power). The resulting spectrum (shown as curve A in Figure 6) exhibited a peak at about 2229 cm-1, in addition to other features. Surprisingly it seemed as if the spectrum did come from the same gas as that over [tmgH]Cl (shown as curve B in Figure 6). The explanation previously given14 regarding the gas phase spectrum over [tmgH]Cl is not likely to be correct because a monomer [tmgH]Br ion pair molecule should vibrate with another spectrum. A DFT ab initio calculation confirmed this: The [tmgH]Br molecule is stable with an optimized total energy of -2937.485883 au, but the calculated vibration of the NH · · · Br is 2556 cm-1 (see Table S4 of the Supporting Information) to be compared with the NH · · · Cl value of 2188 cm-1.14 Instead, decomposition of the samples to form a common gas product is a likely hypothesis. We considered what molecules could have a strong Raman band at 2229 cm-1. According to the book of Socrates,29 the most likely molecules to give strong Raman signals in that range are the aliphatic nitriles, RsCtN. Methyl nitrile, CH3sCtN could not be the particular molecule because it has its strong Raman band at 2249 cm-1 according to our own observations, differing from the 2229 cm-1. Dimethylcyanamide (DMCA). A likely explanation could be a formation of DMCA molecules in the gas phase over heated [tmgH]Cl and [tmgH]Br. The reaction would involve the breakdown of the 1,1,3,3-tetramethylguanidine and the formation of the dimethylammoniumbromide (or chloride) according to the following reaction (eq 1):

13180

J. Phys. Chem. A, Vol. 114, No. 50, 2010

Berg et al.

TABLE 4: Ab Initio Calculated Vibrational Spectra for the Dimethylcyanamide Molecule, ((CH3)2N-CN), Assignments, and Experimental Data for Spectra (See Figure 6)a

mode no.

wave number/cm-1

IR absorption/km mol-1

Raman activity/Å4 AMU-1

depolarization ratio

description of normal mode assigment

symmetry

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

152.5 177.6 210.0 221.4 365.8 542.5 639.1 772.7 1070.2 1072.2 1117.9 1166.0 1232.8 1358.7 1447.9 1470.9 1483.2 1495.0 1501.4 1522.6 2316.9 2998.7 3001.6 3083.7 3085.2 3134.8 3135.8

1.38 0.40 5.14 4.34 0.82 7.69 2.66 8.44 18.38 43.66 3.34 8.29 16.66 25.53 0.09 3.99 4.64 28.26 10.80 11.14 210.27 34.69 81.51 41.84 2.12 4.78 11.88

2.29 0.41 4.02 0.45 1.33 2.10 0.61 10.25 2.72 2.30 1.89 0.95 0.40 0.49 4.21 4.40 14.09 2.26 1.59 12.15 108.36 21.26 299.70 157.27 20.68 39.51 75.41

0.72 0.75 0.75 0.68 0.70 0.73 0.75 0.10 0.75 0.75 0.75 0.71 0.75 0.41 0.75 0.65 0.75 0.74 0.75 0.75 0.24 0.75 0.05 0.37 0.75 0.75 0.67

skeleton oopl bend CH3 twist skeleton ipl bend skeleton oopl bend skeleton sci NCN oopl bend NCN ipl bend NC3 sym str CH3 rock CH3 rock CH3 rock CH3 rock NC2 asym str NC str + CH3 rock CH3 umbrella ooph CH3 umbrella iph CH3 def CH3 def CH3 def CH3 def NCN str CH3 sym str ooph CH3 sym str iph CH3 asym str iph CH3 asym str ooph CH3 asym str ooph CH3 asym str iph

A′ A′′ A′′ A′ A′ A′ A′′ A′ A′′ A′ A′′ A′ A′′ A′ A′′ A′ A′′ A′ A′′ A′ A′ A′′ A′ A′ A′′ A′′ A′

observed IR liquid30 25 °C cm-1

368 m 530 s 630 m 763 s 1060 vs 1147 w 1207 m 1338 s 1418 m 1440 s sh 1455 s 1483 s 2090 w 2160 m sh 2210 vs 2820 m 2858 sh 2915 s 2965 s 3002 m sh

observed Raman satd. gas at 200 °C cm-1

366 w 501 vw 763 s 828 w 1056 w

1452 m 2092 vw 2229 vs 2822 s 2866 s 2913 w 2951 s 2979 s 3021 w sh

a Model: DFT/RB3LYP/6-311+G(d,p). Total energy E(RB + HF - LYP): -227.468050 au. Zero point correction: 0.090617 au. Dipole moment: 5.0302 D. Symmetry: Cs. No. of imaginary frequencies: 0. Abbreviations for approximate vibrations: asym ) asymmetric; bend ) bending; iph ) in phase; ipl ) in plane; ooph ) out of phase; oopl ) out of plane; sci ) scissoring; str ) stretching; sym ) symmetric; twist ) twisting. Codes for band intensity: m ) medium; s ) strong; sh ) shoulder; sp ) sharp; v ) very; w ) weak; br ) broad.

((H3C)2N)2C)NH + HBr f (H3C)2NCN + [(H3C)2NH2]Br (1) A sample of authentic commercial DMCA was transferred to ampules that were sealed. Recorded spectra of the liquid and the gas phase are shown in Figure 6 as curves C and D. A convincing accordance to the curves A and B is obvious. In A and B, the dimethylammonium bromide (or chloride) formed by the reaction (eq 1) probably stayed in the liquid. The DMCA molecule was also subjected to an ab initio calculation of the optimum structure and vibrational spectra. The resulting Raman spectrum is included in Figure 6 as curve E, and the details are given in Table 4. The correspondence between the other curves in Figure 6 seems quite satisfactory. Our spectra for DMCA fit quite well the theoretical results (force field calculation) and previous experiments (IR and Raman on the liquid).30 To investigate if DMCA could really be formed, series of mass spectra were recorded in the following way. Crystals of [tmgH]Br were placed in a u-tube in a furnace are a stream of dry helium was passing through. The furnace was heated to a preselected temperature, and the gas was led to a quadrupole mass spectrometer. At 205 °C, only m/z (mass/charge) peaks for 4, 15-16, and 69 were observed, originating from ionized helium, nitrogen, and DMCA having lost one H, respectively. After further heating at 215 °C, more and stronger peaks were seen for m/z values of 42 and 70. The 42 peak could correspond to ionized CH3NCH, and the 70 peak could be DMCA. At 225 °C, further m/z peaks at 40, 41, and 43 were seen, possibly originating from dimethylamine, and peaks at 53, 58, and 59 also appeared while peaks at 69 and 70 intensified. In conlusion,

the hypothesis that [tmgH]Br can decompose into dimethylcyanamide, dimethylamine, and HBr seems justified. Conclusions The crystal structure of the protic compound [tmgH]Br has been solved and found to be very similar to that of [tmgH]Cl.16 The [tmgH]Br, in different states, was studied by a combination of Raman spectroscopy and ab initio MO calculations. The DFT calculations were found to help the evaluation of chemical information even though the model spectra are not in total agreement with the spectral observations. By taking into account the approximations involved, the limitations, and the likely presence of various unaccounted problems such as methyl group Fermi resonance, anharmonicity, and so forth, the numerical estimation of the calculated frequencies should not be taken too literately. It seems reasonable to conclude that no indication of the existence of a [tmgH]Br monomer ion pair molecule in the gas phase was found. Accordingly the findings in our previous report,14 claiming proof of the existence of a [tmgH]Cl monomer ion pair molecule in the gas phase should also be changed. Hence, it is concluded that the [tmgH]X compounds when heated to temperatures around 175-250 °C decompose forming easily vaporizing dimethylcyanamide gas. The Raman spectrum of the gas was studied and assigned. A characteristic strongly Raman active band was observed at about 2229 cm-1 (a CN stretching mode with a predicted value of ∼2317 cm-1). Acknowledgment. We are grateful to Prof. Irene Shim (DTU Chemistry, Lyngby, Denmark) for help with computational work

[((CH3)2N)2CdNH2]+Br- Structure and Raman Spectra and to Lykke Ryelund (University of Copenhagen, Denmark) for recording FT Raman spectra, respectively. Prof. Jo¨rg Sundermeyer (Philipps-Universita¨t Marburg, Germany) is thanked for suggesting formation of DMCA molecules in the gas phase. Grants from “Lundbeckfonden” (j.nr. 177/06), “Direktør Ib Henriksen’s Fond” and The Danish Agency for Science, Technology, and Innovation contributed funding for maintaining the Raman instrumentation (grant no. 09-065038/FTP). Supporting Information Available: Positional, equivalent isotropic, and anisotropic thermal parameters are listed in Tables S1 and S2. Table S3 lists calculated vibrational spectra and assignments for the [tmgH]2Br2 dimer ion-pair molecule, and Table S4 lists the spectrum of the [tmgH]Br molecule. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Berg, R. W. Monatsh. Chem. 2007, 138, 1045–1075. (2) Earle, M. J.; Esperanc¸a, J. M. S. S.; Gilea, M. A.; Canongia Lopes, J. N.; Rebolo, L. P. N.; Magee, J. W.; Seddon, K. R.; Widegren, J. A. Nature 2006, 439, 831–834. (3) Zaitsau, D. H.; Kabo, G. J.; Strechan, A. A.; Paulechka, Y. U.; Tschersich, A.; Verevkin, S. P.; Heintz, A. J. Phys. Chem. A 2006, 110, 7303–7306. (4) SMART and SAINT, Area Detector Control and Integration Software, version 5.054; Bruker Analytical X-ray Instruments Inc.: Madison, WI, 1998. (5) Sheldrick, G. M. SHELXTL, Structure Determination Programs, version 6.12; Bruker Analytical X-ray Instruments Inc.: Madison, WI, 2001. (6) Armstrong, J. P.; Hurst, C.; Jones, R. G.; Licence, P.; Lovelock, K. R. J.; Satterley, C. J.; Villar-Garcia, I. J. Phys. Chem. Chem. Phys. 2007, 9, 982–990. (7) Strasser, D.; Goulay, F.; Kelkar, M. S.; Maginn, E. J.; Leone, S. R. J. Phys. Chem. A 2007, 111, 3191–3105. (8) Leal, J. P.; Esperanc¸a, J. M. S. S.; Minas da Piedade, M. E.; Canongia Lopes, J. N.; Rebolo, L. P. N.; Seddon, K. R. J. Phys. Chem. A 2007, 111, 6176–6182. (9) Akai, N.; Kawai, A.; Shibuya, K. Chem. Lett. 2008, 37, 256–257. (10) Vitorino, J.; Leal, J. P.; Minas da Piedade, M. E.; Canongia Lopes, J. N.; Esperanc¸a, J. M. S. S.; Rebolo, L. P. N. J. Phys. Chem. B 2010, 114, 8905–8909. (11) Berg, R. W.; Canongia Lopes, J. N.; Ferreira, R.; Rebelo, L. P. N.; Seddon, K. R.; Tomaszowska, A. A. J. Phys Chem. A, 2010, 114, 1083410841. (12) Huang, J.; Riisager, A.; Wasserscheid, P.; Fehrmann, R. Chem. Commun. 2006, 4027–4029.

J. Phys. Chem. A, Vol. 114, No. 50, 2010 13181 (13) Huang, J.; Riisager, A.; Berg, R. W.; Fehrmann, R. J. Mol. Catal. A 2008, 279, 170–176. (14) Berg, R. W.; Riisager, A.; Fehrmann, R. J. Phys. Chem. A 2008, 112, 8585–8592. (15) Berg, R. W. Z. Naturforsch. 2007, 62a, 157–168. (16) Fischer, A. K.; Jones, P. G. Acta Crystallogr. 2002, E 58, o218– o219. (17) Anderson, M. L.; Hammer, R. N. J. Chem. Eng. Data 1967, 12, 442–447. (18) Berg, R. W.; Deetlefs, M.; Seddon, K. R.; Shim, I.; Thompson, J. M. J. Phys. Chem. B 2005, 109, 19018–19025. (19) Brooker, M. H.; Berg, R. W.; von Barner, J. H.; Bjerrum, N. J. Inorg. Chem. 2000, 39, 4725–4730. (20) Berg, R. W.; Maijo´ Ferre´, I.; Cline Scha¨ffer, S. J. Vibrat. Spectrosc. 2006, 42, 346–352. (21) Berg, R. W.; Nørbygaard, T. Appl. Spectrosc. ReV. 2006, 41, 165– 183. (22) Gilbert, B. P.; Berg, R. W.; Bjerrum, N. J. Appl. Spectrosc. 1989, 43, 336–341. (23) Frisch, M. J.; 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 03W, Revision E.01; Gaussian, Inc.: Pittsburgh, PA, 2006. (24) Boyd, S. L.; Boyd, R. J. J. Chem. Theory Comput. 2007, 3, 54–61. (25) Fairchild, S. Z.; Bradshaw, C. F.; Su, W.; Guharay, S. K. Appl. Spectrosc. 2009, 63, 733–741. (26) Berg, R. W.; Riisager, A.; Van Buu, O. N.; Fehrmann, R.; Harris, P.; Tomaszowska, A. A.; Seddon, K. R. J. Phys. Chem. A 2009, 113, 8878– 8886. (27) Allen, F. H. Acta Crystallogr. B 2002, 58, 380–388. (28) Berg, R. W.; Papatheodorou, G. N. Inorg. Chim. Acta 1980, 45, L211–L212. (29) Socrates, G. Infrared and Raman Characteristic Group Frequencies, 3rd ed.; J. Wiley & Co.: Chichester, U.K., 2001; p 36, Chapter 4.00 (30) Ohnet, M.-N.; Jouan, M.; Menard, G.; Spasojevi’c-de Bire´, A.; Dao, N. Q.; Fischer, H.; Reindl, D. Spectrochim. Acta 1996, A 51, 505–526.

JP107152X