A MRAM memory array and MRAM
By employing a multi-layered redundant memory area design in the MRAM memory array, the corner failure problem of the memory array caused by the CMP process was solved, the manufacturing yield was improved and the impact of RC delay was reduced, while maintaining a low-cost manufacturing process.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-30
- Publication Date
- 2026-03-20
AI Technical Summary
During the manufacturing process of MRAM memory arrays, failures can occur due to over-polishing defects at the corners of the memory array caused by the CMP process. While existing technologies can protect the internal memory array to some extent by adding dummy arrays, they cannot effectively block over-polishing when the dummy array is too narrow. Furthermore, too many dummy arrays can increase the chip area and affect RC Delay.
The system employs a multi-layered redundant storage area design, including a first redundant storage area and a second redundant storage area. The first redundant storage area is arranged in a strip around the main storage area. Each strip of the second redundant storage area is attached to the corresponding protruding position of the first redundant storage area. The first redundant storage area is set to be narrower and the second redundant storage area is set to be wider to avoid filling the space of adjacent storage areas and reduce the impact of RC Delay.
It improves product manufacturing yield, avoids the impact of RC delay, and does not increase additional costs, simplifying the manufacturing process and reducing technical costs.
Smart Images
Figure CN115730551B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of semiconductor manufacturing technology, and in particular to a MRAM memory array and MRAM. BACKGROUND
[0002] In the design of a large capacity (more than 64M) MRAM (Magnetoresistive Random Access Memory) memory array, the large capacity memory array is composed of multiple small capacity arrays. Between adjacent small capacity arrays, various logic and peripheral metal traces are inserted to reduce chip area, reduce cost, and facilitate wiring. However, in the process of manufacturing MRAM, due to the working principle of CMP (Chemical Mechanical Polishing) itself and the difference in the selection ratio of polishing solution and material, the Remove Rate (RR) of different materials is different, which will form Over Polish defects in the corners of the memory array. Such defects can cause the entire memory array to fail.
[0003] The prior art usually adds multiple rows of Dummy arrays (redundant memory arrays) around the memory array. Such Dummy arrays can protect the internal memory array to some extent, thereby ensuring the manufacturing yield. However, if the Dummy arrays added around the memory array are too narrow, the Dummy arrays cannot safely block the protection of the internal memory array from Over Polish for a polishing liquid with a large RR. If the space between adjacent memory arrays is filled with Dummy arrays, too many Dummy arrays increase the chip area and increase the cost, and also affect the read and write of peripheral metal and logic lines, causing excessive RC Delay (RC delay refers to the signal delay caused by the charging and discharging process of resistance (R) controlling capacitance (C) in an integrated circuit). SUMMARY
[0004] The present application provides a MRAM memory array and MRAM to overcome the problem of corner failure of the memory array caused by the CMP process, improve the yield of product manufacturing, and at the same time ensure that there is no excessive RC Delay.
[0005] In a first aspect, the present application provides a MRAM memory array, which comprises a plurality of spaced memory areas, wherein each memory area comprises a main memory area, a first redundant memory area and a second redundant memory area. The main memory area is provided with main memory cells. The first redundant memory area is in the form of a belt surrounding the main memory area, and is provided with at least one row of redundant memory cells. The second redundant memory area comprises a plurality of belt-shaped redundant memory areas, each of which is fitted at a corresponding protruding position of the first redundant memory area, and is provided with at least one row of redundant memory cells.
[0006] In the above scheme, by dividing the redundant memory area in each memory area into the first redundant memory area and the second redundant memory area, the first redundant memory area is in the form of a belt surrounding the main memory area, and each belt-shaped redundant memory area of the second redundant memory area is fitted at a corresponding protruding position of the first redundant memory area, so that the first redundant memory area can be set narrower, and each belt-shaped redundant memory area of the second redundant memory area can be set wider, thereby overcoming the problem of corner failure of the memory array caused by the CMP process and improving the product manufacturing yield. At the same time, the redundant memory cells do not fill the space between the adjacent two memory areas, which can avoid the wider first redundant memory area affecting the peripheral circuit, thereby ensuring that there is no excessive RC Delay. Moreover, only the mask used in manufacturing the main memory cells and the redundant memory cells needs to be changed, and other processes such as manufacturing MTJ (magnetic tunnel junction) do not need to be adjusted, so that no additional cost is increased, and the technical cost is relatively low.
[0007] In a specific embodiment, the first redundant memory area is provided with 3-7 rows of redundant memory cells, which are arranged in sequence from the position close to the main memory area to the position away from the main memory area. In this way, the first redundant memory area can be set narrower, so as to ensure that there is no excessive RC Delay.
[0008] In a specific embodiment, the main memory area is in the form of a rectangle, the first redundant memory area is in the form of a rectangular belt, and the second redundant memory area comprises four belt-shaped redundant memory areas corresponding to the four corners of the first redundant memory area. Each belt-shaped redundant memory area is in the form of an L shape, so as to be fitted at the corresponding corner position of the first redundant memory area. By setting the main memory area in the form of a rectangle, the circuit structures such as the main memory cells, word lines and bit lines arranged in an array in the main memory area can be facilitated.
[0009] In one specific embodiment, the main storage area includes a plurality of main storage cell arrays arranged in an array, each main storage cell array has a plurality of main storage cells arranged in an array; and any two adjacent main storage cell arrays are separated by a wiring area. In order to set up various logic and metal wiring inside the main storage area.
[0010] In one specific embodiment, the cross-sectional shape of the main storage cell is circular or square, and the cross-sectional shape of the redundant storage cell is circular or square. Thus, the shapes of the main storage cell and the redundant storage cell can be the same or different, which can be flexibly set.
[0011] In one specific embodiment, the arrangement density of the redundant storage cell in the first and second redundant storage areas is the same as the arrangement density of the main storage cell in the main storage cell array. In order to process main storage cells and redundant storage cells with the same density.
[0012] In one specific embodiment, the cross-sectional shape of the main storage cell is circular or square, and the cross-sectional shape of the redundant storage cell in the first redundant storage area is circular or square. The cross-sectional shape of the redundant storage cell in the second redundant storage area is L-shaped; and the plurality of L-shaped redundant storage cells in each strip-shaped redundant storage area are arranged in sequence from inside to outside based on the corresponding corner of the first redundant storage area. In order to simplify the structure of the redundant storage cell located at the corner and reduce the processing difficulty.
[0013] In one specific embodiment, the arrangement density of the redundant storage cell in the first redundant storage area is the same as the arrangement density of the main storage cell in the main storage cell array. The distance between two adjacent redundant storage cells in each strip-shaped redundant storage area is 1-2 times the distance between two adjacent main storage cells in the same row or column of the main storage cell array. In order to ensure that the second redundant storage area located at the corner has sufficient width.
[0014] In one specific embodiment, 10-30 rows of redundant storage cells are arranged in each strip-shaped redundant storage area, and the 10-30 rows of redundant storage cells are arranged in sequence from inside to outside based on the corresponding corner of the first redundant storage area. The lengths of the two edges of each L-shaped strip-shaped redundant storage area are equal, and the lengths of the two edges are 10-30 times the distance between two adjacent main storage cells in the same row or column of the main storage cell array. Ensure that the second redundant storage area located at the corner has appropriate width and length, so as to overcome the storage array corner failure problem caused by the CMP process, improve the yield of product manufacturing, and also reduce the design area of the storage cell and reduce the cost.
[0015] In one specific embodiment, the metal layer occupied by all the main storage units in the main storage area is the same as the metal layer occupied by the redundant storage units in the first and second redundant storage areas. And the material used by all the main storage units in the main storage area is the same as the material used by the redundant storage units in the first and second redundant storage areas. The main storage units and the redundant storage units are made of the same material and are arranged in the same metal layer, so that the main storage units and the redundant storage units can be processed at the same time.
[0016] In a second aspect, the present application also provides a MRAM comprising any of the MRAM storage arrays described above. By dividing the redundant storage area in each storage area into the first redundant storage area and the second redundant storage area, the first redundant storage area is arranged in a belt shape around the main storage area, and each belt-shaped redundant storage area in the second redundant storage area is fitted at the corresponding protruding position of the first redundant storage area. Therefore, the first redundant storage area can be arranged narrower, and each belt-shaped redundant storage area in the second redundant storage area can be arranged wider. Thus, the problem of corner failure of the storage array caused by the CMP process can be overcome, and the manufacturing yield of the product can be improved. At the same time, the redundant storage units do not fill the space between the adjacent two storage areas, so that the wider first redundant storage area does not affect the peripheral circuit, thereby ensuring that there is no excessive RCDelay impact. Moreover, only the mask used in the manufacturing of the main storage units and the redundant storage units needs to be changed, and other processes such as the manufacturing of MTJ do not need to be adjusted, so that no additional cost is increased, and the technical cost is relatively low. BRIEF DESCRIPTION OF DRAWINGS
[0017] Figure 1 A layout diagram of a MRAM storage array provided for an embodiment of the present application;
[0018] Figure 2 A layout diagram of a storage unit in a storage area provided for an embodiment of the present application;
[0019] Figure 3 Another layout diagram of a storage unit in a storage area provided for an embodiment of the present application;
[0020] Figure 4 Another layout diagram of a storage unit in a storage area provided for an embodiment of the present application;
[0021] Figure 5 A cross-sectional view of a MRAM storage array provided for an embodiment of the present application.
[0022] REFERENCE NUMERALS:
[0023] 10 - main storage area
[0024] 11 - wiring area
[0025] 21 - first redundant storage area
[0026] 22 - strip redundant storage area DETAILED DESCRIPTION
[0027] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.
[0028] In order to facilitate the understanding of the MRAM storage array provided by the embodiments of the present application, the application scenario of the MRAM storage array provided by the embodiments of the present application will be described first below. The MRAM storage array is applied to MRAM and serves as a storage medium for storing data of MRAM. The MRAM storage array will be described in detail below with reference to the drawings.
[0029] Reference Figure 1 and Figure 2 The MRAM storage array provided by the embodiments of the present application includes a plurality of spaced storage areas. As shown in Figure 2 , the number of storage areas can be 2, and of course the number of storage areas can also be 3, 4, 5, 10, 12, 16, etc. any value not less than 3. The plurality of storage areas can be arranged in an array, and the two adjacent storage areas are spaced.
[0030] When each storage area is set, with reference to Figure 2 , each storage area includes a main storage area 10, a first redundant storage area 21 and a second redundant storage area. Among them, the shape of the main storage area 10 can be any polygon, and the main storage unit is arranged in the main storage area 10. For example, a rectangular main storage area 10 can be designed, specifically, in some embodiments, the shape of the main storage area 10 can be set to a square; in other embodiments, the shape of the main storage area 10 can be set to a rectangle. Specifically, in some embodiments, the shape of the main storage area 10 can be designed as a square area as shown in Figure 1 ; in other embodiments, the shape of the main storage area 10 can be designed as a rectangular area. With reference to Figure 2The first redundant storage area 21 is in a band shape surrounding the main storage area 10 to set the redundant storage array as the redundant storage area around the main storage area 10. The shape of the first redundant storage area 21 is specifically related to the shape of the main storage area 10 to be in a band shape around the edges of the main storage area 10. For example, when the shape of the main storage area 10 is the above-mentioned rectangular shape, the shape of the first redundant storage area 21 can also be set as a rectangular band to surround the four edges of the rectangular main storage area 10. At least one row of redundant storage units is arranged in the first redundant storage area 21. Specifically, the number of rows of redundant storage units arranged in the first redundant storage area 21 can be any value not less than 1 row, such as 1 row, 2 rows, 3 rows, 4 rows, etc. It needs to be explained that the first redundant storage area 21 is a band-shaped area, and each row of redundant storage units distributed therein is also in a band shape, which surrounds the main storage area 10. Therefore, the number of rows of redundant storage units in the first redundant storage area 21 refers to the number of rows of redundant storage units in the cross-sectional direction of the band shape surrounding the main storage area 10.
[0031] With continued reference to Figure 2, the second redundant storage area includes a plurality of strip-shaped redundant storage areas 22, each of which is fitted at a corresponding protruding position of the first redundant storage area 21. That is, the second redundant storage area is composed of a plurality of strip-shaped redundant storage areas 22, and each of the plurality of strip-shaped redundant storage areas 22 is fitted at a corresponding protruding position of the first redundant storage area 21. The number of strip-shaped redundant storage areas 22 in the second redundant storage area is specifically equal to the number of protrusions at the edge of the first redundant storage area 21, and there is a strip-shaped redundant storage area 22 fitted at each protruding position at the edge of the first redundant storage area 21. At least one row of redundant storage units is arranged in each strip-shaped redundant storage area 22. Specifically, the number of rows of redundant storage units in each strip-shaped redundant storage area 22 can be any value not less than 1 row, such as 1 row, 2 rows, 3 rows, 4 rows, etc. It also needs to be explained that the same row of redundant storage units in each strip-shaped redundant storage area 22 is also arranged along the edge of the corresponding protruding position of the first redundant storage area 21, so the number of rows of redundant storage units in the strip-shaped redundant storage area 22 is also the number of rows of redundant storage units in the cross-sectional direction of the strip-shaped redundant storage area 22. By dividing the redundant storage area in each storage area into the first redundant storage area 21 and the second redundant storage area, the first redundant storage area 21 is in the form of a strip surrounding the main storage area 10, and each strip-shaped redundant storage area 22 in the second redundant storage area is fitted at a corresponding protruding position of the first redundant storage area 21, so that the first redundant storage area 21 can be set relatively narrow, and each strip-shaped redundant storage area 22 in the second redundant storage area can be set relatively wide, thereby being able to overcome the problem of corner failure of the storage array caused by the CMP process and improve the product manufacturing yield. At the same time, the redundant storage units also do not fill the space between adjacent two storage areas, which can avoid the wider first redundant storage area 21 affecting the peripheral circuit, thereby ensuring that there will be no excessive RCDelay impact. And only the mask used in manufacturing the main storage unit and the redundant storage unit needs to be changed, and other processes such as manufacturing MTJ (magnetic tunnel junction) do not need to be adjusted, so as not to increase additional cost, making the technical cost lower.
[0032] In addition, when the main storage area 10 is rectangular and the first redundant storage area 21 is rectangular, the number of the strip-shaped redundant storage areas 22 in the second redundant storage area can be four, and each of the four strip-shaped redundant storage areas 22 corresponds to a corner of the first redundant storage area 21, so that each of the four strip-shaped redundant storage areas 22 is attached to the corresponding corner of the first redundant storage area 21. When each of the strip-shaped redundant storage areas 22 is arranged, the shape of each of the strip-shaped redundant storage areas 22 can be L-shaped, and the two edges of the L-shaped area of each of the strip-shaped redundant storage areas 22 are parallel to and adjacent to the two edges of the corresponding corner of the first redundant storage area 21, so that each of the strip-shaped redundant storage areas 22 is attached to the corresponding corner of the first redundant storage area 21 and protects the main storage unit at the corner of the main storage area 10. When the main storage area 10 is rectangular, the main storage units, word lines, bit lines and other circuit structures arranged in the main storage area 10 are arranged in an array. Of course, when the main storage area 10 is not rectangular but other polygons, the shape of the first redundant storage area 21 and the strip-shaped redundant storage areas 22 in the second redundant storage area are adjusted correspondingly, so that the first redundant storage area 21 surrounds the main storage area 10 in a strip shape, and each of the strip-shaped redundant storage areas 22 is attached to the corresponding protruding position of the first redundant storage area 21.
[0033] When the main storage units are arranged in the main storage area 10, the main storage area 10 can include a plurality of main storage unit arrays arranged in an array, and each of the main storage unit arrays includes a plurality of main storage units arranged in an array. Any two adjacent main storage unit arrays are separated by the wiring area 11. The plurality of main storage units in the main storage area 10 are also divided into a plurality of small storage arrays, and the plurality of small storage arrays are arranged in an array and spaced apart. The space between the two adjacent main storage unit arrays is used as the wiring area 11, so that various logic and metal wires can be arranged inside the main storage area 10. Figure 2 Figure 3 Figure 4 When the main storage units are arranged in the main storage area 10, the main storage area 10 can include a plurality of main storage unit arrays arranged in an array, and each of the main storage unit arrays includes a plurality of main storage units arranged in an array. Any two adjacent main storage unit arrays are separated by the wiring area 11. The plurality of main storage units in the main storage area 10 are also divided into a plurality of small storage arrays, and the plurality of small storage arrays are arranged in an array and spaced apart. The space between the two adjacent main storage unit arrays is used as the wiring area 11, so that various logic and metal wires can be arranged inside the main storage area 10.
[0034] When the number of rows of redundant storage units in the first redundant storage area 21 is determined, the first redundant storage area 21 can include 3-7 rows of redundant storage units arranged from the position close to the main storage area 10 to the position far from the main storage area 10. Specifically, the number of rows of redundant storage units in the first redundant storage area 21 can be 3, 4, 5, 6 or 7, so that the first redundant storage area 21 is relatively narrow and the RC Delay is not too large.
[0035] When the cross-sectional shape of the main storage unit is determined, the cross-sectional shape of the main storage unit can be Figure 2 Figure 3 Figure 4 The cross-sectional shape of the main storage unit can be circular, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the main storage unit are also circular. Of course, the cross-sectional shape of the main storage unit can also be square, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the main storage unit are also square.
[0036] In determining the cross-sectional shape of the redundant storage unit, the cross-sectional shape of the reference layer, the free layer, and other layer structures in the redundant storage unit are also determined. Figure 1 and Figure 2 The redundant storage units in the first and second redundant storage areas 21 and 22 can be set to the same shape to simplify the shape of the redundant storage units and facilitate the processing of the mask. Furthermore, the cross-sectional shape of the redundant storage units in the two redundant storage areas and the cross-sectional shape of the main storage units in the main storage area 10 can be set to the same shape to further simplify the shape of the storage units and facilitate the processing of the mask. For example, the cross-sectional shape of the main storage units in the main storage area 10 and the cross-sectional shape of the redundant storage units in the first and second redundant storage areas 21 and 22 are all circular, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the redundant storage units in the two redundant storage areas are also circular. Of course, the cross-sectional shape of the redundant storage units in the two redundant storage areas and the cross-sectional shape of the main storage units in the main storage area 10 can be set to different shapes, but the cross-sectional shape of the redundant storage units in the two redundant storage areas is the same. For example, the cross-sectional shape of the main storage units in the main storage area 10 can be circular, and the cross-sectional shape of the redundant storage units in the two redundant storage areas can be square. The cross-sectional shape of the redundant storage units in the first redundant storage area 21 and the main storage units in the main storage area 10 can be the same, but the cross-sectional shape of the redundant storage units in the second redundant storage area is different. For example, the cross-sectional shape of the main storage units in the main storage area 10 and the cross-sectional shape of the redundant storage units in the first redundant storage area 21 are both circular, but the cross-sectional shape of the redundant storage units in the second redundant storage area is square. Figure 2 The cross-sectional shape of the main storage unit can be circular, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the main storage unit are also circular. Of course, the cross-sectional shape of the main storage unit can also be square, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the main storage unit are also square. Figure 3 The cross-sectional shape of the main storage unit can be circular, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the main storage unit are also circular. Of course, the cross-sectional shape of the main storage unit can also be square, and the cross-sectional shape of the reference layer, the free layer, and other layer structures in the main storage unit are also square.
[0037] In determining the array arrangement density of the redundant storage units and the main storage units, the arrangement density of the redundant storage units in the first and second redundant storage areas 21 and 22 can be the same as the arrangement density of the main storage units in the main storage unit array. This facilitates the processing of main storage units and redundant storage units with the same density. Of course, the arrangement density of the redundant storage units in the two redundant storage areas can be the same, but not the same as the arrangement density of the main storage units in the main storage area 10. The arrangement density of the storage units in the first redundant storage area 21 and the main storage area 10 can be the same, but not the same as the arrangement density of the redundant storage units in the second redundant storage area.
[0038] In addition, the redundant storage units in the second storage area can be arranged in other shapes other than a circle or a square, for example, the cross-sectional shape of the redundant storage units in the second redundant storage area can be arranged in an L shape. Specifically, the cross-sectional shape of the main storage units can be arranged in a circle or a square, the cross-sectional shape of the redundant storage units in the first redundant storage area 21 can be arranged in a circle or a square, and the cross-sectional shape of the redundant storage units in the second redundant storage area can be arranged in an L shape. When a plurality of rows of redundant storage units are arranged in each strip-shaped redundant storage area 22, the plurality of L-shaped redundant storage units in each strip-shaped redundant storage area 22 can be arranged in a nested manner from the inside to the outside based on the corresponding corner of the first redundant storage area 21, that is, in each pair of adjacent rows of redundant storage units, one row is close to the corresponding corner of the first redundant storage area 21, and the other row is away from the corresponding corner of the first redundant storage area 21. In this way, the structure of the redundant storage units located at the corners is simplified, and the processing difficulty is reduced.
[0039] At this time, the arrangement density of the redundant storage units in the first redundant storage area 21 can be arranged to be the same as the arrangement density of the main storage units in the main storage unit array. However, the distance between two adjacent redundant storage units in each strip-shaped redundant storage area 22 is 1-2 times the distance between two adjacent main storage units in the same row or column of the main storage unit array, that is, the distance between two adjacent L-shaped redundant storage units in the same strip-shaped redundant storage area 22 is 1 times, 1.5 times, 2 times, or any value between 1-2 times the distance between two adjacent main storage units in the same row or column of the main storage unit array, so as to ensure that the second redundant storage area located at the corner has sufficient width.
[0040] When determining the width of the two edges of each L-shaped strip-shaped redundant storage area 22, 10-30 rows of redundant storage units can be arranged in each strip-shaped redundant storage area 22, and the 10-30 rows of redundant storage units are arranged in a nested manner from the inside to the outside based on the corresponding corner of the first redundant storage area 21. That is, the number of rows of redundant storage units arranged in parallel along the cross-sectional direction of each strip-shaped redundant storage area 22 is 10 rows, 15 rows, 20 rows, 25 rows, 30 rows, or any value between 10-30 rows. It should be noted that the number of rows of redundant storage units arranged in each strip-shaped redundant storage area 22 is determined according to the severity of Over Polish and the impact on the final yield. Figures 2 to 4The lengths of the two sides of each L-shaped strip redundant storage area 22 can be set to be equal to provide the same level of protection in both the horizontal and vertical directions. Furthermore, the length of each side of each strip redundant storage area 22 can be 10 to 30 times the distance between two adjacent main storage cells in the same row or column of the main storage cell array. Specifically, the length of each side of the strip redundant storage area 22 can be any value between 10 and 30 times the distance between two adjacent main storage cells in the same row or column of the main storage array. This ensures that the strip redundant storage areas 22 located at the corners have appropriate width and length, thereby overcoming the corner failure problem of the storage array caused by the CMP process, improving product manufacturing yield, and reducing the design area of the storage cells, thus lowering costs.
[0041] refer to Figure 5 The metal layers occupied by all main memory cells in the main memory area 10 can be the same as those occupied by redundant memory cells in the first redundant memory area 21 and the second redundant memory area. That is, the reference layer, free layer, and other layer structures in the main memory area 10 are located in the same metal layer position as the corresponding reference layer, free layer, and other layer structures of the redundant memory cells in the two redundant memory areas, so that the redundant memory cells and metal layers can be processed together in the same process. Figure 5 In the diagram, Mn refers to the nth metal layer, Mn+1 refers to the (n+1)th metal layer, and Via refers to a via connecting different metal layers. The bottom electrode layer, reference layer, free layer, and top electrode layer in both the main memory cell and the redundant memory cell can be placed on the same metal layer. Furthermore, the material used for all main memory cells in the main memory region 10 can be the same as the material used for the redundant memory cells in the first redundant memory region 21 and the second redundant memory region. By using the same material and placing the main memory cells and redundant memory cells on the same metal layer, it is easier to fabricate both main memory cells and redundant memory cells simultaneously.
[0042] By dividing the redundant storage area in each storage area into the first redundant storage area 21 and the second redundant storage area, the first redundant storage area 21 is in the form of a belt surrounding the main storage area 10, and each belt-shaped redundant storage area 22 in the second redundant storage area is buckled at the corresponding protruding position of the first redundant storage area 21, so that the first redundant storage area 21 can be set narrower, and each belt-shaped redundant storage area 22 in the second redundant storage area can be set wider, so that the problem of corner failure of the storage array caused by the CMP process can be overcome, and the product manufacturing yield can be improved. At the same time, the redundant storage unit also does not fill the space between the adjacent two storage areas, which can avoid the influence of the wider first redundant storage area 21 on the peripheral circuit, thereby ensuring that there will be no excessive RC Delay influence. And only the mask used in the manufacture of the main storage unit and the redundant storage unit needs to be changed, and other processes such as the manufacture of MTJ do not need to be adjusted, so that no additional cost is increased, and the technical cost is relatively low.
[0043] In addition, the embodiment of the present application also provides an MRAM, which refers to Figure 1 The MRAM includes any one of the MRAM storage arrays described above. By dividing the redundant storage area in each storage area into the first redundant storage area 21 and the second redundant storage area, the first redundant storage area 21 is in the form of a belt surrounding the main storage area 10, and each belt-shaped redundant storage area 22 in the second redundant storage area is buckled at the corresponding protruding position of the first redundant storage area 21, so that the first redundant storage area 21 can be set narrower, and each belt-shaped redundant storage area 22 in the second redundant storage area can be set wider, so that the problem of corner failure of the storage array caused by the CMP process can be overcome, and the product manufacturing yield can be improved. At the same time, the redundant storage unit also does not fill the space between the adjacent two storage areas, which can avoid the influence of the wider first redundant storage area 21 on the peripheral circuit, thereby ensuring that there will be no excessive RC Delay influence. And only the mask used in the manufacture of the main storage unit and the redundant storage unit needs to be changed, and other processes such as the manufacture of MTJ do not need to be adjusted, so that no additional cost is increased, and the technical cost is relatively low.
[0044] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. An MRAM storage array, characterized in that, It includes multiple spaced storage areas, each of which includes: The main storage area, wherein a main storage unit is provided in the main storage area; The first redundant storage area surrounds the main storage area in a strip shape, and at least one row of redundant storage units is provided in the first redundant storage area. The second redundant storage area includes multiple strip-shaped redundant storage areas, each strip-shaped redundant storage area being snapped onto the corresponding protruding position of the first redundant storage area; and each strip-shaped redundant storage area is provided with at least one row of redundant storage units. The main storage area is rectangular in shape; The first redundant storage area is rectangular in shape; The second redundant storage area includes four strip-shaped redundant storage areas, each corresponding to one of the four corners of the first redundant storage area; each strip-shaped redundant storage area is L-shaped, so as to fit into the corresponding corner position of the first redundant storage area. The cross-sectional shape of the main storage unit is circular or square, and the cross-sectional shape of the redundant storage units in the first redundant storage area is circular or square. The cross-sectional shape of the redundant storage cells in the second redundant storage area is L-shaped; and the multiple L-shaped redundant storage cells in each strip redundant storage area are arranged sequentially from the inside to the outside based on the corresponding angle of the first redundant storage area.
2. The MRAM storage array as described in claim 1, characterized in that, The first redundant storage area is provided with 3 to 7 rows of redundant storage units, and the 3 to 7 rows of redundant storage units are arranged sequentially from the direction closest to the main storage area to the direction furthest from the main storage area.
3. The MRAM storage array as described in claim 1, characterized in that, The main storage area includes multiple main storage cell arrays arranged in an array, and each main storage cell array has multiple main storage cells arranged in an array. Furthermore, any two adjacent main memory cell arrays are separated by wiring.
4. The storage array as described in claim 3, characterized in that, The main storage unit has a circular or square cross-sectional shape, and the redundant storage unit has a circular or square cross-sectional shape.
5. The MRAM storage array as described in claim 4, characterized in that, The arrangement density of redundant storage cells in the first and second redundant storage areas is the same as the arrangement density of main storage cells in the main storage cell array.
6. The MRAM storage array as described in claim 1, characterized in that, The arrangement density of redundant storage cells in the first redundant storage area is the same as the arrangement density of main storage cells in the main storage cell array. The spacing between two adjacent redundant storage cells in each strip redundant storage area is 1 to 2 times the spacing between two adjacent main storage cells in the same row or column of the main storage cell array.
7. The MRAM storage array as described in any one of claims 4 to 6, characterized in that, Each strip of redundant storage area is provided with 10 to 30 rows of redundant storage units, and the 10 to 30 rows of redundant storage units are arranged from the inside to the outside based on the corresponding corner of the first redundant storage area; Each L-shaped strip of redundant storage area has two sides of equal length, and the length of each side is 10 to 30 times the distance between two adjacent main storage cells in the same row or column of the main storage cell array.
8. The MRAM storage array as described in claim 1, characterized in that, The metal layer occupied by all main storage cells in the main storage area is the same as the metal layer occupied by redundant storage cells in the first redundant storage area and the second redundant storage area. Furthermore, the materials used by all main storage units in the main storage area are the same as those used by the redundant storage units in the first redundant storage area and the second redundant storage area.
9. An MRAM, characterized in that, Including the MRAM storage array as described in any one of claims 1 to 8.
Citation Information
Patent Citations
Apparatus, system, and method for allocating storage
CN102598019A
MRAM memory array
CN111739567A