Molecular structures of 2-methoxyphenol and 1,2-dimethoxybenzene as studied by gas-phase electron diffraction and quantum chemical calculations

Journal of Molecular Structure 933 (2009) 132–141

Contents lists available at ScienceDirect

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

Molecular structures of 2-methoxyphenol and 1,2-dimethoxybenzene as studied by gas-phase electron diffraction and quantum chemical calculations Olga V. Dorofeeva a,*, Igor F. Shishkov a, Nikolai M. Karasev a, Lev V. Vilkov a, Heinz Oberhammer b a b

Department of Chemistry, Lomonosov Moscow State University, Moscow 119991, Russia Institut für Physikalische und Theoretische Chemie, Universität Tübingen, 72076 Tübingen, Germany

a r t i c l e

i n f o

Article history: Received 30 April 2009 Received in revised form 6 June 2009 Accepted 7 June 2009 Available online 13 June 2009 Keywords: 2-Methoxyphenol 1,2-Dimethoxybenzene Gas-phase electron diffraction Molecular structure Conformation Quantum chemical calculations

a b s t r a c t The molecular structure and conformational properties of 2-methoxyphenol (2-MP) and 1,2-dimethoxybenzene (1,2-DMB) have been studied by gas-phase electron diffraction (GED) and quantum chemical calculations (B3LYP and MP2 methods with 6-31G(d,p) and cc-pVTZ basis sets). Of the three stable conformers predicted for 2-MP by quantum chemical calculations, the lowest energy form possesses a planar structure with an intramolecular hydrogen bond between phenolic hydrogen and methoxy oxygen (anti– syn conformer of Cs symmetry). The calculated concentration of this conformer is about 99% and this is confirmed by the GED data. Quantum chemical calculations predict three stable conformers for 1,2DMB: anti–anti (C2v symmetry), anti–gauche (C1 symmetry), and gauche–gauche (C2 symmetry). The GED data were well reproduced for the mixture of these conformers with the relative abundance of 50 ± 12%, 36 ± 16%, and 14%, respectively. A similarly good agreement is also obtained for the single anti–gauche conformer. The experimental structural parameters agree well with results of B3LYP/cc-pVTZ calculations. Ó 2009 Elsevier B.V. All rights reserved.

1. Introduction Experimental and theoretical studies for anisole, C6H5OCH3, lead to rather contradicting results concerning the conformational properties of this compound. Photoelectron spectra [1] and NMR measurements [2] have been interpreted in terms of a mixture of planar and perpendicular conformers (O–CH3 bond in the ring plane or perpendicular to it), in agreement with several quantum chemical calculations using the HF approximation [2–5]. On the other hand, the gas-phase electron diffraction (GED) [6] leads to single conformer with near-planar structure. Microwave spectroscopy [7] and high-level quantum chemical calculations which include electron correlation effects [8], clearly demonstrate the presence of a single conformer with planar orientation of the O–CH3 bond. The barrier to internal rotation around the C(sp2)–O bond is predicted to be between 2 and 3 kcal/mol. Similarly, the conformational properties of 1,2-dimethoxybenzene (1,2-DMB), or veratrole, C6H4(OCH3)2, have been subject of much controversy over the years. These properties have attracted considerable interest, since this compound is a part of many natural products with biological activities, such as papaverine, denopamine, verapamil, glaucin, etc. [9]. The bioactivity of these compounds is closely related to their conformational properties * Corresponding author. Tel.: +7 495 939 4021; fax: +7 495 939 1308. E-mail addresses: [email protected], [email protected] (O.V. Dorofeeva). 0022-2860/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.molstruc.2009.06.009

[10,11]. The conformation of 1,2-DMB is characterized by the two torsional angles around the C(sp2)–O bonds, u1(C2–C1–O7–C9) and u2(C1–C2–O8–C10). For atom numbering see Fig. 1. The four conformations of 1,2-DMB shown in Fig. 2, possessing C2v, Cs, C1 or C2 symmetry, are most feasible. In the C2v structure with the two O–CH3 bonds antiperiplanar to each other both torsional angles are 180°. From photoelectron spectroscopic studies it was concluded that gaseous 1,2-DMB exists predominantly as nonplanar conformer with either C1 symmetry [11] or with C2 symmetry and mean torsional angles of about 80° [12]. On the other hand, optical spectroscopic data were interpreted in terms of a planar structure (C2v or Cs symmetry) [13]. A similar controversy exists for the liquid phase and for solutions. Studies involving partition coefficients [11], dipole moments [14,15], dielectric relaxation times [16] or 17O NMR spectra [17,18] result in nonplanar conformations (C2 or C1 symmetry). 1H and 13C NMR spectra, however, are consistent with a planar structure [19] as well as rotationally resolved fluorescence excitation spectra which are interpreted in terms of a planar conformation with C2v symmetry [20]. 1H NMR spectra in a nematic solvent result in a mixture of planar C2v (75%) and nonplanar C1 conformers (25%) [21]. From the splitting of torsional vibrations in the low-temperature Raman spectrum the presence of a mixture of planar and nonplanar conformers was concluded for solid 1,2-DMB [22]. A low-temperature crystallographic study results in the presence of a single conformer with near-planar structure close to C2v symmetry [23]. The C(sp2)–O torsional angles deviate by less than 10° from the exact

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141

133

Fig. 1. Most stable conformers of 2-MP and 1,2-DMB with atom numbering.

Fig. 2. Conformations of 2-MP and 1,2-DMB.

planar orientation. This result was confirmed by 13C CP MAS spectroscopy [23]. Several quantum chemical calculations have been reported so far in the literature. The HF/STO-3G method predicts the presence of three stable conformers with C2v, C1, and C2 symmetry. Using partially optimized geometries all three conformers possess very similar relative energies between 0.0 kcal/mol for the C1 structure and 0.2 kcal/mol for the C2 form [11]. Without geometry optimization preference of the planar C2v conformer is predicted with the nonplanar C1 and C2 conformers being slightly higher in energy by 0.5 and 1.2 kcal/mol, respectively [24]. In both calculations exactly perpendicular orientation of the O–CH3 bonds was assumed for nonplanar conformers. HF/D95 calculations predict as well three stable structures, however, with a strong preference of the nonplanar C2 conformer [18]. Optimized torsional angles u1 and u2 are 74.5°. C2v and C1 conformers are calculated to be higher in energy by 3.9 and 3.8 kcal/mol, respectively. On the other hand, an MP2/6-311G(d,p) study predicts the C2v and C1 conformers to be the only stable structures, both differing in energy by 0.16 kcal/mol [25]. Very recently B3LYP/6-311++G(d,p) calculations have been reported which result in three stable conformers with the C2v structure to be lowest in energy and the C1 and C2 forms higher by 1.1 and 2.4 kcal/mol, respectively [26].

Numerous studies of 2-methoxyphenol (2-MP), or guaiacol, C6H4(OH)(OCH3), concerning the strength of the intramolecular hydrogen bond have been published, see [27] and references therein. Most of these investigations suppose that the planar anti–syn form is the most stable conformer of 2-MP due to strong hydrogen bonding O7H10 (Figs. 1 and 2). The geometry of this conformer was optimized at the HF/STO-3G level of theory [28]. The potential energy surface of 2-MP as a function of out-of-plane rotation of methoxy and hydroxy group has been calculated using the molecular mechanics method [29]. The global minimum on this surface corresponds to anti–syn conformer. From mass-resolved excitation spectra of jet-cooled 2-MP [29], two rotational isomers, anti–syn and anti–anti were identified. The conformational behavior of 2-MP has been studied by ab initio (HF, MP2, and MP4) [30] methods. Three stable conformers of 2-MP, anti–syn, anti–anti, and gauche–anti (Fig. 2) exist according to the potential energy surface of 2-MP calculated at the HF/6-31G level [30], whereas the syn–anti conformation corresponds to a transition state. The relative stabilities of 2-MP conformers were determined by the HF, MP2, and MP4 methods with 6-31G(d,p), 6-311G(d,p), 6-311++G(2d,p), 6-311++G(2d,2p), and 6-311++G(df,p) basis sets. According to all calculations the anti–syn conformation is the global minimum on the conforma-

134

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141

Fig. 3. Calculated (MP2/6-31G(d)) energy surface in kcal/mol as a function of the two torsional angles around the C(sp2)–O bonds in 2-MP. Three minima occur for u1/u2 = 180°/0°, 180°/180°, and 60°/180°.

tional energy surface. The next conformers, anti–anti and gauche–anti, are by 4.5 and 5.0 kcal/mol less stable (the average values for MP2 and MP4 calculations are given), what suggests

Fig. 4. Calculated (MP2/6-31G(d)) energy surface in kcal/mol as a function of the two torsional angles around the C(sp2)–O bonds in 1,2-DMB. Energies larger than 4.5 kcal/mol, which occur in the region of small torsional angles u, have been set to 4.5 kcal/mol. The calculated energy at u1 = u2 = 0° is 20.4 kcal/mol.

the 99% concentration of the main conformer in the equilibrium mixture.

Table 1 Molecular structure of anti–syn and anti–anti conformers of 2-MP obtained by gas-phase electron diffraction and quantum chemical calculations. Parameter

Anti–syn

Anti–anti

GED rh1 (\h1 )

B3LYP/cc-pVTZ re (\e )

MP2/cc-pVTZ re (\e )

GED rh1 (\h1 )

B3LYP/cc-pVTZ re (\e )

Independent parameters C1–C2 C2–C3 C3–C4 C4–C5 C5–C6 C1–C6 C1–O7 C2–O8 C9–O7 O8–H10 (C–H)av C1–C2–C3 C2–C3–C4 C3–C4–C5 C2–C1–O7 C1–C2–O8 C1–O7–C9 C2–O8–H10 C–C–H O–C–H15 O–C–H16(17) C1–O7–C9–H16 u1(C2–C1–O7–C9) u2(C1–C2–O8–H10)

1.411(2)a 1.390(2)a 1.399(2)a 1.391(2)a 1.402(2)a 1.392(2)a 1.378(4)b 1.368(4)b 1.424(4)b 1.001(6)c 1.119(6)c 119.9(2)d 120.2(2)d 120.3(2)d 113.8(5)e 120.0(5)e 118.2(5)e –f –f –f –f –f 180.0 0.0

1.405 1.384 1.393 1.385 1.396 1.386 1.372 1.362 1.418 0.966 1.084 119.8 120.1 120.2 114.0 120.2 118.5 107.6 118.6–120.5 106.2 111.3 61.1 180.0 0.0

1.404 1.387 1.396 1.390 1.398 1.390 1.370 1.361 1.418 0.967 1.083 119.8 120.0 120.2 113.7 120.0 116.2 106.2 118.5–120.4 106.2 111.0 61.0 180.0 0.0

1.411(2)a 1.388(2)a 1.397(2)a 1.386(2)a 1.398(2)a 1.393(2)a 1.372(5)b 1.377(5)b 1.428(5)b 1.001(8)c 1.123(8)c 119.6(2)d 120.8(2)d 119.5(2)d 114.9(4)e 116.4(4)e 117.5(4)e –f –f –f –f –f 180.0 180.0

1.409 1.386 1.394 1.383 1.396 1.390 1.360 1.365 1.416 0.962 1.084 119.7 120.9 119.6 115.7 117.2 118.2 108.8 119.0–120.5 106.0 111.6 61.2 180.0 180.0

Dependent parameters C4–C5–C6 C1–C6–C5 C2–C1–C6

119.7(8) 120.4(11) 119.6(8)

120.1 119.8 120.1

120.2 119.5 120.3

120.5(9) 120.2(12) 119.5(9)

120.1 120.7 119.1

RL RS Rtot

3.2 4.7 4.0

5.2 5.2 5.2

Bond lengths are in Å, bond angles and torsional angles are in degrees. Values in parentheses are three times the standard deviations. Together with total value of the disagreement factor (Rtot), the R-factors (in %) are given for long (RL) and short (RS) camera distances. a Refined in one group. Differences between parameters in the group were assumed at the values from B3LYP/cc-pVTZ calculation. b See footnote a. c All C–H bond lengths were refined in one group; their average value is given in the Table. d See footnote a. e See footnote a. f Assumed at the values from B3LYP/cc-pVTZ calculation.

135

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141 Table 2 Equilibrium distribution of 1,2-DMB conformers at 333 K predicted from quantum chemical calculations.a Method

Conformer (symmetry)

u1

u2

B3LYP/6-31G(d)

Anti–anti (C2v) Anti–gauche (C1) Gauche–gauche (C2)

180 70.3 70.7

180 178.0 70.7

B3LYP/cc-pVTZ

Anti–anti (C2v) Anti–gauche (C1) Gauche–gauche (C2)

180 70.7 72.7

MP2/6-31G(d)

Anti–anti (C2v) Anti–gauche (C1) Gauche–gauche (C2)

MP2/cc-pVTZb

Anti–anti (C2v) Anti–gauche (C1) Gauche–gauche (C2)

DE 0

S333

DG333

p333

0.0 0.9 1.6

98.8 102.6 102.3

0.0 0.4 0.4

30 54 16

180 177.6 72.7

0.0 1.2 2.2

98.8 102.6 102.6

0.0 0.1 1.0

41 49 10

180 73.2 70.2

180 177.5 70.2

0.0 0.1 0.3

100.9 104.5 103.3

0.0 1.1 1.1

9 46 45

180 73.1 71.0

180 177.6 71.0

0.0 1.0 1.8

100.9 104.5 103.3

0.0 0.1 1.0

41 51 8

a Torsional angles u1 and u2 in degrees; E0 = Ee + ZPE (in kcal/mol), where Ee is the electronic energy, ZPE is the zero point energy correction calculated from scaled vibrational frequencies; S333 (in cal/(K mol)) is the entropy value at 333 K calculated using scaled frequencies and including the entropy of mixing (Rln 2) for conformers with C1 and C2 symmetry, which are present as an equimolar mixture of the two optical isomers; DGT(i)  DE0(i) + TDST(i) is the Gibbs free energy (kcal/mol); P pT ðiÞ ¼ ½expðDGT ðiÞ=RTÞ=½ i ½expðDGT ðiÞ=RTÞ is the mole fraction of conformer (%). b ZPE and vibrational frequencies from MP2/6-31G(d) calculation are used.

Table 3 Molecular structure of anti–gauche (C1) conformer of 1,2-DMB obtained by gas-phase electron diffraction and quantum chemical calculations. Parameter Independent parameters C1–C2 C2–C3 C3–C4 C4–C5 C5–C6 C1–C6 C1–O7 C2–O8 C9–O7 C10–O8 (C–H)av C1–C2–C3 C2–C3–C4 C3–C4–C5 C2–C1–O7 C1–C2–O8 C1–O7–C9 C2–O8–C10 C–C–H O–C–H15(18) O–C–H16(17,19,20) u1(C2–C1–O7–C9) u2(C1–C2–O8–C10) Mole fraction Anti–gauche Anti–anti Gauche–gauche Dependent parameters C4–C5–C6 C1–C6–C5 C2–C1–C6 RL RS Rtot

GED single conformer model rh1 (\h1 )

GED mixture of conformers rh1 (\h1 )

B3LYP/cc-pVTZ re (\e )

MP2/cc-pVTZ re (\e )

1.416(2)a 1.400(2)a 1.401(2)a 1.392(2)a 1.400(2)a 1.393(2)a 1.371(6)b 1.362(6)b 1.428(6)b 1.417(6)b 1.115(5)c 119.3(1)d 120.4(1)d 120.4(1)d 120.3(4)e 115.1(4)e 115.2(4)e 117.3(4)e –f –f –f 66.5(27) –f

1.411(10)a 1.395(10)a 1.396(10)a 1.387(10)a 1.395(10)a 1.388(10)a 1.375(9)b 1.366(9)b 1.431(9)b 1.421(9)b 1.105(6)c 119.4(5)d 120.6(5)d 120.5(5)d 121.1(7)e 115.9(7)e 115.9(7)e 118.1(7)e –f –f –f 65.2(55) –f

1.409 1.392 1.393 1.385 1.392 1.386 1.372 1.363 1.428 1.418 1.087 119.3 120.4 120.4 121.4 116.3 116.3 118.4 117.8–120.2 106.2, 106.0 110.6–111.6 70.7 177.6

1.409 1.395 1.397 1.389 1.395 1.389 1.370 1.361 1.428 1.418 1.085 119.3 120.2 120.5 120.7 116.0 112.8 116.2 117.7–120.4 106.4, 105.9 110.4–111.4 73.1 177.6

100

50(12) 36(16) 14

119.2(7) 121.2(11) 119.4(7) 3.3 5.0 4.2

118.6(16) 122.2(27) 118.7(16) 3.4 5.1 4.3

119.4 121.0 119.6

119.4 120.8 119.8

Bond lengths are in Å, bond angles and torsional angles are in degrees, mole fraction is in %. Values in parentheses are three times the standard deviations. Together with total value of the disagreement factor (Rtot), the R-factors (in %) are given for long (RL) and short (RS) camera distances. a Refined in one group. Differences between parameters in the group were assumed at the values from B3LYP/cc-pVTZ calculation. b See footnote a. c All C–H bond lengths were refined in one group; their average value is given in the Table. d See footnote a. e See footnote a. f Assumed at the values from B3LYP/cc-pVTZ calculation.

In the present study, the gas-phase structures and conformation of the title compounds have been investigated by means of GED

and quantum chemical calculations in order to provide more details about their geometries.

136

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141

ti–syn and anti–anti conformers and nonplanar gauche–anti conformer, respectively (Fig. 2). The fourth conformer, syn–anti (u1/ u2 = 0°/180°), is a transition state with one imaginary frequency. The anti–syn conformer is the global minimum according to all calculations, whereas the anti–anti and gauche–anti structures are the next stable conformers with relative energies of 4.2–5.0 kcal/mol and 5.3–5.6 kcal/mol, respectively, depending on the method of calculation. The mole fraction of these conformers estimated from the theoretical relative Gibbs free energies is expected to be less than 1%. This conclusion about stability of 2-MP conformers is in good agreement with previous results obtained by MP2 and MP4 methods [30]. The results of B3LYP and MP2 geometry optimization with ccpVTZ basis set for anti–syn and anti–anti conformers compared with the final experimental GED results are given in Table 1.

2. Quantum chemical calculations In the first step the two-dimensional energy surfaces as a function of two torsional angles u1 and u2 in steps of 30° were calculated by the MP2/6-31G(d) method. For minima detected on this surface, the geometry was fully optimized at the B3LYP and MP2 level of theory with basis sets 6-31G(d) and cc-pVTZ. The vibrational frequencies were calculated by all methods with the exception of MP2/cc-pVTZ. All calculations were performed with the Gaussian 03 program package [31]. 2.1. 2-Methoxyphenol Three minima with u1/u2 = 180°/0°, 180°/180°, and 60°/180° on the potential energy surface (Fig. 3) correspond to planar an-

Table 4 Molecular structure of anti–anti (C2v) conformer of 1,2-DMB obtained by gas-phase electron diffraction and quantum chemical calculations. Parameter Independent parameters C1–C2 C2–C3@C1–C6 C3–C4@C5–C6 C1–O7@C2–O8 C9–O7@C10–O8 (C–H)av C1–C2–C3@C2–C1–C6 C1–C6–C5@C2–C3–C4 C2–C1–O7@C1–C2–O8 C1–O7–C9@C2–O8–C10 C–C–H O–C–H15(18) O–C–H16(17,19,20) u1(C2–C1–O7–C9) u2(C1–C2–O8–C10) Mole fraction Anti–gauche Anti–anti Gauche–gauche Dependent parameters C4–C5 C3–C4–C5@C4–C5–C6 RL RS Rtot

GED single conformer model rh1 (\h1 )

GED mixture of conformers rh1 (\h1 )

B3LYP/cc-pVTZ re (\e )

MP2/cc-pVTZ re (\e )

1.414(4)a 1.387a 1.397a 1.372(5)b 1.428b 1.114(7)c 119.4(2)d 120.6d 115.4(4)e 118.0e –f –f –f 180.0 180.0

1.417(10)a 1.390a 1.399a 1.363(9)b 1.419b 1.105(6)c 119.6(5)d 120.7d 115.3(7)e 117.9e –f –f –f 180.0 180.0

1.414 1.387 1.397 1.360 1.416 1.087 119.4 120.6 115.6 118.3 119.4–120.0 106.0 111.6 180.0 180.0

1.413 1.391 1.399 1.358 1.416 1.085 119.6 120.4 115.3 116.0 119.4–120.2 105.9 111.4 180.0 180.0

100

50(12) 36(16) 14

1.383(23) 120.0(4) 5.4 6.1 5.7

1.404(36) 119.7(10) 3.4 5.1 4.3

1.381 120.0

1.385 120.0

Bond lengths are in Å, bond angles and torsional angles are in degrees, mole fraction is in %. Values in parentheses are three times the standard deviations. Together with total value of the disagreement factor (Rtot), the R-factors (in %) are given for long (RL) and short (RS) camera distances. a Refined in one group. Differences between parameters in the group were assumed at the values from B3LYP/cc-pVTZ calculation. b See footnote a. c All C–H bond lengths were refined in one group; their average value is given in the Table. d See footnote a. e See footnote a. f Assumed at the values from B3LYP/cc-pVTZ calculation.

Table 5 Experimental conditions of gas-phase electron diffraction experiment. 2-MP

Camera distance (mm) Nozzle temperature (K) Accelerating voltage (kV) Electron wavelength (Å) Number of films used Range of s valuea (Å1) Scale factor a

1,2-DMB

Long camera

Short camera

Long camera

Short camera

362.28 351 60 0.049585 2 4.0–18.0 0.716(9)

194.94 351 60 0.049718 3 8.0–33.0 0.742(10)

362.28 333 60 0.049595 2 4.0–18.0 0.714(9)

194.94 333 60 0.049750 2 8.0–32.8 0.712(11)

s = 4pk1sin h/2, where h is the scattering angle and k is the electron wavelength.

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141

2.2. 1,2-Dimethoxybenzene Three minima with u1/u2 = 180°/180°, 70°/180°, and 70°/70° on the potential energy surface (Fig. 4) correspond to three stable 1,2-DMB conformers of C2v, C1, and C2 symmetry, respectively (Fig. 2). The conformer of Cs symmetry (u1/u2 = 180°/0°) does not correspond to a minimum on the energy surface and possesses an imaginary vibrational frequency. The existence of three stable conformers predicted by the MP2/6-31G(d) method is in agreement with the results of previous HF and B3LYP calculations [11,18,24,26], but is in contrast to the MP2/6-311G(d,p) calculation [25] which predicts only two stable forms. The relative stability of these structures depends strongly on the computational method. Torsional angles, relative energies, entropies, relative Gibbs free energies and contributions of the three conformers at the temperature of the experiment are summarized in Table 2. All methods predict the C2v conformer to be lowest in energy, except for the MP2/6-31G(d) method, which gives a slightly lower energy for C2 conformer. The Gibbs free energies, which determine the composition of the mixture, differ considerably from the relative energies due to difference in the entropic terms of the conformers. As is seen from Table 2, the mole fraction of the C1 conformer is largest in all predicted conformational compositions, whereas the relative contributions of C2v and C2 conformers depend strongly on the computational method. Therefore, different mixtures of the three possible conformers were considered in this study together with the geometric structures of single conformers. The results of B3LYP and MP2 geometry optimization with ccpVTZ basis set for the most abundant anti–gauche conformer of C1 symmetry compared with the final experimental GED results are given in Table 3. Table 4 contains the corresponding results for the anti–anti conformer of C2v symmetry, whose contribution to the equilibrium distribution is also rather substantial according to all of quantum chemical calculations, except the MP2/6-31G(d) method (see Table 2).

137

position of hydrogen atoms in the methyl groups, are not given in Tables 1, 3, and 4. Their values are close to 180° and 60°, which correspond to staggered conformation observed usually for methyl groups. The choice of independent parameters is slightly different for 1,2-DMB conformers (Tables 3 and 4) because of their different symmetry. The amplitudes of vibrations and harmonic vibrational corrections were calculated from the B3LYP/cc-pVTZ force field using the SHRINK program [33]. The values of vibrational amplitudes were not refined. 3.2.1. 2-Methoxyphenol The results of the least squares refinement for the planar anti– syn and anti–anti conformers of Cs symmetry are given in Table 1, where the corresponding values from B3LYP/cc-pVTZ and MP2/

3. Electron diffraction analysis 3.1. Experimental

Fig. 5. Experimental (open cycles) and theoretical (solid line) molecular intensities sM(s) and the difference curve for the anti–syn conformer of 2-MP.

Commercial samples of 2-MP and 1,2-DMB with purity of 98% and 99% were obtained from Alfa Aesar GmbH & Co. KG and Aldrich Chemical Co., respectively, and used without further purification. Electron diffraction intensities were recorded using the electron diffraction apparatus at Lomonosov Moscow State University. Information about the experimental conditions for all data sets used in the present investigation is given in Table 5. The wavelength of electrons was calibrated with CCl4 scattering patterns. The optical densities were measured using an EPSON PERFECTION 4870 PHOTO scanner and the data were processed with a program system UNEX [32]. The resulting modified intensity curves are shown in Figs. 5 and 6. 3.2. Structural refinements The analysis of GED data was carried out by applying the least squares method to the molecular intensities using UNEX program [32]. In this program, the molecular geometry is specified in a format of a Z-matrix. The independent parameters used in the structural analysis of 2-MP and 1,2-DMB are given in Tables 1, 3, and 4. Bond lengths (C–H/O–H, C–C, C–O) and bond angles (C–C–C, C–C– O/C–O–C) were refined in groups with the differences constrained to calculated values (see Tables 1, 3, and 4). For C–H bond lengths, the average values are shown in these Tables, however, these parameters were refined by constraining the differences between them to calculated values. The dihedral angles, which define the

Fig. 6. Experimental (open cycles) and theoretical (solid line) molecular intensities sM(s) and the difference curve for the mixture of anti–gauche and anti–anti conformers of 1,2-DMB.

138

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141

cc-pVTZ calculations are also presented. Initial values of geometrical parameters and the geometric constraints were taken from both theoretical calculations. For the anti–syn conformer the best agreement between theory and experiment with the R-factor of 4.0% was obtained with the values for angles of hydrogen atoms and constraints of difference in the principal parameters accepted from B3LYP/cc-pVTZ calculation. The use of MP2/cc-pVTZ parameters led to a slightly worse agreement between experimental and calculated GED intensities (R = 5.5%). The experimental bond distances and bond angles shown in Table 1 agree with calculated values practically within their experimental error limits. Two planar conformers of 2-MP, the anti–syn and anti–anti forms, differ only in the position of the phenolic hydrogen atom. Since the GED data are poorly sensitive to the position of hydrogen atoms, it is not surprising that the anti–anti conformer also agrees well with the GED experiment (see Table 1 and Fig. 7). Therefore, according to GED data alone, it is impossible to give an unambiguous preference to one or the other of the two planar conformers. On the other hand, although a rather low R-factor was obtained also for the nonplanar gauche–anti conformer (R = 6.3%), there is a conspicuous difference between experimental and theoretical radial distribution curves in the 3.0–5.5 Å region for this model (Fig. 7). This suggests that the gauche–anti conformer can be excluded on the basis of the GED data.

Fig. 7. Comparison between experimental (open cycles) and theoretical (solid line) radial distribution curves f(r) for anti–syn, anti–anti, and gauche–anti conformers of 2-MP. The R-factors of 4.0%, 5.2%, and 6.3% correspond to three above models.

The resulting radial distribution curve for the anti–syn conformer, which is in best agreement with results of both GED analysis and quantum chemical calculations, is shown in Fig. 8. 3.2.2. 1,2-Dimethoxybenzene In the preliminary stage, the conformational analysis was carried out by assuming the existence of only a single conformer. Although rather low R-factors were obtained for the C1, C2v, and C2 conformers (4.2%, 5.4%, and 5.4%, respectively), there is a distinctive difference between experimental and theoretical radial distribution curves in the region which depends on the orientation of the CH3 groups (Fig. 9), especially for the two latter conformers. The best fit to the experimental data is obtained with the C1 conformer, in agreement with its highest concentration in the equilibrium mixture according to quantum chemical calculations (see Table 2). Models with different composition of three stable conformers of 1,2-DMB were used in further refinements. Only the molecular parameters of the major conformer were refined in the analysis and the values of the other species were deduced by adding the calculated differences to the refined values of the major conformer. The mole fractions of the C1 and C2v conformers were determined as additional adjustable parameters. The best agreement between theory and experiment was obtained for the models with C1 and C2v conformers dominating. Because of strong correlation between the mole fractions of C1 and C2v conformers, their values could not be determined with high accuracy. However, within the experimental uncertainties, the GED data are in agreement with B3LYP/ cc-pVTZ and MP2/cc-pVTZ calculations and do not support the results of MP2/6-31G(d) calculations, according to which the equilibrium mixture consists mainly of C1 and C2 conformers (Table 2). The final results of structural analysis for the models of single C1 and C2v conformers and the mixture of three conformers are given in Tables 3 and 4, where the structural parameters obtained from

Fig. 8. Experimental (open cycles) and theoretical (solid line) radial distribution curves f(r) with the difference curve for the anti–syn conformer of 2-MP. The distance distribution is indicated by vertical bars; unassigned bars at r > 2.5 Å are CH and OH distances.

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141

139

Fig. 9. Comparison between experimental (open cycles) and theoretical (solid line) radial distribution curves f (r) for anti–gauche (symmetry C1), anti–anti (symmetry C2v), gauche–gauche (symmetry C2) conformers and the mixture of three conformers of 1,2-DMB. The R-factors of 4.2%, 5.4%, 5.4%, and 4.3% correspond to four above models.

the B3LYP/cc-pVTZ and MP2/cc-pVTZ calculations are also presented. The initial values of geometrical parameters were taken from both theoretical calculations. As in the case of 2-MP, the B3LYP/cc-pVTZ geometric constrains and values of nonrefined parameters associated with hydrogen atoms led to a slightly better agreement between theory and GED data. The structural parameters of anti–gauche (C1) conformer obtained for single conformer model and for mixture of conformers are close to each other and to results of theoretical calculations (Table 3). Good agreement between the experimental and theoretical parameters is also observed for the anti–anti (C2v) conformer, with the exception of C–O bond lengths, whose values are overestimated by 0.012 Å in single conformer model compared to B3LYP/cc-pVTZ values.

A much worse agreement between the GED and theoretical parameters was obtained for gauche–gauche (C2) conformer in the case of a single conformer model. A low R-factor (5.4%) for this model was achieved by substantial changes in some parameters as compared to theoretical ones. These differences amount to 0.02 Å for C9–O7 bond length, 3° for C–C–O angle, 2° for C–O–C angle, and 16° for torsional angle u1 = u2. Therefore, we conclude that the C2 conformer agrees poorly with GED data, whereas the single C1 conformer model or the mixture of conformers give equally good agreement between theory and experiment. The resulting radial distribution curve for the model of conformer mixture, which is in the best agreement with results of both GED analysis and quantum chemical calculations, is given in

140

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141 Table 6 Comparison of structural parameters of 2-MP, 1,2-DMB, phenol, and anisole obtained by GED. Parameter 2

C(sp )–OH C(sp2)–OCH3 3

1.424(4)

C1–C2–O8(H) C1–C2–O8(CH3) C2–C1–O7(CH3) C–O–C

120.0(5)

a

Fig. 10. To simplify the figure, the distance distribution is given for symmetrical C2v conformer. 4. Results and discussion Among the three stable conformers predicted for 2-MP by quantum chemical calculations the lowest energy form possesses a planar structure with intramolecular hydrogen bond between phenolic hydrogen and methoxy oxygen (anti–syn conformer of Cs symmetry). The strength of this hydrogen bond is demonstrated by the short OH distance of 2.103 Å and by the very small C2–C1– O7 angle (113.8(5)°) which demonstrates a strong OH attraction. According to theoretical calculations carried out previously [30] and in this work, the concentration of this conformer is about 99% and it shows the best agreement with GED data (Table 1). Although all theoretical calculations predict three stable conformers of C2v, C1, and C2 symmetry for 1,2-DMB, their relative abundance depends on the computational method. According to B3LYP/6-31G(d), B3LYP/cc-pVTZ, MP2/cc-pVTZ (Table 2) and B3LYP/6-311++G(d,p) [26] calculations, the most abundant conformers are those of C2v and C1 symmetry. However, the MP2/6-31G(d) method (Table 2) predicts the conformers of C1 and C2 symmetry to be the most abundant. The best agreement with GED data was obtained for C1 conformer or for a mixture with the prevailing amount of C1 and C2v conformers (Tables 3 and 4). Therefore, GED data do not agree with results of MP2/6-31G(d) calculations. Unfortunately, because of close distance distributions in C1 and C2v conformers, accurate information about abundance of these conformers could not be obtained by GED. The geometry of 2-MP and 1,2-DMB calculated at the B3LYP/ccpVTZ level is in good agreement with the structural parameters obtained from the GED analysis (Tables 1, 3, and 4). Comparison of some structural parameters of 2-MP and 1,2-DMB with those of phenol and anisole is given in Table 6. Substitution of the orthohydrogen atom at the benzene ring in phenol or anisole by a meth-

1.368(4) 1.378(4)

C(sp )–O

b

Fig. 10. Experimental (open cycles) and theoretical (solid line) radial distribution curves f(r) with the difference curve for the mixture of conformers of 1,2-DMB. The distance distribution for the symmetric anti–anti conformer is indicated by vertical bars; unassigned bars at r > 4.0 Å are CH and OH distances.

2-MP

113.8(5) 118.2(5)

1,2-DMB

Phenola

Anisoleb

1.380(4) 1.366(9) 1.375(9) 1.421(9) 1.431(9)

1.361(15) 1.423(15) 121.2(12)

115.9(7) 121.1(7) 118.1(7) 115.9(7)

120.0(20)

Ref. [34]. Ref. [6].

oxy group causes changes in some geometric parameters. As evident from Table 6, the C–O bond of the hydroxyl group in 2-MP is by 0.012 Å shorter than that in phenol, whereas the length of the C(sp2)–O bond in the methoxy group is 0.017 Å longer that that in anisole. At the same time, the C(sp2)–O bond lengths in 1,2-DMB are close to that in anisole. The values of the C(sp3)–O bond lengths are very similar in all compounds. The C–O–C angles in 2-MP (118.2°) and 1,2-DMB (118.1° and 115.9°) are somewhat smaller than that in anisole (120°). These angles for the CH3 group lying in the plane of benzene ring are very close in both molecules. Acknowledgments This research was supported by the Russian Foundation for Basic Research under Grant Nos. 08-03-00507 and 09-03-91340 and by Deutsche Forschungsgemeinschaft under Grant No. DFG 436 RUS 113/69/0-7. We are grateful to Vasiliy G. Zverev for processing the electron diffraction data. References [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]

H. Friege, M. Klessinger, Chem. Ber. 112 (1979) 1614. T. Schaefer, R. Sebastian, Can. J. Chem. 67 (1989) 1148. T. Schaefer, G.H. Penner, J. Mol. Struct. (Theochem.) 152 (1987) 179. M. Klessinger, A. Zywietz, J. Mol. Struct. (Theochem.) 90 (1982) 341. D.C. Spellmeyer, P.D.J. Gootehuis, M.D. Miller, L.F. Kuyper, P.A. Kollmann, J. Phys. Chem. 94 (1990) 4483. H.M. Seip, R. Seip, Acta Chem. Scand. 27 (1973) 4024. M. Onda, A. Toda, S. Mori, I. Yamaguchi, J. Mol. Struct. 144 (1986) 47. S. Tsuzuki, H. Houjou, Y. Nagawa, K. Hiratani, J. Phys. Chem. A 104 (2000) 1332. The Merck Index, fourteenth ed., Merck & Co., Inc., USA, 2006. M. Shamma, J.L. Moniot, Isoquinoline Alkaloids Research, 1972–1977, Plenum Press, New York, 1978. G.M. Anderson III, P.A. Kollman, L.N. Domelsmith, K.N. Houk, J. Am. Chem. Soc. 101 (1979) 2344. A.I. Vokin, Yu.L. Frolov, S.A. Medvedeva, S.G. D’yachkova, Izv. Akad. Nauk: Ser. Khim. 12 (1993) 1757. P.J. Breen, E.R. Bernstein, H.V. Secor, J.I. Seeman, J. Am. Chem. Soc. 111 (1989) 1958. L.M. DiBello, H.M. McDevitt, D.M. Roberti, J. Phys. Chem. 72 (1968) 1405. O. Exner, V. Jehlicka, Collect. Czech. Chem. Commun. 48 (1983) 1030. M. Dutkiewicz, J. Koput, J. Mol. Struct. 354 (1995) 153. M.A. Wysocki, P.W. Jardon, G.J. Mains, E.J. Eisenbaum, D.W. Boykin, Magn. Res. Chem. 25 (1987) 331. J.C. Facelli, J. Mol. Struct. (Theochem.) 276 (1992) 307. M. Makriyannis, S. Fesik, J. Am. Chem. Soc. 104 (1982) 6462. J.T. Yi, J.W. Ribblett, D.W. Pratt, J. Phys. Chem. A 109 (2005) 9456. J.W. Emsley, S. Hadenfeldt, T.J. Horne, G. Celebre, M. Longeri, J. Chem. Soc., Perkin Trans. 2 (1991) 583. H. Konschin, H. Tylli, B. Westermark, J. Mol. Struct. 102 (1983) 279. M. Gerzain, G.W. Buchanan, A.B. Driega, G.A. Facey, G. Enright, R.A. Kirby, J. Chem. Soc., Perkin Trans. 2 (1996) 2687. H. Konschin, J. Mol. Struct. (Theochem.) 168 (1988) 293. S. Tsuzuki, H. Houjou, Y. Nagawa, K. Hiratani, J. Chem. Soc., Perkin Trans. 2 (2002) 1271. C. Vande Velde, E. Bultinck, K. Tersago, C. Van Alsenoy, F. Blockhuys, Int. J. Quant. Chem. 107 (2007) 670. H. Tylli, H. Konschin, C. Grundfelt-Forsius, J. Mol. Struct. 77 (1981) 37.

O.V. Dorofeeva et al. / Journal of Molecular Structure 933 (2009) 132–141 [28] H. Konschin, J. Mol. Struct. 105 (1983) 213. [29] S.W. Konijn, R.J.J.M. Steenvoorden, P.G. Kistemaker, T.L. Weeding, J. Phys. Chem. 98 (1994) 5399. [30] C. Agache, V.I. Popa, Monatsh. Chem. 137 (2006) 55. [31] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.

141

Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision B.03, Gaussian, Inc., Pittsburgh, PA, 2003. [32] Yu.V. Vishnevskiy, UNEX: United Nuclear EXperiments, 2008. Available from: . [33] V.A. Sipachev, J. Mol. Struct. (Theochem.) 22 (1985) 143. [34] K. Tamagawa, T. Iijima, M. Kimura, J. Mol. Struct. 30 (1976) 243.