Multi-Strike Copper-Aluminum Bonding for 3D Integration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The semiconductor industry faces limitations in increasing integration density due to physical constraints in two-dimensional integrated circuits, leading to increased circuit RC delay and power consumption, prompting the exploration of three-dimensional integrated circuits through wafer bonding techniques.

Innovation Solution

A multi-strike process is employed for bonding package components, where copper bumps with oxide layers strike aluminum pads, breaking the oxide layers to expose un-oxidized surfaces, followed by annealing to form an intermetallic compound bond without the need for solder or nickel layers, simplifying the bonding process and reducing costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional solder bonding with nickel layers is used, then bonding reliability is achieved, but device complexity and manufacturing cost increase

Engineering Contradiction:
Improvebonding reliabilityVSAvoidbonding structure complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent removes the nickel layer from the traditional solder bonding process, extracting only the essential copper bump and solder ball components. This simplification maintains bonding reliability while reducing structural complexity and manufacturing cost by eliminating the intermediate nickel layer that requires additional deposition processes.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of following the traditional sequence of copper bump with nickel layer then solder ball, the patent inverts the approach by using copper bumps directly bonded with solder balls without nickel intermediate layers. This inversion simplifies the bonding structure while achieving reliable electrical and mechanical connections through direct copper-solder bonding.

Inventive Principle:
Principle #13The other way round (Inversion)

2Quantity of substance

If more devices are integrated into one chip, then integration density improves, but circuit RC delay and power consumption increase

Engineering Contradiction:
Improveintegration densityVSAvoidpower consumption
Core Design Contradiction:
Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

The patent enables three-dimensional integration by stacking multiple device layers vertically through direct copper-solder bonding. This dimensional transition from 2D to 3D integration allows higher device density without proportionally increasing interconnection length, thereby controlling power consumption and RC delay despite increased integration density.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Object-affected harmful factors

If oxide layers on copper bumps are present, then oxidation protection is provided, but bonding quality deteriorates

Engineering Contradiction:
Improveoxidation protectionVSAvoidbonding quality
Core Design Contradiction:
Object-affected harmful factorsVSReliability

Solution Approach 1:

The patent applies preliminary protective measures by forming a thin barrier layer or controlling oxide formation before bonding. This preliminary action allows the copper bump to maintain oxidation protection while ensuring that the oxide layer does not interfere with subsequent solder bonding, thus preserving both oxidation protection and bonding quality.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent controls the oxidation state of copper bumps by adjusting processing parameters such as atmosphere control, timing, and temperature. By changing these parameters, the oxide layer is controlled to provide necessary oxidation protection during storage and handling while being thin enough or removable enough to allow high-quality solder bonding.

Inventive Principle:
Principle #35Parameter changes

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 method enhances bonding reliability and reduces the need for chemical solutions, improving throughput and simplifying the structure formation, allowing direct bonding of copper and aluminum without additional layers, thus overcoming the limitations of two-dimensional integration.

Implementation Method 1

A multi-strike process is employed for bonding package components, where copper bumps with oxide layers strike aluminum pads, breaking the oxide layers to expose un-oxidized surfaces

Methodology Applied
Scientific EffectImpact force: Impact Force

Implementation Method 2

followed by annealing to form an intermetallic compound bond without the need for solder or nickel layers

Methodology Applied
Scientific EffectAnnealing: Annealing

Implementation Method 3

annealing to form an intermetallic compound bond

Methodology Applied
Scientific EffectDiffusion: Diffusion

Data Source

PatentUS10068868B2Multi-strike process for bonding packages and the packages thereof
Publication Date: 2018.09.04 TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
  • US10068868B2 patent drawing
  • US10068868B2 patent drawing
  • US10068868B2 patent drawing

AI summary

A method includes performing a first strike process to strike a metal bump of a first package component against a metal pad of a second package component. A first one of the metal bump and the metal pad includes copper. A second one of the metal bump and the metal pad includes aluminum. The method further includes performing a second strike process to strike the metal bump against the metal pad. An annealing is performed to bond the metal bump on the metal pad.