JP2007507615A - Method and system for automatically controlling the distribution of current to a multi-anode structure during metal plating on a substrate surface - Google Patents

Abstract

The electroplating tool operates in conjunction with a controller that automatically determines the individual currents of the multi-anode structure of the electroplating tool. The calculation of the anode current is based on the desired target profile in addition to the sensitivity data and the measurement data so that a quick response to process variations can also be realized for a plating tool including a plurality of process chambers.

Description

  The present invention relates to a process for depositing metal on the surface of a substrate using a reactor for electroplating, and more particularly, by adjusting the current supplied to the multi-anode structure of the plating tool. It relates to a process for obtaining the desired thickness profile of the metal.

  In many technical fields, depositing a metal layer on the surface of a substrate is a frequently used technique. Electroplating or electroless plating type plating for efficiently depositing a relatively thick metal layer on a substrate surface has proven to be a viable and cost effective method. Therefore, plating has become an attractive vapor deposition method in the semiconductor industry.

  Currently, copper is considered the preferred metal for forming metallization layers in advanced integrated circuits. This is because, for example, copper and copper alloys have superior characteristics in terms of conductivity and resistance to electromigration, compared to commonly used aluminum. Since copper is deposited very inefficiently by physical vapor deposition, eg sputter deposition, with a layer thickness on the order of 1 μm or more, electroplating of copper and copper alloys currently forms metal layers It is a preferable vapor deposition method. Although copper electroplating is a well known technique, it is a difficult task for process engineers to reliably deposit copper on a large diameter substrate having a patterned surface including trenches and vias. . For example, forming a metal layer in an ultra-large scale integrated device requires that a wide trench with a width on the order of a few microns be filled, and a via having a diameter of 0.2 μm or less and The trench is also required to be filled surely. The situation is further complicated as the diameter of the substrate tends to increase. At present, 8-inch or 10-inch wafers are generally used in semiconductor process lines. Therefore, great efforts are being made in the technical field of copper plating to provide a copper layer of the desired profile across the substrate surface. At first glance, the metal thickness profile across the substrate surface seems advantageous in that it is formed as non-uniformly as possible. However, in the post-plating process, variously shaped profiles are required to ensure proper device functionality of the finished integrated circuit. For example, excess copper is removed during the formation of the copper-based metal layer. Copper removal is currently often achieved by chemical mechanical polishing (CMP) of metal surfaces. Because CMP processes are themselves very complex processes that often exhibit inherent process non-uniformities, ie, non-uniform removal rates across the substrate surface, the metal thickness profile is applied to post-plating processes. Thus, it is desirable to achieve overall improved process uniformity after completion of the post-plating process. Thus, electroplating tools are often configured to allow deformation of the metal profile, but at present, controlling the final profile is cumbersome and time consuming.

  Referring to FIG. 1, a general prior art electroplating system is described, illustrating the problems associated with metal plating copper in more detail.

  FIG. 1 a shows a typical conventional electroplating system 100. This system includes a reactor vessel 101 having a first electrode 102. In this case, the first electrode 102 is an anode, and includes a plurality of anode parts 102A,..., 102N that can be individually driven, and a multi-anode arrangement is defined. Is done. In this example, a so-called fountain type reactor is considered. In this fountain reactor, the electrolyte solution is directed upward from the bottom of the reactor 101 and recirculated by a pipe 103 connecting the drain 104 to the storage tank 107. The storage tank is connected to the anode 102 via an inlet 105 provided as a passage. System 100 includes a substrate holder 108 that is configured to support a substrate 109, such as a semiconductor wafer, so that a surface of interest can be exposed to an electrolyte solution. Furthermore, the substrate holder 108 functions as a second electrode, in this case a cathode (cathode), and may be configured to provide electrical connection to the power supply 110. The power source 110 is configured to supply each current having a predetermined magnitude to each of the anode portions 102A,... 102N.

  FIG. 1b schematically shows an overhead view of the electrode 102 including the multi-anode structures 102A,...

  Prior to attaching the substrate 109 to the substrate holder 108, a thin current distribution layer, which can include a seed layer, is formed by sputter deposition on the surface of the substrate 109 that receives the metal layer. The substrate 109 is then attached to the substrate holder 108 and a small contact area (not shown for simplicity) provides electrical contact to the power supply 110 through the substrate holder 108. A pump (not shown) is activated, and an appropriate voltage is applied between the anode 102, that is, the multi-anode structure 102A,... 102N, and the substrate holder 108 that generates the respective currents. Electrolyte flow is generated in the electrolyte. The electrolyte solution entering the reactor 101 from the inlet 105 is directed toward the substrate 109, and the deposition of the metal on the substrate 109 is determined by the flow of the electrolyte solution and the arrangement of the multi-anode structures 102A,. The This is because the local metal deposition rate on a specific region on the surface of the substrate 109 is determined by the number of ions reaching this region. Thus, by selecting the current set supplied to the multi-anode structure 102A,... 102N, the final resulting thickness profile is determined, to influence the flow of ions and / or electrolyte solution. Any additional means may be inserted, for example in the form of diffusion plating.

  When the appropriate current set is adjusted by the power supply 110, the resulting thickness profile is determined by the characteristics of the reactor 101, the electrolyte solution, the current set, and the plating time. Thus, variations in one of these characteristics can cause drift in the resulting thickness profile. The situation is further complicated in the electroplating tool 100 including a plurality of reactors 101 corresponding to a plurality of multi-anode structures 102A,... 102N. This is because all subtle process variations in these reactors can cause very complex interactions of the associated process characteristics or result in very complex interactions that result in process stability. This is because the properties are impaired. Therefore, since a plurality of test substrate runs are generally performed periodically, time and labor are required, resulting in a decrease in yield and quality of the plating process.

  Further, when the metal layer is formed by a technique called damascene technique, the via and the trench are filled with metal. Some extra metal must be provided to ensure that the vias and trenches can be filled. Subsequently, excess metal must be removed to ensure electrical isolation between adjacent trenches and vias and to provide a flat surface for further metal layer formation. A preferred technique for removing excess metal and planarizing the substrate surface includes chemical mechanical polishing (CMP). The substrate material that is removed in the CMP technique is exposed to a chemical reaction and simultaneously mechanically removed. As already explained, the CMP process is very complex and can exhibit non-uniformity. This non-uniformity can be at least partially compensated by adapting the electroplating process thickness profile to the CMP non-uniformity. However, the process parameters associated with the generation of the thickness profile, particularly the plating tool with multiple reactors, can deviate significantly from the desired thickness profile, thus effectively compensating for CMP non-uniformities. Not.

  In view of the above-mentioned problems, there is a need in the plating tool for a technique to quickly and efficiently adjust the thickness profile that can eliminate or at least reduce some or all of these problems. Yes.

  In general, the present invention relates to techniques for controlling individual currents supplied to each anode of a multi-anode structure in an electroplating tool. The individual currents for the multi-anode structure, also called current set, are automatically calculated taking into account the desired thickness profile. As a result, even if the plating tool under consideration includes a plurality of process chambers with a plurality of multi-anode structures, process variations of the plating process itself and / or post-plating and / or pre-plating processes ( It is possible to respond to the pre-plating process in a manner that does not substantially delay.

  According to an exemplary embodiment of the present invention, a method includes sensitivity data for quantitatively relating a current set of a multi-anode structure of an electroplating tool to a thickness of a metal layer formed on a substrate by electroplating. data) is included. The method further includes determining an updated current set of the multi-anode structure based on the sensitivity data of the second substrate processed in the electroplating tool.

  In a further embodiment, the method further includes determining an acceptable range for each current in the updated current set.

  In a further embodiment, the updated current set is determined under a second condition that each current in the updated set is within a respective tolerance range.

  In a further embodiment, the method selects the desired amount of metal to be deposited and determines the total current value and processing time required to actually deposit the desired amount of metal on the substrate. A step.

  In a further embodiment, the updated current set is determined under a second condition that the total amount of each current in the updated set is equal to the total current value.

  In a further embodiment, the updated current set is determined by calculating the minimum thickness profile data and the desired profile difference.

  In a further embodiment, the method determines a position independent portion of the metal thickness and calculates an updated processing time for the second substrate when calculating the minimum value. And using the position-independent portion.

  In a further embodiment, the method further includes controlling a thickness profile of the plurality of second substrates based on the updated current set.

  In a further embodiment, the method obtains a thickness profile from the second substrate after processing the second substrate with the updated current set, and based on the thickness profile data of the second substrate. Determining an updated current set.

  In a further embodiment, the electroplating tool includes at least one additional multi-anode structure, and an updated current set is determined for the at least one additional multi-anode structure.

  In a further embodiment, the desired thickness profile is selected based on the specific characteristics of at least one process of the process to which the second substrate is exposed after completion of the electroplating process.

  In a further embodiment, a specific feature of the at least one process is a removal rate distribution across the substrate by a chemical mechanical polishing process.

  In a further embodiment, the method obtains removal rate distribution data from the second substrate after polishing the second substrate, and selects a desired thickness profile based on the removal rate distribution data; Is further included.

  In a further embodiment, the desired thickness profile is selected based on the specific characteristics of at least one process of the process to which the second substrate is exposed prior to the electroplating process.

  In a further embodiment, the specific feature of the at least one process relates to sputter deposition of at least one of a barrier layer and a seed layer.

  According to a further exemplary embodiment of the present invention, a method is provided for depositing metal with an electroplating tool having at least one process chamber comprising a multi-anode structure. This method quantifies the relationship between the desired thickness profile, thickness profile data obtained from at least one substrate processed in an electroplating tool, and the current and thickness profile supplied to the multi-anode structure. And determining a current set of the multi-anode structure based on the model described in (1). Finally, the metal is deposited on one or more metals while using the determined current set.

  In a further embodiment, the method obtains a thickness profile from at least one substrate and uses the obtained thickness profile of the at least one substrate to determine a current set of substrates to be processed in the electroplating tool. The method further includes the step of using the data as profile data.

  In a further embodiment, the method further comprises determining an updated plating process time for depositing metal on the one or more substrates.

  In a further embodiment, the updated processing time is a change in the thickness of the plated metal over the processing time previously used and the plating time required to produce a change in the thickness of the plated metal. It is determined based on a sensitivity factor representing

  According to yet another exemplary embodiment of the present invention, a method is provided for controlling an electroplating tool that includes a plurality of process chambers having a multi-anode structure. The method includes calculating a current set for each multi-anode structure and processing at least one substrate in each of the plurality of process chambers using the determined current set.

  In a further embodiment, the total amount of current in each set is substantially equal to a predetermined target value.

  In a further embodiment, the method determines sensitivity data that quantitatively relates the reference current set of the multi-anode structure to the thickness of a metal layer formed on a substrate processed in at least one process chamber; and Determining a current set of the multi-anode structure based on sensitivity data of a plurality of second substrates processed in the process chamber.

  In a further embodiment, the method further includes obtaining thickness profile data from at least one substrate processed in the electroplating tool and determining a current set based on the thickness profile data. .

  In a further embodiment, the method further includes selecting a desired thickness profile and determining a current set based on the desired thickness profile.

  In a further embodiment, the method further includes obtaining a reference current data set from a plurality of substrates already processed in the plurality of process chambers and determining a current set based on the reference current data.

  In a further embodiment, the method further comprises determining a respective current tolerance for each current set.

  In a further embodiment, the current set is determined under a second condition that each current in the set is within an acceptable range.

  In a further embodiment, the method comprises selecting a desired amount of metal to be deposited and determining a total amount of current and processing time actually required to deposit the desired amount of metal on the substrate. In addition.

  According to yet another exemplary embodiment of the present invention, the controller of the electroplating tool is configured to determine at least one multi-anode structure current set for a substrate processed by the electroplating tool. Includes units. The calculation by this calculation unit is based on the desired thickness profile.

  In a further embodiment, the computing unit is configured to determine a plurality of current sets of the plurality of multi-anode structures before processing the substrate in each of the plurality of multi-anode structures.

  Further advantages, objects and embodiments of the present invention are defined in the appended claims and will become more apparent with the following detailed description when taken in conjunction with the accompanying drawings.

  While the invention will be described with reference to the embodiments presented in the following detailed description in conjunction with the drawings, the following detailed description will be limited to the specific embodiments disclosed in the same manner as the drawings. Rather, the exemplary embodiments described are merely illustrative of various aspects of the invention, and the spirit of the invention is limited only by the appended claims. Will be understood.

  Since the present invention is particularly useful in processing sequences with sensitive post-plating processes such as CMP, the detailed description will be directed to a metal such as copper on a substrate as commonly used in semiconductor manufacturing. Reference is made to electroplating. However, it is readily apparent that the present invention can be applied to any plating process (electroplating) using an externally applied current on any type of substrate that requires a specific deposition profile on the substrate surface or part thereof. . In addition, the present specification refers to a fountain-type plating reactor, for example as schematically illustrated in FIG. 1a, but is similarly used for other types of reactors such as electrolyte baths and the like. May be. Accordingly, the present invention is to be understood as not limited to a particular type of electroplating reactor, unless limitations on the electroplating reactor are set forth in the claims.

  One distinguishing feature provided by the present invention is that it can respond in a short time to changes in processing conditions within the electroplating tool or in any process before or after the plating process. Such changes in processing conditions include, for example, variations in the properties of the plating solution due to subtle variations in the properties of the sensitive additive contained in the plating solution, or degradation of any consumables in the plating tool or post-plating CMP tool. , Etc. The change in processing conditions may be a well-known change in wear of the CMP tool, for example, or may be detected by any suitable sensor element. Other changes in processing conditions are themselves "invisible" and cannot be compensated in an effective manner using conventional plating tools. This is because, after recognition of process drift, a time-consuming readjustment with a subsequent test period is generally required, including subsequent test periods. In view of the problems described above with respect to FIGS. 1a and 1b, the present invention provides a predetermined control scheme. In this control scheme, one or more current sets can be recalculated on a predetermined time scale to perform operation of one or more multi-anode structures. Since this time scale is negligible compared to the time required to process and process the substrate in the electroplating tool, it can respond to process variations with virtually no delay. In certain embodiments, corresponding current sets of various multi-anode structures can be recalculated based on the occurrence of any readily detectable process variation. For example, the calculation is performed under the following conditions, that is, the total amount of current supplied to each of the multiple multi-anode structures (total current amount) is substantially the same, and as a result, the deposited metal Ensures that the total amount is substantially the same for the same processing time. On the other hand, the automatic recalculation of the anode current allows an effective and, if desired, substantially continuous response to any invisible change that can only be identified by deviations from the desired target profile. It becomes possible.

  Thus, in other embodiments, the electroplating control scheme is based on the concept of maintaining a deviation from the desired thickness profile within a predetermined tolerance. The desired thickness profile of the metal layer deposited on the substrate surface is represented by the function T (r), where T represents the thickness value at position r on the substrate surface. The variable r may represent a predetermined position on a flat or non-flat substrate surface, but in the following description, this r is the distance from the center of a substantially disk-shaped substrate, such as a wafer in the semiconductor industry. Represents. Thus, although it is assumed that the desired thickness profile T (r) is axisymmetric, it should be understood that in the principles of the present invention, an arbitrary function T (r) also applies. Similarly, the actual thickness profile of the metal layer is denoted M (r), where M represents the thickness of the metal at location r. Thus, according to some embodiments of the present invention, the actual thickness profile from the desired thickness profile T (r), regardless of variations in the electroplating process or any pre-plating and post-plating processes. It is considered appropriate to maintain the deviation of M (r) within a predetermined tolerance.

As already described in connection with FIGS. 1a and 1b, the actual thickness profile M (r) is affected by the current supplied to the particular anode portion 102A,..., 102N. On the premise that the influence is negligible due to slight processing fluctuations of the electroplating tool, the degree of the influence, that is, the sensitivity of each of the anode portions 102A,. The influence of the current supplied to each anode section 102A,..., 102N is hereinafter referred to as “sensitivity”, and the corresponding data may be referred to as “sensitivity data”. By determining the sensitivity data of a particular electroplating tool, such as the tool 100 described in connection with FIGS. 1a and 1b, the corresponding anode current set has been processed with the desired thickness profile T (r). It can be calculated based on the actual thickness profile M (r) that can be obtained from the substrate in the form of individual measurement data. Generally, a thickness measurement after plating of at least a predetermined representative portion of the substrate is performed. This can be used in accordance with the present invention to determine an updated current set of one or more multi-anode structures, such as structure 102. Individual measurement data of the actual thickness profile can be easily converted into a substantially continuous function M (r) by interpolation, data fitting processes, and any data manipulation techniques known in the prior art. The same applies to sensitivity data. Sensitivity data is provided as individual measurements for a plurality of different current values by a plurality of positions and individual anode portions 102A,. For example, corresponding sensitivity data can be obtained by varying the current of a particular anode section and by measuring the actual metal thickness obtained after a certain deposition time. Thereafter, the deposition process is resumed or restarted using another anode portion 102A,... 102N. From each of the multiple measurements, a continuous function for the sensitivity of the individual anode section, called S _i (r), is given by one of the data manipulation methods described above. The index i represents one of the anode portions 102A,... 102N. Thus, an updated current set based on the current set used to generate the actual thickness profile M (r) is a clear tolerance for the expected thickness profile of the substrate being processed by the plating tool 100. It can be calculated to have a deviation from the desired thickness profile T (r) within the range.

To this end, the difference between the acquired actual thickness profile M (r) and the desired thickness profile T (r) is calculated as follows for the desired ideal metal layer by the following equation: Conveniently indicated by a position-dependent term and a position-independent term representing a constant offset of the layer.
M (r) -T (r) = E (r) + M ^Offset (1)
E (r) represents position dependent deviation or excess material, and M ^Offset represents position independent deviation. Therefore, the sum of the position-dependent portions E (r) over the entire substrate surface A is zero.

基板に蒸着された金属量はアノード部に供給された総電流量及びメッキ時間に実質的に依存するので、位置非依存項Ｍ^Offsetはゼロに近づいてよい。その結果、個々のアノード部１０２A、...、１０２Nへの電流を変化させることによって、位置依存項Ｅ（ｒ）に本質的な影響を及ぼす。一方で、メッキ時間のわずかな変動は、上記の仮定によれば、位置非依存項Ｃ^Offsetに主に影響を及ぼし、その結果、この値をゼロに近づくようにわずかに変動させる。その結果、一実施形態では、個々のアノード部１０２Ａ、...、１０２Ｎに供給されるアップデートされた電流セットは、等式（１）における、Ｅ（ｒ）の合計を減らす或いは最小にする、という概念に基づいてよい。 Assuming a substantially constant plating time and a substantially constant total current amount I ^SUM representing the total amount of each current supplied to the anode portions 102A,.

Since the amount of metal deposited on the substrate substantially depends on the total amount of current supplied to the anode and the plating time, the position independent term M ^Offset may approach zero. As a result, the position-dependent term E (r) is essentially affected by changing the current to the individual anode portions 102A,..., 102N. On the other hand, slight fluctuations in the plating time mainly affect the position-independent term C ^Offset according to the above assumption, and as a result, this value is slightly changed to approach zero. As a result, in one embodiment, the updated current set supplied to the individual anode portions 102A,..., 102N reduces or minimizes the sum of E (r) in equation (1). It may be based on the concept of

Ａ^Wは金属メッキされる基板の領域全体を表す。 For this purpose, the updated current set is denoted as I ^updated = (I _102A ^updated ,... I _102N ^updated ), and the current used to generate the actual thickness profile of the already processed substrate. The set is represented as I ^O = (I _102A ^O ,... I _102N ^O ). Similarly, the corresponding function representing the thickness profile of an already processed substrate is denoted M ^O (r). As already explained, the function M ^O (r) can be obtained by a corresponding set of possible measurements at a plurality of different positions r according to standard thickness measurement procedures. The continuous function or quasi-continuous function M ^O (r) may be obtained by interpolation, data fitting, and any known data manipulation process. In obtaining the thickness profile M ^O (r), assuming a substantially constant plating time, the position-dependent term M ^Offset (see equation (1)) expected to approach zero is , Equation (1) may be used to calculate the sum over the entire substrate area from the difference between M ^O (r) and T (r), ie, the integral value. The reason is that, as already explained with respect to equation (1), the sum of the position-dependent parts E (r) is zero (see equation (1 ′)). Accordingly, a relatively small position-independent portion M ^Offset can be obtained by equation (3) in the following manner.

A ^W represents the entire area of the substrate to be metal plated.

Since M ^Offset is determined by Equation 3 for a class of allowable function representing the position-dependent deviation E (r), the corresponding thickness profile for the subsequent substrate denoted as M (r) is , May be calculated based on equation (1).

インデックスｉは、各アノード部１０２Ａ、...１０２Ｎを表す。その他の各関係は、これらの各関係が各アノード電流の変化による影響を所定の厚さプロファイルに表して、その結果新たな厚さプロファイルを生成する限りは用いることができる。等式（４）の厚さプロファイルＭ^O（ｒ）は、好ましくは、既に処理された基板から取得された測定データに基づく。算出された厚さプロファイルＭ（ｒ）に基づいて現在処理される基板の遅延は、どのような処理の変動に対しても短い応答時間で応答できるように、適度に小さい。 Control variable of the plating process, the sensitivity function _{S 102A (r), ...,} S 102N (r) the current set associated with the thickness profile M (r) by I _102A, ..., since it is I _102N , The already processed substrate M ^O (r) and the sensitivity function S _{102A (r} so that an updated thickness profile M (r) can be obtained as a function of the individual anode currents I _{102A ,. )} ,..., This plating process may be modeled by establishing a correlation between _{S102N (r)} . In one embodiment, a linear relationship can be used for the plating model, for example in the form as given by equation (4) below.

The index i represents each anode part 102A,... 102N. Each other relationship can be used as long as each of these relationships represents the effect of each anode current change in a predetermined thickness profile, resulting in a new thickness profile. The thickness profile M ^O (r) in equation (4) is preferably based on measurement data obtained from an already processed substrate. The delay of the currently processed substrate based on the calculated thickness profile M (r) is reasonably small so that it can respond to any process variation with a short response time.

However, in other embodiments, it may be appropriate to select the function M ^O (r), such as based on averaged measurement data and / or data that is not based on a predetermined test. For example, in the first stage of the plating process where measurement data is not available, appropriate reference data may be used for the function M ^O (r) or the desired target profile T (r).

In one particular embodiment, the stability of the control process is increased by appropriately selecting the acceptable range of each anode current I _102A ,..., I _102N . That is, an upper limit for each of the anode currents I _102A ,..., I _102N so that the resulting thickness profile M (r) is obtained by a current set in which the respective anode current is within its tolerance. And the lower limit can be selected. For example, a target value for each anode current may be determined based on experience gathered from previous plating processes and an acceptable range for each anode current may be set. In other cases, the influence of each anode current can be conveniently investigated during the operation phase of the plating tool, so that the respective target values and associated tolerances are based on measured data acquired while determining the sensitivity function. May be determined. Target values and corresponding tolerances for each anode portion 102A,..., 102N may also be determined based on tool specifications and tool and / or processing requirements. Generally, a variation of about 10-20% relative to the target value for each anode current results in sufficient control stability.

In the following description, the upper and lower limits of each anode current I _i , I = 102A,..., 102N are represented by I _i ^L and I _i ^H. Thus, an updated current set corresponding to a predetermined allowable position dependent deviation E (r) can be calculated based on equations (1), (3), (4).

In one particular embodiment, the updated current set can be obtained by determining the required thickness profile M (r) with the smallest deviation E (r). That is, an ^updated current set denoted as I ^updated = I _102A ^updated ,..., I _102N ^updated is obtained by solving equation (5) below.

In one embodiment, a second condition is used for each anode current such that they are within an acceptable operating range, eg, as defined above, and / or each anode current. Is set to be substantially equal to a predetermined value. Accordingly, these second conditions are represented by the following equations (6) and (7).

  In equation (5), the integration may be performed as a one-dimensional integration from the center of the substrate to the end of the substrate denoted D, or as a two-dimensional integration over the entire surface of the substrate. In one embodiment, shown in equation (5), the substrate center contribution is highlighted in a one-dimensional representation, favoring critical thickness profiles near the substrate center. Therefore, one-dimensional integration is used. For example, post-plating CMP processing may exhibit significant removal rate fluctuations near the center of the substrate so that a correspondingly sensitive plating profile is advantageous.

  The updated anode current can be calculated by inserting equation (4) into equation (5) and using various integration methods. For example, a personal computer with the appropriate instruction set installed, for example in the form of a Matlab® plug-in, may be used to obtain updated anode current values. Depending on the hardware requirements of the corresponding control unit and the desired accuracy of the calculations performed, any suitable implementation may be used, for example a corresponding programmed microcomputer, and analog and / or digital design Any suitably arranged circuit may be used including: As described with respect to FIG. 2, the calculated and updated current set can be sent to the corresponding plating tool so that it can be sent to the plating tool within a time span sufficient to perform the desired control action. An operably connected remote device may be used.

γはメッキ時間の変化に応じたメッキされた金属の厚さの感度因子を表す。γのそれぞれの数値は、１又は複数の明確なメッキ時間間隔内で厚さの増加を測定することによって簡単に取得することができる。従って、厚さプロファイルがターゲットプロファイルＴ（ｒ）に対して実質的に全体にシフトされた厚さプロファイルのあらゆる偏差は、アップデートされたメッキ時間Ｔ^updatedをそれに応じて再計算することによって実効的に補償することができる。既に説明したように、メッキ時間のわずかな変化は、位置依存偏差Ｅ（ｒ）に実質的な影響を与えず、位置依存項Ｍ^Offsetによるメッキ時間の変動もまた小さく、対応するアップデートバージョンのメッキ時間が、この制御方式の安定性若しくは上述の方法で決定される場合に、アップデートされたアノード電流の安定性に、過度に影響を及ぼさないように、依然として非常に小さいままにされる。 As already explained, it may be convenient to assume that the plating time on the substrate processed by the electroplating tool 100 is substantially constant. In other embodiments, a constant variation in plating time may be taken into account by recalculating the plating time T ^{updated updated} using the position-independent deviation M ^Offset already determined. For example, the updated plating time T ^updated can be calculated from the plating time T ^{O of the} already processed substrate using a pre-established relationship. In one example, the ^updated plating time T ^updated may be linearly related to the position independent deviation M ^Offset and the previous plating time T ^O as shown in equation (8) below.

γ represents the sensitivity factor of the thickness of the plated metal according to the change of the plating time. Each value of γ can be easily obtained by measuring the increase in thickness within one or more distinct plating time intervals. Thus, any deviation of the thickness profile in which the thickness profile is substantially shifted relative to the target profile T (r) is effectively achieved by recalculating the ^updated plating time T ^updated accordingly. Can be compensated. As already explained, slight changes in the plating time do not substantially affect the position-dependent deviation E (r), and the variation in the plating time due to the position-dependent term M ^Offset is also small. If time is determined in this way, or the stability of the updated anode current when determined in the manner described above, it is still kept very small so as not to unduly affect the stability of the updated anode current.

  Referring to FIGS. 1a and 2, the implementation of one or more of the above identified control schemes in a plating tool is described in further detail below.

FIG. 2 schematically illustrates a plating tool 200. The plating tool 200 can include one or more reactors that include a multi-anode structure, as described above with respect to FIGS. 1a and 1b. Accordingly, the corresponding reactor is shown as 101 and its description is made in FIG. The reactors 101a, 101b,... Are connected to respective controllable power supplies 201a, 201b,..., Which supply current sets in the reactors 101, 101b,. It is configured to supply a corresponding multi-anode structure. For convenience, each current set supplied to the corresponding reactor 101a, 101b,... Is denoted as I _102A ,..., I _102N , and the number of each anode portion of the reactor is the design of the plating tool 200 It is decided by. As described with reference to FIGS. 1a and 1b, the four anode portions 102A,..., 102N are provided with axial symmetry, and in other plating tools, the number of anode portions is 2 or less or 4 or more. Good. The plating tool 200 may be considered as a system of a plurality of different plating tools, each having a differently designed reactor, and the number of anode portions may be different for some of the plurality of plating tools. Similarly, reactors 101a and 101b may differ from each other in their reactor design, number of anode sections, and the like. The multi-anode structure of each reactor 101a, 101b,... Does not necessarily have to be axially symmetric, but can have any geometry that is considered appropriate. For example, in some cases, a thickness profile that is not axisymmetric is desirable, so a corresponding multi-anode structure may be provided. For example, as shown in FIGS. 1a and 1b, the provisions of the multi-anode structures 102A,..., 102N may be provided in circular sections, each cross-section being mutually It is separate and represents a part of the multi-anode structure. In such a case, the control scheme already described can be used in the same way, and the corresponding position-dependent functions and terms are represented in two-dimensional coordinates rather than one-dimensional radical components. There is a need.

The plating tool 200 is operably connected to the controller 250. This operational connection is indicated by 251 and is meant to represent at least all connections that allow data transfer from the controller 250 to the plating tool 200. In certain embodiments, connection 251 represents a data communication line in a wired or wireless format and sends appropriate control signals to each power source 210a, 210b,. Thus, the power supplies 210a, 210b,... Output the current sets I _{102A ,.} The controller 250 can be implemented in a workstation, PC, or facility management system. They are provided with a corresponding arithmetic unit so that one or more of the above-described embodiments can be implemented to establish an updated current set. The controller 250 may also include the interface and communication units necessary to provide information regarding the updated current set to the power supplies 210a, 210b,... Via the operational connection 251. In other embodiments, the controller 250 may be configured to control conventional operations, typically in the form of a suitably programmed microprocessor, ASIC (application specific integrated circuit), or the like. It may be provided by being mounted or added to a control unit (not shown) included in the plating tool. The computing power of the controller 250 determines, for example, the accuracy and speed of numerical calculations when solving equation (5), but the controller 250 generally does Compared to time intervals, the results are supplied within negligible time intervals. The controller 250 is further configured to receive data from an environment based on the updated anode current calculation. In one particular embodiment, the controller 250 is configured to receive sensitivity data from an external source such as an operator, computer, measurement device, or the like. As a result, the sensitivity data may be provided in the form of individual measurements, individual theoretical values, mathematical functions. Alternatively, the sensitivity data may be other suitable information formats that relate several anode sections 102A,... 102N to their effect on operational thickness profiles with respective anode currents I _{102A ,.} May be offered at. The controller 250 may be configured to record and convert the sensitivity data in any suitable format, depending on the format of the sensitivity data, and the sensitivity data is made available and updated in the computing unit of the controller 250. An anode current is established.

  In other embodiments, the controller 250 may be configured to receive thickness profile data in the form of individual measurements, a substantially continuous function, etc., the format of the thickness profile data described above. Can be converted into all the appropriate representations necessary to perform the calculated calculations. Similarly, the controller 250 may be configured to receive externally supplied profile data representing a desired thickness profile and / or the controller 250 may conveniently display one or more desired thickness profiles. It may be included and used as required by the control action. In one particular embodiment, the controller 250 receives the measurement data directly from an already processed substrate and, as a result, provides a closed loop control function (see FIG. (Not shown) operatively connected. The closed loop response time is substantially determined by the time delay to provide thickness measurement data to the controller 250. As pointed out above, since the updated anode current is automatically calculated, the time to establish the anode current is processed by the plating tool 200 or any measurement system required for the control operation. Compared to any time you do, it is negligible. As pointed out above, since the updated anode current is automatically calculated, the time to establish the anode current is processed by the plating tool 200 or any measurement system required for the control operation. Compared to any time, this time is negligible.

  In other embodiments, the controller 250 may be configured to receive additional information. Further information includes post-plating process data representing subsequent process characteristics, such as CMP processes commonly used in the manufacture of copper-based metal layers, or deposition of current distribution layers, seed layers, etc. Such processing data before plating relating to information about processing before plating processing. Further, the status information of the plating tool 200 may be supplied to the controller 250 and used when establishing a new anode current. In one embodiment, the sensitivity data may be associated with status information of the plating tool 200 to reduce sensitivity data drift relative to the tool status. For example, it is well known that sensitivity data can depend on the overall processing time, for example because certain properties change, such as the properties of the plating solution change over time. Therefore, by adapting the sensitivity data according to the change, it is possible to easily incorporate a known dependency corresponding to the above control method, and as a result, it is possible to further improve the stability of the control operation.

While the tool 200 operates in cooperation with the controller 250, the updated current set can be determined by the controller 250 according to one or more of the above control schemes. Depending on the structure of the tool 200, a plurality of each updated current set is supplied to each multi-anode structure. As a result, product yields can be greatly increased when compared to conventional tools that typically use the same current set for multiple substantially identical reactors. According to the present invention, the updated current set is determined individually for each reactor 101a, 101b,... At a predetermined time interval. This predetermined time interval is a time interval in which the respective reactors 101a, 101b,... Can be operated simultaneously based on the individually determined updated anode current. In certain embodiments, the control action is based on sensitivity functions S _{102A (r)} ,..., S _{102N (r)} and individual thickness profile data. This thickness profile data, previously denoted M ^O (r), can be established and acquired for each reactor 101a, 101b,.

  In other embodiments, the desired thickness profile may be selected based on, for example, a pre-plating process or a post-plating process. Further, the controller 250 is not delayed with respect to the entire processing time of each of the reactors 101a, 101b,... In a time that is negligible compared to the time interval, the corresponding updated current set is provided. For example, if the post-plating CMP process indicates that the removal rate at the center of the substrate is faster than the peripheral region of the substrate, a corresponding new desired thickness profile may be selected and the controller 250 may The reactors 101a, 101b,... Can be immediately supplied with a corresponding updated anode current.

FIG. 3 represents measured thickness profile data M ^O (r) for a plurality of substrates processed in the tool 200. In this embodiment, copper is deposited based on the control scheme described with respect to equations (5), (6), and (7), and a 200 mm substrate has been processed. The desired target thickness profile is represented by a dome-shaped profile so that post-plating CMP processing requirements exhibiting an increased removal rate at the substrate center can be met. In the control method used, the plating time was kept constant. Curve C in FIG. 3 represents a substrate processed in reactor 101a, and curve B and curve A represent a substrate processed in reactor 101b for a time span of about 2 hours. The remaining unsigned curves in FIG. 3 represent a substrate that has been processed in a further reactor (not shown) of the tool 200. As shown in FIG. 3, the controller 250 substantially maintains the desired thickness profile, and the deviation in this example is within about 200 mm at the center of the substrate deposited at the maximum thickness. The systematic shift of each profile A, B, and C is reduced, for example, when the plating time is also updated by the control operation described in equation (8).

  As a result, the present invention calculates individual anode currents and / or updated plating times based on specific criteria without substantially delaying the typical processing time of the electroplating tool. As a result, a technique is provided that enables a quick response to any process variation and enables effective control of the multi-anode structure of the electroplating tool. According to the control scheme described above, the controller 250 performs a control operation based on the measurement result of the preceding substrate, and calculates an updated anode current of one or more substrates to be processed. Since the anode current can be determined in a substantially non-delayed manner, the multiple multi-anode structures are controlled, resulting in a higher quality of the plated substrate, which can greatly increase yield. In addition, each reactor of the one or more plating tools can extend the life of consumables, such as wet ring contacts, anodes, etc., thereby reducing processing tool downtime. The process conditions in each reactor may be optimized automatically, and may be controlled simultaneously. In addition, the concepts of the present invention can be easily implemented in conventional plating tools, thereby increasing the efficiency and throughput of these tools without incurring additional expenditure than necessary. In other embodiments, the anode current and plating time may be updated periodically with or without reference to measurement data. This updated anode current may be established based on various criteria, such as a constant total amount of current supplied to each multi-anode structure.

  Further modifications and variations of the present invention will be apparent to those skilled in the art in view of this description. Accordingly, this description is to be construed as illustrative and is for the purpose of teaching those skilled in the art the general manner of carrying out the invention. The form of the invention shown and described herein is to be understood as the presently preferred embodiment.

  The present invention relates to processes used to manufacture microelectronic devices. Therefore, industrial applicability is clear.

1 is a schematic view of a conventional electroplating tool having a multi-anode structure. 1b is a schematic top view of the multi-anode structure of the tool of FIG. 1 is a schematic diagram of an electroplating system including a controller for automatically determining a current set based on various criteria according to an exemplary embodiment of the present invention. FIG. FIG. 6 is an exemplary diagram of a graph of measurement results of thickness profiles of a plurality of substrates processed in different process chambers according to a control scheme of an exemplary embodiment of the present invention.

Claims (9)

Determining sensitivity data that quantitatively relates the current set for the multi-anode structure of the electroplating tool to the thickness of the metal layer formed on the substrate by electroplating;
Determining an updated current set of the multi-anode structure based on the sensitivity data of a second substrate processed with the electroplating tool.
Method. Obtaining a thickness profile from at least one substrate processed with the electroplating tool; and determining the updated current set based on the thickness profile data.
The method of claim 1. Based on the sensitivity data, determining a second updated current set for a second multi-anode structure such that the total amount of updated current is equal to the total amount of second updated current. Further including
The method of claim 1. Selecting a desired thickness profile;
Determining the updated current set based on the desired thickness profile; and
The method of claim 1. Obtaining a reference current data set from the at least one substrate;
Determining the updated current set based on the reference current data set;
The method of claim 4. A method of depositing metal with an electroplating tool having at least one process chamber comprising a multi-anode structure, comprising:
Quantitatively describe the relationship between the desired thickness profile, the thickness profile obtained from at least one substrate processed by the electroplating tool, and the current and thickness profile supplied to the multi-anode structure Determining a current set for the multi-anode structure based on a model;
Depositing metal on one or more substrates while using the determined current set;
Method. The model is based on sensitivity data relating current variations to thickness profiles during the deposition of the metal with the electroplating tool.
The method of claim 7. The electroplating tool includes at least one further process chamber having a multi-anode structure;
Further current sets include a desired thickness profile, a thickness profile obtained from at least one substrate processed by the electroplating tool, and a relationship between the current supplied to the multi-anode structure and the thickness profile. Determined prior to processing at least one substrate in at least one further said process chamber based on a model that quantitatively represents
The method of claim 7. A method for controlling an electroplating tool comprising a plurality of process chambers having a multi-anode structure, comprising:
Calculating a current set for each multi-anode structure;
Simultaneously processing each substrate of the plurality of process chambers with the determined current set;
Method.

Images