AI-Based Supervision for a Stirred Extraction Column Assisted with Population Balance-Based Simulation

Solvent extraction as environmental benign separation technique can be modeled in physical detail by population balance of the droplet size distribution. However, much information on the droplet generation and coalescence is necessary for representative results. In this contribution, we present a comparison of AI-evaluated experimental and simulated data on the behavior of a stirred solvent extraction column with an inner diameter of 32 mm. Lab experiments were performed using the standard test system with n-butyl acetate, acetone, and deionized water. A digital camera is placed in front of the middle section as well as the head of the column. Droplet size evaluation is performed using a retrained neural net (Mask R-CNN). The stirred DN32 extraction column is modeled and simulated using a 1D CFD population balance software. The simulation allows for behavior analysis, trends comparison, and validation of the hydrodynamics and mass transfer performances.     

AI in Process Industries – Current Status and Future Prospects

The chemical industry is one of the key industrial sectors in Germany and at the same time one of the largest consumers of energy and raw materials. A successful energy transition and the development of a circular economy can only succeed if they are actively supported and shaped by the chemical industry – through the redesign of existing production processes and the exploration and implementation of new process routes. The challenge is to realize this transformation within a very short time and for many production processes, whereby a much larger number of process routes must be explored. Digital technologies are key to master this transformation towards more sustainability, climate, and environmental protection. The KEEN project aims to explore and leverage artificial intelligence (AI) opportunities in process industry. The newly developed AI methods are tested wherever possible in real working environments and production plants to prove the economic benefit, applicability, and reliability of the methods and technologies.     

Entwicklung und Erprobung eines Wäschersystems zur CO2‐Abreinigung aus Rauchgasen mithilfe alkalischer Reststoffe

An innovative, technical approach for the reduction of CO₂ emissions is presented that utilizes alkaline wastes to capture CO₂ from flue gases in stable mineral form. Comprehensive pilot‐scale experiments were conducted with the developed flue gas scrubbing system at a power plant site. By optimizing the process parameters gas flux, CO₂ partial pressure, circulation flux and suspension liquid‐to‐solid ratio, a CO₂ binding of 40 – 90 g kg^–1 waste could be reached and up to 25 % of the CO₂ could be captured. The new technique is economically advantageous especially when both alkaline waste and CO₂ are produced on site and when the carbonated products can be used as secondary resources.     

Monomers for Adhesive Polymers, 10‐Synthesis,Radical Photopolymerization and Adhesive Properties of Methacrylates Bearing Phosphonic Acid Groups

Syntheses of eight novel methacrylates bearing phosphonic acid groups were synthesized in three to five steps. The interaction of these monomers with hydroxyapatite was investigated using ¹³C‐NMR spectroscopy. Free radical homopolymerizations were carried out in a mixture ethanol/water (2.5/1, v/v) using 2,2′azo(2‐methylpropionamidine) dihydrochloride as initiator. The copolymerization of these monomers with a mixture HEMA/GDMA (5/3, mol/mol) was investigated by photo‐DSC. Dentin shear bond strength measurements showed that 2‐methacryloyloxy‐3‐(1,1,2,2‐tetrafluoroethoxy)propylphosphonic acid 4b , 2,3‐dimethacryloyloxypropylphosphonic acid 18 and 3‐(methacryloyloxy)‐2,2‐(di[(dihydroxyphosphoryl)methyl])propyl methacrylate 23 are promising candidates for dental adhesives.

     

Remark on the modified kiefeer-weiss problem for exponntial families

During his investigations of the modified kiefer-weiss problem Lorden (1980) considered the (auxiliary) problem of minimizing a weighted sum of the error Probabilites and the expected sample size. He derived a compiete solution for one-parameter exponential famlies and "inner" parameter points. Here we treat the remaining case of external points. It turns out that the optimal stopping rules fail to be finite a.s.. An additional example shows that this may happen also for other types of distributions.     

Affinity chromatography of tryptases: design, synthesis and characterization of a novel matrix-bound bivalent inhibitor

beta-Tryptases are mast cell-derived serine proteases that are enzymatically active in the form of an oligomer consisting of four subunits each with trypsin-like activity. The active-site clefts, which are directed toward the central pore of the tetramer, form spatial arrays of four negatively charged S1 binding pockets. Therefore, dibasic inhibitors of appropriate geometry can bind in a bivalent fashion to neighboring subunits. We have recently identified a potent bivalent inhibitor (K(i)=18 nM), based on the bifunctional scaffold cyclo-(-D-Asp-L-Asp-) and the arginine mimetic dl-3-aminomethyl-phenylalanine methyl ester as a ligand for S1 pockets that takes advantage of the this unique tetrameric geometry. To generate an affinity matrix, the bivalent ligand was modified and immobilized on a Sepharose matrix by use of the PEG derivative Jeffamine ED 900 as spacer. This matrix selectively recognizes and binds beta-tryptase from crude protein mixtures and thus is useful as a geometry-driven means of isolating and purifying human mast cell tryptases.     

Application of a New Method for Radiative Heat Transfer to Flat Glass Tempering

Temperature distributions within semi-transparent materials like glass strongly determine their behavior. It is fundamental to take radiation into account to determine exact temperature distributions. The Rosseland Approximation for radiation is usually an appropriate method, but is only valid with an optically thick glass. In the present paper, we propose an improved approximation that is both efficient and sufficiently accurate even for the semi-transparent region. This new radiation model is used to determine the temperature along the flat glass thickness during tempering. The impact of temperature evaluation on residual stresses is discussed.     

Visualisation of the structure transfer between an oriented polymer melt and the semi-crystalline state

Structure evolution of highly oriented polyethylene during cautious melting and crystallization is investigated with both high time resolution and high signal-to-noise ratio by means of small-angle X-ray scattering (SAXS) and wide-angle X-ray scattering (WAXS). The two-dimensional SAXS patterns are transformed to the multidimensional chord distribution function (CDF) in physical space. The results are continuous and smooth movies of the nanostructure, which elucidate the mechanisms of the evolution of semicrystalline structure.We find that in our material crystallization is preceeded by a rather diffuse mesomorphic nanostructure. Based on its variation in relation to other observed features like row nuclei and crystalline lamellae, we propose to associate it to phase separated regions of entangled and disentangled chain segments, respectively. The movies show that the mesophase structure holds the key for the understanding of crystallite orientation and arrangement in the fibre.     

Using machine learning models to explore the solution space of large nonlinear systems underlying flowsheet simulations with constraints

Flowsheet simulations of chemical processes on an industrial scale require the solution of large systems of nonlinear equations,so that solvability becomes a practical issue.Additional constraints from technical,economic,environmental,and safety considerations may further limit the feasible solution space beyond the convergence requirement.A priori,the design variable domains for which a simulation converges and fulfills the imposed constraints are usually unknown and it can become very time-consuming to distinguish feasible from infeasible design variable choices by simply running the simulation for each choice.To support the exploration of the design variable space for such scenarios,an adaptive sampling technique based on machine learning models has recently been proposed.However,that approach only considers the exploration of the convergent domain and ignores additional constraints.In this paper,we present an improvement which particularly takes the fulfillment of constraints into account.We successfully apply the proposed algorithm to a toy example in up to 20 dimensions and to an industrially relevant flowsheet simulation.