KR20020040913A - Method for preventing damage to wafers in a sequential multiple steps polishing process - Google Patents

Abstract

In the present invention, the method of preventing the drying of residues on the wafer surface in the multi-stage polishing operation increases the wafer processing efficiency, thereby extending the useful life of the polishing pad used in the CMP operation. The method includes detecting a polishing endpoint for the polished wafer in a second polishing step subsequent to the first polishing step for the wafer. Endpoint detection terminates polishing of another wafer in the first polishing operation and triggers overpolishing of the preceding wafer in the second polishing operation. After a set period, overpolishing of the previous wafer in the second polishing operation ends. Finally, each wafer moves to a subsequent processing operation, which may include a polishing operation or a buffing operation. One of the advantages of the present invention is that the total processing time is increased for the same polishing pad to maximize the total platen pad usage time between platen pad exchanges, thereby increasing the number of wafers between platen pad exchanges. To enable polishing.

Description

METHOD FOR PREVENTING DAMAGE TO WAFERS IN A SEQUENTIAL MULTIPLE STEPS POLISHING PROCESS}

The semiconductor industry manufactures a large number of devices for use in many products available to consumers. These devices include microprocessors fabricated on semiconductor substrate materials. The substrate material consists of silicon produced in a thin section, which in most cases is commonly referred to as a wafer. Various electronic circuit components are fabricated on these wafers, which are divided into dies or chips, each containing many circuit components.

At various points in the manufacturing process, the wafer may require processing to provide a flat surface to locate the electronic circuit components of the semiconductor device. This treatment should provide a high degree of control in producing the desired substrate planar surface without damaging the wafer surface. Various polishing methods are used to obtain a flat substrate surface. Chemical mechanical polishing (CMP) is one of the widely used methods in the industry for wafer processing.

CMP is performed on a polishing device having one or more platens for holding a polishing pad and a plurality of heads for holding a wafer. The polishing pad includes a particulate slurry that polishes the wafer held against the pad by the polishing head as the platen and the pad rotate. When a wafer is polished in multiple platen CMP devices with multiple polishing heads, such as the device available from Applied Material Mirra, Santa Clara, Calif., The entire polishing sequence of this particular wafer is performed on the same polishing head. However, this head is rotated to a different polishing platen. The first portion of the polishing sequence is performed on one platen ("platen-1"). When the first portion of the polishing sequence is completed, the head and wafer move to the second polishing platen (platen-2) for the second portion of the polishing sequence. While the original wafer is in platen-2 receiving the second portion of the polishing sequence, the second polishing head holding the second wafer performs the first portion of the polishing sequence of platen-1. The first head ("head-1") completes the second portion of the polishing sequence at platen-2 and the second head ("head-2") completes the first portion of the polishing sequence at platen-1 At one time, the head-1 moves to platen-3, the head-2 moves to platen-2, and the head changes back the platen so that head-3 moves to platen-1. In a conventional polishing device configuration, polishing (oxide removal) is not performed on platen-3. Instead, polishing occurs only on platens-1 and 2, and platen-3 is used as a post-CMP buffing operation ("buff") to remove chemical residues. .

CMP operation is controlled using an optical laser and an optical radiation detector located at the polishing platen, and the endpoint of polishing is detected. Transparent windows are embedded within the platen surface for radiation transmission. The polishing pad of each platen has a matching embedded pad ("windowed pad") of material that enables the transmission of laser radiation. The window pads are aligned on the platen, allowing radiation to be transmitted through the platen window and the pad window. The aligned platen window and pad window are collectively referred to as the "endpoint window." As the platen rotates, the endpoint window encounters the wafer once per revolution, allowing the rotation to pass through the window, reflect off the wafer and back through the window to the detector.

During polishing of a transparent film coated on an opaque substrate (such as silicon dioxide on silicon), the intensity of radiation at the detector indicates periodicity when the film is removed from the surface. This periodicity is determined by the equation d = λ / (2n cos θ), where d is the distance through the film between peak maxima, n is the refractive index of the film with respect to the emission wavelength, θ is the collection angle, λ is the emission wavelength. Graphing the emission intensity versus polishing time yields an interferometery curve, which can be used for polishing endpoint detection. This curve is called an "endpoint trace." The signature of the endpoint trace is indicative of the film thickness and the polishing rate. From the relationship between trace signature, polishing rate, and film thickness to be provided, the polishing “end point” can be determined at or near a particular node of the trace.

The endpoint traces of each polishing platen visited by a single wafer during processing can be electronically combined into a virtual single trace for the wafer. If the buffing step is performed at Platen-3 as described above, only the endpoint traces from Platen-1 and 2 are collected and combined. In a two step polishing sequence, the polishing time for the wafer is divided into approximately equal parts, about half of the total polishing time performed on platen-1 and about half of the total polishing time performed on platen-2. This division is accomplished by one of the methods described below.

In single platen endpoint mode, endpoint traces are collected from platen-1 and platen-2, both merged into one virtual trace. However, the trace is used to trigger the polishing endpoint only on the second platen. The polishing time on Platen-1 is fixed at approximately half of the expected polishing time based on the polishing rate and the film thickness to be removed. The polishing time on platen-2 is then determined by optical endpoint analysis.

In dual platen endpoint mode, endpoint traces are collected from platen-1 and platen-2, both of which are combined into one virtual trace. The trace is used to trigger the polishing endpoint on both polishing platens. The polishing time on the first platen is determined by the detection of a particular feature in the trace that is determined to be approximately midway between the polishing time, based on the relationship of the endpoint trace and the thickness to be removed. The polishing time on platen-2 is again determined by optical endpoint analysis.

When part of the polishing sequence for a particular platen is completed, the head is lifted from the platen, the wafer is held by the vacuum to the head, and the polishing pad until the sequence for all other stations in the device is complete. Is rich on. For example, when the polishing endpoint reaches the wafer on platen-2, the head lifts the wafer from the platen and holds the wafer on the platen until the polishing sequence for the other head is also completed. All heads meet their part of the polishing sequence, and when all the wafers are lifted from the platen, the heads all rotate to the next station of the polishing sequence. While the wafer is floating on the polishing platen, some slurry residue may be present on the wafer surface. If the slurry residues are dried during this idle time after polishing on platen-2 or buffing on pleton-3, this may cause deleterious scratching of the wafer surface during subsequent processing operations.

The introduction of water mist into the head and wafer regions during idle time will prevent slurry drying on the wafer surface. However, adding moisture to the polishing system can affect the polishing performance and worsen the process.

Summary of the Invention

The present invention is directed to a method and system that meets the above and other needs associated with preventing wafer damage in a sequential multistep polishing process by preventing the polishing slurry from drying on the wafer surface between polishing steps. The invention is illustrated in many implementations and applications, some of which are summarized below.

According to one aspect of the present invention, a method of preventing drying of residues on a wafer surface in a sequential multistep polishing operation is disclosed. The method includes detecting a polishing endpoint for the polished first wafer in a second polishing operation subsequent to a first polishing operation of the first wafer. Detecting an endpoint terminates polishing of the second wafer in the first polishing operation, and then overpolishing the first wafer in the second polishing operation. After a predetermined time period, overpolishing of the first wafer in the second polishing operation ends. Finally, each aper may include a buffing operation or an additional polishing operation, or simply move to a subsequent wafer processing operation that may simply continue the wafer processing line. In applications related to this, the first wafer is polished and moved to the second polishing position before both wafers are polished simultaneously in each of the two polishing positions. The polishing operation at the first position ends as a function of time when the endpoint reaches the first wafer to be polished at the second position.

According to another aspect of the present invention, a method of extending a wafer polishing pad useful for a plurality of polishing pad devices is disclosed. The method includes providing at least first and second polishing pads, polishing the second wafer with the first polishing pad, and polishing the first wafer with the second polishing pad. Then, the polishing of the second wafer by the first polishing pad is terminated when the polishing endpoint for the first wafer is detected. The first wafer is then overpolished with a second polishing pat, after which the overpolishing operation ends. Each wafer is then moved to a subsequent wafer processing operation. In a related embodiment, the cumulative wear time of two or more pads is tracked to determine the time for exchange.

The above summary of the present invention does not describe each illustrated embodiment or every implementation of the present invention. The drawings and the detailed description will more particularly exemplify these embodiments.

The invention will be more fully understood upon consideration of the following detailed description of various embodiments of the invention in conjunction with the accompanying drawings.

TECHNICAL FIELD The present invention relates generally to semiconductor devices and their fabrication, and more particularly, to a method of preventing damage to a wafer in a polishing process.

1 shows an example of two platen polishing systems comprising two heads and two polishing pads in accordance with an embodiment of the present invention;

2 is a flow chart showing how the polishing pad is controlled to prevent drying of residue on the wafer, but rather to extend the life of the polishing pad of the polishing device;

3A is a summary table for the processing of a batch of 24 wafers using a single platen endpoint mode and fixed polishing time for initial polishing operation, FIG.

FIG. 3B illustrates a batch of 24 wafers in accordance with an exemplary embodiment of the present invention. FIG.

The present invention may be modified in various modifications and alternative forms, the details of which will be illustrated in the drawings and described in detail. It should be understood, however, that the intention is not to limit the invention to the particular embodiments described. Rather, the invention includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined in the appended claims.

The present invention relates generally to a method and system for preventing drying of residues on a wafer surface in a polishing operation. The present invention is also suitable for extending the useful life of a plurality of polishing pads used simultaneously for polishing a plurality of wafers. Although the present invention is not necessarily limited to wafer processing applications, the present invention will be better understood using the discussion of exemplary embodiments in this particular situation.

In an exemplary embodiment, a method of preventing drying of residues on a wafer surface in a multi-stage polishing device and increasing wafer yield is described herein. In particular, in a polishing device comprising three polishing platens, namely platen-1, platen-2 and third buffing platen (platen-3), the polishing endpoint detection program is carried out at the polishing step on platen-2. After is completed, it is deformed to prevent idle time for the head and wafer located on platen-2. When the polishing endpoint is detected on Platen-2, for example by analysis of endpoint interference measurement curves, it simultaneously triggers events on Platen-1 and Platen-2. Polishing of the wafer on Platen-2 is terminated and a short duration overpolishing step on Platen-2 is triggered. Overpolishing after detection of the polishing endpoint compensates for with-wafer polishing rate non-uniformity. The overpolishing duration allows sufficient time for the head and wafer at platen-1 to retrace from the polishing pad. In this way, the wafer does not float on platen-2 during idle time while waiting for the completion of a fixed polishing time on platen-1. Thus, slurry drying on the wafer surface due to idle time before the buffing step is minimized to reduce scratching of the wafer on buffing platen-3. There is no platen-2 endpoint detection for the first wafer of the batch polished in platen-1. However, the polishing time on Platen-1 is simply fixed at approximately half of the expected total polishing time based on the polishing rate and the amount of film to be removed from each wafer.

Referring now to the drawings, FIG. 1 shows a polishing system 100 arranged in accordance with one embodiment of the present invention. In this example, the system 100 includes two polishing pads 102A and 104A located on each other platens 102B and 104B having a polishing slurry 106 positioned thereon. Located on each pad is a polishing head 108 and 112 that holds a pair of wafers 110 and 114 to be polished. Platen 104B and pad 104A are aligned and have an embedded endpoint window 116 that is selectively coupled to device 118. Device 118 is coupled to both polishing heads 108 and 112 in communication with computer device 124 that controls their up and down movement towards the polishing pad. In another embodiment, an optional endpoint window 122 and corresponding device 124 are provided for detecting the endpoint more precisely when polishing a first wafer of a batch. Once the endpoint is detected at the first wafer 114 of the head 112, this information is sent to the computer 120 via the device 118 and back to the head 108 to send the signal from the polishing pad 102A to the wafer. Is communicated to pull 110.

2, a flowchart 200 illustrates an example flow of operations for polishing a plurality of wafers in accordance with an embodiment of the present invention. In step 202, a batch of wafers is provided to the polishing device to initiate a CMP polishing operation. In step 204, the first wafer is placed in a first polishing position (eg, head 108) for a predetermined time. In this example, the time is set to about half of the total expected polishing time for the wafer. In step 206, the second wafer is polished at the first polishing position and the first wafer is moved to the second polishing position for polishing. At this point, both wafers are polished simultaneously in both locations. In step 208, polishing of the second wafer ends as a function of time when the polishing endpoint of the first wafer is detected at the first polishing location.

The assembly at the first polishing position is lowered, the wafer is lifted from the polishing pad, and the second polishing position performs an overpolishing operation on the first wafer for a predetermined time. In this example, the predetermined time is about 5 seconds, but this time is typically calculated as 5% of the total polishing time on both platens. Once both polishing heads are lifted from the platen, the wafer is moved to another wafer processing operation as shown in step 212. In this example, the first wafer is moved to the third platen for buffing, while the second wafer is moved to the second polishing position and the third wafer is moved to the first polishing position. Once all the wafers in the batch have completed their polishing sequence, they are placed back in the cassette and moved to the next operation of the wafer processing line in step 214.

In another embodiment of the present invention, a method is provided for extending the useful life of a polishing pad on a polishing device. This is done by reducing the difference in total polishing time between the two polishing platens, thereby reducing the difference in pad wear between the platens. After a certain amount of wear, pad performance degrades and the polishing rate drops, resulting in excessive with-wafer nonuniformity.

The dual-platen endpoint mode described above uses optical endpoint analysis to control polishing on both platen-1 and platen-2. Likewise, the single platen endpoint mode described above sets the polishing time on Polishing-1 to approximately half of the total polishing time, and the polishing time on Polish Platen-2 is determined by optical endpoint analysis. In order that there is no idle time after polishing on platen-2, the polishing time on platen-1 has a fixed polishing duration significantly less than the expected total polishing time. This approach to time causes more pads on platen-2 to wear out than pads on platen-1, so that the performance of platen-2 pads degrades faster if the polishing time is divided evenly between the two pads. Be sure to Thus, the platen-2 pads require a small number of wafers to be changed after polishing, and due to the significant equipment down time required to change and modify the polishing pads, the pads on both platens Is changed at the same time. If only the pads on Platen-2 have reached the end of their life, both Platen-1 and Platen-2 have to be replaced, the potential availability of Platen-1 pads is lost. The end result is more frequent pad changes and higher pad utilization rates for the same number of processed wafers.

The present invention controls the wear rate for various polishing pads by creating a correlation between the polishing pads. Instead of polishing the wafer at the first polishing position for a fixed time, the polishing time is controlled by detection of the endpoint of the wafer at the second or nearest polishing position. In an exemplary embodiment, a CMP device with four heads and four platens was used to process two 24 wafer batches. The wafer requires a polishing time sufficient to remove an oxide film of 4850 ms. The polishing endpoint is from each wafer with an overpolishing duration on Platen-2 of 5% of the total polishing time of Platen-1 and Platen-2 (e.g., platens 102B and 104 of FIG. 1). It was set to remove an oxide film of 4650 kPa. The polishing rate of head 1 (eg, head 108) is about 40 kW / sec, the polishing rate of head 2 (eg, head 112) is about 39 kW / sec, and head 3 (not shown) ), The polishing rate is about 42 kW / sec, and the head 4 (not shown) is about 40 kW / sec.

FIG. 3A provides a summary table of CMP processing of a batch of 24 wafers using a single platen endpoint mode having a first polishing step (fixed time period) of about 55 seconds and a second polishing step of optical endpoint plus overpolishing of 5 seconds. . The difference in average polishing time per wafer of Platen-1 to Platen-2 is 10.25 seconds (65.25 seconds versus 55.00 seconds). The cumulative difference in pad wear time from FIG. 3A is 246 seconds. In single platen endpoint mode, a fixed time of 55 seconds is selected for platen-1. Increasing platen-1 time to 56 seconds results in undesirable idle time on platen-2. Reducing the platen-1 time smaller reduces production and increases the difference in pad wear between platen-1 and platen-2.

In this example, both platen polishing operations have approximately the same duration until the polishing operation is performed by platen-2. 3B provides another summary table for CMP processing of batches of 24 wafers using the endpoint mode of the present invention. Endpoint mode includes overpolishing on Platen-2 for 5 seconds. The difference in average polishing time per wafer for Platen-1 vs. Platen-2 is now 7.25 seconds (63.75 seconds versus 56.50 seconds). The cumulative difference in pad wear time from FIG. 3B is now 174 seconds, improved by 72 seconds for the single platen endpoint mode shown in FIG. 3A.

Typically, 300 wafers are processed between pad changes. In the processing of wafers containing mixtures of high-volume and low-volume ASIC chips, lot sizes vary from as little as 4 wafers per lot to as much as 24 wafers per lot. Most lots have 24 wafers. 3A and 3B show that polishing a 24 wafer lot in accordance with the present invention reduces the pad wear time difference by approximately 72 seconds and reduces the total wear time on platen-2 by 36 seconds. Thus, the overall wear time for processing 12 lots of 24 wafers was 432 seconds on the Platen-2 pad. The reduced wear time is enough time to process an additional 6 wafer lot that requires 384 seconds wear time on Platen-2 before the pad is changed. As a result, if the total platen pad use time between platen pad exchanges is specified in terms of total processing time, not the total number of wafers processed, the present invention provides an additional six wafers between pad exchanges for every 12 lots of 24 wafers processed. Enables polishing of the lot.

Although the present invention has been described with reference to some specific exemplary embodiments, those skilled in the art will recognize that many changes may be made without departing from the spirit and scope of the invention as set forth in the appended claims.

Claims (20)

A method of preventing drying of residues on a wafer surface with a sequential multiple step polishing operation, Detecting a polishing endpoint for the first wafer 114 polished in a second polishing operation following a first polishing operation of the first wafer 114; Terminating polishing of the second wafer 110 in the first polishing operation and overpolishing the first wafer in the second polishing operation; Terminating overpolishing of the first wafer 114 in the second polishing operation; Moving each of the wafers 114, 110 to a subsequent wafer processing operation Residual drying prevention method on the wafer surface comprising a. The method of claim 1, Polishing the second wafer 110 in the first polishing step and polishing the first wafer 114 to the endpoint in the second polishing step have approximately the same duration to prevent residue drying on the wafer surface. . The method of claim 1, Detecting the polishing endpoint comprises determining a film thickness on the surface of the first wafer (114). The method of claim 1, Detecting the polishing endpoint comprises analyzing an endpoint trace from a detection device (118,120). The method of claim 1, Detecting the polishing endpoint comprises analyzing an interferometry curve from a detection device (118, 120). The method of claim 1, Detecting the polishing endpoint comprises analyzing a laser beam reflected from the surface of the first wafer (114). The method of claim 1, Overpolishing the first wafer (114) in the second polishing step occurs during a selected time period. The method of claim 7, wherein Overpolishing the first wafer 114 for a selected time period includes a duration sufficient to prepare the second wafer 110 in the first polishing step to be moved to a subsequent polishing operation. A method for preventing residue drying on the wafer surface. The method of claim 1, Moving each of the wafers (114,110) to subsequent operations comprises a subsequent polishing step. The method of claim 1, Moving each of the wafers (114,110) to subsequent operations comprises a subsequent buffing operation. The method of claim 1, Terminating polishing of the second wafer 110 in the first polishing operation is a residue on the wafer surface that is a function of time when the polishing endpoint for the first wafer 114 is detected in the second polishing operation. How to avoid drying. A method of preventing drying of residues on a wafer surface in a multi-step polishing operation, Polishing the first wafer 114 at the second polishing position simultaneously with polishing the second wafer 110 at the first polishing position, Terminating polishing of the second wafer 110 as a function of time at the first polishing location when a polishing endpoint for the first wafer 114 is detected at the second polishing location; Overpolishing the first wafer 114 for a predetermined time period at the second polishing location Residual drying prevention method on the wafer surface comprising a. The method of claim 12, Prior to said simultaneously polishing, further comprising polishing said first wafer (114) at said first polishing location for a predetermined period of time. A system for preventing drying of residues on a wafer surface with a sequential multiple stage polishing process, Means (108, 112, 120) for polishing the second wafer (110) at the first polishing position (108) and simultaneously polishing the first wafer (114) at the second polishing position (112); Terminating polishing of the second wafer 110 at the first polishing location 108 as a function of time when a polishing endpoint for the first wafer 114 is detected at the second polishing location 112. Means 120, Means 112 for overpolishing the first wafer 114 at the second polishing location 112 for a predetermined time period. Residual drying prevention system on the wafer surface comprising a. A method of extending the wafer polishing pad useful life of a plurality of polishing pad devices, the method comprising: Providing at least first and second polishing pads 104A, 102A; Polishing the second wafer 110 with the first polishing pad 104A, polishing the first wafer 114 with the second polishing pad 102A, and When the polishing endpoint for the second wafer 110 is detected, polishing of the second wafer 110 by the first polishing pad 104A is terminated, and the first polishing pad 102A is used to terminate the polishing. Overpolishing the wafer 114, Terminating overpolishing of the first wafer by the second polishing pad 102A; Moving each of the wafers 110, 114 to a subsequent wafer processing operation Method for extending the wafer polishing pad useful life comprising a. The method of claim 15, Polishing the first wafer (114) with the first polishing pad (104A) for a predetermined period of time. The method of claim 16, And over-polishing with the second polishing pad (102A) for a predetermined period of time. The method of claim 17, Polishing with the first polishing pad (104A) and polishing with the second polishing pad (102A) to an endpoint have approximately the same duration. The method of claim 15, Tracking a cumulative wear time difference between at least the two polishing pads (102A, 104A). The method of claim 15, Detecting the polishing endpoint comprises analyzing endpoint traces from the detection device (120).