DETERMINATION OF WAFER ANGULAR POSITION FOR IN-SEQUENCE METROLOGY

Abstract

A polishing apparatus includes a carrier head configured to hold a wafer in a first plane, the wafer having a perimeter and a fiducial, a drive shaft having an axis perpendicular to the first plane and configured to rotate the carrier head about the axis, a light source configured to direct light onto an outer face of the wafer at a position adjacent the perimeter of the wafer; a detector configured to detect the light collected from the wafer while the drive shaft rotates the carrier head and the wafer; and a controller configured to receive a first signal indicating an angular position of the drive shaft and receive a second signal from the detector, the controller configured to determine based on the first signal and the second signal an angular position of the fiducial with respect the carrier head.

Description

TECHNICAL FIELD

This disclosure relates to rotational alignment of a wafer for in-sequence metrology in a chemical mechanical polishing (CMP) system.

BACKGROUND

An integrated circuit is typically formed on a substrate by the sequential deposition of conductive, semiconductive, or insulative layers on a silicon wafer. One fabrication step involves depositing a filler layer over a non-planar surface and planarizing the filler layer. For certain applications, the filler layer is planarized until the top surface of a patterned layer is exposed. A conductive filler layer, for example, can be deposited on a patterned insulative layer to fill the trenches or holes in the insulative layer. After planarization, the portions of the metallic layer remaining between the raised pattern of the insulative layer form vias, plugs, and lines that provide conductive paths between thin film circuits on the substrate. For other applications, such as oxide polishing, the filler layer is planarized until a predetermined thickness is left over the non-planar surface. In addition, planarization of the substrate surface is usually required for photolithography.

Chemical mechanical polishing (CMP) is one accepted method of planarization. This planarization method typically requires that the substrate be mounted on a carrier or polishing head. The exposed surface of the substrate is typically placed against a rotating polishing pad. The carrier head provides a controllable load on the substrate to push it against the polishing pad. An abrasive polishing slurry is typically supplied to the surface of the polishing pad.

Variations in the slurry distribution, the polishing pad condition, the relative speed between the polishing pad and the substrate, and the load on the substrate can cause variations in the material removal rate. These variations, as well as variations in the initial thickness of the substrate layer, cause variations in the time needed to reach the polishing endpoint. Therefore, determining the polishing endpoint merely as a function of polishing time can lead to overpolishing or underpolishing of the substrate. Various in-situ monitoring techniques, such as optical or eddy current monitoring, can be used to detect a polishing endpoint.

SUMMARY

In some systems, a spectrum from a substrate is measured by an in-sequence metrology station. That is, the spectrum can be measured while the substrate is still held by the carrier head, but at a metrology station positioned between polishing stations or between a transfer station and a polishing station. Due to slippage during polishing, the rotational orientation of the substrate relative to the carrier head may not be known. The light beam that is used to measure the spectrum can also be used to find the fiducial of the wafer, permitting the spectrum to be matched to a position on the substrate.

In one aspect, a polishing apparatus includes a carrier head configured to hold a wafer in a first plane with an inner face of the wafer abutting a support structure of the carrier head, the wafer having a perimeter and a fiducial defined by removal of a portion of the wafer or by having an optically distinct marking at a specific angular position. The polishing apparatus includes a drive shaft having an axis perpendicular to the first plane, the drive shaft being connected to the carrier head and configured to rotate the carrier head about the axis. The polishing apparatus includes a light source configured to direct light onto an outer face of the wafer at a position adjacent the perimeter of the wafer. The polishing apparatus includes a detector configured to detect the light collected from the wafer while the drive shaft rotates the carrier head and the wafer; and a controller configured to receive a first signal indicating an angular position of the drive shaft and receive a second signal from the detector, the controller configured to determine based on the first signal and the second signal an angular position of the fiducial with respect the carrier head.

Implementations may include one or more of the following features. The detector is a spectrometer and the second signal includes a spectrum. The controller computes an average intensity from the spectrum and matches the average intensity to the angular position of the wafer. The controller computes using an algorithm the angular position of the head at which the fiducial is directly over the detector. The algorithm is a valley-finding algorithm. The drive shaft is configured to rotate at a first angular rate and the detected light is configured to be collected at a second rate. The first angular rate is matched to the second rate so that continuous detection of light is made along positions adjacent the perimeter of the wafer. The controller is configured to match spectral data to specific spatial positions on the wafer after the position of the fiducial is determined. The fiducial is a notch. The fiducial is a flat.

In one aspect, a method includes holding a wafer in a first plane using a carrier head, an inner face of the wafer abutting a support structure of the carrier head. The method includes rotating the carrier head along an axis perpendicular to the first plane using a drive shaft, directing light from a light source onto an outer face of the wafer at a position adjacent a perimeter of the wafer, detecting light collected from the wafer while the drive shaft rotates the carrier head and the wafer. The method includes receiving, at a controller, a first signal indicating an angular position of the drive shaft, receiving, at the controller, a second signal from the detector, and determining a position of the fiducial with respect to the angular position of the carrier head, based on the first signal and the second signal.

Implementations may include one or more of the following features. The method further includes using the controller to compute an average intensity from a spectrum. The detector is a spectrometer and the second signal comprises the spectrum. The method further includes matching the average intensity to the angular position of the wafer, computing the angular position of the carrier head, using an algorithm, at which the fiducial is directly over the detector. The drive shaft rotates at a first angular rate and the detected light is collected at a second rate. The method includes making continuous detection of light along positions adjacent the perimeter of the wafer by matching the first angular rate to the second rate. The method further includes matching spectral data to specific spatial positions on the wafer after the position of the fiducial is determined. The spectral data is collected before the position of the fiducial is determined. The spectral data is collected after the position of the fiducial is determined. The method further includes generating a third signal by matching the spectral data to the specific spatial positions, and feeding the third signal forward to a downstream polishing station configured to polish the wafer. The method further includes generating a third signal by matching the spectral data to the specific spatial positions, and feeding the third signal backward to an upstream polishing station configured to polish a subsequent wafer. Implementations may include one or more of the following potential advantages. Better control of the polishing operation can be obtained when the measured spectrum is matched to wafer coordinates from which measurement light for the spectrum was acquired. Within-wafer non-uniformity (WIWNU) and wafer-to-wafer non-uniformity (WTWNU) can be reduced.

The details of one or more implementations are set forth in the accompanying drawings and the description below. Other aspects, features and advantages will be apparent from the description and drawings, and from the claims.

DESCRIPTION OF DRAWINGS

FIG. 1 is a schematic plan view of an example of a polishing apparatus.

FIG. 2 is a schematic cross-sectional view of an example of a polishing apparatus.

FIG. 3 is a schematic cross-sectional view of an example of an in-sequence optical metrology system.

FIG. 4 illustrates another implementation of a polishing apparatus.

FIG. 5 illustrates an example spectrum

FIGS. 6A-6D show different substrates and methods of rotationally aligning a substrate.

FIG. 7 is a schematic cross-sectional view of a wet-process optical metrology system.

FIG. 8 is a schematic cross-sectional view of another implementation of a wet-process optical metrology system.

FIG. 9 is a schematic top view of a substrate.

Like reference symbols in the various drawings indicate like elements.

DETAILED DESCRIPTION

As integrated circuits continue to develop, line widths continue to shrink and layers in the integrated circuit continue to accumulate, requiring ever more stringent thickness control. Thus, polishing process control techniques, whether utilizing in-situ monitoring or run-to run process control, face challenges to maintain keep the post-polishing thickness within specification.

For example, when performing in-situ spectrographic monitoring of a multi-layer product substrate, an incident optical beam from the spectrographic monitoring system can penetrate several dielectric layers before being reflected by metal lines. The reflected beam can thus be a result of the thickness and critical dimensions of multiple layers. A spectrum resulting from such a complex layer stack often presents a significant challenge in determining the thickness of the outermost layer that is being polishing. In addition, the outermost layer thickness is an indirect parameter for process control. This is because in many applications the metal line thickness—a parameter that may be more critical to yield—can vary even if the outermost layer thickness is on target, if other dimensions such as etch depth or critical dimension vary.

A control scheme for determining a polishing endpoint incorporates wet metrology between CMP steps and feedforward or feedback control. The dimensional variations in the substrate are captured after each polishing step at an in-sequence metrology station and used either to determine whether there is a need to rework the substrate, or fed forward or fed back to control the polishing operation or endpoint at a previous or subsequent polishing station.

The polishing apparatus is configured such that a carrier head holds a substrate during polishing at the first and second polishing stations and moves the substrate from the first polishing station to the second polishing station. The in-sequence metrology station is situated to measure the substrate when the carrier head is holding the substrate and when the substrate is not in contact with a polishing pad of either the first polishing station or the second polishing station.

FIG. 1 is a plan view of a chemical mechanical polishing apparatus 100 for processing one or more substrates. The polishing apparatus 100 includes a polishing platform 106 that at least partially supports and houses a plurality of polishing stations 124. The number of polishing stations can be an even number equal to or greater than four. For example, the polishing apparatus can include four polishing stations 124a, 124b, 124c and 124d. Each polishing station 124 is adapted to polish a substrate that is retained in a carrier head 126.

The polishing apparatus 100 also includes a multiplicity of carrier heads 126, each of which is configured to carry a substrate. The number of carrier heads can be an even number equal to or greater than the number of polishing stations, e.g., four carrier heads or six carrier heads. For example, the number of carrier heads can be two greater than the number of polishing stations. This permits loading and unloading of substrates to be performed from two of the carrier heads while polishing occurs with the other carrier heads at the remainder of the polishing stations, thereby providing improved throughput.

The polishing apparatus 100 also includes a transfer station 122 for loading and unloading substrates from the carrier heads. The transfer station 122 can include a plurality of load cups 123, e.g., two load cups 123a, 123b, adapted to facilitate transfer of a substrate between the carrier heads 126 and a factory interface (not shown) or other device (not shown) by a transfer robot 110. The load cups 123 generally facilitate transfer between the robot 110 and each of the carrier heads 126.

The stations of the polishing apparatus 100, including the transfer station 122 and the polishing stations 124, can be positioned at substantially equal angular intervals around the center of the platform 106. This is not required, but can provide the polishing apparatus with a good footprint.

Each polishing station 124 includes a polishing pad 130 supported on a platen 120 (see FIG. 2). The polishing pad 130 can be a two-layer polishing pad with an outer polishing layer 130a and a softer backing layer 130b (see FIG. 2).

For a polishing operation, one carrier head 126 is positioned at each polishing station. Two additional carrier heads can be positioned in the loading and unloading station 122 to exchange polished substrates for unpolished substrates while the other substrates are being polished at the polishing stations 124.

The carrier heads 126 are held by a support structure that can cause each carrier head to move along a path that passes, in order, the first polishing station 124a, the second polishing station 124b, the third polishing station 124c, and the fourth polishing station 126d. This permits each carrier head to be selectively positioned over the polishing stations 124 and the load cups 123.

In some implementations, each carrier head 126 is coupled to a carriage 108 that is mounted to an overhead track 128. By moving a carriage 108 along the overhead track 128, the carrier head 126 can be positioned over a selected polishing station 124 or load cup 123. A carrier head 126 that moves along the track will traverse the path past each of the polishing stations.

In the implementation depicted in FIG. 1, the overhead track 128 has a circular configuration (shown in phantom) which allows the carriages 108 retaining the carrier heads 126 to be selectively orbited over and/or clear of the load cups 122 and the polishing stations 124. The overhead track 128 may have other configurations including elliptical, oval, linear or other suitable orientation. Alternatively, in some implementations the carrier heads 126 are suspended from a carousel, and rotation of the carousel moves all of the carrier heads simultaneously along a circular path.

Each polishing station 124 of the polishing apparatus 100 can include a port, e.g., at the end of an arm 134, to dispense polishing liquid 136 (see FIG. 2), such as abrasive slurry, onto the polishing pad 130. Each polishing station 124 of the polishing apparatus 100 can also include pad conditioning apparatus 132 to abrade the polishing pad 130 to maintain the polishing pad 130 in a consistent abrasive state.

As shown in FIG. 2, the platen 120 at each polishing station 124 is operable to rotate about an axis 121. For example, a motor 150 can turn a drive shaft 152 to rotate the platen 120.

Each carrier head 126 is operable to hold a substrate 10 against the polishing pad 130. Each carrier head 126 can have independent control of the polishing parameters, for example pressure, associated with each respective substrate. In particular, each carrier head 126 can include a retaining ring 142 to retain the substrate 10 below a flexible membrane 144. The substrate 10 has an inner face 101 abutting a support structure such as the retaining ring 142. Each carrier head 126 also includes a plurality of independently controllable pressurizable chambers defined by the membrane, e.g., three chambers 146a-146c, which can apply independently controllable pressurizes to associated zones on the flexible membrane 144 and thus on the substrate 10. Although only three chambers are illustrated in FIG. 2 for ease of illustration, there could be one or two chambers, or four or more chambers, e.g., five chambers.

Each carrier head 126 is suspended from the track 128, and is connected by a drive shaft 154 to a carrier head rotation motor 156 so that the carrier head can rotate about an axis 127. Optionally each carrier head 126 can oscillate laterally, e.g., by driving the carriage 108 on the track 128, or by rotational oscillation of the carousel itself. In operation, the platen is rotated about its central axis 121, and each carrier head is rotated about its central axis 127 and translated laterally across the top surface of the polishing pad. The lateral sweep is in a direction parallel to the polishing surface 212. The lateral sweep can be a linear or arcuate motion.

A controller 190, such as a programmable computer, is connected to each motor 152, 156 to independently control the rotation rate of the platen 120 and the carrier heads 126. For example, each motor can include an encoder that measures the angular position or rotation rate of the associated drive shaft. The associated drive shafts can have respective reference angular positions that are recognized by the encoder to measure the number of revolutions of the drive shafts. Similarly, the controller 190 is connected to an actuator in each carriage 108 to independently control the lateral motion of each carrier head 126. For example, each actuator can include a linear encoder that measures the position of the carriage 108 along the track 128.

The controller 190 can include a central processing unit (CPU) 192, a memory 194, and support circuits 196, e.g., input/output circuitry, power supplies, clock circuits, cache, and the like. The memory is connected to the CPU 192. The memory is a non-transitory computable readable medium, and can be one or more readily available memory such as random access memory (RAM), read only memory (ROM), floppy disk, hard disk, or other form of digital storage. In addition, although illustrated as a single computer, the controller 190 could be a distributed system, e.g., including multiple independently operating processors and memories.

This architecture is adaptable to various polishing situations based on programming of the controller 190 to control the order and timing that the carrier heads are positioned at the polishing stations.

For example, some polishing recipes are complex and require three of four polishing steps. Thus, a mode of operation is for the controller to cause a substrate to be loaded into a carrier head 126 at one of the load cups 123, and for the carrier head 126 to be positioned in turn at each polishing station 124a, 124b, 124c, 124d so that the substrate is polished at each polishing station in sequence. After polishing at the last station, the carrier head 126 is returned to one of the load cups 123 and the substrate is unloaded from the carrier head 126.

An advantage of this mode of operation is that it can provide high throughput at a reasonable footprint of the base 106, while avoiding problems such as coordinating endpoint control and cross-contamination that can occur when multiple substrates are polished on the same polishing pad.

Referring to FIGS. 1, and 3, the polishing apparatus 100 can also include one or more in-sequence (also referred to as in-line) metrology systems 160 (see FIG. 3), e.g., optical metrology systems, e.g., spectrographic metrology systems. An in-sequence metrology system is positioned within the polishing apparatus 100, but does not performs measurements during the polishing operation; rather measurements are collected between polishing operations, e.g., while the substrate is being moved from one polishing station to another. Alternatively, one or more of the in-sequence metrology systems 160 could be a non-optical metrology system, e.g., an eddy current metrology system or capacitive metrology system.

In some implementations, the polishing system includes two in-sequence metrology systems. The two in-sequence metrology systems could be on the path on opposite sides of a polishing station. For example, in some implementations (shown in FIG. 1) the polishing system 100 includes a first metrology system with a first probe 180a located between the third polishing station 124c and the fourth polishing station 124d, and a second metrology system with a second probe 180b located between the fourth polishing station 124d and the transfer station 122. As another example, in some implementations (shown in FIG. 4) the polishing system 100 includes a first metrology system with a first probe 180a located between the transfer station 122 and the first polishing station 124a, and a second metrology system with a second probe 180b located between the first polishing station 124a and the second polishing station 124b.

Each in-line metrology system 160 includes a probe 180 supported on the platform 106 at a position on the path taken by the carrier heads 126 and between two of the stations, e.g., between two polishing stations 124, or between a polishing station 124 and the transfer station stations 122. In particular, the probe 180 is located at a position such that a carrier head 126 supported by the track 128 can position the substrate 10 over the probe 180.

In some modes of operation, the substrate is measured at an in-sequence metrology station 160 before polishing at a station. In this case, in some implementations, the probe 180 of the metrology station 160 can be positioned on the path after the polishing station. Thus, the carrier head 126 with an attached substrate is moved along the path past the polishing station 124 to the probe 180 of the in-sequence monitoring station, the substrate is measured with the probe 180, and the carrier head is moved back along the path (in a reverse direction) to the polishing station 124.

In some modes of operation, the substrate is measured an in-sequence metrology station 160 after polishing at a station. In this case, in some implementations, the probe 180 of the metrology station 160 can be positioned on the path before the polishing station. Thus, the carrier head 126 with an attached substrate is moved along the path past the probe 180 of the in-sequence monitoring station to the polishing station 124, the substrate is polished at the polishing station 124, the carrier head is moved back along the path (in a reverse direction) to the probe 180, the substrate is measured, and the carrier head is forward again along the path past the polishing station 124 to the next station.

In some implementations, the probe 180 of the metrology station 160 can be positioned on the path after the polishing station and be used for a measurement after polishing of the substrate at the polishing station. For example, in the implementations shown in FIG. 1, the first probe 180a and second probe 180b can be used for measuring the second substrate and first substrate after polishing at the third polishing station 124c and fourth polishing station 124d, respectively.

In some implementations, the probe 180 of the metrology station 160 can be positioned on the path before the polishing station and be used for a measurement before polishing of the substrate at the polishing station.

Referring to FIG. 4, in some implementations, the polishing system 100 includes four in-sequence metrology stations. For example, the polishing system 100 can include a first probe 180a between the second load cup 123b and the first polishing station 124a, a second probe 180b between the first polishing station 124a and the second polishing station 124b, a third probe 180b between the third polishing station 124c and the fourth polishing station 124d, and fourth probe 180d between the fourth polishing station 124d and the first load cup 123a.

An advantage of having two (or four) in-sequence metrology stations 160 is that measurements can be performed simultaneously on the two substrates. However, the techniques of moving a carrier head backward on the path to a probe or a polishing station can be applied even if there is only one in-sequence metrology station. In addition, although this examples focus on a polishing system with four polishing stations, the techniques can be applied to nearly any system with multiple polishing stations.

For example, a polishing system could include the four platens as shown in FIG. 1, but only a single in-sequence metrology station, e.g., with the probe positioned between the third polishing station 124c and the fourth polishing station 124d. In this case, for a measurement before the second polishing step, the first substrate would be measured with the probe and then move forward along the path to the fourth polishing station 124d, whereas the third substrate would be measured with the probe and then move backward along the path to the third polishing station 124c.

As another example, a polishing system could include the four platens as shown in FIG. 1, but only a single in-sequence metrology station, e.g., with the probe positioned between the first polishing station 124a and the second polishing station 124b. In this case, for a measurement after the first polishing step, the first substrate would move backwards from the second polishing station 124b to the probe, be measured with the probe and then move forward along the path to the fourth polishing station 124d, whereas the third substrate would move forward from the first polishing station 124a, be measured with the probe and then move forward to the third polishing station 124c.

In some implementations, the probe 180 should be positioned adjacent a station at which the filler layer is expected to be cleared. For example, where the controller 190 is configured with a recipe to perform bulk polishing (but not clearance) of the filler layer at the first and second polishing stations, and removal or clearing of an underlying layer at the third and fourth polishing stations, the probe 180 can be positioned adjacent either the third or fourth polishing stations.

In operation, the metrology station 161 could simply be used to measure the substrate between polishing operations at the first station 124a and the second polishing station 124b. However, the backtracking approach discussed above can also be applied. For example, a carrier heads could be moved back along the track 128 after polishing at the second station 124b to measure the substrate at the station 161, and then the carrier head 126 can be transported forward to the third station 124b. As another example, a carrier head could be moved forward along the track past the first station 124a prior to polishing at that station, the substrate could be measured at the metrology station 161, and then the carrier head can be transported back along the track 128 to the first station 124a.

Although only one probe 180a is illustrated in FIG. 3, the metrology station 160 could include two probes for two separate in-sequence metrology systems to permit two substrates to be measured simultaneously at the metrology station 160. In addition, the metrology station 160 could be positioned between the second station 124b and the third station 124c, with appropriate modification of the order of transfer between the stations.

The optical metrology system 160 can include a light source 162, a light detector 164, and circuitry 166 for sending and receiving signals between the controller 190 and the light source 162 and light detector 164.

One or more optical fibers can be used to transmit the light from the light source 162 to the optical access in the polishing pad, and to transmit light reflected from the substrate 10 to the detector 164. For example, a bifurcated optical fiber 170 can be used to transmit the light from the light source 162 to the substrate 10 and back to the detector 164. The bifurcated optical fiber can include a trunk 172 having an end in the probe 180 to measure the substrate 10, and two branches 174 and 176 connected to the light source 162 and detector 164, respectively. In some implementations, rather than a bifurcated fiber, two adjacent optical fibers can be used.

In some implementations, the probe 180 holds an end of the trunk 172 of the bifurcated fiber. In operation, the carrier head 126 positions a substrate 10 over the probe 180. Light from the light source 162 is emitted from the end of the trunk 172, reflected by an outer face 102 of the substrate 10 back into the trunk 172, and the reflected light is received by the detector 164. In some implementations, one or more other optical elements, e.g., a focusing lens, are positioned over the end of the trunk 172, but these may not be necessary.

The probe 180 can include a mechanism to adjust the vertical height of the end the trunk 172, e.g., the vertical distance between the end of the trunk 172 and the top surface of the platform 106. In some implementations, the probe 180 is supported on an actuator system 182 that is configured to move the probe 180 laterally in a plane parallel to the plane of the track 128. The actuator system 182 can be an XY actuator system that includes two independent linear actuators to move probe 180 independently along two orthogonal axes.

The output of the circuitry 166 can be a digital electronic signal that passes to the controller 190 for the optical metrology system. Similarly, the light source 162 can be turned on or off in response to control commands in digital electronic signals that pass from the controller 190 to the optical metrology system 160. Alternatively, the circuitry 166 could communicate with the controller 190 by a wireless signal.

The light source 162 can be operable to emit white light. In one implementation, the white light emitted includes light having wavelengths of 200-800 nanometers. A suitable light source is a xenon lamp or a xenon mercury lamp.

The light detector 164 can be a spectrometer. A spectrometer is an optical instrument for measuring intensity of light over a portion of the electromagnetic spectrum. A suitable spectrometer is a grating spectrometer. Typical output for a spectrometer is the intensity of the light as a function of wavelength (or frequency). FIG. 5 illustrates an example of a measured spectrum 300.

As noted above, the light source 162 and the light detector 164 can be connected to a computing device, e.g., the controller 190, operable to control their operation and receive their signals. The computing device can include a microprocessor situated near the polishing apparatus, e.g., a programmable computer. With respect to control, the computing device can, for example, synchronize activation of the light source with the motion of the carrier head 126.

For some process control algorithms, it would be useful to know the positions of the spectra relative to the orientation of the dies on the substrate. For example, matching corresponding substrate coordinates from which the probe 180 acquired light for the measured spectrum improves the accuracy of control algorithm because information regarding the state of the substrate layer at different spatial locations on the substrate 10 are known with greater precision.

The substrate 10 to be polished generally includes a fiducial that allows the substrate 10 to be angularly oriented, for example, by a specific angle from the fiducial. The fiducial is generally defined by removal of a portion of the substrate. For example, the fiducial may be a notch 310 in the substrate (shown in FIG. 6A) or the fiducial may be a flat 320 (shown in FIG. 6B). Other than the fiducial, the substrate 10 can be a circular disk. Alternatively, the fiducial can be defined by some region of the substrate being optically distinguishable from the rest via a marking or lack of patterning in the region. In general, the fiducial can be any element that is distinct from the pattern of the dies themselves.

As described above, the inner face 12 of the substrate abuts a support structure of the carrier head 126, such as the flexible membrane 144. In addition, the outer edge 14 of the substrate can abut the inner surface of the retaining ring 142 of the carrier head 126.

Because the substrate 10 is held by the carrier head 126 and transported along the track 128, the geometric (i.e., radial) distance between a center 103 of the substrate 10 and the probe 180 can be directly determined based on the encoder measurements in conjunction with the known dimensions of the components of the polishing system.

The encoder for the drive shaft 156 outputs a signal indicating the angular position of some arbitrary (but fixed) point on the drive shaft. Once the angular position of the carrier is known when the fiducial is detected, the angular position of other spectral measurements on the substrate can be determined relative to the fiducial. In conjunction with the radial positions, this gives coordinate positions of the measurements on the substrate in a frame of reference that is repeatable between different wafers and/or different steps of polishing of the same wafer.

After polishing at each polishing station, the angular position of the substrate with respect to the carrier head can change, e.g., due to slippage between the substrate and the flexible membrane or other support structure of the carrier head, it is preferable to detect the fiducial for each measurement. The determination of the angular position of the substrate can be carried out either before or after measurements are made by the probe 180 It is advantageous for detection of the angular position of the substrate to be made without additional dedicated instruments, such as, for example, a notch aligner. For example, the footprint of the metrology stations can be reduced when the notch aligner is not used. In addition, a notch aligner may not be conveniently deployed without making significant modifications to the system because the substrate is held in the carrier head and the carrier head blocks measuring beams of the notch aligner from reaching the substrate.

A spectrograph or spectrometer used to collect spectra can be simultaneously used for determining the angular position of the substrate. A first step in rotational alignment involves determining the position of the fiducial with respect to the angular reference position 370 of the drive shaft 156 of the carrier head 126. The carrier head 126 is first moved to a position such that light from the light source 162 relayed by the probe 180 is directed onto a region 331 on an outer face 102 of the substrate 10 adjacent a perimeter 312 of the substrate. FIG. 6D is the view of the outer face 103 of the substrate 10 when viewed from the polishing pad towards the carrier head 126. Since the probe 180 relays light in both directions—the probing light from the light source 162 to the substrate 10 and also collects detected light from the substrate 10 back to the detector 164, the region 331 is also above the light detector 164 which is connected to the probe 180.

The drive shaft 154 receives a signal 191 from the controller 190 to rotate at a first angular rate, for example, in a counter-clockwise direction 322. The first angular rate should be matched to the data acquisition rate of the light detector 164. For example, a diameter 351 of light emanating from probe 180 and striking region 331 may be 100 microns. The first angular rate may be, for example 0.05 rad/s, and the substrate may have a diameter of 200 mm. If the data acquisition rate of the light detector is 100 Hz, every acquired spectrum of the light detector will detect light from regions that are spaced 100 micron apart, forming a continuous measurement spots 331, 332, and 333 on the substrate, without leaving space between the spots. The measurement spot 332 will be rotated into the position of region 331, which is directly above the probe 180 at the next (i.e., second) data acquisition period of the light detector. Subsequently, the measurement spot 333 will be rotated into the position of region 331, at the third data acquisition period of the light detector. The angular rate should be chosen such that there is no area at the edge of the substrate left unmeasured by the optical probe. One also may choose an angular rate lower than necessary to achieve such coverage in order to oversample such a region, which may be of benefit if the spot size of the optical probe is larger than the size of the fiducial feature.

When the substrate is rotated to an angle such that the fiducial is directly above the light detector 164, the light detector 164 will detect a change in optical characteristics in the spectrum. Such an optical characteristic may include the intensity of the light recorded by the spectrometer. Information regarding the angular position of the drive shaft is fed to the controller 190 as a first signal. The spectrometer also output the measured spectrum as a second signal to the controller 190. The controller 190 can determine an average intensity of the spectrum from the signal for spectra collected at each angular position of the driver shaft. An algorithm, for example, a valley-finding algorithm, can be used to determine the angular position of the carrier head at which the fiducial would be directly over the probe 180. This then determines an angular offset 365 between the angular reference position 370 of the drive shaft 156 head and the angular position of the fiducial. FIG. 6C illustrates this by showing a view of the inner face 101 of the substrate 10. Also shown is the drive shaft 156 and the angular reference position 370 on the drive shaft, not shown in the drawing is the head carrier connected to the drive shaft 156, and the membrane between the head carrier and the inner face 101.

In this scheme, no additional fiducial detection hardware is required. The spectra collected by the spectrometer during the rotational alignment process can be stored in the controller 190 so that data from these spectra can be matched to specific spatial positions on the wafer once the position of the fiducial is determined. There is no limitation on whether the spectral data is collected before or after the position of the fiducial is determined. Spectral data from the wafer at positions not collected during the rotational alignment process can be collected after the fiducial position is determined. Alternatively, the coordinates from which stored spectral data were collected can be back-calculated once the fiducial position had been determined. This helps to prevent the rotational alignment process from impacting the measurement throughput of the instrument. The use of such rotational alignment techniques is not limited to metrology systems in an in-sequence polishing station. It can be used in any optical metrology system where spectral acquisition occurs on a short time scale.

Optionally, the in-sequence metrology system 160 can be a wet metrology system. In a wet-metrology system, measurement of the surface of the substrate is conducted while a layer of liquid covers the portion of the surface being measured. An advantage of wet metrology is that the liquid can have a similar index of refraction as the optical fiber 170. The liquid can provide a homogeneous medium through which light can travel to and from the surface of the film that is to be or that has been polished. The wet metrology system 169 can be configured such that the liquid is flowing during the measurement. A flowing liquid can flush away polishing residue, e.g., slurry, from the surface of the substrate being measured.

FIG. 7 shows an implementation of a wet in-sequence metrology system 160. In this implementation, the trunk 172 of the optical fiber 170 is situated inside a tube 186. A liquid 188, e.g., de-ionized water, can be pumped from a liquid source 189 into and through the tube 186. During the measurement, the substrate 10 can positioned over the end of the optical fiber 170. The height of the substrate 10 relative to the top of the tube 186 and the flow rate of the liquid 188 is selected such that as the liquid 188 overflows the tube 186, the liquid 188 fills the space between the end of the optical fiber 170 and the substrate 10.

Alternatively, as shown in FIG. 8, the carrier head 126 can be lowered into a reservoir defined by a housing 189. Thus, the substrate 10 and a portion of the carrier head 126 can be submerged in a liquid 188, e.g., de-ionized water, in the reservoir. The end of the optical fiber 170 can be submerged in the liquid 188 below the substrate 10.

In either case, in operation, light travels from the light source 162, travels through the liquid 188 to the surface of the substrate 10, is reflected from the surface of the substrate 10, enters the end of the optical fiber, and returns to the detector 164.

Referring to FIG. 9, a typical substrate 10 includes multiple dies 12. In some implementations, the controller 190 causes the substrate 10 and the probe 180 to undergo relative motion so that the optical metrology system 160 can make multiple measurements within an area 18 on the substrate 10. In particular, the optical metrology system 160 can take multiple measurements at spots 184 (only one spot is shown on FIG. 9 for clarity) that are spread out with a substantially uniform density over the area 18. The area 18 can be equivalent to the area of a die 12. In some implementations, the die 12 (and the area 18) can be considered to include half of any adjacent scribe line. In some implementations, at least one-hundred measurements are made within the area 18. For example, if a die is 1 cm on a side, then the measurements can be made at 1 mm intervals across the area. The edges of the area 18 need not be aligned with the edges of a particular die 12 on the substrate.

In some implementations, the XY actuator system 182 causes the measurement spot 184 of the probe 180 to traverse a path across the area 18 on the substrate 10 while the carrier head 126 holds the substrate 10 in a fixed position (relative to the platform 106). For example, the XY actuator system 182 can cause the measurement spot 184 to traverse a path which traverses the area 18 on a plurality of evenly spaced parallel line segments. This permits the optical metrology system 160 to take measurements that are evenly spaced over the area 18.

In some implementations, there is no actuator system 182, and the probe 180 remains stationary (relative to the platform 106) while the carrier head 126 moves to cause the measurement spot 184 to traverse the area 18. For example, the carrier head could undergo a combination of rotation (from motor 156) translation (from carriage 108 moving along track 128) to cause the measurement spot 184 to traverse the area 18. For example, the carrier head 126 can rotate while carriage 108 causes the center of the substrate to move outwardly from the probe 180, which causes the measurement spot 184 to traverse a spiral path on the substrate 10. By making measurements while the spot 184 is over the area 18, measurements can be made at a substantially uniform density over the area 18.

In some implementations, the relative motion is caused by a combination of motion of the carrier head 126 and motion of the probe 180, e.g., rotation of the carrier head 126 and linear translation of the probe 180.

The controller 190 receives a signal from the optical metrology system 160 that carries information describing a spectrum of the light received by the light detector for each flash of the light source or time frame of the detector. For each measured spectrum, a characterizing value can be calculated from the measured spectrum. The characterizing value can be used in controlling a polishing operation at one or more of the polishing stations.

One technique to calculate a characterizing value is, for each measured spectrum, to identify a matching reference spectrum from a library of reference spectra. Each reference spectrum in the library can have an associated characterizing value, e.g., a thickness value or an index value indicating the time or number of platen rotations at which the reference spectrum is expected to occur. By determining the associated characterizing value for the matching reference spectrum, a characterizing value can be generated. This technique is described in U.S. Patent Publication No. 2010-0217430, which is incorporated by reference. Another technique is to analyze a characteristic of a spectral feature from the measured spectrum, e.g., a wavelength or width of a peak or valley in the measured spectrum. The wavelength or width value of the feature from the measured spectrum provides the characterizing value. This technique is described in U.S. Patent Publication No. 2011-0256805, which is incorporated by reference. Another technique is to fit an optical model to the measured spectrum. In particular, a parameter of the optical model is optimized to provide the best fit of the model to the measured spectrum. The parameter value generated for the measured spectrum generates the characterizing value. This technique is described in U.S. Patent Application No. 61/608,284, filed Mar. 8, 2012, which is incorporated by reference. Another technique is to perform a Fourier transform of the measured spectrum. A position of one of the peaks from the transformed spectrum is measured. The position value generated for measured spectrum generates the characterizing value. This technique is described in U.S. patent application Ser. No. 13/454,002, filed Apr. 23, 2012, which is incorporated by reference.

As noted above, the characterizing value can be used in controlling a polishing operation at one or more of the polishing stations. The controller can, for example, calculate the characterizing value and adjust the polishing time, polishing pressure, or polishing endpoint of: (i) the previous polishing step, i.e., for a subsequent substrate at the polishing station that the substrate being measured just left, (ii) the subsequent polishing step, i.e., at the polishing station to which the substrate being measured will be transferred, or (iii) both of items (i) and (ii), based on the characterizing value.

In some implementations, prior to the first CMP step, substrate dimension information (layer thickness, critical dimensions) from upstream non-polishing steps, if available, is fed forward to the controller 190.

After a CMP step, the substrate is measured using wet metrology at the in-sequence metrology station 180 located between the polishing station at which the substrate was polishing and the next polishing station. A characterizing value, e.g., layer thickness or copper line critical dimension, is captured and sent to the controller.

In some implementations, the controller 190 uses the characterizing value to adjust the polishing operation for the substrate at the next polishing station. For example, if the characterizing value indicates that the etch trench depth is greater, the post thickness target for the subsequent polishing station can be adjusted with more removal amount to keep the remaining metal line thickness constant. If the characterizing value indicates that the underlying layer thickness has changed, the reference spectrum for in-situ endpoint detection at the subsequent polishing station can be modified so that endpoint occurs closer to the target metal line thickness.

In some implementations, the controller 190 uses the characterizing value to adjust the polishing operation for a subsequent substrate at the previous polishing station. For example, if the characterizing value indicates that the etch trench depth is greater, the post thickness target for the previous polishing station can be adjusted with more removal amount to keep the remaining metal line thickness constant. If the characterizing value indicates that the underlying layer thickness has changed, the reference spectrum for in-situ endpoint detection at the previous polishing station can be modified so that endpoint occurs closer to the target metal line thickness.

In some implementations, the controller 190 analyzes the measured spectra and determines the proper substrate route. For example, the controller 190 can compare the characterizing value to a threshold, or determine whether the characterizing value falls within a predetermined range. If the characterizing value indicates that polishing is incomplete, e.g., if it falls within the predetermined range indicating an underpolished substrate or does not exceed a threshold indicating a satisfactorily polished substrate, then the substrate can be routed back to previous polishing station for rework. For example, Once the rework is completed, the substrate can be measured again at the metrology station, or transported to the next polishing station. If the characterizing value does not indicate that polishing is incomplete, the substrate can be transported to the next polishing station.

For example, a parameter such as metal residue can be measured using wet metrology at the in-sequence metrology station 180. If metal residue detected, the substrate can be routed back to previous polishing station for rework. Otherwise, the substrate can be transported to the next polishing station.

In order to detect metal residue, the controller 190 can evaluate the percentage of the area that is covered by the filler material. Each measured spectrum 300 is compared to a reference spectrum. The reference spectrum can be the spectrum from a thick layer of the filler material, e.g., a spectrum from a metal, e.g., a copper or tungsten reference spectrum. The comparison generates a similarity value for each measured spectrum 300. A single scalar value representing the amount of filler material within the area 18 can be calculated from the similarity values, e.g., by averaging the similarity values. The scalar value can then be compared to a threshold to determine the presence and/or amount of residue in the area.

In some implementations, the similarity value is calculated from a sum of squared differences between the measured spectrum and the reference spectrum. In some implementations, the similarity value is calculated from a cross-correlation between the measured spectrum and the reference spectrum.

For example, in some implementation a sum of squared differences (SSD) between each measured spectrum and the reference spectrum is calculated to generate an SSD value for each measurement spot. The SSD values can then be normalized by dividing all SSD values by the highest SSD value obtained in the scan to generate normalized SSD values (so that the highest SSD value is equal to 1). The normalized SSD values are then subtracted from 1 to generate the similarity value. The spectrum that had the highest SSD value, and thus the smallest copper contribution, is now equal to 0.

Then the average of all similarity values generated in the prior step is calculated to generate the scalar value. This scalar value will be higher if residue is present.

As another example, in some implementation a sum of squared differences (SSD) between each measured spectrum and the reference spectrum is calculated to generate an SSD value for each measurement spot. The SSD values can then be normalized by dividing all SSD values by the highest SSD value obtained in the scan to generate normalized SSD values (so that the highest SSD value is equal to 1). The normalized SSD values are then subtracted from 1 to generate inverted normalized SSD values. For a given spectrum, if the inverted normalized SSD value generated in the previous step is less than a user-defined threshold, then it is set to 0. The user-defined threshold can be 0.5 to 0.8, e.g., 0.7. Then the average of all values generated in the prior step is calculated to generate the scalar value. Again, this similarity value will be higher if residue is present.

If the calculated scalar value is greater than a threshold value, then the controller 190 can designate the substrate as having residue. On the other hand, if the scalar value is equal or less than the threshold value, then the controller 190 can designate the substrate as not having residue.

If the controller 190 does not designate the substrate as having residue, then the controller can cause the substrate to be processed at the next polishing station normally. On the other hand, controller 190 designates the substrate as having residue, then the controller can take a variety of actions. In some implementations, the substrate can be returned immediately to the previous polishing station for rework. In some implementations, the substrate is returned to the cassette (without being processed at a subsequent polishing station) and designated for rework once other substrates in the queue have completed polishing. In some implementations, the substrate is returned to the cassette (without being processed at a subsequent polishing station), and an entry for the substrate in a tracking database is generated to indicate that the substrate has residue. In some implementations, the scalar value can be used to adjust a subsequent polishing operation to ensure complete removal of the residue. In some implementations, the scalar value can be used to flag the operator that something has gone wrong in the polishing process, and that the operator's attention is required. The tool can enter into a number of error/alarm states, e.g. return all substrates to a cassette and await operator intervention.

In another implementation, the calculated similarity value for each measurement value is compared to a threshold value. Based on the comparison, each measurement spot is designated as either filler material or not filler material. For example, if an inverted normalized SSD value is generated for each measurement spot as discussed above, then the user-defined threshold can be 0.5 to 0.8, e.g., 0.7.

The percentage of measurement spots within the area 18 that are designated as filler material can be calculated. For example, the number of measurement spots designated as filler material can be divided by the total number of measurement spots.

This calculated percentage can be compared to a threshold percentage. The threshold percentage can be calculated either from knowledge of pattern of the die on the substrate, or empirically by measuring (using the measurement process described above) for a sample substrate that is known to not have residue. The sample substrate could be verified as not having residue by a dedicated metrology station.

If the calculated percentage is greater than the threshold percentage, then the substrate can be designated as having residue. On the other hand, if the percentage is equal or less than the threshold percentage, then the substrate can be designated as not having residue. The controller 190 can then take action as discussed above.

In some implementations a probe 180′ of an optical metrology system 160 is positioned between the loading and unloading station and one of the polishing stations. If the probe 180′ is positioned between the loading station and the first polishing station, then a characterizing value can be measured by the metrology system and fed forward to adjust polishing of the substrate at first polishing station. If the probe 180′ is positioned between the last polishing station and the unloading station, then a characterizing value can be measured by the metrology system and fed back to adjust polishing of a subsequent substrate at the last polishing station, or if residue is detected then the substrate can be sent back to the last polishing station for rework.

The control schemes described above can more reliably maintain product substrates within manufacture specification, and can reduce rework, and can provide rerouting of the substrate to provide rework with less disruption of throughput. This can provide an improvement in both productivity and yield performance.

The above described polishing apparatus and methods can be applied in a variety of polishing systems. For example, rather than be suspended from a track, multiple carrier heads can be suspended from a carousel, and lateral motion of the carrier heads can be provided by a carriage that is suspend from and can move relative to the carousel. The platen may orbit rather than rotate. The polishing pad can be a circular (or some other shape) pad secured to the platen. Some aspects of the endpoint detection system may be applicable to linear polishing systems (e.g., where the polishing pad is a continuous or a reel-to-reel belt that moves linearly). The polishing layer can be a standard (for example, polyurethane with or without fillers) polishing material, a soft material, or a fixed-abrasive material. Terms of relative positioning are used; it should be understood that the polishing surface and substrate can be held in a vertical orientation or some other orientations.

While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Claims

1. A polishing apparatus, comprising

a carrier head configured to hold a wafer in a first plane with an inner face of the wafer abutting a support structure of the carrier head, the wafer having a perimeter and a fiducial defined by removal of a portion of the wafer or by having an optically distinct marking at a specific angular position;

a drive shaft having an axis perpendicular to the first plane, the drive shaft being connected to the carrier head and configured to rotate the carrier head about the axis;

a light source configured to direct light onto an outer face of the wafer at a position adjacent the perimeter of the wafer;

a detector configured to detect the light collected from the wafer while the drive shaft rotates the carrier head and the wafer; and

a controller configured to receive a first signal indicating an angular position of the drive shaft and receive a second signal from the detector, the controller configured to determine based on the first signal and the second signal an angular position of the fiducial with respect the carrier head.

2. The polishing apparatus of claim 1, wherein the detector is a spectrometer and the second signal comprises a spectrum.

3. The polishing apparatus of claim 2, wherein the controller computes an average intensity from the spectrum and matches the average intensity to the angular position of the wafer.

4. The polishing apparatus of claim 1, wherein the controller computes using an algorithm the angular position of the head at which the fiducial is directly over the detector.

5. The polishing apparatus of claim 4, wherein the algorithm is a valley-finding algorithm.

6. The polishing apparatus of claim 1, wherein the drive shaft is configured to rotate at a first angular rate and the detected light is configured to be collected at a second rate.

7. The apparatus of claim 6, wherein the first angular rate is matched to the second rate so that continuous detection of light is made along positions adjacent the perimeter of the wafer.

8. The apparatus of claim 1, wherein the controller is configured to match spectral data to specific spatial positions on the wafer after the position of the fiducial is determined.

9. The apparatus of claim 1, wherein the fiducial is a notch.

10. The apparatus of claim 1, wherein the fiducial is a flat.

11. A method, comprising:

holding a wafer in a first plane using a carrier head, an inner face of the wafer abutting a support structure of the carrier head,

rotating the carrier head along an axis perpendicular to the first plane using a drive shaft,

directing light from a light source onto an outer face of the wafer at a position adjacent a perimeter of the wafer,

detecting light collected from the wafer while the drive shaft rotates the carrier head and the wafer;

receiving, at a controller, a first signal indicating an angular position of the drive shaft,

receiving, at the controller, a second signal from the detector, and

determining a position of the fiducial with respect to the angular position of the carrier head, based on the first signal and the second signal.

12. The method of claim 11, further comprising using the controller to compute an average intensity from a spectrum, wherein the detector is a spectrometer and the second signal comprises the spectrum.

13. The method of claim 12, further comprising, matching the average intensity to the angular position of the wafer.

14. The method of claim 11, further comprising computing the angular position of the carrier head, using an algorithm, at which the fiducial is directly over the detector.

15. The method of claim 11, wherein the drive shaft rotates at a first angular rate and the detected light is collected at a second rate, the method further comprising making continuous detection of light along positions adjacent the perimeter of the wafer by matching the first angular rate to the second rate.

16. The method of claim 11, further comprising matching spectral data to specific spatial positions on the wafer after the position of the fiducial is determined.

17. The method of claim 16, wherein the spectral data is collected before the position of the fiducial is determined.

18. The method of claim 16, wherein the spectral data is collected after the position of the fiducial is determined.

19. The method of claim 16, further comprising generating a third signal by matching the spectral data to the specific spatial positions, and feeding the third signal forward to a downstream polishing station configured to polish the wafer.

20. The method of claim 16, further comprising generating a third signal by matching the spectral data to the specific spatial positions, and feeding the third signal backward to an upstream polishing station configured to polish a subsequent wafer.