This section "Chemistry", which appears for the first time in the Highlights, stresses the importance of synchrotron radiation for a number of studies in the chemistry field.

Third generation synchrotron radiation sources particularly open the way to time-dependent experiments at different time scales. For example, it will soon become possible, on ID9, with the help of a newly developed X-ray streak camera, to follow structural changes in proteins or chemical systems triggered by a laser flash.

A number of examples assembled here clearly indicate the range of techniques that can be applied to chemical and biochemical problems. Such techniques include SAXS/WAXS, single crystal diffractometry, Compton scattering, EXAFS and surface studies.



The mechanism of organic-mediated zeolite crystallization

The assembly of materials by nature is often a complicated process with laws and rules depending on length scales. For example, reactions at molecular scales, influenced by process parameters, are often completely different to reactions of colloidal particles at nanometer or micrometer scales. Therefore elucidation of the properties of reacting systems in a broad range of length scales is a prerequisite to understanding their synthesis and to being able to change these materials. Zeolites are typical examples of such materials (Figure 26). The zeolites found in nature meet the properties needed for a large variety of applications but, for new processes (inspired by existing ones), new combinations of characteristics are desired. To be able to assemble materials with the desired properties, one has to understand the rules which nature applies. This project deals with the elucidation of the formation of zeolites, with an underlying goal to coming closer to their rational design.

In order to understand the assembly mechanism of zeolites, information is needed on a nanometer length scale, considerably larger than the ions in solution (order of a few Å). However, most spectroscopic methods do not give information on such length scales. An additional and very serious experimental problem is the vulnerability of the fragile intermediates in the supersaturated solution. Therefore it is preferable to probe the intermediates in situ under synthesis conditions. Furthermore, when studying transformation processes, time-resolved information is necessary. All the above problems can be solved by a combination of ultra-small (USAXS) small- (SAXS) and wide-angle (WAXS) X-ray scattering techniques using high-brilliance synchrotron radiation.

As a typical example the formation of pure-silica zeolite ZSM-5 (MFI) using tetrapropylammonium (TPA) as the template is presented. During the first phase of the reaction, two different populations of precursors are observed: primary units (2.8 nm), consisting of template molecules surrounded by a silica shell [1], and aggregates composed of these primary units (~ 10 nm) (Figure 27). These precursors are consumed during the formation of the crystalline product as shown by SAXS.

The second part of the reaction is characterized by the formation of nuclei. Experiments with varying alkalinity showed that only the 10 nm sized aggregates are converted into nuclei. The 2.8 nm primary units can be considered to be both building blocks for the 10 nm aggregates and the crystal growth units. This crystal growth process is typically on larger length scales (10-6000 nm). By the application of time-resolved in situ USAXS/SAXS experiments performed on ID2 [2, 3] it could be shown that the building blocks of the crystals are not silicate monomers/oligomers, but the much bigger primary units (see Figure 28). By a combination of USAXS (growth crystals), SAXS (consumption precursors) and WAXS (Bragg reflections for crystalline zeolite lattice) this process was studied over the complete range of relevant length scales for the assembly process.

By applying USAXS/SAXS/WAXS with synchrotron radiation, the nanometer-scale precursors during zeolite synthesis have been identified for the first time, and their formation and consumption have been monitored during the complete course of the crystallization process covering extended length scales.


[1] B.J. Schoeman, In: Progress in zeolite and microporous materials Ed.: H. Chon, S.-K. Ihm, Y.S. Uh, Studies in surface science and catalysis, Elsevier Science B.V., 1997, 105, 647-654, and B.J. Schoeman, Zeolites 1997, 18, 97-105.
[2] P.-P.E.A .De Moor, T.P.M. Beelen, B.U. Komanschek, O. Diat, R.A. Van Santen, J. Phys. Chem. B, 101 (1997) 11077-11086.
[3] P.-P.E.A. De Moor, T.P.M Beelen, R.A. Van Santen, Microporous Mater., 9 (1997) 117-130.

P-P.E.A. de Moor (a), T.P.M. Beelen (a), R.A. van Santen (a), and O. Diat (b), to be published.

(a) Schuit Institute of Catalysis, Eindhoven University of Technology, Eindhoven (The Netherlands)
(b) ESRF




Pentahydridorhenium: crystal and molecular structure of ReH5(PPh3)3 determined by synchrotron X-rays

Transition metal complexes are frequently used as catalysts in homogeneous (same phase) systems. The reason for this is that transition metals form stable complexes with a variety of ligands. These ligands can be activated by coordination to the metal. The availability of the multiple valence states of the metals can be used to activate the ligands towards specific nucleophilic or electrophilic attacks, thus making it possible to control the end-product of the catalytic process. In particular complexes of transition metals with metal hydrido (Me-H) bonds are of importance for many catalytic reactions. The formation of such bonds on metal surfaces is without doubt responsible for the catalytic activities of metals such as Fe, Co, Ni, Pd and Pt in hydrogenation, C-H bond cleavage and many other reactions. Notably rhenium atoms have a complex chemistry with Re-H bonds exhibiting complexes from pure hydrido complexes such as ReH9-2 to complexes with three to eight hydrogens. The essential structural information on these complexes comes from diffraction experiments. Due to the large number of electrons around transition metals atoms (75 in a neutral Re atom), the precise location of the hydrogen atoms has not been possible by conventional X-ray diffraction studies. The information has been obtained almost exclusively by neutron diffraction where the relative scattering factors Me/H is close to 1 whereas the X-ray scattering ratio is 752/1. The disadvantage of neutron diffraction arises from the low neutron flux, the need for large samples (> 1 mm3) and the long data collection times. Weeks to months are required with conventional detectors, making the method rather restricted and exclusive. Using the high-flux, high-energy wiggler beamline ID11 (the Materials Science beamline) at the ESRF, much smaller samples (down to a micron size) can now be studied in a short period of time employing modern CCD detectors and short wavelengths to minimize the absorption problems. Figure 29 illustrates the molecular structure of ReH5(PPh3)3 obtained from a six hour experiment at room temperature on the ID11 beamline using a Bruker SMART CCD detector. The approximate crystal volume was 0.001 mm3. A wavelength of 0.4425 Å was used to collect 102327 reflexions of which 38654 were unique and 21963 had intensities above two standard deviations. The five hydrido atoms bonded to the rhenium could easily be located from the difference Fourier maps after refinement of the heavy atom structure. An example of a Fourier map through the plane containing three of the hydrido atoms is given in Figure 30. The height of the hydrido peaks is0.53e/Å3 and the hydrido atoms could easily be refined giving Re-H distances in the range 1.59-1.71 Å, in agreement with Re-H bond distances in the range 1.68-1.70 Å determined from a similar compound (ReH5(PMePh2)3 [1] by neutron diffraction. This study gives clear indications that many of these important metal complexes, even those including heavy metals such as the period 6 elements, Re,Pd,Pt etc, can now be characterized structurally by synchrotron radiation in a tractable time period using tiny samples. This gives new opportunities for the rapid evaluation and interpretation of many novel catalysts.

[1] T. Emge, T.F. Koetzle, J.W. Bruno and K.G. Caulton, Inorg.Chem.(1984),4012-4017.

Å. Kvick (a), G.B.M, Vaughan (a), A. Puig (a), B. Brummersted Iversen (b), F. Krebs Larsen (b), T.F. Koetzle (c) and R.H. Crabtree (d), to be published.

(a) ESRF
(b) Aarhus University, Aarhus (Danemark)
(c) Brookhaven National Laboratory (USA)
(d) Yale University (USA).


Bonding electrons viewed by synchrotron radiation

X-rays are diffracted by the electrons in the crystalline material being irradiated. Single-crystal diffractometry thus gives the opportunity to map out the electron distribution in atoms and molecular assemblies. Careful studies can reveal important information on chemical bonding and other electronic features in the molecules. The first studies date back to the 1950s and there was renewed vigour in the 1970s when attempts were made to use this method to solve chemical problems. Precision in the experimental studies is of particular importance since the valence electrons, which form a major determinant of the chemical properties, constitute only a small part of the scattering electrons. Systematic errors in the experiments such as absorption, extinction or multiple scattering phenomena clouded the early studies and the field progressed rather slowly. It was, however, realized early on that the use of short-wavelength X-rays, only available at the synchrotron radiation facilities, could drastically reduce these systematic errors. The availability of this radiation has now provided the opportunity to raise the level of confidence in much studies. Furthermore the use of modern CCD cameras gives the opportunity to collect highly redundant data to a very high spatial resolution in a short period of time. Figure 31 illustrates a recent experimental charge density map from aluminium oxide (Al2O3) in the plane of the aluminium oxygen bonds. This compound is notorious for large extinction and was chosen as a test case for the use of ultra-short X-rays in charge density studies. The wavelength at the Materials Science beamline ID11 was systematically lowered from about 1 Å to a final wavelength of 0.214 Å and the reduction in extinction was followed. At the shortest wavelength no systematic errors resulting from extinction could be detected and a complete data set of 5223 reflections merged into 432 unique reflections to a resolution of 0.4 Å was collected at a temperature of 120 K. The structure was refined using multipolar refinement to a reliability index of 1.35%. The resulting electron map reveals clearly the bonding electrons; the longer Al-O (1.966 Å) being less populated than the shorter bond (1.849 Å). The non-bonding density around the oxygen atoms is also clearly seen. In conclusion these high-energy data show that we have now reached the necessary experimental accuracy to be able to test Hartree-Fock (HF) or Local Density Approximation (LDA) theory on a periodic simple system and to characterize complicated molecules not yet amenable to theoretical calculations.

H. Graafsma (a), M. Souhassou (b), A. Puig-Molina (a), Å. Kvick (a), and C. Lecomte (b), Acta Cryst., B54 (1998), 196.

(a) ESRF
(b) Université Henri Poincaré, Nancy (France).




X-ray absorption spectroscopy study of zinc coordination in Langmuir-Blodgett layers from phospholipids and myelin basic protein

Zinc is the second most abundant of the nine trace elements required for human life after iron, with specific functions and an essential role in the activity of many enzymes both at a catalytic and a structural level. The aim of this work was to investigate the influence of zinc on the molecular organization within the myelin sheath, the tightly wrapped proteo/lipid multilayer which is essential to facilitate signal transduction along the axon.

Myelin basic protein (MBP) is the predominant protein component of the myelin sheath and it is thought to play an important role in the formation and maintenance of the structure of the myelin membrane. In some diseases such as multiple sclerosis a demyelinating process takes place: the MBP is released; the myelin membrane is broken down and signal conduction is distrupted. It was found, recently, that zinc ions contribute to the integrity and to the compactness of the whole myelin and inhibit the release of MBP from the membrane during demyelinating process. The mechanisms by which zinc exerts these effects are not yet clear.

In order to get insight into the interplay of lipids, MBP and zinc ions in myelin membrane, the molecular environment of zinc ions in multilayers consisting of phospholipids and MBP was investigated using X-ray absorption spectroscopy (XAS). The samples were assembled by Langmuir Blodgett (LB) deposition of the phospholipid monolayers at the air/water interface, where the zinc and the protein were bound to the lipid monolayer from the aqueous subphase. Phospholipid multilayers with and without the protein were prepared. XAS measurements at the Zn k-edge were performed in fluorescence mode on the CRG beamline GILDA (BM8) at the ESRF. Spectra were recorded in air and in vacuum. The analysis of the spectra indicated that the zinc included in the multilayers was surrounded by a first shell of four oxygen atoms at a mean distance of 1.96 Å. This coordination is quite different from that of zinc ions in solution where octahedral symmetry with an average distance of 2.08 Å is present. The coordination in the presence of the protein was very similar to that of the pure phospholipids: no significant changes of the coordination number, the mean distance and the Debye-Waller factor were found. A peak splitting (Figure 32) in the near edge region of the absorption spectra, in the presence of the protein and after the application of the vacuum, indicated a distortion of the geometry of the first coordination shell. From the results, it has been deduced that in all cases the zinc present in the multilayers is bound mainly to the phosphatidic head groups of the lipids and also to water molecules (Figure 33). The main effect of the protein was a distortion of the coordination shell around the zinc. No indication of zinc exclusively binding to the protein, with a significantly altered environment, was found yet.


H. Haas (a), S. Nuzzo (b), S. Pascarelli (c), P. Cavatorta (d), S. Morante (e), P. Riccio (f) and S. Mobilio (g), to be published.

(a) Department of Chemistry, University of São Paulo (Brazil)
(b) Department of Physics, University of Napoli (Italy)
(c) Istituto Nazionale di Fisica della Materia (INFM) c/o GILDA CRG ESRF
(d) Department of Physics, University of Parma (Italy)
(e) Department of Physics, University Tor Vergata, Rome (Italy)
(f) Department of Biology, D.B.A.F.,University of Basilicata, Potenza (Italy)
(g) Department of Physics, University Rome III (Italy)




Chemical shift X-ray standing wavefield study of PF3 decomposition on Ni(111)


The combination of high flux and high spectral resolution at ID32 offers the possibility of monitoring the local adsorption sites on surfaces using normal incidence X-ray standing waves (NIXSW [1]). This can be done in both an element-specific fashion and with chemical state specificity, by measuring the intensity of the "chemically shifted" core level photoemission signals from the different states. To explore the potential of this idea the co-adsorbed PFx species on Ni(111) produced by synchrotron radiation-induced decomposition of PF3 adsorbed on this surface have been studied. Initially measurements were made of the P 1s photoemission spectra as a function of the time of exposure to the monochromatic beam at a nominal photon energy of 3 keV. At room temperature only two distinct states were observed, attributed to PF3 and PF, the latter growing with time as the former decayed. At low temperature (140 K) four distinct chemically-resolved states were seen, initially attributed to PF3, PF2, PF and P. NIXSW measurements were then made at both the (111) and (-111) reflections to triangulate the adsorption sites of each of these species. Figure 34 summarizes some of the data obtained from the low-temperature decomposed surface, showing the (111) XSW absorption profiles obtained from the four distinct peaks in the photoemission energy distribution curve. Clearly these differ significantly, implying different local adsorption sites for the four states. Most of the results were consistent with expectations based on earlier chemical shift photoelectron diffraction data [2] and electron stimulated ion angular distribution measurements [3]. In particular, the PF3 species was found to be bound to discrete Ni atoms, while the PF2 data is consistent with previous evidence for bridge site occupation. The state tentatively attributed to atomic P corresponds to a P atom adsorbed in an 'fcc hollow' site directly above a third layer Ni atom. The fourth state (nominally PF but with a slightly (0.3 eV) different photoelectron binding energy from the equivalent state seen at room temperature) cannot to reconciled with any simple adsorption site on an unreconstructed Ni(111) surface. As may be seen in Figure 34, the XSW profile for this state is essentially inverted relative to that found for a typical substrate absorption profile, indicating a P-Ni layer spacing of approximately one half of the substrate layer spacing (2.04 Å). The (-111) coherent fraction for this species is close to zero, implying that no single high symmetry site is occupied. We suggest that this species is the P atom which does not bond to the surface in a P2Fx species, the bonding P atom being that initially identified as atomic P. A possible species is P-PF, although the number of associated F atoms is unknown. The P-Ni layer spacing can then be reconciled with a tilt of the P-P axis relative to the normal surface of about 50°. Figure 35 shows a schematic hard-sphere model of the surface showing the co-adsorbed species in these geometries.

[1] D.P. Woodruff, Prog. Surf. Sci. 57 (1998) 1.
[2] K-U. Weiss, R. Dippel, K-M. Schindler, P. Gardner, V. Fritzsche, A.M. Bradshaw, D.P. Wooduff, M.C. Asensio and A.R. González-Elipe, Phys.Rev.Lett. 71 (1993) 581.
[3] M.D. Alvey and J.T. Yates, Jr., J.Am.Chem.Soc. 110 (1988) 1782.

G.J. Jackson (a), J. Lüdecke (a), D.P. Woodruff (a), R.G.Jones (b), N. Singh (b), A.S.Y. Chan (b), J.Mc Combie (b), B.C.C. Cowie (c), V. Formoso (d), to be published.

(a) Physics Department, University of Warwick, Coventry (UK)
(b) Chemistry Department, University of Nottingham (UK)
(c) Daresbury Laboratory, Warrington (UK)
(d) ESRF