process impacts of chemistry preparation on dhf process · pdf filef 1043 - version 01 ....
TRANSCRIPT
F 1043 - Version 01 . AP&S International GmbH
Process impacts of chemistry preparation on DHF process
Thomas Kopp
Levitronix User Conference
Zürich, April 11th 2013
AP&S International GmbH
F 1043 - Version 01 . AP&S International GmbH
Agenda
Company
Products & Applications
Products - Impressions
Motivation
Initial situation
Root cause analysis
New solution
Test results
Summary
F 1043 - Version 01 . AP&S International GmbH
Worldwide Locations in
Europe & Asia
AP&S International GmbH
Obere Wiesen 9
D-78166 Donaueschingen
Germany
www.ap-s.de | www.ap-s.cn | www.ap-s.sg | www.semi-parts.com
Company
F 1043 - Version 01 . AP&S International GmbH
Products & Applications
Clean
• SC1
• TMAH/H2O2
• DSP
• EKC / ACT a.o.
• SOM, DIO3
• SPM
• Scrubber
• Post CMP
• Megasonic
• Etc.
Etch
• DHF
• BOE
• HF 49%
• DHF/HCI
• KOH
• H3PO4
• P-Etch
• Etc.
Metal etch
• Poly-Si
• Alu
• Ti, TiN
• Ni, NiCr
• BiSb
• Co
• CoSi
• Cu
• Etc.
PR Strip
• SPM
• SOM
• Etc.
Eless Plating
• Nickel-Gold
• Nickel-Palladium-Gold
Lift-off
• Metal • Etc.
Drying
• Marangoni • Spin • Hot air
Coating and Developing
• Positive resist • Negative resist • SU8
Wet bench
Single wafer
Parts cleaning
Facility
Refurbishment
Lotus systems products
Customized solutions
F 1043 - Version 01 . AP&S International GmbH
Products - Impressions
F 1043 - Version 01 . AP&S International GmbH
DHF process in semiconductor industry
DHF ≡ Diluted Hydro-Fluoric Acid
DHF is used in various applications in semiconductor industry. Most important applications are:
• Removal of natural Si-Oxide from wafer surface (natural etch
stop when Si-Oxide is removed)
• Etching of a defined layer-size of Si-Oxide (controlled by time)
• Saturation of wafer surface with hydrogen
Motivation
F 1043 - Version 01 . AP&S International GmbH
Motivation
Process parameters
∆ 𝐸𝑡𝑐ℎ 𝑟𝑎𝑡𝑒 ~ 𝑝 ∗ ∆ 𝑇
- Etch rate - Uniformity
tim
e
etchrate vs. DHF concentartion
0
5
10
15
20
25
30
0 20 40 60 80 100 120 140 160 180 200
mixing ratio HF:H2O (1 : X)
etc
h r
ate
[n
m/m
in]
Etch rate vs. DHF concentration
0
20
40
60
80
100
120
0 200 400 600 800 1000
etch rate vs. process time
∆ 𝐸𝑡𝑐ℎ 𝑟𝑎𝑡𝑒 ~ 𝑝 ∗ ∆ 𝑇
F 1043 - Version 01 . AP&S International GmbH
Motivation
In standard wet process equipment, mostly the SiO2-layer removal is controlled by the etch time (process time). This requires a very exact etch rate and therefore a stable status for:
• DHF concentration
• DHF mixture homogenity
• Temperature
F 1043 - Version 01 . AP&S International GmbH
Initial situation - Problem
Problems, that could be seen in this process:
• Etch rate fluctuation over days; the etch rate differs between different days
• Etch rate fluctuation over lots; the etch rate differs between different lots
• Uniformity of etched wafer surface within a wafer or within one lot of wafers doesn‘t fulfill the requirements
F 1043 - Version 01 . AP&S International GmbH
Initial situation - Hardware setup
Preparation
tank
5% HF dilution
DIW
5% HF dilution
0,5%/0,2%
HF dilution
Spiking pump
Static mixer
Static immersion tank
Facility supply
F 1043 - Version 01 . AP&S International GmbH
Initial situation – Measurment results
Fluctuation of oxide-removal day to day
Etch rate fluctuation day to day
8
8.5
9
9.5
10
10.5
11
avg. etch rate [A°] vs. different days [date] A°/min
∆ ~ 𝟐𝟎 𝑨°
Avg. Etch [A°] rate vs. Process time [sec] – Linearity curve
Std.dev.: > 0,5
F 1043 - Version 01 . AP&S International GmbH
Root cause analysis
Following possible root causes where verified:
• Process time – OK; no fluctuation
• Temperature fluctiation – OK; no fluctuation that could cause the fluctuation of the etch rate
• HF concentration preparation tank (5%) – OK; measured with online analytic measurement device
• HF concentration in bath (0,5%/0,2%) – NOK; fluctation observed
• DIW supply – NOK; significant fluctuation of DIW input pressure & flow observed
F 1043 - Version 01 . AP&S International GmbH
Root cause analysis - Measurement results
6.000
7.000
8.000
9.000
10.000
11.000
12.000
16:1
8:5
6
16:1
9:0
4
16:1
9:1
2
16:1
9:1
9
16:1
9:2
7
16:1
9:3
5
16:1
9:4
3
16:1
9:5
0
16:1
9:5
8
16:2
0:0
6
16:2
0:1
3
16:2
0:2
1
16:2
0:2
9
16:2
0:3
6
16:2
0:4
4
16:2
0:5
2
16:2
0:5
9
16:2
1:0
7
Series1
Fluctuation of DIW input flow was caused by very high fluctuation of DIW input pressure at process module between approx.
1 bar and 3,5 bar.
DIW Flow [l/min]
F 1043 - Version 01 . AP&S International GmbH
New solution - requirements
1. Suitable for wide input pressure range 0,8 to 3,7 bar
2. Variable mixing rates (500:1; 200:1)
3. Repeatability of Mixing Ratio ±1%
4. Absolute Accuracy of Mixing Ratio ±2%
5. Compact design
6. Extremely low particle generation
7. Homogeneous mixture of DIW and 5% DHF dilution
8. Smooth, continuous flow without pressure pulsation
F 1043 - Version 01 . AP&S International GmbH
New solution (1) – Pre-preparation tanks
Advantages: • Accuracy an Repeatability • Independend of DIW input pressure
Disadvantages: • No compact design • No mixing function/inhomogeneous mixture of HF and DIW • For variable mixing rates a) different pre-mixing tanks or b) waste of chemistry
Pre mixing of Chemistry in additional tanks
5%HF DIW
0,5
% H
F d
ilu
tio
n
Prep- tank 0.5% HF
5%HF DIW
0,2
% H
F d
ilu
tio
n
Prep- tank 0.2% HF
Static immersion tank
F 1043 - Version 01 . AP&S International GmbH
New solution (2) – Pressure regulator
Advantages: • Compact design • Very stable output pressure
Disadvantages: • Only stable output pressure within a small range
of input pressure variation • No linearity over complete area of operation • No mixing function
Pressure regulator for DIW
F 1043 - Version 01 . AP&S International GmbH
New solution (3) – Blending system
Single Pump system for DIW
Control loop
Advantages: • Compact design • Very stable flow
Disadvantages: • Only stable output flow for
one fluid • No mixing function
F 1043 - Version 01 . AP&S International GmbH
New solution (4) – Blending system
Disadvantages: • Costs
Double Pump system for DIW&HF5%
Advantages: • Compact design • Very stable flow • Two flows
controlled • Mixing function
to static immersion tank
F 1043 - Version 01 . AP&S International GmbH
New solution – Overview
Solution DIW input pressure 0,8 – 3,7 bar
Variable mixing rates
Repeatability / Accuracy
Compact design
Mixing function
Pre-preparation tank OK Premixing not
depending on DIW pressure
a) different tanks for pre mixture
b) Waste of chemistry in case of using one tank
NOK NOK NOK
Precision pressure regulator
For stable output pressure 2,5 – 4,5
bar required
OK (spiking pump)
OK OK NOK
Blending system with one pump
OK OK NOK Only one chemistry
flow controlled.
OK NOK
Blending system with two pumps
OK OK OK OK OK
F 1043 - Version 01 . AP&S International GmbH
New solution – Decision
Based on:
• system requirements
• specifications of blending systems with one and two control loops
Decision:
Solution 4 (Blending System with two pumps/control loops) will be used for further tests.
Hardware:
LeviFlowsensors (Levitronix): LFS-50-Z;LFS-04-Z; 2x LFC-1C
Pumps (Levitronix): BPS-600.2; BPS-200.2; LPC-600; LPC-200
F 1043 - Version 01 . AP&S International GmbH
Test results – Etch rate stability
7
7.5
8
8.5
9
9.5
10
13.03.2010 18.03.2010 23.03.2010 28.03.2010 02.04.2010 07.04.2010 12.04.2010 17.04.2010 22.04.2010
Avg. Etch [A°] rate vs. different days [date] with Blending System A°/min
Standard deviation: 0,18
8
8.5
9
9.5
10
10.5
11
Avg. etch [A°] rate vs. different days [date]- original A°/min
Etch rate stability over different days can be improved significantly
Std.dev.: < 0,18
Std.dev.: > 0,5
F 1043 - Version 01 . AP&S International GmbH
Test results – Etch rate linearity
0
20
40
60
80
100
120
140
160
0 100 200 300 400 500 600 700 800 900 1000
Oxid
e l
os
s (
A°)
Process time (sec)
Avg. Etch [A°] rate vs. Process time [sec] with Blending System ( Linearity curve)
Week 14
Week 18
Stability of oxide-removal can be improved significantly
Avg. Etch [A°] rate vs. Process time [sec] – Original (Linearity curve)
∆ ~ 20 𝐴°
∆ < 1 𝐴°
F 1043 - Version 01 . AP&S International GmbH
Summary
Etching a defined oxide layer in a DHF process, only controlled by time, requires an absolute stable process environment.
The blending system in this test showed
• Repeatability of Mixing Ratio better than ±1%
• Absolute Accuracy of Mixing Ratio better than ±2%
• Homogeneous mixture of DIW and 5% DHF dilution guaranteed
• Variable mixing rates (requires fluid calibration)
Thanks to Levitronix for providing hardware and technical support for these tests !
F 1043 - Version 01 . AP&S International GmbH
Thank you…
Please, contact us for
more information!
Europe
AP&S International GmbH
Obere Wiesen 9 . Aasen
78166 Donaueschingen . Germany
Phone +49-771-8983-0
Fax +49-771-8983-100
www.ap-s.de
China
AP&S Semiconductor Equipment
(Shanghai) Co., Ltd.
Phone +86-21-5042-9053 ext. 107
Fax +86-21-5042-9053 ext. 105
www.ap-s.cn
Europe
SemiParts
An AP&S Business Unit
Obere Wiesen 9 . Aasen
78166 Donaueschingen . Germany
Phone +49-771-8983-200
Fax +49-771-8983-100
www.semi-parts.com
Singapore/Malaysia
AP&S Asia Pte. Ltd.
Phone +60-403-9007
Fax +60-408-0031
www.ap-s.sg