and Metal Gate Applications - NCCAVS Usergroups · Hour 1.52 A165 Hour 2.53 A165 Hour 3.54 A165 Hour 5.55Hour 3.68 A165 0 Hour 7.56 100 200 300 400 500 600 700-150)-100 -50 0 50 100

3M Electronics & Energy Business Group

1 17 March 2014 . All Rights Reserved. © 3M

3MTM TrizactTM Pad Conditioners for FEOL Tungsten Plug

and Metal Gate Applications

March 19, 2014

J. Zabasajja, V. Laraia, M. Fritz, J. Sokol and Junqing Xie 3M Electronics Markets Material Division, St. Paul, MN 55144, USA,



W plug process with TrizactTM B5-M990-5S2

― 30 hour marathon ( TOX dummy wafers and W monitor wafers measured every hour),

• Consumable set B (ex-situ conditioning) :

– Slurry B/Pad B with

3M Trizact B5-M990-5S2 disk (commercial)

Sintered abrasive A165 disk ( POR)

W metal gate process with TrizactTM B6-M990-5S2

― 20 hour marathon ( TEOS and W monitor wafers measured every hour)

• Consumable set B ( ex-situ conditioning):

– Slurry B/Pad B with

3M Trizact B6-M990-5S2 ( experimental)

Sintered abrasive A165 disk ( POR)

3M™ Trizact™ Pad Conditioner Marathon Testing



Conditioner Description

• B5-M990-5S2 (MC3113)

Nominal Primary Tip height 160µm

~5% Primary tips

Complex Multi-faceted structure

Aspect ratio: sharp

Commercial Experimental

• B6-M990-5S2 (MC3112)

Nominal Primary Tip height 160µm

~5% Primary tips

Complex Multi-faceted structure

Aspect ratio: sharper



TrizactTM B5-M990-5S2 and B6-M990-5S2 PWR and SF BenchTop Data

PWR, SF: Trizact B6 > Trizact B5 .

0

1

2

3

4

5

6

7

8

0

200

400

600

800

1000

1200

1400

B5-M990 B6-M990 MC3112

Sa

(µm

)

PW

R (

µm

/ho

ur)

DIW Pad Cut Test, Benchtop Pad, 6lbs DF, Benchtop Tool

PWR(µm/h)

Sa (µm)



Polish Head and Platen Settings

Platen Speed 73 rpm

Head Speed 67 rpm

High wafer df RR=3.8, Z1=3.3, Z2=1.6,

Z3=1.4, Z4=1.3, Z5=1.3 psi

Slurry Flow Rate 300 mL/min

Head Sweep Sinusoidal sweep, 10 zones

Zones 7.00 to 8.02 inches

Slurry W slurry B

Pad Pad B

Polish Head Contour

Dummy Wafers Thermal Oxide

Rate Monitor W wafer

Tool Conditions (Bulk W Test) (P1 and P2) – 300 mm AMAT Reflexion™

Conditioner Settings

Head Speed 81 rpm

Sweep Linear sweep, 19 sweeps/min

Zones 2.5" (a) to 12.75" (b)

Conditioning Time 60 secs (100% ex-situ)

Cond downforce 9 lbs

Break-in Cond downforce 9 lbs for 30 mins (in DIW, ex-situ)

Note (a) may require adjustment to avoid

collision with polish head

Note (b) disk must not travel beyond 12.75"

All consumables and parameters including the downforce on the wafer ( 1.3 psi vs

2.5 psi) for this test selected based on customer’s input



Marathon Test With Slurry B and Pad B TrizactTM B5-M990-5S2 and SA A165

(ex-situ)

W Slurry B and Pad B ( ex-situ conditioning) – Longer life and more stable RR/NU

with Trizact B5 vs SA A165.

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0

100

200

300

400

500

600

700

800

900

0.00 5.00 10.00 15.00 20.00 25.00 30.00

No

n-U

nif

orm

ity

( %

)

W R

emo

val R

ate

( A/m

in)

Total Number Of Conditioning Hours (hrs)

Extended Run Process Results: W RR and NU, 3M Trizact B5, SA A165, 9 lbs CDF, 1.4 psi Wafer df, W Slurry B, Pad B

Trizact B5 RR SA A165 RR Trizact NU SA A165 NU



Pad Wear Rate (PWR) and Surface Finish Using Laser

Groove Depth Analysis (LGDA) for TrizactTM B5-M990-5S2,

and SA A165 with Pad B and Slurry B at 30 hrs

Pad wear rate was at least 3 to 4 times higher with SA A165 vs Trizact B5. Surface

finish was the same.

New VP3500 Pad

A165 at 11hours

B5 at 30 hours

PWR (µm/hr) 36.5 12

Sa (µm) 18.3 16.3 16.6

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

0

5

10

15

20

25

30

35

40

Sa

(µm

)

PW

R (

µm

/hr)

Pad Wear Rate and Surface Finish for Trizact B5-M990-5S2, SA A165, Pad B, Slurry B Using LGDA ex-situ

PWR (µm/hr)

Sa (µm)



SEM Images – TrizactTM B5 after Marathon Test ( Ex-situ Conditioning)

CVD Coating is intact after 30 hours of ex-situ conditioning ( 100%) with Trizact B5

SE imaging BSE imaging



Pad Asperity Distribution Analysis ( A165, TrizactTM B5 –M990-5S2) ,

Pad B, Slurry B, Ex-situ at 30 hours

0.001

0.01

0.1

-30 -20 -10 0 10 20 30

Su

rfac

e H

eig

ht

Pro

bab

ility

Den

sity

F

un

ctio

n (1

/um

)

Surface Height (um)

A165 and B5 Exsitu

Trizact B5 A165

Pad Abruptness, lambda

A165 2.5 ± 0.3

B5 4.2 ± 0.6

Pad abruptness, lambda is higher and asperity truncation is smaller with pad associated with Trizact B5-

M990-5S2 compared to pad conditioned with SA A165 at 30 hours.



W Metal Gate or Buff ( 20% ex-situ conditioning; 9 lbs df, 1.4 psi wafer df)

― Slurry B/Pad B/ Trizact B6-M990-5S2

― Benchmark: A165 conditioning disk (POR)

― Performance Target: Stable TEOS RR ( 250 A/min)

― Consumable Life Target: Conditioner: 20 hours; 7000 wafers ( Current baseline

(~3 hours; 1000 wafers)

Trizact™ Pad Conditioner: Tungsten Gate CMP Process Development



W CMP Gate Buff Process With TrizactTM B6 and SA A165 With Slurry B and

Pad B

0

10

20

30

40

50

60

70

80

200

210

220

230

240

250

260

270

280

290

300

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00

No

n-U

nif

orm

ity (

%)

TE

OS

Re

mo

va

l R

ate

(A

/min

)

Conditioning Time (hours)

TEOS Removal Rate vs Conditioning Time Trizact B6 vs SA A165

Pad B, Slurry B, 9lbs DF, 20% exsitu conditioning

B6 RR

A165 RR

B6 WIWNU

A165 NU

Trizact B6 rate remained stable throughout the test close to the target rate of 250 A/min. The

rate drops off with A165 primarily due to pad balding.



W Metal Gate - TEOS Removal Rate Profiles – Pad B, Slurry B,

Similar TEOS RR profiles between Trizact B6 and SA A165 (POR)

0

100

200

300

400

500

600

700

-150 -100 -50 0 50 100 150

TE

OS

Re

mo

va

l R

ate

( A

/min

)

Wafer Location

TEOS Wafer Profiles for A165 with Pad B and Slurry B, 9bs DF, 20% exsitu Conditioning, 1.4 psi,

3 mm EE A165 Hour 0.51

A165 Hour 1.52

A165 Hour 2.53

A165 Hour 3.54

A165 Hour 5.55

A165 Hour 7.56

0

100

200

300

400

500

600

700

-150 -100 -50 0 50 100 150

TEO

S R

em

ova

l Rat

e (A

/min

)

Wafer Location

TEOS Wafer Profiles, 3M Trizact B6 Pad B, Slurry B, 9lbs DF, 1.4 psi, 20% exsitu

conditioning

Hour 0.54

Hour 1.58

Hour 2.63

Hour 3.68

Hour 4.74

Hour 5.82

Hour 7.93



0

10

20

30

40

50

60

70

80

0

50

100

150

200

250

300

350

400

450

500

0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00

No

n-U

nif

orm

ity (

%)

TE

OS

Re

mo

va

l R

ate

( A

/min

)

Conditioning Time (hours)

TEOS Removal Rate vs Conditioning Time 3M Trizact B6

Pad B, Slurry B, 9lbs DF, 1.4 psi, 20% exsitu conditioning

Removal Rate Non-Uniformity

EOL Marathon Test With W Metal Gate Process With TrizactTM B6-M990-5S2

Disk, Pad B and Slurry B – Oxide RR and NU

Stable Oxide RR and NU with Trizact B6 disk at target rate = 250 A/min



EOL Marathon: Metal Gate Process With TrizactTM B6-M990-5S2 Disk

Pad B, Slurry B- W RR and NU

Stable W RR and NU with Trizact B6 disk at target rate ~ 600 A/min

0.00

20.00

40.00

60.00

80.00

100.00

0

100

200

300

400

500

600

700

800

900

1000

0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00

No

n-U

nif

orm

ity

(%)

Rem

ova

l Rat

e o

f Tu

ng

sten

(A

/min

)

Hours of Conditioning ( hrs)

Tungsten Removal Rate, Non-Uniformity vs Conditioning Time 3M Trizact B6

Pad B, Slurry B, 9lbs DF, 1.3 psi, 20% exsitu conditioning

Removal Rate

Non-Uniformity



SEM Image of TrizactTM B6-M990-5S2 after EOL Marathon Test

SE imaging BSE imaging

CVD coating showing minimal wear after 20 hours of ex-situ conditioning with

Trizact B6-M990-5S2



The utility of 3MTM Trizact Pad Conditioners for W plug and metal gate

processes is demonstrated.

A 3MTM TrizactTM Pad Conditioner reduced the pad wear rate by a

factor of 3 or more, compared to a traditional SA conditioner, while

maintaining the same W rates and non-uniformities for a W plug

process with ex-situ conditioning. A corresponding improvement in pad

life is expected.

For W metal gate process, the more aggressive TrizactTM B6 Pad

Conditioner was needed to polish and maintain stable TEOS rate.

A TrizactTM B6 Pad Conditioner was used for about 18 hours in a W

metal gate process, maintaining the same TEOS and W removal rates

under typical POR conditions. This is expected to lead to a dramatic

increase, > 3X, in the number of wafers processed per pad , compared

to a conventional SA conditioner.

SUMMARY

Documents

and Metal Gate Applications - NCCAVS Usergroups · Hour 1.52 A165 Hour 2.53 A165 Hour 3.54 A165 Hour 5.55Hour 3.68 A165 0 Hour 7.56 100 200 300 400 500 600 700-150)-100 -50 0 50 100