Arcing prevention by dry clean optimization at Shallow

01.03.2001Page 1

Michael HartenbergerUniversity of Cottbus

Lars ChristophAndreas SteinbachCenter forDevelopment andInnovation

Infineon TechnologiesDresden

ProjectPULSAR

APCAPC

2nd AEC/APC Conference Europe, April 18th-20th 2001, Dresden

Arcing prevention by dry clean optimizationat Shallow Trench Isolation (STI) Etch inAMAT MxP by use of plasma parameters

Michael Hartenberger [email protected]

University of Cottbus www.tu-cottbus.de

Lars Christoph [email protected] Steinbach [email protected] for Development and InnovationInfineon Technologies Dresden www.infineon.com

01.03.2001Page 2




ProjectPULSAR

APCAPC


Acknowledgement

q The authors of this poster would like to thankSiegfried BernhardInes UhligAstrid KretschmarThomas KuenzelmannInfineon Technologies Dresden, Germany

q The work for this paper was supportedØ by the EFRE fund of the European CommunityØ and by funding of the State Saxony of the Federal Republic of

Germany, project PULSAR, project number 5706.

q The authors are responsible for the content of the paper.

Acknowledgement

01.03.2001Page 3




ProjectPULSAR

APCAPC


Outline

q Introduction –Shallow Trench Isolation (STI) etch process

q Arcing at STI etch in AMAT MxP chamberq SEERS measurement techniqueq Development of a new dry clean using plasma

parameters for real time process monitoringØ Basic experiments (not shown in this publication)Ø Optimization of clean stepØ Optimization of conditioning stepØ Optimization of clean-/conditioning timeØ Long term evaluation

q Summary

Outline

01.03.2001Page 4




ProjectPULSAR

APCAPC


STI etch process - Overview

ResistARCSiN

Si

deeptrench

a) Starting profile b)

ResistARCSiN

Si

Mask-open step

ResistARCSiN

Si

c) Trench etch step

SiN

Si

Oxid

d) Oxide CVD f) A-B:top-CD bottom-CDOxide

SiN

Si

Oxide

e) Planarization

A

B

Introduction - STI etch process

01.03.2001Page 5




ProjectPULSAR

APCAPC


Arcing – what´s that ?

q Arcing in terms of technology:= higher particle density caused by aluminum particles

q Arcing in terms of physics:= temporary local instability of plasma

Arcing at STI etch in MxP

Fig.: Arcing generatedAluminum particle on wafersurface

01.03.2001Page 6




ProjectPULSAR

APCAPC


Why Arcing at STI etch in our AMAT MxPchambers appeared

q Basic reasons for arcing:Ø Hardware defects (i.e. scratches)Ø Normal waste of anodizationØ Potential differences between chamber partsØ Critical process conditions (i.e. high RF power, polymers)

q Main arcing reason at STI etch in our MxP chambers:Ø Change of product mix: a new product uses a high polymerizing

etch chemistryè The usual dry clean was not efficient enough for the new

process mix, thicker polymer layers at the chamber wallè Arcing !


01.03.2001Page 7




ProjectPULSAR

APCAPC


How to get rid of arcing at STI etch in AMAT MxPchamber ?

q Extending clean time or increasing gasflows for the standardclean recipe is not a viable solution. It caused two effects:Ø Less arcing, yes.Ø But also a very strong first wafer effect on critical dimensions of

product wafers, which could not be tolerated !

è Development of a new dry clean recipe was necessary:ØWith effective polymer reduction on chamber walls.Ø And with small tolerable CD degradation on product wafers.

è Development of a new dry clean using a different gas mixture.


01.03.2001Page 8




ProjectPULSAR

APCAPC


Strategy to develop a new dry clean

q 1. Basic experiments:Ø Test of other gas mixtures, e.g. CF4/O2, using resist test wafersØ Target: High polymer etch rate

q 2. Evaluation of the new dry clean on product wafers:Ø Target: First wafer effect (critical dimensions !) negligible.

q 3. Long term evaluation of new dry clean on product wafers:Ø Target: New dry clean must be efficient enough

q Optimization between long term clean effect and first wafereffect on critical dimensions

q For the efficient development of the new dry clean we useda „quick and dirty“ real time in- situ process monitor -SEERS.

Development of new dry clean

01.03.2001Page 9




ProjectPULSAR

APCAPC


Principle of SEERS

q Self Excited Electron PlasmaResonance SpectroscopyØ Passive electrical methodØ Integral physical parameters

RF currentRF voltage

RF currentRF voltage

FFTFFT

Model SEERSModel SEERS

Electron collision rateElectron density

Bulk powerDC bias voltage

Electron collision rateElectron density

Bulk powerDC bias voltage


01.03.2001Page 10




ProjectPULSAR

APCAPC


Experiments to optimize the new dry clean

q The new CF4/O2 dry clean recipe consists of two parts (like old one):Ø 1. Clean step – to remove the polymerØ 2. Conditioning step – to minimize the first wafer effects on the following

product wafersq Experiments:

Ø Optimization of clean step time, without subsequent conditioning step,using resist test wafers to monitor first wafer effect

Ø Application of standard conditioning step, using product wafersØ Fine tuning of conditioning time / recipe parameters on product wafersØ Long term evaluation of new dry clean

q Targets of all these optimization experiments:Ø Maximize the polymer reduction at chamber wallØ Minimize the first wafer effect on critical dimensions


01.03.2001Page 11




ProjectPULSAR

APCAPC


electron collision rate vs. time [median]

0.00E+00

5.00E+06

1.00E+07

1.50E+07

2.00E+07

2.50E+07

3.00E+07

14:52 15:21 15:50 16:19 16:48 17:16 17:45

time

med

ian

ele

ctro

n

colli

sio

n r

ate

[s-1

]

2.30E+07

2.80E+07

3.30E+07

3.80E+07

4.30E+07

4.80E+07


0.00E+00

5.00E+06

1.00E+07

1.50E+07

2.00E+07

2.50E+07

3.00E+07

15:21 15:50 16:19 16:48 17:16 17:45

time

med

ian

ele

ctro

n

colli

sio

n r

ate

[s-1

]

2.30E+07

2.80E+07

3.30E+07

3.80E+07

4.30E+07

4.80E+07

Optimization of clean step time(without conditioning step) using test wafers

q Clean 1 is longer than Clean 2q Clean step is performed without following conditioning step to monitor the

impact of the cleaning process on chamber conditions.q First wafer effect depends on cleaning process step time:

ØDry clean 1: first wafer effect on 2 ... 4 following resist dummiesØDry clean 2: first wafer effect on 1 following resist dummy only

Clean

Resist Dummy

One point – one waferOne point – one wafer

Optimization of clean step

1 2

01.03.2001Page 12




ProjectPULSAR

APCAPC


Conclusions fromoptimization of clean step time

q Dry clean 2:Ø Has a significant lower impact on chamber conditions.Ø Therefore we decided to use it for further optimization.

q Conditioning step after cleaning process:Ø Is applied to remove the first wafer effect completely.Ø But in case of to much conditioning the whole clean becomes

inefficient, arcing could appear again.è Finally long term tests of new cleaning process are necessary.

Optimization of clean step

01.03.2001Page 13




ProjectPULSAR

APCAPC


top-CD vs. time [mean]

17:45 18:14 18:43 19:12 19:40 20:09 20:38 21:07

time

top-

CD


3.00E+07

3.50E+07

4.00E+07

4.50E+07

5.00E+07

5.50E+07

6.00E+07

17:45 18:57 20:09 21:21

time

med

ian

ele

ctro

n

colli

sio

n r

ate

[s-1

]

2.30E+07

2.70E+07

3.10E+07

3.50E+07

3.90E+07

Application and optimization of conditioning step

§ Squares in diagrams represent first product wafer after each dry clean

q Electron collision rate indicates first wafer effect on product wafersq Electron collision rate also indicates drift of chamber conditions

(decline of average represented by falling line in diagram)q First wafer effect is also indicated by critical dimensions, measured

on product wafers

Clean


Product

Optimization of conditioning step

01.03.2001Page 14




ProjectPULSAR

APCAPC


Conclusions fromapplication of conditioning step

q Tested on product wafersq First wafer effect, caused by dry clean, is indicated by electron

collision rate.q Additionally electron collision rate indicates drift of chamber

conditions (declining average).q First wafer effect is indicated by critical dimensions as,

measured on product wafers.q Chosen dry clean process with standard conditioning step has

still significant impact on critical dimensions at STI etch.èConditioning time and / or conditioning recipe has to be

changed to reduce influence on critical dimensions.


01.03.2001Page 15




ProjectPULSAR

APCAPC



2.80E+07

3.00E+07

3.20E+07

3.40E+07

3.60E+07

3.80E+07

14:52 15:21 15:50 16:19 16:48 17:16 17:45 18:14 18:43 19:12

time

med

ian

ele

ctro

n

colli

sio

n r

ate

[s-1

]

Modified conditioning step recipe


q Now conditioning step has no CF4/O2, but clean-step is still with CF4/O2

è No significant first wafer effect after each dry cleanØ Now dry clean does not seem to have a big influence on critical dimensions &

electron collision rate

q But: is dry clean still as effective?Ø (Polymers are generated in the conditioning step and without CF4 / O2 even moreè possibly the clean effect is spoiled by the following conditioning step ?)

DRAM product


top-CD vs. time [mean]

14:52 15:21 15:50 16:19 16:48 17:16 17:45 18:14 18:43 19:12

time

top

-CD

[n

m]


01.03.2001Page 16




ProjectPULSAR

APCAPC


electrondensity vs. time [median]

3.00E+08

3.40E+08

3.80E+08

4.20E+08

4.60E+08

5.00E+08

14:24 14:52 15:21 15:50 16:19 16:48 17:16

time

med

ian

ele

ctro

n

den

sity

[cm

-3]

1.80E+09

1.84E+09

1.88E+09

1.92E+09

1.96E+09

2.00E+09

electron density vs. time [median]

3.00E+08

3.40E+08

3.80E+08

4.20E+08

4.60E+08

5.00E+08

14:24 14:52 15:21 15:50 16:19 16:48 17:16

time

med

ian

ele

ctro

n

den

sity

[cm

-3]

1.80E+09

1.84E+09

1.88E+09

1.92E+09

1.96E+09

2.00E+09

Clean- and conditioning time optimizationlong vs. short conditioning


q Optimization from economic point of view: Influence ofcleaning- and conditioning time on chamber conditions

q Longer conditioning time does not seem to cause changes intrend of electron density for following resist wafers here

long clean / long conditioning long clean / short conditioning


Time optimization

01.03.2001Page 17




ProjectPULSAR

APCAPC



3.00E+08

3.40E+08

3.80E+08

4.20E+08

4.60E+08

5.00E+08

17:24 17:52 18:21 18:50 19:19 19:48 20:16

time

med

ian

ele

ctro

n

den

sity

[cm

-3]

1.80E+09

1.84E+09

1.88E+09

1.92E+09

1.96E+09

2.00E+09


3.00E+08

3.40E+08

3.80E+08

4.20E+08

4.60E+08

5.00E+08

14:24 14:52 15:21 15:50 16:19 16:48 17:16

time

med

ian

ele

ctro

n

den

sity

[cm

-3]

1.80E+09

1.84E+09

1.88E+09

1.92E+09

1.96E+09

2.00E+09

Clean- and conditioning time optimizationlong vs. short clean


q Shorter clean time reduces first wafer effect on following resisttest wafers

q But shorter clean time means less polymer reduction atchamber walls !

long clean / short conditioning short clean / short conditioning


Time optimization

01.03.2001Page 18




ProjectPULSAR

APCAPC


Long term evaluation of dry clean

q Long term evaluation of new dry clean in normal productionshows no significant influence of dry clean on electroncollision rate of DRAM- product wafers

q Only normal scattering (maybe because of pre- processtolerances) in electron collision rate

Long term evaluation

1.00E+07

1.50E+07

2.00E+07

2.50E+07

3.00E+07

3.50E+07

4.00E+07

4.50E+07

5.00E+07

19.12.00 19.12.00 19.12.00 20.12.00 20.12.00

Med

ian

Ele

ctro

n C

olli

sio

n

Rat

e [s

-1]

21.12.00 21.12.00 21.12.000.00E+00

2.00E+06

4.00E+06

6.00E+06

8.00E+06

1.00E+07

1.20E+07

1.40E+07

1.60E+07

1.80E+07

2.00E+07

Date One point – one wafer

DRAM product

Dry clean

Electron Collision Rate [Median] vs. Date (IT-Mask-Step)

01.03.2001Page 19




ProjectPULSAR

APCAPC


Dry clean impact on critical dimensions -Comparison of old and new dry clean

q New dry cleanØ Critical dimensions in

spec tooØ New dry clean has no

significant influence oncritical dimensions

q Old dry cleanØ Critical dimensions in

spec

Long term evaluation

CD width vs. wafer

200 220 240 260 280 300wafer

CD

wid

th

DRAM product lower Spec upper Spec

CD width vs. wafer

700 720 740 760 780 800wafer

CD

wid

th

DRAM product lower Spec upper Spec

01.03.2001Page 20




ProjectPULSAR

APCAPC


Summary I: Benefit of plasma parameters

q Electron collision rate and electron density have been used tooptimize a new clean recipe at Shallow Trench Isolation etch inAMAT MxP chamber.

q Both plasma parameters show in real timeØ Dry clean impact on chamber conditions and first wafer effect on

product wafersØ And superimposed long term drift effects of chamber conditions

q Because plasma parameters do indicate chamber conditiondrifts, they can be used to monitor cleaning efficiency withrespect to wall polymers.

q Therefore plasma parameter measurements can significantlyhelp to improve efficiency & reduce costs of dry clean processdevelopment.

Summary

01.03.2001Page 21




ProjectPULSAR

APCAPC


Summary II: Dry clean optimization

q Benefits of new dry clean:Ø Has no significant impact on critical dimensions of following

product wafers (as shown in short- and long- time observations)Ø Is shorter than old dry clean è higher throughputØ Better clean efficiency than old dry clean (as observed during

chamber opening for maintenance purposes)Ø Chamber up time has been extended by longer dry clean period

è All targets of dry clean optimization (maximize polymerreduction from walls and minimize first wafer effect on criticaldimensions) have been reached in short period of time usingin-situ plasma monitoring techniques.

è Risk of arcing reduced.

Summary

Documents

Arcing prevention by dry clean optimization at Shallow