Single die imager
•All blocks in the same process node
•Higher footprint
•Not optimized cost
A new world of advanced human/machine interface is emerging leading to brand new applications such as augmented reality (AR), virtual reality (VR), and facial-recognition security systems.
3D cameras are the backbone of this revolution.
At NIR wavelength, CMOS image sensor still represents the most viable solution to address mass market applications despite the relative transparency of Silicon to photons. In comparison, compound (like InGaAs) or Quantum Dot solutions represent more expensive and less mature solutions.
Nevertheless, because of this transparency of silicon in infrared, several technics co-exist to increase the pixel sensitivity:
The 3D integration has demonstrated significant advantages:
Higher Performance and Density
Higher Functionality
Smaller Form Factor
Cost reduction
Enabling optimized partitioning
Available in 300mm
Top silicon from 12nm to 500nm
BOX from 15nm to 145nm
•All blocks in the same process node
•Higher footprint
•Not optimized cost
•Innovative partitioning and enhanced density
•More functionality into the same footprint
•Small Form Factor
•Cost reduction with optimized process node
For 3D-stacked imager the SmartCut™ technology allows multiple die stacking either by:
Bonding SOI wafers and using BOX as an etch stop layer at backgrinding step
Transferring a mono-crystalline layer on top of patterned wafers