April 11, 2005 ITWG1 Technical issues for the future Silicon Wafers *Kiyoshi Takada: Shin-Etsu Handotai Co., Ltd. (a former SSi General Manager) Masato.

April　11, 2005 ITWG 1

Technical issues for the future Silicon Wafers

*Kiyoshi Takada: Shin-Etsu Handotai Co., Ltd.

(a former SSi General Manager)

Masato Imai: Komatsu Electronic Metals Co., Ltd.

(a former SSi Epitaxial Dept. Manager)

Nobuyuki Hayashi: Sumitomo Mitsubishi Silicon Corp.

(SSi Administration Dept., General Manager)

April　11, 2005 ITWG 2

Contents of presentation

Q1. Why was the SSi consortium formed? Why was 400mm studied?

Q2. What have been the SSi main technical success?

Q3. How is the intellectual property rights performance going on?

Q4. What are the key points for sending Technical Information to the Next Wafer Diameter?

April　11, 2005 ITWG 3

Q1. Why was the SSi consortium formed ?

S = Super = Super Heavy and Large+

Si = Silicon itself

SSi

April　11, 2005 ITWG 4

Q1. Why was 400mm studied?

(Quotation from the regime of SELETE at 1996)

At the SSi establishment, it was a general agreement in the semiconductor industry that the next generation of 300mm was 400mm.

April　11, 2005 ITWG 5

Q2. What have been the SSi main technical success?

A2.

1) Monster CZ crystal pulling furnace by SSi original design.

(Verification of “Maruyama Patent”, Cusp typed MCZ, etc.)

2) Wafer shaping process by SSi original Grinding Machines.

(High flatness grinding technology in the lapping-less and etching-less process)

3) Low temperature Epitaxial growth by SSi original design.

(Highly uniform growth of the super thin layer)

April　11, 2005 ITWG 6

World largest CZ crystal pulling furnace

Furnace Height: 12 m Weight: 36 ton Hot zone: 40 inch Cusp-type super conductive magnetCrystal Diameter: 400 mm Weight: over 400 kg Body length: over 1m

April　11, 2005 ITWG 7

Crystal Supporting System

Schematic drawing of built-in systemin SSi furnace M. Maruyama: Patent No. 1851653

April　11, 2005 ITWG 8

The world’s first Si single crystal weighing more than 400Kg in a 400mm diam

eter

400mm Si single crystal, which has a straight length of 110cm and a total weight of 411kg, is shown in contrast with normal 200mm crystal.

April　11, 2005 ITWG 9

Simplified wafer shaping process by Grinding Technology

Schematic view of Ductile Mode Double Disk Grinding (DDG) Machine for Super-Large and Super-Flat Silicon wafer.

Trigonal Prism type Pentahedral Structure

*Bed and base columns are cast as one component

Deionized water type Linear actuator

Bed

Twin double V guide ways

Base column

Four Trigonal Braces Top column

Deionized water type hydraulic grinding spindle

Silicon wafer Saddle

Grinding wheel

April　11, 2005 ITWG 10

SEM photograph of cutting chip

300 nmBy the extremely high stiffness , motion accuracy, feeding resolution and fine mesh grinding wheel, a portion of Ductile-mode grinding was searched.

April　11, 2005 ITWG 11

Q3. How is the intellectual property rights performance going

on?

A3. Gas inlet nozzle

Heater

Gas exhaust nozzle

Heater cover

Holder supportRotation table

Wafer support

Gas inlet nozzle

Heater

Gas exhaust nozzle

Heater cover

Holder supportRotation table

Wafer support

Among the many patents of SSi, the basic patents of Epi reactor are licensed to Tera Semicon (Korea).

April　11, 2005 ITWG 12

Results of 400mm epitaxial wafers

Distance from wafer center, (mm)

Thi

ckne

ss, (

µm

)

0

1

2

3

4

5

-100 0 100 200-200

± 3%± 11%± 22%± 31%

Thickness deviation

Distance from wafer center, (mm)

Thi

ckne

ss, (

µm

)

0

1

2

3

4

5

-100 0 100 200-200

± 3%± 11%± 22%± 31%

Thickness deviation

Distance from wafer center, (mm)

Thi

ckne

ss, (

µm

)

0

1

2

3

4

5

-100 0 100 200-200

± 3%± 11%± 22%± 31%

Thickness deviation

0

1

2

3

4

5

-100 0 100 200-200

± 3%± 11%± 22%± 31%

Thickness deviation

± 3%± 11%± 22%± 31%

Thickness deviation

1014

1015

1016

1017

1018

1019

1020

0.5 1.0 1.5 2.0

T.W.=0.1 µm

Epitaxial layer

Depth, (µm)

Bor

on c

onc.

, (at

oms/

cm3 )

1014

1015

1016

1017

1018

1019

1020

0.5 1.0 1.5 2.0

T.W.=0.1 µm

Epitaxial layer

Depth, (µm)

Bor

on c

onc.

, (at

oms/

cm3 )

Note that the thickness distributions change drastically depending on the gas flow rate adjustment of each inlet nozzle. Thickness uniformity commensurate with industrial levels has been achieved. The transition width of 0.1m is a very steep one compared to that of approximately 0.5m for wafers grown by conventional SiHCl3 process.

April　11, 2005 ITWG 13

Q4. What are the key points for sending Technical Information to the Next Wafer

Diameter ?A4.

Ο SSi information will be much effective.Δ Some of SSi information will be suggestive.Χ New technology should be searched.

Remarks:

ΟNano-Metrology System measuring the price detection of wafer shape.

ΧDefect & Particle Detection Technology.Metrology

ΟEpitaxial Furnace Design with the intention of Cost reduction.

Ο ~ ΔLow Temperature Epitaxial Growth Technology.Epitaxial Growth

ΧHigh purity cleaning technology.

Ο ~ ΔSimplified Wafer Shaping process by Grinding Technologies with Ductile -mode.

Wafer Shaping

ΟFurnace system concept including MCZ, Crystal Suspending system, Larger Quartz Crucible, Safety system, etc.

Δ ~ ΧStandardization of Numerical Simulation and Thermal Properties.Crystal Growth

RemarksTechnological IssuesField

April　11, 2005 ITWG 14

Commercial Viability for Next Diameter crystal growth is

questionable

1) Additional technologies such as Monster Puller installation with MCZ, giant Quartz Crucible, Suspending System,Remote Control,

etc. will be needed. ⇒Higher cost 2) Productivity together with slower growth rate will drastically fall down.

⇒ Higher cost 3) Unusable Si materials such as shoulder, tail, residue, etc. will increase. φ400mm φ200mm ⇒Higher cost

Weight loss >30% 10%

April　11, 2005 ITWG 15

Concluding Remarks

1. SSi’s activity was a milestone to adapt our concept and/or strategy to the new circumstance.

2. SSi patents and know-how would be promised to improve the 300mm Si production and to the next generation wafer.

3. However, the fulfillment of the next wafer would be questionable without Devices, Equipment, Materials and Wafer each makers’ concrete cooperation and appropriate sharing of monster investment in advance.