JP2004529485A - Method and apparatus for defining automated process inspection and hierarchical substrate testing - Google Patents

Abstract

The present invention generally provides an apparatus and method for inspecting a substrate in a processing system. In one embodiment, the substrate processing inspection system includes a plurality of optical inspection systems, each configured to perform an optical inspection process at a first degree of optical resolution, and an optical inspection process at a second degree of optical resolution. An inspection platform configured to perform. Each of the plurality of optical inspection systems includes a transmitting device unit and a receiving device unit. The substrate processing inspection system further includes a controller system connected to the plurality of optical inspection systems and the inspection platform. The controller system processes (i) optical signal information indicative of a phase state on a substrate to be inspected by at least one of the plurality of optical inspection systems, and (ii) responsive to the phase state, It is configured to perform one of the substrate handling steps, the first substrate handling step further comprising transferring the substrate to an inspection platform for optical inspection.

Description

[0001]
This invention is a continuation-in-part of U.S. application Ser. is there.
[0002]
(Field of the Invention)
The present invention relates to an inspection method and apparatus. In particular, the present invention relates to a substrate inspection method and apparatus for specifying a defect or a state related to processing and handling.
[0003]
(Background of related technology)
Chip manufacturing facilities consist of a wide range of technologies. Cassettes containing semiconductor substrates are sent to various stations in the manufacturing facility.
[0004]
Semiconductor processing typically includes depositing material on a substrate and removing material from a substrate ("etching"). Typical processes include chemical vapor deposition (CVD), physical vapor deposition (PVD), electroplating, chemical mechanical polishing (CMP), etching, and the like. During processing and handling of the substrate, the substrate undergoes various structural and chemical changes. Examples of the change include a change in the thickness of the layer deposited on the substrate, the material of the layer formed on the substrate, the surface morphology, and the element pattern. These changes must be controlled in order for the elements formed on the substrate to obtain the desired electrical characteristics. For example, in the case of etching, an endpoint detection method is used to determine when the required amount of material has been removed from the substrate. More generally, successful processing requires assuring an accurate processing recipe and controlling processing excursions (eg, gas flow, temperature, pressure, electromagnetic energy, required time, etc.).
[0005]
Also, the processing environment must be sufficiently stable and free of contamination. Sources of contamination include wear due to mechanical motion, seal degradation, contaminated gases, other contaminated substrates, small debris from process chambers, nucleation of reactive gases, agglomeration during chamber pump down, plasma chamber And the formation of electric arcs. These contaminants create foreign matter that comes into contact with the substrate, resulting in defective elements. Thus, in current semiconductor manufacturing, substrates are regularly inspected for foreign matter to identify "dirty" processes and equipment.
[0006]
Also, locating the center of the substrate and arranging the substrate are necessary steps in a process for generating positional information about the substrate. In conventional systems, these steps are performed at designated locations in the processing system. Therefore, in order to execute each step, it is necessary to reciprocate the substrate to the designated position, which lowers the throughput of the system.
[0007]
Another situation that is driving processing costs is the inadequate routing of substrates within chip manufacturing facilities. Substrates are often improperly transported to processing chambers where processing conditions can cause unstable reactions, thereby damaging the substrates and processing chambers. For example, consider the case where a substrate having a photoresist layer is accidentally sent to a PVD chamber. Processing this substrate in a PVD chamber has been known to cause severe damage to the chamber and considerable repair and replacement costs. Current processing systems are not configured to prevent erroneous route designations, increasing owner costs.
[0008]
Currently, comprehensive inspection analysis of substrates for processing integrity and contamination requires that one or more substrates be periodically or constantly removed from the processing environment and sent to the inspection environment. Thus, during substrate transfer and inspection, the manufacturing flow is effectively interrupted. As a result, conventional measurement and inspection methods have greatly increased the overhead time associated with chip manufacture. In addition, since such inspection methods are only performed with periodic sampling as they have a negative effect on throughput, a large number of contaminated substrates are processed without inspection, resulting in the production of defective elements. If the substrate is redistributed from a batch, the problem becomes more severe and it becomes difficult to trace back the source of contamination.
[0009]
Thus, the integrated properties that allow the inspection of the substrate as an integral part of the processing system for selected properties, including foreign matter, spalling in processing, placement, centering, reflectivity, substrate type, discontinuities, etc. There is a need for measurement and processing systems, "watcher" devices and methods. Such an inspection is preferably performed before, during, or after substrate processing, so that the state of the substrate before and after processing is determined in real time.
[0010]
Other functions that are regularly performed in conventional processing and inspection systems include calibration of robots and inspection equipment. Current calibration methods have a negative impact on throughput because production must be stopped in order to perform the calibration. Deterioration is usually not detected until severe failure occurs. Preferably, the processing system integrates or embeds equipment that continuously monitors the status of the robot and the inspection system and facilitates automatic correction processing. In this case, the processing system can be further integrated to increase the throughput. Preferably, such an integrated device can monitor the behavior of the robot. Robot behaviors of interest include acceleration, speed, repeatability, and stability. Further, it is preferred that such an integrated device be able to determine the presence of dirt on the robot blade supporting the substrate during transfer. The presence of such contamination indicates that micro-scratching on the back surface of the substrate or accumulation of processing by-products occurred during the substrate handling process. However, it has not previously been known that such integrated devices or methods exist in processing systems.
[0011]
Another problem with conventional inspection systems is that the cost of the system is very high. Current systems are typically expensive stand-alone platforms that occupy clean room space. The large area or "footprint" required of a stand-alone inspection platform results in high costs of owning and operating such a system. Regarding foreign matter detection, the cost is further increased because an electro-optical device is used. This device is configured to detect small foreign objects at high resolution and requires a high fidelity mechanism, but such a mechanism is expensive to operate. In addition, the consideration of reduced throughput described above further increases the cost of conventional inspection systems.
[0012]
Accordingly, there is a need for an integrated system that can quickly inspect a semiconductor substrate and determine one or more states of the substrate to detect anomalies and facilitate subsequent substrate handling decisions. ing.
[0013]
(Summary of the Invention)
The present invention generally provides a substrate inspection system used in a substrate processing system. According to one aspect of the invention, at least one optical inspection system is used to inspect a surface topography of a processed substrate. The optical inspection system sends a signal representative of the surface topographic properties of the substrate to a process monitoring controller configured to operate one or more optical inspection systems in a particular process.
[0014]
In one embodiment, the process monitoring controller determines the condition of a particular substrate surface with respect to a reference substrate value. If the substrate characteristics exceed a predetermined value, the substrate is sent to a secondary measurement / inspection process for a finer and deeper analysis. Further, the process monitoring controller may use the information to optimize or change the substrate manufacturing process of the processing system.
[0015]
In another embodiment, multiple optical inspection systems are used to monitor substrate surface topography at various inspection locations located along multiple substrate transport paths with a process monitoring controller. Such a process monitoring system can optimize the throughput of the substrate processing by continuously performing monitoring and analysis. With the integrated inspection, it is possible to optimize the processing recipe by monitoring the change of the processing recipe in almost real time, and to optimize the subsequent influence on the processing.
[0016]
One embodiment of the invention includes a system for software control of process inspection and system throughput enhancement. This system includes a data processing system including a controller including a program for monitoring processing. The program upon execution of the data processing system includes configuring the optical inspection system in response to a system configuration event; adjusting settings for the optical inspection system; and transferring topography information of the substrate surface from the optical inspection system. Receiving; determining whether the substrate topography state exceeds a predetermined value; determining if the substrate requires another deeper analysis if the substrate topography state exceeds the predetermined value. It is configured to perform steps.
[0017]
One embodiment of the present invention includes a program product having a program for optical character recognition and foreign matter and defect detection, wherein the program when executed by the controller configures an optical inspection system in response to a system configuration event, Adjusting the settings for the system, from the signal source, the signal illuminates the substrate, and the signal receiver receives the reflected signal and / or the scattered signal from the substrate surface to detect the surface topography defect Providing a signal to the receiving device.
[0018]
Preferably, the program product includes substrate position information, substrate reflectance information, reflection information, spectrum information, three-dimensional image, substrate defect information, substrate damage information, foreign substance contamination information on the substrate supporting member and the substrate disposed on the supporting member. , Alphanumeric character information, robot behavior information, calibration information for any one and all of the robot, transmitter unit, and receiver unit, and any combination thereof.
[0019]
The invention briefly described above will be described more specifically with reference to embodiments shown in the accompanying drawings. However, as the invention applies to other equally effective embodiments, the accompanying drawings show only exemplary embodiments of the invention and, therefore, are not intended to limit the scope of the invention. Note that you should not consider that
[0020]
(Detailed description of preferred embodiments)
I. Processing system
Embodiments of the present invention are particularly effective in multi-chamber processing systems, such as cluster tools. One example of a multi-chamber processing system commonly used in the semiconductor industry that is suitable for supporting the detection devices described herein is known as a cluster tool. A cluster tool is a modular system with multiple chambers that perform a variety of functions including locating the center of a substrate and placing, degassing, annealing, growing, etching, etc., the substrate, or any of them. The plurality of chambers are mounted to a central transfer chamber containing a robot adapted to shuttle between the chambers. The transfer chamber is typically maintained in a vacuum and an intermediate stage for reciprocating the substrate from one chamber to another and / or a load lock chamber located at the front end of the cluster tool. I will provide a.
[0021]
FIG. 1A is a plan view of an exemplary processing system 100 for semiconductor processing in which the present invention may be used to advantage. Two such platforms are respectively Centura® and Endura®, both available from Applied Materials, Inc. of Santa Clara, California. One detail of such a stepped vacuum substrate processing system is described in Tepman et al., Entitled "Stepped Vacuum Wafer Processing System and Method," issued Feb. 16, 1993, U.S. Pat. No. 5,186. , 718. This patent is incorporated herein by reference. The exact placement and combination of the chambers will vary depending on the application performing the steps of the manufacturing method.
[0022]
In accordance with the present invention, the processing system 100 generally includes a plurality of chambers and a robot, and preferably includes a processing system controller 102 that is programmed to perform various processing methods performed within the processing system 100. Good to have. A front end environment 104 (also referred to herein as a factory interface or FI) is arranged to selectively communicate with a pair of load lock chambers 106. Pod loaders 108A, 108B located within the front end environment 104 are linear, rotatable, and vertically movable to reciprocate a substrate between the load lock 106 and a plurality of pods 105 mounted on the front end environment 104. is there. Load lock 106 provides a first vacuum interface between front end environment 104 and transfer chamber 110. Two load locks 106 are provided to increase throughput by alternately communicating with the transfer chamber 110 and the front end environment 104. Thus, while one load lock 106 is in communication with the transfer chamber 110, the second load lock 106 is in communication with the front end environment 104. A robot 113 is located in the center of the transfer chamber 110 to transfer substrates from the load lock 106 to one of the various processing chambers 114 and one of the service chambers 116. The processing chamber 114 performs any number of processes, such as physical vapor deposition, chemical vapor deposition, etching, while the service chamber 116 is adapted for degassing, placement, and cooling. A number of viewports 120 allow the inside of transfer chamber 110 to be viewed.
[0023]
Embodiments of the present invention include an optical inspection subsystem (OIS) 150 configured to collect optical data. As will be described, the OIS 150 typically includes a transmitter unit adapted to provide a signal and a reflective (brightfield illumination) and / or scattering portion of the signal (darkfield illumination) from the substrate surface. A receiver unit adapted to receive.
[0024]
A plurality of OISs 150 may be located anywhere on the processing system 100 for process monitoring, for example, inside the factory interface 104, transfer chamber 110, processing chamber 114, service chamber 116, and connected to the processing system 100. Or may be either placed or connected. The information received from each OIS 150 can then be processed to determine various states of the substrate moving within the processing system 100.
[0025]
Embodiments of the OIS 150 include various receivers and transmitters. In one embodiment, the receiving device is a charge-coupled device (CCD) camera used to detect light in the visible spectrum. Alternatively, the receiving device may be a spectrometer used to receive incident light and output data indicating various light wavelengths and their intensities. For example, red light has a higher intensity than the light component intensity grouped within the lower wavelengths of the visible spectrum. Spectrometers typically have an optical prism (or grating) interface that optically splits the incident signal into components and then projects it onto a linear CCD detector array. One embodiment of the spectrometer comprises a CCD detector array having thousands of individual detector elements and receiving a spectrum obtained from a prism (or grating). The relative intensities of the optical spectral components are representative of, for example, the substrate surface, the substrate surface pattern, or the overall color of the plasma treatment.
[0026]
In another embodiment, the receiving device is an optical character recognition receiver (OCR). OCR receivers are used to detect optical characters and substrate patterns and distinguish between them. Substrates often include specific information engraved on the surface of the substrate. Typically, the specific information is a series of alphanumeric characters. In one embodiment, the OCR receiver detects the characters to determine if a substrate having a particular OCR marking is being handled properly (eg, being sent to an appropriate chamber for processing). Used effectively. In another embodiment, OCR is used to correlate relevant measurement information.
[0027]
Each of the receiving devices includes an optical element for focusing and / or collecting the reflected signal and / or the scattered signal from the substrate. For example, a spectrometer may utilize a fiber optic collector to receive and direct the light spectrum, or a "lens perspective system" such as a "fisheye lens" that collects a large amount of light, or a specific substrate inspection. A special lens that can modify the field of view for use may be used. An array of lenses is used to increase or focus light. By collecting more light over a wider field of view, an average of local substrate surface variations is obtained.
[0028]
The transmitter of the OIS 150 may be of various configurations adapted for the particular application being processed. For example, a line light source may be used to irradiate only a part of the substrate, or a flash device may be used to emit a flash from a strobe to the substrate during a process of collecting spectral information. All of the transmitting devices may be provided with a beam focusing optical system, and may concentrate and project energy on the substrate.
[0029]
Various receiver and transmitter embodiments may be used alone or in combination throughout processing system 100 to collect processing data. The data collected from the various embodiments is used to monitor substrate processing at various stages of the manufacturing process. The operation of the OIS 150 is controlled by a process monitoring controller (PMC) 86. As shown in FIG. 1A, the PMC 86 is electrically connected to each OIS 150 by an IO (input / output) cable 90 adapted to provide command signals to the respective OIS 150. Further, the PMC 86 receives an output signal from each OIS 150 via the IO cable 90. Although the processing system controller 102 in the processing system 100 is preferably separate from the PMC 86, in one embodiment, the PMC 86 functions as a control unit for the processing system 100, thus eliminating the need for a new control unit. .
[0030]
As the user initiates operation, the PMC 86 continues to monitor boards in view of the OIS 150. When the PMC 86 detects a monitored condition (eg, a contaminated substrate), it alerts the user by displaying a warning message on a display unit (not shown). Additionally or alternatively, the processing system controller 102 instructs the PMC 86 to transfer the substrate to a specific location for final placement, cleaning, or another inspection. For example, in one embodiment, the processing system 100 has an inspection platform 135 for detecting foreign matter, but the functionality of the inspection platform 135 can be extended to include an OIS 150 located anywhere on the FI 104 or on the processing system 100. .
[0031]
The inspection platform 135 can be adapted to perform various measurement tasks such as surface abnormality detection, pattern recognition, spectral analysis, as well as foreign object detection. In the figure, the inspection platform 135 is a configured integrated foreign matter monitoring (IPM) system for an embodiment of the present invention. Generally, the IPM is an inspection platform for foreign substance detection commonly used on cluster tools. One known IPM available is Excite ™ available from Applied Materials, Inc. of Santa Clara, California. When implemented in accordance with the present invention, the IPM acts as a process monitor capable of performing any of the processes of the present invention, including any equivalent processes. Therefore, the inspection platform 135 can perform processing more than mere foreign substance detection.
[0032]
In one embodiment, it is determined whether a particular location is accessible for an OIS 150, while the particular location is determined by routing a substrate within the processing system 100 (to effectively use the robot). Is done. For example, viewport 120 may provide a view of robot blade 48 entering or exiting a chamber (eg, load lock 106, cooling chamber 116, or processing chamber 114) or moving between two locations within transfer chamber 110. As such, the viewport 120 provides a preferred location for an OIS 150. In the figure, arrows 122 and 124 indicate points where optical inspection of the substrate can be performed using port 120. In the figure, arrow 122 represents the point where robot blade 48 extends or contracts. Arrows 122 represent points where the robot blade 48 is rotated, and arrows 124 represent points where the robot blade 48 is expanded or retracted. One embodiment in which the OIS 150 is located on the transfer chamber 110 is described below with reference to FIG.
[0033]
In some embodiments, OIS 150 is used anywhere in processing system 100 where the substrate is moving. In one embodiment, the factory interface 104 adds an inspection site. In the figure, the inspection sites are represented by arrows 132, 134, 136 and 138. Arrows 132, 134, 136, 138 further represent substrate transfer paths. Arrow 136 represents substrate movement between factory interface 104 and pod 105. Arrows 132 and 138 represent substrate movement between factory interface 104 and load lock 106. Arrow 134 represents substrate movement within factory interface 104, particularly between pod loaders 108A and 108B. The OIS 150 may be located along each of the transfer paths 132, 134, 138 to monitor the manufacturing process. However, as described later, an embodiment in which the substrate under inspection does not move is also considered.
[0034]
FIG. 1B is a high-level view of the board inspection system. In one embodiment, the inspection system includes a PMC 86 connected to the processing system controller 102, a factory interface controller 159, a plurality of OISs 150, and an inspection system 135. PMC 86 is used to control the process monitoring of OIS 150. The auxiliary user interface 152 allows for stand-alone operation such that an operator can control the PMC 86. In one embodiment, interface 152 may be a graphical user interface (GUI). The obtained process monitoring information is sent to the board manufacturing data server 162 via the cable 90.
[0035]
For example, in one embodiment, OIS 150 includes a group of light sources and one receiver. FIG. 1C shows a plurality of light sources 56 and fiber optic cables 143A-143H connected to an optical signal multiplexer 141 that communicates with a receiving device 58. The optical cable is positioned to collect reflected and / or scattered light via window 120 and send the optical signal to multiplexer 141. Multiplexer 141 multiplexes the optical data and sends it to spectrum analyzer 58 so that a single receiver can effectively monitor multiple inspection sites. In one embodiment, receiving device 58 is a spectrometer.
[0036]
II. Factory interface
FIGS. 1, 2 and 4 show that the transmitter and receiver units of the OIS 150 are located on or within the FI 104 along various paths of movement of the substrate as indicated by arrows 132, 134, 136, 138. Although an embodiment is shown, other embodiments may be used effectively. FIG. 2 shows the FI 104 including the pod loaders 108A and 108B, the OIS 150A, the mounting member 160, the transfer support member 166 including the substrate alignment detector 168, and the substrate holder 162. The FI further comprises a frame 218. Mounting member 160 is secured to frame 218 by conventional means such as bolts, clamps, and other fasteners known in the art.
[0037]
FIG. 2 illustrates one embodiment of an OIS 150 for use in scanning a substrate along a process arrow 134. The OIS 150A includes a pair of light sources 56A and 56B (that is, a transmitting device) and a receiving device 58A. The light source and the receiving device are arranged on a mounting member 160. The light sources 56A and 56B are arranged at an angle that allows irradiation at two angles. In the figure, the light source 56A illuminates the front of the substrate, and the light source 56B illuminates the rear of the substrate. In one embodiment, the angle between the light sources 56A and 56B is approximately right, so that the amount of illumination overlap (ie, illumination crosstalk) is minimized. Although not shown, the light sources 56A and 56B may include a beam shaping optical element, for example, a lens or the like so that the beam provided by each light source can be miniaturized and adjusted. The light receiving device 58 is positioned to receive a reflection portion and / or a scattering portion of light from the surface of the substrate 37. The optical receiver 58 is selected according to the data to be collected. In the illustrated embodiment, the light receiving device 58 includes a CCD device, a time delay integration (TDI) camera, a photomultiplier tube (PMT), a spectrometer, an OCR camera, and the like. Optical receiver 58 may further include optical elements adapted to assist in information collection. For example, in one embodiment, the optical receiving device 58A includes an optical fiber bundle for acquiring and routing the received light from the inspection point to the receiving device 58.
[0038]
In operation, the substrate 37 is moved between two locations in the FI by the pod loaders 108A and 108B. Next, the OIS 150A scans the substrate 37 so that the light receiving device 58 receives the reflected light and / or the scattered light for the processing described below.
[0039]
In the figure, when the substrate 37 is reciprocating between the load lock 106 and the substrate holder 162, the substrate 37 is positioned and scanned by one of the pod loaders 108A and 108B below the OIS 150A. Preferably, the substrate 37 is scanned during normal substrate handling. For example, in a normal board exchange process between the pod loaders 108A and 108B, the OIS 150A may scan the board while moving the board 37.
[0040]
For example, consider a case where the substrate 37 is moving from the pod loader 108A to the pod loader 108B. First, the substrate 37 is carried by the pod loader 108A and positioned in the substrate holder 162. The substrate holder 162 facilitates the passage / exchange of the substrate between the pod loaders 108A and 108B. During this movement, the substrate 37 can be scanned by the forward irradiation light 56A. In one embodiment, the substrate moves past the midpoint and the entire substrate is scanned. 3A to 3C illustrate the movement of the substrate 37 when scanning with the forward irradiation light 56A. FIG. 3A shows the pod blade 48 immediately after the substrate 37 has begun to move toward the substrate holder 162 such that the leading edge of the substrate 37 is in the path of the signal 54. Therefore, in the drawing, the light projecting portion 53 formed on the light source 56 </ b> A, which is represented by the shaded region, crosses the front edge of the substrate 37. As shown in FIGS. 3B to 3C, while the pod blade 48 continues to move, the light projection unit 53 scans the upper surface of the substrate 37. The receiving device 58 receives the reflected light and / or the scattered light, and performs a process described later. The pod loader 108B also scans the substrate 37 using the same processing as described above.
[0041]
In another embodiment, it is contemplated that a single pod loader 108 may be used for the movement utilized by the two light sources 56A and 56B. For example, when a substrate is loaded into the substrate holder 162 by the pod loader 108A, the OIS 150A scans the substrate using the forward illuminating light source 56A and the receiving device 58. When the substrate is removed from the substrate holder 162 by the pod loader 108A, the light source 56B is activated, and rear irradiation is performed to scan the substrate when the substrate is removed from the substrate holder 162. When the substrate is placed on the substrate holder 162 and taken out by the pod loader 108A, the receiving device 58A receives reflected light and / or scattered light from the substrate.
[0042]
FIG. 4 shows another embodiment of the OIS 150B used when a substrate takes a transfer path as indicated by arrows 132,138. The OIS 150B includes a single light source 56A having beam shaping optics and an optical receiver 58. The OIS 150C is disposed on the mounting member 160. The light source 56A irradiates the substrate and may have a beam shaping optical element. In one embodiment, as the substrate 37 moves from the factory interface 104 into the load lock 106, reflected and / or scattered light is received by the receiving device 58, which receives the signal as described below. To the PMC 86 for further processing.
[0043]
5A to 5C show a scanning sequence using the OIS 150B. FIG. 5A shows the pod blade 48 immediately after the pod 180A has begun to move toward the load lock 106 such that the leading edge of the substrate is in the path of the signal 54. Therefore, in the drawing, the light projecting unit 53 represented by the shaded area crosses the front edge of the substrate 37. As shown in FIGS. 5B to 5C, while the pod blade 48 continues to move, the light projection unit 53 scans the upper surface of the substrate 37. The receiving device 58 receives the reflected light and / or the scattered light, and performs a process described later.
[0044]
As described above, the receiving device in each embodiment of the OIS 150 may select any one of a CCD device, a time delay integration (TDI) camera, a photomultiplier tube (PMT), a spectrometer, an OCR camera, and the like. In general, CCDs, TDIs, and PMTs are configured to collect image intensity changes across the substrate for reflection signature analysis and small flaking / foreign contamination associated with the variations. The spectrometer is configured to collect the light signals reflected / scattered from the surface of the substrate 37 and generate an output representing the color spectral components of the signal. The OCR is configured to collect the optical signals reflected / scattered from the surface of the substrate 37 to generate an output representing an optical character disposed on the surface of the substrate 37. In any case, the optical element is effectively used. For example, in the case of a spectrometer, the received optical signal may propagate from the collection point to the spectrometer via an optical fiber bundle. Alternatively, a diffuser lens may be coupled to the fiber optic bundle to effectively enlarge the field of view and transmit the average light spectrum of the image to the spectrometer.
[0045]
Alternatively, the receiver of OIS 150 may be a combination of detectors. For example, one embodiment of the receiving device includes a spectrometer and a CCD camera. In such a case, the spectrometer may be arranged adjacent to the CCD camera so that both devices of the spectrometer and the CCD camera can share the same field of view of the substrate 37. With such an embodiment, it is possible to perform a plurality of processing monitoring methods without prolonging the data acquisition time. The illustrated process monitoring method will be described in more detail later.
[0046]
It should be understood that the relationship between the transmitting device (ie, light source) and the receiving device is only an example. In any of the above embodiments, the positions of the receiving device and the transmitting device may be reversed. For example, with respect to FIG. 2, OIS 150A may include a pair of receivers mounted at locations where light sources 56A and 56B are shown. The light source can be mounted where the receiver 58A is shown. In such an embodiment, the receiver is positioned at a substantially oblique angle with respect to the upper surface of the substrate located directly below the light source. In operation, the receiver is positioned to receive reflected and / or scattered portions of light from the surface of the substrate 37. When the substrate 37 moves between the loaders 108A and 108B, the OIS 150A scans the substrate 37, and the reflected light and / or the scattered light are received by the receiving device and used for processing described later. The movement of the substrate 37 between the pod loaders is the same as that shown in FIGS. 3A to 3C.
[0047]
III. Transfer chamber
FIG. 6 is a perspective cutaway view of a processing system 100 of the present invention with a transfer chamber 42 and a vacuum chamber 44 mounted thereon (see FIGS. 3A-3C). The transfer chamber 42 and the vacuum chamber 44 can be selectively communicated with a sealable opening 46 by conventional equipment such as a slit valve window located above a slit valve door (not shown). The size of the opening 46 is set according to the transfer of the substrate through the opening. The robot 50 includes a blade 48 disposed at the center of the transfer chamber 42 and configured to hold the substrate 37 coupled to the robot hub 51 by a frog foot linkage 39. The robot 50 allows the blade 48 to rotate and move radially along the transfer surface, thereby moving the substrate back and forth between various locations in the system. Transfer chamber 42 and vacuum chamber 44 are preferably components of processing system 100 as shown in FIG. 1A. Thus, vacuum chamber 44 is a load lock chamber 106 that provides an interface chamber between the front end environment and transfer chamber 42, while transfer chamber 42 provides a vacuum environment that can communicate with various peripheral chambers. Alternatively, the vacuum chamber 44 may be a processing chamber such as a cooling chamber or a directing chamber as shown in embodiments described below with reference to FIGS. 12A to 12C.
[0048]
As shown in FIG. 6, the transmitting device unit 56 and the receiving device unit 58 are mounted on the outside of the lid 52 of the transfer chamber 42. In one embodiment, the transmitter unit 56 includes a light source 60 and beam shaping optics 62 and is positioned such that the signal 54 is transmitted into the cavity 41 of the transfer chamber 42 via the viewport 64. The viewport 64 has an opening formed in the lid 52 and is sealed with a plate 66 made of a material that transmits the signal 54 of the light source 60. In one embodiment, plate 66 may be comprised of any energy transmitting medium adapted to allow light transmission, such as, for example, Quartz Glass ™, glass, transparent polymer, GaAr, and the like.
[0049]
In operation, the signal 54 propagates parallel to the x-axis shown in FIG. 2 and on the upper surface of the substrate 37 that rotates (or otherwise moves relative to the signal 54) through the gap 41 of the transfer chamber 42. Sent to The signal defines the light projecting unit 53 falling on the substrate 37. As will be described in detail later, the spot diameter of the light projection unit 53 may be changed according to the size of the substrate by adjusting the positions of the beam shaping optical element 62 and the light source 60.
[0050]
In one embodiment, light source 60 may be a coherent light source, such as a laser, a non-coherent, broad spectrum light source, or any other non-coherent, narrow spectrum light source, such as infrared light. In other embodiments, light source 60 may include an illumination signal, such as a radio frequency, microwave, or the like. Generally, the light source 60 is selected according to scattering intensity, brightness, and cost. If a laser source is used, it is desirably operable at about 808 nm. However, another laser source such as a laser source having a wavelength of 650 nm or 680 nm may be used.
[0051]
In general, the spot diameter of the light projecting unit 53 is substantially determined by the beam shaping optical element 62 and the position of the transmitter unit 56 with respect to the substrate surface. The beam shaping optical element 62 is selected to have a spot diameter according to the size of the substrate. In one embodiment, the spots are concentrated in a narrow beam. The width of the beam can accommodate blade movement (eg, vertical movement) across the beam. For example, in the case of a substrate of 260 mm, on the upper surface of the substrate, the spot diameter of the light projecting section 53 is preferably about 1 mm (width), and preferably has a width of 260 mm (length, y-axis). Thus, in operation, the entire width of the 260 mm substrate is exposed to the signal 54 after one scan. However, in another embodiment, only a portion of the substrate is exposed to signal 54. For example, with regard to foreign object detection, a typical source of severe processing chamber contamination, such as flaking (also known as chamber excursions), produces hundreds of foreign objects that settle on the processing surface of the substrate. Although it is preferable to inspect the entire substrate, even if the presence of the contaminated substrate is confirmed only in a contaminated part, the processing inspection often works. Process monitoring for other substrate characteristics (eg, film thickness, end point confirmation, etc.) may also be performed by inspecting a limited number of substrates.
[0052]
The receiver unit 58 is mounted in a viewport 70 formed in the lid 52 and defines a signal path 61 to the substrate 37 that moves through the gap 41. The receiver unit 58 is fixed on an energy transmitting plate 72 made of a material selected according to the operating wavelength of the signal 54, preferably the same material as the plate 66 arranged in the viewport 64. For example, if signal source 60 is a laser source operating at about 808 nm, the materials of plates 66 and 72 will be selected to receive the 808 nm signal. The receiver unit 58 is arranged to receive the scattering part 74 of the signal 54 from the substrate 37 during operation. Both or one of the scattering part and the reflection part 74 is represented by the multiplicity of arrows pointing to various angles with respect to the upper surface of the substrate 37, and may be an obstacle such as contamination by foreign matter or an element pattern disposed on the upper surface of the substrate 37. Indicates the presence of an object. The reflector 69 of the signal 54 propagates at an angle to the substrate 37 substantially equal to the angle of incidence α. The reflection part 69 represents a part of the signal 54 that is not substantially hindered even if the upper surface of the substrate 37 is blocked.
[0053]
The receiver unit 58 has an optical element assembly 80 with one or more lenses and one detector 82. The detector 82 of the receiver unit 58 preferably comprises a charge coupled device (CCD) line camera. The CCD line camera is a preferable detector because the angle relationship between the light source and the receiving device is maintained throughout the substrate, and an irradiation / reception environment is always formed. An image is formed by successfully scanning a moving substrate in a receiving environment using a CCD line camera. However, although a CCD detector is preferred, other detectors, such as a time delay integration (TDI) camera or a photomultiplier tube (PMT), may be effectively used in embodiments of the present invention. In another embodiment, receiver unit 58 is selected from one or more of a spectrometer and an OCR receiver.
[0054]
The above description of the positioning of the transmitting device unit 56 and the receiving device unit 58 is merely an example, and other embodiments are possible. For example, FIG. 6 shows the transmitter unit 56 and the receiver unit 58 placed outside the gap 41 of the transfer chamber, but in another embodiment, the transmitter unit 56 and the receiver unit 58 41, and thus is positioned in a vacuum state. In another embodiment, a mirror surface can be placed in the gap 41 to achieve a more critical angle.
[0055]
The operation of the embodiment shown in FIG. 6 is illustrated in FIGS. 7A through 7C, which are top views of the processing system 100 showing the blade 48 and the substrate 37 in various positions during rotation through the transfer chamber. FIG. 3A shows the blade 48 immediately after starting to rotate counterclockwise so that the leading edge 92 of the substrate 37 is in the path of the signal 54. Therefore, in the figure, a part of the light projecting portion 53 represented by the shaded region blocks the front edge 92 of the substrate 37. As shown in FIGS. 3B to 3C, while the blade 48 continues to rotate, the light projection unit 53 scans the upper surface of the substrate 37. The light projection unit 53 illuminates an obstacle 75 on the substrate 37 that causes scattering and / or reflection of the signal 54. Obstacle 75 is in the micrometer range, but is shown with significant emphasis for clarity. Obstacles 75 may be a spall in the processing chamber, surface defects (erosion, dishing, etc.), or the intended characteristics of the device. After that, the receiving unit 58 controls the scattering unit and / or the reflecting unit of the signal 54. Although the detector 82 is a CCD, the scattering part and / or the reflection part 74 are focused by the optical element 82 of the receiving device, imaged on the element of the CCD, converted into an electric signal, and processed for processing. Is transmitted to the PMC 86.
[0056]
It can be appreciated that the above sequence may be performed before and / or after the substrate 37 undergoes a processing cycle in the processing chamber. For example, as shown in FIGS. 3A to 3C, a substrate may be transferred from a load lock chamber to a processing chamber along processing inspection paths 122 and 124. Alternatively, the processed substrate may be transferred to a cooling chamber or returned to the load lock 106 after processing, as shown in FIGS. 7A to 7C.
[0057]
In another embodiment, the processing inspection is performed while the robot 50 is retracting into or extending from the processing or service chamber, or into or out of a load lock along the inspection path 124. This is done during the extension. 8A to 8C are top views of the processing system 100 showing the blade 48 and the substrate 37 in various positions during linear movement out of the chamber 44 via the transfer chamber 42. FIG.
[0058]
8A to 8C are top views of the processing system 100 showing the blades 48 and the substrate 37 that is placed while the blades 48 extend from the transfer chamber 42 into the chamber 44 via the openings 46. FIG. Chamber 44 may be any type of chamber, such as, for example, a processing chamber, a cooling chamber, a measurement chamber, or a substrate placement chamber. In FIG. 8A, the blade 48 is shown immediately after a linear movement from the vacuum chamber 44 to the transfer chamber 42 has been started so that the leading edge 92 of the substrate 37 is positioned in the path of the signal 54. Therefore, in the figure, a part of the light projecting portion 53 represented by the shaded region blocks the front edge 92 of the substrate 37.
[0059]
To maximize the exposed surface area of the substrate 37, the signal 54 preferably blocks the substrate as it exits the chamber 42 at the opening 46. This positioning ensures that substantially the entire top surface of substrate 37 is exposed after blade 48 is fully retracted, thereby maximizing the surface area of substrate 37 scanned by signal 54. 8B to 8C, the light projection unit 53 scans the upper surface of the substrate 37 while the blade 48 keeps moving linearly.
[0060]
The embodiments shown in FIGS. 4, 7A to 7C and 8A to 8C are only examples. In alternative embodiments, a pair of transmitter units 56 and a pair of receiver units 58 are used in combination to monitor the substrate during linear and rotational movement, respectively. With this arrangement, the detection accuracy is improved. Other embodiments will be apparent to those skilled in the art. Furthermore, although the surface scanning of a substrate by a signal is highly accurate using the process monitoring method of the present invention, another method may be used to improve the process monitoring accuracy. For example, the robot blade 48 may be shaken, vibrated, or repositioned so that an obstruction can move into another CCD detector in the array of elements, perhaps in the field of view of an auxiliary high resolution camera. Also, a multi-mode camera whose focus, field of view, and the like automatically changes so as to improve the observation of the obstacle may be used. Inserting before and after multiple detector elements with a robot adds resolution to perform various monitoring methods.
[0061]
In another embodiment, the OIS 150 is configured to first receive reflected light (as opposed to scattered light) in the gray region configuration from the transfer chamber 110 from the transfer chamber 110. An embodiment is shown in FIGS. 9 and 10 show partial cross-sectional views of the processing system 100, particularly the processing chamber 114 and the transfer chamber 110. FIG. In the figure, the robot 113 carries the substrate 37 and is positioned adjacent to the entrance of the processing chamber 114. The receiving device unit 58C and the transmitting device unit 56 in FIG. 9 are mounted on the member 176 and mounted in the viewport 120 with the field of view of the substrate 37. In another embodiment, the receiving device unit 58D and the transmitting device unit 56 in FIG. 10 are mounted on the member 176, and are arranged at the viewpoint 120 in the field of view of the substrate. Preferably, the receiver unit 58C or 58D is a device adapted to receive the spectral data.
[0062]
When the substrate 37 is located below the OIS 150C as shown, the light emitted from the transmitter unit 56 is reflected and scattered from the upper surface of the substrate 37, or is either reflected or scattered. In some embodiments, substrate 37 is scanned with illumination source 56 in the manner described above. In other cases, the entire or substantially entire substrate is illuminated with a broadband flash device or similar device. In either case, the reflective portion is collected at receiver unit 58C in FIG. 9 or receiver unit 58D in FIG. 10 and transmitted to PMC 86 for processing. In particular, the collected information is processed to identify the spectral characteristics of the substrate 37. The spectrum analysis will be described later.
[0063]
In one embodiment, as shown in FIG. 9, the receiver unit receives the incident energy / signal and outputs the distribution of the spectral components and their intensities with a spectrometer 58C designed to determine the components of the energy. I just need. The spectrometer 58C includes an optical element assembly 170 including a fiber optic cable and / or a fisheye lens and all or only the diffuser lens and other optical elements 170 or only the diffuser lens and other optical elements. Coupled to the viewport 66, the signal is focused, shaped, and controlled so that the spectrometer 58C can view the entire substrate 37 and / or a portion thereof.
[0064]
In another embodiment, as shown in FIG. 10, the receiver unit 58 shown in FIG. 10 includes an OCR camera 58D. The OCR camera 58D is configured to acquire and specify a character image located on the upper surface of the substrate 37. Characters include barcodes and other symbol specific markings. Such a device is described above and need not be described in detail.
[0065]
FIGS. 6, 7A to 7C, 8A to 8C, 9 and 10 show embodiments in which the transmitter unit 56 and the receiver unit 58 are provided on or inside the transfer chamber. Other embodiments may be used effectively. In general, the transmitter unit 56 and the receiver unit 58 transmit the signal 54 toward the upper surface of the moving or non-moving substrate, and receive the scattered signal and / or the reflected signal 74. It may be arranged at any point on the processing system 100 as detected by the device unit 58. Multiple transmitters / receivers may be located throughout the system.
[0066]
IV. Lid assembly
11 and 12A to 12C show a lid assembly 1100 illustrating one embodiment of the OIS 150. FIG. Lid 1100 is adapted to be disposed on a processing or service chamber (otherwise as part of a processing or service chamber). The lid assembly 1100 has various devices and components that operate as part of the OIS. Lid assembly 1100 generally has a body defining three ports 1110, 1112, 1114 therein. In one embodiment, the first port 1110 provides a view to a receiver unit or camera 1116, such as a charge coupled device (CCD), for receiving light input from inside the chamber. The second and third ports 1112 and 1114 are formed in the lid 1100 to allow one or more transmitter units or light sources 1118 and 1120 to be coupled into the chamber through their respective ports, and the light sources 1118 and 1120 , That is, an exit of one or both of reflected light and scattered light from the substrate surface. In one embodiment, light sources 1118 and 1120 are a combination of a fiber optic bundle and a light source configured to form a horizontal beam. By attaching the brackets 1152, 1153, 1154, the camera 1116 and the light sources 1118, 1120 are all fixed to the chamber. In one embodiment, light sources 1118 and 1120 are halogen light sources or other light sources operable in the 400-700 nm region. Each of the ports 1110, 1112, 1114 has an energy transmission window 1122, 1124, 1126 mounted therein to create a vacuum separation inside the chamber where the lid assembly is located. In the particular embodiment illustrated, lid assembly 1100 also includes optical element assemblies 1121 and 1123 provided between windows 1124, 1126 and light sources 1118, 1120, respectively. Optical element assemblies 1121 and 1123 can include any combination of filters, diffusers, lenses, and the like.
[0067]
The lid assembly 1100, which carries the windows 1122, 1124, 1126, the CCD or camera 1116, and the light sources 1118, 1120, is made of aluminum or other machinable material to meet the degassing and porosity requirements of a vacuum environment. Can be manufactured. Preferably, each surface within the lid assembly 1100 is machined or polished to obtain the desired surface reflectivity. In some embodiments, the lid assembly 1100 is made of a metallic material to provide a surface 1130 with a suitable roughness or polished condition. The surfaces of both the light sources 1118 and 1120 and the port for the CCD 1116 are polished. For example, the CCD port 1110 is finished to 32 Ra, the surface in the first light source port 1114 is finished to 16 Ra, and the surface in the second light source port 1112 is finished to 8 Ra. The surface is finished to have a surface smoothness that minimizes scattering of secondary light into the optical environment of the chamber.
[0068]
The highly reflective surface is positioned such that reflected light is effectively emitted from the chamber through windows 1126 and / or 1124.
[0069]
12A to 12C are cross-sectional views of a lid assembly 1100 located at the upper end of a processing chamber 1101, such as a cooling chamber. The chamber 1101 typically has a chamber body having a side wall 1102 and a bottom surface 1104. The support member 1106 may be arranged via the bottom surface of the chamber to receive and support the substrate 37 introduced into the chamber 1101. The support member 1106 may include a cooling system, such as a fluid path or a cooling fluid source, for cooling the substrate, for example. A lid assembly 1100 on which the OIS 150E is mounted is provided on the upper surface of the chamber wall to form a seal with the upper surface.
[0070]
The port 1110 for the CCD 1116 has a first angle θ with respect to a horizontal line (that is, the substrate transfer path) when the substrate 37 is introduced into the chamber 1101. ₁ It is arranged in. This angle allows the camera 1116 to look at the substrate 37 and observe the substrate 37 when the substrate 37 enters the chamber 1101 on the robot blade 48. Ports 1110, 1112, 1114 for light sources 1118, 1120 respectively provide a CCD 1116 with a first light source 1118 for illuminating the substrate forward and a second light source 1120 for the substrate 37, which is introduced into the chamber 1101. It is arranged at an angle that gives back illumination. The front and rear irradiations may be performed simultaneously or separately when the substrate is introduced or pulled back into the chamber 1101, for example, one may be performed at the time of introduction and the other may be performed at the time of pulling back. The port 1112 for the first light source 1118 has an angle θ between the port 1110 and the camera 1116. ₁ Deviated from the second angle θ ₂ It is arranged in. The port 1114 for the second light source 1120 (ie, the back illuminating light source 1120) has a first and second angle θ with respect to the substrate transfer surface. ₁ And θ ₂ Yet another third angle θ deviated from ₃ It is arranged in. The signal reflecting surface 1128 is provided on one wall of the port 1114 so that the second light source 1120 can be disposed at a more acute incident angle with respect to the substrate 37 while performing the intended light incidence on the substrate 37. Placed on top. As shown in FIG. 12A, the incident angle of the beam emitted from the light source 1120 with respect to the substrate 37 is θ. ₄ It is.
[0071]
In some embodiments, the orientation of camera 1116 and light sources 1118, 1120 is adjusted automatically (rather than manually). For example, although not shown, a servo or similar actuator connected to a control system may be used to move various components, adjust aperture size, and focus from a remote location. In one embodiment, two cameras 1116 may be arranged side-by-side to better observe the substrate and use cameras in series to improve image resolution.
[0072]
Ports 1112 and 1114 for the first and second light sources 1118, 1120, respectively, are arranged to act as light traps for other light sources. That is, the port 1114 for the second light source 1120 acts as an optical trap for light from the first light source 1118 that has been reflected and scattered from the substrate surface, or has been reflected or scattered. Similarly, the port 1112 for the first light source 1118 functions as a light trap for light emitted from the second light source. The polished surface of the lid 1130 adjacent to the light source port 1126 also effectively acts as a light trap in the line of sight of the CCD 1116 by minimizing light backscatter. The inner surface of the lid 1130 adjacent to the light source port 1126 may be polished to minimize light backscatter to the camera on the main axis.
[0073]
The CCD 1116 and the ports 1112, 1114, 1110 for the light sources 1118, 1120 may each have an optical filter, such as a polarizer, color spectral filter, or other band-selective media. The filter can be positioned on the atmospheric side of windows 1126, 1122, 1124 provided at the openings in ports 1112, 1114, 1110, respectively. The filter may be provided inside the windows 1126, 1122, and 1124 or integrally with the windows 1126, 1122, and 1124.
[0074]
The use of a filter increases the contrast between the patterned background of the substrate 37 and the obstacles on the substrate and rejects long wavelength, poorly focused light or reduces the light reflection content and scatters. By increasing the components, the deterioration of the image can be minimized. For example, in one embodiment, a color spectral filter may be used to increase or select the energy (ie, signal energy) collected by the image associated with the obstruction. If the material on the substrate, e.g., the photoresist is blue and the disturbance is red, a red spectral filter can be used in the light source and camera to minimize the wavelength intensity associated with the photoresist and increase the wavelength associated with the disturbance. I can do it.
[0075]
If a linear polarization filter is used, the reflection component and the scattering component of the received light can be distinguished. For example, light scatter components are first filtered in a pattern viewed through two linear polarizing filters positioned at 90 ° to each other. The change in the scattering component indicates a change in the substrate structure pattern and / or the contamination. Filters can be used to enhance optical recognition (or OCR) in various applications. For example, one filter can be used to increase the degree of foreign matter detection, and another filter can be used to improve the degree of recognition of characters such as board-specific characters. In one embodiment, the light sources 1118, 1120 and the receiver CCD 1116 may include a plurality of different filters to provide different images to perform multiple substrate scans to improve processing inspection. In another embodiment, filters (not shown) are provided between the CCD 1116 and the port 1124 so that the various filters can be mounted independently of the operation of the CCD 1116.
[0076]
As shown in FIGS. 12A to 12C, the first light source 1118 for forward irradiation is directed at an angle away from the line of sight of the CCD 1116. The angle of incidence of light source 1118 on substrate 37 is determined such that the reflection from light source 1118 enters the optical trap formed by port 1114 for second light source 1120. Therefore, the CCD 1116 collects only the light reflected or scattered or reflected or scattered by the obstacle on the surface of the substrate 37 in the irradiation. The remaining reflected light is absorbed or routed through light traps 1112 and 1114, or reflects off surface 1130.
[0077]
The second light source 1120 is arranged to project light at an angle to the CCD 1116. By irradiating the substrate 1108 at this angle, back irradiation is obtained. As shown in FIGS. 12A to 12C, the second light source 1120 is arranged at an angle opposite to the direction in which the light source 1120 is projected. The first light source 1118 strikes the inner surface 1128 of the signal port 1114 and is reflected from the substrate 37 at a desired angle. The reflected light along the normal projection of the second light source 1120 reflects into the light trap formed by the first light source 1112. The light reflected by the irregularities on the surface of the substrate 37 is reflected along the line of sight of the CCD 1116.
[0078]
12A to 12C show both the first and second light sources 1118 and 1120. However, a single light source, either the first light source 1120 or the second light source 1118, may be effectively used depending on the needs of the particular process and substrate characteristics. For example, when detecting a smooth substrate, irradiation from a single light source is required. On the other hand, on a substrate on which a pattern is formed, it is optimal to inspect using two light sources and a polarizer. By providing contrast to two images formed using both front and back illumination, embodiments with two light sources could be used effectively. Information that needs to be received and interpreted by the detection system is reflected along the line of sight of the camera or CCD, providing enough information to make a sufficient decision for a particular process. In addition, one of the light sources 1118 and 1120 is used to scan the substrate 37 at the same time as the substrate 37 is inserted into the chamber, and another scan is performed when the substrate 37 is removed from the chamber 1101. 37 can be observed twice separately.
[0079]
Based on the above, three positions of the blade 48 extending into the chamber 1101 shown in FIGS. 12A to 12C indicate a scanning process, and the upper surface of the substrate 1108 is scanned by the scanning process. In operation, the optical information collected by the receiver unit 1116 (ie, the OCR camera) is used to analyze characteristics of the substrate 37 such as foreign matter, defects, surface damage, patterns, and identifier information (eg, alphanumeric characters). A method for processing such information will be described later.
[0080]
In another embodiment, lid assembly 1100 further includes a spectrometer 1150. Accordingly, FIGS. 12A to 12C show a spectrometer 1150 mounted on the mounting member 1152 adjacent to the light source 1120. Spectrometer 1150 is coupled to window 1126 via fiber optic cable 1155. Although the center of the fiber optic cable is off center, it is positioned close to the principal axis of the reflected light from both or one of light sources 1120 and 1118. In this configuration, the fiber optic cable 1155 obtains a large amount of broadband light sources used to perform another board inspection. It is also conceivable to place the spectrometer via fiber optic cable for any of the windows 1122, 1124, 1126 and couple to that window.
[0081]
V. Scanner
13A-13C are cross-sectional views of a lid assembly 1300 illustrating one embodiment of an OIS 150 for a processing chamber 114, such as a cooling chamber. The chamber 114 typically includes a chamber body having a side wall 1302 and a bottom surface 1304. The support member 1306 may be arranged via the bottom surface of the chamber to receive and support the substrate 37 introduced into the chamber 114. The support member 1306 may include a cooling system such as a fluid path or a cooling fluid source for cooling the substrate, for example.
[0082]
The OIS 150F including the receiving device 58, the transmitting device 56, and the connecting member 1308 is provided on the upper surface of the chamber wall, and is seated on the upper surface so as to be able to seal. The lid assembly typically includes a body defining an internal port 1310 to provide a line of sight for the receiving device 58 and the transmitting device 56. In one embodiment, port 1310 is sealed with an energy permeable medium 1312 so that signal 54 can be transmitted from transmitter 56.
[0083]
The receiving device 58 and the transmitting device 56 are connected to each other via a connecting member 1308 forming a scanning assembly 1325. Preferably, the transmitting device 56 and the receiving device 58 are angled such that reflected light and / or scattered light 1307 enter the receiving device 58 to enhance dark field illumination of the light field. Is good. Scanning assembly 1325 includes a motor (not shown) and is adapted to traverse along the length and width of lid assembly 1300 on rod assembly 1318. Preferably, the rod assembly 1308 allows the support member 1308 to traverse the substrate 37 entirely.
[0084]
13A to 13C show the operation of the processing inspection embodiment. 13A to 13C are side views of the chamber 114 showing the scanning assembly 1325 in various positions during movement along the processing chamber 114. FIG.
[0085]
FIG. 13A is a side view of the assembly 1300 showing the support member 1308 and the substrate 37 disposed on the assembly 1300 in the processing chamber 114. FIG. 13A shows the scanning assembly 1325 immediately after starting a linear movement so that the leading edge 92 of the substrate 37 is in the path of the signal 54.
[0086]
The processing chamber 114 is any type of chamber, such as, for example, a cooling member or a substrate placement chamber. Further, the OIS 150F is arranged at an arbitrary place on the tool 100 or in a dedicated inspection / measurement chamber to monitor the process.
[0087]
Scanning assembly 1325 continues to move from one side of chamber 114 to the other. In order to increase the resolution by a series of scans, the series of scans may be performed up to a new area of the substrate while increasing the OIS150F while proceeding. Thus, the signal 54 traverses the substrate 37, thereby exposing the upper surface of the substrate 37 to the signal 54. Obstacles in the substrate on the top surface of the substrate 37 (foreign matter, patterns, dishing, etc.) cause the signal 54 to be scattered and reflected, as shown by arrow 74, or to be either scattered or reflected. The scattering portion 74 of the signal 54 is collected at the receiver unit 58, converted to an electrical signal, and transmitted to the PMC 86 for processing. 13B to 13C show the scanning assembly 1325 continuing to move linearly, with the top surface of the substrate 37 being illuminated by the signal 54. FIG.
[0088]
Although the embodiments shown in FIGS. 11 and 13 are described with reference to the processing chamber 114, they may be used in other areas of the tool, such as the transfer chamber 110, load lock interface, and factory interface 104. At these locations, the substrate can be inspected before and after processing without affecting system throughput.
[0089]
VI. Board alignment and detection
The location of a defect on a substrate can be determined by identifying certain features on the substrate or blade. For example, in one embodiment, the leading edge, i.e., the curvature of the substrate that first enters the field of view of the receiver unit, and the lagging edge, i.e., the last curvature detected by the receiver, may be used during linear or rotational movement of the substrate. The PMC 86 may be programmed so as to detect this. Since the edge of the substrate is a reference point and the acquisition speed and field of view of the CCD detector element are known, the reference point may be used to form one of two coordinates, X and Y. The acquisition speed refers to the line acquisition frequency of the camera when forming an image while the substrate is moving within the field of view of the CCD detector. Preferably, a continuous image is formed so that the substrate has no overlapping or missing portions. Thereby, the processed output of the CCD detector becomes a "photograph" of the entire substrate surface. The position of the defect / obstruction may be determined using a detector array of a CCD detector.
[0090]
FIG. 14 illustrates a substrate detection system 1400 that includes two or more sensors for detecting a substrate. In one embodiment, the detection system includes a first sensor and a second sensor. The sensors detect the edge of the blade 48 and the substrate as the substrate moves across the sensors 1410A and 1410B.
[0091]
FIG. 14 shows the operation of the detector. FIG. 14 shows the substrate supported by blade 48 and moving toward sensors 1410A and 1410B. As the substrate enters the optical path of the signals of sensors 1410A and 1410B, the waveform output changes. Using the change in output, the presence, position, and speed of the substrate are determined. In particular, the detector output is an electrical signal indicating that it is as logically high as the positive detection criterion.
[0092]
At position A, the output signal is low and T = t ₁ It becomes high with. This indicates that the substrate has just been detected. The high output signal of 1410A indicates that T = t when the blade is detected by sensor 1410A. ₁ Indicates that it becomes higher. Position B shows the edge of blade 48 traversing the front of sensor 1410A, where T = t ₂ It is shown that the waveform at the time of is high. A low output for sensor 1410B indicates that blade 48 has not been detected by sensor 1410B. Positions C and E show the edge of blade 48 moving toward sensor 1410B and crossing in front of 1410B. T = t ₃ , Which indicates that the blade edge has been detected by the sensor 1410B.
[0093]
T = t ₂ And T = t ₄ At this time, the blade has completely traversed the detector and the substrate is detected. If the substrate has a non-uniform reflective surface, false triggers may be identified in areas 1430A and 1430B. t ₁ , T ₂ , T ₃ And t ₄ Then, 1430A and 1430B do not consider the area. These possible erroneous signals are identified by checking and comparing timing intervals specific to the blade edge speed, width, distance and angle between sensors.
[0094]
The direction of movement of the substrate is detected by processing which sensor 1410A or 1410B first detected the substrate. The speed is determined by knowing the distance between the two sensors and dividing by the total time.
[0095]
VII. Summary of device description
In the above embodiments, the detection apparatus and method capable of monitoring the substrate in progress and on-site by the processing system are described. On-site and on-the-fly testing minimizes the need to use a conventional stand-alone testing platform with dedicated actuation mechanisms as commonly used in the art. In other embodiments, the inspection is performed while the substrate is stationary, eg, in a cooling chamber. Further, embodiments of the present invention effectively utilize components typically included in any conventional processing system, such as robot 113 (shown in FIG. 1A), enabling a non-stepped inspection system. . In each case, the process impact can be minimized during the required sequence of operations by monitoring the process at various points in the processing system without having to transfer the substrate to a separate stand-alone inspection platform. Can be stopped. Thus, each substrate traveling through the entire processing system can be inspected, which improves upon conventional systems and processes that only allow for periodic sampling in view of the adverse effect on throughput.
[0096]
At the time of manufacture, embodiments of the present invention provide a viable method for determining whether to stop manufacturing or to more carefully inspect a particular substrate for contamination, processing related defects and routing errors. I do. Therefore, only the selected substrate needs to be further inspected. If the substrate is proven to be ok in the system, no further alignment or other alignment is required. In addition, by using conventional components such as transfer chambers, viewports provided in transfer chambers, processing chambers such as cooling chambers and placement chambers, and transfer robots, valuable time must be spent on refurbishment, cleaning, and reconfirmation. Rather, it is easier to retrofit existing systems with embodiments of the invention.
[0097]
It should be noted that while embodiments of the invention facilitate testing while in progress, other embodiments may be utilized to effectively use a dedicated metrology platform, such as the testing platform 135 described above (shown in FIG. 1A). Good. In particular, a process monitoring method described below, such as analysis of both or any one of reflection and scattering and spectrum analysis, does not depend on a specific mechanism and does not need to be executed in progress.
[0098]
VIII. Embedded processing monitoring:
The inventors have described the inspection apparatus described herein in many of the inventive uses required in processing systems, such as calibrating robots and inspection equipment, monitoring robot behavior, as well as contamination, reflectance, (reflection or It has been discovered that it can be adapted to determine selected characteristics of the substrate, including the type of substrate (scattering), discontinuities, placement, center detection, etc. In the following description, various embodiments of the present invention will be described. However, those skilled in the art can easily understand other possible embodiments, and are not limited to the embodiments described here.
[0099]
The present invention also allows the inspection system 135 to perform process monitoring and subsequent further evaluation. As the substrate is processed, each OIS 150 produces a spectral signature, color, etc., indicative of the state / status of the substrate. In one embodiment, at each processing step, PMC 86 monitors the status using data obtained from each OIS 150. If a defect is detected during the condition, the PMC 86 will need to send the board to the inspection platform 135 for further analysis, or request operator intervention to remove the obstructing board in the system. It is determined whether to continue the next processing step. In one embodiment, the inspection platform 135 is adapted to perform the same tests as the OIS 150 but at a much higher resolution. All substrates can be inspected, or inspection can be concentrated on a specific area. If there is no severe processing event that causes the processing to be stopped, the measurement unit inspects the suspicious substrate while continuing the processing, so that the throughput is not seriously affected.
[0100]
Inspection platform 135 merely represents one site for another inspection. Typically, the system may have any number of inspection sites, not just OIS 150. Thus, in some embodiments, a secondary, or even tertiary, platform may be provided.
[0101]
Accordingly, the present invention provides an apparatus and method for generating real-time information about a selected property of a substrate. The substrate inspection is preferably performed before and after processing. The preferred operation of the present invention will be understood with reference to FIG. 1A. Optical inspection of the substrate is first performed by the OIS 150 located within the factory interface 104. Therefore, the substrate can be analyzed before the substrate enters the vacuum environment 110 where the processing chamber and the service chamber of the processing system are arranged. After transferring the substrate from the transfer chamber 110 into the processing chamber 114 or the service chamber 116 by the robot 113, embodiments of the present invention scan the substrate or otherwise acquire an image of a portion or the entire substrate. It works as follows. Subsequent to processing, the substrate may be rescanned while the substrate is being withdrawn from the processing or service chamber. Further, a determination regarding the processing result may be made. For example, information on processing uniformity may be generated using the collected substrate image scanning, or it may be confirmed that processing has been performed at the end point.
[0102]
Therefore, it is possible to continuously monitor the substrate at various stations in the processing system 100. For example, from the resulting images, processing uniformity, smoothness, substrate type, placement, center detection, discontinuities / edge defects (eg, structural defects in the substrate due to heat transfer that cause the substrate to break), Information is generated about the evidence of reflection and / or scattering, the presence of foreign matter, and other processing conditions. In one embodiment, the OISI 150 and the PMC 86 operate to form a mapping of the substrate topography (eg, a two-dimensional or three-dimensional mapping). The mapping is analyzed for texture properties such as flatness, uniformity, and thickness. Further, any substrate damage or defect such as a chip or a crack may be detected and mapped. The analysis can be improved by using a color CCD detector and / or a spectrometer.
[0103]
Another substrate characteristic that can be determined is the optical surface characteristic of the substrate. Using information about optical surface properties, it can be determined whether a certain processing state, such as the end point of an etching process, has been successfully achieved. Because the endpoint information is available in near real time, i.e., near the processing chamber substantially simultaneously with the end of processing, the substrate being processed may be returned immediately for further processing. Traditionally, substrates have been sent to a remote location for endpoint inspection. Subsequent determinations as to whether the substrate is in process may require the substrate to be discarded because it typically takes too much time to return the substrate for further processing or to grow native oxide. Was. In addition, processing times are often extended, and avoiding substrates during processing reduces system throughput.
[0104]
Thus, the present invention provides near real-time pre- and post-processing information about the properties of the substrate being processed. Because the information is near real-time, decisions on how to handle the failure base can be made immediately and cost-effectively. In addition, instead of selecting a board from a series of boards, each board is inspected almost immediately after processing in the system, so additional board and information that can be quickly corrected for issues identified in the processing environment. Can be used. This allows process monitoring to be near real-time, optimizing process recipes, bringing them very close to process tolerances, and thus increasing throughput.
[0105]
Thus, the apparatus functions as a "watcher" processing system that plays a continuous first level of substrate and system characteristics. If the acceptance criteria are not met, further analysis can be performed on the inspection platform 135 or a similar inspection platform. Such a system increases the opportunity to identify a wide range of processes and handle the routing of the original defect without adversely affecting throughput.
[0106]
A. Reflection analysis
In one embodiment, substrate characteristics are analyzed using both or one of the reflection information and the scattering information. Such embodiments are particularly applicable to patterned substrates that include topographic variations that can cause scattering of incident light. When inspecting a substrate on which a pattern is formed, the present invention utilizes a unique intensity distribution of a trace generated by irradiating the substrate. The unique signature is the result of the pattern / structure formed on the substrate. Since the topography due to the pattern on the substrate that has undergone a particular process is substantially repeated, the signature is almost consistent in each of the processed substrates, but still varies from one processed substrate to another. Therefore, the unique trace is stored in the memory and used to compare the surface condition of the substrate being manufactured. Further, an average signature of the last "n" processed substrates, where n is an integer greater than 1, may be used as a dynamic reference (or calibration) substrate.
[0107]
FIG. 15 shows a substrate image scan consisting of 30,000 data points showing reflection traces and / or scatter traces 150 for a calibration substrate scanned in accordance with the techniques described above. The number of occurrences (y-axis) at a certain intensity (x-axis) level, that is, the number of readings by the detector is plotted. Next, two different test substrates were similarly scanned to obtain two different, distinct reflection signatures. The traces for the two test substrates were compared to the trace 150 for the calibration substrate trace 150 to determine the relative condition of the substrate surface. Graphs 152 and 154 shown in FIG. 16 are results obtained by subtracting the number of occurrences at a certain intensity on the calibration substrate from the number of occurrences at the same intensity on the two test substrates. Thus, the first graph 152 represents the difference in the recorded intensity output of the detector between the first test substrate and the calibration substrate. The second graph 154 represents the difference in the recorded output of the detector between the second test substrate and the calibration substrate and shows a significant variation indicating the difference in surface condition between the compared substrates.
[0108]
Both reflection information and / or scattering information can be collected using the above-described embodiment, for example, provided with a line camera (eg, CCD 1116 shown in FIGS. 11 and 12A-12C). In such embodiments, a line camera is positioned to collect light scattered from the substrate. In the case of a CCD camera, the CCD detector of the camera comprises, for example, 4096 pixel elements arranged in a linear array. These elements serve to provide light intensity in the x-axis. The y-axis value is generated by collecting a continuous data set (x value) while the robot moves the substrate into (or out of) the cooling chamber. Each pixel value in the data set ranges from 0 to 255 and represents the intensity level of light collected at a particular portion of the substrate. In this case, the scanning process resulted in a data array of more than 165,000 intensity values for each substrate. In particular, the data array can be used for foreign object detection and processing consistency confirmation using analysis of reflection traces and / or scattering traces.
[0109]
In one embodiment, the number of occurrences of an intensity value from 0 to 255 can be represented in an intensity distribution histogram. Since the intensity distribution histogram is directly related to the structure on the substrate surface, it can be considered as a unique reflection and / or scattering signature of the substrate type processed in a particular way. If that amount of data is included, even small changes in the substrate surface can cause significant changes in the reflection and / or scattering signatures. By plotting such a change, it is possible to show the difference between the reflection trace and / or the scattering trace for the same substrate. The reference signature subtracted from the other signatures represents the ideal / desired processing object. If it is known that a certain defect mode exists for a particular process, the evidence can be used as a reference. The resulting difference plot represents structural changes on the substrate surface. FIG. 17 includes the difference between the reflected and / or scattered traces over a series of overetch times with respect to the nominal overetch time of 40 seconds.
[0110]
For clarity, the first 100 out of 255 intensity bins / ranges are shown. As shown in FIG. 17, a change in the reflection trace and / or the scattering trace is large, and the change is caused by a different processing state. For example, for an intensity bin of less than 18, an etch time of less than 40 seconds nominally occurs about 260,000 more times than an etch time of more than 40 seconds. With an intensity bin of about 18, an etch time shorter than the nominal 40 seconds will generate hundreds of thousands less times than an etch time longer than 40 seconds. These signature changes are, in other words, due to changes in features / structures on the substrate resulting from the processing. Below 40 seconds, the structure does not appear completely, and at about 40 seconds, the etching process affects the photoresist and aluminum lines. For clarity, FIG. 18 shows a conventional endpoint plot with overetch intervals marked.
[0111]
To simplify the display presented to the user, the treatment of reflection traces and / or scattering traces can be expressed as an average luminance value (FIG. 19). The system operator observes a running average as the system processes substrates. Marking the warning and alarm areas also gives the user quick feedback on processing integrity.
[0112]
FIG. 20 shows the average intensity value of the substrate calculated as the delta average for the reference reflection trace and / or the scattering trace. The delta average represents an average weighted average luminance associated with the intensity distribution of the substrate with respect to the reference average luminance. FIG. 20 shows the continuous delta average value on the substrate with respect to the reference substrate average value. As noted above, marking the warning and alarm control limits can provide quick feedback to the user on process integrity.
[0113]
Although the description herein has focused on reference reflection traces and / or scattering traces from an etch process that already has an endpoint system in place, embodiments of the invention may include other structural changes in the substrate. The present invention can be similarly applied to the case where the processing step is monitored. For example, FIG. 21 reflects the changes in the reference reflection signature and / or the scattering signature seen as the strip time increases.
[0114]
Example
Substrates were processed in a Centura system with a DPS metal etch chamber and an ASP strip / passivation chamber available from Applied Materials, Inc. of Santa Clara, California. After etching and stripping, the hot substrate is cooled to approximately room temperature in a cooling chamber having the apparatus system according to one embodiment of the present invention.
[0115]
The processed substrate consisted of an EPIC-generated substrate, and Applied Materials equipment was used in all steps except the lithography step. The photoresist / metal stack on the substrate was: 8000A DUV photoresist / 250A TiN ARC / 5500A Al-Cu 0.5% / 250A TiN / 200A Ti / 3000A thermal oxide. A dense pattern of 0.25 micrometers lines / space occupied about 50% of the substrate.
[0116]
All substrates were etched using a typical recipe. This etched the entire metal stack, leaving the majority of the photoresist at about 5200A on the patterned components (3700A at the shoulders of the patterned lines).
[0117]
As the geometry of the chip shrinks, a greater variety is observed between thin metal stacks and (especially theoretically) higher levels of metal. The contact tip is the most extreme, the aluminum metal 1 is 4000 A and the thickness of the aluminum top metal is often 4 (m), so the yield is more sensitive to corrosion and photoresist remaining on the substrate while the ASP chamber The ability must be compatible with all metal stacks.
[0118]
An ASP recipe was used for this experiment. This recipe is H.sub.2 at a lower pressure than recipes typically used in ASP chambers. ₂ It consisted of only one processing step in which only O was flowed. The new recipe is as follows: temperature 250C, pressure 0.5 torr, H2O flow 750 seem, microwave power 1400 W, one step with variable time specified in the matrix.
[0119]
The etched substrate was measured using an embodiment of the present invention at various ASP processing times from 0, 10, 20, 26.40, 50 seconds up to 200 seconds. FIG. 22 shows a state in which the intensity scan changes at various times up to 50 seconds of the ASP processing. On the scale of FIG. 22, a peeling time longer than 40 seconds looks the same as 40-50 seconds, and is replotted in another graph as shown in FIG. From this figure, it can be seen that the photoresist was completely removed in 40 seconds.
[0120]
Since no obvious trends can be seen in these curves, it is likely that they are noise. To check the noise level, one peeled substrate is inspected six times in a row to determine repeatability. This information is shown in FIG. Assuming that the repeatable scan is similar to the scan of FIG. 23, the substrate is completely stripped in 40 seconds, the noise level is limited to an envelope of approximately 15,000 occurrences at zero intensity, and the luminance It can be concluded that the value decreases as the value increases. This envelope is multiplied by 3 to obtain a 3-sigma envelope. Any value exceeding this envelope indicates statistically valid data. At the manufacturing facility, sigma is measured for each type of product, and therefore includes a systematic review of sources of variation.
[0121]
In each case, at a strip time of 26 seconds, the peak at intensity bin 10 is 70,000, more than twice the projected 3-sigma envelope. This is the reference measurement for the remaining photoresist. Since this signal is greater than 3 sigma (> 3 sigma), the photoresist should be detectable on the stripped substrate in 35 seconds. Thus, a 700 A photoresist can be detected on the substrate.
[0122]
In another embodiment, a moving average of the averages, ie, a boxcar average, is used. The average of successive averages obtained during the manufacturing process is obtained by adding the amount of the previous average and dividing by the number of continuous averages added. Once the average of the averages is obtained, the average of the averages is used as a moving reference value and compared with the new substrate average. If the new board average is within the acceptable range for the average, replace the average of the average as a new reference with one of the previous averages and the current average from the current board; It is calculated by calculating as described above. The warning and alarm control limits are maintained so that the average of the average values does not "gradually" increase or decrease to exceed the processing allowance.
[0123]
By using the optimally processed substrate as a reference signature, a series of recipe experiments can be performed immediately and the results can be observed without having to remove the substrate for independent verification. Scanning may be performed until the trace difference is relatively reduced.
[0124]
Also, an image of the patterned substrate, including useful information obtained according to embodiments of the present invention, can be used to measure parameters associated with other processing on the substrate. For example, it is possible to detect a photoresist remaining on a substrate on which a pattern is formed after metal etching and stripping / passivation processing.
[0125]
In a metal etching process, the substrate is etched and stripped in place. In-situ stripping and passivation processes are also necessary to prevent corrosion of the etched aluminum structures on the processed substrate. All photoresist must be removed from the substrate since the chlorine-containing photoresist causes corrosion of the aluminum after the substrate is evacuated to a manufacturing atmosphere containing some moisture. In a typical manufacturing situation, the substrate is exposed to air for one hour to one day, followed by further processing to stop corrosion.
[0126]
In a production environment, the OIS 150 of the present invention can be used as a photoresist detector to prevent any substrate with residual photoresist from being vented to the atmosphere where corrosion can occur. Using the embodiments of the present invention, many substrate states, such as large foreign matter, residual photoresist, corners of the substrate, oxide thickness, etc., can be measured. Other processes can be achieved using embodiments of the present invention, but only to the extent that structural changes occur on the substrate surface.
[0127]
In one embodiment, many processing systems 100 with associated inspection equipment can be networked to track individual substrate data to detect processing integration problems. For example, the thickness of the oxide can be tracked using spectral scanning of the corners of the substrate in steps from oxide growth to oxide etching / stripping and metal etching / stripping. Similarly, metal stacking changes in thickness and uniformity can be tracked. Large contaminants can be tracked by various processing steps so that the hardware that generates the contaminants is accurate and readily identifiable. The networked inspection web forms an integrated process monitoring capability of the substrate using the production line from beginning to end of the manufacturing process.
[0128]
In at least one embodiment of the present invention, a two-stage foreign object detection method is used. In the first stage, various stain analysis techniques are used to determine the approximate size and location of the foreign object as represented by pixel generation. This information can also be used to determine the approach of these pixel occurrences. In the second stage, the warning and alarm states are determined using the number of pixel occurrences equal to or greater than the intensity threshold. Using such a method, the approximate size and position of the foreign object can be determined, and individual pixel intensity thresholds can trigger an alarm.
[0129]
In one embodiment, the contrast of the obstruction may be maximized by selecting a polarization source and a linear polarization filter configured to select light rotated about 90 degrees from the polarization direction of the polarization source. As a result of this "cross-polarization" filtering method, the scattered light is the main component incident on the receiver. The signal scatter represents a randomly rotated signal. Rotation occurs when the signal illuminates features / structures on the substrate surface. By using mutual polarization, it is possible to selectively remove the reflected component of the reflected light pattern from the substrate that has the scattered light and enhances it, thereby increasing the contrast of the obstacle. This method serves to enhance the contribution of light from the patterned substrate structure, thereby reducing processing problems, substrate layer and thickness problems, and / or routing errors. The sensitivity to intensity fluctuations caused by the cause is improved.
[0130]
In another embodiment, the detection of changes in the intensity of the reflected signal is used to enhance the reflected light, possibly caused by small mirrors formed by the "dishing" substrate surface. The dished surface separates the signal from the receiver and directs it to the receiver.
[0131]
B. Spectrum analysis
In other embodiments, the spectral data obtained from a substrate is an image, which is used effectively. Spectral data may be collected using a spectrometer, color CCD camera, or other device known in the art. An exemplary embodiment using a color acquisition device has been described above. For example, FIG. 9 and FIGS. 12A through 12C show an OIS 150 using a spectrometer.
[0132]
In one embodiment, spectral data collected from a substrate is used to generate spectral signatures in much the same way as reflective signature generation (described above). The spectral signature represents the intensity of the color components and the substrate. As described with respect to the reflection trace, the color trace may be further compared with the reference color trace to determine characteristics such as the type of the substrate and the residual material during processing.
[0133]
FIG. 25 shows various regions of the substrate 37 that have identified processing problems using spectral data. For example, the substrate region 2505 is defined by a shaded region indicating that various electronic elements are etched and processed during processing to obtain a substrate. Substrate area 2510 is an enlarged view of the processed area of the substrate. Substrate area 2520 is an open area of substrate 37 that has been processed but the circuits have not been etched. The substrate region 2530 is an overall view of the substrate 37 and captures all of the substrate in one view. The substrate region 2540 is a region of the substrate as viewed from the OIS 150 receiving element as described above.
[0134]
In any of these regions, spectral information can be obtained by irradiating with a signal such as a light source. Different color / spectral signatures may be used to determine local or global processing abnormalities. For example, consider the case where substrate region 2510 is normally green when viewed with a spectrometer. During processing inspection with OIS 150, the spectrometer obtains a picture of the substrate area 2520 for the new substrate and discovers that it has blue components. It is known that the color changes as the thickness of the substrate changes, so changing from green to bluish green may be a processing problem, but it indicates that the thickness of the substrate layer is incorrect. Is shown.
[0135]
Observed areas 2510, 2520, 2530, and 2540 have different problems that can be considered processing monitoring points. For example, the coloring of substrate region 2520 indicates a change in plasma density during processing. Even in over-etching, the spectrum traces are different.
[0136]
In one embodiment, successive color scans of the same substrate and overlapping them provide a color contour map of the substrate. Color contour mapping improves process inspection by effectively indicating color changes across the substrate. For example, if the substrate has a uniform and smooth surface, there is almost no change in color variation. The color profile mapping of the substrate exhibiting color variations is related to changes in the thickness and / or uniformity of the substrate during the plasma processing step.
[0137]
In another embodiment, the amount of energy received is a function of the size of the fiber optic cable. For example, as shown in FIG. 9, fiber optic cable 170 is used to direct the reflected light from light source 56. An optical fiber cable consists of a chain of a plurality of optical fiber filaments. Increasing the number of chains increases the amount of collected signal energy transmitted to the spectrometer 58.
[0138]
In one embodiment, the angle of the detector with respect to the fiber optic cable is moved relative to the principal axis of the reflected light to enhance a certain color spectrum. The different spectra relate to substrate topography differences and / or material differences, such as substrate thickness, obstacles, and substrate material type.
[0139]
C. Board type identification and routing
In one embodiment, the invention determines the type of substrate. As mentioned above, the pattern on the substrate produces unique reflections and spectral signatures. Accordingly, the present invention may be used to scan a substrate in the manner described above, transmit the received signal to the PMC 86 for processing, and recognize the substrate based on evidence. As described with reference to FIG. 9, the receiving device 58 receives the light reflection and scattering portions and determines the color of the substrate surface. Next, the type of substrate is determined by comparing the scanned pattern with stored color and / or reflection traces. By using in this way, it is possible to detect a substrate on which an incorrect route has been set across the system. For example, the OIS 150 may detect and reject a substrate with photoresist, which was incorrectly sent to a physical vapor deposition (PVD) chamber. This will prevent the processing chamber and the substrate from being damaged. Further, the processing recipe can be automatically changed according to the type of the substrate by using the recognition of the substrate pattern. Furthermore, board recognition allows the device to automatically select appropriate test / monitoring criteria for warning and alarm conditions.
[0140]
D. 3D image formation
The invention can also monitor faults in three dimensions (3D). Referring again to FIG. 2, in one embodiment, the substrate is scanned in two or more directions by light source 56A and then in two or more directions by light source 56B. The receiving device 58 acquires the reflected signal and / or the scattered signal from the light sources 56A and 56B including two or more different images. If the images 56A and 56B are at right angles to each other, the light illumination will strike a substrate surface obstruction from two different angles (ie, perspective). Thus, an image from light source 56A obtains information relating to one side of the obstruction, while an image from light source 56B obtains information relating to the opposite side of the obstruction.
[0141]
Scanning by light sources 56A and 56B is used for surface and substrate movement in all directions. For example, the substrate is moved in one direction and scanned by the light source 56A, and then moved in the same direction or a different direction and scanned by the light source 56B. In another embodiment, the substrate is scanned using multiple light sources at different angles.
[0142]
If data integration and image processing techniques are used, an appropriate three-dimensional display of the substrate surface can be obtained. This image allows the process monitoring system to locate the size, height, and / or depth of an obstacle or a patterned feature. Thus, by forming a three-dimensional image, the surface topography of the substrate can be mapped and the mapping can be compared with other substrate topography to determine the relative uniformity of the substrate.
[0143]
E. FIG. Optical character recognition
In another embodiment, the invention provides optical character recognition (OCR). OCR refers to the detection and processing of alphanumeric characters by image formation. The substrate may be identified by a character typically imprinted on its surface. As shown in FIG. 10, the transmitter unit and receiver unit 58 of the present invention, which are hidden in the figure, provide a device that can detect and illuminate characters and direct signals to the PMC 86 for processing. The receiver unit 58 is positioned to receive reflected light and / or scattered light. In operation, the substrate is scanned in the manner described above. During scanning, the signal 54 hits a character on the substrate and reflects / scatters according to the geometry of the character. As noted above, reflection / scattering is unique to each character arrangement and configuration. OCR techniques typically utilize a pattern recognition algorithm adapted to recognize and read images, such as characters, symbols, bar codes, and the like.
[0144]
F. Placement and center detection
In another embodiment, the invention is used to determine the placement and center of a substrate. Positioning and center detection are necessary to ensure that the substrate is properly positioned within the chamber for processing. For example, etching often uses a mask, guard ring or clamp to cover certain portions of the substrate surface. In order for the mask, guard ring, or clamp to be located in the proper portion of the substrate, the center of the substrate must be accurately located in the processing chamber. Therefore, the centering / positioning of the substrate can be performed using the curvature of the substrate edge. Further, the arrangement can be confirmed using a flat portion or a notch typically provided on the substrate.
[0145]
When locating the center of the substrate, the center of the substrate is determined using one or more sensors. Utilizing the present invention provides substrate center detection capability, thereby minimizing the need for additional sensors. In particular, the substrate can be scanned in FI 104 for process monitoring and foreign object detection. Thus, information received during the scan can be processed to determine center and / or placement.
[0146]
In one embodiment, the center and / or location is found while the substrate is moving. As described above, the substrate is illuminated and scanned by the OIS 150 (ie, the transmitter unit 56 and the receiver unit 58) while pulling back, extending, and / or rotating the blade. Thus, in one embodiment, PMC 86 can determine the diameter, and thus the center, of the substrate. For example, as the substrate moves into the signal path, the leading edge of the substrate is detected by reflected light. Once the board has passed the signal and over it, the receiver unit 58 stops detecting the signal. Record the time between the first detection of the signal and the abort detection. At known robot speeds, the recorded time may be used to calculate the diameter of the substrate. If it is determined that the substrate is not centered with respect to the calibration value, the destination coordinates of the robot are adjusted to correct the deviation. The particular method of calculating the substrate center is not a limitation of the present invention, and those skilled in the art will appreciate other possibilities. For example, in another embodiment, the detection of the leading and trailing edges of the substrate is related to the encoder value of the robot at the time of detection. The encoder value must be compared to a calibration value for a substrate of the same diameter to determine the offset and adjust for this offset.
[0147]
In another embodiment, the substrate is positioned in a chamber, such as a cool chamber, a degas chamber, or any of the chambers of the processing system as shown in FIG. 1A (ie, the substrate is stationary). , The placement and center of the substrate can be found. If the substrate is positioned within the field of view of the receiver unit 58, center detection and arrangement can be performed simultaneously. 13A to 13C illustrate one embodiment of OIS 150 adapted for placement and center detection when the substrate is not moving. The substrate surface is illuminated by moving the components of the OIS rather than moving the substrate. Alternatively, referring to FIGS. 9 and 10, the transmitter unit may be configured to illuminate a sufficient portion of the substrate without moving any of the OIS components or the substrate (eg, the transmitter unit). Is provided with a flash element and an optional optical element). As described above, the chamber not only performs a processing function such as cooling or degassing, but also functions as an area in substrate analysis. As a result, the analysis can proceed without affecting the throughput of the processing system.
[0148]
G. FIG. Equipment calibration
In addition to substrate inspection, the invention is adapted for calibration. In one embodiment, the invention is used to calibrate the detection device. Since the illumination and detection optics of the present invention are not uniform, operation must be standardized. In one embodiment, the normalization is performed as follows. The OIS 150 (transmitter unit and receiver unit) is first installed and the substrate is placed upside down on the robot blade to hit the scattering surface. Next, the robot blade moves the substrate under the OIS 150. During blade rotation and linear movement, the OIS 150 measures vertex to vertex root mean square (RMS) over the entire substrate surface and sends it to the PMC 86. Next, the average readings from each OIS150 measurement are compared to determine the correction factor required for system standardization. In addition, the substrate is removed and the rigid portion of the robot blade, ie, the portion without holes and edges, is scanned in a similar manner. Thereafter, the vertex and root mean square values from vertex on the blade are compared to the standardized correction factor to determine an appropriate blade correction factor for the entire journey. Once the blade normalization factor is obtained, it is possible for the blade to function as a unique calibration standard. Therefore, the OIS 150 together with the PMC 86 can monitor the empty blade in normal operation to determine whether the receiving device and the transmitting device are functioning properly. If the receiving device and the transmitting device are contaminated, they are detected by the background test described above. The flatness and consistency of the blade is also monitored and confirmed.
[0149]
H. Blade contamination
Further, contamination on the blade surface is also detected by the test described in the above embodiment. Contamination on the blade indicates that the backside of the substrate has been scratched at some point during substrate handling and / or that residual processing byproducts have collected on the substrate. Thus, if contamination on the blade is detected, the system can be shut down for inspection, thereby preventing further contamination of the processing environment.
[0150]
I. Robot calibration
In another embodiment, OIS 150 facilitates robot calibration. Processing system robots, such as transfer chamber robot 113, must be periodically calibrated to ensure correct placement and alignment. Because the OIS 150 is mounted at a fixed location in the processing system, the OIS 150 can provide a reference point to the PMC 86 for transfer chamber robot calibration. Once the blade standardization factor is determined as described above, the position of the robot can be ascertained by detecting the characteristics of the blade. By monitoring speed and vibration, the proportional, integral and derivative (PID) values of the motion control system can be monitored / adjusted. A sufficient variation between the detected position value and the calibrated position value stored in PMC 86 indicates that the blade is misaligned. Therefore, misalignment can be corrected automatically.
[0151]
J. Robot behavior
In the present invention, the behavior of the robot can also be monitored. For example, if the robot blade is rotating in the optical path 61 (shown in FIG. 6) of the receiver unit 58, the blade edge closest to the center of rotation will enter the optical path first. This edge then in turn enters the field of view (FOV) of each detector element at a speed determined by the blade speed. This allows the OIS 150 to independently measure / monitor the behavior of the robot including features such as settling time and stability. Using the collected data, the PID parameters of the robot 113 can be set manually or automatically.
[0152]
Other possible uses are not described in detail here. For example, the present invention can be used to determine whether a substrate is within a blade clamp used to secure the substrate during blade movement, and to detect that a substrate is present on the robot blade. Good. Those skilled in the art will appreciate other possible applications of the present invention.
[0153]
Thus, the present invention facilitates the integration of the myriad functions currently achieved by different components within a typical processing tool. One or more OIS units 150, effectively located, such as in a transfer chamber, can perform multiple process monitoring functions. Accordingly, the present invention provides a multipurpose device that can achieve system integration at a relatively high level and reduce system operating costs.
[0154]
K. First wafer effect
One common state in semiconductor processing is known as the "first wafer effect." The effect of the first wafer (or substrate) is the effect of a clean chamber condition on substrate processing. The chamber must be periodically cleaned to remove residual build-up that has accumulated over time inside the chamber surface. However, it was found that the result of the substrate processing before the cleaning was different from the result of the processing after the cleaning. In particular, the N substrates after the cleaning cycle differed in properties from the subsequently processed substrates. Thus, the cleaned chamber is typically subjected to a seasoning process, which allows the chamber to reach an equilibrium such that the substrate is processed substantially uniformly. For seasoning, the chamber is operated under processing conditions (or modified processing conditions to promote the desired result) to coat the chamber surface with material. However, one problem with the seasoning process is deciding when to fully season the chamber.
[0155]
According to the embodiment of the present invention, the end point of the seasoning process can be detected. Specifically, during seasoning of the chamber, the substrate is processed in the cleaned chamber. Thereafter, each substrate is inspected using one or more OISs 150. Once the processed substrate exhibits the desired properties, it is known that the chamber has been sufficiently seasoned.
[0156]
L. Throughput monitoring
In another embodiment, the throughput is monitored. Throughput can be monitored by determining when a substrate enters the processing system, when the substrate exits the processing system, or when it completes a processing phase in the system. Also, the time during which the substrate is in the cooling chamber can be observed and recorded. Using data collected for a plurality of substrates according to this method, peak throughput, average throughput, arrival frequency, frequency variability and other related information can be determined.
[0157]
M. Standardization of processing monitoring
In one embodiment, process monitoring differences are standardized. The processing system 100 has a plurality of OISs 150, each monitoring a different area of processing. Each OIS 150 contributes to the reading of the baseline process, which varies as the substrate is processed. Over time, fluctuations can vary significantly from OIS 150 to OIS 150, which can cause errors in processing degradation readings and alarms. To ensure that alarm errors are minimized, standardized substrates are sent into the processing system and each OIS 150 is counted. Any variation between OIS 150 is normalized to the standardized substrate.
[0158]
In one embodiment, a reference wafer is used to calibrate other processing systems 100. For example, if the processing system 100 has a known value based on a reference substrate, it is considered that the same or similar value is given to another identical processing system.
[0159]
IX. Data processing system
A program product readable by PMC 86 determines which tasks can be performed on the board. Preferably, the program product is software readable by the PMC 86 and includes at least position information, substrate reflectance information, reflection information and / or scattering information, substrate defect information, substrate damage information, smooth substrate or It is preferable to have a code for generating foreign matter contamination information, robot behavior information, calibration information of the robot and the detection device on the substrate on which the pattern is formed, and any combination of these information.
[0160]
In one embodiment, the invention is implemented as a computer program product for use with a control PMC 86. A program (one or more) that defines the functions of the embodiments described herein can be provided to a computer via a signal-bearing medium. As the signal mounting medium, (i) non-writable storage medium information (for example, a read-only memory element in a computer such as a read-only CD-ROM readable by a CD-ROM or a DVD drive); (ii) a writable storage medium ( Changeable information stored on, for example, a floppy disk in a diskette drive or hard disk drive); (iii) information sent to a computer by a communication medium, such as via a computer including a wireless communication or a telephone communication. However, the present invention is not limited to these. If such a signal-bearing medium carries a computer-readable instruction for instructing the functions of the present invention, it is another embodiment of the present invention. Although the components of the product program may be developed and executed individually, they are combined to form an embodiment of the present invention.
[0161]
FIG. 26 is a high-level architecture of one embodiment of a system 2600 configured to perform a foreign object detection or other process monitoring method. Although described as being separate from the PMC 86, the system 2600 may be an embodiment of the PMC 86 or may be integrated with the PMC 86.
[0162]
System 2600 typically includes an application layer 2602, a driver layer 2604, a hardware interface layer 2606, an internal hardware layer 2608, an external hardware layer 2610, and a factory interface layer 2612. Each layer of system 2600 may be a combination of hardware and software adapted to support a certain function. Generally, the application layer 2602, driver layer 2604, hardware interface layer 2606, and internal hardware layer 2608 are components of the optical inspection system (OIS) 150 described above. External hardware layer 2610 represents any system on which OIS 150 is implemented. For example, the external hardware layer 2610 may be the cluster tool described with reference to FIG. The factory interface layer 2612 represents an access point between the peripheral device, the OIS 150, and the external hardware layer 2612. In one embodiment, the factory interface layer 2612 has a host and a data collection server.
[0163]
The application layer 2602 includes a GUI (graphical user interface) 2620, a server process 2622, a foreign object detection process 2624, a process monitoring process 2626, and a database 2628. The GUI 2620 is a task configured to serve as an interface between the user and the server process 2622. In some embodiments, the GUI 2622 includes a dialog box that can be displayed on a monitor that conveys or requests information from a user. In particular, GUI 2622 operates to generate a local request in response to a command issued by the user. The commands are entered into a keyboard, keypad, light pen, touch screen, trackball, or any input device such as a speech recognition unit, audio / video player, or the like.
[0164]
Local requests generated by GUI 2620 are sent to server process 2622 for processing. In addition to receiving requests from local clients, such as GUI 2622, server operation 2622 also receives requests from distant clients. In FIG. 26, the distant client is represented by an external hardware layer 2610 and an external interface layer 2612.
[0165]
In response to various client requests, server operation 2622 performs processing to generate a response or otherwise handle the response. For example, the server process 2622 calls the foreign object detection process 2624 or the process monitoring process 2626 according to a command transmitted by the user sent by the GUI 2620. The foreign substance detection processing 2624 and the processing monitoring processing 2626 will be described later in more detail. The server process 2622 also serves to communicate with other layers of the system 2600. The server operation 2622 also performs initialization of the system 2600. Initialization includes reading a configuration file that contains information about the configuration of system 2600. The configuration file may be stored in the database 2628. Thus, server process 2622 serves as a central information management entity for application layer 2602.
[0166]
Generally, the foreign object detection process 2624 assists in inspecting the board image for foreign objects and creating a report. To this end, the foreign object detection process 2624 executes one or more foreign object detection algorithms. Examples of the algorithm include a “Blob” analysis algorithm and a “pixel” analysis algorithm. The substrate inspected by the process 2624 may be a smooth (unpatterned) substrate or a patterned substrate.
[0167]
The process monitoring process 2626 performs one or more process monitoring processes such as an average intensity analysis. In general, the process monitoring process 2626 assists in inspecting the histogram and generating a comparison plot. Embodiments of the foreign object detection processing 2624 and the processing monitoring processing 2626 will be described below with reference to FIGS.
[0168]
Application tier 2602 communicates with other tiers of system 2600 via one of various interfaces. In the figure, the application layer 2602 has a semiconductor device communication standard (SECS) interface 2630, a reporting device interface 2632, a camera interface 2634, and a light source interface 2636.
[0169]
SECS interface 2630 re-initializes and communicates information between server process 2622 and remote clients. The remote client is represented by an external hardware layer 2610 and an external interface layer 2612.
[0170]
The reporting device interface 2632 is an interface task that serves to generate a report each time one board is inspected and save it to a local or remote disk storage location. In one embodiment, the storage location is part of the factory interface layer 2612.
[0171]
Camera interface 2634 is an interface task configured to support the operation of the above-described receiver unit (as well as other detection devices such as substrate sensors), and is represented by internal hardware layer 2608. In general, the camera interface 2634 assists in board image acquisition, process monitoring and performing foreign object detection, and receiver unit setup. The setup of the receiver unit includes downloading the receiver settings, focusing the receiver, adjusting the position (arrangement) of the receiver, and the like. In addition, the camera interface 2634 starts a request for generating a report handled by the reporting device interface 2632. In operation, the camera interface 2634 receives instructions from the server process 2622 about the operation of the receiver unit. Next, the instruction is sent to the internal hardware layer 2608 via the driver layer 2602 and the hardware layer 2606. The command response is subsequently received from the driver layer 2602 to the camera interface 2634 and sent to the server process 2622. In one embodiment, the driver layer 2604 comprises a National Instrument Image Driver (NIIMAQ) driver and the hardware interface layer 2606 comprises a RS232 and a frame grabber card with TTL ports, each with a camera interface 2634 and an internal hardware interface 2608. Calibrated to support messages routed between
[0172]
Light source interface 2634 is an interface task configured to support the operation of the transmitter unit described above and represented by internal hardware layer 2608. In general, the functions of light source interface 2634 include determining the current light intensity and adjusting the light intensity. In operation, information received from the server process 2622 is transmitted to the driver in the driver layer 2604 via the light source interface 2634 and then to the card in the hardware layer 2606. In one embodiment, the driver is an Omega ADLIB driver and the card is a DI / DO card. Responses from internal hardware layer 2608 are sent to light source interface 2634 in reverse order.
[0173]
Although shown as a single system, the components of the application layer 2602 may be distributed in a networked environment. For example, the GUI 2620 may be located at a workstation that has a server process 2622, a foreign object detection application and a process monitoring application, and is connected to a remotely located computer via a network.
[0174]
FIG. 27 is a flowchart of a program control method 2700 for process monitoring and foreign substance detection using the system 2600. For simplicity, process monitoring is limited to reflection analysis. However, those skilled in the art will appreciate that other process monitoring embodiments of the present invention, including spectral analysis, are applicable.
[0175]
The method 2700 enters step 2705 when the system 2600 is activated. At step 2710, the system 2600 is initialized and is ready to receive a program input event. At step 2712, method 2700 receives the event.
[0176]
At step 2715, the method determines whether a system configuration has occurred. System configuration events include the configuration of the data storage directory, the establishment of defect mapping, the setting of alarms, the defect count threshold and other program settings. If no system configuration event has occurred, the method 2700 proceeds to step 2725, described below. If a system configuration event has occurred, the method 2700 proceeds to step 2720, where the system configuration is obtained and set. At step 2720, adjusting program behavior, setting execution time counters to track program operation, and setting other system configurations via system parameters contained within one or more data structures. I can do it. In one embodiment, the data structure containing the system configuration may be stored in a database 2628 shown in FIG.
[0177]
At step 2725, method 2700 determines whether a system setup event has occurred. If not, the method 2700 proceeds to step 2735. System setup events include receiver calibration (eg, alignment and focus), receiver sampling rate settings, receiver parameter settings, and other system adjustments. If a system setup event has occurred, the method 2700 proceeds to step 2730 to set up the system. As an example, step 2730 includes providing the user with meaningful feedback to adjust the alignment characteristics of the receiver 616 by showing the user the respective statistical information of the reflected image and the signal pattern and having them adjusted. Can be done. Further, at step 2730, the receiver sampling rate is set. In one embodiment, step 2730 includes allowing the user to independently align and adjust the transmitter unit to provide the desired illumination of the substrate.
[0178]
At step 2735, method 2700 determines whether the event is a foreign object detection or process monitoring event. If not, the method 2700 proceeds to step 2745. If the event is for foreign object detection, the method 2700 proceeds to step 2740 to detect foreign objects on the substrate. One embodiment of a method for foreign object detection / processing monitoring is described below with reference to FIG.
[0179]
At step 2745, method 2700 determines whether the event is a termination event, such as a user closing an active application. If so, the method 2700 exits at step 2755. If not, the method 2700 handles the event at step 2557 and then returns to step 2712 to get the next event.
[0180]
FIG. 28 is a flowchart illustrating a method 2800 for foreign object detection and process monitoring. The method 2800 enters from step 2740 to step 2805. At step 2810, the PMC 86 operates to detect an incoming substrate. One method of substrate detection has been described above with reference to FIG. At step 2815, the substrate is scanned or illuminated and the resulting signal pattern is stored. Examples of methods of scanning / irradiating the substrate have been described above.
[0181]
At step 2820, method 2800 generates a histogram of pixel intensities (also referred to herein as a current histogram) based on the information collected at step 2815. Further, an average intensity value for the substrate is calculated.
[0182]
Thereafter, method 2800 proceeds along two logical paths according to the desired process. For foreign object detection, method 2800 proceeds along logic path 2822 to step 2825. For process monitoring, method 2800 proceeds along logical path 2824 to step 2850.
[0183]
As the method 2800 proceeds along the path 2822, the information is binarized at step 2825. Binarization is a binarized representation of a grayscale substrate image represented as an array of pixel values, for example, from 0 to 255, where 0 is black and 255 is white. The gray scale image is binarized by selecting a threshold intensity such that a value smaller than the threshold is black and a value larger than the threshold is white. The value of the threshold can be determined empirically according to the particular application.
[0184]
Further, in step 2825, a morphological operation is performed to manipulate and enhance the image obtained by filtering noise and other extraneous signal information. Noise is caused by deformation or vibration of the substrate, such as by accidental variations in signal scattering / reflection, such as power modulation, for example, when the substrate is not completely planarized or immobilized. Good. In addition, changes and contamination in the chamber, such as foreign matter floating in the chamber, optical deterioration of the window, and heat flow in the chamber, also change with time, thus causing noise. Further, electrical noise (for example, electronic noise, white noise, pink noise, etc.) is also provided from the transmitter unit 56 and the receiver unit 58.
[0185]
Signal noise filtering methodologies are known, but vary according to the type of noise being filtered. Methods include filtering the received signal using digital signal processing (DSP), electronic filters (eg, low-pass filters, high-pass filters, etc.), signal sampling, and the like.
[0186]
Image enhancement may be achieved using known spectral selection algorithms that can enhance image contrast and color, or by blocking unwanted spectra with spectral filters. Further, image enhancement may be achieved by a DSP or other digital enhancement techniques known in the art.
[0187]
At step 2830, method 2800 searches for a substrate image for the foreign object based on the illumination intensity and size. In the Blob analysis, the intensity and proximity of a pixel are used as means for specifying the size and position of a foreign substance in a region on a substrate. Steps 2825 and 2830 constitute a "blob" analysis.
[0188]
Then, the determination result of the “blob” analysis is analyzed in step 2835. In step 2835, the state of the substrate is rendered. In one embodiment, the state of the substrate includes a passing state, a warning state, a failure state, and the like. The user is reported the status, for example, by an audible or visible signal. Upon analyzing the histogram data and the foreign matter data, the method 2800 proceeds to step 2840, where a report including an output board image, a histogram data report and a defect summary is generated for the end user. Thereafter, the method exits step 2845 and returns to method 2700.
[0189]
Returning to step 2820, if it is chosen to proceed along logical path 2824, method 2800 proceeds to step 2850. At step 2850, the current histogram data generated at step 2820 is plotted. At step 2852, the histogram plot is compared to the reference histogram and the result is plotted to represent the difference between the current histogram and the reference histogram. At step 2854, method 2800 plots the mean intensity trend of the current histogram relative to the reference histogram. An example of a graph showing this tendency has been described above with reference to FIG. At step 2856, the results of steps 2852 and 2854 are analyzed to determine the state of the substrate. In one embodiment, the substrate state includes a pass state, a warning state, and a failure state. The method 2800 then proceeds to step 2840, where a report is created that can be viewed by the end user. The report may include an output board image, a histogram data report and a summary of the defects.
[0190]
In one embodiment, an alarm is raised to the user when a predetermined event occurs indicating a defective substrate. The user can define an alarm criterion based on the number of occurrences above a user-defined intensity threshold. Thus, the alarm definition threshold can be moved closer to the noise floor created by the patterned substrate. The sum of occurrences of intensity above the threshold is summed and compared to a user defined count threshold. If the total number of occurrences exceeds the count value, an alarm is generated. For example, the intensity threshold level may be set such that 3,500 counts (+/− 50 counts) are accumulated for the entire image. When it encounters contaminated water and reaches 3,555 counts, an alarm condition will occur with 5 counts more than 3,500 counts. As the count becomes greater than the count threshold, the reliability of the alarm increases accordingly. This represents the secondary decision quality score and can be used to establish a confidence interval.
[0191]
In another embodiment, an automatic mode is performed to define alarm and warning thresholds. In the automatic mode, statistics such as the mean and standard deviation of one or more reference substrates were used. The alarm threshold is based on some multiple of the standard deviation that is selected by the user or preset based on empirical data. Since the standard deviation in smooth wear is much smaller than in patterned wear, the detection threshold will be much closer to the average than in patterned wear. Such an automated method avoids any undesired effects caused by subjective input by the user setting of alarms and warning thresholds.
[0192]
While the preferred embodiment of the present invention has been described, other embodiments and alternative embodiments of the invention may be devised without departing from the basic scope of the invention, and the scope is defined by the following claims. To decide.
[Brief description of the drawings]
FIG. 1A
FIG. 1A is a plan view of a typical processing system for semiconductor processing in which the present invention can be effectively used.
FIG. 1B
FIG. 1B is a diagram showing a high-level processing inspection system.
FIG. 1C
FIG. 1C illustrates a processing inspection system comprising one embodiment of the present invention having a plurality of optical inspection systems connected to an optical signal multiplexer in communication with a receiving device for use in the present invention.
FIG. 2
FIG. 2 is a perspective view of a factory interface used in the present invention, comprising an embodiment of the present invention comprising two transmitter units and one receiver unit for use in the present invention.
FIG. 3A
FIG. 3A is a top view of the processing system of FIG. 2 showing a specific position of a substrate placed on the pod blade during linear movement of the blade.
FIG. 3B
FIG. 3AB is a top view of the processing system of FIG. 2 showing a specific position of a substrate placed on the pod blade during linear movement of the blade.
FIG. 3C
FIG. 3C is a top view of the processing system of FIG. 2 showing a particular location of a substrate placed on the pod blade during linear movement of the blade.
FIG. 4
FIG. 4 is a perspective view of a factory interface used in the present invention, which comprises one embodiment of the present invention comprising one transmitting device unit and one receiving device unit.
FIG. 5A
FIG. 5A is a top view of the processing system of FIG. 4, showing a particular location of a substrate placed on the pod blade during linear movement of the blade.
FIG. 5B
FIG. 5B is a top view of the processing system of FIG. 4, showing a particular location of a substrate placed on the pod blade during linear movement of the blade.
FIG. 5C
FIG. 5C is a top view of the processing system of FIG. 4 showing a specific position of a substrate placed on the pod blade during linear movement of the blade.
FIG. 6
FIG. 6 is a partial perspective view of a transfer chamber used in the present invention, which comprises one embodiment of the present invention comprising one transmitting device and one receiving device unit.
FIG. 7A
FIG. 7A is a top view of the processing system of FIG. 6, showing a particular position of a substrate placed on the blade during rotation of the blade.
FIG. 7B
FIG. 7B is a top view of the processing system of FIG. 6, showing a particular position of a substrate placed on the blade during rotation of the blade.
FIG. 7C
FIG. 7C is a top view of the processing system of FIG. 6, showing a particular position of a substrate placed on the blade during rotation of the blade.
FIG. 8A
FIG. 8A is a top view of the processing system of FIG. 6, showing a particular location of a substrate placed on the pod blade during linear movement of the blade.
FIG. 8B
FIG. 8B is a top view of the processing system of FIG. 6, showing a particular location of a substrate placed on the pod blade during linear movement of the blade.
FIG. 8C
FIG. 8C is a top view of the processing system of FIG. 6, showing a particular location of a substrate placed on the pod blade during linear movement of the blade.
FIG. 9
FIG. 9 is a cross-sectional view of the chamber and lid assembly illustrating one embodiment of the optical inspection system.
FIG. 10
FIG. 10 is a cross-sectional view of the chamber and lid assembly illustrating one embodiment of the optical inspection system.
FIG. 11
FIG. 11 is a perspective cross-sectional view of a chamber and lid assembly illustrating one embodiment of an optical inspection system.
FIG. 12A
FIG. 12A is a cross-sectional view of the chamber and lid assembly of FIG. 11 showing a particular location of a substrate placed on the blade during linear movement of the blade.
FIG. 12B
FIG. 12B is a cross-sectional view of the chamber and lid assembly of FIG. 11 showing a particular location of a substrate placed on the blade during linear movement of the blade.
FIG. 12C
FIG. 12C is a cross-sectional view of the chamber and lid assembly of FIG. 11 showing a particular location of a substrate placed on the blade during linear movement of the blade.
FIG. 13A
FIG. 13A is a cross-sectional view of a chamber and lid assembly illustrating one embodiment of an optical inspection system, showing specific locations of a substrate placed on a blade during a substrate surface scanning sequence.
FIG. 13B
FIG. 13B is a cross-sectional view of the chamber and lid assembly illustrating one embodiment of the optical inspection system, showing specific locations of the substrate placed on the blade during a substrate surface scanning sequence.
FIG. 13C
FIG. 13C is a cross-sectional view of the chamber and lid assembly illustrating one embodiment of the optical inspection system, showing specific locations of the substrate placed on the blade during a substrate surface scanning sequence.
FIG. 14
FIG. 14 is an embodiment of a movement diagram of the substrate detection system used in the present invention.
FIG.
FIG. 15 is a graph of the reflection intensity distribution of the reflection from the patterned substrate irradiated by the light source.
FIG.
FIG. 16 is a comparison graph of the reflection intensity distribution on the pattern-formed substrate.
FIG.
FIG. 17 is a plot showing the reflection trace difference for an overetch time of 40 seconds.
FIG.
FIG. 18 is an etching end point plot based on a change in the processing plasma intensity.
FIG.
FIG. 19 is a plot showing the change in average intensity value for various overetch times.
FIG.
FIG. 20 is a plot showing delta average intensity values for successive test substrates.
FIG. 21
FIG. 21 is a plot showing the change in the reflection trace with the time of stripping the etching photoresist.
FIG.
FIG. 22 is a plot showing an intensity plot for the peel time for the reference substrate.
FIG. 23
FIG. 23 is a plot of the intensity scan for strip time after the photoresist has been removed.
FIG. 24
FIG. 24 is a graph of a plot of the repeatability difference for determining system noise.
FIG. 25
FIG. 25 is an overall substrate surface view for spectrum analysis.
FIG. 26
FIG. 26 is a diagram illustrating a high-level architecture of one embodiment of a system configured to perform foreign object detection and other process monitoring methods.
FIG. 27
FIG. 27 is a flowchart of a program control method for processing monitoring and foreign substance detection using the system.
FIG. 28
FIG. 28 is a flowchart for processing monitoring and processing report generation.
[Explanation of symbols]
56 light sources
58 Receiver
86 PMC (Process monitoring controller)
90 cable
100 processing system
102 Processing system controller
104 Factory Interface
105 pod
106 load lock chamber
108A, 108B Pod loader
110 transfer chamber
113 Robot
114 Processing Chamber
116 Service Chamber
120 viewports
122 arrow
124 arrow
135 Integrated Measurement / Process Inspection System
136 arrow
137 Inspection system controller
138 Inspection system controller
141 signal exchange unit
143A, 143H Optical fiber cable
150 OIS
155 auxiliary user interface
159 Factory interface controller
162 Substrate manufacturing (data server)

Claims (21)

A controller system coupled to an inspection platform configured to perform the optical inspection process at a first degree of optical resolution and a plurality of optical inspection systems configured to perform the optical inspection process at a second degree of optical resolution. Wherein each of the plurality of optical inspection systems is located at a different location on the substrate processing system, the method comprising:
Receiving, from one of the plurality of optical inspection systems, a processing data read comprising optical signal signature information indicative of a phase state on a substrate to be inspected by the one optical inspection system;
Processing the processing data read to determine a next substrate handling step;
A method comprising: Processing the processing data read to determine the next substrate handling step:
Determining whether the processing data reading exceeds a predetermined value;
Determining that an unacceptable phase state exists on the substrate if the processing data reading exceeds a predetermined value;
Transferring the substrate to an inspection platform;
The method of claim 1, comprising: Optical signal signature information includes substrate reflectance information, reflection information, spectrum information, substrate defect information, substrate damage information, foreign matter contamination information, alphanumeric character information, non-uniform plasma growth, and any of these. 2. The method of claim 1, comprising at least one of the following combinations: The method of claim 1, wherein processing the processing data read to determine a next substrate handling step further comprises determining whether the substrate should be transferred to an inspection platform for optical inspection. The method according to claim 1, wherein the next processing step is a processing end step. The method of claim 1, wherein processing the processing data read to determine a next substrate handling step comprises determining a location of the substrate in the processing system and determining a routing sequence for the substrate. The described method. A plurality of optical inspection systems, each configured to perform an optical inspection process at a first degree of optical resolution, each comprising a transmitter unit and a receiver unit;
An inspection platform configured to perform an optical inspection process at a second degree of optical resolution;
Connected to multiple optical inspection systems and inspection platforms,
(I) processing optical signal information indicative of a phase state on a substrate to be inspected by at least one of the plurality of optical inspection systems;
(Ii) a controller system configured to perform one of a plurality of substrate handling steps according to a phase state, wherein the first substrate handling step inspects the substrate for further optical inspection; A substrate processing inspection system comprising: The system of claim 7, wherein each of the optical inspection systems is disposed on a processing system along a substrate transfer path. The system of claim 7, wherein the receiver unit comprises at least one of a charge coupled device (CCD) camera and a spectrometer. The system according to claim 7, further comprising an input unit for inputting control information used to operate the controller system. The controller system is
Determining whether optical inspection data collected by at least one of the plurality of optical inspection systems exceeds a predetermined value;
If so, by determining that an unacceptable substrate processing condition exists on the substrate,
The system of claim 7, wherein the system is configured to perform a first substrate handling step. Optical inspection data includes substrate reflectance information, reflection information, spectrum information, substrate defect information, substrate damage information, foreign matter contamination information, alphanumeric character information, non-uniform plasma growth, and any of these. The system of claim 11, comprising a combination. The system of claim 11, wherein the controller system is configured to initiate a system shutdown process sequence if an unacceptable substrate processing condition exists on the substrate. A processing system comprising a cluster tool and an optical inspection system, the optical inspection system comprising:
(A) a plurality of units each comprising a transmitter unit and a receiver unit, each configured to perform an optical inspection process at a first degree of optical resolution, each being located at a different location on the cluster tool; An optical inspection system;
(B) an inspection platform connected to the cluster tool and configured to perform an optical inspection process at a second degree of optical resolution;
(C) connected to a plurality of optical inspection systems and inspection platforms,
(I) processing optical signal information indicative of a phase state on a substrate to be inspected by at least one of the plurality of optical inspection systems;
(Ii) a controller system configured to execute one of a plurality of substrate handling steps according to a phase state, wherein the first substrate handling step inspects the substrate for further optical inspection; Transferring to a platform;
(D) an input device configured to allow an operator to operate the controller. The cluster tool includes a transfer chamber and a processing chamber connected to the transfer chamber, wherein at least one of the plurality of optical inspection systems is located on the transfer chamber and at least one of the plurality of optical inspection systems is on the processing chamber. The system according to claim 14, wherein the system is installed on a computer. 15. The system of claim 14, wherein each of the optical inspection systems is disposed on a processing system along a substrate transfer path. 15. The system of claim 14, wherein the receiver unit comprises at least one of a charge coupled device (CCD) and a spectrometer. 15. The system according to claim 14, further comprising an input unit for inputting control information used to operate the controller system. The controller system is
Determining whether optical inspection data collected by at least one of the plurality of optical inspection systems exceeds a predetermined value;
If so, by determining that an unacceptable substrate processing condition exists on the substrate,
15. The system of claim 14, wherein the system is configured to perform a first substrate handling step. Optical inspection data includes substrate reflectance information, reflection information, spectrum information, substrate defect information, substrate damage information, foreign matter contamination information, alphanumeric character information, non-uniform plasma growth, and any of these. 20. The system of claim 19, comprising a combination. 20. The system of claim 19, wherein if an unacceptable substrate processing condition exists on the substrate, the controller system is configured to initiate a shut down process sequence to remove the substrate from the processing system.

Images