Etching a substrate: processes – Gas phase etching of substrate – With measuring – testing – or inspecting
Reexamination Certificate
1998-09-30
2001-05-08
Gulakowski, Randy (Department: 1746)
Etching a substrate: processes
Gas phase etching of substrate
With measuring, testing, or inspecting
C216S059000, C438S014000, C438S016000
Reexamination Certificate
active
06228278
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to the manufacture of semiconductor devices. More particularly, the present invention relates to improved techniques for ascertaining the end of an etch process for endpointing purposes while etching through a selected layer on a substrate.
In the manufacture of semiconductor devices, such as integrated circuits or flat panel displays, the substrate (e.g., the wafer or the glass panel) may be processed in a plasma processing chamber. Processing may include the deposition of layers of materials on the substrate and the selective etching of the deposited layer(s). To prepare a layer for etching, the substrate surface is typically masked with an appropriate photoresist or hard mask. During etching, a plasma is formed from the appropriate etchant source gas to etch through regions unprotected by the mask. The etching is terminated once it is determined that the target layer is etched through. This termination of the etch is typically referred to as the etch “endpoint.”
To determine when to terminate an etch, many techniques have been employed in the art. By way of example, the etch may be terminated upon the expiration of a predefined period of time. The predefined period of time may be empirically determined in advance by etching a few sample substrates prior to the production run. However, there is no allowance made for substrate-to-substrate variations as there is no feedback control.
More commonly, the end of an etch process may be dynamically ascertained by monitoring the optical emission of the plasma. When the target layer is etched through, the optical emission of the plasma may change due to the reduced concentration of the etch byproducts, the increased concentration of the etchants, the increased concentration of the byproducts formed by reaction with the material(s) of the underlayer, and/or due to the change in the impedance of the plasma itself.
It has been found, however, that the optical emission-based technique has some disadvantages. By way of example, the use of some etchants and/or additive gases interferes with the optical emission endpoint technique, giving rise to inaccurate readings. As a further example, as the feature sizes decrease, the amount of film exposed to the plasma through openings in the mask is also reduced. Accordingly, the amount of byproduct gases that is formed from reactions with the exposed film reduces, rendering signals that rely on plasma optical emission less reliable.
It has been found that, as the target layer etch is completed and the underlayer is exposed to the plasma, the self-induced bias of the substrate may change. By way of example, for the etch of a dielectric target layer, the self-induced bias of the substrate is observed to change as a conductive underlayer is exposed to the plasma. As a further example, for the etch of a conductive target layer, the self-induced bias of the substrate is observed to change when a dielectric underlayer is exposed to the plasma. By monitoring the change in the self-induced bias of the substrate, the end of the etch process may be ascertained for endpointing purposes.
To facilitate discussion,
FIG. 1
illustrates a typical endpointing arrangement wherein the self-induced bias on the wafer is monitored to determine when the target layer is etched through for the purpose of endpointing the etch. As shown in
FIG. 1
, a wafer
102
is shown disposed on an electrode
104
, which is typically made of a metallic material. Electrode
104
, which functions as a chuck in this example, is energized by an RF power source
106
through a capacitor
108
. During etching, the self-induced bias on wafer
102
is detected at a node
110
through a monitoring circuit
112
. Monitoring circuit
112
include a low pass filter
114
, which blocks the RF component of the signal and allows only the DC component to pass through. Since the self-induced bias on the wafer tends to be in the hundreds of volts, the signal that is passed through low pass filter
114
is typically stepped down through a voltage divider circuit to allow the monitoring electronics (not shown to simplify the discussion) to monitor the change in the self-induced bias on wafer
102
. This information pertaining to changes in the self-induced bias on the wafer allows the endpointing electronics to determine when the etch should be terminated.
However, the sensitivity and accuracy of the monitoring technique discussed in
FIG. 1
may degrade as the percentage of the target film exposed to the plasma decreases and/or if the DC conductivity between the plasma and the electrode is decreased (e.g., due to the presence of a dielectric layer underlying the target layer to be etched). Furthermore, the monitoring technique of
FIG. 1
is typically ineffective when electrostatic chucks are employed. This is because electrostatic chucks typically employ a dielectric layer between the conductive chuck body and the substrate. The presence of this dielectric layer interferes with the current path between the plasma and the chuck, rendering it very difficult to accurately determine the self-induced bias on the wafer at node
110
. Furthermore, the relationship between the voltage detected at node
110
and the self-induced bias on wafer
102
is not linear. By way of example, the resistance of the electrostatic chuck depends, in part, on the voltage existing on the chuck. Accordingly even if a signal can be detected at node
110
, it is difficult to correlate the signal detected with the self-induced bias on the substrate for endpointing purposes.
In view of the foregoing, there are desired improved techniques for detecting the end of a plasma etch process for endpointing purposes.
SUMMARY OF THE INVENTION
The invention relates to methods and apparatus for ascertaining the end of an etch process while etching through a target layer on a substrate in a plasma processing system. This invention exploits the change in the electric potential of the substrate which, for many different etch applications, corresponds to the end of the etch process. In one embodiment, the endpointing arrangement includes a current monitoring circuit configured to monitor the current flowing to a pole of the electrostatic chuck to detect a pattern indicative of the end of the etch process. Upon ascertaining the pattern indicative of the end of the etch process in the current signal, a control signal is produced to terminate the etch.
In another embodiment, the chuck represents a bipolar electrostatic chuck and currents flowing to both poles of the electrostatic chucks are monitored for the aforementioned pattern indicative of the end of the etch process in order to terminate the etch. In yet another embodiment, the differential of the currents supplied to the poles of the electrostatic chuck is monitored for the aforementioned pattern indicative of the end of the etch process in order to terminate the etch.
In yet another embodiment, the electrostatic power supply includes a bias compensation power supply, which monitors currents supplied to the electrostatic chuck poles and outputs a compensation voltage responsive thereto. The compensation voltage is then input into the chuck power supply in order to keep the currents supplied to the poles substantially equal but opposite in sign throughout the etch. In this embodiment, the compensation voltage is monitored for the aforementioned pattern indicative of the end of the etch process in order to terminate the etch.
These and other advantages of the present invention will become apparent upon reading the following detailed descriptions and studying the various drawings.
REFERENCES:
patent: 4381965 (1983-05-01), Maher, Jr. et al.
patent: 4473435 (1984-09-01), Zafiropoulo et al.
patent: 5013400 (1991-05-01), Kurasaki et al.
patent: 5045149 (1991-09-01), Nulty
patent: 5198072 (1993-03-01), Gabriel
patent: 5248371 (1993-09-01), Maher et al.
patent: 5308414 (1994-05-01), O'Neill et al.
patent: 5557215 (1996-09-01), Saeki et al.
patent: 5665166 (1997-09-01), Deguchi et al.
pate
Dassapa M. J. Francois Chandrasekar
Hudson Eric A.
Wiepking Mark
Winniczek Jaroslaw W.
Ahmed Shamim
Beyer Weaver & Thomas LLP
Gulakowski Randy
Lam Research Corporation
LandOfFree
Methods and apparatus for determining an etch endpoint in a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for determining an etch endpoint in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for determining an etch endpoint in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2449117