Fund Project:Project supported by the National Natural Science Foundation of China (Grant Nos. 51607061, 51677061) and the Fundamental Research Funds for the Central Universities, China (Grant No. 531107040929).
Received Date:11 January 2019
Accepted Date:26 February 2019
Available Online:01 May 2019
Published Online:05 May 2019
Abstract:Streamer is a strong ionizing region which advances very quickly in gases, liquids and solids. Streamer is a low-temperature plasma, which produces a variety of chemically reactive substances efficiently. So, streamer discharge has been widely adopted in industry. Furthermore, streamer is the initial stage of electric breakdown in long air gap. Studying the streamer discharge characteristics and its mechanism is the basis of external insulation in power transmission systems.Streamer branching is a significant characteristic during its development. Lichtenberg figure is the first clear recording of the filamentary structure of streamers. One of acceptable explanations is that the random fluctuations of the electron density ahead of streamer trigger branching. Furthermore, photoionization provides the necessary free electrons for the development of positive streamers. The experimental results show that the branching characteristics are closely related to the photoionization rate in streamer head. The streamer shows higher possibility of branching if the photoionization rate decreases. Since previous experiment is indirect evidence of this deduction, we turn to numerical models to study the influence of photoionization rates on positive streamer branching in atmospheric air. A three-dimensional particle-in-cell model with Monte Carlo collision (PIC-MCC) scheme called Pamdi3D (Teunissen J, Ebert U 2016 Plasma Sources Sci. Technol.25 044005) is employed in this paper. The development and branching of positive streamersin a millimeter-scale needle-plane gap are simulated at atmospheric pressure. Different streamer branching behaviors are investigated by artificially changing the nitrogen-oxygen ratio, the absorption cross section of oxygen, and the photoionization efficiency coefficient.The effects of different photoionization parameters are systematically studied. When the nitrogen-oxygen ratio, photon absorption cross section or photoionization efficiency coefficient are reduced, the streamer branching occurs earlier in three cases after reducing the photoionization rate. These results imply that the streamer shows higher possibility of branching if the photoionization rate decreases. When the streamer propagates in a non-uniform electric field region and the photoionization rate decreases to a certain value, it is believed that the seed electron distribution is more susceptible to random fluctuations. It will lead to instability in the space charge layer of streamer, thus causing the streamer to branch. Hence it is proposed that streamer branch will be triggered more easily if the photoionization rate in the streamer head decreases, in the case without considering other seed electron sources. Keywords:streamer discharge/ branching/ photoionization/ PIC-MCC/ atmospheric pressure plasma
首先改变氮气和氧气的比例, 光电离速率随着氮氧气体比例变化而改变. 当氧气浓度由20%减少至1%时, 氮气浓度变化不大, 被光子电离产生电子的氧气浓度减小了一个数量级, 因此光电离速率减小; 相似地, 当氧气浓度由20%增大至99%时, 由于氮气浓度变化了一个数量级, 提供电离的光子数大幅减小, 光电离速率也减小. 综上分析, 仿真研究了氮气-氧气比例分别为80% : 20%, 99% : 1%以及1% : 99%下流注发展情况. t = 9 ns时不同的氮气-氧气比例下流注电子密度的三维仿真结果见图3. 图中大于2.0 × 1019 m–3的电子密度区域是不透明的, 并且当密度降低时其显示的透明度增加. 在三种条件下, 所有流注都呈现有分支现象, 尤其在1% : 99%的氮气-氧气中特别清楚. 图 3 不同氮气-氧气比例下流注电子密度三维仿真结果(t = 9 ns) (a) 80% : 20%; (b) 99% : 1%; (c) 1% : 99% Figure3. Three-dimensional simulation results of electron density at t = 9 ns for different nitrogen-oxygen ratio: (a) 80% : 20%; (b) 99% : 1%; (c) 1% : 99%.
图4为三种情况下通过针尖对称轴平面上的电子数密度和电场强度在不同时刻的变化趋势. 图4(a)中流注最大场强位置离尖端1.4 mm, 流注头部电荷层在9 ns开始出现断续情况, 代表流注此时将要发展出分支; 而对于另外两种氮气-氧气比例, 图4(b)和图4(c)流注在6.0 ns时刻已经开始分支, 并且对于氮气-氧气比例1%和99%情况, 流注头部出现多个分支, 而不是像氮气-氧气比例99%和1%情况仅有两个分支. 由上述仿真结果可知, 改变氮气-氧气浓度间接地影响光电离的效率, 降低氧气浓度后会更早引发流注分支, 也进一步证实分支机理, 减小光电离速率会加速分支. 图 4 不同氮气-氧气比例下流注发展仿真结果对比 (a) 80% : 20%; (b) 99% : 1%; (c) 1% : 99% Figure4. Electron density and electric field in simulated region at different moments. It shows the comparison of streamer branching results for different nitrogen-oxygen ratio: (a) 80% : 20%; (b) 99% : 1%; (c) 1% : 99%.
23.2.氧气吸收光子电离截面系数的影响 -->
3.2.氧气吸收光子电离截面系数的影响
根据氧气中光电离的光谱研究, 电离氧气的最大辐射光波长为1025 ?. 增大光电离碰撞截面系数, 等效于减小光电离吸收平均自由程, 理论上可以增大流注头部同等距离区域内光电离产生电子数量, Zheleznyak光电离模型中最小氧气吸收光电离截面系数χmin = 0.035 Torr–1·cm–1 (1 Torr = 1.33322 × 102 Pa), 人为地将此截面系数减小1/10以及放大20倍, 取χmin = 0.0035 Torr–1·cm–1和0.7 Torr–1·cm–1, 对比两种情况下流注发展情况, 图5为通过针尖对称轴平面上的电子数密度和电场强度在不同时刻的变化趋势. 图 5 不同氧气吸收光子电离截面下流注发展 (a) χmin = 0.0035 Torr–1·cm–1; (b) χmin = 0.7 Torr–1·cm–1 Figure5. Electron density and electric field in simulated region at different moments. It shows the comparison of streamerbranching results for different absorption cross sections: (a) χmin = 0.0035 Torr–1·cm–1; (b) χmin = 0.7 Torr–1·cm–1.
其中${\upsilon _*}$为氮气有效激发系数, ${\upsilon _{\rm{i}}}$为总电离频率系数, $\xi $为波长980—1025 ?光子中平均光电离效率. Zheleznyak光电离模型和试验中提供了$P{I_{{\rm{eff}}}}$取值表[11,31], 为约化电场的函数, 如图6所示. 此效率系数直接决定光电离速率的大小. 图 6 光电离效率系数关于约化场强的取值 Figure6. Photoionization efficiency coefficient as a function of reduced electric field.
我们仿真对比了1/10和2倍$P{I_{{\rm{eff}}}}$下流注发展情况, 图7为两种情况下电子数密度和电场强度在不同时刻的变化趋势. 增大光电离效率系数如图7(a)所示, 流注的头部强电场区域变大, 到7 ns时仍远离尖电极朝外扩散发展; 但当减小光电离效率系数后, 流注头部更为紧凑发展, 7 ns后强场区域出现不连续情况(见图7(b)), 代表此时开始发生分支. 仿真结果说明, 当增大光电离效率系数后, 流注头部更稳定, 更难发生分支, 也直接验证了减小光电离速率会更早引发分支. 图 7 不同光电离效率系数下流注发展仿真结果对比 (a) 2$P{I_{{\rm{eff}}}}$; (b) 0.1$P{I_{{\rm{eff}}}}$ Figure7. Electron density and electric field in simulated region at different moments. It shows the comparison of streamer branching results for different photoionization efficiency coefficient: (a) 2$P{I_{{\rm{eff}}}}$; (b) 0.1$P{I_{{\rm{eff}}}}$.