The diffraction quality of crystals of macromolecules can be improved by controlled dehydration. A number of methods exist to change the relative humidity that surrounds crystals, but for reproducible results, with complete characterisation of the changes induced, a precise humidity control device coupled with an X-ray source is required. The EMBL Grenoble has developed a humidity control device, the HC1, now available on the ESRF MX beamlines [1,2]. Using this device, many systems have shown significant improvements in their diffraction quality. The first step of a dehydration experiment is to determine the relative humidity (RH) in equilibrium with the mother liquor of the crystal being studied. If the RH to which the crystal is exposed is too high it will dissolve and if too low, changes may be induced in the crystal. Until now, the equilibrium RH has been found experimentally by placing a drop of the mother liquor in a loop and monitoring the size of the drop using specific image processing software. This stage of the experiment can be quite time consuming as an initial starting point is often unknown. In order to simplify this process, we have measured the equilibrium RH for a range of concentrations of the most commonly used precipitants (salts and polymers such as polyethylene glycol – PEG). The data provide a starting point for most dehydration experiments and Raoult’s Law for the equilibrium vapour pressure of water above a solution [3] can be used to understand the observations and make predictions for precipitant concentrations commonly in use.

Fig. 25: Plot showing the measured equilibrium relative humidity for PEG concentrations commonly used in macromolecular crystallogenesis

We found that increasing the molecular weight of the PEG (for a given w/w concentration) increased the RH in equilibrium with the solution and observed a steep increase in RH equilibrium point with decreasing PEG concentration (Figure 25). Measurements were also made of typical buffer solutions (100 mM) and detergents (1% (w/v)) and these were found to have an RH equilibrium point of 100%; therefore, only the main precipitant affects the RH equilibrium point. Using Raoult’s Law, the equilibrium relative humidity can be predicted for many of the precipitants used in macromolecular crystallogenesis. Raoult’s Law has two aspects that are counter intuitive and lead to some surprising observations. The first is that the number of equivalent molecules in solution must be accounted for. This means that for sodium chloride, each ion in solution counts as a molecular equivalent. The second is that it is the number of species, and not the nature of the species, in solution that affects the equilibrium vapour pressure. This means that one molecule of PEG 200 has the same contribution as, for example, a sodium ion. Raoult’s law starts to break down for PEG solutions over a molecular weight of 1000 Da, but this can be corrected using the Flory-Huggins model for the entropy of mixing of polymers as shown in Equation 1.

Eq. 1

In this equation, RH represents relative humidity, x is the mass fraction of solute and n is the molecular weight of the polymer (as in PEG n). The parameter m comes from the Flory-Huggins model for the entropy of mixing of polymers. This is such that the ratio n/m is the number of polymer segments, each of which takes up one space in the disordered lattice. For n>>m the dependence on n diminishes – therefore, the RH equilibria for all PEGs of >1000 Da will be equal. For the precipitants most often used in crystal growth experiments (typically 10-30% PEG (w/w)), the equilibrium point will be around 99.5%.

An online RH calculator is available, based on the equations derived from this work, to allow users to predict RH starting points in advance of experiments [4] and has made these experiments much faster as well as being useful for vapour diffusion crystallisation experiments. Determining the starting point for these experiments is the first step in automation. A new workflow interface [5] for these experiments, through the beamline GUI MXCuBE, coupled to on-line data analysis, is also available and should increase the use of the device and the number of successful cases.


Principal publication and authors

M.J. Wheeler (a), S. Russi (a,b), M.G. Bowler (c) and M.W. Bowler (a,b), Acta Cryst. F68, 111-114 (2012).

(a) ESRF

(b) EMBL, Grenoble (France)

(c) Department of Physics, University of Oxford (UK)


[1] J. Sanchez-Weatherby et al., Acta Cryst. D65, 1237-1246 (2009).

[2] S. Russi et al., J. Struct. Biol. 175, 236-243 (2011).

[3] F-M. Raoult, C. R. Acad. Sci. Paris, 104, 1430-1433 (1887).


[5] S. Brockhauser et al., Acta Cryst. D68, 975-984 (2012).