US20050067290A1

US20050067290A1 - Method and system for automatically controlling a current distribution of a multi-anode arrangement during the plating of a metal on a substrate surface

Info

Publication number: US20050067290A1
Application number: US10/861,997
Authority: US
Inventors: Matthias Bonkass; Dirk Wollstein; Axel Preusse
Original assignee: Individual
Current assignee: Advanced Micro Devices Inc
Priority date: 2003-09-30
Filing date: 2004-06-04
Publication date: 2005-03-31
Also published as: DE10345376B4; DE10345376A1; CN1860259A; TW200514872A

Abstract

An electroplating tool is operated in combination with a controller which automatically determines the individual currents for a multi-anode configuration of the plating tool. The calculation of the anode currents may be based on sensitivity data and measurement data as well as on a desired target profile, so that a fast response with respect to process variations may be achieved even for a plating tool including a plurality of process chambers.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention
The present invention relates to the process of depositing a metal on a substrate surface using a reactor for electroplating, and, more particularly, to the adjustment of the current supplied to a multi-anode arrangement of a plating tool to obtain a desired thickness profile of the metal across the substrate surface.
2. Description of the Related Art
In many technical fields, the deposition of metal layers on a substrate surface is a frequently employed technique. For efficiently depositing relatively thick metal layers on a substrate surface, plating, in the form of electroplating or electroless plating, has proven to be a viable and cost-effective method. Thus, plating has become an attractive deposition method in the semiconductor industry.
Nowadays, copper is considered a preferred candidate in forming metallization layers in sophisticated integrated circuits due to the superior characteristics of copper and copper alloys in view of conductivity and resistance to electromigration compared to, for example, the commonly used aluminum. Since copper may not be deposited very efficiently by physical vapor deposition, for example by sputter deposition, with a layer thickness on the order of 1 μm and more, electroplating of copper and copper alloys is the currently preferred deposition method in forming metallization layers. Although electroplating of copper is a well-established technique, reliably depositing copper over large diameter substrates, having a patterned surface including trenches and vias, is a challenging task for process engineers. For example, forming a metallization layer of an ultra-large scale integration device requires the reliable filling of wide trenches with a width on the order of micrometers and also requires the filling of vias and trenches having a diameter or width of 0.2 μm or even less. The situation gains even more in complexity as the diameters of the substrates tend to increase. Currently, eight or even ten inch wafers are commonly used in a semiconductor process line. Thus, great efforts are being made in the field of copper plating to provide the copper layer in a desired profile across the substrate surface. At a first glance, it appears to be advantageous that the metal thickness profile across the substrate surface may be formed as uniformly as possible. However, post-plating processes may require a differently shaped profile so as to assure proper device functionality of the completed integrated circuits. For instance, during the formation of copper-based metallization layers, excess copper may be removed, which is presently often achieved by chemical mechanical polishing (CMP) of the metal surface. Since the CMP process is per se a highly complex process frequently exhibiting an intrinsic process non-uniformity, i.e., a non-uniform removal rate across the substrate surface, it may be preferable to adapt the metal thickness profile to the post-plating process to achieve in total an improved process uniformity after completion of the post-plating process. Therefore, electroplating tools are often configured so as to allow a variation of the metal profile, wherein the control of the finally obtained profile presently is, however, cumbersome and time consuming.
With reference to FIG. 1, a typical prior art electroplating system will now be described to illustrate the problems involved in electroplating copper in more detail. In FIG. 1 a, there is shown a typical conventional electroplating system 100 including a reactor vessel 101 with a first electrode 102, in this case the anode, having a plurality of individually drivable anode portions 102A . . . 102N, thereby defining a multi-anode configuration. In this example, a so-called fountain type reactor is considered, in which an electrolyte solution is directed from the bottom of the reactor vessel 101 to the top side and is then re-circulated by a pipe 103 connecting an outlet 104 with a storage tank 107, which in turn is connected to an inlet 105 provided as a passage through the anode 102. The system 100 further comprises a substrate holder 108 that is configured to support a substrate 109, such as a semiconductor wafer, so as to expose a surface of interest to the electrolyte. Moreover, the substrate holder 108 may be configured to act as a second electrode, in this case the cathode, and to provide the electrical connection to a power source 110, which is configured to enable the supply of individual currents of defined magnitude to each of the anode portions 102A . . . 102N.
FIG. 1 b schematically shows a top view of the electrode 102 including the multi-anode configuration 102A . . . 102N for four individual anode portions.
Prior to installing the substrate 109 on the substrate holder 108, a thin current distribution layer, possibly including a seed layer, typically provided by sputter deposition, is formed on the surface of the substrate 109 that will receive the metal layer. Thereafter, the substrate 109 is mounted on the substrate holder 108, wherein small contact areas (not shown for the sake of simplicity) provide electrical contact to the power source 110 via the substrate holder 108. By activating a pump (not shown) and applying appropriate voltages between the anode 102, that is, the multi-anode configuration 102A . . . 102N, and the substrate holder 108 that creates respective currents, an electrolyte flow is created within the reactor vessel 101. The electrolyte entering the reactor vessel 101 at the inlet 105 is directed towards the substrate 109, wherein the deposition of metal on the substrate 109 is determined by the flow of electrolyte and the arrangement of the multi-anode configuration 102A . . . 102N, since the local deposition rate of metal on a specific area of the surface of the substrate 109 depends on the number of ions arriving at this area. Hence, by selecting a set of currents supplied to the multi-anode configuration 102A . . . 102N, the finally-obtained thickness profile may be determined, wherein, optionally, additional means for influencing the ion and/or electrolyte flow may be inserted in the form of, for instance, a diffuser plate.
Once an appropriate set of currents is adjusted in the power supply 110, the resulting thickness profile is determined by the characteristics of the reactor vessel 101, the electrolyte solution, the set of currents and the plating time. Hence, a variation of one of these characteristics may lead to a drift of the finally-obtained thickness profile. The situation is even more complex for an electroplating tool 100 including a plurality of reactor vessels 101 with a corresponding plurality of multi-anode configurations 102A . . . 102N, since then any subtle process fluctuation in any of these reactor vessels may occur and may result in a highly complex mutual interaction of the involved process characteristics, thereby compromising process stability. Thus, a plurality of test substrate runs is typically performed on a regular basis, thereby requiring time and manpower and hence reducing the yield and quality of the plating process.
Furthermore, in forming metallization layers by the so-called damascene technique, vias and trenches are filled with metal and a certain degree of excess metal has to be provided so as to reliably fill the vias and trenches. Subsequently, the excess metal has to be removed to ensure electrical insulation between adjacent trenches and vias and to provide a planar surface for the formation of further metallization layers. A preferred technique for removing excess metal and planarizing the substrate surface is chemical mechanical polishing (CMP), in which the surface material to be removed is subjected to a chemical reaction and is simultaneously mechanically removed. As previously explained, the CMP process is highly complex and may exhibit a non-uniformity, which may at least partially be compensated for by adapting the thickness profile of the electroplating process to the CMP non-uniformity. However, the plurality of process parameters involved in creating the thickness profile, especially for plating tools having a plurality of reactor vessels, may lead to significant deviations from the desired thickness profile, thereby rendering the compensation of CMP non-uniformities inefficient.
Accordingly, in view of the above problems, a need exists for a technique enabling rapid and efficient adjustment of a thickness profile in a plating tool so as to eliminate or at least reduce some or all of these problems.

SUMMARY OF THE INVENTION

Generally, the present invention relates to a technique for controlling the individual currents supplied to individual anodes of a multi-anode configuration in an electroplating tool, wherein the individual currents, also referred to as a set of currents, for the multi-anode configuration are calculated in an automated fashion in view of a desired thickness profile, thereby providing the potential for responding to process fluctuations of the plating process itself and/or to a post-plating process and/or to a pre-plating process in a substantially non-delayed manner, even if the plating tool under consideration includes a plurality of process chambers with multi-anode configurations.
According to one illustrative embodiment of the present invention, a method comprises determining sensitivity data quantitatively relating a set of currents for a multi-anode configuration of an electroplating tool to a thickness of a metal layer formed on a substrate by electroplating. The method further includes determining an updated set of currents for the multi-anode configuration on the basis of the sensitivity data for a second substrate to be processed in the electroplating tool.
In a further illustrative embodiment of the present invention, a method of depositing metal in an electroplating tool having at least one process chamber including a multi-anode configuration is provided. The method comprises determining a set of currents for the multi-anode configuration on the basis of a desired thickness profile, thickness profile data obtained from at least one substrate processed in the electroplating tool and a model quantitatively describing a relation between current supplied to the multi-anode configuration and a thickness profile. Finally, metal is deposited on one or more substrates while using the determined set of currents.
According to yet another illustrative embodiment of the present invention, a method of controlling an electroplating tool including a plurality of process chambers having a multi-anode configuration is provided. The method comprises calculating a set of currents for each multi-anode configuration and processing at least one substrate in each of the plurality of process chambers with the determined sets of currents.
According to still another illustrative embodiment of the present invention, a controller for an electroplating tool comprises a calculation unit configured to determine, for a substrate to be processed in the electroplating tool, a set of currents for at least one multi-anode configuration, wherein the calculation is based on a desired thickness profile.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention may be understood by reference to the following description taken in conjunction with the accompanying drawings, in which like reference numerals identify like elements, and in which:
FIG. 1 a schematically depicts a conventional electroplating tool having a multi-anode configuration;
FIG. 1 b schematically shows a top view of the multi-anode configuration of the tool of FIG. 1 a;
FIG. 2 schematically illustrates an electroplating system including a controller for automatically determining a set of currents on the basis of various criteria in accordance with illustrative embodiments of the present invention; and
FIG. 3 is a graph illustrating results of thickness profile measurements of a plurality of substrates processed in different process chambers in accordance with a control strategy of illustrative embodiments of the present invention.
While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and are herein described in detail. It should be understood, however, that the description herein of specific embodiments is not intended to limit the invention to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.

DETAILED DESCRIPTION OF THE INVENTION

Illustrative embodiments of the invention are described below. In the interest of clarity, not all features of an actual implementation are described in this specification. It will of course be appreciated that in the development of any such actual embodiment, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which will vary from one implementation to another. Moreover, it will be appreciated that such a development effort might be complex and time-consuming, but would nevertheless be a routine undertaking for those of ordinary skill in the art having the benefit of this disclosure.
The present invention will now be described with reference to the attached figures. Although the various regions and structures of a semiconductor device are depicted in the drawings as having very precise, sharp configurations and profiles, those skilled in the art recognize that, in reality, these regions and structures are not as precise as indicated in the drawings. Additionally, the relative sizes of the various features and doped regions depicted in the drawings may be exaggerated or reduced as compared to the size of those features or regions on fabricated devices. Nevertheless, the attached drawings are included to describe and explain illustrative examples of the present invention. The words and phrases used herein should be understood and interpreted to have a meaning consistent with the understanding of those words and phrases by those skilled in the relevant art. No special definition of a term or phrase, i.e., a definition that is different from the ordinary and customary meaning as understood by those skilled in the art, is intended to be implied by consistent usage of the term or phrase herein. To the extent that a term or phrase is intended to have a special meaning, i.e., a meaning other than that understood by skilled artisans, such a special definition will be expressly set forth in the specification in a definitional manner that directly and unequivocally provides the special definition for the term or phrase.
It is further to be noted that the detailed description will refer to electroplating of a metal, such as copper, on substrates such as those typically used in semiconductor fabrication, since the present invention is particularly useful in a process sequence with sensitive post-plating processes, such as CMP. It will be readily appreciated, however, that the present invention is applicable to any plating process with an externally impressed current (electroplating), of any types of substrates requiring a specified deposition profile on a substrate surface or a portion thereof. Moreover, although the description refers to a fountain type plating reactor, for example as schematically illustrated in FIG. 1 a, other types of reactors, such as electrolyte baths and the like, may be used as well. Thus, the present invention is to be understood as not being restricted to a specific type of electroplating reactor unless such restrictions are explicitly set forth in the claims.
One particular feature provided by the present invention is the potential for rapidly responding to a change of process conditions within an electroplating tool or in any other process preceding or following the plating process. Such changing process conditions may be, for instance, a variation of the characteristics of a plating solution owing to a subtle fluctuation of characteristics of sensitive additives included in the plating solution, or the deterioration of any consumables in the plating tool or in a post-plating CMP tool. A change of process conditions may be either known in advance, for instance the change of a consumable in a CMP tool, or may be detected by any appropriate sensor element. Other changes of process conditions may not be per se “visible” and may not be compensated for in an efficient manner in conventional plating tools, since typically a time-consuming readjustment with a subsequent test period is required after recognizing the process drift. In view of the problems previously explained with reference to FIGS. 1 a and 1 b, the present invention provides a control strategy in which one or more sets of currents for operating one or more multi-anode configurations may be recalculated on a time scale that is negligible compared to the time required for handling and processing a substrate in an electroplating tool, thereby providing the potential for responding to process fluctuations in a substantially non-delayed fashion. In some embodiments, the recalculation of corresponding sets of currents for various multi-anode configurations may be based on the occurrence of any readily detectable process variations, wherein, for instance, the calculation may be performed under the condition that the total current supplied to each of a plurality of the multi-anode configurations is substantially the same, thereby assuring that the amount of metal deposited is substantially the same for identical process times. On the other hand, the automated recalculation of the anode currents also enables an efficient and, if desired, a substantially continuous response to any non-visible changes that may only be identified by a deviation from a desired target profile.
Therefore, in other embodiments, the control strategy for an electroplating process is based on the concept of maintaining a deviation from a desired thickness profile within a predefined allowable range. The desired thickness profile of a metal layer to be deposited on a substrate surface may be expressed by a function T(r), wherein T is assumed to represent a thickness value at a position r on the substrate surface. Although the variable r may represent any position on a planar or non-planar substrate surface, it is assumed in the following that r represents a distance from the center of a substantially disc-shaped substrate, such as a wafer in the semiconductor industry. Hence, the desired thickness profile T(r) is assumed to have an axial symmetry, wherein, however, it is to be understood that the principles of the present invention also apply to an arbitrary function T(r). Similarly, an actual thickness profile of a metal layer may be denoted as M(r), wherein M is to represent a metal thickness at the position r. Hence, according to some embodiments of the present invention, it may be considered appropriate to maintain a deviation of the actual thickness profile M(r) from the desired thickness profile T(r) within a predefined allowable range, irrespective of fluctuations of an electroplating process or any pre- and post-plating processes.
As previously explained with reference to FIGS. 1 a and 1 b, the actual thickness profile M(r) may be influenced by the current supplied to a specified anode portion 102A . . . 102N, wherein the degree of influence or sensitivity of each of the anode portions 102A . . . 102N may be determined in advance, assuming that the influence is affected by subtle process variations of the electroplating tool in a negligible manner only. The influence of the current supplied to an individual anode portion 102A . . . 102N may be referred to hereinafter as sensitivity, and corresponding data may be denoted as sensitivity data. By determining the sensitivity data for a specified electroplating tool, such as the tool 100 described with reference to FIGS. 1 a and 1 b, a corresponding set of anode currents may be calculated on the basis of the desired thickness profile T(r) and the actual thickness profile M(r) that may be obtained in the form of discrete measurement data from a previously processed substrate. Typically, post-plating thickness measurements are performed on at least some representative positions of a substrate which may, according to the present invention, be used to determine an updated set of currents for one or more multi-anode configurations, such as the configuration 102. The transformation of discrete measurement data for an actual thickness profile into a substantially continuous function M(r) may readily be obtained by interpolation, data fit procedures, and any other data manipulation techniques as are well known in the art. The same holds true for the sensitivity data, which may be provided as discrete measurement values for a plurality of positions and for a plurality of different current values for each of the individual anode portions 102A . . . 102N. For instance, corresponding sensitivity data may be obtained by varying the current for a specified anode portion and measuring the obtained actual metal thickness after a specified deposition time. Thereafter, the deposition process may be resumed or restarted with another one of the anode portions 102A . . . 102N. From the plurality of discrete measurement values, a continuous function for the sensitivity of each individual anode portion, referred to as S_i(r), may be derived by one of the above-mentioned data manipulation methods, wherein the index i represents one of the anode portions 102A . . . 102N. Hence, based on the set of currents used for generating the actual thickness profile M(r), an updated set of currents may be calculated such that an expected thickness profile for a substrate to be processed by the plating tool 100 has a deviation from the desired thickness profile T(r) that is within the well-defined allowable range.
To this end, it may be convenient to express the difference between an actual thickness profile M(r) to be obtained and the desired thickness profile T(r) by a position-dependent term and a position-independent term representing a constant offset of the actual metal layer with respect to the desired ideal metal layer according to the following equation: $\begin{matrix} M (r) - T (r) = E (r) + M^{Offset} & (1) \end{matrix}$
wherein E(r) represents the position dependent deviation or excess material, and M^Offsetrepresents the position-independent deviation. Hence, the sum of the position dependent portion E(r) across the entire substrate surface A equals zero, that is: $\begin{matrix} \int_{W} E (r) ⅆ A = 0 & (1^{'}) \end{matrix}$
With the assumption of a substantially constant plating time and with a substantially constant total current I^SUMrepresenting the sum of the individual currents supplied to the anode portions 102A . . . 102N, that is: $\begin{matrix} \sum_{i = 102 A}^{102 N} I_{i} = I^{SUM} & (2) \end{matrix}$
the position-independent term M^Offsetmay be close to zero since the quantity of metal deposited onto the substrate substantially depends on the total current supplied to the anode portions and the plating time. Consequently, changing the currents to the individual anode portions 102A . . . 102N may essentially affect the position dependent term E(r), whereas minor fluctuations of the plating time will, according to the above-stated assumptions, mainly affect the position independent term M^Offset, causing this value to slightly vary around zero. Consequently, in one embodiment an updated set of currents supplied to the individual anode portions 102A . . . 102N may be based on the concept to reduce or minimize the amount of E(r) in Equation 1.
To this end, the updated set of currents may be denoted as I^updated=(I_102A ^updated, . . . I_102N ^updated), while the set of currents used for creating an actual thickness profile of a previously processed substrate may be denoted as I^O=(I_102A ^O, . . . , I_102N ^O). Similarly, the corresponding function representing the thickness profile of the previously processed substrate may be denoted as M^O(r), wherein, as previously explained, the function M^O(r) may be obtained by a corresponding set of measurement values that may be taken on a plurality of different positions r in accordance with standard thickness measurement procedures. The continuous or quasi-continuous function M⁾(r) may then be obtained by interpolation, data fit, and any other well-established data manipulation procedures. With the assumption of a substantially constant plating time of the plating process in obtaining the thickness profile M^O(r), the position dependent term M^Offset(see Equation 1), although expected to be close to zero may be calculated by using Equation 1 as the sum or integral taken over the entire area of the substrate from the difference of M^O(r) and T(r), since the sum of the position dependent part E(r) is zero (see Equation 1′), as previously explained with reference to Equation 1. Thus, the relatively small position independent portion M^Offsetmay be obtained by Equation 3 in the following way: $\begin{matrix} M^{Offset} = \frac{1}{A^{W}} \int_{W} (M^{O} (r) - T (r)) ⅆ A & (3) \end{matrix}$
wherein A^Wrepresents the total area of the substrate on which metal is plated.
Since M^Offsetmay be calculated according to Equation 3 for a class of allowable functions representing the position dependent deviation E(r), a corresponding thickness profile for a subsequent substrate, indicated as M(r), may then be calculated on the basis of Equation 1.
Since the control variable of the plating process is the set of currents I_102A, . . . , I_102N, which may be correlated to the thickness profile M(r) by means of the sensitivity functions S_102A(r), . . . , S_102N(r), the plating process may be modeled by establishing a correlation between the thickness profile of the previously processed substrate M^O(r) and the sensitivity functions S_102A(r), . . . , S_102N(r)so as to obtain the updated thickness profile M(r) as a function of the individual anode currents I_102A, . . . , I_102N. In one embodiment, a linear relationship may be used for the plating model in a form as, for example, given by the following Equation 4: $\begin{matrix} M (r, I_{102 A}, \dots, I_{102 N}) = M^{O} (r) + \sum_{i = 102 A}^{102 N} S_{i} (r) (I_{i} - I_{i}^{O}) & (4) \end{matrix}$
wherein the index i represents the individual anode portions 102A . . . 102N. It should be noted that other relationships may be used, as long as these relationships express the influence of a changing of the individual anode currents on a given thickness profile, thereby creating a new thickness profile. In Equation 4, the thickness profile M^O(r) is preferably based on measurement data obtained from a previously processed substrate, wherein the delay with respect to a substrate that is currently to be processed on the basis of the calculated thickness profile M(r) is moderately small so as to provide the potential for a short response time to any process fluctuations.
In other embodiments, however, it may be considered appropriate to select the function M^O(r) on the basis of averaged measurement data and/or predetermined non-experimental data and the like. For instance, at an initial phase of the plating process in which measurement data may not be available, appropriate reference data may be used for the function M^O(r) or the desired target profile T(r) may be used.
In one particular embodiment, the stability of the control process may be enhanced in that an allowable range for the individual anode currents I_102A, . . . ,I_102Nis appropriately selected. That is, for each of the anode currents I_102A, . . . , I_102N, an upper limit and a lower limit may be selected such that the resulting thickness profile M(r) is obtained by a set of currents with each individual anode current being within its allowable range. For instance, based on experience gathered from previous plating processes, target values for the individual anode currents may be determined and a respective allowable range for each anode current may then be set. In other cases, respective target values and associated allowable ranges may be determined on the basis of the measurement data obtained during determining the sensitivity functions, since, in this operation phase of the plating tool, the influence of the individual anode current may conveniently be investigated. The target value and the corresponding allowable range for each anode portion 102A . . . 102N may also be determined on the basis of tool specifications and tool and/or process requirements. Typically, a range of approximately 10-20% variation with respect to the respective target value of the individual anode currents may result in sufficient control stability.
In the following, the upper and lower limits for the respective anode currents I_i, I=102A . . . 102N, are denoted as I_i ^Land I_i ^H, respectively. Hence, for a given allowable position dependent deviation E(r), a corresponding updated set of currents may be calculated on the basis of Equations 1, 3 and 4.
In one particular embodiment, the updated set of currents may be obtained by determining the required thickness profile M(r) with a minimal deviation E(r). That is, the updated set of currents, denoted as $I^{updated} = I_{102 A}^{updated}, \dots, I_{102 N}^{updated}$
may be obtained by solving the following Equation 5: $\begin{matrix} \min_{102 A, \dots, 102 N} \int_{0}^{D} \langle M (r, I_{102 A}, \dots, I_{102 N}) - M^{Offset} - T (r) \rangle ⅆ r & (5) \end{matrix}$
In one particular embodiment, secondary conditions for the individual anode currents are used and are set such that the individual currents are within an allowable operating range, for example as determined above, and/or that the sum of the individual anode currents is substantially equal to a predefined value. Hence, these secondary conditions may be represented by the following Equations 6 and 7:
I_i ^L≦I_i≦I_i ^U (6)
$\begin{matrix} \sum_{i = 102 A}^{102 N} I_{i} = I^{sum} & (7) \end{matrix}$
In Equation 5, the integral may be performed as a one-dimensional integral from the substrate center to the edge of the substrate, indicated as D, or the integral may be performed as a two-dimensional integral over the entire substrate surface. In one embodiment as shown in Equation 5, the one-dimensional integral is used since the contribution of the substrate center is emphasized in the one-dimensional representation, which may be advantageous for thickness profiles that are critical in the vicinity of the substrate center. For instance, a post-plating CMP process may exhibit a significant removal rate variation in the vicinity of the substrate center so that a correspondingly sensitively adapted plating profile may be advantageous.
By inserting Equation 4 into Equation 5 and using numerical integration methods, the updated anode currents may be calculated. For instance, a personal computer that has an appropriate instruction set installed therein, for example in the form of Matlab® plug-in, may be used to obtain numerical values for the updated anode currents. Depending on the hardware requirements for a corresponding control unit and depending on the desired accuracy of the calculations performed, any other appropriate implementation may be used, such as a correspondingly programmed microcomputer, or any other appropriately arranged circuitry, including analog and/or digital designs, may be used. As will be explained with reference to FIG. 2, a remote device may also be used that is operatively connectable to a corresponding plating tool so as to convey the calculated, updated set of currents to the plating tool within a time span that is sufficient to allow the desired control operation.
As previously explained, it may be advantageous to assume that the plating time is substantially constant for the substrate to be processed by the electroplating tool 100. In other embodiments, a certain degree of variation of the plating time may be taken into account by correspondingly recalculating an updated plating time T^updatedby using the previously determined position-independent deviation M^Offset. For instance, the updated plating time T^updatedmay be calculated from the plating time T^Ofor the previously processed substrate by means of a pre-established relationship. In one example, the updated plating time T^updatedmay linearly be related to the position independent deviation M^Offsetand the previous plating time T^Oin the form of the following Equation 8: $\begin{matrix} T^{updated} = T^{O} - \frac{1}{γ} M^{Offset} & (8) \end{matrix}$
wherein γ represents a sensitivity factor of the plated metal thickness with respect to a change of the plating time. A respective numerical value for γ may readily be obtained by measuring the increase in thickness within one or more well-defined plating time periods. Hence, any deviations of the thickness profile in which the thickness profiles practically shifted as a whole with respect to the target profile T(r) may efficiently be compensated for by correspondingly recalculating the updated plating time T^updated. As previously explained, a minor change of the plating time does not substantially affect the position dependent deviation E(r), while the position independent term M^Offsetstill remains very small, so that the variation in the plating time is also small and corresponding updated versions of the plating time may not unduly influence the stability of the control strategy or the updated anode currents when determined in the above-described manner.
With reference to FIGS. 1 a-1 b and 2, implementations of one or more of the above-identified control strategies into a plating tool will now be described in more detail.
FIG. 2 schematically shows a plating tool 200 which may comprise one or more reactor vessels including multi-anode configurations, such as previously described with reference to FIGS. 1 a-1 b. Hence, the corresponding reactor vessels may be denoted as 101 and a description thereof is given with respect to FIGS. 1 a-1 b. The reactor vessels 101 a, 101 b . . . are connected to respective controllable power supplies 201 a, 201 b . . . that are configured to provide a set of currents to the corresponding multi-anode configurations within the reactor vessels 101 a, 101 b . . . For convenience, a respective set of currents supplied to a corresponding reactor vessel 101 a, 101 b . . . is referred to as I_102A, . . . , I_102N, wherein the number of anode portions in each of the reactor vessels depends on the design of the plating tool 200. As explained with reference to FIGS. 1 a-1 b, four anode portions 102A . . . 102N may be provided having an axial symmetry, wherein in other plating tools the number of anode portions may be as low as two and may be well above four. The plating tool 200 may also be considered as a system of a plurality of different plating tools, each having a differently designed reactor vessel, wherein the number of anode portions may also be different in at least some of the plurality of plating tools. Similarly, the reactor vessels 101 a, 101 b may differ from each other in the reactor design, the number of anode portions, and the like. It should also be pointed out that the multi-anode configurations of the individual reactor vessels 101 a, 101 b . . . may not necessarily exhibit an axial symmetry, but may have any geometric arrangement as is deemed appropriate. For instance, in some cases, a thickness profile may be desirable that exhibits non-axial symmetry, and hence a corresponding multi-anode configuration may be provided. For example, the multi-anode configuration 102A . . . 102N as shown in FIGS. 1 a-1 b may be provided in the form of sections of a circle, wherein individual sections are isolated from each other and represent one portion of the multi-anode configuration. In such a case, the previously explained control strategy may be used as well, wherein the corresponding position dependent functions and terms have to be expressed in terms of two-dimensional coordinates rather than by a one-dimensional radial component.
The plating tool 200 is operatively connected with a controller 250, wherein the operative connection is indicated by 251 and is intended to represent any connection enabling a data transfer at least from the controller 250 to the plating tool 200. In particular embodiments, the connection 251 represents a data communication line, in a wired or wireless form, to communicate an appropriate control signal to the respective power supplies 210 a, 210 b . . . , which in turn cause the power supplies 210 a, 210 b . . . to output a respective set of currents I_102A, . . . , I_102Nfor each of the reactor vessels 101 a, 101 b . . . according to calculation results performed in the controller 250. The controller 250 may be implemented within a work station, a PC, or a facility management system, which have a corresponding calculation unit provided therein so as to perform one or more of the embodiments described above to establish an updated set of currents. The controller 250 may also include any interface and communications sections that are necessary to provide information on the updated set of currents to the power supplies 210 a, 210 b . . . via the operative connection 251. In other embodiments, the controller 250, for instance in the form of an appropriately programmed microprocessor, an ASIC (application specific integrated circuit), and the like, may be implemented into or may be provided in addition to a control unit (not shown) as is typically included in conventional electroplating tools for controlling the operation thereof. Although the computational capabilities of the controller 250 determine accuracy and speed of the numerical calculations, for instance in solving Equation 5, the controller 250 typically provides the results within a time interval that is negligible compared to any time intervals related to the processing of a substrate in the plating tool 200. The controller 250 is further configured to receive data from the environment on which the calculation of updated anode currents may be based. In one particular embodiment, the controller 250 is configured to receive sensitivity data from an external source, such as an operator, a computer, a measurement device, and the like. Hereby, the sensitivity data may be provided in the form of discrete measurement values, discrete theoretical values, a mathematical function, or in the form of any other appropriate information relating the effect of at least some of the anode portions 102A . . . 102N on a thickness profile upon operation with respective anode currents I_102A. . . I_102N. Depending on the format of the sensitivity data, the controller 250 may be configured to store and convert the sensitivity data in any appropriate form enabling the usage thereof in the calculation unit of the controller 250 for establishing updated anode currents.
In other embodiments, the controller 250 may be configured to receive thickness profile data in the form of discrete measurement values, a substantially continuous function, and the like, wherein the format of the thickness profile data may be converted into any appropriate representation required for performing the calculations explained above. Similarly, the controller 250 may be configured to receive externally supplied profile data representing a desired thickness profile and/or the controller 250 may include one or more desired thickness profiles in any convenient representation, which may be used upon request for the control operation. In one particular embodiment, the controller 250 is operatively connected to a thickness profile measurement system (not shown) so as to directly receive measurement data from previously processed substrates, thereby providing the potential for a closed loop control function, wherein the response time of the closed loop is substantially determined by the time delay for providing thickness measurement data to the controller 250. As previously pointed out, due to the automatic calculation of updated anode currents, the time for establishing the anode currents is negligible compared to any other times in processing a substrate in the plating tool 200 or in any other measurement system involved in the control operation.
In other embodiments, the controller 250 may be configured to receive additional information, such as post-plating process data representing process characteristics of subsequent processes, such as a CMP process as is typically used in the fabrication of copper-based metallization layers, or pre-plating process data relating to information on processes prior to the plating process, such as the deposition of current distribution layers, seed layers, and the like. Moreover, status information of the plating tool 200 may be supplied to the controller 250 and may be used in establishing new anode currents. In one embodiment, the sensitivity data may be correlated to the status information of the plating tool 200 so as to reduce a drift of the sensitivity data with respect to the tool status. For example, it may be known that the sensitivity data may depend on the overall process time owing, for instance, to a change of certain characteristics, such as a change of the characteristics of the plating solution over time. A corresponding well-known dependency may therefore be readily incorporated in the above-described control strategies by correspondingly adapting the sensitivity data, thereby further enhancing the stability of the control operation.
During the operation of the tool 200 in combination with the controller 250, updated sets of currents may be determined by the controller 250 in conformity with one or more of the above-identified control regimes, wherein, depending on the configuration of the tool 200, a plurality of respective sets of updated currents are supplied to the respective multi-anode configurations, thereby significantly enhancing production yield compared to conventional tools in which typically the same current settings are used for a plurality of substantially identical reactor vessels. According to the present invention, the set of updated currents may be determined individually for each reactor vessel 101 a, 101 b . . . , in a time interval that enables a simultaneous operation of each reactor vessel 101 a, 101 b . . . on the basis of individually determined updated anode currents. In particular embodiments, the control operation is based on the sensitivity functions, S_102A(r) . . . S_102N(r) and the thickness profile data, previously denoted as M^O(r), which may be established and obtained for each individual reactor vessel 101 a, 101 b . . . .
In other embodiments, a desired thickness profile may be selected, for instance on the basis of a pre-plating process or a post-plating process, and the controller 250 provides a correspondingly updated set of currents without time delay with respect to the overall process time in the individual reactor vessels 101 a, 101 b . . . , i.e., within a time interval that is negligible compared to a time interval for instance required for loading a substrate into each of the reactor vessels 101 a, 101 b . . . For instance if a post-plating CMP process indicates that the removal rate at the center of a substrate changes more rapidly than at a peripheral region of the substrate, a corresponding new desired thickness profile may be selected and the controller 250 may immediately provide corresponding updated anode currents to the plurality of reactor vessels 101 a, 101 b . . . .
FIG. 3 schematically represents measured thickness profile data M^O(r) for a plurality of substrates processed in the tool 200. In the present example, 200 mm substrates have been processed to deposit copper based on a control strategy as described with reference to Equations 5, 6 and 7, wherein the desired target thickness profile is represented by a dome-shaped profile so as to meet the requirements of a post-plating CMP process exhibiting an increased removal rate at the substrate center. In the control strategy used, the plating time was kept constant. Curve C in FIG. 3 represents a substrate processed in the reactor vessel 101 a, whereas curves B and A represent substrates that are processed in the reactor vessel 101 b within a time span of approximately two hours. The remaining non-indicated curves in FIG. 3 represent substrates that are processed in further reactor vessels of the tool 200 (not shown). As is indicated in FIG. 3, the controller 250 substantially maintains the desired thickness profile, wherein a deviation in this example is within approximately 200 Å at the center of the substrate where the maximum thickness is deposited. It should be noted that a systematic shift of the profiles A, B and C may be reduced, when the plating time is also updated, for instance according to a control operation as described with Equation 8.
As a result, the present invention provides a technique that enables efficient control of a multi-anode configuration in an electroplating tool in that the individual anode currents and/or an updated plating time are calculated on the basis of specific criteria substantially without any time delay with respect to a typical process time of the electroplating tool, thereby enabling a fast response to any process variation. By means of the control strategies explained above, the controller 250 may perform a control operation on the basis of measurement results of a preceding substrate to calculate updated anode currents for one or more substrates to be processed. Due to the potential for determining the anode currents in a substantially non-delayed manner, a plurality of multi-anode configurations may be controlled, thereby significantly enhancing yield as the quality of the plated substrates increases. Moreover, individual reactor vessels of one or more plating tools may be controlled simultaneously and the process conditions within each individual reactor vessel may be optimized in an automated fashion, so that the lifetime of consumables, such as wet ring contacts, anodes and the like, may be increased, thereby significantly reducing the down time of the process tool. Furthermore, the concept of the present invention may be readily implemented into conventional plating tools, thereby contributing to an increased efficiency and throughput of these tools without unduly causing additional expense. In other embodiments, the anode currents and plating times may be updated on a regular basis with or without referring to measurement data, wherein the updated anode currents may be established on the basis of various criteria, such as a constant total current supplied to each multi-anode configuration.
The particular embodiments disclosed above are illustrative only, as the invention may be modified and practiced in different but equivalent manners apparent to those skilled in the art having the benefit of the teachings herein. For example, the process steps set forth above may be performed in a different order. Furthermore, no limitations are intended to the details of construction or design herein shown, other than as described in the claims below. It is therefore evident that the particular embodiments disclosed above may be altered or modified and all such variations are considered within the scope and spirit of the invention. Accordingly, the protection sought herein is as set forth in the claims below.

Claims

1. A method, comprising:

determining sensitivity data quantitatively relating a set of currents for a multi-anode configuration of an electroplating tool to a thickness of a metal layer formed on a substrate by electroplating; and

determining an updated set of currents for said multi-anode configuration on the basis of said sensitivity data for a second substrate to be processed in said electroplating tool.

2. The method of claim 1, further comprising obtaining thickness profile data from at least one substrate processed in said electroplating tool and determining said updated set of currents on the basis of said thickness profile data.

3. The method of claim 1, further comprising determining a second updated set of currents for a second multi-anode configuration on the basis of said sensitivity data, wherein a sum of said updated set of currents is substantially equal to a sum of said second updated set of currents.

4. The method of claim 1, further comprising selecting a desired thickness profile and determining said updated set of currents on the basis of said desired thickness profile.

5. The method of claim 4, further comprising obtaining a set of reference current data from said at least one substrate and determining said updated set of currents on the basis of said reference current data.

6. The method of claim 1, further comprising determining an allowable range for each current of said updated set of currents.

7. The method of claim 6, wherein said updated set of currents is determined under the secondary condition that each current of said updated set is within its respective allowable range.

8. The method of claim 7, further comprising selecting a desired amount of metal to be deposited and determining a total current value and a process time required to actually deposit said desired amount of metal on a substrate.

9. The method of claim 8, wherein said updated set of currents is determined under the secondary condition that a sum of individual currents of said updated set is equal to said total current value.

10. The method of claim 9, wherein said updated set of currents is determined by calculating a minimum of a difference between said thickness profile data and said desired profile.

11. The method of claim 10, further comprising determining a position independent portion of metal thickness when calculating said minimum and using said position independent portion to determine an updated process time for said second substrate.

12. The method of claim 1, further comprising controlling the thickness profile for a plurality of second substrates on the basis of said updated set of currents.

13. The method of claim 1, further comprising obtaining thickness profile data from said second substrate after processing said second substrate with said updated set of currents and determining a new updated set of currents on the basis of the thickness profile data of said second substrate.

14. The method of claim 1, wherein said electroplating tool comprises at least one further multi-anode configuration and wherein an updated set of currents is determined for said at least one further multi-anode configuration.

15. The method of claim 4, wherein said desired thickness profile is selected on the basis of at least one process specific characteristic of a process to which said second substrate is subjected after completion of the electroplating process.

16. The method of claim 15, wherein said at least one process specific characteristic is a removal rate distribution across a substrate of a chemical mechanical polishing process.

17. The method of claim 16, further comprising obtaining removal rate distribution data from said second substrate after polishing said second substrate and selecting said desired thickness profile on the basis of said removal rate distribution data.

18. The method of claim 4, wherein said desired thickness profile is selected on the basis of at least one process specific characteristic of a process to which said second substrate is subjected before said electroplating process.

19. The method of claim 18, wherein said at least one process specific characteristic relates to sputter deposition of at least one of a barrier layer and a seed layer.

20. A method of depositing metal in an electroplating tool having at least one process chamber including a multi-anode configuration, the method comprising:

determining a set of currents for said multi-anode configuration on the basis of a desired thickness profile, thickness profile data obtained from at least one substrate processed in said electroplating tool and a model quantitatively describing a relation between current supplied to said multi-anode configuration and a thickness profile; and

depositing metal on one or more substrates while using said determined set of currents.

21. The method of claim 20, wherein said model is based on sensitivity data relating a thickness profile to a current variation during the deposition of metal in said electroplating tool.

22. The method of claim 20, wherein said electroplating tool comprises at least one further process chamber including a multi-anode configuration and a further set of currents is determined, on the basis of a desired thickness profile, thickness profile data obtained from at least one substrate processed in said electroplating tool and a model quantitatively describing a relation between current supplied to said multi-anode configuration and a thickness profile, prior to processing at least one substrate in said at least one further process chamber.

23. The method of claim 20, further comprising obtaining thickness profile data from said at least one substrate and using said obtained thickness profile data of said at least one substrate as the thickness profile data for determining said set of currents for a substrate to be processed in said electroplating tool.

24. The method of claim 20, further comprising determining an updated plating process time for depositing metal on said one or more substrates.

25. The method of claim 24, wherein said updated process time is determined on the basis of a previously used process time and a sensitivity factor expressing a change of a thickness of plated metal over plating time required to create said change.

26. A method of controlling an electroplating tool including a plurality of process chambers having a multi-anode configuration, the method comprising:

calculating a set of currents for each multi-anode configuration; and

simultaneously processing a substrate in each of the plurality of process chambers with said determined sets of currents.

27. The method of claim 26, wherein a sum of currents for each set is substantially equal to a predefined target value.

28. The method of claim 26, further comprising:

determining sensitivity data quantitatively relating a set of reference currents for said multi-anode configuration to a thickness of a metal layer formed on a substrate processed in at least one of said process chambers; and

determining said sets of currents for said multi-anode configurations on the basis of said sensitivity data for a plurality of second substrates to be processed in said process chambers.

29. The method of claim 28, further comprising obtaining thickness profile data from at least one substrate processed in said electroplating tool and determining said sets of currents on the basis of said thickness profile data.

30. The method of claim 26, further comprising selecting a desired thickness profile and determining said sets of currents on the basis of said desired thickness profile.

31. The method of claim 28, further comprising obtaining a set of reference current data from a plurality of substrates previously processed in said plurality of process chambers and determining said sets of currents on the basis of said reference current data.

32. The method of claim 31, further comprising determining an allowable range for each individual current in each of said sets of currents.

33. The method of claim 32, wherein said sets of currents are determined under the secondary condition that each current of said sets is within its respective allowable range.

34. The method of claim 33, further comprising selecting a desired amount of metal to be deposited and determining a total current value and a process time required to actually deposit said desired amount of metal on a substrate.

35. A controller for an electroplating tool, comprising:

a calculation unit configured to determine, for a substrate to be processed in said electroplating tool, a set of currents for at least one multi-anode configuration on the basis of a desired thickness profile.

36. The controller of claim 35, wherein said calculation unit is configured to determine a plurality of sets of currents for a plurality of multi-anode configurations prior to processing a substrate in each of the plurality of multi-anode configurations.