Yasar Krysiak

Solving the Hydrogen and Lithium Substructure of Poly(triazine imide)/LiCl Using NMR Crystallography

Poly(triazine imide) with incorporated lithium chloride has recently attracted substantial attention due to its photocatalytic activity for water splitting. However, an apparent H/Li disorder prevents the delineation of structure-property relationships, for example, with respect to band-gap tuning. Herein, we show that through a combination of one- and two-dimensional, multinuclear solid-state NMR spectroscopy, chemical modelling, automated electron diffraction tomography, and an analysis based on X-ray pair distribution functions, it is finally possible to resolve the H/Li substructure. In each cavity, one hydrogen atom is bound to a bridging nitrogen atom, while a second one protonates a triazine ring. The two lithium ions within each cavity are positioned between two nitrogen atoms of neighbouring triazine rings. The thereby induced local dipole moments cause slight buckling of the framework and lateral displacements of the Cl- ions at a coherence length below 2 nm. Nevertheless, the average structure conforms to space group P21 21 21 . In this way, we demonstrate that, in particular, the above-mentioned techniques allow for smart interplay in delineating the real structure of PTI/LiCl.

Structural insights into M2O-Al2O3-WO3 (M = Na, K) system by electron diffraction tomography.

The M2O-Al2O3-WO3 (M = alkaline metals) system has attracted the attention of the scientific community because some of its members showed potential applications as single crystalline media for tunable solid-state lasers. These materials behave as promising laser host materials due to their high and continuous transparency in the wide range of the near-IR region. A systematic investigation of these phases is nonetheless hampered because it is impossible to produce large crystals and only in a few cases a pure synthetic product can be achieved. Despite substantial advances in X-ray powder diffraction methods, structure investigation on nanoscale is still challenging, especially when the sample is polycrystalline and the structures are affected by pseudo-symmetry. Electron diffraction has the advantage of collecting data from single nanoscopic crystals, but it is frequently limited by incompleteness and dynamical effects. Automated diffraction tomography (ADT) recently emerged as an alternative approach able to collect more complete three-dimensional electron diffraction data and at the same time to significantly reduce dynamical scattering. ADT data have been shown to be suitable for ab initio structure solution of phases with large cell parameters, and for detecting pseudo-symmetry that was undetected in X-ray powder data. In this work we present the structure investigation of two hitherto undetermined compounds, K5Al(W3O11)2 and NaAl(WO4)2, by a combination of electron diffraction tomography and precession electron diffraction. We also stress how electron diffraction tomography can be used to obtain direct information about symmetry and pseudo-symmetry for nanocrystalline phases, even when available only in polyphasic mixtures.

Ab initio structure determination and quantitative disorder analysis on nanoparticles by electron diffraction tomography

Nanoscaled porous materials such as zeolites have attracted substantial attention in industry due to their catalytic activity, and their performance in sorption and separation processes. In order to understand the properties of such materials, current research focuses increasingly on the determination of structural features beyond the averaged crystal structure. Small particle sizes, various types of disorder and intergrown structures render the description of structures at atomic level by standard crystallographic methods difficult. This paper reports the characterization of a strongly disordered zeolite structure, using a combination of electron exit-wave reconstruction, automated diffraction tomography (ADT), crystal disorder modelling and electron diffraction simulations. Zeolite beta was chosen for a proof-of-principle study of the techniques, because it consists of two different intergrown polymorphs that are built from identical layer types but with different stacking sequences. Imaging of the projected inner Coulomb potential of zeolite beta crystals shows the intergrowth of the polymorphs BEA and BEB. The structures of BEA as well as BEB could be extracted from one single ADT data set using direct methods. A ratio for BEA/BEB = 48:52 was determined by comparison of the reconstructed reciprocal space based on ADT data with simulated electron diffraction data for virtual nanocrystals, built with different ratios of BEA/BEB. In this way, it is demonstrated that this smart interplay of the above-mentioned techniques allows the elaboration of the real structures of functional materials in detail - even if they possess a severely disordered structure.

Investigation of layered and porous nanomaterials by electron diffraction tomography

Yasar Krysiak1, Haishuang Zhao1, Bastian Barton1, Jürgen Senker2, Reinhard B. Neder3, Ute Kolb1 1Inst. Of Inorg. Chemistry And Analyt. Chemistry, Johannes Gutenberg University, Mainz, Germany, 2Inorganic Chemistry III, University of Bayreuth, Bayreuth, Germany, 3Chair for Crystallography and Structural Physics, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany E-mail: krysiak@uni-mainz.de

Optimization of automated electron diffraction tomography for challenging applications

Ten years after the development of automated electron diffraction tomography (ADT) [1, 2] the analysis of the sequentially scanned reconstructed reciprocal space and the use of extracted reflection intensities for direkt crystal structure analysis has gained strong attention in many subjects. Especially in combination with precession electron diffraction allowing for reflection integration (ADT/PED) it was possible to extract fine structural details often explicitly important for physical properties of the material. The variety of nanocrystalline materials solved with ADT on the basis of quasi-kinematical electron diffraction intensity data covers alloys, large cell porous minerals, zeolites, beam-sensitive metal-organic frameworks and small organic molecules. Some originating from single nanocrystals down to 30 nm, strongly agglomerated particles or FIB lamellae. In combination with other approaches like HR-TEM imaging, X-ray diffraction methods analysing reflections (XRPD) or total scattering information (PDF) and neutron diffraction (ND) or spectroscopic measurements like solid state nuclear magnetic resonance (SS-NMR) it was possible to describe the crystal structure solutions gained for twinned, pseudo-symmetric or disordered material even more detailed [3]. ADT data can be in principle collected using standard TEMs but in order to collect electron diffraction data in a quality necessary to achieve high-end crystal structure solutions the data collection should be tailored to the material and the problem to be solved. Here we focus on the description of different approaches and the applicability to the various material classes.

Solving Challenging Crystallographic Problems with Automated Electron Diffraction Tomography (ADT)

Many important materials, ranging from minerals or catalysts to framework compounds and pharmaceuticals are not suitable for growing large crystals prohibiting single crystal X-ray analysis. Yet, introduction of nano crystallinity and special crystallographic features like disorder, defects, pseudo symmetry or stress/strain effects creates new or allows optimizing existing physical properties. With increasing complexity of the structures and special structural features as well as with decreasing size of crystalline domains, X-ray powder diffraction becomes more and more difficult for structural characterization, which is fundamental for understanding material properties. High-resolution transmission electron microscopy (HR-(S)TEM) allows visualizing structural features directly at the atomic scale but requires high electron dose of several thousand e/Ås causing beam damage. In contrast, electron diffraction needs only a fraction of this electron dose. For a complete structure solution, delivering atomic positions in sub Ångstrom accuracy, needs three-dimensional experimental data with high completeness. Data collection from oriented nano crystals limits the amount of measurable reflections significant and thus, delivers mostly heavy atom positions but hardly lighter atoms. Dynamical scattering effects strongly enhanced in oriented zones and may be reduced by electron beam precession technique [1]

Electron and X-ray diffraction – two worlds united

Small crystals structure solution usually done with X-ray powder diffraction (XRPD) provides bulk information and is powerful for insitu investigations. Peak overlap in the one-dimensional data causes problems e.g. for polyphasic or impure samples and large cell parameters thus peak indexing and intensity extraction are the main issues where x-ray powder data may be supported by extra information. Electrons sample smaller volumes but strong coulombic interaction cause multiple scattering effects changing intensities often so strong that a structure solution is becoming impossible. Nevertheless, oriented electron diffraction patterns may provide sufficient information to support indexing or the assignment of impurity peaks in the case of low quality x-ray powder pattern. Reciprocal space tomography [1] uses a series of non-oriented diffraction patterns for which dynamical effects are significantly reduced and an enhanced amount of independent reflections sampled allows “ab-initio” crystal structure solution using established Xray structure solution packages. Although structure refinement based on kinematical intensities is stable, achievable R values of 1030% are high and final refinement may be performed based on X-ray powder data. Scanning transmission electron microscopy (STEM) for crystal tracking and nano electron diffraction (NED) is suitable for beam sensitive material, agglomerated particles, twins or intergrown phases on crystals down to 30nm size [2, 3]. Interesting properties of nanocrystalline materials are driven mainly by twinning, defects, disorder in one or two dimensions down to the amorphous state. Here low data completeness or uncertain intensity determination causes problems in structure solution. Here a mean structure may be determinable serving as a basis for disorder description and being used as a starting model being refined onto X-ray powder data maybe supported by a combination of the diffraction methods or by adding extra information.