5 Werkstofftechnik
Filtern
Dokumenttyp
- Posterpräsentation (24) (entfernen)
Referierte Publikation
- nein (24)
Schlagworte
- Glass (8)
- Crack growth (4)
- Crystallization (3)
- Soda-lime silicate glass (3)
- Vickers (3)
- Bildanalyse (2)
- Crystal orientation (2)
- DCB (2)
- Ontologie (2)
- Oxidglas (2)
Organisationseinheit der BAM
- 5.6 Glas (24) (entfernen)
The system Na2O.B2O3-SiO2 (NBS) is the basis of many industrial glass applications and therefore one of the most studied systems at all. Glass formation is possible over a wide compositional range, but the system also contains ranges of pronounced phase separation and crystallization tendency. Despite its importance, experimental data are limited to few compositional areas. The general understanding and modelling of glass formation, phase separation, and crystallization in this system would therefore be easier if small step melt series could be studied. The efficient melting of such glass series is now possible with the new robotic glass melting system at the Federal Institute for Materials Research and Testing (BAM, Division Glasses). Using three exemplary joins within this NBS system, the small step changes of glass transition temperature (Tg), crystallization behavior as well as glass density (Roh) was studied. Additionally, experimental Tg and Roh data were compared with their modeled counterparts using SciGlass and a newly developed DFT model, respectively.
GlasDigital
(2023)
Der aktuelle Stand des MateriaDigital1 Projektes GlasDigital wird vorgestellt. Hierbei wird allgemein die Problem- und Zielstellung präsentiert, als auch auf 2 separaten Postern die Ergebnisse. Diese beinhalten zum Einen die smarte Gestaltung der robotergestützten Glasschmelzanlage der BAM inkl. Analytik und zum Anderen die Digitalisierungsbestrebungen im Bereich Glas, d.h. ML-gestützte C-S-P-Simulation, Ontologie für den Werkstoff Glas, Digitaler Zwilling des Gießprozesses.
Enhancing the fracture toughness is still one of the major challenges in the field of oxide glasses. To screen different glass systems for promising candidates, a theoretical expression for the fracture surface energy, G, linked to the fracture toughness, KIc, is thus of interest. Extending our earlier work on nucleation and surface energies [1], we present a simple approach for predicting the fracture surface energy of oxide glasses, G using readily available crystallographic structure data and diatomic bond energies. The proposed method assumes that G of glass equals the surface fracture energy of the weakest fracture (cleavage) plane of the isochemical crystal. For non-isochemically crystallizing glasses, an average G is calculated from the weighed fracture energy data of the constitutional crystal phases according to Conradt [2]. Our predictions yield good agreement with the glass density- and chemical bond energy-based prediction model of Rouxel [3] and with experimentally obtained G values known at present.
[1] C. Tielemann, S. Reinsch, R. Maass, J. Deubener, R. Müller, J. Non-Cryst. Solids 2022, 14, 100093
[2] R. Conradt, J. Non-Cryst. Solids 2004, 345-346, 16
[3] R., Tanguy, Scripta Materialia 2017, 109-13, 137
Up to now, oriented surface crystallization phenomena are discussed controversially, and related studies are restricted to few glasses. For silicate glasses we found a good correlation between the calculated surface energy of crystal faces and oriented surface nucleation. Surface energies were estimated assuming that crystal surfaces resemble minimum energy crack paths along the given crystal plane. This concept was successfully applied by Rouxel in calculating fracture surface energies of glasses. Several oriented nucleation phenomena can be herby explained assuming that high energy crystal surfaces tend to be wetted by the melt. This would minimize the total interfacial energy of the nucleus. Furthermore, we will discuss the evolution of the microstructure and its effect on the preferred crystal orientation.
As part of a joint project involving the Fraunhofer Institute for Silicate Research (ISC), the Friedrich Schiller University of Jena, the Clausthal University of Technology and the Federal Institute for Materials Research and Testing (BAM), digital tools are to be created for the development of new types of glass materials. Current processes for the production of glasses with improved properties are usually very cost- and energy-intensive due to the low degree of automation and are subject to long development cycles. The use of robotic synthesis processes in combination with self-learning machines is intended to overcome these problems in the long term. The development of new types of glass can then not only be accelerated considerably, but also be achieved with much less effort.
In this talk, data generation via a robotic high-throughput glass melting system is presented, which should be the experimental basis for the ontology developed within the project GlasDigital.
Melting of Na2CO3-ZnO-B2O3 batches containing up to 16.8 wt% AgNO3 (5 mol% Ag2O in the target glass composition) was observed in situ by means of hot stage microscopy. In all batches metallic silver precipitation took place as most of the silver nitrate was reduced to metallic silver before Ag+ ions could be dissolved in the evolving borate melts. In turn, only traces of Ag+ (<300 ppmw) were dissolved in the sodium zinc borate glass melts under study. It is assumed that the oxidation to Ag+ was limited due to poor availability of reducible oxygen in the glass melts and presence of Na2O being a stronger base than Ag2O. Thus, the precipitated metallic silver formed droplets of different sizes. The larger droplets (d > 20 µm) were already settled at the bottom of the container and remained constant in size upon dwelling for 1 h at 1050 °C of about one hour and the subsequent cooling (45 K/min) to room temperature, whereas the smaller droplets (d < 20 µm) were mobile in the borate melt due to Marangoni and Stokes motion. For the latter droplets, coalescence was observed in situ. A growth of larger droplets at the expense of smaller ones, i.e., Ostwald ripening was also expected but could not be studied with the used experimental equipment.
Low melting zinc borate glasses awake interest to replace lead borate glasses in the silver metallization pastes for solar cells or microelectronics. In the current study, characteristic properties of alkali zinc borate glasses (X2O-ZnO-B2O3, X = Li, Na, K, Rb) were compared to an earth alkali zinc borate glass (CaO-ZnO-B2O3). Additionally, zinc oxide is partially substituted by lead oxide or cooper oxide in the borate glasses (Li2O-PbO-B2O3, Na2O ZnO CuO-B2O3). The alkali zinc borate glasses indicate less differences in Raman spectra, and thus in structural properties, in comparison to the Ca and Pb ions influence. LPbB (Tg = 401 °C) has a lower viscosity than LZB (Tg = 468 °C) and CaZB has the highest glass transition temperature (Tg = 580 °C). The Angell plot for the alkali zinc borate glasses shows a high fragility m = 80. Besides Tg, the density measured by means of the Archimedean principle, molar volume, and coefficient of thermal expansion (CTE) of the glasses were investigated. Trends could be found according to alkali ions or intermediate oxides. The density increases with decreasing alkali ion size from KZB (2.632 g/cm3) to LZB (2.829 g/cm3) and increases from LZB to LPbB (3.764 g/cm3). CTE ranges between 7.09 10-6 K-1 for CaZB and 11.5 10 6 K 1 for KZB and RZB. The differential thermal analysis (DTA) and X ray diffraction (XRD) indicate crystallization of various crystalline phases during heating with 5 K/min in most cases.
A heat treated glass that formed crystals under controlled conditions can be a glass ceramic. A proper understanding of crystallization in glassy melts is needed yet still lacking. Involved processes include the mobility of Elements within the glass during heat treatment and the reorganization of atomic bonds during crystallization. This change in coordination number of certain elements is easily observable in bulk glass samples during heat treatment with XAS. Our plan was to monitor these reorganization processes in situ at the immediate surface during heat treatment with PEEM to get time and spatial resolved data.
Fresnoite glass with excess SiO2 exhibits oriented surface crystallization, in contrast to the stoichiometric glass composition. Recent EBSD studies documented that the crystals in BTS (2BaO-TiO2-xSiO2, x=0-3) can occur in a distinct [101]-orientation perpendicular to the surface and claimed that this orientation is not a result of growth selection. During these previous studies, however, the effect of surface preparation and surrounding atmosphere during the crystallization experiments were not considered. As these parameters may influence crystal orientation, we studied the surface crystallization of a BTS glass (2BaO-TiO2–2.75SiO2) under controlled conditions with the help of light, electron and polarisation microscopy as well as EBSD. Heat treatments for one hour at 840°C of fractured BTS glass surfaces in air resulted in a large number of not-separable surface crystals. This large number of crystals can be caused by dust particles, which act as nucleation agents. As crystal growth velocity could further be influenced by humidity, our experiments are performed in a filtered and dried air atmosphere. The crystal morphology and orientation will be analysed in dependence of the sample preparation and a differing surrounding atmosphere.
Characterization of early crystallization stages in surface-crystallized diopside glass-ceramics
(2019)
Structure formation in glass-ceramics by means of surface crystallization is a challenging open question and remains elusive to definite answers. In several glass-ceramic systems, oriented crystal layers have been observed at the immediate surface, including diopside and some fresnoite systems. However, it is still open to debate, whether oriented surface crystallization is the result of oriented nucleation or growth selection effects. In the same vein, there is still discussion whether surface nucleation is governed by surface chemistry effects or by defects serving as active nucleation sites.
In order to help answer these questions, annealing experiments at 850°C have been performed on a MgO·CaO·2SiO2 glass, leading to the crystallization of diopside at the surface. Different annealing durations and surface treatment protocols (i.a. lapping with diamond slurries between 16 µm and 1 µm grain size) have been applied. Particular focus has been put on earliest crystallization stages, with crystal sizes down to about 200 nm. The resultant microstructure has been analyzed by electron backscatter diffraction (EBSD) and two different kinds of textures have been observed, with the a- or b-axis being perpendicular to the sample surface and the c-axis lying in the sample plane. Even at shortest annealing durations, a clear texture was present in the samples. Additionally, selected samples have been investigated with energy-dispersive x-ray spectroscopy in the scanning transmission electron microscope (STEM-EDX). The diopside crystals have been found to exhibit distinguished submicron structure variations and the glass around the crystals was shown to be depleted of Mg.
Morphologies of Fresnoite surface crystals of BT0.75S glass under different atmospheric conditions
(2019)
Fresnoite glass-ceramics are characterized by piezoelectric, pyroelectric and non-linear optical properties. These properties can be adjusted by orienting the fresnoite crystals during crystallization. Crystallization begins at the surface. The first surface crystals are not oriented perpendicular and, therfore, produce a surface layer that has not the intended properties. To overcome this issue, the formation of fresnoite surface crystals at different atmospheric conditions was studied.
High conductive silver metallization pastes are key components in advanced electronics and photovoltaics. Increasing demands on efficiency, miniaturization and ever shorter time-to-market require tailored glass-silver-pastes. In these pastes, low-melting glasses act as a sintering aid achieving better sintering, adhesion and contact formation for solar cells. Yet, the related liquid phase sintering of silver-glass-composites and the underlying mechanism of silver dissolution, transport and reprecipitation are rarely investigated. In this study, systematically varied low melting alkaline zinc borate, alkaline earth borate, and Pb- and Bi-glasses are investigated. Glass transition and crystallization are studied with dilatometry, DTA and XRD. Sintering of the pure glasses, pure silver and silver-glass-composites is analyzed with Hot Stage Microscopy, optical and electron microscopy. Since oxygen dissolved in silver powders can affect the silver dissolution as silver oxide in the matrix oxide glasses, the O2-content of silver powders is determined by Vacuum Hot Extraction. The glass transition temperature of the glasses under study varies between 370 °C and 590 °C whereas the sinter onset largely ranges between 400 °C and 600 °C. On the other hand, it scattered between 200 °C and 450 °C for selected commercial Ag-powders of different particle size and morphology.
Crack healing in glass ceramic solid oxide fuel cell (SOFC) sealants is of utmost importance as cracks caused by thermal cycling remain a bottleneck in developing durable SOFC. Whereas no or low crystal volume fraction seems most favorable for viscous crack healing, it does not for load bearing and undesired diffusion. On the other hand, crystals or filler particles can make the sealant less prone to these disadvantages but it could increase the effective composite viscosity and retard crack healing.
Against this background, the influence of crystal volume fraction, phi, on viscous crack healing in glass matrix composites prepared from soda lime silicate glass and zirconia filler particles was studied. Vickers indention induced radial cracks were healed isothermally during interrupted annealing steps and monitored with optical microscopy. Due to the slow crystallization of the glass under study, phi could be kept constant during crack healing.
For bulk glass samples (phi = =), the decrease in radial crack length was retarded by an initial increase in crack width due to crack rounding. Up to phi = 0.15 the increase in effective viscosity retarded this crack broadening thereby yielding faster crack healing. For phi > 0.15, crack broadening was progressively suppressed but the same was true for crack healing, which was fully prevented above phi = 0.3. Results indicate that optimum micro structures can prevent crack broadening limited by the global effective composite viscosity and this way promote crack healing limited by local glass viscosity.
Glass strength and fatigue is limited by surface cracks. As subcritical crack growth (SCCG) is governed by ambient humidity, stress corrosion at the crack tip is widely accepted to be the underlying mechanism. However, as water is known to have decisive effect on glass properties and can rapidly enter the crack tip near glass region, SCCG could be affected by such water related phenomena. We tried to mimic these effects studying water dissolution and speciation, mechanical properties, and SCCG in water-bearing glasses. For this purpose, glasses up to 8 wt% water have been prepared by means of high-pressure melting of glass powder - water mixtures.
As part of this effort, SCCG in dry and hydrous commercial micros¬cope slide glass (CW = 6 wt%) was studied in double cantilever beam (DCB) geometry and sub-Tg relaxation was measured by Dynamic Mechanical Analysis (DMA).
For SCCG in ambient air (24% r.h.), SCCG was promoted by the presence of 6wt% bulk water with respect to the dry glass. On the other hand, stress intensity values, KI, required to cause slow crack growth (v < 10-6 ms-1) resemble literature findings for float glass of similar composition in liquid water, which might represent the maximum possible promoting effect of ambient water on SCCG.
For SCCG in vacuum (10-3 mbar), dissolved bulk water causes even more pronounced effects. Most strikingly, it strongly decreases the slope of the log v(KI)-curve, which is a measure of dissipated energy during fracture. A strong increase of sub-Tg relaxation with increasing water content was confirmed by DMA. As a consequence, slow crack growth occurs at KI values as measured in the dry glass whereas fast crack growth occurs at much larger KI than that of the dry glass. Kinks and shoulders shown by the inert log v(KI)-curve indicate that bulk water does not simply affect bulk mechanical properties.
Sintered bioactive glass scaffolds of defined shape and porosity, e.g. made via additive manufacturing, must provide sufficient bioactivity and sinterability. As higher bioactivity is often linked to high corrosion and crystallization tendency, a certain compromise between sintering ability and bioactivity is therefore required. Groh et al. developed a fluoride-containing bioactive glass (F3), which allows fiber drawing and shows a bioactivity well comparable to that of Bioglass®45S5.
To study whether and to what extent the sinterability of F3 glass powder is controlled by particle size, coarse and fine F3 glass powders (300-310µm and 0-32µm) were prepared by crushing, sieving and milling. Sintering, degassing and phase transformation during heating were studied with heating microscopy, vacuum hot extraction (VHE), DTA, XRD, and SEM.
For the coarse glass powder, sintering proceeds slowly and is limited by surface crystallization of primary Na2CaSi2O6 crystals. Although the crystallization onset of Na2CaSi2O6 is shifted to lower temperature, full densification is attained for the fine powder. This finding indicate that certain porosity might be tuned via particle size variation. Above 900°C, intensive foaming is evident for the fine powder. VHE studies revealed that carbon species are the main foaming source.
Premature failure of glass under load is caused by sub-critical crack growth (SCCG) originate from microscopic flaws at the surface. While SCCG is related to the humidity of the ambient atmosphere, leading to stress corrosion phenomena at the crack tip, the detailed mechanism and the effect of different network formers are still not fully understood. For more clarity, various soda silicate glasses with a second network former were investigated by double cantilever beam technique: Na2O*Al2O3*SiO2 (NAS), Na2O*B2O3*SiO2 (NBS), Na2O*PbO*SiO2 (NPbS).
Three effects on the crack growth velocity, v, versus stress intensity, KI, curves were found out. The slope in region I, which is limited by corrosion, increases in the order NAS < NBS ≲ NPbS. The velocity range of region II reflecting the transition between corrosion effected and inert crack growth (region III), varies within one order of magnitude between the glasses. The KI region of inert crack growth strongly scatters between 0.4 and 0.9 MPam1/2. For comparison, crack growth at different humidity in commercial soda lime silicate glass (NCS) was measured.
Statistical analysis of Vickers induced subcritical crack growth in soda-lime silicate glasses
(2018)
Studies on Vickers induced subcritical crack growth are controversially discussed since the stresses that drive the crack growth are distributed three dimensionally within the material and cannot be retraced by available methods. Hence, empirical approaches are used to calculate mechanical material parameters such as the stress intensity factor KI. However, the results of these approaches show large deviations from those measured by standardized techniques such as double cantilever beam (DCB) or double cleavage drilled compression (DCDC). Yet, small specimen sizes and low specimen quantities can prevent the execution of DCB and DCDC measurements. Here we present an approach that is based on a statistical analysis of Vickers induced radial cracks. For this purpose more than 150 single radial cracks were analyzed. The cracks were generated in a commercial soda-lime silicate glass. The experiments were performed in a glovebox purged with dry nitrogen gas to minimize the influence of atmospheric water on crack growth. The temporally resolved evolution of the radial cracks was monitored in-situ using an inverted microscope equipped with a camera system directly below the Vickers indenter. An automated image analysis software was used to determine the crack length over time. The data show that the crack propagation and thereby the crack velocities are not uniformly but statistically distributed. These findings allow, using the statistical mean value of the distributions in combination with DCB data, a precise formulation of KI for each measured crack length.