Red Phosphorus Doped Silicon Crystal Defect Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge is to produce an epitaxial silicon wafer with extremely low substrate resistivity while minimizing the generation of stacking faults (SF) and light point defects (LPD), which are exacerbated by the use of red phosphorus as a dopant, leading to poor quality wafers due to the formation of oxygen and red phosphorus clusters.

Innovation Solution

The method involves controlling the solidification rate and thermal history of the single crystal growth process by adjusting the withdrawal time and temperature of the single crystal from the dopant-added melt, specifically maintaining the temperature within 570±70 degrees Celsius for a controlled duration, and omitting the polysilicon back seal to prevent the formation of SF, thereby maintaining gettering ability without increasing LPD.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If red phosphorus is used as a dopant to achieve low substrate resistivity, then the substrate resistivity decreases to 0.9 mΩ·cm or less, but stacking faults and light point defects are generated due to oxygen and red phosphorus cluster formation

Engineering Contradiction:
Improvesubstrate resistivityVSAvoidstacking faults and light point defects
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent changes the thermal parameters during crystal growth, specifically controlling the temperature profile and solidification rate. By optimizing these parameters, the patent suppresses the formation of oxygen-red phosphorus clusters while maintaining low substrate resistivity through red phosphorus doping, thus resolving the contradiction between achieving low resistivity and preventing defect formation

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies preliminary actions during the crystal growth process by controlling the thermal history and solidification conditions before the epitaxial growth stage. This preliminary control of doping uniformity and cluster prevention during ingot formation enables subsequent low-defect epitaxial growth with low resistivity

Inventive Principle:
Principle #10Preliminary action

2Reliability

If polysilicon back seal is applied to improve gettering ability, then oxygen precipitates are controlled, but stacking faults are generated on the epitaxial film

Engineering Contradiction:
Improvegettering abilityVSAvoidstacking faults on epitaxial film
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent extracts and eliminates the polysilicon back seal process from the manufacturing flow. By removing this step that causes stacking faults, the patent achieves both low defect density and maintains adequate gettering ability through alternative thermal and compositional controls during crystal growth

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent converts the potential harm of red phosphorus volatility and cluster formation into benefit by optimizing the thermal profile. The controlled thermal history transforms what would be defect-causing conditions into beneficial doping uniformity, achieving low resistivity without stacking faults

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

3Reliability

If arsenic or antimony is used as n-type dopant for resistivity adjustment, then volatility is reduced, but dopant concentration cannot be sufficiently increased

Engineering Contradiction:
Improvedopant concentrationVSAvoidextreme volatility
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent changes the physical state and incorporation mechanism of the dopant by using red phosphorus in a controlled solidification process rather than relying on volatile doping during epitaxial growth. This parameter change in doping methodology enables high dopant concentration achievement with red phosphorus while avoiding the volatility problems of arsenic and antimony

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 approach effectively reduces the number of SF and LPD on the epitaxial silicon wafer, achieving a low resistivity of 0.9 mΩ·cm or less while ensuring high-quality wafers with minimal defects, improving yield and production efficiency.

Implementation Method 1

bringing the seed crystal into contact with the dopant-added melt... such that a resistivity of the single crystal is 0.9 mΩ·cm or less

Methodology Applied
Scientific EffectSolidification: Freezing

Implementation Method 2

withdrawing the single crystal from the dopant-added melt in a state that a temperature of an upper end of the straight body is 590 degrees C. or more

Methodology Applied
Scientific EffectThermal diffusion: Diffusion

Data Source

PatentUS10100429B2Method for producing a silicon single crystal doped with red phosphorous with reduced number of stacking faults and method for producing a silicon wafer using the same
Publication Date: 2018.10.16 SUMCO CORP
  • US10100429B2 patent drawing
  • US10100429B2 patent drawing
  • US10100429B2 patent drawing

AI summary

A method for producing a single crystal includes: bringing a seed crystal into contact with a dopant-added melt, in which a red phosphorus is added to a silicon melt, such that a resistivity of the single crystal is 0.9 mΩ·cm or less and subsequently pulling up the seed crystal, to form a straight body of the single crystal; and withdrawing the single crystal from the dopant-added melt in a state that a temperature of an upper end of the straight body is 590 degrees C. or more.