Dipole moment calculation in solution for some liquid crystalline molecules

Journal of Molecular Structure 1059 (2014) 44–50

Contents lists available at ScienceDirect

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

Dipole moment calculation in solution for some liquid crystalline molecules M. Włodarska ⇑ Institute of Physics, Technical University of Lodz, Wolczanska 219, 90-924 Lodz, Poland

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 We computed geometry and dipole

moment of four LC compounds in non-polar solvent.  For highly polar molecules, PCM model is closer to experiment than Onsager model.  Orientation of symmetric terminal polar groups affects the direction of dipole moment.  Adding polar terminal groups changes the dielectric anisotropy in the LC phase.

a r t i c l e

i n f o

Article history: Received 29 September 2013 Received in revised form 4 November 2013 Accepted 5 November 2013 Available online 16 November 2013 Keywords: Conformational analysis Theoretical calculations Solvent effects Dipole moment Liquid crystals

a b s t r a c t Theoretical and experimental studies were used to determine the dipole moment for four liquid crystalline materials. Ab initio calculations in vacuum led to elimination of a few potential conformers in each molecule. These results were used as initial data to obtain the geometry of the molecules in non-polar solvent according to Onsager’s or PCM model. After employing both models, slight increase of the dipole moment was noticed for all the molecules. The results are similar for both the models, with one exception for the strongly polar molecule where the Onsager’s model led to much higher dipole moment. The dipole moment was also determined experimentally for all the materials. The experimental data were compared with the results from calculations. For the nematic materials, the theoretically predicted sign of the dielectric anisotropy was compared with experimental results. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction Connecting theoretical computations of molecular structure with experiment is a constant subject of scientific interest [1–3]. Recent theoretical studies also consider changes in the molecular structure resulting from interactions with the surrounding medium (e.g. a solvent) [3–6]. One group of materials in which the molecular structure as well as the interactions between the molecules have large impact on the physical properties in the condensed phase ⇑ Fax: +48 (42) 631 36 39. E-mail address: [email protected] 0022-2860/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.molstruc.2013.11.015

are liquid crystals [7–10]. The fundamental parameters describing a liquid crystal are the temperatures of phase transitions and the type of the mesophase [11–13], which are also related to the structure of the molecules. Thus, precise determination of the parameters describing internal structure of the investigated material may help to predict some physical properties of the mesophase. A very important property of compounds used in devices based on electrostatic control of molecular orientation (mainly nematics) is also the dielectric anisotropy, which determines the character and speed of changes in the molecular alignment in response to the applied electric field [7,8]. The Maier–Meier equations successfully provide qualitative explanation of the observed changes in the electric

M. Włodarska / Journal of Molecular Structure 1059 (2014) 44–50

permittivity of nematic liquid crystals built of polar molecules. The best example is the relation between dielectric anisotropy De and the angle b between the direction of the dipole moment and the long axis of the molecule. When 3cos2 b = 1 (b  55°), the contributions of the dipole moment to e|| and e\ are equal. De is then determined by (positive) anisotropy of molecular polarizability. The dipolar contribution to De is positive for b < 55° and negative for b > 55°. In the latter case, the actual sign of De will depend on the absolute values of contributions from the anisotropy of molecular polarizability Da and the orientational polarization [7]. Qualitative analysis based on Maier–Meier equations may be useful for initial estimation of the dielectric anisotropy of a given material, based on the dipole moments of functional groups building up the compound. For instance, starting with a non-polar compound with small positive dielectric anisotropy and substituting one of the functional groups with a group having strong dipole moment roughly parallel to the long molecular axis, one may expect the dielectric anisotropy to increase significantly due to the large growth of e||. If the introduced functional group has small dipole moment which makes a large angle with the long axis of the molecule then a small increase of e|| may be expected, due to the component l|| and increased polarizability in the direction of the long axis. However, since the component e\ will also increase due to the appearance of the perpendicular component of the dipole moment, the dielectric anisotropy may be expected to remain unchanged or even decrease slightly. A behavior qualitatively consistent with this description has been observed e.g. in some derivatives of azobenzene [7]. This means that by knowing the value of the dipole moment and the angle between its direction and the long molecular axis in a nematic material it is possible to predict the sign of the dielectric anisotropy and therefore determine how the ordering of the given material will change in an external electric field. The quantum–mechanical calculations can be used to determine the dipole moment and its inclination angle to the long axis. In earlier theoretical studies we used ab initio calculations in vacuum to determine basic conformers and their dipole moments for four of the compounds investigated in the present paper [14]. In the present work, the molecular geometry obtained from those calculations was used as a starting point for further optimization, using also the RHF method with 6-31G basis set, but including the interaction between the molecules and the solvent (benzene) according to Onsager’s model of polar molecules in solution [15] and PCM model [16]. The theoretically determined dipole moments of the investigated compounds were compared with the dipole moments obtained from the experiment. For the nematic materials, the sign of the dielectric anisotropy predicted by the calculations was compared with the experimental data.

2. Materials The general structure of the monomers investigated in this work consists of a rigid central segment and two flexible endings. The central segments are typical for mesogenic compounds. Indeed, five of these six monomers appear in liquid crystalline state in a wide temperature range. The molecular structure of the monomers is shown below, along with the acronyms introduced for greater clarity of the text (Fig. 1). The compounds were synthesized recently and were subjected to detailed analysis, including confirmation of the molecular structure (FTIR, NMR, and elemental analysis) and determination of phase transitions and type of mesophase (DSC, optical observations, and X-ray diffraction studies). The results of these investigations have been presented elsewhere [17–19]. It has been established that the compound B1 has no mesophase, BU1 exhibits two smectic phases at cooling: SmB and – in a narrow temperature

45

range – SmE. The compounds M1, MU1 exist in the nematic phase in a wide temperature range (115–72 °C), both at heating and cooling [17]. 3. Experimental Dielectric measurements in low frequency range were carried out using a Novocontrol system containing a high-resolution Alpha dielectric analyzer. The refractive index was measured with CarlZeiss Abbe refractometer with sodium vapor lamp as the light source. The investigated materials were dissolved in benzene at room temperature and the solutions with different concentration were studied. The measurements of the dielectric anisotropy of the compounds were made with a Solartron 1260 Impedance Analyser. The samples were placed inside a flat capacitor made of glass plates covered with a conducting ITO layer, and oriented in a magnetic field of 1T. Ab initio calculations of molecular structures and dipole moments were performed using the Gaussian 03 program suite on a PC workstation [20]. Molecular energies and wavefunctions were computed by closed-shell Hartree–Fock method (RHF) with standard basis set 6-31G. Structural parameters of the studied molecules were determined by means of the Berny algorithm of geometry optimization, using a program developed by Schlegel with later modifications [21]. As a final step, single-point energy calculations were performed for all the optimized geometries, using the RHF/6-31G model with second-order Møller–Plesset (MP2) energy correction for electron correlation. The dipole moments were computed as first derivatives of energy with respect to the applied electric field. Furthermore, interactions between the molecules and the non-polar surrounding medium (solvent) were included in the calculations, according to Onsager model [15] and polarizable continuum model (PCM) [16]. As in the earlier cases, the molecular energy was calculated using restricted Hartree–Fock (RHF) method with 6-31G basis set. The electric permittivity of the solvent was taken from literature data for benzene. 4. Results 4.1. Theoretical study Earlier theoretical studies on the structure of the monomers, using ab initio calculations in vacuum [14], resulted in determining basic conformers and their dipole moments. In the course of the calculations the molecules were split into three segments (one central part and two identical endings) in a view to enable detection of more than one possible structure. Indeed, it has been found that the central segment of the monomers B1/BU1 may exist in at least two forms, which have substantially different dipole moments but very close values of energy. Consequently, two structural isomers were found for B1 and BU1 compounds and possible conformers of M1 and MU1 molecules were also presented. In addition, possible existence of several conformations differing in the mutual orientation of segments was discussed [14]. In our previous paper [14], a two-step method for geometry optimization of the monomers was proposed. In the first step, the investigated structures were split into three segments: the central part and the two endings (cf. Fig. 1), which were optimized as if they were separate molecules. In the next step, the optimized segments were used to reconstruct the original molecules, taking into account several possible orientations of the terminal segments with respect to the central segment, and these structures were further optimized. Based on the results of the calculations, which were carried out for four of the compounds studied in the present paper, possible co-existence of several isomers in real materials

46

M. Włodarska / Journal of Molecular Structure 1059 (2014) 44–50

EN B1

B

EN

O CH2=CH (CH2)2 C O

O O C (CH2)2 CH=CH2

EO BU1

EO

O CH2 CH (CH2)2 C O

O O C (CH2)2 CH CH2 O

O

M

EN M1

O CH2=CH (CH2)2 C O

EN O O C (CH2)2 CH=CH2

O O C

EO MU1

EO

O CH2 CH (CH2)2 C O

O O C (CH2)2 CH CH2

O O C

O

O Fig. 1. Chemical structure of the investigated compounds.

segment B

-835.438 -835.439 -835.44

ERHF [a.u.]

was suggested [14]. Following this suggestion, we decided to carry out subsequent detailed studies on this issue, using the same twostep method. Throughout the rest of the text, the central segment of molecules B1 and BU1 (composed of two phenylene rings and two ester groups) will be denoted as segment ‘B’. Similarly, the rigid core of molecules M1 and MU1 (which differs from segment ‘B’ in the additional ester group connecting the rings) will be denoted as segment ‘M’. The terminal aliphatic chains with epoxy and vinyl groups will be labeled with symbols ‘EO’ and ‘EN’, respectively (cf. Fig. 1). Each of the segments was optimized using restricted Hartree–Fock (RHF) method with 6-31G basis set. One of the main observations from the previous work [14] was that the spatial configuration of a molecule may be determined not only by relative orientation of the three segments, but also by mutual orientation of phenylene rings and ester groups within the central segment, as they do not lie in a single plane and several configurations corresponding to local minima of energy may be possible. This is important because ester groups are polar and so conformational changes of the central segment may lead to significant variations of its dipole moment, having considerable effect on the physical properties of the material. Indeed, two isomers of segment B were found and their dipole moments were quite different [14]. A closer comparison of the structural parameters of both isomers shows that the only significant difference between them is in the value of the dihedral angle between the planes of the two aromatic rings (d). Exploring this further, we analyzed the dependence of the segment’s energy on the value of this angle so as to locate all the existing local minima. The plot showing the variations of the potential energy is presented in Fig. 2. It can now be seen that there are in fact four local minima and a closer look reveals that the absolute value of the angle between the rings is virtually the same in all the four cases and is equal to about 46°, as in other molecules based on a biphenyl core [22]. Energy values corresponding to these four structures are very close, within a range of about 0.0005 a.u., whereas their dipole moments differ significantly – from 0.1 to 3.4 D. However, full optimization of the molecules built from the four variants of the central segment resulted in some of initially different structures having converged to the same final structure. Ultimately, two different conformations

-835.441 -835.442 -835.443 -835.444 -835.445 -835.446 -180

-90

0

90

180

torsion angle [o] Fig. 2. Dependence of potential energy (EHF) of the segment B on the torsion angle of the C–C bond between the phenylene rings. Four local minima can be seen at about ±46° and ±133°.

of the central segment were found in the case of BU1 molecule and three conformations in the case of B1. Their energy values and dipole moments are shown in Table 1. Similar analysis of molecules M1/MU1 revealed existence of several minima as well. However, in contrast to the B1/BU1 case in which the minima were relatively well-defined, the minima were in this case quite broad and flat which made it difficult to determine the correct values of the dihedral angles corresponding to these minima (as shown in Fig. 3). It suggests that internal rotation about this bond may occur relatively easy. Similar results for other molecules containing aromatic rings linked by an ester group can also be found in the literature [23]. In order to overcome this difficulty we used another approach. Beginning with the initial structure of the central segment M as it was determined in the paper [14], we subsequently created different variants of this segment by changing the dihedral angles between planes defined by different parts of the segment (aromatic rings and ester groups). The changes were made in such a way that absolute values of the angles between the planes were preserved and only the spatial orientation of the groups changed. Taking into account the symmetry, 16 different variants of the segment M can be created in that way, which were subsequently fully optimized. The optimization

47

M. Włodarska / Journal of Molecular Structure 1059 (2014) 44–50 Table 1 Results of final ab initio calculations performed at RHF + MP2 level of theory with 631G(d) basis set for different structural isomers obtained in the course of geometry optimization. Molecule

ERHF (a.u.)

EMP2 (a.u.)

l (D)

h (°)

BU1c BU1d B1b B1c B1d M1a M1b MU1a

1295.044807 1295.044793 1145.366520 1145.366642 1145.366636 1332.973561 1332.973533 1482.652306

1298.885859 1298.885880 1148.842472 1148.842682 1148.842716 1336.913800 1336.913748 1486.958521

1.96 2.03 0.44 2.19 2.21 1.48 2.06 3.15

90 90 90 90 90 24 51 74

decreased the total number of isomers to 8 as some initially different structures ultimately converged to identical final structures. Among these eight variants it is possible to identify a group of four monomers having very similar values of energy, while the other four have noticeably higher energies. Consequently, we suggest that structures from this group are energetically preferred and the other four are rather insignificant: even though the isomers containing these variants of the central segment may exist in a real material, their population is likely to be relatively low. Subsequent optimization of complete molecules built from the selected conformers of the central segment (again, labeled with greek letters) resulted in several structures, of which we chose three having somewhat lower energies than the other ones. Once again, elimination of structures with higher energies seems to be justified by relatively low probability of encountering them in real material. Summarizing the above analysis it can be said that except the structures already reported in the paper [14] a few more structural isomers should be considered, particularly when dipole moment is of importance, as in spite of very close values of their energy the conformers have substantially different dipole moments. The energies and dipole moments as well as the angles between the direction of the dipole moment and the main molecular axis – obtained from the calculations carried out for all the conformers – are listed in Table 1. All the calculations reported so far were performed for free molecules, neglecting any possible interactions with a surrounding medium. In the next step, the molecular geometry obtained from those calculations was used as a starting point for further optimization, using also the RHF method with 6-31G basis set, but including the interaction between the molecules and the solvent (benzene) according to Onsager’s model of polar molecules in solution [15] and PCM model [16]. Different models of solute–solvent dielectric interactions usually assume that a polar molecule is located within a cavity in a polarizable medium. The molecular dipole leads to polarization of the medium and the local electric field effectively modifies the molecular dipole moment. The difference between various models

-1023.02

(b) -1023.02

-1023.03

-1023.03

ERHF [a.u.]

ERHF [a.u.]

(a)

is in the definition of the cavity. The Onsager’s model assumes the spherical shape neglecting the actual shape of the molecule. By adding terms describing the molecule-solvent interactions to the molecular Hamiltonian it is then possible to follow the usual routines not only to calculate single-point molecular properties, but to optimize the geometry in the presence of the medium as well. The values of the dipole moment calculated for molecules in solution are presented in Table 2. It is noteworthy that influence of the solvent on the molecular dipole moment is much more prominent in the case of highly polar molecules (cf. Tables 1 and 2). In MU1, for Onsager model, the solvent effect changes the value of the dipole moment from 3.15 D to 4.89 D (increase by more than 50%) whereas the dipole moment of one of the conformers of B1 changes only slightly, from 0.44 D to 0.48 D (10% increase). In the case of MU1 material, the detailed comparison of the spatial structures obtained for an isolated molecule and for a molecule interacting with the solvent shows that the conformation corresponding to the energy minimum in vacuum is no longer optimal after taking into account the influence of the solvent. In result of further energy minimization, torsion angles of a few bonds underwent a substantial change (25–30°) leading to a modification of the mutual orientation of polar groups in the molecule, which is responsible for such a big change of the value of the dipole moment (Fig. 4). An interesting fact is that the structural change occurred within the central segment whereas the orientation of the flexible terminal groups with respect to the central segment remained practically unchanged. This is consistent with the earlier observation (shown in Fig. 3) that changing the torsion angles of bonds inside the segment M around their optimal values have very little effect on the potential energy of the molecule – so these bonds are likely to be sensitive to external influences. For PCM model, the highest increase is also seen in the MU1 molecule, but it is not as spectacular as for Onsager’s model. For the other molecules the dipole moment slightly increases (similarly to the previous case), but for two conformers we also noticed a decrease for PCM model. In the light of these results it is clear that interactions with the solvent cannot be neglected, especially if the molecules have large dipole moments. On the other hand, we should keep in mind that the model used for calculations contains significant simplifications (e.g. the spherical shape of the cavity) and thus the results of the calculations may not be very accurate. For both the methods, the calculated angle between the main molecular axis and the direction of the dipole moment (which is an important parameter from the perspective of commercial applications) does not differ much from the value obtained for the same molecule in vacuum. In the case of B1 and BU1 molecules this angle equals 90°, and for the MU1 molecule its value is around 74° (Tables 1 and 2). The largest and most interesting differences are in the results of the calculations performed for the M1 molecule,

-1023.04 -1023.05 -1023.06 -200

-1023.04 -1023.05

-100

0

torsion angle [o]

100

200

-1023.06 -200

-100

0

100

200

torsion angle [o]

Fig. 3. Dependence of potential energy (EHF) of the central segment M on torsion angles of selected bonds: (a) for d2 – the angle between the planes of two aromatic rings and (b) for d3 – the angles between the planes of the aromatic rings and the planes of the ending groups.

48

M. Włodarska / Journal of Molecular Structure 1059 (2014) 44–50

Table 2 Theoretical calculations of RHF energy and dipole moment in solution, according to Onsager and PCM models of solvent effects. h – angle between the direction of the dipole moment and the long molecular axis. Molecule

B1b B1c B1d BU1c BU1d M1a M1b MU1a

Onsager model

PCM model

ERHF (a.u.)

ltotal (D)

h (°)

ERHF (a.u.)

ltotal (D)

h (°)

1145.3665264 1145.3668244 1145.3667878 1295.0449483 1295.0449070 1332.9736256 1332.9737193 1482.6527398

0.48 2.59 2.52 2.42 2.18 1.78 3.01 4.89

90 90 90 90 90 26 60 73

1145.3771125 1145.3771115 1145.3770787 1295.0589354 1295.0520705 1332.9862689 1332.9862238 1482.6681986

1.63 1.49 3.20 2.49 1.60 1.94 2.28 3.93

90 90 90 90 90 50 54 77

Fig. 4. Modification of MU1 molecular geometry caused by taking interactions with the solvent into account during the computations: (a) structure obtained for a molecule in vacuum and (b) structure obtained for a molecule of the material dissolved in benzene. The main change is in the orientation of the indicated polar groups with respect to the rest of the molecule.

where this angle is around 50°. In the case of MU1 and M1 materials it means that replacing the additional symmetric polar groups with non polar flexible endings changes the dielectric anisotropy of the material in the nematic phase (which is positive for angles smaller than 55°). This is an important conclusion because it means that dielectric properties in liquid crystalline phase – even for materials with symmetric molecules – depend not only on the mesogen but also on the orientation of flexible polar terminal groups. This conclusion can be drawn from all the performed computations. The theoretical results obtained in this part of the paper are compared with the experimental data in the next paragraph. 4.2. Experimental dipole moment in solution The dipole moments of the investigated compounds were analyzed both experimentally and theoretically. The measurements of the dipole moment were carried out for compounds dissolved in a non-polar solvent (benzene) so as to minimize interactions between neighboring molecules. The values of the dipole moment were calculated in the framework of Onsager’s local field model. Despite Onsager’s original assumption of spherical shape of the molecules, which is not true for the majority of liquid crystalline compounds, his model has been successfully used in practice to describe many dielectric properties of such compounds [24,25].

Within this model, the apparent dipole moment of polar molecules dissolved in a non-polar solvent can be expressed with the formula [26]:

l2s ¼

9e0 kT ðe0  e1 Þð2e0 þ e1 Þ cN A e0 ðe1 þ 2Þ2

ð1Þ

where ls is the dipole moment of a free molecule, k the Boltzmann’s constant, T the temperature, c the molar concentration, NA the Avogadro’s number, e0 the electric permittivity of the vacuum, e0 the electric permittivity of the solution, and e1 is the high-frequency electric permittivity of the solution (often replaced by the square of the refractive index, n2). It is assumed that the energy of interaction between polar molecules is small as compared to the energy of thermal motion, which can be written as cNAl2  kT. In our calculations e1 was replaced by n2, e0 was measured at low frequencies. The frequency range was selected so as to avoid the influence of DC conduction – in this range e0 was constant. The relation (1) can be transformed to

D¼

l2 Na C; 9e0 kT

where

D¼

ðe0  n2 Þð2e0 þ n2 Þ

e0 ðn2 þ 2Þ2

ð2Þ

49

M. Włodarska / Journal of Molecular Structure 1059 (2014) 44–50

Hence, the plot of D against C should yield a straight line with slope coefficient dependent on the square of the dipole moment of a molecule. The results obtained in measurements carried out for the investigated compounds in different concentrations are shown in Fig. 5. It is noteworthy, that the greater slope of the D (C) plot, the greater the dipole moment of the molecule. Hence, by looking at Fig. 5 one can immediately conclude that MU1 molecules have the greatest dipole moment of all, B1 molecules – the least and M1 and BU1 have comparable dipole moments. The exact values calculated from experimental data plotted in Fig. 5 are listed in Table 3. The experimental values of the dipole moment are presented in Table 3 and can be compared with the computed values shown in Table 2. Both results are in reasonable accordance in the case of M1, B1 and BU1 molecules, but calculation based on the Onsager’s model seems to be closer. Assuming co-existence of different conformers in a real sample one may expect the apparent dipole moment to have an intermediate value between the lowest and the highest of the values obtained theoretically. In our case the experimental values tend to be somewhat higher than the values expected from both calculations. The only difficult case is MU1, for which the calculations led to a very high value of the dipole moment (4.9 D) in Onsager’s model, significantly higher than the experimental value of 3.7 D. One possible explanation is that despite the more systematic approach used in this work we may still have missed some other stable conformations of MU1 molecule having lower dipole moments. The experimental data may also contain errors, both purely experimental and those resulting from the Onsager’s theory, which assumes spherical shape of the molecules. It may be supposed that using a better model of molecule-solvent interactions would result in smaller deviations from theoretical expectations. For this molecule (MU1), PCM model gives results much closer to experimental data, which suggests that this model is a better approximation for molecules with high dipole moment. The experimental data show clearly that replacing vinyl endings with epoxy ones increases the value of the dipole moment. The theoretical results are similar, particularly when the dipole moment of the central segment is small (e.g. one of the conformers of B1).

Table 3 Experimental values of dipole moment of investigated compounds. BU1

M1

MU1

Experimental dipole moment l (D)

2.2

3

2.9

3.7

Dielectric anisotropy of the nematic materials M1 and MU1 was derived from the values of e|| and e\ measured with magnetic fields oriented – respectively – in the direction parallel and perpendicular to the direction of the electric field. The measurements were made in the course of cooling from isotropic liquid state. The results obtained for the MU1 material (Fig. 6a) show that it has a negative dielectric anisotropy. Such a result was expected on

(a)

7.0

nematic

ε' 6.0 ε||

5.5

ε⊥

5.0 50

70

90

110

130

150

T [°C]

(b)

nematic

isotropic

6 5

ε' 4 ε||

3

ε⊥ 2 50

70

90

110

130

150

T [°C] Fig. 6. Experimental dependencies e\ (T) and e||(T) for the nematic phase of the materials MU1 (a) and M1 (b).

B1

100 80

60

Δ

40 20

60 40 20

0

0 0

0.2

0.4

0.6

0

0.8

0.2

C [mol/dm3]

0.4

0.6

0.8

C [mol/dm3]

MU1

Δ

isotropic

6.5

80

Δ

B1

4.3. Dielectric anisotropy in nematic materials

M1

100

Compound

BU1

100

100

80

80

60

Δ

40

60 40 20

20

0

0 0

0.2

0.4

C [mol/dm3]

0.6

0.8

0

0.2

0.4

0.6

0.8

C [mol/dm3]

Fig. 5. Values of permittivity-dependent quantity D, defined by formula (2), plotted against the molar concentration C of compounds M1, B1, MU1 and BU1 dissolved in benzene.

50

M. Włodarska / Journal of Molecular Structure 1059 (2014) 44–50

the grounds of earlier theoretical ab initio calculations, which produced a large value of the angle b between the direction of the dipole moment and the long molecular axis, which – according to Maier–Meier interpretation – should lead to negative dielectric anisotropy. Fig. 6b presents the measurement results of the electric permittivity of the M1 material in the nematic phase. In this case, the dielectric anisotropy is positive, in agreement with the theoretical computations which predicted a dipole moment not deviating much from the direction of the long molecular axis (b < 55°, which in the Maier–Meier theory leads to a positive dielectric anisotropy). The experimental studies therefore confirm the earlier conclusion from the theoretical computations, that dielectric properties in liquid crystalline phase depend not only on the rigid mesogen (central segment) but also on the orientation of the flexible polar terminal groups. 5. Conclusions Full optimization of molecular geometry was carried out, including interactions between the molecules and the non-polar surrounding medium (solvent) according to the Onsager and PCM models. For both the models, the calculation in solution led to a somewhat greater dipole moment in comparison with the calculation in vacuum. For molecules with small dipole moment, optimization with Onsager model slightly changed the geometry of the molecules and the dipole moment, but after using PCM model the changes were much more visible. One exception was the molecule with high dipole moment, where the situation was opposite. Comparing the experimental and theoretical results one can conclude about the participation of particular conformations in real material. For the molecules with high dipole moment the PCM model seems to be closer to experiment than Onsager model. All the conducted studies also allow to conclude that orientation of the symmetric terminal polar groups affects the total molecular dipole moment and some physical properties. It was observed that modification of the molecular structure of the material by adding polar terminal groups changed the dielectric anisotropy of the material in liquid crystalline phase. Acknowledgments The author is grateful to Prof. G.W. Bak and Prof. W. Bartczak  for helpful discussions and encouragement to further work.

References [1] V. Udayakumar, S. Periandy, M. Karabacak, S. Ramalingam, Spectroscopy 83 (1) (2011) 575–586. [2] M. Bartkowiak, G.J. Kearley, M. Yethiraj, A.M. Mulders, Phys. Rev. B: Cond. Matter Mater. Phys. 83 (6) (2011) 064102. [3] Y. Futami, Y. Morisawa, Y. Ozaki, Y. Hamada, M.J. Wojcik, Y. Ozaki, J. Mol. Struct. 1018 (2012) 102–106. [4] F. Safinejad, M. Asghari-Khiavi, J. Comp. Methods Sci. Eng. 9 (3) (2009) 155– 164. [5] U. Umamaheswari, N. Ajeetha, D.P. Ojha, Crystallogr. Rep. 54 (7) (2009) 1202– 1207. [6] M. Monajjemi, M. Khaleghian, N. Tadayonpour, F. Mollaamin, Int. J. Nanosci. 9 (5) (2010) 517–529. [7] W.H. de Jeu, Physical Properties of Liquid Crystalline Materials, Gordon and Breach Sci. Pub., London, 1980. _ [8] J. Jadzyn, G. Czechowski, C. Legrand, R. Douali, Proc. SPIE 4147 (2000) 184. [9] G.W. Gray, Liquid crystals and molecular structure: nematics and cholesterics, In: The Molecular Physics of Liquid Crystals, G.R. Luckhurst, G.W. Gray (Eds.), Academic Press, London, 1979. [10] A. Melloni, R.R. Paccani, D. Donati, V. Zanirato, A. Sinicropi, M.L. Parisi, E. Martin, M. Ryazantsev, W.J. Ding, L.M. Frutos, R. Basosi, S. Fusi, L. Latterini, N. Ferre, M. Olivucci, J. Am. Chem. Soc. 132 (2010) 9310–9319. [11] P.G. de Gennes, The Physics of Liquid Crystals, Clarendon Press, Oxford, 1974. [12] D. Demus, L. Richter, Textures of Liquid Crystals, Verlag Chemie, Weinheim, New York, 1978. [13] S. Sharma, D. Lacey, P. Wilson, Liq. Cryst. 30 (4) (2003) 451. [14] M. Włodarska, G.W. Ba˛k, W. Bartczak, J. Mol. Struct. (Theochem) 619 (2002) 59. [15] M.W. Wong, M.J. Frisch, K.B. Wiberg, J. Am. Chem. Soc. 113 (1991) 4776. [16] S. Miertus, E. Scrocco, J. Tomasi, Chem. Phys. 55 (1981) 117. [17] M. Włodarska, B. Mossety-Leszczak, H. Galina, G.W. Ba˛k, T. Pakuła, Liq. Cryst. 31 (4) (2004) 525. [18] B. Mossety-Leszczak, P. Wojciechowski, H. Galina, J. Ulan´ski, Polimery 5 (2001) 374. [19] J. Fabia, H. Galina, B. Mossety-Leszczak, J. Ulan´ski, P. Wojciechowski, Proc. SPIE 4759 (2002) 60. [20] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery Jr., R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, J.L. Andres, C. Gonzalez, M. Head-Gordon, E.S. Replogle, J.A. Pople, Gaussian 98, Gaussian Inc., Pittsburgh, PA, 1998. [21] H.B. Schlegel, J. Comp. Chem. 3 (1982) 214. [22] R.A. Vaia, D. Dudis, J. Henes, Polymer 39 (24) (1998) 6021. [23] J.-Y. Yeh, D.B. Saebø, S. Saebø, J. Mol. Struc. (Theochem) 388 (1996) 27. [24] M. Dutkiewicz, J. Małecki, J. Mol. Struc. 416 (1997) 105. _ [25] P. Ke˛dziora, J. Jadzyn, Mol. Cryst. Liq. Cryst. 192 (1990) 31. [26] H. Fröhlich, Theory of Dielectrics, Clarendon Press, Oxford, 1986.