Design of a Plasma Actuator Based on Dielectric Barrier Discharge (DBD)

Proyecto Fin de Carrera Ingeniería de Telecomunicación Formato de Publicación de la Escuela Técnica Superior de Ingeniería Autor: F. Javier Payán Somet Tutor: Juan José Murillo Fuentes Dep. Teoría de la Señal y Comunicaciones Escuela Técnica Superior de Ingeniería Universidad de Sevilla Sevilla, 2013 Master’s Thesis MsC Degree in Aeronautical Engineering Design of a Plasma Actuator Based on Dielectric Barrier Discharge (DBD) Author: MªScherezade Balsera Barquero Tutors: José Cotrino Bautista, Chair in Physiscs Ana María Gómez Ramírez, PhD in Physics Antonio Franco Espín, PhD in Aeronautical Engineering Dept. Aerospace Engineering and Fluid Mechanics Higher Technical School of Engineering University of Seville Sevilla, 2018 Master’s Thesis MsC Degree in Aeronautical Engineering Design of a Plasma Actuator Based on Dielectric Barrier Discharge (DBD) Author: MªScherezade Balsera Barquero Tutors: José Cotrino Bautista, Chair in Physiscs Ana María Gómez Ramírez, PhD in Physics Antonio Franco Espín, PhD in Aeronautical Engineering Dept. Aerospace Engineering and Fluid Mechanics Higher Technical School of Engineering University of Seville Sevilla, 2018 Master’s Thesis: Design of a Plasma Actuator Based on Dielectric Barrier Discharge (DBD) Author: MªScherezade Balsera Barquero Tutors: José Cotrino Bautista, Chair in Physiscs Ana María Gómez Ramírez, PhD in Physics Antonio Franco Espín, PhD in Aeronautical Engineering The tribunal appointed to judge the work indicated above, which is composed of the following professors: President: Vocal/s: Secretary: agree to grant her the qualification of: The secretary of the tribunal Date: Acknowledgement First of all, thanks to the Surface Nanotechnology Area of the Isla de la Cartuja Scientific Research Center for having allowed me to use its facilities and equipment for the development of this Master’s Thesis, and thanks to all its researchers I met because I have enjoyed a lot contemplating the work ambient of research world. Special thanks to the Professor José Cotrino Bautista, for accepting being my tutor almost a year ago when I went to the Faculty of Physics to ask him if I could develop a project with plasma, transmitting his knowledge with a facilitating role and proactive attitude; and of course, to Paula de Navascués Garvín, for her support; I learned a lot from you, Paula. Thank you very much to you both. I have been enthusiasted with every step that the work went through. Two years ago I wrote the acknowledgments of my Final Bachelor’s Degree Project from this same table, excited for a new stage, but also aware of the effort that a master’s degree of this size would require; and at that moment, I promised myself that during those two years I would squeeze the time to the maximum, and at the same time that I am moving a little further towards my dream of building spaceship engines, I was not going to abandon any of the other aspects of life that are also part of what I am. Today, proud, I can say now that during the two years that this Master’s Degree has taken –that brought me to new classrooms, new streets and new people– that I achieved it. So, thanks for the engineering knowledge and all the opportunities for personal and professional growth provided and taken, for having felt part of this university, for having known great professionals (both proffessors and colleagues), and because these years have led me to have better criteria and to feel more confident with my ability to work. Again, I have enjoyed a lot learning and doing. It was an honor for me to have been able to make the speeches for both Aerospace Engineering and Aeronautical Engineering, because they are my sincere thanks to life and my reconciliation with all the futures that come, came or will come. To Sevilla, for all the hearts and minds that it has given me the opportunity to meet, really thanks; from the full days in Cartuja to the nightly promenade along the bank of the Guadalquivir, from which we have built years with only hours rediscovering ourselves. Thank you, Luis, for your support ’to the vacuum and back’; always I will have a memories-stop in my Elves Place. Thanks to my family, for a support and motivation full of love; to her, our angel, to my four “frústrate” siblings, Almudena, Carolina, Inés y Pedro Antonio, to my grandmother Inés and, without doubt, to my mother, Toni. Absolutamente gracias, mamá, como siempre. Good luck, Scherezade. I Abstract Related with recent investigations, plasma actuator or plasma controller has been the most promising flow control device, and specifically the Dielectric Barrier Discharge flow controller (usually called DBD plasma actuator). The popularity of these devices comes from the advantages offered by a lower weight and size, absence of moving parts that would mean, for example, less drag and possible reduction in maintenance. The experimental study carried out during the last months has been exposed in this document, through which a a large number of different devices have been built, seeking to identify which factors contribute to a maximization of the efficiency in terms of different parameters, of plasma generation devices through Dielectric Barrier Discharge (DBD). For this purpose, the operation and performance of the different configurations of the DBD plasma actuator were carried out, having into account frequency, pressure, substrate materials and geometrical configuration of the electrodes; and they have been characterized and quantified from a fundamentally electrical point of view. In addition, optical spectroscopic analyzes of the plasma generated were done. The use of ferroelectric materials such as lithium niobate allows to provide new approaches to traditional studies. Also, it has been able to successfully compare the conclusions derived from this work with other published studies. Keywords DielectricBarrierDischarge,Plasma,Actuator,Capacitance,microplasma III XNotation εPermittivity φPhase difference PPower Xbd Referred to “breakdown” XeReferred to electrons XiReferred to ions XnReferred to neutral particles εrRelative Permittivity or Dielectric constant RResistance γes Secondary electron emission coefficient in the cathode SSurface/A of the dielectric tTime/Thickness ε0Vacuum permittivity VVoltage λWavelength 1 Introduction Laws of physics never stop surprising us as we go deeper into their knowledge, while it offers us new and fascinating challenges to embark on after each advance, especially to scientists and engineers. Within these innumerable fascinating fields of study, both technological and purely scientific, to which physics open its doors, lies the generation of artificial plasmas and all the practical possibilities that their properties allow. Thus, this work will focus on taking advantage of certain properties to incorporate them into aeronautical engineering, specifically the control of aerodynamic flow; regarding which it is remarkable the number of investigations and relevant developments have taken place in recent years. Related with those investigations, plasma actuator or plasma controller has been the most promising flow control device, and in particular the Dielectric Barrier Discharge flow controller (usually called DBD plasma actuator). 1.1 Objective As a theoretical concept, the DBD plasma actuator stands out as a good alternative to, for example, the use of mobile parts in the wing for the control of surrounding flow; however, it is not yet used as an application for real use. That is why, despite the fact that there are numerous publications about it, it is still necessary to continue delving into the analysis of its performance, effectiveness and efficiency; since there are still limits to improve and barriers to overcome. This has been the reason that has led to this master’s thesis to choose this line of research as a topic, pretending to contribute to the advances in this field. The questions that are asked must be answered in a more precise and systematic way, for which different paths can be followed. As can be seen from the reference literature, the importance of offering systematic and quantitative answers regarding the operation of plasma actuators arises from the existing need that requires objective parameters that allow comparing the different designs that are studied between the different groups of research, alluding to the advantages that could be derived from the use of specific dimensionless parameters for the DBD actuator, since without them the comparison between the different published studies becomes more complex. In turn, the advance 1 2Chapter 1. Introduction in numerical models on plasmas so that the simulations are more detailed and realistic, while considering the operational conditions of the DBD flow controllers, represents an important line of research in which to advance. Throughout this work, the characterization and quantification of the operation and yield of the DBD controller is examined in detail, through the electrical and spectroscopic magnitudes, thus employing a macroscopic and optical vision respectively of the concept as a tool to quantify its performance. The experimental study has been developed by varying the frequency of the discharge in a range of low frequencies (1-15 kHz): the type of dielectric material used as a substrate or barrier, the geometrical configuration of the electrodes (specially to study the influence of the gap between active and encapsulated electrode) and the pressure, having experimented with near than twenty different devices that have been manufactured for that purpose. So, summarizing, these are the objectives of this project: 1. Review of existing academic literature on this issue. 2. Learn how to develop a DBD actuator. 3. Build and test different DBD actuator designs varying electrodes geometry, dielectric material, frequency and pressure. 4. Use different programs to analyze the voltage and current signals of the generated DBD plasma. All measurements will be saved in the voltage value in which it is noticeable to the naked eye the appearance of plasma in the devices. This point will be called interchangeably "breakdown" value or ignition value of the plasma. It is important to note that in this document when referring to the concept of " efficiency ", this is related to the voltage necessary to achieve the ignition of the plasma; the more efficient the device will be, the lower the voltage it needs to produce plasma. 1.2 Methodology After presenting a broad study about the theoretical basis that support this experimental project, this document presents everything related to manufacturing and experimentation with different devices, including reasons, results and conclusions about each step, in order to give response to the claims presented in Section 1.1. As explained in the previous section, in this work one can find both electrical and spectroscopic analysis of these DBD plasma actuators, and the study has been developed according to the frequency of the input signal (in a range from 1 kHz to 15 kHz), the pressure of the ambient (from ambient conditions to values of reduced pressure of 20 mbar), the type of substrate or barrier dielectric material used in the device (quartz, alumina and lithium niobate) and the geometrical configuration of its electrodes (with special attention in the influence of gap). To achieve the proposed objectives, the experiments have been framed within four different phases of study, in which particular approaches are had into account.These phases and their purposes are the following: 1.2 Methodology 3 •Phase 1 : Basis functional study of DBD actuator with the aim of generating the largest possible amount of surface covered with plasma In this first phase it has been studied, mainly, the efficiency of different configurations of a DBD Actuator with, as explained, the aim of generating the largest possible amount of surface covered with plasma, on the dielectric layer. For this purpose, reactors with three different electrode configurations (A, B y C) have been designed, manufactured and powered, and in addition, the effect on two different dielectric materials has been checked: quartz (SiO 2 ) and lithium niobate (LiNbO 3 ). At the end, 4 devices in total have been developed and tested: Experiments 1 to 4 –see Table 3.1–. •Phase 2: Influence of dimension of gap To this goal, reactors that have been fed they share the same configuration regarding: the shape, number and function of the electrodes used, composed of an active electrode with a narrow band on the upper face and another with the same characteristics on the lower face but in this case connected to ground, that is, Type D configuration. This configuration of electrodes has been mounted on 3 different types of substrates, thus having 3 different device species, to which the size of the gap between both electrodes has been varied; for each device, 4 measurements have been made according to the gap: 5 mm, 2,5 mm, 1 mm and canceling the gap. The substrates used were: quartz (SiO 2 ), alumina (Al2O 3 ) and lithium niobate (LiNbO 3 ). In total, 12 different devices have been built and tested: Experiments 5 to 16 –see Table 3.1–. •Phase 3: Active muti-electrode configurations Phase 3 is presented as an evolution of phase 2. In it, active multi-electrode configurations are studied, as a result of a series of type D configurations adjacent to each other, it means, alternating conductive and insulating bands. At the end, 3 devices in total have been developed and tested: Experiments 17 to 19 –see Table 3.1–. •Phase 4: Experimentation with airflow and at reduced pressure All the experiments from Phase 1 to 3 have been tested in atmospheric pressure and calm atmosphere (laboratory conditions), except finally in Phase 4, in which the multi-band device on lithium niobate had been, on the one hand faced with an air flow and on the other hand tested at reduced pressure. Experiments 20 and 21, respectively –see Table 3.1–. The fact of sometimes maintaining the same geometric structure and varying the substrate material, as well as, conversely, maintaining the same barrier material and studying different electrode configurations, makes possible to establish comparisons between the different results. The chosen mode of comparative analysis is based on electrical and spectroscopic characterization, for which the necessary equipment circuits are assembled and explained in the Section 3.4. The electrical characterization will be carried out from measurements of current and voltage as a function of time obtained for each of the experiments tested as a function of frequency. Both signals have been analyzed, and from which the rest of the electrical magnitudes have been calculated and the appropriate graphs have been constructed. 4Chapter 1. Introduction Regarding the spectroscopic characterization, it has been done by using a monochromator to, in this way, collect the optical emission spectrum of the air plasma excited by the surface DBD discharge; and the objective is to identify the composition of the plasma generated. Respect to the way to expose the results, they have been collected in the same chapter (Chapter 4), and explained differentiating by operational parameter of study. Finally, in Chapter 5, about conclusions, the results obtained during this master‘s thesis are compared to other published studies (or completing conclusions with them), as well as they have been related to Paschen representations. Excel Spreadsheets, Matlab and OriginPro have been used to process and visualize the data. 1.3 Structure of the document This document is divided into five chapters and three appendices, whose content is briefly explained in the following enumeration: •Chapter 1: Introduction Include objectives –Section 1.1–, methodology –Section 1.2– and structure of the document –Section 1.3–. •Chapter 2: Theoretical basis In this chapter, as an introduction, a summary of the main characteristics that define, firstly, the plasma state will be presented –Section 2.1–, to later describe the operation of an aerodynamic flow controller through Dielectric Barrier Discharge –Section 2.2– and, finally, specify what are the general terms of the electrical and spectroscopic analysis that will be carried out –Section 2.3–. With this knowledge base, it is tried to familiarize ourselves with the classifications of different types of DBD plasma controller, and thus be able to understand their different applications based on quantifiable parameters. •Chapter 3: Design and Construction of DBD plasma controllers In this chapter, specifically in Section 3.1, are explained all the study phases –and consequently, its devices– and the reasons why the different configurations were carried out; that is, the design and construction of the DBD Plasma Actuator. Each one of these devices used will be presented in order to characterize them to understand the differences between the data obtained from their analysis, which will be shown in Chapter 4. Also, in Section 3.2 and 3.3 a series of considerations regarding the manufacturing of the devices and about the experimentation are collected respectively; and, finally the experimental procedure and its necessary resources and considerations will be shown in Section 3.4, including electrical and spectroscopic procedures and equipment. •Chapter 4: Analysis and Results In this chapter the results obtained after carrying out all the experiments and their respective analysis are presented, from Phase 1 to 4, including comparisons according to the substrate material, the geometric 1.3 Structure of the document 5 configuration of the devices and influence of pressure and frequency. The results are exposed differentiating by operational parameter of study. •Chapter 5: Conclusions In this last chapter are presented the conclusions of this work according to the objectives presented in Section 1.1: influence of dielectric material, geometrical electrode configuration, pressure and frequency in the plasma ignition, and conclusions about the spectroscopic analysis. Appendices Finally, appendices A, B and C, respectively, include representations in reference to Paschen’s curves, a summary of some interesting data from others SDBD studies and possible lines with which to continue this investigation. 2 Theoretical Basis In this chapter, a summary of the main characteristics that define, the plasma state will be presented –Section 2.1–, to later describe the operation of an aerodynamic flow controller through Dielectric Barrier Discharge –Section 2.2– and, finally, specify what are the general terms of the electrical and spectroscopic analysis that will be carried out –Section 2.3–. With this knowledge base, it is tried to familiarize ourselves with the classifications of different types of DBD plasma controller, and thus be able to understand their different applications based on quantifiable parameters such as frequency, gap or pressure. 2.1 Plasma physics Physics has taught us that gases containing electrically charged particles offer an excellent way of manipulating and transforming materials in a controlled manner. Plasmas are the object of study of the academic field of Plasma Science or Plasma Physics, including sub-disciplines such as the following fields of active research: • Plasma theory (Plasma equilibria and stability, Plasma interactions with waves and beams, Guiding center, Adiabatic invariant, Debye sheath, Coulomb collision, ...), • Plasma in nature (Astrophysical plasma, Northern and southern (polar) lights, The Earth’s ionosphere, Interplanetary medium, Planetary magnetospheres, Space plasma, ...), • Industrial plasmas (Plasma chemistry, Plasma processing, Plasma spray, Plasma display, Plasma sources, Dusty plasmas, ...), • Plasma diagnostics (Thomson scattering, Langmuir probe, Ball-pen probe, Faraday cup, Spectroscopy, Interferometry, Ionospheric heating, Incoherent scatter radar, ...) or 7 8Chapter 2. Theoretical Basis • Plasma applications (Dielectric Barrier Discharge, Enhanced oil recovery, Fusion power, Ion implantation, Ion thruster, MAGPIE (Implosion experiments), Plasma ashing, Food processing, Plasma arc waste disposal, recycling, Plasma acceleration, Plasma medicine, Plasma window, ...). In this Section, a general review of plasma physics will be carried out by answering a series of questions that present both its definition and its main characteristics that allow us to understand how plasma is classified. These questions are three: 1. What is plasma? –Section 2.1.1–. 2. What are the main characteristic magnitudes of the plasma state? 3. How can different plasma states be classified? Through question 2 you will be responding to question 3 in the aspects that interest us, so both will be answered in Section 2.1.2. 2.1.1 What is plasma? Langmuir, in his works initiated in the twenties to develop valves that allowed the passage of high currents, introduced in 1929 the word plasma to describe ionized gases. Everything seems to indicate that 99% of the visible matter of Universe is in plasma state. He described the plasma he observed as follows: “Except near the electrodes, where there are sheaths containing very few electrons, the ionized gas contains ions and electrons in about equal numbers so that the resultant space charge is very small. We shall use the name plasma to describe this region containing balanced charges of ions and electrons.”[ 2 ]. The term “plasma” refers to total or partially ionized gases, that presents collective behavior, corresponding to the fourth state of matter[3]. This overall behavior is due to the long range of electric forces and the existence of mobile charges can cause a slight local loss of spatial neutrality, which causes electric fields to which the loads are subject. Also as a consequence of the movement of them, magnetic fields are produced that will exert dynamic actions on the charges.[ 4 ] As to how this state is reached, the transition from gas to plasma takes place in a smooth way, whereby the plasma state is generally constituted by a gas of electrons, ions and neutral particles. At a macroscopic scale, plasmas present quasi-neutrality 1 [ 6 ]; a quasi-neutral state refers to the concentration of positive and negative charges is approximately equivalent. These particles are unbound, but it doesn’t mean that they are free in the sense of not experiencing forces, because moving charged particles generate an electric current within a magnetic field, and any movement of 1 There are also non-neutral or unstable plasmas . The strength and range of the electric force and the good conductivity of plasmas usually ensure that the densities of positive and negative charges in any sizeable region are equal ("quasineutrality"). A plasma with a significant excess of charge density, or, in the extreme case, is composed of a single species, is called a non-neutral plasma. In such a plasma, electric fields play a dominant role. Examples are charged particle beams, an electron cloud in a Penning trap and positron plasmas.[5] 2.1 Plasma physics 9 a charged plasma particle affects and is affected by the fields created by the other charges. In turn this fact governs collective behavior with many degrees of variation. So, plasma is a state of matter in which an ionized gaseous substance becomes highly electrically conductive to the point that long-range electric and magnetic fields dominate the behavior of the matter[ 7 ][ 8 ]. Most of matter in the visible Universe is in a plasma state; unlike other states of matter, plasma is rare on the Earth’s surface under normal conditions, and is mostly artificially generated from neutral gases. Three factors define a plasma: the plasma approximation, bulk interactions and plasma frequency [ 9 ], but they will be discussed in Section 2.1.2. Thus, regarding ionization, ionized gases cover a wide range of thermodynamic properties and regimes depending on the values that can be associated with important properties that characterize them, such as electronic density, ne , ion density, ni , and temperature of the electrons, Te ; and these different regimes to which reference is made, can be achieved both naturally and by industrial processes. It is important, therefore, to develop a characterization of types of plasmas, so that they can be identifiable and, therefore, also lead us to develop the most desired configuration of DBD controller through the representative parameters of each different plasma. It is important to be aware that in spite of being able to define relatively simple mathematical models that characterize plasma physics, in reality, the behavior of the plasma can be extraordinarily varied and subtle, and the appearance of unexpected behaviors from a simple model it leads us to develop complex systems . These systems are, in some way, at the limit between ordered and disordered behavior, since they can not normally be described by simple and fluid mathematical functions, nor by sheer randomness. The spontaneous formation of interesting spatial features on a wide range of length scales is one manifestation of plasma complexity. Examples of complexity and complex structures in plasmas include: filamentation (striations or string-like structures also known as Birkeland currents), Dusty plasma/grain plasma, non-neutral plasma –already described– or impermeable plasma[ 10 ]; however, it is not the purpose of this work to delve into these aspects. Derived from nature of this state of matter, there are quite interesting properties . Because they are charged, plasmas can be manipulated with electric and magnetic fields, and can be used to effect technologically useful physical and chemical transformations in controllable ways. two of them will be highlighted right now, such as being able to control the direction/confinement of the plasma through electromagnetic fields; and the generation of the "sheath” effect in the contact zone between plasma and solids. Respect to the first, the literature [ 4 ] says that the name "plasma" seems to be related to the possibility of shaping ("plasmar" in Spanish) using electric and magnetic fields, thus avoiding the need for a container or container, as is the case with fluids; this first property has many applications, for example, the concept applied by ionic propellers, which can dispense with physical nozzles by replacing them with an electromagnetic field that directs the output flow, thus avoiding temperature limitations on the flow that the materials of nozzles require to be able to resist. Regarding the second property, it will be explained in a little more detail below. 16 Chapter 2. Theoretical Basis Thus, generally a criterion widely used to classify plasmas is its kinetic temperature and its density of charged particles, as can be seen in Fig. 2.2 developed from the diagrams of Boeuf[ 43 ], Braithwaite[ 36 ], Jolibois[ 37 ] and Smirnov[ 38 ]. The formula for α is in equation 2.7, only depends on ion density, ni and neutral-particles density, nn. [41] α=ni ni+nn (2.7) figure 2.2 Classification of plasmas characterized by electronic density, ne, and electronic temperature, Te.Source of the image: [43][36][37][38]. One can differentiate between fully ionized plasmas, partially or weakly ionized gases. In a plasma, the electronion collision frequency, νe,i , is much greater than the electron-neutral collision frequency, νe,n . Therefore, with a weak degree of ionization the electron-ion collision frequency can equal the electron-neutral collision frequency: νe,i=νe,n, being the limit separates a fully ionized plasma from a weakly ionized plasma. • Fully ionized plasmas: the term was introduced by Lyman Spitzer, and it means that the plasma is in a Coulomb-collision dominated regime, i.e., νe,i>νe,n , which can correspond to a degree of ionization as low as 0.01%.[42] • Partially ionized plasmas: the plasma is not dominated by Coulomb collisions, i.e. when νe,i<νe,n , this means that the screening of the Coulomb interaction of ions and electrons by other ions and electrons is negligible 2.1 Plasma physics 17 ( α1 ). In this case, it can be assumed as a hypothesis that collisions mainly occur between charged particles and neutral atoms, and that collisions between charged particles can be practically disregarded. Most of technological (engineered) plasmas are weakly ionized gases. Relating the density to the speed or frequency of collision , logically, the amount of movement that is transferred in the collisions per unit of time increases with the number of molecules in a specific volume of the gas; and in addition, the collision frequency, νm , is proportional to the density, nn , and consequently to the gas pressure, p , nevertheless, the dependency with the collision speed can be neglected in first approximation –relation is in equation 2.8–. [50] [51] νm∝nn∝p(2.8) By comparing different collisions with each other, loads will surely be accelerated in random directions. In this way, provided an uniform external electric field, −→ E , this random movement will be governed by a systematic movement in the direction of the field, which is known as drift. If this movement is averaged over numerous collisions, the drag or drift velocity, −→ vd, is obtained –equation 2.9–. −→ vd=         −µ−→ Efor positive particles −µ−→ Efor negative particles (2.9) where µ is mobility like µ=e/mνm , valid for any value of the electric field, since it is assumed that in first approximation the mobility is only a function of the collision frequency[ 33 ][ 51 ]. Because the mass, and its corresponding inertia, are larger in ions than in electrons, the mobility of the ions is several orders of magnitude less than the mobility of the electrons: µiµe. As one has seen, different approaches can be used to classify Plasmas. 18 Chapter 2. Theoretical Basis 2.2 DBD Plasma Actuator In recent times, the development of plasmas artificially (industrially), the so-called technological plasmas , has become a reality with many future possibilities in very varied applications (surface engineering, biomedical, environmental, manufacturing, propulsive, energy sources, ...). This vast field of possibilities is favored by, as one has already seen, the amount of possible alternatives in which plasma can present itself, with a wide range of temperatures and pressures. Most artificial plasmas are generated by application of electric and/or magnetic fields through a gas, causing its ionization, or to a solid to achieve an ionizing sublimation. Different energetic approaches are applied to achieve this objective. In general, plasma generated in a laboratory setting and for industrial use can be categorized by[52]: •The type of power source used to generate the plasma •The pressure they operate at •The degree of ionisation within the plasma •The temperature relationships within the plasma •The electrode configuration used to generate the plasma •The magnetization of the particles within the plasma The work objective of this project is focused on the generation of technological plasma through Dielectric Barrier Discharge (hereinafter “DBD”) reactors –see page 7–, studying the influence of gap, pressure, frequency and material; for possible aerodynamic flow control purposes. So, in this part of the document a review of the physics of discharges on gases will be shown to understand how plasma is generated –Section 2.2.1– to later expose, specifically, the bases of generation of plasma by Dielectric Barrier Discharge –Section 2.2.2–, and finally go on to explain the general concept of the device and its applications –Section 2.2.3–. 2.2.1 Electrical discharges in gases In this Section, basic ideas about the mechanism and discharge status are exposed, to understand the operation of the plasma flow controller generated by gas discharge. Three parts are going to be distinguished: basic aspects –Section 2.2–, voltage-current characteristics –Section 2.2– and breakdown voltage and Paschen curve –Section 2.2–. Basic aspects of discharges in gases Electric discharge in gases occurs when electric current flows through a gaseous medium due to ionization of the gas. Depending on several factors, the discharge may radiate visible light (glow discharge). 2.2 DBD Plasma Actuator 19 First investigations on discharges in gases were mainly concentrated in low pressure discharge tubes, whose scheme can be seen in Fig. 2.3. Unlike them, plasma controllers are devices that work with alternating current, AC, and in a range of frequencies between 50Hz and 30 kHz, with an inhomogeneous and at least two-dimensional electric field. This implies more complexity in the corresponding download mechanisms and therefore does not allow an exact definition of the properties of the download. Despite this difference, some aspects of the direct current case DC discharge mechanism can also be used for a description of the fundamental aspects of the AC case. [ 0 ][ 54 ]. figure 2.3 Discharge tubes. Low pressure discharge tubes are the famous fluorescent tubes today. Source of the image: [53]. The fact that plasma controllers can operate at non-reduced pressure offers a lot of possibilities. Voltage-Current characteristics of discharges in gases The increase in voltage between the two electrodes in a low pressure discharge tube leads to a non-linear voltagecurrent relationship –see Fig. 2.4–. There are three types of discharge regimes, these are: Dark Discharge (dark current mode), Glow Discharge (normal glow mode) and Arc Discharges (arc mode). [55][56][0] 1. Dark current mode: Apart from the exception of the corona discharge, this discharge regime remains invisible to the naked eye and is therefore called dark discharge. •A-B: During the background ionization stage of the process the electric field applied along the axis of the discharge tube sweeps out the ions and electrons created by ionization from background radiation. Background radiation from cosmic rays, radioactive minerals, or other sources, produces a constant and measurable degree of ionization in air at atmospheric pressure. The ions and electrons migrate to the electrodes in the applied electric field producing a weak electric current. Increasing voltage sweeps out an increasing fraction of these ions and electrons. I.e., at low voltage there are very few charges that contribute to current flow. In this regime an increasing number of the charges generated are accelerated towards the electrodes as the voltage increases. •B-C: Once all the charges produced reach the electrodes, there is a stagnation in the current in the saturation regime, in which the energy levels are too low to create new secondary loads. Because if the voltage between the electrodes is increased far enough, eventually all the available electrons and ions are swept away, and the current saturates. So, in the saturation region, the current remain constant while 20 Chapter 2. Theoretical Basis figure 2.4 Typical V/I plot of a glow discharge (DC). The main characteristics of the discharge such as the breakdown voltage, the voltage current characteristic and the structure of the discharge depend on the geometry of the electrodes, the gas used, the pressure and the electrode material.[55][56] Source of the image: [102]. the voltage is increased. This current depends linearly on the radiation source strength, a regime useful in some radiation counters. •C-D: If voltage across the low pressure discharge tube is increased beyond point C, the current will rise exponentially. The electric field is now high enough so the electrons initially present in the gas can acquire enough energy before reaching the anode to ionize a neutral atom. As the electric field becomes even stronger, the secondary electron may also ionize another neutral atom leading to an avalanche of electron and ion production. The region of exponentially increasing current is called the Townsend discharge; and in this regime the electric field is high enough so that the primary charges partially ionize the gas; and, like it has been said, due to these new charges acquire enough field energy to ionize the gas, avalanche discharges appear, leading to an exponential growth of the current as the voltage increases. •D-E: Corona (crown) discharges may appear in the later stages of the Townsend regime, where between D and E a localized increase of the field exceeds the breakdown point of the gas. These local concentrations are caused by peaks, side edges or roughness of the electrodes. Definetly, corona discharges occur in Townsend dark discharges in regions of high electric field near sharp points, edges, or wires in gases prior to electrical breakdown; and if coronal cuurents are high enough, corona discharges can be technically “glow discharges”, visible to the eye. However, for low 2.2 DBD Plasma Actuator 21 currents, the entire corona is dark, as appropriate for the dark discharges 9 . E (Electrical breakdown) occurs in Townsend regime with the addition of secondary electrons emitted from the cathode due to ion or photon impact. At the breakdown, or sparking potential,the current might increase by a factor of 104 to 108, and is usually limited only by the internal resistance of the power supply connected between the plates. 2. Normal glow mode: •E-F: Once the breakdown voltage is reached at point E, an electrical break occurs followed by a DISCONTINUOUS transition from E to F to the glow discharge regime. The large potential difference between the electrodes drops due to the significant increase in the current flow, and therefore the plasma becomes visible. •F – G: After reaching point F, the gas enters the normal glow/luminiscent region, in which the voltage is almost independent of the current over several orders of magnitude in the discharge current, and current increases. The electrode current density is independent of total current in this regime. This means that the plasma is in contact with only a small part of the cathode surface at low currents. As the current is increased from F to G, the fraction of the cathode occupied by the plasma increases, until plasma covers the entire cathode surface at point G. •G – H: In the abnormal glow regime above point G, the voltage increases significantly with the increasing total current in order to force the cathode current density above its natural value and provide the desired current, because once the cathode is completely covered, a considerably high voltage rise is needed to achieve an abnormally high current density (and this result is achieved in electrodes heated in succession and in an incandescent cathode at point H). Go on from C and moving to the left to lower currents, a form of hysteresis-cycle is observed in the voltage-current characteristic curve: the discharge maintains itself at considerably lower currents and current densities than at point F and only then makes a transition back to Townsend regime; it means, the luminescent discharge is maintained below F, once it has been established. At point F’the transition back to the Townsend regime occurs without going through the Townsend discharge or crown again. 3. Arc mode: •H-I: The electrodes at point H become sufficiently hot that the cathode emits electrons thermionically. In consequence, if the DC power supply has a sufficiently low internal resistance, the discharge will undergo a glow-to-arc transition (H-I). •I-J: The transition is followed by a regime of non-thermal arcs (I-J), where the voltage required for the discharge decays rapidly due to the additional loads of the thermionic emission: the arc regime, from I 9Related phenomena include the silent electrical discharge, an inaudible form of filamentary discharge, and the brush discharge, a luminous discharge in a non-uniform electric field where many corona discharges are active at the same time and form streamers through the gas. 22 Chapter 2. Theoretical Basis through K is one where the discharge voltage decreases as the current increases, until large currents are achieved at point J, and after that the voltage increases slowly as the current increases. •J-K: Since this influence grows with the heating of the electrons, the effect corresponds to a characteristic negative resistance. Beyond the local voltage minimum at point J, the voltage-intensity curve again reaches the thermal arc regime (J-K) towards thermal equilibrium at point K. As we already know, the above discussion introduces the main characteristics that describe a low pressure discharge. For levels of atmospheric pressure and more inhomogeneous electric fields, direct breakdowns by arcs can occur, which can be maintained with sufficient current applied (indicated in Fig. 2.4 by dotted lines and by the point E’). Depending on the geometrical conditions and the gas, it is possible that there is a direct transition from the Townsend discharge or corona to the arc discharge (E’-I).[57] Breakdown voltage and Paschen curve The measurements of the experiments carried out have been saved at the ignition voltage appreciable by the naked eye. Theorically, the breakdown voltage , Vbd , (ignition/breakdown potential), signaled with point E in Fig. 2.4, is a system instability resulting from discharge and high voltage, and is crucial since before that point (period CE) the discharge is only possible through the emission of secondary electrons. Beyond point E the ionization avalanche provides a sufficient amount of electrons to maintain a self-sustaining discharge. [58] The interest of this project is that said breakdown voltage, and its corresponding electric firing field, depend both on the material of the electrode, the geometry and the distance between electrodes, and on the species of gas and pressure, as well as on the the characteristics of the substrate used in the case in question, DBD actuators. At the end of the 19th century, Paschen formulated a law with the same name (equation 2.10) in which it was established that for a given gas, the ignition voltage is only a function of the product pg (where g the distance between electrodes) and the secondary electron emission coefficient in the cathode γes . [ 62 ][ 135 ] –see Fig. 2.5–. For all gases there is a minimum ignition voltage, depending on pg , which corresponds to the Stoletow benchmark of better efficiency, which implies a minimum power consumption of the discharge. Vbd =Bpg ln(Apg)−ln[ln(1+1 γes )] (2.10) where p is pressure in pascals, g distance between electrodes in meters, is the coefficient of emission of secondary electrons in the cathode, Ais the ionization of saturation in the gas at a particular value of E/p, and Bis related to ionization and excitation energies. The values A and B are determined experimentally, and what is obtained is that they are approximately constant in a strict range of values of E/p for a given gas. For example, For example, in the case of air with values of E/p between 450 and 7500 V(kPa cm) −1 , we have values for the constants of A=112.50 (kPa cm)−1, and B=2737.50 V(kPa cm)−1.[135] [62] 2.2 DBD Plasma Actuator 23 figure 2.5 Paschen curves for different gases. The parameter drefers to the distance between electrodes, in the present document it is represented as g.Source of the image: [60]. . 2.2.2 Dielectric Barrier Discharge Plasmas Theory: Characterization Dielectric Barrier Discharge (DBD) –see Fig. 2.6– is a non-thermal electrical discharge generated by the application of high voltages across small gaps wherein a non-conducting coating prevents the transition of the plasma discharge into an arc10 . It is characterized by having at least one layer of an insulating dielectric barrier located between the electrodes. When a high voltage is applied between the electrodes some arcs are randomly generated among them caused by the accumulation of charges on the surface of the dielectric material –Section 2.2–, this generates thousands of discharges in microseconds which provides the degree of ionization required to sustain a discharge of plasma. figure 2.6 Filamentous Dielectric Barrier Discharge. The photograph shows an atmospheric DBD discharge occurring in between two steel electrode plates, each covered with a dielectric (mica) sheet. The filaments are columns of conducting plasma, and the foot of each filament is representative of the surface accumulated charge. The discharge is taking place in normal atmospheric air, at about 30 kHz, with a discharge gap of about 4 mm. The “foot” of the discharge is the charge accumulation on the barrier surface.Source of the image: [78] . Because the DC current can not pass this dielectric barrier, it is required to operate in AC 11 for the operation 10An electric arc, or arc discharge, is an electrical breakdown of a gas that produces an ongoing electrical discharge. 11 The process normally uses high voltage alternating current, ranging from lower RF to microwave frequencies. However, other methods were developed to extend the frequency range all the way down to the DC. One method was to use a high resistivity layer to cover one of the electrodes. This is known as the resistive barrier discharge. [75][76] 24 Chapter 2. Theoretical Basis of the DBD device. In addition, this fact offers the advantage of locating the electrodes, or at least one of them, separated from the discharge, that is to say without being in contact with it [ 63 ]. This can be very beneficial when, for example, highly aggressive gases are used. So, plasma controllers used in this project act at AC voltages, that is, the electric field periodically changes sign, so when the AC voltage is applied, a discharge appears on the surface of the dielectric. Therefore, in the operation, the discharge appears every half cycle of the applied AC voltage. figure 2.7 Examples of typical configurations of Dielectric Barrier Discharge in volume (VD). Source of the image: Own image . figure 2.8 Examples of typical configurations of Dielectric Barrier Discharge in surface (SD). Source of the image: Own image . Unlike most other non-equilibrium discharges, the Dielectric Barrier Discharge can operate at high pressures, in a range of 10−1−101bar. Also, there are both class types Volume Discharge (VD) and Surface Discharge (SD), like you can see in Fig. 2.7 and 2.8, respectively. The main difference between these two configurations is that in the VD discharge the rupture occurs through the gas in the space between electrodes, while in the SD discharge the rupture grows periodically on the dielectric. Note that the presence of a dielectric layer is essential for the formation of a surface discharge, since the insulation prevents the flow of current.[ 74 ] The figures that have been referred show diagrams with typical constructions of a DBD device, wherein one, usually, of the two electrodes is covered with a dielectric barrier material. The lines between the dielectric and the electrode are representative of the discharge filaments, which are normally visible to the naked eye. This work is based on surface discharges. 2.2 DBD Plasma Actuator 25 The first technological application of the Dielectric Barrier Discharge was made by Ernst Werner von Siemens in 1857[ 64 ], who observed luminescence between two coaxial glass tubes while generating ozone . Originally DBD discharge was called silent (inaudible) –but experiments reveal that the geration of plasma DBD, the ionization of a gas, does sound– discharge and also known as ozone production discharge or partial discharge. It is the most profitable technique to produce it. [65][66][66]. With respect to its applications , currently Dielectric Barrier Discharges constitute a scientific field of great interest, whose success is demonstrated in the wide variety of industrial applications based on this type of plasma, especially focused on the generation of ozone, decontamination of gases, fluorescent lamps, laser discharge, surface treatment, synthetic facrics and plastics, plasma medicine 12 or also used in the design of low temperature plasma jets13.[68][69][71] Interest in plasma actuators as active flow control devices is growing rapidly due to their lack of mechanical parts, light weight and high response frequency.[ 77 ] In consequence, due to the direct relationship with my studies, this Master’s Thesis will focus on the applications of DBD devices as flow controllers , and Section 2.2 contains a list of historical advances regarding this discipline. Dielectric and ferroelectric materials A dielectric (or dielectric material) is an electrical insulator –low electrical conductivity/specific conductance (σ1)– that can be polarized by an applied electric field . Thus, all dielectric materials are insulators but not all insulating materials are dielectric. The term dielectric, which means “through”, was coined by William Whewell (from dia−+electric) in response to a request from Michael Faraday. [153][152] Although the term “insulator” implies in itself having low electrical conductivity; A “dielectric” material refers to materials with a high degree of polarizability. That is, the term insulation is used to indicate electrical obstruction, while the adjective dielectric is used to indicate the energy storage capacity of the material, through polarization. The concept of polarizability is expressed through the magnitude: relative permittivity, εr14 , also is called as dielectric constant. Scheme shown in Fig. 2.9. When a dielectric is placed in an electric field, electric charges do not flow through the material as they do in an electrical conductor but only slightly shift from their average equilibrium positions causing dielectric polarization. Because of dielectric polarization, positive charges are displaced in the direction of the field and negative charges shift in the opposite direction. This creates an internal electric field that reduces the overall field within the dielectric 12 The Dielectric Barrier Discharge was used in the mid-1990s to show that low temperature atmospheric pressure plasma is effective in inactivating bacterial cells. This work and later experiments using mammalian cells led to the establishment of a new field of research known as plasma medicine. [72] 13 These plasma jets are produced by fast propagating guided ionisation waves known as plasma bullets.[73] 14Relative permittivity is its (absolute) permittivity, ε, expressed as a ratio relative to the permittivity of vacuum, ε0. Permittivity is a material property that affects the Coulomb force between two point charges in the material. Relative permittivity is the factor by which the electric field between the charges is decreased relative to vacuum. Likewise, relative permittivity is the ratio of the capacitance of a capacitor using that material as a dielectric, compared with a similar capacitor that has vacuum as its dielectric. Relative permittivity is also commonly known as dielectric constant. 32 Chapter 2. Theoretical Basis Table 2.2 Characteristics and typical parameters of the SDBD plasma controller. Definition range ∼Atmospheric pressure p=10−1−101bar Weakly ionized gas α=10−4−10−3 Non-equilibrium plasma T0 e=1−10eV Ti≈Tn≈300K=0.0259eV Range of Usual Values Operating voltage V=1−50kV Frequency of operation f=0.5−25kHz Wave type sinusoidal, square, triangular load imbalance at the edges of the plasma. This imbalance generates an electric field opposite to the applied external field. Thus, the redistribution of charges continues until the net electric field in the plasma is neutralized.[0][28] Usual values See Table 2.2. Thus, the discharge rate of the plasma controllers is generally classified as a luminescent discharge (slightly filamentary) in the group of non-equilibrium plasmas, where the temperature of the electrons is approximately two orders of magnitude higher than that of the electrons ( T0 e=1−10eV , Ti≈Tn≈300K=0,0259eV ). According to the low degree of ionization ( α=10−4−10−3 ), the DBD plasma controllers produce a thin volume of weakly ionized gas, where the density of charged species is approximately n=1016m−3.[0] In this work the experiments have been developed in electrical values usual in the industry for this type of devices. Generally, an AC voltage of a peak value of 5-20 kV is usually applied to the electrodes (although it is normal to move in a 1-50 kV range) at a frequency range from 3-15 kHz ( analogously, literature can be found that extends this range to 0,5−25kV). The AC signal is usually a sinusoidal, square or triangular wave. [88]-[91];[0] Regarding the pressure of the gas to which the actuator is used, this usually corresponds to reduced pressure ( 10−1bar ) that represents cruise characteristics[ 98 ][ 99 ], or raised to 101bar to reproduce the internal conditions of compressors and turbines.[ 100 ] In this work, the central focus will be on experiments at atmospheric pressure (sea level: 1013 mbar). 2.2 DBD Plasma Actuator 33 2.2.4 DBD applications as flow control devices There are two ways through which to manipulate desire a flow: actively or passively. One knows that with the term “active”, it is understood a flow control device that involves the need to apply energy to the system through an actuator, which needs auxiliary power as well as a control method. There are numerous investigations related to plasma controllers, fruit of a growing interest of the scientific world and the industry in the development of both plasma applications and micro-scale technologies. Consequently, you can access a variety of experimental tests or models of representation that support the path of the potential of plasma devices to become a reality for active flow control in a field as promising as aerodynamics. In order to manipulate a free airflow, three are the main objectives to act on: laminar-to-turbulent transition (delaying it to reduce drag, for example), separation (for take-off, landings...) and turbulence (for reducing noise for example). The main concept is that the body force acting on the flow –see Section 2.3–, as produced by the plasma DBD actuator, is used to change the Kutta condition on the rounded trailing edge of the profile and with that the circulation and corresponding lift. Such a configuration reduces the weight and complexity issues typically associated with classical blowing circulation control concepts and many other concepts aiming at load control[ 1 ]. Specifically “in the middle of the 1990s, Roth’s group perfected and developed a new atmospheric pressure DBD, a surface DBD, established in air between at least two electrodes placed asymmetrically on each side of a dielectric; they protected it by a patent in 1994 and called it ‘One Atmosphere Uniform Glow Discharge Plasma’ (OAUGDPTM).” It waas presented for the first time for airflow control in 1998 [ 149 ], [ 147 ] and published in a journal in 2000 [ 148 ]. It was revealed that the discharge induces a secondary airflow of several m/s tangentially to the wall, an that consequent force increases with the applied voltage, modifying the free air stream, which causes a change in the drag. Going back to half a century ago in history, where one of the first discharge-based flow control articles was published in 1968 by Velko and Ketcham [ 136 ], who demonstrated the effect of electric fields on fluid boundary layers through corona discharge cables, which were placed at a short distance from the surface of the flat plate. It also highlights the advances made in the 90s by the research group of John R. Roth at the University of Tennessee; they developed a new way of generating uniform supeficial Dielectric Barrier Discharge, which was called uniform discharge of luminescent plasma at one atmosphere (OAUGDP). At the end of the century, Roth published the first results based on OAUGDP [137]. At present, this subject arouses great interest in the scientific world, and the generation of plasma by barrier discharge is proposed in a variety of aerodynamic applications, as it is the case that concerns the replacement of moving parts for flow control where, in addition, a fast response time is needed. Some other applications successfully tested in the laboratory are the elevation of the wing section [ 135 ], control of the separation of the blades of a low pressure turbine [ 138 ], control of the aerodynamic flow at the tip of a turbine [ 139 ], generation of unstable vortex [ 140 ], separation control at the tip of a wing profile [ 141 ], application to the aerodynamics of a train [ 124 ], etc. In the final chapter of this work, in which the main conclusions of the study are summarized, some other studies are shown that can be compared with the present work, or that analyze the same concept from different perspectives or 34 Chapter 2. Theoretical Basis models (as experimental as computational), as well as possible future research lines. Important dissertations also are the projects related with the study of the single-dielectric barrier discharge aerodynamic plasma actuator, developed in [ 88 ] and [ 95 ] by Orlov et al., where the physics of the plasma discharge was studied through the time-resolved light intensity measurements of the plasma illumination, and plasma characteristics were obtained and analyzed for a range of applied voltage amplitudes and a.c. frequencies. Based on this data, electro-static and lumped-element circuit models were developed, and, in this way, the time-dependent charge distribution was used to provide boundary conditions to the electric field equation that was used to calculate the actuator body force vector; and it was applied on airfoils. Finally, some very interesting theses, in addition to those that are appearing referenced throughout the work, or that can be found in the historical summary on the plasma actuators made by Moreau [ 94 ], are the following. For example, “Plasma Flaps and Slats: An Application of Weakly Ionized Plasma Actuators” [ 150 ], “Environmental Impacts on Dielectric Barrier Discharge Plasma Actuators” [ 151 ] , application related with anti-icing, for example, in “Experimental study of an anti-icing method over an airfoil based on pulsed dielectric barrier discharge plasma”[ 152 ] or “An experimental study of icing control using DBD plasma actuator ” [ 153 ], about plasma actuators manufacturing to reduce degradation[ 154 ] or, different, for separation control in oscillating profiles[ 155 ]. It is also interesting to pay attention to the publications developed in the framework of the Delft University of Technology about plasma actuators in recent years. 2.3 Mechanical, Electrical and Spectroscopic Analysis 35 2.3 Mechanical, Electrical and Spectroscopic Analysis It is intended to model the plasma actuator in such a way that its effectiveness in controlling flow separation is as successful as possible. For this, most experiments have focused on the modification of the physical dimensions and macroscopic parameters of the plasma actuator system. There are many parameters involved, so controlling in detail the dynamic behavior of the system becomes extremely complicated in short time scales. However, the macroscopic behavior of the device is highly influenced by its microscopic structure, and the fact that the scales of measurement of the device elements and the discharges move in small units (the devices are usually several millimeters in height and the discharge is composed of multiple micro-shocks of short duration, of the order of nanoseconds that vary significantly over the period of the AC voltage source) –see Chapter 3–, causes that directly measure these properties microscopic is also of great difficulty, properties such as charge density or electric field. In this work the DBD devices are going to be characterized, above all, through an electrical analysis, and for this it is going to study aspects such as the influence of the configuration of the electrodes, the dielectric material or the gap, whose results are presented in the following chapters. In the consulted bibliography, as it has been summarized in Section 2.2, there have been very diverse studies in which other parameters have been varied; that is why it is recommended that the reader go to different bibliography to deepen in the understanding of this type of reactors. An efficient alternative to measure these micro-properties is numerical modeling, thus, numerical modeling and simulations are useful and flexible tools for modeling the development of Dielectric Barrier Discharge controllers in spatial and temporal dimensions, in addition to being able to continuously monitor the evolution of complex flow control systems. But, like it has been explained, this is not the objective of this work –see Section 1.1–. In this part of the document, firstly, fundamental concepts about plasma body force are collected in Section 2.3; accordingly to the power characterization, in Section 2.3 the aspects of the electrical analysis that correspond to this work are explained; likewise, spectroscopic analysis is included, whose bases are found in Section 2.3. 2.3.1 Ionic wind The ionized air that appears on the dielectric surface is the basic mechanism that gives it the ability to manipulate or induce a flow configuration to the DBD plasma actuator or controller, since in the presence of the electric field produced by the geometry of the electrode, a force of body that acts on the environment, not charged air; because it has already been seen that the collision processes lead to a transfer of momentum between the particles involved, because as the charged particles move under the Coulombian forces of the electric field, between the exposed electrode and the surface of the dielectric, they collide with the neutral particles that make up the gas, transferring amount of movement, these collisions being called “Lorentzian collisions”. This effect is classically termed as an 36 Chapter 2. Theoretical Basis electric wind, directed from the exposed electrode toward the dielectric surface, that modifies the boundary layer, and allowing for the active manipulation of the airflow [ 93 ][ 94 ][ 95 ]; the act of producing a mechanical output or kinetic energy, for an electrical input or electric energy, without the use of mechanical moving parts, classifies the plasma actuator as a Micro-Electro-Mechanical-System (MEMS). [94] figure 2.15 Scheme of the flow controller where the direction of flow induced can be seen. . Thus, ion wind –ionic wind, coronal wind or electric wind– describe the airflow induced by electrostatic forces linked to corona discharge arising at the tips of some sharp conductors (such as points or blades) subjected to high voltage relative to ground , as each collision contributes to the force exerted on the gas and in this way flow velocities are induced on the surface of the electrodes. The induced flow is characterized by a tangential velocity from the exposed electrode towards the encapsulated electrode –see Fig. 2.15–. Ion wind (this motion induced in the gas, created by the Coulombian forces acting on the ions and electrodes, and the Lorentzian collisions between the particles) is an electrohydrodynamic phenomenon; and Ion wind generators can also be considered electrohydrodynamic thrusters. [97] [32] As already mentioned, the DBD controllers generate an EHD force or plasma body force that depends entirely on the electric field, −→ E(−→ r,t) , defined by the geometry of the electrode and the charged particles density, −→ ni(−→ r,t) ,as a function of position −→ r and time t , integrated into the plasma volume V –see equations 2.11 and, summed over all species, 2.12–.[28] −→ Fi(t) = Zv qini(−→ r,t)−→ E(−→ r,t)dt(2.11) −→ FEHD(t) = ∑ i −→ Fi(t)(2.12) The phenomenon of electric wind, is referred to the “movement of gas induced by the repulsion of ions from the 2.3 Mechanical, Electrical and Spectroscopic Analysis 37 vecinity of a high voltage electrode” [ 46 ], and it was firstly reported at the beginning of XVIII century and Faraday explained it more than a hundred years later. So, this wind, as Moreau says in [ 94 ], “is due to the collisions between the ions that drift and the neutral particles in the electrode gap region, because as the electron velocity is much higher than the ion one, one can neglect the role of electrons because their mass is very low compared with the ion one.” Expression 2.3[46] contains the firs relation for electric wind: νG=kqi ρµ In consequence, plasma actuators consist of using the discharge-induced electric wind within the boundary layer to change its properties, and in this way transform the airflow actively. During this project, the most usual configuration also has been built and tested, composed of two electrodes flush-mounted at the wall of a flat plate (simulating the wing profile), between which a high voltage is applied, generating a non-thermal surface plasma. When discharge is on and is established, the airflow is deviated tangentially very close to the wall and accelerated within its boundary layer as consequence of the depression induced at the anode by the electric wind.[ 94 ] In these conditions of the devices that are going to be exposed in this worl, and above the ignition voltage, a plasma sheet appears on both sides of the dielectric, consisting of an electric wind on both sides of the dielectric substrate. As one can check in this document, electrical and mechanical ..as bibliography tell us[ 94 ]– characteristics of this plasma depend strongly on different parameters, such as electrode width, electrode gap, dielectric thickness or the nature of the dielectric. Over the years, various body force models have been proposed , and these models can be classified into two categories: Phenomenological (or Simplified), and First- Principles-Based models. Phenomenological models involve approaches with reduced complexity providing reasonable qualitative correspondence with experimental observations. In reference [ 28 ], is explain to us that such models are limited in their ability to accurately predict, and explain the physics involved. Meanwhile, First-Principles-Based models distinctly address the necessary knowledge required towards understanding the fundamental physical processes involved. [91]-[96] This work will not have simulation studies, but the reader can go to, for example, the reference [ 28 ] to know the conclusions drawn from the analysis developed to understand the plasma actuators through: a fluid electrostatic model by Orlov[ 88 ], and a hybrid PIC model formulated by the author. Both models are based on first-principles, whose objective will be to provide a comparative study of the two models in their application to the SDBD plasma actuator flow problem focusing specifically on the formation of the plasma body force. 2.3.2 Electrical characterization: Accumulated charge and Consumed power The yield of a plasma controller is mainly characterized by the electrical power that is consumed during the discharge. The plasma parameters, such as electron density, sheath thickness and sheath voltage, are not easy to be probed experimentally, while the electrical characteristics, such as impedance, resistance and reactance, are relatively convenient to be measured, and there are several studies that that relate both parameters through theoretical models. [146] 38 Chapter 2. Theoretical Basis In Section 3.4 the schematization of the experimental procedure will be presented, while in the present part of the work the underlying theory is indicated. As a result of the experiment, the oscilloscope returns two output files with the electrical conditions of the storage point, that is, the breakdown point of the plasma; These correspond to the current and voltage signal that the device is experiencing at that moment, and which contain as many signal periods as specified on the screen –see Fig. 4.1 and 4.2– ; the signal can be saved either averaged or not averaged. Thus, one has two files that contain the discretization of Vbd(t) e Ibd(t) , and from them all the electrical parameters that are presented in this Section are calculated/obtained. Voltage and current are defined as follows: •Voltage, electric potential difference, electric pressure or electric tension (formally denoted ∆V or ∆U , but more often simply as V or U , for instance in the context of Ohm’s or Kirchhoff’s circuit laws) is the difference in electric potential between two points. The voltage between two points is equal to the work done per unit of charge against a static electric field to move a test charge between two points. This is measured in units of volts (V) (a joule per coulomb); moving 1 coulomb of charge across 1 volt of electric potential requires 1 joule of work.[126] •Electric current , I is a flow of electric charge. In electric circuits this charge is often carried by moving electrons in a wire. It can also be carried by ions in an electrolyte, or by both ions and electrons such as in an ionised gas (plasma).[127] The SI unit for measuring an electric current is the ampere (A), which is the flow of electric charge across a surface at the rate of one coulomb per second. •Phase difference or Phase shift as it is also called of a Sinusoidal Waveform is the angle φ , in degrees or radians that the waveform has shifted from a certain reference point along the horizontal zero axis. In other words phase shift is the lateral difference between two or more waveforms along a common axis and sinusoidal waveforms of the same frequency can have a phase difference.[108] The three main components in an AC circuit which can affect the relationship between the voltage and current waveforms, and then their phase difference, by defining the total impedance of the circuit are the resistor, the capacitor and the inductor. The generated current in plasma reactor is sinusoidal with a phase shift φ with respect to the voltage signal–see Fig. 4.1 and 4.2–, because there are not only resistive elements, and we are interested in moving away from a behavior of the device as if it were a capacitor 16 . Related to this, for the further development of the power concept, it is important to know that the intensity has two components. Active component of the intensity, Ia=Icos φ , is defined to the component of it that is in phase with the voltage, and reactive component, Ir=Isenφ , to which it is in quadrature with it –see Fig. 2.16.a–; so in order 16Why is there a phase shift between current and voltage in an AC circuit? Related to the impedance associated with the passage of current in the case of capacitors and coils. The capacitor, when it is initially discharged, offers little resistance to current flow; However, as time goes by, it becomes charged and thus increases the resistance to the passage of current. When it is fully charged, its behavior is as if it were an open circuit. The fact of using alternating current causes that it also generates the same effect when the capacitor is discharged, at the beginning of the negative half cycle. When the voltage enters in a negative half-cycle, the capacitor still has remains from the charge of the first half cycle. This is what causes 90 to be lagged. In the case of the coils, the behavior is completely opposite to that of the capacitor. In a coil, its first behavior is of high impedance, and with the passage of time, this resistance begins to decrease 2.3 Mechanical, Electrical and Spectroscopic Analysis 39 for us to calculate the total consumed power, it is needed to know the phase difference between the sinusoidal waveforms of the voltage and current. (a) Current triangle.. (b) Impedance triangle.. (c) Power triangle.. figure 2.16 RLC circuit in AC: complex current, complex impedance and complex power. A phasor is a graphic representation of a complex number that is used to represent an oscillation, so that the phasor sum of several phasors can represent the magnitude and phase of the oscillation resulting from the superposition of several oscillations in an interference process.. The signals that are received from the oscilloscope in a discrete manner, are first centered so that the calculations of the integrals in which they are later involved are more precise; for this the following assumption is taken: the average of each of the signals during a period must be zero. In addition, after that they are smoothed through the smooth Matlab function. Ohm’s law is a basic law of electrical circuits. It establishes that the potential difference V that is applied between the ends of a given conductor is proportional to the intensity of the current I that circulates through said conductor. Ohm completed the law introducing the notion of electric resistance R ; which is the proportionality factor that appears in the relationship between Vand I: V=R·I(2.13) where V corresponds to the potential difference, R to the resistance e I to the intensity of the current. The units of these three magnitudes in the international system of units are, respectively, volts (V), ohms (Ω) and amps (A). Some parameters used in this projects are explained in the following sections: •Voltage and Current, peak-to-peak and RMS values •Accumulated Charge •Lissajous Figures and Zero Charge 40 Chapter 2. Theoretical Basis •Impedance Components •Consumed Power •Power Factor and Phase Shift Voltage and Current, peak-to-peak and RMS values The amplitude of a periodic variable is a measure of its change over a single period (such as time or spatial period). There are various definitions of amplitude (peak-to-peak amplitude, peak amplitude, semi-amplitude, root mean square amplitude (RMS) or pulse amplitude), which are all functions of the magnitude of the difference between the parameter’s extreme values. In general, the use of peak amplitude is simple and unambiguous only for symmetric periodic waves.[129] •Peak-to-peak amplitude is the change between peak (highest amplitude value) and trough (lowest amplitude value, which can be negative). With appropriate circuitry, peak-to-peak amplitudes of electric oscillations can be measured by meters or by viewing the waveform on an oscilloscope. •Peak amplitude , useful in telecommunications and other areas where the measurand is a signal that swings above and below a reference value but is not sinusoidal; if the reference is zero, this is the maximum absolute value of the signal; if the reference is a mean value (DC component), the peak amplitude is the maximum absolute value of the difference from that reference. •Peak amplitude, semi-amplitude or amplitude means 1/2 (half) of the peak-to-peak amplitude. •Root mean square (RMS) amplitude: The quadratic mean value of an electric quantity is called the effective value. The concept of effective value is used especially to study the periodic waveforms, despite being applicable to all waveforms, constant or not. It is sometimes referred to as RMS (from the English, root mean square). So, this value is used especially in electrical engineering, and is defined as the square root of the mean over time of the square of the vertical distance of the graph from the rest state, i.e. the RMS of the AC waveform (with no DC component). For complicated waveforms, especially non-repeating signals like noise, the RMS amplitude is usually used because it is both unambiguous and has physical significance. For example, the average power transmitted by an acoustic or electromagnetic wave or by an electrical signal is proportional to the square of the RMS amplitude (and not, in general, to the square of the peak amplitude). For alternating current electric power, the universal practice is to specify RMS values of a sinusoidal waveform. One property of root mean square voltages and currents is that they produce the same heating effect as direct current in a given resistance. 2.3 Mechanical, Electrical and Spectroscopic Analysis 41 The peak-to-peak value is used, for example, when choosing rectifiers for power supplies, or when estimating the maximum voltage that insulation must withstand. VRMS =Vpeak √2(2.14) IRMS =Ipeak √2(2.15) •Pulse amplitude , in telecommunication, is the magnitude of a pulse parameter, such as the voltage level, current level, field intensity, or power level. Pulse amplitude is measured with respect to a specified reference and therefore should be modified by qualifiers, such as average, instantaneous, peak, or root-mean-square. Accumulated Charge Electric charge , Q , is the physical property of matter that causes it to experience a force when placed in an electromagnetic field. There are two types of electric charges, that is, positive and negative (commonly carried by protons and electrons respectively). Electric charge is a conserved property: the net charge of an isolated system, the amount of positive charge minus the amount of negative charge, cannot change. An object with an absence of net charge is referred to as neutral. [130] The coulomb (C) is the International System of Units (SI) unit of electric charge. It is the charge transported by a constant current of one ampere in one second. In this paper the concept of accumulated electric charge Qa will be used and, specifically, of accumulated centered charge Qc (although for simplification the subscript can be eliminated from now on), whose mathematical expressions are as follows, equation 2.16 and 2.17, respectively: Qa(t) = ∑ t I(t)·dt(2.16) where the summation has been calculated in Matlab through the cumsum() function, and the time differential d t is calculated as the time interval between two measurements, directly from the outputs of the oscilloscope, and Qc(t) = Qa(t)−max{Qa(t)}+min{Qa(t)} 2(2.17) and finally, Qc(t) is smoothed through the smooth() function, which smooths the data in the column vector, what is in function’s parentheses, using a moving average filter, being the default span for the moving average 5. 48 Chapter 2. Theoretical Basis UV-visible. A molecule absorbs a photon only if the energy of the incident radiation exactly matches the separation between two energy levels of the molecule. This condition is necessary, although not sufficient, since another set of requirements must be fulfilled (selection rules).[113] Molecules are small accumulators of energy, and can modify their energy by collisions with other molecules, or by emission or absorption of radiation. There is a set of spectroscopies called emission – often called Optical emission spectroscopy , due to the nature of the light that is emitted–, in which the molecules are previously excited, analyzing then the radiation they emit upon returning to their ground state, and this is the case that occupies this work as can be known. In this work the molecules have been excited by the passage of electrical current. Then, also called field spectroscopy or emission spectroscopy, optical spectroscopy for the study of plasmas is a characterization technique that allows to know the properties of a plasma. So, Emission optical spectrometry is a spectroscopic technique that analyzes the wavelengths of photons emitted by atoms or molecules during their transition from an excited state to a state of lower energy. Each element emits a characteristic set of discrete wavelengths according to its electronic structure. By observing these wavelengths, the elemental composition of the sample can be determined. Emission spectrometry was developed at the end of the 19th century, and theoretical efforts to explain the atomic emission spectra led to quantum mechanics. Thus, the spectral analysis is based on detecting the absorption or emission of electromagnetic radiation at certain wavelengths, in relation to the energy levels involved in a quantum transition. By means of a spectrophotometer the spectrum of the light is measured (intensity of the light absorbed, reflected or emitted as a function of the frequency or of the wavelength); and the spectra differ considerably from element to element. Previous concepts Quantum mechanics is a discipline of physics responsible for providing a fundamental description of nature at small spatial scales. Quantum mechanics can not predict the exact location of a particle in space, only the probability of finding it in different places; and, in consequence, there is the term “atomic orbital” is a mathematical function that describes the wave-like behavior of either one electron or a pair of electrons in an atom. This function can be used to calculate the probability of finding any electron of an atom in any specific region around the atom’s nucleus; it may also refer to the physical region or space where the electron can be calculated to be present, as defined by the particular mathematical form of the orbital.[131] Each orbital in an atom is characterized by a unique set of values of the three quantum numbers n , l and m , which respectively correspond to the electron’s energy, angular momentum, and an angular momentum vector component, the magnetic quantum number. Each such orbital can be occupied by a maximum of two electrons, each with its own spin quantum number s . The simple names s orbital, p orbital, d orbital and f orbital refer to orbitals with angular momentum quantum number l equal to 0, 1, 2 and 3 respectively. These names, together with the value of n , are used to describe the electron configurations of atoms. 2.3 Mechanical, Electrical and Spectroscopic Analysis 49 The ground state of a quantum mechanical system represents its lowest possible energy state; also known as the zero point energy of the system. By excited or vibrational state is understood any state with energy superior to that of the ground state. If there is more than one fundamental state, it is understood that they are degenerate. According to the third principle of thermodynamics a system at absolute zero temperature is in its ground state, since its entropy is determined by the degeneracy of the ground state. Many systems, such as a perfect crystal structure, have a unique ground state, and therefore have zero entropy when they are at absolute zero –it is also possible for the state of greatest excitation to have absolute zero temperature in the systems that exhibit negative temperature–. The ground state in physics opened the doors of quantum mechanics at the beginning of the 20th century, when the Bose Einstein Condensate was predicted; since when the atoms lose their individual identity at extremely low temperatures, it is established that the atoms are at the same quantum level. In this, as can be understood, energy is a fundamental pillar, since at very low temperatures atoms are also at their lowest energy level, while at normal temperatures atoms are at different energy levels, i.e., in an excited (vibrational) state. Electronic and Vibrational Structure of the N2 The developed experiments have been carried out with air, being the composition of this one of 78% of nitrogen, 21% of oxygen and the remaining 1% is composed of gases such as carbon dioxide, argon, neon, helium, hydrogen, others gases and water vapor. That is why it is mainly interesting to know the spectroscopic properties of N2 . This high concentration is the result of the balance between the fixation of atmospheric nitrogen by bacterial, electrical (lightning) and chemical (industrial) action and its release through the decomposition of organic matter by bacteria or by combustion.18 The nitrogen atom is a chemical element with atomic number 7 and the first element of group 15 of the periodic table. [114] Firstly, homonuclear molecules/species are composed of only one element. They may consist of several atoms, depending on the properties of the element, and some of these may have several allotropes19. Noble gases are exceptional cases of monoatomic molecules, because they rarely form bonds. However, most homonuclear molecules are diatomic: hydrogen ( H2 ), oxygen ( O2 ), nitrogen ( N2 ) and all halogens; however, not all diatomic molecules are homonuclear.[115] •Electronic structure: The usefulness of homonuclear diatomic molecules comes from the fact that they provide an excellent basis for understanding the composition of molecular orbitals by means of combinations of atomic orbitals: wave functions superimposed linearly or constructively (union orbitals) or destructive form (antibonding orbitals), 18Information of this section extracted meanly from [133]. 19Allotrope: each of two or more different physical forms in which an element can exist. For example, graphite, charcoal, and diamond are all allotropes of carbon. 50 Chapter 2. Theoretical Basis which are governed by the symmetry of the molecule and the degree of superposition atomic. [35] Regarding atomic orbitals, they can be mixed in symmetric ( σ ) or antisymmetric ( π ) molecular orbitals under rotations with respect to the internuclear axis; and, likewise, molecular orbitals can be symmetric ( g ) or antisymmetric (u) under inversion with respect to the center of mass of the molecule. The most external energy level in an atom is known as the valence band, with the electrons placed in this layer named as valence electrons. These electrons are involved mainly in the formation of bonds and in the chemical reaction with other atoms, thus being responsible for different chemical and physical properties of the element. In the case of nitrogen, as its atomic number 7, its electronic configuration is 1s2 2s2 2p3, having as valencies 1,2,+3,-3,4,5 to achieve a stable configuration with its full valence band. In the case of nitrogen N 2 , the orbitals of the 1s electrons do not overlap, however the remaining 10 valence electrons occupy molecular orbitals that are the result of the mixture of the 2s and 2p orbitals, including s-p interactions due to their proximity, as we can see in Fig. 2.18.a. (a) Electronic configuration.. (b) Vibrational states.. figure 2.18 Electronic and Vibrational structure of N2molecule. Regarding a), nitrogen molecules perform vibrations according to the particular occupation of the electrons in the binding or anti-binding molecular orbitals determining the length and strength of the bond. The energy levels of the corresponding quantized vibrations of the molecule, described by the vibrational quantum number of the molecule v, are superimposed on the levels of electronic energy. Source of the image: [0] Respect to b), the red lines represent the curves of potential energy of the vibrational states C3Πand B3Π. They are the transitions responsible for the most intense peaks that are observed in the spectrum, and which are represented with the number ˜ ν0ν00.Source of the image: [116] . 2.3 Mechanical, Electrical and Spectroscopic Analysis 51 •Vibrational structure: In addition, the discrete energy structure of the molecular electron can be divided in a superposed spectrum of vibrational states. Due to the fact that each electronic state is characterized by different lengths and link strengths (by different electronic distributions), the molecular oscillator will have a variety of potential energy curves associated with the different vibrational states, each with a range of vibrational levels spaced differently, organisationally represented by quantum numbers: ν=0,1,2... The designation negative and positive groups (or bands) refer to the occurrence of these bands in the negative glow or the positive column, respectively, of an electric discharge. The positive groups are due to the neutral molecule, the negative groups to the singly positively charged molecular ion.[134] In the present work, the Second Positive System of vibrational states of nitrogen will be studied: C3Πu and B3Πg, whose potential energy curves are illustrated in Fig. 2.18.b. So, to observe the vibrational levels, the nitrogen molecule must be first excited into metastable radiative states, such as via a transition from the molecular ground state X1Rginto C3Πu. As already explained in Section 2.2, the discharges in gases get to excite the species causing collisions between electrons and ions, and as a consequence, the species decay energetically through the emission of light forming glow discharges. If that potential difference is maintained so that the plasma self-sustains, new accelerated free electrons will be enveloped in new impacts by excitation and ionizations. Nitrogen is easily excited at one of its ν0 levels of the C 3Πu state. The discharge region may also contain other species of molecular nitrogen, such as positive ions (N + 2 ) produced by direct electron impact ionization of the neutral molecules from the ground state X 1∑g , which is predominant in the gas. However, depending on the energy of the electron impact, the ions may be knocked into their vibrational ground state X 2∑g (which could be later excited) or directly into an excited B2∑ustate: N2(X,ν=0)+e→N+ 2(X,ν0)+2e(2.36) where electron energy is higher than 15.6eV. N2(X,ν=0)+e→N+ 2(B,ν0)+2e(2.37) where electron energy is higher than 18.5eV. Derived from this reasoning, the excited states will deca in a spontaneous way spawning an emission sprectrum with line intensities proportional to the population of the upper bands N2(C, ν0) or N2(B, ν0): N+ 2(C,ν0)→N2(B,ν00) +hν(2.38) N+ 2(B,ν0)→N+ 2(X,ν00)+hν(2.39) 52 Chapter 2. Theoretical Basis The vibronic radiative transitions occurring between various vibrational states of N 2 and N + 2 will be recorded into a spectrum representing the emission bands in Section 4.2. How to analyze the data from spectroscopy Regarding how “to read” the nitrogen emission spectra, it is to be understood that this is a source of information about the potentials associated with the vibrational states of the molecules. Therefore, the spectrum transition characteristics must first be extracted and organized; for whose correct reading it is first necessary to know the selection rules that govern the transitions leading to the pattern of intensity peaks shown within each emission band progression. The probability of transition between vibrational levels depends on the degree of overlap between the wave functions of two states, in addition to obvious contributions such as the electron occupancy of the level ν0 and the probability of electronic transition (which is constant for transitions within the same system, as for example, the second positive system C3Πu→B3Πg ). The overlap is quantified by Franck-Condon factors weighting the vibronic transition probability. So, to identify the different head bands of the nitrogen, it has to be compared the wavelength of these results with the tabulated table included in Fig. 2.19[ 142 ], which represents the pattern that nitrogen (major molecule in an air plasma, easily excitable) always follows. 2.3 Mechanical, Electrical and Spectroscopic Analysis 53 figure 2.19 Tabulated table for Nitrogen emission spectrum for second positive system states. Source of the image: [142].. 3 Design and construction of DBD plasma controllers Different devices have been built, from which electrical and spectroscopic characterizations have been developed, in order to draw conclusions. All of them have been the result of four different phases of study, in which a series of different characteristics have been taken into account as a research subject, and whose conclusions have defined the line of action of the next phase, respectively. So, this series of devices have been designed based on the conclusions that were drawn from the previous analyzes, thus continuing with the most fruitful research line. In this chapter, specifically in Section 3.1, are explained all the phases –and consequently, its devices– and the reasons why the different configurations were carried out; that is, the design and construction of the DBD Plasma Actuator. Each one of these devices used will be presented in order to characterize them to understand the differences between the data obtained from their analysis, which will be shown in Chapter 4. Also, in Section 3.2 and 3.3 some considerations regarding the manufacturing of the devices and about the experimentation are collected respectively, and, finally the experimental procedure and its necessary resources and considerations will be shown in Section 3.4. 3.1 Phases of the study Four phases of study have been considered to develop this work, through which one wanted to find the most powerful and energy efficient configuration, finding the device that achieves a surface covered with plasma of greater length when it is sufficiently energized as function of the magnitudes/properties that have been studied in this project. All of them, except experiments of Phase 4, have been studied at atmospheric pressure and atmosphere in calm as general strategy. The description of these phases is presented below in the next sections. Before presenting the mentioned enumeration of the phases, remember that the reader can go to Section 2.2.2 to know the physical characteristics of the different materials that are used in the construction of the devices. Besides, after the presentation of the different phases, in Table 3.1 all built devices are presented. Having said this, the different 55 56 Chapter 3. Design and construction of DBD plasma controllers phases of study are called like the following way –included in Section 3.1.1-3.1.4. –: •Phase 1: Basis functional study of DBD actuator with the aim of generating the largest possible amount of surface covered with plasma •Phase 2: Influence of dimension of gap •Phase 3: Active multi-electrode configurations •Phase 4: Experimentation with airflow and reduced pressure The following devices included in Table 3.1 have been built, and, in general, the nomenclature used to size the devices is indicated in Fig. 3.1. The dimensions that characterize the devices built in Phase 1 and Phase 2 are detailed in Table 3.2, whose structures can be seen in the schemes contained in Fig. 3.2; dimensions of the rest are detailed in their corresponding Section. Table 3.1 List of experiments and its corresponding devices.. DEVICES EXPERIMENT Substratum Conductive Configuration Picture Exp.1, Phase 1|Device 1 Quartz (SiO2) Type A Exp.2, Phase 1|Device 2 Quartz Type B Exp.3, Phase 1|Device 3 Quartz Copper (Cu) Type C Fig. 3.3 Exp.4, Phase 1|Device 4 lithium niobate (LiNbO3) Type A Exp.5, Phase 2|Device 5 Quartz Type D, gap=5mm Exp.6, Phase 2|Device 6 Alumina (Al2O3) Type D, gap=5mm Exp.7, Phase 2|Device 7 lithium niobate Type D, gap=5mm Exp.8, Phase 2|Device 8 Quartz Type D, gap=2.5mm Exp.9, Phase 2|Device 9 Alumina Type D, gap=2.5mm Exp.10, Phase 2|Device 10 lithium niobate Type D, gap=2.5mm Exp.11, Phase 2|Device 11 Quartz Copper Type D, gap=1mm Fig. 3.4 Exp.12, Phase 2|Device 12 Alumina Type D, gap=1mm Exp.13, Phase 2|Device 13 lithium niobate Type D, gap=1mm Exp.14, Phase 2|Device 14 Quartz Type D, gap=0mm Exp.15, Phase 2|Device 15 Alumina Type D, gap=0mm Exp.16, Phase 2|Device 16 lithium niobate Type D, gap=0mm Exp.17, Phase 3|Device 17 Quartz TFO, Fluor tin oxide (SnO2−F) Type E Fig. 3.5 Exp.18, Phase 3|Device 18 Quartz TFO Type F Fig. 3.5 Exp.19, Phase 3|Device 19 lithium niobate Copper Type G Fig. 3.6, 3.7 Exp.20 Phase 4|Device 19 Device 19 faced with airflow Fig. 3.8 Exp.21 Phase 4|Device 19 Device 19 at reduced pressure Fig. 3.9 3.1.1 Phase 1: Basic functional study of DBD actuator In this first phase it has been studied, mainly, the efficiency of different configurations of a DBD Actuator with, as explained, the aim of generating the largest possible amount of surface covered with plasma, on the dielectric layer. For this purpose, reactors with three different configurations have been powered, and in addition, the effect on two different dielectric materials has been checked: quartz (SiO2) and lithium niobate (LiNbO3). Three types of devices (Type A, B and C –see Fig. 3.2 to know the dimensions–) have been built and powered according to the configuration of the electrodes –in all devices, has been avoided that electrodes extend to the limits of the substrate to ensure that they do not interact, creating edge effects which are not desired–. That is: 3.1 Phases of the study 57 figure 3.1 Nomenclature to size the devices. The green zone represents the face of the active electrode where the plasma is generated most immediately.. Table 3.2 Dimensions of devices of Phase 1 and 2. Thickness of the dielectric discs used (h) are, respectively for lithium niobate, quartz and alumina: 1mm, 1mm and 2mm. The values belong to the same order of magnitude, and their influence on the discharge has not been studied, however the influence is not believed to be significant, as it does not affect the conclusions of the trend included in this paper.. DIMENSIONS OF DEVICE CONFIGURATIONS Electrode material, thickness (t) Copper, 0.025mm Configuration type Type A Type B Type C Type D Width of the exposed electrode(s) (a) 12mm 12mm 10mm 10mm, 10mm 10mm 10mm Width of the encapsulated electrode (b) R>9cm R>9cm R>9cm R>9cm 10mm 10mm 10mm Gap (g) - - 5mm 5mm 5mm,2.5mm,1mm,0mm Length of discharge zone (s) >7cm >7cm 6.5cm 6.5cm 7cm 7cm 7cm Dielectric material Quartz lithium niobate Quartz Quartz Quartz lithium niobate Alumina Type A: Device with a single active electrode, located on the upper face of the substrate, centered and shaped like a rectangular strip, and the ground electrode under the circle-shaped substrate that occupies practically the entire surface –see Fig. 3.2.a–. The objective of this configuration is, fundamentally, to verify that the discharge causes the generation of plasma around the active electrode, observing the characteristics of the same depending on the dimension of the different magnitudes that determine the energization chosen to apply to the device during the experiment. Type B: Device with three electrodes in the form of rectangular bands on the upper face of the substrate, of which the two external bands are active and the center is a ground electrode, and a single ground electrode, with a circular shape, at the bottom of the substrate as the Type A configuration –see Fig. 3.2.b–. The purpose of this device is to try to cover the space between the different electrodes of the upper face of the substrate, in the form of "bridges" between them. In the beginning it was decided that the electrode of the 64 Chapter 3. Design and construction of DBD plasma controllers of the dielectric, it may or may not be welded on the copper once it is already adhered to the dielectric to avoid damage; since, for example, the first time I worked on the lithium niobate, it could not withstand the thermal shock when this was done, breaking the disc and having to be replaced, so after this incident the welding was carried out on the copper prior to its assembly on these discs in question. 3. Not all the cables used have been of the same type. Within Phase 1, for the Type A device, common grounding wires were used for both electrodes, while for Type B and C devices, thinner cables were used. For Phase 2, specific cables for high voltage have been used –see Fig. 3.4.c–. 4. As I was progressing in the construction of the devices, I tried to achieve smaller welds, so that the area of contact with the high voltage was smaller, and interfere, as a result, as little as possible in the ignition of the plasma. 5. In general, the substrate discs have been reused for the different experiments. 6. The dimensions of the discs have not been strictly taken into account, since they did not necessarily have to be the same in all the materials, since this aspect did not affect the experiment. 7. The discs of dielectric material in the laboratory all have the same surface finish on both sides, with the exception of lithium niobate, in which one of the faces is polished and the other is not; for this reason, the active electrode has always been placed on the unpolished surface. Regarding the quartz discs, both faces are polished, while in alumina none of them is, although the roughness is minimal and comparable to the unpolished face of the lithium niobate discs. 8. Ground electrode is not embedded in the substrate, but adhered to the underside of the disc. 9. The electrodes never extend to the edge of the dielectric to avoid unwanted edge effects. 10. Related to nomenclature aspects, when one talks about the upper face of the substrate it refers to the face where the active electrodes are located, while the lower face or underside is the face where only the ground connection is located. 3.3 Considerations regarding the experiments 65 3.3 Considerations regarding the experiments 1. The generated plasma emits fluorescent light of violet color due to nitrogen mainly. 2. Regarding how to recognize/establish the breakdown point in which to save measures, a visual criterion has been applied to perceive a significant level of luminosity, this being the minimum to which I can perfectly perceive the discharge lines between the active electrode and the substrate if I observe the disc in profile (that is, facing the vision to the thickness of the device), since it is more difficult to try to be exact in said recognition if I observe the disc in plant (from above), especially the less translucent is the substrate material. It has to be commented that due to the existence of concentrators such as the corners of the electrodes, the welding zone to the high voltage cables, the possible irregularities and roughnesses that the copper film presents, fruits of the construction of the device, or the degradation of the adhesive that keeps the electrodes adhered to the discs either by overheating or by drying them during the time they remain inactive stored in the laboratory, these elements constitute uniformities in the electrodes that affect the way in which the current is distributed over them , as well as, consequently, the moment in which the plasma appears around each local point of the perimeter of the electrodes. Consequently, in all experiments it is common that approximately half a kilovolt –at least– before the breakdown point considered by my criteria, plasma appeared in very localized points of the electrodes progressively, as are the corners mentioned above. That is why these previous appearances have not been considered representative of the phenomenon studied, and were taken simply as a reference (although observed at all times to analyze its repercussion). The breakdown point was considered when a significant part of the contour was turned on, and when said parameter was repeated throughout the experiment, but logically said pattern is not the same for each configuration, but a level considered equivalent. In some cases this critic level corresponded almost instantaneously to practically the whole contour, but in other occasions, if one waited for the contour to fill up, there were large areas that remained long (from a relative perspective) time on, reaching greater luminosity; for what was the appreciable ignition of these concrete areas, whenever their extension was significant, the one considered as point of ignition. It was also thought that perhaps a finer design of the electrodes would improve the uniformity in the generation of the plasma. It is useful to cover the methacrylate box with an opaque material so that the reactor remains dark in the viewing angle and the discharge is easily noticeable at its minimum level of light intensity. The atmosphere of the room should be as dark as possible.This fact adds a certain degree of subjectivity to the experiment since it is not possible to maintain an ambience in the laboratory completely without light. Derived from the above considerations, the most complex measure to determine, for me, has always been the first one at a frequency of 1 kHz. 3. The experiment generates ozone, so precautions must be taken to divert this flow as much as possible and thus aspire as little as possible. 66 Chapter 3. Design and construction of DBD plasma controllers 4. Due to the ionization of the gas, there is emission of sound from the experiments from a certain level of charging, especially intense and acute in the case of experiments with lithium niobate, but this aspect has not been considered as a useful reference as measuring tool. 5. The heating of the disc or level of polarization achieved can influence the breakdown point of the experiments when it remains energized for a long time, since the experiments extend at least one hour. That is why the experiments should be done with the maximum agility possible. At least, while the plasma is active, the process of discharge the device must also be slow, progressively decreasing the polarization, avoiding forcing the crystalline network of the substrate. 6. Prior to starting each study at a certain frequency, the device must be electrically discharged. That is why after loading the reactor for each measurement, it must be completely discharged. The charging process must be done at a moderate speed (with a sensitivity level in the ascent of the order of ten volts, minimum), taking care that the substrates do not suffer damage or short circuits. 7. For the development of experiments at reduced pressure it was necessary to take special care in opening the air inlet valve of the camera between experiments, to clean the atmosphere in which they were carried out, and thus avoid an atmosphere enriched in ozone. 3.4 Experimental Procedure 67 3.4 Experimental Procedure The performance of the process was analyzed by varying both the input voltage, V(t) , and the characteristic frequency, f . The experimental system used for the electrical characterization is showed in Fig. 3.10: a function generator (Stanford Reserach System, Model DS345) and a high voltage generator (Trek, Model 20/20 C) have been needed to power the controller, and as a reader of voltage and current signals in time, V(t) e I(t) respectively, an oscilloscope (Agilent Technologies Oscilloscope DSO X 3024A) was used. The DBD controller is inserted in a methacrylate box to carry out the experiments as a safety measure. Likewise, this is usually covered so that the external light of the ambient does not penetrate into the cavity and the plasma and, more specifically, the breakdown point can be appreciated (since although the discharge is perceptible to the naked eye, the necessary conditions to appreciate it in detail require an atmosphere of darkness, for which the laboratory ambience was adapted closing all possible light sources although this can not be total), moment in which the measures are saved –see Fig. 3.11–. For the case of the experiments under reduced pressure, these were carried out inside the spheric camera which can be seen in the Fig. 3.9, marketed by Kurt J. Lesker Company. figure 3.10 Experimental system for electrical characterization. A function generator (Stanford Reserach System, Model DS345) and a high voltage generator (Trek, Model 20/20 C) have been needed to power the controller, and as a reader of voltage and current signals in time, V(t)eI(t)respectively, an oscilloscope (Agilent Technologies Oscilloscope DSO X 3024A) was used.. Following with the typical performance values of the DBD controllers –see page 32, Fig. 2.2– we have worked in a range of frequencies of 1-15 kHz taking measurements of each unit of kHz, this is, 15 measurements for each device; with a sinusoidal signal with offset and phase both null. Thus studying the influence of frequency on the electrical results of the experiments. Considerations to take into account for the visual recognition of the breakdown point during the experiment –see Fig. 3.11 to check these tips–: 1. Ensure that the monochromator remains in a stable position and correctly pointing to the target. 2. Guarantee dark ambient, especially important to perform the spectroscopic analysis. We try to keep the 68 Chapter 3. Design and construction of DBD plasma controllers (a) Plasma off.. (b) Plasma on.. (c) Plasma on.. figure 3.11 DBD reactor connected for the experiment in different positions. Pictures taken in different experiments to appreciate the way in which the box was covered (in the image it is partially covered). Likewise, the relative position occupied by the reactor inside the box is indifferent to the experiment, this depends in each case of the place where the connection cables have been placed, or the size of the device, always trying to find the more comfortable and stable position, which guarantees the correct execution of the experiment and for visual recognition of the breakdown point. The two images starting on the left correspond to one of the experiments performed on a Type A device mounted on quartz, while the first starting on the right is a Type D configuration on lithium niobate.. 3.4 Experimental Procedure 69 laboratory dark during the experiment, using an opaque and insulating cover on the methacrylate box if necessary. 3. Position the device so that the active electrode is always visible, as well as orientate it so that the area where the plasma is most immediately generated is in front of the observer; this is, in the case of Type D devices, around the face of the active electrode that is closest to the ground electrode, from a plan view, indicated in green in Fig. 3.1. 4. Ensure that the field of vision has access to a perspective of the profile of the device in the thickness direction, so that the start of the generated discharges can be appreciated more easily, when these are not of great intensity. Especially in less translucent devices, where it is more complicated to perceive said point of ignition of the plasma. 5. One must ensure the stability of all the elements involved in each experiment, both to ensure the general safety of the experiments, and to ensure that the spectroscopy is done correctly. Different insulating support structures have been used (not designed for this purpose but that already existed previously in the laboratory for uses unrelated to this application), to facilitate locating the device in the way that best suited each occasion based on these conditions and of the geometry of each device, as can be seen inside the methacrylate box. Regarding the emission optical spectroscopy carried out in some of the experiments that have been carried out, this test was developed with a Jobin-Yvon FHR640 monochromator 2 . Respect to the resolution of the measures saved, the spectroscopy was carried out at three different levels of resolution, whose results and argumentation are reflected in Section 4.2. The degree of precision of measurement of the spectroscopy depends mainly on the configuration of these parameters, as well as on the calibration of the equipment; as well as it is recommended to do a background spectroscopy to ensure that the only thing that is captured is noise or identify possible elements that are foreign to the plasma that will later be measured –see Fig. 3.12–. For measures referred to quartz-device number 3 and lithium-niobate-device number 16 was used a diffraction network centered at 330nm and with a density of lines in the network of 1201lines/mm. The opening/gap of the entrance and exit slits was 1mm; the resolution of the monochromator (step value, i.e., how much interval in wavelength a point is referred to) was 0.5nm and the integration time was 0.5 s. For spectroscopy of lithium-niobate-device number 19 to measure the general spectra, the values were: •Slit in, slit out: 1 mm •Step: 0.5 nm •Integration time: 0.5 s 2 A monochromator is an optical device that transmits a mechanically selectable narrow band of wavelengths of light or other radiation chosen from a wider range of wavelengths available at the input. 70 Chapter 3. Design and construction of DBD plasma controllers and finally, with high-resolution for the 360-390 λ interval (to have accurately values for calculating vibrational temperature), for the five measures of device 19 depending of voltage those parameters were: •Slit in, slit out: 0.5 mm •Step: 0.2 nm •Integration time: 0.5 s To perform the spectroscopic measurements, the different devices were placed inside a dark chamber, for this purpose it was ensured that the methacrylate box in which the experiments were developed was completely covered with opaque material. In this way it is ensured that the optical measurements of the discharge were not disturbed by external signals. It was made use of a fiber optic cable connected to the monochromator specified above; The monochromator is responsible for digitizing the signals collected with the fiber. Subsequently, using the OriginPro software, the spectroscopic graphics were represented. figure 3.12 Background spectroscopy.. 4 Analysis of the results The oscilloscope provides two output files that contain, respectively, the discretization of the Voltage V(t) and Current I(t) signals at the point of choice, for as many periods as indicated on the screen at the time of saving the data. From these data, Matlab has been used to convert them to international system measurement units, treat them and obtain other significant electrical magnitudes for the experiment, and finally the output files of said data processing have been open with OriginPro, a computer program for interactive scientific graphing and data analysis produced by OriginLab Corporation, through which to build the different graphs that are shown in the document. In this chapter, the results and conclusions of the electrical and spectroscopic characterization are presented, respectively in Section 4.1 and 4.2 –it is remembered that Table 3.1 contains an index of experiments classified by phase of study, with the main characteristics of the devices–. Finally, in Section 4.3, some things to keep in mind about the materials during the experiments was collected. Reminding that for electrical analysis the measures were saved at appreciable ignition point of plasma for each selected frequency (from 1 kHz to 15 kHz). Even though an exhaustive comparative analysis of all the experiments has been carried out, in the report will show only a representative sample of them, to avoid an extremely long report. These results will be exposed differentiating by operational parameter. In reference to the spectroscopic analysis , the obtained results will be exposed in the Section 4.2. These data, unlike the electrical analysis, do not necessarily had to be recorded at the point of ignition of the plasma, but at a level in which sufficient luminosity is guaranteed so that the spectroscopy can be successfully developed. All the measures have been saved averaged and without averaging; however, for the data processing, averaged outputs have been used whenever possible. Likewise, a smoothing of said outputs has been used (through the function of matlab smooth()). It is recommended to see the content of Fig. 4.1 and 4.2. It is recalled that the fact of using alternating current (AC) allows to work continuously charging and discharging the dielectric successively. According to [ 125 ], the discharge current is composed by the displacement current and conduction current, where the magnitude of the latter is smaller than that of the former in the test. Meanwhile, 71 Analysis of results72 Chapter 4. Analysis of the results figure 4.1 Some aspects about comparison between Voltage and Current signals I. Comparison of measures smoothed, averaged and without averaging. As the frequency increases, the differences decrease; but always differences are more significant for the current signal than for the voltage signal. Table 3.1 contains an index of experiments. . figure 4.2 Some aspects about comparison between Voltage and Current signals I. Offset between V(t)and I(t), and variation by frequency increase. Table 3.1 contains an index of experiments. . intensity of the discharge is stronger during the positive half-cycle. The current leads the voltage by 90, performing a typical characteristic of the capacitive discharge. During these experiments, the current signal does not suffer large deformations due to the characteristics of the dielectric, obtaining a more homogeneous discharge than in another bibliography –see Fig. 4.3.a–, being in our case the filaments of the current signal very small. Likewise, in these experiments there is no noise in the signal either. However, it was appreciated that the greater the efficiency of the device, the faster its shape varies, tending to get square-form. This was most noticeable especially when the experiments were carried out under reduced pressure on the multi-band configuration on lithium niobate (device 19, Experiment 21). It is recommended to see Fig. 4.3.b to verify this affirmation. Analysis of results Analysis of results 4.0 Experimental Procedure 73 (a) Typical waveforms of discharge voltage and current. Source of the image: [125].. (b) Comparison of deformation of Current signal by generation of plasma in different experiments.. figure 4.3 Informtation about deformation of the Current signal. Table 3.1 contains an index of the experiments carried out.. Analysis of results 80 Chapter 4. Analysis of the results 4.1.2 Results Next, in the following sections, the behavior patterns that are extracted from analyzing all the operating parameters found in Table 4.1 are exposed. It has been discovered that there is an optimum frequency around 5 kHz, for which the accumulated charge is maximum , and that can be inferred from the electrical behavior of the devices for each parameter studied. Table 4.1 A. Operational parameters analyzed with the electric analysis. Parameters Frequency V Power factor PF Voltage V Phase difference φ Current I Gap g Accumulated charge QcDielectric material - Consumed power P Number of electrodes - Effective capacitance C Pressure p Impedance Z A. Behavior of Voltage and Current with frequency Ignition voltage5 decreases when frequency of the signal that is applied to the experiment increases. The decrease is more pronounced between low frequencies and ends up drastically decreasing its slope around 5 kHz for all the cases studied, reaching a zone of apparent stabilization. Voltage and current have a proportional behavior. The ignition current increases when frequency of the signal that is applied to the experiment increases. The increase more pronounced between low frequencies and ends up slightly decreasing its slope around 5 kHz. It must be remembered that the electric current value shown corresponds to a total value and not a real value; since, as we saw in the theory of Chapter 2, electric current has two components, a conductive due to the movement of electrons (known as real) and another of displacement ∂E/∂t due to polarization. The lithium niobate presents a much more efficient behavior from the point of view of the voltage consumed at the time of ignition. However, the generated plasma is located around the same current values for the three dielectrics studied. That is, voltage and current increase with frequency until reaching a saturation value, and around 5 kHz its variation slope changes. These trends occur independent of the type of device. Below, in Fig. 4.7- 4.9, are figures that corroborate this conclusion for phase 1 –showing the peak-to-peak values of both parameters– 0mm-of-gap 5 The measurement of the signals tends to be more stable as the frequency of the experiments increases, since the appreciable breakdown point appears more intense and abrupt. 4.1 Electrical analysis results 81 differentiating by material, and lithium niobate differentiating by gap, respectively. figure 4.7 Voltage and Current results: example with Phase 1 outputs. Table 3.1 contains an index of experiments.. figure 4.8 Voltage and Current results: example with Phase 2 outputs for 0 mm of gap, differentiating by material: quartz, alumina and lithium niobate. Table 3.1 contains an index of experiments.. 82 Chapter 4. Analysis of the results figure 4.9 Voltage and Current results: example with Phase 2 outputs for lithium niobate. Table 3.1 contains an index of experiments.. 4.1 Electrical analysis results 83 B. Lissajous figures. Behaviour of Accumulated charge and Capacitance with frequency The shape of the Lissajous figures varies with frequency in a pattern that is repeated throughout the rest of the experiments; in general, they have an elliptical appearance. For representations of I(t) vs. V(t) , the shape of the figure is more flattened in the direction of the current and more elongated along the horizontal axis representing the voltage for low frequency values, which is equivalent to higher ignition voltages and low currents, and this appearance is gradually reversed as the frequency increases, becoming more energy efficient as the current through the device grows and the ignition voltage decreases. Regarding the representation of Q(t) vs. V(t) , although the shape of the figure is still approximately elliptical, the evolution with the frequency is different from the previous representation due to the fact that the accumulated load decreases as frequency increases, thus also corroborating an efficient growth in the performance of the device, which means that the area enclosed by these figures –and equivalent to, being multiplied by frequency, the average power P – grows with frequency increase. Also, as a consequence of the behavior of the electric current, it is noticeable that as the frequency increases, the maximum accumulated electric charge points are not presented for the maximum voltage amplitude of said signal. The variations between the different figures with the frequency is more prominent for low frequencies, while it tends to adopt a more similar behavior starting from, approximately, 5 kHz. And respect to the zero-charge6 , it has a trend of increasing-decreasing when frequency increases, and the turning point (a maximum) is located around 5 kHz for all the cases. So, it is clear that said frequency is the optimum frequency from a point of view of energy consumption because the accumulated charge in the device is maximum when there is no applied voltage, on this way less voltage is needed to generate plasma. To illustrate these comments, Fig. 4.10 shows a representative selection of Lissajous figures that have been built from the electrical signals saved from the experiment. The trend of zero-charge, Q0can be also checked in Fig. 4.11-4.12. This fact is significant since the accumulated charge is the factor that decreases the ignition voltage. In this way, although the ferroelectrics have a dielectric constant that varies with the voltage-dependent ε(V) (which leads therefore to a capacitance also dependent on the voltage C(V) ), it takes place a maximum charge capacity reached with this optimum frequency, around 5 kHz –see equation 4.1–. The devices have a maximum value of accumulated charge when the device is off (called zero charge Q0 ), that is, they present a maximum capacitance at the optimum frequency value, and in consequence they need less voltage to generate plasma when the discharge cycle starts again. C=Q V;C=ε0εr S d(4.1) where C is capacitance, Q is electric charge, V is voltage, ε0 is the vacuum dielectric constant, εr is the dielectric constant, dis the thickness of the dielectric and Sis the dielectric surface/area. 6Referred to zero-accumulated charge, charge when voltage is null. 84 Chapter 4. Analysis of the results figure 4.10 Lissajous figures. Representative results. Experiments 8 and 10 for all the frequencies tested. The figures of Lissajous are collected for a range of frequencies fof 1 to 15 kHz. The representation of Charge Q(t) vs. Voltage V(t) has vital importance since its enclosed area offers directly the value of the Average Power P of the period for each frequency if this area is multiplied by frequency. Table 3.1 contains an index of experiments.. 4.1 Electrical analysis results 85 figure 4.11 Zero Charge results: Phase 1 sample. Table 3.1 contains an index of experiments.. figure 4.12 Zero Charge results: Examples of Phase 2 results –0 mm-gap sample in function of dielectric and lithium niobate sample in function of gap. Table 3.1 contains an index of experiments.. 86 Chapter 4. Analysis of the results C. Behavior of Impedance and Consumed Power with frequency The devices act as capacitors, in this way it has already been said that there are two types of current, one due to the polarization of the dielectric and another corresponding to the current of electrons in the circuit. Consumed power is presented as a monotonically increasing parameter with the increase in frequency, and responds to the fact that, like the current values shown, it represents the total power that passes through the circuit, not to the real component (IR 2 ), since this total consumed power contains the sum of both components. That is to say, the part of power associated to the displacement current due to polarization ( ∂E/∂t ) –and not to the electrons– that travel the circuit is the responsible for this result, since it is translated in terms of power as reactive power. It can also be verified how correct experimental results have been obtained since the power extracted from the area enclosed in the Lissajous figures coincides practically in full for all the experiments developed with the integral way to obtain the power. Both aspects can be checked in Fig. 4.13 and 4.14. Obviously, although the trends are repeated for all the experiments, the values are different according to the experiment, the lower the dielectric constant, the higher the power consumption, as well as the power consumed the lower the gap for the same material –see Fig. 4.14–. figure 4.13 Consumed power: Integral process vs. Lissajous figure’s area enclosed multiplied by frequency. Phase 1. Table 3.1 contains an index of experiments.. The behaviors with the frequency of the voltage, current and power studied do not allow to see so clearly the existence of an optimal operating point in terms of the frequency of ignition, in comparison to other study parameters. However, the behavior of the impedance makes it possible to know the existence of an optimum frequency value according to the evolution of its real and imaginary components with frequency, that in this occasion they can be studied individually, and reflect the behaviour of the components of the current. While the real component of the impedance remains practically stable regardless of the frequency value, the imaginary component that has a high value at low frequencies decreases sharply (with a high value of the slope for the lower frequencies), until both 4.1 Electrical analysis results 87 figure 4.14 Consumed power: Integral process vs. Lissajous figure’s area enclosed multiplied by frequency. Some examples of Phase 2. Table 3.1 contains an index of experiments.. impedance components of the impedance are crossed around 5 kHz of frequency, the optimum frequency. From this point on, the real component has a higher value than the imaginary one, and the variation of both is practically zero during the rest of the tested frequency range up to 15 kHz. This fact tells about an increase in the efficiency of the device in generating plasma from a certain frequency value. See Fig. 4.15 and 4.16, for results of Phase 1 and 0mm-gap-devices of Phase 2 respectively, to check this affirmation in a representative sample of the experiments. From an energy efficiency point of view, this event is tremendously positive, since from the optimum frequency, the device works mainly with real and not imaginary impedance. Regarding the power factor , it is as representative as the impedance, since they come from the same, it simply constitutes a different way of showing the results. That is why it has finally been decided to include a single example of representation, contained in Fig. 4.15. it increases progressively but it has a slope change around 5 kHz, from 88 Chapter 4. Analysis of the results which its growth begins to approximate towards a horizontal asymptote. The closer that value is to the unit, the more efficient the device will be, because it implies that the number of watts of consumed power is the same as the number of volt-amperes consumed –that is, a unit value for the power factor is equivalent to a percentage of 100% in which there are no energy losses–. At 5 kHz, there is an inflection point where also the "crossing" occurs with the representation of the phase shift, which decreases with frequency increase, change its slope at 5kHz, and logically to become practically zero for frequencies around 15 kHz. figure 4.15 Impedance and Power factor results: Phase 1 sample. Table 3.1 contains an index of experiments.. figure 4.16 Impedance results: Phase 2 sample on 0mm-gap-devices. Table 3.1 contains an index of experiments.. 4.1 Electrical analysis results 89 D. Influence of Gap Gap significantly affects the ignition values, being the experiments developed at null gap those of greater electrical efficiency. This tendency is reflected in all the data analyzed, breakdown voltage sharply decreases as gap decrease – see Fig. 4.17 representing this influence at optimum frequency sample–. So, a decrease of gap favors more capacitance for the device, because capacitance is obtained as indicated by expression 4.1. Likewise, it has not been successful for the conditions of the study to generate plasma surface greater than approximately one millimeter. figure 4.17 Influence of gap in the ignition voltage.Example at optimum frequency of 5 kHz.. 96 Chapter 4. Analysis of the results 4.2 Spectroscopic analysis results As for why optical emission spectroscopy (OES) is useful, as we have already read in Section 2.3 it allows to directly obtain the composition of the plasma from its “bands or lines” of intensity. In addition, if more exhaustive analyzes are applied, as mentioned, it offers information about the excitation or ionization processes and plasma temperatures. Thus, depending on the type or value of temperature that is established, different properties of the plasma can be characterized. The objective of this spectroscopic analysis is not to develop a complete study of the field, but to demonstrate the existence of active species, generated by the plasma, in the air surrounding the device; allowing thus to develop a deeper characterization of the generated plasma; as long as air plasma has been generated, and therefore it must be obtained nitrogen spectra, because N2species are, mostly, easy to excite. Each device generates a different plasma that can be characterized with spectroscopy. At atmospheric pressure, in a dark ambient and at a plasma intensity level considered high enough for the fiber make a correct measurement, spectroscopic measurements were taken of the emission spectra of devices number 3, 16, and 19. In the case of Experiment 3, measurements were taken at 7, 8, 9 and 10 kHz; in the case of Experiment 16 the measurement took place at 15 kHz frequency; and for the last case spectroscopy was carried out at five different voltages for the same frequency, the optimum frequency (5 kHz). In the optical emission spectrum, the band heads associated with the element constituting the plasma are to be observed8. 4.2.1 Results At the end of the spectroscopic analysis, Section 4.2 contains the Table 4.4 with the main results obtained. As a representative sample, is included in Fig. 4.25 the optical emission spectrum 9 of the plasma generated with device 19, indicating in it the corresponding assignment to each nitrogen’s vibrational transition 10 . In this figure, there is the spectra corresponding to device 19 at optimum frequency. The vibrational quantum numbers ν0−ν00 of each transition are indicated above the band head. Under the spectrum is the diagram of the transitions observed in it between the vibrational levels of N2. 8 What is really observed is the envelope of a considerable number of bands of smaller size representing both the vertical and rotational displacements of the electrons, because both (and especially the latter, the rotational ones) are very complicated to observe with greater degree of detail with a usual monochromator. To get an idea, we would have to have a monochromator of measurement characteristic of around a minimum of two meters, to obtain a resolution at that level with respect to rotational vibrations. This is why we usually work with band heads. 9To know the resolution data of measuring of each spectroscopy showed in this section, it is recommended to go to page 69. 10 It is important to say that the values contained in the tables included in Fig. 2.19 correspond to the equilibrium, so they are only useful for comparative purposes. 4.2 Spectroscopic analysis results 97 figure 4.25 Assigned head bands for Nitrogen second positive system states in the plasma’s emission spectrum. Device 19: lithium niobate, type G. Table 3.1 contains an index of experiments.. In Fig. 4.26 and 4.27, a study is collected according to the voltage for that same device and optimum frequency with an increase in resolution of the measure in order to, also, to calculate the vibrational temperature according to Section 4.2. It has been found that the pattern of placement of the band heads with respect to the wavelength never varies, since it is characteristic of the element in question, in this case, nitrogen. Likewise, the increase in frequency or voltage translates into a significant increase in the intensity of emission. 98 Chapter 4. Analysis of the results (a) Device 19. Zoom of b).. (b) Device 19. 370-390nm.. figure 4.26 Head bands for Nitrogen second positive system states in the plasma’s emission spectrum: Detail in function of voltage at high-resolution (370-390nm of wavelength). Device 19: lithium niobate, type G. Electrical value for saved measures collected in Fig. 4.27. Table 3.1 contains an index of experiments.. 4.2 Spectroscopic analysis results 99 figure 4.27 Head bands for Nitrogen second positive system states in the plasma’s emission spectrum: Electrical value for saved measures of Fig. 4.26. Device 19: lithium niobate, type G. Table 3.1 contains an index of experiments.. Examples of the influence of frequency that corroborate the previous statement (an increase in frequency results in an increase in intensity) can be found in Fig. 4.28 developed from measurements taken on device 3 (quartz, type C) for different values of frequency and voltage. It is important to know that in this case the experiments were not performed at the same voltage but at a similar level of brightness that would allow obtaining the spectroscopy correctly. That is why the results must be studied qualitatively. Because, in addition, this spectroscopic analysis, unlike those contained in Fig. 4.25 and 4.26, in which the device was only turned on for that purpose, were instead made throughout the electrical characterization experiment of the device in question, so that the device was on for a considerable time as well as subjected to many charges and discharges in a short time. The voltage data to which each measurement was saved are included in the caption of the figure. Finally, Fig. 4.29, which compares the spectra obtained for different devices (19 and 16) that both use lithium niobate for the barrier. Regarding the influence of material of the barrier, lithium niobate presents greater intensity for same values of voltage than quartz, due to it gets more current for same voltage; and in an analogous way, Fig. 4.29 shows the influence of the electrode configuration regarding said reasoning. 100 Chapter 4. Analysis of the results figure 4.28 Head bands for Nitrogen second positive system states in the plasma’s emission spectrum: Detail in function of frequency at low-resolution. Device 3: quartz, type C. Table 3.1 contains an index of experiments. The voltage data to which each measurement was saved are the following: 6.844 kV pp -0.0305A pp (measured at 10 kHz), 6.91 kVpp-0.02975App (measured at 9 kHz), 8.256 kVpp-0.03372App (measured at 8 kHz) and 8.332 kVpp-0.03186App (measured at 7 kHz).. 4.2 Spectroscopic analysis results 101 figure 4.29 Head bands for Nitrogen second positive system states in the plasma’s emission spectrum: Comparison between different electrode configurations, both in lithium niobate. Device 16 and 19. Table 3.1 contains an index of experiments. It is important to know that in this case the experiments were not performed at the same current conditions. That is why the results must be studied qualitatively. . Vibrational Temperature According to the study found in the reference [ 143 ], under conditions of atmospheric discharge, the temperature of the gas can be estimated through the rotational temperature of a molecule, because rotational energy levels are closely spaced (10 −3eV ) and they allow quick transfers of energy between two energy modes. This article tells “The population distribution inthe rotational energy level of N 2 is also believed to fit Boltzmann distribution due to the dense collisions between molecules. [...] By assuming a rotational temperature and considering the dipole radiation probability and the response function of the monochrometer, one can calculate the profile of a certain emission band. The actual rotational temperature (T r ) can be determined through comparing the experimental measurement and theoretical calculation.[...] Here, in surface discharge, the vibrational excitation of nitrofen seems to be the most important. N 2 (C 3Πu ), which is not a metastable state, is generated from the ground-state electron impact excitation an the cascading effect is not important for the N 2 (C 3Πu ) state population. Therefore, the vibrational temperature can be determined according to the ratio of two lines in the N 2 (C) second positive system. The spectra lines at 371.1nm and 380.5nm are selected to calculate the vibrational temperature (tV), as follows in equation 4.3.´´ For this reason a series of spectroscopy was measured with an increase in resolution, to obtain data contained in Table 4.3 used to build Fig. 4.30. In it, for the case of lithium niobate device 19, it can be verified how, from a certain voltage, the electronic vibrational temperature associated with the second positive state of the nitrogen states 102 Chapter 4. Analysis of the results remains approximately stable with the voltage growth (although with a slight decrease, which means a decrease in the movement –energy– of the electrons), which corroborates the conclusions presented in other studies consulted [0][143]. The consulted bibliography tells us that the rotational temperature is very sensitive to both: applied voltage and frequency. In addition, some bibliography teaches that the electronic vibrational temperature is independent of the material used as a barrier[0]. It is important to bear in mind that the calculated temperatures are an effective temperature measurement, with the aim of qualitative analysis, since the plasma we have is out of equilibrium. 1.1384 ·e−0.4952eV /T0 e=I371.1 I380.5 (4.3) Table 4.3 Vibrational temperature data for air plasma according to [143]. Results extracted from Fig. 4.26, and these data are represented in Fig. 4.30.. Intensity [CPS] Vibrational temperature [eV] Voltage [ kV] I371.1nm I380.5nm A=I371.1 I380.5T=−0.4952/ln(A/1.1384) 2 372 416 0.89423077 2.05123809 2.5 472 1190 0.39663866 0.469671758 3 920 7204 0.12770683 0.226362469 3.5 3974 23902 0.16626224 0.257405507 4 9614 64096 0.14999376 0.244327796 figure 4.30 Vibrational temperature for nitrogen of a air plasma, at 5 kHz and atmospheric pressure. Generated with device 19. Data from Table 4.3, including what is Ratio A referred to. Table 3.1 contains an index of experiments. . 4.2 Spectroscopic analysis results 103 Main spectroscopy results: summary After analyzing the emission bands of the generated surface plasma, one has the main results included in Table 4.4. Table 4.4 Summary of main spectroscopy results.. MAIN RESULTS OF SPECTROSCOPIC ANALYSIS The head bands optical spectrum of the nitrogen molecule are perfectly identified, corresponding to the vibrational transitions associated with the first and second positive states –Fig. 4.26–. Parameter Behavior with intensity Figure Voltage (V) The increase in voltage translates into a significant increase in the intensity of emission. 99 Frequency (f) The increase in frequency translates into a significant increase in the intensity of emission. 100 Behavior of Vibrational temperature with voltage From a certain value of voltage, the electronic vibrational temperature associated with the second positive state of the nitrogen states remains approximately stable with the voltage growth,although with a slight decrease, which means a decrease in energy of the electron. 102 104 Chapter 4. Analysis of the results 4.3 Things to keep in mind about the materials during the experiments During the course of this experimental work have taken place the following facts or appreciations regarding the materials of discharge barrier, such as wear or degradation and even breakage. It is important to know how the materials used as a substrate respond, in order to choose the most appropriate material to develop a device to market, where the operating conditions can be, in addition to prolonged, more extreme than those carried out in this experimental work . The most notable observations are listed below: 1. It is necessary to be careful with the realization of welds on the copper already adhered to the dielectric discs, especially for the case of lithium niobate, avoiding in this case if it is possible to carry them out on it. This has to do with the thermal shock that the material receives, which together with the pressure of the welder, can cause the material to break, as happened on one occasion with one of the devices on lithium niobate while it was being manufactured. In addition, in general, special care must be taken in the handling of lithium niobate discs due to their fragility. 2. The discs do not show a great deal of degradation after the experiments, at least not enough to alter the results of the experiments excessively. Although we must be aware that we must always try to ensure that the measurements are taken correctly the first time (for example, avoiding repeating a measurement or having to discharge the disc to return to a breakdown point that we would not have realized), since the heating of the disc as well as the accumulation of charge in ferroelectrics would introduce new variables that would not correspond with the rest of the measurements taken in the rest of the experiments. 3. In order to preserve the subtracts of the devices in the best possible state, the experiments must be carried out as quickly as possible, although always carrying out the electric charge process in a moderate and controlled manner. It is as much about avoiding the excessive heating of the disc as of the excessive accumulation of charge in the case of ferroelectrics, which leads to an affectation in the results introducing other parameters of influence to the study. Regarding the electric arcs, only two short circuits were experienced during this work: one of them took place at the beginning of experiment 2 and it was considered that the reasons had nothing to do with the device itself, and during the development of the study of the influence of the voltage in the spectroscopy performed on device 19, which broke during the application of a peak-to-peak voltage of 4.5 kV at 5 kHz of frequency (twice its breakdown voltage approximately at this frequency), because it had been on for a long time and probably would have a high value of accumulated charge. After the mentioned breaks of the lithium niobate discs, they were unusable. Regarding its break during the spectroscopy experiment, the reasons can be based on the fact that when applying such a high voltage and due to its crystalline structure, a piezoelectric effect appears. As the frequency increases, this effect becomes 4.3 Problems with the materials used 105 increasingly greater until the crystals do not support the stresses that are produced and therefore break; said ruptures may be due to defects of the crystalline network produced during the manufacture of the material. It was not the object of this study to bring the materials to a limit state. List of Tables 2.1 Dielectric constants. 26 2.2 Characteristics and typical parameters of the SDBD plasma controller 32 3.1 List of experiments and its corresponding devices. 56 3.2 Dimensions of devices of Phase 1 and 2. 57 4.1 Operational parameters analyzed with the electric analysis. 80 4.2 Electrical analysis summary 95 4.3 Vibrational temperature data for air plasma. 102 4.4 Spectroscopic analysis summary 103 113 A. In relation to Paschen curves Related to the theory exposed in Subsection “Breakdown voltage and Paschen curves” in Page 22 a similar representation has been made with data extracted from experiments of Phase 2 on lithium niobate in function of frequency. The representations, Voltage vs. pg, are included in Fig. 1 and 2. (a) Phase 2-Lithium niobate for different gaps and frequencies, at atmospheric pressure. Type D configuration, Experiments 10, 13 and 16.. figure 1 Representations related to Paschen curves I. Examples with lithium niobate. Table 3.1 contains an index of experiments.. 115 116 A. In relation to Paschen curves (a) Figure a) compare to quartz (Exp. 8, 11 and 14) and alumina (Exp. 9, 12 and 15) for 5 kHz.. (b) Comparison between device 19 in function of pressure and frequency (Exp.19 and 21).. figure 2 Representations related to Paschen curves II. Examples with lithium niobate. Table 3.1 contains an index of experiments.. B. Other thesis about DBD actuators Respect to conclusions extracted from works about SDBD actuators (not necessary referred to breakdown values), is interesting to know that it has been verified the following: •The power consumption increases with the increasing applied voltage [124] –see Fig. 1–. (a) Q-VLissajous Figure.. (b) Power.. figure 1 Power consumption of SDBD actuator with different applied voltage. Source of the image: [124].. •The surface temperature of dielectric increases with the increasing applied voltage [124] –see Fig. 2–. • The induced flow velocity increases and then decreases with the widening gap between two electrodes. With the increase in voltage, the induced airflow velocity has also an increasing tendency [ 124 ] –see Fig. 3 and 4–. •The SDBD plasma actuator can effectively suppres the flow separation[124] [0] –see Fig. 5 and 6–. 117 118 B. Other thesis about DBD actuators figure 2 Temperature distribution of dielectric surface with different applied voltages. Source of the image: [124].. • With the enlargement of the electrode gap, the discharge voltage increases slightly as a result of the increasing actuator impedance; while the discharge current increases first and then decreases by the combined effects of applied voltage and electrode gap [125] –see Fig. 7 and 8–. •Variation of electric power consumption with the electrode gap shows a similar trend with that of discharge current. As for a fixed applied voltage, the critical electrode gap corresponding to the maximum power is slightly larger than that of the maximum current [125] –see Fig. 9–. • Variations of the velocity of induced flow are in accordance with that of the discharge power. Optimal electrode gaps corresponding to the maximum induced flow velocity are obtained, related with the different applied voltage [125]. B. Other thesis about DBD actuators 119 (a) Induced flow velocity with the widening gap between two electrodes.. (b) Induced flow velocity with different applied voltages.. figure 3 Induced flow velocity. Source of the images: [124].. 120 B. Other thesis about DBD actuators figure 4 Smoke visualization of induced flow with different applied voltages. Source of the image: [124].. B. Other thesis about DBD actuators 121 (a) Schematic of SDBD flow control.. (b) Train head.. (c) Train rear.. figure 5 Smoke visualization of gas flow around the train head/rear (with the SDBD installed) with different applied voltages. Source of the images: [124].. 128 Bibliography [12] Kizilyalli, M.; Corish, J.; Metselaar, R. Definitions of terms for diffusion in the solid state (IUPAC Recommendations 1999). Pure Appl. Chem. 71 (7):1307-1325. doi:10.1351/pac199971071307. 1999. [13] Langmuir, Irving. Positive Ion Currents from the Positive Column of Mercury Arcs. Volume 58. Issue 1502. pp. 290-291. Science. 1923. [14] Albert W. Hull and Irving Langmuir. Control of an Arc Discharge by Means of a Grid. Proc Natl Acad Sci USA. 15(3):218-225. 1929. [15] J. R. Hollahan. Techniques and Applications of Plasma Chemistry. John Wiley &Sons Inc. 1974. [16] Nuclear Engineering Course Notes. Massachusetts Institute of Technology. [17] Temerin, M.; Mozer, F. S. Double Layers Above the Aurora. NASA Conference Publication, # 2469. 1987. [18] Block, L. P. A double layer review. Astrophysics and Space Science, vol. 55. no.1. pp. 59-83. 1978. [19] Bulgakova, Nadezhda M. et al. Double layer effects in laser-ablation plasma plumes. Physical Review E (Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics), Volume 62. Issue4. pp. 5624-5635. 2000. [20] Joos, G. Theoretical Physics. London &Glasgow: Blackie &Son Ltd. p. 271. 1951 [22] Peter Andrew Sturrock. Plasma Physics: An Introduction to the Theory of Astrophysical, Geophysical and Laboratory Plasmas. Stanford University, California. 1994. [23] Chen, Francis.F. Introduction to Plasma Physics and Controlled Fusion. Springer, New York. ISBN 0306413329- 0CLC 9852700. 2016. p. 2. doi:10.1007/978-3-319-22309-4. ISBN 978-3-319-22309-4. [24] M. Rodríguez Vidal y J. A. Vallés Abarca. Curso Electromagnetismo, Unidad Didáctica 3, Ondas electromagnéticas libres, Temas X y XI. UNED. 1977. [25] P. C. Clemmow y J. P. Dougherty. Electrodynamics of Particles and Plasmas. Addison-Wesley. 1969. [27] Quasi-neutrality - The Plasma Universe theory. www.plasma-universe.com. [28] Ajay Peter Manuel. Numerical Investigations of the Plasma Actuator Flow Problem. [29] Merlino, Robert L. Plasma Oscillations— An application of electrostatics and classical mechanics. University of Iowa. omepage.physics.uiowa.edu. 2012. [30] J. R. Reitz, F. J. Milford y R. W. Christy. Fundamentos de la teoría electromagnética, 4 º ed, Cap. 14. Addison-Wesley Iberoamericana. 1996. [31] Nicholson, Dwight R. Introduction to Plasma Theory. John Wiley & Sons. ISBN 978-0-471-09045-8. 1983. [32] J. P. Boeuf, Y. Lagmich, T. Challegari, and L. Pitchford. Electrohydrodynamic force and acceleration in surfaces discharges. AIAA 2006-3574; 37th AIAA Plasmadynamics and Lasers Conference, San Francisco, California, USA. 2006. Bibliography 129 [33] Y. P. Raizer. Gas Discharge Physics. Springer. 1991. [34] E. E. Kunhardt. Generation of large-volume, atmospheric-pressure, non-equilibrium plasmas. IEEE Transactions on Plasma Science. 28:189200. 2000. [35] P. C. Clemmow y J. P. Dougherty. Electrodynamics of Particles and Plasmas. Addison-Wesley. 1969. [36] N. S. J. Braithwaite. Plasma Sources Science and Technology. Introduction to gas discharges. 9(4):517527, 2000. [37] J. Jolibois. Etude et développement dún actionneur plasma à décharge à barrière diélectrique application au contrôle d’écoulement sur prol dáile. PhD thesis. Université de Poitiers. 2008. [38] B. M. Smirnov. Physics of Ionized Gases. Wiley Interscience. 2001. [39] Alexander A. Fridman. Plasma Chemistry. Cambridge, UK: Cambridge University Press. p. 94. ISBN 0-521-84735-4. 2008. [41] F. Delobeau. The environment of the earth. Reidel. 1972. [42] Chapman, Brian. Chapter 3: Plasmas. Glow Discharge Processes: Sputtering and Plasma Etching. New York: John Wiley &Sons. p. 49. ISBN 978-0471078289. 1980. [43] J. P. Boeuf. Applications and modeling of non-equilibrium plasmas. In Mathematical. [44] Introduction to Plasma Physics and controlled fusion. Chen, Francis F. Springer International Publishing. pp. 2–3. ISBN 9781475755954. 1984. [45] Freidberg, Jeffrey P. Plasma Physics and Fusion Energy. Cambridge University Press. p. 121. ISBN 9781139462150. 2008. [46] Robinson M. Movement of air in the electric wind of the corona discharge. AIEE Trans. 1961. [47] Slater, J. C.Atomic Shielding Constant. Phys. Rev. 36 (1): 57-64. doi:10.1103/PhysRev.36.57. 1930. [48] PC Clemmow & JP Dougherty. Electrodynamics of particles and plasmas. Redwood City CA: Addison- Wesley. pp. § 7.6.7, p. 236 ff. ISBN 0-201-47986-9. 1969. [49] Alexander A. Fridman. Plasma Chemistry. Cambridge, UK: Cambridge University Press. p. 94. ISBN 0-521-84735-4. 2008. [50] L. G. Christophorou and J. K. Oltho. Fundamental Electron Interactions with Plasma Processing Gases. Physics of Atoms and Molecules. Kluwer Academic/Plenum Publishers. 2004. [51] B. Smirnov. Physics of weakly ionized gases. Mir Publishers Moscow. 1981. [52] Hippler, R.; Kersten, H.; Schmidt, M.; Schoenbach, K.M. Plasma Sources. Low Temperature Plasmas: Fundamentals, Technologies, and Techniques (2nd ed.). Wiley-VCH. Wiley-VCH. ISBN 978-3-527- 40673-9. 2008. 130 Bibliography [53] S. FlüggeThe glow discharge at low pressure. G. Francis. Encyclopedia of Physics, volume Volume 22: Gas Discharges 2, pages 53-61. Editor: Springer. 1956. [54] R. Gadri and J. Roth. Glow discharge-like characteristics of a oaugdp revealed by computer modeling. 25th Anniversary. IEEE Conference Record Abstracts. 1998 IEEE International on Plasma Science, page 288, 1998. [55] Thornhill, W. The Z-Pinch Morphology of Supernova 1987A and Electric Stars. IEEE Transactions on Plasma Science, vol. 35, issue 4, pp. 832-844. 2007. [56] Structure of a Glow Discharge. Princeton Plasma Physics Laboratory. [57] L. B. Loeb. Electrical breakdown of gases with steady or direct current impulse potentials. [58] C. Stees. Erste Ansätze einer neuen Strategie zur ezienten Modellierung eines Plasma Aktuators auf Basis des Verständnisses bestehender Modelle. Master’s thesis. TU Darmstadt. 2008. 2008. [59] F. Paschen. Ueber die zum funkenübergang in luft, wassersto und kohlensäure bei verschiedenen drucken erforderliche potentialdierenz. Annalen der Physik. 273(5):6996. 1889. [60] K. Vollrath and G. Thomer. Funkenlichtquellen und Hochfrequenz- Funkenkinematographie. K. Vollrath. Editors: Kurzzeitphysik. Springer. pages 76-165. 1967. [61] Friedrich Paschen. Ueber die zum Funkenübergang in Luft, Wassersto und Kohlensäure bei verschiedenen Drucken erforderliche Potentialdierenz. Annalen der Physik. 273 (5): 69-75. 1889. [62] F. Paschen. Ueber die zum funkenübergang in luft, wassersto und kohlensäure bei verschiedenen drucken erforderliche potentialdierenz. Annalen der Physik. 273(5):69-96, 1889. 1889. [63] Y. P. Raizer, M. N. Shneider, and N. A. Yatsenko. Radio-Frequency Capacitive Discharges. CRC Press. 1995. [64] W. Siemens. Ueber die elektrostatische Induction und die Verzögerung des Stroms in Flaschendrähten. Annalen der Physik. 178(9):66-122, 1857. [65] Matsuno, Hiromitsu, Nobuyuki Hishinuma, Kenichi Hirose, Kunio Kasagi, Fumitoshi Takemoto, Yoshinori Aiura, and TatsushiIgarashi. Dielectric Barrier Discharge lamp. United States Patent 5757132. [66] Dhali, S.K. and I. Sardja. Dielectric-barrier discharge for the removal of SO2 fromflue gas. IEEE International Conference on Plasma Science. IEEE Conference Record - Abstracts, 1989. [67] Kogelschatz, Ulrich, Baldur Eliasson, and Walter Egli. From ozone generators to flat television screens: history and future potential of dielectric-barrier discharges. Pure Applied Chemistry, Vol. 71. No. 10, pp. 1819-1828, 1999. [68] U. Kogelschatz. Dielectric-Barrier Discharges: Their History, Discharge Physics, and Industrial Applications. Plasma Chemistry and Plasma Processing. 23:146, 2003. Bibliography 131 [69] Leroux, F.; Perwuelz, A.; Campagne, C.; Behary, N. Atmospheric air-plasma treatments of polyester textile structures. Journal of Adhesion Science and Technology. 20 (9): 939–957. 2006. [70] Leroux, F. D. R.; Campagne, C.; Perwuelz, A.; Gengembre, L. O. Polypropylene film chemical and physical modifications by Dielectric Barrier Discharge plasma treatment at atmospheric pressure. Journal of Colloid and Interface Science. 328 (2): 412–420. 2008. [71] Leroux, F. D. R.; Campagne, C.; Perwuelz, A.; Gengembre, L. O. Polypropylene film chemical and physical modifications by Dielectric Barrier Discharge plasma treatment at atmospheric pressure. Journal of Colloid and Interface Science. 328 (2): 412–420. 2008. [72] Laroussi, M. Sterilization of contaminated matter with an atmospheric pressure plasma. IEEE Transactions on Plasma Science. 24 (3): 1188. 1996. [73] Lu, X.; Naidis, G.V.; Laroussi, M.; Ostrikov, K. Guided ionization waves: Theory and experiments. Physics Reports. 540 (3): 123. 2014. [74] V. I. Gibalov and G. J. Pietsch. The Development of Dielectric Barrier Discharges in Gas Gaps and on Surfaces. Journal of Physics D: Applied Physics. 33(20):2618, 2000. 2000. [75] Aerosol charge distributions in Dielectric Barrier Discharges. European Aerosol Conference 2009 Karlsruhe. 2009. [76] M. Laroussi, I. Alexeff, J. P. Richardson, and F. F. Dyer. The Resistive Barrier Discharge. IEEE Trans. Plasma Sci. 30, 158. 2002. [77] Roth, J. Reece. Chapter 15.3 Atmospheric Dielectric Barrier Discharges (DBDs). Industrial Plasma Engineering: Volume 2: Applications to Nonthermal Plasma Processing (1st ed.). CRC Press. ISBN 978- 0750305440. 2001. [78] Source of the image: en.wikipedia.org/wiki/File:Filamentous_Dielectric_Barrier_Discharge.JPG [79] Kao, Kwan Chi. Dielectric Phenomena in Solids. London: Elsevier Academic Press. 2004. [80] Source of the image: https://en.wikipedia.org/wiki/File:Capacitor_schematic_with_dielectric.svg [81] Daintith, J. Biographical Encyclopedia of Scientists. CRC Press. p. 943. ISBN 0-7503-0287-9. 1994. [82] William Whewell. The Correspondence of Michael Faraday, Volume 3. Letter. William Whewell to Faraday. 1798. James, Frank A.J.L. The Institution of Electrical Engineers. London, United Kingdom. 996. ISBN 0-86341-250-5. 1996. [83] Frederick Seitz; T. P. Das; David Turnbull; E. L. Hahn.Ferroelectrics and Antiferroelectrics. Werner Känzig. Solid State Physics. 4. Academic Press. p. 5. ISBN 0-12-607704-5. 1957. [84] M. Lines; A. Glass. Principles and applications of ferroelectrics and related materials. Clarendon Press, Oxford. ISBN 0-19-851286-4. 1979. 132 Bibliography [85] Enloe, C., McLaughlin, T., VanDyken, R., and Fuscher, J. Plasma structure in the aerodynamic plasma actuator. AIAA Paper 2004-0844, 2004. [86] http://www.lambdaphoto.co.uk/pdfs/Inrad_datasheet_LNB.pdf [88] Dmitriy M. Orlov. Modelling and Simulation of Single Dielectric Barrier Discharge Plasma Actuators. PhD thesis. University of Notre Dame, Graduate School of Aerospace, and Mechanical Engineering. 2006. [89] Gad el Hak. Flow Control: Passive, Active, and Reactive Flow Management. Cambridge University Press, New York. 2000. [90] Stephen P. Wilkison Thomas C. Corke, C. Lon Enloe. Dielectric barrier discharge plasma actuators for flow control. Annu. Rev. Fluid Mech. 42:505– 529, 2010. [91] Adil Bouchmal. Modeling of Dielectric Barrier Discharge actuator - implementation, validation, and generalization of an electrostatic model. Master’s thesis. Delft University of Technology. 2011. [92] http://www.moreelectricaircraft.com/ [93] G. Touchard. Plasma actuators for aeronautics applications - state of art review. I.J. PEST, 2:1–25, 2008. [94] Eric Moreau. Airflow control by non-thermal plasma actuators. J. Phys. D: Appl. Phys, 40:605–636, 2007. [95] Dmitriy M. Orlov. Thomas C. Corke, Martiqua L. Single dielectric barrier discharge plasma enhanced aerodynamics: physics, modeling, and applications. Post. Exp Fluids, 46:1–26, 2009. 2009. [96] Wei Shyy Balaji Jayaraman. Modeling of Dielectric Barrier Discharge-induced fluid dynamics and heat transfer. Progress in Aerospace Sciences, 44:139– 191, Progress in Aerospace Sciences. 2008. [97] Zyga, Lisa. What causes ionic wind? Phys.org. 2018. [98] N. Bénard, N. Balcon, and E. Moreau. Electric wind produced by a surface dielectric barrier discharge operating in air at dierent pressures: aeronautical control insights. Journal of Physics D: Applied Physics. 41(4):042002 (5pp). 2008. [99] Enloe, C., McLaughlin, T., VanDyken, R., and Fuscher, J. Plasma structure in the aerodynamic plasma actuator. AIAA Paper 2004-0844, 2004. [100] P. Versailles, V. Gingras-Gosselin, and H. Vo. Impact of Pressure and Temperature on the Performance of Plasma Actuators. AIAA Journal. 48:859863. 2010. [101] Al-Khazali, Hisham A. H.; Askari, Mohamad R. Geometrical and Graphical Representations Analysis of Lissajous Figures in Rotor Dynamic System. IOSR Journal of Engineering.2 (5): 971–978. 2012. [102] Sozer, E.B. Gaseous Discharges and Their Applications as High Power Plasma Switches. 2008. Bibliography 133 [103] Ge Lei and Zhang Yuantao. A Simple Model for the Calculation of Plasma Impedance in Atmospheric Radio Frequency Discharges. Plasma Science and Technology, Volume 16, Number 10. [104] Demetrius T. Paris and F. Kenneth Hurd. Basic Electromagnetic Theory. McGraw-Hill, New York. ISBN 0-07-048470-8, pp. 512, 546. 1969. [105] Electromagnetism for Engineers. P. Hammond. Pergamon Press. OCLC 854336.1969. [106] Bagotskii, Vladimir Sergeevich. Fundamentals of electrochemistry.p. 22. ISBN 978-0-471-70058-6. 2006. [107] Horowitz, Paul; Hill, Winfield.The art of electronics (3rd ed.). Cambridge University Press. ISBN 978-0-521-80926-9. 2015. [108] www.electronics-tutorials.ws/accircuits/phase-difference.html [109] Croft, Terrell; Summers, Wilford I. Reports on August 2003 Blackout, North American Electric Reliability Council website. [110] American Electricians’Handbook (Eleventh ed.). New York: McGraw Hill. SBN 0-07-013932-6. 1987. [111] Fink, Donald G.; Beaty, H. Wayne. Standard Handbook for Electrical Engineers (Eleventh ed.). New York: McGraw Hill. ISBN 0-07-020974-X. 1978. [112] www.electronics-tutorials.ws [113] www.uco.es/organiza/departamentos/quimica-fisica/quimica-fisica/QuiFis/L14_QF_10_11.pdf [114] https://www.lenntech.es/periodica/elementos/n.htm [115] What are Homonuclear Molecules and Hetero-nuclear Molecules? TheBigger.com [116] A. Lofthus and P. H. Krupenie. J. Phys. Chem. The spectrum of molecular nitrogen. Data 6, 113307. 1977. [117] G. Herzberg. Spectra of Diatomic Molecules. Molecular Spectra and Molecular Structure. Van Nostrand, Princeton. 1950. [118] Kennelly, Arthur. Impedance. AIEE, 1893. [119] Ortega, Manuel R. Lecciones de Física (4 volúmenes). Monytex. ISBN 84-404-4290-4, ISBN 84-398- 9218-7, ISBN 84-398-9219-5, ISBN 84-604-4445-7. 1989-2006. [120] Resnick,Robert & Krane, Kenneth S. Physics. Nueva York: John Wiley & Sons. ISBN 0-471-32057-9. 2001. [121] Guoqiang Gao, Lei Dong, Kaisheng Peng, Wenfu Wei, Chunmao Li, and Guangning Wu. Comparison of the surface Dielectric Barrier Discharge characteristics under different cond. 134 Bibliography [122] J. P. Richardson M. Laroussi, I. Alexeff and F. F. Dyer. The resistive barrier discharge. IEEE Trans. Plasma Sci, 30, 2002. [123] D.D. Sijacic U. Ebert, I.R. Rafatov. Structure formation in a dc-driven ”barrier” discharge stability analysis and numerical solutions. 28th ICPIG. 2007. [124] Lei Dong, Guoqiang Gao, Kaisheng Peng, Wenfu Wei, Chunmao Li, and Guangning Wu. Effects of surface Dielectric Barrier Discharge on aerodynamic characteristic of train. AIP Advances 7, 075112 (2017); doi: 10.1063/1.4995985. 2017. [125] Guoqiang Gao, Lei Dong, Kaisheng Peng, Wenfu Wei, Chunmao Li, and Guangning Wu. Comparison of the surface Dielectric Barrier Discharge characteristics under different electrode gaps. Physics of Plasmas 24, 013510; doi: 10.1063/1.4974037. 2017 [126] P. Hammond. Electromagnetism for Engineers. Pergamon Press 1969 OCLC 854336. [127] Purcell, E. M. Electricidad y Magnetismo. Reverté. p. 141. ISBN 9788429143195. 1988. [128] Física general. Burbano de Ercilla, Santiago y Carlos Gracía Muñoz [129] Regents of the University of California. Universe of Light: What is the Amplitude of a Wave? 1996. [130] The development of the concept of electric charge: Electricity from the Greeks to Coulomb. Roller, Duane; Roller, D.H.D. Cambridge, MA: Harvard University Press. 1954. [131] Feynman, Richard; Leighton, Robert; Sands, Matthew. Section 2-5: energy levels, 19: hydrogen atom. The Feynman Lectures on Physics 3. 1965. [132] Fundamentos de Espectroscopía Raman. Capítulo 1. UPC Universitat Politècnica de Catalunya. [133] S. B. Bayrama and M. V. Freamatb. Vibrational spectra of N 2 : An advanced undergraduate laboratory in atomic and molecular spectroscopy. Physics Department, Miami University, Oxford, Ohio 45056. 2012. [134] G. Herzberg. Molecular Spectra and Molecular Structure, I. Spectra of Diatomic Molecules. Van Nostrand, Princeton. 1950. [135] Friedrich Paschen. Ueber die zum Funkenubergang in Luft, Wassersto und Kohlensaure bei verschiedenen Drucken erforderliche Potentialdifferenz. Annalen der Physik 273 (5): 69-75.. 1889. [136] H. R. Velkoff and J. Ketcham. Effect of an Electrostatic Field on Boundary-Layer Transition. AIAA. 6:1381-1383, 1968. [137] J. R. Roth, D. Sherman, and S. P. Wilkinson. Boundary Layer Flow Control with a One Atmosphere Uniform Glow Discharge Surface Plasma. 1998-0328; 36th AIAA Aerospace Science Meeting and Exhibit; Reno, Nevada, USA. AIAA. 1998. Bibliography 135 [138] Huang, J., Corke, T., and Thomas, F. Separation control over low pressure turbine blades. American Physical Society Division of Fluid Dynamics. vol. 47. 2002. [139] Morris, S. C., Corke, T. C., VanNess, D., Stephens, J., and Douville, T. Tip Clearance Control Using Plasma Actuators. AIAA Paper 2005-0782, 2005. [140] Visbal, M. R., and Gaitonde, D. V. Control of Vortical Flows Using Simulated Plasma Actuators. AIAA Paper 2006-505, 2006. [141] Corke, T. C., He, C., and Patel, M. Plasma aps and slats: an application of weaklyionized plasma actuators. AIAA Paper 2004-2127, 2004. [142] The Identification of Molecular Spectra. R.W.B. Pearse and A.G. Gaydon. Third Edition. Chapman & Hall LTD. 1965. [143] Max Mulder. Ying-hong Li, Yun Wu, Hui-min Song, Hua Liang and Min Jia. Plasma Flow Control, Aeronautics and Astronautics, Prof. 473-3, InTech.2011. [144] A Short History of Ferroelectricity. Talari. 2009. [145] Ferroelectridad y piezoelectricidad. Universidad de Navarra. Lesson 7. www4.tecnun.es/asignaturas/PFM_Mat/Prog/Ferroelecv2.pdf [146] Source: http://iopscience.iop.org/article/10.1088/1009-0630/16/10/05/pdf [147] Roth J R. Electrohydrodynamically induced airflow in a one atmosphere uniform glow discharge surface plasma. 25th IEEE Int. Conf. Plasma Science (Raleigh, USA). 1998. [148] Roth J R, Sherman D M and Wilkinson S. P. Electrohydrodynamic flow control with a glow discharge surface plasma. AIAA. 2000. [149] Roth J R, Sherman D M and Wilkinson S. P. Boundary layer flow control with a one atmosphere uniform glow discharge surface plasma. AIAA Meeting (Reno, USA) 1998. [150] Chuan He, Thomas C. Corke, and Mehul P. Patel. Plasma Flaps and Slats: An Application of Weakly Ionized Plasma Actuators. Plas, Journal of Aircraft, Vol. 46. 2009. [151] Sebastian Bürkle. Environmental Impacts on Dielectric Barrier Discharge Plasma Actuators. Master- Thesis. 2013. [152] Yongqiang TIAN, Zhengke ZHANG, Jinsheng CAI, Leilei YANG Lei KANG. Experimental study of an anti-icing method over an airfoil based on pulsed dielectric barrier discharge plasma. Chinese Journal of Aeronautics Volume 31. 2018. [153] Jinsheng Cai, Xuanshi Meng, Yongqiang Tian, Xuzhao Han. An experimental study of icing control using DBD plasma actuator. Article in Experiments in Fluids 58(8). August 2017. 136 Bibliography [154] Nicole M. Houser. Manufacturing of DBD plasma actuators for degradation resistance. Dissertation. University of Toronto. 2013. [155] Martina L. Post. Plasma actuators for separation control in stationary and oscillating airfoils. Dissertation. M.S.M.E. Notre Dame, Indiana. 2004.