Microscopic structure and dynamics of a partial bilayer smectic liquid crystal Yves Lansac, Matthew A. Glaser, and Noel A. Clark Condensed Matter Laboratory, Department of Physics, and Ferroelectric Liquid Crystal Materials Research Center, University of Colorado, Boulder, Colorado 80309 Received 30 April 2001; published 16 October 2001 Cyanobiphenyls nCB’srepresent a useful and intensively studied class of mesogens. Many of the peculiar properties of nCB’s e.g., the occurence of the partial bilayer smectic-A d phaseare thought to be a manifes- tation of short-range antiparallel association of neighboring molecules, resulting from strong dipole-dipole interactions between cyano groups. To test and extend existing models of microscopic ordering in nCB’s, we carry out large-scale atomistic simulation studies of the microscopic structure and dynamics of the Sm- A d phase of 4-octyl-4 ' -cyanobiphenyl 8CB. We compute a variety of thermodynamic, structural, and dynamical properties for this material, and make a detailed comparison of our results with experimental measurements in order to validate our molecular model. Semiquantitative agreement with experiment is found: the smectic layer spacing and mass density are well reproduced, translational diffusion constants are similar to experiment, but the orientational ordering of alkyl chains is overestimated. This simulation provides a detailed picture of molecular conformation, smectic layer structure, and intermolecular correlations in Sm- A d 8CB, and demon- strates that pronounced short-range antiparallel association of molecules arising from dipole-dipole interactions plays a dominant role in determining the molecular-scale structure of 8CB. DOI: 10.1103/PhysRevE.64.051703 PACS numbers: 61.30.-v,64.70.Md I. INTRODUCTION Liquid crystals LC’s, broadly defined as state of matter intermediate between crystalline solid and isotropic liquid are fascinating materials both from a fundamental and an applied point of view 1. Due to their rich phase behavior, resulting from the delicate interplay of fluidity and self orga- nization, LC’s play an important role in biological systems for example, lipid membranesand in technological appli- cations, in particular electro-optic displays utilizing nematic and, more recently, ferroelectric LC’s 2. Despite its key importance for the synthesis of LC materials having optimal properties for specific technological applications, the rela- tionship of macroscopic properties to molecular structure and microscopic organization is poorly understood. This is mainly due to the extreme sensitivity of the LC properties to small changes in the molecular architecture, itself a conse- quence of the intricate interplay of a variety of energetic and entropic effects responsible for the spontaneous formation of mesophases. Due to the rapid growth in available computer power, as well as the development of efficient algorithms, Monte Carlo MCand molecular dynamics MDsimulations of chemi- cally realistic molecular models hold considerable promise for investigating the microphysics of LC’s and for illuminat- ing the relationship between molecular architecture and mac- roscopic behavior. Owing to the technical challenges associ- ated with atomistic simulation of LC’s, however, there have been only a relatively limited number of atomistic simulation studies of thermotropic LC’s 3–26. The quality of these studies varies widely, and whether or not specific studies meet the criterion of ‘‘chemical realism’’ is debatable. An early reported atomistic simulation of a thermotropic LC 3 was of total duration 60 ps, far too short a time to equilibrate even a small sample 26. Subsequent work has focused on the development of improved molecular models and on the simulation of larger systems over greater spans of time. The main fundamental obstacle to the quantitative model- ing of LC phases is the need for highly accurate models of molecular interactions, due to the sensitivity of LC properties to details of molecular structure. The central aims of this paper are to develop and validate the computational infra- structure for achieving realistic modeling of LC materials, and to test methodologies for deriving suitable interaction potentials for LC’s. For this purpose, a well-studied experi- mental system is needed to check the quantitative predictive capabilities of our approach for specific LC compounds. The partial bilayer phase (Sm- A d ) of a of 4-octyl- 4 ' -cyanobiphenyl 8CBas been chosen for this study be- cause it is the prototypal partial bilayer smectic. The layer spacing of a partial bilayer smectic is intermediate between that of a monolayer smectic ( d l , where l is the molecular lengthand a bilayer smectic ( d 2 l ). The layer spacing is d =31.432 Å at T =24°C 27, which is approximatively 1.4 times the fully extended molecular length of 22.1 Å, and d increases with increasing temperature, in contrast to most monolayer SmA materials. In fact, the Sm- A d phase of 8CB is the ‘‘fruit fly’’ of smectic LC science, being undoubtedly the most studied smectic phase. The chemical structure and phase diagram of 8CB is shown in Fig. 1. The most notable feature of the chemical FIG. 1. Chemical structure and phase diagram of 8CB. PHYSICAL REVIEW E, VOLUME 64, 051703 1063-651X/2001/645/05170312/$20.00 ©2001 The American Physical Society 64 051703-1