Fault classification in a plasma process chamber

Details Fault classification in a plasma process chamber Fault classification in a plasma process chamber

2002-02-14

2004-11-30

Barlow, John (Department: 2863)

Data processing: measuring, calibrating, or testing

Measurement system in a specific environment

Mechanical measurement system

C324S459000, C702S085000

Reexamination Certificate

active

06826489

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method of fault classification in a plasma process chamber powered by an RF source.
2. Prior Art
Many thin film processes use plasma processes to facilitate the rapid and accurate fabrication of minute structures with desired properties. Plasma processes include the deposition and etching of insulators, conductors and semiconductors on a substrate, for example, a silicon wafer. The plasma process usually involves placing the substrate in a vacuum chamber, introducing process gases and applying radio-frequency (RF) power, typically 0.1 to 200 MHz, to create a plasma. The plasma consists of ions, electrons, radical gas species and neutral gas, all of which permit the desired reaction to proceed. The plasma reaction has many inputs, including RF power, gas type and flow rates, chamber pressure, substrate and wall temperatures, chamber wall conditions, electrode spacing, and so on.
Control of the complex plasma process is the key to improved manufacturing, i.e. to have accurate and repeatable processing the plasma itself should be repeatable.
Unfortunately there are few schemes in existence for direct plasma monitoring and control. It is more usual to monitor or control process inputs such as gas flow, power output from RF generator, chamber pressure or temperature, etc., using statistical process control charts. However, since the plasma process depends directly on the plasma parameters, measuring these indirect variables is generally not sufficient.
In the ideal production scenario, the plasma chamber operates continually, only stopping for scheduled maintenance. However, because of the complexity of the process, faults do occur, resulting in unscheduled tool downtime. To maximize productivity, these faults need to be repaired as rapidly as possible. Faults generally appear either as aborts on process set points or as out of control product metrology, which includes fails in parameters such as etch rate on test wafers as well as device yield drops.
Because of the dearth in tool control and monitoring, faults are often addressed using a cause and effect approach. For example, during production monitor checks it may be that there is an out-of-control etch-rate. The technician will consider all inputs (causes) which affect etch rate, including plasma power, pressure, gas flows, cathode position and others. Generally there will be few clues as to which has changed so all may have to be checked. This is costly in terms of labour and lost production time. Furthermore, the drive to return the process tool to production as rapidly as possible may involve unnecessary parts replacement, again incurring significant cost.
The RF powered plasma represents a non-linear complex load in electrical terms. This results in the generation of harmonics of the RF driving signal. These harmonics, known as Fourier components, are very sensitive to changes both in the plasma process and the process parameters.
It is generally accepted that monitoring the Fourier components of the RF power signal provides a useful way to monitor the plasma process. These components are a more direct measurement of the plasma process since they are more directly related to fundamental plasma parameters.
It is known to use an RF sensor to monitor and control RF plasmas by measuring the Fourier components of voltage and current. The sensor can be used in closed or open loop control, as for example, in etch end-point control or as in-situ monitoring of the plasma process. In either case the plasma can be terminated when one or more of the RF Fourier components reaches pre-determined limits.
Unfortunately, when problems occur in the plasma reactor due to undesired changes in input parameters, such as changes in gas flow rate, the RF sensor as previously described can only determine that a fault has occurred, i.e., it cannot determine the exact fault mechanism or faulty component.
There is a need, therefore, for an improved method for fault identification on plasma processing chambers.
SUMMARY OF THE INVENTION
Accordingly, the present invention provides a method of fault classification in a plasma process chamber powered by an RF source, comprising the steps of:
a) running a plurality of different baseline processes on the chamber,
(b) in respect of each baseline process, determining the magnitudes of a plurality of Fourier components of delivered RF power and storing the magnitudes as reference data for that baseline process, and
c) when a fault is to be classified, repeating at least one of the said baseline processes according to a predetermined decision tree to classify the fault by comparing the current magnitudes of the said Fourier components with the corresponding reference data.
In the present specification a baseline process is a plasma process with pre-determined values for the process input parameters. It is therefore a datum from which inferences regarding the plasma process may be drawn.
An embodiment of the invention involves the maintenance of a set of baseline impedance fingerprints for a given plasma chamber. This set includes an impedance fingerprint of the chamber running a typical process recipe. Also included is an impedance fingerprint of the chamber alone, without any plasma. Also included is a fingerprint of the plasma chamber running an inert gas plasma. All of these impedance fingerprints includes a set of electrical signals associated with the RF power delivery. They may include RF fundamental and Fourier components of voltage, current and phase and derived impedance.
This set of impedance fingerprints is recorded and maintained regularly. When a fault occurs on the chamber, a systematic approach to fault finding is employed. Each individual impedance fingerprint is retaken as necessary and compared to each of the baseline impedance fingerprints. The approach allows the user to classify the fault as either a hardware component, or a process component.
The invention therefore provides a technique that allows an operator to diagnose the cause of a fault, or at least eliminate a large number of probable causes, which would otherwise prove costly to investigate.


REFERENCES:
patent: 5576629 (1996-11-01), Turner et al.
patent: 5642296 (1997-06-01), Saxena
patent: 5691642 (1997-11-01), Dobkin
patent: 5808415 (1998-09-01), Hopkins
patent: 6252354 (2001-06-01), Collins et al.
patent: 6441620 (2002-08-01), Scanlan et al.
patent: 6603538 (2003-08-01), Oluseyi et al.
patent: 6656843 (2003-12-01), Bol
patent: 82556 (2002-10-01), None

Affiliated with

Also associated with

Rating

Fault classification in a plasma process chamber does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Fault classification in a plasma process chamber, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fault classification in a plasma process chamber will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3345453