| Removal of oxide layer from a steel substrate is called descaling, which is a necessary procedure in metallurgical industry and can affect the quality of the steel product. Conventional methods for descaling (mechanical descaling, chemical descaling) will cause many kinds of environmental pollution such as water pollution and noise. Due to the local extremely high temperature produced by the cathode spots, the oxide at the spots will be instantaneously evaporated or heated to explosion. Because the processing is done in a closed space, it neither generates dust or noise nor causes water pollution. The evaporated material is collected using a filter, which enables recycling of the material by melting the filter. Therefore, low-pressure arc is a new environment-friendly technology for descaling.In this dissertation, the descaling process, discharge characteristic and plasma properties in descaling with low pressure dc arc plasma are studied. This work lays the foundation for the industrial application of arc descaling.A set of apparatus is built for arc descaling, experimental investigation on cylindrical electrode, sphere electrode, graphite electrode and other electrodes is carried out. The static and dynamic descaling for the samples is conducted. Descaling for plate samples of different configurations and tubular samples are achieved.The discharge properties of descaling are investigated. Descaling rate is affected by pressure, electrode gap length and current. The arc voltage and voltage fluctuation change with the property of cathode surface and discharge parameters. The voltage fluctuation is associated with the non-stationary explosive characteristic which is inherent to the arc spot. The images of the cathode spots are captured by a high speed camera, from which we can see that the spots under different conditions tend to get together or are apt to disperse. The size and the luminance vary too. It is found that the descaling rate increase with the decrease of the voltage fluctuation. We derive from the explosive model (ecton model) that the descaling rate is proportional to arc current, which is consistent with the experimental result.Electron temperature (Te) is obtained using the Fe I lines based on Boltzmann plot. For arc burns on the surface of the oxide layer, the Te at the positive column is about4000-5100K, while it is4000-6000K for on the surface of the iron. Te varies with the parameters such as current and pressure, and with the power fed into the plasma and the collision of the electron. The variation of Te at the different positions of positive column is mainly affected by cathode and anode sheath, and by the diffusion and recombination of the electrons.Electron temperature exhibits obvious fluctuation even in the same discharge process. This fluctuation is not caused by the change of the discharge conditions or the measuring error, but is associated with the non-stationary cathode process of the arc spots. At the same condition, the standard deviation of the Te for the case arc burns on the iron surface is larger than that on the oxide layer, which manifests that the arc on the surface of the iron is more instable.The oxide layer on the sample surface is not descaled at once. We observed a phenomenon of two-step scanning for plasma spots with various time scales on the sample surfaces. Spots move on the entire surface of the sample with strong evaporation and explosion of oxide. After then, a fast scanning of spots takes place on the surface once again and the clean surface of bulk metal is present. It is believed that the phenomenon of two-step plasma scanning is because of the layered features of oxides, which results in the self-adjusting of cathode spots on oxidized surface.Millisecond arc discharge is achieved, the colored random noise (CRN) of which is studied. When the arc burns on the sample surface, the arc voltage fluctuates. The frequency and amplitude of the voltage noise changes with properties of the sample surface and the discharge parameters. The voltage waveform is analyzed by fast Fourier transform (FFT), and the acquired spectral power is according with power law. The variation of the exponent a is associated with the motion of the arc spots and it’s substructure. The motion of arc spots and the substructures are corresponding to the low-frequency and high-frequency components of the waveform. The relative weight of the two components determines the exponent a. The amplitude of the voltage noise is related to the velocity and the lifetime of the spot. The interfacial property between materials affects the motion and the lifetime of the cathode spot, consequently affects the amplitude of the voltage noise. |