Search

Thermal decomposition experiments with monosilane diluted in hydrogen have been conducted in a free-space reactor with an extendable reaction zone, allowing for easy variation of residence time. Reactor effluent was analyzed by gas-chromatography combined with mass-spectrometry (GC-MS). The applied analysis technique enables detection of silanes with up nine silicon atoms, as well as absolute quantification of the concentrations of mono-, di-, and trisilane. For all the detected silanes, our gas analyses show a peak in reactor outlet concentration as function of temperature whose position and shape depend on the number of silicon atoms (nSi) contained in the silane species. The peak width decreases and the peak position shifts to higher temperatures with increasing nSi. At increased residence time, the concentration peak shifts to lower temperatures and the SiH4 consumption rate increases. This is consistent with the expected behavior for a system described by Arrhenius kinetics. The maximum outlet concentrations of all the measured silanes decrease with increasing residence time. However, the dependence of silane concentrations on temperature and residence time is not trivial: At a fixed temperature the measured outlet concentrations will increase with increasing residence time in some temperature regions and decrease with residence time in other temperature regions. By mapping outlet concentrations as function of temperature and residence time we attempt to decouple the effect of these two parameters and to untangle their effect from that of reactor geometry and operation.

SUMMARY Using an industry-relevant method of production we have demonstrated production of doped silicon nanoparticles for Li-ion batteries. The particles have been characterized and tested in Li-ion battery half-cells for demonstration of the performance as Li ion storage material in the anode. In this presentation we will show the results regarding the characterization of the particles as well as the performance results of the material in Li-ion half cells. BACKGROUND The use of silicon as part of the anode in Li-ion batteries enhances the potential storage capacity greatly. While the graphite anode used in standard Li-ion batteries has a theoretical capacity of 372 mAh/g, the theoretical capacity for Si is 3,572 mAh/g [1]. However, to be able to absorb the sufficient amount of Li ions reversibly, the silicon material needs to be nanostructured [2]. EXPERIMENTAL IFE has earlier demonstrated production of silicon particles prepared using a silane-based free-space reactor, which allows control of particle size distribution and crystallinity and a high production capacity [3]. In this work, the setup was modified to allow for introduction of phosphine gas in the production process. The intention was to introduce phosphorous dopant atoms to modify the electrical conductivity of the silicon material. The current particles were produced with a silane flow rate of 5 slm, and 5 slm of 3% phosphine diluted in He. Thus, the P concentration in the process gases was 3% with respect to Si. The production rate at 100% yield would be about 400 g/h, while the product recovered from the process typically was between 100 and 200 g/h. The processing temperature was varied from 475 °C to 600 °C to produce particles with different properties. All electrochemical tests were performed in a crimped 2032 coin cell with lithium metal used as a counter electrode, with a polymer separator (Celgard 3401) and 1 M LiPF6 in 1:1 EC/DMC electrolyte (LP30, BASF). For some experiments, 10 wt% fluoroethylene carbonate (FEC) was used as electrolyte additive. The cells were cycled at 25 °C between 0.05 and 1.0 V with a constant C-rate of C/10 (after initial 3 cycles at C/20) using an Arbin Battery cycler (Arbin Instruments). RESULTS AND CONCLUSIONS The phosphorus content of the particles was by inductively coupled plasma mass spectrometry (ICP-MS) and energy-dispersive x-ray spectroscopy (EDS) measured to be in the range of 1.7-1.8% with respect to silicon. Thus, it seems that the utilization of the phosphine is not complete, or phosphorous binds to materials which are not recovered from the production process. The particle size distribution was centered around 600 nm as measured by laser scattering in a Malvern Mastersizer 2000. SEM and TEM revealed that the particles consisted of aggregates of spherical silicon particles with a primary particle size of 50-400 nm. X-ray diffraction showed that we could produce amorphous or crystalline particles depending on the temperature of the process. Cycling of the particles in lithium battery half cells showed a life time of about 300 cycles for doped and undoped particles. Some measurements indicate a better conductivity of the doped material but the results were not conclusive. The cells were cycled with a limited capacity regime where the capacity was set to 1000 mAh/g(Si) and thus the end-voltage could vary. The most pronounced difference in the cycling performance between doped and undoped silicon particles was a shift in the increase in end voltage for the doped particles: While end-voltage increased from 0.4 V to 0.6 V over the first 25 cycles, this development is stretched out over the first 50 cycles or more for doped particles. See the attached figure for further details. References [1] D. Larcher, S. Beattie, M. Morcrette, K. Edström, J.-C. Jumas, and J.-M. Tarascon, “Recent findings and prospects in the field of pure metals as negative electrodes for Li-ion batteries,” J. Mater. Chem., vol. 17, no. 36, p. 3759, 2007. [2] H. Kim, M. Seo, M. H. Park, and J. Cho, “A critical size of silicon nano-anodes for lithium rechargeable batteries,” Angew. Chemie - Int. Ed., vol. 49, no. 12, pp. 2146–2149, 2010. [3] H. F. Andersen, W. O. Filtvedt, J. P. Mæhlen, T. Mongstad, M. Kirkengen, and A. Holt, “Production of Silicon Particles for High-Capacity Anode Material, Yielding Outstanding Production Capacity,” ECS Trans., vol. 62, no. 1, pp. 97–105, 2014. Figure 1

Today's most frequently used production path for polysilicon and solar grade silicon (SG-Si) is based on thermal decomposition of trichlorosilane (SiHCl 3 ) in a Siemens-type reactor. Due to the batch-wise mode of its operation and its requirement for heating and cooling, the Siemens technology is an energy demanding procedure. As an alternative to this process, SG-Si can be produced through the thermal decomposition of monosilane (SiH 4 ), for example in a fluidized bed reactor or a centrifugal chemical vapour deposition reactor. In order to avoid production of fine particulates and related clogging, this process requires knowledge of the nucleation properties of silane. In this work, the critical concentration for particle nucleation from SiH 4 , diluted in H 2 has been determined for temperatures in the range 400 - 700 °C. For this purpose, we used a free space reactor and an optical particle detector of own design for in-situ monitoring of the reactor exhaust. In a plot of logarithmic critical nucleation concentration against inverse temperature, an almost linear relation was found for temperatures below approximately 500 °C and above approximately 600 °C. Between 500 and 600 °C, a less temperature dependent behavior is observed. Apparent activation energies are estimated to be 32 kcal/mol for temperatures below 500 °C, 8 kcal/mol for temperatures between 500 and 600 °C and 23 kcal/mol for temperatures above 600 °C. The first and last of these values agree well with existing literature. The low temperature dependence in the range 500 °C to 600 °C, however represents a deviation from earlier studies. Investigations of reactor exhaust by GC-MS (gas chromatography - mass spectrometry), allowing detection of the concentrations of higher order silanes (disilane (Si 2 H 6 ), trisilane (Si 3 H 8 ), etc) as function of reactor temperature, suggest that the low temperature dependence of the nucleation concentration for some conditions is related to an increase of the concentration of Si 2 H 6 and Si 3 H 8 at the same temperature range.

For an installed silicon-based solar cell panel, about 40% of the energy needed for fabrication is consumed for production of the silicon feedstock. Reducing the energy consumption in this step is therefore crucial in order to minimize the energy payback time and cost of the technology. The most promising alternative to the conventional methods is to use fluidized bed reactors. In this article, we report the results from a novel reactor layout with a selectively cooled distribution arrangement. Important aspects of fludized bed monitoring and operation are described. Two different operation regimes are stated yielding hence porous and dense growth. Further, the method of encapsulating fines by means of more dense depositions is verified, and the nature of the scavenged material is characterized. Also, two different types of fines formation are identified and accounted for.

For an installed silicon based solar cell panel, about 40% of the energy costs involved in the production of the panels can be attributed to the production of the silicon feedstock itself (poly production and crystal growth). Hence reducing the energy consumption in these steps is crucial in order to minimize the energy payback time of installed capacity. For the first step, viz., the poly production, the most promising cost reduction alternative is the fluidized bed reactors (FBR) using silane as a precursor rather than trichlorosilane (TCS) since for TCS the reverse reactions makes the theoretical trichlorosilane conversion substantially lower. Use of silane has, however, many challenges and scaleup to larger capacity can be achieved if associated risks are properly dealt with. This paper outlines some of these challenges and provides a detailed survey of the current status on the growth mechanism and kinetics of silane pyrolysis. The paper also provides a summary of modeling of fluidized bed reactors (FBR) in some depth and give some empirical insight to key aspects in FBR scaleup design.

For an installed silicon based solar cell panel, about 40% of the energy consumed producing the cell was used during production of the silicon feedstock. Reducing the energy consumption in this step is therefore crucial in order to minimize the energy payback time of installed capacity. The most promising alternative to the conventional methods is fluidized bed reactors (FBR). The development of FBR for the production of polysilicon is here reviewed. Through the last half century several research groups and companies have contributed to the development of such reactors to a point where the method is competitive on the terms given in today's marked. Most work within the field of research has been done by the industry and this has resulted in few articles and a substantial amount of patents on the topic. This article aims to summarize the development and to paint the complete picture on what concepts have been investigated, what challenges arose and where the development stands today.

The article presents a novel design for a distribution plate. The solution is suitable for a reactor vessel where a reactant gas needs to be maintained at a different temperature from the reaction chamber in order to avoid unwanted occurrences, such as clogging of the distribution plate. A normal procedure involves cooling of the distribution plate which is reported to either increase heat loss substantially or yield insufficient temperature in parts of the reaction chamber. The problem is especially important for reactors where the difference in reactant inlet temperature and desired reaction temperature is large. The investigated design utilized materials of very different thermal conductivity to only cool specific parts of the distribution arrangement and thereby minimize heat loss. Our system is a distribution plate for use in a fluidized bed reactor for silane pyrolysis. However, the solution is general and may be utilized in many types of vessels and chemical reactors.

Silicon is proven to have a great potential as an anode material in lithium-ion batteries due to its high theoretical electrochemical capacity. The standard, commercial graphite anode has a theoretical capacity of less than 400 mAh/g, whereas the silicon anode can potentially deliver a tenfold capacity as a result of multiple Li-ion incorporation in the structure. However, silicon anodes deteriorate quickly during cyclic charging and discharging, rendering them useless in only a few cycles. This has been attributed to stresses induced by the large volume change of the material during cycling. The methods explored in order to overcome these problems such as using lithography, advanced nanotechnology, incorporation of silicon in carbon nanotubes or similar methods are too slow and too expensive for commercial use. This work presents results from using a silane-based decomposition reactor in order to produce silicon particles with a suitable nanostructure for use in lithium-ion battery anodes. The silane gas is decomposed in a controlled environment at a temperature of 500-600 °C. The current pilot reactor has demonstrated production of up to 350 g/hour in an easily up-scalable lab version. Particles of diameter ranging from 50 nm and up to 500 nm have been produced with relatively narrow size distribution. This method may produce both amorphous and crystalline particles and the surface of the particles can be terminated by hydrogen or other elements if desired. The silicon particles were mixed with an organic binder in an aqueous slurry and coated on a Cu-foil, The electrochemical performance was tested with CR-2032 coin cells. In the course of the presented work studies of cyclic voltammetry, cycling stability (Figure 1), voltage profiles and electrochemical impedance were performed. Besides electrochemical methods, SEM (Figure 2), XRD, ICP-MS and particle size distribution measurements were implemented. The silicon particles achieved a high capacity, relatively good stability, as well as a high yield and production capacity. Further developments on the silicon particles, such as doping of Si and in-line surface coating, are feasible.