Its reactivity is comparable to oxygen. At higher temperatures, NF3 can dissociate into NF2 free radicals and fluorine atoms F, with reactivity equivalent to F, thus becoming a strong oxidant that can react violently with many substances such as water and oils. When used as an oxidant, NF3 can also serve as a source of NF2 free radicals during reactions [4,5]. The central nitrogen atom in nitrogen trifluoride adopts sp3 hybridization, possessing a lone pair of electrons, forming a trigonal pyramidal molecule. The boiling point (1013 kPa) is -1290℃, and the melting point (1013 kPa) is -2068℃, with critical constants. Main experimental equipment and reagents include carbon electrodes, electrolytic cells, control cabinets, rectifier cabinets, GC-14C gas chromatograph, electrochemical workstation, X-ray photoelectron spectrometer; ammonium bifluoride (NH4HF2), high-purity HF. The performance parameters of carbon electrodes are shown in Table 1 (fixed carbon, sulfur content, and ash content are expressed as mass fractions). Table 1 Carbon Electrode Performance Parameters Ash% True Density/(gcm-3) Bulk Density/(gcm-3) Flexural Strength/MPa Compressive Strength/MPa Resistivity/(μΩm) Fixed Carbon% Porosity% Sulfur% Carbon Electrode. Carbon Electrode. Orthogonal Experiment and Analysis of NF3 Preparation Using Carbon Electrodes 31 Orthogonal Experiment There are many factors affecting the electrolytic preparation of NF3. Under the condition of using non-graphite carbon electrodes as anodes, experiments were conducted on the main factors affecting the electrolytic results, selecting the operating temperature of the electrolytic cell, electrolyte ratio, and voltage (specific values are confidential) as research objects, with the electrolytic cell current and the content of NF3 and CF4 in the anode gas as evaluation targets. Orthogonal experiments were conducted under various conditions with carbon electrodes as anodes within the operating temperature range of 90-140℃, electrolyte mass ratio of 5-10, and cell voltage of 5-10 V. Due to technical confidentiality, specific experimental results are omitted. Within the above experimental conditions, the volume fractions of NF3 and CF4 in the anode gas varied between 390%-740% and. During the electrolytic preparation of NF3 using carbon electrodes, the anode generates gases containing. and CO2, among which NF3, N2, and N2O are the main components, with other gases less than 1% (volume fraction. and CO2 are easily separated from NF3; CF4 and NF3 have a boiling point difference of only 1℃, making them difficult to separate, so the amount of CF4 in the electrolytic gas must be below 50010-6 (volume fraction) to reduce costs. To maintain the current efficiency of NF3 production using carbon electrodes, the volume fraction of NF3 in the crude gas must not be less than 60%. During the operation of carbon electrodes, their surfaces fluorinate at a certain rate, and when the current density is below 50 mA/cm2, the fluorination rate significantly increases. To maintain a certain space-time yield of the electrolytic cell, the current of the experimental electrolytic cell is preferably above 60 mA/cm2. Orthogonal experiments also found that voltage, temperature, and the interaction between temperature and electrolyte ratio have a significant impact on the electrolytic current, with the current increasing with the increase of voltage and temperature, but the effect of electrolyte ratio and the interaction between voltage and electrolyte ratio on the electrolytic current is minimal. The content of NF3 decreases with the increase of temperature and voltage, and increases with the increase of electrolyte ratio; and voltage has the greatest impact on the content of NF3. Within the experimental electrolyte ratio range, the impact of electrolyte on the content of NF3 is smaller than that of voltage and temperature. Among the above factors, voltage has the greatest impact on the content of CF4 in the electrolytic gas, followed by the interaction between temperature and electrolyte ratio, the interaction between voltage and electrolyte ratio, then temperature, and the electrolyte ratio has the least impact on the content of CF4 in the electrolytic gas, far below the first three. 32 The Impact of Different Carbon Electrodes on the Content of CF4 in the Electrolytic Gas Under the same experimental conditions, the impact of carbon electrode 1 and carbon electrode 2 on the content of CF4 in the electrolytic gas is shown in Figure 1 (each set of time series data is 2 hours apart). The most significant factor affecting the content of CF4 in the electrolytic gas is the nature of the carbon electrode itself. Comparing the two different electrodes, it is found that the hardness of the electrode has a significant impact on the content of CF4 in the electrolytic gas. From Figure 1, it can be seen that the high-hardness carbon electrode 1 initially shows a high 151 Volume 25, Issue 2 Ji Yanzhi et al.: Application of Carbon Electrodes in the Electrolytic Preparation of Nitrogen Trifluoride.

  Carbon Electrode 1 Carbon Electrode 2 Figure 1 Changes in CF4 Content in Electrolytic Gas Prepared by Two Different Carbon Electrodes CF4 content, after the electrode surface stabilizes, the CF4 content remains at a relatively low value; for the low-hardness carbon electrode 2, the surface remains relatively active, with high CF4 content and significant fluctuations. The hardness conditions of domestic carbon electrodes cannot yet compare with foreign carbon electrodes [7], making it difficult to produce 99.99% pure NF3 using carbon electrodes. 33 The Impact of Carbon Electrodes on Conductivity Experiments were conducted using carbon electrodes with different performances listed in Table 1, and their impact on conductivity is shown in Figures 2 and 3. (The time interval for data is approximately 24 hours. The ratio of electrolytic areas corresponding to the two figures is about 15:10). Figure 2 Conductivity Change of Carbon Electrode 1 Figure 3 Conductivity Change of Carbon Electrode 2 The impact of carbon electrodes on conductivity is mainly reflected in the anode effect of carbon electrodes. Before the anode effect occurs, the conductivity of the electrolytic system is high, but after the anode effect occurs, the conductivity drops sharply (see Figures 2 and 3). Porous carbon electrodes have a large surface area, which is beneficial for improving the space-time yield of the electrolytic cell, but porous carbon electrodes are prone to anode effects, and the hardness of porous carbon electrodes is relatively low, resulting in less NF3 generated by porous carbon electrode 2 than by carbon electrode 1. Therefore, increasing the surface area under the premise of meeting the hardness requirements of carbon electrodes can maximize the reduction of electrolytic costs. The anode effect is mainly due to the carbon-fluorine film on the surface of carbon electrodes. Although the carbon-fluorine film is conductive, anode gases easily adhere to the carbon-fluorine film, causing a sharp drop in the conductive current of the entire electrolytic system. Using 1010-2mol/L potassium ferricyanide. Potassium ferrocyanide (K4Fe(CN)6) as the electrolyte, saturated KCl solution as the supporting medium, and saturated calomel electrode as the reference electrode, the potential scan rate was 20 mV/s. The potential-current curves of carbon electrode A (new carbon electrode 1) and carbon electrode B (carbon electrode 1 with anode effect) were tested, and the curves are shown in Figures 4 (C to F mass ratio from X-ray photoelectron spectroscopy) and 5. By comparing Figures 4 and 5, it can be seen that the carbon-fluorine film is conductive. Electrode surface: m(C)/m(F) is 100:113 Figure 4 Potential-Current Curve of Carbon Electrode A Figure 5 Potential-Current Curve of Carbon Electrode B152.

  March 2008 34 Comparison of N2O Content in Anode Gas Under the same conditions, the different contents of N2O in the electrolytic gas prepared using carbon electrode 1 and nickel electrode as anodes were measured, and the data are listed in Table 2. Table 2 Different Contents of N2O in Electrolytic Gas Prepared by Different Electrodes Test Number N2O Volume Fraction 106 Nickel Electrode Carbon Electrode. Average Value 2435 3 4255 The X-ray photoelectron spectroscopy of carbon electrode B (same as in Figure 5) is shown in Figure 6. Figure 6 X-ray Photoelectron Spectroscopy From Table 2, it can be seen that under normal electrolytic conditions, the N2O content in the electrolytic gas prepared using carbon electrodes is much higher than that prepared using nickel electrodes. After the same electrolyte undergoes small current water removal for the same time through nickel electrodes, the ratio of NH4HF2 to HF in the electrolyte and the water content in the electrolyte are basically the same, so part of the oxygen in the total N2O may come from the carbon electrode itself. From the chemical process of carbon electrode manufacturing, it can also be found that carbon powder (petroleum coke crushed) is bonded together with asphalt as a binder, and asphalt contains a certain amount of oxygen-containing compounds, which are not removed during the carbon electrode manufacturing process [see Figures 6(a) and (b)], but during the electrolytic preparation of NF3, oxygen may gradually migrate to the electrode surface and combine with nitrogen in the electrolyte to generate N2O. This can be seen from the lower intensity ratio of carbon to oxygen in Figure 6(a) compared to Figure 6(b). Experimental studies have shown that the characteristics of carbon electrodes required for the preparation of NF3 are: porosity, high compressive and flexural strength; carbon electrodes themselves contain a certain amount of oxygen during processing, and during electrolysis, the oxygen reacts with the electrolyte to generate N2O. When preparing carbon electrodes, the oxygen content in the material should be reduced; by further improving the purity of carbon electrode materials and selecting appropriate preparation methods, it is feasible to replace nickel electrodes with carbon electrodes.

  Reprint Statement: This article is reprinted from online resources, and I bear no responsibility for this article. If anyone suffers any damage, personal injury, or any other harm due to this article, all consequences shall be borne by the original author (I cannot identify the author, please cooperate with public security and internet supervision authorities to find the author).