CN107153719A - Method and system for filling redundant metal - Google Patents

Abstract

The present invention provides a redundant metal filling method and system, the method comprising: analyzing the layout density of the layout to be filled, confirming the density to be filled of each grid; setting the filling density according to the density to be filled of each grid The redundant metal side length and spacing of each grid; according to the set redundant metal side length and spacing, index array analysis is performed on each grid to confirm the array to be filled in each grid; the to-be-filled array in each grid is The array is integrated to obtain the area to be filled in each grid; redundant metal filling is performed in the area to be filled in each grid. Since the index array analysis is performed on each grid according to the set redundant metal side length and spacing, and the array to be filled is integrated to obtain the area to be filled, the present invention can quickly complete redundant metal filling without missing filling. .

Description

Method and system for a redundant metal filling method

technical field

The invention relates to the field of semiconductor manufacturing, in particular to a method and system of a redundant metal filling method.

Background technique

With the continuous development of semiconductor technology, semiconductor process nodes are becoming smaller and smaller, and chemical mechanical polishing (CMP) is used for global planarization, which becomes more and more important. During the manufacturing process of the CMP process, since metals and oxides and other media have different material properties, they have different material removal rates during the CMP process, resulting in dishing and erosion defects. A butterfly defect is the difference in thickness between the dielectric layer in the blank area and the metal in the trench. Erosion defects refer to the thinning of the oxide and metal in the pattern area, which is the difference in the thickness of the oxide layer before and after polishing. The prior art eliminates butterfly defects and erosion defects mainly through redundant metal filling.

The main principle of redundant metal filling (Dummy Fill) technology is to change the metal pattern density of the grid by filling the grid with small metal blocks that are not connected to the circuit, thereby improving the uniformity of the pattern density of the layout and improving the flatness of the chip degree and electrical performance. There are two core technologies: 1) layout density and filling quantity analysis technology; 2) margin analysis technology.

At present, the research on layout density and filling quantity analysis technology is mainly based on the layout grid density analysis to determine the filling area and filling density. For example, the min-variation proposed by Kahng et al. and the min-fill linear programming model proposed by Tian et al. There are two main algorithms in the research of margin analysis technology, array index algorithm and scan line algorithm. Among them, the array index algorithm divides the grid into n*n index arrays, and fills the arrays that meet the requirements. Its advantage is that the calculation speed is fast, but its disadvantage is that the size of the array depends heavily on the redundant metal filling parameter settings. When When the filling parameters set are inappropriate, it will lead to missing filling phenomenon, and the traditional arrangement after filling, the parasitic effect is obvious, and the adjustability is poor; the scan line algorithm divides the grid into several areas through the scan line, and then redundancies these areas Metal filling, the advantage of this method is that it avoids the partial missing filling problem of the array indexing algorithm. There are various filling methods, which can effectively reduce the parasitic effect. The disadvantage is that it cannot handle the layout of complex geometric structures, and the calculation speed is much lower than that of the array indexing algorithm. .

Contents of the invention

The present invention provides a method and system of a redundant metal filling method, which solves the problem that the redundant metal filling cannot be quickly completed without missing filling through the existing margin analysis technology.

The present invention provides a method for redundant metal filling method, comprising the steps of:

Perform layout density analysis on the layout to be filled to confirm the density to be filled of each grid;

Set the redundant metal side length and spacing for filling each grid according to the density to be filled of each grid;

Perform index array analysis on each grid according to the set redundant metal side length and spacing, and confirm the array to be filled in each grid;

Integrate the arrays to be filled in each grid to obtain the area to be filled in each grid;

Redundant metal filling is performed in the area to be filled in each grid.

Preferably, performing layout density analysis on the layout to be filled, and confirming the density to be filled of each grid includes:

Step A1: Using chemical mechanical planarization CMP-design for manufacturability DFM tool to obtain the density value D _ij in each grid, the grid size is G×G;

Step A2: Determine whether there is a grid to be filled that exceeds the preset density lower limit DI, and if so, obtain the density D1 of the grid to be filled, so that the density of the grid to be filled is adjusted to the preset density lower limit DI; if If not, go directly to step A3;

Step A3: sequentially calculate the maximum density D _max of each grid and its adjacent grids from one side to the opposite side;

Step A4: Determine whether there is a grid to be filled whose difference from D _max is greater than the preset allowable density fluctuation ΔD, and if so, update the density D1 of the grid to be filled so that the density of the grid to be filled is adjusted is D _max -△D; if not, go to step A5;

Step A5: Obtain the to-be-filled density of each grid sequentially from one side to the opposite side, and turn to the next row when the opposite side is reached.

Preferably, the index array analysis is performed on each grid according to the set redundant metal side length and spacing, and it is confirmed that the array to be filled in each grid includes:

The grid to be filled determined by step A2 and step A4 is divided into arrays with L1 as the side length, wherein, W and S are the initially set redundant metal side length and spacing;

It is judged whether each array overlaps with the metal line, if yes, the array is marked as an array not to be filled; if not, the array is marked as an array to be filled.

Preferably, the method also includes:

Before judging whether each array overlaps with the metal line, the critical size L0 of the defect is simulated through the CMP process;

Expand the metal lines in each grid by L0/2;

Described judging whether each array overlaps with the metal line, if so, then marking the array as an array not to be filled; if not, marking the array as an array to be filled includes:

It is judged whether each array overlaps with the expanded metal line, if yes, the array is marked as an array not to be filled; if not, the array is marked as an array to be filled.

Preferably, the area to be filled is rectangular or square, and the arrays to be filled in each grid are integrated to obtain the area to be filled in each grid including:

Step C1: start traversing from the array at one corner of the grid, skip if the current array N _{i, j} is not an array to be filled, otherwise enter step C2, the traversal includes: advancing from one side to the other side, and advancing to Turn to the next row or column when opposite sides;

Step C2: From one side of the grid, when the current array N _{i, j} is an array to be filled, judge whether N _{i+1, j} or N _{i, j+1} is an array to be filled, and if so, set N _{i, j} , N _{i+1, j} , or N _{i, j} , N _{i, j+1} are merged into (i, j) area M _{i, j} to be filled, and continue to judge the array N _{i+2, j} or N _{i, j+ 2} Whether it is an array to be filled, and so on, until N _{i+k, j} or N _{i, j+k} is not an array to be filled or the opposite side of the grid, where i is the number of columns, j is the number of rows, and k is positive integer;

Step C3: Judging whether there are arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} that are not to be filled, if not, the array N _{i, j+1} to N _{i+k, j+1} or the array N _{i+1, j} to N _{i+1, j+k} are merged into (i, j) area M _{i, j} to be filled, and continue to judge Whether the arrays N _{i, j+2} to N _{i+k, j+2} or the arrays N _{i+2, j} to N _{i+2, j+k} are merged into (i, j) area M _{i, j} to be filled, so that By analogy, when arrays N _{i, j+p} to N _{i+k, j+p} or arrays N _{i+p, j} to N _{i+p, there are non-to-be-filled arrays in j+k} , if so, do not adjust ( i, j) area to be filled M _{i, j} ; wherein p is a positive integer;

Step C4: Repeat steps C1 to C3 until traversing all the arrays in the current grid to obtain each area to be filled in the grid;

Step C5: Repeat steps C1 to C4 until the area to be filled in each grid is obtained.

Preferably, the redundant metal filling in the regions to be filled in each grid includes:

Determine the geometric coordinates of each vertex of the area to be filled;

Calculate the actual filling coordinates of redundant metal according to the geometric coordinates of each vertex in the area to be filled;

Output the actual filling coordinates of the redundant metal to the layout file to complete the redundant metal filling.

A system of redundant metal filling methods comprising:

The density confirmation module is used to analyze the layout density of the layout to be filled, and confirm the density to be filled of each grid;

A redundant metal setting module is used to set the redundant metal side length and spacing for filling each grid according to the density to be filled of each grid;

The array confirmation module is used to analyze the index array of each grid according to the set redundant metal side length and spacing, and confirm the array to be filled in each grid;

An area acquisition module, configured to integrate the arrays to be filled in each grid to obtain the area to be filled in each grid;

The filling module is used for redundant metal filling in the area to be filled in each grid.

Preferably, the density confirmation module includes:

A density value acquisition unit, configured to use the CMP-DFM tool to obtain the density value D _ij in each grid, where the grid size is G×G;

The first judging unit is used to judge whether there is a grid to be filled exceeding the preset density lower limit DI, and if so, obtain the density D1 of the grid to be filled, so that the density of the grid to be filled is adjusted to the preset density Lower limit DI; if not, execute calculation unit;

A calculation unit is used to sequentially calculate the maximum density D _max of each grid and its adjacent grids from one side to the opposite side;

The second judging unit is used to judge whether there is a grid to be filled whose difference with D _max is greater than the preset allowable density fluctuation ΔD, if so, update the density D1 of the grid to be filled, so that the grid to be filled The density of is adjusted to D _max -△D; if not, the acquisition unit is executed;

The acquisition unit is used to obtain the density to be filled of each grid sequentially from one side to the opposite side, and transfer to the next line when the opposite side is reached.

Preferably, the array confirmation module includes:

The array division unit is used to divide the grid to be filled determined by step A2 and step A4 into arrays with L1 as the side length, wherein, W and S are the initially set redundant metal side length and spacing;

The array marking unit is used for judging whether each array overlaps with the metal line, and if so, marks the array as an array not to be filled; if not, marks the array as an array to be filled.

Preferably, the system also includes:

The critical dimension acquisition module is used to simulate the critical dimension L0 of defects generated by the CMP process before judging whether each array overlaps with the metal line;

The metal wire expansion module connected with the array confirmation module is used to expand the metal wires in each grid to L0/2;

The array marking unit is specifically used for judging whether each array overlaps with the expanded metal line, and if so, marks the array as an array not to be filled; if not, marks the array as an array to be filled.

Preferably, the area acquisition module includes:

The traversal unit is used to start traversing from the array at a vertex of the grid, skip if the current array is not an array to be filled, otherwise execute the area merging unit, the traversal includes: advancing from one side to the opposite side, advancing to the opposite side Go to the next row or column when the side is turned;

The area merging unit is used to determine whether N _{i+1, j} or N _{i, j+1} is an array to be filled from one side of the grid, and if the current array N _{i, j} is an array to be filled, if so, N _ij , N _{i+1, j} , or N _ij , N _{i, j+1} are merged into (i, j) area M _{i, j} to be filled, and continue to judge the array N _{i+2, j} or N _{i, j+2} Whether it is an array to be filled, and so on, until N _{i+k, j} or N _{i, j+k} is not an array to be filled or the opposite side of the grid, where i is the number of columns, j is the number of rows, and k is positive integer;

The area adjustment unit is used to determine whether there are arrays not to be filled in arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} , if not, Then merge arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} into (i, j) area M _{i, j} to be filled, And continue to judge whether the arrays N _{i, j+2} to N _{i+k, j+2} or the arrays N _{i+2, j} to N _{i+2, j+k} are merged into (i, j) area M _{i to be filled, j} , and so on, when arrays N _{i, j+p} to N _{i+k, j+p} or arrays N _{i+p, j} to N _{i+p, j+k} have arrays not to be filled, if so, then Do not adjust (i, j) area M _{i, j} to be filled; where p is a positive integer;

The third judging unit is used to judge whether to traverse all the arrays in the current grid, if so, execute the fourth judging unit, otherwise execute the traversing unit;

The fourth judging unit is used for judging whether to traverse all grids, if so, end, if not, execute the traversing unit.

Preferably, the filling module includes:

The area coordinate determination unit is used to determine the geometric coordinates of each vertex of the area to be filled;

The filling coordinate technical unit is used to calculate the actual filling coordinates of the redundant metal according to the geometric coordinates of each vertex in the area to be filled;

The filling unit is used to output the actual filling coordinates of the redundant metal to the layout file to complete the redundant metal filling.

The method and system of a redundant metal filling method provided by the present invention, the method confirms the to-be-filled density of each grid through layout density analysis, and sets the redundant metal side length and spacing according to the to-be-filled density, and then according to the set The redundant metal side length and spacing are analyzed by index array for each grid to obtain the array to be filled in each grid, and then the array to be filled in each grid is integrated to obtain the area to be filled, and finally in each grid The inner area to be filled is filled with redundant metal. Since the index array analysis is performed on each grid according to the set redundant metal side length and spacing, and the array to be filled in each grid is confirmed, it can avoid the missing filling caused by the existing array index algorithm when the set filling parameters are inappropriate. In addition, the arrays to be filled in each grid are integrated to obtain the area to be filled in each grid, which effectively improves the processing speed and enables the present invention to quickly complete redundant metal filling without missing filling .

Further, the area to be filled is rectangular or square, and a step of forming a rectangular or square area to be filled is provided, so that the present invention can obtain the actual filling coordinates of redundant metal filling by obtaining the coordinates of the vertex corners of the area to be filled , further increasing the redundant metal filling speed.

Further, the metal lines in each grid are expanded by L0/2, where L0 is the critical size of defects generated by CMP process simulation, so that the method provided by the present invention takes into account the critical size of defects generated by CMP process, and can expand the subsequent semiconductor preparation. The process window, and since it is further determined that the redundant metal filling area is not required, the area to be filled is reduced, and the redundant metal filling speed can be increased.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings that are required in the embodiments. Obviously, the accompanying drawings in the following description are only described in the present invention For some embodiments of the present invention, those skilled in the art can also obtain other drawings according to these drawings.

Fig. 1 is the flowchart of a kind of redundant metal filling method provided according to the present invention;

FIG. 2 is a logic diagram of a layout density analysis of a layout to be filled according to the present invention;

Fig. 3 is a logic diagram of a grid margin analysis technique provided according to the present invention;

FIG. 4 is a schematic diagram of array division and merging provided according to the present invention;

5 is a schematic diagram of redundant metal filling provided according to the present invention;

Fig. 6 is a schematic structural diagram of a redundant metal filling system according to the present invention.

detailed description

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention. Furthermore, the present invention may repeat reference numerals and/or letters in different instances. This repetition is for the purpose of simplicity and clarity and does not in itself indicate a relationship between the various embodiments and/or arrangements discussed.

There are mainly two algorithms for margin analysis techniques of existing redundant metal filling methods, array index algorithm and scan line algorithm. Among them, the array index algorithm has a fast calculation speed, but because the array size is heavily dependent on the redundant metal filling parameter settings, when the set filling parameters are inappropriate, it will lead to missing filling and poor adjustability; the scan line algorithm can solve the above problems, However, it cannot handle the layout of complex geometric structures, and the calculation speed is much lower than that of the array index algorithm.

The method and system of a redundant metal filling method provided by the present invention, because the index array analysis is performed on each grid according to the set redundant metal side length and spacing, and the array to be filled in each grid is confirmed, it can avoid the Inappropriate filling parameters lead to missing filling, and the array to be filled is integrated to obtain the area to be filled in each grid, which effectively improves the processing speed, so that the present invention can quickly complete the redundant metal without missing filling. filling.

In order to better understand the technical solutions and technical effects of the present invention, a detailed description will be made below in conjunction with a flow chart and specific embodiments. The flow chart is shown in FIG. 1 .

Step S01 , perform layout density analysis on the layout to be filled, and confirm the density to be filled of each grid, as shown in FIG. 2 .

In this implementation, the layout density analysis of the layout to be filled, and confirming the density of each grid to be filled includes: Step A1: Using the CMP-DFM tool to obtain the density value D _ij in each grid, the grid size is G×G; Step A2: Determine whether there is a grid to be filled that exceeds the preset density lower limit DI, and if so, obtain the density D1 of the grid to be filled, so that the density of the grid to be filled is adjusted to the preset density Lower limit DI; if not, go directly to step A3; step A3: calculate the maximum density D _max of each grid and its adjacent grid sequentially from one side to the opposite side; step A4: judge whether there is a difference with D _max The grid to be filled whose value is greater than the preset allowable density fluctuation ΔD, if so, update the density D1 of the grid to be filled so that the density of the grid to be filled is adjusted to D _max - ΔD; if not, then Go to Step A5; Step A5: Obtain the filling density of each grid sequentially from one side to the opposite side, and turn to the next line when reaching the opposite side. Among them, the preset density lower limit DI and the preset allowable density fluctuation △D can be determined according to experience or actual use effects, and can also be obtained through simulation; from one side to the opposite side can be from left to right, from right to left, from From top to bottom or from bottom to top, in this embodiment, take from left to right as an example for illustration.

In a specific embodiment, performing layout density analysis on the layout to be filled, and confirming the density to be filled of each grid includes:

Step A1: Set the lower density limit DI and the allowable density fluctuation ΔD;

Step A2: Use the CMP-DFM tool to obtain the density value in each grid. At this time, the layout is divided into several grids with a size of G×G;

Step A3: Determine whether there is a grid exceeding the lower limit of density, if so, add redundant metal to make the grid density reach DI, and obtain the density D1 to be adjusted, such as D1=DI-D _ij , otherwise, directly enter step A4; It should be noted that DI can also be a range, that is, when D1 is within the setting range of DI;

Step A4: traverse from left to right, and calculate the maximum density D _max of each grid and its adjacent grids;

Step A5: Determine whether the difference between the maximum value D _max and the current grid density is greater than ΔD, if it is greater, adjust the current grid density to D _max - ΔD, and update the density to be adjusted D1 = D _max - ΔD - D _ij , Go to step A6, if less than or equal to, no adjustment will be made;

Step A6: Advance the grid calculation from left to right, and turn to the next row when it reaches the rightmost end.

The foregoing acquisition of the density to be filled is merely exemplary, and methods for obtaining the density to be filled in the prior art are all applicable, and are not limited here.

Step S02, setting redundant metal side lengths and spacings for filling each grid according to the density to be filled of each grid.

In this embodiment, the redundant metal side length and spacing can be calculated in real time according to the density to be filled and the geometric constraints of the layout, etc., or it can be a redundant metal template constructed in advance, for example, from the template library of the filled template Select the template whose metal density is closest to the density to be filled. The input of the template can be the density to be filled, etc., and the output is the redundant metal side length and spacing, etc., which are not limited here.

In a specific embodiment, the side length and spacing of the redundant metal are W and S respectively, and the metal density thereof is W ² /(W+S) ² .

Step S03, perform index array analysis on each grid according to the set redundant metal side length and spacing, and confirm the array to be filled in each grid, refer to Fig. 3 to Fig. 4 .

In this embodiment, the index array analysis is performed on each grid according to the set redundant metal side length and spacing, and it is confirmed that the array to be filled in each grid includes: the grid to be filled determined in step A2 and step A4 Divide into arrays with L1 as the side length, where, It is judged whether each array overlaps with the metal line, if yes, the array is marked as an array not to be filled; if not, the array is marked as an array to be filled.

Further, in order to expand the process window of the subsequent semiconductor preparation and reduce the area to be filled, so as to increase the filling speed of redundant metal, the method further includes: before judging whether each array overlaps with the metal line, it is generated by CMP process simulation The critical size L0 of the defect; expand the metal lines in each grid by L0/2; determine whether each array overlaps with the metal lines, if so, mark the array as an array not to be filled; if not, then Marking the array as an array to be filled includes: judging whether each array overlaps with the expanded metal line, if so, marking the array as a non-filling array; if not, marking the array as an array to be filled .

In a specific embodiment, the index array analysis is performed on each grid according to the set redundant metal side length and spacing, and it is confirmed that the array to be filled in each grid includes:

Step B1: CMP simulation to determine the critical size L0 of defects;

Step B2: expand the metal lines in the grid by L0/2;

Step B3: divide the grid to be filled determined in step A3 and step A5 into an array with L1 as the side length, wherein, Represents the rounding down symbol;

Step B4: Determine whether the array intersects with the virtual metal line, if so, mark it as not an array to be filled, otherwise, mark it as an array to be filled.

Step S04, integrate the arrays to be filled in each grid to obtain the area to be filled in each grid, as shown in FIG. 4 and FIG. 5 .

In this implementation, the area to be filled is rectangular or square, and the integration of the arrays to be filled in each grid to obtain the area to be filled in each grid includes: Step C1: from a corner of the grid The array starts to traverse, if the current array is not an array to be filled, it is skipped, otherwise it enters step C2, and the traverse includes: advancing from one side to the opposite side, and turning to the next row or column when advancing to the opposite side; step C2: From one side of the grid, when the array N _{i, j} is the array to be filled, judge whether N _{i+1, j} or N _{i, j+1} is the array to be filled, and if so, set N _ij , N _{i+1, j} , or N _ij , N _{i, j+1} are merged into (i, j) area to be filled M _{i, j} , and continue to judge whether the array N _{i+2, j} or N _{i, j+2} is an array to be filled, By analogy, until N _{i+k, j} or N _{i, j+k} is the non-to-be-filled array or the opposite side of the grid, where i is the number of columns, j is the number of rows, and k is a positive integer; step C3: judge Arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k,} whether there is an array not to be filled, if not, the array N _{i, j +1} to N _{i+k, j+1} or the array N _{i+1, j} to N _{i+1, j+k} are merged into (i, j) area M _{i, j} to be filled, and continue to judge the array N _i, Whether _j+2 to N _{i+k, j+2} or the array N _{i+2, j} to N _{i+2, j+k} are merged into (i, j) area M _{i, j} to be filled, and so on, when Arrays N _{i, j+p} to N _{i+k, j+p} or arrays N _{i+p, j} to N _{i+p, j+k} have arrays not to be filled, if so, do not adjust (i, j) Area to be filled M _{i, j} ; where p is a positive integer; Step C4: Repeat Step C1 to Step C3 until traversing all arrays in the current grid to obtain each area to be filled in the grid; Step C5: Repeat Step C1 to Step C4 Until the area to be filled in each grid is obtained. In this embodiment, the traversal starts from the upper left corner of the array as an example for illustration.

In a specific embodiment, first, step C1: starting from the upper left corner of the grid, traversing from left to right, turning to the next line when it is at the rightmost end, skipping directly if an array is not fillable, otherwise, enter step C2;

Step C2: Array N _{i, j} is the array to be filled, judge whether N _{i+1, j} is the array to be filled, if so, merge N _ij and N _{i+1, j} into M _{i, j} , and continue to judge the array N _{i+2, j} , and so on, until N _{i+k, j} is an unfillable array or the rightmost end of the grid;

Step C3: Assuming that M _{i, j} is a collection of grids between arrays N _{i, j} to N _{i+k, j} , then check whether the arrays N _{i, j+1} to N _{i+k, j+1} are all The array can be filled, if so, add it to the collection M _i,j , and continue to check the arrays N _i,j+2 to N _i+k,j+2 , and so on, until N _i,j+p to N _i There is an unfillable array between _{+k, j+p} , otherwise Mi _{, j} will not be adjusted;

Step C4: Repeat steps C1 to C3 until all arrays in the grid are traversed, and the example in Figure 4 is finally merged into 8 areas to be filled;

Through the above steps, a rectangular or square area to be filled can be formed, which is convenient for subsequent metal redundant filling according to the corner coordinates of the area to be filled.

It should be noted that the above method of integrating the array to be filled is only a preferred method. In addition, it can also be the method of obtaining the area to be filled commonly used in the prior art, and select the metal redundant filling method accordingly in the subsequent steps. , is not limited here.

Step S05 , perform redundant metal filling in the area to be filled in each grid, as shown in FIG. 5 .

In this implementation, the filling of redundant metal in the area to be filled in each grid includes: determining the geometric coordinates of each vertex of the area to be filled; calculating the actual filling coordinates of the redundant metal according to the geometric coordinates of each vertex of the area to be filled; Output the actual filling coordinates of the redundant metal to the layout file to complete the redundant metal filling.

In a specific embodiment, the redundant metal filling in the regions to be filled in each grid includes:

Step D1: Determination of the geometric coordinates of the area to be filled: in the grid G _{r, s} , assume that the area M _{i, j} to be filled is an array set The formula for calculating the coordinates of its four vertices is shown in formula (1):

Among them, r, s are the serial number of the horizontal grid and the serial number of the vertical grid respectively; k, p are the number of horizontal and vertical arrays from the array N _{i, j} respectively (excluding the array N _{i, j} );

Step D2: Calculation of the actual filling coordinates of redundant metals: As shown in Figure 5, the traditional arrangement is taken as an example to illustrate the calculation of the actual filling coordinates (other arrangements can be calculated using a similar method). For the area M _{i, j} to be filled, the coordinate formula of the four vertices of the redundant metal R _{m, n} is shown in formula (2):

Among them, BS is the buffer distance, that is, the distance from the redundant metal array to the metal line; m is the horizontal redundant metal serial number; n is the vertical redundant metal serial number;

Step D3: Output the actual filling coordinates of redundant metals to the layout file to complete the intelligent filling of redundant metals.

It should be noted that the above filling method of redundant metal is only exemplary, and the array to be filled may also be directly filled with redundant metal, which is not limited here.

In the embodiment of the present invention, the method performs layout density analysis on the layout to obtain the density to be filled of each grid, and then sets the redundant metal size etc. according to the density to be filled, and calculates the Perform index array analysis to obtain the array to be filled, which can avoid the phenomenon of missing filling when the filling parameters set by the existing array index algorithm are inappropriate, and then integrate the arrays to be filled in each grid to obtain the array to be filled Finally, redundant metal filling is performed in the area to be filled in each grid, which effectively improves the subsequent processing speed, so that the present invention can quickly complete redundant metal filling without missing filling.

Correspondingly, the present invention also provides a redundant metal filling method system, as shown in FIG. 6, including:

A density confirmation module 601, configured to analyze the layout density of the layout to be filled, and confirm the density to be filled of each grid;

The redundant metal setting module 602 is used to set the redundant metal side length and spacing for filling each grid according to the density to be filled of each grid;

The array confirmation module 603 is used to perform index array analysis on each grid according to the set redundant metal side length and spacing, and confirm the array to be filled in each grid;

An area acquisition module 604, configured to integrate the arrays to be filled in each grid to obtain the area to be filled in each grid;

The filling module 605 is configured to perform redundant metal filling in the area to be filled in each grid.

In practical applications, the density confirmation module 601 may include:

A density value acquisition unit, configured to use the CMP-DFM tool to obtain the density value D _ij in each grid, where the grid size is G×G;

The first judging unit is used to judge whether there is a grid to be filled exceeding the preset density lower limit DI, and if so, obtain the density D1 of the grid to be filled, so that the density of the grid to be filled is adjusted to the preset density Lower limit DI; if not, execute calculation unit;

A calculation unit is used to sequentially calculate the maximum density D _max of each grid and its adjacent grids from one side to the opposite side;

The second judging unit is used to judge whether there is a grid to be filled whose difference with D _max is greater than the preset allowable density fluctuation ΔD, if so, update the density D1 of the grid to be filled, so that the grid to be filled The density of is adjusted to D _max -△D; if not, the acquisition unit is executed;

The acquisition unit is used to obtain the density to be filled of each grid sequentially from one side to the opposite side, and transfer to the next line when the opposite side is reached.

In this embodiment, the system performs index array analysis on each grid according to the set redundant metal side length and spacing, and confirms the array to be filled in each grid. The array confirmation module 603 includes:

The array division unit is used to divide the grid to be filled determined by step A2 and step A4 into arrays with L1 as the side length, wherein, W and S are the initially set redundant metal side length and spacing;

The array marking unit is used for judging whether each array overlaps with the metal line, and if so, marks the array as an array not to be filled; if not, marks the array as an array to be filled.

In order to further increase the redundant metal filling speed and expand the process window of subsequent semiconductor fabrication, the system further includes:

The critical dimension acquisition module 706 is used to simulate the critical dimension L0 of defects generated by the CMP process before judging whether each array overlaps with the metal line;

The metal line expansion module 707 connected to the array confirmation module 603 is used to expand the metal lines in each grid by L0/2;

The array marking unit is specifically used for judging whether each array overlaps with the expanded metal line, and if so, marks the array as an array not to be filled; if not, marks the array as an array to be filled.

In addition, the area acquisition module 604 includes:

The traversal unit is used to start traversing from the array at a vertex of the grid, skip if the current array is not an array to be filled, otherwise execute the area merging unit, the traversal includes: advancing from one side to the opposite side, advancing to the opposite side Go to the next row or column when the side is turned;

The area merging unit is used to determine whether N _{i+1, j} or N _{i, j+1} is an array to be filled from one side of the grid, when the array N _{i, j} is an array to be filled, and if so, N _ij , N _{i+1, j} , or N _ij , N _{i, j+1} are merged into (i, j) area M _{i, j} to be filled, and continue to judge the array N _{i+2, j} or N _{i, j+2} Whether it is an array to be filled, and so on, until N _{i+k, j} or N _{i, j+k} is not an array to be filled or the opposite side of the grid, where i is the number of columns, j is the number of rows, and k is positive integer;

The area adjustment unit is used to determine whether there are arrays not to be filled in arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} , if not, Then merge arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} into (i, j) area M _{i, j} to be filled, And continue to judge whether the arrays N _{i, j+2} to N _{i+k, j+2} or the arrays N _{i+2, j} to N _{i+2, j+k} are merged into (i, j) area M _{i to be filled, j} , and so on, when arrays N _{i, j+p} to N _{i+k, j+p} or arrays N _{i+p, j} to N _{i+p, j+k} have arrays not to be filled, if so, then Do not adjust (i, j) area M _{i, j} to be filled; where p is a positive integer;

The third judging unit is used to judge whether to traverse all the arrays in the current grid, if so, execute the fourth judging unit, otherwise execute the traversing unit;

The fourth judging unit is used for judging whether to traverse all grids, if so, end, if not, execute the traversing unit.

In this embodiment, the area acquisition module 604 can form a rectangular or square area to be filled. In order to improve the efficiency of redundant metal filling, the filling module 605 includes:

The area coordinate determination unit is used to determine the geometric coordinates of each vertex of the area to be filled;

The filling coordinate technical unit is used to calculate the actual filling coordinates of the redundant metal according to the geometric coordinates of each vertex in the area to be filled;

The filling unit is used to output the actual filling coordinates of the redundant metal to the layout file to complete the redundant metal filling.

Of course, the system may further include a storage module (not shown), which may be used to store the analysis results of the index array, etc., such as the array to be filled and related information on the area to be filled. In this way, it is convenient to automatically process the layout to be processed by computer, and store information related to redundant metal filling results and the like.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts in each embodiment can be referred to each other. In particular, for the system part, since it is a system formed according to the method provided by the present invention, the description is relatively simple, and for the related parts, please refer to the description of the method part. The embodiments described above are only illustrative, and can be understood and implemented by those skilled in the art without creative efforts.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Any person familiar with the art, without departing from the scope of the technical solution of the present invention, can use the methods and technical content disclosed above to make many possible changes and modifications to the technical solution of the present invention, or modify it into an equivalent of equivalent change Example. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention, which do not deviate from the technical solution of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims (12)

1. A method for redundant metal filling method, characterized in that, comprising steps: Perform layout density analysis on the layout to be filled to confirm the density to be filled of each grid; Set the redundant metal side length and spacing for filling each grid according to the density to be filled of each grid; Perform index array analysis on each grid according to the set redundant metal side length and spacing, and confirm the array to be filled in each grid; Integrate the arrays to be filled in each grid to obtain the area to be filled in each grid; Redundant metal filling is performed in the area to be filled in each grid. 2. The method according to claim 1, characterized in that, performing layout density analysis on the layout to be filled, and confirming the density to be filled of each grid comprises: Step A1: Using chemical mechanical planarization CMP-design for manufacturability DFM tool to obtain the density value D _ij in each grid, the grid size is G×G; Step A2: Determine whether there is a grid to be filled that exceeds the preset density lower limit DI, and if so, obtain the density D1 of the grid to be filled, so that the density of the grid to be filled is adjusted to the preset density lower limit DI; if If not, go directly to step A3; Step A3: sequentially calculate the maximum density D _max of each grid and its adjacent grids from one side to the opposite side; Step A4: Determine whether there is a grid to be filled whose difference from D _max is greater than the preset allowable density fluctuation ΔD, and if so, update the density D1 of the grid to be filled so that the density of the grid to be filled is adjusted is D _max -△D; if not, go to step A5; Step A5: Obtain the to-be-filled density of each grid sequentially from one side to the opposite side, and turn to the next row when the opposite side is reached. 3. The method according to claim 2, wherein the index array analysis is carried out to each grid according to the set redundant metal side length and spacing, and it is confirmed that the array to be filled in each grid includes: The grid to be filled determined by step A2 and step A4 is divided into arrays with L1 as the side length, wherein, W and S are the initially set redundant metal side length and spacing; It is judged whether each array overlaps with the metal line, if yes, the array is marked as an array not to be filled; if not, the array is marked as an array to be filled. 4. method according to claim 3, is characterized in that, described method also comprises: Before judging whether each array overlaps with the metal line, the critical size L0 of the defect is simulated through the CMP process; Expand the metal lines in each grid by L0/2; Described judging whether each array overlaps with the metal line, if so, then marking the array as an array not to be filled; if not, marking the array as an array to be filled includes: It is judged whether each array overlaps with the expanded metal line, if yes, the array is marked as an array not to be filled; if not, the array is marked as an array to be filled. 5. The method according to claim 3, wherein the area to be filled is a rectangle or a square, and the arrays to be filled in each grid are integrated to obtain the area to be filled in each grid comprising: Step C1: start traversing from the array at one corner of the grid, skip if the current array N _{i, j} is not an array to be filled, otherwise enter step C2, the traversal includes: advancing from one side to the other side, and advancing to Turn to the next row or column when opposite sides; Step C2: From one side of the grid, when the current array N _{i, j} is an array to be filled, judge whether N _{i+1, j} or N _{i, j+1} is an array to be filled, and if so, set N _{i, j} , N _{i+1, j} , or N _{i, j} , N _{i, j+1} are merged into (i, j) area M _{i, j} to be filled, and continue to judge the array N _{i+2, j} or N _{i, j+ 2} Whether it is an array to be filled, and so on, until N _{i+k, j} or N _{i, j+k} is not an array to be filled or the opposite side of the grid, where i is the number of columns, j is the number of rows, and k is positive integer; Step C3: Judging whether there are arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} that are not to be filled, if not, the array N _{i, j+1} to N _{i+k, j+1} or the array N _{i+1, j} to N _{i+1, j+k} are merged into (i, j) area M _{i, j} to be filled, and continue to judge Whether the arrays N _{i, j+2} to N _{i+k, j+2} or the arrays N _{i+2, j} to N _{i+2, j+k} are merged into (i, j) area M _{i, j} to be filled, so that By analogy, when arrays N _{i, j+p} to N _{i+k, j+p} or arrays N _{i+p, j} to N _{i+p, there are non-to-be-filled arrays in j+k} , if so, do not adjust ( i, j) area to be filled M _{i, j} ; wherein p is a positive integer; Step C4: Repeat steps C1 to C3 until traversing all the arrays in the current grid to obtain each area to be filled in the grid; Step C5: Repeat steps C1 to C4 until the area to be filled in each grid is obtained. 6. The method according to claim 5, characterized in that, performing redundant metal filling in the regions to be filled in each grid comprises: Determine the geometric coordinates of each vertex of the area to be filled; Calculate the actual filling coordinates of redundant metal according to the geometric coordinates of each vertex in the area to be filled; Output the actual filling coordinates of the redundant metal to the layout file to complete the redundant metal filling. 7. A system of redundant metal filling methods, comprising: The density confirmation module is used to analyze the layout density of the layout to be filled, and confirm the density to be filled of each grid; A redundant metal setting module is used to set the redundant metal side length and spacing for filling each grid according to the density to be filled of each grid; The array confirmation module is used to analyze the index array of each grid according to the set redundant metal side length and spacing, and confirm the array to be filled in each grid; An area acquisition module, configured to integrate the arrays to be filled in each grid to obtain the area to be filled in each grid; The filling module is used for redundant metal filling in the area to be filled in each grid. 8. The system according to claim 7, wherein the density confirmation module comprises: A density value acquisition unit, configured to use the CMP-DFM tool to obtain the density value D _ij in each grid, where the grid size is G×G; The first judging unit is used to judge whether there is a grid to be filled exceeding the preset density lower limit DI, and if so, obtain the density D1 of the grid to be filled, so that the density of the grid to be filled is adjusted to the preset density Lower limit DI; if not, execute calculation unit; A calculation unit is used to sequentially calculate the maximum density D _max of each grid and its adjacent grids from one side to the opposite side; The second judging unit is used to judge whether there is a grid to be filled whose difference with D _max is greater than the preset allowable density fluctuation ΔD, if so, update the density D1 of the grid to be filled, so that the grid to be filled The density of is adjusted to D _max -△D; if not, the acquisition unit is executed; The acquisition unit is used to obtain the density to be filled of each grid sequentially from one side to the opposite side, and transfer to the next line when the opposite side is reached. 9. The system according to claim 8, wherein the array confirmation module comprises: The array division unit is used to divide the grid to be filled determined by step A2 and step A4 into arrays with L1 as the side length, wherein, W and S are the initially set redundant metal side length and spacing; The array marking unit is used for judging whether each array overlaps with the metal line, and if so, marks the array as an array not to be filled; if not, marks the array as an array to be filled. 10. The system according to claim 8, further comprising: The critical dimension acquisition module is used to simulate the critical dimension L0 of defects generated by the CMP process before judging whether each array overlaps with the metal line; The metal wire expansion module connected with the array confirmation module is used to expand the metal wires in each grid to L0/2; The array marking unit is specifically used for judging whether each array overlaps with the expanded metal line, and if so, marks the array as an array not to be filled; if not, marks the array as an array to be filled. 11. The system according to claim 9, wherein the area acquiring module comprises: The traversal unit is used to start traversing from the array at a vertex of the grid, skip if the current array is not an array to be filled, otherwise execute the area merging unit, the traversal includes: advancing from one side to the opposite side, advancing to the opposite side Go to the next row or column when the side is turned; The area merging unit is used to determine whether N _{i+1, j} or N _{i, j+1} is an array to be filled from one side of the grid, and if the current array N _{i, j} is an array to be filled, if so, N _ij , N _{i+1, j} , or N _ij , N _{i, j+1} are merged into (i, j) area M _{i, j} to be filled, and continue to judge the array N _{i+2, j} or N _{i, j+2} Whether it is an array to be filled, and so on, until N _{i+k, j} or N _{i, j+k} is not an array to be filled or the opposite side of the grid, where i is the number of columns, j is the number of rows, and k is positive integer; The area adjustment unit is used to determine whether there are arrays not to be filled in arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} , if not, Then merge arrays N _{i, j+1} to N _{i+k, j+1} or arrays N _{i+1, j} to N _{i+1, j+k} into (i, j) area M _{i, j} to be filled, And continue to judge whether the arrays N _{i, j+2} to N _{i+k, j+2} or the arrays N _{i+2, j} to N _{i+2, j+k} are merged into (i, j) area M _{i to be filled, j} , and so on, when arrays N _{i, j+p} to N _{i+k, j+p} or arrays N _{i+p, j} to N _{i+p, j+k} have arrays not to be filled, if so, then Do not adjust (i, j) area M _{i, j} to be filled; where p is a positive integer; The third judging unit is used to judge whether to traverse all the arrays in the current grid, if so, execute the fourth judging unit, otherwise execute the traversing unit; The fourth judging unit is used for judging whether to traverse all grids, if so, end, if not, execute the traversing unit. 12. The system of claim 11, wherein the filling module comprises: The area coordinate determination unit is used to determine the geometric coordinates of each vertex of the area to be filled; The filling coordinate technical unit is used to calculate the actual filling coordinates of the redundant metal according to the geometric coordinates of each vertex in the area to be filled; The filling unit is used to output the actual filling coordinates of the redundant metal to the layout file to complete the redundant metal filling.