Atmospheric plasma processes for environmental applications

Plasma chemistry is a rapidly growing field which covers applications ranging from technological processing of materials, including biological tissues, to environmental remediation and energy production. The so called atmospheric plasma, produced by electric corona or dielectric barrier discharges in a gas at atmospheric pressure, is particularly attractive for the low costs and ease of operation and maintenance involved. The high concentrations of energetic and chemically active species (e.g. electrons, ions, atoms and radicals, excited states, photons) present in such plasmas can promote chemical reactions which are otherwise hardly possible under such mild temperature conditions. This thesis deals with the use of atmospheric plasma to activate two different processes: water purification from organic pollutants and carbon dioxide reforming of methane to produce syngas. Both address major environmental issues, specifically the ever growing demand for drinking water and the need to control carbon emissions in the atmosphere. Due to the very different nature of the two investigated processes, different plasma sources, types of discharge, reactors, experimental conditions and analytical procedures had to be developed and adopted. Despite such differences, however, both lines of research stem from a common background and share a common goal: to understand and exploit the great chemical potential of atmospheric plasma. Thus, a common research approach was used, based on extensive investigation of the discharge and plasma features, notably of its reactive species, and of the process efficiency, products and intermediates. The mechanistic investigations involved quantitative product and kinetic studies, spectroscopic determinations and some modeling. An already available prototype reactor was used for water treatment, in which dielectric barrier discharges (DBD) are generated in the air above the liquid. The strong oxidants formed in humid air plasma (OH radicals, atomic oxygen, ozone) interact with the aqueous solution and induce the oxidation of even the most resistant organic pollutants. Phenol was used as a model organic pollutant and found to be decomposed quite efficiently, especially in dilute solutions, the rate of reaction increasing linearly with the reciprocal of phenol initial concentration. Despite its high reactivity air plasma displays some selectivity. The rate of oxidation of monosubstituted phenols (m-(CH3)2N-, m-Cl-, p-NO2- and m-NO2-) depends linearly on the Hammett substituents constant yielding a rho value of -0.48 which is characteristic of electrophilic attack by the OH radical. The main products and intermediates of phenol decomposition were determined quantitatively. The behavior of two such intermediates, maleic acid and fumaric acid, was investigated in detail since they are very common water secondary pollutants formed in the oxidative degradation of most aromatic compounds. The reaction mechanisms and the role of the major oxidizing species – hydroxyl radical and ozone – were investigated in experiments in which the two acids were treated separately and also in mixture under different pH conditions. Most interesting and useful was also the comparison with the results obtained in experiments of ozonation conducted under the same experimental conditions except for the fact that ozone was produced ex situ. These experiments show that under any conditions plasma treatment is more efficient due to the contribution of short lived highly reactive species. As for the oxidation mechanism of the two acids in the plasma system, it is concluded that due to their high reactivity with ozone, the decomposition process of maleic and fumaric acids is mainly due to this species. Depending on the pH of the solution, ozone reacts directly with the organic molecules or is converted to OH radicals. However, additional OH radicals produced directly by the electrical discharge also contribute to the oxidation of maleic and fumaric acids in the air-liquid plasma system, independently on the pH used. Thus, the direct formation of •OH by the discharge in situ constitutes a big advantage of plasma treatment over reaction with ozone produced ex-situ, in particular at acidic pH values for compounds which do not react with ozone itself. In fact, contrary to ozone, OH radicals react efficiently with any organic compound and when directly produced by the discharge their concentration is independent on pH. The obtained results are also very useful to show the importance of ozone mass transfer from the gas phase to the solution. Both in plasma treatment and in ozonation ozone is not accumulated into the solution but reacts as it is transferred in water or directly on its surface. However, comparing the behavior of maleic and fumaric acids in plasma treatment and in ozonation, it was demonstrated that the ion wind present in the DBD reactor, due to the charged species formed by the discharge, plays an important role in mixing the solution. In fact, when ozone produced ex-situ is used magnetic stirring of the solution is required to allow the reaction to take place also in the bulk and not only on the surface of water, while in the case of plasma treatment magnetic stirring increases the rate of the reaction but does not change significantly the shape of the oxidation curves. The reactor and experimental apparatus for performing plasma driven carbon dioxide reforming of methane and product analysis had to be designed and developed from scratch since this line of research started with this Thesis. To allow emission spectroscopy measurements and in view of future investigations on the combination of plasma with heterogeneous catalysis, the reactor was made of quartz: two flanges are welded on the extremities of a tube of 570 mm of length and 37 mm of diameter, while a ring is welded in the middle of the tube to support a stainless steel tip which constitutes the high voltage electrode. The grounded counter electrode has the shape of a funnel and is covered by a stainless steel mesh. Most of the quartz tube is filled with ceramic cylinders, while the plasma zone occupies a volume of about 40 cm3 in the middle of the tube for allowing its heating in a vertical furnace for future investigations with heterogeneous catalysts. The setting-up of the experimental apparatus was a major task which was followed by preliminary tests with different types of discharge for determining the most efficient regime for transformation of methane and carbon dioxide to the mixture of hydrogen and CO. The best results in terms of efficiency and product selectivity were obtained with a spark discharge, self-triggered by a simple and efficient power supplying. The average electron density of the plasma, 5.7 x 1014 cm-3, was measured by emission spectroscopy techniques and the temperature of the bulk gas, approaching 100°C, by a thermocouple. However, the main characteristic of spark is the development of discharge filaments, in which the electron density and the temperature of the species, such as electrons, radicals, ions, but also atoms and molecules, are significantly higher than those of the bulk. In the present reactor these filaments fill completely the plasma zone. Thus, it is assumed that the elementary processes of the reaction between methane and carbon dioxide take place inside the discharge filaments. The major products, hydrogen and carbon monoxide were determined quantitatively by GC/FID/TCD. A few byproducts were also detected in low percentages and identified by means of GC/MS analysis. These include ethane, ethylene and acetylene. Based on quantitative product data and on precise measurements of the input and output flows, the reagents conversion, the products yield and selectivity and the energy efficiency of the process were calculated. The quite high conversion of CH4 (74%) and CO2 (69%), the high selectivity for the desired products (78% H2 and 86% CO) and the good energy efficiency (2.4 mmol/kJ) obtained make this system competitive with other reactors/processes described in the literature. Moreover, no carbon deposition was observed and CO2/CH4 ratios between 0.5 and 1.5 could be used without significant changes in the characteristics of the process. Easy power control and self-triggering of the system eliminate the need for expensive high-voltage switches, making this setup attractive for scaling up and further development