JP2007016320A - Electropolishing assembly and method for electropolishing conductive layer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus and a method for electropolishing a conductive film on a wafer. <P>SOLUTION: The apparatus includes a wafer chuck 1002 for holding the wafer, an actuator 1000 for rotating the wafer chuck and a nozzle 1010 configured to electropolish the wafer. The apparatus may further include a conductive ring or a shroud 1006. The method for electropolishing the conductive film on the wafer comprises a step wherein the wafer chuck is rotated at a sufficient speed such that electrolyte fluid incident upon the wafer flows on the surface of the wafer towards the edge of the wafer. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates generally to semiconductor processing equipment, and more specifically to an electropolishing apparatus for electropolishing a conductive layer on a semiconductor device.

  Transistor elements and interconnect elements are formed by manufacturing or processing semiconductor devices on a semiconductor wafer using a number of different processing steps. In order to electrically connect the transistor terminals associated with the semiconductor wafer, conductive (eg, metal) trenches and vias are formed in the dielectric material as part of the semiconductor device. The trenches and vias couple electrical signals and outputs between transistors, semiconductor device internal circuitry, and circuitry external to the semiconductor device.

  When forming the interconnection element, for example, a masking process, an etching process, and a deposition process are performed on the semiconductor wafer, thereby forming a desired electronic circuit of the semiconductor device. Specifically, by performing a plurality of masking steps and etching steps, it is possible to form a pattern composed of recessed regions in the dielectric layer on the semiconductor wafer. These recessed areas serve as trenches and vias for the interconnects. A deposition process is then performed to deposit a metal layer over the semiconductor wafer, thereby depositing the metal in both the trench and via, and also on the non-recessed area of the semiconductor wafer. In order to insulate interconnects, such as patterned trenches and vias, the metal deposited on non-recessed areas of the semiconductor wafer is removed.

  Conventional methods for removing metal films deposited on non-recessed regions of a dielectric layer on a semiconductor wafer include, for example, chemical mechanical polishing (CMP). The CMP method is widely used in the semiconductor industry to form interconnect lines by polishing and planarizing the metal layers within the trenches and vias along with the non-recessed regions of the dielectric layer.

  For the CMP method, the wafer assembly is positioned on a CMP pad placed on a platen or web. A wafer assembly includes a substrate having one or more layers and / or components, for example, interconnect elements formed in a dielectric layer. A force is then applied to crimp the wafer assembly to the CMP pad. While applying a force for polishing and flattening the wafer surface, the CMP pad and the substrate assembly are pressed against each other and moved relative to each other. Polishing is facilitated by dispensing a polishing solution, often referred to as a polishing slurry, onto the CMP pad. The polishing slurry typically contains an abrasive and can be chemically reacted to selectively remove unwanted materials, such as metal layers, from the wafer more quickly than other materials, such as dielectric materials. .

  However, the CMP method can have some detrimental effects on the underlying semiconductor structure. This is because a relatively strong mechanical force is involved. For example, as the interconnect geometry goes below 0.13 μm, there can be a large difference between the mechanical properties of conductive materials, such as copper, and the mechanical properties of low-k films used in typical damascene processes. For example, the low k dielectric film has a Young's modulus that is 10 orders of magnitude lower than that of copper. As a result, applying relatively strong mechanical forces to the dielectric film and copper, particularly during CMP processing, can lead to stress related defects on the semiconductor structure. These defects include delamination, dishing, erosion, film lift or scratches, and the like.

  Therefore, a new processing technique is desired. For example, the metal layer can be removed or etched from the wafer using electropolishing. In general, in the case of electropolishing, the portion of the wafer to be polished is immersed in an electrolyte fluid solution, and then a charge is applied to the wafer. These conditions result in copper being removed or polished from the wafer.

  In one aspect of the present invention, one embodiment of an apparatus and method for electropolishing a conductive film on a wafer is provided. One embodiment of the apparatus includes a wafer chuck for holding a wafer, an actuator for rotating the wafer chuck, a nozzle configured to electropolish the wafer, and positioning about an edge of the wafer Containing shroud. One embodiment of a method for electropolishing a conductive film on a wafer comprises rotating the wafer chuck at a rate sufficient to allow electrolyte fluid incident on the wafer to flow over the wafer surface toward the wafer edge. Including.

  The present invention is better understood upon consideration of the detailed description below in conjunction with the accompanying drawings and claims.

  Numerous specific details, such as specific materials, parameters, etc., are set forth in the description below to provide a more thorough understanding of the present invention. However, it should be understood that this description is not intended to limit the scope of the invention, but is provided to better illustrate the embodiments.

I. Examples of electropolishing apparatus:
1A and 1B are a cross-sectional view and a plan view showing an embodiment of a wafer electrolytic polishing apparatus that can be used to polish a wafer 1004. FIG. Roughly speaking, the electropolishing apparatus of this embodiment works by directing the flow of electrolyte fluid toward the metal film on the wafer while a charge is applied to the wafer. Due to the charge and the electrolyte fluid, the metal ions in the metal film are dissolved in the electrolyte fluid. The current density of the electrolyte fluid and the metal ion concentration in the electrolyte fluid at least partially determine the polishing rate. Therefore, by controlling the current density, the electrolyte solution concentration, etc., the electrolytic polishing apparatus can accurately polish the metal layer disposed on the semiconductor wafer.

  As shown in FIG. 1A, the electrolytic polishing apparatus can include a chuck 1002, an actuator 1000, and a polishing container 1008. Polishing vessel 1008 can be any material that is electrically insulated and resistant to acid and corrosion, such as polytetrafluoroethylene (commercially referred to as TEFLON®), polychlorinated. It can be formed from vinyl (PVC), polyvinylidene fluoride (PVDF), polypropylene, and the like. Preferably, the polishing container 1008 can be formed from PVDF. Needless to say, however, the polishing container 1008 can be made of different materials depending on the application.

  As shown in FIG. 1A, an electrolyte fluid 1038 can flow into the polishing vessel 1008 through nozzles 1010, 1012 and / or 1014. More specifically, pump 1020 pumps electrolyte fluid 1038 from electrolyte fluid reservoir 1070 through return valve 1024 to pass filter 1018. The pass filter can include a liquid mass flow controller (LMFC). These LMFCs can control the amount and rate of electrolyte fluid 1038 supplied to nozzles 1010, 1012 and 1014. In addition, the pass filter 1018 may filter contaminants from the electrolyte fluid 1038, which may enter the polishing vessel through the nozzles 1010, 1012 or 1014 and degrade the electropolishing process or be used by the LMFC If done, the amount of contaminants that could clog the LMFC can be reduced. In this way, contaminants are prevented from entering the polishing container 1008 and / or clogging the LMFC. In this embodiment, pass filter 1018 suitably removes particles greater than about 0.05 to about 0.1 μm. However, it will be appreciated that various filtration systems can be used depending on the particular application. In addition, while contaminant filtration is advantageous, the pass filter 1018 can be omitted from the wafer polishing assembly in some applications.

The electrolyte fluid 1038 can include any convenient electropolishing fluid, such as phosphoric acid. Preferably, the electrolyte fluid 1038 includes orthophosphoric acid (H ₃ PO ₄ ) at a concentration of about 60 wt% to about 85 wt%, preferably about 76 wt%. Further, the electrolyte fluid 1038 preferably comprises about 10% to about 40% by weight glycol, with the balance comprising about 1% aluminum metal (based on acid weight). However, the concentration and composition of the electrolyte fluid 1038 may vary depending on the particular application.

  Pump 1020 may include any suitable hydraulic pump, such as a centrifugal pump, a diaphragm pump, a bellows pump, and the like. Further, the pump 1020 may be resistant to acid, corrosion and contamination. Although a single pump 1020 is shown, it will be appreciated that any number of pumps 1020 may be used. For example, separate pumps may be used for each nozzle 1010, 1012 and 1014. In addition, in some applications, electrolyte fluid 1038 can flow into polishing vessel 1008 through nozzles 1010, 1012, and 1014 without pump 1020. For example, the electrolyte fluid 1038 can be maintained at a predetermined pressure within the electrolyte fluid reservoir 1070. Alternatively, the supply conduit between the electrolytic fluid reservoir 1070 and the nozzles 1010, 1012 and 1014 can be maintained at a predetermined pressure.

  The LMFC can include any convenient mass flow controller. The mass flow controller is further preferably resistant to acid, corrosion and contamination. In addition, the LMFC can suitably supply the electrolyte fluid 1038 at a flow rate proportional to the cross-sectional area of the nozzles 1010, 1012 and 1014. For example, if the diameter of nozzle 1012 is larger than nozzle 1014, the LMFC may advantageously provide electrolyte fluid 1038 to nozzle 1012 at a higher flow rate. In this embodiment, the LMFC is configured to supply the electrolyte fluid 1038 at a flow rate of 0.5 liters to 40 liters per minute, depending on the size of the nozzle and the distance between the nozzle and the wafer. preferable.

  The electrolyte fluid reservoir 1070 further includes a heat exchanger 1036, a cooler / heater 1034, and a temperature sensor 1032 to control the temperature of the electrolyte fluid 1038 in the electrolyte fluid reservoir 1070. Further, one or more electrodes 1028 may be included in the electrolyte fluid reservoir 1070, which may be connected to a power source 1030. By applying a charge to the electrode 1028, metal ions are removed from the electrolyte fluid 1038, thereby adjusting the metal ion concentration of the electrolyte fluid 1038. A counter charge can be applied to the electrode 1028 to add metal ions to the electrolyte fluid 1038.

  This embodiment of the wafer polishing assembly further includes electrodes disposed within nozzles 1012 and 1014. As described in more detail below, this example includes two nozzles with two electrodes therein, but the number of nozzles used and the number of electrodes per nozzle is less than two. There may be any number, whether or not. In general, as the surface area of the electrode inside the nozzle increases, the current density and electropolishing rate across the flow profile of the electrolyte fluid 1038 increases.

  As shown in FIGS. 1D and 1E, nozzle 1012 includes an electrode 1056 and nozzle 1014 includes an electrode 1060. Electrodes 1056 and 1060 may include any conductive material such as copper, stainless steel, tantalum (Ta), titanium (Ti), TaN, TiN, lead and platinum.

  During the electropolishing process, some of the metal ions migrated from the metal layer on the wafer 1004 may accumulate on the electrodes 1056 and 1060. As described in more detail below, the metal deposit or metal plating can be removed in a plating removal process. For example, when electrodes 1056 and 1060 are positively charged and wafer 1004 is negatively charged, wafer 1004 is electroplated without being electropolished. Thus, and similarly, the metal plated on the electrodes 1056 and 1060 can be removed, ie, plated. Alternatively, the electrodes 1056 and 1060 can be replaced at any suitable time. For example, after processing about 100 wafers, the electrodes 1056 and 1060 can be replaced.

  In some embodiments, the metal layer can include copper. Thus, during the electropolishing process, some of the copper ions from the metal layer being polished migrate to electroplate electrodes 1056 and 1060. However, if the electrodes 1056 and 1060 include copper, the electrodes 1056 and 1060 may dissolve and deform during the plating removal process. Thus, in some embodiments, it is desirable that the electrodes 1056 and 1060 comprise a material that is resistant to being dissolved during the plating removal process. For example, electrodes 1056 and 1060 may include platinum or a platinum alloy. Alternatively, the electrodes 1056 and 1060 may comprise titanium, for example, suitably coated with a platinum layer having a thickness of about 50 μm to about 400 μm.

  In the case of the apparatus of this embodiment, the wafer chuck 1002 suitably holds and positions the wafer 1004 in the polishing container 1008 or above the polishing container 1008. More specifically, the wafer 1004 is preferably positioned within the shroud 1006 opposite the nozzles 1010, 1012 and 1014. A shroud 1006 is optionally included around the wafer 1004 to prevent splashing and the like, as will be described in more detail below.

  After the wafer 1004 is suitably positioned in the polishing container 1008, the electrodes 1056 and 1060 are charged by the power source 1040. Further, the wafer 1004 is also charged by the power source 1040. Alternatively, two or more power sources can be used to charge the electrodes 1056 and 1060 and the wafer 1004. When properly charged and the electrolyte fluid 1038 flows between the electrodes 1056 and 1060 inside the nozzles 1012 and 1014 and the surface of the wafer 1004, an electrical circuit is formed. More specifically, the electrodes 1056 and 1060 have a negative potential compared to the wafer 1004 by being charged. In response to this negative potential at electrodes 1056 and 1060, metal ions move away from wafer 1004 and move into electrolyte fluid 1038, thus electropolishing wafer 1004. However, when the polarity of the circuit is reversed, the metal ions move toward the wafer 1004, thus electroplating the wafer 1004.

  Further, as shown in FIGS. 1A and 1C, the nozzle 1010 includes an injection nozzle 1052 and an end point detector 1016. During the electropolishing process, the spray nozzle 1052 can be configured to supply the electrolyte fluid 1038 and the endpoint detector 1016 can be configured to detect the thickness of the metal layer on the wafer 1004. The endpoint detector 1016 can include various sensors, such as an ultrasonic sensor, a light reflection sensor, and an electromagnetic sensor (eg, eddy current sensor). The electrolyte fluid 1038 supplied by the injection nozzle 1052 can act as a medium in which the endpoint detector 1016 emits a signal to measure the thickness of the metal film. By using the electrolyte fluid 1038 as a single medium for transmitting the signal, the accuracy of the measured value determined by the endpoint detector 1016 is increased. This is because the electrolyte fluid 1038 provides a single phase. In contrast, if the injection nozzle 1052 does not provide the electrolyte fluid 1038, the emitter and measurement body from the endpoint detector will pass through the electrolyte fluid 1038 applied to the wafer 1004 by the nozzle 1012 or nozzle 1014. Various other media, such as air, can pass through. As described below, having updated or real-time characteristics of the electrolyte fluid 1038 that can change over time can also increase the accuracy of the endpoint detector 1016. Further, although one nozzle 1010 is illustrated as having an endpoint detector 1016, any number of nozzles having any number of endpoint detectors may be used.

  Further, as shown in FIG. 1A, the actuator 100 can rotate the chuck 1002 and the wafer 1004 about the z axis. Further, in some applications, the actuator 100 can move the chuck 1002 and the wafer 1004 along the x-direction while the nozzles 1010, 1012 and 1014 remain stationary. For other applications, the nozzles 1010, 1012 and 1014 are movable along the x direction, whereas the chuck 1002 and wafer 1004 remain stationary along the x direction. In yet another application, the actuator 1000 can move the chuck 1002 and the wafer 1004 along the x direction, while the nozzles 1010, 1012 and 1014 also move along the x direction.

  Furthermore, the electroplating apparatus can be oriented in other ways. For example, nozzles 1010, 1012 and 1014 are positioned above wafer 1004, thereby directing electrolyte fluid downward toward wafer 1004. In addition, the wafer 1004 can be oriented vertically such that the nozzles 1010, 1012, and 1014 direct electrolyte fluid toward the wafer 1004.

  For additional discussion of exemplary wafer electroplating equipment, see US Pat.No. 6,395,152 dated July 2, 1999 (titled `` METHODS AND APPRATUS FOR ELECTROPOLISHING METAL INTERCONNECTIONS ON SEMICONDUCTOR DEVICES See Method and apparatus for electropolishing))). The entirety of the above specification is incorporated herein by reference. In addition, for additional discussion of endpoint detector examples, see U.S. Patent Application Publication No. 6,447,688 (titled `` METHODS AND APPRATUS FOR END-POINT DETECTION '' dated May 12, 2000). Device))). The entirety of the above specification is incorporated herein by reference.

II. Protection against Bounce of Electrolyte Fluid An example of an electropolishing method includes rotating the wafer 1004 while the electrolyte fluid 1038 is directed to the surface of the wafer 1004. Wafer 1004 is rotated at a speed sufficient to create a centrifugal force. Due to this centrifugal force, the incident electrolyte fluid 1038 flows across the surface of the wafer 1004 toward the edge of the wafer 1004. Preferably, the electrolyte fluid 1038 flows to the edge of the wafer 1004 before falling from the surface. Directing the flow across the surface of the wafer 1004 prevents the fluid from falling off the wafer surface and disturbs the flow of the electrolyte fluid 1038, or a continuous column of electrolyte fluid in the polishing vessel 1008 ( column) can be prevented. However, this process can cause the electrolyte fluid to splash within the container and escape from the device or disrupt the electrolyte fluid flow. Accordingly, an embodiment of the electropolishing apparatus includes a shroud 1006 positioned around the wafer 1004 so that liquid under centrifugal action may splash or otherwise move away from the polishing container 1008. Reduce or prevent escape.

  1A and 1B show a shroud 1006 configured to surround the wafer 1004 and the chuck 1002. As shown in FIG. 1A, the nozzle 1012 can provide a flow of electrolyte fluid to the surface of the wafer 1004. In order to polish the metal film on the wafer 1004 more uniformly, the wafer 1004 is allowed to flow across the wafer 1004 to the exposed portion of the chuck 1002, and in this case, the electrolyte fluid can be removed from the surface of the wafer 1004. It can be rotated so that it does not fall into the polishing container 1008. If the electrolyte fluid falls from the wafer 1004 and forms a continuous column of electrolyte fluid between the wafer 1004 and the polishing vessel 1008, any such electrolyte fluid will be exposed to the wafer where the column is formed. May cause excessive polishing of 1004. This additional polishing can lead to non-uniform and unpredictable polishing rates of the metal layer.

  In addition, any electrolyte fluid that falls from the wafer 1004 or splashes in the polishing vessel 1008 can interfere with the flow of electrolyte fluid supplied by the nozzle 1012. The shape of the electrolyte fluid 1038 flow, or more specifically, the flow profile of the electrolyte fluid 1038 affects the current density and polishing rate of the electropolishing apparatus. Accordingly, it is desirable to have electrolyte fluid 1038 flow along the surface of wafer 1004 toward the edge of wafer 1004 and away from the flow of electrolyte fluid 1038 directed at wafer 1004.

Depending on the viscosity of the electrolyte fluid used, the wafer 1004 can be rotated at an appropriate rotational speed so that the electrolyte fluid can flow across the wafer 1004 toward the edge of the wafer 1004. The rotational speed is such that the electrolyte fluid 1038 flows across the wafer 1004 and does not fall from the surface of the wafer 1004 to form a continuous column or interfere with the flow of the electrolyte fluid 1038. Should be. Specifically, the lower the viscosity of the electrolyte fluid, the higher the centrifugal acceleration is required. For example, if 85% of the phosphoric acid, the centrifugal acceleration can be selected to be greater than about 1.5 meters / second ^2. In the example method, a 300 mm diameter wafer is rotated in the range of about 100 revolutions per minute (rpm) to about 2,000 rpm or more, preferably in the range of about 1,500 rpm to about 2,000 rpm.

  Typically, nozzle 1012 or 1014 scans the entire surface of wafer 1004, thereby polishing wafer 1004 more uniformly. By rotating the wafer 1004, a constant centrifugal acceleration can be formed on the incident electrolyte fluid 1038 when the nozzle 1012 is scanning different parts of the wafer 1004. For example, centrifugal acceleration is directly proportional to the radial distance from the wafer center and the square of the rotational speed. Thus, the rotational speed of the wafer 1004 can be reduced when the nozzle 1012 or 1014 is polishing a portion of the wafer 1004 near the edge of the wafer 1004, i.e. a large radius, and near the center of the wafer 1004. This can be increased when polishing a portion, ie a small radius.

  Typically, when electrolyte fluid is supplied to the wafer 1004 as described above, the electrolyte fluid flows toward the edge of the wafer 1004 and further through the edge of the wafer 1004 to the wall of the polishing vessel 1008. Can flow toward. If the shroud 1006 is not provided, the electrolyte fluid 1038 will contact the wall of the polishing vessel 1008 and jump within the vessel, thereby disturbing the flow of the electrolyte fluid 1038 or escaping from the polishing vessel 1008. There is.

  As shown in FIGS. 1A and 1B, a shroud 1006 is placed around the wafer 1004 and the chuck 1002, thereby reducing electrolyte fluid 1038 from splashing or escaping from the polishing container 1008. Or it can be prevented. Furthermore, the shroud 1006 can move together in the x direction with the chuck 1002 and the actuator 1000 during the polishing process. Specifically, the shroud 1006 can be attached to the chuck 1002 and / or the actuator 1000 by a mechanical attachment, a joint, or the like. Alternatively, another actuator that synchronizes the movement of the shroud 1006 with the chuck 1002 and the actuator 1000 can drive the shroud separately. The shroud 1006 can be rotated simultaneously with the chuck 1002 or in other ways.

  The shroud 1006 can be formed in any suitable shape, such as circular and polygonal. Preferably, the shroud 1006 is shaped to reduce electrolyte fluid splash after the electrolyte fluid 1038 flows out of the wafer 1004 and to contain the electrolyte fluid 1038 in the polishing vessel 1008. The gap between the chuck 1002 and the shroud 1006 may be in the range of about 1 mm to about 10 mm, preferably about 5 mm. In addition, as shown in FIG. 1A, by forming the cross section of the side wall of the shroud 1006 in an L shape, it is possible to prevent the electrolyte fluid from splashing and exceeding the shroud 1006 or the chuck 1002. However, the cross section of shroud 1006 can have a variety of other shapes. For example, the side wall of the shroud 1006, that is, the vertical portion of the L shape can be formed in other shapes such as a C shape. Furthermore, by shredding the shroud 1006 inward or outward, splashing or the like can be reduced. The shroud 1006 may extend further above or below the wafer 1004 and chuck 1002 than that shown in FIG. 1A.

  The shroud 1006 can be formed from plastics and ceramics, or from corrosion resistant metals or alloys such as tantalum, titanium, and 300 series stainless steel. Further, the shroud 1006 can be coated with an electrolyte fluid-resistant material such as Teflon (registered trademark).

  Needless to say, however, the electropolishing method does not require the electrolyte fluid 1038 to pass through the edge of the wafer and onto the shroud 1006. The problem that the electrolyte fluid 1038 forms a continuous column with the polishing container 1008 and the problem that the electrolyte fluid 1038 jumps in or out of the polishing container 1008 are It can be reduced or prevented without flowing across the wafer 1004. For example, reducing or preventing unwanted effects by simply rotating the wafer 1004 so that the electrolyte fluid flows along the part of the surface of the wafer 1004 toward the edge of the wafer 1004 before falling off the wafer 1004 can do.

III. Reduction of Edge Overpolishing In another aspect, an electropolishing method and electropolishing apparatus for reducing overpolishing at or near the edge of a wafer is described. Typically, a portion of the metal layer at or near the edge of the wafer is polished faster than portions of the metal layer on other areas of the wafer. Electrodes connected to the edge of the wafer may increase the current density in the electrolyte fluid near the edge region of the wafer and consequently increase the polishing rate. In general, a higher current density near the edge of the wafer and by absorbing a portion of the current density through the electrolyte fluid using a conductive member, such as a ring, located at or near the edge of the wafer. The polishing rate can be reduced. To change the amount of current absorbed, the current density near the edge can be adjusted by charging the conductive member, thereby controlling the current density to a higher degree.

  Referring to FIG. 7, an embodiment of an apparatus and method for reducing edge overpolishing is shown. A flow of electrolyte fluid 7080 is applied from nozzle 7054 to wafer 7004. Wafer 7004 is rotated at a rotational speed sufficient to form a thin layer of electrolyte fluid 7081 that can polish the metal layer on wafer 7004. Typically, when an electrode is connected to the edge of the wafer 7004, the metal layer near or at the edge of the wafer 7004 is faster than the metal on other areas of the wafer 7004 and the electrolyte fluid 7081 Polished by thin layer. Therefore, the edge of the wafer 7004 or the metal layer near the edge may be excessively polished.

  The chuck 7002 includes a conductive member 7114. This conductive member 7114 can reduce the amount of excessive polishing at or near the edge of the wafer 7004. For example, both the wafer 7004 and the conductive member 7114 can be connected to a power source 7110 and charged, so that a portion of the polishing current in the thin layer of electrolyte fluid 7081 is absorbed by the conductive member 7114. By absorbing a portion of the polishing current, the conductive member 7114 can reduce the polishing rate of the metal layer at or near the edge of the wafer 7004 and reduce or prevent overpolishing.

  The conductive member 7114 can include a single ring positioned near or at the edge of the wafer 7004. Alternatively, the conductive member 7114 can include two or more sections. These sections are arranged at or near the edge of the wafer 7004. The conductive member 7114 can include a metal or alloy, such as tantalum, titanium, and stainless steel, and other conductive materials suitable for contacting the electrolyte fluid 7081.

  Further, the wafer 7004 can be positioned between the wafer chuck 7002 and the conductive member 7114 as shown in FIG. For example, a robot arm or the like can position the wafer 7004 adjacent to the wafer chuck 7002 or between the wafer chuck 7002 and the conductive member 7114. The wafer 7004 can then be held between them by bringing the wafer chuck 7002 and the conductive member 7114 together or in close proximity. Thus, the assembly of this embodiment includes additional elements such as a holder or positioner for holding wafer chuck 7002 and conductive member 7114 in alignment, as well as conductive member 7114 and wafer 704. An insulating member can be included between the contacts formed to be charged.

  Needless to say, the embodiment of the apparatus shown in FIG. 7 may include other components as shown in FIG. 1A, but these components are omitted to illustrate this example. Has been. For example, shroud 1006 (FIGS. 1A, 1B), as well as various pumps, nozzles, filters, etc. can be used with the apparatus of this embodiment.

  FIG. 8A shows another embodiment of an electropolishing apparatus useful for reducing the polishing rate near the edge of the wafer. A chuck 8002 with a conductive member 8114 is shown. This conductive member can reduce the amount of excessive polishing at or near the edge of the wafer 8004. FIG. 8A is similar to FIG. 7 except that in FIG. 8A, the conductive member 8114 is separated from the wafer 8004 by a spacer element 8118. The spacer element 8118 is, for example, an O-ring. The spacer element 8118 is further electrically insulative and further resistant to acid and corrosion, such as ceramic, polytetrafluoroethylene (commercially known as TEFLON®), poly It may be formed from vinyl chloride (PVC), polyvinylidene fluoride (PVDF), polypropylene, silicon rubber, Viton rubber, and the like. The conductive member 8114 is connected to the power source 8112, and a second conductive member or electrode, for example, a spring member 8119 is connected to the power source 8110. As shown, the current flowing through the conductive member 8114 can be adjusted or controlled by the power supply 8112 to control the polishing rate of the metal layer at or near the edge of the wafer 8004. In general, as the amount of current absorbed by the lower chuck 8114 increases, the polishing rate of the metal layer near or at the edge of the wafer 8004 decreases.

  The power source 8112 may be a DC power source, an AC power source synchronized with the main polishing power source 8110, or the like. The AC power source can also include a forward pulse power source and forward / reverse power sources. Further, the power supply 8112 can be operated in a constant current mode, a constant voltage mode, or a combination of a constant current mode and a constant voltage mode. In this combination, the constant current mode is applied over a part of the polishing time, and the constant voltage mode is applied over the other part of the polishing time. Various electric charges can be applied to the conductive member 8114 by using a variable resistor instead of the power supply 8112 (see FIG. 9A, for example). Further, a variable resistor may be included between the conductive member 8114 and the spring member 8119.

  Conductive member 8114 may similarly include a metal or alloy, such as tantalum, titanium, and stainless steel, as well as other conductive materials. Further, the conductive member 8114 may include one or more sections positioned near or at the edge of the wafer 8004.

  Therefore, in the electropolishing apparatus of this embodiment, the electric charges applied to the wafer 8004 through the spring member 8119 and the conductive member 8114 can be controlled separately by the power supplies 8110 and 8112, respectively. This allows the over-polishing of this edge region to be controlled and reduced by more greatly controlling the current density near the edge region of the wafer 8004.

  FIG. 8B is an enlarged view showing a configuration and connection formed by the conductive member 8114 and the wafer 8004 of FIG. 8A. Specifically, the conductive member 8114 is charged by the power source 8112 and is disposed at a predetermined distance from the wafer 8002 by the spacer element 8118. Wafer 8004 is separately charged by power supply 8110. This power supply 8110 is connected to a spring member 8119 positioned around the edge of the wafer 8004. Spring member 8119 provides charge to wafer 8009. This charge is distributed around the wafer 8004 more evenly than some electrodes positioned around the edge of the wafer 8004, for example. In the case where separate charges are applied to the conductive member 8114 and the spring member 8119, the insulating member 8121 may be positioned between the conductive member 8114 and the spring member 8119. The spring member 8119 may be formed as a coil spring formed in the ring (see, for example, FIG. 8C). However, other cross-sectional profiles are possible, for example an elliptical cross-sectional profile. Furthermore, any number of coil springs can be used depending on the application. The spring member can be formed from any conductive material, such as stainless steel, spring steel, and titanium. The spring member 8119 can be formed from a corrosion-resistant material, or the spring member 8119 can be coated with a corrosion-resistant material such as platinum, TiN, and TaN.

  By changing the number of coils in the spring member 8119, the number of contacts formed between the wafer 8004 and the power source can be changed. Thus, the charge applied to the wafer 8004 can be more uniformly distributed around the outer periphery of the wafer 8004. For example, for a 200 mm wafer, a charge of about 1 to about 10 amperes is typically applied. By configuring the spring member 8119 to form about 1,000 contacts with the wafer 8004, the charge was reduced to about 1 to about 10 milliamperes per contact.

  However, it will be appreciated that the wafer 8004 can be charged by one or more electrical contacts. In addition, any means for distributing charge around the wafer 8004 can be advantageously used.

  If the conductive member 8114 is separated from the wafer 8004 by the spacer element 8118, a short circuit may occur if the spring member 8119 is exposed to the electrolyte fluid. A short circuit of the spring member 8119 may reduce the uniformity of the polishing rate near the edge of the wafer 8004. Thus, in one embodiment, the spacer element 8118 serves as a seal to isolate the spring member 8119 from the electrolyte fluid. The spacer element 8118 may be formed of a corrosion resistant material, such as Viton (fluorocarbon) rubber, silicone rubber, and the like. Further, the spacer element 8118 may have a variety of shapes and configurations depending on the particular application.

  FIG. 8C is an exploded view showing an example of a wafer chuck holder for use with the electropolishing apparatus of this example, useful for reducing the polishing rate near the wafer edge. The wafer chuck includes a body having a base section 8002 located in the upper portion of the body and a conductive member 8114. A wafer 8004 is held between the base section 8002 of the body and the conductive member 8114. The wafer chuck may further include an upper holder (not shown) for clamping or otherwise holding the wafer 8004 and the assembly together. In addition to the first conductive member 8114, the wafer chuck includes a second conductive member, such as a spring member 8119, for applying a charge to the wafer 8004. In some embodiments, the wafer chuck may further include an insulator member 8121 and a spacer member 8118. These members are disposed between the base section 8002 and the conductive member 8114 included in the lower portion of the body. Needless to say, however, as shown in FIG. 7, in some embodiments, the spring member 8119 and the spacer member 8118 can be omitted. When the spring member 8119 is omitted, an electrode or the like may be included as the second conductive member for applying a charge to the wafer 8004.

  In this embodiment, a spring member 8119 is disposed between the wafer 8004 and the spacer member 8118. When pressure is applied to hold conductive member 8114 and base section 8002 together, spring member 8119 is adapted to maintain electrical contact to wafer 8004 (see FIG. 8B). In addition, the spacer member 8118 fits between the conductive member 8114 and the wafer 8004 to form a seal that protects the spring member 8119 from the electrolyte fluid and, if desired, conductive with the spring member 8119. Enables electrical insulation with member 8114.

  The shape of the semiconductor wafer is typically substantially circular. Accordingly, the various components of the wafer chuck are illustrated as having a substantially circular shape. It will be appreciated, however, that the various components of the wafer chuck can include a variety of shapes, depending on the particular application and / or the shape of the wafer. For example, a semiconductor wafer may have a truncated shape, to which the components of the wafer chuck fit.

  Another example of a wafer chuck assembly suitable for the above apparatus and method for holding a wafer and applying a charge to the wafer is disclosed in US Pat. No. 6,248,222 issued Jun. 19, 2001. (Title `` METHODS AND APPRATUS FOR HOLDING AND POSITIONING SEMICONDUCTOR WORKPIECES DURING ELECTROPOLISHING AND / OR ELECTROPLATING OF THE WORKPIECES '') Can be found in The entirety of the above specification is incorporated herein by reference.

  FIG. 9A shows another embodiment of an electropolishing apparatus useful for reducing the polishing rate near the edge of the wafer. Specifically, the wafer chuck 9002 includes a conductive member 9114 that can reduce the amount of excess polishing near or at the edge of the wafer 9004. FIG. 9A is similar to FIG. 8A, except that in FIG. 9A, the conductive member 9114 includes an insulating ring 9115 and a conductive ring 9116 formed within the insulating ring 9115. Insulating ring 9115 can include non-corrosive insulating materials such as plastic and ceramic. The conductive ring 9116 can include a metal or alloy, such as platinum, tantalum, titanium, and stainless steel. The conductive ring 9116 may be connected to the power source 9110 via a variable resistor 9112 or the like. In addition, by placing a spacer element 9118, such as an O-ring, between the conductive member 9114 and the wafer 9004, the portion of the wafer 9004 that is connected to the power supply 9110 via one or more electrodes. It is possible to prevent the electrolyte fluid from coming into contact. Further, a spring member or the like (not shown) may be included to more uniformly distribute the charge to the wafer 9004.

  The device embodiment shown in FIG. 9A allows a small amount of conductive material to be used with conductive material 9114. FIG. This makes the device cheaper, lighter and reduces power consumption during operation. Further, since the surface area of the conductive member 9114 is smaller than that of the conductive member 8114 (FIGS. 8A and 8B), the current density in the edge region of the wafer 8004 can be more greatly controlled. Further, the configuration of FIG. 9A (and FIG. 7) can be advantageously used with the configurations of FIGS. 7 and 8A-8C.

  FIG. 9B is an enlarged view showing another embodiment of the electropolishing apparatus. This embodiment is similar to FIG. 9A, except that in FIG. 9B, the conductive member 9114 includes an insulating member 9121 formed on the lower portion of the conductive member 9114, ie, on the opposite side of the wafer 9004. It is out. Further, the wafer assembly is configured such that the metal layer 9005 on the wafer 9004 is charged near the edge via the conductive spacer element 9118.

Thus, as shown in FIG. 9B, as the electrolyte fluid 9080 is directed near the edge of the wafer 9004, a portion of the current I ₁ flows through the metal layer 9005 and the second portion of the current I ₂ is conductive. Flows to member 9114. The insulating member 9121 formed in the lower portion of the conductive member 9114 serves to reduce the current I ₂ and increase the current I ₁ flowing through the metal layer 9005. Therefore, by adjusting the relative thickness of the insulating member 9121 and the conductive member 9114, the currents I ₁ and I ₂ can be adjusted accordingly.

IV. Electropolishing Method for Fragmented Metal Layer on Wafer The metal layer formed to cover the wafer may be fragmented during the electropolishing process. For example, one or more discontinuous metal regions may occur on the wafer surface. When this occurs, some pieces of the metal layer may become isolated from the edge of the wafer where the electrodes are located. In such cases, the conventional electropolishing method cannot efficiently polish these fragmented sections. This is because the electrode does not charge the fragmented metal layer. In one example method, a thin layer of electrolyte fluid is formed over and contacted with the conductive member by rotating the wafer with a conductive member disposed around the fragmented portion of the metal layer. be able to. A thin layer of electrolyte fluid and a conductive ring allow the fragmented portion to be electropolished.

  As shown in FIGS. 11A and 11B, the metal layer 1150 is adapted to be fragmented, for example, during a polishing process. Fragments of the metal layer 11150 are not connected or arranged at the edge of the wafer 11104 where the electrodes (not shown) are connected to the power supply 11110. Since the fragments of the metal layer 11150 are not located at the edges of the wafer 11004 or are not connected to these edges by metal, current is passed through the fragments to the electrodes provided at the edges of the wafer 11004. I can't guide you. Thus, conventional polishing methods, such as immersing a wafer in a polishing bath, are generally not effective in polishing these pieces.

  The metal layer fragment 11150 may include, for example, an exposed portion of the barrier layer. These exposed portions are left in the semiconductor device portion without the trench after the copper layer is removed by polishing. Furthermore, metal layer fragments 11150 may result from, for example, non-uniform polishing or over-polishing of the edge region.

  Referring to FIG. 11B, an embodiment of an electropolishing apparatus for electropolishing a metal layer piece 11150 on a wafer 11004 is shown. The system includes a chuck 11002, an actuator 11000, a stationary nozzle 11054, and a power source 11110. As stationary nozzle 11054 applies a flow of electrolyte fluid 11080 to wafer 11004, actuator 11000 rotates chuck 11002, which causes electrolyte fluid 11080 to flow across the surface of wafer 11004 as described above, resulting in a metal layer. A thin layer 11081 may be formed extending across the entire fragmented portion of 11150. For example, for a 300 mm diameter wafer, the wafer chuck 11002 can be rotated in the range of about 100 rpm to about 2000 rpm, preferably about 1500 rpm. The thin layer 11081 can conduct a current between the flow of the electrolyte fluid 11080 and the conductive member 11114 of the chuck 11002 by providing a path across the metal layer piece 11150. This current allows the isolated metal layer fragment 11150 on the wafer 11004 to be electropolished.

  In addition, the example apparatus shown in FIG. 11B may be part of a large electropolishing assembly as shown in FIG. 1A. For example, inclusion of the shroud 1006 (FIG. 1) can prevent splashing, non-uniform polishing, or disruption of the polishing flow of the electrolyte fluid 1038. Further, various embodiments of the conductive member 11114 described above with respect to reducing edge polishing can be used with the apparatus of FIG. 11B.

  Another system shown in FIG. 12 can be used to electropolish the metal layer fragments on the wafer 12004. FIG. 12 is similar to FIG. 11, except that in FIG. 12, actuators 12180 and 12182 can move nozzle 12054 along the x direction, while actuator 12000 moves chuck 12002 in a stationary position. Rotate.

  In the system shown in FIGS. 11B and 12, the chuck or nozzle moves along the x-direction, but it should be understood that both the chuck and nozzle can be moved in various directions depending on the particular application. .

V. Control of metal concentration measurement and endpoint detection One factor in making wafer polishing quality more consistently acceptable in a mass production environment is the supply of electrolyte fluid used to polish the wafer It is to control the metal concentration in the part. When the metal concentration in the electrolyte fluid supply reaches a predetermined value, the electrolyte fluid can become very active even when no current is applied. This can cause chemical etching or corrosion of the wafer, for example during post-electropolishing processes. It is therefore desirable to monitor the metal concentration in the electrolyte fluid during process execution and make adjustments based on real time as desired.

  Further, the end point detection sensor typically uses a light detector. The photodetector measures through the electrolyte fluid. Thus, the measured value depends at least in part on the optical properties of the electrolyte fluid. However, the optical properties of the electrolyte fluid can change over time depending on the concentration of metal dissolved in the electrolyte fluid, as well as other factors such as contaminant particles in the electrolyte fluid, hydrogen bubble formation, and the like. Therefore, as the optical properties of the electrolyte fluid change during process execution, the measurement values obtained from the endpoint detector can be adjusted accordingly, thereby increasing the accuracy of the endpoint detection measurement.

  FIG. 10A shows an example of a system that can be used to measure metal concentration, such as within a supply of electrolyte fluid 10030, eg, in an electrolyte fluid reservoir 1070 (FIG. 1A). The system of this example includes a fiber probe 10102, a fiber optic sensor 10104, and a reflector 10100. The fiber probe 10102 and the reflector 10100 can be immersed in the electrolyte fluid 10030, and the light emitted from the fiber probe 10102 is reflected by the reflector 10100 to reach the fiber probe 10102 at maximum light intensity. The fiber probe can be positioned relative to the reflector 10100 to allow it to return. For example, as shown in FIG. 10A, the fiber probe 10102 can be positioned to emit light in a direction perpendicular to the surface of the reflector 10102.

  In addition, the distance H between the reflector 10100 and the fiber probe 10102 can provide measurement accuracy of the metal concentration in the electrolyte fluid. For example, the distance H can be selected such that when the metal concentration reaches a minimum concentration within the supply of electrolyte fluid 10030, the intensity of light received by the optical sensor 10104 reaches a maximum value. Needless to say, other paths between the optical sensor 10104 and the reflector 10100 can be selected, such as multipath and multiple reflection depending on the application and the length of the desired path. Including the path to have. The fiber probe 10102 may be located outside the fluid reservoir, where the path traverses a portion of the electrolyte fluid 1038. Further, an optical sensor positioned to detect the light intensity received by the optical sensor 10104 can be used in place of the reflector 10100.

In general, the color of the electrolyte fluid depends on the type and concentration of metal ions dissolved in the electrolyte fluid. For example, the copper ion in phosphoric acid (H ₃ PO ₄ ) has a blue color. In addition, the intensity of light passing through the electrolyte fluid can be attenuated depending on the color of the electrolyte fluid. In general, as the concentration of metal ions in the electrolyte fluid increases, the light intensity attenuation also increases.

For the system shown in FIG. 10A, the relationship between metal concentration in the electrolyte fluid and light decay is tabulated as follows for the specific metals and electrolyte fluids used with this system:

  This information shown in the table can be stored in the computer 10105. Using the information in the table, the computer can automatically calculate the metal concentration in the electrolyte fluid based on the light intensity detected by the light sensor 10104 using interpolation, rounding or other approximate methods. Can do. Although specific values are listed in the table above for metal concentration (% by weight), any value can be used, or any number of values can be used.

  By selecting the color of the light emitted by the fiber probe 10102, the sensitivity of the measurement detected by the optical sensor 10104 can be increased. Specifically, by making the color of the light emitted by the fiber probe 10102 different from the color of the metal ions in the electrolyte fluid supply unit, the sensitivity to specific metal ions can be increased. For example, for copper ions in the phosphoric acid supply unit, by emitting red light, the sensitivity to copper ion concentration can be made higher than by emitting green light, and by emitting green light, blue light is emitted. Sensitivity can be higher than when light is emitted. However, white light can be emitted for any color of metal ions in the electrolyte fluid.

  FIG. 10A also illustrates another aspect of the system of the above example, which can be used to remove metal ions from the supply of electrolyte fluid 10030. The system further includes two electrodes 10022 and 1000029 and a power supply 10030. When the optical sensor 10104 measures that the metal ion concentration in the supply portion of the electrolyte fluid 10030 has reached the first preset value, the computer 10105 applies a voltage to the electrodes 10022 and 10020 to apply metal ions from the electrolyte fluid supply portion. The power supply 10030 can be instructed to remove. When a voltage is applied to the electrodes 10022 and 1000029, metal ions from the supply portion of the electrolyte fluid 10030 begin to be plated on the electrode 10027. When the optical sensor 10104 detects that the metal ion concentration has fallen below a second preset value, the computer 10105 stops the application of voltage to the electrodes 10022 and 1000029 to cause the metal from the supply of electrolyte fluid 10030 to Power supply 10030 can be instructed to stop the removal of ions. Thus, the concentration of metal ions in the supply portion of the electrolyte fluid 10030 can be maintained between the first preset value and the second preset value, for example, during the electropolishing process.

  The end point detector 1016 (FIGS. 1A, 1B) can also be assisted using the concentration value of the metal ions in the electrolyte fluid 10030. An endpoint detector 1016 can be used to measure the thickness of the metal layer on the wafer 1004. This information can be used by the electropolishing apparatus to determine when to continue or interrupt the electropolishing process on a particular area of the wafer 1004. This information can also be used to determine a suitable polishing rate. The end point detector 1016 may include various sensors, such as an ultrasonic sensor, an optical sensor, and an electromagnetic sensor. If the electrolyte fluid 1038 is used as a medium for transmitting a signal and obtaining a measurement value, the measurement accuracy is improved. This is because it is not necessary to consider the medium interface, for example, the interface between air and the electrolyte fluid 1038. However, if the characteristics of the electrolyte fluid 1038 that can affect the sensor change, the measured values may become inaccurate over time. Accordingly, the endpoint detector measurement can be improved by taking into account changes in the properties of the electrolyte fluid 1038.

  FIG. 10B shows another embodiment of a system for monitoring the optical properties of an electrolyte fluid. This system can be used, for example, to adjust endpoint detector measurements. FIG. 10B is similar to FIG. 10A, except that in this FIG. 10B, a second optical sensor 10204 and a second optical fiber 10202 are included. The optical sensor 10104, the optical fiber 10102, and the reflector 10100 operate in the same manner as described with reference to FIG. 10A. The second optical sensor 10204 and the second optical fiber 10202 operate similarly to the optical sensor 10104 and the optical fiber 10102, but the second optical sensor 10204 and the second optical fiber 10202 Measure optical properties. For example, during the electropolishing process, hydrogen bubbles are often formed at the electrode. Bubbles can adversely affect the endpoint detector by diffracting the measurement beam in the electrolyte fluid and reducing the intensity of the measurement beam. The decrease in intensity affects the measurement of metal ion concentration, but the metal ion concentration can be accurately measured by using multiple detectors that are sensitive to different properties.

  In embodiments where the optical properties of the electrolyte fluid due to air bubbles are determined, again the sensitivity of the measurements detected by the optical sensor 10204 can be increased by selecting the color of the light emitted by the fiber probe 10202. In this case, selecting the color of the light emitted by the fiber probe 10202 as the same color as the metal ions in the electrolyte fluid supply can increase the sensitivity to bubbles and reduce the sensitivity to metal ions. For example, with respect to copper in the phosphoric acid supply unit, emitting blue light has higher sensitivity to bubbles than emitting white light, lower sensitivity to copper ions, and emitting white light emits red light. Is more sensitive to bubbles and less sensitive to copper ions.

  In addition, the intensity of red light from the fiber probe 10102 is also reduced by any bubbles in the electrolyte fluid, which makes copper ion concentration measurements inaccurate. However, the second optical sensor 10204 exhibits a reduced intensity portion mainly due to bubbles and not due to the copper ion concentration. This is because the sensitivity of the fiber probe 10202 is selected to be insensitive to the copper ion concentration. Therefore, the decrease in the intensity of red light can be determined by considering the portion caused by the bubbles, which is determined by the second photosensor 10204. In addition, endpoint detector 1010 (FIG. 1A) can retrieve the optical properties of the electrolyte fluid from computer 10105 and accurately measure the metal thickness of wafer 1004 (FIG. 1A). In this way, the second optical sensor 10204 can improve the accuracy of both the endpoint detector measurement value and the metal ion concentration measurement value.

  Needless to say, various properties of the electrolyte fluid can be measured by using any number of sensors. Various properties, such as optical properties, can then be stored and used to adjust or determine endpoint detector measurements and the like.

VI. Nozzle Configuration According to another aspect, an embodiment of a method and apparatus for electropolishing a metal film on a wafer includes the use of multiple sized nozzles having different polishing rates. In general, large nozzles allow for higher polishing rates of metal films, such as copper, formed on the wafer, while small nozzles allow for lower polishing nozzles. Thus, the electropolishing process can be more accurately controlled by using a large nozzle to coarsely polish the metal layer and then using a small nozzle. Thus, multiple nozzles are advantageous to more accurately polish different wafer areas. However, since the space in the clean room is limited, for example, an apparatus having a plurality of nozzles is desirably compact. An embodiment of a device with multiple nozzles configured on a rotating nozzle holder allows for the use of multiple nozzles in a compact space.

  13A, 13B, 13C, 13D, and 13E show an example of an electropolishing assembly that includes an assembly of multiple rotating nozzles. 13A-13E are similar to FIGS.1A-1E, except that in FIGS. 13A-13E, a plurality of optical end-point detectors 1016 and eddy current thickness / end-point detectors 1009 are positioned adjacent to each other. A rotating nozzle 1012 having a nozzle is added. As indicated by the arrows in FIG. 1A, the rotating nozzle 2012 can direct the flow of electrolyte fluid 1038 to the wafer 1004 by rotating and positioning nozzles 1014 of different sizes and / or shapes. Accordingly, the pump 1018 directs electrolyte fluid only to the nozzle 1010 of the endpoint detector 1016 and the single nozzle 1014. In contrast, in FIG. 1A, electrolyte fluid 1038 is directed to each individual nozzle used in FIG. 1A.

  The end point detector 1009 can be operated to measure the thickness of the metal film formed on the wafer 1004. The end point detector 1009 can measure the thickness of the metal film before, during and after the electropolishing process. In one embodiment of the method, endpoint detector 1009 is used to measure the thickness of the metal film across wafer 1004 prior to electropolishing using, for example, an eddy current endpoint detector. This metal film thickness can then be used to control the local polishing rate corresponding to various locations on the wafer 1004 by controlling the current density and / or flow profile. The distance between the endpoint detector 1009 and the wafer 1004 is, for example, in the range of about 5 to about 1000 μm. Film thickness across the wafer can be measured by rotating the wafer 1004 and moving the chuck 1002 in the horizontal direction while simultaneously allowing the endpoint detector 1009 to scan the entire surface of the wafer 1004. However, it will be appreciated that in another embodiment, the endpoint detector 1009 can scan a stationary wafer 1004.

  The rotating nozzle 2012 can then rotate to select the desired nozzle 1014 depending on the portion of the wafer 1004 being polished, the metal film thickness, and the like. For example, a large nozzle can be used in a region where the metal layer is thick, and a small nozzle can be used in a region where the metal layer is thin. Thus, the inclusion of multiple nozzles of various sizes and profiles within a simple and compact electropolishing assembly that can be interchanged quickly and easily improves the polishing accuracy.

  Referring to FIG. 14A, an example of a rotating nozzle holder 2012 that holds a plurality of nozzles is shown in cross-section. The rotating nozzle holder 2012 holds a plurality of nozzles 2014. The drive means 2070 rotates the rotating nozzle holder 2012 via the drive coupling 2068 when positioning a new nozzle for directing the flow of electrolyte fluid. For example, an O-ring 2066 seals the drive coupling 1068. The driving means 2070 may be a stepping motor, a servo motor, a pneumatic (compressed gas or liquid) driving rotating means, or the like. The nozzle 2014 in the rotating nozzle holder 2012 includes an electrode 2056. These electrodes may be electrically connected to an external power source 1040 (FIG. 13A) through a current feedthrough 2062. The rotating nozzle holder 2012 is placed on the plate 2084. This plate is sealed to the chamber 1008 by an O-ring 2072 and bolts 2074.

  The nozzle holder 2012 may be made of plastic, such as PVC, PVD, Teflon and polypropylene, or the rotating nozzle holder 2012 is coated with a generally insulating and non-corrosive material. Has been. The nozzle 2014 may be made of tantalum, titanium, platinum, stainless steel, or the like.

  FIG. 14C shows an example of a process for electropolishing a metal film from a wafer 1002 using the apparatus of FIG. 14A. In block 1, the metal film thickness profile is measured, for example, by an endpoint detector 1009 that translates in the x direction as the wafer 1004 is rotated as described above. In block 2, the large nozzle 2014 can be used to initially polish the metal film at a high polishing rate. By rotating the rotating nozzle holder 2012 following the high polishing rate, a low polishing rate can be utilized with the small nozzle 2014 in the block 3. After the initial polishing in block 1 and / or block 2, the residual metal thickness profile can be measured using an endpoint detector 1009 in block 4, such as an overcurrent endpoint detector or an optical endpoint detector. . Based on the residual metal thickness profile measured in block 4, the polishing current is adjusted or adjusted in block 5, thereby polishing the thick film spot at high speed, polishing the thin film spot at low speed, and the film thickness Polishing can be stopped at zero points. The polishing current can be adjusted, for example, by using different nozzles 2014 and / or changing the charge supplied by the power source. In block 6, the thickness profile measurement is repeated (ie, block 4). If the metal layer thickness reaches a preset value, the polishing process can be stopped. However, if the metal layer thickness does not reach the preset value, block 5 can be repeated until the desired thickness is reached.

  Needless to say, many variations and variations on this process described with reference to FIG. 14C are possible. In addition, many other processes can be used in conjunction with the apparatus embodiment of FIG. 14A.

  Referring to FIG. 14B, another embodiment of a rotating nozzle assembly holding a plurality of nozzles is shown. The rotating nozzle assembly shown in FIG. 14B is similar to that shown in FIG. 14A, except that in FIG. 14B magnetic coupling couplings 2078 and 2082 are provided instead of the drive coupling 2068. Yes. The advantage of using magnetic coupling couplings 2078 and 2082 is that the drive coupling 2078 is not directly coupled to the nozzle holder 2012 and the O-ring 2066 of FIG. 14A can be omitted. This reduces the risk that electrolyte fluid 1038 will leak through drive coupling 2068. Of course, various ways of coupling the drive coupling 2068 to the rotating nozzle assembly 2112 are possible.

  Referring to FIG. 15, a linearly movable nozzle assembly holding a plurality of nozzles is shown. The linearly movable nozzle assembly operates in a manner similar to the rotating nozzle assembly 2012 of FIGS. 13A-13E, except that in FIG. 15, the nozzle moves linearly rather than rotationally. The linearly movable nozzle assembly includes a nozzle 3054 that includes an electrode 3056, a nozzle 3222 that includes an electrode 3220, and a nozzle 3226 that includes an electrode 3224. These three nozzles 3054, 3222 and 3226 may be configured to have different profiles, eg, different diameters, and thus provide different polishing rates.

  The nozzles 3054, 3222 and 3226 are movable in the horizontal direction, that is, in the x direction via the nozzle holder 3180 and the movement guide 3182. The electrodes 3056, 3220 and 3224 are further connected to a power source 3110 via an electrical feedthrough (not shown). Electrolyte fluid 3080 is supplied to nozzles 3054, 3222 and 3226 via nozzle holder 3180. As described with reference to FIG. 14C, during the electropolishing process, differently sized nozzles 3054, 3222 and 3226 can be used interchangeably to remove the metal film disposed on the wafer 1004. it can. In general, when the metal film is thick, the metal film can be polished at a high polishing rate by using a large nozzle, and when the metal film is thin or when it is desired to remove a small amount of metal In this case, the metal film can be polished at a low polishing rate by using a small nozzle.

  16A-16E illustrate an example of an electropolishing assembly that includes a rotating nozzle assembly that holds a plurality of nozzles. FIGS. 16A to 16E are similar to FIGS. 13A to 13E, except that in FIGS. 16A to 16E, a linearly movable base 4180 on which rotating nozzles 4012 and 4014 are mounted and an exercise guide 4182 are added. ing.

  Specifically, a plurality of rotating nozzles 4014, an optical end point detector 4016, and an eddy current type thickness / end point detector 4060 are mounted on a base 4180 capable of linear motion. The linearly movable base member can be moved along the exercise guide 4182 in the horizontal direction, ie, the x direction. This assembly allows multiple nozzles to be contained within a compact space.

  The structure and operation of the multiple nozzles 4014 are similar to those shown in FIGS. 14A and 14B, but the rotational drive means, drive coupling, current feedthrough and electrolyte fluid feedthrough are for illustrative purposes only. It is omitted.

VII. Nozzle Automatic Cleaning Process An embodiment of a process for automatically cleaning an electropolishing nozzle will be described according to another aspect. During a typical electropolishing process, metal dissolved in the electrolyte fluid may become plated on the nozzle electrodes. The plated metal can limit or deform the nozzle opening, thereby changing the shape and / or direction of the electrolyte fluid flow. If this flow shape changes, the current density of the flow may change, and as a result, the polishing rate of the electropolishing apparatus may change. By applying a reverse voltage to the nozzle and dissolving the metal ions back into the electrolyte solution, the plating can be removed from the nozzle or the nozzle can be cleaned. For example, the metal can be plated on another nozzle or anticorrosion material.

  Referring again to FIGS. 1A-1E, the metal from the metal layer polished from the wafer 1004 becomes dissolved in the electrolyte fluid 1308 so that a portion of the dissolved metal is removed from the nozzle electrode 1056. And / or become plated on 1060. In order to remove this metal from the nozzle electrode 1056 and / or 1060, a reverse voltage can be applied to the nozzle electrode 1056 and / or 1060. A reverse voltage can be applied by using a DC power source or an AC power source. For one example process, a reverse voltage is applied to dissolve the metal deposit in the electrolyte fluid. In another example process, a reverse voltage is applied to plate the metal deposit on the disposable wafer. In yet another embodiment, a reverse voltage is applied to plate the metal deposit on the block.

A. Dissolving metal deposits in the electrolyte fluid using a DC power source Referring to Figure 1A, using a DC power source, the metal deposits on the nozzle 1012 are polished away and removed in the electrolyte fluid 1038. Can be dissolved. More specifically, since the conductor C can be connected to the conductor b and the conductor B can be connected to the conductor a, the nozzle electrode 1056 (FIGS. 1B to 1E) acts as an anode, and the nozzle electrode 1060 Acts as a cathode. By supplying electrolyte fluid 1038 through nozzles 1012 and 1014, an electrical circuit can be formed between electrodes 1056 and 1060, which removes metal deposits on nozzle 1012 from nozzle 1012; Allowing it to be dissolved in the electrolyte fluid 1038. A portion of the metal dissolved in the electrolyte fluid 1038 can be plated onto the nozzle 1014.

  Although this process appears to move metal from one nozzle and only plate to another nozzle, the majority of the metal removed from nozzle 1012 remains dissolved in electrolyte fluid 1038. Since the metal concentration in the electrolyte fluid 1038 corresponding to one embodiment of the electropolishing process is typically low, for example less than 3% by weight, the electropolishing is driven by charging the electrodes 1012 and 1014 and the electrolyte It is not driven by the chemistry of solution 1038. Accordingly, the amount of metal polished from the nozzle 1012 is greater than the amount of metal plated on the nozzle 1014. For example, for every metal ion plated on the nozzle 1014, ten metal ions can be removed from the nozzle 1012 so that most of the metal ions are dissolved in the electrolyte fluid 1038.

  With continued reference to FIG. 1A, the process can be reversed to use a DC power source to polish away metal deposits on the nozzle 1014 and dissolve them in the electrolyte fluid 1038. More specifically, since lead B can be connected to lead b and lead C can be connected to lead a, nozzle electrode 1060 (FIGS. 1B-1E) acts as an anode, and nozzle electrode 1056 Acts as a cathode. By supplying electrolyte fluid 1038 through nozzles 1012 and 1014, an electrical circuit can be formed, which removes metal deposits on nozzle 1014 from nozzle 1014 and dissolves in electrolyte fluid 1038. Make it possible. A portion of the metal dissolved in the electrolyte fluid 1038 can be plated onto the nozzle 1012.

  By repeating this process, the step of reversing the voltage corresponding to each nozzle used in the apparatus, the nozzles can be cleaned. In one example process, a continuous wafer is electropolished by first removing the plating from nozzle 1012 and then plating nozzle 1014 followed by removal of plating from nozzle 1014 and plating nozzle 1012. As a result, the nozzle is quickly cleaned. Both nozzles are effectively cleaned. This is because most of the metal is dissolved in the electrolyte fluid 1038 rather than being plated on the opposing nozzle as described above.

  2 and 3 show two embodiments of the configuration of nozzles 1012 and 1014 during the cleaning process. Nozzles 1012 and 1014 are positioned close to each other, allowing electrolyte fluid to flow through nozzles 1012 and 1014 to form a membrane or path of electrolyte fluid between these nozzles. As shown in FIG. 3, when the nozzles 1012 and 1014 are positioned closer together, the two membranes or paths of the electrolyte fluid 1080 flowing between the nozzles 1012 and 1014 are combined to form a single path. Can be formed. This single path reduces the length of the electrical circuit, thereby increasing the efficiency of the metal deposit removal process. Needless to say, the method of this embodiment can also be adopted by using three or more nozzles.

B. Plating Metal Deposits on Wafer Using DC Power Source Referring to FIG. 1A, referring to another example process, using a DC power source, the metal deposits on nozzle 1012 are polished away using a DC power source. This can be plated on the wafer 1004. More specifically, since lead A can be connected to lead b and lead B can be connected to lead a, wafer 1004 acts as the cathode and nozzle electrode 1056 (FIGS. 1B-1E) is the anode Acts as By supplying the electrolyte fluid 1038 to the wafer 1004 through the nozzle 1012, an electrical circuit can be formed, which allows the metal deposit on the nozzle 1012 to be plated on the wafer 1004. . The nozzle 1012 or other nozzles in the electropolishing apparatus can be cleaned in parallel or in series with the nozzle 1014. After the metal deposit on the nozzle 1012 is removed, the wafer 1004 can be discarded.

  Referring also to FIG. 1A, a DC power source can be used to polish away metal deposits on nozzle 1014 and plate them on wafer 1004. More specifically, since lead A can be connected to lead b and lead C can be connected to lead a, wafer 1004 acts as the cathode and nozzle electrode 1060 (FIGS. 1B-1E) is the anode Acts as By supplying electrolyte fluid 1038 to wafer 1004 through nozzle 1014, an electrical circuit can be formed, which allows metal deposits on nozzle 1014 to be plated on wafer 1004. . The nozzle 1012 or other nozzles in the electropolishing apparatus can be cleaned in parallel or in series with the nozzle 1014. After the metal deposit on the nozzle 1014 is removed, the wafer 1004 can be discarded.

C. Plating metal deposit on block using DC power source Referring to FIG. 4, according to another example process, the metal deposit 1057 on nozzle 1012 is polished using DC power source. It can be removed and plated onto the block 1082. More specifically, since lead B can be connected to lead a and lead D can be connected to lead b, block 1082 acts as a cathode and nozzle electrode 1056 acts as an anode. By supplying electrolyte fluid 1038 (FIG. 1) through nozzle 1012 and allowing electrolyte fluid 1038 to come into contact with block 1082, an electrical circuit can be formed through electrolyte fluid 1038, which is connected to nozzle 1012. Of metal deposits can be plated on the block 1082. The block 1082 can be discarded after the metal deposit on the nozzle 1012 is removed or otherwise convenient.

  Referring also to FIG. 4, a DC power source can be used to polish away metal deposit 1057 on nozzle 1014 and plate it on block 1082. More specifically, since lead C can be connected to lead a and lead D can be connected to lead b, block 1082 acts as a cathode and nozzle electrode 1060 acts as an anode. Electrolyte fluid 1038 (FIG. 1) is supplied through nozzle 1014, allowing electrolyte fluid 1038 to come into contact with block 1082 to form an electrical circuit, which is a metal deposit on nozzle 1014. Allows to be plated on block 1082. The block 1082 can be discarded after the metal deposit on the nozzle 1014 has been removed or otherwise convenient. Further, the electrodes 1056 and 1060 can be cleaned in series or in parallel.

  D. Removal of metal deposits using AC power

  In another example nozzle cleaning process, an AC power source, rather than a DC power source, may be used with any of the above configurations to remove metal deposits from nozzles 1012 and 1014. Specifically, by using an AC power source, the metal accumulation is dissolved in the electrolyte fluid, the metal accumulation is plated on the wafer to be discarded, or the metal accumulation on the block or sacrificial material. Plating objects.

  As the metal concentration in the electrolyte fluid decreases, it becomes more efficient to remove metal deposits from the nozzle by an AC power source. Thus, the metal concentration may typically range from about 0.1% to about 5% by weight during the removal process, preferably less than about 0.5%.

  VIII. Nozzle shape

  In any of the above embodiments, variously shaped nozzles can be advantageously employed. Different types of nozzles, such as nozzles having different sizes, profiles and cross-sectional shapes, provide different polishing characteristics, and such nozzles can be advantageously used depending on the particular application. For example, as shown in FIG. 1B, an electropolishing apparatus embodiment may include two different sized nozzles 1012 and 1014. By using these nozzles, different sections of the wafer 1004 can be electropolished.

  In addition, FIGS. 5A-5H show examples of various nozzles having various shapes and configurations. The shape of the nozzle, such as the shape of the passage and the distal end, can change the profile of the electrolyte fluid exiting the nozzle, the current density in the electrolyte fluid flow, and the like. FIGS. 5A-5E show various configurations and shapes of nozzles including insulators 5054 and electrodes 5056. FIG. 5F to 5H show a nozzle without an insulator. Some nozzle configurations have a curved electrode 5056 near the opening. The curved electrode 5056 prevents electrical peaks at sharp points within the electrode. This curved electrode helps to make the current density in the electrolyte fluid flow more uniform. The nozzle illustrated in FIG. 5H includes an electrode 5056 and a rod 5058. This rod is placed near the center of the nozzle, thereby increasing the surface area of the electrode and creating a more uniform current density.

For each of the above nozzles, electrode 5056 can comprise a metal or alloy, such as tantalum, titanium, and stainless steel. In addition, the insulator 5054 can include plastics such as PVC, PVD and Teflon, or ceramics such as Al ₂ O ₃ , ZrO ₂ and SiO ₂ . Therefore, since metals and alloys are typically easier to form in various shapes than plastics and ceramics, nozzles having electrodes with curved and tapered shapes and insulators with straight shapes. Can be manufactured at a lower cost than other shapes. Further, a nozzle having only electrode 5056, as shown in FIGS. 5F, 5G and 5H, can have a simpler shape and a larger surface area. In addition, the nozzle shown in FIG. 5H includes a rod 5058 as part of the electrode 5056. This rod provides an electrode with a larger surface area and can distribute the potential more evenly throughout the electrolyte fluid 1038 exiting the nozzle (FIG. 1A). If the potential is more evenly distributed, the wafer 1004 will be more uniformly electropolished.

  FIGS. 6A and 6B show another embodiment of a nozzle having an insulator 6054, an electrode 6056, and a conductive internal structure 6086. FIG. The internal structure 6086 includes a metal or alloy, such as tantalum, titanium, and stainless steel. In addition, the internal structure 6086 includes a plurality of passages. These passages can increase the surface area of the electrode and can distribute the potential more evenly throughout the nozzle 6056. The size of the passage may range from about 0.1 mm to about 10 mm, depending on the nozzle diameter and the particular application. Preferably, the size of each passage may be about one tenth of the nozzle diameter.

  As shown in FIGS. 6B-6I, the passages can be formed in various cross-sectional shapes. For example, the passages can be squares, fiber shapes, straight slots, metal rods, corrugated slots, rectangles, and honeycombs. Furthermore, although specific cross-sectional shapes are shown in FIGS. 6B-6I, these passages can be formed in any cross-sectional shape, such as a triangle, polygon, and ellipse.

  The above detailed description is provided to illustrate the embodiments and is not intended to limit the invention. As will be apparent to those skilled in the art, many variations and modifications are possible within the scope of the present invention. For example, different electropolishing apparatus embodiments, such as shrouds, conductive members, various nozzles and endpoint detectors can be used together in a single assembly or can be used separately and conventional. The electropolishing apparatus can be improved. Accordingly, the present invention is defined by the appended claims and should not be limited by the description herein.

FIG. 1A is a cross-sectional view showing an embodiment of a semiconductor processing apparatus including a shroud. FIG. 1B is a top view showing an embodiment of a semiconductor processing apparatus including a shroud. FIG. 1C is a cross-sectional view showing an embodiment of a nozzle of a semiconductor processing apparatus. FIG. 1D is a cross-sectional view illustrating an example of a nozzle of a semiconductor processing apparatus. FIG. 1E is a cross-sectional view showing an embodiment of a nozzle of a semiconductor processing apparatus. FIG. 2 is a cross-sectional view showing an embodiment of a nozzle of a semiconductor processing apparatus. FIG. 3 is a cross-sectional view showing an embodiment of a nozzle of a semiconductor processing apparatus. FIG. 4 is a cross-sectional view showing an embodiment of a nozzle and a block of a semiconductor processing apparatus. FIG. 5A is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5B is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5C is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5D is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5E is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5F is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5G is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 5H is a cross-sectional view showing an example of the shape and configuration of the nozzle. FIG. 6A is a cross-sectional view showing an example of a nozzle structure. FIG. 6B is a top view showing an example of a nozzle structure. FIG. 6C is a top view showing an example of a nozzle structure. FIG. 6D is a top view showing an example of a nozzle structure. FIG. 6E is a top view showing an example of a nozzle structure. FIG. 6F is a top view showing an example of a nozzle structure. FIG. 6G is a top view showing an example of a nozzle structure. FIG. 6H is a top view showing an example of a nozzle structure. FIG. 6I is a top view showing an example of a nozzle structure. FIG. 7 is a cross-sectional view showing an embodiment of a semiconductor processing apparatus including a conductive member. FIG. 8A is a cross-sectional view showing an example of a semiconductor processing apparatus including a conductive member. FIG. 8B is a cross-sectional view showing an example of a semiconductor processing apparatus including a conductive member. FIG. 8C is an exploded view showing an embodiment of a wafer chuck assembly including a conductive member. FIG. 9A is a cross-sectional view showing an example of a semiconductor processing apparatus including a conductive member. FIG. 9B is a cross-sectional view showing an example of a semiconductor processing apparatus including a conductive member. FIG. 10A is a cross-sectional view showing an embodiment of a semiconductor processing apparatus including one sensor. FIG. 10B is a cross-sectional view showing an example of a semiconductor processing apparatus including two sensors. FIG. 11A is a top view showing an embodiment of a semiconductor processing apparatus. FIG. 11B is a cross-sectional view showing an example of a semiconductor processing apparatus. FIG. 12 is a cross-sectional view showing an embodiment of a semiconductor processing apparatus. FIG. 13A shows an electropolishing assembly with a rotating assembly that holds a plurality of nozzles. FIG. 13B shows an electropolishing assembly with a rotating assembly that holds a plurality of nozzles. FIG. 13C shows an electropolishing assembly with a rotating assembly that holds a plurality of nozzles. FIG. 13D shows an electropolishing assembly with a rotating assembly that holds a plurality of nozzles. FIG. 13E shows an electropolishing assembly with a rotating assembly that holds a plurality of nozzles. FIG. 14A is a cross-sectional view illustrating an example of a rotating nozzle assembly that holds a plurality of nozzles. FIG. 14B is a cross-sectional view illustrating an example of a rotating nozzle assembly that holds a plurality of nozzles. FIG. 14C shows an example of a process for electropolishing a conductive layer on a wafer. FIG. 15 is a cross-sectional view of an electropolishing chamber with an example of a linearly movable nozzle assembly that holds a plurality of nozzles. FIG. 16A is a diagram showing an example of an electropolishing apparatus including a plurality of rotating nozzles capable of linear motion. FIG. 16B is a diagram showing an example of an electropolishing apparatus including a plurality of rotating nozzles capable of linear motion. FIG. 16C is a diagram showing an example of an electropolishing apparatus including a plurality of rotating nozzles capable of linear motion. FIG. 16D is a diagram showing an example of an electropolishing apparatus including a plurality of rotating nozzles capable of linear motion. FIG. 16E is a diagram showing an example of an electropolishing apparatus including a plurality of rotating nozzles capable of linear motion.

Explanation of symbols

1000 Actuator 1002 Chuck 1004 Wafer 1008 Polishing Container 1020 Pump 1038 Electrolytic Fluid 1070 Reservoir

Claims (74)

In an apparatus for electropolishing a fragmented metal layer on a semiconductor wafer:
A wafer chuck for holding the wafer;
A conductive member that circulates around the periphery of the wafer chuck;
A nozzle configured to direct a flow of electrolyte fluid to the surface of the wafer;
An actuator configured to rotate the wafer chuck at a rotational speed sufficient to electrically connect the fragmented metal layer by forming a thin film of electrolyte fluid over the entire surface of the wafer;
An apparatus for electropolishing a fragmented metal layer on a semiconductor wafer.   The apparatus of claim 1, wherein the electrolyte fluid thin film forms a path for conducting current between the electrolyte fluid and the conductive member.   The apparatus of claim 1, wherein the wafer is oriented face down, and the flow of electrolyte fluid incident on the wafer flows to the edge of the wafer before falling off the surface of the wafer.   The apparatus of claim 1, wherein the actuator is configured to vary the rotation of the chuck in response to the portion of the wafer being electropolished.   The apparatus of claim 1, wherein the actuator is configured to rotate the chuck at a higher speed when the electropolishing portion of the wafer is near center.   The apparatus of claim 1, wherein the wafer chuck is configured to translate the wafer relative to the nozzle.   The apparatus of claim 1, wherein the nozzle is configured to move relative to the wafer chuck.   The apparatus of claim 1, further comprising a shroud surrounding the wafer chuck.   The apparatus of claim 8, wherein the shroud moves relative to the nozzle with the wafer chuck.   The apparatus of claim 8, wherein the shroud and the wafer chuck are mechanically coupled for relative movement together with respect to the nozzle.   9. The apparatus of claim 8, wherein the shroud is positioned about 1 mm to about 10 mm from the edge of the wafer chuck.   9. The apparatus of claim 8, wherein the shroud is positioned about 5 mm from the edge of the chuck.   9. The apparatus of claim 8, wherein the shroud sidewall includes an L-shaped cross section.   The apparatus of claim 8, wherein the shroud sidewalls are tapered.   9. The apparatus of claim 8, wherein the shroud sidewall extends above or below the chuck.   The apparatus of claim 8, wherein the shroud comprises a plastic material or a ceramic material.   The apparatus of claim 8, wherein the shroud comprises a corrosion resistant metal or alloy.   9. The apparatus of claim 8, wherein the shroud is coated with an anti-electrolytic fluidic material. In a method of electropolishing a fragmented metal layer on a semiconductor wafer:
Holding the wafer with a wafer chuck including a conductive member positioned around the periphery of the wafer;
Electropolishing the wafer with a flow of electrolyte fluid;
Rotating the wafer such that electrolyte fluid incident on the wafer forms a thin film of electrolyte fluid on the surface of the wafer;
A method of electropolishing a fragmented metal layer on a semiconductor wafer, comprising:   20. The method of claim 19, wherein the wafer is rotated at a rotational speed sufficient to allow electrolyte fluid incident on the wafer to flow toward the edge of the wafer without leaving the surface of the wafer. .   20. The method of claim 19, further comprising varying the rotation of the chuck depending on the portion of the wafer that is being electropolished.   The method of claim 21, wherein the wafer is rotated at a higher speed when the electropolished portion of the wafer is near center.   20. The method of claim 19, further comprising translating the wafer relative to a nozzle.   20. The method of claim 19, further comprising positioning a shroud around the wafer chuck.   The method of claim 19, wherein the electrolyte fluid flows through the edge of the wafer and impinges on the shroud.   20. The method of claim 19, further comprising moving the shroud relative to the nozzle.   20. The method of claim 19, further comprising moving the shroud and the wafer relative to the nozzle together.   20. The method of claim 19, further comprising moving the nozzle relative to the wafer. In an apparatus for electropolishing a wafer, comprising a nozzle holder configured to hold two or more nozzles adjacent to an electrolyte fluid supply conduit:
One or more of two or more nozzles are connected to the electrolyte fluid supply conduit by moving at least one of the nozzle holder and the supply conduit relative to the other; Electropolishing equipment.   An actuator comprising: an actuator configured to connect one of the two or more nozzles by rotating the nozzle holder. 29. The apparatus according to 29.   30. The apparatus of claim 29, wherein the nozzle holder comprises an insulating material.   30. The apparatus of claim 29, wherein the nozzle holder comprises a non-corrosive material.   30. The apparatus of claim 29, wherein the nozzle holder is made of plastic.   30. The apparatus of claim 29, wherein the nozzle holder and the one or more nozzles are integrally formed.   30. The apparatus of claim 29, wherein the two or more nozzles comprise at least two different profiles.   30. The apparatus of claim 29, further comprising an endpoint detector positioned adjacent to the nozzle holder.   30. The apparatus of claim 29, further comprising a movable base, wherein the nozzle holder is coupled to the movable base.   38. The apparatus of claim 37, wherein the movable base is configured to move in a linear direction and the nozzle holder is configured to rotate. In the method of electropolishing a semiconductor wafer:
Preparing a wafer;
Providing an electrolyte fluid supply;
Providing two or more nozzles mechanically coupled together; and
Movably positioning one of the two or more nozzles relative to the electrolyte fluid supply to direct an electrolyte fluid flow toward the wafer;
A method of electropolishing a semiconductor wafer, comprising:   40. The method of claim 39, wherein the two or more nozzles are mechanically coupled via a nozzle holder.   40. The method of claim 39, wherein the step of movably positioning one of the two or more nozzles comprises rotating the nozzle holder.   40. The method of claim 39, wherein the step of movably positioning the nozzle comprises translating the nozzle holder in a linear direction.   40. The method of claim 39, wherein the two or more nozzles comprise at least two different nozzle profiles. Measuring the profile of the metal layer on the wafer;
Directing the flow of electrolyte fluid to the metal layer with various nozzle profiles, depending on the specific profile of the metal layer;
40. The method of claim 39, further comprising:   45. The method of claim 44, wherein the various nozzles include two or more different nozzle profiles.   45. The method of claim 44, wherein the different nozzles produce different polishing rates.   45. The method of claim 44, wherein the various nozzles are selected to include a relatively high polishing rate on a thick portion of the metal layer and a relatively low polishing rate on a thin portion of the metal layer. The method described.   45. The method of claim 44, wherein the metal layer profile is measured with an endpoint detector positioned adjacent to the two or more nozzles. In a nozzle for electropolishing a semiconductor wafer, comprising a passage with side walls and a distal opening for directing the flow of electrolyte fluid:
The passage includes a conductive material; and
A nozzle for electropolishing a semiconductor wafer, wherein the sidewall is curved near the distal opening.   50. The nozzle of claim 49, further comprising an insulator disposed outside the sidewall with respect to the passage.   50. A nozzle according to claim 49, wherein the passage comprises a cylindrical shape.   50. A nozzle according to claim 49, wherein the passage comprises a conical / cylindrical shape.   50. The nozzle of claim 49, further comprising a conductive structure disposed within the passage.   54. A nozzle according to claim 53, wherein the conductive structure comprises a rod.   54. The nozzle of claim 53, wherein the conductive structure includes a rod positioned within the passage.   54. The nozzle of claim 53, wherein the conductive structure includes a plurality of passages.   57. The nozzle of claim 56, wherein the plurality of passages have a cross-sectional size in the range of about 0.1 to 10 mm.   57. A nozzle according to claim 56, wherein the cross-sectional size of the plurality of passages is approximately one tenth of the diameter size of the passages. In a method of removing plating from a nozzle used in an electropolishing apparatus:
Providing an electrolyte fluid through the nozzle;
Applying a charge configured to remove metal ions from the nozzle to the nozzle;
A method of removing plating from a nozzle used in an electropolishing apparatus.   60. The method of claim 59, further comprising applying a second counter charge to a second nozzle, wherein the electrolyte fluid forms a path between the first nozzle and the second nozzle. Method.   61. The method of claim 60, wherein the nozzle and the second nozzle are brought close together during the plating removal process.   61. The method of claim 60, wherein the nozzle is an anode and the second nozzle is a cathode.   61. The method of claim 60, wherein a portion of the unplated metal is dissolved in the electrolyte fluid.   61. The method of claim 60, wherein the metal ion concentration in the electrolyte fluid is about 3% by weight or less.   61. The method of claim 60, further comprising removing plating from the second nozzle by reversing the charge applied to the nozzle and the second nozzle.   61. The method of claim 60, wherein the nozzle is an anode and the second nozzle is a cathode.   60. The method of claim 59, wherein the nozzle is charged with a DC power source.   60. The method of claim 59, wherein the nozzle is charged with an AC power source.   60. The method of claim 59, further comprising applying a second counter charge to the wafer.   70. The method of claim 69, wherein a portion of the unplated metal is dissolved in the electrolyte fluid.   70. The method of claim 69, wherein the metal ion concentration in the electrolyte fluid is about 3% by weight or less.   70. The method of claim 69, wherein the nozzle is charged with a DC power source.   70. The method of claim 69, wherein the nozzle is charged with an AC power source.   60. The method of claim 59, further comprising applying a second counter charge to the conductive material.

Images