Semiconductor Device with Laterally Varying Doping for Thin-Wafer Technology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for forming field-stop regions in power semiconductor devices face challenges in thin-wafer technology due to difficulties in masking proton implantation and high temperature loads required for drive-in processes, which affect the manufacturing process and device performance.
Innovation Solution
A method involving proton implantation into a semiconductor wafer with subsequent masked hydrogen plasma treatment to create laterally varying n-type doping concentrations, allowing for the formation of structured field-stop regions with reduced wafer bowing and lower temperature loads, suitable for thin-wafer technology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If masked implantation with high energy protons is used to form deep field-stop regions, then the penetration depth is sufficient (30-60 μm), but significant wafer bowing occurs and manufacturing becomes difficult
Solution Approach 1:
The doping process is segmented into multiple implantation steps with different proton energies. First, low energy protons (e.g., 100 keV) are implanted to create a shallow highly-doped region. Then, high energy protons (e.g., 2 MeV) are implanted to create a deeper lower-doped region. This segmentation allows each step to use appropriate masking techniques without causing excessive wafer bowing, while achieving the required total penetration depth of 30-60 μm.
Solution Approach 2:
Different regions of the semiconductor wafer are given different doping qualities through selective masking. The field-stop region receives specific doping concentrations and profiles that differ from surrounding areas. By using masks during implantation, locally optimized doping characteristics are achieved - with the field-stop region having appropriate depth and concentration distribution - while minimizing overall wafer deformation through controlled local processing.
2Length of stationary object
If thick masks are used on the wafer backside for proton implantation, then deep penetration is achieved, but significant wafer bowing occurs
Solution Approach 1:
The single thick mask approach is segmented into multiple thinner mask applications corresponding to different implantation energy steps. Each thin mask application causes minimal wafer bowing, while the cumulative effect of multiple implantation steps achieves the required deep penetration depth. This eliminates the need for a single thick mask that would cause excessive bowing.
Solution Approach 2:
A shallow highly-doped region is formed first using low energy protons before the deep field-stop region is created with high energy protons. This preliminary action establishes a foundation doping profile that facilitates subsequent deep implantation without requiring excessive mask thickness, thereby reducing wafer bowing while still achieving the necessary final penetration depth.
3Shape
If thinner implantation masks are used for n-type dopants, then wafer bowing is reduced, but higher temperature loads are required for drive-in processes
Solution Approach 1:
The doping method transitions from thermal drive-in processes to proton implantation, fundamentally changing the parameter space. Proton implantation delivers dopants directly to desired depths through kinetic energy control, eliminating the need for high temperature thermal diffusion. This parameter change allows the use of thin masks without requiring compensating high temperature loads, as the implantation energy (e.g., 100 keV to 2 MeV) directly controls penetration depth at lower temperatures.
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 enables the creation of semiconductor devices with improved softness and high voltage blocking capability by forming laterally varying n-type doping concentrations, enhancing the switching behavior and reducing leakage current while maintaining compatibility with thin-wafer technology.
Implementation Method 1
forming the second n-type semiconductor layer includes implanting protons of a first maximum energy into the first n-type semiconductor layer
Implementation Method 2
locally treating the second surface with a masked hydrogen plasma
Data Source
AI summary
A semiconductor device includes a semiconductor body including a first surface having a normal direction defining a vertical direction, a first n-type semiconductor region arranged below the first surface and having a first maximum doping concentration and a second n-type semiconductor region arranged below the first n-type semiconductor region and including, in a vertical cross-section, two spaced apart first n-type portions each adjoining the first n-type semiconductor region, having a maximum doping concentration which is higher than the first maximum doping concentration and having a first minimum distance to the first surface, and a second n-type portion adjoining the first n-type semiconductor region, having a maximum doping concentration which is higher than the first maximum doping concentration and a second minimum distance to the first surface which is larger than the first minimum distance. A p-type second semiconductor layer forms a pn-junction with the second n-type portion.