Extended Via Inductor Structure for Higher On-Chip Q Factor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in integrated circuit design is to increase the quality factor (Q factor) of on-chip inductors to improve signal selectivity and reduce substrate loss and self-resonance, which are hindered by substrate loss and self-resonance factors.
Innovation Solution
Increasing the oxide thickness of on-chip inductors by extending upper conductive vias through at least two dielectric layers, and optionally through an etch stop layer, to reduce substrate loss and self-resonance, thereby enhancing the Q factor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If oxide thickness is increased by extending upper conductive vias through at least two dielectric layers, then Q factor is improved, but device complexity increases
Solution Approach 1:
The patent extends the upper conductive vias vertically through multiple dielectric layers (first dielectric layer, second dielectric layer, and optionally third dielectric layer) to increase oxide thickness. This vertical extension into the third dimension increases the distance between the inductor and substrate, thereby reducing substrate loss and self-resonance effects while improving the Q factor of the on-chip inductor.
Solution Approach 2:
The patent divides the dielectric structure into multiple distinct layers (first dielectric layer with first oxide thickness, second dielectric layer with second oxide thickness, and optionally third dielectric layer with third oxide thickness). Each layer can be independently formed and controlled, allowing the total oxide thickness to be increased through stacking multiple thinner layers rather than requiring a single thick layer, thus improving Q factor while managing fabrication complexity.
2Loss of energy
If upper conductive vias are extended through multiple dielectric layers, then substrate loss is reduced, but manufacturing precision requirements increase
Solution Approach 1:
The total oxide thickness is achieved by stacking multiple dielectric layers with individually controllable thicknesses. The first dielectric layer has a first oxide thickness, the second dielectric layer has a second oxide thickness, and the optional third dielectric layer has a third oxide thickness. This segmentation allows each layer to be formed with standard precision tolerances while achieving greater total thickness than a single layer would permit, thereby reducing substrate loss without excessively increasing manufacturing precision requirements.
Solution Approach 2:
The patent utilizes the vertical dimension by extending conductive vias through multiple stacked dielectric layers. This vertical stacking approach allows cumulative oxide thickness to increase (reducing substrate loss) while each individual layer thickness remains within manufacturable precision ranges. The multi-layer structure transforms a single-dimension thickness challenge into a multi-dimensional solution.
Data Source
AI summary
A device includes a substrate, a first conductive layer on the substrate, a first conductive via, and further conductive layers and conductive vias between the first conductive via and the substrate. The first conductive via is between the substrate and the first conductive layer, and is electrically connected to the first conductive layer. The first conductive via extends through at least two dielectric layers, and has thickness greater than about 8 kilo-Angstroms. An inductor having high quality factor is formed in the first conductive layer and also includes the first conductive via.


