A new multivariate population balance modeling for the homogenous nucleation from bismuth vapor is tested against experimental results. However, the small differences of some key parameters could lead to the ill-posed problem, such as the surface tension. In this study, the parameter fluctuations set according to the response surface method are employed in the population balance simulation to identify their importance. Subsequently, the quadratic polynomials are established to replace the simulation and the fluctuations are evaluated with the characteristic parameters of particle evolution. It is found that the surface tension tends to be the most significant factor determining the particle evolution, which is also influenced by the coefficient in condensation rates and fractal dimension in the coagulation. To get more accurate modeling and key parameters, the comprehensive sum of mean square error is calculated based on all the fluctuations and the appropriate value for the surface tension is 0.466 N/m.

Aerosol evolution is a sensitive process which strongly depends on the spatiotemporal environment, especially for the nucleation serving as the source term. Challenges and interests are rising when insights are needed to evaluate the dynamic events within the process. In our recent study, a multivariate population balance Monte Carlo (PBMC) simulation coupled with computational fluid dynamics (CFD) was verified with experimental results measured in the homogenous nucleation from bismuth vapor. Therefore, a further analyze relating the main mechanisms is established, which covers the nucleation, condensation, coagulation and wall deposition. The nucleation is found the most significant and great discrepancies in aerosol parameters are contributed by different theories. Competition of vapor consumption between the nucleation and condensation determines the concentration of primary particles and following aggregate growth. The coagulation dominates the evolution after the nucleation, while the radial migration leading to wall deposition helps in averaging the spatiotemporal aerosol concentration.

Particle formation from an iron-based precursor dissolved in ethanol and 2-ethylhexanoic acid was studied via population balance simulations of the SpraySyn burner. Monte-Carlo population balance modeling was used to estimate droplet evaporation and breakup, while particle nucleation and growth were calculated using a pivot method. To investigate common particle formation pathways a precursor chemistry model was formulated and discussed for the cases of instantaneous and absent thermal decomposition in the liquid phase. Following this, the droplet breakup time was calculated to determine when precursor and particle transfer into the gas phase occurs. The simulation results show good agreement with experimental data from literature for different precursor concentrations. However, in the cases where thermal decomposition is absent in the liquid phase, the model underestimates particle size and polydispersity. The primary conclusion is that nanoparticles smaller than 10 nm most likely formed in the liquid phase. Moreover, particle formation in the liquid phase increases polydispersity through the formation of an accumulation mode near the droplet surface.

The fractionation of airborne particles based on multiple characteristics is becoming increasingly significant in various industrial and research sectors, including mining and recycling. Recent developments aim to characterize and fractionate particles based on multiple properties simultaneously. This study investigates the fractionation of a technical aerosol composed of a mixture of micron-sized copper and silicon particles by size and material composition using a classifying aerodynamic lens (CAL) setup. Particle size distribution and material composition are analyzed using scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX) for samples collected from the feed stream (upstream of CAL) and product stream (downstream of CAL) at varying operational pressures. The experimental findings generally agree with the predictions of an analytical fractionation model but also point to the importance of particle shape as a third fractionation property. Moreover, the results suggest that material-based fractionation is efficient at low operational pressures, even when the aerodynamic properties of the particle species are similar. This finding could have significant implications for industries where precise particle fractionation is crucial.

The software’Particle Droplet Population-Balance Simulation’ (Particle-Droplet-PBS) simulates the time of particle-shell formation and particle growth in burning spray droplets. Application examples are provided to demonstrate how the code can be used to incorporate a variety of particles, precursor chemicals and solvents. The software solves the population balance equations for particle coagulation and nucleation, while applying an adaptive grid method requiring only two cells per droplet. Thus, the computation time is drastically reduced. This enables users without access to high-performance work stations to run droplet simulations in a timely manner, making it a valuable tool for research and process engineering.

​​​​​​​The application of ultraviolet (UV)-light-based air disinfection methods holds promise but also presents several challenges. Among these, the quantitative determination of the required UV radiation dose for aerosols is particularly significant. This study explores the possibility of determining the UV dose experienced by aerosols without the use of virus-containing aerosols, circumventing associated laboratory safety issues. To achieve this, we developed a model system comprised of UV-sensitive dyes dissolved in di-ethyl-hexyl-sebacate (DEHS), which facilitates the generation of non-evaporating and UV-degradable aerosols. For the selection of UV-sensitive dyes, 20 dyes were tested, and 2 of them were selected as being the most suitable, according to several selection criteria. Dye-laden aerosol droplets were generated using a commercial aerosol generator and subsequently exposed to UV-C radiation in a laboratory-built UV irradiation chamber. We designed a low-pressure impactor to collect the aerosols pre- and post-UV exposure. Dye degradation, as a result of UV light exposure, was then analyzed by assessing the concentration changes in the collected dye solutions using a UV-visible spectrophotometer. Our findings revealed that a UV dose of 245 mW s cm−2 resulted in a 10 % degradation, while a lower dose of 21.6 mW s cm−2 produced a 5 % degradation. In conclusion, our study demonstrates the feasibility of using aerosol droplets containing UV-sensitive dyes to determine the UV radiation dose experienced by an aerosol.

Formation of hetero-contacts between particles of different materials in nanoparticle hetero-aggregates can lead to new functional properties. Improvement of the functional behavior requires a detailed characterization of mixing between the two types of particles, in order to correlate different mixing with the performance of the material. Scanning transmission electron microscopy (STEM) is an option for this task. To obtain statistically relevant results, STEM-images of many hetero-aggregates have to be acquired and evaluated. This can be time-consuming if it is done manually. In the present work, the applicability of convolutional neural networks for the automated analysis of STEM-images acquired from TiO2-WO3 nanoparticle hetero-aggregates is investigated. Hetero-aggregates are obtained in a double flame spray pyrolysis (DFSP) setup, in which a variation of setup parameters is expected to affect the mixing of TiO2 and WO3. Mixing is investigated by a measurement of cluster sizes (the number of connected particles of the same material within an aggregate) and coordination numbers (the number of particle contacts with particles of the same or the different material). Results show that the distribution of measured values is wide for both quantities, rendering it challenging to correlate mixing with parameters varied in the DFSP setup.

Thermally-induced breakup of metal-precursor-laden droplets in spray-flame synthesis occurs via a rapid and disruptive disintegration, i.e., "puffing" and "micro-explosion". To assess the temporal evolution and statistics of droplet disruption, LED-illuminated droplet shadowgraphs were imaged with a microscope onto a high-speed camera and morphological image analysis was applied. The atomized liquid was a mixture of 35 vol.-% ethanol and 65 vol.-% 2-ethylhexanoic acid mixed with iron(III) nitrate nonahydrate (INN) as a precursor. Droplet evaporation and disruption were also simulated with a population balance model. The model finds solid precipitates forming in the droplets because of the decomposition of the precursor intermediate iron(III) 2ethylhexanoate. The precipitates form a particle shell, which favors the superheating of the droplets' interior, and they facilitate heterogeneous bubble nucleation. Imaging experiments and modelling find that per 10 & mu;s lifetime of a droplet, the probability for disruption increases from 5 to 13% and 5 to 19%, respectively, when increasing the INN concentration from 0.05 to 0.5 mole/l. The probability of disruption suggests that throughout their lifetime in the spray flame, nearly all droplets will undergo disruption and many of them multiple times. In the experiment, droplets before disruption are 15% smaller than regular, non-disrupting droplets. Once disrupted, the droplets have a 45% smaller mean diameter than regular droplets. Under all conditions, disrupting and disrupted droplets are slower than regular droplets while the disruption does not significantly accelerate disrupted droplets.

The SpraySyn burner is a new system recently developed at the University of Duisburg-Essen to investigate experimentally nanoparticle synthesis in spray flames for a variety of materials. The current project aims at performing direct numerical simulations with detailed physicochemical models of configurations closely related to this burner. The effect of using different solvents to produce titanium-dioxide (TiO2) nanoparticles is discussed in this work. The two solvents considered are o-xylene and ethanol mixed in liquid state with tetraisopropoxide to form TiO2. The liquid is injected into a pilot flame as dispersed spray with a carrier flow (dispersion gas). The resulting particle size distribution is examined as well. It is in particular observed that using ethanol leads to faster agglomeration and larger nanoparticles. This effect is qualitatively similar to that found when injecting smaller liquid spray droplets.

Describing spatiotemporal evolution and characteristics of dispersed systems using the population balance equation (PBE), examples including sectional and moment methods are fraught with numerous issues. Hence, this study develops an accurate method by combining computational fluid dynamics and population balance-Monte Carlo method (CFD-PBMC) with a moderate computational cost. An efficient sub-model for particle migration was proposed to simulate the convection and diffusion processes of particulate flows. A graphics processing unit (GPU)-based parallel computation was performed to accelerate the high-dimensional CFD-PBMC. Several classical cases with analytical or benchmark solutions were simulated, and a comprehensive comparison was made using the classical weighted random walk method. Good agreements were obtained, except in the case of radial migration, the reasons for which are explained in detail. The measured speedups on the GPU showed a factor of similar to 450 for pure migration and similar to 50 for the CFD-PBMC method when compared with a standard high-performance computer.

The major challenges in producing highly electrically conductive copper films are the oxide content and the porosity of the sintered films. This study developed a multilayer sintering method to remove the copper oxides and reduce copper film porosity. We used a self-built arc discharge reactor to produce copper nanoparticles. Copper nanoparticles produced by arc discharge synthesis have many advantages, such as low cost and a high production rate. Conductive inks were prepared from copper nanoparticles to obtain thin copper films on glass substrates. As demonstrated by scanning electron microscopy analyses and electrical resistivity measurements, the copper film porosity and electrical resistivity cannot be significantly reduced by prolonged sintering time or increasing single film thickness. Instead, by applying the multilayer sintering method, where the coating and sintering process was repeated up to four times in this study, the porosity of copper films could be effectively reduced from 33.6% after one-layer sintering to 3.7% after four-layer sintering. Copper films with an electrical resistivity of 3.49 ± 0.35μΩ·cm (two times of the bulk copper) have been achieved after four-layer sintering, while one-layer sintered copper films were measured to possess resistivity of 11.17 ± 2.17μΩ·cm.

Identifying structure–property relationships of polycrystalline microstructures demands an accurate and precise quantification of their features. Measuring grain sizes is tedious and creates a non-transparent bias when being performed by different individuals, impairing comparability with references. Here, we present a novel use of region-based convolutional neural networks (R-CNNs) to quantify several microstructural characteristics and their distributions: Feret diameter, axis length, area, circumference, dihedral angle and coordination number. We utilize a two-step approach: (i) a semi-automatic annotation tool to generate training data for (ii) a fully automated R-CNN, quantitatively evaluating images. Using Al-doped ZnO as a model system, we trained two R-CNNs, one for ZnO and one for precipitated ZnAl2O4. The R-CNN performs well in evaluating grain size characteristics from images with low contrast and in differentiating uni-, and bimodal grain size distributions on the sub-micron, and nanoscale. An extended statistical analysis of the distributions is performed to extract microstructural parameters quantitatively. This innovative solution makes grain size measuring amenable, time-effective, less biased, consistent, and statistically more precise.

Because numerical challenges are encountered when dealing with multiple variates for the population balance equation (PBE), the Monte Carlo (MC) simulation becomes promising to provide comprehensive records of the aerosol evolution. Based on a previous sub-model for particle migration coupled with computational fluid dynamics (CFD), an integrated CFD-PBMC method is developed to describe the multiple particle attributes in practical bismuth vapor-to-aggregate cases. Relevant sub-models are decoupled with appropriate timestep strategies, including the nucleation, condensation, particle transport and other dynamics processes. Parallel computing on the graphics processing unit (GPU) is applied to accelerate the simulation. Compared with cor-responding experiments and simulation, a larger nucleation zone is observed with slight differences in the nucleation rates, which suits the "cut-off" experiment better. Multiple variates of particles are well predicted while a slightly overestimated aerosol concentration is found, which are analyzed in detail on the basis of the spatiotemporal evolution.

Powders and their mixtures are elemental for many industries (e.g., food, pharmaceutical, mining, agricultural, and chemical). The properties of the manufactured products are often directly linked to the particle properties (e.g., particle size and shape distribution) of the utilized powder mixtures. The most straightforward approach to acquire information concerning these particle properties is image capturing. However, the analysis of the resulting images often requires manual labor and is therefore time-consuming and costly. Therefore, the work at hand evaluates the suitability of Mask R-CNN—one of the best-known deep learning architectures for object detection—for the fully automated image-based analysis of particle mixtures, by comparing it to a conventional, i.e., not machine learning-based, image analysis method, as well as the results of a trifold manual analysis. To avoid the need of a laborious manual annotation, the training data required by Mask R-CNN are produced via image synthesis. As an example for an industrially relevant particle mixture, endoscopic images from a fluid catalytic cracking reactor are used as a test case for the evaluation of the tested methods. According to the results of the evaluation, Mask R-CNN is a well-suited method for the fully automatic image-based analysis of particle mixtures. It allows for the detection and classification of particles with an accuracy of 42.7% for the utilized data, as well as the characterization of the particle shape. Also, it enables the measurement of the mixture component particle size distributions with errors (relative to the manual reference) as low as −2±5 for the geometric mean diameter and −6±5% for the geometric standard deviation of the dark particle class of the utilized data, as well as −8±4% for the geometric mean diameter and −6±2% for the geometric standard deviation of the light particle class of the utilized data. Source code, as well as training, validation, and test data publicly available.

A model flow reactor provides a narrow particle temperature-residence time distribution with well-defined conditions and is mandatory to measure changes of the particle structure precisely. The experimental data of iron and iron oxide agglomerates are used to determine the sintering kinetics considering the temperature-time history of the particles. Thousand particle trajectories are tracked in a validated CFD model at three different furnace temperatures each. Strongly agglomerated particles with a small primary particle size (similar to 4 nm) are synthesized by spark discharge and are size-selected (25-250 nm) before sintering. The structure development is measured simultaneously with different online instrumentations and the structure calculated by means of structure models. A simple sintering model, based on the reduction of surface energy, is numerically quantified with the experimental results. The surface of the particles is strongly dependent on the primary particle size and the agglomerate structure. The chemical phase is analyzed using the offline techniques XANES, XRD, and EELS. It is observed that the addition of hydrogen led to a reduction of iron oxide to iron nanoparticles and to changes of the sintering kinetics. The sintering exponent m = 1 was found to be optimal. For Fe, an activation energy E-a of 59.15 KJ/mol and a pre-exponential factor A(s) of 1.57 10(4) s/m were found, for Fe(3)O(4 )an activation energy E-a of 55.22 kJ/mol and a pre-exponential factor A(s) of 2.54 10(4) s/m.

高温物理气相合成是合成纳米颗粒的重要方法之一,该过程包含复杂的成核、生长、凝并等颗粒动力学事件,对其进行精准、快速的数值模拟具有重要的指导意义.基于群平衡Monte Carlo方法,本研究采用GPU统一计算设备架构(CUDA)并行算法模拟Ag纳米颗粒的高温物理气相合成过程.为了提高计算效率,颗粒凝并事件考虑了快速Monte Carlo方法.该方法能够很好地描述纳米颗粒的高温物理气相合成过程,与实验测量对比表明,模拟结果基本吻合颗粒群尺度分布,计算速率平均提高13.8％.

CrTiN thin films are known to form a solid solution independent from the Ti content. Using a novel spatially separated synthesis approach, consisting of magnetron sputtering and atmospheric-pressure arc evaporation, artificial CrTiN nanocomposites were deposited. For the nanocomposite formation, TiN nanoparticles were synthesized using a transferred arc reactor and directly injected into growing CrN thin films using an aerodynamic lens system. The CrN and CrTiN thin films were deposited using various deposition conditions, such as heating power, substrate rotation velocity, nanoparticle injection distance, and cathode setup. The deposited thin films were analyzed regarding their physical structure, microstructure and mechanical properties. Based on the investigations, between 0.02 and 0.11 at.-% of TiN nanoparticles are embedded in the CrN matrix dependent on the deposition parameters. 2D GI-XRD experiments using synchrotron radiation confirm the nanocomposite structure for the two thin films with the highest TiN nanoparticle content. The crystallite size of the CrN thin film decreases from 9.4 +/- 2.3 nm to 5.3 +/- 1.2 nm due to the embedding of the nanoparticles. Concerning the physical structure, the nanoparticle injection leads to a change of the texture, as shown by the Debbye-Scherrer rings. Based on TEM-investigations, TiN nanoparticle agglomerates lead to a coarser microstructure of the CrN matrix. The hardness of the thin films is not significantly affected by the nanoparticle embedment. The nanoparticle injection distance and cathode setup reveal the highest impact on the film properties.

In this work, we introduce a new particle mass spectrometer (AP-PMS) that is able to detect particle-size distributions at ambient pressure using a three-stage pumping design. This device is demonstrated for direct sampling from the particle formation in spray-flame synthesis of iron oxide nanoparticles. Aerosol sampling is performed by a probe with integrated dilution that has been characterized and configured by computational fluid dynamics simulations and the chamber-skimmer system has been investigated by schlieren imaging. The system was validated by detailed characterization of a standardized sooting flame and by iron oxide nanoparticles generated in the SpraySyn burner from iron nitrate dissolved in a mixture of ethanol and 2-ethylhexanoic acid. The PMS results are compared to additional inline measurements with SMPS and ELPI + as well as with TEM measurements of thermophoretically sampled materials from the same location in the spray flame.

The synthesis of nanocomposites is limited to thermodynamically immiscible phases or to phase separation by exceeding the limits of solution. Hence, the formation of nanocomposites based on transition metals, revealing a nanocrystalline Metal-Nitride/nanocrystalline Metal-Nitride structure, is restricted. These restrictions can be overruled by a spatially separated synthesis of the two phases and a recombination during the deposition. With this approach, the limits of current systems can be expanded, enabling the synthesis of artificial nanocomposites based on a variety of materials. We demonstrate the synthesis of a composite of two nanocrystalline phases of the miscible transition metal-nitrides CrN and TiN. TiN nanoparticles were synthesized using an atmosphericpressure arc reactor and in-situ injected into a growing CrN thin film. The thin films are analyzed regarding their physical- and microstructure using two-dimensional GIXRD, XPS based on synchrotron radiation and TEM. The CrTiN thin film reveals a two-phase structure consisting of nanocrystalline CrN and TiN phases with crystallite sizes of 9 nm and 4 nm according to GIXRD. XPS indicates bonding of Cr-N, Cr-Cr, and Ti-N. No hint for Cr-Ti bonding was found, excluding (Cr,Ti)N solid solution formation. Based on the TEM-investigations, TiN nanoparticles are embedded as agglomerates in the CrN matrix.

Fiber-shaped materials (e.g. carbon nano tubes) are of great relevance, due to their unique properties but also the health risk they can impose. Unfortunately, image-based analysis of fibers still involves manual annotation, which is a time-consuming and costly process. We therefore propose the use of region-based convolutional neural networks (R-CNNs) to automate this task. Mask R-CNN, the most widely used R-CNN for semantic segmentation tasks, is prone to errors when it comes to the analysis of fiber-shaped objects. Hence, a new architecture - FibeR-CNN - is introduced and validated. FibeR-CNN combines two established R-CNN architectures (Mask and Keypoint R-CNN) and adds additional network heads for the prediction of fiber widths and lengths. As a result, FibeR-CNN is able to surpass the mean average precision of Mask R-CNN by 33% (11 percentage points) on a novel test data set of fiber images. (C) 2020 Elsevier B.V. All rights reserved.