Epitaxial Barrel Susceptor Thickness Uniformity

Solution Overview

Problem

Conventional barrel reactors struggle to meet current thickness uniformity requirements for epitaxial deposition, resulting in wafers with thickness variations exceeding two percent, which is not acceptable for modern semiconductor production.

Innovation Solution

A modified barrel susceptor design featuring a polyhedral body with trapezoidal faces, shallow recesses, and ledges that support wafers, allowing for improved gas circulation and reduced boundary layer thickness, thereby enhancing epitaxial thickness uniformity across the wafer surface.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a conventional barrel susceptor with planar surfaces is used, then multiple wafers can be processed simultaneously (high throughput), but the epitaxial thickness uniformity deteriorates (variations exceed 2%)

Engineering Contradiction:
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The susceptor surface is modified with recesses and ledges that create different local geometries. The recesses position wafers at specific depths while the ledges provide localized support, creating non-uniform local conditions that collectively improve overall thickness uniformity across all wafers while maintaining simultaneous processing capability

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The invention transitions from a flat two-dimensional susceptor surface to a three-dimensional structure with recesses and ledges. This adds vertical dimensionality to the susceptor design, allowing precise control of wafer positioning and gas flow paths, thereby improving thickness uniformity while maintaining high throughput

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Device complexity

If a conventional barrel susceptor design is used, then the device structure remains simple, but the boundary layer thickness increases reducing gas circulation efficiency

Engineering Contradiction:
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The susceptor surface is segmented into multiple recesses and ledges rather than being a continuous flat surface. This segmentation creates multiple small gas flow channels between and around the ledges, improving gas circulation efficiency and reducing boundary layer thickness while adding only moderate structural complexity

Inventive Principle:
Principle #1Segmentation

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 modified susceptor achieves epitaxial thickness profiles with variations of less than one micrometer over a 200 mm diameter wafer, significantly improving uniformity and maintaining thickness near the edges, compared to conventional designs which had variations exceeding 1.2 micrometers.

Implementation Method 1

Chemical vapor deposition is a process for growing a thin layer of material on a semiconductor wafer so lattice structures of the deposited material are identical to the wafer

Methodology Applied
Scientific EffectChemical vapor deposition: Chemical Vapour Deposition

Implementation Method 2

The epitaxial deposition process begins by introducing a cleaning gas, such as hydrogen or a hydrogen and hydrochloric acid mixture, to a front surface of the wafer (i.e., a surface facing away from the susceptor) to pre-heat and clean the front surface of the wafer

Methodology Applied
Scientific EffectThermal heating: Heating

Implementation Method 3

During both steps, the susceptor rotates in the reaction chamber

Methodology Applied
Scientific EffectRotational motion:

Data Source

PatentUS8404049B2Epitaxial barrel susceptor having improved thickness uniformity
Publication Date: 2013.03.26
  • US8404049B2 patent drawing
  • US8404049B2 patent drawing
  • US8404049B2 patent drawing

AI summary

A barrel susceptor for supporting semiconductor wafers in a heated chamber having an interior space. Each of the wafers has a front surface, a back surface and a circumferential side. The susceptor includes a body having a plurality of faces arranged around an imaginary central axis of the body. Each face has an outer surface and a recess extending laterally inward into the body from the outer surface. Each recess is surrounded by a rim defining the respective recess. The susceptor also includes a plurality of ledges extending outward from the body. Each of the ledges is positioned in one of the recesses and includes an upward facing support surface for supporting a semiconductor wafer received in the recess. Each of the support surfaces is separate from the outer surface of the respective face.