Layout splitting method, computer program product and computer readable medium

Abstract

The invention provides a layout splitting method, and the method comprises the steps: employing a strategy network, obtaining the feature information of a current to-be-dyed node based on the dyeing information of the periphery of the current to-be-dyed node, and obtaining the feasible dyeing type of the current to-be-dyed node based on a preset DSA rule, and selecting an optimal color for dyeing. Therefore, the model does not need to be retrained for each new DSA process, the DSA rule and the learning model can be decoupled, and the generalization of the layout splitting algorithm is greatly enhanced.

Description

Technical Field

[0001] The present invention relates to a layout splitting method, in particular to a layout splitting method for a mask plate used in a guided self-assembly technology. Background Art

[0002] With the rapid rise of emerging technology industries such as 5G, artificial intelligence, and the metaverse, massive data has spawned a huge demand for computing power, which in turn has driven high-performance, high-computing chips to become the driving force and new trend of industry development. While the market demand for high-performance chips is growing, it has also further accelerated the iteration cycle of chips.

[0003] For example, in the process nodes of 7nm and below for logic chips, the minimum design rules for contact hole / via layers have exceeded the resolution limit of mainstream 193nm immersion lithography machines. In order to achieve small-period contact hole / via patterns, DUV 193i multi-patterning lithography technology is needed to meet the demand.

[0004] Multiple Patterning Lithography (MPL) is a technology that decomposes the layout with lithography conflicts into different masks, and performs multiple exposures and processing on each mask to achieve higher lithography resolution. However, this technology requires multiple masks, and the increase in the number of masks will lead to increased manufacturing costs and increased difficulty in the manufacturing process.

[0005] Directed Self-Assembly (DSA) lithography is a new type of graphic process with great development potential. It can realize the preparation of periodic nano-patterns. Compared with traditional lithography, it has the advantages of high resolution and low cost. DSA lithography is based on the molecular weight and microphase separation strength of block copolymers (BCP). After annealing, it can form high-resolution and high-density patterns and has been successfully applied to the research of various devices. With its advantages of low cost and high throughput, Directed Self-Assembly (DSA) technology can reduce the number of mask splits and provide a cost-effective solution for layout splitting. Summary of the invention

[0006] However, the existing layout splitting strategies are generally only applicable to layouts that can be composed of standard cells or sparse and nearly planar layouts, and do not consider the new conflicts introduced by the DSA template, and cannot be simplified into precise coverage problems. In addition, the existing DSA layout splitting model needs to be retrained for each new DSA process. It needs to learn specific DSA rules, such as whether the current node can be combined with other nodes to form a DSA point pair, which makes the DSA layout splitting model extremely complicated.

[0007] The present invention is completed to solve the above-mentioned problems, and its purpose is to provide a layout splitting method, which selects the optimal color for dyeing according to the feature information of the current node to be dyed obtained based on the dyeing information around the current node to be dyed, and the feasible dyeing types of the current node to be dyed obtained based on the predetermined DSA rules by utilizing the strategy network. There is no need to retrain the model for each new DSA process, and the DSA rules can be decoupled from the learning model, which greatly enhances the generalization of the layout splitting algorithm.

[0008] Technical solutions to technical problems

[0009] In order to solve the above problem, the layout splitting method involved in the first aspect of the present invention includes the following steps:

[0010] (1) Modeling the original layout into a conflict structure graph with through holes as nodes and conflict relationships between the through holes as edges;

[0011] (2) obtaining characteristic information of the node to be colored according to the coloring information around the node to be colored;

[0012] (3) Based on a predetermined rule, determining a feasible coloring type for the current node to be colored;

[0013] (4) Using a strategy network, according to the feature information of the node to be colored and the feasible coloring types of the node to be colored, the optimal color is selected for coloring.

[0014] Furthermore, the method further comprises the following steps:

[0015] (5) For all nodes to be colored, repeat steps (2), (3), and (4);

[0016] (6) Splitting the structure graph after all nodes to be colored into multiple split structure graphs so that nodes of the same color are located on the same split structure graph, thereby obtaining multiple split layouts corresponding to the multiple split structure graphs.

[0017] Furthermore, in the step (4), the coloring uniformity of the current structure graph is judged based on the color selected according to the feature information of the current node to be colored and the feasible coloring types of the current node to be colored, and evaluation information is fed back to the strategy network based on the coloring uniformity, and the strategy used in the strategy network is updated based on the evaluation information, so as to select the optimal color for coloring.

[0018] Further, the evaluation information is information showing a difference between the coloring uniformity of the current structure diagram and the average coloring uniformity.

[0019] Furthermore, in the step (4), when the color selected based on the feature information of the current node to be dyed and the feasible dyeing types of the current node to be dyed is not included in the feasible dyeing types, evaluation information is fed back to the strategy network, and the strategy used in the strategy network is updated based on the evaluation information, so as to select the optimal color for dyeing.

[0020] Furthermore, the predetermined rules are rules guiding the self-assembly process.

[0021] The rules in the guided self-assembly process include the number of split layouts, the conflict distances between multiple nodes, and the conditions for combining multiple nodes into point pairs.

[0022] Furthermore, the coloring information around the current node to be colored includes the coloring information of the nodes within the conflict range around the current node to be colored, and the information of the edges corresponding to the conflict relationship between the through holes.

[0023] Furthermore, the feature information is a high-dimensional feature vector.

[0024] Furthermore, the layout is a layout used in a guided self-assembly process.

[0025] The computer program product involved in the second aspect of the present invention includes a computer program, which implements the above-mentioned layout splitting method when executed by a processor.

[0026] Effects of the Invention

[0027] According to the layout splitting method of the present invention, by utilizing a strategy network, the optimal color is selected for dyeing based on the feature information of the current node to be dyed obtained based on the dyeing information around the current node to be dyed, and the feasible dyeing types of the current node to be dyed obtained based on a predetermined DSA rule. Therefore, there is no need to retrain the model for each new DSA process, and the DSA rule can be decoupled from the learning model, which greatly enhances the generalization of the layout splitting algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Therefore, in order to understand the above-mentioned features of the present invention in detail, the present invention can be described in more detail by referring to the embodiments, and a more specific description of the disclosure briefly described above can be made by referring to the embodiments, some of which are illustrated in the accompanying drawings. However, it should be noted that the drawings only illustrate exemplary embodiments and therefore should not be considered as limiting the scope thereof, and other equivalent embodiments may be allowed.

[0029] Figure 1 It is a schematic diagram showing a specific framework example of the layout splitting method involved in the embodiment of the present invention.

[0030] Figure 2 It is a flowchart showing an example of specific steps of the layout dividing method involved in the embodiment of the present invention.

[0031] Figure 3A It is a flowchart showing an example of a feedback process implemented in the step of selecting the optimal color for coloring according to the feature information of the current node to be colored and the feasible coloring types of the current node to be colored in the layout splitting method involved in the embodiment of the present invention.

[0032] Figure 3B It is a flowchart showing another example of the feedback process implemented in the step of selecting the optimal color for coloring according to the feature information of the current node to be colored and the feasible coloring types of the current node to be colored in the layout splitting method involved in the embodiment of the present invention.

[0033] Figure 4 It is a schematic diagram showing an example of multiple split structure graphs obtained by splitting a structure graph after all nodes to be colored are colored using the layout splitting method involved in an embodiment of the present invention.

[0034] Figure 5 It is a schematic diagram showing an example of a structural diagram obtained by coloring a structural diagram including two nodes (including diagonal ones) constituting a point pair and three nodes (including diagonal ones) constituting a point pair using the layout splitting method involved in an embodiment of the present invention.

[0035] Description of symbols

[0036] 11 Figure Network Neural Network

[0037] 12First Environment Module

[0038] 13Feature Storage Module

[0039] 14Strategy Network

[0040] 15 Second Environment Module DETAILED DESCRIPTION

[0041] The present invention is described in detail below in conjunction with the accompanying drawings and specific implementations. This implementation is implemented based on the technical solution of the present invention, and a detailed implementation and specific operation process are given, but the protection scope of the present invention is not limited to the following implementation.

[0042] For ease of description, spatially relative terms, such as "under," "below," "down," "above," "on," etc., may be used herein to describe the relationship of one element or feature relative to another element or feature as shown in the figures. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation other than the orientation shown in the figures. For example, if the device in the figure is turned over, elements described as being "under" or "below" other elements or features will be oriented "above" the other elements or features.

[0043] Unless otherwise defined, the terms used herein have the same meanings as those generally understood by those skilled in the art to which the present invention belongs. The terms should be understood to have the same meanings as those in the context of the relevant technology and should not be understood in an idealized or overly formal sense unless explicitly defined in this manner.

[0044] <Layout Splitting Method>

[0045] Below, in conjunction with the accompanying drawings, an example of a layout splitting method involved in an embodiment of the present invention is described in detail.

[0046] As an example, the layout splitting method involved in the first aspect of the present invention includes the following steps:

[0047] (1) Modeling the original layout into a conflict structure graph with through holes as nodes and conflict relationships between the through holes as edges;

[0048] (2) obtaining characteristic information of the node to be colored according to the coloring information around the node to be colored;

[0049] (3) Based on a predetermined rule, determining a feasible coloring type for the current node to be colored;

[0050] (4) Using a strategy network, according to the feature information of the node to be colored and the feasible coloring types of the node to be colored, the optimal color is selected for coloring.

[0051] Figure 1 It is a schematic diagram showing a specific framework example of the layout splitting method involved in the embodiment of the present invention. Figure 2 It is a flowchart showing an example of specific steps of the layout dividing method involved in the embodiment of the present invention.

[0052] like Figure 2As shown, first, in step ST101, the original layout is modeled as a conflict structure graph with through holes as nodes and conflict relationships between through holes as edges. Specifically, for example, the through holes in the original layout can be modeled as nodes in the conflict structure graph, so that the through holes in the original layout and the nodes in the conflict structure graph form a one-to-one correspondence. Figure 1 The edges in the figure represent the conflict relationship between through-holes. For example, when there is a gray edge between two nodes, it means that there is a conflict relationship between the through-holes, and the two nodes need to be assigned to different split structure graphs. When there is no gray edge between the two nodes, it means that there is no conflict relationship between the through-holes, and the two nodes can be assigned to the same split structure graph.

[0053] In addition, as an example, in the initial state, the nodes in the conflict structure graph are gray, for example, and first, according to the number of split structure graphs to be split (for example Figure 1 The original conflict structure graph is initialized and colored by a predetermined initial algorithm, wherein there are nodes to be colored. The red, yellow, and blue here do not mean that the nodes in the conflict structure graph are to be colored visually red, yellow, and blue, but are only to distinguish the different split structure graphs to be assigned later. For example, all red nodes can be assigned to the first split structure graph, all yellow nodes can be assigned to the second split structure graph, and all blue nodes can be assigned to the third split structure graph.

[0054] Then, in step ST102, characteristic information of the current node to be colored is obtained based on the coloring information around the current node to be colored. Specifically, for example, a node to be colored in the original conflict structure graph can be selected as the current node to be colored, and the graph neural network 11 obtains characteristic information of the current node to be colored based on the coloring information around the current node to be colored, and inputs the characteristic information into the characteristic storage module 13.

[0055] As an example, the coloring information around the current node to be colored includes the coloring information of the nodes within the conflict range around the current node to be colored and the edge information corresponding to the conflict relationship between the through holes. For example, when the conflict distance between two nodes in the DSA rule is 108nm, the above-mentioned conflict range is a circular range with a radius of 108nm centered on the current node to be colored.

[0056] In addition, as an example, the feature information of the current node to be colored is, for example, a high-dimensional feature vector (refer to Figure 1 below the “+” in the feature storage module 13).

[0057] Then, in step ST103, the first environment module 12 determines the feasible dyeing type of the current node to be dyed based on a predetermined rule. The predetermined rule may be, for example, a rule in a DSA process, and the rule in the DSA process may include, for example, the number of split layouts, the conflict distance between multiple nodes, and the condition for combining multiple nodes into a point pair.

[0058] As an example, Figure 1 As shown, based on the rules in the DSA process, it can be determined that the current node to be dyed can be dyed red or blue, but cannot be dyed yellow (refer to Figure 1 ) Specifically, when the current node to be dyed is dyed red, it satisfies the rules in the DSA process, so it is judged to be dyed red. In addition, when the current node to be dyed is dyed blue, it can be combined into a point pair with the adjacent blue, which satisfies the rules in the DSA process, so it is judged to be dyed blue. In addition, when the current node to be dyed is dyed yellow, it does not meet the rules in the DSA process because the distance between it and the adjacent yellow is less than the conflict distance, so it is judged not to be dyed yellow.

[0059] Then, in step ST104, the strategy network 14 is used to select the optimal color for coloring according to the feature information of the current node to be colored and the feasible coloring types of the current node to be colored.

[0060] As an example, the policy network 14 selects red as the optimal color for coloring based on the feature information (high-dimensional feature vector) of the current node to be colored and the feasible coloring types (red or blue) of the current node to be colored, and inputs it into the second environment module 15.

[0061] Next, use Figure 3A and Figure 3B , two feedback processes that can be used in step ST104 are explained. Figure 3A It is a flowchart showing an example of a feedback process implemented in the step of selecting the optimal color for coloring according to the feature information of the current node to be colored and the feasible coloring types of the current node to be colored in the layout splitting method involved in the embodiment of the present invention. Figure 3B The flowchart is another example of the feedback process implemented in the step of selecting the optimal color for coloring according to the feature information of the current node to be colored and the possible coloring types of the current node to be colored according to the layout splitting method involved in the embodiment of the present invention. In addition, at least one of the following first feedback process and second feedback process can be adopted in step ST104.

[0062] As the first feedback process, Figure 3AAs shown, in step ST1041a, the second environment module 15 determines the coloring uniformity of the current structure diagram based on the color selected according to the characteristic information of the current node to be colored and the feasible coloring types of the current node to be colored. The coloring uniformity here refers to whether the proportions of red, blue, and yellow are uniform (i.e., whether the number of each color is the same).

[0063] Then, in step ST1042a, the second environment module 15 feeds back evaluation information to the strategy network based on the dyeing uniformity. As an example, when the proportion of red when the current node to be dyed is dyed red is large (i.e., the number of red is large), the evaluation information fed back to the strategy network 14 shows that the reward is negative. When the proportion of red when the current node to be dyed is still small (i.e., the number of red is small), the evaluation information fed back to the strategy network 14 shows that the reward is positive. In addition, as an example, the evaluation information can also be information showing the difference between the dyeing uniformity of the current structure diagram and the average dyeing uniformity. Thus, the influence of the uniformity (density) of the DSA layout on manufacturability is also considered in the layout splitting process.

[0064] Then, in step ST1043a, the strategy used in the policy network is updated based on the evaluation information, so as to select the optimal color for dyeing. The purpose of the strategy update is to increase the probability of actions with high rewards, so that the rewards obtained by the actions of the policy network are getting higher and higher, thereby selecting the optimal color for dyeing.

[0065] As the second feedback process, Figure 3B As shown, in step ST1041b, the second environment module 15 feeds back evaluation information to the policy network when the color selected according to the characteristic information of the current node to be dyed and the feasible dyeing types of the current node to be dyed is not included in the feasible dyeing types. As an example, when the color selected by the policy network 14 according to the characteristic information of the current node to be dyed and the feasible dyeing types of the current node to be dyed is yellow, since yellow is not included in the feasible dyeing types (red or blue), the evaluation information fed back to the policy network 14 is shown as a large penalty.

[0066] Then, in step ST1042b, the strategy used in the strategy network 14 is updated based on the evaluation information, so as to select the optimal color for coloring. The purpose of the strategy update is to reduce the probability of actions with large penalties and increase the probability of actions with high rewards, so that the rewards obtained by the actions of the strategy network are getting higher and higher, thereby selecting the optimal color for coloring.

[0067] Next, return to Figure 2The description continues with step ST105.

[0068] In step ST105, it is determined whether all nodes to be colored have been colored. If not all nodes to be colored have been colored, steps ST102, ST103, and ST104 are repeated for all nodes to be colored. If all nodes to be colored have been colored, the process proceeds to step ST106.

[0069] In step ST106, the structure graph after all nodes to be colored are colored is split into multiple split structure graphs so that nodes of the same color are located on the same split structure graph, thereby obtaining multiple split layouts corresponding to the multiple split structure graphs.

[0070] Figure 4 FIG. 1 is a schematic diagram showing an example of a plurality of split structure graphs obtained by splitting a structure graph after all nodes to be colored are colored using the layout splitting method according to an embodiment of the present invention. Figure 4 As shown, as an example, the multiple split structure diagrams obtained by splitting the structure diagram after dyeing all the nodes to be dyed are purple, black, green, red, and yellow. Among them, the time complexity of dyeing all the nodes to be dyed is O(N) (i.e. linear), and the time consumption of dyeing each node is, for example, 0.0008s. In addition, Figure 4 In the figure, the conflict distance between nodes is 108nm, each node is covered with a circular hole with a diameter of 50nm, and there are no two nodes with a minimum distance less than 50nm on the same split layout.

[0071] also, Figure 5 1 is a schematic diagram showing an example of a structure diagram obtained by coloring a structure diagram including two nodes (including oblique ones) constituting a point pair and three nodes (including oblique ones) constituting a point pair using the layout splitting method involved in the embodiment of the present invention. Figure 5 As shown, as an example, by using the layout splitting method involved in the embodiment of the present invention, two nodes / three nodes constituting point pairs in the horizontal, vertical and diagonal directions can be dyed the same color ( Figure 5 blue, green, and red in the middle).

[0072] Therefore, according to the layout splitting method of the present invention, by utilizing the strategy network, the optimal color is selected for dyeing according to the feature information of the current node to be dyed obtained based on the dyeing information around the current node to be dyed, and the feasible dyeing types of the current node to be dyed obtained based on the predetermined DSA rules. Therefore, there is no need to retrain the model for each new DSA process, and the DSA rules can be decoupled from the learning model, which greatly enhances the generalization of the layout splitting algorithm.

[0073] In addition, the second aspect of the present invention also provides a computer program product, including a computer program, which implements the above-mentioned layout splitting method when executed by a processor.

[0074] It should be understood that the above description is illustrative and not restrictive. For example, the above embodiments (and / or their various aspects) can be used in combination with each other. In addition, without departing from the scope of the present invention, many modifications can be made to adapt specific conditions or materials to the teachings of the various embodiments of the present invention. Although the size and type of the material described herein are used to define the parameters of the various embodiments of the present invention, the various embodiments are not meant to be restrictive, but exemplary embodiments. In the case of reading the above description, many other embodiments are obvious to those skilled in the art. Therefore, the scope of the various embodiments of the present invention should be determined with reference to the attached claims, and the full range of equivalent forms claimed by these claims.

[0075] Industrial Applicability

[0076] The layout splitting method of the present invention can be widely used in the fields of semiconductor technology, chip manufacturing, etc., and has broad research and application value.

Claims

1. A layout splitting method, characterized in that: The steps include: (1) Modeling the original layout into a conflict structure graph with through holes as nodes and conflict relationships between the through holes as edges; (2) obtaining characteristic information of the node to be colored according to the coloring information around the node to be colored; (3) Based on a predetermined rule, determining a feasible coloring type for the current node to be colored; (4) Using a strategy network, according to the feature information of the node to be colored and the feasible coloring types of the node to be colored, the optimal color is selected for coloring.

2. The layout splitting method according to claim 1, characterized in that: The following steps are also included: (5) For all nodes to be colored, repeat steps (2), (3), and (4); (6) Splitting the structure graph after all nodes to be colored into multiple split structure graphs so that nodes of the same color are located on the same split structure graph, thereby obtaining multiple split layouts corresponding to the multiple split structure graphs.

3. The layout splitting method according to claim 1 or 2, characterized in that: In the step (4), the coloring uniformity of the current structure graph is judged based on the color selected according to the feature information of the current node to be colored and the feasible coloring types of the current node to be colored, and evaluation information is fed back to the strategy network based on the coloring uniformity. The strategy used in the strategy network is updated based on the evaluation information, so as to select the optimal color for coloring.

4. The layout splitting method according to claim 3, characterized in that: The evaluation information is information showing a difference between the coloring uniformity of the current structure diagram and the average coloring uniformity.

5. The layout splitting method according to claim 1 or 2, characterized in that: In the step (4), when the color selected based on the feature information of the current node to be dyed and the feasible dyeing types of the current node to be dyed is not included in the feasible dyeing types, evaluation information is fed back to the strategy network, and the strategy used in the strategy network is updated based on the evaluation information, so as to select the optimal color for dyeing.

6. The layout splitting method according to claim 1 or 2, characterized in that: The predetermined rules are the rules guiding the self-assembly process, The rules in the guided self-assembly process include the number of split layouts, the conflict distances between multiple nodes, and the conditions for combining multiple nodes into point pairs.

7. The layout splitting method according to claim 1 or 2, characterized in that: The coloring information around the node to be colored currently includes the coloring information of nodes within the conflict range around the node to be colored currently, and the information of the edges corresponding to the conflict relationship between the through holes.

8. The layout splitting method according to claim 1 or 2, characterized in that: The feature information is a high-dimensional feature vector.

9. The layout splitting method according to claim 1 or 2, characterized in that: The layout is a layout used in a guided self-assembly process.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the layout splitting method described in any one of claims 1 to 9 is implemented.

Images