particle adhesion and removal in post-cmp applications378435/fulltext.pdfair (jerkov) bond strength...

Microcontamination Research LabMicrocontamination Research Lab

George Adams, Ahmed A. Busnaina and George Adams, Ahmed A. Busnaina and Sinan MuftuSinan Muftuthe Microcontamination Research Laboratorythe Microcontamination Research Laboratory

Mechanical, Industrial, and ManufacturingMechanical, Industrial, and Manufacturing Eng Eng..DepartmentDepartment

Northeastern University, Boston, MA 02115-5000Northeastern University, Boston, MA 02115-5000

PARTICLE ADHESION ANDPARTICLE ADHESION ANDREMOVAL IN POST-CMPREMOVAL IN POST-CMP

APPLICATIONSAPPLICATIONS


OUTLINEOUTLINEqGoals and ObjectivesqApproachqPreliminary ResultsvParticle Removal MechanismvDouble layervCovalent bonds between for silica on

thermal oxide films and glass particles onglass substratesvAdhesion induced deformation

qKey Preliminary Research Results


Surface Preparation TechnologySurface Preparation TechnologyRequirements -- Long TermRequirements -- Long Term****

YEAR 1999 2000 2002 2005 2008 2011 2014TECHNOLOGY NODE 180nm 130nm 100nm 70nm 50nm 35nm

FEOL Particle Size(nm) 90 82.5 60 50 35 25 18

FEOL Particles (#/cm2) 0.064 0.06 0.064 0.051 0.052 0.052 0.052

BEOL Particle Size(nm) 180 165 130 100 70 50 36

BEOL Particles (#/cm2) 0.064 0.06 0.064 0.051 0.052 0.052 0.052

Surface Roughness (nm) 0.15 0.14 0.12 0.1 0.08 0.08 0.08

Critical surface metals(*109) 9 7 4.4 2.5 -- -- --

Organics ( *1013 atoms/cm2) 7.3 6.6 5.3 4.1 -- -- --

** ---- The International Technology Roadmap for Semiconductors, 2000


Goals and ObjectivesGoals and Objectivesu To determine the adhesion force for different particles on

different substrates in different solutions experimentally.u To understand and determine the onset of large adhesion

forces after polishing such as the development of covalentbonds.

u Study the removal and adhesion forces for alumina andsilica slurry particles from silicon wafers (with differentfilms, TOX, W, Cu, TaN, BPSG, etc.).

u Develop better cleaning guidelines and techniques toreduce surface defects after polishing.

u In addition, the effect of polishing pressure on the slurryparticle adhesion force will be determined.


ApproachApproachuu Fundamental ApproachFundamental Approachuu Experimental and modeling approach toExperimental and modeling approach to

determine, understand and predict:determine, understand and predict:uu Particle Adhesion ForceParticle Adhesion Forceuu Adhesion force increase due to:Adhesion force increase due to:

uu Particle DeformationParticle Deformationuu Covalent BondsCovalent Bondsuu CMPCMPuu Slurry and Cleaning Solution ChemistrySlurry and Cleaning Solution Chemistryuu Pad hardness, compressibility and tensile strengthPad hardness, compressibility and tensile strengthuu Particle shapeParticle shapeuu Surface RoughnessSurface Roughness


ApproachApproach

uu Fundamental ApproachFundamental Approachu Key Particle Removal Parameters

u Particle shapeu Polishing pressureu Particle sizeu Particle compositionu Slurry and cleaning solution chemistryu Chemical reactions giving rise to covalent bondsu particle deformationu Double layer effectu Cleaning liquid surface tensionu Surface and particle surface energy (hydrophilic or hydrophobic)


PARTICLE REMOVALPARTICLE REMOVALq Nano-scale particles will be a challenge to current

cleaning techniques.

q The most widely used cleaning techniques:ü non-contact method (megasonic cleaning)ü contact cleaning method (brush scrubbers)

q The two basic elements that need to beunderstood are :ü Particle Adhesionü Particle Removal


Particle AdhesionParticle Adhesionu Adhesion between a contaminant particle and a

wafer occurs in three basic steps:

u Long-range attraction forces draw the particletowards the surface

u The contaminant particle touches the wafer at onepoint of contact

u Short range van der Waals force dominates nearthe surface and the contact area increases.

u Deformation may occur due to adhesion-induceddeformation thereby increasing the adhesion forcethe deformation stops and the system is inequilibrium.


Removal Percentage vs. MomentRemoval Percentage vs. MomentRatio (Silica Removal Experiment)Ratio (Silica Removal Experiment)

Removal Percentage Moment Ratio

0

10

20

30

40

50

60

70

80

90

100

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Moment Ratio

Rem

ova

l P

erce

nta

ge


Removal Efficiency vs. Aging Timeaged at 25 degree C, 95%RH

0.840.86

0.880.9

0.920.940.96

0.981

0 1 2 3 4 5 6 7

Aging Time (weeks)

Rem

ova

l E

ffic

ien

cyAging Effect on Glass Particle Removal from FPD

RESULTS


Imprints on the glass surface ( megasonic cleaned after aging)

RESULTS


SEM image of glass chips(large bonding areas matched-flatsurfaces in contact low profile)

RESULTS


AFM micrograph of surface imprintsAFM three dimensional view of imprints

RESULTS


1E+3

1E+4

1E+5

1E+6

1 10 100 1000

Microns

PS

I -

Sh

ea

r

Experimental Data

Soda-lime fiber inair (Griffith)

Borosilicate fiber inair (Jerkov)

Bond strength vs. particle diameter ( Experiment datacompared with Griffth's tensile strength of glass fiber and

with Jerkov's tensile strength of glass fiber)

RESULTS


1.58um Silica Particle on Thermal Oxide Wafer

One Week Aging

Six weeks aging

Dry surface, 55%RH aging Wet surface, 55%RH aging Wet surface, 100%RH aging

RESULTS


RESULTS

1.58µm silica particle on slurry wettedoxide surface within one hour


Removal Efficiency vs. Aging Time

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 1 2 3 4 5 6 7

Aging Time(week)

Rem

ova

l Eff

icie

ncy

dry 55%RH

wet 55%RH

wet 100%RH

RESULTS


Adhesion Force vs. Aging

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0 1 2 3 4 5 6 7

Aging Time(week)

Ad

hes

ion

Fo

rce(

dyn

)

dry 55%RH

wet 55%RH

wet 100%RH

van der Waalsforce(withoutdeformation)

van der Waalsforce(withdeformation)

RESULTS


40% Relative Humidity Effect on the Ratio of Contact Diameter/Particle Diameter as a Function of Time

0.35

0.37

0.39

0.41

0.43

0.45

0.47

0.49

0.51

0 1 2 3 4 5 6 7 8

Aging Time(day)

Co

nta

ct D

iam

ter/

Par

ticle

Dia

met

er

400nmexperiment data


RESULTS


95%RH Relative Humidity Effect on the Ratio of Contact Diameter/Particle Diameter as a Function of Time

0.3

0.35

0.4

0.45

0.5

0.55

0.6

0 2 4 6 8

Aging Time( day)

Co

nta

ct D

iam

eter

/Par

ticl

e D

iam

eter

400nmexperiment data.


RESULTS


RESULTS

Adhesion Force (400nm PSL Particles)

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

two day 40%RHAging

Seven Day40%RH Aging

two Day 95%RHAging

seven Day95%RH Aging

Adh

esio

n Fo

rce

(dyn

)

MP model

Capillary Force

Measured Force


SUMMARY

u Moisture plays a significant role in silica particles’adhesion on silicon oxide substrate. The adhesionforce significantly increases with moisture.

u The large meniscus indicates that initially, hydrogenbond forms and as moisture reacts between thesurface and the particle, covalent bond forms aroundthe periphery of the particle. If considering practicalsurface condition, the measured adhesion force fallbetween hydrogen and covalent bonds.


u Contact diameter between the particles and the substrate increasewith time logarithmically.

u For the same aging time, high humidity (95%RH) shows significantincrease in particle deformation compared to the 40%RH conditions.Capillary force tends to accelerate the adhesion-induced deformationprocess.

u For the same aging conditions, smaller particles encounter largerdeformation.

u The plastic deformation is caused by adhesion and capillary stressesthat exceed the yield strength of PSL particles.

SUMMARYThe average pressure exerted by surface force between PSL particle andsilicon oxide surface is

P is found to be 1.6×109 Pa, which is comparable to the Young’s modulus ofthe polystyrene (2.55 ×109 Pa) and far in excess of its yield strength (10.8×106 Pa). This confirms that it is plastic deformation taken place.

0

2zw

P A=

particle adhesion and removal in post-cmp applications378435/fulltext.pdfair (jerkov) bond strength...

Documents