@inproceedings{AckerDuckeRietigetal., author = {Acker, J{\"o}rg and Ducke, Jana and Rietig, Anja and M{\"u}ller, Tim and Eisert, Stefan and Reichenbach, Birk and L{\"o}ser, Wolfgang}, title = {Segregation, grain boundary milling, and chemical leaching for the refinement of metallurgical-grade silicon for photovoltaic application}, series = {Silicon for the Chemical and Solar Industry XII, Trondheim, 2014}, booktitle = {Silicon for the Chemical and Solar Industry XII, Trondheim, 2014}, editor = {Oye, Harald A. and Brekken, Harald and Rong, Harry and Tangstad, Merete and Tveit, Halvard}, publisher = {Department of Materials Science and Engineering, Norwegian University of Science and Technology}, address = {Trondheim}, isbn = {978-82-997357-8-0}, pages = {177 -- 188}, abstract = {The present work describes a completely new approach to the solidification refinement of metallurgical-grade silicon. The new process comprises the following steps: (i) The first step involves adding auxiliary metals to the molten silicon in order to segregate the metallic and non-metallic impurities in the secondary phase after cooling. (ii) The melt is rapidly cooled in the cellular solidification regime. This generates a Si microstructure with a defined cell size in which all cell boundaries are surrounded by the secondary phase. Furthermore, the secondary phase should form an interconnected three-dimensional network. (iii) The solids are crushed by shockwaves using electrohydraulic fragmentation techniques. The shockwaves lead to preferential crushing at the interface between the silicon and the secondary phase. (iv) The secondary phases are fast and effectively removed by microwave-assisted high-pressure leaching that was newly developed for this process. The potential of the new refinement procedure is demonstrated with auxiliary metals Ca, Al, and Ti. This new procedure yields a significant decrease in phosphorous and metal impurities.}, language = {en} } @misc{AckerSieberLangneretal., author = {Acker, J{\"o}rg and Sieber, Tim and Langner, Thomas and Herold, Steven}, title = {The impact of lattice strain on the reactivity of silicon}, series = {Silicon for the Chemical and Solar Industry XIV}, journal = {Silicon for the Chemical and Solar Industry XIV}, editor = {Andresen, Birger and Nygaard, Lars and Rong, Harry and Tangstad, Merete and Tveit, Halvard and Page, Ingrid Gamst}, publisher = {The Norwegian University of Science and Technology}, address = {Trondheim}, pages = {11 -- 20}, abstract = {The present study is focused on the question of how lattice strain mechanically introduced into silicon alters the chemical reactivity of the silicon atoms that are affected by the strain field on a microscopic length scale. The magnitude and local distribution of lattice strain are extracted from confocal Raman microscopy measurements. The reactivity of Si is expressed by the etch rate of Si after treatment with HF-HNO3-H2SiF6 mixtures. Then, the local etch rate is calculated from the local etch depth as determined by confocal microscopy. It has been found that tensile strain leads to the highest enhancement of the etch rate, followed by a compressive strain increase in the etch rate.}, language = {en} } @misc{LangnerSieberAcker, author = {Langner, Thomas and Sieber, Tim and Acker, J{\"o}rg}, title = {Etching Shapes the Topography of Silicon Wafers: Lattice-Strain Enhanced Chemical Reactivity of Silicon for Efficient Solar Cells}, series = {ACS Applied Nano Materials}, volume = {1}, journal = {ACS Applied Nano Materials}, number = {8}, doi = {10.1021/acsanm.8b00906}, pages = {4135 -- 4144}, abstract = {Multiwire sawing of silicon (Si) bricks is the state-of-the-art technology to produce multicrystalline Si solar wafers. The massive indentation of the abrasive Si carbide or diamond particles used leads to a heavily mechanically damaged layer on the wafer surface. Etching the surface layer using typical HF-HNO3-H2SiF6 acid mixtures reveals an unevenly distributed etch attack with etch rates several times higher than known for bulk Si etching. The present study follows the hypothesis that lattice strain, introduced by the sawing process, leads to an increase of the etch rate and determines the topography of the etched wafer, the so-called texture. Scratches were introduced into single crystalline Si surfaces in model experiments, and the magnitude and local distribution of lattice strain were extracted from confocal Raman microscopy measurements. The essential parameter used to describe the local reactivity of Si is the local etch rate, which was derived by confocal microscopy from the local height before and after etching. It was found that the reactivity of Si increases linearly with the magnitude of lattice strain. An increase in tensile strain raises the reactivity of Si significantly higher than an increase of compressive strain. The second decisive parameter is the reactivity of the etch mixture that correlates with the total concentration of the acid mixtures. Diluted acid mixtures with a low reactivity attack only the highest strained Si, whereas more concentrated and, therefore, more reactive acid mixtures can attack even slightly strained Si. Side effects, such as the behavior of amorphous or nanocrystalline Si and the generation of highly reactive intermediary species while etching, are discussed. The presence of unevenly distributed lattice strain of different magnitude and the resulting unevenly distributed reactivity of Si explain the features of a heterogeneous etch attack observed and the resulting topography of the etched wafer surface.}, language = {en} } @inproceedings{LangnerSieberAcker, author = {Langner, Thomas and Sieber, Tim and Acker, J{\"o}rg}, title = {Lattice strain controls the etching of solar wafer surfaces}, series = {9th European Silicon Days, 9-12 September 2018, Saarbr{\"u}cken, Germany, Book of abstracts}, booktitle = {9th European Silicon Days, 9-12 September 2018, Saarbr{\"u}cken, Germany, Book of abstracts}, editor = {Scheschkewitz, David and Kickelbick, Guido}, publisher = {Universit{\"a}t des Saarlandes}, address = {Saarbr{\"u}cken}, pages = {S. 201}, abstract = {Multi-wire sawing using an abrasive SiC slurry or diamond wires constitutes the main slicing techniques for multi- and monocrystalline silicon crystals in photovoltaics. The massive mechanical load during the sawing process creates a wafer surface layer characterized by lattice defects, pits, fractures, rifts, cracks, amorphous Si and even some high-pressure Si modifications, otherwise known as saw damage.[1] This highly defect-rich surface causes the rapid recombination of electron-hole pairs, requiring that it be removed by etching in order to manufacture solar cells and to generate a surface morphology having a low reflectivity which directly affects the solar cell's efficiency. However, etching of the saw damage features of a heterogeneous and laterally unevenly distributed etch attack and a significantly higher etch rate compared to the underlying bulk silicon.[2,3] The present study is focused on the question of how mechanically introduced lattice strain in single-crystalline silicon alters the chemical reactivity of the silicon atoms affected by the strain field on a microscopic length scale. The magnitude and local distribution of lattice strain were extracted from confocal Raman microscopy measurements according to Ref. 4. One of the parameters used to describe the reactivity of silicon is the local etch rate, which was derived from the local removal before and after etching by confocal microscopy. Wet-chemical etching was performed with HF-HNO3-H2SiF6 acid mixtures of different concentrations. It was found, that the reactivity of silicon increased linearly with the magnitude of lattice strain. In particular, an increase in tensile strain led to a higher increase in reactivity compared to the increase observed with growing compressive strain. The second decisive parameter is the reactivity of the etch mixture. Diluted acid mixtures with a low reactivity attack only the highest strained Si, whereas more concentrated and therefore more reactive acid mixtures are able to attack even slightly strained Si. Side effects, such as the behavior of amorphous or nanocrystalline Si and the generation of highly reactive intermediary species while etching, are discussed.}, language = {en} } @inproceedings{HeroldAcker, author = {Herold, Steven and Acker, J{\"o}rg}, title = {The influence of lattice deformations on the etch rates of potassium hydroxide on silicon}, series = {9th European Silicon Days, 9-12 September 2018, Saarbr{\"u}cken, Germany, Book of abstracts}, booktitle = {9th European Silicon Days, 9-12 September 2018, Saarbr{\"u}cken, Germany, Book of abstracts}, editor = {Scheschkewitz, David and Kickelbick, Guido}, publisher = {Universit{\"a}t des Saarlandes}, address = {Saarbr{\"u}cken}, pages = {S. 198}, abstract = {One of the most common methods to analyse silicon materials is Raman-spectroscopy.[1] More recently there is a focus in the analysis of lattice deformations during sample preparations.[2] These deformations not only increase the chance of crack formation but also increases the chemical reactivity of the silicon towards etch processes or oxidation processes.[3] In this contribution, we will present our latest results on the influence of lattice deformations induced by scratches and saw damage on the etch rate of potassium hydroxide. The lattice deformation will be measured using Raman-microscopy and correlated with the height difference measured with confocal microscopy. It will be shown that the etch rate strongly depends on the lattice deformation and that the etch rate will increase by up to the factor 10 in areas with tensile deformation.}, language = {en} } @inproceedings{RietigLangnerAcker, author = {Rietig, Anja and Langner, Thomas and Acker, J{\"o}rg}, title = {Dissolution of silicon in HF/HNO3 mixtures: A revised model}, series = {9th European Silicon Days, 9-12 September 2018, Saarbr{\"u}cken, Germany, Book of abstracts}, booktitle = {9th European Silicon Days, 9-12 September 2018, Saarbr{\"u}cken, Germany, Book of abstracts}, editor = {Scheschkewitz, David and Kickelbick, Guido}, publisher = {Universit{\"a}t des Saarlandes}, address = {Saarbr{\"u}cken}, pages = {S. 199}, abstract = {The dissolution of Si in HF/HNO3 consists of a set of complex reactions and thus a large number of reaction products. The most comprehensive picture of this reaction, the role of the involved reaction products and the reactivity of the HF/HNO3 mixtures depending on their composition is drawn by Steinert et al..[1]-[3] Based on the first systematic investigations on hydrogen formation by Hoffmann et al.[4], Acker et al. succeeded a first mass and electron balance for the reaction of silicon in HF/HNO3.[5] However, there is still a lack in interpretation of the mass end electron balances arising from several nitrous oxides. So far, the identified nitrogen oxides NO, NO2 and N2O were considered in sum[5] and neither separated nor individually studied or quantified. The aim of this work is to complete the mass and electron balance by the contribution of the individual nitrous oxides and to identify their individual formation pathways. Kinetic measurements of the NO and NO2 formation during the dissolution of Si, NO2 turns out as a result from the oxidation of the primary product NO by the HNO3 in the etching mixtures. Subsequently, NO and NO2 react to N2O3 dissolved in the acid mixture. The kinetics of both reactions were individually studied by bubbling NO in HF/HNO3 mixtures of different composition. The already identified intermediary species N4O62+ turns out to be formed by disproportionation of dissolved NO2 via N2O4 without dissolution of silicon. A detailed kinetic studied showed, that only dissolved N2O3 and not the intermediate N4O62+ contribute to the dissolution rate of silicon in HF/HNO3 acid mixtures. Finally, kinetic measurements revealed that the formed H2 reduces gaseous NO yielding to the final gaseous reaction products N2, N2O as well as to ammonium ions which all are formed with identical reaction rate. This reaction is assumed to proceed via NH2OH as intermediate. As result of the identification and quantification of intermediary and final reaction products a new reaction scheme needs to be established leading to a new approach to the mass and electron balance for the oxidation of silicon during the dissolution in HF/HNO3 mixtures.}, language = {en} } @inproceedings{AckerLangnerKoschwitz, author = {Acker, J{\"o}rg and Langner, Thomas and Koschwitz, Tim}, title = {Lattice-strain induced chemical reactivity of silicon}, series = {8th European Silicon Days 2018, Conference Proceedings}, booktitle = {8th European Silicon Days 2018, Conference Proceedings}, editor = {Marciniec, Bogdan}, publisher = {Wielkopolska Centre for Advanced Technologies, Adam Mickiewicz University}, address = {Poznań, Poland}, pages = {S. 124}, abstract = {Silicon wafer for solar application are produced by multi-wire sawing from 12x12 cm2 silicon bricks. After slicing the wafer surface consists of a several micrometer surface layer of very heterogeneous constitution, the so called saw damage. The topmost layer of the saw damage consists of debris, amorphous silicon and high-pressure silicon phases followed by a very defect-rich and lattice-strained region of fractures, cracks, and rifts caused by the rupture of the silicon lattice during the slicing process [1,2]. Such a damaged surface exhibits very poor semiconductor properties; therefore the saw damage is removed by chemical etching using mixtures of HF, HNO3 and H2SiF6 in order to produce solar cells. Recent investigations showed that the etching of the saw damage is a very heterogeneous process [3-5]. The surface consists of spots at which the etching forms very rapidly deep grooves while other spots remain unetched over a considerably long time. Some of the rapidly formed grooves seem to remain their shape while others grow because of an attack of the side walls, however, without making these grooves significantly deeper. So far there is no explanation for this behavior. The present work is the first study about the locally resolved etching behavior of lattice-strained silicon. The surface of polished single-crystal wafers were scratched with a diamond tip under defined conditions. By means of confocal Raman microscopy the local state of the silicon lattice in and nearby the scratches were characterized in terms of compression and tensile stress with a lateral resolution of 2 μm. Then, the scratches were etched stepwise using HF-HNO3-H2SiF6 mixtures and measured by confocal microscopy to quantify the local removal of silicon and measured by confocal Raman microscopy to monitor the lattice state. For the first time this study reveals and quantifies the impact of tensile and compressive lattice stress on the etch rate of silicon with the major outcome, that stress leads to a significantly anisotropic etching behavior. From the time dependent development of the scratch profiles and the topography of the surrounding wafer areas a detailed picture about the formation of highly reactive species and etching behavior against unstrained silicon is deduced.}, language = {en} } @misc{RietigLangnerAcker, author = {Rietig, Anja and Langner, Thomas and Acker, J{\"o}rg}, title = {A revised model of silicon oxidation during the dissolution of silicon in HF/HNO₃ mixtures}, series = {Physical chemistry, chemical physics}, volume = {21}, journal = {Physical chemistry, chemical physics}, issn = {1463-9076}, doi = {10.1039/c9cp04429a}, pages = {22002 -- 22013}, abstract = {The stoichiometry of wet chemical etching of silicon in concentrated HF/HNO₃ mixtures was investigated. The formation of nitrogen species enriched in the etching mixture and their reactivity during the etching process was studied. The main focus of the investigations was the comprehensive quantification of the gaseous reaction products using mass spectrometry. Whereas previously it could only be speculated that nitrogen was a product, its formation was detected for the first time. The formation of hydrogen, N₂, N₂O and NH₄⁺ showed a dependence on the etching bath volume used, which indicates the formation of nitrogen compounds by side reactions. Simultaneously, the ratio of the nitrogen oxides, NO and NO₂, formed decreases with increasing etching bath volume, while nitric acid consumption increases, so that the formation of NO₂ could also be identified as a side reaction. Based on the stoichiometries obtained, a new reaction scheme for the reduction of nitric acid during etching in HF/HNO₃ mixtures and an electron balance for the oxidation of silicon is presented.}, language = {en} } @misc{LangnerRietigAcker, author = {Langner, Thomas and Rietig, Anja and Acker, J{\"o}rg}, title = {Raman spectroscopic determination of the degree of dissociation of nitric acid in binary and ternary mixtures with HF and H2SiF6}, series = {Journal of Raman Spectroscopy}, volume = {51}, journal = {Journal of Raman Spectroscopy}, number = {2}, doi = {10.1002/jrs.5769}, pages = {366 -- 372}, abstract = {The oxidizing effect of nitric acid in aqueous solutions depends on the concentration of undissociated nitric acid. This makes the concentration of undissociated nitric acid an essential parameter to monitor and control the quality of silicon etching in the industrial manufacturing of solar cells. In the present study, a method known already is extended in such a way that the degree of dissociation of nitric acid can be determined by Raman spectroscopy in HF/HNO3/H2SiF6 acid mixtures over a broad concentration range for the first time and without using an internal or external standard to compensate the typical time-dependent drift of a Raman spectrometer. The method developed requires the calculation of a peak area ratio from the areas of the unimpeded Raman signals assigned to nitrate (νN - O) at 1,048 cm-1 and to undissociated HNO3 (νN - OH) at 957 cm-1. The correlation between the peak ratio and the degree of dissociation of nitric acid revealed can be described by a simple empirical equation. Using this equation, the degree of dissociation of nitric acid can be determined over a broad concentration range in binary and ternary mixtures of HNO3 with HF and H2SiF6. The impact of the acids HF and H2SiF6 and the total water content in the degree of dissociation of nitric acid is discussed.}, language = {en} } @misc{RietigLangnerAcker, author = {Rietig, Anja and Langner, Thomas and Acker, J{\"o}rg}, title = {Advanced insights into the stoichiometry and kinetics of the reaction of silicon in HF/HNO3 and HF/HNO3/H2SiF6 mixtures}, series = {Silicon for the Chemical and Solar Industry XV}, journal = {Silicon for the Chemical and Solar Industry XV}, editor = {Andresen, Birger and Rong, Harry and Tangstad, Merete and Tveit, Halvard and Page, Ingrid}, publisher = {The Norwegian University of Science and Technology}, address = {Trondheim}, isbn = {978-82-997357-9-7}, pages = {145 -- 159}, abstract = {The stoichiometry and kinetics of wet chemical etching of silicon in HF/HNO3 and HF/HNO3/H2SiF6 mixtures was investigated. The side reactions and main reaction pathways were identified by quantifying all reaction products. The relationship between the concentration of undissociated HNO3 and the consumption of HNO3, as well as the formation of H2 as a function of the mixing ratios were found by varying the etching mixture composition systematically. Based on the etching rates determined, kinetic models for the reaction- and diffusion-controlled reaction mechanism are presented as well as the interrelation between the etchant composition and the transition between reaction- and diffusion-controlled etching.}, language = {en} } @misc{SchoenekerlAcker, author = {Sch{\"o}nekerl, Stefan and Acker, J{\"o}rg}, title = {The Kinetics and Stoichiometry of Metal Cation Reduction on Multi-Crystalline Silicon in a Dilute Hydrofluoric Acid Matrix}, series = {Nanomaterials}, volume = {10}, journal = {Nanomaterials}, number = {12}, issn = {1747-681X}, doi = {10.3390/nano10122545}, pages = {34}, abstract = {In this study, the process of metal cation reduction on multi-crystalline silicon in a dilute hydrofluoric acid (HF) matrix is described using Ag(I), Cu(II), Au(III) and Pt(IV). The experimental basis utilized batch tests with various solutions of different metal cation and HF concentrations and multi-crystalline silicon wafers. The metal deposition kinetics and the stoichiometry of metal deposition and silicon dissolution were calculated by means of consecutive sampling and analysis of the solutions. Several reaction mechanisms and reaction steps of the process were discussed by overlaying the results with theoretical considerations. It was deduced that the metal deposition was fastest if the holes formed during metal ion reduction could be transferred to the valence bands of the bulk and surface silicon with hydrogen termination. By contrast, the kinetics were lowest when the redox levels of the metal ion/metal half-cells were weak and the equilibrium potential of the H3O+/H2 half-cells was high. Further minima were identified at the thresholds where H3O+ reduction was inhibited, the valence transfer via valence band mechanism was limited by a Schottky barrier and the dissolution of oxidized silicon was restricted by the activity of the HF species F-, HF2- and H2F3-. The findings of the stoichiometric conditions provided further indications of the involvement of H3O+ and H2O as oxidizing agents in addition to metal ions, and the hydrogen of the surface silicon termination as a reducing agent in addition to the silicon. The H3O+ reduction is the predominant process in dilute metal ion solutions unless it is disabled due to the metal-dependent equilibrium potential of the H3O+/H2 half-cell and the energetic level of the valence bands of the silicon. As silicon is not oxidized up to the oxidation state +IV by the reduction of the metal ions and H3O+, water is suspected of acting as a secondary oxidant. The stoichiometric ratios increased up to a maximum with higher molalities of the metal ions, in the manner of a sigmoidal function. If, owing to the redox level of the metal half-cells and the energetic level of the valence band at the metal-silicon contact, the surface silicon can be oxidized, the hydrogen of the termination is the further reducing agent.}, language = {en} } @misc{HeroldAcker, author = {Herold, Steven and Acker, J{\"o}rg}, title = {Lattice strain enhanced acidic etching on as cut sawn silicon wafer}, series = {Materials Science in Semiconductor Processing}, volume = {123}, journal = {Materials Science in Semiconductor Processing}, issn = {1873-4081}, doi = {10.1016/j.mssp.2020.105575}, pages = {11}, abstract = {The mechanical processing of silicon wafers leads to a heterogeneous lateral strain distribution and various modifications of the silicon, both of which influence the resulting topography after acid etching. In this study we investigate the influence of local strains and the initial topography of slurry and diamond wire saw wafers on the acid etching mechanism. The strain distribution is quantified and qualified by Raman microscopy before and after thermal treatment, while the topography is characterized by confocal microscopy before and after etching. The thermal treatment was used to selectively relax strains and investigate the effect of the individual strains on the etching mechanism. We found that amorphous silicon and compressive strained silicon are mainly present on the top most surface of the saw damage and do not influence the chemical reactivity of acid etching. In contrast, highly reactive tensile strained silicon is found up to 2.7 μm deep in the saw damage and acts as an etching mask. Rapid etching of the tensile strained silicon by HF/HNO3/H2SiF6 leads to the formation of cracks with high local concentrations of intermediate species. These strains induced cracks are etched out together with the original saw damage induced cracks and trenches and form the final surface after etching. Furthermore, we can show how the tensile strain strength must have a relative Raman shift of at least -2 cm-1 to have an effect on the local etch rate. Our data demonstrate how mechanical treatment in combination with thermal treatment and acidic etching can be used to optimize the resulting topography for applications like photovoltaics. In addition, it provides a deeper insight into the acid etching mechanism for non-planar silicon wafers.}, language = {en} } @misc{SchoenekerlAcker, author = {Sch{\"o}nekerl, Stefan and Acker, J{\"o}rg}, title = {The Role of the Molecular Hydrogen Formation in the Process of Metal Ion Reduction on Multi-crystalline Silicon in a Hydrofluoric Acid Matrix}, series = {Nanomaterials}, volume = {11}, journal = {Nanomaterials}, number = {4}, issn = {2079-4991}, doi = {10.3390/nano11040982}, abstract = {Metal deposition on silicon in hydrofluoric acid (HF) solutions is a well-established process for the surface patterning of silicon. The reactions behind this process, especially the formation or the absence of molecular hydrogen (H2) are controversially discussed in the literature. In this study, several batch experiments with Ag+, Cu2+, AuCl4- and PtCl62- in HF matrix and multi-crystalline silicon were performed. The stoichiometric amounts of the metal depositions, the silicon dissolution and the molecular hydrogen formation were determined analytically. Based on these data and theoretical considerations of the valence transfer, four reasons for the formation of H2 could be identified. Firstly, H2 is generated in a consecutive reaction after a monovalent hole transfer (h+) to a Si-Si bond. Secondly, H2 is produced as a result of a monovalent hole transfer to the Si-H bonds. Thirdly, H2 occurs if Si-Si back bonds of the hydrogen-terminated silicon are attacked by Cu2+ reduction resulting in the intermediate species HSiF3, which is further degraded to H2 and SiF62-. The fourth H2-forming reaction is the reduction of oxonium ions (H3O+) on the silver/, copper/ and gold/silicon contacts via monovalent hole transfer to silicon. In the case of (cumulative) even-numbered valence transfers to silicon, no H2 is produced. The formation of H2 also fails to appear if the equilibrium potential of the 2H3O+/H2 half-cell does not reach the energetic level of the valence bands of the bulk or hydrogen-terminated silicon. Non-hydrogen-forming reactions in silver, copper and gold deposition always occur with at least one H2-forming process. The PtCl62- reduction to Pt proceeds exclusively via even-numbered valence transfers to silicon. This also applies to the reaction of H3O+ at the platinum/silicon contact. Consequently, no H2 is formed during platinum deposition.}, language = {en} } @misc{RietigGrafeAcker, author = {Rietig, Anja and Grafe, Hans-Joachim and Acker, J{\"o}rg}, title = {New insights into boron species in acidic digestion solutions of boron-doped silicon}, series = {Journal of Analytical Atomic Spectrometry}, volume = {36}, journal = {Journal of Analytical Atomic Spectrometry}, number = {11}, issn = {1364-5544}, pages = {2492 -- 2500}, abstract = {The exact and precise determination of the boron concentration in silicon is still a challenge. A systematic investigation dealing with the digestions of 60 silicon samples with HF-HNO3 and subsequent boron determination by ICP-OES revealed that the concentration found could be up to 60\% lower than the actual boron concentration depending on the composition of the sample solution. As the original boron-silicon compound that was identified was colloidally precipitated in the presence of an excess of hydrofluoric acid and then partially retained by filtration or by the sample introduction system, systematic lower boron concentrations were determined. In acidic, HF-free digestion solutions, this compound existed in a soluble form parallel to the borate in B(OH)4-. In an excess of hydrofluoric acid, the compound was converted into the colloidal form and, in parallel, B(OH)4- was converted to tetrafluoroborate, BF4-. For the composition of the colloidal compound, a molar ratio of boron to silicon of 1 : 4 could be determined. 11B-ss-NMR analysis revealed a tetrahedral geometry compound with a central boron atom surrounded by four silicon atoms. It is assumed that a soluble form with four -Si(OH)3 groups was present in the hydrofluoric acid-free solutions, while an insoluble form with four -SiF3 groups was present in HF-containing solution.}, language = {en} } @misc{RietigLangnerAcker, author = {Rietig, Anja and Langner, Thomas and Acker, J{\"o}rg}, title = {Comprehensive stoichiometric studies on the reaction of silicon in HF/HNO3 and HF/HNO3/H2SiF6 mixtures}, series = {Physical chemistry, chemical physics}, volume = {24}, journal = {Physical chemistry, chemical physics}, number = {5}, issn = {1463-9076}, doi = {10.1039/d1cp05418j}, pages = {3094 -- 3108}, language = {en} }