Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Atomic theory wikipedia , lookup
Electrical resistivity and conductivity wikipedia , lookup
Photoredox catalysis wikipedia , lookup
Electrolysis of water wikipedia , lookup
Synthesis of carbon nanotubes wikipedia , lookup
Electrochemistry wikipedia , lookup
Toxic hotspot wikipedia , lookup
Gaseous detection device wikipedia , lookup
Inductively coupled plasma mass spectrometry wikipedia , lookup
Plasma polymerization wikipedia , lookup
Plasma Treatment for Environment Protection Project PlasTEP – Dissemination and fostering of plasma-based technological innovations for environment protection in the Baltic Sea region 1 Plasma Treatment for Environment Protection The pollution control is a transnational request of all countries and a strategic aim of the European Union. This is also reflected in the increasing tightening of the exhaust emission standards particularly in the Baltic Sea region (BSR) countries. The objective of the project PlasTEP is to push plasma-based cleaning technologies of atmospheric air and water treatment to a visible practical application. The project would like to raise wide awareness about the practical applications of plasma technology for environment protection. PlasTEP contributes to a better future by providing solutions for cleaning of exhaust gases and wastewater. The project disseminates and fosters plasma-based technological innovations for the environment protection in the BSR. PlasTEP also builds up a network to combine the existing knowledge about plasma technologies with partners from industry, science and policy. PlasTEP is contributing to the EU Strategy for the BSR and it is partfinanced by the European Union (European Regional Development Fund). The project aims to bring the idea of investing in plasma technology and therewith in future research into the minds of decision makers and to show them: Plasma opens new ways for environment protection! Edited by: Indrek Jõgi, Ronny Brandenburg, Alexander Schwock, David Cameron, Justyna Jaskowiak and Katherina Ulrich. All rights reserved. Printed in Estonia Tartu, 2012 2 Contents Foreword…………………………………………………………………………………………………………… 5 1. Pollution and methods for pollution control.........….………………………………………. 7 1.1 Pollutants (NOx/SOx, VOCs). Sources and harmful effects, present and future regulations 7 1.2 Methods for air and exhaust gas depollution and role of plasma technologies………………… 14 2. Plasma-based technologies ……………………………………………………………………………. 21 2.1 Introduction to plasma science……………………………………………………………………………………………. 21 2.2 Microwave driven plasmas for plasma depollution……………………………………………..………………. 27 2.3 Corona and Barrier Discharges……………………………………………………………………….……………………. 33 2.4 Hollow cathode discharges………………………………………………………………………………………………….. 39 2.5 Electron beam generated plasmas for gas depollution………………………………………………………… 42 2.6 Power supplies and power determination on electrical discharges……………………………………… 47 2.7 Hybrid methods for pollution removal……………………………………………………………………………..…. 56 2.8 Eco-efficiency and cost-benefit analysis of plasma technologies………………………………………….. 64 3. Practical examples of plasma technology………………………………………………………. 71 3.1 PlasmaAir AG……………………………………………………………………………………………………………………….. 72 3.2 Bioclimatic GmbH……………….……………………………………………..………………………………………………… 77 3.3 MCT GmbH…………..………………………………………………………………………………………………………..……. 83 3.4 BÄRO GmbH & Co. KG………………………………………………………………………………………………………….. 88 3.5 Rafflenbeul Anlagenbau GmbH……………………………………………………………………………………………. 92 3.6 Applied Plasma Physics AS (APP)…………………………………………………………………………………………. 94 3 Appendices………………………………………………………………………………………………………… 98 Appendix A. Plasma chemical reactions.......................................................................................... 98 Appendix B. Different types of discharges...................................................................................... 101 4 Foreword For more than a century, industrialization has been the driving force for increased living standards in Europe and consequently also in the Baltic Sea region. At the beginning of this process, the smoking stacks of factories were the sign of the progress but the improvements in living standards due to industrialization have not come without cost. Industry has a strong environmental impact with effects on our health and on the surrounding nature. As a result the negative aspects of industrialization have started to counteract the positive aspects of our living standards. The negative environmental effects of industrialization are mostly due to the pollution which can come from very different sources. Large power plants are needed to fulfill our energy demand but they also emit large amounts of gaseous pollutants. The transportation system, which is necessary for keeping our economy running, also has a significant share in both to the air and water pollution. The production of various commodities, as well as livestock and other agricultural activities is an additional source of pollution with a very large variety of pollutants. The pollution from these sources can lead to various health problems such as cancer, allergies etc. which directly affect our lives. There are also many indirect ways in which the pollution affects us. Some pollutants are the precursors of acid rain and smog which can damage both the buildings and plants. The pollution can also impact the fertility of animals and put them in danger of extinction. Moreover, the damage to nature has also economic impact leading to lost opportunities in tourism because, up to now, pristine nature has been one of the important attractions in the Baltic Sea region. Since the development of industrialization cannot be stopped thus we need to look for possible ways to keep under control the pollution originating from the industrialization. There are several general routes which can be used to limit the pollution from various sources. One route is to make the technology of energy production, transportation and manufacturing more efficient. In the long term future, the technologies which allow the efficient harvesting of wind and solar energy together with energy storage technologies will certainly lead to zero-emission societies with smaller environmental impact. However, these technologies are not yet available in the large scale and the transition to new 5 technologies will demand large financial investments and time. Thus, we need to find efficient midterm solutions to prevent serious damage of our environment. The elimination of the pollutants before emission to the atmosphere or water can be considered as one such solution. There are various pollution removal technologies available (e.g. thermal incinerators or wet scrubbers) but they are often not sufficiently efficient for a significant decrease of the environmental impact of industrial wastes in near future. Thus we need new more efficient cleaning technologies. Application of various plasma technologies is one way to obtain higher efficiencies of pollution removal. A universal plasma source, suitable for application to of all types of pollutant sources does not exist. There are a number of different plasma sources which can be used for pollution control depending on the nature of the pollutant and pollution source. Significant progress has been achieved in the scientific understanding of the Fused Hollow Cathode strengths and weaknesses of various plasma technologies Plasma Group at the Ångström Laboratory, Uppsala University applicable for pollution control. Unfortunately, up to now this knowledge has been available only to a narrow group of specialists in the field of plasma science while public awareness is still at a very low level. However, an increased level of awareness is important because it allows both the overcoming of possible fears connected with the application of plasma technologies and the unlocking of investments. The present book is intended to improve the public understanding of plasma and its applicability in pollution control. The expected audience of the book is those who do not have a solid background in physics but due to their professional duties want to know the basic concepts of plasma physics and get a general understanding of the working principles of plasma devices and the impact on the environment. Thus, the handbook should be useful for those who are dealing with problems of air and water cleaning and environmental protection (engineers, salesmen, teachers, students, civil servants and journalists). The handbook will start with an overview of the main pollution sources, their impact on our health and the present situation in the pollution control. A short introduction to the principles behind the production of plasmas is given before introducing various plasma types. This introductory part could be skipped by those who are already familiar with the topic and would like to learn about the applicability of these plasma sources for the removal of different classes of pollutants, which is the main emphasis of the handbook. The last part of the book shows existing state-of-the-art technologies and the companies supplying them. 6 1. Pollution and methods for pollution control 1.1 Pollutants (NOx/SOx, VOCs). Sources and harmful effects, present and future regulations Saulius Vasarevičus1, Andra Blumberga2, Dagnija Blumberga2 1 Vilnius Gediminas Technical University, Vilnius, Lithuania Riga Technical University, Riga, Latvia 2 Introduction Air pollution is the contamination of atmosphere by chemicals, particulate matter, or biological materials which causes harm or discomfort to humans and to the environment. The anthropogenic sources are regarded as the main reason for air pollution but some of these pollutants have also natural origin. Measurable effects on humans and environment are: 1) indoor air pollution 2) radioactivity 3) urban photochemical smog 4) acid rain 5) visibility reduction 6) greenhouse warming 7) depletion of the ozone layer 8) climate change due to anthropogenic aerosols 7 Air pollution problems can be schematically presented as a system consisting of three basic components: emission sources, atmosphere and receptors. Various types of emission sources produce different pollutants which are emitted into the atmosphere. Major sources are transportation, industrial, domestic fuel burning and industrial processes. The atmosphere is the medium which transports the pollutants to the receptors. The atmosphere additionally acts as a medium for the transformation of emitted pollutants to new types will take place through various complex mechanisms. The receptors are the targets of the pollution and they can be humans, animals, plants, soil, water or materials. Classification of pollutants and their sources There are several possible ways to classify air pollutants. Some types of classifications are shown in the table 1.1.1. In the following chapters, the main focus will be on gaseous pollutants with several different chemical compositions. The most important sources are shown for these pollutants. The collected data (2011) for main pollution sources in the Baltic Sea region (BSR) can be found in the report “Analysis of main pollution source of NOx, SOx, VOC/odour and waste water in the BSR” which is available on PlasTEP’s website: http://www.plastep.eu/fileadmin/dateien/Outputs/Emission_sources.pdf. Table 1.1.1. Classification types of pollution. Chemical composition Physical state Way of reaching to atmosphere Space scales Sources Sulphur-containing Gaseous Primary Indoor Power generation Nitrogen-containing Liquid Secondary Regional Transportation Carbon-containing Solid Global Consumer products Halogen-containing Commercial products Toxic metals Residential heating Radioactive Agriculture The standards and regulations of air pollutants In most countries worldwide the air pollution is regulated and monitored by specific bodies, agencies and organisations. In European countries including the Baltic Sea region, the European Commission directives are used as the means to regulate air pollution. The directive 2008/50/EC has merged most of the existing legislations into one directive but there are additional directives concerning national 8 ceilings (2001/81/EC), large power plants (2001/80/EC), waste incineration plants (2000/76/EC) and pollutant emissions from ships (MARPOL 73/78). Table 1.1.2 presents some limit values for most important pollutants according to directive 2008/50/EC. The European Environment Agency (EEA) provides independent information on the environment and collects data concerning air pollution. The European environment information (EIONET) is a supporting network for EEA comprising from EEA, six European Topic Centres and a network of around 1.000 experts from 39 countries. The European Pollutant Release and Transfer Register (E-PRTR) is a new Europe-wide register that provides key environmental data from industrial facilities in European Union member states and some other European countries. There are other organisations influencing the regulations and data collection e.g. the European Integrated Pollution Prevention and Control (IPPC) Bureau which produces reference documents on best available techniques and Convention on Longrange Transboundary Air Pollution (CLRTAP) which comprises eight protocols on the reduction of specific air pollutants. Table 1.1.2. Limit values of SO2, NOx, benzene, CO and PM10 for the protection of human health according to directive 2008/50/EC, Annex XI. Chemical Time span Limit values and times exceeded in calendar year hourly 350 µg/m3, not more than 24 times daily 125 µg/m3, not more than 3 times hourly 200 µg/m3, not more than 18 times yearly 40 µg/m3 Benzene yearly 5 µg/m3 CO 8 h mean 10 mg/m3 PM10 daily 50 µg/m3, not more than 35 times yearly 40 µg/m3 SO2 NOx The document “Analysis of main pollution source of NOx, SOx, VOC/odour and waste water in the BSR” based on EEA data is available on PlasTEP’s website: http://www.plastep.eu/fileadmin/dateien/Outputs/Emission_sources.pdf. 9 Main pollutants The composition of pure air is shown in the right. The main pollutants for which ambient air standards have been set to protect human health and welfare are listed below: • Ozone, O3 • Carbon monoxide, CO • Sulphur dioxide, SO2 • Nitrogen oxides, NOx • Volatile organic compounds, VOCs • Heavy metals, like Pb, Hg, Cd etc. • Particulate matter, PM • Halogen-containing compounds, like HCl, CH3Cl etc. Composition of pure air Element Parts per million Nitrogen 780 800 Oxygen 209 500 Argon 9 300 Carbon dioxide 300 Neon 18.2 Helium 5.2 Krypton 1.1 The main pollutants where the plasma technology has been applied are described below together with the effect on health and environment, main sources and regulations. The sources of pollutants in European countries according to European Environment Agency (EEA) 2008 year data are also shown. Sulphur dioxide (SO2) is a colourless gas with a pungent, suffocating odour. Short-term exposure to SO2 (5 minutes to 24 hours) results in adverse respiratory effects including bronchoconstriction and increased asthma symptoms. SO2 and other sulphur-oxide compounds can react with other compounds in the atmosphere to form small particles which can penetrate deeply into the lungs and cause or worsen respiratory disease such as emphysema and bronchitis and can aggravate existing heart disease. SO2 is formed primarily by combustion of sulphur-containing fuels such as coal, oil etc. About 95 % of sulphur in fuels is emitted in the form of SO2. Once in atmosphere, SO2 reacts slowly with cloud and fog droplets forming sulphuric acid which results in acidified soils, lakes and streams, accelerates corrosion of buildings and monuments and reduces visibility. The main sources of SO2 in EEA member countries are shown in the sector diagram above, the SO2 is mostly produced by large stationary sources and it is in particular present in coal and in charcoal and burning processes. The main sources vary depending on countries and some examples are available at PlasTEP’s website in the document “List of the emission processes in the Baltic Sea region”: http://www.plastep.eu/fileadmin/dateien/Outputs/OP41.1_List_of_the_emission_processes_in_the_Baltic_Sea_region.pdf 10 Nitrogen oxide (NO) and nitrogen dioxide (NO2) are collectively termed as nitrogen oxides (NOx). NO is colourless and odourless gas while NO2 is a brownish gas with a distinct sharp biting odour. NO is not soluble in water or on surfaces while NO2 is water-soluble. Both compounds but particularly NO2 penetrate into lung tissues and damage them, causing prepature death in extreme cases. In addition, respiratory diseases such as emphysema and bronchitis can follow and it may also aggravate existing heart disease. Nitrogen oxides are formed during the combustion of all types of fuels directly from air and from fuel nitrogen compounds. During combustion, NO is primarily formed while in atmosphere it is converted to NO2. NOx reacts with common organic chemicals and ozone to form a wide variety of other toxic products: nitroarenes, nitrosamines and nitrate radicals. It is also a precursor for acid rain alongside with SO2. NOx is also an important factor determining the ozone concentration. The main sources of NOx according to EEA are shown in the diagram (2008). The sources of NOx are much more diverse compared to SO2 but road transport can be taken as the main source for NOx together with energy production and distribution. Thus, a large part of NOx is produced by mobile sources while stationary sources are also important. The main sources for various BSR countries are available at PlasTEP’s website in the document: “List of the emission processes in the Baltic Sea region” http://www.plastep.eu/fileadmin/dateien/Outputs/OP41.1_List_of_the_emission_processes_in_the_Baltic_Sea_region.pdf Volatile organic compounds (VOCs) are a diverse group of organic compounds - there are several thousand different compounds classified as VOCs. According to European legislation, VOC is any organic compound emanating from human activities, other than methane, which is capable of producing photochemical oxidants by reacting with nitrogen oxide in the presence of sunlight. Many VOCs have known adverse health effects (cancer, allergy, respiratory diseases, headaches etc.) and are regulated as toxic air pollution. Some examples of 11 such VOCs are acetaldehyde, aniline, benzene, chloroform, formaldehyde, hexane, methanol, naphthalene, phenol, styrene, toluene, xylenes, etc. According to EEA (2009), the VOCs are mostly used as solvents and as chemical feedstock in industrial processes. Transportation is also an important source of VOCs where they are emitted as unburned fraction of fuel. A variety of indoor sources can also be listed: paints, lacquer, synthetic carpets, cosmetics, markers etc. Ozone (O3) is a colourless gas with a distinct smell. It is a secondary pollutant which forms in the lower atmpsphere by photochemical reactions of nitrogen oxides, VOCs and carbon monoxide. In the indoor environment it may be produced by photocopiers, for example. It is also formed in electrical discharges (lightning and artificial plasmas) which is an important aspect to consider when applying plasma technologies for air treatment. Ozone is harmful to the respiratory system and exposure to it may cause premature death, asthma, bronchitis, heart attacks and other cardiopulmonary problems. The allowed level of ozone is 120 µg/m³ (60 nmol/mol). Other important hazardous compounds are carbon monoxide (CO), hydrogen chloride (HCl) and hydrogen fluoride (HF), hydrogen sulfide (H2S), methyl mercaptan (CH3SH) released from incomplete burning of fuels, waste incinerators, pulp mills, fuel-fired boilers, metallurgy, petrochemical processes etc. Particulate matter (PM), a mixture of small particles and aerosols, is considered as the most serious air pollution health risk in the EU, leading to premature mortality. The particles with size below 10 micrometers are especially dangerous because they can reach all parts of the lungs after inhalation. 12 These particulates will cause lung diseases but the finest particles affect also the heart. The particulates are further divided as inhalable coarse particles with size between 2.5 and 10 µm and fine particles which are smaller than 2.5 µm. Both types of particles are produced from various sources. The coarse particulates are for example mineral dust, dirt, soot or smoke produced from dusty industries or roadways. Smoke or haze are composed of the fine particles. Conclusions The dominant sources of the various pollutants are somewhat different: SO2and NOx are produced together in large power plants while a large part of the NOx is also produced by transportation. VOCs are also produced by plants and transportation, however industrial use of solvents and indoor production are also very important sources. The principles of their removal methods also vary due to the type of source and chemical composition. For VOCs, oxidation to CO2 and H2O is the most common and desirable method. The oxidation of SO2/NOx on the other hand is not sufficient but may be one important step in the process. Mineralization of these species to fertilizers is one possible way to remove them from the exhaust. The reduction of NOx to N2 and O2 is also a possible solution, especially in transportation where SO2 is usually less abundant. Different concepts for removal methods are one reason why SOx/NOx removal and VOC oxidation are often dealt with separately. In the following section the conventional methods are shortly described. References/further reading European Envrionment Agency (EAA) website: www.eea.europa.eu United States Environmental Protection Agency (EPA) website: www.epa.gov EPA Air Pollution Training Institute course 415 „Control of Gaseous Emissions”: http://www.epa.gov/apti/catalog/cc415.html 13 1.2 Methods for air and exhaust gas depollution and role of plasma technologies Indrek Jõgi1, Saulius Vasarevičus2 1 University of Tartu, Tartu, Estonia Vilnius Gediminas Technical University, Vilnius, Lithuania 1 Introduction There are several ways to prevent the pollution which can be taken as 3P – Philosophy of Pollution Prevention: • modify the process: use different raw materials • modify the process: increase efficiency • recover and reuse: less waste means less pollution The recovery and reuse type of methods include also one common solution: end-of-pipe treatment which means collection of waste streams and add-on equipment at emission points which deals with the pollution on-site. The methods which are discussed below belong to the last type of methods. In general, the end-of-pipe treatment methods can be divided to mechanical, chemical and biological. Mechanical control is mostly usable for particulate matter while chemical and biological methods are suitable for removal of gaseous components. Control of gaseous pollutants There are several methods for the removal of gaseous pollutants which can be either combustion based methods or other methods. Combustion based methods remove pollutants by oxidizing them to less harmful substances. These methods are suitable for the combustible pollutants, mostly VOCs. Combustion methods are energy demanding but some of the energy can be recovered and catalysts can be used to further decrease the thermal energy budget. Other types of methods include the collection of the gaseous pollutants by absorption into liquids or by adsorption on solid surfaces. Condensation of vapours to liquid form can also be regarded as a collection method. These methods do not allow the removal of pollutants but in some cases the collected pollution can be reused or decomposed more efficiently by combustion based methods. Biofiltration and plasma-based 14 technologies are somewhat similar to the combustion methods but the oxidation occurs at low temperatures. Most of these methods will be explained more thoroughly in the following passages while the plasma-based technologies are covered in next chapters. Thermal oxidation Most of the organic substances can be decomposed and oxidized into CO2 and H2O at high temperatures in the presence of oxygen. Usually, the harmful gases can be decomposed at 700-780°C with sufficient efficiency while in some cases higher temperatures are necessary. With proper design of the oxidation plant, the gases have to spend about 0.5 to 1.5 sec in the chamber for sufficient removal. Thermal oxidation is more efficient at high concentration of pollutants because the amount of energy used for the heating of the gas is practically independent on the amount of pollutants. Furthermore, at higher concentration of pollutants the energy spent for oxidation of one pollutant decreases. In addition, at higher concentration of organic components, the exothermic combustion energy can also be used and in such cases the thermal oxidation becomes cost-efficient when the concentration is > 8 g/m3 (Rafflenbeul, 2010). Part of the heat energy can be recovered by heat exchangers where primary heat exchanger heats the untreated gas above 450 °C. Often, the secondary heat exchanger can be used for the recovering of additional part of the heat energy. One problematic aspect of thermal oxidation is the production of various nitrogen oxide compounds NOx which may be even more harmful than the initial solvent. Obviously, these types of plants can not be used for the removal of NOx/SOx without additional measures. One option for NOx removal is the so-called selective non-catalytic reduction (SNCR) where ammonia (NH3) or urea is injected to the hot exhaust (870-1100 °C). Part of the NO (20-60 %) is then reduced to N2 and H2O. The adoption of the method is rather complicated due to the need to attach injection nozzles to a space of narrow temperature range. Too low temperatures do not allow complete reaction of NH3 (so-called ammonia slip) and too high temperatures result in the oxidation of NH3 to additional NOx. Catalytic thermal oxidation Catalyst is a material that accelerates the chemical reactions without undergoing a change by itself. This allows the decrease of the decomposition temperatures to the range of 300 to 350 °C and considerably decreases the energy consumption of the process. Catalysts can be precious metals, non- 15 precious metals or metal-oxides dispersed on substrates with large surface area. The catalyst cost is often about half the cost of the system (Moulijn, 2001). The catalyst can only decompose a certain amount of pollutants in a reasonable time period and thus the concentrations of hazardous gases should not exceed 4 g/m3 at these temperatures (Rafflenbeul, 2010). When higher concentrations are present, higher temperatures have to be used. Similarly to the thermal oxidation, some of the heat can be recovered by heat exchanger and the energy from burning of organic substances can cover most of the energy demand when sufficiently high concentrations are used. However, at the start of the removal process the temperature is not sufficiently high for the oxidation and it is still necessary to use a burner. In addition to the removal of organic compounds, the catalysts can also be used for the removal of NOx from power plants and engines. In these Selective Catalytic Reduction (SCR) devices, the NOx can be reduced to N2 and H2O with the aid of reducing agents (NH3, urea or hydrocarbons). The overall NOx removal efficiencies reach 90 % at optimized conditions. The SCR with NH3 works in a relatively narrow temperature range of 290 to 400°C. At higher temperatures, the ammonia is oxidized to NOx while at low temperatures the reactions do not proceed sufficiently fast and the catalyst surface may become contaminated with NH3 salts. Ammonia slip is also a possible problem in such cases. In ordinary working conditions the temperature of the exhaust can be held in the optimal range of operation but the initial heating period of engines is often problematic when SCR is used. The catalysts have an inherent problem of surface contamination or so-called surface poisoning where some of the reaction products or particulate matter will remain on the surface and block the area for further catalysis. Sulphur poisoning is a common problem when the exhaust stream contains sulphurous gases. The contamination leads to slowl decrease of the efficiency of the catalyst and it is necessary to replace or regenerate it after certain operation periods. This increases the maintenance costs of the technology. The contamination with intermediate reaction products can be especially problematic at low temperatures. 16 Adsorption Adsorption of gaseous pollutants on the surfaces of highly porous materials can be used for collection of the harmful contaminants. The collection is done at low temperatures below 50°C until the surface becomes covered with the contaminants. A short regeneration step is usually following at higher temperatures or decreased pressures to clean the surface for the next adsorption cycle. Activated carbon has been used as adsorbent for more than 100 years but there are a number of other suitable materials including molecular sieves and silica gel. The surface area of these materials is generally more than 500 m2/g. Large exhaust gas masses can be removed with this method (> 30.000 m3/h) and the concentration of 1-4 g/m3 is economically feasible (Rafflenbeul, 2010). The initial investment costs are rather large making the method more cost-efficient for large emissions. The adsorption method becomes especially attractive when the pollutants can be recovered and reused. The adsorbents are usable for the collection of many organic compounds, but also mercury vapours and acid gases (hydrogen chloride etc). The method is generally suitable for compounds with molecular mass between 50 and 200 Da because lighter molecules do not adsorb on the surfaces while heavy molecules cannot be removed afterwards. Steam can be used for the regeneration of the adsorbent when the species are not soluble in water. Inert gases have to be used for cleaning when the contaminants at high concentration are prone to explosions. The regeneration constitutes important part of the maintenance costs in case of adsorption methods. Most desired way to deal with the adsorbed pollutants is to recover them by distillation of steam or by condensation. Another way to deal with the adsorbed pollutants is to apply thermal oxidation. The efficiency of oxidation becomes much higher at large concentration of pollutants. Non-thermal plasma oxidation of adsorbed organic compounds was also tested for regeneration of adsorbent and the energy efficiency was found to be very high (Kim, 2011). The NOx storage and reduction (NSR) systems for diesel engines use separate adsorption phase before reduction of NOx (Liu, 2011). The NSR is similar to SCR but instead of continuous catalytic reactions, the NO is oxidized and adsorbed on the metal-oxide surfaces at long oxygen rich conditions. Short regeneration step is carried out in fuel rich conditions which reduces the NOx to N2. This approach allows running the engine at more efficient conditions and avoids the use of separate reducing agent. The adsorption process becomes more complicated when there are several different pollutants in the gas stream. Heavy compounds, SOx and particulate matter which can not be removed in the 17 regeneration step slowly deactivate the surface. This may limit the lifetime of the adsorbent and sometimes additional methods are applied to collect the particulate matter before it reaches the adsorbent. Absoprtion In absorption plants, the exhausts are either directed through certain liquids which can absorb the gaseous contaminants or washed with streams of these liquids. As a consequence, the absorption methods are generally applicable for compounds that are soluble in aqueous liquids. For inorganic noxious gases (SOx, HCl, NH3 etc.) and several VOC-s, the plants can allow relatively cheap way for purification of high gas flows. The contaminants can either dissolve in the liquid or react with other chemicals in the liquid. Many soluble organic molecules are removed by the first method and the solubility of the molecules is limiting the removal. Additional chemicals are usually necessary for removal of acid gases and SOx whereas the amount of reactant in the liquid becomes the limiting factor. Lime slurry or NaOH are often used as neutralising materials and are part of the maintenance costs. Several washers with different agents can also be used in series to increase the efficiency. With varying loads, the control may become rather complicated and the wastewater may pose a problem afterwards. Absorption of NOx is also costly due to poor solubility and special chemicals (metal salts) are necessary to make the method usable. Biofilters Biological methods are usable when purified exhaust has low concentrations of contaminants <500 mg/m3. It can be interpreted as a type of absorption system which allows the decomposition of the absorbed contaminants. Biofiltration does not need high energy input and is especially suitable where water is cheap. One use of biofiltration is to wash the exhaust air through water or active charcoal suspension which contains an absorption agent with specific micro-organisms. After saturation with the solvents, the wastewater as an intermediate product is moved to a reaction basin. The solvent is treated there by the micro-organisms in several hours. Another method is to use towers with packing containing biological agents which can absorb solvents from humid air. Compost is often used as the substance for biological matter. The size of these plants has to be rather high (about 1 m2 per 100 m3/h of gas flow) for efficient removal of exhausts, especially solvents. The solubility of pollutants is another issue. With low solubility, the space loads have to be less than 40 g/m3·h. Bioreactors which are based on tower and 18 can work with moist air may reduce the necessary space but are more expensive. Odours are especially hard to remove with this type of filters and some contaminants can result in the inactivation of the filter. Control of particulate matter (PM) Particulates have been divided into two types by regulations: 10 µm (PM10) and 2.5 µm(PM2.5). The sources of particulates can be for example wood processing, rock quarries and coal power plants. Aerosols are micro-droplets which have diameter of less than 0.3 µm. The movement of aerosols becomes more similar to the gas molecules with decreasing diameter of aerosols and not all of the methods suitable for particulates are applicable for aerosols. Due to the diverse nature of particulates, mechanical methods for catching the particulates are often applied. The methods can be divided into three categories: use of mass inertia, use of electrostatic forces and use of filtration. These various methods include settling chambers, high-efficiency cyclones, electrostatic precipitators, spray towers, venturi scrubbers and bag house fabric filters. Cyclone Mass inertia is used in cyclones where the gas is forced to move in circular motion. In simplest cases, the contaminated gas is entering a tower from top and moves downwards in large circles at the outer part of the tower. At the conical end, the flow is constricted and forced to move upwards in smaller circles inside the downward spiral. Heavier particles are forced towards the walls by the mass inertia and they will fall down close to the walls due to smaller gas flow in there. The gas purity of < 20 mg/m3 is possible with these devices. There is a lower limit to the size of the particles which can be removed by cyclones called the cut-off diameter. This diameter depends on the parameters of the tower and on the working condition. Sometimes the cyclone is used together with spray towers to collect the mist of light aerosols which would otherwise slip out from the spray tower. Filtration In the case of filtration, the particles and aerosols are collected on filter surfaces which let through gas and a fraction of smaller particles. The particles can be collected by different ways. Larger particles cannot penetrate the openings of filters while smaller ones can be forced on the fibres by mass inertia or incidentally when the size is especially small. Textile fabrics are often used as filter materials. The gas purity of < 5 mg/m3 can be achieved with such filters. The filters have to be 19 cleaned after certain time periods which may range from few minutes to more than one hour. The cleaning can be achieved by using compressed air pulse or by shaking. More thorough cleaning has to be made in a yearly basis or even longer time periods. Electrostatic Precipitators In electrostatic filters, the particles and aerosols are charged by ion flow and then dragged towards the walls by electric field. This method is especially suitable when high gas flows are used and it can allow to collect even particles with <0.1 µm. Aerosols may also be removed by the electrostatic filters. Even though it is only usable for dust and aerosols, the filters are found as a part of the removal setup where gaseous products are also removed. It may be used to protect other parts e.g. adsorbent or catalyst from dust and mist. It may be also used as a final stage to collect the solid reaction products of some NOx/SOx removal devices. Conclusions None of these methods is usable for all types of pollutants and working conditions. Some of these are energy consuming (e.g. thermal oxidation methods), while some are costly due to the use of expensive materials (catalysts and adsorbers). The devices are often bulky and lack the control over the working parameters (biofilters). The use of plasma technologies allows to overcome some of these shortcomings. The plasma technologies may consume less energy at low pollutant concentration, are often more compact and their working parameters can be controlled by electrical energy input. Best results are often obtained by combining plasma technology with other methods listed in this chapter. Various plasma technologies and useful combinations with other technologies are explained more thoroughly in the following chapters. References/further reading Kim, H.-H. & Ogata, A. (2011). Nonthermal plasma activates catalyst: from current understanding and future prospects. The European Physical Journal Applied Physics, Vol. 55, No. 1, pp. 13806 (11), ISSN 1286-0042 Liu, G. & Gao P.-X. (2011). A review of NOx storage/reduction catalysts: mechanisms, materials and degradation studies, Catalysis Science & Technology, Vol 1, No 4, pp. 552-568, ISSN 2044-4753 Moulijn, J.A.; Makkee, M. & Diepen, A. (2001). Chemical Process Technology, Wiley, London, ISBN 978-0-47163062-3 Air Pollution Training Institute (APTI), U.S. Environmental Protection Agency 20 2. Plasma based technologies 2.1 Introduction to plasma science Indrek Jõgi1, Matti Laan1, Ronny Brandenburg2 1 University of Tartu, Tartu, Estonia INP Greifswald (Leibniz Institute for Plasma Science and Technology), Greifswald, Germany 2 The plasma state: definitions and its generation In physics and chemistry, plasma is an ionised gas containing free electrons, ions and neutral species (atoms and molecules) characterized by collective behaviour. Plasma is often referred as the “4th state of matter” since it has unique physical properties distinct from solids, liquids and gases. In particular, due to the presence of charge carriers, plasmas are electrically conductive and respond strongly to electromagnetic fields. It can contain chemically reactive media as well as excited species and emits electromagnetic radiation in various wavelength regions. Figure 2.1.1. Plasma as the fourth state of matter. 21 The majority of matter in the visible universe (stars, interplanetary and interstellar medium) is in the plasma state. Lightning, sparks, St. Elmo’s fire and the polar aurorae are examples for natural terrestrial plasmas. Furthermore, for more than 150 years plasmas have been generated artificially by supplying energy to gases, liquids or solids. Such plasmas are used and under investigation for various applications, e.g. surface modification, chemical conversion, light generation or controlled nuclear fusion. Natural as well as artificial plasmas cover an extremely wide range of parameters such as temperatures, particle densities and pressures. Broadly speaking, plasmas can be distinguished into thermal and non-thermal plasmas. In thermal plasmas all present species (electrons, ions and neutral species) are in the local thermal equilibrium, i.e. all species have the same mean free kinetic energy (temperature). Such plasmas are produced in fusion experiments and plasma arcs with temperatures higher than 104 K. Contrary, in other situations most of the coupled energy is primarily released to the free electrons which exceed the temperatures of the heavy plasma components (ions, neutrals) by orders of magnitude. Such mixtures of energetic electrons in a relatively cold mass of ions and neutrals are called non-thermal or non-equilibrium plasmas. If the gas temperature stays nearly at or slightly above room temperature the plasma is termed “cold plasma”. Even in non-equilibrium plasmas the gas temperature can increase to some 103 K. In such cases it is called “hot non-thermal plasma” or “translational plasma” since it marks the transition to the thermal regime. In fact cold as well as translational plasmas are used for gas depollution. Figure 2.1.2. Division of different types of plasmas by the temperature of plasma species. The most common method of plasma generation for technological and technical application is by applying an electric field to a neutral gas (Fig. 2.1.3). If the applied field exceeds a certain threshold (breakdown field strength) a gas discharge and thus plasma is formed. There are many different designs of plasma sources for depollution and the most important will be described in the next subsections. Alternatively by the interaction of an electron beam with gaseous medium plasma can be 22 generated. Such electron beam generated plasmas are used in the so-called electron beam flue gas treatment. The type of plasma and the gas temperature can be controlled by the electrical current flowing through the plasma or by the frequency of the applied voltage. The transition between different types of discharges at increasing currents is described more thoroughly in Appendix A. Dielectric barrier discharges (DBD) and Corona discharges (section 2.3) are examples of “cold” non-thermal plasmas where electrical currents remain low. With increasing current values, the temperature of gas and electrodes also increases and at certain threshold values there is transition to an arc discharge (section 2.2). The term discharge was adapted when charged capacitors where discharged through the air gap between the electrodes. When the gap between electrodes became sufficiently thin, electrical breakdown occurred. Figure 2.1.3. Low pressure discharge tube. The space between the electrodes, so-called discharge gap can be several tens of centimeters while the diameter of the electrodes may be several centimeters. The tube is usually evacuated to 100-10000 Pa. The frequency of applied electric field also affects the type and temperature of the plasma (Fig. 2.1.4). The electrical breakdown process is usually completed in the time interval of 10-8 to 10-6 s. As a result, up to the frequencies of several tens of kHz, the basic mechanisms will remain essentially the same: during each half-cycle the discharge is ignited again and a self-sustained discharge forms. At low alternating field the only effect is the exchange of the position of anode and cathode after each half cycle. At higher frequencies covering the radiofrequency and microwave range, there is a change in plasma breakdown mechanism which diminishes the role of electrodes and is used in so-called plasma torches (section 2.1). The temperature of the gas also increases with increasing frequencies and such high-frequency plasmas belong usually to translational or thermal plasmas. Figure 2.1.4. Division of different types of plasmas by the temperature of plasma species. 23 Plasma chemistry Non-thermal plasmas represent environments where very energetic chemical processes can occur at low ambient temperatures. In non-thermal or cold plasmas the energy of the electric field is preferentially deposited into the electron component of the plasma. As a result of the large mass difference between them, electrons gain much higher energy compared to ions and neutrals which remain close to room temperature (Fig. 2.1.5). Alternatively, high-energy electron beam may be injected from a separate source as already mentioned. Chemical processes in non-thermal plasmas are based on non-thermal activation of particles via collisions by these high-energy electrons. The quality and quantity of collisions is determined by the density of colliding particles and the kinetic parameters (e.g. mean velocity, collision frequency). In general three different phases have to be distinguished. The first phase is characterized by the electrical breakdown of the gas where free electrons with high kinetic energies are produced via ionising Figure 2.1.5. Plasma-chemical processes taking collisions. These electrons undergo further place in non-thermal plasmas. electron-molecule collisions, producing ionisation, dissociation, excitation and electron attachment (see Appendix A for further details). The second stage of non-thermal plasma chemistry is the radical formation and removal stage, where a multitude of inorganic reactions takes place. In particular radicals are generated through direct electron impact molecule dissociation and ionisation but other reactions (see Appendix A) may also be important (Chang, 2008). In air plasmas reactive oxygen species of O(3P), O(1D) and O2(1a) are generated by various processes from O2 and H2O (Appendix A). Furthermore in non-thermal plasmas generated in oxygen containing atmospheres at low gas temperatures, ozone, which is a strong oxidizing agent like atomic oxygen O, hydroxyl radicals •OH and peroxy radicals HO•2, will be formed. Many pollutant molecules are readily attacked by these free radicals. Decomposition of hazardous compounds is achieved without heating of the flue or off gas. Due to the presence of oxygen, water vapour and ozone, oxidizing reaction are dominant. The resulting chemistry is quite complex and depends on the gas mixture itself as well as the temperature. The hydrocarbons for example are oxidized to CO, CO2 and H2O while some partially oxidized species may also remain in the exhaust (see Appendix A). Plasma-based flue gas treatment for NO and SO2 proceeds also by oxidative processes which lead to the formation of NO2, N2O5 , HNO3, SO3 and H2SO4. A complete description of all 24 processes is outside the scope of this chapter and only the main important aspects will be discussed in the following. For more detailed and comprehensive information the reader is referred to several books and review papers, e.g. (Fridman, 2008; Penetrante & Schultheiss, 1993; Kim, 2004; Chang, 2008). Following the removal stage aerosol particles are formed through reaction of larger radicals with cluster ions and molecules. Aerosol formation is a quite important process which promotes the removal processes of pollutants due to heterogeneous reactions. Figure 2.1.6. Time-scale of plasma processes (adapted from Holzer, 2003). The time-scales of the various processes occurring in plasma can vary in wide range (Fig. 2.1.6). Electron induced processes like excitation and ionisation are very fast, tens of nanoseconds or less. Energy distribution of electrons reaches equilibrium with the external field also within picoseconds at elevated pressures. Electron induced dissociation processes take place in nanoseconds to microseconds. The time-scale for chemical reactions involving ground-state species is usually in milliseconds or seconds while radical reactions take place in micro- to milliseconds. The mean energy of electrons gained between the collisions depends on the ratio of electric field strength to the density of the gas: E/N which is called reduced electric field strength. This value has great importance in the studies of plasma! 25 Outlook As seen from previous section, plasma is an efficient source of highly active radicals (O, OH, HO2). Depending on the mean electron energy and gas composition, up to 30 % of the input energy can be used for the production of certain radicals, for example O or N. The reaction rates between these radicals and removed pollutants proceed often very quickly and almost 100 % of radicals can be used for pollutant conversion. Together with high efficiency of radical production this results in high energy efficiency of pollutant removal compared to heating where all gas species absorb energy and only a small amount of this energy is used for removal of pollutants. Dielectric barrier discharge and Corona discharge are good examples of such plasmas (see section 2.3 for further details). There are also pollutants which do not react efficiently with radicals in cold plasmas but at the higher temperatures obtained in translational plasmas (section 2.2) these reactions proceed more efficiently. Translational and hot plasmas are also useful due to their higher plasma density which is necessary for decomposition of pollutants by electron impact or ions. References/further reading Chang, J. S. (2008). Physics and chemistry of plasma pollution control technology. Plasma Sources Science and Technology, Vol. 17, 045004 (6 pp), ISSN 0963-0252 Fridman, A. (2008). Plasma Chemistry, Cambridge University Press, Cambridge, USA, ISBN-13 978-0-521-84735-3 Hammer, T. (1999). Applications of plasma technology in environemtal techniques. Contributions to Plasma Physics, Vol. 39, No. 5, pp. 441-462, ISSN 0863-1042 Holzer, F. (2003) Oxidation von organischen Verbindungen unter Nutzung von porösen und unporösen Feststoffen im nichtthermischen Plasma, Dissertation Universität Halle-Wittenberg Kim, H.-H. (2004). Nonthermal plasma processing for air-pollution control: A historical review, current issues, and future prospects. Plasma Processes and Polymers, Vol. 1, pp. 91-110, ISSN 1612-8850 Kogelschatz, U. (2004). Atmospheric-pressure plasma technology. Plasma Physics and Controlled Fusion, Vol. 46, 12B, pp. B63-B75, ISSN 0741-3335 Mizuno, A. (2007). Industrial applications of atmospheric non-thermal plasma in environmental remediation. Plasma Physics and Controlled Fusion, Vol. 49, pp. A1-A15, ISSN 0741-3335 Müller, S. & Zahn, R.-J. (2007). Air Pollution Control by Non-Thermal Plasma. Contributions to Plasma Physics, Vol. 47, No. 7, pp. 520-529, ISSN 0863-1042 Penetrante, B. M. & Schultheis, S. E. (1993). Non-Thermal Plasma Techniques for Pollution Control, NATO ASI Series, Vol. G34, Springer, Part A and B, Berlin, Germany, ISBN 0258-2023 26 2.2 Microwave driven plasmas for plasma depollution Miroslaw Dors, Mariusz Jasinski and Jerzy Mizeraczyk Szewalski Institute of Fluid Flow Machinery, Gdansk, Poland Plasma-based depollution by means of “hot” plasmas Decomposition of very stable pollutants (e.g. halogenated hydrocarbons and aromatic chlorofluorocarbons) by plasma species needs high gas temperatures which renders non-thermal plasmas inefficient. Because most of the energy deposited into plasma results in the heating of the gas plasma devices which allow higher energy input become necessary. There are several suitable plasma devices available, typical examples being high-intensity arcs and plasma torches. An overview is given in (Hammer, 1999). (a) (b) (c) Figure 2.2.1. General schemes of thermal and translational plasmas (a) free burning arc discharges in vertical and horizontal configurations; (b) plasma torch; (c) gliding arc. Electric arc discharges are driven between two electrodes (see Fig. 2.2.1 a) by high current (10 to 1000 A). Thus in arc plasmas high energy and current densities are reached (107–109 J m−3; 107–109A m−2). High-current arcs at atmospheric pressure can be characterized as thermal plasmas reaching temperatures in the range 5,000–50,000 K (Kogelschatz, 2004), which makes them very 27 useful for material processing (welding, cutting, spraying) and waste treatment. In plasma torches (also referred to as plasmatrons or plasma guns) the electrical energy is coupled into the working gas inside a nozzle and a high gas flow leads to the expansion outside the nozzle as a plasma jet (Fig. 2.2.1 b). A large variety of plasma torches has been developed. The majority of commercial torches use direct current arc, inductively coupled radio frequency discharges or microwave excited plasmas as the heat source and atmospheric pressure air as a working medium. Gliding arcs (Fig. 2.2.1 c) are another example for translational plasmas studied for gas depollution and other applications. They consist of at least two diverging electrodes which are passed by a gas flow. The discharge starts at nearest distance between the electrodes and is spreading by gliding along the electrodes in the direction of the gas flow which leads to cooling of the plasma. Microwave driven plasmas Microwave driven plasmas at atmospheric pressure are typical examples of transitional plasmas (nonequilibrium plasmas at elevated gas temperatures up to 4,000 K). However, the gas temperature is high enough to decompose stable organic molecules. The main parts of the microwave plasma sources (MPS) are a microwave generator (magnetron working usually at 2.45 GHz), microwave supply and measuring system, and a gas supply system. In most of the MPS systems, the microwave power is supplied from a magnetron to the discharge generator by a rectangular waveguide. At atmospheric pressures, the microwave plasmas are usually sustained within a dielectric (e.g. fused silica) tube inserted into the microwave field applicator (Fig. 2.2.2 and 2.2.3). Plasma is often induced in the form of a plasma “flame” at the tip of a field-shaping structure – a nozzle (e.g. Yamamoto & Murayama, 1967; Moisan et al., 1994, 2001). Large power densities deposited into the plasma result in deterioration of the dielectric tube and nozzle. High gas flow through the nozzle (Fig. 2.2.2) or an additional gas flow, so-called gas swirl, is used together with the main gas stream to protect the tube walls (Fig. 2.2.3). High-powers (up to 6 kW) and flow rates (hundreds of L/min) can be applied to these MPSs. Yet another type is the surface wave sustained MPS where the discharge in the form of a plasma column is generated inside a quartz tube cooled with a lossless liquid. 28 Figure 2.2.2. Photo and sketch of a waveguide-based nozzle-type MPS operated at high gas flow rate. The processed gas flow is introduced centrally to the discharge zone through a conical nozzle. The nitrogen flow is supplied tangentially through the four inlets creating a spiral vortex flow in the glass cylinder. This stabilizes the discharge in the centre of the cylinder and protects the wall from the discharge heat. 29 Figure 2.2.3. Photo and sketch of the waveguide-based nozzleless cylinder-type MPS. The plasma generated inside the glass discharge cylinder is stabilized by injecting a swirl gas which creates a vortex flow in the cylinder. The gas swirl also protects the cylinder wall from the discharge heat. 30 Decomposition of VOC-s by microwave plasmas The atmospheric-pressure plasmas generated by waveguide-based nozzle-type MPS have been tested for the decomposition of several chlorine and fluorine based VOCs in their mixtures with synthetic air or nitrogen (Jasinski et al., 2002a; Jasinski et al., 2002b; Mizeraczyk et. al., 2005). The VOCs included carbon tetrachloride CCl4 and chloroform CHCl3, aromatic CCl3F (CFC-11) and CCl2F2 (CFC-12), hydrochlorofluorocarbon CHClF2 (HCFC-22), hydrofluorocarbon C2H2F4 (HFC-134a) and fluorocarbon C2F6 (CFC-116) which are used as refrigerants, fire extinguishers and propellants. The concentrations of the processed VOCs were relatively high, i.e. usually a few percent but sometimes up to 50 %. The CFC-11 decomposition approached 100 % when CFC-11 initial concentration was 10 %, almost regardless of the gas flow rate. Even for pollutant concentration as high as 50 %, high decomposition rates up to 100 % were achieved at microwave power of 200-400 W and lowest flow rate of 1 L/min. With increasing gas flow rate, the resident time of the processed gases in the plasma decreased, resulting in lower value of the decomposition efficiency (60% at 3 l/min). The waveguide-based nozzleless cylinder-type MPS were used for destruction of refrigerant HFC-134a (Jasinski et al., 2009) with destruction mass rate and corresponding energetic mass yield of up to 34.5 kg h-1 and 34.4 kg per kWh of microwave energy absorbed by the plasma, respectively. The MPSs are also used in semiconductor industry for abatement of etch process effluent gases containing perfluorinated compounds (PFCs) and SF6. References/further reading Hammer, T. (1999). Applications of plasma technology in environmental techniques. Contributions to Plasma Physics, Vol. 39, No. 5, pp. 441-462, ISSN 0863-1042 Jasinski, M.; Mizeraczyk, J.; Zakrzewski, Z.; Ohkubo, T., & Chang, J.S. (2002a). CFC-11 Destruction by Microwave Torch Generated Atmospheric-Pressure Nitrogen Discharges. Journal of Physics D: Applied Physics, Vol. 35, No. 18., pp. 2274-2280, ISSN 0022-3727 Jasinski, M.; Mizeraczyk, J.; Zakrzewski, Z. (2002b). Decomposition of C2F6 in Atmospheric-Pressure Nitrogen Microwave Torch Discharge. Czechoslovak Journal of Physics, Vol. 52, Suppl. D, pp. D743-749 Jasinski, M.; Dors, M. & Mizeraczyk, J. (2008). Production of hydrogen via methane reforming using atmospheric pressure microwave plasma. Journal of Power Sources, Vol. 181, No. 1, pp. 41-45 Jasinski, M.; Dors, M. & Mizeraczyk, J. (2009). Destruction of Freon HFC-134a using a nozzleless microwave plasma source. Plasma Chemistry and Plasma Processessing, Vol. 29, No. 5 , pp. 363-372, ISSN 0272-4324 Kogelschatz, U. (2004). Atmospheric-pressure plasma technology. Plasma Physics and Controlled Fusion, Vol. 46, 12B, pp. B63-B75, ISSN 0741-3335 Mizeraczyk, J.; Jasiński, M.; Zakrzewski, Z. (2005). Hazardous gas treatment by atmospheric pressure microwave discharges. Plasma Physics and Controlled Fusion, Vol. 47, Suppl. 12B, pp. B589-B602 Moisan, M.; Sauve, G.; Zakrzewski, Z. & Hubert, J. (1994). An atmospheric pressure waveguide-fed microwave 31 plasma torch: the TIA design. Plasma Sources Science and Technology, Vol. 3, No. 4, pp. 584-592, ISSN 0963-0252 Moisan, M.; Zakrzewski, Z. & Rostaing, J.C. (2001). Waveguide-based single and multiple nozzle plasma torches: the TIAGO concept. Plasma Sources Science and Technology, Vol. 10, No. 3, pp. 387-394, ISSN 0963-0252 Yamamoto, M. & Murayama, S. (1967). UHF torch discharge as an excitation source. Spectrochimia Acta A, Vol. 23, No. 4, pp. 773-776, ISSN 0370-8322, 0584-8539, 1386-1425 32 2.3 Corona and Barrier Discharges Matti Laan1 and Ronny Brandenburg2 1 University of Tartu, Tartu, Estonia INP Greifswald (Leibniz Institute for Plasma Science and Technology), Greifswald, Germany 2 Introductory remarks Non-thermal plasmas in gas streams at atmospheric pressure can be generated either with the injection of a high energetic electron beam or the generation of a gas discharge by means of a sufficiently high voltage applied to two electrodes (gas discharges). In discharge generated plasmas the electrons have lower mean energies than in electron beam produced plasmas and as a consequence, plasma chemical reactions and energetic efficiency can differ substantially. However, discharge generated plasmas give the chance to construct more compact end-of-pipe treatment systems for small and medium size gas streams. Thus they are used in many gas depollution processes, e.g. deodorization. In cold non-thermal plasmas the free energetic electrons are able to produce radicals and other reactive species (e.g. ions) which react with the pollutant molecules or particles (section 2.1 and appendix A). Furthermore, if ions can be extracted from the discharge, fine particles can be charged and thus filtered electrically from the flue gas (Grundmann et al., 2007). Additionally, biological decontamination of air due to plasma treatment has been reported (e.g. Müller & Zahn, 2007). Several possibilities exist for the generation of plasmas by gaseous discharges (Becker et al., 2005; Fridman, 2008; Kogelschatz, 2004). The most common discharge types are dielectric barrier discharges (DBDs) and corona discharges. Both types of discharges are characterized by non-uniformly distributed thin plasma channels or discharge filaments with very short duration (10-100 ns) which prevents the development of a hot arc (see also appendix B). Corona discharge Corona discharges are generated in highly divergent electric field geometries. The primary ionisation processes are confined to the regions close to the high-field electrodes. The non-uniform configuration of the electric field is achieved by special electrode geometries, e.g. point-to-plane, wire-to-plane (see Fig. 2.3.1 a) or coaxial wire-in-cylinder configurations (see Fig. 2.3.1 b). 33 Corona is called unipolar if the high field region exists only close to one of the electrodes (active electrode). The discharge gap of unipolar corona has two characteristic regions: the ionization and drift zones. In the ionization zone, ionization and excitation by electron impact takes place while, in the drift region, reactions between the heavy species dominate. Depending on the polarity of the high field electrode, in the case of DC corona we are speaking about positive or negative corona discharges. Positive and negative coronas are sources of unipolar fluxes of positive and negative ions. (a) (b) Figure 2.3.1. Typical configurations of corona discharges for gas treatment (a) multi-needle-plate arrangement; (b) cylindrical wire-in-tube arrangement. The corona discharges of each polarity are characterized by different working modes depending on applied voltages. At conditions which are relevant for applications, the coronas work either in pulsed mode with distinctly visible streamers at lower voltages or in visually uniform glow mode at higher voltages (Korge et al., 1979 & 1993; Laan et al., 1994). These streamers are characterized by electrical currents which have very fast rise time (2 ns), last about 100 ns and have typically maximum current of 10 mA. At very high voltages (> 15 kV) the streamers reach the opposite electrode and arcs may form. This limits the power which can be deposited into the plasma. Corona discharges are usually DC-driven discharges, but for environmental applications they are often driven by high voltage pulses with rapid voltage rise (several kV per ns) and short duration (some tens of ns). This concept also referred to as pulsed corona discharges (PCD). Short pulses allow the increase of the average electron energy and thus, the efficiency of chemical processes (Naidis, 2012). Electrostatic precipitators and corona radical shower discharge DC-driven corona discharges are established in pollution control as electrostatic precipitators (ESP) for dust and aerosol removal of flue gases. In this application the active plasma is restricted to the region around the wire electrode. Between this so-called active zone and the opposite electrode (so-called 34 collecting electrode made as a plate or cylinder) a passive zone of low conductivity is formed. Ions generated in the active plasma zone enter the passive zone and drift to the collecting electrode. On their way they charge solid particles or droplets which migrate to the collecting electrode. The charged particles precipitate onto the collecting surfaces, are neutralized, dislodged and removed. Various types of dust, mist, droplet etc. down to submicron size can be removed under dry and wet conditions with high efficiency and low pressure drop (Kogelschatz, 2004). Thus, ESP technology uses the physical aspects of the corona discharge and not the chemical processes, although the promotion of plasma chemistry is possible, too. To decrease the power consumption and to overcome the “back corona effect”, pulsed operation was proposed (Mizuno, 2007; Kim, 2004). The back corona effect is caused by high resistivity dust (e.g. cement particles), which forms insulating dust layers on the collecting electrode and reduces the emissions of ions. Alternatively sulphur trioxide can be injected into the flue gas stream to lower the resistivity of the particles. An interesting concept of corona discharge is the (corona) radical shower discharge, which was developed in particular for NOx- and later for combined NOx- and SOx-removal (Okubo et al., 1996; Park et al., 1999). The discharge only treats a portion of the total contaminated exhaust flow. The treated gas with plasma generated active species is then injected in the total exhaust gas flow like a shower. Barrier discharge Barrier discharges, also referred to as silent discharges or dielectric barrier discharges (DBDs) are characterized by the presence of at least one dielectric layer between the electrodes (Kogelschatz, 2004; Wagner et al., 2003). Typical materials for dielectric barriers are glass, quartz and ceramics. Different arrangements and types exist as shown in Fig. 2.3.2. Besides the gas treatment, DBDs are used for industrial ozone generation and surface treatment (activation). Fig. 2.3.2a shows a so-called volume DBD in cylindrical geometry. The discharge gap is usually in the range of 1 mm. Fig. 2.3.2 b is a planar surface DBD, i.e. both electrodes (metal meshes) are in direct contact with the dielectric plates. Another type of DBD is the so-called coplanar discharge where both electrodes are embedded in the dielectric material. Special type of DBDs are so-called packed bed reactors (Fig. 2.3.2, c), where dielectric or ferroelectric pellets (e.g. alumina oxide Al2O3, titanium oxide, TiO2 or barium titanate BaTiO3) are packed between two electrodes (Holzer et al., 2005; Yamamoto et al., 1992). The pellets can be used as a catalyst (also see section 2.7) enabling direct interaction between plasma and catalyst and leading to uniform distribution of gas flow and plasma in the reactor. A disadvantage of packed bed reactors is relatively high pressure drop. 35 (a) (b) (c) Figure 2.3.2. Typical configurations of barrier discharges for gas treatment: (a) cylindrical asymmetric volume barrier discharge; (b) plate-like surface barrier discharge; (c) packed bed reactor. Due to the capacitive coupling of the insulating material to the gas gap DBDs can only be driven by alternating feeding voltage or pulsed DC voltages. When a sufficient voltage is applied to the electrodes, electrical breakdown occurs most commonly as number of individual discharge filaments or microdischarges (Kogelschatz, 2002). In packed bed reactors spontaneous polarization of ferroelectric pellets lead to high electric field at their contact points resulting in microdischarge formation. Microdischarges have a small duration (tens of nanoseconds in air), small size (diameter about 100 µm) (Brandenburg et al., 2005) and are distributed over the whole surface area. Due to the local charging of the dielectric surface after microdischarge inception the local electric field is weakened leading to the extinction of the microdischarge after several ten nanoseconds. Thus the barrier prevents the formation of a spark or arc discharge, keeping the plasma in the non-thermal regime. Despite the numerous applications of DBDs the knowledge of microdischarge development and thus plasma parameters and elementary processes within these microplasmas is not sufficient, although it is known that the properties of the subsequent microdischarges determines the efficiency and selectivity of the exhaust gas treatment. 36 Environmental application Both discharge types are used for the treatment of gases and are the objects of intensive research. The effect of gaseous discharge plasmas has been reported in numerous papers and textbooks as given in the references, e.g., Fridman, 2008; Kim, 2004; Veldhuizen, 2000; Mizuno, 2007. The plasma chemistry in contaminated gas containing oxygen leads to the generation of ozone as well as the oxidation of nitride oxides. The energy required for complete removal depends on the gas composition, the amount of contamination, the temperature, humidity and is influenced by the type and operation parameters of the discharge device. The main application of these gas discharges is deodorization where the plasma stage is combined with a catalyst or adsorber (e.g., active carbon). Organic odour molecules are indeed adsorbed but free radicals and ozone attack the odorous compounds too. The most effective non-thermal plasma treatment is achieved at low concentrations of contaminants. The combination with catalyst or adsorption offers synergies between both methods. Gas phase chemistry reduces the number of adsorbed species and thus the load on the adsorbing medium (adsorbent), and adsorbed molecules can react with active species from the plasma, leading to conversion of the adsorbate and regeneration of the adsorbent. This in turn increases the time of use for the adsorbent and thus reduces service costs. Another application of DBDs and PCD is low temperature oxidation technology. Further details are given in section 2.7. References/further reading Brandenburg, R.; Wagner, H.-E.; Morozov, A.M. & Kozlov, K.V. (2005). Axial and radial development of the microdischarges of barrier discharge in N2/O2 mixtures at atmospheric pressure. Journal of Physics D: Applied Physics, Vol. 38, No. 11, pp. 1649-1657, ISSN 0022-3727 Fridman, A. (2008). Plasma Chemistry, Cambridge University Press, Cambridge, USA, ISBN-13 978-0-521-84735-3 Grundmann, J.; Müller, S. & Zahn, R.-J. (2007). Extraction of Ions from Dielectric Barrier Discharge Configurations. Plasma Processesses and Polymers, Vol. 4, pp. S1004–S1008, ISSN 1612-8850 Holzer, F.; Kopinke, F.D. & Roland, U. (2005). Influence of ferroelectric materials and catalysts on the performance of non-thermal plasma (NTP) for the removal of air pollutants. Plasma Chemistry and Plasma Processing, Vol. 25, No. 6, pp. 595-611, ISSN 0272-4324 Kim, H.-H. (2004). Nonthermal plasma processing for air-pollution control: A historical review, current issues, and future prospects. Plasma Processes and Polymers, Vol. 1, pp. 91-110, ISSN 1612-8850 Kogelschatz, U. (2002). Filamentary, patterned, and diffuse barrier discharges. IEEE Transactions on Plasma Science, Vol. 30, No. 4, pp. 1400-1408, ISSN 0093-3813 Kogelschatz, U. (2004). Atmospheric-pressure plasma technology. Plasma Physics and Controlled Fusion, Vol. 46, No. 12B, pp. B63-B75, ISSN 0741-3335 Korge H.; Kudu K. & Laan M. (1979). The discharge in pure nitrogen at atmospheric pressure in point-to-plane discharge gap. Proc. 3rd Int. Symp. High Voltage Engineering , Milan, 1979, paper 31.04, pp. 1-4 37 Korge H., Laan M., Paris P. (1993). On the formation of the negative corona. Journal of Physics D: Applied Physics, Vol. 26, pp. 231-236, ISSN 0022-3727 Laan M., Paris P. (1994) The multiavalanche nature of streamer formation in inhomogeneous field. Journal of Physics D: Applied Physics, Vol. 27, No. 5 , pp. 970-978, ISSN 0022-3727 Mizuno, A. (2007). Industrial applications of atmospheric non-thermal plasma in environmental remediation. Plasma Physics and Controlled Fusion, Vol. 49, pp. A1-A15, ISSN 0741-3335 Müller, S. & Zahn, R.-J. (2007). Air Pollution Control by Non-Thermal Plasma. Contributions to Plasma Physics, Vol. 47, No. 7, pp. 520-529, ISSN 0863-1042 Naidis, G. V (2012). Efficiency of generation of chemically active species by pulsed corona discharges. Plasma Sources Science and Technology, Vol. 21, pp. 042001 (5 pp), ISSN: 0963-0252 Okubo, T.; Kanazawa, S.; Nomoto, Y.; Chang, J. S. & Adachi, T. (1996). Time dependence of NOx removal rate by a corona radical shower system. IEEE Transaction on Industry Applications, Vol. 32, pp. 1058-1062, ISSN 0093-9994 Park, J.Y.; Tomicic, I.; Round, G.F. & Chang J.S. (1999). Simultaneous removal of NOx and SO2 from NO-SO2-CO2N2-O2 gas mixtures by corona radical shower systems. Journal of Physics D: Applied Physics, Volume: 32, 9, pp. 1006-1011, ISSN 0022-3727 Van Veldhuizen (Ed.), E.M. (2000): Electrical Discharges for Environmental Purposes. Nova Biomedical. ISBN/EAN: 978-1-56072-743-9 Wagner, H.-E.; Brandenburg , R.; Kozlov, K.V.; Sonnenfeld, A.; Michel, P. & Behnke, J.F. (2003). The barrier discharge: basic properties and applications to surface treatment. Vacuum, Vol. 71, pp. 417-436, ISSN: 0042207X Yamamoto, T.; Ramanathan, K.; Lawless, P.A.; Ensor, D.S.; Newsome, J.R.; Plaks, N. & Ramsey, G.H. (1992). Control of volatile organic-compounds by an ac energized ferroelectric pellet reactor and a pulse corona reactor. IEEE Transactions on Industry Applications, Vol. 28, 3, pp. 528-534, ISSN: 0093-9994 38 2.4 Hollow cathode discharges Hana Barankova and Ladislav Bardos Uppsala University, Uppsala, Sweden Hollow cathodes The hollow cathode exhibits the so-called hollow cathode effect: a large increase in current density with reduced separation of the two cathodes. The configuration of the hollow cathode can be realized by different geometries: by two parallel plates at the same potential or by a cylindrical shape. The hollow cathode can operate in DC, pulsed DC, AC and radio frequencies (RF, i.e. MHz-range) regimes. The concept of the RF hollow cathode for activation of gas was patented in 1985 (Bardos et al., 1985). The hollow cathodes are versatile and scalable and can be used both at reduced and atmospheric pressures. In principle, the hollow cathode is a non-equilibrium plasma source because of the population of high energy electrons. The pendulum motion of the electrons between repelling potentials of the opposite space charge sheaths enhances ionisation and excitation of gas species and results in high plasma density. Figure 2.4.1 shows a typical V-I characteristics of the hollow cathode (Barankova & Bardos, 1999, 2006). 350 300 V [V] 250 200 150 100 0 0.5 1 1.5 I [A] 2 2.5 3 Figure 2.4.1. V-I characteristics measured for the rf cylindrical hollow cathode at pressure of 67 Pa (0.5 Torr). 39 At low levels of delivered RF power, there exists only a RF discharge, which has a similar role to a predischarge in the DC hollow cathodes (Schoenbach, 1993). Above a certain power threshold the breakdown of the plasma-sheath takes place. The breakdown is accompanied by a voltage drop and the hollow cathode plasma is developed. When the temperature at the electrode is high enough (caused by the ion bombardment), the discharge is transformed into the hollow cathode arc. The decrease of the voltage takes place again. The hollow cathode effect occurs only within a certain range of inner diameters (wall separations) at a definite pressure. Atmospheric pressure requires the inner diameter/wall separation correspondingly reduced as the space charge sheath becomes thinner with the pressure (Barankova & Bardos, 2000). The atmospheric pressure hollow cathode discharge has been successfully ignited and operated in noble gases, in nitrogen and air and in different gas mixtures. Typical inner diameters are 300-400 µm. Both the V-I characteristics and peak-to-peak voltage versus RF power curves exhibit the transitions from the predischarge to the hollow cathode discharge, similar to those in the reduced pressures regimes. Fused Hollow Cathodes The Fused Hollow Cathode (FHC) represents a new concept in non-thermal plasma sources. The FHC cold atmospheric plasma source is based on the simultaneous generation of multiple hollow cathode discharges in an integrated open structure with flowing gas (Barankova & Bardos, 2000), see Fig. 2.4.2. The operation stability of FHC systems is excellent; the plasma is uniform and does not exhibit streamers. The FHC system allows generation of cold plasma in both monoatomic and molecular gases (Barankova & Bardos 2003). Figure 2.4.2. Fused Hollow Cathode. 40 The upstream FHC concept with aerodynamic stabilization and pulsed DC generation was successfully tested for conversion of NOx in air mixtures. The discharge works as an 100% oxidation catalyst, converting NO to NO2 completely, without any additives and/or catalysts over a certain range of the repetition frequencies (Barankova & Bardos, 2010). This range, forming the processing window, increases with the oxygen content in the mixture. The electrode material plays an important role in the plasma chemical kinetics as it brings about its own material constants, e.g. work function, secondary electron emission coefficient and also catalytic activity. Due to an optimized geometry and efficient transfer of power to the electrons in the system, the power consumption for gas conversion is extremely low. Typical specific energy inputs within the processing window are around 5 J/l, i.e. 0.14 kWh/100 m3. Hybrid Hollow Electrode Activated Discharge (H-HEAD) The H-HEAD source combines a microwave antenna with a hollow cathode powered by a RF or pulsed DC generator (Bardos & Barankova, 2005). The microwave plasma is produced by surface waves propagating along the antenna. The gas flows through this microwave antenna that serves at the same time as the hollow cathode. The combination of microwave excitation and hollow cathode plasmas enables a very efficient control of plasma parameters. The H-HEAD source can produce long plasma columns and it works both with monoatomic and molecular gases. References/further reading Bardos, L. et al. (1985), Czech Patent PV 4407-85 Schoenbach, K. H. (1993). The effect of magnetic fields on hollow cathode discharges, Proc. 21st Int. Conf. on Phenomena in Ionized Gases (ICPIG), Bochum, Germany, September 1993, Vol. 3, p. 287 Baránková, H. & Bárdos, L. (1999). Effect of gas and cathode material on the r.f. hollow cathode reactive PVD, Surface Coating Technology, Vol.120-121, pp. 704-708, ISSN 0257-8972 Baránková, H. & Bárdos, L. (2000). Fused hollow cathode cold atmospheric plasma, Applied Physics Letters, Vol. 76, pp. 285-287, ISSN 0003-6951 Baránková, H. & Bárdos, L. (2003). Hollow cathode cold atmospheric plasma sources with monoatomic and molecular gases, Surface Coating Technology, Vol. 649, pp. 163–164, ISSN 0257-8972 Bárdos, L. & Baránková, H. (2005). Characterization of the cold atmospheric plasma hybrid source, Journal of Vacuum Science and Technology, Vol. A23, pp. 933-937, ISSN 0734-2101 Baránková, H. & Bárdos, L. (2006), Hollow cathode and hybrid plasma processing, Vacuum, Vol. 80, pp. 688-692, ISSN 0042-207X Baránková, H. & Bárdos, L. (2010). Effect of the electrode material on the atmospheric plasma conversion of NO in air mixtures, Vacuum, Vol. 84, pp. 1385-1388, ISSN 0042-207X 41 2.5 Electron beam generated plasmas for gas depollution Andrzej Pawelec and Andrzej G. Chmielewski Institute of Nuclear Chemistry and Technology, Warsaw, Poland Electron beam generated plasmas Plasma used for removal of harmful pollutants from off gases may be effectively generated by electron beams. To generate plasmas with electron beams special electron accelerator units are needed (see Fig. 2.5.1.). Electrons are produced from the cathode by heating and then accelerated inside the vacuum tube. The electron beam transits from the beam generation environment at vacuum pressure (10-5 mbar) into the flue gas stream at atmospheric conditions via a beam window and then through a secondary window (Chmielewski et al., 1995). Due to a beam alignmentsteering system the beam will scan across or along the flue gas stream. Beam scanning and window cooling is necessary to avoid destruction of the titanium windows. The beam acceleration ranges from 0.7 to 1.2 MeV, allowing the beam to penetrate the windows without excessive energy loss. The maximum power per accelerator available nowadays is up to 400 kW, total beam power in installations can exceed 1 MW (Department of Energy, 2010). Next generation electron beam techniques use radio frequency cavity systems instead of DC transformers (Edinger, 2008). This enables pulsed driven beams with optimized energy control. 42 Figure 2.5.1. Electron accelerator in Pomorzany. Electron beam flue gas treatment (EBFGT) chemistry EBFGT is a dry-scrubbing process of multipollutant emission control. The most important advantages of the process are simultaneous removal of SO2, NOx, VOC and others in one process, high effectiveness of the process, lack of problematic wastewaters, no generation of waste due to agricultural use of the by-products as fertilizers and simplicity of the construction and operation of the facility which makes retrofitting easy. The main components of flue gases are N2, O2, H2O, and CO2, with SOx and NOx in much lower concentrations. Ammonia NH3 may be introduced as an additive to support the removal of SOx and NOx. The electron energy is transferred to the gas components present in the mixture in proportion to their mass fraction. Fast electrons interact with gas creating various ions and radicals, the primary species formed include N2+, N+, O2+, O+, H2O+, OH+, H+, CO2+, CO+, N2*, O2*, N, O, H, OH, and CO. In case of high water vapour concentration the oxidizing radicals •OH, HO2• and O(3P) as well as excited ions are the most important products (Person & Ham, 1988). These species take part in a variety of ionmolecule reactions, neutralization reactions, dimerization etc. SO2, NO, NO2, and NH3 cannot compete with the fast reactions because of very low concentrations, but react with N, O, •OH, and HO•2 radicals. Ozone, O3, which forms by reaction between O radical and O2 molecule, is also important in oxidation of NO. The EBFGT process consists of four main stages – cooling and humidification, ammonia injection, irradiation, by-product collection (see Fig. 2.5.2). Ammonia is the main reagent of the process and its role is to support the removal of SOx and NOx. Irradiation of flue gas by electron beam is necessary for oxidation and further removal of pollutants by a number of chemical reactions. Reaction pathways There are several known pathways of NO oxidation (appendix A). In the case of EBFGT the most common are involving the oxidation of NO to NO2 by O(3P), O3 and HO•2 as a first step (Tokunaga & Suzuki, 1984). The oxidation product NO2 is further converted to nitric acid HNO3 in the reaction with • OH radicals. Finally, HNO3 aerosol further reacts with NH3 giving ammonium nitrate, NH4NO3. NO is also partly reduced to atmospheric nitrogen by N radicals (Namba et al., 1990). The main parameter in NOx removal is the dose. The rest of parameters play minor role in the process. Nevertheless in a real, industrial process, dose distribution and gas flow conditions are important from the technological point of view. There can be also several pathways of SO2 oxidation depending on the conditions. In the EBFGT process the most important are radio-thermal and thermal reactions (Namba, 1993). Radio-thermal reactions proceed through oxidation of SO2 by •OH radical whereby the HSO3 formed creates ammonium sulphate, (NH4)2SO4, in the following steps involving molecules of O2, H2O and NH3 43 (appendix A). The thermal reaction is based on the process where SO2 reacts directly with NH3 and the (NH3)2SO2 formed is then converted to (NH4)2SO4 by reactions with O2 and H2O. The total yield of SO2 removal consists of the sum of the yields of thermal and radio-thermal reactions (Chmielewski, 1995). The yield of the thermal reaction depends on the temperature and humidity and decreases with the temperature increase. The yield of the radio-thermal reaction depends on the dose, temperature and ammonia stoichiometry. In the case of VOCs decomposition, the process itself is based on the similar principles as the primary reactions concerning SO2 and NOx removal i.e. free radicals attack organic compounds chains or rings causing VOC decomposition. For aromatic hydrocarbons, VOC decomposition will mainly go through two channels. The first one is due to positive ions charge transfer reactions: M+ + RH → M + RH+ (RH=VOC, eg. benzene or PAHs). Part of the VOC will be decomposed by rapid charge transfer reactions because RH has lower ionisation energy than most primary positive ions. The second channel is by radical – neutral particles reactions initiated by •OH radicals (appendix A). The addition of an •OH radical to the aromatic ring (eg. toluene) •OH + C6H5CH3 = R1• results in H atom abstraction either as a •H radical or H2O which in turn leaves behind a new radical R2•. Radicals (R1•, R2•) formed above go through very complex reactions, namely O2 addition, O atom release, aromatic – CHO (-dehydes), -OH (-ol) compounds formed or ring cleavage products. New radicals are formed in the process which results in chain reaction of decomposition of VOC. EBFGT process and installations The EBFGT process was invented in Japan in 1970's. In that phase, the fundamentals of the process were elaborated. During the next twenty years pilot plant scale experiments were performed in Germany (Karlsruhe), USA (Indianapolis), Japan (Nagoya) and Poland (Kaweczyn near Warsaw). The results were used for design and construction of industrial scale EBFGT plants. Two such installations were constructed by the Ebara corporation in China (Chengdu and Hangzou), however after a short time of exploitation they were closed down. These installation were designed for mostly SO2 removal and NOx removal was very low (10 – 15%). The first EBFGT industrial installation for simultaneous SO2 and NOx removal was constructed in 2000 in Pomorzany Power Plant in Szczecin, Poland. Total capacity of the plant is 270,000 Nm3/h of flue gas, with SO2 removal efficiency above 95% and NOx removal above 70%. A dose of up to 10 kGy (1 kGy = 1 kJ/kg flue gas) is required for NOx removal, while SO2 can be removed in proper conditions at lower energy consumption. Nowadays most technical problems occurring in the prototype installations have been solved (Chmielewski et al., 2004). At the moment design works on another implementation of this technology in Bulgaria are in progress. 44 The overall construction of the facility is similar in all installations of this kind. According to the constructional and functional point of view, the installation may be divided in four main parts: • cooling and humidification of flue gases, • ammonia supply system, • reaction unit and • byproduct filtration. Clean gas Figure 2.5.2. Schematic picture of EBFGT process. After the boiler, fly ash is removed from the flue gas by an electrostatic precipitator (ESP) and cooled down and humidified in spray towers. Cooled and humidified gases are than exposed to the electron beam radiation after the injection of ammonia. The high-energy electrons form the plasma and initiate a series of the above listed reactions which lead to the removal of the SOx and NOx by forming ammonium sulphate (NH4)2SO4 and ammonium nitrate NH4NO3 respectively. The reacted gas then passes through a particulate removal device (e.g. ESP) to remove the ammonium sulphate and ammonium nitrate which are used as fertilizers. Future developments Although already implemented, the technology is relatively young and further development is under progress. In recent investigations the decrease in electrical energy consumption and increase in availability are two of the most important tasks. In addition, the removal of VOCs, dioxins, mercury and other pollutants from flue gases using EBFGT has been investigated. The most important development concerns the application of electron beams for the reduction of polychlorinated dibenzodioxin (PCDD, so-called dioxins) and polychlorinated dibenzofuran (PCDF) emission from municipal solid waste incinerators (Hirota et al., 2003). Electron beam irradiation has also demonstrated high levels of mercury oxidation at the bench scale, and the technology might help to 45 improve mercury removal in wet scrubbers or wet ESPs when employed as a primary or secondary mercury oxidation technique (J.C. Kim et al., 2010). The technology has proved its ability for industrial scale applications and may be regarded as a good promoter of environmental plasma technologies. References/further reading Chmielewski, A. G.; Zimek, Z.; Panta, P. & Drabik, W. (1995). The double window for electron beam injection into the flue gas process vessel. Radiation Physics and Chemistry, Vol. 45, No. 6, pp. 1029-1033, ISSN 0969-806X Chmielewski, A.G.; Licki, J.; Pawelec, A.; Tyminski, B. & Zimek, Z. (2004). Operational experience of the industrial plant for electron beam flue gas treatment. Radiation Physics and Chemistry, Vol. 71, pp. 439-442, ISSN 0969806X Department of Energie [DOE] (2010) Accelerators http://www.acceleratorsamerica.org/report/index.html for America’s Future, Washington, DC Edinger, R. (2008). Reduction of SOx and NOx emissions by electron beam flue gas treatment. available at http://www.ebfgt.com/file_library/userfiles/file/ Hirota, K.; Hakoda, T.; Taguchi, M.; Takigami, M.; Kim H. & Kojima, T. (2003) Application of electron beam for the reduction of PCDD/F emission from municipal solid waste incinerators. Environmental Science and Technology, Vol. 37, No. 14, pp. 3164-3170, ISSN 0013-936X Kim, J.C.; Kim, K.H.; Al Armendariz & Al-Sheikhly, M. (2010). Electron Beam Irradiation for Mercury Oxidation and Mercury Emissions Control. Journal of Environmental Engineering, Vol. 136, No. 5, pp. 554-559, ISSN 0733-9372 Namba, H., Tokunaga, O., Suzuki, R. & Aoki, S. (1990) Material balance of nitrogen and sulfur components in simulated flue gas treated by an electron beam. Applied Radiation and Isotopes, Vol. 41, No. 6, pp. 569-573, ISSN 0969-8043 Tokunaga, O. & Suzuki, N. (1984). Radiation chemical reactions in NOx and SO2 removals from flue gas. Radiation Physics and Chemistry, Vol. 24, No. 1, pp. 145-165, ISSN 0969-806X 46 2.6 Power supplies and power determination on electrical discharges Marcin Hołub West Pomeranian University of Technology, Szczecin, Poland Power supplies for dielectric barrier discharge and pulsed corona reactors Dielectric Barrier Discharge (DBD) reactors and Pulsed Corona (PC) reactors, which are most widely used as non-thermal plasma sources in pollution control, require typically different supply waveforms. DBD reactors are most often supplied using alternating, sinusoidal voltage while the corona discharge systems have pulsed supply. In the case of DBD, classical 50 or 60 Hz supplies with high-voltage transformers are predominantly used (Sasoh et al., 2007; Kostov et al., 2009). However, for adequate control of the chemical processes, it is often necessary to employ higher operation frequencies and to control the average power. Therefore, modern supply system designs include power amplifiers with high-voltage transformers (Francke et. al., 2003; Mok et al., 2008) or a number of solid-state switch based power electronic converter topologies, often resonant ones (Casanueva et al., 2004). Resonant operation complicates the fluent control of the output power and often a time-averaged burst (often called pulse density modulation - PDM) technique is used instead (Fujita et al., 1999). Considering the supply voltage waveforms, a set of typical patterns can be defined. An overview of common voltage waveforms is given in Fig. 2.6.1. Most common AC sinusoidal high voltage waveforms at 50 or 60 Hz are shown in Fig. 2.6.1-1. In order to influence the average reactor power, pulse density modulation (PDM) techniques can be used (Fig. 2.6.1-2). Optimization of effectiveness as well as voltage potential distribution levelling sometimes results in a discontinuous, bipolar waveform with a pre-ionisation phase (Fig. 2.6.1-3). Pulsed supplies are also used both in the case of DBD and PC reactors either in unipolar (Fig. 2.6.1-4, 2.6.1-5) or bipolar mode (Fig. 2.6.1-6). A different group of supplies uses DC high voltage (Fig. 2.6.1-7), or pulsed supply with a DC bias (Fig. 2.6.1-8). 47 Figure 2.6.1. DBD reactor supply voltage waveforms. For the generation of these waveforms, a large variety of different constructions of supply systems can be distinguished. Only a very brief overview of existing designs will be presented in the following. Basic configurations of non-thermal plasma supply systems working at low and high frequencies are depicted in Fig. 2.6.2. In the case of low frequency AC supply systems, current limiting resistors may be used at the primary or secondary transformer side (Rp or Rs). In the case of DC power supplies a reactor current-limiting resistor (RDC) is sometimes implemented (Fig. 2.6.2 a). In controllable systems an adjustable transformer may be used too. The usual shortcomings of low-frequency systems are the limited efficiency ratings (about 40% for low power systems) and large weight/volume consumption. High frequency supplies have usually a rectifier as the first power electronic converter (Fig. 2.6.2 b). Then different configurations and topologies are used. For example, a high frequency – high voltage transformer (HF, HV) is often present in such systems. Additional pulse forming networks are occasionally implemented in order to shape the output voltage waveform. 48 (a) (b) Figure 2.6.2. Basic configurations of power supplies: (a) low frequency systems; (b) high frequency systems. In the case of high-frequency AC supply systems a power electronic converter is usually implemented on the primary side of a high voltage – high frequency transformer. An example of such a system is given in Fig. 2.6.3. In the simplest cases, the secondary side of the transformer is connected directly to the DBD discharge reactor. In other cases (as presented in Fig. 2.6.3) an additional converter is present to enable direct control of the high voltage parameters. In the given example, a series resonant converter is implemented and the primary side control is only used to define the DC link voltage value and therefore enables output voltage peak–to–peak value control. It must be emphasised that in the case when many resonant circuits are used, the output frequency of the system is fixed by the values of the components used as well as by the parasitic capacitance of the reactor itself. Figure 2.6.3. Power electronic, high frequency supply system with secondary resonant converter. 49 Pulsed supply systems are constructed in a variety as large as in the case of AC sources. Classical constructions often implement the so-called Marx generator topology (Marx, 1928) and Fitch generator topology (Fitch et al., 1968). Both constructions are presented in Fig. 2.6.4. Large installations have very high peak values of voltage (up to several MV) and current (up to 0.5 MA) and pulse modulators are therefore constructed implementing pulsed thyristors, gas switches (thyratrons, krytrons) or spark gap switching apparatus. As a downside, these technologies allow only a low frequency of operation and a limited lifetime. (a) (b) Figure 2.6.4. Typical pulsed system topologies: (a) Marx generator: SG - spark gap switch; R – charging resistor; C – storage capacitor; (b): Three stage Fitch generator principal construction: S – power switch, C – capacitor storage; L – recharge inductor; R – charging resistor. 50 Solid state technology enables much higher operating frequencies and very long lifetime but has a limitation on the maximum allowable blocking voltage and maximum repeatable peak current per single power semiconductor. High voltage MOSFET transistors and HV IGBT transistors are typically used for power electronic supply systems. In order to overcome single element limitations power switching stacks are produced. Typical efficiency values of different designs are summarized in the table below. Table 2.6.1. Typical efficiency ratings of plasma system power supplies Supply system Character Efficiency [%] Pulsed Magnetic pulse compression Up to 76 Resonant AC output Up to 96 Pulsed Bipolar output waveform Up to 76,9 LF AC HV LF transformer ~ 40 Methods of plasma power estimation in DBDs A literature study shows that four approaches are used for plasma power estimation. Current and voltage waveform based methods Most natural from the electrical point of view is the definition-based measurement of active power delivered to the load (the gas discharge plasma) based on the momentary values of voltage and current: P= 1 T t +T ∫ u ⋅ idt (1) t This approach is sometimes not easy to implement practically due to heavy distortion of the filamentary current waveform (most of all in sinusoidal-supply systems) and the necessity to include all microdischarges in a single discharge period. This implies the use of high bandwidth current sensor as well as high frequency digital oscilloscope. Careful notice has to be paid to the proper span of integration in order to integrate full periods of the supply waveforms. Analytical approach An early publication in the field of electrochemistry gave an analytical solution to the topic of plasma power. Most commonly used in the so-called Manley equation (Manley, 1943): 51 P = 4U ig C D f CD (U peak − U ig ) C D + CG (2) taking into consideration the main reactor and power supply parameters as Uig – reactor’s critical voltage (minimum external voltage at which microdischarges are observed), f – supply electrical frequency, Upeak – peak reactor voltage, CD, CG – reactor dielectric and gas gap capacitances. Such an approach may result in large errors in the power estimation due to nonlinear changes of the reactor parameters (capacitances) as a function of reactor temperature, dielectric material losses and reactor parameter estimation itself. Voltage-charge-plots or Lissajous figure approach A variation of Manley’s approach is often used called the Lissajous figure approach. Based on the relation (3): P = f ⋅ E = f ⋅ ∫ U (t ) T dQ dt = f ⋅ C P ∫ U (t )dU P , dt (3) the value of power can be obtained, Cp is an additional measurement capacitor (Fig. 2.6.6 a) connected in series with the measured reactor (with CP >> CR; CR is the total capacity of the reactor consisting of CD and CG). Based on capacitor voltage, which represents the dissipated charge Q, the energy E can be determined and power power can be calculated. Typical Lissajous curve is given in Fig. 2.6.5 b. The Lissajous figure also allows the identification of the main reactor parameters as CD and CG (see Fig. 2.6.5 b with graphical overview of main, possible readings). (a) (b) Figure 2.6.5. Lissajous figure measurement method: (a) measurement scheme; (b) exemplary Lissajous figure. Comparative approach The comparative approach to power determination uses a dummy, namely a high voltage - high 52 frequency measurement capacitor with equal capacitance to the DBD reactor being considered. The results of capacitor and reactor operation are compared in order to determine the input power. The input power from the power supply is taken for comparison assuming constant and equal power supply and reactor efficiency. (a) (b) Figure 2.6.6. (a) Typical voltage and current waveforms during experiments; (b) momentary power resulting from voltage and current waveforms. Fig. 2.6.6. depicts typical momentary voltage, current and power waveforms of a DBD reactor. The reactor current consists of the sinusoidal displacement current through the capacitance CD and the active discharge current through the gas gap. The latter appears when the momentary voltage reaches 53 the critical value for electrical breakdown and stops when the voltage peak value is reached. Only the active discharge current should contribute to the momentary reactor power, as shown in Fig. 2.6.6 b. The mean value of the momentary power is the plasma power. Measured values obtained by all above mentioned methods are summarized in the table 2.6.2. As can be noticed, there are some drifts of reactor parameters with the operation time. This is related to the increase of reactor temperature (starting at room temperature) with operation time. Nevertheless, the overall reactor capacitance does not change significantly. Table 2.6.2. Measured reactor parameters in dependence of reactor operation time and operating voltage. Measurement number 1 – starting DBD 2 – after 5 minutes of operation 3 – lower voltage Measured average power [W] 44.73 42.3 CR [nF] CD [nF] CG [nF] 1.347 1.356 2.13 2.08 3.66 3.9 11.37 0.79 1.59 1.57 Very comparable results were obtained in case of Lissajous figure approach and definition based measurement, 3.3% was the largest difference in readings. Both methods are recommended for plasma power measurements. Definition-based methods are also a little simpler due to the fact that no CP capacitor is necessary. Attention has to be drawn to proper measurement sensor selection and oscilloscope sampling frequency. As already mentioned, the time span of time integral has to ensure a full number of voltage periods. References/further reading Casanueva, R.; Azcondo, F. J. & Bracho S. (2004). Series–parallel resonant converter for an EDM power supply. Journal of Materials Processing Technology , Vol. 149, pp. 172-177, ISSN 0924-0136 Fitch R. A. (1968). Electrical Pulse Generators. US Patent US 3,366,799. 30 Jan. 1968 Francke, K. P.; Rudolph, R. & Miessner, H. (2003). Design and operating characteristics of a simple and reliable DBD reactor for use with atmospheric air. Plasma Chemistry and Plasma Processing. Vol. 23, No. 1, pp. 47-57, ISSN 0272-4324 Fujita, H. & Akagi, H. (1999). Control and performance of a pulse-density-modulated series resonant inverter for corona discharge processes. IEEE Transactions on Industry Applications. Vol. 35, No. 3, pp. 621-627, ISSN 00939994 Kostov, K.G.; Honda, R. Y.; Alves, L.M.S. & Kayama, M.E. (2009). Characteristics of dielectric barrier discharge reactor for material treatment. Brazilian Journal of Physics, Vol. 39, No. 2, pp. 322-325, ISSN 0374-4922, 01039733 54 Manley, T.C. (1943). The Electric Characteristics of the Ozonator Discharge. Journal of The Electrochemical Society, Vol. 84, No. 1, pp. 83-96, ISSN 1945-7111 Marx, E. (1928). Verfahren zur Schlagpruefung von Isolatoren und anderen elektrischen Vorrichtungen. Patentschrift Nr. 455933. 13 Feb. 1928 Mok, Y. S.; Lee, S.-B.; Oh, J.-H.; Ra K.-S. & Sung B.-H. (2008). Abatement of Trichloromethane by Using Nonthermal Plasma Reactors. Plasma Chemistry and Plasma Processing. Vol. 28, No. 6, pp. 663-676, ISSN 02724324 Sasoh, A.; Kikuchi K. & Sakai, T. (2007). Spatio-temporal filament behaviour in a dielectric barrier discharge plasma actuator. Journal of Physics D: Applied Physics, Vol. 40, pp. 4181-4184, ISSN 1361-6463 55 2.7 Hybrid methods for pollution removal Indrek Jõgi1, David Cameron2, Eugen Stamate3 1 University of Tartu, Tartu, Estonia Lappeenranta University of Technology, (ASTRal), Mikkeli, Finland 3 Risø National Laboratory for Sustainable Energy, Technical University of Denmark, Roskilde, Denmark 2 Plasma and catalysis for VOC and NOx removal Non-thermal plasma is suitable for the removal of a number of pollutants but for certain pollutants the removal efficiency is not sufficiently good. The oxidative nature of plasmas in exhaust gases results in further oxidation of the inorganic species NOx and SOx and complementary methods are necessary to remove these species from the exhaust. In the case of organic pollutants, the oxidation by active oxygen species (O, OH) may be too slow and competing reactions consume these radicals before they can be used for the decomposition of the pollutant (Rosocha, 2000). One important reaction which consumes O radicals alongside the oxidation of pollutants is the production of ozone (O + O2 + M → O3 + M). This reaction proceeds relatively fast (characteristic times less than millisecond) and often depletes O radicals before they can react with pollutants. Ozone itself is usually not sufficiently reactive and will remain in the exhaust gas acting as additional pollutant. Consequently, the energy spent for the production of radicals is not completely used for the removal of pollutants and the efficiency of the process decreases. Production of small amounts of NOx in the plasma (Kim, 2011) is also a problem which has to be considered. Complementary techniques like absorption and catalysis have to be employed to solve these problems (Van Durme, 2008). Plasma and catalysis for VOC and NOx removal One way to improve the energy efficiency is to combine the plasma with catalyst (Fig. 2.7.1). The catalysts can be placed inside the plasma zone (so-called single-stage arrangement) or downstream from the plasma (two-stage arrangement). In a single stage system, the conditions (temperature, etc) within the plasma chamber may not be optimum for the catalyst but radiation and short-lived active species from the plasma may increase the activation of the catalysts. In a two-stage system, the conditions for the plasma and the catalyst can be optimised independently but there may be a loss of 56 radicals which could otherwise improve the catalytic performance. The synergy between plasma and catalyst may arise from a number of effects. In both types of reactors, the catalyst adsorbs radicals produced by plasma, preventing the loss of radicals in gas phase reactions (Kim, 2011; Van Durme, 2008). The radical reactions on surfaces often proceed more favourably with respect to final products and backward Figure 2.7.1. Single- and two-stage plasma-catalytic reactors. reactions. Furthermore, pollutants may adsorb on the catalyst which increases their retention time. Many catalysts decompose ozone by detaching oxygen radicals from ozone on the surface of catalyst. This allows a decrease of the concentration of ozone in the exhaust gas and a recovery of oxygen radicals lost to ozone production. In a two-stage reactor, ozone is the main reagent which reaches the catalyst while in a single-stage reactor, short lifetime species (O radicals, metastables, ions etc.) can also take part in reactions on the catalyst surface. In addition to the improved utilization of active species produced in the plasma, the catalyst in a single-stage reactor may also recover some of the energy deposited into the plasma in the form of heat or radiation (Van Durme, 2008). The UV radiation emitted by the plasma (Kim, 2011) is absorbed by photocatalysts and used for production of additional oxidizing species. Part of the thermal energy may be used to activate the catalysts thermally. The catalyst may also affect the plasma properties in single-stage reactors. Catalysts can additionally be used to trap small soot particles which are then oxidized by the plasma in single-stage configurations. In two-stage reactor, it is possible to separate the oxidative reactions which dominate in the plasma from reduction reactions on the catalyst. This is an important feature which has found application in diesel exhaust treatment by selective catalytic reduction (SCR). For some pollutants, reactions with ozone proceed sufficiently fast on catalysts and then the ozone can be produced from clean air separately by plasma and injected into the catalytic reactor together with the polluted stream. Examples of some of these processes are given below and in the next section. Odour removal by non-thermal plasma and catalysts For deodorization applications, non-thermal plasma is often combined with catalytic or absorption methods. In the first stage the polluted gas is stripped of solids, aerosols and particulates by means of a pre-filter (Müller et al., 2006). Appropriate filter media such as bag filters for damp or oily air are used according to the air impurities to be removed. A surface barrier discharge or corona serves as the 57 second stage, where pre-filtered air is subjected to reactive radicals and ions initiating oxidation reactions and the decomposition of VOCs and other contaminants. Finally, compounds not yet oxidised are retained in an activated carbon bed, which is described as a storage reactor that, among other effects, converts residual ozone back to oxygen. The economical and long serviceable life of the activated carbon, as it regenerates itself during the process, is promoted as one of the main special characteristic of this technology. It is successfully used in gastronomy and kitchens (large scale and private households) as well as the food processing industry. For example, the exhaust from 1.5 MW ovens for convenience products made of meat, generating an exhaust stream of 8,000 Nm3/h can be deodorized (Langner, 2009). Plasma processes can also be combined with a biofilter as pre-filter and an oxidation catalyst is employed as a post-filter (see also chapter 3). Biodegradable compounds in higher concentrations are decomposed in the biofilter, while the subsequent plasma unit partially oxidizes non-biodegradable pollutions which are finally decomposed in the oxidation catalyst section. The German company Envisolve describes several commercialised combinations of non-thermal plasmas with catalysts or molecular sieves for waste management facilities, paintshops and other industrial applications generating exhaust streams of up to 300,000 Nm3/h (Rafflenbeul 1998, 2008). In many installations the exhaust gas is not treated directly by the plasma. A stream of plasma treated clean air is injected into the exhaust stream just after the gas discharge stage. This so-called injection method prevents the deposition of aerosols or films (by products of a direct plasma treatment) in the plasma stage, thus increasing the service life time and reducing the operations costs. The concept of the Dutch company Aerox B.V. enable the treatment of exhaust air flows up to 80,000 Nm3/h but does not apply any aftertreatment process (Aerox, 2012). Purification of diesel exhausts from NOx by plasma-enhanced selective catalytic reduction Non-thermal plasma alone is not suitable for the destruction of NOx to N2 and O2 because the presence of oxygen in exhaust gases favours oxidation of NO to NO2, N2O5, and HNOx. However, these oxidized NOx species are more easily removed by complementary processes like wet-treatment, adsorption or SCR (see also section 1.2) compared to NO which is the main NOx component in initial flue gases. Therefore, non-thermal plasma increases the removal efficiency of NOx of other methods and may be utilized as one step in a hybrid process. Additionally, non-thermal plasma allows the removal of particulate matter and VOCs which are also present in diesel exhausts and which could make the SCR process or adsorption inefficient. 58 The SCR allows the decomposition of NOx to N2 and O2 on various catalyst surfaces with the aid of reducing agents (NH3, urea or hydrocarbons). In transportation, the decomposition of NOx is the most viable solution (only gaseous products are formed) and most of the studies concerning NOx removal by SCR have been carried out for diesel car exhausts. Recently, plasma-enhanced SCR has also been tested for marine exhausts (Cha, 2007; McAdams, 2008) and oil-fired Figure 2.7.2. Schematic explanation of SCR process. boilers (Park, 2008). The SCR process does not work properly at temperatures below 200-300°C which is a serious draw-back when the engine is cold. The main reason why the catalysts do not work at low temperatures is the decreased NO reduction capability. The reduction of NO2 proceeds also at lower temperatures and oxidation of NO by plasma allows the extension of the temperature range of the catalytic reduction method towards lower temperatures. Further enhancement of NOx reduction by SCR is achieved by hydrocarbon species. In the plasma stage, the hydrocarbon radicals that are formed improve the oxidation of NO. The chemistry involving hydrocarbon radicals is rather complex (see also appendix A for further details). In principle, hydrocarbons react quickly with O radicals and form partly oxidized hydrocarbons CxHyOz and HO2 radicals. These species efficiently oxidize NO to NO2. The OH radicals formed in the reactions will further oxidize hydrocarbons and thus, one O radical can oxidize many NO molecules while backreactions with NO2 are prevented. Oxidation of SO2 is also prevented in this process and this gives additional benefit because the oxidation products SO3 and H2SO4 degrade the catalyst (Penetrante 1999). In the catalyst stage, the partially oxidized hydrocarbons improve the reduction of NOx to N2 and O2. In the case of the optimized burning process, there are usually not enough hydrocarbons in the engine exhaust for efficient reduction of NOx. Thus it is necessary to inject additional reducing agent into the exhaust gas (Fig. 2.7.2). NH3 or urea are most often used as reducing agents even though they have to be carried in a separate tank while hydrocarbons can be obtained directly from the fuel. The production of HCN can be problematic when hydrocarbons are used while the ammonia slip and catalyst poisoning by NH4NO3 have to be considered when ammonia or urea is used. The available electrical energy which can be used for plasma-assisted SCR is limited. Additional fuel has to be used for the generation of the electricity and this additional fuel consumption should remain below 5 %. This corresponds to the energy consumption of 15-60 J/L depending on the diesel engine. In the limits of these input energy values, about 80 % removal of NOx has been achieved in simulated flue gases with space velocities up to 15000 h-1 (Hammer, 2000; Mizuno, 2007; Lee, 2007). With the use of multiple stages of plasma and catalyst reactors the removal of 90 % of NOx at reasonable input 59 energies has been achieved (Tonkyn 2003). In real diesel exhausts the removal of NOx remains somewhat smaller (50-70 %) due to the negative effect of soot which converts some of the NO2 back to NO (Dorai, 2000; Miessner, 2002; Mizuno, 2007). Hovewer, removal of soot may be regarded as an additional benefit of plasma. A great deal of effort is devoted to the treatment of particulate matter in flue gas from diesel engines. In (Yamamoto et al., 2003) the diesel particulate filter (DPF) regeneration for real diesel engine emissions at low temperatures by means of indirect or direct non-thermal plasma treatment was demonstrated. In other studies (Chae et al., 2001; Mok & Huh, 2005) corona and barrier discharge reactors were successfully used for the removal of smoke and particulate matter from diesel engines. In (Müller & Zahn, 2007) a reactor combining a barrier discharge with a wall flow filter for soot reduction is described. In this system one electrode is porous and gas-permeable. The flue gas is let out through the porous electrode, which filters and accumulates the soot particles. The accumulated soot is decomposed due to a cold oxidation process initiated by active plasma species leading to constant regeneration of the filter at low temperatures during all engine operation conditions. Injection methods combined with wet scrubbing For some pollutants (most notably NOx), ozone is an efficient oxidation agent and could be utilized with high efficiency. Ozone, O3, is produced in plasma from O radicals by the reaction O+O2→O3. The presence of NOx in the flue gas results in O recombination through the chain reactions O + NO→NO2 and O + NO2 → NO+O2. Thus the production of ozone is not efficient in NOx containing exhaust gas and energy used for production of O radicals becomes wasted. Figure 2.7.3. Schematic pictures of low temperature oxidation (LTO) by ozone injection and wet pulsed corona discharge. Higher NO oxidation efficiency can be achieved by producing ozone separately from clean air or pure oxygen and injecting the resulting ozonised gas into the exhaust. The low temperature oxidation (LTO) 60 of NOx by ozone injection is the most well known example of this type of injection method where NO is oxidized by ozone to NO2 and further to NO3. The NO2 and NO3 quickly form N2O5 which is easily soluble in water. The NOx removal efficiency has found to be maximum at 100 °C and the addition of small water droplets improves the NOx removal rate by converting N2O5 to HNO3 (Stamate et al., 2010). Complementary techniques such as wet scrubbing, electrostatic precipitation, catalysis or adsorbers are subsequently applied to remove the oxidized NOx (Fig. 2.7.3). Untreated NOx contains mostly NO which has very low solubility in water and does not adsorb on the surfaces of catalyst or adsorbers and thus the oxidation step is crucial for the utilization of these complementary techniques. A commercial system applying ozone injection is available under the trademark LoTox (Low Temperature Oxidation). The process works within the Electro-dynamic Venturi (EDV) wet scrubbing system in order to achieve a combined reduction of PM, SOx and NOx of stationary emission sources, especially refinery applications (Confuorto & Sexton, 2005). Ozone is generated on site and on demand and injected after the dry ESP directly into the wet scrubber. N2O5 is converted to HNO3 and finally neutralized by the scrubber alkali reagent to NaNO3. Other pollutants such as SO2 and HCl are removed in the wet scrubbing process simultaneously. There exist a number of commercial installations in the USA and in Asia on different emission sources. NOx removal higher than 90 % has been reported. The removal of mercury was demonstrated, too. Several refinery installations have demonstrated LoTox performance and reliability on an applicable scale, the process is available from DuPont BELCO Clean Air Technologies. There are several advantages combining plasma treatment of gases with scrubbing processes. Gutsol et al. reported a wet or spray pulsed corona discharges study for the removal of VOC from paper mill exhaust gases (Fridman, 2008). In spray corona water is injected to the corona discharge from a shower, while in wet corona a thin water film flows on the outer wall electrode (Fig. 2.7.3). In such an arrangement soluble VOCs adsorb on the water droplets or film while non-soluble VOCs can be converted to soluble compounds (e.g. peroxides and peroxide radicals) by means of plasma treatment and subsequently scrubbed within the same arrangement. This results in much lower energy requirements. Furthermore, plasma-stimulated oxidation continues after adsorption resulting in a larger adsorbing capacity of the water and thus decreases water consumption. However such a process is only applicable where large amounts of polluted water are already generated and requires effective water cleaning. The ECO (Electro-Catalytic Oxidation) process is another example for a commercialized plasmaassisted depollution process combined with scrubbing (Boyle, 2005). The process is designed for installation downstream of a dry ESP or fabric filter (ash removal). The flue gas is directly exposed to a barrier discharge and oxidizes pollutants to soluble or capturable compounds (e.g. NO to NO2; SO2 to SO3; Hg to mercury oxide HgO) and forms particulate matter and aerosol mist. SO2, NO2 and HgO are removed in a subsequent absorber vessel (two-loop scrubber). Ammonia is added to the scrubber to 61 maintain the pH of the solution for keeping a high SO2 scrubbing rate. NO2 formed in the ECO reactor is scrubbed by sulphite ions, which are formed by SO2. Finally (NH4)2SO4 and NH4NO3 are formed as well. Several preliminary designs for coal-fired electric utility applications ranging from 175 – 1000 MW has been developed and long time performance and reliability tests were successfully completed. The process is available from the American company Powerspan Corporation and was recently combined with post-combustion CO2 capture technology. References/further reading “The Aerox-Injector Odor Control” Brochure of Aerox B.V. available at http://www.aeroxinjector.com/fileadmin/files/aerox/PDF_s/Brochure_EN.pdf Boyle, P. D. (2005). Multi-Pollutant Control Technology for Coal-Fired Power Plants, Proceeding of the Clean Coal and Power Conference, Washington, DC, November 21-22, 2005; available at http://www.powerspan.com/technology/eco_overview.shtml Cha, M.S.; Song, Y.-H.; Lee, J.-O. & Kim, S.J. (2007). NOx and Soot Reduction Using Dielectric Barrier Discharge and NH3 Selective Catalytic Reduction in Diesel Exhaust. International Journal on Plasma Environmental Science and Technology, Vol. 1, pp. 28-33, ISSN 1881-8692 Chae, J.O.; Hwang, J.W.; Jung, J.Y.; Han, J.H.; Hwang, H.J.; Kim, S. & Demidiouk, V.I. (2001). Reduction of the particulate and nitric oxide from the diesel engine using a plasma chemical hybrid system. Physics of Plasmas, Vol. 8, pp. 1403-1411, ISSN 1403-1410, ISSN 1070-664X Confuorto, N. & Sexton, J. (2007). Wet Scrubbing Based NOx Control Using LoTOx™Technology - First Commercial FCC Start-up Experience. Proceedings of NPRA 2007 Environmental Conference, Austin/Texas, September 24-25, 2007 Dorai, R.; Hassouni, K. & Kushner, M.J. (2000). Interaction between soot particles and NOx during dielectric barrier discharge plasma remediation of simulated diesel exhaust. Journal Applied Physics, Vol. 88, pp. 60606071, ISSN 0021-8979 Durme, J.; Dewulf, J.; Leys, C. & Langenhove, H. (2008). Combining non-thermal plasma with heterogeneous catalysis in waste gas treatment: A review, Applied Catalysis B: Environmental, Vol. 78, No. 3-4, pp. 324-333, ISSN 0926-3373 Fridman, A. (2008). Plasma Chemistry, Cambridge University Press, Cambridge, USA,ISBN-13978-0-521-84735-3 Hammer, T.; Kishimoto, T.; Miessner, H. & Rudolph, R. (1999). Plasma Enhanced Selective Catalytic Reduction: Kinetics of Nox-Removal and Byproduct Formation. SAE Technical Paper 1999-01-3632, ISSN 0148-7191 Kim, H.-H. & Ogata, A. (2011). Nonthermal plasma activates catalyst: from current understanding and future prospects. The European Physical Journal Applied Physics, Vol. 55, No. 1, pp. 13806 (11), ISSN 1286-0042 Kim. H.-H. (2004). Nonthermal Plasma Processing for Air-Pollution Control: A Historical Review, Current Issues, and Future Prospects. Plasma Processes and Polymers, Vol. 1, No. 2, pp. 91-110, ISSN 1612-8850 Langner, M.H. (n.d.). Image brochure of airtec competence GmbH, Recke, Germany; available at http://www.plasmanorm.de, conducted March 2009 Lee, J.O.; Song, Y.-H.; Cha, M.S. & Kim, S.J. (2007). Effects of Hydrocarbons and Water Vapor on NOx Using V2O5WO3/TiO2 Catalyst Re nmduction in Combination with Nonthermal Plasma, Industrial & Engineering Chemistry 62 Research, Vol. 46, No. 17, pp. 5570-5575, ISSN 0888-5885 McAdams, R.; Beech, P. & Shawcross, J.T. (2008). Low temperature plasma assisted catalyticreduction of NOx in simulated marine diesel exhaust. Plasma Chemistry and Plasma Processing, Vol. 28, pp. 159-171, ISSN 0272-4324 Miessner, H.; Francke, K.-P. & Rudolph, R. (2002) Plasma-enhanced HC-SCR of NOx in the presence of excess oxygen. Applied Catalysis B: Environmental, Vol. 36, pp. 53-62, ISSN 0926-3373 Mizuno, A. (2007). Industrial applications of atmospheric non-thermal plasma in environmental remediation. Plasma Physics and Controlled Fusion, Vol. 49, pp. A1-A15, ISSN 0741-3335 Mok, Y.S. & Huh, Y.J. (2005). Simultaneous Removal of Nitrogen Oxides and Particulate Matters from Diesel Engine Exhaust Using Dielectric Barrier Discharge and Catalysis Hybrid System. Plasma Chemistry and Plasma Processing, Vol. 25, pp. 625-639, ISSN 0272-4324 Müller, S.; Zahn, R.-J.; Grundmann, J. & Langner, M. (2006). Plasma Treatment of Aerosols and Odours. Proceedings 3rd International Workshop on Microplasmas IWM3, Greifswald, Germany, May 9-11, 2006, p. 121 Müller, S. & Zahn, R.-J. (2007). Air Pollution Control by Non-Thermal Plasma. Contributions to Plasma Physics, Vol. 47, No. 7, pp. 520-529, ISSN 0863-1042 Park, B. R. & Deshwal, S.H. (2008) NOx removal from the flue gas of oil-fired boiler using a multistage plasmacatalyst hybrid system, Fuel Processing and Technology, Vol. 89, pp. 540-548, ISSN 0016-2361,0378-3820 Penetrante, B.M.; Brusasco, R.M.; Merritt, B.T.& Vogtlin, G.E. (1999). Environmental applications of lowtemperature plasmas, Pure and Applied Chemistry, Vol. 71, No. 10, pp. 1829-1835, ISSN 0033-4545 Rafflenbeul, R. (1998). Nicht-thermische Plasmaanlagen (NTP) zur Luftreinhaltung in der Abfallwirtschaft. Müll und Abfall 1/1998, pp. 38-44 (in german), ISSN 0027-2957 Rafflenbeul, R. (2008). Geringe Kosten für gering konzentrierte Abluft. Wasser, Luft und Boden (wlb) 6/2008, pp. 36-41, (in german); available at http://www.envisolve.com, ISSN 1421-8615 Rosocha L.A. & Korzekwa R.A. (2000). Removal of volatile organic compounds (VOC's) by atmospheric-pressure dielectric-barrier and pulsed-corona electricxal discharges, in Electrical Discharges for Environmental Purposes – Fundamentals and Applications, Veldhuizen, E.M; Ed., Huntington, NY: Nova Science, pp. 245-278 Stamate, E.; Chen, W.; Jørgensen, L.; Jensen, T.K., Fateev, A. & Michelsen, P.K. (2010). IR and UV gas adsorption measurements during NOx reduction on an industrial gas fired power plant. Fuel, Vol. 89, No. 5, pp. 978-985, ISSN 0016-2361 Tonkyn, R.G.; Barlow, S.E. & Hoard, J.W. (2003). Reduction of NOx in synthetic diesel exhaust via two-step plasma-catalysis treatment. Applied Catalysis B: Environmental, Vol. 40, No. 3, pp. 207-217, ISSN 0926-3373 Yamamoto, T.; Okubo, M.; Kuroki, T. & Miyairi, Y. (2003). Nonthermal plasma regeneration of diesel particulate filter. SAE Technical Paper 2003-01-1182, SAE International, ISSN 0148-7191 63 2.8 Eco-efficiency and cost-benefit analysis of plasma technologies Dainius Martuzevicius1, Inga Stasiulaitiene1, Andra Blumberga2, Dagnija Blumberga2 1 Kaunas University of Technology, Kaunas, Lithuania Riga Technical University, Riga, Latvia 2 Introduction The data presented in this section prove that plasma-based technologies have a strong potential for the implementation in the market as technically sound and efficient technologies. At the same time, other aspects of the implementation of such technologies must be considered, such as overall environmental performance and cost efficiency. The following question should be asked: • Are plasma-based technologies environment friendly? • How much environmental burden do they create throughout their entire life cycle? • Does the high-voltage-based plasma generation cause a higher impact on the environment compared with other technologies? • Is the implementation of these technologies economically feasible? These questions may be answered by comparing plasma-based technologies to pollutant treatment technologies, based on physico-chemical processes. In this section the term "sustainable development" is defined as the right to meet the development aspirations of the present generation without compromising the rights of future generations to meet their developmental needs. Viewing from the perspective of the concepts of sustainable development, plasma-based technologies are usually classified as end-of-pipe technologies. In the environmental application field, they do not suggest how to minimize the formation of pollutants, but rather treat the already formed ones. On the other hand, in case of processes where the minimization of pollution is too difficult to achieve in current economics (such as energy production by combustion, generation of odours in farming, etc.), plasma technologies may suggest the treatment of pollutants which is more effective, consumes less input materials and generates less waste than conventional end-of-pipe technologies. The environmental performance of the pollutant treatment technologies may be assessed by 64 numerous techniques, such as Environmental Impact Assessment, Risk Assessment, Material Flow Analysis, Life Cycle Assessment, Eco-Efficiency Assessment, and others. Life Cycle Assessment (LCA) analyses the impact to the environment during the entire life cycle of a product/service/technological process and quantifies the resulting potential impact on the environment, while eco-efficiency assessment adds customer value approach to the LCA results. The efficiency of a particular investment or technology in economic terms from the perspective of sustainable development is determined by cost-benefit analysis. The starting point for the analysis of economic efficiency is the financial analysis. The financial performance of the investment or technology is determined only on the basis of cash flows occurring in the entity performing the investment project. The analysis of economic efficiency takes additionally into account the benefits and costs from the perspective of a wider area - the region, country or the European Union. During cost-benefit analysis the costs and benefits of economic, social and environmental impacts are analyzed, quantified (attempt to present numeric values) and monetized (express the previously quantified factors in monetary values). The location of investment has a strong impact on the social costs and benefits as opposed to financial analysis making the cost efficiency also dependent on the location of investment. The study described in this section presents the results from the life-cycle analysis (LCA) based quantitative assessment of environmental impact and cost-benefit analysis of several plasma-based and conventional end-of-pipe technologies. The full “Report on Eco-Efficiency of Plasma-Based technologies for Environmental Protection” and the “Report on cost-benefit analysis of plasma-based technologies“ are available at the PlasTEP project website (http://www.plastep.eu/fileadmin/dateien/Outputs/OP3-2.1_Eco-efficiency_report.pdf, http://www.plastep.eu/fileadmin/datein/Outputs/120208_CBA.pdf). The processes which have been researched include the treatment of fuel (coal) combustion exhaust gases, mostly targeting the reduction of NOX and SOX concentrations, and ventilation exhaust gases, targeting the abatement of volatile organic compounds (VOCs). The selected technologies for the NOX and SOX reduction included comparison of Electron Beam Flue Gas Treatment (EBFGT) versus Wet Flue Gas Desulphurization with Selective Catalytic Reduction (WFGD+SCR). In case of VOC reduction, technologies based on dielectric barrier discharge (DBD) versus biofiltration, adsorption by zeolite (for LCA) and molecular sieves (for CBA) have been investigated. EBFGT technology employs electron beam injection as a special type of plasma generation combined with dry-scrubbing process for simultaneous SO2 and NOX removal. The irradiation of stack gases with the electron beam induces chemical reactions making the removal of SO2 and NOX easier. Ammonia NH3 is injected at the next stage and (NH4)2SO4, (NH4)NO3 are collected as reaction products which can be sold as fertilizers (see section 2.5 and appendix A for further details). WFGD-SCR technology is a 65 simultaneous NOX/SOX removal process, where NH3 is injected into the flue gas upstream of a catalyst, with the aim to reduce NO to N2 (see section 1.2). In a subsequent wet scrubbing process SO2 is converted to CaSO4 using limestone or lime as a reagent. The resulting waste is usually stored in landfills. As a DBD-based technology the so-called plasmaNorm® was chosen (Langner, 2009). It was developed for the complete removal of cooking odours. DBD-based gas cleaning technology treats intake, ambient or exhaust air for the environmentally safe removal of the smallest gaseous, organic carbon compounds such as odour molecules, viruses, bacteria, spores, etc. The DBD plasma stage is usually combined with pre-treatment for particle removal and post-treatment with adsorbing material (e.g., active carbon) or catalysts. An alternative conventional technology of Biofiltration uses biological organisms to oxidize the VOCs, consuming them as a food source. The adsorption-based process consists of a rotating bed of adsorbers (activated carbon, zeolite, polymer or a combination) with which the odour or solvent is first adsorbed and then desorbed, after which the cycle starts again. Life-cycle assessment The inventory data for the flue gas treatment technologies were collected using technological questionnaires or based on values published in literature. The functional unit (a value that all the impacts were compared to) was set as 1,000 Nm3 of treated flue gases, that is, all inputs and outputs of the processes were normalized to the above quantity of exhaust gas. System boundaries were set for the processes of the polluted media treatment technologies. The materials needed for transportation of supply materials and by-product elimination was not included due to the extensive amount of data. Moreover, neither were the materials used for manufacturing the flue gas treatment devices and supply infrastructure evaluated. Five main categories (global warming, ozone layer depletion, acidification, eutrophication and human toxicity, based on the CML2001 method) were used for environmental impact evaluation. The weighting procedure (CML 2001 experts IKP, Central Europe) was applied to evaluate the total impact. The environmental analysis based on the LCA technique revealed that within the chosen system boundaries the plasma-based EBFGT performed better from the environmental impact viewpoint with respect to acidification and human toxicity potential, but the WFGD+SCR technology was more favourable from the global warming potential, ozone layer depletion and eutrophication potential. Most of the impacts were associated with the utilization of ammonia in both processes. In addition, formation and deposition of gypsum sludge scored low in case of WFGD+SCR technology. 66 1200 By-products Material resources By-products Material resources Electricity Electricity 41.1 40 CML 2001, Experts IKP (Central Europe) CML 2001, Experts IKP (Central Europe) 1000 50 1099.7 800 600 400 30 26.1 20 10 200 4.2 74.4 0 0 EBFGT WFGD + SCR DBD plasma Adsorption (zeolite rotor) Biofilttration Figure 2.8.1. Total weighed environmental impact of plasma and non-plasma end-of-pipe pollutant treatment technologies. Left: the comparison of technologies for SOx/NOx removal. Right: the comparison of technologies for VOCs removal. Among the studied technologies for the removal of VOC from the ventilation exhaust gases, plasmabased (DBD) technology performed better from the environmental impact viewpoint with respect to the ozone layer depletion, acidification, eutrophication and human toxicity potential. The usage of electricity and pre-filtration stage caused plasma technology to be less favourable compared to biofilters with respect to the global warming potential. This may be improved by selecting more environment-friendly pre-filtration material. Figure 2.8.1 shows the comparison of the plasma and non-plasma end of pipe treatment technologies, with respect to their overall environmental performance score. Plasma-based technologies were revealed as equally competitive ones compared to other end-of-pipe methods. The relatively high demand of electrical energy causes lower positioning of plasma technologies in cases where no other materials are utilized and no major waste is formed. On the other hand, many traditional end-of pipe technologies are associated with high amounts of process waste, which provides plasma technologies with an opportunity to establish themselves in the market as more efficient, and in many occasions more environment-friendly ones. 67 Cost-benefit analysis The financial analysis was made for the reference period of 15 years, discount rate of 5.5 % and for investor who can deduct VAT on capital expenditure and operation costs. All cash flow streams for all periods covered by the financial analysis are negative and the calculation of the indicators IRR and benefit/cost is impossible. Due to the fact that the investor is liable for VAT, capital expenditures and operating costs are net prices and there is no need to adjust capital expenditures and operating costs for VAT. For this reason, adjustments were made only for: income tax from individuals and legal entities; social security contributions. The Polish fiscal and social security system was adopted to determine the fiscal effects. The share of wage costs in capital investment and in operating costs was 10 %. The main benefits after implementing the plasma technologies are: avoided payments for air pollution, avoided external costs and health benefits. Costs are: taxes for air pollution, capital, administrational and operational costs. Cost calculation of EBFGT was carried out for a selected oil refinery using flue gas flows between 65,000 and 130,000 Nm3/hour and availability of 8,320 hour/year. The total investment costs for EBFGT installation are between 9,580,000 EUR and 13,328,000 EUR while the annual operating costs are 387,000 – 473,000 EUR/year. The expenses for EBFGT are partly compensated by the sale of byproducts (covers the expenses of raw materials) and the income will reach 358,200 or 717,000 EUR/year assuming ammonium sulfate cost of 94 EUR/ton. The investment and operation costs for a 300 MW unit are shown in table 2.8.1 together with the same data for Wet SO2 scrubbing combined with either selective catalytic reduction or selective non-catalytic reduction methods (see also 1.2 and 2.8). In addition, the cost comparison of EBFGT and WFGD-SCR methods for 120 MW coal fired boilers based on an earlier study is given in table 2.8.2. Table 2.8.1. Investment and operation costs of EBFGT and combination of conventional deSO2 and deNOx methods (Calinescu, 2008; Basfar, 2009). Flue gas treatment method >300 MW unit Investment costs €/kW Annual operation costs €/MW 32-45 1290-1577 Wet deSO2 + SCR 176-247 4786-5350 Wet deSO2 + SNCR 144-190 3870-4223 Investment costs €/kW Annual operation costs €/MW EBFGT 113 5167 WFGT+SCR 162 5343 EBFGT Emission control method 120 MW unit 68 The financial analysis was performed for the sample system CHP “Pomorzany” with power of 120 MW. All cash flow streams for all periods were negative and the financial net present value was calculated to be -20,993 EUR. External benefits of these technologies are due to the decrease of costs of environmental damage caused by energy production from fossil sources as a result of the implementation of the investment. The biggest part of the economic costs is the amount of health care cost which depends on the specific country. External costs of energy production in Poland was estimated based on Radowice’s 2002 report Assessment of external costs of power generation in Poland. It was estimated that health costs and the greenhouse effect outweigh the other effects, contributing about 98 %, in which health costs associated with the emission of SO2, NOx and volatile substances are 15 %. Total average external cost calculated on a kWh of electricity in Poland in 2004 was 0.0466 €/kWh. For the calculation of economic benefits for this project 55 % of this value was adopted. In order to bring the level of external costs of electricity generation from 2004 to the base year (2010), GDP growth data were used. Economic benefits include: • • • Wages of employees directly involved in the implementation and operation of investment and employees of companies - contractors of the project. The value of this benefit was estimated based on the share of wage costs (in net terms) in the total cost. In total, this benefit was estimated at 841,950 EUR (investment phase) and at 35,340 EUR per year (operational phase). Income of companies involved in the implementation and operation of the investment and contractors and subcontractors who implement the project. In total, this benefit was estimated at 1,196,370 EUR (investment phase) and at 50,220 EUR per year (operational phase). The multiplier effect of investment spending in the vicinity of the project. It is assumed that the multiplier effect due to employees and contractors spending part of the income in the region will bring an additional benefit of 5 % of the net cost of investment - a total of 696,880 EUR. With the accepted discount rate, the economic indicator NPV value (ENPV) for the project is 47,140,500 EUR and the economic internal rate of return was 51.8 %. Cost/benefit ratio was 3.4. All these indicators are closely linked to one another and the values obtained indicate that with the accepted assumptions the project is economically efficient. For odour reduction, the analysis was made to assert economic efficiency of molecular sieve duplex technology compared with non thermal plasma technique in flavour processing industry with exhaust flow of 50,000 m3N/hour and for [Corg] < 100 mg/m3. 69 Regarding the financial performance of the two analyzed technologies, molecular sieve duplex technology performs better at the financial level. The value of FNPV for non thermal plasma technique is 1,070.91 thousand Euros and for molecular sieve duplex it is 972.78 thousand Euros. The external benefits of both technologies were not analyzed because the degree of odour reduction is very similar and their values do not affect the comparative analysis. However, the negative external effect, which is carbon dioxide emission, was taken into account and analyzed. The value of the ENPV index (excluding the external benefits) for the non thermal plasma technique is 1,073.3 thousand Euros and for the molecular sieve duplex it is 875.4 thousand Euros. Due to the exclusion of external benefits, the each year’s streams of economic flows in the analysis are negative and the calculation of EIRR and benefit/cost ratio is impossible. The cost-benefit analysis shows that the use of EBFGT technology is efficient in economic terms, based on sustainable development approach as well as the sum of all the costs and benefits for all groups which the project may affect. Taking into account the amount of capital expenditures and operating costs for comparable methods, it can be explicitly stated that the EBFGT is advantageous compared with alternative technologies. The non thermal plasma technique for VOC removal is somewhat more expensive compared to molecular sieve duplex (adsorption) technology but surpasses all other analyzed technologies. It should be noted that the capital expenditure in the molecular sieve duplex is doubled compared to the non thermal plasma technique. Therefore, an investor with limited financial capabilities will be more willing to use technology based on non thermal plasma, in spite of less favourable rate of NPV. References/further reading Basfar A. et al. (2009). Electron Beam Flue Gas Treatment (EBFGT) Technology for Simultaneous Removal of SO2 and NOx from Combustion of Liquid Fuels: Technical and Economic Evaluation. International Topical Meeting on Nuclear Research Applications and Utilization of Accelerators Calinescu I. et al (2008). Electron beam technologies for reducing SO2 and NOx emissions from thermal power plants. WEC regional energy forum – Foren 2008 Guinée, J.B.; Gorrée, M.; Heijungs, R.; Huppes, G.; Kleijn, R.; Koning, A. de; Oers, L. van; Wegener Sleeswijk, A.; Suh, S.; Udo de Haes, H.A.; Bruijn, H. de; Duin, R. van; Huijbregts, M.A.J. (2002) Handbook on life cycle assessment. Operational guide to the ISO standards. I: LCA in perspective. IIa: Guide. IIb: Operational annex. III: Scientific background. Kluwer Academic Publishers, Dordrecht, 692 pp. ISBN 1-4020-0228-9 Langner, M.H. (n.d.). Image brochure of airtec competence GmbH, Recke, Germany; available at http://www.plasmanorm.de, downloaded May 2009 70 3. Practical Examples of Plasma Technology Preface There are several companies and suppliers for plasma-based or plasma-enhanced exhaust depollution technologies available in Europe and in Balti Sea Region. Their activities cover the cleaning and deodorization of exhaust gase in many fields, e.g. food production, tobacco drying, flavour production, oil seed proceesing, slaughterhouses, rubber production and pharmaceutics. To give an impression of the current state-of-the-art of industrial application we offered to companies active in this field the ability to present themselve with a short profile in this publication. The respective companies themselve are responsible for the presented material and content on the following pages. We hope that these company profiles give a good overview on the practicability and chances of environmental plasma technologies and related fields. We are gratefull to the companies and colleagues who responded on our request and agreed to contribute to this publication. 71 3.1 PlasmaAir AG Dr. Ing. Bernd Glocker Introduction Green house effect and global warming are the critical issues in this new aera. The influence is not only the environment, but also the economic and national safety. The majority gas, PFCs and SF6, the GHG(Green House Gas) defined in Kyoto Protocol are generated gas out from the semiconductor and optical electronics dry etching procedure accumulated and mixed with CVD gas for chamber clean. The special PFCs gases from the tools such as SF6, CF4, C2F6, C3F8, CHF3 and NF3 have to be treated before emission. A plasma scrubber system was developed and tested as Point of Use abatement system for the treatment of PFC loaded gases coming from a production line from semiconductor industry. The system consists of an arc heated water steam plasma unit followed a combustion chamber and a quench-scrubber-system. For chemical destrucion of PFC material, a water plasma is favorable because both, the oxygen and the hydrogen, which is necessary to produce the desired products after the high temperature zone-CO, CO2, HF and HCl- are supplied by the heat carrier itself. Therefore a new plasma torch system using water as plasma gas was developed and tested under the specification to achieve a long lifetime of the electrodes. The system was developed and tested under industrial condition at a process line in semiconductor industry. It was demonstrated, that, with this process a reduction of PFC in the exhaust gas of > 99.9 % can be achieved. Description of the technology The Plasma Process For destruction of fluorinated compounds such as PFCs/HFCs (SF6, CF4, C2F6, C3F8, CHF3 and NF3, etc) a special plasma source is needed. Plasma generators for noble gases and gases like nitrogen are well known and often used because of little erosion and a long lifetime for the electrodes. The cracking process is a high-temperature pyrolysis process. The strong chemical bond between C and F or Cl (halogen) will be broken, reacted with oxygen and hydrogen and produces hydrogen fluoride and hydrogen chloride as stable compounds. Especially for the CF4 which has the most stable and strong C-F bond, it needs more than 1600 °C for pyrolysis process. In the mean time, the oxygen and 72 hydrogen have to be provided to react with F. Otherwise it return to stable CF4 again. Therefore, we need to supply O and H free radical at the same time while cracking PFCs/HFCs to increase the destruction efficiency. It is not helpful to use standard plasma technologies for the destruction of PFCs. The problem was solved by developing a special high-temperature pyrolysis process using steam plasma in which the PFCs/HFC molecules are cracked by heat and O and H radicals into smaller units, in order to obtain hydrofluoric acid and hydrochloric acid for recovery as valuable chemicals. Therefore the water plasma is generated in an arc heated plasma torch. The PFC containing gas is injected into the reactive plasma in a mixing chamber. To ensure a full reaction, the mixture is fed through a high temperatur reaction chamber. The Plasma Torch Conventionaly used arc heated plasma torches work with argon as a plasma gas. For some applications Nitrogen is used. For the here described PFC reduction process an oyidiziend plasma gas is required to assure the chemical reaction process. The main problem using oxidizing gases is the short life time of the electrodes. This limts the operation time of the system. Building a exhaust gas cleaning system, a continous operating system that operates several 100 hours is required. Therefore, a new plasma torch was developed. The focus on the development was to achieve a lifetime of min. 500 hours using water steam as plasma gas. The power level was fixed between 5 and 20 kW. This is the level that is required to treat typical exhaust gas streams containing PFC coming from production processes or from processes in semiconductor industry. Parallel to the torch the complete equipment necessary to operate the torch was developed. A special evaporator, cooling cycle and the power supply were developed and optimized for long continuous operation. Figure 3.1.1. Steam plasma torch in operation. The same torch can be operated without changes with several other plasma gases. It was continously tested with water steam, air, pure oxygen and nitrogen. Instead of the argon, hydrogen can be used to protect the cathode from erosion. 73 The Plasma Scrubber System Plasma cracking is a non-incineration thermal process using temperatures more than 2000°C by cracking high stability materials like fluorinated compounds into smaller molecules. The process gas is mixed into a steam plasma jet. Through the high temperatures and the available radical the PFC is cracked and reacts with the oxygen and the hydrogen coming from the water plasma to CO2 and HF. The exhaust gases from the high temperature reactors are cooled and cleaned in a specially developed quench and water scrubber stage. In the first stage, the gas stream is quenched down by a injected water stream. In parallel to the quench effect one gets a first cleaning of the gas stream from the acid components. Temperature of the stream can be cooled to 40 °C after quenching. After the first stage, additional cleaning stages are followed by using recycling water for gases scrubbering. For achieving the required clean gas concentration, the last fine cleaning stage is done with clean city water. Figure 3.1.2. Scheme of the Plasma scrubber. Examples of the test results With a prototype the PFC decomposition was investigated under industrial condition. It was positioned in the exhaust gas line of a production process in semiconductor industry. For the first qualification test a typical production in semiconductor industry was investigated. With a FT-IR system the concentration of the green house gases were tested and recorded. The results are shown in table 3.1.1. It can be seen 74 that the Destruction Removal Efficiency (DRE) was very high compared to conventional used systems even for the thermal and chemical very restistant substances. Site Operation Test #1: - Process: Etch (Equip: Lam Research) - POU Abatement System: ZRI-XP-20 - Site Operation 7 Chambers Etch treated in 2 set of XP-20 Operation: 1-Operation, 1-Standby - Operation Condition: Total Gas Flow into Scrubber: 288.5 SLM Plasma Power Level: 16 KW Steam flow: 1m3/hr (over heated) Argon flow: 0.7 L/min Table 3.1.1. Destruction Removal Efficiency Test #1. Conc. (PPMv) DRE% Gas Recipe Inlet Outlet SF6 2488 1674 3.28 99.74% C2F6 2494 1294 1.81 99.80% CF4 416 4.89 98.35% HF 210 0.16 99.90% HCl 324 0.01 99.99% * CF4 is C2F6 byproduct Summary A steam plasma system for as exhaust gas POU abatements system was develped and tested under industrial condition. A very high destruction efficiency of the system was demonstrated under typical operation conditions. The continous operation time of the system before some mantanance was deemed was extended up to more than 800 hours. These systems are sucessfully operated in semiconductor industry for more than 3 years until now. It will help to reduce the green house gases coming out from semiconductor industry. The system can be used in chemical industry either to clean gasstreams containing PFC gases up to high concentration. 75 Contact data Contact person: Dr. Ing Bernd Glocker Address: Am Lindenberg 8 71263 Weil der Stadt Phone: +49 7033 3098840 [email protected] E-mail: Notes to copyright All photos in this chapter were provided by the company PlasmaAir AG and are subjected to the copyright by the company PlasmaAir AG. This is regardless of whether the image has an apparent marking, record or a logo visible integrated in the image or an invisible digital watermark. The use of images on other web sites for personal, commercial or promotional purposes requires the written permission and/or fees of the company PlasmaAir AG Unauthorised download for other media, print, digital presentations, image montages, collages and similar will not happen. This refers also to the generally applicable copyright law, including related rights. 76 3.2 Bioclimatic GmbH Dipl.-Ing. Nicole Achilles The Goal A human’s basic needs are 1kg of food, 2kg of fluids and 15kg of air per day (Fig. 3.2.1) Given these numbers, taking care of the air quality is of great importance. As air can be contaminated in various ways – by particles, chemicals and microorganisms – the How to clean this air becomes of particular interest. Here the bioclimatic Bipolar Ionisation offers a technology which is versatile in its application while at the same time being long-lasting, environmental friendly and low in energy consumption as well as easy to maintain. Bipolar Ionisation is capable of abating odorous and Figure 3.2.1. Basic needs of a human. persistent volatile organic compounds (VOCs) as well as killing bacteria and fungi (moulds) by oxidation without being specifically engineered to just one class of VOCs or bacteria. This broad spectrum makes this technology interesting for numerous applications: food industry, animal keeping, restaurants, offices, laboratories, or hospitals, to name just a few. Technology Bipolar Ionisation is a plasma technology employing the principle of dielectric barrier discharge to create “activated oxygen”, a mix of positive and negative ions as well as a few radicals and a bit of ozone derived from the oxygen and water molecules which are naturally present in air, as reactants with which to abate the air contaminants (Fig. 3.2.2). Figure 3.2.2. Dielectric Barrier Discharge. 77 All bioclimatic ionisation units comprise a high voltage transformer and a minimum of one ionisation tube. The use of a high voltage transformer as part of the unit allows employing the ionisation units at the standard current available from any ordinary power socket. The ionisation tube connected with the transformer consists of an inner metal electrode, a glass tube as dielectric barrier and an outer metal electrode (wire mesh cylinder), all recyclable materials. By applying alternating current, transformed to voltages between 1250V and 3000V, to the electrodes electrical discharge is being forced. The energy thereby released is absorbed by the oxygen and water molecules in the surrounding air and their bonds are dissolved, thus creating the reactants which form the activated oxygen. These reactants are then carried with the air by way of natural air circulation or a ducted air stream to mix with and come into contact with the contaminants. In case of VOCs the reactants’ oxidation will abate the contaminants to CO2 and water while with microorganisms the reactants access the water molecules of the microorganisms’ cells, thereby disrupting the cell structure and either kill the microorganisms outright or damage them in a way which prevents their reproduction. Figure 3.2.3. Bioclimatic ionisation units. 78 Because of different situations there are ionisation units available for installation in rooms as well as for installation in ventilation ducts and air handling units (Fig. 3.2.3). A typical schematic set-up of a ventilation system equipped with ionisation units can be seen in figure 3.2.4. Here, duct-mountable aerotron units are integrated in the supply air duct after particle removing pre-filters, thereby treating both return air and incoming outdoor air with ionisation. The created activated oxygen not only abates VOCs incoming with the supply air, but is also carried with the air flow stream into the rooms, thereby treating contaminants inside these rooms which have entered through doors or windows or stem from indoor sources. This is a distinct advantage over stationary systems such as activated carbon. Figure 3.2.4. Schematic set-up of duct mounted ionisation system. As air contamination levels may vary from day to day or even throughout a day, any bioclimatic ionisation system can be operated at different ionisation intensity levels to meet the ionisation needs of the actual moment. Setting of ionisation intensity level can either be done manually or by automatic controllers based on sensors. A sensor-based automatic control set-up employing an air quality sensor and an ozone sensor is depicted in figure 3.2.4. With the ionisation system set to the appropriate intensity level to meet the existing or expected contamination, the Bipolar Ionisation can be operated constantly as indeed it is designed to do. This 79 may raise some concerns regarding power consumption, but considering that when operated at standard 230V the bioclimatic ionisation units have the power consumption ranging from 6W for the smallest units like the AirDeco Pyramid to a maximum of 100W for the largest aerotron units, fully equipped with 14 ionisation tubes and operated at highest intensity level, the power consumption is no more than that of light bulbs. Adding to this the fact that a unit such as the aforementioned aerotron with 14 ionisation tubes can treat more than 8,000m³/h supply air in standard environment while at the same time causing no pressure drop in the ducted ventilation system, equipping a ventilation system with ionisation does hardly add to the energy costs of the system as a whole. Another matter of concern when constantly operating the Bipolar Ionisation units is the generation of ozone. As mentioned above, the activated oxygen contains also some ozone. Ozone, while being a good oxidation agent is also hazardous to the health if encountered in too high concentrations. Already in lower concentration, at which ozone does not pose a health risk, it is irritating to eyes and nose and can cause a slight headache. On the other hand, ozone is constantly present in our environment without causing people to suffer, so, like with many other things, concentration makes the difference. As Bipolar Ionisation with the activated oxygen generated adds to the natural concentration of ozone present in the air, it is important to know how much ozone the ionisation units are generating and even more so to keep it well below any threshold where it could become harmful. Here bioclimatic already 25 years ago had their equipment tested in a field experiment set up in the customers’ hall of a health insurance agency and had it monitored by the TÜV. The result was that even at the highest intensity level, where the customers and employees already detected the tell-tale swimming-pool smell of ozone, the ozone concentration was well below any thresholds above which the ozone would pose a health risk (see also Tab. 3.3.1). Since then the bioclimatic Bipolar Ionisation technology has been refined and sensor based controlling has been added to ensure as safe a product as may be. At the same time bioclimatic Bipolar Ionisation units are easy to maintain. Besides regular cleaning (cleaning intervals depend on the mounting situation and possible pre-filtration of the air and can be as rare as once per year for duct-mounted systems) the only required maintenance work is the exchange of the ionisation tubes after approx. 24,000 operation hours (equalling about 3 years). Table 3.2.1. Test results ozone concentration. Ionisation Intensity medium intensity level Ozone concentration < 0.01ppm highest intensity level 0.02ppm EU threshold for prolonged duration of ozone exposition 0.05ppm 80 Facts and Figures Ozone is not the only substance where tests with the bioclimatic ionisation units were conducted. Over the years multiple tests have been carried out in laboratories all over the world as well as in the technical centre of bioclimatic in Bad Nenndorf, Germany to determine the efficiency of the Bipolar Ionisation technology with regards to bacteria, moulds and different VOCs. Among the microorganisms tested were prominent species such as MRSA. The test results showed that while not an instantaneous solution Bipolar Ionisation with its advantage of constant operation is a lasting solution to this medical threat (see also Fig. 3.2.5). Figure 3.2.5. MRSA concentration after 48 hours of incubation – with and without ionisation. While abatement of contaminants often is of course the main goal for employing the Bipolar Ionisation technology, there are also economical advantages where the employment of Bipolar Ionisation allows the optimisation of the ventilation system. An example for this is shown in table 3.2.2. Here the key figures and costs of operation of a cheese maturing chamber are compared. In this application the use of Bipolar Ionisation allowed to change the ventilation system from strictly fresh air – exhaust air to a system which also employed return air, which positively affected the necessary air flow volume but also the energy necessary for cooling the supply air. Furthermore, as the bipolar ionisation abated the threat of mould infestation, temperature could be slightly increased while at the same time the amount of detergents could be decreased. The overall change resulted in annual 81 savings of approx. 75,000 € while at the same time the one-time investment costs amounted to approx. 60,000 € which included the remodelling of the ventilation system to employ return air, thereby allowing for an amortisation within the first year. Table 3.2.2. Cheese maturing chamber, savings by using bipolar ionisation. air flow volume cooling temperature energy consumption (cooling) savings energy savings detergents total savings without ionisation 75,000 m³/h 12 °C 89 kW with ionisation 56,500 m³/h 15 °C 79 kW ≈ 19,000 €/a ≈ 56,000 €/a ≈ 75,000 €/a Summary Bipolar Ionisation, a plasma technology employing the principle of dielectric barrier discharge, is a versatile technology for the abatement of odorous and persistent VOCs as well as microorganisms such as bacteria and mould in the air. The bioclimatic ionisation units are low in energy consumption, easy to maintain, long-lasting and environmental friendly by using recyclable materials. Furthermore this technology is neither limited to stationary effect nor engineered to just one class of VOCs or microorganisms, thereby making it attractive for a multitude of applications. Contact data Contact person: Nicole Achilles Address: Bioclimatic GmbH Im Niedernfeld 4 31542 Bad Nenndorf Germany Phone: +49 5723 9440 0 E-mail: [email protected] Notes to copyright All photos in this chapter were provided by the company Bioclimatic GmbH and are subjected to the copyright by the company Bioclimatic GmbH. This is regardless of whether the image has an apparent marking, record or a logo visible integrated in the image or an invisible digital watermark. The use of images on other web sites for personal, commercial or promotional purposes requires the written permission and/or fees of the company Bioclimatic GmbH. Unauthorised download for other media, print, digital presentations, image montages, collages and similar will not happen. This refers also to the generally applicable copyright law, including related rights. 82 3.3 MCT GmbH Armin Eschenhof Introduction The Nice-R™-Air Cleaner was developed to neutralize odors in the air; e.g. cooking odors, odors out of the fermentation of organic substances (household garbage), lingering pet odors, smells which arise from bonding floors and much more. Moreover, bacteria and germs can be eliminated as well as the effectiveness of pollen as allergens will be prevented by using the Nice-R™-Air-Cleaner. The units were developed for private individuals, nursing homes, kindergartens, schools, trade, offices, conference rooms etc.. They are equipped with ventilators of 115 m³/h and 320 m³/h. The connected load lies between ca. 7 W and ca. 55 W and the units are low-noise. The product line is by now certified with the VDE/GS and EMC sign by the VDE institution. Description of the technology The technology is based on the principle of low-temperature plasma, generated by voltages between 3.5-5 kV. The unit consists of a pre-filter, ventilator, plasma electrodes and an active carbon block. 83 The air which should be cleaned is pulled through the pre-filter (sediment filter) by the ventilator and flows around the electrodes. A short circuit protected transformer powers the electrodes with a voltage of 3 kV. Depending on the quantity and capacity of the electrodes, the supply voltage at the electrodes increases to 3.5-5 KV and plasma is generated here. Pollutants will be reduced while the polluted air flows around the electrodes (plasma purification step). It may occur that a part of the pollutants circulating in the air will only be cracked, i.e., oxidatively degraded to fragments of lower molecular mass. The air including these fragments flows subsequently through a downstream active carbon block which enriches the cracked pollutants such that they react with the components in ionized gas. 84 Examples of the test results Test scores: a) Pyridine ( C5 H 5 N ) From a chemical point of view, Pyridine is an aromatic hydrocarbon with a very low odor threshold of 10 ppm. The odor threshold of 10 ppm is equivalent to a concentration of 35,3 mg/m³. In general, a smell is said to be intolerable from a concentration of ca. 30 ppm onwards. During the test, 10 g pyridine was added to a 90 m³ sealed room filled with air such that an intolerable odor with a concentration of 111 mg/m³ existed. The Nice-R was switched on and a gas sample was retrieved every 20 minutes. As a result, the Nice-R was able to completely clean the air polluted with this very high concentration of an intensive and malodorous substance within 8 hours. No decomposition products of pyridine were detectable after that time. 85 b) Pollen of bushes and grasses The pictures shown below were captured with a high-definition scanning electron microscope (SEM) at the “Friedrich-Bauer-Forschungsinstitut für Biomaterialen” in Bayreuth. The pollen was exposed to a plasma treatment in the Nice-R device for short period of time. The level of damage is clearly distinguishable. Due to this irreversible damage of the surface the effectiveness of the pollen as allergen does not exist anymore (source: Hänsel / Stricher, pharmacognosy – phytopharmacy). The lock-and-key principle of the allergic reaction, i.e., the adsorption of the allergen in the mucous membranes of the respiratory system is prevented. Summary With the Nice-R ™-Air-Cleaner, a product line based on low-temperature plasma technology has been developed which can be applied in various areas, e.g., in the private sector, the administration departments, the trade and industry etc.. These devices are highly energy-efficient, e.g., for the noncommercial and business sector up to ca. 7 W. Nice R provides ambient air with a reduced concentration of pollutants, odors, and bacteria and thus is beneficial to health accompanied by a higher quality of life and an improved well-being. 86 Contact data Contact person: Armin Eschenhof Address: Oberurseler Strasse 61 – 63 61440 Oberursel Phone: +49 61 71 / 501 - 0 Fax: +49 61 71 / 501 – 312 E-mail: [email protected] MCTGmbH mag. Komponenten - Module - Systeme Notes to copyright All photos in this chapter were provided by the company MCT GmbH and are subjected to the copyright by the company MCT GmbH. This is regardless of whether the image has an apparent marking, record or a logo visible integrated in the image or an invisible digital watermark. The use of images on other web sites for personal, commercial or promotional purposes requires the written permission and/or fees of the company MCT GmbH. Unauthorised download for other media, print, digital presentations, image montages, collages and similar will not happen. This refers also to the generally applicable copyright law, including related rights. 87 3.4 BÄRO GmbH & Co. KG Dr. Martin Kirsten Introduction As one of the leading European suppliers in the field, since 1993 BÄRO has offered an extensive range of efficient air hygiene solutions. These include powerful systems for UV-C disinfection of supply air, circulating air, surfaces and water in the food industry and general ventilation and air conditioning technology. Plasma technics and KitTech from BÄRO are innovative technologies for effectively treating air containing germs, odours and/or grease. Plasma technology is based on a purely physical principle and works without using any chemicals. Odour and grease molecules and the cell structure of bacteria and viruses are destroyed and rendered harmless. Plasma technology forms the basis for an extensive product range for many different kinds of different applications. For example, these include eliminating odours from the exhaust air emitted by canteens, restaurants and hotels as well as cigarette smoke, bacteria, viruses and other pollutants in indoor air. Products with plasma technology are tailored to individual customer requirements and oriented to specific on-site conditions. All these products offer sustained effectiveness and energy efficiency and comply with the provisions of the Energy Saving Directive (EnEV 2010). Description of the technology In a process with intake air, circulating air or exhaust air Plasma technology eliminates tiny gaseous and organic carbon compounds, for example odour and fat molecules, viruses, bacteria and spores (Fig. 3.4.1). The air to be purified passes through three stages: 1. Pre-separation (pre-filter) The pre-filter keeps back the largest air impurities. This protects the following stages of the Plasma technology so that the plasma electrodes, fan and active carbon do not need to be cleaned. The prefilter is generally a high-performance aerosol separator. 2. Oxidation (Plasma stage) The plasma stage triggers an oxidation and decomposition process in the fraction of a second which eliminates almost all odours and germs in the pre-filtered exhaust air. The grease particles in the 88 exhaust air are also completely destroyed, reducing the fire load and dispensing with the need for expensive cleaning of the exhaust air duct. 3. Final stage (Activated carbon filter) An activated carbon filter as the final stage ensures that compounds previously not oxidised are kept back and decomposed. The activated charcoal filter is characterised by an extremely long service life. It regenerates itself during the process, which means that it does not need to be replaced. The air exiting the unit is completely purified and free of grease, odours and germs. Figure 3.4.1. Schematic figure of the technology. Examples of the test results For more than 75 years Hidding butcher’s shop has stood for quality, a pleasant atmosphere and customer service. In addition to meat and sausage products the branch in the old part of Münster offers a range of lunch specials. To minimise odours from the kitchen exhaust air, the proprietors opted for plasma technology from BÄRO.“Our butcher’s shop with its hot snack counter gets lots of customers every day, especially at lunchtime. We offer our customers typical Westphalian food and delicious home-style cooking. Everything is freshly prepared and comes from our own production. Customers know where it originates from”, says owner Andrea Runge. However, the daily cooking and 89 frying meant that the proprietors had increasing problems with smells from the kitchen exhaust air. They initially hoped that a “normal” ventilation system would remedy the situation. But the problems remained. The shop is placed in an area of high building density and the shopping street is very busy. Our neighbours kept complaining and our operating licence was at risk. In the search for a solution the owner came across plasma technology from BÄRO. Following a consultation with BÄRO air hygiene and an on-site analysis, the owners decided to retrofit an exhaust air unit with plasma technology in her branch (the family has four further butcher’s shops in and around Münster). Together with an air conditioning and ventilation specialist from Münster BÄRO developed a customised solution to effectively eliminate odour emissions. A plasma technology module was fitted in the existing exhaust air system in the intermediate ceiling in very cramped conditions. It was specially designed for the exhaust air requirements of the butcher’s shop and achieves an air purification rate of 3,000 m³ per hour. Including ventilation the system consumes just 1.5 kW/h per hour, which means that it does not only work cleanly - it is also extremely energy efficient. The exhaust air is cleaned in three stages of operation. To start with, the exhaust air is drawn into the system by a fan directly above the cooking areas where it is passed through an easily removable filter that frees it from grease, solids, aerosols and very small particles. In the second stage, the plasma 90 stage, the odours in the previously filtered exhaust air are destroyed by an oxidation and decomposition process. Finally, the exhaust air passes through an active charcoal filter that – as a reaction platform only – keeps back any compounds that have not yet been oxidised and decomposes them. The result is odourless and grease-free kitchen exhaust air on a purely physical basis. The elimination of grease residue in the exhaust air duct also has benefits in terms of fire protection. What’s more, the method is very environmentally friendly. There is no longer any need for timeconsuming and expensive removal of grease residue. After being cleaned by plasma technology the exhaust air that is now free from odours, germs and grease is then routed outside to the street. Andrea Runge is delighted with the result: “The exhaust air is almost completely odourless and we have not had any complaints since. And we no longer have to worry about losing our operating licence. The planning and installation were completed on time and very professionally. The investment has most definitely paid off.” Contact data Contact person: Dipl.-Ing. Martin Ferres E-mail: [email protected] Address: Phone: E-mail: Website: BÄRO GmbH & Co. KG Wolfstall 54-56 42799 Leichlingen . Germany +49 2174 799 0 [email protected] www.baero.com Notes to copyright All photos in this chapter were provided by the company BÄRO and are subjected to the copyright by the company BÄRO. This is regardless of whether the image has an apparent marking, record or a logo visible integrated in the image or an invisible digital watermark. The use of images on other web sites for personal, commercial or promotional purposes requires the written permission and/or fees of the company BÄRO. Unauthorised download for other media, print, digital presentations, image montages, collages and similar will not happen. This refers also to the generally applicable copyright law, including related rights. 91 3.5 Rafflenbeul Anlagenbau GmbH Bernd Hansel Introduction The company Rafflenbeul Anlagenbau GmbH (until May 2012 Rafflenbeul Ingenieure and NIPAG, Nichtthermische Plasma Aktiengesellschaft) works since more than 25 years in the fields of exhaust cleaning, solvent recovery, energy composite systems, modernization of printing machines, design and engineering for low-temperature or non-thermal plasma systems (NTP) for waste gas cleaning and odor removal also by using additional molecular sieve buffers. Description of the activities We do the project management from development to testing (various mobile test equipment) for the desired application using modern processes for more than 700 customers in Europe, Brazil, South Africa and the United States. We devise plants and processes for waste air purification like scrubbers, aerosol- and dust precipitators, molecular sieve adsorption systems and non-thermal plasma plants. Emissions – for instance volatile organic compounds (VOC), dusts, dioxin, hydrogen sulfide, hydrochloric acid and other harmful gases are degraded by specifically developed processes. A special advantage is the add-on system that enables combining processes among each other. The second focus of our work is developing improved measures concerning energy efficiency. Ground water cooling systems are applied instead of compression refrigeration plants. During summer waste heat is used by absorption refrigeration units for air-conditioning the production plants or for process chilling. Hybrid operated cogeneration plants take solvent containing waste air as combustion air for operating engines to generate electricity and an optimum heat supply. The third key aspect of our activities is the procedural optimization of dryers and production plants. For enabling a low-cost operation of dryers and thermal systems (e. g. phenolic resin application) energy interconnections, waste heat recovery systems and drying air multi-loops are created. What we have done Until today we have built more than 15 non-thermal plasma units that are in use in several industrial applications like coating, food and feed and water treatment industry. We treat exhaust flows from several hundred to tens of thousands of cubic meters waste air per hour. With non-thermal plasma it is possible to eliminate odors not only absolutely effectively and inexpensively but in combination with 92 molecular sieves or catalysts also organic compounds (such as benzene) are removed safely from the air. Same number of built molecular-sieve-buffer stations reduce emission peaks so smooth that downstream combustion systems can be operated cost neutral up to bringing revenue instead of high running costs. Molecularsieve buffer for 5,000 m³/h waste air NTP-unit of plasma-catalytic waste air abatement plant Contact data Address: Phone: Fax: E-mail: Rafflenbeul Anlagenbau GmbH Voltastrasse 5, 63225 Langen +49 6103/30 09 78 +49 6103/280664 [email protected] Notes to copyright All photos in this chapter were provided by the company Rafflenbeul Anlagenbau GmbH and are subjected to the copyright by the company Rafflenbeul Anlagenbau GmbH. This is regardless of whether the image has an apparent marking, record or a logo visible integrated in the image or an invisible digital watermark. The use of images on other web sites for personal, commercial or promotional purposes requires the written permission and/or fees of the company Rafflenbeul Anlagenbau GmbH. Unauthorised download for other media, print, digital presentations, image montages, collages and similar will not happen. This refers also to the generally applicable copyright law, including related rights. 93 3.6 Applied Plasma Physics AS (APP) Jon Are Beukes Introduction Increasingly, food & feed and tobacco manufacturers are being required to address the issue of release of odour and contaminants into their surroundings. Local communities, special interest groups and government legislation all bring pressure to bear on factories to minimize real or perceived risks of contamination. There are no universally applied standards for odour and Volatile Organic Compounds (VOC) emissions. Providing solutions that fit specific factory needs can be complex. Figure 3.6.1. Six NTP Units for treating an emission of 120.000 m³/h. 94 APP specializes in delivering direct corona discharge non-thermal plasma (NTP) systems for solving a range of industrial scale air-pollution control issues. Two main issues that have been addressed are odour control and particulate capture. Since the start in 1997, APP has delivered several hundred cost efficient systems worldwide for emission air volumes from 3 000 m3/h up to 2 000 000 m3/h. For odour applications the typical odour concentrations to treat varies from around 10 000 OUE/m3 to several millions OUE/m3. The system typically reduce odour by 90-98% on applications from the selected emissions, and the technology has been evaluated by the European Union through the Integrated Pollution Prevention and Control (IPPC) as Best Available Technology (BAT) for the food drink and milk sector. Cold Plasma Technology – Nature in a box The APP non-thermal plasma is an advanced odour removal system which works by some key mechanisms. These mechanisms are mimicking the processes that happen in the atmosphere, and the most important system specific mechanisms being: - Indirect radical based oxidation and reduction by radicals and subsequent particle growth - Capture of smelly aerosols and particles - Direct cleavage of odorous molecules by energetic electrons The use of direct plasma systems ensures that the whole emission is passing the plasma reactor. This is important, as some of the initial plasma processes happen in the very first microseconds and would not otherwise contribute directly in the emission. The air pollution control (APC) systems are powered by high frequency switch mode power supplies (SMPS) that have been developed by APP. The power supplies, ModuPowerTM, are modular and each module can deliver up to 30 kW and be operated in parallel. The typical operational voltage is 40-120 kV, and due to the modularity there is virtually no upper limit on the current it can deliver. When the power supplies are used for large electrostatic precipitators (ESPs), then a typical delivery consist of a few tens up to several hundred individual generators delivering up to several megawatts in total. The largest ModuPowerTM installation at a single facility can deliver more than 10MW. These Power supplies are superior to the conventional transformer/rectifier power supplies, as they give much improved collection of small particles. 95 Figure 3.6.2. Installation example. Left: In this example, 5 reaction chambers are installed in parallel, dimensioned to handle up to 100.000 m³/h. Untreated emission enters inlet stage of the reaction chambers from a lower duct and escapes vertically through the upper visible duct. Closing valves on the inlet- and outlet duct can isolate each reaction chamber. Right: Internal view of the hexagonal structure inside the reaction chamber. The low pressure-drop APC reactor has a compact design with a footprint of less than 4 m2, and is minimizing space requirements for an end-of-pipe solution. It has low running cost; typically the running cost is electric energy consumption of less than 1Wh/m3, which is comparable to system fans to push emissions through competitive higher pressure drop systems. The reactor requires a minimal level of maintenance, which is mainly connected to inspection and occasional replacement of electrodes and manual cleaning, which is normally done automatically. The APC system operates at normal manufacturing emission temperatures up to 70 °C and does not need additional cooling, such as is the case with e.g. biological abatement technology, that operates below 40 °C. The system works in low humidity and as well as in air saturated with water, which is in itself is actually contributing to even better performances. There are no needs for chemical additions and no liquid effluent waste, except when using the optional cleaning with water or by some condensation of humid emissions. Also on/off is instant and no warm up time is needed. The modular construction gives ease of relocation if change of production location and ease the ability to add on modules according to required capacity. Summary The APP systems, based on corona discharge plasma, have been delivered for several industries for odour and particle control for large volume emission applications for odour and particle removal applications. The modularity of the systems makes them easy to integrate on any emission volumes. 96 The systems are easy in use with a low pressure drop and few requirements other than electrical energy during operation. The technology uses mechanisms that mimic the natural processes in the atmosphere, and does not require water or chemical additives. There is little maintenance required, and also low costs of operation. System efficiencies have showed results on odour reductions up to 98%. Contact data Address: Applied Plasma Physics AS Bedriftsveien 25 Po. Box 584 4305 Sandnes Norway Phone: E-mail: +47 5160 2200 [email protected] Notes to copyright All photos in this chapter were provided by the company Applied Plasma Physics AS and are subjected to the copyright by the company Applied Plasma Physics AS. This is regardless of whether the image has an apparent marking, record or a logo visible integrated in the image or an invisible digital watermark. The use of images on other web sites for personal, commercial or promotional purposes requires the written permission and/or fees of the company Applied Plasma Physics AS . Unauthorised download for other media, print, digital presentations, image montages, collages and similar will not happen. This refers also to the generally applicable copyright law, including related rights. 97 Appendices Appendix A. Plasma chemical reactions Breakdown of the gas Electrons with high kinetic energies are produced during the electrical breakdown of the gas and they undergo further electron-molecule collisions, namely ionisation (1, 3), dissociation (2, 3), excitation (4) and electron attachment (7). Furthermore Penning-ionisation and dissociation (5, 6); charge transfer (8) and ion reactions (9-12) are possible. Ionisation: Dissociation: Dissociative ionisation: Excitation: Penning-Ionisation: Penning-Dissociation: Attachment: Charge transfer: Recombination: Ion-Molecule reaction: Dissociative recombination: Detachment: AB + e- → AB+ + 2eAB + e- → A + B + eAB + e- → A+ + B + 2eAB + e → AB* + eM* + A2 → A2+ + M M* + A2 → 2A + M AB + e- → ABAB + e- → A- + B AB+ + C → AB + C+ AB+ + e- → AB A+ + B- → AB + I + AB → products AB+ + e- → products (10) (11) AB- → A + B + e- (12) (1) (2) (3) (4) (5) (6) (7) (8) (9) The probability of a particular collision process depends on the energy threshold of the process and the average energy of electrons. The dissociation occurs at energies between 3 and 10 eV, while ionisation requires energies more than 10 eV and electron attachment happens at energies of some eV or lower. The unit eV (electron volts) for the characterization of energies or temperatures (mean kinetic energy) is quite common in plasma science. 1 eV equals to 11604 K. The exact values are 98 determined by the electronic configuration of the molecule being considered. The reaction rates of these collision processes further depends on the gas temperature which depends on the vibrational excitation level of molecules. Radical formation and removal Radicals are forming through direct electron impact molecule dissociation and ionization as well as ion-molecule reactions (10), dissociative recombination of ions and electrons (11), attachment and detachment reactions (12) (Chang, 2008). In air plasmas reactive oxygen species are generated by direct electron collisions (13-16), via Penning-processes (17-20) and charge exchange (21) with subsequent ion-molecule reaction (22) from O2 and H2O. Furthermore in non-thermal plasmas generated in oxygen containing atmospheres at low gas temperatures ozone, which is a strong oxidizing agent like O, •OH and HO•2 will be formed. e- + O2 → 2 O(3P) + ee- + O2 → O(3P) + O(1D) + ee- + O2 → O2(1a) + ee- + H2O → O• + •OH + e2 N( D, 3P) + O2 → O(3P) + NO N(2D) + H2O → •OH + NH O(1D) + H2O → 2 •OH N2(A) + H2O → •OH + H + N2(X) M+ + H2O → M + H2O+ H2O+ + H2O → •OH + H3O+ O(3P) + O2 + M → O3 + M O3 + •OH → HO•2 + O2 H + O2 + M → HO•2 + M (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) The removal of saturated hydrocarbons (denoted as RH, e.g. alkane) starts with dehydrogenization reactions (26, 27) followed by the oxidation of the remaining organic radical R• (28). The latter reaction result in the formation of peroxy radicals RO•2 (28) which are further oxidized down to CO2 and H2O (total oxidation) or trigger a radical chain reaction with alkyl hydroperoxide radicals R-OOH (29). In case of unsaturated hydrocarbons additionally radical addition following oxidation, radical chain reaction or polymerisation of hydrocarbons is taking place. An alternative pathway, which is important for aromatic hydrocarbons, goes through the positive ion charge transfer (30) which is followed by decomposition of the hydrocarbon. Rapid charge transfer reactions can take place when RH has lower ionisation energy than primary positive ions (> 11 eV) and this condition is fulfilled for large number of VOCs (e.g. Benzene: 9.24 eV; PAHs: <10 eV). R-H + O• → R• + •OH 99 (26) R-H + •OH → R• + H2O R• + O2 → R-O-O• Ri-O-O• + Rj-H → Ri-OOH + Rj• (27) (28) (29) M+ + RH → M + RH+ (30) In plasma-based flue gas treatment for NO and SO2 removal desired reductive reaction paths are of minor importance. Oxidative processes (31 - 33) lead to the formation of NO2. The oxidation up to N2O5 is possible. If hydrocarbons are present (e.g. ethene, propene, propane) HO•2 and peroxy radicals become the dominant oxidizers (31, 32) and the energy required to oxidize an NO molecule can be reduced. The presence of H2O or OH radicals further leads to the formation of HNO3 (35). However, to remove NOx from the gas a heterogeneous chemical process for NO2 reduction must follow the plasma treatment. As an alternative, ammonia is injected to obtain NH4NO3 which can be sold as fertilizer (36). In a similar way SO2 oxidation to SO3 and by means of plasma treatment is possible (37-38), while SO3 needs to be removed chemically (39-40). NO + O(3P) + M → NO2 + M NO + O3 + M → NO2 + O2 + M NO + HO•2 + M → NO2 + •OH +M NO + R-O-O• → NO2 + R-O• NO2 + •OH + M → HNO3 + M HNO3 + NH3 → NH4NO3 (31) (32) (33) (34) (35) (36) SO2 + •OH + M → HSO3 + M HSO3 + O2 → SO3 + HO•2 SO3 + H2O → H2SO4 H2SO4 + 2NH3 → (NH4)2SO4 (37) (38) (39) (40) Formation of aerosols Following the removal stage aerosol particles are formed through reaction of larger radicals with cluster ions and molecules. Aerosol formation is a quite important process since aerosol surface reaction rate is a few orders of magnitude higher than the electronic, ionic and radical reactions. The removal processes are promoted due to heterogeneous reactions. Regarding SO2 the stimulation of chain oxidation mechanism by plasmas in liquid droplets or ionic clusters at humid gas conditions is known (see Fridman, 2008 and references therein). References/further reading Chang, J. S. (2008). Physics and chemistry of plasma pollution control technology. Plasma Sources Science and Technology, Vol. 17, 045004 (6 pp), ISSN 0963-0252 Fridman, A. (2008). Plasma Chemistry, Cambridge University Press, Cambridge, USA, ISBN-13 978-0-521-84735-3 100 Appendix B. discharges Different types of The simplest way to clarify the regularities of ionisation is to use a closed tube with a certain gas and apply DC voltage on the metal electrodes inserted into the tube (Fig. B.1). Starting from a voltage, the ionisation starts, and the gas can undergo a transition to the plasma state. This transition is called electrical breakdown. Most of plasma types which are created by other setups can be explained on the basis of this simplest case. Figure B.1. Discharge tube. The space between electrode is known as the discharge gap and can be several tens of centimeters while the diameter of electrodes may be several centimeters. The tube is usually evacuated to 100-10000 Pa. The curve describing the current dependence on applied voltage (I-V curve) has a complex shape (Fig. B.2) and there are several regions which belong to different types of discharge. It has to be pointed that all these types have been found applications in solving environmental problems. In the region of lowest voltages (A-B), the current depends on the concentration of charged particles which are produced by external sources like cosmic rays and/or by the illumination of the cathode (photoemission) and there is only a small increase of current with increasing voltage. This regime, called sometimes Geiger regime, is used in Geiger counters where high-energy particles will create current pulses which are proportional to the intensity of the radiation intensity. At higher voltage (B-C), ionisation by the electric field becomes significant and in a narrow voltage range the current increases superexponentially. This region in the currentvoltage curve is the Townsend discharge region. Because only a small amount of light is emitted in this region, it is also known as dark discharge. It should be pointed out that in this region, the discharge is still not self-sustaining, i.e. it exists only because of an external ionisation source. In the region C-D, the Townsend discharge is self-sustained. In this region, the current increases by Figure B.2. I-V curve for discharge several orders of magnitude almost at a constant plasma voltage in neon at 133 Pa. value. An external resistor is necessary to control the current in this regime. In the region D-E the current growth is related to a plasma voltage decrease. In this region 101 the initially homogeneous electric field E becomes distorted because of the space charges accumulated in the discharge gap and the Townsend discharge changes to normal glow discharge (EF). The gas temperature of the glow discharge remains close to the room temperature while the electron temperature is tens of thousands of Kelvins (few eV). With the growth of current the plasma voltage starts to increase again (abnormal glow, F-G). Voltage fall in the region G-H is caused by the heating of cathode and the arc discharge forms. The gas temperature could reach ten thousand Kelvin while electron temperature decreases. These different regimes of the plasma will be discussed further in detail. The resistor R in figure B.1 has an important role in the stabilization of the discharge by controlling the current through the discharge because a range of possible current values may correspond to a fixed discharge voltage (Fig. B.2). In such cases, a stationary current value is established by the external resistor R as I = (Vappl-VD)/R where Vappl is the applied voltage and VD is the discharge voltage. Townsend discharge and Townsend breakdown When the discharge current is in the region of the I-V curve corresponding to the Townsend discharge (Fig. B.2 B-C and C-D) the multiplication of charge carriers is the most important feature. After passing an average distance λi = 1/α, an electron obtains energy which is sufficient to create a new electronion pair (α is the first Townsend ionisation coefficient). As a result, we have two electrons which after passing the distance λi are again able to produce new electron-ion pairs. An exponentially increasing concentration of electrons will emerge in the direction of anode (Fig. B.3). Because of the resemblance to a snow avalanche, the exponential multiplication of electrons is called the avalanche. In the region B-C of I-V curve Figure B.3. Exponential growth of electron concentration (Fig. B.2), the electrons are liberated (avalanches) inside the discharge gap. from the cathode by an external source. In the region C-D, an additional mechanism of electron production arises at the cathode: electrons are liberated due to the positive ions, which are attracted to the negative cathode, and/or photons produced in previous avalanche. This concept is known as the Townsend breakdown. The probability that an ion (photon) liberates an electron from the cathode is the coefficient of secondary emission γi (γph). New (secondary) electrons give the beginning of new avalanches. As α is itself an exponential function of electric field E, the number and magnitude of new avalanches increases quickly with the 102 growth of the electric field. At a certain voltage (point D in Fig. B.2), a new avalanche has the same size as the preceding one i.e. the discharge is self-sustained. The corresponding voltage is called the breakdown voltage Vb. In the case of fixed gas medium and cathode material, Vb depends only on the product pd. This important dependence known as the Paschen curve is shown in the figure B.4. At a certain pd value, Figure B.4. Breakdown voltage as a function of a minimum of breakdown voltage is obtained. The the product pd: Paschen curve. minimum value for breakdown voltage depends on the gas (parameters A and B) as well as on the cathode material (γ). The minimum breakdown value is smaller for inert gases (Ar, He, Kr etc.) and larger for molecular gases. Streamer breakdown The Townsend breakdown is not the only mode of breakdown. Especially at high pressures, the number of collision of electrons with the background gas particles is so large that the electric fields due to the high charge density in an avalanche head reaches the value corresponding to that of the external electric fields. Secondary avalanches which are started by photoionisation or due to background ionisation will penetrate towards primary avalanche if its number of charge carriers (positive ions) reaches about ~108 (so-called Raether-Meek criterion, see fig. B.5). This results in a perturbation of the electric field in the form of an ionisation front or wave, called a streamer. This streamer propagates towards the cathode due to the increase of local space charges. The velocity of streamers can reach ~108 m/s and during a few tens of nanosecond, the discharge transits from point D to E in the I/V curve (Fig. B.3) and a new stationary form of discharge, the glow discharge establishes. The streamer mechanism leads to the constriction of the plasmas into distinct filaments. Most non-thermal plasmas in air operate in a filamentary regime. 103 (a) (b) (c) (d) Figure B.5. Formation of streamers (a) primary avalanche; (b) primary avalanche and secondary avalanches cause perturbation of electric field; perturbation of electric field by space charges increases (c) and propagates towards the cathode (d). DC glow discharge – the classics of plasma Most of regularities of different plasma types, including the pulsed forms of plasmas, can be understood on the basis of the glow discharge formed between plate electrodes at low pressures. After establishment of a high current self-sustained discharge (Fig. B.2 point E) a redistribution of the electric field takes place (Fig. B.6 b) due to the space charges between the electrodes (Fig. B.6 a). Figure B.6. Visual appearance of a glow discharge with (a) the electron and positive ion densities and (b) electric field distribution. Contrary to the Townsend discharge where the intensity of light increases exponentially towards the anode, the visual appearance of glow discharge is more complicated. Zones with intensive light are separated by darker zones and the various distinct regions have characteristic names some of which are shown in the figure B.6. The existence of different zones is caused by space charge effects. Near the cathode, the concentration of positive ions np, surpasses considerably the concentration of electrons ne resulting in 104 a high electric field where the intensive multiplication of charge carriers takes place. As the number of electrons is relatively small and the electric field is too strong for efficient excitation (Fig. B.6), the light intensity in this region is comparatively low. The electron density ne increases towards the anode and at some distance x reaches the value of positive ions: ne ≈ np. The electric field diminishes but the energy of electrons entering this region is still high enough for excitation of atoms and intensive negative glow appears. In the Faraday dark space, the energy of electrons becomes too small for excitations of atoms. These cathode zones are followed by the positive column where np ≈ ne and the electric field is just high enough to compensate the loss of charge carriers and to emit light. In the region close to the anode, the positive space charge prevails again causing a small increase of electric field again. Enlargement of the distance between electrodes does not affect the dimensions of the cathode regions and only the prolongation of the positive column takes place. Growth of the current increases the area of the cathode occupied by the discharge but in a large range of current values the voltage drop of the discharge will stay constant (Fig. B.2 E-F). After the whole area of the cathode becomes covered by discharge, further growth of the current leads to the increase of the discharge voltage VD. This region is known as abnormal glow. With the increase of pressure, the negative glow and dark spaces surrounding it will shrink towards the cathode and at atmospheric pressure it is not possible to distinguish the cathode regions of the glow discharge. Growth of the current and/or pressure lead to the contraction of the positive column. At atmospheric pressures, the glow discharge can be obtained only under certain circumstances and external circuit parameters while the cathode has to be cooled to prevent transition to an arc discharge. Arc discharge The current increase from 0.1 to 1 A (Fig. B.2 G-H) leads to a considerable decrease of the discharge voltage and the transition to the arc discharge occurs. High current will result in the heating of the gas medium and the cathode. At increased cathode temperatures, the secondary electrons are liberated from the cathode by thermionic emission. This mechanism has a considerably higher coefficient of secondary emission γ and thus the voltage necessary to sustain the discharge will decrease and will be up to 10 times smaller. In addition, the high discharge current causes a growth of the gas temperature. The arc usually has very bright core and less bright surroundings resembling a hot flame. An arc keeps changing its position due to heating of the gas. Arcs can easily operate at atmospheric pressures and the main reason why the glow discharge is usually not obtainable at atmospheric pressures is that the arc will evolve directly after the breakdown of the gas. Arcs can also be obtained with external heating 105 or by field emissions from high electric fields. Compared with the glow discharge, the difference between gas and electron temperature decreases considerably i.e. the plasma becomes closer to thermal equilibrium. It should be pointed that the common characteristics of the I-V relationship as well as the light distribution between electrodes do not depend on the gas composition and its pressure. At the same time, the current and voltage values and dimensions of different spatial regions of discharge can differ by several orders of magnitude. Frequency effect on discharges The classical glow discharge tube was operated with direct current but a similar discharge can be obtained also in the AC regime. The breakdown process is usually completed in the time interval of 10-8 to 10-6 s and for the frequencies up to several kHz, the basic mechanisms will remain essentially the same: during each half-cycle the discharge is ignited again and a self-sustained discharge (Townsend and/or glow discharge) forms. At low alternating field the only effect is the exchange of the position of anode and cathode after each half cycle. At lower AC frequencies, during a half-cycle there is sufficient time for the removal of charge carriers from the gap between electrodes before the next half cycle. Starting from a certain frequency fmax the space charge of ions cannot be removed quickly enough from the gap and this causes breakdown at lower electric fields as compared to DC discharges. This change usually occurs at frequencies around 10 -100 kHz. At a still higher frequency even the electrons cannot reach the electrodes during a half cycle and they are lost mainly due by diffusion to the walls. This frequency gives a distinct boundary between two breakdown mechanisms, mobility controlled breakdown at lower frequencies and a diffusion controlled mechanism at higher frequencies. Increase of the frequency diminishes the role of the electrodes. This change in mechanism will manifest itself in large drop of breakdown voltage and usually it takes place in the radio frequency range (larger than MHz). At near-atmospheric pressures the microwave breakdown (GHz region) is also diffusion controlled. References/further reading Becker, K.H.; Kogelschatz, U.; Schoenbach, K.H. & Barker, R.J. (2005). Series in Plasma Physics: Non-Equilibrium Air Plasmas at Atmospheric Pressure, Institute of Physics Publishing Ltd, Bristol and Philadelphia, USA, ISBN 07503-0962-8 Fridman, A. (2008) Plasma Chemistry, Cambridge University Press, Cambridge, USA, ISBN-13 978-0-521-84735-3 Raizer, Y.P. (1991) Gas Discharge Physics, Springer-Verlag, Berlin, ISBN-3-540-19462-2 106