Electric arc | EPFL Graph Search

An electric arc (or arc discharge) is an electrical breakdown of a gas that produces a prolonged electrical discharge. The current through a normally nonconductive medium such as air produces a plasma, which may produce visible light. An arc discharge is initiated either by thermionic emission or by field emission. After initiation, the arc relies on thermionic emission of electrons from the electrodes supporting the arc. An arc discharge is characterized by a lower voltage than a glow discharge. An archaic term is voltaic arc, as used in the phrase "voltaic arc lamp". Techniques for arc suppression can be used to reduce the duration or likelihood of arc formation. In the late 19th century, electric arc lighting was in wide use for public lighting. Some low-pressure electric arcs are used in many applications. For example, fluorescent tubes, mercury, sodium, and metal-halide lamps are used for lighting; xenon arc lamps have been used for movie projectors. Electric arcs can be uti

On the Powder Formation in the Plasma Enhanced Chemical Vapor Deposition Process for the Deposition of SiOx Barrier Coatings

Marina Ricci

Nowadays thin films play an important role in everyday life and in industries. Thin film technology has been developed primarily for the growing demand for development of smaller and smaller devices that require advanced materials and new processing techniques suitable for micro and nano technology. One of the methods of producing thin films is by means of plasma technology. This is a technique by which plasma, also known as 'the fourth state of matter', is used to generate ions, electrons and radicals which deposit at a surface where they will form a thin film. The source of these depositing species could be either gaseous, liquid or solid in origin. Most of the time a plasma is generated by means of an electrical discharge. Plasma technologies are used for various applications. One of these is Plasma Enhanced Chemical Vapor Deposition (PECVD). PECVD systems have widespread applications in the electronic sector, because of their exibility in depositing different films such as silicon oxide thin films (SiOx). These films, currently deposited by organosilicon compounds, such as hexamethyldisiloxane (HMDSO), have dielectric and good barrier properties. The aim of the work at CRPP consists in optimizing PECVD processes for the deposition of these films on polymers. In this study, the species in gaseous phase and the powder produced in oxygen-diluted HMDSO plasmas were experimentally characterized in a radiofrequency (RF) capacitively-coupled reactor at 13.56 MHz. The gaseous phase of these plasmas and the particle formation were studied by in-situ infrared absorption spectroscopy and optical emission spectroscopy. The study of the powder formation is important because the particle deposition imposes an upper limit on the deposition rate. Particle formation was firstly studied as a function of HMDSO/O2 ratio. Analysis of the time evolution of the infrared absorption spectra gives the development of the size, the number density, the structure and the particle composition. At high oxygen dilution the formation of large particles is observed. The analysis of HMDSO/O2 plasma at low O2 content suggests that at the beginning of the process very small particles are formed by polymerization; the same particles do not grow much and keep sizes around 50 nm. The admixture of inert or non-reactant gases, such as Ar or N2, into the plasma promotes polymerization of the monomer; a high value of one of these inert or non-reactant gases leads to small powder particle size, since only polymerization takes place in this case. Particle size around 50 nm is also found in this case. At the same time we see a decrease of the carbon content and a weak reduction of the stoichiometry in SiOx indicated by the shift of the SiOSi asymmetric stretching vibration. This reduction also influences the deposited layers, less oxidized and, thus, quite far from a stoichiometric character. Hexamethyldisilazane (HMDSN) and tetramethylsilane (TMS) are other reagents that have been identified as alternatives to HMDSO. The in-situ infrared absorption spectra show that HMDSO/O2 and HMDSN/O2 plasmas are quite similar, but in the second case a lower SiOSi stretching peak intensity, due to NH and SiN peaks, is seen. TMS/O2 and HMDSN/O2 plasma spectra show a low SiOSi bending and stretching peak intensity, due also to the oxygen atom absence in their chemical structure. Two separate chapters are respectively devoted to the silicon nitride thin films (SiNx) and plasma crystal formation. SiNx layers are deposited by three different organosilicon compounds, such as HMDSN, TMS and bis(dimethylamino)dimethylsilane (BDMADMS). No powder formation is visualized in this contest. The ex-situ infrared absorption spectra show the difficulty to produce stable SiNx films; they present a low SiN peak intensity and are quite hygroscopic. Furthermore, in HMDSO/O2 plasmas at high oxygen dilution and high pressure, later in time into the discharge, accretion leads to larger and larger particles which can finish up in particles forming a plasma crystal. The study of RF harmonics shows the presence of instabilities, clear indications of the passage from small particulate to plasma crystal structure.

EPFL2011

Characterization and Fatigue of the Converse Piezoelectric Effect in PZT Films for MEMS Applications

Davide Balma, Nachiappan Chidambaram, Ramin Matloub Aghdam, Andrea Mazzalai, Paul Muralt

A measurement setup for the detailed study of the transverse piezoelectric coefficient e(31,f) in the converse (actuator) mode was developed. It allows the assessment of the piezoelectric stress in thin films on silicon cantilevers and provides for a correlation of this stress with large and small signal responses to ferroelectric polarization and dielectric response, both as a function of slowly sweeping electric field. This test is important for the understanding of piezoelectric thin films in microelectromechanical systems. The method is illustrated at hand of sol-gel lead-zirconate-titanate (PZT) thin films, and verified also with AlN and AlN-ScN alloy thin films. A 1-mu m thick, sol-gel derived PZT(53/47) gradient-free sample showed a response of -18.3 C/m(2) at 100-kV/cm electric field. Reliability tests of PZT thin films were carried out with the same setup in an accelerated manner. The piezoelectric activity did not degrade significantly up to 109 unipolar pulses at 100 kHz with an amplitude of -150 kV/cm. The increase in leakage toward the end of the cycles was explained by a thermal runaway effect.

Institute of Electrical and Electronics Engineers2015

On the Measurement and Calculation of Horizontal Electric Fields From Lightning

Marcos Rubinstein, Abdolhamid Shoory

In this paper, we discuss two issues related to the measurement and calculation of the horizontal electric fields from lightning. On the one hand, there is an inherent difficulty inmeasuring the horizontal electric field component from lightning because of the overshadowing effect of the vertical electric field component which, depending on the distance to the lightning channel, the ground conductivity, and the height of the observation point can be one to two orders of magnitude larger than the horizontal electric field component. Consequently, even a small tilt of the measuring antenna would result in a noticeable contamination of the measured horizontal waveform. This may explain the fact that data on horizontal electric fields are very scarce. Numerical simulations show that, for a ground conductivity of 0.0025 S/m, the resulting error for a one-degree sensor tilt in the field peak is about 20% for distances ranging from 60 to 500 m. For strikes to a 100-m-tall tower, the resulting errors are found to be slightly smaller (about 10% to 15% for the first peak). The second issue dealt with in this paper is the computationmethods of the horizontal electric field. In this regard, some authors have emphasized the importance of taking into account the so-called conduction current in the calculation of nearby horizontal electric fields. We show in this paper that the conducted contribution is automatically taken into account when using the general solutions of Maxwell's equations or obtained by exact numerical simulations. In this case, there is no need to consider separately any other contributions because the solution yields the total horizontal electric field, taking into account both the radiation from the channel and the current flowing into the ground. However, a "conducted contribution" needs indeed to be taken into account separately when high-frequency approximate solutions are used for the evaluation of the electric field. At very close distances to the lightning strike location, the current distribution in the ground may be highly nonuniform because of surface arcing and plasma channel formations. Given the random nature of these phenomena, it is virtually impossible to gain detailed knowledge of the current distribution and, hence, to evaluate the resulting horizontal electric field near the strike point. Based on the results and discussion presented in this paper, we recommend taking special care when measuring the horizontal electric field from lightning to minimize the contaminating effect of the vertical electric field. It is important that both components (vertical and horizontal) be measured simultaneously to evaluate possible contamination of the horizontal field.