Methods of forming sputtering targets, and sputtering targets formed thereby

[0002] The invention pertains to methods of forming sputtering targets, and further encompasses the targets formed by the methods.

[0003] A sputtering method is described with reference to FIG. 1, which illustrates a sputtering target 10 spaced from a substrate 12 by a distance T/S. Distance TIS is referred to as the target substrate distance. Substrate 12 can comprise, for example, a semiconductive material wafer. Target 10 can comprise numerous materials known to persons of ordinary skill in the art, such as, for example, metallic materials (e.g. one or more of aluminum, copper, titanium, tantalum, tungsten, cobalt, nickel, etc.), or ceramic materials (e.g., BaTiO₃, Pb(Zr, Ti)O₃, BiSrTaO₃, etc.). Also, target 10 can comprise numerous shapes. For instance, FIG. 2 illustrates that target 10 can comprise a circular shape

[0004] Referring again to FIG. 1, a shield 14 is provided over a peripheral region of target 10. Shield 14 can comprise, for example, stainless steel or aluminum.

[0005] In operation, material from target 10 is sputter-deposited onto substrate 12. More specifically, target 10 has a face surface 16 which is exposed to high energy ions and/or atoms. The high energy ions and/or atoms eject atoms from surface 16, and the ejected atoms are subsequently deposited onto substrate 12. Shield 14 protects peripheral edges of target 16 from being exposed to the high energy ions and/or atoms. One of the goals in target fabrication is to deposit a uniform film of material over substrate 12. One aspect of achieving a uniform film is to have an appropriate T/S distance between target surface 16 and substrate 12, as well as to maintain a substantially common T/S distance from the entirety of the sputtered target face 16 and substrate 12. Shield 14 is provided to alleviate problems which could occur if the sloped regions of target 10 were exposed to high energy ions and/or atoms during a sputtering process.

[0006] FIG. 3 illustrates target 10 after the target has been subjected to the wear of having material sputtered therefrom. Specifically, FIG. 3 illustrates a wear profile formed across sputtered face surface 16. The illustrated wear profile is for exemplary purpose only. The shape of an actual wear profile can depend on, for example, the magnet type and target life of materials used in a sputtering process. A dashed line 18 is provided in FIG. 3 to illustrate the starting position of the face surface when target 10 was new (i.e., the face surface shown in FIG. 1). As shown in FIG. 3, a number of troughs (i.e., sputter tracks) are formed within face surface 16 during the sputtering operation. Accordingly, the target does not wear uniformly across the surface 16.

[0007] Attempts have been made to improve target lifetime by adding additional material to a target to compensate for the uneven wear pattern of FIG. 3. For instance, FIG. 4 illustrates a target 20 which attempts to compensate for the uneven wear of FIG. 3. Target 20 is shown with a dashed line 18 illustrating the position of original face 16 in the target 10 of FIGS. 1-3. FIG. 4 also shows additional material 22 provided over original position 18, and in locations which compensate for the uneven wear profile of FIG. 3. Accordingly, target 20 has a face surface 24 which effectively comprises a mirror image of the wear profile of FIG. 3.

[0008] FIG. 4 is one embodiment of prior art processes for compensating for the uneven wear profile of FIG. 3. Another embodiment is to simply form additional material 22 over various regions of 18, without necessarily creating a mirror image of the wear of FIG. 3. Regardless of which of the prior art techniques is utilized, the result is a target having relatively large peaks at positions in which wear has been most significant in prior targets. A difficulty with the processing of FIG. 4 is that target 20 has large variations in thickness across its surface, and accordingly a T/S distance relative to face 24 of target 20 varies significantly across the face. Accordingly, the uniformity of:film deposition from target 20 can be significantly less than the uniformity of film deposition from a target having a planar face. Thus, even though lifetime can be improved utilizing the target 20 of FIG. 4 instead of the target 10 of FIGS. 1-3, the loss in uniformity can render target 20 less desirable than previous targets 10 of FIGS. 1-3.

[0009] It would be desirable to develop techniques for forming targets having improved lifetimes, and which can be utilized to uniformly sputter-deposit materials on substrates.

[0010] In one aspect, the invention encompasses a method of forming a sputtering target. A wear profile for a sputtering target surface is determined. The wear profile corresponds to a shape of the target surface after the target is subjected to the wear of having material sputtered therefrom. It can be preferred to determine a wear profile from a target which has been exposed to an anticipated semiconductor wafer fabrication process (specifically, an anticipated sputtering process), for a maximum anticipated lifetime of the target. The maximum anticipated lifetime can vary depending on, for example, the sputtering chamber configuration, the target composition, and the target configuration. The wear profile is divided amongst a plurality of datapoints across the target surface. A difference in height of the target surface after the wear relative to a height of the target surface prior to the wear is calculated. The difference in height calculations generate a plurality of wear definition datapoints. Target lifetime datapoints are calculated using the wear definition datapoints, and sputtering uniformity datapoints are also calculated using the wear definition datapoints. A difference between the target lifetime datapoints and sputtering uniformity datapoints is calculated. A constant corresponding to the difference between a target lifetime datapoint and a sputtering uniformity datapoint is added to the sputtering uniformity datapoints to generate a desired profile for a sputtering target sputtering surface. A sputtering target is formed having a sputtering surface with the desired profile.

[0011] The invention encompasses another method of forming a sputtering target. A wear profile for a sputtering target surface is determined. The wear profile is divided amongst a plurality of datapoints to generate datapoints {S₁. . . S_i}, where “i” is a positive integer. Also, datapoints are generated to define the target surface prior to the wear, with the datapoints being {R₁. . . Ri}. Difference datapoints {A₁. . . A_i} are generated, with each datapoint A_nbeing defined as R_n−S_n. Target lifetime datapoints {B₁. . . B_i} are calculated. Each datapoint B_nis defined as ((A_n* y)+Q); where y is a constant greater than 0, and Q is a constant which can be 0. Sputtering uniformity datapoints {C₁. . . C_i} are calculated, with each datapoint C_nbeing defined as ((A_n*z)+P); where z is a constant greater than 0 and less than y, and where P is a constant which can be 0. Difference datapoints {D₁. . . D_i} are calculated, with each difference datapoint D_n, being defined as (B_n−C_e). The difference datapoint having the greatest magnitude is determined, and is defined as D_max. A desired profile dataset {E₁. . . E_i} is generated, with each datapoint E_nbeing defined as (C_n+D_max). A sputtering target is formed to have a sputtering surface with a profile corresponding to the desired profile dataset.

[0012] Preferred embodiments of the invention are described below with reference to the following accompanying drawings.

[0013] FIG. 1 is a diagrammatic, cross-sectional view illustrating a prior art sputtering target and substrate.

[0014] FIG. 2 is a view along the line 2—2 of FIG. 1.

[0015] FIG. 3 is a diagrammatic, cross-sectional view of a prior art sputtering target after the target has been subjected to the wear of a sputtering operation.

[0016] FIG. 4 is a diagrammatic, cross-sectional view of a prior art sputtering target illustrating prior art methodology for increasing the lifetime of a sputtering target.

[0017] FIG. 5 is a flow chart diagram describing a method encompassed by the present invention.

[0018] FIG. 6 is a diagrammatic, cross-sectional view of a target which has been subjected to the wear of a sputtering operation, and specifically illustrates the target of FIG. 3. FIG. 6 also illustrates datapoints defined across the target surface for utilization in embodiments of the present invention.

[0019] FIG. 7 illustrates a curve generated in accordance with the present invention from the datapoints of FIG. 6, and specifically illustrates a curve comprising target lifetime datapoints.

[0020] FIG. 8 illustrates a second curve generated in accordance with the present invention from the datapoints of FIG. 6, and specifically illustrates a curve comprising sputtering uniformity datapoints.

[0021] FIG. 9 illustrates a difference curve generated by subtracting the FIG. 8 curve from the FIG. 7 curve.

[0022] FIG. 10 illustrates a desired target surface profile determined by adding a parameter determined from the FIG. 9 curve to the datapoints of the FIG. 8 curve.

[0023] FIG. 11 illustrates a cross-sectional view of sputtering target having a surface with the desired profile of FIG. 10.

[0024] The invention encompasses methodology for target design which can be utilized for designing desired profiles for sputtering target design so that the sputtering targets will have improved lifetime, and so that the sputtering targets will also sputter material to a desired uniformity. Methodology of the present invention can be utilized for improving sputtering targets of any shape, and comprising any material. The invention is described with reference to a flow chart of FIG. 5, and illustrations of FIGS. 6-11.

[0025] Referring to FIG. 5, an initial step of the invention is to measure a wear profile from a used sputtering target. Such is illustrated in FIG. 6, wherein the used sputtering target described previously with respect to FIG. 3 is shown. A wear profile of the sputtering target can be measured with, for example, a coordinate measuring machine. FIG. 6 shows the target 10 of FIG. 3, and shows a distance “X” relative to an initial upper surface 18 of target 10. Distance “X” can comprise, for example, about 1 ½ inches. FIG. 6 also shows a plurality of datapoints {A₁. . . A_i} (wherein i is an integer greater than 0). FIG. 6 actually shows only the four datapoints {A₁. . . A₄}, but it is to be understood that numerous other datapoints could be acquired and processed in addition to the shown datapoints. The datapoints can be across an entirety of a sputtering target face. It is noted that although the invention is described with reference to a process wherein a target wear profile is measured, it is to be understood that a target wear profile can be determined in other ways, such as, for example, by a computer-generated model of the wear profile rather than an actual measurement of the wear profile.

[0026] The datapoints {A₁. . . A₄} correspond to differences between the worn surface 16 and the initial surface 18. In practice, surface 18 is initially divided amongst a plurality of datapoints {R₁. . . R_i}, which, for the shown planar surface 18 will be the same as one another. Also, the worn surface 16 is divided into a plurality of datapoints {S₁. . . S_i}, and the datapoints {A₁. . . A_i} are determined as a difference between the datapoints R and the datapoints S. Specifically, a datapoint A_nis defined as R_n−S_n. The calculations of {A₁. . . A_i} correspond to step 102 of FIG. 5, which indicates that an amount of wear is calculated at a plurality of regions of a target by determining the amount of target removed at the regions. Each of the datapoints {A₁. . . A_i} corresponds to one of the regions referred to in step 102. In the shown embodiment, A₁, A₂, A₃and A₄have values of 4, 9, 4 and 6, respectively. The values are relative values of A₁, A₂, A₃and A₄to one another, and are provided for comparison of A₁, A₂, A₃and A₄. The values have units of length, but no particular units are assigned to the values used herein as the values are for illustrative purposes only and do not correspond to actual measured values.

[0027] The number of datapoints {A₁. . . A_i} that are calculated can vary depending on the processing equipment and time available. It can be desired to utilize a large number of datapoints {A₁. . . A_i} in that such will generally better characterize target wear than will fewer datapoints. However, a large number of datapoints will take more processing time than will fewer datapoints. In an exemplary embodiment, the number of datapoints is chosen so that spacing between adjacent datapoints is from about 0.05 inch to about 0.5 inch.

[0028] Referring to step 104 of FIG. 5, a plurality of target lifetime datapoints are calculated for the regions corresponding to {A₁. . . A_i}. Such calculation generates the curve shown in FIG. 7. More specifically, FIG. 7 illustrates a curve comprised of a plurality of target lifetime datapoints {B₁. . . B₁}. Each of the datapoints B_nis defined as ((A_n*y)+Q), where y is a constant greater than 0, and Q is a constant which can be 0. The constant y is defined as a target lifetime parameter. The target lifetime parameter can be from greater than 0 to 1, and is typically from 0.2 to 0.5, with 0.33 being an exemplary number. Ultimately, the target lifetime parameter can determine how much material is added to a target to increase the target lifetime. It can be preferred that a target lifetime be an integral of the lifetime of shields used around a periphery of the target (such as, for example, shields 14 of FIG. 1). For instance, if the shields have a lifetime of about 300 kilowatt hours it can be desirable that a target have a lifetime of either 600 kilowatt hours, 900 kilowatt hours, or 1200 kilowatt hours. Prior art targets have been produced having uncertain lifetimes. It would be desirable to develop targets having substantially exact lifetimes triple or quadruple the lifetime of shields. Methodology of the present invention can enable quality targets to be produced which have lifetimes of triple or more the lifetime of shields. The target lifetime parameter enables a lifetime of a target to be manipulated.

[0029] In the embodiment shown in FIG. 7, the target lifetime parameter is 0.5 and the constant Q is 0. Accordingly, in the shown embodiment in which A₁, A₂, A₃and A₄are 4, 9, 4 and 6, respectively; B₁, B₂, B₃and B₄are 2, 4.5, 2, and 3, respectively. The curve B is shown drawn relative to a dashed coordinate 30. Coordinate 30 is defined by the parameter “Q”. The constant Q can correspond to, for example, the distance “X” of FIG. 6, or can be any other number.

[0030] Referring to step 106 of FIG. 5, sputtering uniformity datapoints are determined for the various regions defined by datapoints {A₁. . . A_i}. Such is illustrated in FIG. 8, wherein the sputtering uniformity datapoints are shown as {C₁. . . C₄}. In practice, a plurality of datapoints {C₁. . . C_i} are defined from the plurality of datapoints {A₁. . . A_i}. Each datapoint C_nis defined as ((A_n*z)+P), where z is a constant greater than 0 and less than y, and where P is a constant which can be 0. The constant z is defined as a sputtering uniformity parameter. In practice, z is usually from about 0.001 to 1, and can be from about {fraction (1/16)} to about {fraction (1/16)}. The magnitude of z can depend on, for example, one or more of a magnet type utilized in a sputtering process, a target-to-substrate distance utilized in a sputtering process, a sputtering chamber configuration, and a target composition The datapoints {C₁. . . C_i} define a curve which could be utilized to form a target surface that would lead to a high uniformity of deposited material on a substrate. However, such target surface would not have a lifetime significantly improved relative to the original target surface 18 (FIG. 6).

[0031] FIG. 8 shows a curve generated using z=⅛, and specifically shows C₁, C₂, C₃and C₄equal to 0.5, 1.125, 0.5 and 0.75, respectively. The curve of FIG. 8 is shown relative to a coordinate 32. Coordinate 32 is defined by constant P and can, for example, correspond to the value “X” of FIG. 6. It can be preferred that coordinate 32 be identical in magnitude to coordinate 30 of FIG. 7, and accordingly it can be preferred that the constant P utilized to generate datapoints {C₁. . . C_i} be identical to the constant Q utilized to generate datapoints {B₁. . . B_i}.

[0032] Referring to step 108 of FIG. 5, a maximum difference between the target lifetime datapoints and the sputtering uniformity datapoints is determined. Such is illustrated in FIG. 9, wherein a curve is generated by subtracting the curve of FIG. 8 from that of FIG. 7. Specifically, a plurality of values {D₁. . . D_i} are generated with each value D_ncorresponding to B_n−C_n. The shown curve comprises D₁D₂, D₃and D₄corresponding to 1.5, 3.38, 1.5 and 2.25, respectively. The largest difference is referred to as D_max, and in the shown embodiment corresponds to the 3.38 of D₂. The curve of FIG. 9 is shown relative to a coordinate 34. Coordinate 34 is determined by the difference between Q and P. If parameter Q equals parameter P then coordinate 34 will be 0. If Q is different than P, coordinate 34 will have a value, and coordinate 34 can comprise either positive or negative value. It can be preferred for P to equal Q, and accordingly for coordinate 34 to equal 0.

[0033] Referring to step 110 of FIG. 5, the value D_maxis added to the uniformity datapoints of FIG. 8 to generate a desired target surface profile. The desired target surface profile is shown in FIG. 10, and comprises a plurality of datapoints {E₁. . . E_i}. Each of the datapoints E_nis calculated as C_n+D_max, with the values C_nbeing those shown in FIG. 8. In the shown embodiment, E₁, E₂, E₃and E₄correspond to 3.88, 4.51, 3.88 and 4.13, respectively. It is noted that values other than D_maxcan be added to the uniformity parameters of FIG. 8 to generate a desired target profile. However, if values less than D_maxare utilized, the target lifetime will be less than if D_maxwere used; and if values greater than D_maxare utilized, the resulting target may be too thick to be used in desired applications. In the embodiment described herein, D_maxis an additive value calculated from the target lifetime datapoints, and is added to the uniformity datapoints of FIG. 8 to generate a desired target surface profile. It is to be understood that the invention encompasses utilization of additive values other than values calculated from target lifetime datapoints in generating a desired target profile from sputtering uniformity datapoints, but such can be less preferred in that it can render it difficult to accurately control target lifetime.

[0034] Referring to step 112 of FIG. 5, the data from FIG. 10 is utilized to form a target having a surface with a desired target surface profile. Such is shown in FIG. 11, wherein a target 50 is shown having a surface 52 generated with the profile of FIG. 10. Target 50 has a shape corresponding to that of the target that generated the wear pattern of FIG. 3 with additional material defined by the data from FIG. 10 provided to form surface 52. More specifically, a dashed line 18 is shown to illustrate where the initial target of FIG. 3 would have had an upper surface. Additional material 54 is shown provided over dashed line 18, with additional material 54 corresponding to the profile of FIG. 10. Additional material 54 has the surface 52. Surface 52 defines a maximum target thickness determined by the target lifetime parameter y (assuming that D_maxis used with the curve of FIG. 8 to generate the desired target surface profile), and accordingly will lead to a target having a desired lifetime. Further, profile 52 has a surface planarity defined by the target uniformity parameter z, and accordingly will sputter deposit-material to a desired uniformity on a substrate. Accordingly, methodology of the present invention can provide a target having a desired lifetime, and also a desired sputtering uniformity. The parameters y and z can be determined to match desired specifications for particular target applications.

[0035] In compliance with the statute, the invention has been described in language more or less specific as to structural and methodical features. It is to be understood, however, that the invention is not limited to the specific features shown and described, since the means herein disclosed comprise preferred forms of putting the invention into effect. The invention is, therefore, claimed in any of its forms or modifications within the proper scope of the appended claims appropriately interpreted in accordance with the doctrine of equivalents.