Optics: measuring and testing – Dimension – Thickness
Reexamination Certificate
2000-04-26
2003-09-16
Pham, Hoa Q. (Department: 2877)
Optics: measuring and testing
Dimension
Thickness
C451S006000, C451S303000, C451S041000, C438S690000
Reexamination Certificate
active
06621584
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to the field of semiconductor substrate processing and, more particularly, to the monitoring of material being removed during chemical-mechanical polishing of a semiconductor substrate.
The manufacture of an integrated circuit device requires the formation of various layers (both conductive, semiconductive, and non-conductive) above a base substrate to form necessary components and interconnects. During the manufacturing process, removal of a certain layer or portions of a layer must be achieved in order to planarize or in order to form the various components and interconnects. Chemical-mechanical polishing (CMP) is being extensively pursued to planarize a surface of a semiconductor substrate, such as a silicon substrate, at various stages of integrated circuit processing. It is also used in polishing optical surfaces, metrology samples, micro-machinery, and various metal and semiconductor based substrates.
CMP is a technique in which a polishing agent is used along with a polishing pad to polish away materials on a semiconductor substrate. The mechanical movement of the pad relative to the substrate, in combination with the chemical reaction of the polishing agent, provides an abrasive force with chemical erosion to planarize the exposed surface of the substrate (or a layer formed on the substrate).
In the most common method of performing CMP, a rotating wafer holder supports a wafer, and a polishing pad rotates relative to the wafer surface. The wafer holder presses the wafer surface against the polishing pad during the planarization process and rotates the wafer about a first axis relative to the polishing pad (see, for example, U.S. Pat. No. 5,329,732). The mechanical force for polishing is derived from the speed of the polishing pad rotating about a second axis different from the first and the downward force of the wafer holder. A polishing agent is constantly transferred under the wafer holder, and rotation of the wafer holder aids in polishing agent delivery and averages out local variations across the substrate surface. Since the polishing rate applied to the wafer surface is proportional to the relative velocity between the substrate and the polishing pad, the polish rate at a selected point on the wafer surface depends upon the distance of the selected point from the two primary axes of rotation—that of the wafer holder and that of the polish pad. This results in a non-uniform velocity profile across the surface of the substrate, and therefore, in a non-uniform polish. Additionally, it is generally accepted by those experienced in the art of CMP that a higher relative velocity between the wafer and the polish pad is desired for superior planarization performance (see, for example, Stell et al., in “Advanced Metallization for Devices and Circuits—Science, Technology and Manufacturability” ed. S. P. Murarka, A. Katz, K. N. Tu and K. Maex, pg 151). However, a higher average relative velocity in this configuration leads to a less desirable velocity profile across the surface of the substrate, and therefore, poor uniformity of polish.
This problem is solved by using a linear polisher. In a linear polisher, instead of a rotating pad, a belt moves a pad linearly across the substrate surface to provide a more uniform velocity profile across the surface of the substrate. The substrate is still rotated for averaging out local variations as with a rotating polisher. Unlike rotating polishers, however, linear polishers result in a uniform polishing rate across the substrate surface throughout the CMP process for uniformly polishing the substrate.
Additionally, linear polishers are capable of using flexible belts, upon which the pad is disposed. This flexibility allows the belt to flex, which can cause a change in the pad pressure being exerted on the substrate. A fluid bearing formed by a stationary platen can be utilized to control the pad pressure being exerted on a substrate at various locations along the substrate surface, thus controlling the profile of the polishing rate across the substrate surface.
Linear polishers are described in a patent application titled “Control of Chemical-Mechanical Polishing Rate Across A Substrate Surface;” Ser. No. 08/638,464; filed Apr. 26, 1996 and in a patent application titled “Linear Polisher and Method for Semiconductor Wafer Planarization;” Ser. No. 08/759,172; filed Dec. 3, 1996. Fluid bearings are described in a patent application titled “Control Of Chemical-Mechanical Polishing Rate Across A Substrate Surface For A Linear Polisher;” Ser. No. 08/638,462; filed Apr. 26, 1996 and in U.S. Pat. No. 5,558,568.
Rotating CMP systems have been designed to incorporate various in-situ monitoring techniques. For example, U.S. Pat. No. 5,081,421 describes an in-situ monitoring technique where the detection is accomplished by means of capacitively measuring the thickness of the dielectric layer on a conductive substrate. U.S. Pat. Nos. 5,240,552 and 5,439,551 describe techniques where acoustic waves from the substrate are used to determine end point. U.S. Pat. No. 5,597,442 describes a technique where the end point is detected by monitoring the temperature of the polishing pad with an infrared temperature measuring device. U.S. Pat. No. 5,595,526 describes a technique where a quantity approximately proportional to a share of the total energy consumed by the polisher is used to determine end point. U.S. Pat. Nos. 5,413,941, 5,433,651 and European Patent Application No. EP 0 738 561 A1 describe optical methods for determining end point.
In U.S. Pat. No. 5,413,941, a laser light impinges onto an area of the substrate at an angle greater than 70° from a line normal to the substrate, the impinged laser light predominantly reflecting off the area as opposed to transmitting through. The intensity of the reflected light is used as a measure of a change in degree of planarity of the substrate as a result of polishing. In U.S. Pat. No. 5,433,651, the rotating polishing table has a window embedded in it, which is flush with the table as opposed to the polishing pad. As the table rotates, the window passes over an in-situ monitor, which takes a reflectance measurement indicative of the end point of the polishing process. In European Patent Application No. EP 0 738 561 A1, the rotating polishing table has a window embedded in it, which, unlike the one in the ′651 patent, is flush with or formed from the polishing pad. A laser interferometer is used as the window passes over an in-situ monitor to determine the end point of the polishing process.
A linear polisher capable of in-situ monitoring for end point detection using a laser interferometer is described in U.S. patent application Ser. No. 08/869,655, assigned to the assignee of the present application.
Laser interferometry, however, has some inherent disadvantages. First, it measures absolute intensity of light emitting from an overlying substrate layer, and is dependent upon the material being polished. Second, in laser interferometry the operator cannot directly determine whether the film thickness being measured by the incident light is actually the desired finished thickness or some integer multiple thereof.
Additionally, an inherent limitation of these end point detection monitoring systems is that one has to analyze the interference curve and fit it to a reasonable approximation. Thus, depending upon the wavelength used and the film properties, there is a finite amount of removal (2000-4000 Å) before the interference curve can be fitted to a reasonable amount of accuracy. Further, using a single wavelength can, at best, only provide the removal rate, and based on the removal rate and prior knowledge of the initial thickness of the oxide, one can estimate the residual thickness of the oxide. Usually in a production fab, the initial thickness of the dielectric varies within the control limits of the deposition/growth process. Therefore, the assumption of a particular initial thickness of oxide will create at least an error equivalent
Chadda Saket
Jairath Rahul
Krusell Wilbur C.
Pecen Jiri
Brinks Hofer Gilson & Lione
Lam Research Corporation
Pham Hoa Q.
LandOfFree
Method and apparatus for in-situ monitoring of thickness... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for in-situ monitoring of thickness..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for in-situ monitoring of thickness... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3011362