Stacked via rivet of chip hot spot

Abstract

A structure is described that includes a plurality of dielectric regions. The structure may include a rivet cell. The rivet cell may include a set of stacked vias. The rivet cell may extend through a stress hotspot of the structure. A length of the rivet cell may penetrate at least one of the plurality of dielectric regions. The rivet cell may be one of a plurality of rivet cells inserted at the stress hotspot. The stress hotspot may be one of a plurality of stress hotspots present throughout the structure. A length of the rivet cell may be based on a model of a relationship between a length of the rivet cell and an energy release rate of the structure. The rivet cell may penetrate an interface between a first dielectric region and a second dielectric region having different dielectric constants.

Description

Technical Field

[0001] This application relates to a structure including a partial stack of vias penetrating one or more dielectric layers of the structure.

Background Art

[0002] Semiconductor devices can include structures (e.g., back-end-of-line (BEOL) structures) that include multiple dielectric regions or layers having various values of dielectric constant k. The interface between dielectric layers of different k can crack or delaminate due to various factors that cause crack propagation. The forces causing this include differences in thermal expansion (or contraction) between various materials in the package or chip structure, residual stress in the materials, etc. In addition to the forces caused by the chip or package, a thermal expansion mismatch can occur between the metal and dielectric in the BEOL structure of the chip. The operation of a semiconductor device with cracks or delamination at the interface can be vulnerable to defects. In some examples, a crack can cause an open circuit between different BEOL levels (e.g., between one of the metal wirings and a via), rendering the entire chip inoperable. In some examples, a crack can also increase the resistance of the current path formed in the BEOL structure, and the increased resistance can generate undesirable heat within the semiconductor device.

[0003] Some conventional solutions for dealing with cracks and delamination may include, for example, minimizing stress by adjusting the mismatch in the coefficient of thermal expansion (CTE) between materials, or changing the deposition conditions to vary the residual stress in the materials. Other process techniques include improving the interfacial properties between various materials such that the tendency for cracks to occur in the materials is reduced. However, the stress within the structure is not always completely removed, and there may still be a possibility that cracks will propagate. Other conventional solutions may include, for example, minimizing the number of via-to-via interfaces within the structure to reduce the likelihood of cracks or delamination or both. However, the possibility of cracks still exists. Another conventional solution may include, for example, inserting bond pads or connectors between dielectric regions or between layers, but the locations at which these bond pads and connectors are inserted are not specified. SUMMARY OF THE INVENTION

[0004] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell may extend through a stress hot spot of the structure. The length of the rivet cell may be able to penetrate at least one of the plurality of dielectric regions.

[0005] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell may extend through a stress hot spot of the structure. The length of the rivet cell may be able to penetrate at least one of the plurality of dielectric regions. The rivet cell may be a first rivet cell disposed at a first lateral distance from a first corner of the structure. The structure may further include a second rivet cell disposed at a second lateral distance from a second corner of the structure.

[0006] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can be one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells can be based on the size of the stress hot spot.

[0007] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The stress hot spot can be one of a plurality of stress hot spots present throughout the structure, and each stress hot spot can include a respective set of rivet cells.

[0008] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The length of the rivet cell can be based on a model of the relationship between the length of the rivet cell and the energy release rate of the structure.

[0009] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The structure can be a back-end-of-line (BEOL) structure of a semiconductor device.

[0010] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can penetrate an interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions. The first dielectric region and the second dielectric region may have different dielectric constants.

[0011] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can penetrate an interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions. The first dielectric region and the second dielectric region may have different dielectric constants. The first dielectric region may include a hard dielectric material, and the second dielectric region may include a soft dielectric material.

[0012] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of a stress hot spot in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of the plurality of dielectric regions of the structure.

[0013] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of a stress hot spot in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of the plurality of dielectric regions of the structure. The rivet cell data may further indicate the insertion positions of a plurality of rivet cells, and the plurality of insertion positions are within the stress hot spot.

[0014] In some examples, a method for determining the position of rivet cells in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The modeling of the relationship may be performed based on the prediction of anomalies at the interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell may penetrate at least a portion of the first dielectric region and at least a portion of the second dielectric region. The energy release rate of the structure may be based on the position of the predicted anomaly.

[0015] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. The predicted anomaly may be a crack at the interface between a first dielectric region and a second dielectric region.

[0016] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. The first dielectric region and the second dielectric region may have different values of dielectric constant.

[0017] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The stress hot spot may be one of a plurality of stress hot spots present throughout the structure, and each stress hot spot may include a respective set of rivet cells.

[0018] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The rivet cell may be one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells may be based on the size of the stress hot spot.

[0019] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. Identifying the optimal length of the rivet cell may include identifying the length of the rivet cell at which the energy release rate is lowest.

[0020] In some examples, a method for determining the position of rivet cells in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cells and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cells based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cells. The insertion position may be based on the position of the stress hot spots. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include rivet cells at the insertion position. The rivet cells may extend through the stress hot spots of the structure and may penetrate at least one of the plurality of dielectric regions of the structure. The method may further include generating a three-dimensional (3D) model of the structure. The method may further include executing instructions to perform a stress analysis on the 3D model. The method may further include identifying the position of the stress hot spots based on the output of the stress analysis.

[0021] In some examples, a computer program product for determining the position of rivet cells in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of the device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of the plurality of dielectric regions of the structure.

[0022] In some examples, a computer program product for determining the position of rivet cells in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the present structure according to the rivet cell data. The present structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the present structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the present structure. The modeling of the relationship may be performed based on the prediction of an anomaly at an interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell may penetrate at least a part of the first dielectric region and at least a part of the second dielectric region. The energy release rate of the present structure may be based on the position of the predicted anomaly.

[0023] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product can include a computer-readable storage medium having program instructions recorded therein. The program instructions can be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions can be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions can be executable by a processing element of the device to cause the device to identify the optimal length of the rivet cell based on the modeled relationship. The program instructions can be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position can be based on the position of the stress hot spot. The program instructions can be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure can be constructed to include a rivet cell at the insertion position. The rivet cell can extend through the stress hot spot of the structure and can penetrate at least one of a plurality of dielectric regions of the structure. The modeling of the relationship can be performed based on the prediction of anomalies at the interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell can penetrate at least a portion of the first dielectric region and at least a portion of the second dielectric region. The energy release rate of the structure can be based on the position of the predicted anomaly. The predicted anomaly can be a crack at the interface between the first dielectric region and the second dielectric region.

[0024] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The modeling of the relationship may be performed based on a prediction of an anomaly at an interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell may penetrate at least a portion of the first dielectric region and at least a portion of the second dielectric region. The energy release rate of the structure may be based on the position of the predicted anomaly. The first dielectric region and the second dielectric region have different dielectric constant values.

[0025] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The rivet cell may be one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells may be based on the size of the stress hot spot.

[0026] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of the device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify the optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. The stress hot spot may be one of a plurality of stress hot spots present throughout the structure, and each stress hot spot may include a respective set of rivet cells.

[0027] In some examples, a computer program product for determining the position of rivet cells in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify the length of the rivet cell at which the energy release rate is the lowest to identify the optimal length of the rivet cell.

[0028] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The program instructions may be executable by a processing element of the device to generate a three-dimensional (3D) model of the structure. The program instructions may be executable by a processing element of the device to execute instructions to perform a stress analysis on the 3D model. The program instructions may be executable by a processing element of the device to identify the position of the stress hot spot from the output of the stress analysis.

Brief Description of the Drawings

[0029]

Figure 1A

Figure 1B

Figure 2

Figure 3A

Figure 3B

Figure 4

Figure 5

Figure 6

Figure 7

DETAILED DESCRIPTION OF THE INVENTION

[0030] Next, the present application will be described in more detail by referring to the following explanations and drawings attached to the present application. Note that the drawings of the present application are provided for illustrative purposes only, and thus, it should be noted that the drawings are not drawn to scale. It should also be noted that similar elements and corresponding elements are referred to by the same reference numerals.

[0031] In the following description, numerous specific details are set forth, such as specific structures, components, materials, dimensions, processing steps, and techniques, in order to provide an understanding of various embodiments of the present application. However, it will be understood by those skilled in the art that the various embodiments of the present application may be practiced without these specific details. In other instances, well-known structures or processing steps are not described in detail so as not to obscure the present application.

[0032] When an element is referred to as being "on" or "over" another element as a layer, region, or substrate, it will be understood that it may be directly on the other element or intervening elements may be present. On the other hand, when an element is said to be "directly on" or "directly over" another element, no intervening elements are present. When an element is said to be "beneath" or "under" another element, it will be understood that it may be directly beneath or under the other element or intervening elements may be present. On the other hand, when an element is said to be "directly beneath" or "directly under" another element, no intervening elements are there.

[0033] More specifically, as described below, the structures described herein include one or more specially designed partial stack via sets placed in the hot spot regions. Further, the methods and systems described herein can identify the regions and exact locations of the partial stack via sets by using various modeling techniques while also determining how many of these multiple partial via sets can be placed. The partial stack via sets may be referred to as "rivet cells" and can be fixed, in particular, through soft dielectrics and may be intended to function as anchors for multiple dielectric layers. Further, since the rivet cells are inside the metal stack and can extend through multiple metal levels under bond pads or C4 (Controlled Collapsed Chip Connection) structures, they do not need to be constrained to contact the bond pads. Further, a number of rivet cells can be used in various regions of the chip and may not need to be treated as a network and may not need to reach bond pads or device contacts. The stress hot spot regions near the corners of the chip can be modeled so that rivet cells can be inserted into the hot spot regions before cracks occur, thereby providing support. Further, the size of the rivet cells is determined based on a model of the relationship between the size of the rivet cells and the possible reactions of the structure to environmental changes (e.g., thermal expansion and contraction, stress, force, etc. with respect to temperature changes).

[0034] In some examples, a structure including rivet cells is generally described. The structure may include multiple dielectric regions. The rivet cells may include a set of stack vias. The rivet cells can extend through the stress hot spots of the structure. The length of the rivet cells can penetrate at least one of the multiple dielectric regions. The rivet cells in the structure can prevent the propagation of cracks that may exist at the interface between a pair of dielectric regions of at least one of the dielectric regions.

[0035] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stack vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can be the first rivet cell disposed at a first lateral distance from a first corner of the structure. The structure may further include a second rivet cell disposed at a second lateral distance from a second corner of the structure. The first and second rivet cells of the structure can prevent propagation of cracks that may exist across a plurality of stress hot spots of the structure.

[0036] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stack vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can be one of a plurality of rivet cells inserted into a stress hot spot, and the number of rivet cells can be based on the size of the stress hot spot. If there are that number of rivet cells in the stress hot spot, propagation of cracks that may exist in that stress hot spot of the structure, or across a plurality of stress hot spots, or both, can be prevented.

[0037] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell may extend through a stress hot spot of the structure. The length of the rivet cell may penetrate at least one of the plurality of dielectric regions. The stress hot spot may be one of a plurality of stress hot spots present throughout the structure, and each stress hot spot may include a respective set of rivet cells. The rivet cells of the structure can prevent the propagation of cracks that may exist across a plurality of stress hot spots of the structure.

[0038] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell may extend through a stress hot spot of the structure. The length of the rivet cell may penetrate at least one of the plurality of dielectric regions. The length of the rivet cell may be based on a model of the relationship between the length of the rivet cell and the energy release rate of the structure. The model of the relationship can provide an optimal size of the rivet cell in the structure and an optimal position of the rivet cell.

[0039] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell may extend through a stress hot spot of the structure. The length of the rivet cell may penetrate at least one of the plurality of dielectric regions. The structure may be a back-end-of-line (BEOL) structure of a semiconductor device. The rivet cells in the structure can prevent the propagation of cracks that may exist in the BEOL structure.

[0040] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can penetrate an interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions. The first dielectric region and the second dielectric region can have different dielectric constants. The rivet cell in the structure can prevent the propagation of cracks that may exist at the interface between the first dielectric region and the second dielectric region.

[0041] In some examples, a structure including a rivet cell is generally described. The structure may include a plurality of dielectric regions. The rivet cell may include a set of stacked vias. The rivet cell can extend through a stress hot spot of the structure. The length of the rivet cell can penetrate at least one of the plurality of dielectric regions. The rivet cell can penetrate an interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions. The first dielectric region and the second dielectric region can have different dielectric constants. The first dielectric region may include a hard dielectric material, and the second dielectric region may include a soft dielectric material. The rivet cell in the structure can prevent the propagation of cracks that may exist at the interface between the hard dielectric region and the soft dielectric region.

[0042] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. By inserting a rivet cell into the structure, propagation of cracks that may exist at the interface between a pair of dielectric regions of at least one of the dielectric regions can be prevented.

[0043] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of the plurality of dielectric regions of the structure. The rivet cell data may further indicate a plurality of insertion positions of a plurality of rivet cells, and the plurality of insertion positions are within the stress hot spot. The presence of the rivet cell at the plurality of insertion positions can prevent the propagation of cracks that may exist across the plurality of stress hot spots of the structure.

[0044] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of a stress hot spot in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The modeling of the relationship may be performed based on predicting an anomaly at an interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When a rivet cell is inserted, the rivet cell may be able to penetrate at least a portion of the first dielectric region and at least a portion of the second dielectric region. The energy release rate of the structure may be based on the position of the predicted anomaly. The model of the relationship may provide an optimal size and an optimal position of the rivet cell in the structure.

[0045] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The predicted anomaly is a crack at the interface between the first dielectric region and the second dielectric region. The rivet cell of the structure can prevent crack propagation and delamination.

[0046] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The first dielectric region and the second dielectric region may have different dielectric constant values. The rivet cell in the structure may be capable of preventing the propagation of cracks that may exist at an interface between the first dielectric region and the second dielectric region.

[0047] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of a stress hot spot in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The stress hot spot may be one of a plurality of stress hot spots present throughout the structure, and each stress hot spot may include a respective set of rivet cells. The rivet cells of the structure may be capable of preventing the propagation of cracks that may exist across a plurality of stress hot spots of the structure.

[0048] In some examples, a method for determining the position of rivet cells in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicative of the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of the plurality of dielectric regions of the structure. The rivet cell may be one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells may be based on the size of the stress hot spot. If there are that number of rivet cells in the stress hot spot, propagation of cracks that may exist in the structure, in that stress hot spot, or across multiple stress hot spots, or both, can be prevented.

[0049] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. Identifying the optimal length of the rivet cell may include identifying the length of the rivet cell at which the energy release rate is lowest. The model of the relationship may provide an optimal size of the rivet cell in the structure and an optimal position of the rivet cell.

[0050] In some examples, a method for determining the position of a rivet cell in a structure is generally described. The method may include modeling the position of stress hot spots in the structure. The method may further include modeling the relationship between the length of the rivet cell and the energy release rate of the structure. The method may further include identifying an optimal length of the rivet cell based on the modeled relationship. The method may further include generating rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The method may further include transmitting the rivet cell data to a device to instruct the device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the structure and may penetrate at least one of the plurality of dielectric regions of the structure. The method may further include generating a three-dimensional (3D) model of the structure. The method may further include executing an instruction to perform a stress analysis on the 3D model. The method may further include identifying the position of the stress hot spot based on the output of the stress analysis. This stress analysis can provide the position of the stress hot spot that can be used to identify the optimal position of the rivet cell in the structure.

[0051] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. The rivet cell in the structure may be capable of preventing the propagation of cracks that may exist at an interface between a first dielectric region and a second dielectric region.

[0052] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spots. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The modeling of the relationship may be performed based on a prediction of an anomaly at an interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell may penetrate at least a portion of the first dielectric region and at least a portion of the second dielectric region. The energy release rate of the structure may be based on the position of the predicted anomaly. The model of the relationship may provide an optimal size and an optimal position of the rivet cell in the structure.

[0053] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of the device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify the optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to transmit the rivet cell data to another device and instruct the other device to construct the present structure according to the rivet cell data. The present structure may be constructed to include a rivet cell at the insertion position. The rivet cell can extend through the stress hot spot of the present structure and can penetrate at least one of the plurality of dielectric regions of the present structure. The modeling of the relationship may be performed based on the prediction of anomalies at the interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell can penetrate at least a part of the first dielectric region and at least a part of the second dielectric region. The energy release rate of the present structure may be based on the position of the predicted anomaly. The predicted anomaly may be a crack at the interface between the first dielectric region and the second dielectric region. The rivet cell of the present structure can prevent crack propagation and delamination.

[0054] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include the rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the structure. The modeling of the relationship may be performed based on a prediction of an anomaly at an interface between a first dielectric region and a second dielectric region of the plurality of dielectric regions at the position of the stress hot spot. When the rivet cell is inserted, the rivet cell may penetrate at least a portion of the first dielectric region and at least a portion of the second dielectric region. The energy release rate of the structure may be based on the position of the predicted anomaly. The first dielectric region and the second dielectric region have different dielectric constant values. The rivet cell in the structure may prevent the propagation of cracks that may exist at the interface between the first dielectric region and the second dielectric region.

[0055] In some examples, a computer program product for determining the position of rivet cells in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify the optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the present structure according to the rivet cell data. The present structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through the stress hot spot of the present structure and may penetrate at least one dielectric region of a plurality of dielectric regions of the present structure. The rivet cell may be one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells may be based on the size of the stress hot spot. If there are that number of rivet cells in the stress hot spot, propagation of cracks that may exist in the present structure within the stress hot spot, across multiple stress hot spots, or in both forms can be prevented.

[0056] In some examples, a computer program product for determining the position of rivet cells in a structure is generally described. The computer program product can include a computer-readable storage medium having program instructions recorded therein. The program instructions can be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions can be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions can be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions can be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position can be based on the position of the stress hot spot. The program instructions can be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the present structure according to the rivet cell data. The present structure can be constructed to include a rivet cell at the insertion position. The rivet cell can extend through a stress hot spot of the present structure and can penetrate at least one dielectric region of a plurality of dielectric regions of the present structure. The stress hot spot can be one of a plurality of stress hot spots present throughout the present structure, and each stress hot spot can include a respective set of rivet cells. The rivet cells of the present structure can prevent the propagation of cracks that can exist across a plurality of stress hot spots of the structure.

[0057] In some examples, a computer program product for determining the position of rivet cells in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating the insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to send the rivet cell data to another device and instruct the other device to construct the structure according to the rivet cell data. The structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the structure and may penetrate at least one of a plurality of dielectric regions of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify the length of the rivet cell at which the energy release rate is lowest to identify the optimal length of the rivet cell. The model of the relationship can provide the optimal size of the rivet cell in the structure and the optimal position of the rivet cell.

[0058] In some examples, a computer program product for determining the position of a rivet cell in a structure is generally described. The computer program product may include a computer-readable storage medium having program instructions recorded therein. The program instructions may be executable by a processing element of a device to cause the device to model the position of stress hot spots in the structure. The program instructions may be executable by a processing element of the device to cause the device to model the relationship between the length of the rivet cell and the energy release rate of the structure. The program instructions may be executable by a processing element of the device to cause the device to identify an optimal length of the rivet cell based on the modeled relationship. The program instructions may be executable by a processing element of the device to cause the device to generate rivet cell data indicating an insertion position of the rivet cell. The insertion position may be based on the position of the stress hot spot. The program instructions may be executable by a processing element of the device to cause the device to transmit the rivet cell data to another device and instruct the other device to construct the present structure according to the rivet cell data. The present structure may be constructed to include a rivet cell at the insertion position. The rivet cell may extend through a stress hot spot of the present structure and may penetrate at least one of a plurality of dielectric regions of the present structure. The program instructions may be executable by a processing element of the device to generate a three-dimensional (3D) model of the structure. The program instructions may be executable by a processing element of the device to execute instructions to perform a stress analysis on the 3D model. The program instructions may be executable by a processing element of the device to identify the position of the stress hot spot from the output of the stress analysis. This stress analysis may provide the position of the stress hot spot that can be used to identify the optimal position of the rivet cell in the present structure.

[0059] FIG. 1A is a cross-sectional view of an exemplary structure including a rivet cell in one embodiment. The structure 102 shown in FIG. 1A can be, for example, part of a back-end-of-line (BEOL) structure of a semiconductor device 100. The semiconductor device 100 can include additional layers or structures such as a middle-of-the-line (MOL) structure and a front-end-of-line (FEOL) structure, but these MOL and FEOL structures are not shown in the drawings of this application. The structure 102 can include a plurality of dielectric regions 101a - 101f having different values of k, and at least one soft dielectric material 101d sandwiched between two hard dielectric materials 101e, 101c. The soft dielectric is an ultra low-k (ULK) material having a dielectric constant ranging from about 1.5 to about 3.5, or between about 2.0 and about 2.5. Exemplary ULK dielectrics can include porous inorganic materials such as silicon-containing materials such as compositions of Si, C, O, and H including (SiCOH), also known as C doped oxide (CDO) or organosilicate glass (OSG). Other low-k materials can also include SilK (a trademark of Dow), a low-k organic polymer. Non-porous ULK materials can also be used in this application. The hard dielectric can have a dielectric constant greater than that of the ULK material. Examples of hard dielectrics can include, for example, silicon dioxide, silsesquioxane, C doped oxide (i.e., organosilicate) containing atoms of Si, C, O, and H, thermosetting polyarylene ether, or multilayers thereof. The term "polyarylene" is used in this application to represent an aryl moiety or an inert substituted aryl moiety linked together by bonds, fused rings, or inert linking groups such as oxygen, sulfur, sulfone, sulfoxide, carbonyl, etc.

[0060] Structure 102 may include one or more sets 104 of stacked vias and metal that can be used as conductive paths spanning between dielectric regions 101a - 101f. The six dielectric regions (101a - 101f) shown in this figure are merely examples. Such dielectric regions with different dielectric thicknesses along with different dielectric constants may exist more or less. Structure 102 may further include at least one rivet cell such as rivet cell 110, and rivet cell 110 may include a set of stacked vias and metal that penetrate one or more layers of the dielectric regions of structure 102. This rivet cell does not necessarily have to be in an active state and does not have to be a conductive path for the operation of a circuit or chip. Rivet cell 110 may be disposed in structure 102 to prevent cracks and delamination at the interfaces of the dielectric regions of structure 102. For example, the rivet cell 110 shown in FIG. 1A may be disposed to prevent cracks that may potentially occur at interface 106 between dielectric regions 101d and 101e. In one example, the position of rivet cell 110 in structure 102 can be estimated based on 1) the evaluation of one or more hot spot regions such as stress hot spot 105 (or "hot spot 105") detected on semiconductor device 100, and 2) a model that describes the relationship between the length of rivet cell 110 (e.g., number of stacked vias) and the energy release rate (ERR) of structure 102. In one exemplary embodiment, multiple rivet cells may be disposed within a large continuous hot spot region. In one example, rivet cell 110 may be composed of a conductive metal or a conductive metal alloy such as copper (Cu), aluminum (Al), tungsten (W), or an alloy thereof, such as a Cu - Al alloy. In one example, the metal and via stack of rivet cell 110 may also include a liner material such as TaN or TiN that may be part of a typical BEOL construction process for each metal level and via level. Structure 102 may be formed using techniques well known to those skilled in the art.For example, the various dielectric regions can be formed using deposition processes such as CVD (chemical vapor deposition), PECVD (plasma-enhanced chemical vapor deposition), spin coating, etc. The rivet cell 110 and the stack via can be formed using a damascene process.

[0061] In the example shown in FIG. 1, the rivet cell 110 can extend through the stress hot spot 105, span from a part of the dielectric region 101c, penetrate through the entire thickness of the dielectric regions 101d and 101e, and can stop at a part of the dielectric region 101f. Further, in some examples, the structure 102 can include a plurality of rivet cells. For example, the rivet cell 110 can be disposed at a lateral distance L1 from the corner 107 of the structure 102, and another rivet cell 111 can be disposed at a lateral distance L2 from the corner 108 of the structure 102. In one example, the rivet cells can be disposed in proximity to each corner of the structure 102. The lateral distances L1 and L2 can be the same or different, and each of the lateral distances can be based on the position of the stress hot spot of the semiconductor device 100. Further, each rivet cell can have a different length and size. For example, the rivet cell 110 can be longer (e.g., having more stack vias) than the rivet cell 111. Further, each rivet cell can extend through the various dielectric regions to reduce the possibility of cracks and delamination occurring at separate interfaces between the dielectric regions. For example, the rivet cell 110 can be disposed to prevent cracks that may potentially occur at the interface 106 between the dielectric regions 101d and 101e, and the rivet cell 111 can be disposed to prevent cracks that may potentially occur at another interface between the dielectric regions 101e and 101f.

[0062] Figure 1B is a cross-sectional view of the structure 102 of Figure 1A with additional rivet cells added, in one embodiment. In the example shown in Figure 1B, a hot spot region (or "hot spot") 150 can be present near or adjacent to the corner 108 of the structure 102. The hot spot 150 can have a relatively large width (e.g., horizontally, side to side) compared to the width of a single rivet cell (e.g., rivet cell 110). In an exemplary embodiment, the width of the hot spot 150 can span an area on the order of about 3 millimeters (mm) × 3 mm (e.g., measured horizontally, side to side), and the width of the rivet cell can be a relatively small size such that it spans an area on the order of about 0.5 micrometers (μm) × 0.5 μm. In one example, due to the relatively large width difference between the hot spot and the rivet cell, a fairly large number of rivet cells (e.g., hundreds, thousands, millions) may be inserted at the location of one hot spot. The hot spots near each of the four chip corners of the structure 102 (e.g., corner 107, corner 108, etc.) can have the same width or different widths, and it should be noted that each rivet cell within the structure 102 can have the same width or different widths. In one example, a plurality of rivet cells 152 can be inserted at the location of the hot spot region 150 in order to cover the entire 3 mm × 3 mm hot spot region 150 as large as possible. In some examples, an operating circuit may overlap the location of the hot spot, and it is not always desirable to insert a rivet cell at the location where it overlaps the operating circuit. Therefore, the methods and systems described herein can determine the exact insertion location and size of the rivet cell such that the inserted rivet cell does not overlap or interfere with the operating circuit of the structure 102. In another exemplary embodiment, a plurality of hot spot regions can be near the corners of the structure 102. For example, as shown in Figure 1B, the hot spot region 150 and another hot spot region 160 can be present near the corner 108.Hot spot regions 150 and 160 can each have a respective set of rivet cells, such as rivet cell 152 disposed in hot spot region 150 and rivet cell 162 disposed in hot spot region 160. By inserting a plurality of rivet cells into structure 102, delamination and cracking at the dielectric interface throughout the length (e.g., vertical direction) of the hot spot region identified in structure 102 can be prevented.

[0063] Figure 2 is a diagram showing an exemplary system that can be implemented to construct a rivet cell in one embodiment. System 200 can include a processor 202, a memory 204, and a device 220. Processor 202 can be configured to communicate with memory 204 and device 220. In some examples, processor 202 and memory 204 can be components of the same device, such as a computing device. Device 220 can be, for example, a device or machine (e.g., a dicing machine) configured to perform various manufacturing techniques for constructing structure 102.

[0064] Memory 204 may be configured to store an instruction set 206. The instructions 206 may include code such as source code, executable code, or both, which may be executable by the processor 202, thereby generating rivet cell data 208. In some examples, the instructions 206 may be an executable application (e.g., software) that may be executable by the processor 202, thereby executing various modeling techniques (described below). The rivet cell data 208 may include data regarding one or more rivet cells (e.g., the rivet cells 110, 111, 152, 162 shown in FIG. 1A or FIG. 1B or both), and may include instructions (e.g., executable code) that may be executable by the device 220, thereby constructing a structure 102 having the rivet cell 110. For example, the rivet cell data 208 may include data such as the length of the rivet cell 110 (e.g., the number of vias in the stack via of the rivet cell 110), the position of the hot spot 105 (e.g., the distance of the hot spot from the corner 107), one or more dielectric regions up to the position where the rivet cell 110 penetrates, or other types of data regarding the rivet cell 110, or a combination thereof. Further, data indicating the distance between the rivet cells (e.g., the rivet cells 110, 111, 152, 162) and the corner that is most likely to be closest to these rivet cells may be included in the rivet cell data 208. The instructions included in the rivet cell data 208 may be executable code that may be executable by the device 220, such as filler insertion code, and may fill the structure 102 with one or more rivet cells (e.g., the rivet cells 110, 111, 152, 162, etc.) at the positions indicated by the rivet cell data 208. In some examples, the rivet cell data 208 may be provided to the device 220 as an input to an application being executed by the device 220, and the application may be executed by the device 220 to construct the structure 102 and the rivet cell 110.Furthermore, the application being executed by device 220 can construct structure 102, as well as / or various rivet cells such as 110, 111, 152, 162, and other rivet cells that need to be placed in all hot spot regions in structure 102.

[0065] Figures 3A and 3B are diagrams showing exemplary implementations of stress hot spot modeling that can be used to identify stress hot spots in a structure in one embodiment. In one example, the generation of rivet cell data 208 can include stress hot spot modeling to determine the location of stress hot spot 105 and energy release rate (ERR) modeling to determine the length of rivet cell 110. In the example shown in Figure 3A, to perform stress hot spot modeling and energy release rate (ERR) modeling, processor 202 (shown in Figure 2) can be configured to execute instruction 206 (shown in Figure 2) to generate a model 300 of at least a portion of a chip package including semiconductor device 100. In some examples, model 300 can be a virtual three-dimensional (3D) model of at least a portion of a chip package including semiconductor device 100. In some examples, processor 202 can execute instruction 206 to execute an application (or software) to generate model 300. Examples of such applications can include, but are not limited to, software such as SIMULIA ABAQUS(R). In the example shown in Figure 3A, model 300 can be, for example, a quarter cutout of the entire chip package located near plane 351 and plane 352. Model 300 can include modeled components such as lid 330, printed circuit board (PCB) 332, and die 334. In one example, semiconductor device 100 of Figure 1A or Figure 1B can be a part of die 334 in model 300.

[0066] In some examples, the processor 202 may receive an input (e.g., a user input) indicating dimensions of various aspects or components of the model 300. For example, the semiconductor device 100 may be a chip package, and the model 300 may be a virtual model of the chip package. The input received by the processor 202 may include, but is not limited to, the size of the PCB, the size of the die, the size of the lid, the size of the underfill, the size of the TIM (thermal interface material), etc., along with material properties such as those of the PCB, die, lid, underfill, TIM.

[0067] The processor 202 can execute the instruction 206 to perform a virtual thermal stress analysis on the model 300, thereby generating an output indicating how various spots of the semiconductor device 100 (modeled as the model 300) may react to thermo - mechanical stress. For example, the output 302 may include a contour plot representing a top - view of a portion of the model 300. The contour plot can be coded with different shadings to represent various levels of thermo - mechanical stress. The processor 202 can identify the hot spot 105 in the output 302, which may have a significantly different shading compared to other regions of the output 302. In the example shown in FIG. 3A, the hot spot 105 may be identified at a diagonal distance of about r units from the corner 107.

[0068] In one example, the processor 202 can determine the size of the m×n region that can sufficiently cover the hot spot 105. For example, the processor 202 can be initially set to m = 1 millimeter (mm) and n = 1 mm, and determine whether the 1 mm×1 mm region is sufficient to cover the hot spot 105 shown in the output 302. In response to the 1 mm×1 mm region being insufficient to cover the hot spot 105, the processor 202 can increment the values of m and n, such as m = 2 mm and n = 2 mm, and determine whether the 2 mm×2 mm region is sufficient to cover the hot spot 105 shown in the output 302. Upon receiving the determination of the values of m and n that are sufficient to cover the hot spot 105, the processor 202 can include the values of m, n, and r in the rivet cell data 208. Note that the values of m and n can define the cross-sectional area in the lateral direction of the rivet cell 110. In some examples, the instruction 206 can define limits (e.g., 3 mm) for the values of m and n so as not to potentially overlap or contact other vias or components or both in the structure 102. The processor 202 can further identify hot spots adjacent to all corners (e.g., four corners) of the model 300, determine the corresponding values of m, n, and r, and include the determined values in the rivet cell data 208.

[0069] FIG. 3B shows a plot 320 that can model the relationship between the level of stress and the distance from the corner of the semiconductor device 100. In plot 320, axis 321 corresponds to the distance (e.g., in millimeters) from the corner (e.g., corner 107) of the semiconductor device 100, and axis 322 represents the level of normalized stress where 1 represents the maximum stress. Axis 321 can be along the radius marked r from corner 107 as seen in FIG. 3A. Curve 323 represents the variation of the stress level for various values of the distance from the corner. As shown in plot 320, curve 323 increases significantly between approximately 1.0 mm and 1.8 mm from the corner but can decrease after passing 1.8 mm. In the example shown in FIG. 3B, curve 323 can be within the normalized stress range from 0 to 1. The hot spot region can be defined where the normalized stress is greater than the stress threshold 325 (e.g., 0.6). In this case, the hot spot region can be between approximately 1.3 mm and 3 mm from corner 107. The behavior of curve 323 can vary depending on the shape of the package and material properties and can be determined by running an appropriate model.

[0070] FIG. 4 is a diagram showing an exemplary model that can be used to perform energy release rate modeling in accordance with the present disclosure. To perform energy release rate (ERR) modeling, the processor 202 (shown in FIG. 2) can be configured to execute instructions 206 (shown in FIG. 2) for extracting one or more submodels from the model 300. In the example shown in FIG. 4, the processor 202 can extract a submodel 410 from the model 300, and the submodel 410 can be a two-dimensional (2D) or three-dimensional (3D) model showing a certain cross-sectional area of the model 300 from a perspective of one side (e.g., front, rear, left, right). The example shown here is for obtaining a representative 2D model. The processor 202 can further extract another submodel 420 from the submodel 410, and the submodel 420 can be another 2D model showing a part of the submodel 410. The processor 202 can further extract another submodel 430 from the submodel 420, and the submodel 430 can be another 2D model showing a part of the submodel 420.

[0071] The processor 202 can be configured to extract various submodels based on a scale factor. For example, the submodel 410 can be a 2D model having a millimeter-scale mesh size (e.g., the size of the grid or nodes used to render the submodel), the submodel 420 can be a 2D model having a micrometer-scale mesh size, and the submodel 430 can be a 2D model having a nanometer-scale mesh size. In some examples, the processor 202 can receive the number of submodels to be generated, the scale difference between submodels, the dimensions of various aspects or components of the model 300, the materials (and their properties) used in the construction of the structure 102 and / or the semiconductor device 100, and / or inputs (e.g., user inputs) indicating the properties of the materials. For example, the input can indicate the process-of-record (POR) properties of the BEOL material, the plastic effect of the material, boundary conditions (such as free surface or fixed surface), loading conditions (e.g., cool-down within a temperature range), etc.

[0072] Based on the model 300 or based on one of the generated sub - models, the processor 202 can further predict anomalies. In the example shown in FIG. 4, the processor 202 can predict a crack 432 in one of the interfaces between consecutive dielectric regions in the sub - model 430. In some examples, the processor 202 can extract the sub - models 410, 420, 430 based on the stress hot - spots identified by the above - mentioned stress hot - spot modeling. For example, when identifying the stress hot - spot 105 close to the corner 107, the processor 202 can extract a part of the model 300 that includes the position of the stress hot - spot 105, such as the sub - model 410. The processor 202 can further extract subsequent sub - models that include the position of the stress hot - spot 105, such as the sub - models 420 and 430. Based on the position of the stress hot - spot 105, the processor 202 can predict a crack 432 at the position of the stress hot - spot 105. When predicting an anomaly (e.g., crack 432) based on the model 300, the processor 202 can execute the instruction 206 to model the relationship between the length of the rivet cell 110 and the energy release rate of the semiconductor device 100.

[0073] FIG. 5 is a diagram showing an exemplary implementation of energy release rate modeling that can be used to determine the length of a rivet cell in one embodiment. The processor 202 can execute the instruction 206 to perform an energy release rate (ERR) modeling 500. The output of the ERR modeling 500 can include data representing the relationship between the length of the rivet cell 110 (e.g., the number of stacked vias in the rivet cell 110) and the energy release rate (ERR) of the semiconductor device 100. In particular, the ERR modeling 500 executed by the processor 202 can be performed for a particular interface between a pair of dielectric regions under the prediction of an anomaly such as the crack 432 at the hot spot 105. For example, in the example shown in FIG. 5, a crack 432 can be predicted at the interface between the dielectric regions 101d and 101e (see FIG. 1).

[0074] In one example, when the ERR modeling 500 that predicts the crack 432 at the interface between the dielectric regions 101d and 101e is completed, the processor 202 can remove the predicted crack 432 based on the submodel 430 (in FIG. 4) and predict another anomaly at another interface between another pair of dielectric regions. The processor 202 can perform the ERR modeling 500 for each pair of dielectric regions or for the interfaces between some selected pairs of dielectric regions based on the anomaly prediction at the specified stress hot spots (e.g., near each corner) of the semiconductor device 100.

[0075] In another example, the processor 202 can perform ERR modeling on a selected interface. For example, referring to FIG. 1A, the dielectric region 101f can have a dielectric constant value of k = 2.7, the dielectric region 101e can have a dielectric constant value of k = 2.7, the dielectric region 101d can have a dielectric constant value of k = 2.55, the dielectric region 101c can have a dielectric constant value of k = 2.7, and the dielectric region 101b can have a dielectric constant value of k = 3.0. The values of k shown herein are merely exemplary values, and there can be other values of k corresponding to low-k, ULK, hard dielectrics, etc. In one example, based on the arrangement of the dielectric regions of the structure 102, the processor 202 can identify dielectric regions that may be relatively more vulnerable to cracks or delamination. For example, the processor 202 can compare the values of k between adjacent dielectric regions. Based on the comparison between the dielectric regions 101e (k = 2.7) and 101d (k = 2.55), and the comparison between 101d (k = 2.55) and 101c (k = 2.7), the processor 202 can determine that the dielectric region 101d is "sandwiched" between two dielectric regions having relatively high k. Based on the identification result of the vulnerable region (e.g., the dielectric region 101d), the processor 202 can determine that the interface in contact with the dielectric region 101d may be relatively more vulnerable to cracks. In another example, when a first dielectric region and a second dielectric region are adjacent to each other (e.g., laminated to each other) and the first and second dielectric regions are classified as having different rigidities (e.g., a hard dielectric and a soft dielectric), the processor 202 can determine that the interface between the first dielectric region and the second dielectric region may be relatively more vulnerable to cracks. The processor 202 can perform the ERR model using the predicted anomalies at the interface 106 between the dielectric regions 101e and 101d. The processor 202 can further perform the ERR model using the predicted anomalies at the interface between the dielectric regions 101d and 101c. In the example shown in FIG. 5, the processor 202 can predict a crack 432 at the interface 106 (see FIG. 1) and perform the ERR model.In another example, the processor 202 may determine that the interface between two low-k regions, such as the dielectric regions 101e and 101f, may not be as fragile as to cause cracks because the dielectric constants are the same (e.g., the dielectric regions 101e and 101f have the same stiffness). Thus, the processor 202 does not necessarily need to execute the ERR model for the modeled structure having the predicted anomaly at the interface between the dielectric regions 101e and 101f.

[0076] In the example shown in FIG. 5, the processor 202 can start the ERR modeling 500 by generating a modeled structure 510, which can be a virtual prototype of a plurality of dielectric regions of the structure 102 having a predicted crack 432 at the location of the hot spot 105. The ERR modeling 500 can proceed through different modeled structures having different numbers of vias of the stack via 508. For example, the ERR modeling 500 can start with the one having the fewest number of vias among the stack vias 508 and proceed in order up to the one having the most number of vias among the stack vias 508. In the example shown in FIG. 5, the ERR modeling 500 can proceed in the order of the modeled structures 511, 512, 513, 514 and end at 515. The ordered modeled structures 511, 512, 513, 514, 515 are shown here, but additional modeled structures can be included before, after, or in between any one of the shown modeled structures 511, 512, 513, 514, 515 of this ERR modeling 500. The rivet cell with the smallest ERR among the modeled structures 511, 512, 513, 514, 515 can be chosen or selected as the optimal rivet cell (e.g., having an optimal length). In the example of FIG. 5, the ERR of the modeled structure 513 is the smallest. The ERR can be a measure indicating the tendency of crack generation. The greater the ERR, the higher the likelihood of crack occurrence. Thus, for example, the rivet cell of the modeled structure 513 can be an optimal choice. In one example, the ERR is the energy released per unit area of crack propagation. Numerically, the ERR can represent the energy available at the crack tip due to the crack shape, material properties, and external stress occurring in the free body. Further, each material can have a unique property called the "critical" energy release rate. The crack can propagate if and only if the ERR at the crack tip exceeds the critical ERR of the material. Therefore, in the design and manufacture of the chip, it may be desirable to have the lowest possible ERR for a given crack shape and material conditions.

[0077] In one example, instruction 206 may include an algorithm or code or both related to finite element analysis such that the processor 202 can execute instruction 206 to perform finite element analysis on structure 102. Finite element analysis (FEA) can be regarded as a computerized method for predicting how structure 102 may react to real-world forces such as vibration, heat, fluid flow, or other forces, or combinations thereof. The finite element analysis performed by processor 202 can provide a prediction of how structure 102 may react when the number of vias included in via 508 changes. One of the reactions that can be monitored is ERR. ERR can indicate a negative impact on structure 102 in response to the presence of cracks or delamination. Therefore, it may be desirable to construct a structure with a low ERR.

[0078] To initiate ERR modeling 500, the processor 202 can incrementally add vias to the stack via 508 and calculate ERR in response to the added vias. In the example shown in FIG. 5, axis 501 can represent the value of ERR of the modeled structure of FIG. 5, and axis 502 can represent the number of vias in the stack via 508. Curve 503 can represent the relationship between the observed ERR and the length or number of vias in the stack via 508. Curve 503 can represent the relationship between the observed ERR and the length or number of vias in the stack via 508 under the condition that the modeled structure 510 includes the predicted crack 432. As shown in FIG. 5, the ERR of the modeled structure 510 that does not include the predicted crack 432 is expected to decrease as vias are added to the stack via 508, but when the modeled structure 510 includes the predicted crack 432, the observed ERR increases when a certain number of vias are added to the stack via 508. In one example, this minimum value can be caused by mechanisms such as cracks caused by thermo-mechanical loads due to the difference in thermal expansion of different materials in the structure of the chip package (e.g., the chip package modeled as the model 300 in FIG. 3A), and the resulting chip-level hotspots. In the example shown in FIG. 5, the modeled structure 515 includes a rivet cell that penetrates the entire structure (e.g., the entire set of dielectrics), which can cause the vertical tension that can increase the tendency of cracks to occur. Thus, with the modeling examples shown in FIGS. 2 and 4, the system 200 may be able to identify the optimal size of the rivet cell (e.g., the modeled structure 513) that minimizes the ERR metric. Technologies with different BEOL thicknesses having different dielectric stacks can utilize the modeling process described herein to identify hotspots and prevent crack generation and crack propagation.

[0079] Based on the relationship modeled by curve 503, the processor 202 can identify the optimal length of the rivet cell 110. For example, the processor 202 can identify the minimum point of curve 503 corresponding to the minimum ERR. In response to identifying the minimum point, the processor 202 can identify the modeled structure corresponding to the identified minimum ERR. In the example shown in FIG. 5, the modeled structure 513 appears to correspond to the minimum ERR on both sides of curve 503. The processor 202 can set the length of the rivet cell 110 based on the stack vias 508 of the modeled structure 513. In one example, the length of the rivet cell 110 can have the same number of vias as the stack vias 508 of the modeled structure 513. In another example, the processor 202 can remove one or more vias from the stack vias 508 of the modeled structure 513 based on factors such as the BEOL stack used in the technology, the various dielectrics used, and their respective dielectric thicknesses, as well as local via density limits, ERR, and other thermo-mechanical stresses in the modeled domain. After setting the length of the rivet cell 110, the processor 202 can include the length of the rivet cell 110 in the rivet cell data 208.

[0080] FIG. 6 is a flowchart regarding stack via rivets of chip hot spots in one embodiment. The process 600 of FIG. 6 can be executed, for example, using the computer system 200 described above. The exemplary process can include one or more operations, actions, or functions as illustrated by one or more of blocks 602, 604, 606, 608, and / or 610. Although shown as individual blocks, depending on the desired implementation, the various blocks can be divided into additional blocks, combined into fewer blocks, excluded, or executed in parallel.

[0081] Process 600 can be a process for determining the positions of rivet cells in a structure. Process 600 can start from block 602, and the device can model the positions of stress hot spots in the structure. In some examples, the device can generate a three-dimensional (3D) model of the structure, execute instructions for performing stress analysis on the 3D model, and identify the positions of stress hot spots from the output of the stress analysis.

[0082] Process 600 can proceed from block 602 to block 604. In block 604, the device can model the relationship between the length of the rivet cell and the energy release rate of the structure. In some examples, the device can model the relationship based on the prediction of anomalies at the interface between a first dielectric region and a second dielectric region among a plurality of dielectric regions at the position of the stress hot spot. In some examples, the energy release rate of the present structure can be based on the predicted position of the anomaly. In some examples, the predicted anomaly can be a crack at the interface between the first dielectric region and the second dielectric region. In some examples, the first dielectric region and the second dielectric region can have different dielectric constant values. In some examples, the rivet cell can be one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells can be based on the size of the stress hot spot.

[0083] Process 600 can proceed from block 604 to block 606. In block 606, the device can identify the optimal length of the rivet cell based on the modeled relationship. In some examples, the device can identify the optimal length of the rivet cell by identifying the length of the rivet cell that results in the lowest energy release rate.

[0084] Process 600 can proceed from block 606 to block 608. At block 608, the device can generate rivet cell data indicating the insertion positions of the rivet cells. The insertion positions can be based on the positions of the stress hot spots. In some examples, the rivet cell data can further indicate the insertion positions of a plurality of rivet cells. The plurality of insertion positions can be within the stress hot spots. In some examples, the insertion positions can be arranged at a certain lateral distance from the corners of the structure.

[0085] Process 600 can proceed from block 608 to block 610. At block 610, the device can send the rivet cell data to the device and instruct the device to construct the structure according to the rivet cell data. The present structure can be constructed to include rivet cells at the insertion positions. The rivet cells can extend through the stress hot spots of the present structure and can penetrate at least one of the plurality of dielectric regions of the present structure. In some examples, when a rivet cell is inserted, the rivet cell can penetrate at least a part of the first dielectric region and at least a part of the second dielectric region.

[0086] FIG. 7 is a schematic diagram of an exemplary computer or processing system that can perform the generation of chip hot spot stack via vias in one embodiment. This computer system is merely an example of a suitable processing system and is not intended to imply any limitation as to the scope of use or functionality of the embodiments of the methods described herein. The illustrated processing system is operable using a number of other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, or configurations, or combinations thereof, with which the use associated with the processing system shown in FIG. 7 may be suitable include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, cellular phones, tablet computers, wearable devices, virtual reality devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, supercomputers, and distributed cloud computing environments including any of the above systems or devices.

[0087] A computer system may be described in the general context of computer system executable instructions, such as program modules. Generally, program modules may include routines, programs, objects, components, logic, data structures, etc. that perform particular tasks or implement particular abstract data types. A computer system may be practiced in a distributed cloud computing environment where tasks are performed by remote processing devices linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.

[0088] The components of a computer system can include, but are not limited to, one or more processors or processing devices 12, a system memory 16, and a bus 14 that couples various system components including the system memory 16 to the processor 12. The processor 12 can include a module 30 (e.g., a rivet cell module 30) that executes the methods described herein. The module 30 may be programmed in the integrated circuit of the processor 12, or may be loaded from the system memory 16, a storage device 18, or a network 24 or a combination thereof.

[0089] The bus 14 can represent one or more of a plurality of types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus that uses any of a variety of bus architectures. By way of example and not limitation, such architectures include ISA(R) (Industry Standard Architecture) bus, MCA(R) (Micro Channel Architecture) bus, Enhanced ISA(R) (EISA) bus, VESA(R) (Video Electronics Standards Association) local bus, PCI(R) (Peripheral Component Interconnects) bus, and USB(R) (Universal Serial Bus).

[0090] The computer system can include various computer system readable media. Such media can be any available media accessible by the computer system, and can include both volatile and nonvolatile media, removable and non-removable media.

[0091] System memory 16 may include a computer system-readable medium in the form of volatile memory such as random access memory (RAM) or cache memory or both. The computer system may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage device 18 may be provided for reading from and writing to a non-removable non-volatile magnetic medium (e.g., a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a “floppy (R) disk”), and an optical disk drive for reading from and writing to a removable non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media may be provided. Other examples of non-volatile memory or storage media may include, for example, flash memory, magnetoresistive random access memory (MRAM). In such examples, each may be connected to bus 14 by one or more data media interfaces.

[0092] The computer system may also communicate with one or more external devices 26 such as a keyboard, a pointing device, a display 28; one or more devices that enable a user to interact with the computer system; or any device that enables the computer system to communicate with one or more other computing devices (e.g., a network card, a modem, etc.); or a combination thereof. Such communication may be through an input / output (I / O) interface 20.

[0093] Furthermore, the computer system can communicate with one or more networks 24, such as a local area network (LAN), a general wide area network (WAN), Wi-Fi(R), a cellular network (e.g., 3G, 4G, 5G, Long Term Evolution (LTE)), or a public network (e.g., the Internet), or a combination thereof, via the network adapter 22. As shown, the network adapter 22 communicates with other components of the computer system via the bus 14. Although not shown, it should be understood that other hardware components or software components or both can also be used in combination with the computer system. Examples include, but are not limited to, microcode, device drivers, redundant processing devices, external disk drive arrays, RAID (redundant array of independent disks) systems, tape drives, data archive storage systems, etc.

[0094] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, segment, or portion of instructions, which includes one or more executable instructions for performing the specified logical function. In some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially simultaneously, or the blocks may be executed in the reverse order, depending upon the functionality required. It should also be noted that each block of the block diagrams or flowchart diagrams, or combinations of blocks in the block diagrams or flowchart diagrams, can be implemented by a special purpose hardware-based system that performs the specified function or acts, or combinations of special purpose hardware and computer instructions.

[0095] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. As used herein, the terms "comprise" or "comprising" or both are defined to specify the presence of the stated features, integers, steps, operations, elements, or components, or combinations thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or collections thereof.

[0096] Although the present application has been particularly shown and described with respect to its preferred embodiments, it will be understood by those skilled in the art that the foregoing and other changes in form and detail may be made without departing from the scope of the present application. Accordingly, the present application is not intended to be limited to the exact forms and details shown and described, but is intended to be within the scope of the appended claims.

Claims

1. A structure comprising a plurality of dielectric regions and a rivet cell including a set of stack vias, wherein the rivet cell extends through a stress hot spot of the structure, the length of the rivet cell penetrates at least one of the plurality of dielectric regions, and the length of the rivet cell is based on a model of the relationship between the length of the rivet cell and the energy release rate of the structure.

2. The structure according to claim 1, wherein the rivet cell is a first rivet cell disposed at a first lateral distance from a first corner of the structure, and the structure further comprises a second rivet cell disposed at a second lateral distance from a second corner of the structure.

3. The structure according to claim 1 or claim 2, wherein the rivet cell is one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells is based on the size of the stress hot spot.

4. The structure according to any one of claims 1 to 3, wherein the stress hot spot is one of a plurality of stress hot spots existing in the whole structure, and each stress hot spot includes a respective set of rivet cells.

5. The structure according to any one of claims 1 to 4, wherein the structure is a wiring process (BEOL) structure of a semiconductor device.

6. The structure according to any one of claims 1 to 5, wherein the rivet cell penetrates an interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions, and the first dielectric region and the second dielectric region have different dielectric constants.

7. The structure according to claim 6, wherein the first dielectric region includes a hard dielectric material and the second dielectric region includes a soft dielectric material.

8. A method for determining the position of a rivet cell in a structure, comprising: modeling the position of a stress hot spot in the structure; modeling the relationship between the length of the rivet cell and the energy release rate of the structure; identifying the optimal length of the rivet cell based on the modeled relationship; and generating rivet cell data indicating the insertion position of the rivet cell, wherein the insertion position is based on the position of the stress hot spot. ​ Transmitting the rivet cell data to the device to instruct the device to construct the structure according to the rivet cell data, whereby the structure is constructed to include the rivet cell at the insertion position, and thus the rivet cell extends through the stress hot spot of the structure and penetrates at least one of the plurality of dielectric regions of the structure, said transmitting A method comprising. **Claim 9** The method according to claim 8, wherein the rivet cell data further indicates a plurality of insertion positions of a plurality of rivet cells, and the plurality of insertion positions are within the stress hot spot. **Claim 10** Modeling the relationship is performed based on prediction of an anomaly at an interface between a first dielectric region and a second dielectric region among the plurality of dielectric regions at the position of the stress hot spot, When the rivet cell is inserted, the rivet cell penetrates at least a part of the first dielectric region and at least a part of the second dielectric region The method according to claim 8, wherein the energy release rate of the structure is based on the predicted position of the anomaly. **Claim 11** The method according to claim 10, wherein the predicted anomaly is a crack at the interface between the first dielectric region and the second dielectric region. **Claim 12** The method according to claim 10, wherein the first dielectric region and the second dielectric region have different dielectric constant values. **Claim 13** The method according to claim 8, wherein the stress hot spot is one of a plurality of stress hot spots present throughout the structure, and each stress hot spot includes a respective set of rivet cells. **Claim 14** The method according to claim 8, wherein the rivet cell is one of a plurality of rivet cells inserted into the stress hot spot, and the number of rivet cells is based on the size of the stress hot spot. **Claim 15** The method according to claim 8, wherein specifying the optimal length of the rivet cell includes specifying the length of the rivet cell at which the energy release rate is the lowest. **Claim 16** Modeling the stress hot spot involves Generating a three-dimensional (3D) model of the structure; and Executing instructions to perform a stress analysis on the 3D model identifying the location of the stress hot spot based on the output of the stress analysis The method according to claim 8, comprising: **Claim 17** A computer program for determining the location of a rivet cell in a structure, the computer program causing a computer to execute the method according to any one of claims 8 to 16.

Images