A hybrid photocatalytic-electrostatic reactor for nitrogen oxides removal

Size: px

Start display at page:

Download "A hybrid photocatalytic-electrostatic reactor for nitrogen oxides removal"

Crystal Palmer
5 years ago
Views:

American Journal of Environmental Engineering and Science 2015; 2(1): 7-13 Published online January 30, 2015 (http://www.openscienceonline.

1 American Journal of Environmental Engineering and Science 2015; 2(1): 7-13 Published online January 30, 2015 ( A hybrid photocatalytic-electrostatic reactor for nitrogen oxides removal Gabriele Fava *, Mattia Pierpaoli Department of Materials, Environmental Sciences and Urban Planning (SIMAU), Marche Polytechnic University, Ancona, Italy address g.fava@univpm.it (G. Fava), mattia.pierpaoli@gmail.com (M. Pierpaoli) To cite this article Gabriele Fava, Mattia Pierpaoli. A Hybrid Photocatalytic-Electrostatic Reactor for Nitrogen Oxides Removal. American Journal of Environmental Engineering and Science. Vol. 2, No. 1, 2015, pp Abstract A combined photocatalytic-electrostatic apparatus for the removal of indoor levels of nitrogen oxides (NOx) has been evaluated. Titanium Dioxide (TiO 2 ) photocatalysis has been reported as inexpensive promising method to reduce gaseous environmental pollutants while electrostatic precipitation (ESP) is a highly efficient process for removing fine particles through the action of an induced electrostatic field. This article is aimed to study the synergic effect of the two processes combined into one reactor. In particular it has been studied: (i) the efficiency of NO removal and the selectivity for nitrogen dioxide (NO 2 ), (ii) their dependence from the plate-wire configuration and the inlet mass flow, (iii) the generation of ozone by ESP and its reduction by the photo activated TiO 2, (iv) the photo inhibitory effect on TiO 2 by adsorbed nitrates. A simplified path flow reaction will be also presented. Keywords TiO 2 Catalysis, ESP, NOx, Nitrates Inhibition, Ozone Reduction 1. Introduction The reduction of nitrogen oxides (NOx) is an important issue for the global environment and NO 2 is an important air pollutant because it takes parts in forming photochemical smog, reacting in the atmosphere to form ozone (O 3 ) and it is responsible, along with SO 2, as acid rain forming. Photocatalytic oxidation (PCO) over TiO 2 is reported as a promising technique for decomposition of various hazardous compounds [1]. In PCO, an ultraviolet (UV) radiation with the same energy of the TiO 2 band gap, promotes the formation of the electron/hole couple on catalyst surface and induces the formation of active species, such as OH radicals, which take part in the oxidative process. Nowadays, TiO 2 based paints attract wide interest for their beneficial abilities to provide self-cleaning and pollutants-removal benefits [2]. Nitrate is the desired end product of NOx oxidation because it is an available soluble form of nitrogen for organisms, while NO 2 and HONO are undesirable products of the oxidation. Long-term studies have shown that the overall efficiency of these paints is renewed providing that the surfaces are washed with water. Electrostatic precipitation is a technique usually selected to remove suspended particles for air cleaning. The high collection efficiency, low pressure drop and low operating costs, lead the electrostatic precipitators (ESP) to be a choice widely selected by industries for removing particulate matter from gas flow, or for cleaning indoor environments like offices, single houses, hospital or other public facilities. In this paper we report an ionizing electrostatic process combined with photocatalytic oxidation. For this study we built a simple configuration made of an aluminum plate coated with TiO 2 TiO 2 catalyst as depicted in the schematic description of Figure 1. When a high voltage is applied to the discharge electrode (tungsten wire) a positive corona discharge takes place. Ions, electrons and excited species are produced at the corona, and an ionic wind flows through the space between two electrodes. Charged particles are then picked up on the collecting electrode [3]. At the corona, low amount of NOx are generated and NO oxidize into NO 2. Some precautions to reduce this aspect are: (i) use negative instead of a positive corona discharge and (ii) lower relative humidity [4]. Another

2 8 Gabriele Fava and Mattia Pierpaoli: A Hybrid Photocatalytic-Electrostatic Reactor for Nitrogen Oxides Removal drawback of ESP is the generation of ozone [5]. In some cases it has been reported a production able to rise indoor ozone levels which approach or exceed health limits, especially when they are used continuously. The aim of this work is also to limit the NOx and O 3 productions through the photocatalytic oxidation operated by the TiO 2. To describe this combined process, we have studied separately the main effects of each stage. Figure 1. Schematic description of the ESP reactor in use in this experiment 1.1. NO Photocatalytic Oxidation The general mechanism of NO oxidation by photocatalysis has been described as follows. An electron-hole pair ( h+) generated upon UV-A excitation is trapped at the TiO 2 surface. Hydrogen ions (H+) and hydroxide ions (OH-) are dissociated from water. The active oxygen (O 2 -) species are produced on the TiO 2 surface. The nitric monoxide is oxidized mainly to nitrous acid by hydroxyl radicals and then to nitric dioxide; the overall reaction: NO + 2OH NO + H O is considered to be slow. Subsequently NO 2 reacts with another OH. and it is transformed to HNO 3, which is accumulated on the catalyst surface ESP Hybrid Photocalysis High-energy electrons generated in proximity of the wire initiate the corona reactions with the atmospheric molecular nitrogen (N 2 ) and oxygen (O 2 ). O and N radical can: (i) combine to form NO and NO 2 (in the described experiments a constant NO 2 production of 30µg/min has been noticed, depending on the electrode distance and material); (ii) recombine releasing heat, (iii) generate ozone, which may oxidize NO directly to NO 2. This single step, shorter than the full photocalysis pathway, consents a faster oxidative kinetic and higher efficiency on NO removal. The selectivity drifts anyway results higher, which means that the catalyst surface is inhibited by the fast nitrates accumulation. 2. Experimental Section 2.1. Apparatus The experimental apparatus is illustrated in Figure 2. The photo reactor has a volume of 3L and it consists of a Pyrex glass cylinder. Inside is placed a thin aluminum plate, coated on one side with the TiO 2 catalyst. This plate is laid on a poly methyl methacrylate (PMMA) support, above which the tungsten wire is placed at a variable distance from the plate. Three PMMA support with fixed distance between the wire and plate of 1.5cm, 3cm, 4cm were used. Retention time inside the reactor has been imposed between 3 and 1.5 min, depending on the inlet flux. The UV lamp is located at the center of the reactor, over the sample. The lamp is an UVA metal-halogen quartz lamp with mercury vapor, peak at 360 nm and adsorbed power of 400 W. Irradiance, which was measured with a photoradiometer Delta Ohm HD2102.2, is kept constant at 20W/m2. The probe of the photoradiometer is centered in the field of UVA with a resolution of W/m2. The NOx flux inside the reactor is kept constant with a permeation and dilution system (Calibrator 8188) which is alimented with a NOx tank (499ppb NO). Dilution is obtained by mixing with atmospheric air at room temperature (20±2 C) and relative humidity between 40-60%. NOx concentrations are determined using a chemiluminescent analyzer (Monitor Labs, Nitrogen oxides analyzer model 8841). To monitor the ozone concentration, a UV Photometric Ozone Monitor (API Ozone Monitor Model 450) has been used, connecting directly the outlet of the plug-flow reactor, through a bypass of the NOx analyzer. Figure 2. Experimental apparatus: NOx and ozone analysis are carried out separately, by means of a bypass downstream the plug-flow reactor 2.2. Sample Characterization The catalyst used is TiO 2 P25 Aeroxide. A TiO 2 paint was obtained mixing a primer, dispersant and water, and then applied on aluminum support. The aluminum plate is 10x10x0.1cm and the exposed surface was finely grated to achieve better adhesion of the paint on the support Experimental Procedure Experiments were run according to a fixed procedure; each test is divided in five parts, of different duration. A plot of NO,

3 American Journal of Environmental Engineering and Science 2015; 2(1): NO 2, NOx concentrations of a typical test is shown in Figure 3 for a visual simplification. Within the first minutes constant NO and NO 2 concentrations established. By the next 30 minutes the UV lamp is turned on, to activate the TiO 2 photocatalysis. ESP is turned on, without turning off the UV lamp, for the next 30 minutes, to have the combined effect. UV lamp is turned off. Only the ESP is still on. This phase lasts for 20-40min, depending on the stabilization of the concentration values. ESP is turned off. NO and NO 2 stabilize at the initial concentrations. Figure 3. Concentration of NOx (black), NO 2 (red), NO (blue), during one single test. It is possible to observe the 5 different phases. For each active stage (UV only, ESP only, UV+ESP) the following parameters are calculated: Efficiency of NO Reduction = Where is the initial concentration of NO while both ESP an UV are turned off (ppb), is the concentration of NO while ESP or UV or ESP+UV are turned on (ppb). Negative values indicate that not a reduction, but a production of NO is underway, due to an oxidation of the atmospheric nitrogen. Further detail will be given in paragraph NO 2 Selectivity = Where, are the initial concentrations of NO and NO 2 while both ESP an UV are turned off (ppb);, are the concentrations of NO and NO 2 while ESP or UV or ESP+UV are turned on (ppb); therefore, a negative selectivity indicates a reduction of NO 2 instead of a production. Selectivity and efficiency of conversion are calculated for each stage (UV only, ESP only, UV+ESP). 3. Results and Discussion 3.1. NO Efficiency At first the NO reduction efficiency will be discussed. Many tests have been conducted, varying: (i) the ESP wire-plate distance; (ii) the inlet NOx mass flux; (iii) the empty bed residence time Dependence on the Plate and Wire Distance The distance between the plate and wire has been varied keeping constant the direct applied voltage at 5kV. This modification changed the strength of the electric field and the ozone production. From Figure 5 we can notice that: 1. PCO-only: NO removal efficiency is constant (Figure 5-a); 2. ESP-only: NO removal efficiency highly increases with decreasing distance between the wire and plate (Figure 5-b); 3. PCO+ESP: increasing distance between electrodes, efficiency decreases because of the lower effect of ESP on the global process (Figure5-c); 4. Introducing ESP, NO 2 selectivity increases. In case of ESP-only, values higher than 100% are observed. With the combined use of ESP+PCO, selectivity remains close to zero; 5. In Figure 4 it is shown how PCO+ESP is more efficient than PCO-only. In abscissa wire-plate distances are reported; in ordinates the increment of efficiency computed as: Increment of eficiency = η #$%&'(# η #$%)*+, η #$%)*+, 100 Figure 4. Increment of efficiency of UV+ESP, compared to UV-only, for different wire/plate configuration.

4 10 Gabriele Fava and Mattia Pierpaoli: A Hybrid Photocatalytic-Electrostatic Reactor for Nitrogen Oxides Removal Figure 5. Efficiency of NO removal and NO 2 selectivity for different wire/plate configuration Figure 6. Efficiency of NO removal and NO 2 selectivity for different mass flow input NOx Mass Flow Effect In this set of experiments, the inlet mass flow rate was varied in the range µgnox min-1. Figure 6 illustrates the results obtained. It can be observed that: Overall NOx reduction efficiency is comparable in either case. As already mentioned, combining ESP with UV gives the best results: high removal rate (in the range of 80-90%) and low NO 2 selectivity (an average of 7%). Selectivity in ESP-only mode reaches and exceeds value of 100%. A possible explanation is given by (i) the ESP ozone production reacts with NO to produce NO 2, (ii) room temperature plasma of oxygen and nitrogen radicals and their recombination in forming nitrogen oxides Ozone Production As widely known [6] [7] the ESP wire-plate configuration generates ozone. Ozone can result in a number of negative health effects including respiratory symptoms, decrements in lung function and airways inflammation [8]. Some limit values of ozone concentrations are set at 51ppb (WHO), 75ppb (USA) and 61ppb (EU). On the other hand, ozone is very reactive specie able to oxidize different organic pollutants [9]. Analyses have been conducted with the purpose of determining the ESP ozone production with and without the UV illumination over the TiO 2 catalyst. As reported in literature [10], ozone is destroyed under UV illumination; a possible explanation found in literature [11] is that ozone reacts with O2-, formed superficially by irradiated TiO 2, to give additional OH. Table 1 reports Q the ESP ozone generation rates with and without the presence of PCO. The experiments were conducted varying distance between plate and wire. All generation rates were observed constant with the time. Table 1. Ozone generation rate for every different plate/wire configuration, during ESP-only and ESP+PCO phase. Wire-plate distance (cm) ESP-only ON ESP+PCO ON (µg min-1) (µg min-1) As we can see from Table 1, at a fixed potential of 5kV, increasing the electrode distance the ozone production rates diminishes. The limit of ESP influence is located between 3 and 4 cm, since no ozone concentrations can be observed. Figure 7. UV- only, ESP-only and UV+ESP ozone evolution in the presence or absence of NOx.

5 American Journal of Environmental Engineering and Science 2015; 2(1): Combining ESP and UV a maximum ozone reduction of 62% could be obtained. Figure 7 shows the ozone concentration at different working conditions. The highest O 3 concentrations are reached during ESP-only mode in absence of NOx, while during UV-only mode, ozone levels are comparable to those naturally present in the atmosphere. Combining ESP and UV the presence of nitrogen oxide, halves the O 3 levels (since part of it oxidize to NO 2 ) Oxidation of NO by Ozone In Figure 3 it is possible to observe that during UV+ESP phase, NO 2 concentrations tend to rise slightly. For this reason, additional tests were conducted maintaining for 3 hours the active phases (ESP + UV or UV-only). The results shown in Figure 8 point toward a drift proportional to the production of ozone. A confirms is given by the ozone mass balance. The ozone produced is calculated by its production rate reported on Table 1, multiplied by the test duration. From the regression reported in Figure 9 it is possible to infer a reaction stoichiometry 3 to 1 as indicates by the stoichiometry of the reaction 3NO + O 1 3NO. This findings point to ozone as responsible for the increased material reactivity since it constitutes a preferential path in the oxidation of nitric oxide to nitrogen dioxide, without passing through the intermediate oxidation state of nitrous acid. Test Table 2. Ozone - Nitrogen dioxide mass balance Wire-plate distance Tot. ozone produced (moles) NO 2 produced (moles) UV-only E E-07 UV+ESP 3cm 1.00E E-07 UV+ESP 4cm 5.94E E-06 Figure 9. O3 generated by ESP in presence of TiO 2 under illumination, versus NO 2 produced Nitrate Effect on TiO 2 Catalysis As previously suggested, the ozone generated directly oxidizes NO to NO 2 within the gas phase. The shorter pathway avoids the production of NO 2 characteristic of the photocalytic process. Further evidence comes from the nitrates to nitrites ratio found analyzing water extraction of the surface samples. The analyses show that the relationship between nitrate and nitrite depends on ozone produced: the higher the amount produced, the higher the ratio (Figure 10). Figure 10. O 3 generated by ESP in presence of TiO 2 under illumination, versus nitrate over nitrite ratio. Nitrate and nitrite have been determined after washing with pure water the surface of the catalyst. Figure 8. Concentration evolutions of NOx (black), NO (blue), NO 2 (red) for UV+ESP (wire-plate distance = 3cm) and UV only, for 3 hours of continuous run. (UV+ESP with the electrode configuration at 4cm is not shown). The effects of nitrate ions (NO 3 -) on the selective catalytic NOx reduction by TiO 2 catalyst were studied, preparing an aqueous solution of Mg(NO 3 ) 2 (840mg/l). The TiO 2 painted surfaces were added with different amount of the solution and dried using an infrared lamp. The efficiency and selectivity measured indicate that higher is the amount of nitrates added, lower is the NO reduction and higher the NO 2 selectivity as shown in Figure 11.

6 12 Gabriele Fava and Mattia Pierpaoli: A Hybrid Photocatalytic-Electrostatic Reactor for Nitrogen Oxides Removal 4. Conclusions Figure 11. Reduction of NO removal efficiency and increase of NO 2 selectivity related to the catalyst inhibition by nitrates. These findings agree with the work of other authors [12][13] who have reported that the presence of some inorganic salts may reduce the efficiency of the TiO 2 catalyst. In the literature, the inhibition of photocatalytic support in presence of non-organic ions is often explained by the scavenging of OH radical to NO 3 radical followed by NO 3 radical hydrolysis: NO 1 OH NO 1 OH 3.4. Reaction Pathway Photocatalytic oxidation starts by producing electron/hole pairs under UV irradiation. Part of them recombines but a small fraction survives on the TiO 2 surface. Here, electrons and holes react with water and oxygen leading to the formation of oxidizing species, like OH and superoxide radicals. These species initiate NOx oxidation. ESP was introduced by us in order to accumulate reactive species over the catalytic surface by the moment that the secondary ESP aspects proceed from the corona formation. A simplified NOx reaction pathway inside the ESP-hybrid photocatalytic reactor is proposed and depicted in Figure 12. It can be divided into two parts: one carried out by UV photocatalysis (slashed line) and one by ESP (continuous line). Literature reported reactions give an overview of the complex processes occurring at different reaction stages (UV-only, ESP-only, UV+ESP)[14] [15][16]. H The photocatalytic NOx conversion to NO 3 - by TiO 2 coated surfaces has been proposed for reducing the environmental pollution. The aim of this work was to investigate if and how it was possible to increase the efficiency of this process, combining it with the electrostatic process in a hybrid technology. The results obtained show that: Metal plate coated with TiO 2 proved to be a good and inexpensive material as hybrid electrostaticphotocatalytic surface. NO removal efficiency by photocatalytic oxidation reached value of 60% while NO 2 selectivity is generally negative. NO removal efficiency is strongly connected to the distance between wire and plate. Smaller the distance, higher the removal efficiency, but also higher the NO 2 selectivity. A mass balance shows that NO 2 is mostly generated by ozone oxidation directly in the gas phase. By adjusting voltage and wire configuration, it is possible to limit this unwanted effect. Combining PCO and ESP, furnished a compromise between NO removal efficiency and NO 2 selectivity. Our best results were a 93.3% of NO removal with 12% of selectivity or efficiency up to 84.3% with -5% of NO 2 selectivity. With the hybrid configuration assembled only NO 2 and NO 3 - were observed as final products. Catalyst de-activation has been attributed to the accumulation of nitrate on the active surface and we have shown that also selectivity remains affected by this buildup. Coupling PCO to ESP processes provide a new interesting and inexpensive way for the control of NOx oxidation. New concerns will be directed towards the research for synergic methods to reduce fouling of the catalyst and to achieve selectivity towards molecular nitrogen as end product. References [1] F. Akira, Z. Xintong and A. T. Donald, TiO 2 photocatalysis and related surface phenomena, Surface Science Reports, no. 63, [2] G. Kasthurirangan, J. S. Wynand and H. John, Climate Chage, Energy, Sustainability and Pavements, Springer. [3] T. Wen-Jey Liang, The Characteristics of Ionic Wind and Its Effect on Electrostatic Precipitators, Aerosol Science and Technology, vol. 20, no. 4, [4] A. Katatani, H. Yahata and A. Mizuno, Reduction of NOx Generation from Electrostatic Precipitators, International Journal of Plasma Environmental Science & Technology, vol. 4, no. 1, Figure 12. Simplified UV+ESP model [5] D. H. Rim, D. G. Poppendieck, L. L. Wallace and A. K. Persily, Effectiveness of an in-duct electrostatic precipitator in nanoparticle removal with consideration of ozone emissions, ASHRAE IAQ 2013, 2013.

7 American Journal of Environmental Engineering and Science 2015; 2(1): [6] J. Chang, P. Lawless and T. Yamamoto, Corona discharge processes, Plasma Science, IEEE Transactions on, vol. 19, no. 6, pp , [7] T. Ohkubo, S. Hamasaki, Y. Nomoto and J. Chang, The effect of corona wire heating on the downstream ozone concentration profiles in an air-cleaning wire-duct electrostatic precipitator, Industry Applications, IEEE Transactions on, vol. 26, no. 3, [8] EPA, Health Effects of Ozone in the General Population, [Online]. Available: [Accessed December 2014]. [9] S. Kazuhiko, S. Aya and S. Kazuhiko, Degradation of toluene with an ozone-decomposition catalyst in the presence of ozone, and the combined effect of TiO 2 addition, Catalysis Communications, vol. 4, no. 5, [10] K.-P. Yu and G. W. Lee, Decomposition of gas-phase toluene by the combination of ozone and photocatalytic oxidation process (TiO 2 /UV, TiO 2 /UV/O 3, and UV/O 3 ), Applied catalysis B: environmental, vol. 75, no. 1. [11] K. M. Bulanin, J. C. Lavalley and A. A. Tsyganenko, Infrared study of ozone adsorption on TiO 2 (anatase), The Journal of Physical Chemistry, vol. 99, no. 25, [12] G. Chantal, P. Eric, L. Hinda, H. Ammar and H. Jean-Marie, Why inorganic salts decrease the TiO 2 photocatalytic efficiency, INTERNATIONAL JOURNAL OF PHOTOENERGY, vol. 7, [13] S. Nikitenkoa, L. Venaultb and P. Moisyb, Scavenging of OH radicals produced from H2O sonolysis with nitrate ions, Ultrasonics Sonochemistry, vol. 11, no. 3-4, [14] B. Michael, L. Jose, B. Kurt and T. Hauke, Complex Plasmas: Scientific Challenges and Technological Opportunities, Springer, [15] M. B. McElroy, The Atmospheric Environment: Effects of Human Activity, Princeton University Press, [16] M. Ballari, M. Hunger, G.Husken and H. Brouwers, NOx photocatalytic degradation employing concrete pavement containing titanium dioxide, Applied Catalysis B: Environmental, vol. 95, no. 3-4, [17] R. S. o. C. (. Britain), 100 Years of Physical Chemistry, Royal Society of Chemistry.

Chapter - III THEORETICAL CONCEPTS. AOPs are promising methods for the remediation of wastewaters containing

Chapter - III THEORETICAL CONCEPTS 3.1 Advanced Oxidation Processes AOPs are promising methods for the remediation of wastewaters containing recalcitrant organic compounds such as pesticides, surfactants,