Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
2001-03-29
2004-04-20
Picard, Leo (Department: 2125)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
C438S691000, C438S692000, C438S959000, C341S108000, C341S126000, C341S131000, C341S155000
Reexamination Certificate
active
06725120
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to high performance systems and techniques for polishing workpieces. Specifically, the present invention relates to chemical mechanical polishing (CMP) apparatus and methods for improving the accuracy of conversion of data representing required CMP pressures to data representing CMP forces to be applied by a polishing (or planarization) head to a workpiece such as a semiconductor wafer, wherein quantization errors are minimized even though components having average resolution are used to provide some of the data used in the conversion operations.
BACKGROUND OF THE INVENTION
DESCRIPTION OF THE RELATED
In the fabrication of semiconductor devices, CMP operations are performed for buffing, cleaning, planarization, and polishing of wafers. A typical semiconductor wafer may be made from silicon and may be a disk that is 200 mm or 300 mm in diameter. The term “wafer” is used below to describe and include such semiconductor wafers and other planar structures, or substrates, that are used to support electrical or electronic circuits.
As integrated circuit device complexity increases, there is an increased need to improve the accuracy of CMP operations for planarizing dielectric materials deposited onto wafers. Also, as more metallization line patterns are formed in the dielectric materials, there is an increased need for higher accuracy in CMP operations that remove excess metallization.
In a typical CMP system, a wafer is mounted on a carrier with a surface of the wafer exposed. The carrier and the wafer rotate in a direction of rotation. The CMP process may be achieved, for example, when the exposed surfaces of the rotating wafer and of a polishing pad are urged into contact with each other by a polishing force, and when the wafer and the polishing pad move laterally relative to each other.
Two aspects of achieving accuracy of the polishing force applied to a wafer are of interest. Once a value of a required polishing pressure is specified, that value must first be accurately converted to a corresponding required force and then to a required force signal that accurately represents the required force. The force signal is applied to a force-producing device. Secondly, the actual force applied by the force-producing device must be measured and fed back to adjust the force signal. Improvements have been made to facilitate making repeatable measurements of the actual polishing forces applied to the wafer. However, there is still a need to more accurately convert the value of the required pressure to the value of the force signal. Such need exists, for example, in CMP systems in which the value of the required CMP force must be rapidly changed in relation to rapidly changing values of the exposed area of the wafer that is in contact with the polishing pad as the lateral position of the polishing pad changes relative to the wafer. CMP systems and methods in which the value of the required polishing forces are rapidly changed according to such rapidly changing values of the contact areas are described in co-pending U.S. patent application Ser. No. 09/748708, filed Dec. 22, 2000, entitled “POLISHING APPARATUS AND METHODS HAVING HIGH PROCESSING WORKLOAD FOR CONTROLLING POLISHING PRESSURE APPLIED BY POLISHING HEAD,” by Miguel A. Saldana and Damon V. Williams (the Prior Application). Such Prior Application is hereby incorporated by reference.
The CMP systems and methods of the Prior Application implement a recipe (or set of instructions) for laterally moving the polishing pad relative to a wafer carrier and to a retaining ring on the carrier. The relative movement results in the rapidly changing values of the contact area between the polishing pad and the exposed surface of the wafer, and between the pad and a conditioning puck. Feedback of polishing pad position is coordinated with determinations of required values of the variable force by which such different contact areas are separately urged into contact with the polishing pad so that the pressure on each such different contact area may be controlled. The feedback is generated by an encoder that indicates the actual successive lateral positions of the polishing pad relative to the wafer, for example. The different value of each such separate contact area is determined based on the output of the encoder. For each respective pair of one such contact area and one such pressure to be applied to that contact area, a force signal is output (commanded) to represent a corresponding requested force. Each respective force signal is applied to the force-producing device (e.g., an actuator) which provides the force by which the one such contact area of the wafer, for example, is separately urged into contact with the polishing pad at the particular time at which the actual lateral position is measured.
Even though the invention of the Prior Application enables conversions of the value of the required pressure to the force signal, there is a need to increase the resolution of the commanded force signal when the actuator that is used displays analog controllability better than that of conventional digital control methods. For example, conventional pneumatic actuators have a low (or coarse) resolution, which provides steps or increments of 2.5 pounds of force. With such coarse resolution, the actuator may be used with the conventional digital control methods having a 10 bit resolution, for example. In detail, a range of polishing pressure may be 10 psi for a 200 mm wafer that has an area of about 50.26 square inches. The maximum force is 502.6 pounds (10 psi×50.26 sq. in.). Force increments corresponding to the 10 bits are about 0.49 pounds (the force divided by the 1024 steps of the resolution). Thus, the increments of the mechanical resolution are more coarse than the 10 bit digital increments. However, when the actuator is a high resolution actuator capable of applying force in increments substantially less than 2.5 pounds (e.g., much less than the above exemplary 0.49 pounds), the conventional digital control methods do not provide the small increments of the commanded force signal that are necessary to take advantage of the high actuator resolution.
Another example illustrates errors that may result from use of devices having too low a resolution. Resolution is generally defined as 2 bit, 4 bit, n bit, etc. The number of output signals (or counts or steps) is 2 to the nth power. Thus, the very low 2 bit resolution corresponds to four counts or steps. In the context of the above-described required pressure, the resolution of the above-described digital methods dictates aspects of the force computation for converting the required pressure to the required force and to the value of the required force signal, and those aspects have an effect on accuracy. For example, the very low 2 bit resolution would correspond to a very low 2 bit computational resolution. Use of the 2 bit computational resolution would provide that a 10 psi pressure range be divided into four parts, such as discrete steps at 2.5 psi intervals, i.e., pressure values of 0, 2.5, 5.0, 7.5, and 10 psi. If the CMP system performs the conversion computations with respect to a required pressure having a value of 8.25 psi, for example, the increments (or steps) of the pressure may be 0.25 psi, which may be referred to as a parameter resolution increment. Also, 7.5 psi would be the value of the available output pressure step that is closest to the required 8.25 psi pressure. An accuracy problem resulting from such low resolution is shown by an example in which the required pressure value of 8.25 psi is to be input for processing. The conversion computation must convert the value of the required pressure (e.g., from psi to counts to voltage to counts and back to psi). Ideally, after the conversions, the required pressure would be output as exactly 8.25 psi. However, if the very low 2 bit resolution is used, the value of the required pressure would not exactly match the absolute value of any of the 0, 2.5, 5.0, 7.5, or 10 psi val
Kasenge Charles
Lam Research Corporation
Martine & Penilla LLP
Picard Leo
LandOfFree
Apparatus and methods with resolution enhancement feature... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and methods with resolution enhancement feature..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and methods with resolution enhancement feature... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3206696