How the feed gas composition influences the chemistry induced by non-thermal plasma in water solution

How the feed gas composition influences the chemistry induced by non-thermal plasma in water solution FINAL REPORT - ERASMUS+ INTERNSHIP Author: Octavian Lombart Supervisors: Dr. Cristina Canal, Dr. Francesco Tampieri Session: June 2021 2 3 1 - Introduction We start by defining what a plasma is. It is one of the four fundamental states of matter (Fig. 1.1), first systematically studied by Irving Langmuir in the 1920s. It consists of a gas of ions - atoms or molecules which have one or more electrons stripped (or, rarely, an extra electron attached) - and free electrons [Wiki 2021]. Figure 1.1: The four states of matter. Plasmas can be artificially generated by heating a neutral gas or by subjecting it to a strong electromagnetic field. The presence of free charged particles makes plasma electrically conductive, with the dynamics of individual particles and macroscopic plasma motion governed by collective electromagnetic fields and very sensitive to externally applied fields. Plasmas can be natural or artificial. Some examples of natural plasmas are: ➢ Astrophysical plasma ➢ Stars, gas nebulae, quasar, pulsar ➢ Northern lights ➢ Lightning ➢ Ionosphere ➢ Solar wind ➢ Tail of comets ➢ Trail of shooting stars ➢ Heart of flames Examples of artificial plasma are: ➢ The electrical discharges (arcs as in the high-voltage circuit breakers or torches, or other types of discharges as in gas lamps, microwave discharge, or the generators of X-ray) ➢ Treatment plasmas for deposition, etching, surface modification or doping by ion implantation ➢ Plasma televisions ➢ Propulsion plasmas ➢ Nuclear fusion Plasmas can be classified in two types: thermal plasma and non-thermal plasma. The temperature of a plasma is given by the kinetic energy of electrons and heavy particles such as ions and neutral atoms and/or molecules. In this report, only non-thermal plasma will be considered, in which the temperature of the electrons is way higher than the temperature of ions and neutral species. Nonthermal plasmas can be generated at atmospheric pressure and room temperature by applying an 4 electric current produced by a high voltage generator. This excites the gas and causes it to pass from the gaseous to the plasma state. This excited state of the gas disappears as soon as the current supply is stopped. Figure 1.2: Components of non-thermal plasma. A plasma is an extremely reactive environment (Fig. 1.2). The primary reactive species that are generated (free electrons and radical ions) quickly react to generate secondary reactive species or recombine to restore the original neutral atom/molecule. The nature of these primary and secondary reactive species depends on the plasma feed gas. When a plasma is generated in presence of air (or using air as a feed gas) a cocktail of reactive oxygen and nitrogen species (RONS) is generated. The most typical are ozone (O3), atomic oxygen (O), singlet oxygen (1O2), superoxide (O2-·) and nitrogen oxides (NO, NO2). If the plasma is generated in contact with an aqueous solution, or in presence of water vapour in the air, many RONS derive from the interaction with water. The most typical are hydroxyl radicals (·OH), hydrogen peroxide (H2O2), nitrite (NO2-) and nitrate (NO3-) ions. 5 Figure 1.3: Typical Reactive Oxygen and Nitrogen Species generated when an air plasma is in contact with water [Samukawa 2012]. Thanks to the presence of these cocktails of RONS, whose nature and concentration can be tuned by changing some experimental parameters - like the applied voltage, feed gas flow rate and composition, atmosphere, plasma-to-target distance, etc - non-thermal plasmas find promising applications in many fields as energy production, agriculture, environment, manufacturing and medicine (Fig. 1.4). Figure 1.4: Application of non-thermal plasmas. 6 7 2 - Aim of the work The nature and amount of plasma-generated reactive species in the gas-phase and in water solution are strongly dependent on many experimental parameters. According to [Neretti 2018], the helium flow rate has an impact on the reactive species observed. In a general way, the amount of the studied species (OH and N2) increases with the flow. Some late studies analyzed the influence of the feed gas composition on the concentration of reactive species generated in water. A recent paper by B. Myers [Myers 2021] showed that the amount of OH radicals and O atoms generated by the same plasma source is strongly dependent on the addition of traces of molecular oxygen or water vapour to the feed gas (helium) and that this has a paramount influence on the subsequent water chemistry. The objective of this work is to study the dependence of the plasma-generated reactive species in the gas-phase and in aqueous solution, as a function of the composition of the feed gas using an home-made atmospheric-pressure plasma jet that was previously studied only with pure helium. The gas composition will be modified by doping pure helium with small amounts of oxygen or nitrogen. At first qualitative experiments will be performed on the plasma plume in the gas-phase using optical emission spectroscopy (OES). These will allow us to analyze the excited species generated by plasma and their dependence on the gas composition. Then, using some selected gas compositions a detailed quantitative analysis of reactive species generated in the treated liquid as a function of the plasma treatment time will be performed. 8 9 3 - Experimental Section 3.1 - Materials Sodium dihydrogen phosphate dihydrate (NaH2PO4·2H2O, >98.0%), disodium hydrogen phosphate dodecahydrate (Na2HPO4·12H2O, >98.0%), sulfanilamide (C6H8N2O2S, ≥99%), N-(1- naphthyl)ethylenediamine dihydrochloride (C12H14N2·2HCl, >98%), hydrogen peroxide solution (H2O2, 30%), sodium nitrite (NaNO2, >99.9%), potassium nitrate (KNO3, >99.0%), titanium(IV) oxysulfate - sulfuric acid solution (27-31% H2SO4 basis), 2-hydroxyterephtalic acid (C8H6O5, 97%) and potassium indigotrisulfonate (C16H7K3N2O11S3) were purchased by Sigma Aldrich. Disodium terephthalate (C6H4(COONa)2, >99%) was purchased by Alfa Aesar. Nitrate test (Spectroquant, method: photometric, DMP) was purchased by Supelco. Ultrapure water was obtained by filtration using 0.22 μm pore size MILLEXGP filter unit (Merck Millipore Ltd., Ireland). Helium gas (99.998%), oxygen and nitrogen were provided by Praxair, Spain. 3.2 - Plasma source and treatments The plasma source used in this work is an atmospheric pressure plasma jet (APPJ) that was already described in detail in previous publications [Zaplotnik 2015, Canal 2016]. The active electrode is a copper wire (0.1 mm diameter) embedded inside a quartz tube (ID = 1.2 mm) and connected to a high voltage power supply. The discharge was operated with a sinusoidal waveform at 23 kHz with (U) ∼ 2 kV and (I) ∼ 3 mA. The average power delivered to the discharge was 1 W. Helium (pure or doped with oxygen or nitrogen) flows through the tube and serves as plasma feed gas. Two Bronkhorst EL-FLOW Select flow controllers were used to set the flow rate of the gases. All the treatments, if not otherwise stated, were done using the following conditions. He flow rate: 1 L min-1 (+ small contribution of the dopant gas); nozzle-to-target distance: 10 mm and 5 mm (fixed); sample volume: 1 mL in 24-well plate for liquid-phase analysis and 1.9 mL in 48-well plate for gas-phase analysis. Treatment time: between 60 to 600 s. The gas flow was started 15-20 min before each treatment in order to ensure the purge of the gas line. The plasma was started at least 5-10 min before the treatment to let it stabilize. Each plasma treatment was repeated three times to ensure reproducibility. 3.3 - Optical Emission Spectroscopy experiments Optical emission spectroscopy experiments were performed using a StellarNet Black C-25 LT-14 Spectrometer equipped with an Ocean Optics QP600-2-SR optical fibre and SpectraWiz Software (Fig. 3.1). The head of the optical fibre was mounted perpendicular to the plasma plume, 3-4 mm horizontal distance. The distance was the lowest possible to maximize the signal without generating sparks. The vertical position of the optical fibre head was changed with respect to the plasma plume to explore different regions. During the optical emission spectroscopy measurements, 1.8 mL of water in a 48-well plate (full well) were positioned 5 or 10 mm below the plasma nozzle. The well was refilled when needed to compensate for the water lost due to evaporation. 16 Many OES spectra were collected varying the feed gas composition and the plasma-to-target distance. In the following figures are reported the intensity of the main species just described as function of the feed gas composition (increasing percentage of nitrogen or oxygen) in all the conditions studied. All spectra were acquired in duplicate and the intensity values reported in the figures are the average of the two experiments. Helium doped with nitrogen Figure 4.2: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: nitrogen; plasma-to-target distance: 5 mm; z = 1 mm. For the helium doped with nitrogen at z = 1 mm with a plasma-to-target distance of 0.5 cm (Fig. 4.2), the intensity of NO signals increases a bit at low dopant percentage (0.05%) and stays constant until 0.3% and then decreases and stays constant. OH signals increase until 0.1%, then stay constant until 0.3% and then decrease. N2 signals start to increase until 0.3% and then stay constant. N2+ signals increase at 0.05% and then decrease slowly until 1%. He and O signals have the same profile: they increase for low doping percentage and then almost disappear at nitrogen content higher than 0.4%. Figure 4.3: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: nitrogen; plasma-to-target distance: 10 mm; z = 1 mm. 17 By increasing the plasma-to-target distance to 1 cm (Fig. 4.3), all the nitrogen species show more or less the same trends. On the other side, the signals of He and O are very low, even at low dopant percentage. Figure 4.4: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: nitrogen; plasma-to-target distance: 5 mm; z = 4 mm. For the helium doped with nitrogen at z = 4 mm with a plasma-to-target distance of 0.5 cm (Fig. 4.4), all the species show the same trend as with z = 1 mm. Figure 4.5: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: nitrogen; plasma-to-target distance: 10 mm; z = 4 mm. For a plasma-to-target distance of 1 cm and z = 4 mm (Fig. 4.5), NO signals increase at low dopant percentage until 0.2% and then stay constant. OH signals increase until 0.5%, stay constant until 0.2% and then decrease by further increasing the dopant percentage. N2 signals increase at low dopant percentage until 0.4% and then stay constant. N2+ signals increase from 0% to 0.05% and then decrease slowly by further increasing the dopant percentage. He and O signals stay at a low value (almost zero), regardless of the composition of the gas. 18 Helium doped with oxygen Figure 4.6: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: oxygen; plasma- to-target distance: 5 mm; z = 1 mm. For the helium doped with oxygen at z = 1 mm with a plasma-to-target distance of 0.5 cm (Fig. 4.6), NO signals are very low and constant regardless of the gas composition. OH signals decrease quickly from the pure He condition until 0.2% oxygen and then stay constant. N2 signals decrease quickly until 0.08% oxygen and then stay constant. N2+ signals decrease quickly until 0.03% and then stay constant. He signals increase from 0 to 0.03% oxygen, stay constant until 0.2% oxygen and then decrease until 0.4% and then stay at a low signal. O signal increases from 0 to 0.06%, stays constant until 0.2% and then decreases at 0.3% and then stays constant. Figure 4.7: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: oxygen; plasma- to-target distance: 10 mm; z = 1 mm. For a plasma-to-target distance of 1 cm (Fig. 4.7), no NO signal was evident in the spectra. OH signals decrease at the beginning until 0.2% and then stay constant. N2 signals decrease at the beginning until 0.15% and increase until 0.2% and stay constant until 0.4% and then decrease and stay constant from 0.5%. N2+ signals decrease from 0 to 0.2% oxygen and then stay constant at a low signal. He signals increase slightly until 0.05% and then decrease auntil 0.2% and stay at a low signal. O signal increases with the presence of oxygen and decreases from 0.2% until 0.5% and stays at a low signal. 19 Figure 4.8: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: oxygen; plasma- to-target distance: 5 mm; z = 4 mm. For the helium doped with nitrogen at z = 4 mm with a plasma-to-target distance of 0.5 cm (Fig. 4.8), NO signals are constant at a low value, regardless of the composition of the gas. OH signals are decreasing until 0.3% and then stay constant at a low value. N2 signals are increasing until 0.06% and then decrease until 0.3% and then stay constant until 1%. N2+, He and O signals are constant at a low value, regardless of the composition of the gas. Figure 4.9: Intensity of the main excited species in the plasma plume as function of the feed gas composition. Dopant: oxygen; plasma- to-target distance: 10 mm; z = 4 mm. For a plasma-to-target distance of 1 cm (Fig. 4.9), no NO signal was evident in the spectra. OH signals decrease until 0.2% and then stay constant at a low value. N2 signals decrease with the oxygen percentage until reaching a low value. N2+ signals increase a bit with oxygen but decrease from 0.02% and then stay at a low signal. He and O signals are constant at a low value, regardless of the composition of the gas. All the data presented in this section allow us to select some interesting conditions for the subsequent quantitative analysis of plasma-generated species in water solution: ➢ For helium doped with oxygen, 0.1 and 0.3 % are selected because they correspond to a higher generation of oxygen atoms with respect to the pure helium condition. ➢ For helium doped with nitrogen, 0.2 % is selected because it corresponds to a maximum in the N2+ trend and 0.5 % because it corresponds to maxima in N2 and NO trends. 20 4.2 - Quantitative analysis of plasma-generated species in water solution We know from previous studies [Tampieri 2021] that the evaporation rate of water due to plasma treatment for a plasma-to-target distance of 10 mm is (2.10 ± 0.09)·10-4 s-1. We obtained the same parameter for 5 mm distance. We treated 1 mL aliquots of phosphate buffer solutions in a 24-well plate with plasma source and we weighted the liquid at the end of the treatment. All measurements were done in triplicate. We plotted the relative evaporation as a function of the plasma treatment time (Fig. 4.2) and we obtained the evaporation rate by fitting the experimental data with a linear function. For a plasma-to-target distance of 5 mm we obtained an evaporation rate of 2.282 x 10-4 mL.s-1. The evaporation rates have been used to correct the data of concentration of RONS in water solution. Figure 4.10: Evaporation as function of the treatment time for experiments with plasma-to-target distance of 0.5 cm. In the following figures are reported the concentration of plasma generated reactive species in water obtained using the method reported in the previous chapter. All experiments have been done in triplicate and the data were corrected taking into account for the evaporation. The experimental data have been interpolated with straight lines and the slopes of these lines have been collected in tables. The slopes represent the rates of formation of the reactive species. Ozone Quantitative analysis for ozone was performed as described in the experimental section but in all the conditions studied the amount of ozone generated was lower than the detection limit of our quantification method. In all cases the absorbance of indigo solution after addition of plasma treated solution was not different from the one after addition of untreated solution. Therefore, no figure or table is presented in this section. Hydrogen peroxide Hydrogen peroxide in water is generated by recombination of OH radicals. 21 Figure 4.11: Concentrations of hydrogen peroxide as a function of plasma concentration treatment time with different feed gas conditions and a plasma-to-target distance of 0.5 cm Figure 4.12: Generation rate of hydrogen peroxide for different feed gas conditions at a plasma-to-target distance of 0.5 cm In the case of the 0.5 cm plasma-to-target distance (Fig. 4.11 and 4.12), the amount of hydrogen peroxide generated is nearly the same in each gas composition, and it is increasing with the treatment time, in the same way (slopes are similar), but a bit more in the case of He +0.5% of N2. Figure 4.13: Concentrations of hydrogen peroxide as a function of plasma concentration treatment time with different feed gas conditions and a plasma-to-target distance of 1 cm 22 Figure 4.14: Generation rate of hydrogen peroxide for different feed gas conditions at a plasma-to-target distance of 1 cm In the case of the 1 cm plasma-to-target distance (Fig. 4.13 and 4.14), the concentrations of hydrogen peroxide in the case of pure He and He + 0.2% of N2 are increasing linearly with the treatment time up to 10 min treatment, and they reach approximately the same value (about 0.6 mM), which are the highest that we could quantify between al the conditions studied. In the case of He + 0.1% of O2 and He + 0.5% of N2, the concentrations are increasing slower with the treatment time compared to the previous ones, and they reach a plateau condition after 5 min treatment. For He + 0.2% of O2, the concentration of hydrogen peroxide remains very low for all treatment times. Regarding the distance, it appears that with pure helium or adding a small amount of nitrogen, the concentration of hydrogen peroxide is higher if the plasma is 10 mm from the water surface. For all the other conditions (addition of oxygen or high amount of nitrogen) the concentration is higher when the plasma is closer to the water surface. Nitrite ions Nitrite ions are generated by the reaction of NO with water. In this way nitrous acid is produced and it can dissociate in water to generate nitrite ions. Figure 4.15: Concentrations of nitrite ions as a function of plasma concentration treatment time with different feed gas conditions and a plasma-to-target distance of 0.5 cm 23 Figure 4.16: Generation rate of nitrite ions for different feed gas conditions at a plasma-to-target distance of 0.5 cm In the case of the 0.5 cm plasma-to-target distance (Fig. 4.15 and 4.16), the concentration of nitrite in the case of pure He and He doped with oxygen, is very low, even at high treatment time. When doping He with N2, the concentration of nitrites increases with respect to pure helium is increasing. The increase is much higher in the case of 0.5% nitrogen. Figure 4.17: Concentrations of nitrite ions as a function of plasma concentration treatment time with different feed gas conditions and a plasma-to-target distance of 1 cm Figure 4.18: Generation rate of nitrite ions for different feed gas conditions at a plasma-to-target distance of 1 cm For 1 cm plasma-to-target distance (Fig. 4.17 and 4.18), the concentrations of nitrites in the case of He doped with oxygen and He + 0.5% of N2, is very low for all treatment times. For pure He, the concentration is increasing until 6 min of treatment and then reaches a plateau (about 80 uM). For He + 0.2% of N2, the concentration is increasing with the treatment time and is the highest one (150 uM at 10 min treatment). In general, the amount of nitrites obtained at a plasma-to-target distance of 1 cm are higher for the pure He and He doped with a small amount of nitrogen. In all the other conditions, the short distance gives higher concentration. 24 Nitrate ions Nitrate ions are generated by the reaction of NO2 with water. In this way nitric acid is produced and it can dissociate in water to generate nitrate ions. Nitrates can also be generated as a product of the reaction of nitrites and hydrogen peroxide in presence of an acid. Figure 4.19: Concentrations of nitrate ions as a function of plasma concentration treatment time with different feed gas conditions and a plasma-to-target distance of 0.5 cm Figure 4.20: Generation rate of nitrate ions for different feed gas conditions at a plasma-to-target distance of 0.5 cm In the case of the 0.5 cm plasma-to-target distance (Fig. 4.19 and 4.20), all the concentrations are increasing with treatment time. In the case He doped with oxygen, the concentrations are increasing slower than pure He, and in the case of He doped with nitrogen, the concentrations are increasing faster. But for He + 0.2% of N2, the concentration is increasing until 6 min of treatment and then reaches a plateau. For He + 0.5% of N2, the concentrations are the highest keep increasing with the treatment time. 25 Figure 4.21: Concentrations of nitrate ions as a function of plasma concentration treatment time with different feed gas conditions and a plasma-to-target distance of 1 cm Figure 4.22: Generation rate of nitrate ions for different feed gas conditions at a plasma-to-target distance of 1 cm For 1 cm plasma-to-target distance (Fig. 4.21 and 4.22), the concentrations in the case of He doped with oxygen and He + 0.5% of N2, are very low and stay constant with the treatment time. But for pure He and He + 0.2% of N2, the concentration is increasing until 6 min of treatment and then reaches a plateau. The concentrations with He + 0.2% of N2 are higher than with pure He. The highest concentration of nitrates have been observed for 1 cm plasma-to-target distance. OH radicals and O atoms A recent published work [Myers 2021] reported that the product of the reaction between hydroxyl radicals and terephtalate and the product of the reaction between O atoms and terephtalate are the same, hydroxyterephtalate. Therefore, using TPA as a probe during the plasma treatment should allow us to indirectly quantify the sum of the contribution of both species.