@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Chemistry, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Danilovič, Zorana"@en ; dcterms:issued "2009-11-21T20:39:51Z"@en, "2004"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Orientational ordering in nematic liquid crystal phases arises from the presence of anisotropic intermolecular forces. To date NMR experiments, theory and Monte Carlo simulations indicate the importance of two main contributions to orientational ordering of small solutes in various liquid crystals and liquid crystal mixtures. The first contribution is well defined and involves short-range interactions that depend on the size and the shape of the solute. The second contribution, which accounts for long-range (electrostatic) interactions, is believed to have lesser impact on the molecular ordering. Which of the electrostatic interactions (induction, electric quadrupole or polarization) are most important is still debated. In order to investigate the impact of electrostatic interactions on molecular ordering, small symmetric molecules with the same size and shape, and therefore the same short-range interactions, but different electrostatic properties were dissolved in various liquid crystals and mixtures of liquid crystals. Second rank orientational order parameters of solutes in various liquid crystal phases are obtained from analysis of high-resolution NMR spectra. For high-spin systems, initial spectral parameters needed to solve very complicated high-resolution spectra are estimated from selective multiple-quantum NMR spectra, collected using a 3D selective MQ-NMR technique. Structural: parameters of the solutes are calculated -using non-vibrationally corrected nuclear dipolar coupling constants accurately obtained from analysis of highresolution NMR spectra. The contribution of the electrostatic interactions to the orientational ordering of small solutes in liquid crystal phases is discussed in terms of different solutes and different types of liquid crystals by comparing experiment with theoretically determined order parameters. Those comparisons seem to suggest that dipoles have the least impact on orientational ordering of small molecules in nematic liquid crystals. Quadrupole contribution results predict opposite signs of the electric field gradient to the one obtained in similar previous studies. Experiments with zero-electric-field-gradient mixtures ('magic mixture') show no significant contributions of the electrostatic long-range interaction to the orientational mechanism in the special mixture. The polarizability effect appears strongly dependent on molecular geometry and in this study appears to be an important electrostatic mechanism of orientation."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/15464?expand=metadata"@en ; dcterms:extent "6100382 bytes"@en ; dc:format "application/pdf"@en ; skos:note "ORIENTATION AL ORDERING OF SMALL MOLECULES IN NEMATIC LIQUID CRYSTALS By Zorana Danilovic B . S c , The University of Belgrade, 2001 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE R E Q U I R E M E N T S FOR THE D E G R E E OF M A S T E R OF S C I E N C E in THE F A C U L T Y OF G R A D U A T E STUDIES D E P A R T M E N T OF CHEMISTRY We accept this thesis as conformation to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 2004 © Zorana Danilovic, 2004 JUBCL THE UNIVERSITY OF BRITISH COLUMBIA FACULTY OF GRADUATE STUDIES Library Authorization In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Name of Author (please print) Date (dd/mm/yyyy) Title of Thesis: MJ AM/r?T7C JjQU/H QZYSIMS Degree: Department of The University of British Columbia Vancouver, BC Canada Year: grad.ubc.ca/forms/?formlD=THS page 1 of 1 last updated: 20-M-04 Abstract Orientational ordering in nematic liquid crystal phases arises from the presence of anisotropic intermolecular forces. To date NMR experiments, theory and Monte Carlo simulations indicate the importance of two main contributions to orientational ordering of small solutes in various liquid crystals and liquid crystal mixtures. The first contribution is well defined and involves short-range interactions that depend on the size and the shape of the solute. The second contribution, which accounts for long-range (electrostatic) interactions, is believed to have lesser impact on the molecular ordering. Which of the electrostatic interactions (induction, electric quadrupole or polarization) are most important is still debated. In order to investigate the impact of electrostatic interactions on molecular ordering, small symmetric molecules with the same size and shape, and therefore the same short-range interactions, but different electrostatic properties were dissolved in various liquid crystals and mixtures of liquid crystals. Second rank orientational order parameters of solutes in various liquid crystal phases are obtained from analysis of high-resolution NMR spectra. For high-spin systems, initial spectral parameters needed to solve very complicated high-resolution spectra are estimated from selective multiple-quantum NMR spectra, collected using a 3D selective MQ-NMR technique. Structural: parameters of the solutes are calculated -using non-vibrationally corrected nuclear dipolar coupling constants accurately obtained from analysis of high-resolution NMR spectra. The contribution of the electrostatic interactions to the orientational ordering of small solutes in liquid crystal phases is discussed in terms of different solutes and different types of liquid crystals by comparing experiment with theoretically determined order parameters. Those comparisons seem to suggest that dipoles have the least impact on orientational ordering of small molecules in nematic liquid crystals. Quadrupole contribution results predict opposite signs of the electric field gradient to the one obtained in similar previous studies. Experiments with zero-electric-field-gradient mixtures ('magic mixture') show no significant contributions of the electrostatic long-range interaction to the orientational mechanism in the special mixture. The polarizability effect appears strongly dependent on molecular geometry and in this study appears to be an important electrostatic mechanism of orientation. Contents Abstract ii Contents iii List of Tables v List of Figures vi List of Abbreviations vii Acknowledgement viii Dedication ix 1 Introduction 1 1.1 Liquid Crystals 1 1.2 Nematic Liquid Crystals 2 1.3 Orientational Ordering and Intermolecular Forces in Nematic Liquid Crystals 3 1.4 Orientational Distribution Function and Order Parameter 4 1.5 Orientational Ordering from Experiments 6 1.5.1 NMR Spectra of Oriented Molecules 7 1.5.2 Solutes as Probes of Orientational Order and Intermolecular Forces 9 1.5.3 Multiple-Quantum NMR 10 1.6 Orientational Ordering from Theoretical Calculation 12 1.6.1 Statistical Treatment of Orientational Ordering Using Mean-Field Approach 12 1.6.2 Anisotropic Intermolecular Interactions 14 1.6.3 Molecular Models of Orientational Order 15 A. Short-Range Models 15 B. Long-Range Models 17 1.6.4 Previous Predictions on Orientational Ordering 18 1.7 Outline of the Thesis 19 References 20 in 2 Experiment 22 2.1 Sample Preparation 22 2.2 Experimental Conditions 24 References 26 3 Spectral Analysis 27 3.1 Introduction 27 3.2 Spectral Analysis with Aid of MQ NMR 27 3.3 Molecular Structure and Order Parameters 40 3.4 Summary 44 References 44 4 Orientational Ordering in Nematic Liquid Crystals 45 4.1 Introduction 45 4.2 Determination of a Consistent Set of Order Parameters 47 4.3 Qualitative Comparison between Experimental Order Parameters 50 4.4 Comparison between Experimental and Theoretically Calculated Order Parameters 52 A. Short-Range Model Comparison 57 B. Long-Range Model Comparison 58 C. Dipolar Interaction Contribution 62 D. Quadrupolar Interaction Contribution 62 E. Polarizability Interaction Contribution 65 4.4 Summary 67 References 68 5 Conclusion 69 IV L i s t of T a b l e s 1.1 Independent Order Parameters as a Function of Molecular Symmetry 5 1.2 Maximum Number of Transitions in One Quantum Spectrum of a Partially Oriented Molecules as a Function of Number of Spins with / = \\ 10 3.1 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of TCB 3-spin System 30 3.2 Fitting Parameters from Analysis of MQ and High-Resolution Spectra of DOT in E B B A 32 3.3 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of DOT 6-spin System 37 3.4 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of C L M X 9-spin System 38 3.5 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of MESIT 12-spin System 39 3.6 Structural Parameters from Fits to Dipolar Couplings for DCT, C L M X and MESIT 41 3.7 Order Parameters from Fits to Dipolar Couplings 43 4.1 Scaled and Non-Scaled Order parameters 48 4.2 Anisotropy in Scale and Non-Scaled Order Parameters 51 4.3 Molecular Electrostatic Parameters 53 4.4 Adjusted Parameters in the Fitting Procedure 54 V List of Figures 1.1 Liquid crystal phase on temperature scale 1 1.2 Example of nematic liquid crystal E B B A 2 1.3 Molecular organization in nematic phase 3 1.4 The nematic molecule fixed axis system 5 1.5 Poor resolved high-resolution spectrum of nematogen ZLI 1132 9 1.6 Three pulse sequence use to generate and observe MQ coherences 12 1.7 Elastic model for short-range interaction between solute and liquid crystal 16 2.1 ZLI 1132 liquid crystal 23 2.2 E B B A liquid crystal 23 2.3 Solutes 23 2.4 Single pulse sequence 24 2.5 3D MQ NMR pulse sequence 25 3.1 High-resolution proton NMR spectra of DCT, C L M X and MESIT in ZLI 1132 at 300K 29 3.2 All positive MQ spectra of DCT in E B B A at 300K 33 3.3 Experimental and calculated +5Q spectra of DCT in E B B A 34 3.4 Experimental and calculated +4Q spectra of DCT in E B B A 35 3.5 Experimental and calculated high-resolution proton spectra of DCT in E B B A at 300K 36 3.6 Atom labeling of solute molecules 42 4.1 Chosen set of solutes with same size and shape but different electrostatic properties presented in their coordinate system 46 4.2 Non-scaled (series A) and scaled (series B) experimental order parameters 49 4.3 Anisotropy in scaled order parameters 51 4.4 Differences between S^\"' and S^' as a function of S*™'r'' for short-range models 57 4.5 Differences between S^'led and as a function of S^'!ed for long-range models 60 4.6 Different pair of short- with long-range interaction fits for magic mixture 61 vi 4.7 Differences between . V a n d S%?c as a function of S™led for dipole interaction contribution 63 4.8 Differences between S™led and S'f as a function of S^\"led for quadrupole interaction contribution 64 4.9 Differences between Ss'\"'e\" and S _ f as a function of for AA XX polarizability interaction contribution 66 List of Abbreviations NMR MQ Nematogen ZLI 1132 E B B A TCB DCT C L M X MESIT Efg n Sap x,y,z X.Y.Z Nuclear magnetic resonance Multiple quantum Nematic liquid crystal molecule Eutectic mixture of frans-4-n-alkyl-(4-cyanophenyl)-cyclohexane (alkyl=propyl, pentyl, heptyl) and trans—4-n-pentyl-(4'-cyanobiphenyl-4)-cyclohexane N-(p-EthoxyBenzylidene)-p'-n-ButylAniline 1,3,5 trichlorobenzene 3,5 dichlorotoluene 5 chloro m-xylene 1,3,5 trimethylbenzene (mesytilene) Electric field gradient The director of liquid crystal Order parameter Molecule fixed axis system Laboratory fixed axis system VI1 Acknowledgements I would like to thank my supervisor Prof. Elliott Burnell for his patience, encouragement, superb guidance and support over the past few years. It has been a great pleasure and privilege to work with him. I also wish to thank my colleagues Anand Yethiraj, Joseph Lee, Andrew Lewis, Ray Syvitski, Aman Taggar and Chris Campbell from which I have learned a lot and enjoyed interacting with. Thanks to Dr. Nick Burlinson for enthusiasm, moral support and resourceful discussions. Most of the experimental work wouldn't be possible without the excellent help from the people in the electronic shop, mechanical shop, chemistry store and NMR facility. I would also like to thank Scott Kroeker from University of Manitoba for donating one of the solutes, 3,5-dichlorobenzene. Financial assistance in the form of a teaching and research assistantship is highly appreciated. Special thanks to my family for their love and support, since without their inspiration none of this would have been possible. Finally I would like to express my gratitude to my friends and loved one Ken for their constant support, help and patience. vni Dedication To my lovely parents Ljiljana and Tetar IX 1. INTRODUCTION 1.1 LIQUID CRYSTALS The liquid crystal phase is known as a mesophase (u.eooa in Greek means middle), a phase between liquid and solid phases (Fig. 1.1). When heated from the solid phase at the first transition point they appear as a cloudy liquid that upon further heating above the clearing transition point transform into a clear isotropic liquid. Two transitional temperatures define the region in which mesophases are thermodynamically stable. Liquid crystals show the translational and rotational mobility of liquids and the optical properties of solids (like birefringence). They typically consist either of rod like or disk like molecules. Solid Liquid Crystal Liquid Melting point Clearing point TEMPERATURE Figure 1.1 Liquid crystal phase on temperature scale The main characteristic of liquid crystals is long-range orientational ordering. Some of them display some positional ordering as well. The tendency to align along a preferred direction is caused by intermolecular forces acting between liquid crystal molecules. As a consequence liquid crystals are categorized as anisotropic phases in which measured properties depend on the direction in which they are measured. The anisotropic nature of liquid crystals is responsible for their unique properties exploited in a variety of applications. Introduction Liquid crystals are classified into two main categories: thermotropic and lyotropic. Thermotropic liquid crystals are produced by thermal process as while lyotropic liquid crystal phases depend on solvent composition. The present study focuses only on nematic liquid crystals, the simplest thermotropic liquid crystal that posses only orientational order. 1.2 NEMATIC LIQUID C R Y S T A L S Nematic (vefiazoa in Greek means threadlike) liquid crystals are composed of rod-shape molecules with molecular length several times the molecular diameter. These molecules consist of semi-rigid cores like benzene rings with polar and nonpolar flexible ending groups Figure 1.2 Example of nematic liquid crystal EBBA (N-(p-EthoxyBenzylidene)-p'-nButyl .Aniline); stable in the,temperature range from 308 to 352 K Nematic molecules {nematogens) are arranged such that there is no positional order of their centers of mass, meaning that molecules diffuse randomly and rapidly like in isotropic liquids. When a solid melts the thermal energy is sufficient to destroy positional order but not enough to disrupt orientational order. This results in a nematic phase in which molecules group in clusters, and tend to partially align their molecular axes along a preferred direction, called the director n (Fig. 1.3). The cluster size can be as large as 1 fim [1]. In the absence of an external field, the orientation of the director axis in each cluster varies throughout the sample. The cloudy appearance of a macroscopic sample comes from the random scattering of light as it penetrates the sample between the clusters with different directors. In the presence of a magnetic field all clusters' directors reorient along the main magnetic field direction in just a few seconds. For almost all nematics known nowadays, the nematic phase is uniaxial. (Fig. 1.2). 2 Introduction Therefore they are cylindrically symmetric and all measured properties are invariant to rotations about the director n . Also the directions -n and ware indistinguishable making the phases apolar. 1.3 ORIENTATION AL ORDERING AND INTERMOLECULAR F O R C E S IN NEMATIC LIQUID C R Y S T A L S Orientational order is one of the basic characteristics common to all the mesophases. It is a result of anisotropic intermolecular forces acting between liquid crystal molecules. Understanding the relationship between orientational order and anisotropic intermolecular forces that cause the molecular orientation is one of the fundamental questions in the study of liquid crystals that still hasn't been fully answered. Several approaches (experimental, theoretical and computational) have been utilized in order to give some insight into the problem. In this study, NMR was used to extract information about order parameters that were compared with theoretically predicted order parameters calculated from theoretical models (which will be described in 1.6). Director n Figure 1.3 Molecular organization in nematic phase 3 Introduction 1.4 ORIENTATIONAL DISTRIBUTION FUNCTION AND O R D E R P A R A M E T E R /quantitative description of orientational order/ The orientational behaviour of a rigid, axially symmetric nematogen is fully described by the single orientational distribution function f(£l), where Q. represents the Eulerian angles (q>,d,\\j/) that describe the orientation of the molecule frame system (x, y, z) relative to the nematic director system (laboratory frame system (X, Y, Z)) (Fig. 1.4). Taking into account the cylindrical symmetry and uniaxiality of the phase, the orientational distribution function f(Q), expanded in terms of spherical harmonics, reduces to [2]: f(Q)= X ^ ( P / ) P ; ( c o s a ) = l + | < ^ > ^ ( c o s 0 ) + | (P 4 )P 4 ( cosc9) + ... (1.1) / even ^ 1 2 . I where (/>) = JP, (cos 0)7(0) sin 0 yz 5 Introduction (1.4) and for off-diagonal order parameter elements 4 S ' V S 4 (1.5) When all molecules are oriented parallel to the magnetic field and therefore the nematic director (8 = 0 or 9 = K ), the order parameter is Saa = 1 indicating perfect alignment while for perpendicular alignment (8 = — ), the order parameter becomes Saa If molecules tumble freely and isotropically orientation is totally random, the average value of the cosine squared function is ^ c o s 2 f ^ = j and Saa = 0, i.e. the system shows no orientational order at all. 1.5 O R I E N T A T I O N A L O R D E R I N G F R O M E X P E R I M E N T S There are many experimental techniques that can provide information about orientational ordering. For example polarized Raman scattering [6], Electron Paramagnetic Resonance (EPR) [7], quasielastic scattering of X-rays [8] or neutrons [9] can give up to fourth rank orientational order parameters. Their instrumental and analytical limitations make this task really difficult and incomplete. NMR has proved to be an excellent non-destructive technique for studying orientational order in anisotropic systems. Its only limitation lies in the fact that only the second rank order parameters can be extracted from experimental NMR spectroscopic data therefore giving a slightly modest representation of the orientational distribution function. The theoretical background and the relationship between order parameters and accessible spectral parameters will be described in the following section 1.6.1. Spectral step by step analysis and determination of order parameters will be covered in Chapter 3. 6 Introduction 1.5.1 NMR S P E C T R A OF ORIENTED M O L E C U L E S In the high-field limit* the spin Hamiltonian for a collection of spins ( / = | ) in a uniaxial anisotropic environment is given as: H = H 7 + Hj + H D (1.6) where Hz is the Zeeman, Hj is the scalar (indirect) coupling and HD is the direct dipolar Hamiltonian. The Zeeman Hamiltonian represents the interaction between nuclei and the main external magnetic field. Expressed in Hz units, it takes the following form Hz = - _ > , 7 , , z ^ - ^ 7 , ( 1 - 0 - , ^ ) / , , (1.7) where vj are the chemically shifted frequencies of the /' spin, B7 is the static external magnetic field defined to be along the Z-axis, li7 is the Z-component of the spin operator for the /'spin, and a i 7 7 is the ZZ-component of the chemical shift tensor for spin /'. The scalar coupling Hamiltonian has the general form where the scalar coupling constant Jv is a second-rank tensor that describes the indirect interaction between spins defined through spin operators 7(. and 7.. Another approximation is applied here: for most protons the anisotropy in the Ji} tensor is small and is ignored. * The high-field limit approximation takes into account that the Zeeman term in the Hamiltonian dominates over other terms in the Hamiltonian; for example the Zeeman term in a 100 MHz field is of order ~108 Hz while dipolar and quadrupolar contributions do not exceed 104 Hz; this simplifies the expression for the Hamiltonian. 7 Introduction The direct dipolar Hamiltonian ^ = X ^ ( 3 / , A z - A - ^ - ) <1-9> '8 have a complex Hamiltonian and high-resolution NMR spectra which are challenging to analyze even for symmetric molecules. For these systems which have so closely packed spectra, quite accurate initial spectral parameters are required in order to solve the spectrum. In that case known spectral parameters ( J ' v and cr,) of molecules with similar size and shape for the same liquid crystal are used as starting parameters in spectral analysis. In most cases this does not lead to a satisfactory solution so multiple quantum NMR is used. Multiple quantum spectra involve transitions between energy levels with the difference in M quantum numbers A M = 0,±\\,....±(N-\\),±N. Higher order multiple quantum spectra are quite simple containing fewer transitions than lower quantum ( A M < ±(N - 2)) spectra and therefore easier to analyze. Transition frequencies are governed by the same spectral parameters as conventional high-resolution spectra. In principle analyzing the N, N-1 and N-2 multiple quantum spectra is sufficient enough to gather all spectral parameters. The limitation of MQ NMR lies in the fact that multiple quantum spectra are much broader and less resolved than single quantum spectra leading to less precise spectral parameters. That is why spectral parameters determined from MQ experiments are used as initial spectral parameters in analysis of high-resolution spectra. Generally MQ coherences are created and observed using a three pulse sequence (Fig. 1.5) [13]. Before firing the first pulse the spin system is in equilibrium and only longitudinal magnetization I7 is present. The first pulse flips I7 magnetization into the x-y plane, which evolves in preparation time % (fixed) among one-quantum coherences. The second pulse transforms one-quantum coherences into all possible MQ coherences that evolve in t1 evolution time under the internal spin Hamiltonian (direct dipolar, indirect scalar) and finally the third pulse partially converts MQ coherences back into observable l7 one-quantum coherences that evolve in time t2. 11 Introduction nil* nil. n!2r A C Q U I R E / t 2 Figure 1.6 Three pulse sequence used to generate and observe MQ coherences More on how different MQ coherences are selectively detected using phase cycling and how MQ spectra are transformed will be explained in Chapter 2. 1.6 ORIENTATIONAL ORDERING FROM THEORETICAL C A L C U L A T I O N 1.6.1 STATISTICAL T R E A T M E N T OF ORIENTATIONAL ORDERING USING M E A N -FIELD A P P R O A C H The orientational distribution function f(Q.) fully describes the orientational order of a rigid solute in the nematic liquid crystal as well as the connection with experimentally measurable spectral parameters. Using statistical mechanics, the orientational distribution function f(Q) can be expressed as a singlet distribution function, that is kT kT dQ. (1.13) In the mean-field approximation all interactions in the system can be represented as interactions between a single molecule and the average field, i.e t / (Q ) i s the mean-field potential that governs the orientational order of.the molecule.. 12 Introduction The physical picture behind this simple philosophy is that interactions between the solute and the liquid crystal environment (solvent) are described as interactions between a solute property (such as dipole) and the averaged field of the solvent, so that V(0) = ASo!u,e(0)-FLiquicl co,skd . (1.14) The assumption that the averaged field is not influenced by the solute presence means that there is a clear and distinct separation of both solute and solvent contributions to the potential (no correlations between the two contributions). Statistically speaking, any measurable property taken as an average (for example Sap ) can be then expressed as ^ = J / ( Q ) . ( A c o s 0 a c o s 0 / , - ^ ) - d Q . (1.15) Combining Eq. 1.3, 1.13 and 1.15 gives Qcalc >~>„a — J(-f cos0 a cos 6^ kr dQ. -t/(Q) (1.16) dQ Equation 1.16 is the basis for the statistical treatment of the orientational ordering, because it connects the intermolecular potential with the orientational ordering. It gives the opportunity to investigate various intermolecular potentials, by comparing calculated and experimental order parameters and therefore create a better picture of the orientational behavior of molecules in nematic liquid crystals. Representation of the intermolecular interactions can be done either using statistical theories'(Maier-Saupe [14,15], Onsager [16]), simulation'methods [17-25] or phehomenological models [10-12,26-29]. In the present study the simple phenomenological approach will be utilized. 13 Introduction 1.6.2 ANISOTROPIC INTERMOLECULAR INTERACTIONS Anisotropic intermolecular interactions are responsible for the orientational ordering of the solutes in the liquid crystal systems. The effective orienting mean-field potential can be divided into two contributing terms; one that presents short-range interactions and the other that presents long-range interactions, so that Short-range interactions consist of both an attractive and repulsive part. At short distances, the attractive part is ignored. Only the dominant repulsive part that is a consequence of overlapping of the electron clouds between neighbour molecules is considered. Short-range interactions are considered as interactions that are highly affected by the molecular structure. Long-range interactions-involve distances much larger that the molecular dimensions. They are attractive or repulsive interactions that depend on electrostatic properties of the molecules. , . • 1.6.3 M O L E C U L A R M O D E L S OF ORIENTATIONAL O R D E R A. S H O R T - R A N G E M O D E L S A1. CI model One of the short-range potentials used is a phenomenological model called Cl by Burnell and co-workers [10], a combination of a circumference (C) model and an integral (I) model [26, 27] in which LR (1.17) (1.18) 14 Introduction where k7Z is a solvent parameter that determines the mean field influence on the solute, k and ks are proportionality constants; C(Q) is the circumference of the projection of the solute at orientation Q. onto a plane perpendicular to the director (that is along the Z axis in the laboratory frame) and C(Q,Z) is the circumference of the projection of the solute at position Z and orientation Q onto a plane perpendicular to the director; so that C(Q,Z) dZ is an infinitesimal thin ribbon that traces out the molecule at position Z and orientation Q . The first term in Eq. 1.18 (C model) can be interpreted as a Hooke's elastic law where the liquid crystal is treated as an elastic continuum and the solute as its distortion. The second term (I model) can be seen as an anisotropic interaction between solute surfaces and solvent averaged field. The presented model treats molecules as a collection of van der Waals hard spheres placed at the atomic sites (Fig. 1.7). A2. SS model Another type of short-range potential written as an expansion in spherical harmonics truncated at the first non-zero term [11] is ^(Mapk77)(3coseacosp-dap) (1.19) a.fi=x,y,z where M a j 3 are the traceless tensor components related to size and shape, and k77 is the same as in the previous model, a liquid crystal parameter related to the degree of orientation of the nematogen. 15 Introduction Figure 1.7 Elastic model for short-range interaction between solute and liquid crystal The potential described by Eq. 1.18 depends on the orientation Q of the molecule throughout C(Q) and C ( Z , Q ) ; C (O) is the circumference of the projection of the solute at orientation Q. onto a plane perpendicular to the director; C(Z,Q) is the circumference of the projection of the solute at position Z and orientation Q onto a plane perpendicular to the director. The Z axis corresponds to the direction of the liquid crystal director. The molecules are modeled as a collection of van der Waals hard spheres. 16 B. L O N G - R A N G E M O D E L S Introduction For long-range models only dipoles, quadrupoles and polarizabilities are suspected to be the most important contributors to orientational ordering and therefore will be utilized. B1. Dipole model The dipole interaction in the mean-field approximation, defined as an interaction between a permanent solute dipole moment and the averaged electric field of the liquid crystal, would be zero due to the apolar nature of the liquid crystal, i.e (is) = 0 . Nevertheless, the permanent dipole moment can induce a dipole moment in the liquid crystal, creating the so called reaction field (in the liquid crystal) that will in return react with the permanent dipole. The magnitude of this kind of interaction is proportional to the magnitude of the created field as can be seen from: tf(\")£U=-;r-7 S ^ ^ - ^ X S c o s ^ c o s ^ - ^ ) (1.20) where }ia,jip are dipole moments of the solute along the a,B molecule fixed directions and (R77 -RXX^J is the average value of the difference in the electric field between the Z and X laboratory directions [11,12]. B2. Quadrupole model The quadrupole interaction, an interaction between the quadrupole moment of the solute and the electric field gradient of the liquid crystal {efg), is given by m)t^=1-rr\\ Sca^xscose.cose,-sa l i ) ' 0 .21) where Qap is the afi quadrupole tensor component of the solute and F77 is the ZZ component of the electric field gradient, traceless second-rank tensor parallel to the nematic director [11]. 17 Introduction B3. Polarizability model The local electric fields within the liquid crystals could change the electronic charge distribution in the solute molecules that could lead to a change in their orientation. The potential corresponding to this kind of interaction is: UMZor**!*, =~T—T S«^.(^ - ^ ^ c o s ^ c o s ^ -dap) (1.22) where cua/3 is the a/3 molecular polarizability tensor component of the solute and (E77 -Elx^ is the average value of the squares of the electric field between the Z and X laboratory directions [11]. 1.6.4 PREVIOUS PREDICTIONS ON ORIENTATIONAL ORDERING Some important experiments and predictions from theory/models over the course of the past few decades will be presented so as to gain a better perspective on where the science stands on the fundamental question of orientational ordering. Phenomenological investigation of orientational ordering has been developed in two different ways. One in which interactions between solutes and liquid crystals are described as mean-field interactions meaning that all solutes feel the same averaged liquid crystal field [12, 30-36]. The other states that the mean-field picture is too simple and that there are correlation effects between solutes and liquid crystals leading to the fact that different solutes feel different liquid crystal fields [37-39]. This is supported by the possibility that one solute molecule could spend more time on average in the aromatic regions of the liquid crystals while the other solute would prefer more the hydrocarbon chain region of the liquid crystals. Burnell and co-workers have utilized the mean-field approach and investigated the orientational behaviour of small symmetric molecules D2 and HD [30-33] and methane [34-36] in a special mixture called magic mixture of 55 wt% 1132/EBBA and its pure components. From magic mixture experiments it was determined that there is no long-range 18 Introduction interaction contributions to the ordering, leaving the short-range interaction as the dominant contribution. The same results were obtained studying a whole range of small molecules dissolved in the magic mixture [12]. A computer simulation study of hard ellipsoids was performed to test the previously stated findings, and confirmed that short-range interactions indeed have a dominant effect on the orientational ordering of molecules in liquid crystals [19]. Studies of D2 and HD in pure component liquid crystals show some evidence of quadrupole-liquid crystals efg interactions. The determined electric field gradient of E B B A was shown to be negative not only from D2 and HD but also from benzene and benzene derivatives studies [12]. The sign of the efg of E B B A is the same as one predicted from the mean-field quadrupole moment/efg model demonstrating that the quadrupole interaction is an important orientational mechanism. Photinos et al. [37-39] on the contrary predicted theoretically that short- and long-range interactions contribute equally to the orientation and that long-range interactions arise mainly from dipoles and quadrupoles. Emsley, Luckhurst and co-workers have shown, using the statistical Maier-Saupe theory of nematics [40-42] that the quadrupole is the lowest order multipole that contributes to a non-vanishing efg and that the efg strongly depends on solute and solvent molecular properties. Overall agreement of different results from experiments, theories and computer simulations is that short-range interactions, size and shape dependent, are the dominant orientational mechanism. The importance of different electrostatic interactions is still an open question and is the main object of the presented study. 1.7 THE OUTLINE O F THE THESIS In order to gain a better understanding about anisotropic intermolecular forces within liquid crystals, small molecules were dissolved in liquid crystal solvents and used to probe the anisotropic intermolecular forces. Chosen solute molecules have the same size and shape, same short-range contribution to the ordering but different electrostatic properties that enable testing of the effects of the additional long-range interactions. Also, the choice of liquid crystals is important in the sense of separately exploring the effects of different electrostatic effects. In magic mixture liquid crystals it was found [30-33] that the efg is zero, i.e. quadrupole interactions are annulled and therefore other electrostatic effects (like dipole and polarizability) if effective should clearly be visible. 19 Introduction Chapter 1 contains all the theory basics needed to understand the experimental and theoretical determination of the spectral and order parameters and how valuable information about intermolecular forces that causes the orientational behavior of the molecules in liquid crystals can be extracted from those parameters. Chapter 2 presents the technical aspect of experiments while Chapter 3 focuses on analysis methodology and determination of spectral and structural parameters from non-vibrationally corrected dipolar couplings. A set of self-consistent order parameters is obtained in Chapter 4 and it was used to examine the effects of various short- and long-range interactions models on the ordering by comparing them with the calculated order parameters from those models. Chapter 5 summarizes the important results of the presented study with future work proposal. References: [1] de Gennes, P. and Prost, J . , 1993, The Physics of Liquid Crystals, Claredon Press, Oxford, 2 n d edition. [2] Emsley, J . W., 1985, Nuclear Magnetic Resonance of Liquid Crystals, C. Riedel Press [3] Saupe, A., 1964, 2. Naturforsch., 19, 161. [4] Diehl, P., Khetrapal, C. I., 1969, NMR Basic Principles and Progress, Vol . 1, Springer-Verlag, Berlin. [5] Bunnell, E. E., de Lange, C. A., 2003, NMR of Ordered Liquids, Kluwer Academic Publishers. [6] Vartogen, G. and de Jeu, W. H., 1988, Thermotropic Liquid Crystals, Fundamentals, Springer-Verlag. [7] Berliner, L. J . , 1976, Spin Labelling:Theory and Application, Academic Press, London. [8] Zannoni, C. and Guerra, M., 1981, Mol. Phys., 44, 849. [9] Leadbetter, A., 1979, The Molecular Physics of Liquid Crystals, Academic Press. [10] Burnell, E. E. and de Lange, C , 1998, Chem. Rev., 98, 2359. [11] Syvitski, R. and Burnell, E. E., 1997, Chem. Phys. Letters, 281, 199. [12] Syvitski, R. and Burnell, E. E., 2000, J. Chem. Phys., 113, 3452. [13] Syvitski, R., Burlinson; N., Burnell, E. E. and Jeeiier., J . , 2002, J.Mag. Res., 155, 251. [14] Maier, W. and Saupe, A., 1959, Z. Naturforsch. A, 14, 287. 20 Introduction [15] Maier, W. and Saupe, A., 1960, Z. Naturforsch. A, 15, 882. [16] Onsager, L , 1949, N.Y. Acad. Sci., 51, 627. [17] Hashim, R., Luckhurst, G. R. and Romano, S., 1985, Mol. Phys., 56, 1217. [18] Luzar, M. Rosen, M. E. and Caldarelli, S., 1996, J. Phys. Chem., 100, 5098. [19] Poison, J . M. and Bunnell, E. E., 1996, Mol. Phys., 88, 767. [20] Poison, J . M. and Burnell, E. E., 1997, Phys. Rev. E, 55, 4321. [21] Syvitski, R. T., Poison, J . M. and Burnell, E. E., 1999, Int. J. Mol. Phys. C, 10, 403. [22] Burnell, E. E., Berardi, R., Syvitski, R. T. and Zannoni, C , 2000, Chem. Phys. Letters, 331 ,455. [23] Celebre, G., 2001, Chem. Phys. Letters, 342, 375. [24] Celebre, G., 2001, J. Chem. Phys., 115, 9552. [25] Lee, J . ' -S . J . , Undergraduate thesis, Department of Chemistry, University of British Columbia, 2001. [26] van der Est, A. J . , Kok, M. Y. and Burnell, E. E.,1987, Mol.Phys., 60, 397. [27] Zimmerman, D. S. and Burnell, E. E., 1990, Mol. Phys., 69, 1059. [28] Zimmerman, D. S. and Burnell, E. E., 1993, Mol. Phys., 78, 687. [29] Ferrarini, A., Moro, G. J . , Nordio, P. L. and Luckhurst, G. R., 1992, Mol. Phys., 77,1. [30] Burnell, E. E., de Lange, C. A. and Snijders, J . G., 1982, Phys. Rev., A25, 2339. [31] Burnell, E. E., van der Est, A. J . , Patey, G. N., de Lange, C. A. and Snijders, J . G., 1987, Bull. Mag. Reson., 9, 4. [32] van der Est, A. J . , Burnell, E. E. and Lounila, J . , 1988, J. Chem. Soc. Faraday Trans. 2, 84, 1095. [33] Burnell, E. E., de Lange, C. A., Serge, A. L., Capitani, D., Angelini, G., Lilla, G. and Barnhoorn, J . B. S., 1997, Phys. Rev., E55, 496. [34] Snijders, J . G. , de Lange, C. A. and Burnell, E. E., 1983, Israel. J. Chem., 23, 269. [35] Snijders, J . G., de Lange, C. A. and Burnell, E. E., 1982, J. Chem. Phys, 77, 5386. [36] Snijders, J . G., de Lange, C. A. and Burnell, E. E., 1983, J. Chem. Phys, 79, 2964. [37] Terzis, A. and Photinos, D., 1994, Mol. Phys., 83, 847. [38] Photinos, D., Poori, C , Samulski, E. and Toiumi, H., 1992, J. Phys. Chem, 96, 8176. [39] Photinos, D., Poon, C , Samulski, E., 1993, J. Chem. Phys., 98, 10009. [40] Emsley, J . , Palke, W. and Shilstone, G., 1991, Liquid Crystals, 9, 643. [41] Emsley, J . , Heeks, S., Home, T., Howells, M., Moon, A., Palke, W., Patel, S., Shilstone, G. and Smith, A., 1991, Liquid Crystals, 9, 649. [42] Emsley, J . , Luckhurst, G. and Sachdev, H., 1989, Mol. Phys., 67, 151. 21 2. EXPERIMENT 2.1 S A M P L E PREPARATION The nematic liquid crystals used as solvents are: - ZLI-1132, a eutectic mixture of frans-4-n-alkyl-(4-cyanophenyl)-cyclohexane (alkyl=propyl, pentyl, heptyl) and rrans-4-n-pentyl-(4'-cyanobiphenyl-4)-cyclohexane (Fig. 2.1) purchased from Merck. : . , '• ~ E B B A (N-(p-EthoxyBenzylidene)-p'-n-ButylAniline) (Fig. 2.2) synthesized according to a reference book [1] by Ray Syvitski - a mixture 55 wt% ZLI-1132/EBBA called 'magic mixture' a nematic phase that has zero external electric field gradient (efg) at approximately 301.4 K [2,3] prepared from pure liquid crystals The solutes 1,3,5-trichlorobenzene (TCB), 3,5-dichlorotoluene (DCT), 5-chloro m-xylene (CLMX) and mesytilene (MESIT) (Fig. 2.3) were commercially available except 3,5-dichlorotoluene, which was donated by Scott Kroeker from the University of Manitoba. All liquid crystals and solutes were used with no further purification. 3,5-dichlorotoluene, 5-chloro m-xylene and mesytilene were dissolved in liquid crystal solvents in about - 5 % mol concentration in 5 mm outer diameter NMR tubes. In each sample - 1 % mol concentration of T C B was added as an internal orientational standard for scaling purposes. The mixtures were then heated to the isotropic phase and mixed thoroughly on a vortex stirrer repeatedly until samples became homogeneous. The NMR tubes were equipped with a capillary tube filled with acetone-d6, centered in the middle using teflon spacers. The acetone-d6 was used as a deuterium lock (signal). 22 Experiment H trans-4-n-propyl-(4-cyanopheny)-cyclohexane 24% NC-H H trans-4-n-pentyl-(4-cyanophenyl)-cyclohexane 36% C 5 H 1 1 ^ trans-4-n-heptyl-(4-cyanophenyl)-cyclohexane 25% C 5 H 1 1 15o/0 trans-4-n-pentyl-(4'-cyanobiphenyl-4)-cyclohexane Figure 2.1. ZLI 1132 liquid crystal H 5 C — O- \\ // H X ^ C 4 H 9 Figure 2.2 E B B A liquid crystal ( N-(p-EthoxyBenzylidene)-p'-n-ButylAniline ) Cl C H C H . C H . C |' \"ci c r \" Cl C l ' T C B DCT C L M X Figure 2.3 Solutes ^ C H 3 C H 3 CH MESIT 23 Experiment 2.2 e x p e r i m e n t a l c o n d i t i o n s Proton high-resolution and multiple-quantum NMR experiments were performed on a Bruker AMX-500 spectrometer equipped with a high resolution probe operating at 11.75 T (corresponding to a proton resonance frequency of 500 MHz). The temperature in the probe was calibrated using the difference in the proton chemical shifts of 80% Ethylene-Glycol/DiMethylSulfOxide-d6 and it was controlled by a Bruker temperature unit using air-flow. The dial temperature of 300.7 K was calibrated to 300.0 K +/_ 0.1 K. All samples were heated up to the isotropic phase and vortexed thoroughly before placing in the NMR probe. They were let to sit in the probe for half an hour in order for them to reach equilibrium. High-resolution proton NMR spectra were acquired using a simple one pulse sequence (Fig. 2.4) with 32K data points that are zero filled to 64K points in the ^dimension. The phase of the pulse (0) and the receiver (6) were cycled according to the cycling scheme in Fig. 2.4 in order to reduce the effects of pulse and receiver imperfections. The 90° degree pulse width varied between 11 u.s and 13 u,s; recycle delay was 2 s; spectral width varied from 13 kHz to 20 kHz depending on the sample, and the number of scans was 100. FID Phase Phase cycling 4 e 0, n, n, 0, n/2, 3n/2, 3TT/2, n/2 0, JI, 7 i , 0. n/2, 3TC/2, 3TX/2, n/2 Figure 2.4 Single pulse sequence 24 Experiment All multiple quantum (MQ) spectra were acquired simultaneously as separate slices of a three-dimensional spectrum using the 3D MQ-NMR pulse sequence (Fig. 2.5) [4]. MQ coherences excited by the second pulse evolve in variable time t{ and are converted into observable one-quantum coherences by the third pulse. Individual echoes S(t2) are acquired as a function of time /, and phase

,/ ,• . / + 2 ^ ) / , , / / . z + | ( . y , ; - o , ) ( / , / - + / , / ; ) ] (3.1) * / '\" '/ j>i where /\"*\" and I~ are raising and lowering spin operators, v, is the resonance frequency of nucleus /'; and Ji} and ZX are the indirect and dipolar coupling constant between nuclei / and j in the molecule. The eigenstates and eigenvalues that govern spectral frequencies and intensities can be calculated from diagonalization of the Hamiltonian for a given set of spectral parameters. Therefore an initial set of spectral parameters is needed to simulate the experimental spectrum. 27 Spectral analysis One way of choosing an appropriate initial set of parameters for a molecule is the choosing a set for the same liquid crystal solvent from a molecule with similar size and shape. This approach is only successful in a small number of simple cases (like low-spin symmetrical systems). For more complicated cases (Fig. 3.1) MQ spectra were analyzed first. Due to the poor resolution of MQ spectra (up to 100 Hz) the obtained set of spectral parameters is rather imprecise but it serves as a good starting point for analysis of the high-resolution spectra (where observed line widths were of the order of a few Hz) (see Table 3.2). Using the initial set of spectral parameters a trial spectrum was calculated. Calculated frequencies were then assigned to the experimental ones using cursor control in the graphical program called S M [1] with macros written by Ray Syvitski and spectral parameters were adjusted in a least-square fitting routine. Assignment and reassignment of frequencies and adjustment of spectral parameters were repeated until a reasonable fit was achieved (a fit that is of the order of the digital resolution of the experimental spectra). During the assignment process non-resolvable and low intensity spectral lines were assigned last to avoid a misleading and meaningless fit. The program used to calculate spectra and iteratively adjust spectral parameters was- L E Q U O R [2] for high-resolution spectral analysis and its modified version for MQ analysis [3]. 28 Spectral analysis . , . i , . . . i . , . i -5000 0 5000 Frequency (Hz) Figure 3.1 High-resolution proton NMR spectra of D C T (A), C L M X (B) and MESIT (C) in ZLI 1132 liquid crystal at 300 K with total number of calculated spectral lines of 69, 782 and 3819. Spectral lines marked with an (*) belong to the internal standard T C B . The rolling baseline is the unresolved spectrum of the ZLI 1132 liquid crystal. 29 Spectral analysis The high-resolution spectrum of T C B (3-spin system) is a 1:2:1 triplet with a splitting of 3-DHH and chemical shift that corresponds to the position of the middle peak (Table 3.1). Table 3.1 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of T C B 3-spin System Solutes Parametera ZLI 1132 Liquid Crystal 55 wt% 1132/EBBA EBBA T C B with DCT -199.68(01) -171.92(02) -138.19(03) 3536.85(02) 1875.15(04) 3920.67(06) R M S error 0.020 0.045 0.064 T C B with C L M X -208.27(04) -163.52(04) -144.15(05) °nb 2187.67(09) 2590.95(10) 3783.51(11) R M S error 0.095 0.099 0.113 T C B with MESIT -212.13(04) -167.65(00) -144.14(02) OH\" 2586.97(10) 2528.62(00) 2740.47(05) R M S error 0.097 0.001 0.051 a Parameter in Hz b Frequency is referenced to an arbitrary zero. Spectra were acquired at 500.13 MHz. Determination of spectral parameters of DCT (6-spin system) is a challenging task since DCT has 3 different chemical shifts and 5 independent dipolar coupling constants. In this case an initial set of spectral parameters was obtained from the (N-1) and (N-2)Q spectra (Fig. 3.2). First the (N-1)Q spectrum was fitted using a modified version of the program L E Q U O R that allows either independent adjustment of 5(>. and/or Dtj within the fitting 30 Spectral analysis routine. Knowing the geometry of the molecule and the relation between Sy and ZX (Eq. 1.11), by adjusting the Sy, D.tj is adjusted automatically. In some cases varying order parameters is an advantage since a smaller number of parameters are fitted to and therefore the spectral analysis is simplified. In the case of the +5Q spectrum of DCT, instead of fitting to 5 independent dipolar coupling parameters only two independent order parameters (Szz and Sxx - order parameters in the plane of the benzene ring (Fig. 3.6)) were adjusted within the fitting routine. Since the geometry of DCT hasn't been determined previously, the known geometry of TCB a molecule with similar size and shape, taken from ref. [4] was used as an initial geometry needed to calculate independent /^parameters. The +5Q and +4Q spectra were easily fitted (Fig. 3.3 and Fig. 3.4) with a R M S error of 4 Hz and the set of spectral parameters obtained (Table 3.2) were then used to simulate the high-resolution spectrum. A total of 68 lines were assigned in the high-resolution spectrum of DCT in E B B A with a R M S error of 0.343 Hz. The fitted high-resolution spectrum is shown in Fig. 3.5 and fitting parameters are presented in Table 3.3. In the calculated spectrum a Lorentzian line broadening of 2 Hz was used to incorporate the natural line broadening effect due to motion and collision in the system. Similar spectral strategy was applied in solving the high-resolution spectra of C L M X and MESIT molecules. Fitted spectral parameters of those molecules are presented in Tables 3.4 and 3.5. 31 Spectral analysis Table 3.2 Fitting Parameters and R M S Errors from Analysis of MQ and High-Resolution Spectra of DCT in E B B A EBBA EBBA EBBA Parametera +5Q +4Q +1Q spectrum spectrum spectrum Dl2 -120.94 -123.95 -126.39(17) -153.60 -157.74 -157.90(17) - 399.42 -396.10 -394.60(08) -63.18 -63.15 -63.58(043) A»5 1161.58 1174.49 -\" 1173.95(08) 01 6 3861.57 3831.77 3836.03(20) 3913.77 3882.17 3891.44(23) 1686.31 1706.67 1710.42(15) R M S error (Hz) 3.092 4.094 0.343 Number of 6 14 70 assigned lines a Parameter in Hz b Frequency is referenced to an arbitrary zero. Spectra were acquired at 500.13 MHz. 32 Spectral analysis + 6 Q + 5 Q + 4 Q + 3 Q + 2 Q +1Q + 0 Q 20000 10000 . 0 10000 20000 Frequency (Hz) Figure 3.2 All positive MQ spectra of D C T (6-spin system) in E B B A liquid crystal at 300 K With T =12 ms, recycle delay of 2 s, two scans for each n=16 phase increments per r,, 1024 tx increments, F2 spectral width of 25 kHz and Fl spectral width of 50 kHz. The +0Q spectrum strong central line was cut off for clarity 33 Spectral analysis 10000 12000 14000 16000 Frequency (Hz) 18000 20000 Figure 3.3 Experimental (top) and calculated (bottom) +5Q spectra of DCT in E B B A Line width in the experimental spectrum is approximately 80 Hz. Two lines around 12000 Hz in calculated spectrum are not visible separated. The intensities of the-calculated spectrum do not correspond with those of the experimental spectrum 34 Spectral analysis I I I L I ,, I I I I I I I I I I 5000 . ... 10000 . . 15000 . Frequency (Hz) ' Figure 3.4 Experimental (top) and calculated (bottom) +4Q spectra of DCT in E B B A Line width in the experimental spectrum is approximately 70 Hz and the intensities of the calculated spectrum do not correspond with those of the experimental spectrum 35 Spectral analysis Figure 3.5 Experimental (top) and calculated (bottom) high-resolution proton spectra of DCT in E B B A at 300 K. Spectral lines marked with an (*) belong to the internal standard TCB and resonances marked with an (•) are from unknown impurities (the DCT solute is an old sample that has probably degraded a little bit over a long period of time); less intense resonances (+) belong to the lock solvent (acetone-d6 in a capillary tube); the rolling base line is the unresolved spectrum of E B B A liquid crystal. In the experimental spectrum the line width at half-maximum height varied between 2-8 Hz. Lorenzian line broadening used in the calculated spectrum is 2 Hz. The intensities of the calculated spectrum closely correspond with those of the experimental spectrum. 36 Spectral analysis Table 3.3 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of DCT 6-spin System Solute DCT Parametera ZLI1132 Liauid Crystal 55 wt% 1132/EBBA EBBA Dn -255.66(19) -201.31(19) -126.39(17) Dn -176.15(21) -162.89(21) -157.90(17) -493.48(08) -440.10(08) -394.60(09) -56.26(36) -56.39(36) -63.58(44) A.4 1030.85(08) 1039.46(09) 1173.95(08) ° 1 2 - - -J n 2.06(44) 1.85(40) 2.20(32) -0.70(14) -0.95(14) -0.72(17) -0.38(65) -0.85(69) -0.50(84) O\", 3569.70(21) 1862.77(23) 3836.03(20) 0-3 3411.09(24) 1787.45(27) 3891.44(73) 1534.72(15) -266.03(16) 1710.41(15) R M S error 0.327 . 0.546 0.343 Number of 68 61 70 assigned lines a Parameter in Hz * b Jl2 coupling constant was not determined due to insensitivity of the fitting process 0 Frequency is referenced to an arbitrary zero. Spectra were acquired at 500.13 MHz. 37 Spectral analysis Table 3.4 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of C L M X 9-spin System Solute Parameter' ZLI1132 Liquid Crystal 55 wt% 1132/EBBA EBBA C L M X H H. H, C l H. H. 0 , 2 0 ,4 0,7 D 34 D 45 D, ,2 ,3 J J J J O\", O-3 O-4 R M S error Number of assigned lines 34 47 -160.34(11) -70.42(05) -659.22(05 -414.89(07) 1292.43(03) -60.26(03) 1.99(18) 1.53(10 -0.51(10) -0.66(11) -0.75(13) -0.18(06) 2099.63(09) 2214.41(12) 91.23(05) 0.560 445 -144.20(10) -59.49(07) -495.41(07) -374.65(08) 1089.89(03) -54.35(03) 1.97(18) 1.75(13) -0.43(14) -0.69(16) -0.76(17) -0.22(06) 2497.30(10) 2546.08(13) 422.84(06) 0.401 396 -167.69(09) -61.95(09) -403.10(07) -439.08(06) 1141.24(03) -63.74(03) 2.01(18) 1.37(15) -0.78(17) -0.53(17) -0.77(14) -0.21(06) 3675.27(10) 3657.00(13) 1578.88(06) 0.453 379 a Parameter in Hz b Frequency is referenced to an arbitrary zero. Spectra were acquired at 500.13 MHz. 38 Spectral analysis Table 3.5 Fitting Parameters and R M S Errors from Analysis of High-Resolution Spectra of MESIT 12-spin System Solute Parameter1 ZLI1132 Liquid Crystal 55 wt% 1132/EBBA EBBA MESIT H H H. H , H, 9, S H H 1 2 H „ D l 2 -220.49(06) -179.26(06) -168.75(06) 0 , 4 -571.94(03) -465.51(03) -439.58(03) 0 , 7 -82.99(06 -67.583(07) -63.79(07) 044 1518.15(02) 1234.96(03) 1166.50(03) 8 9 047 -82.97(02 -67.53(02) -63.82(02) 1.72(11) 1.57(13) 1.76(12) -0.71(05) -0.66(07) -0.67(07) AI -0.44(10) -0.41(01) -0.45(14) *^47 -0.22(04) -0.19(05) -0.22(05) 2510.77(08) 2420.58(10) 2592.37(09) 0-4 492.02(05) 367.53(07) 537.91(06) R M S error 0.474 . 0.341 0.371 Number of 494 355 381 assigned lines a Parameter in Hz b Frequency is referenced to an arbitrary zero. Spectra were acquired at 500.13 MHz. 39 Spectral analysis 3.3 M O L E C U L A R S T R U C T U R E AND O R D E R P A R A M E T E R S The relative positions of nuclei and order parameters for a given solute were calculated from a simultaneous fit to the dipolar couplings obtained for all three liquid crystals. For this purpose a slightly modified version of a program called S H A P E [3] was used. The program calculates the dipolar couplings according to Eq. 1.11 using initial inputted geometry and it performs the least-square minimization routine NL2SNO [5] that minimizes the difference between calculated and experimentally determined dipolar couplings (which are part of the input data). The exact geometry of solutes used in this study, except TCB, hasn't been measured yet so the initial geometry used in the fitting routine was chosen to be geometry of the molecule with similar size and shape (such as T C B (ref. [4])). For molecules with rotating groups, such as a methyl group, dipolar interactions must be averaged over a whole rotation. Each methyl group was modeled with a potential function V - i ; ( l ~ c o s 6 a ) / 2 . ' (3.2) where V6 is fixed at 60 J/mol [6]. The potential minimum corresponds to the position of the proton in a methyl group that is perpendicular to the benzene ring. Each methyl group is rotated independently through 360° in 10° steps. For each position dipolar couplings between hydrogens were calculated and averaged over all possible ~v/-conformations, with each conformation weighted by the Boltzman factor (~ e /kT). As a result of the minimization routine, relative structural parameters and order parameters of the studied solutes DCT, C L M and MESIT were successfully determined and are presented in Tables 3.6 and 3.7. Order parameters were determined with high accuracy. That is quite important since the values of the order parameters of a solute may be quite small and the prime goal of this study is to explore various models for the anisotropic potential. 40 Spectral analysis Table 3.6 Structural Parameters from Fits to Dipolar Couplings for DCT, C L M X and M E S I T a Parameter 3 TCB d DCT e CLMX e MESIT e r(Cx-C2) 1.3908 1.3920(05) 1.3870(19) 1.3861(37) r ( C 3 - C 4 ) 1.3908 1.3898(08) 1.3922(20) 1.3861(37) r ( C 4 - C 3 ) 1.3908 1.38909 1.3919 9 1.3861(37) r(Cx-H) 1.0940 1.0893(14) 1.0865(24) 1.0881(18) r(C5-H) 1.0940 1.0914(14) 1.0903(25) 1.0881(18) r(C2-X c) 1.7326 1.5283(18) 1.7326 f 1.5267(27) r(C4-X h) 1.7326 1.7326 f 1.5218(30) 1.5267(27) r(C-H)mulhyl - 1.0997(13) 1.1114(19) 1.1056(22) r(C6 — Cmethyi) - 1.7326 f 1.5281(30) 1.5267(27) Z ( C , C 2 C 3 ) 122.00 121.77 9 121.98 9 121.64(14) Z(C2C3C4) 118.00 118.01(07) 118.22(12) 118.36(14) Z(C,C4C5) 122.00 122.21(15) 121.72(12) 121.64(14) *(C4C5C6) 118.00 117.78 9 118.13 9 118.36(14) Z(C5C(X h) 121.00 119.00' 118.90(12) 119.18(07) Z(C4C3H) 121.00 120.77(05) 121.35(13) 120.82 9 Z(C4C5H) ' 121.00 •' 121.'11 9 ' 120.939' \" 120.82 9 Z(CXC2X\") 119.00 119.11(05) 119.01(12) 119.18(07) ^( C6CmelhylHmethyl) - 110.47(07) 111.22(16) 110.94(14) R M S error - 0.321 0.548 0.566 41 Spectral analysis 3 See figure 3.6 for structure and atom labeling of molecules. Bond distances are in (A) bond angles in degrees and errors are in Hz b X=CI for TCB and CLMX; X=CH3 for DCT and MESIT 0 X=C/ for TCB and DCT; X=CH3 for CLMX and MESIT d Geometry taken from Ref. [4] e Initial a priori geometries are TCB geometries taken from Ref. [4] 1 Parameter not varied during fit 9 Parameter calculated from the bond angles and lengths of the carbon skeleton z C L M X \" i M E S I T Figure 3.6 Atom labeling of solute molecules 42 Spectral analysis Table 3.7 Table of Order Parameters from Fits to Dipolar Couplings Solute Order Parameter + 1132 Liauid Crystal 55wt% 1132/EBBA EBBA TCB/DCT \" 0.13480(01) 0.11606(01) 0.09329(02) -0.26960(01) -0.23212(03) -0.18658(04) szz 0.13480(01) 0.11606(01) 0.09329(02) T C B / C L M X ~ 0.14060(03) 0.11039(03) 0.09731(03) -0.28120(05) -0.22078(05) -0.19462(05) szz 0.14060(03) 0.11039(03) 0.09731(03) TCB/MESIT ~ 0.14321(00) 0.11318(00) 0.09731(01) -0.28641(00) -0.22635(00) -0.19462(03) szz 0.14321(00) 0.11318(00) 0.09731(01) DCT 0.17265(80) 0.13162(50) 0.08347(50) -0.27022(11) -0.23001(82) -0.19459(86) szz 0.09757(31) 0.09838(32) 0.11112(36) C L M X s,. 0.10627(40) 0.09565(35) 0.11186(39) -0.28468(03) -0.22793(83) -0.21573(79) szz 0.17841(63) 0.13228(48) 0.10387(40) MESIT ~ 0.14467(66) 0:11770(54) 0.11117(51) -0:28935(32) -0.23541(08)' -0.22234(02) szz 0.14467(66) 0.11770(54) 0.1117(51) * Order parameter that is perpendicular to the benzene ring + sxx+svy+s_.=o AX yy 43 3.4 SUMMARY Spectral analysis In this study spectral, structural and orientational order parameters for TCB, DCT, C L M X and MESIT dissolved in three different liquid crystals were determined. An initial set of spectral parameters, used in the analysis of the high-resolution spectrum, was estimated from the analysis of the (N-1) and (N-2)Q spectra adjusting either Sy or Dtj independently. With this approach analysis was simplified and therefore analysis time was significantly reduced. Highly accurate spectral parameters were extracted by analyzing the high-resolution spectrum. Accurate order parameters and relative positions of nuclei were obtained by simultaneously fitting to all dipolar coupling parameters of the solute in all liquid crystals. . , : References: [1] Graphical Interface program S M , Edition 2.2.0. Jan. 1992, Lupton, R. and Monger, P. [2] Diehl, P., Kellerhals, H. and Niederberger, W., 1971, J. Magn. Resonance, 4, 352. [3] Syvitski, R. and Burnell, E.E., 2000, J. Chem. Phys., 113, 3452. [4] Almenninbgen, A. and Hargittai, I., 1984, J. Mol. Struct., 116, 119. [5] Dennis, J . E., Gay, D. M. and Welsch, R. E., 1981, ACM Trans. Math. Software, 7, 3. [6] Lister, D., MacDonald, J . and Owen, N., 1978, Internal Rotation and Inversion, Academic Press, London. 44 4. ORIENTATIONAL ORDERING IN NEMATIC LIQUID CRYSTALS Investigation of different contributions to the orientational mechanisms in nematic liquid crystals 4.1 INTRODUCTION Previous studies on orientational mechanisms in nematic liquid crystals gave great insight into the nature of the physical interactions responsible for orientational behaviour in these phases. Generally they show that the main orientational ordering mechanism comes from short-range interactions [1-7], i.e. the repulsive interactions that are closely correlated with the size, shape and flexibility of the molecules. Contributions from the long-range interactions, interactions that are due to properties that describe the distribution of the charge over a molecule (like dipole, quadrupole, polarizability) seem to have a less dominant effect on the orientational ordering [8-11]. To what extent each electrostatic long-range interaction contributes to the orientational mechanisms is still an open debate. The purpose of this study is to determine the effects of permanent dipoles, quadrupoles and molecular polarizabilities on orientational ordering. In order to reduce the effect of dominant short-range interactions and to emphasize different electrostatic interactions, solutes with similar size and shape but different electrostatic properties are chosen as probe solutes in various liquid crystals. The set of chosen solutes with similar size arid shape (Fig. 4.1) have either C2 or C3 type symmetry. If they had the same size and shape and if the size and shape is the only orientational mechanism then the anisotropy of the experimentally determined order parameters (Sxx-S__ ) would be zero due the same symmetry. And if any other orientational mechanisms beside size and shape also contributes to the orientational ordering then the anisotropy in the order parameters would be influenced by them, and they should differ from zero. Particularly the anisotropy in the order parameters in the magic mixture (zero electric field gradient and quadupole interaction) will allow the clear manifestation of dipole and polarizability effects. 45 Orientational ordering in nematic liquid crystals Comparing the anisotropy of the experimentally determined order parameters (SXX -S-Z) in a series of solutes with the same size and shape but different electrostatic properties, a qualitative picture about long-range electrostatic contributions to the orientational mechanisms can be drawn. It should be stressed that behind this way of examining the experimental results is the assumption that all solutes have the same size and shape. A more quantitative approach to learning about intermolecular interactions is to compare the experimental order parameters with calculated order parameters from theory or models, or those determined from computer simulations. As described in the introductory chapter two short-range potential (Cl and size and shape) models and three long-range potential (dipole, quadruple and polarizability) models were utilized to investigate the orientational mechanism in nematic liquid crystals. H . H C L M X • M E S I T ~ x Figure 4.1 Chosen set of solutes with the same size and shape but different electrostatic properties presented in their coordinate system 46 Orientational ordering in nematic liquid crystals 4.2 DETERMINATION O F A CONSISTENT S E T O F O R D E R P A R A M E T E R S When solutes are dissolved in liquid crystal solvents they tend to slightly perturb the liquid crystal environment. The perturbation effect depends on solute properties and solute concentration in a given sample. If one needs to compare the solute orientational order parameters in different samples then all the above mentioned factors must be compensated for. In other words a consistent set of order parameters must be obtained. That can be done in a few different ways [1,7,11-13]. T For example, spectra can be recorded at the same reduced temperature (TR= , where • ' ' \" ' TN1 TN1 is the nematic-to-isotropic phase transition temperature) or order parameters of each solute in the liquid; crystal can be measured as a function of concentration and then extrapolated to zero concentration. In the present study a constant real temperature scaling approach, shown to give the most consistent results, [13] is utilized. Order parameters were scaled using the internal standard TCB and the simple equation: c r. scaled _ ij( solute) ^ ij (solute) ~ „ ' ij (TCB-reference) (4-1) ^ij(TCB) where S!J(sulule) and SiJ(TCB) are the order parameters of the solute and internal standard T C B in the same sample tube and Sij(TCIS_re/erence) is the order parameter of internal standard T C B used as a reference tube (arbitrarily chosen to be T C B with MESIT sample tube) in the same liquid crystal. The set of scaled and non-scaled order parameters is presented in Table 4.1. Differences between scaled and non-scaled values of order parameters of different solutes (DCT, C L M X and MESIT) in the same liquid crystal are less then 6 % using this scaling method (see Fig. 4.2). Also comparing scaled and non-scaled values of order parameters of TCB coodisolved with various solutes for the same liquid crystal the discrepancies are less than 6% (Table 47 Orientational ordering in nematic liquid crystals 3.7). Both discrepancies imply that the main contribution to the difference comes from solute properties since the concentration of solutes is very small and roughly the same. Table 4.1 Scaled and Non-Scaled Order Parameters Solute Scaled/ Non-scaled Order Parameter' 1132 Liauid Crystal 55wt% 1132/EBBA EBBA T C B + 0.14321/0.14321 0.11318/0.11318 0.09731/0.09731 C< scaled j r t -0.28642/-0.28642 -0.22636/-0.22636 -0.19462/-0.19462 S^'\"\"1 ISZZ 0.14321/0.14321 0.11318/0.11318 0.09731/0.09731 DCT g scaled j Q 0.18342/0.17265 0.12835/0.13162 0.08707/0.08347 o scaled i r t *yy '^yy -0.28707/-0.27022 -0.22429/-0.23000 -0.20297/-0.19459 £ i scaled j Q 0.10365/.09757 0.09594/0.09838 0.11590/0.11112 C L M X g scaled j g 0.10824/0.10627 0.09807/0.09565 0.11185/0.11186 ri scaled / r t ^ '^yy -0.28996/-0.28468 -0.23369/-0.22793 -0.21572/-0.21572 ITI scaled j azz T C B 0.025 h -0.017 \" 0.349' -0.690 0.341 ' 21.634' 9.438 21.663 DCT 0.147\" -7.188 -1.555 -1.407 2.962 22.225 10.131 20.303 C L M X 0.011 * 7.024 2.418 -1.973 -0.445 20.342 10.909 21.799 MESIT 0.125\" -0.021 \" 1.139' -2.146 1.007' 20.883' 11.529 20.515 a Values calculated with respect to the center of mass using Gaussian 98 with B3LYP/6-311++G** (theory/basis set). Axis labeling is shown in Figure 4.1. \"Units of 1(T30 Cm (\\ Cm = 2.99793- \\Q\" esu ). c Units of 10\" 3 9 C-m\\ 1 C• m2 = 2.99793-1013esu • cm ). \"Units of K T 1 0 Cm2V-\\ 1 C - m2 • F\" 1 = 8.9878 • 10 1 5cm 3). e For all solutes the fi component is zero. ' For all solutes the Q and Qy, components are zero. 9 For all solutesi the a vi, and oc components are zero. h Values taken as zero (by symmetry) in the fitting routine. ' Due to a D3h type symmetry, xx and,zz tensor components of molecular parameters should be the same. Therefore their averaged values were used in the fitting routine. 53 Orientational ordering in nematic liquid crystals Table 4.4 Adjusted Parameters in the Fitting Procedure a/ Cl model combined with/without long-range models Cl fitting Dipole Quadrupole Polarizability SS fitting SS fitting SS fitting Liquid crystal parameter {R7,7. R \\ \\ } \" F c M a M» 8 MB a Error' k 3 parameter parameter paramater parameter parameter parameter Fit #1 1132 1.46 (05) e 6 ~ 9 e e e MM 0.96 (03) e e _ e e e e 1.32 EBBA 0.84 (03) e e e e e e Fit #2 1132 1.46 (05) -2.9 (8.0) e e e _ e e MM 0.96 (04) 1.7 (8.5) _ e _ e _ e e e 1.29 EBBA 0.83 (03) 8.4 (8.7) e _ e e e e Fit #3 1132 1.48 (03) e -2.4 (0.7) e e e e MM 0.96 (02) e -0.1(0.7) e e e e 0.86 EBBA 0.80 (02) e 4.0 (0.7) _ e e e e Fit #4 1132 1.07 (12) e e 68.9 (19.9) e e e MM 1.02 (11) e e -9.9 (20.4) e e e 0.79 EBBA 1.58 (12) e e -135.3(21.0) e e e Fit #5 1132 1.48 (04) 1.9 (5.8) -2.4 (0.7) e _ e e e MM 0.96 (03) 2.1 (6.2) -0.1 (0.8) e e e e 0.86 EBBA 0.80 (02) 0.5 (6.3) 4.0 (0.8) e e e e Fit #6 1132 1.07 (13) 1.1 (5.3) e 69.9 (21.5) e e e MM 1.01 (12) 1.2 (5.6) _ e 8.8 (22.1) _ e e e 0.79 EBBA 1.57 (13) . 1.1 (5.7) e -134.4(22.6) e .. e e Fit #7 1132 3.18 (61) e -11.5(9.8) -289.5(103.6) _ e e e MM 2.95 (40) _ e -33.0 (6.7) -340.5(69.1) _ e _ e _ e 0.46 EBBA 2.83 (38) _ e -7.1 (2.1) -348.7 (65.5) _ e _ e e Fit#8 1132 3.31 (60) 3.8 (3.1) -12.5 (3.2) -312.8(102.2) _ e e e MM 3.03 (40) 4.7 (3.3) -11.6 (2.2) -354.6 (67.8) _ e _ e _ e 0.42 EBBA 2.88 (38) 3.5 (3.3) -7.6 (3.0) -358.6 (64.6) e e e 54 Orientational ordering in nematic liquid crystals bl S S model combined with/without long-range models Cl fitting Dipole Quadrupole Polarizability SS fitting SS fitting SS fitting Liquid crystal parameter (*,/ * , v ) 9 F 0 Mxx a Myy a Error' parameter parameter paramater parameter parameter parameter Fit m 1132 1.48 (09) e e e MM 1.00 e e e -628 (27) 1257 (56) -630 (29) 1.62 EBBA 0.89 (06) e e _ e Fit #10 1132 1.49 (10) -3.1 (1.3) e e MM 1.00 1.2 (11.8) e e -628 (27) 1256(60) -628 (33) 1.60 EBBA 0.88 (06) 7.5 (11.8) _ e e Fit #11 1132 1.51 (04) e -6.2 (0.6) e MM 1.00 e -2.6 (0.6) e -643 (12) 1298 (25) -655 (13) 0.66 EBBA 0.83 (02) e 1.9 (0.6) e Fit #12 1132 1.04 (15) e e 126.0 (13.1) MM 1.00 e _ e 44.8 (13.4) -467 (48) 942 (97) -475 (49) 0.72 EBBA 1.53 (19) e e -47.2 (13.6) Fit #13 1132 1.51 (04) -8.6 (5.3) -6.2 (0.6) _ e MM 1.00 6.6 (4.9) -2.7 (0.6) _ e -648 (12) 1294 (26) -646 (14) 0.61 EBBA 0.83 (02) 4.6 (4.9) 1.9 (0.6) e Fit #14 1132 1.04 (16) 4.0 (5.3) _ e 134.6(13.7) MM 1.00 4.0 (5.5) e 45.8 (14.1) -466 (51) 931 (103) -465 (51) 0.70 EBBA 1.54 (20) 5.6 (6.2) _ e -46.0 (14.1) Fit #15 1132 1.03 (26) _ e -6.9(2.7) -19.3 (58.2) MM 1.00 e -7.1 (1.9) -97.1 (38.4) -1018 (159) 2058 (320) -1039 (162) 0.52 EBBA 0.95 (19) e -3.3(1.8) -110.6(36.5) Fit #16 1132 1.12 (23) 10.5 (3.8) -9.1 (2.4) -63.6 (50.8) MM 1.00 10.4 (3.7) -8.2(1.7) -114.5(33.8) -1095 (132) 2189 (263) -1093(132) 0.40 EBBA 0.94 (16) 8.9 (3.7) -4.1 (1.5) -124.8(30.8) 55 . Orientational ordering in nematic liquid crystals cl long-range models alone and combined Cl fitting Dipole Quadrupole Polarizability SS fitting SS fitting SS fitting Liquid crystal parameter (Rzz - Rxx ) \" (E?.z-Exx)\" MM a Mvv 3 Error' k 8 Kr/. parameter parameter paramater parameter parameter parameter Fit #17 1132 e 167.1 (106.9) e e e e e MM e 133.5 (108.6) _ e e e e e 16.41 EBBA _ e 130.0 (108.8) e e e e _ e Fit #18 1132 e _ e 27.3 (9.3) e e e _ e MM e e 23.4 (9.4) e _ e _ e e 13.98 EBBA e e 24.5 (9.6) e e e _ e Fit #19 1132 e _ e e 246.4 (15.8) e _ e e MM e _ e e 169.4 (12.1) e _ e e 2.81 EBBA e _ e e 146.2 (10.9) e _ e e Fit #20 1132 e 56.8 (124.0) 25.0 (11.1) e _ e _ e e MM e 36.8 (121.9) 21.8 (11.2) _ e e e _ e 13.89 EBBA e 31.0 (118.6) 23.2 (11.3) e _ e e e Fit #21 1132 e 12.1 (17.4) _ e 245.8 (16.2) _ e e _ e MM e 12.5 (18.1) _ e 168.4 (12,5) e _ e e 2.72 EBBA e 18.5 (18.6) _ e 144.5 (11.3) _ e e e Fit #22 1132 _ e e 5.6 (1.0) 252.8 (8.5) _ e e _ e MM e e 5.3 (1.0) 166.0 (6.0) e e _ e 1.22 EBBA e _ e .8.6(1.0) .,137.8(5.7) _ e _ e _ e fif#23 \" 1132 8 0.44 (8.2) 5^ 6 (1.1) 252.8(9.0) _ e e _ e MM e 0.7 (8.7) 5.3(1.1) 166.0 (6.4) e e e 1.22 EBBA e -0.6 (8.9) 8.6 (1.1) 137.8 (6.0) e e _ e a Units of 10~23 J. b Unitsof 1036 V/Cm2. c Units of 1017 V/m2. \" Units of 1017 V2/m2. e Parameter was not adjusted. f Units of 10'2. 56 Orientational ordering in nematic liquid crystal A. Short-Range Model Comparison scaled Examining the difference between S^\"HL1 and S\"XCorder parameters as a function of (Fig. 4.4, Fit #1 and Fit#9) that result from fits to short-range models alone it can be stated that in magic mixture the Cl model is excellent (Fig. 4.4, A 2 , with errors up to 5-6%) while the S S model is less satisfying (Fig. 4.4, B 2 , with errors up to 14%). ZLI 1132 55 wt% 1132/EBBA EBBA 008 010 012 014 016 0 16 0 20 022 008 010 0 12 0.14 0 16 0 18 0 20 0 22 008 010 0 12 0 14 0 16 0 18 0 20 0 22 CO u 0.06 C O 012 0 14 0 16 0.18 0 20 0 22 „ scaled 0.06 0 10 0.12 0.14 0.16 016 0.20 0.22 „ seated 0.06 0.10 012 0.14 0 16 0 18 0 20 0.22 _ scaled Figure 4.4 Differences between . V a n d S^'1' as a function of S[:\"HJ for short-range models alone. Series A are results from Fit #1 (Cl model) with RMS= 1.32-10 2 and series B are results from Fit #9(SS model) with RMS=1.62 10\" 2. • TCB T DCT • C L M X MESIT 57 Orientational ordering in nematic liquid crystal The Cl model uses geometries determined individually for each solute from high-resolution spectral analysis. On the contrary the S S model assumes that all solutes have the same geometry, and it doesn't take into account that DCT and C L M X geometries slightly deviate from D 3 h symmetry. That could explain why the Cl model fits better than the S S model and therefore in the following investigations only the Cl model in combination with long-range models will be considered. Not surprisingly the fits for pure liquid crystals are poorer (Fig. 4.4 A 3 , Eh, B3)) (with discrepancies up to 30%, see DCT in A 3 , or B^ . The same trends are seen for S ^ ' a n d gscaled j ^ j s a | | 0 w s f o r investigation of what other contributions to the short-range interaction might improve the fit and to what extent they might contribute to the orientational mechanism. In that manner of thinking different long-range contributions will be additively combined with short-range ones and their combined contributions will be examined in part C-E . B. Long-Range Model Comparison Although short-range models reasonably well describe the orientational mechanism, especially in the magic mixture, let's see what information we can get if we assume that the orientational mechanism can be described with only long-range electrostatic interactions. Judging by the R M S of the Fits #17-19 (Table 4.4) that belong to various electrostatic models alone and by comparing the differences between experimental and calculated order parameters, dipole (Fit #17) alone (Fig. 4.5, series C) and quadrupole (Fit #18) alone (Fig. 4.5, series D) are significantly worse than short-range (Fit #1) (Fig. .4.5, series A). The Fit #19 (Fig. 4.5, series E) that accounts for polarizability effects alone with R M S of 2.8110\" 2 is qualitatively and quantitatatively better than other long-range fits alone but still worse than short-range fits alone.(RMS(Fit#1)=l.32lO- 2rRMS(Fit#9)=l.62lO- 2). Almost the same R M S and fitted potential parameters are reached when the Fits #17-19 were repeated only for the magic mixture. The anisotropy in the polarizability tensor is strongly affected by size and shape anisotropy. That connection between polarizability with size and shape is even more stressed by the fact that the polarizability fit alone, although worse than the size and shape model, explains relatively well the experimental results. More about polarizability effects to the 58 Orientational ordering in nematic liquid crystal orientational ordering mechanism will be discussed later when comparing short-range with long-range contributions in an additive manner. Although Fits #22 and #23 (combination of two/three long-range models) show small R M S the fits are meaningless since fitted F77 values have very large associated errors due to the large number of adjusting parameters and possible correlation between potential models. Therefore those fits won't be taken into consideration. Using only magic mixture spectral parameters in the fitting routine where the quadrupole contribution is expected to be zero, the aim was to investigate the effects of other long-range potentials like dipole and polarizability. Fig. 4.6 contains fits of different pairs of short-range with long-range interactions only for the magic mixture. Fit F belongs to short-range (Cl) model alone, fit G is combination of C l model with dipole model, fit H is Cl model with quadrupole model and fit I is Cl model with polarizability model. The equality of the Fits A and H indicates that the quadrupole interaction in magic mixture is annulled and that potential parameter F77 ~0. It is also obvious that overall all fits look the same (Fig. 4.6) with the same RMS=0.94 10~2, the same kZ7 and zero Fzz , (RZZ -RXX) and (E77 -E2^ parameters. From all of the above, we can say that in magic mixture long-range interactions do not play an important role in the orientational mechanism and that short-range interactions are the only dominant contributor to the orientational ordering. 59 Orientational ordering in nematic liquid crystal 55 wt% ZLI 1 1 3 2 / E B B A 0.3 03 Xi 0 0 m # K1 OB Q1D 012 014 0*6 OIG Q3D K 3 scaled scaled 008 010 012 014 018 018 030 on o scaled Fig 4.7 Differences between Stf''\"J and S^f as a function of Ss™,ed Series A and B are results from Fit #1 (Cl model) and Fit #9 (SS model); Series J and K are results from Fit #2 (Cl+Dipole) and Fit #10 (SS+Dipole) • T C B V D C T • C L M X MESIT 63 Orientational ordering in nematic liquid crystal 55wt%2LI 1132/EBBA EBBA V 005 _ o m • O • • •0 05 • • -OOS • T one oio gti O H 010 010 0 20 0 22 G1C 0.10 0.20 022 ooe a 10 012 a 4 010 0 10 020 0.22 005 . o 005 • M _ ...» • L 3 oat 010 G12 O H 010 UC ..1. ooe aia 012 a 4 0 10 0 10 020 022 oos . o 0.06 v m M 2 ... • y M 3 000 0.10 012 O H ftlfl • 10 .• 022 OtS 0.1S 0.20 0 22 o Jf oos - o DOS . o • ¥ OOt 0.10 012 014 O.ie Nl CO 0 20 0.22 -ace o.oo oto 0.12 an N 2 gto 0.1C 020 022 -005 • • 0 00 0 10 0 12 u N 3 scaled scaled scaled Fig 4.8 Differences between and S ' f as a function of M XX XX XX Series A and L are results from Fit #1 (Cl model) and Fit #3 (Cl+Quadrupole); Series M and N are results from Fit #2 (Cl+Dipole) and Fit #5 (CI+ Dipole+Quadrupole) • T C B T DCT • C L M X MESIT 64 Orientational ordering in nematic liquid crystal Certain Fits (#7, #8, #15, #16) that contain a quadrupole contribution in combination with other long-range interactions are discarded as meaningless, since the potential parameters F77 obtained have large calculated errors. The explanation for this finding might be the existence of a correlation between the different potential models. The reason for getting a good R M S in these cases is the fact that by adding another potential more available fitting parameters are introduced into the fitting routine. E. Polarizability Interaction Contribution Comparing the R M S errors of Fits #1, #4 and #6 or Fits #9, #12 and #14 significant change is noticed when the polarizability interaction is added to the fitting. This is more obvious when comparing series A (Cl model), O (Cl with Polarizability) and P (Cl with Dipolar and Polarizability model) (see Fig. 4.9). In the magic mixture the polarizability potential parameters (E27,— E2^ are zero within the calculated errors (Table 4.4), once more agreeing with the results found in magic mixture fits (section B, page 59, 61), i.e. that in the magic mixture long-range interactions are not affecting the ordering mechanism since the Cl short-range model so wonderfully explains the experimental results. In the component liquid crystals the polarizability model significantly improves the fit. All this indicates that the polarizability effect could be considered as one of the contributors to the orientational ordering mechanism. 65 Orientational ordering in nematic liquid crystal ZLI 1132 55 wt% ZLI 1132/EBBA EBBA 0.05 0.05 o • o • • -0.05 •0.05 • A l A 3 006 010 012 0.14 0.16 018 0.20 0.22 008 0 10 0 12 0 14 016 016 0 20 0 22 0.05 0.05 O • 4 • • •0.05 \" • * •0.05 • • Oi 0 2 0.06 0.10 0 12 0 14 0 16 Q. 16 0 20 0 22 0.06 010 012 014 0.16 0.18 0.20 0.22 008 0 10 0.12 0.14 0.16 0.18 0.20 0.22 0.05 0.05 • o T • am -0.05 CLOG -0.05 Pi P2 P3 0.08 0.10 0 12 0 14 0.16 0.18 0.20 0.22 II if 010 012 Q.14 0.16 0.16 0.20 C.22 0 06 010 012 0.14 0 16 018 0.20 022 o scaled ^ scaled o scaled Fig 4.9 Differences between S^** and S™'C as a function of Series A, O and L are results from Fit #1 (Cl model), Fit #4 (Cl+Polarizability) and Fit #6 (Cl+Dipole+Polarizability); • TCB T DCT • C L M X MESIT 66 Orientational ordering in nematic liquid crystal 4.5 SUMMARY Short-range interactions are the dominant orientational mechanism in liquid crystal environments. The importance of various electrostatic contributions to the orientational mechanism is still not completely understood. The role of various electrostatic interactions in the orientational mechanism is investigated using small and symmetric molecules with the same size and shape, therefore the same short-range interactions but different electrostatic properties. According to the R M S and differences in experimental and calculated order parameters of different fits to different combinations of long-range and short-range models, the dipole interaction appears to be least important. On the contrary, results for quadrupole interactions imply that improvement of the fits could be attributed to the effect of the quadrupoles on the orientational mechanism; however, predicted signs of the electric field gradients F77 for 1132 and E BBA contradict previous F77 study results for component liquid crystals. This interesting observation needs future investigation possibly using the Monte Carlo simulation method. Single potential fits for all liquid crystals show that the Cl model alone accounts the best for the orientational ordering. In the magic mixture polarizability, alone explains the experimental results better than the other elecrostatic interactions alone but still not as good as does the Cl model alone. Combined short and long range potential investigations show that in the magic mixture not only is the quadrupole interaction annulled but also the other long-range interactions as well. The polarizability effect appears not to be easily separated from short-range interactions since they both strongly depend on the size and the shape of the molecule. However, examining additive contributions of short-range models with and without the polarizability effect, the polarizability contribution should be considered as an important electrostatic effect in the orientational mechanism. 67 Orientational ordering in nematic liquid crystal References: [1] Burnell, E. and de Lange, C , 1998, Chem.Rev., 98, 2359. [2] Gelbart, W., 1982, J.Phys.Chem., 86, 4289. [3] Frenkel, D., 1989, Liq. Crystals, 5, 929. [4] Vertogen, G. and de Jeu, W., 1989, Thermotropic Liquid Crystals, Fundamentals, Springer, Heidelberg, 2 n d edition. [5] Vroege, G. and Lekkerkerker, H., 1992, Rep. Prog. Phys,, 55, 1241. • [6] Poison, J . and Burnell, E., 1996, Mol. Phys., 88,767. [7] Terzis, A., Poon, C , Samulski, E., Luz, Z., Poupko, R., Zimmermann, H., Muller, K., Toriumi, H. and Photinos, D., 1996, J. Am. Chem. Soc, 118, 2226. [8] Terzis, A. and Photinos, D., 1994, Mol. Phys., 83, 847. [9] Photinos, D., Samulski, E. and Toriumi, H., 1990, J. Phys. Chem., 94, 4694. [10] Emsley, J . , Palke, W. and Shilstone, G., 1991, Liquid Crystals, 9, 643. [11] Photinos, D., Poon, C , Samulski, E. and Toriumi, H., 1992, J. Phys. Chem., 96, 8176. [12] Celebre, G., de Luca, G., Longeri, M. and Ferrarini, A., 1994, Mol.Phys., 83, 309. [13] Syvitski, R., Pau, M. and Burnell, E., 2002, J. Chem. Phys., 117, 376. [14] Gaussian 98, Revision A.9, 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, 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, Baboul, 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, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J . L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle, and J . A. Pople, Gaussian, Inc., Pittsburgh PA, 1998. [15] Syvitski, R. and Burnell, E., 1997, Chem. Phys. Letters, 281, 199. [16] Patey, G., Burnell, E., Snijders, J . and de Lange, C , 1983, Chem. Phys. Letters, 99, 271. 68 5. CONCLUSIONS Extensive comparison analysis of experimental and theoretically calculated order parameters for a series of C 2 v and D 3 h type symmetry molecules in a magic mixture support the conclusion that orientation of these solutes is dominated by short-range hard body interactions. In pure liquid crystal components the results suggest that additional long-range interactions might play a role. The dipole interaction appears to be the least important while the quadrupole interaction fits predict different signs of electric field gradient than previous studies [15, 16]. The ambiguity associated with the quadrupole findings should be tested by using computer simulations (MC simulations) as well as checking the validity of statistical approximations employed in the theory. Finally, the trends among experimental order parameters are the most consistent with the polarizability/short-range model implying that polarizability effects should be taken as an important electrostatic orientational mechanism. 69 "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2004-11"@en ; edm:isShownAt "10.14288/1.0061197"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Chemistry"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Orientational ordering of small molecules in nematic liquid crystals"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/15464"@en .