Soft-condensed matter is a field that comprises ordinary polymers as well as biopolymers but also liquid crystals and suspensions. In this respect, it covers several different domains such as life sciences, chemistry, physics and industrial research. The field will obtain more importance in the future when the central beamline for this type of research, ID2, which is dedicated to small-angle scattering at the highest possible angular resolution, becomes fully dedicated. Also other beamlines such as ID1, which allow anomalous scattering investigations and scattering without disturbances from windows and atmospheric scattering, promise to play an important role in this field.

Structural information on the long-range supramolecular order in these systems is obtained from an analysis of the small-angle scattering patterns. Dynamical studies such as shear-induced ordering of micellar systems are rapidly gaining importance.

At the Microfocus beamline, ID13, instrumentation is being developed to study the diffraction patterns and thereby the structure with high spatial resolution. This opens up the field to the study of thin fibers (spider silk) and bio-membranes, which cannot be crystallized into large samples.


Commensurability of the twist grain boundary smectic C phase (TGBC) on a wedge sample

Introduced by de Gennes in 1972, the analogy between the nematic to smectic A transition in liquid crystals and the normal to superconductor transition in metals was beautifully illustrated in 1989 by the discovery of a helicoïdal smectic A phase in chiral compounds by Goodby et al. The highly dislocated structure of these new phases called Twist Grain Boundary phases (TGB) was proposed in 1988 by Renn and Lubensky [1]: slabs of smectic A material of thickness lb are regularly stacked in a helical fashion along an axis P parallel to the smectic layers. Adjacent slabs are continuously connected via grain boundaries constituted of a grid of parallel equi-spaced screw dislocation lines. The nematic director n (or the smectic layer normal N) is rotated across each grain boundary by a finite angle = 2sin-1 (d/2ld) d/ld, where d is the smectic period and ld the distance between parallel dislocation lines. The existence of two TGB phases was predicted, namely TGBA and TGBC, in which the smectic slabs are, respectively, smectic A (n is parallel to N) and smectic C (n is tilted relative to N).

According to the Renn-Lubensky model, the X-ray structure factor of the TGB phase qualitatively depends on the value of the ratio = /2. If is irrational, the scattering is intense on a Bragg cylinder of axis x parallel to P and radius Q0 = 2/d. The TGB structure is then incommensurate. If on the other hand, = p/q (p, q mutually prime integers) the structure is periodic along x and the TGB structure is commensurate. The fundamental set of reciprocal vectors forms a ring of q (2q if q is odd) equi-spaced Bragg spots in a plane perpendicular to the pitch direction.

An experiment undertaken at the microfocus beamline (ID13) was designed to answer unambiguously the question: is the commensurability of the TGBC phase intrinsic or fortuitous [2]? The main idea was to change continously the balance between surface and volume effects (i.e. alignment constraint vs intrinsic TGB properties) by changing the thickness (D) in a well aligned wedged cell (0 µm D 30 µm ).

Three different kinds of patterns were distinguished: In the thin region of the wedge (thickness of order one to two helical pitches, D < 12 µm), Bragg spots corresponding to individual smectic slabs were observed for the first time. They appeared in groups of 2, 3 and 4 spots with increasing thickness corresponding to regions with 2, 3 and 4 helical pitches respectively (Figure 16a). Spots arising from pitch n and n+1 did not superimpose, which is the signature of an incommensurate TGB structure. In the intermediate region (12 µm D 18 µm), the Bragg spots arising from indivudual smectic slabs were too high in number to distinguish any regularity. Diffraction patterns appeared to be incommensurate (Figure 16b). In the thicker region (18 µm < D < 30 µm), the scattering intensity is merged into a small number of equispaced broad spots. The diffraction patterns were periodic around the ring i.e. commensurate (Figure 16c). This measure of the commensurability increases with thickness which suggests that its origin is intrinsic rather than induced by the confinement.


[1] S.R. Renn and T.C. Lubensky, Phys. Rev. A 38, 2132 (1988).
[2] L. Navailles, P. Barois, and H.T. Nguyen, Phys. Rev. Lett. 71, 545 (1993).

L. Navailles (a), P. Barois (b), M. Petit (b), M. Nobili (b,*), H.T. Nguyen (b), F. Heidelbach (c) and R. Pindak * (d), to be published.

(a) GDPC, UMR 5581, CNRS, Université Montpellier II (France),
(b) Centre de Recherche Paul Pascal, CNRS, France,
(c) ESRF,
(d) Lucent, Murray Hill (USA),
* Present address: INLN, CNRS, Valbonne




Kinetics of the microphase separation transition of asymmetric diblock copolymers under pressure

The blending of two immiscible polymers A and B leads to a phase separation on a macroscopic scale. However, in a diblock copolymer the two blocks of the polymers A and B are bonded together and therefore no macroscopic phase separation is possible. In trying to minimise the amount of unfavourable contacts, a microscopic separation on the dimensions of the size of the molecules occurs. The morphology varies with the volume fraction , of the polymer blocks. Poly (styrene-block-butadiene), P (S-b-B), shows an upper miscibility gap and a transition from the phase separated state to the disordered separation transition, MST, taking place with increasing temperatures. This transition can also be induced by the application of pressure. In general, the existence of local concentration fluctuations is assumed to play an important role for this weak first order transition. For asymmetric systems (one polymer block is larger than the other) mean field model predicts two different possible mechanisms for the observed microphase separation process [1, 2]. Close to the transition, a regime of nucleation and growth is predicted. Spinodal decomposition becomes relevant for larger temperature or pressure jumps.

The pressure dependence of the MST and the time development of the scattering intensity for strongly asymmetrical P (S-b-B) ( 0.15) has been explored using the high-pressure cell on the high brilliance beamline ID2. A shift of the MST to higher temperature at increased pressure was observed. On the other hand the morphology turns out to be hexagonal instead of lamellar for symmetric systems ( 0.5).

The relaxation behavior after a jump to higher pressure from the disordered to the phase-separated state seems to be determined only by the value of the final pressure. The farther this point is from the transition region, the faster are the kinetics. A set of measured diffraction patterns is shown in Figure 17 for different pressures jump. The characteristic time of the relaxation process as well as the dimension of growth of the microstructures can be obtained by adapting an Avrami type law to the data. In Figure 18, three pressure jumps are compared. The characteristic times derived from the Avrami fits are 80 s, 360 s and 810 s, respectively. Since the jump to the highest pressure is the fastest, the effect of the increased glass temperature of the polystyrene block due to pressure seems to be negligible in this system. The jump to p = 450 bar has an unexpectedly long relaxation time and may be interpreted by appealing to nucleation and growth as the leading process. This picture fits perfectly to theoretical predictions. While in lamellar systems one-dimensional growth is predominant, the dimension of growth in the asymmetrical system varies between 2 and 3 as obtained from the Avrami fits. A suitable explanation may be found in the different growth processes due to the different morphology. A hexagonal nucleus may grow in higher dimensions than a lamellar one. Pressure-induced phase transitions from one morphology to another have not been observed, but still remain possible for systems with different volume fractions .

[1] L. Leibler, Macromolecules 13 (1980) 1602.
[2] F.S. Bates & G.H. Fredrickson, Ann. Rev. Phys. Chem. 41 (1990) 525.

W. De Odorico (a,b), H. Ladynski (a), M. Stamm (a) and T. Kuhlmann (a), to be published.

(a) MPI für Polymerforschung, Mainz (Germany).
(b) Institut für. Kernphysik, Universität Frankfurt/Main (Germany)


Soft condensed matter at buried interfaces

Soft condensed matter readily occurs in nature as thin films at both liquid/liquid and liquid/solid interfaces. Indeed the human body is 70% water, and thus nearly all of the other components can be considered as interfacial. Using the technique of X-ray reflectivity on the BM32 beamline, studies have been completed on the structures of soft condensed matter at both the liquid/liquid and solid/liquid interfaces.

Previous attempts on X-ray reflectivity at buried interfaces have been hampered by the large absorption cross-sections of the liquid phases. However, using the high energy X-ray beams (~ 20 keV) available at the ESRF, combined with a highly accurate goniometer, it has been possible to achieve well-resolved specular and diffuse scans at ultra-low grazing angles.

Figure 19a shows a schematic diagram of a phospholipid monolayer adsorbed at an alkane/water interface. Such systems can be considered as models for regions of biological membranes, and they are also of great interest to the detergent industry. Sub-microscopic capillary waves are found to be constantly buffeting the adsorbed lipids and the magnitude of these interfacial fluctuations has been measured using diffuse X-ray scattering. Fitting the diffuse scattering with the distorted Born approximation and Helfrich's Hamiltonian [1] for the fluctuations of the interfacial height spectrum provides good agreement with the data.

Polyelectrolytes constitute another, relatively poorly understood, region of soft condensed matter at interfaces. Biological examples may be found in the nucleic acids (e.g. DNA, RNA) and carrageenans (sea weed extract). Their synthetic counterparts such as polystyrene sulphonates (PSS) and polyacrylic acids find applications in such diverse fields as nappies, colloidal stabilization and water recovery. However the detailed behavior of these charged polymers adsorbed onto surfaces remains fairly mysterious, due to the difficulty in performing experiments on such complex fragile systems. Figure 19b indicates a possible conformation of partially charged PSS adsorbed at the solid hydrophobic silane/ water interface. The association of heavy metal ceasium counterions with the polyelectrolyte provides contrast for the reflexion of X-rays and has allowed the thickness of these counterion clouds to be measured. Specifically, the effects of changing the salt concentration in the bulk aqueous phase on the reflectivity curves obtained from a series of partially charged PSSs has been examined. Increasing the screening of the interchain repulsion by an increase in the concentration of ceasium chloride can be seen in Figure 20 to cause a collapse of the layer on to the surface.

[1] W. Helfrich, Z. Naturforschung 28 c 693 (1973).

[1] C. Fradin (a), D. Luzet (a), A. Braslau (a), M. Alba (a), F. Muller (a), J. Daillant (a), J.M. Petit (b), F. Rieutord (c), Langmuir, to be published.
[2] O. Theodoly (d), T.A. Waigh (d), R. Ober (d), C.E. Williams (d), J. Daillant (a), F. Rieutord (c), O. Konovalov (c), in preparation.

(a) DRECAM/SPEC, CEA-Saclay (France)
(b) ESRF
(c) DRFMC-SI3M, CEA-Grenoble (France)
(d) LPMC, Collège de France, Paris (France)




Determination of the structure of binary liquid and amorphous materials using combined neutron and X-ray diffraction

A knowledge of the atomic arrangements in any material is important for understanding its physical and chemical properties. In a crystalline material an analysis of the positions and intensities of the diffraction peaks, along with chemical information, is usually sufficient to determine the atomic arrangements. However, in the case of liquids and amorphous materials, the atoms are no longer arranged in a periodic fashion and the diffraction pattern is no longer characterized by Bragg peaks, but by a smoothly varying intensity over all angles. Nevertheless local ordered arrangements of the atoms remain and still govern the properties of the material. This order is described in terms of the pair distribution g(r) that gives the mean distances and coordination numbers of the atoms in a material. In an n component material there are n(n+1)/2 partial gAB(r) functions for each possible combination of atom pairs. In a diffraction experiment from a disordered material each of these correlations contributes to a function SAB() which contributes to the scattering at all angles. Hence in a single diffraction experiment it is not possible to determine the individual SAB() and hence the individual gAB(r) that we wish to know. With neutron scattering it is possible, by isotopic substitution (NDIS), to change the scattering power of particular elements and hence change the weighting of the individual SAB() to the scattering pattern. In this way it is possible to obtain the SAB() directly if different isotope combinations are used. This technique has been used for many years [1] but the number of materials studied has been limited owing to the number of suitable isotopes available and their cost.

There are several reasons why X-ray diffraction studies of liquids have been of more limited use: the corrections are more complex, the X-ray scattering power depends on angle, X-rays are strongly absorbed by sample containers and furnaces, the X-ray scattering power of an element does not depend on the isotope. However, new developments at synchrotron radiation sources have improved the outlook for X-ray diffraction methods. The availability of high energy X-rays, high fluxes and high quality optics has meant that experiments on high atomic mass samples can be carried out using transmission geometry with difficult sample environments (furnaces etc.). The high quality optics also allows the use of a narrow bandwidth crystal analyzer in the diffracted beam that rejects the inelastic X-ray scattering (e.g. Compton scattering). This greatly simplifies the experimental corrections and improves the quality of the X-ray measurements. Two experiments have been carried out using BM16 and ID1 at the ESRF. Figure 21 shows the diffraction pattern obtained from liquid TlSe at 450°C using BM16. With this X-ray measurement, in combination with two neutron diffraction measurements using isotopic substitution, it has been possible to improve considerably the precision of the partial structure factors that have been measured for this material using NDIS alone (see Figure 22).

It is also possible to vary the X-ray scattering power of an atom by up to 10-20 % using the anomalous scattering technique [2]. Using the ID1 diffractometer an experiment has been performed on glassy phosphorus selenide using this technique.

Figure 23 shows the result of a differential anomalous scattering measurement on this glass. This shows a combination of the P-Se and Se-Se correlations in the glass. In this case one can unambiguously resolve the P-Se bond distance and coordination in the glass. The results also demonstrate the absence of homo-atomic P-P and Se-Se bonds in the glass. Recent work that combines neutron scattering and X-ray anomalous scattering encourages the belief that this could be used as a new method of determining partial structure factors in binary systems.

[1] F.G. Edwards, J.E. Enderby, R.A. Howe and D.I. Page, J. Phys. C: Solid State Physics 8 (1975) 3483.
[2] P.H. Fuoss, P. Eisenberger, W.K. Warburton and A. Bienenstock, Phys. Rev. Letters 46 (1981) 1537.

A.C. Barnes (a), M.A. Hamilton (a) and P. Buchanan (a), to be published.

(a) H.H. Wills Physics Laboratory, Bristol (UK).




Grazing incidence diffraction studies of two-dimensional protein crystals at the air-water interface


Three-dimensional crystallization of proteins is generally considered as a prerequisite for structure determination using X-ray diffraction. Nevertheless, periodic ordering of proteins in two-dimensions as well as along one-dimensional helices have already been used to determine important structural features using transmission electron microscopy and, more recently, atomic force microscopy. Reported here are the first high resolution grazing incidence diffraction studies of protein 2D-crystals performed in situ at the air-water interface, using the focussed monochromatic beam of the TROIKA I beamline (ID10A).

Hydrosoluble proteins may be adsorbed at the surface of water by specific interactions with a monolayer of insoluble ligand lipids. After an incubation time of a few hours, provided the protein concentration in the buffer solution is sufficient, the free surface of water becomes covered with a two-dimensional powder of protein 2D-crystals [1] (see Figure 24). It is worth pointing out that this technique is applicable to all proteins which are expressed with a histidine-tag (His6-) grafted on the N terminus, for the purpose of purification on nickel-columns. Indeed these are able to bind to new Ni-chelating lipids recently synthezised by the group of C. Mioskovski (DBCM/CEA Saclay, France).

Structure determination from a 2D powder of protein crystals with large unit cell parameters requires several conditions to be fulfilled:

(i) A high angular resolution is necessary to separate the numerous Bragg reflections at a relatively low angle. This has been achieved using a Ge (111) analyser crystal and a helium atmosphere to reduce the diffuse scattering at small angles.

(ii) A vertical position sensitive detector has been used to record the intensity profiles of the Bragg rods which contain the information on the vertical structure factors of the 2D-crystals.

(iii) The data collection procedure has been adapted to the time-scale of irradiation damage of the protein 2D-crystals, which appears to be of typically 2 minutes under the full monochromatic power of 6 x 1012 photons mm-2 s-1, at 13 keV.

Grazing incidence diffraction patterns have been recorded for three different proteins of molecular weights between 36 kDa and 60 kDa: Annexin-V bound to negatively-charged PS-phospholipids, His6-HupR (a RNA transcription regulator protein) bound Ni-chelating lipids and Streptavidin bound to a biotinylated lipid. Figure 25 shows the diffraction pattern obtained with Streptavidin at pH 5.5. The maximum angular ranges of measurable diffraction along Qxy and Qz respectively determine the in-plane and out-of-plane spatial resolutions which can be obtained from such a data collection. For streptavidin these are respectively 9 Å and 14 Å, but they are larger so far for the 2 other proteins investigated, although the use of a crosslinker such as glutaraldehyde appears able to increase the in-plane resolution [2]. This suggests that the resolution of such diffraction studies at the surface of water might be intrinsically limited by thermally induced in-plane elastic fluctuations that are specific to 2D crystals.

This new approach in protein crystallographic studies appears very complementary to electron crystallography from 2D-crystals transferred onto TEM grids. Moreover it opens a new field of in-situ studies of protein conformational changes or of complexation with species injected in the buffer solution.

[1] C. Vénien-Bryan (b), P.F. Lenne (a), C. Zakri (a), A. Renault (a), A Brisson (f), J.F. Legrand (d), B. Berge (a), Biophysics Journal 74, p. 2649, (1998).
[2] P.F. Lenne (a), B. Berge (a), A. Renault (a), C. Vénien-Bryan (b), S. Courty (b), N. Boudet (c), O. Konovalov (d), J.F. Legrand (d), F. Balavoine (e), J. Lal (c), G. Grübel (c), W. Bergsma-Schutter (f) and A. Brisson (f), (to be published).

(a) Laboratoire de Spectrométrie Physique, Université Joseph Fourier, Grenoble (France)
(b) Institut de Biologie Structurale (CEA-CNRS), Grenoble (France)
(c) ESRF
(d) SI3M, DRFMC, CEA Grenoble (France)
(e) CEA Saclay, Gif sur Yvette, (Franc
(f) GBB/Biophysical Chemistry, University of Groningen (The Netherlands)