TW200529973A - Real time process control for a polishing process - Google Patents

Abstract

Embodiments of the invention provide a fluid processing system for substrates. The processing system includes a processing cell and a substrate inspection station. During a processing sequence, a substrate is first inspected in the inspection station. The information from the inspection is then fed forward and used to control the operation of the fluid processing cell.

Description

200529973 15414plf.doc IX. Description of the invention: [Technical field to which the invention belongs] The present invention is a method for processing a substrate by a feed-forward control process. In particular, the embodiment of the present invention uses a substrate inspection table to measure the pre-processing Substrate parameters, which are then used as inputs to control the process. [Prior Art] An eddy current sensor used to measure the thickness of a conductive object is a non-contact measurement device. Simply put, an eddy current detector includes a sensing coil. Under the action of alternating current, the sensing coil generates an oscillating magnetic field, which will induce an eddy current near the conductor. Eddy current strength is usually expressed in house amperes (mA) and is related to the magnetic field flux or strength generated by alternating current and the impedance of the conductor. We know that the impedance of a conductor is related to the resistivity of the conductor. Therefore, we can determine the thickness of a conductor from the known resistivity of the conductor and the measured eddy current or impedance. In the process, the eddy flu tester is used to measure the thickness of the film deposited on the substrate (such as a copper layer) or formed on the substrate-the thickness of the film layer can be used to measure the purchase process on the substrate (such as corrosion, polishing, Layer, etc.) Corrective measures are taken to ensure that the desired uniformity is obtained. When waste is eliminated, there is no need for further processing-cost reduction. The substrate can be used by raspberry uranium to measure the guide and measuring device on the substrate — 200529973 15414plf.doc

The accidental position between the sensors is slightly A. The sensor and the substrate are jealous. Therefore, in order to maintain a complex and expensive binary = ;; away, these measurement devices are usually equipped with the [inventive content] the purpose of the embodiment is that the silk board provides a dynamic binary system including an operating room And a substrate detects the operation of 丄, °, and 控 in a timely manner. The object embodiment further provides a method for measuring the thickness of a portion of the conductive layer on the substrate, and a 7VW, a device matching one or more embodiments includes a thirteen Influenza detector, the sensor has-side or both sides of a first measurement substrate. At a critical moment, it is placed at a predetermined distance between the sensors to be sensed. Second, this distance. When the test object reaches this interval :: = ;: Γ The measurement device includes a position sense == /, the position of the sampling point on the object, and the phase of the vortex tester and position system F, the = Yi Zhi also includes a dummy at the sampling point. 3 The eye estimation circuit is used to determine the degree of m-field exhaustion. The 5H measurement device also includes a set of each measurement. Spacing device. Tian and Li are sold in real-time, and the ride also includes a circuit connection. At this time, the evaluation circuit adjusts the sensing; ㈣ = ::: 200529 paper The detected displacement of the test object. A pair of _ is placed on the test object, which is summarized in the test object -ρ binocular η 'test 1 is the first and second components placed on both sides of the test object. This method also includes the calculation of the thickness of the test object by sampling the test object; When the sensor head is in the measurement, the object can move between the sensor heads. In the method of the ^ = or more embodiments, the method also includes measuring the displacement occurring in the direction of the connecting line after the test iron, and J :! to compensate for any detected displacement of the test object. For the above and other special features that are clearly stated in the table, better practices, and in conjunction with the modification and retouching of the attached handle, therefore, the scope of this patent shall prevail. Mao Yuezhibao's 4-track enclosure is attached as an application [Embodiment] The present invention uses a fast vortex flu detector as a reassurance to determine the thickness of the object at each sampling location. The thickness of the front film layer makes the conductive 'throw two or two: perform the light or' child product process', so as to control the first or the child product according to the measurement results. In a nutshell, the device does have two sensor heads placed opposite each other and it; == the second between the two sensor heads; =: 7, should; pass or enter the thickness of the two test objects. The device also contains a set of position sensors for: 2: 20052005973 15414pif.doc position on the object. The two vortex flue probes are placed on the opposite sides of the test object with accuracy. This is due to the accidental movement or vibration of Wei Zhi's opponent near or away from the sensor at the position of Passage_I = I. At the same time, the tilt measurement is still performed, which can be performed. One or more embodiments in the present invention use a sensor in the 2 direction to determine the distance between the test object and the sensor head to determine the correlation with the sensor. Correction coefficient and apply it to the influence of compensation and vibration in the original data, thereby further improving the measurement accuracy. Bayer Fig. 1 shows a schematic diagram of a typical vortex flu probe 1G, which is a thickness measuring device applied to various embodiments of the present invention. The sensor head W includes two pot cores 12 and a coil 14. For example, the pot-shaped iron core 12 is, for example, a split soft magnetic pot-shaped iron core (_ not overlooking the mouth coffee), a diameter of about 9 mm, a height of about 5 mm, or other shapes or iron cores are also suitable. Similarly, for example, the size of the coil 14 is about 2 guagewires and about 10-30 resistors, and the sizes and shapes of other wires are also applicable. ... Tian Yi 丨 When L (AC) electricity passes through the sensing coil 14, this coil generates an oscillating magnetic field, which induces an eddy current on the surface of the test object. The magnitude of the eddy current depends on the impedance of the test object and the AC ( AC) The intensity of the magnetic field (B) that is electrically excited, and the impedance of the test object is related to the thickness and resistance of the test object. In this way, the thickness of the test object can be determined based on the known electrical resistance and the eddy current detected by the sensing coil. 200529973 15414pif.doc We can also use other plastic-like eddy flu probes, such as dual-coil sensor heads. The primary coil is driven by alternating current (AC) current and generates an oscillating magnetic field. The secondary coil receives the test from the test. Signal. FIG. 2 is a block diagram of a device 20 for measuring the thickness of a test object according to one or more embodiments of the present invention. FIG. 3 is a perspective view of certain elements of the device 20. Referring to FIG. 2 and FIG. 3, the device 20 includes a eddy current sensor. The sensor has two sensor heads 24, 26, which can be connected in series or in parallel. The two sensor heads 24, 26 are respectively placed on the bracket 28 at a predetermined distance, and at the same time, an entrance or gap is formed between them, and the width of the entrance can be changed according to the size of the test object. For example, the thickness of a semiconductor layer used to determine the thickness of a semiconductor layer deposited on a substrate ranges from about 2 to 6 mm. In semiconductor processing applications, we found that this range guarantees reasonable dot size, signal strength, and operational reliability. The flu test heads 24 and 26 are connected to the sensor circuit board 30. This circuit 30 generates alternating current (AC) electricity to drive the sensor heads 24 and 26 and receives eddy current signals from the sensor heads 24 and 26, and This signal indicates the thickness of the test object. The received eddy current signal is passed to the controller 32 in the form of voltage. The controller includes an analog-to-digital converter that converts the received signal into a digital signal, as described below. We can use different alternating current (AC) power to drive the coil. For example, the frequency of alternating current can range from about 300 kilohertz (kHZ) to 5 million hertz (MHZ), other values also apply. The T device 20 also includes an array position sensor 34. These position sensors 200529973 15414pif.doc 34 detect the position of the test object as the test object 22 passes through the gap between the two vortex flu probes 24, 26. These position sensor materials are connected to the controller & when measuring the thickness of the test object, the controller 32 determines that on the test object = sampling position. The position sensors arranged in an array can be an optical sense through-beam type sensor (through_beam type sensor). The model J_ = manufactured by Renshi is _: Xin in order to further improve the measurement. Accuracy, one or more of the present invention, uses a displacement sensor 36 in the z_ direction to measure the distance between the two test objects (24, 26) to determine the two related to the displacement. Improve measurement accuracy. Suitable 2_direction. ° You can have a laser displacement sensor, such as the Japanese 0MR0N eight U model, which is a private controller (m_ XZjv) ° N ΗManufacturing ratio · Digital conversion 嶋 嶋 ,, ;; 2T and personal computer⑽ . The analog signal is converted into = %% ==: sensor in the direction _ to get the sensing signal from each sensor, and make measurements and compensation calculations, and (; call to get data, brain display or printer., … If the fruit wheel is output to the output device 33, electricity is counted based on the readings of the eddy flu tester. For example, the method of t square == ,; the thickness of four objects is 10,000, based on empirical data. The positive curve of the number, these experience # material vortex flu meter readings. In # use Xi Zhuangli ± the number of howlings to f to make it so that the reading of the vortex flu meter is 22 A normal curve is used to determine the thickness of the test object. _Two examples Γ 5 'device 2. It is used to determine the thickness of the guide layer on the-substrate 22. The substrate 22 is placed on the electrical arm of the money machine. The wire passes through the system. When the substrate 22 passes between the heads, the position controller 34 is arrayed and the front end of the substrate 22 is slightly

Ensor 34. The substrate 22 touches when passing through the first position sensor; f H is ^^ This sensor contains a eddy current that periodically reads the thickness (such as the sampling rate of reading per second) and the position sensor. 34. Although each time the edge of the soil plate passes through a position, the controller 32 can determine the moving speed of the substrate, and determine the thickness of each sampling position by this return = the position of each sampling point on the substrate. In this way, we can measure the thickness along a certain line of the soil plate. When measuring along other lines, only

The substrate needs to be turned to the position to be measured, while the line to be measured is passed through the device 20. In this test, I is more suitable for measurement during operation, that is, when the substrate is on the axis between two eddy current inductors. It is possible to quickly measure the thickness of the substrate with a high sample rate. For example, according to one or more embodiments of the present invention, it takes only 2 seconds to make 2,000 sampling measurements on a substrate that is approximately 300 mm in length, and other sampling rates may be used. Due to the method of placing two vortex flue probes on the opposite rain side of the test object, the measurement position accidentally approaches or leaves the sensor. The measurement result is not 11 200529973 15414pif.doc caused by d: the accidental movement of the species is caused by the test When the object passes the sensor, two: two r pairs, which can be deleted and faster. "The object passes two thirsty, 'position & system requirements. In addition, when the test can be fast and accurate measurement of two conductive layers We a, ΐ correct measures to ensure that the desired thickness is obtained. For example, the uniformity degree 'and the measurement results show that the thickness is not enough. He processed it for selective chemical mechanical polishing or its f, 4. It is used in a conventional thickness measurement device. It has a schematic diagram of the magnetic field lines generated by the representative = early-heart sensor. This eddy flu is measured as a bio-group magnetic field line. When the test object 22 is a certain distance from the eddy flu sensor, it will cross-quantify the magnetic field lines. If The test object 22 suddenly deflects or deviates from the eddy flu detector (even if a small deviation occurs), the number of magnetic lines intercepted by the test object will change significantly. This change will affect the eddy flu detector's measured value. The measurement accuracy is reduced. Fig. 5 shows that the magnetic field lines generated by the double vortex flu probes 24, 26 are not intended to be used in the device 20 corresponding to the embodiments described above. As shown in the figure, when the test object 22 is biased or deviated from the respective sensor head, the change in the number of magnetic field lines intercepted is significantly reduced. Therefore, the device reduces the sensitivity to the change in the distance between the test object and the sensor head. Figure 6 shows the difference in distance sensitivity between the two sensor heads. One device is a single-sensor head and the other is a dual-sensor head. 200529973 15414pif.doc sensing example shown in the figure ' These dimensions can be changed according to the requirements of the application. Ϋ In the material to compensate for the effects of movement and vibration., Representative distance compensation coefficient values, these values are selected based on the distance traveled by the sensor relative to the sensor head. The numerical value = empirical value, which varies depending on the dress used and the test object. If: ^ ^ ί Flow chart of one or more implementations of the present invention in accordance with the present invention. In steps, when testing Passed: mesh = shell, 'put on test The thickness of the first and second vortex flue probes on both sides of the object to be sampled on the test object; step 110, true = test, the position of the sampling point; step 12G, the reliance object is on the first ^ Brother: The displacement in the direction of the connection line of the sensor head; Step 130, Figure 9 is shown as an exemplary Ecp system of the present invention, and the top view of the system: ECP system 100 includes a workbench 13〇, this workbench is also usually a private board loading station. The workbench 13Q includes a plurality of substrate loading points, and the second loading inflammation point is equipped with a card box 134 for loading the substrate. The robot arm 132 is placed on the workbench by women. 130 and can access the substrate installed in the cassette 134. The robot arm 132 can also enter the channel 115 connected to the table 13 and the processing platform U3; the robot arm 132 can enter the cassette 134 containing the substrate to retrieve the base 13 200529973 15414pif.doc, and send the substrate to the processing chamber 114, 116 on the processing platform 113, or to an annealing station 135 or a substrate inspection station 15; Similarly, the robot arm 132 can also After processing, the substrate is retrieved from the processing chamber ι14, 116 or the annealing table 135. Here, in order to remove the substrate from the system 100, the robot arm 132 sends the substrate to the inspection table 150 or returns to one of the cassettes 134 containing the substrate. The annealing station 135 usually includes a two-position annealing chamber. A cold plate 136 and a hot plate 137 in the annealing chamber are adjacent to the substrate transfer robot 14o. The transfer robot 140 is located between the two plates and transfers the substrate on the hot plate 137 or the cold plate 136. In addition, although the schematic position of the annealing chamber 135 is close to the connection passage U5, the embodiment of the present invention does not limit this. For example, the annealing chamber 135 can be placed in a position directly connected to the Gongpingtang 113, that is, grasped by the robot arm of the processing table, or the annealing chamber 135 can be placed in a position capable of communicating with the processing platform ι13, such as annealing. The chamber 135 can be placed in the same system as the processing platform 113, but the annealing chamber 135 cannot be directly connected to the processing platform 113 and is located in a position where the robot arm 120 of the processing table cannot reach. For example, the annealing chamber 135 can be placed in direct communication with the connection channel 115 and can enter the processing platform 113 = position. Therefore, the annealing chamber 135 can be connected to the processing platform Η]. As described above, the ECP system 100 also includes a processing platform 113, and a flat plate transfer arm 120 is mounted on the flat hall 113. This transfer arm 120 typically includes one or more arms / blades 122, 124 for supporting and transferring the substrate. In addition, the transfer arm 120 and its blades 122 and 124 can be extended, rotated, and moved vertically to transfer the arm. The substrate can be placed on the processing 14 200529ft7r3d0c platform 113-series processing chambers 102, 1 and 10. 6, ⑽, add 112, m, 116 + or remove. Similarly, the robot arm of the workbench can also make the substrate support blades stretch, rotate, move vertically, and move linearly on the track extending from the work to the guard platform 113. Generally, the processing chambers 102, 104, 106, 108, 110, 112, 114, and 116 can be used in two electrochemical chambers and processing chambers. What's more, some of the processing chambers can be used as electrochemical coating chambers, washing chambers, tilted cleaning chambers, spin-rinsing-drying chambers, substrate surface cleaning chambers (including cleaning, rinsing and wet money engraving chambers), non-electrolytic plating chambers, Metrology test benches, and / or other processing rooms useful in relation to the ore layer =. The various processing chambers and robot arms are often connected to the sequence controller 111. The program control is a microprocessor control system for receiving input from the user and / or various sensors on the system 1 () () and appropriately controlling the operation of the system in accordance with the input.

In the coating system shown in FIG. 9, we extend the processing chamber as follows: The processing chamber m'm can be used as a wet processing table in the Jiawei platform 113, a drying processing zone in the connection channel 115, and an annealing chamber. Interchange between 135 and workbench 130. The processing chamber located at the transit point may be a spin-drying chamber and / or a substrate cleaning chamber. In particular, the processing chambers 114 and 116 can be a rotatory washing and drying chamber and a substrate cleaning chamber placed in a piled manner. The processing chambers 102, 104, 110, 112 can be used as a plating chamber (electroless plating chamber or non-electrolytic plating chamber). The processing chambers 106 and 108 can be used as substrate tilt cleaning chambers. US Patent Application No. 10 / 435,121 published on December 19, 2002, "A variety of chemical electrochemical treatment systems, so that other forms and applications of electrochemical treatment systems can be incorporated into this case. 15 200529973 15414pif .doc test Figure 10 shows a schematic diagram of -show = 2 = testing table 4 applied in the processing system 100 of the present invention. The substrate testing table 4GG often has a structure that can enter people in two micrometers: μ points. In particular, Substrate inspection: Π4, two daggers 3, top 402, and a bottom part separated from the top 402. The distance between wide and / can define a substrate receiving slot ^ w, the sensor and / or analysis device can be placed The top part is so that the sensing device 11 / analyzing device can face the substrate placed in the receiving tank. For example, the substrate inspection table 400 can also be placed on the working platform, the connection channel out or the processing platform 。. It can control the transmission of the relay after the 111, such as the system to detect the substrate under the system control system 111, and then the test station 400 transmits the information of the measurement results to the pure control unit Lu to control the post-gamma processing process. The process of the inclination of the concubine in Wei Xue Useful. The paste is' from the f-degree of the seed layer that can be touched before the substrate is plated ', so the thickness of the seed layer measured before recording is feed-forwarded to the mine controller (controller lu)' while being appropriately controlled The growth or deposition process of the seed layer in order to increase the thickness of the seed layer to a preset value. For example, 'a substrate conveyor similar to the robot arm 132 can insert the substrate into the receiving groove 406. The sequence of inserting a substrate is shown in FIG. 10. : Baixian, place the robot arm 132 in position a (this position is outside of the receiving slot 406), and then insert the substrate into the receiving slot 406 when the robot arm 132 moves to position B in the direction of the arrow c. In this structure, the sensor placed on the upper part 402 or the lower part 404 of the inspection table 400 is exposed so that the diameter of the substrate can be measured by 16 200529973 15414plf.doc. Figure ii shows another substrate inspection table of the present invention. Example of 500: The inspection table 500 includes a base 512 on which a detection device is installed. The detection device usually has a top 502 and a bottom 504 with a sufficient distance between them to allow the substrate to be inserted, and more particularly to Center the substrate It is located in the receiving groove 506 formed by the gap between the top 502 and the bottom 504. The inspection platform 500 also includes a substrate supporting device 508, which usually has a substrate supporting plate 153. This supporting plate 153 is attached to a movable supporting device 508 ', and the bearing split 508 can be moved on a linear guideway 51. The shape of this guideway enables the trough on the support plate 153 to be measured. For example, the test bench can also be placed on the connection channel 145 of the work | 130 or platform 100, see picture! . More specifically, the test stand 500 can be placed in an annealing rack 301 at 160. In addition, the detection port 500 is also placed on the processing table 113. The test station 4 is usually electrically controlled by the bathroom controller. In this way, the inspection is used to measure the substrate under the control of the system controller ⑴, and the inspection station then passes the measurement information to the system control n m to control a plus-sublayer repair process. The method for measuring the eddy currents commonly used in the substrate inspection stations 400 and 5000 of the present invention 3 determines the thickness of the deposited layer on the substrate. For example, the top π or bottom of each test station is equipped with a thirsty flu detector. When the measured layer is located at the top of the test station, conversely, when the test is & = * the device is located at the bottom of the test station accordingly. . , Lig, Month, how to measure Generally, eddy current detector works at a frequency that can penetrate all conductive films 17 200529973 15414pif.doc 〇 ~ See that the film often includes the traditional four-point detection eddy current measurement located on the conductive film ^ F dielectric layer. Different from the traditional detection device when there is a dielectric layer on the surface, only ^ = is that when the inherent characteristics of the measurement device under the conductive layer will lead to the traditional four points similar to the transmission resistance of the female device of the four test stations, _ ^^ = '^ _ The thickness of the film is obtained from the surface of the film. Alas, it can be inferred using known mathematical methods: the resistance varies with temperature. Therefore, the present invention makes the film cold or annealed and measures the electrical resistance. For example, when the temperature is two, it ’s on the surface of the film. Each temperature must be compensated. It is also said that,,,, product film, eye 3 "for accurate measurement, can provide repeatable results. The hunting is determined by establishing a time-varying magnetic field to determine the conductive film's ί, =? # Is generated by applying alternating current to the coil: 22: mouth, 1 circle on the loop, the coil radiates energy, such as generating A cyclic = magnetic current generates a magnetic field in the conductive film, and the magnetic field interacts with the line conductor. This effect causes the distribution or correction of the magnetic field of the coil, and then causes the corresponding electrical parameters such as the impedance change of the coil. The change in impedance can be directly scaled and proportional to the thickness of the conductive layer. 18 200529973 15414pif.doc In the coating system shown in the present invention, the inspection table is used to analyze or determine the thickness of the conductive layer on the substrate. The measured thickness is fed forward to a certain mineral layer chamber and is used to control the coating process. . More specifically, the coating chamber can be configured with multiple anodes that are individually controlled during the coating process. In this structure ', in order to obtain a fairly uniformly thick deposited layer, the measured thickness is used to determine the voltage applied to each anode. FIG. 12 does not illustrate another embodiment of the present invention. In this embodiment, the inspection table 1104 is applied to a substrate polishing system 1100, which has at least one suitable for electrochemical deposition and / or chemical mechanical polishing table such as electrochemical mechanical polishing (ECMP) table 1102, and There is at least one commonly used polishing or polishing table 1106, which is arranged on a single platform or tool. A suitable polishing tool of the present invention is a MIRRA® MesaTM chemical mechanical polishing machine manufactured by Applied Materials, Inc. of St. Clair, California. The apparatus 1100 in FIG. 12 includes, for example, two ECMP tables 1102 and a polishing table 1106, and these machines are used to treat the surface of the substrate. For example, for a substrate in which a barrier layer is defined by a feature and a layer of conductive material is deposited on the barrier layer, the conductive material can be removed in two ECMP stations 1102 through two steps and polished. The stage 1106 polishes the barrier layer to form a flat surface. In addition, the polishing system n⑻ may include a Jintong chemical mechanical polishing table, a low-pressure chemical mechanical polishing table, or other polishing and / or planarization machines that can be used for semiconductor material processing. Representative devices 1100 often include a base 110. The base is used to support one or more ECMP tables 1102, one or more polishing tables 1106, a transfer table 1010, and a rotating table 1112. Conveying table 111〇19 200529973 15414pif.doc By the loading machine 1116 and the device is said to advance the substrate " 1114 work $ 112G includes a cleaning module, measurement and one or more substrate storage boxes ⑽: = table should be -One NovaScanTM Smart Thickness 'Detects the measuring instrument company and company position of Lissina State and Phoenix City ^ Aben ff' In other embodiments of the present invention, the detection platform ^ by = m at the thickness of the base layer The electric device is shown in Figure ^ No, the test stand 1104 can be placed in different positions on the system. In addition, there are 1116 (or work 4 112G) clothes-carrying machines on pure money processing tools (not shown); pure gas phase; i, physical gas phase / child product tools, money carving tools, and others. In the known example, the transfer table 1110 includes at least one of an input buffer table 1124, an output buffer table 1126, a transfer machine 1132, and an assembly device U28. The loading machine 1116 places the substrate 1114 on the input buffer table 1124. The conveying machine 1132 has two holding devices, each of which has a relatively large amount of money, and is used to control the substrate in the manner of the edge of the substrate. The conveying machine 1132 clamps the substrate 1114, the rotating jig, and the substrate 1114 from the input buffer table 1124 to place the substrate 1114 on the assembly device 1128, and then places the substrate 1114 on the assembly device 1128. The rotating table 1112 generally supports a plurality of polishing heads 113. During the polishing process, each polishing head abuts a substrate. The rotating table 1112 transfers the polishing head 1130 between the transfer table 1110, one or more ECMP tables 1102, and one or more polishing tables 1106. U.S. Patent No. 5,804,507 describes the rotary table used in the invention 20 200529973 15414pif.doc, which was approved to Tolies et al. On September 8, 1998, and can be incorporated into this case for reference. Generally, the turntable 1112 is placed in the center of the base 110. The rotating table 1112 often includes a plurality of arms 1138, and each arm usually supports a polishing head 1130. The arm 1138 is not shown in FIG. 12 so that the transfer table 1110 can be seen. The rotary table 1112 is rotatable so that the polishing head can move the polishing head 1130 between the ECMp tables 1102, 1106 and the transfer table 111, in the order specified by the user. When the substrate 1114 is placed in the ECMP table 1102 or the polishing table 1100, the polishing head usually abuts the substrate 1114. The arrangement of the ECMP table 1102 and polishing table 1106 on the device 1100 allows the substrates to be mineralized or polished in sequence ', which is achieved by moving the substrate between the tables containing the same polishing head 1130. The polishing head used in the present invention is a substrate carrier with the trademark TITAN HEAD ™, which is manufactured by Applied Materials, Inc. of Santa Gola, California. An embodiment of the polishing head 1130 used in the polishing device n⑻ described herein is described in U.S. Patent No. 6,183,354, which was approved to Zuniga et al. On February 6, 2001, and can be incorporated herein by reference. _ In order to control the polishing device 1100 and the processes performed on the polishing device, a controller composed of a central processing unit (CPU) 1142, a memory 1144, and a supporting circuit 1146 is connected to the polishing device 1100. The CPU 1142 may be any computer processor that is used industrially to control drives and pressurizers. The memory 1144 is connected to the CPU 1142. The memory 1144 or the self-selectable §memory can be one or more readable memories such as a random access memory (Ram), a read-only memory ( ROM), 21 200529973 ^ floppy disk, hard disk, or other forms of digital memory such as regional or remote. The support circuit 1146 is connected to the CPU 1142 in a conventional form to support a processor. These circuits include cache memory, power supplies, clock circuits, input / output circuits, subsystems, and so on. & The power required to operate the polishing rack 1100 and / or the controller 1140 is supplied by a power source 1150. The power source 115o is schematically connected to a plurality of components of the polishing apparatus 1100, which include a transfer table 111o, a work table 112o, a loading machine 1116, and a controller 1140. In other embodiments, two or more components in the polishing apparatus 1100 are separately powered. Fig. 13 shows a cross-sectional view of a polishing head 1130 used on the ECMP table 1102 described above. The ECMp stage 1102 often includes a slot 1202, an electrode 1204, a polishing pad 1205, a plate 1203, and a cover 1208. In the embodiment, the groove 1202 is combined with the bottom 1108 of the polishing device Satoshi. Tank 1202 is usually used as a __liquid such as electricity_122G. The county board 1114 ^; Shao, tungsten, gold, silver or any other material that can be deposited on or removed from the substrate by electricity. The trough 1202 may be in the shape of a bowl, and is made of a material such as Teflon, PFA, PE, ρρς and ten + from the + material. The groove of the heart is equal to its anti-electrolytic polished ten dm 7 shoe 1210, which has an opening 1216 and a row of tubes 1214, and the opening is usually at the center of the bottom and allows the shaft 2 to pass through. Seal ___ old Γ12 can rotate and can prevent the groove_solution from opening == 22 200529973 15414pif.doc The groove 1202 may contain an electrode 1204, a disk 1206, and a polishing pad 1205, and the polishing pad 1205 such as a polishing pad is located on the disk 1206. The electrode 1204 is a counter electrode (counter_electr) for the substrate 1114 and / or the polishing pad 1205 that is in contact with the surface of the substrate. The polishing pad 1205 is at least partially conductive and is combined with the substrate as an electrode in an electrochemical process, such as an electrochemical mechanical plating process (ECMPP). The electroplating process includes electrochemical deposition and chemical mechanical polishing or electrochemical dissolution. The electrode 1204 can be used as a positive or negative electrode, and its polarity is determined by whether the bias voltage between the electrode 1204 and the polishing pad is positive or negative. For example, if the material of the electrolyte is deposited on the substrate surface, the electrode 1204 is used as the anode, and the substrate surface and / or the polishing pad 1250 is used as the cathode. If the material is removed from the surface of the substrate, for example, by dissolving it with a bias voltage, then the electrode 204 is used as the cathode and the substrate surface and / or the polishing pad 1205 is used as the anode. This process is called dissolution. The pole 1204 g is located between the pan 1206 and the bottom 1210 of the tank 1202, and it can be immersed or exposed to the electrolyte 122. The electrode 1204 can be made of a material coupled with a, and this material can be processed into a solid flat plate. Electricity is a ship thing? Transparent buttons, money records, and multi-electrode pads. A permeable membrane (not shown) is placed between the electrode and the electrode, and is polished and polished. It filters air bubbles such as air bubbles from the wafer surface, reduces = formation, and makes the current applied between the electrodes. Either the voltage is more stable or the process of electricity is' electrode 1204 is to be deposited or to be removed 23 200529973 15414pif.doc Materials such as copper, gold, silver, crane or others are dissolved by electrical anodes ==== The material, not the deposited material, is pure, carbon, and molybdenum for copper dissolution. The polishing pad blue is made of a material suitable for the environmental requirements of the solution environment-a cymbal, grid or tape. In the embodiment of Fig. 13, a ring-shaped polishing pad 1205 is placed on the upper end of the groove 12G2, and the surface of the polishing pad Cong is supported by the disc 12G6. Polishing 塾 At least one conductive surface made of conductive material, such as one or more conductive elements used to make contact with the surface of the substrate during the process. The polishing pad 1205 may be a part or all of a conductive polishing material, or a conductive polishing material composition that is a job or disposed on the Jintong polishing material. For example, the "air material" is placed on the "office," located between the disc 1206 and the polishing pad 1205, in order to adapt to the plasticity and / or hardness of the polishing pad 1205 during processing. Slot 1202 The cover 1208 and the disk 1206 are located on the base 110 and are movable. When the rotary table 1112 points to the substrate 1114 between the ECMP and the polishing table 1102, 1106, the groove 1202, the cover 1208, and the disk 1206 are along the axis. Move towards the base 1108 for cleaning of the polishing head 1130. The disk 1206 is located in the groove 1202 and fits with the shaft 1212. The shaft 1212 can be coupled to the motor 1224 located below the base 1108. The motor 1224 responds from the controller i The signal of HO rotates the disc 1206 at a predetermined rate. The disc 1206 is a perforated support made of a material suitable for the electrolyte 1220 and having no harmful effect on polishing. The disc 1206 can be made of a polymer 24 200529973 15414pif.doc such as Made of fluoropolymer, PE, Teflon, pfa, PES, HDPE, UHMW, etc. Fasteners can fix the disc 1206 in the groove 1202. Fasteners can be screws or other means including fasteners or suspended in the interior with Peripheral interference fittings, etc. in electrolyte 1 The premise of placing the disc 1206 in 120 is to provide a wider processing window, thereby reducing the deposition of material on the surface of the substrate and eliminating the sensitivity to the size of the electrode 1204. The disc 1206 can usually be penetrated by the electrolyte 1220. In the embodiment In the tray 1206, there are some penetrating holes or grooves 1222. The penetrating holes can be perforations, holes, openings, or channels that partially or completely pass through an object (such as a polished object). Choose the size and density of the penetrating holes The premise is to ensure that the electrolyte 1220 passing through the plate 1206 to the substrate 1114 can be evenly distributed. The direct control of the penetrating holes on the plate 1206 is about 0.002 inches (0.05 mm) and about 0.04 inches. (Ϊ́ · 〇mm). The density of the penetration holes is between about 20% and 80% of the polishing pad. About 50% of the penetration hole density has been observed to minimize the harmful effects of the electrolyte flow on the polishing process. In general, the through-holes on the disk 1206 and the polishing pad 1205 are aligned to ensure that sufficient electrolyte flows through the disk 1206 and the polishing pad 1205 to reach the substrate surface. The polishing pad 1205 is located on the disk 1206, And mechanically clamped or bonded together by conductive glue. When described here When the polishing pad is used in the electrochemical mechanical polishing (ECMP) process, the present invention considers the use of a conductive polishing pad in other electrochemical processing processes. Examples of the electrochemical process include electrochemical deposition, and this > The product includes the application of a polishing pad 1205 to apply a uniform bias to the surface of the substrate for the deposition of conductive materials, without the use of ordinary bias devices such as edge contacts, and electrochemical deposition and chemical mechanical polishing. 200529973 15414pif.doc Photoelectrochemical Electromechanical Power Process (EQVJPP). During the intercalation process, the polishing pad 1205 was placed on a plate 1206 in the groove 1202, and the plate was in an electrolyte. Substrate 4 on the polishing head = placed in the electrolyte and in contact with the polishing pad 1205. The electrolyte passes through the openings of the pan 1206 and the polishing pad 1205 and distributes the electrolytic solution on the surface of the substrate through the grooves on the substrate. Then, a voltage from a power source is applied to the conductive polishing pad 1205, the electrode 1204, and a conductive material such as copper, and further, the material is removed by dissolving the anode. The electrolyte 1220 flows from the storage tank 1233 to the container 1232 through the nozzle 1270, and prevents the electrolyte from flowing out of the container 1232 through a small hole 1234 in a ring shape. These holes 1234 pass through the cover 1208 to provide a passage for the electrolyte outflow container 1232 and allow the electrolyte to flow into the bottom of the tank 1202. At least a part of the hole 1234 is often located between the lower surface 1236 and the central portion 1252 of the pressure reducer 1258. Because the hole 1234 is often still under the pressure reducer 1236, the electrolyte can fill the container 1232 and contact the substrate 1114 and the polishing pad 1250 . Therefore, the substrate 1114 is in contact with the electrolyte over the entire space between the cover 1208 and the disk 1206. The electrolyte 1220 'collected in the discharge pipe 1214 at the bottom 1210 of the tank 1202 flows into the fluid transfer device 1272. The fluid transfer device 1272 often includes a storage tank 1233 and a pump 1242. The electrolyte flowing into the fluid transfer device 1272 is collected in the storage tank 1233, and the pump 1242 transfers the electrolyte from the storage tank 1233 to the nozzle 1270 through the supply line 1244 to recycle the electrolyte to the ECMP station 1102. The filter 1240 is often placed between the storage tank 1233 and the nozzle 1270 in order to remove particles that may be present in the electrolyte 26 200529973 15414pif.doc and agglomerated material. The electrolytic solution may be an industrially produced electrolytic solution. For example, for the removal of copper-containing materials, the electrolytic solution may be a sulfuric acid-based or phosphoric acid-based electrolytic solution such as acetic acid (K3Pα〇, or a combination thereof); Derivatives of acid-based electrolytes such as scale copper sulfate; the electrolyte may also be a gas-acid-acetic acid solution and its derivatives. In addition, the present invention considers the composition of the electrolyte that is customarily used in electroplating or electrolytic polishing, And customary plating or electrolytic polishing additives such as stabilizers, inhibitors, accelerators, etc. A source of electrolytic solutions used in electrochemical processes such as copper bonding, copper anode dissolution, and both is Rohm and Hass Shipley Leonel Division, headquartered in Philadelphia, Pennsylvania, under the trademark Ultrafill 2000. U.S. Patent Application No. 10/038066 filed on January 3, 2002 Examples of suitable electrolyte compositions are described, and all are incorporated into this case for reference. The electrolytic solution flows to the substrate table at a certain flow rate (such as about 2 min per minute) in an electrochemical cell. Or between the surface of the substrate and an electrode. The flow rate is tuned 'between about 0.5 and 20 plus per minute. The electrolyte flow rate in this range can eliminate polished materials and chemistry from the substrate surface. By-products can also be supplemented with electrolyte to increase the polishing rate. When mechanical polishing is used in the polishing process, the substrate Π14 and polishing potential 1205 are turned towards each other to cause the material to be ground off the substrate surface. Polishing is achieved by physical contact between the conductive polishing material and ordinary polishing materials. The substrate 1114 and the polishing pad 1205 are each 27 20052 μm. They can be rotated at about 5 revolutions per minute or more (such as The rotation speed is about 5 to 100% per minute). In the embodiment, a high-speed polishing process may be adopted. The high-speed polishing process includes: polishing pad 1205 at about 15 rpm or higher ( For example, the rotation speed is about 150 to 750 revolutions per minute, and the substrate 1114 is rotated at about 150 to 500 revolutions per minute (for example, about 300 to 500 revolutions per minute). 2001 Filed on July 25, National Patent Application No. 60/308030 "The method and method of chemical mechanical polishing of semiconductor substrates further describes the polishing objects, processes, and high-speed polishing processes of the device herein. During the processing, it can also be used as Other motions include dominating motions or scanning motions on the substrate surface. When the polishing pad 1205 is in contact with the substrate surface, a pressure of about 6 pounds per square inch or less (such as 2 pounds per square inch or less) is applied. It is applied between the polishing pad 1205 and the surface of the substrate. If a substrate containing a material with a low dielectric constant is polished, then during the substrate polishing process, one is about 2 pounds per square inch or less (such as per square inch (5 lbs or less) pressure the substrate 1114 against the polishing pad 1205. On the other hand, as described herein, a pressure of about 0.1 pounds to 0.2 pounds per square inch can be used to polish a substrate having a conductive polishing pad. During the anodic dissolution process, a potential difference or bias is applied to the polishing surface 31 of the% electrode 1204 as the cathode and the polishing pad 1205 as the anode (see Fig. 13). The substrate in contact with the polishing pad is polarized by the surface of a conductive polishing pad, and at the same time, a bias is applied to the support of the conductive body. The application of a bias voltage can remove a conductive material such as a copper-containing material formed on a substrate surface. Should be 28 200529973 15414pif.doc used on the surface of the substrate can be about 15 volts or lower than i5 volts; at: .1 = the voltage between the temple and 10 volts can be used to dissolve the copper material from the substrate surface In the electrolyte, the bias voltage also produces a current density of about 0.1 mA to 5G mA / em2, and a substrate of about 1 G band is applied with a current of about G1 Ampere to M Ampere. The signal provided by the power source H50 establishes a potential difference and performs an anode dissolution process. The signal can be changed according to the requirement of removing material from the surface of the substrate. = A changing anode signal is provided to the conductive polishing. 205 cuts the electrical pulse. _ Transforms the money and throws money. The electrical pulse modulates a time period. Apply a constant reverse voltage or voltage to the substrate in the second time period of current density or electricity, and repeat the first and second steps. For example, electrical pulse modulation technology can use a varying potential difference (from about volts and volts to about αΐvolts and about 15 volts). It is known from the high edge removal rate and low waste of 2 pairs of conventional edge contacts that the waste is discarded. If used on a county pad, the conductive material will be biased between the substrate and the polishing pad 1205 at night. (Such as 1): Dissolve uniformly from the substrate surface and dissolve into the electrolytic conductive material such as copper-containing material. At least a part of the substrate surface can be removed at a rate of t minutes or lower. Between points. In one embodiment of the present invention, a copper material having a thickness of about 12,000 angstroms is applied to a conductive polishing 29 200529973 15414 pif.doc, which should provide a removal rate between about 100 angstroms per minute and 8000 angstroms per minute. .基板 After the electrolytic polishing process, the substrate may be further polished or polished to remove the transfer layer and surface defects from the dielectric material towel, or to improve the smoothness of the polishing process using the v-electric polishing pad. U.S. Patent Application No. 699699, filed on May η, 2000, provides an example of a suitable polishing process and composition that can be incorporated by reference. U.S. Patent Application No. 10/608513, filed on June 26, 2013, provides information on polishing devices, pads, liner distribution or storage areas or distribution, electrodes, and process processing. _Additional: g reference.幵 Minus text and US Patent Application No. 10/608513 describe the use of multi-area a and test stations. The embodiment of the present invention performs a two-shell type program control for the polishing process. For example, before the polishing device of the present invention is processed, == is first measured to determine the thickness of at least one layer on the substrate, and any one of the detection devices and methods described in the second thickness measurement is completed. The thickness of the stage or measurement is often taken from a few points on the base and used by the program controller to control the polishing process. The controller uses the measured thickness information to increase or decrease the thickness variation of the layer that is being polished in Wulailai;: == In the case, there is an example of the area 4202′4 ° 4. Therefore, the second step is to control the polishing process of the uneven layer. More specifically, the preference of the pad region corresponding to the thick region in the electro-active layer can be accelerated by polishing to make the thickness of the polishing layer uniform. 30 200529973 15414pif.doc In another embodiment of the present invention, the method and process of the present invention can be used in a low pressure chemical mechanical polishing (I-CMP) system. In the I <: Mρ system, removal is achieved by oxidation of a conductive layer (usually copper) on the substrate surface. The bias applied between the substrate and the cathode determines the removal rate. Furthermore, the removal of the conductive layer causes the current between the anode (substrate) and the cathode to be located behind the polishing pad or on the polishing pad. Previous experiments have shown a linear relationship between the current density on the substrate surface and the copper removal rate; experiments have also shown that the cathode is divided into segments or regions that can be applied with different bias voltages Controls the removal curve of the substrate. Embodiments of the present invention apply instant removal curve monitoring during polishing and / or deposition. The monitoring and control process of the present invention requires less correction than conventional monitoring devices. This is because the only parameter that needs to be corrected is the relationship between current and removal rate. Furthermore, the process of the present invention is not affected or limited by pad grinding. Therefore, the present invention is not limited by the spatial resolution and has no limitation in thickness measurement. The control process includes input such as the rotation rate and sweep rate of the substrate during the polishing process, the number of regions or electrodes and their shapes, the characteristics of the pad hole, the location of the thickness measurement, and Other parameters. For example, if the center of the substrate is found to be high using t %% measurement, then the voltage applied to the electrode or area in the center of the plate is increased. For example, it can be true that the edge of the substrate ^ # 1 彳 hexagon or " 卩 points are substrates of other shapes. If a certain area of the substrate is thin, then the corresponding area or electrode is biased at a lower voltage than the low material removal rate. 31 20052 Touch 丄 c to choose if to lie better_ spit, weave and use c :: iL, which departs from the spirit and scope of the present invention, and when Xi Shen Erjian is decorated, the scope of protection of the present invention should be considered The attached application is defined by the patent patent. [Brief description of the figure] Fig. 1 is a schematic diagram of a representative vortex flu probe. Block of the device for the thickness of the test object. Figure 2 shows a measurement diagram according to various embodiments of the present invention.

FIG. 3 is a perspective view of some components shown in FIG. 2. Figures 4 and 4 are not schematic diagrams of magnetic field lines of a representative, early-vortex fluorimeter used in previous measurement devices. Fig. 5 is a schematic diagram of the magnetic field lines of a representative twin-vortex influenza tester shown in Fig. 2. The figure shows that the device of one or more embodiments of the present invention reduces the sensitivity to the change in the distance between the / J wuwu and the vortex flu detector head.

FIG. 7 does not show the distance compensation coefficient value of the ring pattern related to one or more embodiments of the present invention. 〃 FIG. 8 is a flow chart showing the thickness of a rigid object according to one or more embodiments of the present invention. FIG. 9 illustrates a top view of an exemplary ECP system. FIG. 10 is a perspective view of an exemplary substrate inspection station. FIG. 11 is a perspective view of another substrate inspection station. Example FIG. 12 shows an example of a processing room of a polishing system according to the present invention when a test bench is applied to a substrate polishing system ^ 32 200529973 15414pif.doc [Description of symbols of main components] 10 · Full flu probe 12: Pot-shaped iron core 14: Coil 20 · · Device 22: Test object, substrate 24, 26: Sensor head 28: Bracket 30: Sensor circuit board 32: Control Device 33: Output device 34: Position sensor 36: Z-direction sensor 50: Single vortex flue sensor 100 · ECP system 102, 104, 106, 108, 110, 112, 114, 116 · Processing room 111: program controller 113: processing platform 115: connection channel 120: transfer arm 122, 124: arm / blade 130: table 132: robot arm 33 200529973 15414pif.doc 134: cassette 135: annealing station 136: cold plate 137: Hot plate 145: Connection channel 301: Annealing rack 400: Substrate inspection station 402: Top of substrate inspection station 404: Bottom of substrate inspection station 406: Substrate receiving tank 500: Inspection station 502: Top of inspection station 504: Inspection station Bottom 506: groove 508: bearing device 513: substrate support plate 1100: polishing system 1102. ECMP table 1104. inspection table 1106: polishing table 1108: base 1110: transfer table 1112: rotary table 1114: substrate 34 200529973 15414pif .doc 1116: Loading machinery 1118: Storage box 1120 : Processing table 1122: Cleaning module 1124: Input buffer table 1126: Output buffer table 1128: Assembly device 1130: Polishing head 1132: Transfer machine 1138: Arm 1140: Controller 1142: Central processing unit (CPU) 1144: Memory 1146: Support circuit 1150: Power supply 1202.Slot 1204: Electrode 1205: Polishing pad 1206: Disk 1208.Cover 1210: Bottom of the slot 1212.Shaft 1214: Drain pipe 1216: Opening hole 35.

2005mL 1218: Seal 1220 Electrolyte 1222: Hole or groove 1224: Motor 1232: Container 1233: Storage tank 1234: Small hole 1236: Lower surface of the pressure reducer 1240: Filter 1252: Central portion 1258: Pressure reducer 1270: Nozzle 1272: Fluid transfer device

36

Claims (1)

200529973 15414pif.doc 10. Scope of patent application: 1. A method for controlling a polishing process, including: determining a thickness curve of a polishing layer; determining a polishing curve of the polishing layer according to the determined thickness curve; and according to the determined The polishing curve is used to polish the polishing layer. 2. The method for controlling the polishing process as described in the first item of the patent application scope ', wherein the polishing step includes selectively activating separate independent polishing regions calculated from the determined polishing curve. 3. The method for controlling a polishing process according to item 2 of the scope of the patent application, wherein the polishing regions include a cathode electrode located in a polishing chamber. 4. A method for controlling a planarization process, which is suitable for depositing a layer on a substrate. The method for controlling the planarization process includes: measuring a thickness curve of the layer using a vortex fluorimeter; and according to the measured thickness Curve to determine the polishing curve of the layer; polishing the layer in a polishing chamber with multiple polishing electrodes; and controlling each polishing bias applied to the polishing electrodes to obtain a predetermined polishing curve. 5. The control method of the flattening process as described in item 4 of the declared patent scope, wherein the control of the M photoelectrodes includes a bias voltage of a polishing electrode that is high corresponding to a thick region of the layer. 6 The control method of the flattening process as described in item 4 of the patent scope / the control of producing some polishing electrodes is low to correspond to the thin electrode of the layer 37 20052997 ^. 7. The method for controlling the planarization process according to item 4 of the scope of patent application, wherein polishing the layer includes performing at least one CMP, 1-CMP, or ECMP process. 38