Analytische Chemie
Filtern
Dokumenttyp
- Zeitschriftenartikel (16)
- Posterpräsentation (10)
- Vortrag (9)
- Sonstiges (2)
- Beitrag zu einem Sammelband (1)
- Beitrag zu einem Tagungsband (1)
- Forschungsdatensatz (1)
Sprache
- Englisch (40)
Schlagworte
- Electron microscopy (40) (entfernen)
Organisationseinheit der BAM
- 6 Materialchemie (30)
- 6.1 Oberflächen- und Dünnschichtanalyse (30)
- 1 Analytische Chemie; Referenzmaterialien (7)
- 1.2 Biophotonik (5)
- 1.9 Chemische und optische Sensorik (2)
- 5 Werkstofftechnik (2)
- 5.1 Mikrostruktur Design und Degradation (2)
- 6.7 Materialsynthese und Design (2)
- 6.3 Strukturanalytik (1)
- 6.6 Physik und chemische Analytik der Polymere (1)
Paper des Monats
- ja (2)
Eingeladener Vortrag
- nein (9)
We present a workflow for obtaining fully trained artificial neural networks that can perform automatic particle segmentations of agglomerated, non-spherical nanoparticles from scanning electron microscopy images “from scratch”, without the need for large training data sets of manually annotated images. The whole process only requires about 15 minutes of hands-on time by a user and can typically be finished within less than 12 hours when training on a single graphics card (GPU). After training, SEM image analysis can be carried out by the artificial neural network within seconds. This is achieved by using unsupervised learning for most of the training dataset generation, making heavy use of generative adversarial networks and especially unpaired image-to-image translation via cycle-consistent adversarial networks. We compare the segmentation masks obtained with our suggested workflow qualitatively and quantitatively to state-of-the-art methods using various metrics. Finally, we used the segmentation masks for automatically extracting particle size distributions from the SEM images of TiO2 particles, which were in excellent agreement with particle size distributions obtained manually but could be obtained in a fraction of the time.
Core-shell (CS) particles with a polymeric core and a silica shell play an important role in the materials and (bio)analytical sciences. Besides the establishment of reliable synthesis procedures, comprehensive particle characterization is essential for batch-to-batch reproducibility and objective performance assessment across architectures, protocols, and laboratories. Particle characterization with respect to size, size distribution, shell thickness and texture, surface area and roughness or materials composition is commonly conducted with different analytical methods, often on different samples. Our approach uses a dual-mode TSEM/SEM set-up with an EDX detector to obtain a complementary data set with sufficient statistical confidence of one and the same sample on a single instrument. Our protocol reveals information about size, size distribution and shell thickness of the various particles employed from overview images, while an increased field of view (FOV) and high-resolution EDX analysis yields detailed information on shell texture and elemental composition. An image analysis tool was developed to derive and quantify the profile roughness of CS particles from individual beads. Comparison with surface roughness data from AFM showed a similar trend in roughness across the series of particles. Reliable classification into smooth and rough is proposed and roughness changes within different particle batches were tracked systematically.
In this contribution different ways are explored with the aim to generate suitable training data for ‘non-ideal’ samples using various approaches, e.g., computer-generated images or unsupervised learning algorithms such as generative adversarial networks (GANs). We used these data to train simple CNNs to produce segmentation masks of SEM images and tested the trained networks on real SEM images of complex nanoparticle samples. The novel use of CNN for the automated analysis of the size of nanoparticles of complex shape and with a high degree of agglomeration has proved to be a promising tool for the evaluation of particle size distribution on a large number of constituent particles. Further development and validation of the preliminary model, respectively larger training and validation data sets are necessary.
In this contribution different ways are explored with the aim to generate suitable training data for ‘non-ideal’ samples using various approaches, e.g., computer-generated images or unsupervised learning algorithms such as generative adversarial networks (GANs). We used these data to train simple CNNs to produce segmentation masks of SEM images and tested the trained networks on real SEM images of complex nanoparticle samples. The novel use of CNN for the automated analysis of the size of nanoparticles of complex shape and with a high degree of agglomeration has proved to be a promising tool for the evaluation of particle size distribution on a large number of constituent particles. Further development and validation of the preliminary model, respectively larger training and validation data sets are necessary.
As part of the development of a library of accurate and efficient methods for measurement of nanoparticle properties, we develop and optimize a method for the efficient analysis of nanoparticle size distribution from suspensions via microprinting and digital analysis of electron microscopy (SEM and TEM) images, with the ultimate aim of automated quantitative concentration analysis (calculated from drop volume). A series of different nanoparticle suspensions (gold, latex, and SiO2 in varying sizes and concentrations) were printed onto TEM grids in a 4 x 4 array in the concentration range 7x10^8 to 1x10^11 nanoparticles/mL and imaged with SEM. Concentrations and printing conditions (temperature, relative humidity) were varied in order to minimize the coffee-ring effect.
In this paper one refined approach is applied to determine the exact orientation of bipyramidal TiO2 nanoparticles prepared with good dispersion as almost isolated particles on a carbon TEM grid. The advantages of the recently developed high-throughput Transmission Kikuchi Diffraction (TKD) are used to identify quickly and reliably the geometrical orientation of the crystalline TiO2 nanoparticle bipyramids (anatase) on a statistically relevant number of particles.
In this paper one refined approach is applied to determine the exact orientation of bipyramidal TiO2 nanoparticles prepared with good dispersion as almost isolated particles on a carbon TEM grid. The advantages of the recently developed high-throughput Transmission Kikuchi Diffraction (TKD) are used to identify quickly and reliably the geometrical orientation of the crystalline TiO2 nanoparticle bipyramids (anatase) on a statistically relevant number of particles.
Extraction of true, 3D shape (and size) of non-spherical nanoparticles (NPs) is associated with errors by conventional 2D electron microscopy using projection images. Significant efforts within the ISO technical committee TC 229 ‘Nanotechnologies’ are aimed at establishing accurate TEM and SEM measurement of NP size and shape as robust, standard procedures. Study groups have been organizing inter-laboratory comparisons on well-selected NP systems according to the market needs, such as aggregated titania nano-powder for which size and shape distribution of primary crystallites of irregular shape must be measured accurately. To be noticed is e. g. the fact that the measurement procedure allows only manual selection of the particles clearly distinguishable for analysis as well as manual definition of the contour of the imaged NPs.
An inter-laboratory exercise on titania NPs (pure anatase, grown by hydrothermal synthesis) of well-defined non-spherical shape, i.e. bipyramidal has been recently started within ISO/TC 229 under similar conditions as for the irregular shaped titania. Overlapped particles were allowed to be considered, as long as they are clearly distinguishable. One decisive NP selection criterion was to analyze only those NPs with a roundness value below 0.7, i.e. the NPs laying on the support foil and, hence, with projection areas clearly deviating from perfect circles (R=1). The overall evaluation (for 15 labs) of the size descriptors (area, Feret, minFeret, perimeter) and shape descriptors (aspect ratio, roundness, compactness, extent) by analysis of variance is just to be finished and included in ISO/WD 21363 Nanotechnologies -- Protocol for particle size distribution by transmission electron microscopy.
Extraction of true, 3D shape (and size) of non-spherical nanoparticles (NPs) is associated with errors by conventional 2D electron microscopy using projection images. Significant efforts within the ISO technical committee TC 229 ‘Nanotechnologies’ are aimed at establishing accurate TEM and SEM measurement of NP size and shape as robust, standard procedures. Study groups have been organizing inter-laboratory comparisons on well-selected NP systems according to the market needs, such as aggregated titania nano-powder for which size and shape distribution of primary crystallites of irregular shape must be measured accurately. To be noticed is e. g. the fact that the measurement procedure allows only manual selection of the particles clearly distinguishable for analysis as well as manual definition of the contour of the imaged NPs.
An inter-laboratory exercise on titania NPs (pure anatase, grown by hydrothermal synthesis) of well-defined non-spherical shape, i.e. bipyramidal has been recently started within ISO/TC 229 under similar conditions as for the irregular shaped titania. Overlapped particles were allowed to be considered, as long as they are clearly distinguishable. One decisive NP selection criterion was to analyze only those NPs with a roundness value below 0.7, i.e. the NPs laying on the support foil and, hence, with projection areas clearly deviating from perfect circles (R=1). The overall evaluation (for 15 labs) of the size descriptors (area, Feret, minFeret, perimeter) and shape descriptors (aspect ratio, roundness, compactness, extent) by analysis of variance is just to be finished and included in ISO/WD 21363 Nanotechnologies -- Protocol for particle size distribution by transmission electron microscopy.