Prof. Dr. Tobias Kraus

Hybrid dielectrics were prepared from dispersions of nanoparticles with gold cores (diameters from 2.9 nm to 8.2 nm) and covalently bound thiol-terminated polystyrene shells (5000 Da and 11 000 Da) in toluene. Their microstructure was investigated with small angle X-ray scattering and transmission electron microscopy. The particles arranged in nanodielectric layers with either face-centered cubic or random packing, depending on the ligand length and core diameter. Thin film capacitors were prepared by spin-coating inks on silicon substrates, contacted with sputtered aluminum electrodes, and characterized with impedance spectroscopy between 1 Hz and 1 MHz. The dielectric constants were dominated by polarization at the gold–polystyrene interfaces that we could precisely tune via the core diameter. There was no difference in the dielectric constant between random and supercrystalline particle packings, but the dielectric losses depended on the layer structure. A model that combines Maxwell–Wagner–Sillars theory and percolation theory described the relationship of the specific interfacial area and the dielectric constant quantitatively. The electric breakdown of the nanodielectric layers sensitively depended on particle packing. A highest breakdown field strength of 158.7 MV m−1 was found for the sample with 8.2 nm cores and short ligands that had a face-centered cubic structure. Breakdown apparently is initiated at the microscopic maxima of the electric field that depends on particle packing. The relevance of the results for industrially produced devices was demonstrated on inkjet printed thin film capacitors with an area of 0.79 mm2 on aluminum coated PET foils that retained their capacity of 1.24 ± 0.01 nF@10 kHz during 3000 bending cycles.

Self-assembled nanoparticles (NPs) in superlattices are in close contact. Their dense packing and the proximity of aligned facets can facilitate coalescence and enable crystal lattices to fuse at temperatures below the bulk melting point. This phenomenon could be applied in nanodevice manufacture. We study NP fusion in superlattices in liquid and dry environments at controlled temperatures using electron microscopy at minimized electron doses. We found that coalescence of self-assembled gold NPs (AuNPs, diameter 8.1 ± 0.4 nm) depended on their arrangement. A double layer of AuNPs in a hexagonally close packed superlattice started to coalesce within 2 min at a temperature of 70 °C in cyclohexane but remained stable for 30 min at 98 °C when it was dry. AuNPs assembled in hexagonal monolayers coalesced after 5 min at 75 °C in cyclohexane. The mobility of the ligand shells and the interfacial gold atoms and the sparse ligand coverage on (111) facets likely facilitated this AuNP coalescence at low temperatures.

An ontology for the structured storage, retrieval, and analysis of data on lithium-ion battery materials and electrode-to-cell production is presented. It provides a logical structure that is mapped onto a digital architecture and used to visualize, correlate, and make predictions in battery production, research, and development. Materials and processes are specified using a predetermined terminology; a chain of unit processes (steps) connects raw materials and products (items) of battery cell production. The ontology enables the attachment of analytical methods (characterization methods) to items. Workshops and interviews with experts in battery materials and production processes are conducted to ensure that the structure is conformable both for industrial-scale and laboratory-scale data generation and implementation. Raw materials and intermediate products are identified and defined for all steps to the final battery cell. Steps and items are defined based on current standard materials and process chains using terms that are in common use. Alternative structures and the connection of the ontology to other existing ontologies are discussed. The contribution provides a pragmatic, accessible way to unify the storage of materials-oriented lithium-ion battery production data. It aids the linkage of such data with domain knowledge and the automation of data analysis in production and research.

In the field of carbanionic polymerization bifunctional initiators permit the synthesis of complex triblock copolymer structures. Using 1,3-bis(1-phenylethenyl)benzene (PEB), isoprene was polymerized in cyclohexane, yielding a high content of 1,4-PI units of 93%. Subsequently, 3 hydroxyl groups were introduced simultaneously both in α- and ω-position by means of end-functionalization of the living anionic di-lithiated polyisoprene (PI) chains with 1,2-isopropylidene glyceryl glycidyl ether (IGG) and subsequent acidic deprotection. The resulting hexa-hydroxy functional PI-macroinitiators were then used to initiate L-lactide (LLA) in a DBU-catalysed polymerisation, ultimately yielding super-H-shaped (PLLA)3-b-PI-b-(PLLA)3 triblock structures with molecular weights of 23–49 kg mol−1. Narrow molecular weight distributions with dispersity in the range of 1.19–1.35 were obtained, and thermal characterisation revealed two distinct glass transition temperatures (Tg), indicating phase separation. The PI-domains feature a low Tg between −55 °C and −59 °C, whereas the PLLA-domains exhibit a higher Tg of 41 °C to 49 °C. Further, the block copolymers were analyzed by TEM and SAXS, confirming clearly phase-separated cylindrical and lamellar morphologies. The reported bifunctional approach combining carbanionic polymerization with the ROP of lactones represents an efficient and general synthesis pathway for a large variety of complex polymer architectures.

Within the present work, a thermo-responsive ultrafiltration membrane is manufactured based on a polystyrene-block-poly(diethyl acrylamide) block copolymer (BCP). The poly(diethyl acrylamide) block segment features a lower critical solution temperature (LCST) in water, similar to the well-known poly(N-isopropylacrylamide), but having increased biocompatibility and without exhibiting a hysteresis of the thermally induced switching behavior. The BCP is synthesized via sequential “living” anionic polymerization protocols and analyzed by 1H-NMR spectroscopy, size exclusion chromatography, and differential scanning calorimetry. The resulting morphology in the bulk state is investigated by transmission electron microscopy (TEM) and small-angle X-ray scattering (SAXS) revealing the intended hexagonal cylindrical morphology. The BCPs form micelles in a binary mixture of tetrahydrofuran and dimethylformamide, where BCP composition and solvent affinities are discussed in light of the expected structure of these micelles and the resulting BCP membrane formation. The membranes are manufactured using the non-solvent induced phase separation (NIPS) process and are characterized via scanning electron microscopy (SEM) and water permeation measurements. The latter are carried out at room temperature and at 50 °C revealing up to a 23-fold increase of the permeance, when crossing the LCST of the poly(diethyl acrylamide) block segment in water.

Solvent engineering is a powerful and versatile method to tune colloidal stability. Here, we link the molecular structure of apolar ligand shells on gold nanoparticles with their colloidal stability in solvent mixtures. The agglomeration temperature of the particles was measured with small-angle X-ray scattering. It depended on solvent composition and changed linearly for hexane–hexadecane mixtures, but nonlinearly for cyclohexane–hexadecane and hexanol–hexadecane mixtures. Molecular dynamics (MD) simulations indicate that agglomeration is dominated by temperature-dependent ligand order in the alkane mixtures and that the temperature at which the ligand shell orders depends on the solvent composition near the ligands, which can differ substantially from the bulk composition. Small-angle neutron scattering confirmed that, at intermediate solvent compositions above the agglomeration temperature, the fraction of cyclohexane near the ligands was larger than in the bulk. The enrichment of cyclohexane near the ligands stabilized their disordered state, which, consequently, led to the experimentally observed nonlinear trend of the agglomeration temperature. In contrast, hexanol was depleted from the ligand shell at all temperatures. This again stabilized the disordered state. Furthermore, we found that agglomeration at high hexanol fractions was driven by a solvophobic effect that exceeded the influence of ligand order. The results show that strong nonlinearities in the colloidal stability of nanoparticle dispersions in solvent mixtures are directly linked to the molecular details of ligand–solvent and solvent–solvent interactions, which can be used to precisely tune stability.

Current environmental challenges and the shrinking fossil-fuel feedstock are important criteria for the next generation of polymer materials. In this context, we present a fully bio-based material, which shows promise as a thermoplastic elastomer (TPE). Due to the use of β-farnesene and L-lactide as monomers, bio-based feedstocks, namely sugar cane and corn, can be used. A bifunctional initiator for the carbanionic polymerization was employed, to permit an efficient synthesis of ABA-type block structures. In addition, the “green” solvent MTBE (methyl tert-butyl ether) was used for the anionic polymerisation, enabling excellent solubility of the bifunctional anionic initiator. This afforded low dispersity (Đ=1.07 to 1.10) and telechelic polyfarnesene macroinitiators. These were employed for lactide polymerization to obtain H-shaped triblock copolymers. TEM and SAXS revealed clearly phase-separated morphologies, and tensile tests demonstrated elastic mechanical properties. The materials featured two glass transition temperatures, at – 66 °C and 51 °C as well as gyroid or cylindrical morphologies, resulting in soft elastic materials at room temperature.

Self-assembly of nanoscale structures at liquid-solid interfaces occurs in a broad range of industrial processes and is found in various phenomena in nature. Conventional theory assumes spherical particles and homogeneous surfaces, but that model is oversimplified, and nanoscale in-situ observations are needed for a more complete understanding. We used liquid phase scanning transmission electron microscopy (LP-STEM) to examine the interactions that direct the self-assembly of superlattices formed by gold nanoparticles (AuNPs) in non-polar liquids. Varying the molecular coating of the substrate modulated short-ranged attraction and led to switching between a range of different geometric structures including hexagonal close-packed (hcp), simple hexagonal (sh), dodecahedral quasi-crystal (dqc), and body-centered cubic (bcc) lattices, as well as random distributions. Langevin dynamics simulations explain the experimental results in terms of the interplay between nanoparticle faceting, ligand shell structure, and substrate-NP-interactions.

Abstract When searching for new antibiotics against Gram-negative bacterial infections, a better understanding of the permeability across the cell envelope and tools to discriminate high from low bacterial bioavailability compounds are urgently needed. Inspired by the phospholipid vesicle-based permeation assay (PVPA), which is designed to predict non-facilitated permeation across phospholipid membranes, outer membrane vesicles (OMVs) of Escherichia coli either enriched or deficient of porins are employed to coat filter supports for predicting drug uptake across the complex cell envelope. OMVs and the obtained in vitro model are structurally and functionally characterized using cryo-TEM, SEM, CLSM, SAXS, and light scattering techniques. In vitro permeability, obtained from the membrane model for a set of nine antibiotics, correlates with reported in bacterio accumulation data and allows to discriminate high from low accumulating antibiotics. In contrast, the correlation of the same data set generated by liposome-based comparator membranes is poor. This better correlation of the OMV-derived membranes points to the importance of hydrophilic membrane components, such as lipopolysaccharides and porins, since those features are lacking in liposomal comparator membranes. This approach can offer in the future a high throughput screening tool with high predictive capacity or can help to identify compound- and bacteria-specific passive uptake pathways.

We directed the self-assembly of nanoscale colloids via direct nanoimprint lithography to create flexible transparent electrodes (FTEs) with metal line widths below 3 μm in a roll-to-roll-compatible process. Gold nanowires and nanospheres with oleylamine shells were imprinted with soft silicone stamps, arranged into grids, and converted into metal lines in a plasma process. We studied the hierarchical structure and opto-electronic performance of the resulting grids as a function of particle geometry and concentration. The performance in terms of optical transmittance was dominated by the line width. Analysis of cross-sections indicated that plasma sintering only partially removed the insulating ligands and formed lines with thin conductive shells and a non-conductive core. We provide evidence that the self-assembly of high-aspect nanowires can compensate for defects of the stamp and substrate irregularities during imprinting, while spheres cannot. The wire-based electrodes thus outperformed the sphere-based electrodes at ratios of optical transmittance to sheet resistance of up to ≈ 0.9 %·Ωsq-1, while spheres only reached ≈ 0.55 %·Ωsq-1.