Hammerhead MTJ Structure to Reduce MRAM Etch Shorts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reactive ion etch and ion beam processing of magnetic tunnel junction (MTJ) stacks in MRAM devices often lead to shorts due to metal re-sputtering from thick bottom metal layers onto sidewalls, posing a challenge for forming embedded MTJ structures with well-defined interfaces.
Innovation Solution
The MTJ structure is formed with a hammerhead side profile, where the tunneling barrier, free layer, and top electrode have a greater width than the bottom electrode and reference layer, and are formed using a damascene scheme to prevent metal re-sputtering, with specific liners and dielectrics used to control the formation and reduce shorts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional planar MTJ structures are used, then manufacturing is simpler, but metal re-sputtering occurs during etch processes causing shorts and reduced reliability
Solution Approach 1:
The patent applies asymmetry by creating a hammerhead-shaped MTJ structure where the top electrode has a larger footprint than the bottom electrode, forming an inverted trapezoidal profile. This asymmetric geometry prevents metal re-sputtering from reaching the bottom electrode during etch processes, eliminating shorts and improving device reliability while maintaining manageable structural complexity through systematic layer formation
Solution Approach 2:
The patent employs preliminary action by forming spacers on the sidewalls of the MTJ stack before the metal deposition step. These spacers act as protective masks that prevent re-sputtered metal from contaminating the bottom electrode and reference layer, thereby preventing shorts before they can occur and enhancing device reliability
2Productivity
If device dimensions are reduced for scalability, then density increases, but manufacturing precision requirements become more stringent
Solution Approach 1:
The asymmetric hammerhead profile with wider top electrode than bottom electrode creates a geometry that is inherently more tolerant to etch variations at scaled dimensions. The expanding profile upward provides a larger process window for maintaining precision during fabrication, enabling better scalability without proportionally increasing manufacturing difficulty
Solution Approach 2:
By forming spacers as a preliminary protective layer before metal deposition, the patent establishes a safeguard that compensates for potential etch precision variations. These spacers ensure that even with dimensional scaling and associated manufacturing challenges, the bottom electrode remains protected from re-sputtering, maintaining reliability across scaled device sizes
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 reduces metal re-sputtering and extends the scalability of MRAM devices by minimizing shorts between pillars, improving embedded MRAM performance and reducing top contact failures.
Implementation Method 1
separated by a spin conductor layer... the tunneling barrier...
Data Source
AI summary
A magnetic tunnel junction (MTJ) stack with a hammerhead profile, including vertically aligned layers of a top electrode, a free layer, a tunneling barrier, a reference layer and a bottom electrode, where the bottom electrode and the reference layer each include a first width, and the top electrode, the free layer and the tunneling barrier, each include a second width greater than the first width. Forming vertically aligned layers of a bottom electrode and a reference layer on the bottom electrode, of a magnetic tunnel junction (MTJ), where the bottom electrode, the reference layer and the hard mask, each include a first width, and separately forming vertically aligned layers of a tunneling barrier, a free layer and a top electrode on the free layer, where the tunneling barrier, the free layer and the top electrode each include a second width, where the second width is greater than the first width.