In-Line Detection and Measurement of Molecular ...

Characterization and Metrology for ULSI Technology Conference 2005 1

In-Line Detection and Measurement of Molecular Contamination in

Semiconductor Processing Solutions

Jason Wang, Michael West, Ye Han, Bob McDonald, Wenjing Yang,

Bob Ormond and Harmesh Saini

Metara Inc.Sunnyvale, CA, USA


OutlineIntroduction

Why the analysis of molecular contamination is important?

DiscussionUse of the Metara Trace Contamination Analyzer (TCA) for molecular contamination measurementProblem solving examples

1. A nitrogen-containing compound in H2O22. Organic additives in SC-13. Urea in UPW4. Molecular contamination in UPW5. Plasticizers in IPA6. Sulfur-containing compounds in IPA

Conclusion


Molecular Contamination Sources

Process equipmentImpurities in incoming process chemicalsTransfer from earlier process stepsAirborne molecular contaminationDeliberate addition of organics to process chemicals

Surfactants and chelating agents“Proprietary” additives


Gate Oxide Degradation Due To Organic Contamination

* HMDS Monolayer on Oxide

* HMDS =

Hexa-methyl-di-silazane

Cumulative Failure (%)

QBD = Charge-to-Breakdown

measurement

“Cost Effective Cleaning and High-quality Thin Gate Oxides”,IBM J. Res. Develop. Vol. 43, No. 3, May 1999, M. Heyns and et. al.


International Technology Roadmap for Semiconductor (ITRS)

Surface CarbonITRS 2004 Updated, Table 114a Technology Requirements for wafer environmental contamination control

(2003 Edition)

30% H2O2 total oxidizable carbon (ppb) - TBD TBD TBD TBD TBD TBD

IPA: High molecular weight organics (ppb) - TBD TBD TBD TBD TBD TBD

30% H2O2: Resin byproducts (ppb) - TBD TBD TBD TBD TBD TBD

ADD

ADD


TCA (Trace Contamination Analyzer) for Metallic, Organic and Molecular Contaminants

Electrospray Ionization Interface

Mass-Analyzer

Ion-Trap & TOF (Time of Flight)

Sample Sample Preparation Preparation ModuleModule

Mass-Spectrometer

Sample Sample Extraction UnitExtraction Unit

Trace Contaminant Analysis (TCA) Platform

..

..

..

.. ... .

.. .

• Cations

• Anions

• Metallics

• Organics

• 24/7

Bath


Example 1: H2O2 Excursion at a Production Fab

Contaminated H2O2 suspected to cause yield crash at a fab

Analyses of traditional lab methods show inconclusive results between the “Good” & the “Bad” H2O2 samples

ICP-MS (Inductively Coupled Plasma Mass Spectrometry), IC (Ion Chromatography), TOC (Total Oxidizable Carbon) Assay

Results of TCA showed Intensity of peak at m/z 118 was ~ 20x higher in “bad” sample than in “good” sample


Contaminant found from H2O2 Sample by TCA

0

250000

500000

750000

1000000

1250000

1500000

1750000

2000000

50 100 150 200 250 300

MassGEN( 00:03:06 : 00:05:19 )

Am

plitu

de

m/z

Curve 1

“Good” Sample

Intensity of 118 peak was found to be ~ 20x higher in “Bad” sample

0

250000

500000

750000

1000000

1250000

1500000

1750000

2000000

50 100 150 200 250 300

MassGEN( 00:05:19 : 00:07:33 )

Am

plitu

de

m/z

Curve 1

“Bad” Sample118.087


Identification of Contaminant in “Bad” H2O2 Sample

Possible compound: Tri-methyl-glycine

0

100000

200000

300000

400000

500000

600000

700000

25 50 75 100 125 150 175

MassGEN( 00:01:04 : 00:02:32 )

Am

plitu

de

m/z

Curve 1

Suspected TMA (CH3)3NH+

m/z 60.082 m/z 118.087

m/z = 118.087

H+

0

100000

200000

300000

400000

500000

600000

700000

25 50 75 100 125 150 175

MassGEN( 00:01:20 : 00:02:32 )

Am

plitu

de

m/z

Curve 1

Fragment

m/z 59.074

Higher Collision Voltage

m/z = 59.074


Contaminant from Ion Exchange Resin

Structure of Ion Exchange Resin

Tri-methyl-glycine-CH-CH2-CH-

CH2

H3C-N

CH-CH2-CH-

CH3 CH3

A-+

CH2

H3C-N

CH3 CH3

CH3COO -+


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0:00 2:24 4:48 7:12 9:36 12:00 14:24 16:48 19:12 21:36

Time

Peak

Are

a of 1

18 vs

inte

rnal

std

0

1

2

3

4

5

6

7

Peak

Are

a of 1

92 vs

inte

rnal

std

Ratio of 118/internal stdRatio of 192/internal std

Bath Change Bath ChangeBath Change

TCA In-Line Monitoring Molecular Contamination in SC-1 (NH4OH:H2O2:H2O) Bath at a Fab


High Organic Content in High Purity H2O2

5000 ppb max

High Organic Content

Total Organic Carbon (TOC)

Problems

10 ppt max

Low Metallic Content


Example 2: Deliberate Addition of Surfactant or Chelating Agents in Baths

Industrial Trend for Using Diluted Chemistry SC-1 NH4OH:H2O2:UPW from x:1:5 (x = 0.05-3), up to 1:1:500SC2 HCl:H2O2:UPW up to 1:1:1000

Addition of Surfactant or Chelating Agents in Baths

Improve Particle & Metal Removal EfficiencyImprove Surface Wetability for Uniform Wafer Surface Preparation


Chelating Agent found from a SC-1 (NH4OH:H2O2:H2O) by TCA

0

10000

20000

30000

40000

50000

60000

70000

50 100 150 200 250 300 350 400 450 500

MassGEN( 00:00:48 : 00:02:00 )

Am

plit

ud

e

m/z

Curve 1

SC-1 Bath

14

14

16

1416

1414

14

19

C3H7COO-

87.044

Fragment CH2

Fragment O

4644

44

CH2CH2O = 44CH2 = 14O = 16F = 19?

Possible Formulation:

A Mixture of Compounds with Functioning Groups R-COO-

and -(CH2CH2)n-O

Chelating agents/surfactants Need to be completely rinsed off from wafer surface


Example 3: TOC (Total Oxidizable Carbon) Excursions at a Fab

Seasonal TOC excursions at a fabUrea (fertilizer) in UPW (Ultrapure Water) was suspected“No way to confirm suspicions” because “no laboratory methods available to accurately measure low ppb concentrations of urea contamination in water”*

*Ref. “Undetectable TOC in UPW can influence DUV photolithography processes”, J. Rydzewski& R. Godec, proceedings of SPWCC 2002 (Semiconductor Processing with Wet Chemicals Conference)


TCA Quantitative Analysis of Urea in UPW

0

500

1000

1500

2000

2500

3000

3500

25 50 75 100 125 150

MassGEN( 00:04:24 : 00:05:36 )

Am

plitu

de

m/z

Curve 1

0

500

1000

1500

2000

2500

3000

3500

25 50 75 100 125 150

MassGEN( 00:00:40 : 00:01:52 )

Am

plitu

de

m/z

Curve 1

0

500

1000

1500

2000

2500

3000

3500

25 50 75 100 125 150

MassGEN( 00:01:20 : 00:02:32 )

Am

plitu

de

m/z

Curve 1

(NH2)2COH+

m/z = 61.041

(NH2)2COH+

(NH2)2COH+

0 ppb Urea

5 ppb Urea

10 ppb Urea

y = 1211x + 492R2 = 0.9927

0

1000

2000

3000

4000

5000

6000

7000

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5ppb Urea as C12

Pea

kH

eigh

t

DL = 3σ/m = 0.084 ppb C = 84 ppt C, n = 6


TCA Analysis of Urea by Ratio Technique

0

250

500

750

1000

1250

1500

1750

2000

55.0 57.5 60.0 62.5 65.0 67.5 70.0 72.5 75.0 77.5 80.0

MassGEN( 00.40275 - 01.60391 )

Am

plitude

m/z

Curve 1

0 ppb 12C Urea + 10 ppb 13C Urea

13C Urea

12C Urea

0

250

500

750

1000

1250

1500

1750

2000

55.0 57.5 60.0 62.5 65.0 67.5 70.0 72.5 75.0 77.5 80.0

MassGEN( 00.80258 - 02.00375 )

Am

plitude

m/z

Curve 1

13C Urea12C Urea


0

250

500

750

1000

1250

1500

1750

2000

55.0 57.5 60.0 62.5 65.0 67.5 70.0 72.5 75.0 77.5 80.0

MassGEN( 00.80308 - 02.00425 )

Am

plitude

m/z

Curve 1

13C Urea12C Urea



Ratio Measurement of 12C Urea/13C Urea

m/z = 61 (C12)

0

500

1000

1500

2000

2500

3000

0 5 10 20Urea C12 (ppb)

Peak

Inte

nsity

m/z = 62 (C13)

0

200

400

600

800

1000

1200

1400

1600

10 10 10 10Urea C13 (ppb)

Peak

Inte

nsity

m/z = 61 (C12)

0

500

1000

1500

2000

2500

3000

0 5 10 20Urea C12 (ppb)

Peak

Inte

nsity

m/z = 62 (C13)

0

200

400

600

800

1000

1200

1400

1600

10 10 10 10Urea C13 (ppb)

Peak

Inte

nsity

Ratio of 61/62

0.000

0.500

1.000

1.500

2.000

2.500

0ppbC12/10ppbC13 5ppbC12/10ppbC13 10ppbC12/10ppbC13 20ppbC12/10ppbC13

Rat

io


TCA Analytical Results of Urea

( Asp – Ratio ×Bsp )( Ratio ×Bs – As )

( )Cs = CspVsp

Vs

TCA Automatic Quantification

Calculated Results (10ppb C13 Urea Spike)

0.0

5.0

10.0

15.0

20.0

25.0


Cal

cula

ted

Res

ults

(ppb

)

Well Quantified at ppb level

Calculated Results (10ppb C13 Urea Spike)

0.0

5.0

10.0

15.0

20.0

25.0


Cal

cula

ted

Res

ults

(ppb

)

Accurately Quantified at ppb level


Example 4: Molecular Contamination in Pre-Gate Cleaning Processes at a Fab

Yield problems at a FabGate oxide breakdown voltage reduction

Results of TXRF and VPD-ICP/MSNo metallic contamination

Organic or molecular contamination was proposedNo significant suspect by routine lab methods


Contaminants Found from HQDR (Hot Quick Dump Rinse) UPW by TCA

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

1000000

25 50 75 100 125 150 175

MassGEN( 00:00:32 : 00:01:44 )

Am

plitude

m/z

Curve 1

0

250000

500000

750000

1000000

1250000

1500000

1750000

2000000

25 50 75 100 125 150 175

MassGEN( 00:00:32 : 00:01:52 )

Am

plitude

m/z

Curve 1

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

25 50 75 100 125 150 175

MassGEN( 00:01:04 : 00:02:16 )

Am

plitude

m/z

Curve 1

HQDR UPW_2

HQDR UPW_3

HQDR UPW_4NMP-H+

100.076

Scale Different

(H3PO4)H+

98.987

Urea (NH2)2COH+

m/z = 61.040


Possible Contamination Sources in Fab

NMP = N-Methyl-2-pyrrolidone (C5H9NO m/z = 99.068)

• Photoresist stripper• Wafer cleaning• Semi-aqueous defluxing• Degreasing• Coatings (polyamide, epoxy, & polyurethane)

H3PO4 Incomplete rinse from nitride etching?

Urea Source water, re-cycling or reclaimed water?


TCA In-Line Monitoring Phosphorus Species in SC-1 Bath at a Production Fab

0

0.2

0.4

0.6

0.8

1

1.2

0:00 2:24 4:48 7:12 9:36 12:00 14:24 16:48 19:12 21:36

Time

Rat

io

Bath Change Bath Change Bath ChangeBath Change

Proposed as H3PO4NH4+ m/z=106.011


Example 5: Phthalate (Plasticizer) Contamination in IPA (Isopropyl Alcohol)

It has been reported:

Dibutyl phthalate (DBP) in high density polyethylene (HDPE) containers leaching into IPAPlasticizers deposit inside the gas nozzles and chamber in a dryer

Phthalates have deleterious effects on wafers


TCA Analysis of Dioctyl Phthalate (DOP) and Dibutyl Phthalate (DBP) in IPA

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

200 225 250 275 300 325 350 375 400 425 450

MassGEN( 00:00:40 : 00:01:52 )

Am

plitude

m/z

Curve 1

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

200 225 250 275 300 325 350 375 400 425 450

MassGEN( 00:00:32 : 00:01:44 )

Am

plitude

m/z

Curve 1

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

200 225 250 275 300 325 350 375 400 425 450

MassGEN( 00:00:32 : 00:01:44 )

Am

plitude

m/z

Curve 1

DOP H+

m/z=391.285DBP H+

m/z=279.160

Blank

5 ppb

10 ppb

DOP H+

m/z=391.285

DOP H+

m/z=391.285

DBP H+

m/z=279.160

DBP H+

m/z=279.160


Example 6: Molecular Contamination in IPA (Isopropyl Alcohol) at a Fab

0

50000

100000

150000

200000

250000

300000

350000

400000

25 50 75 100 125 150 175 200

MassGEN( 00.80250 - 02.00366 )

Am

plitude

m/z

Curve 1

0

50000

100000

150000

200000

250000

300000

350000

400000

25 50 75 100 125 150 175 200

MassGEN( 00.67033 - 01.87158 )

Am

plitude

m/z

Curve 1

IPA lot 1, 2, 3

(CH3)2CHOSO3-

m/z = 139.007

• IPA from lot 1,2 3 suspected causing multiple excursions

• No significant difference between lot 1,2,3 and lot 4,5,6 by routine lab methods

(CH3)2CHOSO2-

m/z = 123.012

HSO4-

m/z = 96.960

IPA lot 4, 5, 6


Contamination Sources from IPA Manufacturing Processes

di-isopropyl sulfate

(CH3)2CH CH(CH3)2O-S-O=O

=O

isopropyl sulfate

+ (CH3)2CH

m/z = 139.007

_O-S-O=O

=O

propene gas

CH3CH=CH2 + H2SO4

hydrolysed w/ H2O

(CH3)2CHOH

IPAFinal Product =

Isopropyl sulfate is a highly suspect in IPA lot 1, 2, 3


Summary

Using the TCA we have demonstrated the ability to:

Analyze molecular contamination in a variety of process solutions including UPW, SC-1 and IPAIdentify specific molecular contaminantsProvide quantitative concentration measurementMake measurements in-line and in near real-timeProvide process chemistry trends that correlate to Wafer Fab yield problems


Conclusion

We believe this new measurement capability will:

Enable statistically valid real-time process chemistry control decisions Provide advanced warning of excursionsEnable chemical specifications and bath life decisions based on process data and yield correlation

Date post:	15-Apr-2022
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

In-Line Detection and Measurement of Molecular ...

Documents