Semiconductor Device Nickel Barrier Layer Electromigration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The electromigration phenomenon (EM phenomenon) occurs frequently in semiconductor devices due to the small area of intermediate bonding layers, leading to void generation and increased electric resistance, especially when current flows through boundaries between different metals.
Innovation Solution
The solution involves forming nickel layers on the surfaces and side surfaces of the joint parts, and structuring the intermediate bonding layer with a higher Young's modulus material, such as an intermetallic compound of tin, to prevent atom movement and void formation. Additionally, the intermediate bonding layer is made thinner and composed of materials like silver or silver compounds to enhance its modulus.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If the intermediate bonding layer area is reduced to downsize joint parts, then device compactness is improved, but electromigration phenomenon increases due to higher current density
Solution Approach 1:
A nickel layer is introduced as an intermediary substance between the copper joint part and the tin-based solder material. This nickel barrier layer prevents direct contact between copper and tin, blocking the electromigration pathway and atom diffusion, thereby solving the reliability issue while maintaining the reduced joint part size
Solution Approach 2:
The bonding structure uses a composite material system consisting of copper joint part + nickel barrier layer + tin-based solder material. This multi-layer composite structure combines the advantages of each material: copper for conductivity, nickel for electromigration barrier, and tin for bonding, achieving both compactness and reliability
2Reliability
If nickel layers are added to prevent electromigration, then reliability is improved, but manufacturing complexity increases
Solution Approach 1:
The nickel layer is formed on the joint part surfaces before the bonding process. This preliminary preparation of the barrier layer ensures that electromigration prevention is built into the structure from the outset, rather than requiring additional complex processes later in manufacturing
Solution Approach 2:
The invention changes the material parameter by selecting nickel with specific properties (electromigration barrier capability) and controls the layer thickness parameter to be thin enough to maintain compactness while sufficient to provide effective barrier function, optimizing both reliability and manufacturing simplicity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively restrains the progression of the EM phenomenon by preventing atom migration and void generation, maintaining the integrity of the bonding layer and reducing electric resistance, even under high current densities.
Implementation Method 1
When current flows through bonded surfaces made from different metals, a phenomenon occurs in which metal atoms move from one of the metals to the other metal and voids are generated in one of the metals. This phenomenon is called an electromigration phenomenon.
Implementation Method 2
a nickel layer is formed on an upper surface 13a of the first joint part 13, which is bonded to the bonding layer 8g, and a side surface 13b continuous from the upper surface 13a. The nickel layer functions as a barrier that prevents movements of atoms
Implementation Method 3
a nickel layer is formed on an upper surface 13a of the first joint part 13, which is bonded to the bonding layer 8g, and a side surface 13b continuous from the upper surface 13a. Similarly, a nickel layer is formed on a lower surface 26c of the second joint part 26, and a side surface 26b continuous from the lower surface 26c
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
A semiconductor device (2) includes first and second semiconductor elements (3, 5) and first and second conductive members (10, 29). A first electrode (3a) on the first semiconductor element is bonded to a first stack part (12) of the first conductive member by a first bonding layer (8a). A second electrode (5b) on the second semiconductor element is bonded to a second stack part (25) of the second conductive member by a second bonding layer (8f). A first joint part (13) of the first conductive member is bonded to a second joint part (26) of the second conductive member by an intermediate bonding layer (8g). A first surface of the first joint part facing the second joint part, a side surface of the first joint part continuous from the first surface, a second surface of the second joint part facing the first joint part, and a side surface of the second joint part continuous from the second surface are covered by nickel layers (19a, 19b).