A "reference cell" for generating radio-frequency (rf) glow discharges in gases at a frequency of 13.56 MHz is described. The reference cell provides an experimental platform for comparing plasma measurements carried out in a common reactor geometry by different experimental groups, thereby enhancing the transfer of knowledge and insight gained in rf discharge studies. The results of performing ostensibly identical measurements on six of these cells in five different laboratories are analyzed and discussed. Measurements were made of plasma voltage and current characteristics for discharges in pure argon at specified values of applied voltages, gas pressures, and gas flow rates. Data are presented on relevant electrical quantities derived from Fourier analysis of the voltage and current wave forms. Amplitudes, phase shifts, self-bias voltages, and power dissipation were measured. Each of the cells was characterized in terms of its measured internal reactive components. Comparing results from different cells provides an indication of the degree of precision needed to define the electrical configuration and operating parameters in order to achieve identical performance at various laboratories. The results show, for example, that the external circuit, including the reactive components of the rf power source, can significantly influence the discharge. Results obtained in reference cells with identical rf power sources demonstrate that considerable progress has been made in developing a phenomenological understanding of the conditions needed to obtain reproducible discharge conditions in independent reference cells.

Polysilicon etching in a single-wafer, parallel-plate, magneticallyenhanced RIE tool has been examined using two different approaches to the non-physical modeling of the system characteristics. The behavior of both process responses (polysilicon and oxide etch rates) and plasma parameters (voltage and current metrics) have been examined as a function of five variables (rf power, pressure, magnetic field, gas flow rate, and He backside cooling) . The variable-response mapping was examined using both neural network and response surface approaches. The greater fitting power of the former method is demonstrated in a side-by-side, internally consistent comparison of the same data set using these two approaches.

No abstract

As semiconductor processing requirements evolve to meet the demands ofdecreasing geometries, new approaches in plasma metrology will be needed to monitor the performance ofthe equipment and its processes. This performance has traditionally been monitored via Statistical Process Control (SPC) on output parameters such as etch rate and uniformity. These measurements are typically taken on single film wafers which may not be an accurate representation ofproduct. With emerging, non-intrusive, RF sensor technology, equipment and process engineers have access to signals which provide better resolution in determining the health ofthe equipment. This paper will discuss the relationships between machine settings, real-time RF sensor measurements and the etch rate and uniformity metrics typically used in machine/process qualifications. Run to run control algorithms using the RF sensor measurements will also be presented. Finally, the implications ofusing RF sensor measurements to provide real-time closed loop control ofmachine settings will be discussed. THEORYThere are many variables which affect the impedance characteristics ofan RF load. These include gas flow, chamber pressure, RF power, DC Bias, the chemistries involved in the process and the discharge itself The effects ofthese parameters create non linear loads. This means that the amplitude ofan impedance measurement (either current or voltage) is not related to other current or voltage measurements by a single proportionality constant Due to this non-linearity, the input frequency (13.56 MHz) is converted into several different freqen' Consequently, the signals from the plasma are distorted from the original wave form and contain harmonics which, when filtered, can be useful in fingerprinting a particular process condition. A harmonic is a sinusodial quantity whose frequency is an integral multiple (n>1) ofsome fundamental frequency. This can be given as:The Fourth State Technology (FST) Radio Frequency Metrology System (RFMS) consists of a broadband sensor and an electronics unit for analyzing the sensor signals. As shown in figure 1, the sensor which contains current and voltage transducers is placed between the RF impedance matching network and the powered electrode. The matching network is designed to tune the changes in the chamber load impedance so as to provide a good "match" with the source (RF generator). Consequently, the positioning of the sensor is critical so as to allow the monitoring of the plasma without any variation from the matching unit. These harmonic rich signals are passed from the RF sensor through a multiplexed filter bank to allow the capture of the voltage and current harmonic amplitudes versus time. The phase angle between the fundamental current and voltage is directly measured using the sensor signals. The sampling frequency for all of these measurements is approximately 1HZ.

No abstract