Experimental investigation of high efficient inductively coupled plasma by splitting external antenna coil

Summary form only given. We investigate the effect of the discharge coil structure to improve the discharge efficiency in the remote inductively coupled plasma (ICP) source. Experiment was performed with cylindrical quartz tube placed on aluminum chamber where plasma processes are done. Plasma diagnostics were made in the center of Al chamber where wafer is placed and actual plasma process has done. The RF power is delivered by 3 turn helical coil antenna placed externally at the middle of the quartz tube. The antenna coil is split along helical coil where the total coil length is fixed; hence three different type of coil structure was tested. First antenna structure has single coil which has 3 turns. Second structure split the coil into two, of each having 1.5 turns, connected parallel. Third case used three antenna coils each having a single turn connected parallel. Both second and third coils are connected parallel for power distribution from single matching network. Each antenna ends has measured its voltages using a high voltage probe, then balanced via end-capacitor (vacuum variable capacitor)1. RF power sustained at 13.56 MHz delivered up to 3kW to our ICP source. Nitrogen and oxygen were used for the working gases. The chamber pressure was kept constant at 0.2 Torr throughout plasma diagnostics. Plasma density was spatially measured using a floating harmonic probe system. For each antenna configurations, V/I values were measured to compare impedance changes. With given number of coil turns the total inductance is decreased by increasing the number of coil splits, where the lowest voltage and the highest current were measured. E-field, which is created by high voltage over discharge coil, has sustained the plasma by electrostatic coupling. It is experimentally confirmed that EH mode transition power gets lower as the value of antenna impedance got higher i.e., single turn. The measured plasma density is highest for using two turn coils compare to other coil configurations. The result of these experiments describe the resistive loss induced by coil resistance for using longer coil, i.e., single turn, and the inductance loss for using short coil hence reducing the inductive coupling, i.e., three turns, has to be compensated for optimal discharge configurations. These results will be considered to design an optimized discharge coil configuration for the ICP source and to improve the process yield.