Closed Loop Control Of Coppers
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Closed Loop Control Of Coppers as PDF for free.
More details
- Words: 2,580
- Pages: 7
<<< Back | Print
Closed-Loop Control of Copper Damascene Electroplating Solutions
Paul Findeis, Richard Henry and Jim Fluegel, IBM Microelectronics, East Fishkill, N.Y.; Michael Pavlov, Eugene Shalyt and Peter Bratin, ECI Technology, East Rutherford, N.J. -- 11/1/2004 Semiconductor International
Acid copper sulfate baths are employed in the damascene process to electrodeposit copper within fine trenches and vias in dielectric material on semiconductor chips.¹ There is a growing need to control the damascene process using control techniques that are accurate, sensitive, easy to employ and cost-effective. In addition, the analysis must be performed in online mode with automatic reporting of analytical data back to the dosing module of the plating tool to minimize operator error and establish closed-loop control of the plating solution.
At a Glance The control of plating bath components, primarily organic additives, is known to be critical in determining the resultant properties of the copper deposit.
In the copper electrodeposition process, it is critical to achieve uniformity of the copper film across the wafer surface, required mechanical properties of the deposit, and "super-filling" characteristics in the small features such as trenches and vias. These important properties are typically achieved using multi-component plating solutions, which include organic and inorganic components. Copper electroplating baths are normally formulated using highly stable electrolytes containing copper sulfate and sulfuric acid. Copper concentration in these electrolytes is 14-60 g/L and sulfuric acid is 1-240 g/L. Other components added into the bath are present in relatively small amounts. These components are organic additives and chloride ions. The organic additives, depending on the concentration and chemical composition, affect the properties of electrodeposited copper, including uniformity, hardness, ductility, tensile strength, etc. Organic additives added to copper electroplating baths typically include suppressors (polymers such as polyethylene glycols), accelerators (sulfur-contained compounds) and levelers (secondary suppressors). Organic components are consumed during the electroplating process at various rates, thus requiring individual control. Accelerators, for example, are consumed faster than other organic additives. Consumption of organic additives depends on various factors, including conditions of electroplating cell, current density, flow rate, number of wafers plated, etc. To keep the concentration of plating bath constituents in the optimum range for the electroplating process, several approaches can be used. The conventional approach employs automatic dosing of individual components into plating solution(s). The dosing system incorporates the hardware needed for the delivery of individual components into a solution and, if needed, removal of used-up solution from the bath. Replenishment of individual components is controlled via predictive software algorithms, mostly in conjunction with periodic analysis of solution composition. In high-volume production, the plating solution is controlled by continuous feed-and-bleed of virgin makeup solution combined with automatic dosing of organic additives, based on predictive algorithms and frequent solution analysis. This approach enables tight control of plating solution and limits buildup of additive by-products formed during the electroplating process. An advantage of this method, if performed properly, is that the solution is maintained at a virtual steady-state condition, and can be used for a long period of time without change. Despite the sophistication of dosing algorithms, the plating-bath composition can change rapidly and drift out of
operational range. Such bath changes occur because of intermittent or unpredictable operation of the bath, or can be caused by a malfunction of plating-bath delivery components, which can produce overdosing or underdosing of the plating bath. These problems can be prevented and avoided if each component of the plating bath is individually monitored with an online analytical system. The alternative approach of keeping the bath in control is based on short-term usage of the plating solution, which would be dumped once a certain number of wafers passed through the bath. According to the inventors, such an approach does not require analytical control. 2 However, this approach relies on the continuous mixing of the bath from individual components. Any failures in the dosing equipment or deviations in activity of the individual solutions cannot be detected, thus allowing this approach to produce defective wafers that will remain undetected until later stages of production. Another difficulty is that this approach works with continuously changing composition of the solutions, which are constantly discarded, rather than solutions under steady-state conditions and in relatively large reservoir. CVS analysis Among a wide range of analytical approaches that have been used for monitoring concentrations of organic additives in electroplating solutions, only cyclic voltammetric stripping (CVS) has demonstrated the ability to reliably monitor the activity of a wide range of components in the plating baths. In this method, the classical three-electrode cell is used where the main indicator electrode is a platinum rotating disc electrode. 2 CVS involves cycling the potential of a Ptrotating-disk electrode so metal is alternately plated and stripped at the electrode surface. Organic additives are detected by CVS from the effect they exert on the electrodeposition rate measured via the metal stripping peak area (Ar). This approach has become the most popular and effective method for controlling damascene copper deposition processes. Closed-loop plating bath control The first generation of CVS-based analyzers for the semiconductor industry, developed in 1997, was designed to support and assist developers researching the plating process. These first standalone systems did not report analytical information to the dosing unit forming a closed loop with the electroplating tool. First-generation systems have been installed in several research locations, and have produced valuable results for understanding and improvement of the plating process. After a few years of using these analytical systems in the field, confidence in their consistent results was gained, which opened opportunity to finally close a loop. The second generation of online analyzers, introduced in 1999, had a closed-loop communication capability as a feature of the standard software package. One of the earliest closed-loop experiences is shown in Figure 1 . Improvement in the bath control (accelerator component) before and after the closed-loop feature was engaged is depicted. As can be seen, the accelerator activity in the bath showed smaller variation once the closed loop was activated, which is related to continuous adjustment of the dosing algorithm based on analytical results.
1. Enabling closed-loop control function allowed improved precision of obtained results,
which is related to continuous adjustment of the dosing algorithm based on analytical results. A third generation of the CVS analyzer, employing a dual-cell configuration, enabled significant improvements in accuracy, precision, reliability, cost of ownership and time of analysis. It also has an improved communication capability, including a feature to communicate with various dosing systems serving different plating tool manufacturers. The improved communication protocol allows the user to send the data either in real time when analysis of one component is finished or in discrete mode once a complete set of analytical results is available. This generation of the analyzer also has the capability to monitor certain byproducts in the plating solutions. When starting a new plating bath, it is important to find the optimum operating conditions. In most cases during these startups, it is necessary to perform frequent and continuous analyses of the bath. However, once the tool is stabilized and steady state is achieved, continuous monitoring can be replaced by less frequent analyses. The following examples illustrate dynamic changes in the activity of constituents of a typical high-volume plating solution for copper damascene processes. Figures 2 and 3 show data for two components of a typical plating system running under closed-loop control. One component of the organic additive (Fig. 2) is continuously dosed into the bath and frequently analyzed by a chemical monitoring system (CMS). As a result of this approach, the activity of this component in the bath is very stable. Figure 3 shows the response of one inorganic component, which is generally more stable and does not require continuous dosing. However, it was monitored with the same frequency as the organic components. As can be seen in Figure 3 , the concentration of this component slowly changes with time, and since it is not automatically dosed, it must be added manually as needed. For such components with relatively low consumption (organic or inorganic), frequency of the analysis can be tuned by modifying the analysis schedule.
2. One component of the organic additive is continuously dosed into the bath and frequently analyzed by a chemical monitoring system (CMS). As a result of this approach, activity of this component in the bath is stable.
3. Inorganic components are monitored with the same frequency as organic components. Concentration of this component is slowly changing with time, and since it is not automatically dosed, it must be added manually as needed. During routine operation when a plating bath performs well, activity of all components in the bath is stable and no adjustments to the dosing algorithm are required. However, during high-volume use of the plating system, concentration of one or several components might change because of evaporation, drag-out, decomposition, or malfunctioning of dosing or other parts of the plating system. It is for these reasons that analysis is performed more frequently, to detect the problem early and allow ample time for troubleshooting before the plating system must be halted. Some examples of a bath going out of control are shown in Figure 4 , which illustrates the behavior of organic components when the closed-loop feature is turned off. While the closed loop was disabled, the dosing system kept inserting additives into the bath according to a predetermined algorithm. This caused the activity of one organic component to increase rapidly, causing an overdose. Once closed-loop communication was restored, activity slowly returned to the target level, and was maintained there by the closed loop. This demonstrates the ability of the dosing system to appropriately use analytical results and perform immediate corrective actions to adjust activity of any organic component in the bath.
4. The closed loop was disabled and the dosing system kept adding additives into the bath according to a predetermined algorithm. This caused activity of one organic component to increase rapidly, causing an overdose. Other organic components were also analyzed at the same frequency while the closed-loop feature was turned off. As can be seen from Figure 4, their activity did not change significantly since their consumption rate is generally much lower than that of the first component. Considering the consumption of individual components in a plating solution, altering the analysis schedule can optimize monitoring of the frequency of each component. One component can be monitored more often than others. Optimized dosing of all components of the plating bath is shown in Figures 5 and 6. This bath is controlled using an optimized schedule of analysis and closed-loop communication with a plating tool. The
data shown in Figures 5 and 6 were collected during a two-week period.
5. Monitoring frequency of each organic component can be optimized by altering analysis schedule and providing the best replenishment conditions.
6. The bath is controlled using optimized schedule of analysis and closed-loop communication with a plating tool. As we mentioned above, during long-term use of a plating bath, there is a potential to accumulate additive byproducts, which eventually might interfere with the plating process. In the most recent design of the analyzer, monitoring of some byproducts can be accomplished without any changes in hardware and without addition of auxiliary reagents. Monitoring of suppressor byproducts in a copper plating bath for an extended period of time is shown in Figure 7 . Data collection began when the bath was freshly mixed, and then it continued for the next six weeks. Results in this graph show that the relative amount of byproducts in the bath increases rapidly during the first few days, then the bath reaches equilibrium, and no further increase in the level of suppressor byproducts is observed. During this period, there was no visible increase in the defects on the wafers. These results indicate that the initial startup period has the largest change in the composition of the bath, but once the equilibrium is reached, relative levels of suppressor byproducts are low and stable.
7. Data collection began when the bath was freshly mixed, and then continued for the next six weeks. Results show that the relative amount of byproducts in the bath increases relatively rapidly during the first few days, then the bath reaches equilibrium. Reliability of analytical data During closed-loop operation, the reliability of analytical data sent to the dosing system is extremely important. A CMS can have its own outliers and unexpected malfunctions, which can cause erroneous results. With advanced software algorithms, unreliable results must be detected by the analyzer, identified as questionable and the solution must be reanalyzed. If the malfunction is serious and cannot be automatically corrected, the analyzer sends an alarm message to the plating tool and announces an inability to perform analysis. Such an approach assures accurate and reliable dosing of the plating bath when closed-loop communication between the plating tool and CMS unit is used. Once an analyzer is in a closed-loop mode in an electroplating tool and some historical data is available, analysis of previously obtained data allows prediction of an expected concentration of components in the plating solution. A comparison between the newly obtained data and expected value is then used for real-time detection and elimination of possible outliers during the analysis. If a newly obtained value is not in agreement with the predicted concentration, the analysis is repeated and a new comparison is performed. Only after this new verification will the analyzer send the results or error message to the dosing system. Conclusion With the damascene process gaining full acceptance in the high-volume manufacturing process for the next generation of semiconductors, reliable metrology is of utmost importance. CVS has shown that it is a highly accurate and precise method for the control of the organic additives. Closed-loop control enhances dosing system performance and prevents undesirable deterioration in plating solution quality. Author Information Paul Findeis is a plating process engineer at IBM , and has worked on electroless and electrolytic plating processes for 20+ years. He has a B.S. in chemical engineering from Cornell University, and an M.S. in materials science from Columbia University. Richard Henry is a development plating and surface preparation process engineer for IBM Microelectronics. He received a B.S in chemical engineering and an M.S in engineering management from Syracuse University. James Fluegel is a senior lab specialist in manufacturing engineering (plating and wet processing) for IBM Microelectronics, and has worked 35 years in plating. He has an associates degree in applied science from the State University of New York. Michael Pavlov is product manager at ECI Technology . He has a B.S. in metallurgy and an M.S. in electrochemistry from the Moscow University of Steel and Alloys in Russia. Eugene Shalyt is R&D manager at ECI Technology. He has an M.S and Ph.D in electrochemistry from the Mendeleev Russian University of Chemical Technology. Peter Bratin is vice president of ECI Technology. A graduate of Brooklyn College and CUNY, he has been in the plating industry for almost 20 years. He served as chairman of the Analytical Methods Committee and Research Board of AESF, and is currently a member of the Solderability Committee of IPC.
References
1. M. Yang, D. Mao, C. Yu, J. Dukovic and M. Xi, "Sub-100 nm Interconnects Using Multi-Step Plating," Solid State Tech. , October 2003. 2. P. Bratin, G. Chalyt, M. Pavlov and R. Sandor, "Automated On-Line Control of Plating Bath Additives Increases Wafer Yield," Semiconductor Fabtech, Summer 2001, 14th ed.
<<< Back | Print © 2004, Reed Business Information, a division of Reed Elsevier Inc. All Rights Reserved.
Closed-Loop Control of Copper Damascene Electroplating Solutions
Paul Findeis, Richard Henry and Jim Fluegel, IBM Microelectronics, East Fishkill, N.Y.; Michael Pavlov, Eugene Shalyt and Peter Bratin, ECI Technology, East Rutherford, N.J. -- 11/1/2004 Semiconductor International
Acid copper sulfate baths are employed in the damascene process to electrodeposit copper within fine trenches and vias in dielectric material on semiconductor chips.¹ There is a growing need to control the damascene process using control techniques that are accurate, sensitive, easy to employ and cost-effective. In addition, the analysis must be performed in online mode with automatic reporting of analytical data back to the dosing module of the plating tool to minimize operator error and establish closed-loop control of the plating solution.
At a Glance The control of plating bath components, primarily organic additives, is known to be critical in determining the resultant properties of the copper deposit.
In the copper electrodeposition process, it is critical to achieve uniformity of the copper film across the wafer surface, required mechanical properties of the deposit, and "super-filling" characteristics in the small features such as trenches and vias. These important properties are typically achieved using multi-component plating solutions, which include organic and inorganic components. Copper electroplating baths are normally formulated using highly stable electrolytes containing copper sulfate and sulfuric acid. Copper concentration in these electrolytes is 14-60 g/L and sulfuric acid is 1-240 g/L. Other components added into the bath are present in relatively small amounts. These components are organic additives and chloride ions. The organic additives, depending on the concentration and chemical composition, affect the properties of electrodeposited copper, including uniformity, hardness, ductility, tensile strength, etc. Organic additives added to copper electroplating baths typically include suppressors (polymers such as polyethylene glycols), accelerators (sulfur-contained compounds) and levelers (secondary suppressors). Organic components are consumed during the electroplating process at various rates, thus requiring individual control. Accelerators, for example, are consumed faster than other organic additives. Consumption of organic additives depends on various factors, including conditions of electroplating cell, current density, flow rate, number of wafers plated, etc. To keep the concentration of plating bath constituents in the optimum range for the electroplating process, several approaches can be used. The conventional approach employs automatic dosing of individual components into plating solution(s). The dosing system incorporates the hardware needed for the delivery of individual components into a solution and, if needed, removal of used-up solution from the bath. Replenishment of individual components is controlled via predictive software algorithms, mostly in conjunction with periodic analysis of solution composition. In high-volume production, the plating solution is controlled by continuous feed-and-bleed of virgin makeup solution combined with automatic dosing of organic additives, based on predictive algorithms and frequent solution analysis. This approach enables tight control of plating solution and limits buildup of additive by-products formed during the electroplating process. An advantage of this method, if performed properly, is that the solution is maintained at a virtual steady-state condition, and can be used for a long period of time without change. Despite the sophistication of dosing algorithms, the plating-bath composition can change rapidly and drift out of
operational range. Such bath changes occur because of intermittent or unpredictable operation of the bath, or can be caused by a malfunction of plating-bath delivery components, which can produce overdosing or underdosing of the plating bath. These problems can be prevented and avoided if each component of the plating bath is individually monitored with an online analytical system. The alternative approach of keeping the bath in control is based on short-term usage of the plating solution, which would be dumped once a certain number of wafers passed through the bath. According to the inventors, such an approach does not require analytical control. 2 However, this approach relies on the continuous mixing of the bath from individual components. Any failures in the dosing equipment or deviations in activity of the individual solutions cannot be detected, thus allowing this approach to produce defective wafers that will remain undetected until later stages of production. Another difficulty is that this approach works with continuously changing composition of the solutions, which are constantly discarded, rather than solutions under steady-state conditions and in relatively large reservoir. CVS analysis Among a wide range of analytical approaches that have been used for monitoring concentrations of organic additives in electroplating solutions, only cyclic voltammetric stripping (CVS) has demonstrated the ability to reliably monitor the activity of a wide range of components in the plating baths. In this method, the classical three-electrode cell is used where the main indicator electrode is a platinum rotating disc electrode. 2 CVS involves cycling the potential of a Ptrotating-disk electrode so metal is alternately plated and stripped at the electrode surface. Organic additives are detected by CVS from the effect they exert on the electrodeposition rate measured via the metal stripping peak area (Ar). This approach has become the most popular and effective method for controlling damascene copper deposition processes. Closed-loop plating bath control The first generation of CVS-based analyzers for the semiconductor industry, developed in 1997, was designed to support and assist developers researching the plating process. These first standalone systems did not report analytical information to the dosing unit forming a closed loop with the electroplating tool. First-generation systems have been installed in several research locations, and have produced valuable results for understanding and improvement of the plating process. After a few years of using these analytical systems in the field, confidence in their consistent results was gained, which opened opportunity to finally close a loop. The second generation of online analyzers, introduced in 1999, had a closed-loop communication capability as a feature of the standard software package. One of the earliest closed-loop experiences is shown in Figure 1 . Improvement in the bath control (accelerator component) before and after the closed-loop feature was engaged is depicted. As can be seen, the accelerator activity in the bath showed smaller variation once the closed loop was activated, which is related to continuous adjustment of the dosing algorithm based on analytical results.
1. Enabling closed-loop control function allowed improved precision of obtained results,
which is related to continuous adjustment of the dosing algorithm based on analytical results. A third generation of the CVS analyzer, employing a dual-cell configuration, enabled significant improvements in accuracy, precision, reliability, cost of ownership and time of analysis. It also has an improved communication capability, including a feature to communicate with various dosing systems serving different plating tool manufacturers. The improved communication protocol allows the user to send the data either in real time when analysis of one component is finished or in discrete mode once a complete set of analytical results is available. This generation of the analyzer also has the capability to monitor certain byproducts in the plating solutions. When starting a new plating bath, it is important to find the optimum operating conditions. In most cases during these startups, it is necessary to perform frequent and continuous analyses of the bath. However, once the tool is stabilized and steady state is achieved, continuous monitoring can be replaced by less frequent analyses. The following examples illustrate dynamic changes in the activity of constituents of a typical high-volume plating solution for copper damascene processes. Figures 2 and 3 show data for two components of a typical plating system running under closed-loop control. One component of the organic additive (Fig. 2) is continuously dosed into the bath and frequently analyzed by a chemical monitoring system (CMS). As a result of this approach, the activity of this component in the bath is very stable. Figure 3 shows the response of one inorganic component, which is generally more stable and does not require continuous dosing. However, it was monitored with the same frequency as the organic components. As can be seen in Figure 3 , the concentration of this component slowly changes with time, and since it is not automatically dosed, it must be added manually as needed. For such components with relatively low consumption (organic or inorganic), frequency of the analysis can be tuned by modifying the analysis schedule.
2. One component of the organic additive is continuously dosed into the bath and frequently analyzed by a chemical monitoring system (CMS). As a result of this approach, activity of this component in the bath is stable.
3. Inorganic components are monitored with the same frequency as organic components. Concentration of this component is slowly changing with time, and since it is not automatically dosed, it must be added manually as needed. During routine operation when a plating bath performs well, activity of all components in the bath is stable and no adjustments to the dosing algorithm are required. However, during high-volume use of the plating system, concentration of one or several components might change because of evaporation, drag-out, decomposition, or malfunctioning of dosing or other parts of the plating system. It is for these reasons that analysis is performed more frequently, to detect the problem early and allow ample time for troubleshooting before the plating system must be halted. Some examples of a bath going out of control are shown in Figure 4 , which illustrates the behavior of organic components when the closed-loop feature is turned off. While the closed loop was disabled, the dosing system kept inserting additives into the bath according to a predetermined algorithm. This caused the activity of one organic component to increase rapidly, causing an overdose. Once closed-loop communication was restored, activity slowly returned to the target level, and was maintained there by the closed loop. This demonstrates the ability of the dosing system to appropriately use analytical results and perform immediate corrective actions to adjust activity of any organic component in the bath.
4. The closed loop was disabled and the dosing system kept adding additives into the bath according to a predetermined algorithm. This caused activity of one organic component to increase rapidly, causing an overdose. Other organic components were also analyzed at the same frequency while the closed-loop feature was turned off. As can be seen from Figure 4, their activity did not change significantly since their consumption rate is generally much lower than that of the first component. Considering the consumption of individual components in a plating solution, altering the analysis schedule can optimize monitoring of the frequency of each component. One component can be monitored more often than others. Optimized dosing of all components of the plating bath is shown in Figures 5 and 6. This bath is controlled using an optimized schedule of analysis and closed-loop communication with a plating tool. The
data shown in Figures 5 and 6 were collected during a two-week period.
5. Monitoring frequency of each organic component can be optimized by altering analysis schedule and providing the best replenishment conditions.
6. The bath is controlled using optimized schedule of analysis and closed-loop communication with a plating tool. As we mentioned above, during long-term use of a plating bath, there is a potential to accumulate additive byproducts, which eventually might interfere with the plating process. In the most recent design of the analyzer, monitoring of some byproducts can be accomplished without any changes in hardware and without addition of auxiliary reagents. Monitoring of suppressor byproducts in a copper plating bath for an extended period of time is shown in Figure 7 . Data collection began when the bath was freshly mixed, and then it continued for the next six weeks. Results in this graph show that the relative amount of byproducts in the bath increases rapidly during the first few days, then the bath reaches equilibrium, and no further increase in the level of suppressor byproducts is observed. During this period, there was no visible increase in the defects on the wafers. These results indicate that the initial startup period has the largest change in the composition of the bath, but once the equilibrium is reached, relative levels of suppressor byproducts are low and stable.
7. Data collection began when the bath was freshly mixed, and then continued for the next six weeks. Results show that the relative amount of byproducts in the bath increases relatively rapidly during the first few days, then the bath reaches equilibrium. Reliability of analytical data During closed-loop operation, the reliability of analytical data sent to the dosing system is extremely important. A CMS can have its own outliers and unexpected malfunctions, which can cause erroneous results. With advanced software algorithms, unreliable results must be detected by the analyzer, identified as questionable and the solution must be reanalyzed. If the malfunction is serious and cannot be automatically corrected, the analyzer sends an alarm message to the plating tool and announces an inability to perform analysis. Such an approach assures accurate and reliable dosing of the plating bath when closed-loop communication between the plating tool and CMS unit is used. Once an analyzer is in a closed-loop mode in an electroplating tool and some historical data is available, analysis of previously obtained data allows prediction of an expected concentration of components in the plating solution. A comparison between the newly obtained data and expected value is then used for real-time detection and elimination of possible outliers during the analysis. If a newly obtained value is not in agreement with the predicted concentration, the analysis is repeated and a new comparison is performed. Only after this new verification will the analyzer send the results or error message to the dosing system. Conclusion With the damascene process gaining full acceptance in the high-volume manufacturing process for the next generation of semiconductors, reliable metrology is of utmost importance. CVS has shown that it is a highly accurate and precise method for the control of the organic additives. Closed-loop control enhances dosing system performance and prevents undesirable deterioration in plating solution quality. Author Information Paul Findeis is a plating process engineer at IBM , and has worked on electroless and electrolytic plating processes for 20+ years. He has a B.S. in chemical engineering from Cornell University, and an M.S. in materials science from Columbia University. Richard Henry is a development plating and surface preparation process engineer for IBM Microelectronics. He received a B.S in chemical engineering and an M.S in engineering management from Syracuse University. James Fluegel is a senior lab specialist in manufacturing engineering (plating and wet processing) for IBM Microelectronics, and has worked 35 years in plating. He has an associates degree in applied science from the State University of New York. Michael Pavlov is product manager at ECI Technology . He has a B.S. in metallurgy and an M.S. in electrochemistry from the Moscow University of Steel and Alloys in Russia. Eugene Shalyt is R&D manager at ECI Technology. He has an M.S and Ph.D in electrochemistry from the Mendeleev Russian University of Chemical Technology. Peter Bratin is vice president of ECI Technology. A graduate of Brooklyn College and CUNY, he has been in the plating industry for almost 20 years. He served as chairman of the Analytical Methods Committee and Research Board of AESF, and is currently a member of the Solderability Committee of IPC.
References
1. M. Yang, D. Mao, C. Yu, J. Dukovic and M. Xi, "Sub-100 nm Interconnects Using Multi-Step Plating," Solid State Tech. , October 2003. 2. P. Bratin, G. Chalyt, M. Pavlov and R. Sandor, "Automated On-Line Control of Plating Bath Additives Increases Wafer Yield," Semiconductor Fabtech, Summer 2001, 14th ed.
<<< Back | Print © 2004, Reed Business Information, a division of Reed Elsevier Inc. All Rights Reserved.
Related Documents
Closed Loop Control Of Coppers
June 2020 14
Closed Loop
June 2020 16
Closed Loop Power Control For Lte Uplink
June 2020 16
Control Loop Coockbook
May 2020 11
