J. Appl. Cryst. (2000). 33, 569±573 # 2000 International Union of Crystallography  Printed in Great Britain ± all rights reserved 569 conference papers Measurement of sugar depletion from uncharged lamellar phases by SANS contrast variation Bruno Demé a and Thomas Zemb b a Institut Laue-Langevin, B.P. 156, F-38042 Grenoble Cedex 9, France, and b Service de Chimie Moléculaire, Centre d’Études de Saclay, F-91191 Gif sur Yvette Cedex, France E-mail: deme@ill.fr, zemb@nanga.saclay.cea.fr We applied small-angle neutron scattering (SANS) contrast variation to samples where a microphase separation occurred. The samples contain multilamellar vesicles in equilibrium with excess “solvent” that produce a very common powder pattern in SANS: a Debye-Sherrer ring produced by the regular bilayer packing superposed to a sharply decaying Porod behaviour. These two features of the SANS pattern have distinct contrast match points (CMP). We exploit here the small angle signal to determine the partition of sugars between two coexisting microphases. The net result is an exclusion of small sugar molecules from the liquid crystalline domains of the sample. We discuss this exclusion in relation with the observed maximum swelling, headgroup hydration and bilayer softening induced by the presence of the sugar molecules. 1. Introduction Microphase separation is a very common situation in colloidal solutions. For example, the phase equilibria of anisometric colloids (Langmuir, 1938) results in the coexistence of ordered and diluted domains in colloidal solutions. The formation of “tactoids” (Kruyt, 1952), is described as the most common situation where small amounts of a birefringent phase are in equilibrium with a sol. This type of samples have often the appearance of clear gels which are very difficult to separate in two “pure” phases. The equilibrium phase diagram can only be determined by chemical analysis of well separated phases. In some cases, for example when large polyelectrolytes coexist with concentrated clay dispersions, the complete separation of the two phases allows the determination of the osmotic pressure, i.e. the thermodynamic underlying the coexistence of the two distinct microphases (Morvan et al., 1994). A very important case is the analysis of lamellar phases used as model membranes, such as the classical zwitterionic DMPC/water dispersions. The binary phase diagram of this system is well known (Janiak, Small & Shipley, 1976; Smith et al., 1988) : at room temperature, the DMPC lamellar liquid crystal is in equilibrium with excess water at the so-called maximum swelling (D* = 60.4 Å). At that periodicity, the water content is 41 % corresponding to 24.9 Å thick water layers separating the membranes in L α domains. An ubiquitous founding paper of biological membrane physics (LeNeveu et al., 1977) has proposed a quantitative explanation of the observed maximum swelling, using an original experimental procedure. The ansatz is that at the maximum swelling, the difference between the osmotic pressures of the coexisting phases is zero. Now, if one of the phases is pure or almost pure water (lipid at the CMC), then the total osmotic pressure in the concentrated phase is zero as well, resulting from the balance between dispersion forces and short range “hydration” repulsions, both independently measurable (Lis et al., 1982). The critical test of this basic hypothesis is the measurement of the maximum swelling versus sugar content using different sugar concentrations. It has been inferred (LeNeveu et al., 1977) that the main effect of sugar addition was to shield the dispersion force by matching the permittivity, thus producing a minimum in the attractive dispersion force (Parsegian & Weiss, 1981). The associated maximum swelling observed when sugar is added to zwitterionic lecithin/water dispersion has been explicitly calculated by Parsegian and has been shown to be consistent with the experimental observation. However, the identification of the dominant interactions in biphasic samples involving a lamellar phase in coexistence with a “solvent” as a reservoir requires as a first step the knowledge of the exact content of each of the two microphases in equilibrium. When macroscopic separation is possible, the easiest way is a direct dosage of the third component together with a measure of the osmotic pressure. For example, cationic bilayers in the presence of excess salt induce a strong Donnan exclusion mechanism which explains quantitatively the observed stability limit versus dilution of the DDAB/water/salt system (Dubois et al., 1992). In the case of added polymers or complex sugars (Demé, 1995), macroscopic phase separation using centrifugation is not reliable since ultracentrifugation may separate self-assembled aggregates instead of the microphases. The aim of the experimental method described here is to demonstrate that contrast variation can be used to dose the content of the two coexisting phases without requiring the delicate step of macroscopic separation. The ternary system used is DMPC in the presence of an excess “solvent”. The solvent is a concentrated sugar solution, similar to those used a long time ago to increase contrast in small-angle x-ray scattering (SAXS) experiments. The ternary sample is a microphase separated biphasic sample, with a lamellar phase at “maximum swelling” in equilibrium with excess solvent. This lamellar phase appears in the form of multilayer vesicles designed as onions or MLVs (Multilamellar Vesicles) producing Maltese crosses under polarising microscope. These are formed on a mesoscopic scale, i.e. they are too small for easy separation from the pure coexisting solvent, but however large enough to produce sharp Bragg peaks whose width is not limited by the number of layers (finite crystal case), but by the interlayer fluctuation (Dubois & Zemb, 1991). The aim of the SANS contrast variation experiment is to determine directly the sugar content of the “excess solvent” and the sugar content of the water forming the multilayer vesicles suspended in the excess solvent. A priori these two concentrations cannot be considered safely as equal, and the ratio of the two concentrations cannot be estimated from any current solvation predictive theory as in the case of salt in charged phases (Dubois et al., 1992). 2. Materials and methods We used two sugars, glucose and fructose as model host molecules to investigate the swelling behaviour of the L α domains, and partially deuterated sugars: 2D-fructose, 2D- glucose and 7D-glucose for the contrast variation experiment.