Method and system for obtaining etching deviation prediction model, equipment and storage medium

Abstract

A method and system for obtaining an etching deviation prediction model, a device and a storage medium, the method comprising: obtaining a first training data set of first graphs, the first training data set comprising the width of the first graphs, the spacing between adjacent first graphs, and the distribution density of the first graphs; acquiring a second training data set of the ion implantation regions, wherein the second training data set comprises the width of the ion implantation regions, the distance between the adjacent ion implantation regions and the distribution density of the ion implantation regions; obtaining a third training data set of the overlapped graph, wherein the third training data set comprises the width of the overlapped graph; performing linear regression processing on the first training data set, the second training data set and the third training data set to obtain an etching deviation data set; and obtaining an etching deviation prediction model according to the etching deviation data set. And the precision of the etching deviation value obtained according to the etching deviation prediction model is improved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of semiconductor manufacturing, and in particular to a method, system, device, and storage medium for obtaining an etching deviation prediction model. Background Art

[0002] Optical Proximity Correction (OPC) is fundamental to ensuring accurate chip dimensions. As semiconductor technology evolves toward advanced processes and critical pattern dimensions decrease, etch bias (EB) compensation has become a key focus of OPC correction.

[0003] The usual way to calculate EB is: for a fixed pattern, its critical dimension (CD) is measured in the after development inspection (ADI) stage and the after etching inspection (AEI) stage respectively. The difference between the two CD values ​​is the EB.

[0004] Ion implants affect the etch rate, which in turn affects the EB difference between areas with and without implants. If EB compensation doesn't fully account for the implant-induced differences, linewidth will deviate from the target value. Because the implant's impact on etch is complex, rigorous equation derivation cannot guarantee OPC efficiency.

[0005] Therefore, the prediction model to obtain etching deviation still needs to be improved. Summary of the Invention

[0006] The problem solved by the embodiments of the present invention is to provide a method, system, device and storage medium for obtaining an etching deviation prediction model, thereby improving the accuracy of obtaining an etching deviation value.

[0007] To solve the above problems, an embodiment of the present invention provides a method for obtaining an etching deviation prediction model, comprising: providing a pattern to be tested, the pattern to be tested comprising a plurality of spaced-apart ion implantation regions, the pattern to be tested comprising a plurality of spaced-apart first patterns, some of the first patterns being located in the ion implantation regions, and the first patterns located in the ion implantation regions being used as overlapping patterns; obtaining a first training data set of the first patterns, the first training data set comprising the width of the first patterns, the spacing between adjacent first patterns, and the distribution density of the first patterns; obtaining a second training data set of the ion implantation regions, the second training data set comprising the width of the ion implantation regions, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation regions; obtaining a third training data set of the overlapping patterns, the third training data set comprising the width of the overlapping patterns; performing linear regression processing on the first training data set, the second training data set, and the third training data set, respectively, to obtain an etching deviation data set; and obtaining an etching deviation prediction model based on the etching deviation data set.

[0008] Optionally, before performing linear regression processing on the first training data set, the second training data set and the third training data set respectively, it also includes: selecting any first graphic that meets the maximum safety distance from the ion implantation area, and using the first graphic as the anchor graphic; obtaining the post-development size of the anchor graphic and the post-etching size of the anchor graphic; and obtaining the etching deviation of the anchor graphic based on the post-development size and the post-etching size.

[0009] Optionally, linear regression processing is performed on the first training data set, the second training data set and the third training data set respectively, and the step of obtaining the etching deviation data set includes: performing a first linear regression processing on the width of the first graphic to obtain the etching deviation value of the width of the first graphic; performing a second linear regression processing on the spacing of the first graphic to obtain the etching deviation value of the spacing of the first graphic; performing a third linear regression processing on the width of the overlapping graphic to obtain the etching deviation value of the width of the overlapping graphic; performing a fourth linear regression processing on the distribution density of the ion implantation area to obtain the distribution density value of the ion implantation area; performing a fifth linear regression processing on the distribution density of the first graphic to obtain the distribution density value of the first graphic, the etching deviation value of the width of the first graphic, the etching deviation value of the spacing of the first graphic, The etching deviation value of the width of the overlapping graphic, the distribution density value of the ion implantation area, and the distribution density value of the first graphic constitute the etching deviation data set; in the step of obtaining the etching deviation prediction model according to the etching deviation data set, the etching deviation prediction model EB=A0*EB0+A1*bias1+A2*bias2+A3*D+A4*bias3++A5*Im; wherein, A0 refers to the weight ratio of the etching deviation of the anchor point graphic, A1 refers to the weight ratio of the etching deviation of the width of the first graphic, A2 refers to the weight ratio of the etching deviation of the spacing of the first graphic, A3 refers to the weight ratio of the distribution density of the first graphic, A4 refers to the weight ratio of the etching deviation of the width of the overlapping graphic, and A5 refers to the weight ratio of the distribution density of the ion implantation area.

[0010] Optionally, the etching bias value bias1 of the width of the first pattern is obtained as follows: bias1 = (W-W0) / (W end -W0); wherein W refers to the width of the first graphic, W0 refers to the minimum line width allowed by the first graphic, and W end Refers to the ideal width value of the first graphic.

[0011] Optionally, the ideal width value W of the first graphic end =min(W max , 2Rc), where W max It refers to the maximum line width value allowed by the first graphic, and Rc refers to the characteristic radius value of the first graphic that meets the convergence standard.

[0012] Optionally, the etching bias value bias2 of the spacing of the first pattern is obtained as follows: (S-S0) / (S end -S0); wherein S refers to the spacing of the first graphic, S0 refers to the minimum line spacing allowed by the first graphic, S endRefers to the ideal spacing value of the first graphic.

[0013] Optionally, the ideal spacing value S of the first graphic end =min(S max , 2Rc), where S max It refers to the maximum spacing value allowed by the first graphic, and Rc refers to the characteristic radius value of the first graphic that meets the convergence standard.

[0014] Optionally, the distribution density value of the first graphic is D=W / (W+S), wherein W refers to the width of the first graphic and S refers to the spacing of the first graphic.

[0015] Optionally, the distribution density value Im of the ion implantation area is W i / (W i +S i ), where W i refers to the width of the ion implantation area, S i Refers to the spacing of the ion implantation areas.

[0016] Optionally, the etching deviation value of the width of the overlapping pattern Wherein, E refers to the width of the overlapping graphic, and W0 refers to the minimum line width allowed for the first graphic.

[0017] Accordingly, an embodiment of the present invention provides a system for obtaining an etching deviation prediction model, comprising: a providing module for providing a pattern to be tested, the pattern to be tested comprising a plurality of spaced-apart ion implantation regions, the pattern to be tested comprising a plurality of spaced-apart first patterns, some of the first patterns being located in the ion implantation regions, and the first patterns located in the ion implantation regions being used as overlapping patterns; a first acquisition module for obtaining a first training data set of the first patterns, the first training data set comprising the width of the first patterns, the spacing between adjacent first patterns, and the distribution density of the first patterns; a second acquisition module for obtaining a second training data set of the ion implantation regions, the second training data set comprising the width of the ion implantation regions, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation regions; a third acquisition module for obtaining a third training data set of the overlapping patterns, the third training data set comprising the width of the overlapping patterns; a fourth acquisition module for performing linear regression processing on the first training data set, the second training data set, and the third training data set, respectively, to obtain an etching deviation data set; and a fifth acquisition module for obtaining an etching deviation prediction model based on the etching deviation data set.

[0018] Optionally, the system for obtaining the etching deviation prediction model also includes: a sixth acquisition module, used to select any first graphic that meets the maximum safety distance from the ion implantation area, and use the first graphic as the anchor graphic; obtain the post-development size of the anchor graphic and the post-etching size of the anchor graphic; and obtain the etching deviation of the anchor graphic based on the post-development size and the post-etching size.

[0019] Optionally, the fourth acquisition module includes: a first processing unit for performing a first linear regression process on the width of the first graphic to obtain an etching deviation value of the width of the first graphic; a second processing unit for performing a second linear regression process on the spacing of the first graphic to obtain an etching deviation value of the spacing of the first graphic; a third processing unit for performing a third linear regression process on the width of the overlapping graphic to obtain an etching deviation value of the width of the overlapping graphic; a fourth processing unit for performing a fourth linear regression process on the distribution density of the ion implantation area to obtain a distribution density value of the ion implantation area; a fifth processing unit for performing a fifth linear regression process on the distribution density of the first graphic to obtain the distribution density value of the first graphic, the etching deviation value of the width of the first graphic, and the etching deviation value of the spacing of the first graphic. , the etching deviation value of the width of the overlapping graphics, the distribution density value of the ion implantation area, and the distribution density value of the first graphics constitute the etching deviation data set; in the step of obtaining the etching deviation prediction model according to the etching deviation data set, the etching deviation prediction model EB=A0*EB0+A1*bias1+A2*bias2+A3*D+A4*bias3++A5*Im; wherein, A0 refers to the weight ratio of the etching deviation of the anchor point graphics, A1 refers to the weight ratio of the etching deviation of the width of the first graphics, A2 refers to the weight ratio of the etching deviation of the spacing of the first graphics, A3 refers to the weight ratio of the distribution density of the first graphics, A4 refers to the weight ratio of the etching deviation of the width of the overlapping graphics, and A5 refers to the weight ratio of the distribution density of the ion implantation area.

[0020] Accordingly, an embodiment of the present invention also provides a device comprising at least one memory and at least one processor, wherein the memory stores one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method for obtaining an etching deviation prediction model provided in an embodiment of the present invention.

[0021] Accordingly, an embodiment of the present invention further provides a storage medium storing one or more computer instructions, wherein the one or more computer instructions are used to implement the method for obtaining an etching deviation prediction model provided by an embodiment of the present invention.

[0022] Compared with the prior art, the technical solution of the embodiment of the present invention has the following advantages:

[0023] In a method for obtaining an etching deviation prediction model provided by an embodiment of the present invention, a first training data set of the first graphic is obtained, the first training data set including the width of the first graphic, the spacing between adjacent first graphics, and the distribution density of the first graphic; a second training data set of the ion implantation region is obtained, the second training data set including the width of the ion implantation region, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation region; a third training data set of the overlapping graphic is obtained, the third training data set including the width of the overlapping graphic; linear regression processing is performed on the first training data set, the second training data set, and the third training data set to obtain an etching deviation data set; and an etching deviation prediction model is obtained based on the etching deviation data set. Compared to the existing solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation region, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation region. During the subsequent linear regression processing, the second training data set will change according to changes in the width, spacing, and distribution density of the ion implantation region, so that the second training data set is a variable during the linear regression processing, thereby making the obtained etching deviation prediction model more accurate and improving the precision of the etching deviation value obtained according to the etching deviation prediction model. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flow chart of an embodiment of a method for obtaining an etching deviation prediction model according to the present invention;

[0025] Figure 2 2 is a schematic diagram corresponding to an embodiment of a method for obtaining an etching deviation prediction model according to the present invention;

[0026] Figure 3 This is a functional block diagram of an embodiment of the present invention for obtaining an etching deviation prediction model;

[0027] Figure 4 It is a hardware structure diagram of an embodiment of the device provided by the present invention. DETAILED DESCRIPTION

[0028] As can be seen from the background technology, obtaining etching deviation through prediction models still needs to be improved.

[0029] In order to solve the technical problem, the embodiment of the present invention provides a method for obtaining an etching deviation prediction model. Figure 1 , showing a flow chart of an embodiment of a method for obtaining an etching deviation prediction model according to the present invention.

[0030] In this embodiment, the method for obtaining the etching deviation prediction model includes the following basic steps:

[0031] Step S1: providing a pattern to be tested, wherein the pattern to be tested includes a plurality of ion implantation regions distributed at intervals, and the pattern to be tested includes a plurality of first patterns distributed at intervals, wherein some of the first patterns are located in the ion implantation regions, and the first patterns located in the ion implantation regions are used as overlapping patterns;

[0032] Step S21: Acquire a first training data set of the first graphics, where the first training data set includes the width of the first graphics, the spacing between adjacent first graphics, and the distribution density of the first graphics;

[0033] Step S22: obtaining a second training data set of the ion implantation region, wherein the second training data set includes a width of the ion implantation region, a spacing between adjacent ion implantation regions, and a distribution density of the ion implantation region;

[0034] Step S23: obtaining a third training data set of the overlapping graphics, wherein the third training data set includes the width of the overlapping graphics;

[0035] Step S3: performing linear regression processing on the first training data set, the second training data set, and the third training data set respectively to obtain an etching deviation data set;

[0036] Step S4: obtaining an etching deviation prediction model according to the etching deviation data set.

[0037] In an embodiment of the present invention, a first training data set of the first graphic is obtained, where the first training data set includes the width of the first graphic, the spacing between adjacent first graphics, and the distribution density of the first graphic. A second training data set of the ion implantation region is obtained, where the second training data set includes the width of the ion implantation region, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation region. A third training data set of the overlapping graphic is obtained, where the third training data set includes the width of the overlapping graphic. Linear regression processing is performed on the first, second, and third training data sets to obtain an etching deviation data set. An etching deviation prediction model is obtained based on the etching deviation data set. Compared to the existing solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation region, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation region. During the subsequent linear regression processing, the second training data set will change according to changes in the width, spacing, and distribution density of the ion implantation region, so that the second training data set is a variable during the linear regression processing, thereby making the obtained etching deviation prediction model more accurate and improving the accuracy of the etching deviation value obtained according to the etching deviation prediction model.

[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0039] Figure 2 : is a schematic diagram corresponding to an embodiment of a method for obtaining an etching deviation prediction model according to the present invention, wherein Figure 1 It is a flow chart of an embodiment of a method for obtaining an etching deviation prediction model according to the present invention.

[0040] refer to Figure 2 , execute step S1: provide a pattern to be tested 100, wherein the pattern to be tested 100 includes a plurality of ion implantation regions 101 distributed at intervals, and the pattern to be tested 100 includes a plurality of first patterns 103 distributed at intervals, part of the first patterns 103 is located in the ion implantation regions 101, and the first patterns 103 located in the ion implantation regions 101 are used as overlapping patterns 108.

[0041] Specifically, the pattern to be tested 100 provides a basis for subsequently acquiring an etching deviation data set.

[0042] In this embodiment, in the process of forming the semiconductor structure, the ion implantation region 101 is used as an ion doping region.

[0043] In this embodiment, in the process of forming a semiconductor structure, the first pattern 103 is used to form a device structure, such as a gate structure, an interconnection via structure, or a metal line.

[0044] In this embodiment, part of the first pattern 103 is located in the ion implantation region 101 . That is, in the process of forming the semiconductor structure, a doping treatment is performed on a part of the device structure formed by the first pattern 103 .

[0045] It should be noted that using the first figure 103 located in the ion implantation area 101 as the overlapping figure 108 is conducive to the subsequent acquisition of a training data set of the overlapping figure 108, so that the influence of the ion implantation area 101 on the etching deviation value can be taken into account in the prediction model, thereby improving the accuracy of the etching deviation value obtained by the etching deviation prediction model.

[0046] refer to Figure 2 , execute step S21: obtain a first training data set of the first graphic 103, the first training data set includes the width W2 of the first graphic 103, the spacing W5 between adjacent first graphics 103, and the distribution density of the first graphic 103.

[0047] It should be noted that the first training data set provides data support for the subsequent acquisition of the etching deviation data set.

[0048] refer to Figure 2 , execute step S22: obtain a second training data set of the ion implantation region 101, the second training data set including the width W1 of the ion implantation region 101, the spacing W4 between adjacent ion implantation regions 101, and the distribution density of the ion implantation region 101.

[0049] The second training data set provides data support for the subsequent acquisition of the etching deviation data set. At the same time, compared with the solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation region 101, the spacing between adjacent ion implantation regions 101, and the distribution density of the ion implantation region 101. In the subsequent linear regression processing, the second training data set will change according to the changes in the width, spacing and distribution density of the ion implantation region 101, so that the second training data set is a variable in the linear regression processing, thereby making the obtained etching deviation prediction model more accurate, and improving the accuracy of the etching deviation value obtained according to the etching deviation prediction model.

[0050] refer to Figure 2 , executing step S23: obtaining a third training data set of the overlapping graphic 108 , wherein the third training data set includes a width W3 of the overlapping graphic 108 .

[0051] The third training data set provides data support for the subsequent acquisition of the etching deviation data set.

[0052] It should be noted that there is no particular order in which step S21 , step S22 , and step S23 are executed.

[0053] refer to Figure 2 Before subsequently performing linear regression processing on the first training data set, the second training data set, and the third training data set, the method further includes: selecting any first graphic 103 that meets the maximum safety distance from the ion implantation area 101, and using the first graphic 103 as an anchor graphic; obtaining a post-development size of the anchor graphic and a post-etching size of the anchor graphic; and obtaining an etching deviation of the anchor graphic based on the post-development size and the post-etching size.

[0054] Specifically, the anchor point pattern does not overlap with the ion implantation region 101 , and the distance between the anchor point pattern and the ion implantation region 101 satisfies the maximum safety distance.

[0055] Specifically, the maximum safety distance is limited according to the process size requirements.

[0056] It should be noted that obtaining the etching deviation of the anchor point pattern provides data for subsequently obtaining an etching deviation prediction model.

[0057] refer to Figure 2 , executing step S3: performing linear regression processing on the first training data set, the second training data set and the third training data set respectively to obtain an etching deviation data set.

[0058] Specifically, compared with the solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation region 101, the spacing between adjacent ion implantation regions 101, and the distribution density of the ion implantation region 101. In the subsequent linear regression processing, the second training data set will change according to the changes in the width, spacing and distribution density of the ion implantation region 101, so that the second training data set is a variable in the linear regression processing, thereby making the obtained etching deviation prediction model more accurate, and improving the accuracy of the etching deviation value obtained according to the etching deviation prediction model.

[0059] As an example, linear regression processing is performed on the first training data set, the second training data set and the third training data set respectively, and the step of obtaining the etching deviation data set includes: performing a first linear regression processing on the width W2 of the first graphic 103 to obtain the etching deviation value of the width W2 of the first graphic 103.

[0060] Specifically, the etching bias value bias1 of the width W2 of the first pattern 103 is obtained as follows: bias1 = (W-W0) / (W end -W0); wherein W refers to the width W2 of the first graphic 103, W0 refers to the minimum line width allowed by the first graphic 103, and W end Refers to the ideal width value of the first graphic 103.

[0061] As an example, the ideal width value W of the first graphic 103 is end =min(W max , 2Rc), where W max It refers to the maximum line width value allowed by the first graphic 103, and Rc refers to the characteristic radius value of the first graphic 103 that meets the convergence standard.

[0062] In this embodiment, a second linear regression process is performed on the spacing of the first patterns 103 to obtain an etching deviation value of the spacing of the first patterns 103 .

[0063] It should be noted that the etching bias value bias2 for obtaining the spacing of the first pattern 103 is (S-S0) / (S end -S0); wherein S refers to the spacing of the first graphic 103, S0 refers to the minimum line spacing allowed by the first graphic 103, S end Refers to the ideal spacing value of the first graphic 103.

[0064] As an example, the ideal spacing value S of the first graphic 103 is end =min(S max , 2Rc), where S max It refers to the maximum spacing value allowed by the first graphic 103, and Rc refers to the characteristic radius value of the first graphic 103 that meets the convergence standard.

[0065] In this embodiment, a third linear regression process is performed on the width of the overlapping pattern 108 to obtain an etching deviation value of the width of the overlapping pattern 108 .

[0066] It should be noted that the etching deviation value of the width of the overlapping pattern 108 is Here, E refers to the width of the overlapping pattern 108 , and W0 refers to the minimum line width allowed by the first pattern 103 .

[0067] In this embodiment, a fourth linear regression process is performed on the distribution density of the ion implantation region 101 to obtain a distribution density value of the ion implantation region 101 .

[0068] It should be noted that the distribution density value Im of the ion implantation region 101 is Wi / (W i +S i ), where W i Refers to the width W1 of the ion implantation region 101, S i It refers to the spacing W4 of the ion implantation region 101 .

[0069] As an example, the distribution density of the first graphic 103 is subjected to a fifth linear regression process to obtain the distribution density value of the first graphic 103. The etching deviation value of the width W2 of the first graphic 103, the etching deviation value of the spacing of the first graphic 103, the etching deviation value of the width of the overlapping graphic 108, the distribution density value of the ion implantation area 101, and the distribution density value of the first graphic 103 constitute the etching deviation data set.

[0070] It should be noted that the distribution density value D of the first graphic 103 is D=W / (W+S), wherein W refers to the width W2 of the first graphic 103 , and S refers to the spacing W5 of the first graphic 103 .

[0071] refer to Figure 2 , executing step S4: obtaining an etching deviation prediction model according to the etching deviation data set.

[0072] It should be noted that, compared with the solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation region 101, the spacing between adjacent ion implantation regions 101, and the distribution density of the ion implantation region 101. In the subsequent linear regression processing, the second training data set will change according to the changes in the width, spacing and distribution density of the ion implantation region 101, so that the second training data set is a variable in the linear regression processing. Accordingly, in the process of obtaining the etching deviation prediction model based on the etching deviation data set, the etching deviation prediction model can be made more accurate, thereby improving the accuracy of the etching deviation prediction model in obtaining the etching deviation value.

[0073] As an example, in the step of obtaining an etching bias prediction model based on the etching bias data set, the etching bias prediction model EB=A0*EB0+A1*bias1+A2*bias2+A3*D+A4*bias3++A5*Im; wherein, A0 refers to the weight ratio of the etching bias of the anchor point figure, A1 refers to the weight ratio of the etching bias of the width W2 of the first figure 103, A2 refers to the weight ratio of the etching bias of the spacing of the first figure 103, A3 refers to the weight ratio of the distribution density of the first figure 103, A4 refers to the weight ratio of the etching bias of the width of the overlapping figure 108, and A5 refers to the weight ratio of the distribution density of the ion implantation area 101.

[0074] It should be noted that A0 to A5 are assigned values ​​according to their respective proportions.

[0075] Correspondingly, the present invention also provides a system for obtaining an etching deviation prediction model. Figure 3 It is a functional block diagram of the etching deviation prediction model acquisition system of the present invention.

[0076] The system 300 for obtaining an etching deviation prediction model includes: a providing module 301 for providing a pattern to be tested, wherein the pattern to be tested includes a plurality of spaced ion implantation regions, the pattern to be tested includes a plurality of spaced first patterns, some of the first patterns are located in the ion implantation regions, and the first patterns located in the ion implantation regions are used as overlapping patterns; a first acquisition module 302 for obtaining a first training data set of the first patterns, wherein the first training data set includes the width of the first patterns, the spacing between adjacent first patterns, and the distribution density of the first patterns; a second acquisition module 303 for obtaining a second training data set of the ion implantation regions, wherein the second training data set includes the width of the ion implantation regions, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation regions; a third acquisition module 304 for obtaining a third training data set of the overlapping patterns, wherein the third training data set includes the width of the overlapping patterns; a fourth acquisition module 305 for performing linear regression processing on the first training data set, the second training data set, and the third training data set respectively to obtain an etching deviation data set; and a fifth acquisition module 306 for obtaining an etching deviation prediction model based on the etching deviation data set.

[0077] Specifically, compared with the existing solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion injection area, the spacing between adjacent ion injection areas, and the distribution density of the ion injection area. In the subsequent linear regression processing, the second training data set will change according to the changes in the width, spacing and distribution density of the ion injection area, so that the second training data set is a variable in the linear regression processing, thereby making the obtained etching deviation prediction model more accurate, and improving the accuracy of the etching deviation value obtained according to the etching deviation prediction model.

[0078] Specifically, the providing module 301 is used to provide a pattern to be tested, so that the pattern to be tested provides a basis for subsequently obtaining an etching deviation data set.

[0079] In this embodiment, in the process of forming the semiconductor structure, the ion implantation region is used as an ion doping region.

[0080] In this embodiment, in the process of forming a semiconductor structure, the first pattern is used to form a device structure, such as a gate structure, an interconnection via structure, or a metal line.

[0081] In this embodiment, part of the first pattern is located in the ion implantation region. That is, in the process of forming the semiconductor structure, a doping treatment is performed on a part of the device structure formed by the first pattern.

[0082] It should be noted that using the first figure located in the ion implantation area as an overlapping figure facilitates the subsequent acquisition of a training data set of overlapping figures, so that the influence of the ion implantation area on the etching deviation value can be taken into account in the prediction model, thereby improving the accuracy of the etching deviation value obtained by the etching deviation prediction model.

[0083] The first acquisition module 302 is used to acquire a first training data set of the first graphics, where the first training data set includes the width of the first graphics, the spacing between adjacent first graphics, and the distribution density of the first graphics.

[0084] It should be noted that the first training data set provides data support for the subsequent acquisition of the etching deviation data set.

[0085] The second acquisition module 303 is used to acquire a second training data set of the ion implantation region, where the second training data set includes the width of the ion implantation region, the spacing between adjacent ion implantation regions, and the distribution density of the ion implantation region.

[0086] The second training data set provides data support for the subsequent acquisition of the etching deviation data set. At the same time, compared with the solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation area, the spacing between adjacent ion implantation areas, and the distribution density of the ion implantation area. In the subsequent linear regression processing, the second training data set will change according to the changes in the width, spacing and distribution density of the ion implantation area, so that the second training data set is a variable in the linear regression processing, thereby making the acquired etching deviation prediction model more accurate, and improving the accuracy of the etching deviation value obtained according to the etching deviation prediction model.

[0087] The third acquisition module 304 is used to acquire a third training data set of the overlapping graphics, where the third training data set includes the width of the overlapping graphics.

[0088] Specifically, the third training data set provides data support for the subsequent acquisition of the etching deviation data set.

[0089] It should be noted that the fourth acquisition module 305 is used to perform linear regression processing on the first training data set, the second training data set and the third training data set respectively to obtain an etching deviation data set.

[0090] Specifically, compared with the solution in which the second training data set is a fixed value, the second training data set of the embodiment of the present invention includes the width of the ion implantation area, the spacing between adjacent ion implantation areas, and the distribution density of the ion implantation area. In the subsequent linear regression processing, the second training data set will change according to the changes in the width, spacing and distribution density of the ion implantation area, so that the second training data set is a variable in the linear regression processing, thereby making the obtained etching deviation prediction model more accurate, and improving the accuracy of the etching deviation value obtained according to the etching deviation prediction model.

[0091] In this embodiment, the fourth acquisition module 305 includes: a first processing unit configured to perform a first linear regression process on the width of the first pattern to obtain an etching deviation value of the width of the first pattern.

[0092] Specifically, the etching bias value bias1 of the width of the first pattern is obtained as follows: (W-W0) / (W end -W0); wherein W refers to the width of the first graphic, W0 refers to the minimum line width allowed by the first graphic, and W end Refers to the ideal width value of the first graphic.

[0093] As an example, the ideal width value W of the first graphic is end =min(W max, 2Rc), where W max It refers to the maximum line width value allowed by the first graphic, and Rc refers to the characteristic radius value of the first graphic that meets the convergence standard.

[0094] As an example, the second processing unit is configured to perform a second linear regression process on the spacing of the first graphics to obtain an etching deviation value of the spacing of the first graphics.

[0095] It should be noted that the etching bias value bias2 for obtaining the spacing of the first pattern is (S-S0) / (S end -S0); wherein S refers to the spacing of the first graphic, S0 refers to the minimum line spacing allowed by the first graphic, S end Refers to the ideal spacing value of the first graphic.

[0096] As an example, the ideal spacing value S of the first graph is end =min(S max , 2Rc), where S max It refers to the maximum spacing value allowed by the first graphic, and Rc refers to the characteristic radius value of the first graphic that meets the convergence standard.

[0097] As an example, the third processing unit is configured to perform a third linear regression process on the width of the overlapping pattern to obtain an etching deviation value of the width of the overlapping pattern.

[0098] It should be noted that the etching deviation value of the width of the overlapping pattern is Wherein, E refers to the width of the overlapping graphic, and W0 refers to the minimum line width allowed for the first graphic.

[0099] As an example, the fourth processing unit is configured to perform a fourth linear regression process on the distribution density of the ion implantation region to obtain a distribution density value of the ion implantation region.

[0100] It should be noted that the distribution density value Im of the ion implantation area is W i / (W i +S i ), where W i refers to the width of the ion implantation area, S i Refers to the spacing of the ion implantation areas.

[0101] As an example, a fifth linear regression process is performed on the distribution density of the first pattern to obtain a distribution density value of the first pattern. The etching deviation value of the width of the first pattern, the etching deviation value of the spacing between the first pattern, the etching deviation value of the width of the overlapping pattern, the distribution density value of the ion implantation area, and the distribution density value of the first pattern constitute the etching deviation data set. It should be noted that the distribution density value D of the first pattern is D = W / (W+S), where W refers to the width of the first pattern and S refers to the spacing between the first pattern.

[0102] In this embodiment, the fifth processing unit is used to perform a fifth linear regression process on the distribution density of the first graphic to obtain the distribution density value of the first graphic. The etching deviation value of the width of the first graphic, the etching deviation value of the spacing of the first graphic, the etching deviation value of the width of the overlapping graphic, the distribution density value of the ion implantation area, and the distribution density value of the first graphic constitute the etching deviation data set.

[0103] It should be noted that the distribution density value of the first graphic is D=W / (W+S), wherein W refers to the width of the first graphic and S refers to the spacing of the first graphic.

[0104] As an example, the system 300 for obtaining an etching deviation prediction model also includes: a sixth acquisition module 307, used to select any first graphic that meets the maximum safety distance from the ion implantation area, and use the first graphic as an anchor graphic; obtain the post-development size of the anchor graphic and the post-etching size of the anchor graphic; and obtain the etching deviation of the anchor graphic based on the post-development size and the post-etching size.

[0105] Specifically, the anchor point pattern does not overlap with the ion implantation region, and the distance between the anchor point pattern and the ion implantation region satisfies the maximum safety distance.

[0106] Specifically, the maximum safety distance is limited according to the process size requirements.

[0107] It should be noted that the etching deviation of the anchor point pattern provides data for the subsequent acquisition of the etching deviation prediction model.

[0108] The fifth acquisition module 306 is configured to acquire an etching deviation prediction model according to the etching deviation dataset.

[0109] Specifically, in the step of obtaining the etching bias prediction model according to the etching bias data set, the etching bias prediction model EB=A0*EB0+A1*bias1+A2*bias2+A3*D+A4*bias3++A5*Im; wherein, A0 refers to the weight ratio of the etching bias of the anchor point figure, A1 refers to the weight ratio of the etching bias of the width of the first figure, A2 refers to the weight ratio of the etching bias of the spacing of the first figure, A3 refers to the weight ratio of the distribution density of the first figure, A4 refers to the weight ratio of the etching bias of the width of the overlapping figure, and A5 refers to the weight ratio of the distribution density of the ion implantation area.

[0110] It should be noted that A0 to A5 are assigned values ​​according to their respective proportions.

[0111] The embodiment of the present invention further provides a device, which can implement the method for obtaining an etching deviation prediction model provided by the embodiment of the present invention by loading the above-mentioned method for obtaining an etching deviation model in the form of a program. An optional hardware structure of the terminal device provided by the embodiment of the present invention can be as follows: Figure 4 As shown, it includes: at least one processor 01, at least one communication interface 02, at least one memory 03 and at least one communication bus 04.

[0112] In this embodiment, the number of processor 01, communication interface 02, memory 03, and communication bus 04 is at least one, and the processor 01, communication interface 02, and memory 03 communicate with each other through the communication bus 04. The communication interface 02 can be an interface of a communication module for network communication, such as an interface of a GSM module. The processor 01 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement an embodiment of the present invention. The memory 03 may include a high-speed RAM memory, and may also include a non-volatile memory (NVM), such as at least one disk storage. Among them, the memory 03 stores one or more computer instructions, and the one or more computer instructions are executed by the processor 01 to implement the method for obtaining an etching deviation prediction model provided in an embodiment of the present invention.

[0113] It should be noted that the above-mentioned terminal device may also include other devices (not shown) that may not be necessary for understanding the contents disclosed in the embodiments of the present invention; given that these other devices may not be necessary for understanding the contents disclosed in the embodiments of the present invention, the embodiments of the present invention will not introduce them one by one.

[0114] An embodiment of the present invention further provides a storage medium storing one or more computer instructions, wherein the one or more computer instructions are used to implement the method for obtaining an etching deviation prediction model provided by an embodiment of the present invention.

[0115] The embodiments of the present invention can be implemented by various means such as hardware, firmware, software or a combination thereof. In a hardware configuration, the method according to the exemplary embodiment of the present invention can be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, etc. In a firmware or software configuration, the embodiments of the present invention can be implemented in the form of modules, processes, functions, etc. The software code can be stored in a memory unit and executed by a processor. The memory unit is located inside or outside the processor and can send data to the processor and receive data from the processor via various known means.

[0116] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the scope defined by the claims.

Claims

1. A method for obtaining an etching deviation prediction model, characterized in that: include: Providing a pattern to be tested, the pattern to be tested comprising a plurality of ion implantation regions distributed at intervals, the pattern to be tested comprising a plurality of first patterns distributed at intervals, some of the first patterns being located in the ion implantation regions, and the first patterns located in the ion implantation regions being used as overlapping patterns; Acquire a first training data set of the first graphics, where the first training data set includes a width of the first graphics, a spacing between adjacent first graphics, and a distribution density of the first graphics; Acquire a second training data set of the ion implantation region, wherein the second training data set includes a width of the ion implantation region, a spacing between adjacent ion implantation regions, and a distribution density of the ion implantation region; Acquire a third training data set of the overlapping graphics, wherein the third training data set includes the width of the overlapping graphics; Performing linear regression processing on the first training data set, the second training data set, and the third training data set to obtain an etching deviation data set; An etching deviation prediction model is obtained according to the etching deviation data set.

2. The method for obtaining an etching deviation prediction model according to claim 1, wherein: Before performing linear regression processing on the first training data set, the second training data set and the third training data set respectively, it also includes: selecting any first graphic that meets the maximum safety distance from the ion implantation area, and using the first graphic as the anchor graphic; obtaining the post-development size of the anchor graphic and the post-etching size of the anchor graphic; and obtaining the etching deviation of the anchor graphic based on the post-development size and the post-etching size.

3. The method for obtaining an etching deviation prediction model according to claim 2, wherein: The step of performing linear regression processing on the first training data set, the second training data set, and the third training data set to obtain an etching deviation data set includes: performing a first linear regression processing on the width of the first pattern to obtain an etching deviation value of the width of the first pattern; Performing a second linear regression process on the spacing of the first pattern to obtain an etching deviation value of the spacing of the first pattern; performing a third linear regression process on the width of the overlapping pattern to obtain an etching deviation value of the width of the overlapping pattern; performing a fourth linear regression process on the distribution density of the ion implantation region to obtain a distribution density value of the ion implantation region; performing a fifth linear regression process on the distribution density of the first pattern to obtain a distribution density value of the first pattern, wherein the etching deviation value of the width of the first pattern, the etching deviation value of the spacing between the first patterns, the etching deviation value of the width of the overlapping pattern, the distribution density value of the ion implantation region, and the distribution density value of the first pattern constitute the etching deviation data set; In the step of obtaining an etching bias prediction model according to the etching bias data set, the etching bias prediction model EB=A0*EB0+A1*bias1+A2*bias2+A3*D+A4*bias3++A5*Im; wherein, A0 refers to the weight ratio of the etching bias of the anchor point figure, A1 refers to the weight ratio of the etching bias of the width of the first figure, A2 refers to the weight ratio of the etching bias of the spacing of the first figure, A3 refers to the weight ratio of the distribution density of the first figure, A4 refers to the weight ratio of the etching bias of the width of the overlapping figure, and A5 refers to the weight ratio of the distribution density of the ion implantation area.

4. The method for obtaining an etching deviation prediction model according to claim 3, wherein: Obtain the etching bias value bias1 of the width of the first pattern = (W-W0) / (W end -W0); wherein W refers to the width of the first graphic, W0 refers to the minimum line width allowed by the first graphic, and W end Refers to the ideal width value of the first graphic.

5. The method for obtaining an etching deviation prediction model according to claim 4, wherein: The ideal width value W of the first graphic end =min(W max , 2Rc), where W max It refers to the maximum line width value allowed by the first graphic, and Rc refers to the characteristic radius value of the first graphic that meets the convergence standard.

6. The method for obtaining an etching deviation prediction model according to claim 3, wherein: Obtain the etching bias value bias2 of the spacing of the first pattern = (S-S0) / (S end -S0); wherein S refers to the spacing of the first graphic, S0 refers to the minimum line spacing allowed by the first graphic, S end Refers to the ideal spacing value of the first graphic.

7. The method for obtaining an etching deviation prediction model according to claim 6, wherein: The ideal spacing value S of the first pattern end =min(S max , 2Rc), where S max It refers to the maximum spacing value allowed by the first graphic, and Rc refers to the characteristic radius value of the first graphic that meets the convergence standard.

8. The method for obtaining an etching deviation prediction model according to claim 3, wherein: The distribution density value of the first pattern is D=W / (W+S), where W refers to the width of the first pattern and S refers to the spacing of the first pattern.

9. The method for obtaining an etching deviation prediction model according to claim 3, wherein: The distribution density value of the ion implantation area Im=W i / (W i +S i ), where W i refers to the width of the ion implantation area, S i Refers to the spacing of the ion implantation areas.

10. The method for obtaining an etching deviation prediction model according to claim 3, wherein: The etching deviation value of the width of the overlapping pattern Wherein, E refers to the width of the overlapping graphic, and W0 refers to the minimum line width allowed for the first graphic.

11. A system for obtaining an etching deviation prediction model, characterized in that: include: Providing a module for providing a pattern to be tested, wherein the pattern to be tested includes a plurality of ion implantation regions distributed at intervals, the pattern to be tested includes a plurality of first patterns distributed at intervals, some of the first patterns are located in the ion implantation regions, and the first patterns located in the ion implantation regions are used as overlapping patterns; A first acquisition module is configured to acquire a first training data set of the first graphics, where the first training data set includes a width of the first graphics, a spacing between adjacent first graphics, and a distribution density of the first graphics; A second acquisition module is configured to acquire a second training data set of the ion implantation region, wherein the second training data set includes a width of the ion implantation region, a spacing between adjacent ion implantation regions, and a distribution density of the ion implantation region; A third acquisition module is used to acquire a third training data set of the overlapping graphics, wherein the third training data set includes the width of the overlapping graphics; a fourth acquisition module, configured to perform linear regression processing on the first training data set, the second training data set, and the third training data set respectively to acquire an etching deviation data set; The fifth acquisition module is configured to acquire an etching deviation prediction model according to the etching deviation dataset.

12. The system for obtaining an etching deviation prediction model according to claim 11, wherein: The system for obtaining the etching deviation prediction model also includes: a sixth acquisition module, used to select any first graphic that meets the maximum safety distance from the ion implantation area and use the first graphic as the anchor point graphic; obtain the post-development size of the anchor point graphic and the post-etching size of the anchor point graphic; and obtain the etching deviation of the anchor point graphic based on the post-development size and the post-etching size.

13. The system for obtaining an etching deviation prediction model according to claim 12, wherein: The fourth acquisition module includes: a first processing unit, configured to perform a first linear regression process on the width of the first pattern to obtain an etching deviation value of the width of the first pattern; a second processing unit, configured to perform a second linear regression process on the spacing of the first pattern to obtain an etching deviation value of the spacing of the first pattern; a third processing unit, configured to perform a third linear regression process on the width of the overlapping pattern to obtain an etching deviation value of the width of the overlapping pattern; a fourth processing unit, configured to perform a fourth linear regression process on the distribution density of the ion implantation region to obtain a distribution density value of the ion implantation region; a fifth processing unit, configured to perform a fifth linear regression process on the distribution density of the first pattern to obtain a distribution density value of the first pattern, wherein the etching deviation value of the width of the first pattern, the etching deviation value of the spacing between the first patterns, the etching deviation value of the width of the overlapping patterns, the distribution density value of the ion implantation area, and the distribution density value of the first pattern constitute the etching deviation data set; In the step of obtaining an etching bias prediction model according to the etching bias data set, the etching bias prediction model EB=A0*EB0+A1*bias1+A2*bias2+A3*D+A4*bias3++A5*Im; wherein, A0 refers to the weight ratio of the etching bias of the anchor point figure, A1 refers to the weight ratio of the etching bias of the width of the first figure, A2 refers to the weight ratio of the etching bias of the spacing of the first figure, A3 refers to the weight ratio of the distribution density of the first figure, A4 refers to the weight ratio of the etching bias of the width of the overlapping figure, and A5 refers to the weight ratio of the distribution density of the ion implantation area.

14. A device, characterized in that The method comprises at least one memory and at least one processor, wherein the memory stores one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method for obtaining an etching deviation prediction model according to any one of claims 1 to 10.

15. A storage medium, characterized in that: The storage medium stores one or more computer instructions, and the one or more computer instructions are used to implement the method for obtaining an etching deviation prediction model according to any one of claims 1 to 10.

Images