A completely new crystal-growth device has been developed that permits charting a course across the phase diagram to produce crystalline samples optimized for diffraction experiments. The utility of the device is demonstrated for the production of crystals for the traditional X-ray diffraction data-collection experiment, of microcrystals optimal for data-collection experiments at a modern microbeam insertion-device synchrotron beamline and of nanocrystals required for data collection on an X-ray laser beamline.

L'invention concerne un procede de cristallisation d'une substance dissoute dans un solvant, comprenant les etapes suivantes : introduction d'un volume de solvant contenant la substance, dans une chambre a temperature, humidite de l'air et composition gazeuse pre-etablies; addition, au volume de solvant contenant la substance, d'un volume predetermine d'un precipitant; evaporation du solvant tout en observant, par diffusion lumineuse dynamique, les variations de structure dans le volume de solvant renfermant la substance et le precipitant; observation des variations de poids et determination des molarites; coordination des emplacements dans le diagramme de phases, sur la base de mesure DLS et des resultats de la determination de molarite, formation d'un nombre predetermine de centres de cristallisation, par addition de solvant ou par addition de precipitant; passage du volume de solvant contenant le precipitant en un etat metastable par addition de solvant et/ou de solution de proteine, ou par reduction de la concentration de la substance dissoute par formation de noyaux de nucleation; maintien de l'etat metastable par addition d'une quantite predeterminee de substance au volume de solvant contenant la substance et le precipitant, ou evaporation du solvant jusqu'a formation d'au moins un cristal d'une grosseur predeterminee.

A novel principle developed by the authors to measure simultaneously the equilibrium solvent vapour pressure (solvent activity) and the solvent heat of evaporation of aqueous macromolecular (protein) solutions is applied to aqueous insulin solutions. The measuring principle is based on the evaporation of samples in the μl-range. It exploits the effect of the continuously increasing solute concentration provided that, in practice, only the solvent evaporates. On the course of the measurement, an enhancement of the solute concentrations by a factor two up to three, compared with the starting ones, can be reached. The protein concentration-dependent parts μA(s,P) and hA(s,P) of the solvent molecular chemical potential and the solvent molecular enthalpy, respectively, are determined for dilute protein concentrations (the latter for the first time). The data evaluation is based on the predictions of the classical, molecular theory of solutions (as underlying model), and focused on parameters describing excess properties beyond the ideal solution behaviour, in particular, on second-order and third-order contributions in solute concentrations attributed to the protein–protein and protein–precipitant interaction, respectively. Measurements on Insulin Glargine (Hoe 901) solution (starting concentration 0.3 mg/ml) together with 300 mM ammonium sulfate (AS), 50 mM Tris buffer, and 0.0, 0.3 and 1.0% phenol (starting concentrations) as composite precipitant are presented, where we find hA(s,P)/kBT≈10−1 and μA(s,P)/kBT≈10−2 (kB: Boltzmann constant, T: temperature). For comparison, scaling of published data to protein concentrations and precipitant molalities applied in our measurements yields μA(s,P)/kBT≈−10−3 for aqueous solutions of lysozyme/NaCl, and |μA(s,P)/kBT|≤10−4 for various aqueous protein solutions with a second organic solute component, respectively. The order of μA(s,P)/kBT for insulin/AS and lysozyme/NaCl is not explained by the protein–protein interaction only, but by the strength of the protein–precipitant (salt) interaction. Beyond the underlying model, the latter interaction can involve the precipitant-induced change of internal properties of macromolecular units formed by the protein molecules and the surrounding solvation shell, namely the successive incorporation of precipitant and solvent molecules for increasing precipitant concentrations. To be consistent with established models of the phase coexistence protein solution-protein crystal we argue that for both the phases, the protein molecules form macromolecular units with a screened protein net charge, where both the direct protein–precipitant interaction as well as the protein–protein interaction are significantly affected by the protein solvation.

A novel principle to measure simultaneously the equilibrium solvent vapour pressure (solvent activity) and the solvent heat of evaporation of aqueous macromolecular (protein) solutions is presented. These measurements are based on the simultaneous recording and evaluation of both the course of the sample mass and sample temperature during the sample evaporation. A setup is described where the mass and the temperature of the sample are continuously measured by weighing and contactless pyrometry, respectively. The sample in the μl-range is stored in a container placed within a box where temperature and relative humidity is stabilized up to 10−2K and 10−1%, respectively. The incoming data reflect the effect of the continuously increasing solute concentration (changing at a ratio of about 2:1 up to 3:1) provided that, in practice, the solvent only evaporates. The protein concentration-dependent part μA(s,P) and hA(s,P) of the solvent chemical potential and the solvent molecular enthalpy, respectively, can be simultaneously determined (the latter for the first time). For appropriate protein concentrations, the second-order and third-order contributions in solute concentrations can be determined by a regression analysis. The data evaluation utilizes reference measurements on the pure solvent and upstream measurements on the solution of the precipitant only. Supported by FE simulations, a prodecure to minimize statistical errors and to eliminate systematic errors is derived and applied. For conditions of an ordinary laboratory, the equipment described enables a measurement of hA(s,P)/kBT and μA(s,P)/kBT with an absolute accuracy ≤10−1 and ≤10−3, respectively (kB: Boltzmann constant; T: temperature). We present measurements on aqueous solutions of Insulin Glargine (Hoe901) with ammonium sulfate as main precipitant component to demonstrate basic steps of the data evaluation procedure and the use of our measuring method.