Fixed-Abrasive Machining for Semiconductor Wafer Flatness

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for producing semiconductor wafers often result in scratches on the surface due to the use of relatively coarse abrasive grains, and require multiple steps and large facilities, leading to increased installation area requirements as wafer diameter increases.

Innovation Solution

A fixed abrasive-grain processing device that uses fine abrasive grains (diameter of 4 µm or less) dispersed in elastic members, with pressure applied between 250-400 g/cm² to prevent scratches and allow concurrent processing of multiple wafers, reducing the number of steps and facility requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If relatively coarse abrasive grain is used in lapping and grinding steps, then processing rate is improved, but scratches are generated on the wafer surface

Engineering Contradiction:
Improveprocessing rateVSAvoidsurface quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent changes the particle size parameter of abrasive grain from relatively coarse (#1000-#1500) to fine (#5000-#8000) to eliminate scratches while maintaining processing rate through batch processing multiple wafers simultaneously

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent combines multiple wafer processing operations into a single batch lapping step, processing 6-12 wafers concurrently to maintain high productivity while using fine abrasive grain that prevents surface scratching

Inventive Principle:
Principle #5Merging (Combining)

2Manufacturing precision

If multiple processing steps (lapping, etching, mirror-surface polishing, finish grinding) are implemented, then surface quality is improved, but the number of steps and facility requirements increase

Engineering Contradiction:
Improvesurface flatnessVSAvoidnumber of processing steps
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent merges lapping, etching, and mirror-surface polishing operations into a single integrated batch processing step, achieving multiple surface treatment functions simultaneously while using fine abrasive grain to eliminate the need for separate finish grinding

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The lapping device is designed to perform multiple functions (lapping, etching, and mirror-surface polishing) in one batch processing operation, making the equipment universal and reducing the total number of processing steps required

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If wafer diameter is increased, then productivity per wafer is improved, but installation area of facilities must be increased

Engineering Contradiction:
Improvewafer sizeVSAvoidinstallation area
Core Design Contradiction:
ProductivityVSArea of stationary object

Solution Approach 1:

The patent processes multiple wafers of large diameter (300mm-450mm) simultaneously in a single batch lapping operation, achieving high total productivity without requiring proportionally larger facility area, as the equipment footprint does not scale linearly with the number of wafers processed

Inventive Principle:
Principle #5Merging (Combining)

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

The solution achieves preferable flatness and high processing rates with reduced facility needs by using fine abrasive grains in elastic members, preventing scratches and maintaining productivity across multiple wafers, thus reducing the number of steps and facility area requirements.

Implementation Method 1

pressure applied between 250-400 g/cm²

Methodology Applied
Scientific EffectPressure: Pressure Increase

Implementation Method 2

abrasive grains...grind the top and the bottom surfaces

Methodology Applied
Scientific EffectFriction: Friction

Implementation Method 3

fine abrasive grains...grind the top and the bottom surfaces of a number of wafers...to obtain preferable flatness

Methodology Applied
Scientific EffectAbrasion: Abrasion

Data Source

PatentEP2439768B1Fixed-abrasive-grain machining apparatus, fixed-abrasive-grain machining method, and semiconductor-wafer manufacturing method
Publication Date: 2022.02.09 SUMCO CORP
  • EP2439768B1 patent drawingFigure 1
  • EP2439768B1 patent drawingFigure 2~3
  • EP2439768B1 patent drawingFigure 4~5

AI summary

Disclosure relates to a fixed abrasive-grain processing device and a method of fixed abrasive-grain processing used for producing a semiconductor wafer, and a method for producing a semiconductor wafer which make the surface of the semiconductor wafer possible to have preferable flatness and which can prevent the number of steps and the installation area of facilities from increasing. The producing of semiconductor wafers (W) uses a fixed abrasive-grain processing device (1) including a lower fixed abrasive-grain layer (21) that is adjacent to the top surface of the lower surface-plate (2) and that grinds the top surfaces of the plurality of semiconductor wafers (W); an upper fixed abrasive-grain layer (31) that is adjacent to the bottom surface of the upper surface-plate (3) and that grinds the bottom surfaces of the plurality of semiconductor wafers (W); a carrier plate (4) that is horizontally interposed between the lower surface-plate (2) and the upper surface-plate (3) and that includes a plurality of holes (4a) each accommodating one of the plurality of semiconductor wafers (W); and a carrier rotating device (40) that circularly moves the carrier plate (4), wherein the lower fixed abrasive-grain layer (21) and the upper fixed abrasive-grain layer (31) include fixed abrasive grain (21b, 31b) having a diameter of 4 µm or less and being dispersed and fixed in elastic members (21a, 31a).