Upload
truonghanh
View
216
Download
0
Embed Size (px)
Citation preview
©2010 Veeco Instruments Inc2
OutlineOutline
Background
New technologies for PMR pole deposition
Optical films for HAMR write heads
Summary
©2010 Veeco Instruments Inc3
Background – Technology RoadmapBackground – Technology Roadmap
BPMBPMHAMRHAMRPMRPMR
> 20152012 – 20142009 – 2011
2D signal processing
Media patterning
Light delivery
Near-field transducer
Sigma reduction
Damascene writer
Reduced HOC
Key AreasKey Areas
HAMR
Shingled (SWR)
2D read-back
BPM
HAMR
Shingled (SWR)
DFH
PMR
DFHTechnologiesTechnologies
> 5 TB/in21 – 5 TB/in2< 1TB/in2DensityDensity
©2010 Veeco Instruments Inc5
PR PR
Motivation for Damascene ProcessesMotivation for Damascene Processes
Two types of process are used for PMR poles– Subtractive process (etch)– Additive process (Damascene)
PR
Alumina
Pole
RIBE IBE IBE
Alumina
PolePR
Alumina
PR
Alumina
CMP IBE
Damascene Processes are Preferred as Geometries Shrink
©2010 Veeco Instruments Inc6
Requirements for Damascene Pole ProcessesRequirements for Damascene Pole Processes
Key steps in the damascene process– Deposition of seed layer– Plating of magnetic materials– CMP
Seed layer requirements1. Conformal2. Conductive / Low resistivity3. Void free plating of magnetic materials with desired properties4. Deposited at TFMH compatible temperatures (<200ºC)5. Good thickness control6. Adhesion
CVD deposition of Ru meets the above requirements
Fabrication Sequence
Alumina Alumina Alumina
Etch CVD Ru Deposition Plate + CMP
CVD Plated pole
©2010 Veeco Instruments Inc7
Ru CVD for Damascene Writer PoleEnabling FeaturesRu CVD for Damascene Writer PoleEnabling Features
CVD Ru is unique in being able to combine several key features1. Appropriate properties as a seed layer for plating
• Conformal
• Conductive / Low resistivity
2. Appropriate as a seed layer for Magnetic properties of plated layer
• Void free plating of magnetic materials
• High Bs plated layer
3. Provides ability to fine tune final track-width of the PMR pole
• Good thickness control
• Compensate for trench etch variation
Ru CVD also used as a plating seed for wrap around shields– Enables narrower track-widths by reduced cross
track interference
Step coverage95 -105 %
~ 0.4 m trenches
Step coverage95 -105 %
~ 0.4 m trenches
a
a
b
CVD- Rua
a
b
CVD- Ru
©2010 Veeco Instruments Inc8
Ru CVD ChemistryRu CVD Chemistry
Precursor : RuO4– ToRuS Blend from Air Liquide
• Mixture of RuO4 in two fluorinated solvents• Liquid at room temperature• High Vapor Pressure
– Rapid Reaction – practical deposition rates– Low Temperature Process (170°C)– Small molecule that nucleates on all tested surfaces
Two step reaction (heated wafer)
A. RuO4 RuO2 + O2 (not self limiting)B. RuO2 + 2H2 Ru + 2H2O
Accomplished as cyclic CVD in 4 steps – Steps 1,2 : Reaction A: + Inert Gas Purge – Steps 3,4 : Reaction B: + Inert Gas Purge Repeat to get desired thickness
©2010 Veeco Instruments Inc9
System ArchitectureSystem Architecture
Cross flow Design
Heated substrateInlet Outlet
Heated substrate
InletShowerhead Design
Outlet Outlet
RuO4 decomposition reaction to RuO2 is catalyzed by Ru but not self limiting
Cross flow architecture requires self limiting reactions to achieve good WiWuniformity and precursor utilization
Showerhead architecture allows for localized flow optimization to achieve optimum WiW uniformity and precursor utilization
NEXUS Ru CVD uses a showerhead design
180
210
240
270
300
330
1 2 3 4 5 6
ToRuS pulse time (s)
XRF
thic
knes
s (A
)
Lower ToRuS partial pressureHigher ToRuS partial pressure
©2010 Veeco Instruments Inc10
Deposition ConformalityDeposition Conformality
Cross-section SEM ~400Å CVD Ru on
SiO2
Deposition conformality similar at process temperatures from 170-200°C
Deposition conformality similar for features of various sizes
Conformality of adhesion enhancing under-layers can impact stack conformality
©2010 Veeco Instruments Inc11
CVD Ru: Thickness Control, ResistivityCVD Ru: Thickness Control, Resistivity
Rate Linearity of RuCVD on SiO2
0100200300400500600
0 50 100 150 200 250
# of Cycles
XR
F Thic
knes
s (A
)
170C190C y=2.6x - 40.3
y=2.8x - 29.8
C VD R u on S iO2 - R ate vs D ep Temp
2.5
2.6
2.7
2.8
2.9
3
3.1
3.2
3.3
160 170 180 190 200 210 220D ep Temp(degC )
Gro
wth
Rat
e(A
/cyc
le)
Growth rate is linear with number of cycles– Thickness can be accurately targeted– Initial nucleation delay <7 cycles at 200ºC
Good thickness control, appropriate electrical properties for plating
R es is tiv ity vs D ep T em p (~400A o n S iO 2)
17
17 .5
18
18 .5
19
19 .5
20
20 .5
21
160 170 180 190 200 210 220D ep T em p eratu re (d eg C )
Res
istiv
ity (u
Ohm
-cm
) Resistivity change slightly with temperature
– < 20µΩ-cm at 400Å on SiO2 at 170-210°C
Growth rate varies with temperature
©2010 Veeco Instruments Inc12
Film Stress EngineeringFilm Stress Engineering
Ru CVD film stress ~2600MPa Stress can be tuned to desired value by periodic plasma annealing Stress driven more compressive with:
– More plasma anneals– Higher power plasma anneals– Lower pressure plasma anneals
CVD only
CVD withperiodic plasmaanneal
Substrate Substrate
©2010 Veeco Instruments Inc13
Marathon Test Through Ampoule LifeMarathon Test Through Ampoule Life
Data from one 3.6L ampoules with single process recipe on Lab Tool 150mm wafers, 190°C deposition temp 632 wafers processed within tight specifications (black line shows End Of Life)
– Mean thickness at 401.6A with 3variation of 1.45%– Ampoule Utilization at 83A/cc
Sheet Resistance and Non-Uniformity with # of wafers
3.73.83.94.04.14.24.34.44.54.64.7
0 100 200 300 400 500 600 700
Wafer Number
She
et R
esis
tanc
e (o
hm/s
q)0.00.51.01.52.02.53.03.54.0
3s R
s un
if (%
)
Film Thickness (XRF) and Resistivity with # of wafers
350
360
370
380
390
400
410
420
0 100 200 300 400 500 600 700
Wafer Number
Cen
ter X
RF
Thic
knes
s (A
)
16.5
17.0
17.5
18.018.5
19.0
19.5
20.0
20.5
Res
istiv
ity(u
ohm
-cm
)
WTW 0.34% 1σ
WTW 0.36% 1σ
WTW 0.49% 1σ
Mean 1.38%
Good WTW and WIW repeatability across ampoule life< 2% 3 Non-Uniformity up to 650 wafers
©2010 Veeco Instruments Inc14
CVD Ru: Uniformity, Repeatability (200mm)CVD Ru: Uniformity, Repeatability (200mm) 12-wafer lot on Lab Tool: WIW < 3% 3σ, WTW ~1% 3σ Multiple ampoules on Lab Tool: WIW ~3% 3σ, WTW ~2.5% 3σ Data through ampoule life: WIW <6% 3σ, WTW < 2% 3σ
Good uniformity and repeatability through multiple ampoules< 6% 3σ WIW, < 2% 3σ WTW through ampoule life
~400A CVD Ru on 200mm SiO2 - 200C(49pt 6mmEE measurement)
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5
1 2 3 4 5 6 7 8 9 10 11 12
Wafer #
Rs
(ohm
/sq)
00.10.20.30.40.50.60.70.80.91
Nu
(%, 1
s)
Sheet resistanceWIW uniformity
WTW repeatability 0.35% 1σ
Average WIW uniformity 0.81% 1σ
Sheet Resistance and Uniformity vs Ampoule CyclesThree Ampoules - 200mm wafer - 200C - Lab Tool
0
1
2
3
4
5
6
0 2500 5000 7500 10000 12500 15000 17500
Nos. of cycles on three 1.2l ampoules
Rs
(ohm
/sq)
0
1
2
3
4
5
6
WiW
1s
non-
unif
(%)
Average WiW 1σ non-uniformity 0.88% 1σ
WTW repeatabilty 0.83% 1σ
©2010 Veeco Instruments Inc15
SummarySummary
NEXUS CVD Ru meets the film properties required of a conformal seed layer for damascene processing of high areal density write poles and shields– Conformality: >0.98– Deposition temperature: <200°C– Resistivity: <20cm– Uniformity: <3%3– Repeatability: <2%3– High efficiency of chemical use: >80Å/ml
The showerhead architecture is ideal for the ToRuS based CVD Ruprocess
The NEXUS CVD Ru system has been proven in production– Demonstrated >90% uptime– Stable film performance over lifetime of ampoule
©2010 Veeco Instruments Inc17
HAMR TechnologyHAMR Technology
Overview– HAMR is the leading candidate for >1TB/in2 areal
densities for write heads
Concept– At high areal densities, write heads cannot produce
enough field to record data in high coercivity media
– By heating the media locally, the coercivity can be reduced and the data recorded
– HAMR write heads incorporate a laser light and near field transducer to locally heat the media
Challenges1. Integration of light source and delivery mechanism
2. Thermal management
3. Low loss optical materials for light transmission
HAMR is a key technology for >1Tb/in2
Light transmission via low loss waveguides
Media
Head
Maximum Head Field
Field needed for smaller bit size
Mag
netic
Fi
eld
Source: IDEMA Presentation 12/6/07, Mark Re, Seagate
©2010 Veeco Instruments Inc18
HAMR: Waveguide MaterialsHAMR: Waveguide Materials
Laser with optical waveguide – Core/Cladding for waveguides
Requirements– Low and high index films– Low optical loss / defect free– Good Throughput
<10 dB/cm (cladding)<5 dB/cm (core)
Optical Loss
<3% R/M (200mm)Uniformity
100-500 Å/min (core)500-1500 Å/min (cladding)Rate
0 (below ellipsometer limit)k
High n: >2 (Ta2O5) (core)Low n: <2 (Al2O3) (cladding)Index
RequirementParameter
©2010 Veeco Instruments Inc19
PVD Deposition Technologies for Waveguide FilmsPVD Deposition Technologies for Waveguide Films
RateUniformity (R/M 200mm)Loss
Cladding>1000 Å/min<3.5% Al2O3<10 dB/cmPVD1 Metal Mode
100-500 Å/min
60-150 Å/min
Core
Good optical propertiesLow Rate / Poor CoO
Comment
<3% Ta2O5<5 dB/cmPVD1 Dielectric Mode
<3%<0.5 dB/cmIBD
PropertyTechnology
PVD processing has flexibility of hardware and deposition modes– Hot chuck capability (up to 400ºC)– Dielectric and metal modes of deposition– RF diode / DC magnetron
©2010 Veeco Instruments Inc20
Ta2O5 Dielectric/Poisoned Mode FilmsTa2O5 Dielectric/Poisoned Mode Films
No measurable change in optical
properties (ellipsometry) operating at
different O2 flows in poison mode
Processes show acceptable loss
properties
Ta Ox Hyste re sis
300320340360380400420440460480500
0 2 4 6 8 10 12 14 16
Oxyge n flow (s ccm )
Targ
et V
olta
ge(V
)
No t po is one d
Pois one d
0.07
0.07
0.1
Index U%1σ%
1.662.131.01551250Flow 3Process C
1.542.120.51451250Flow 2Process B
1.51 2.120.53503000Flow 1Process A
Process Loss ResultsDb/cm
IndexU% 1σ%200 mm
Deposition Rate (A/min)
Power (W)O2 flow (sccm)
©2010 Veeco Instruments Inc21
Closed-Loop Control “Metal Mode” Al2O3Deposition for High Deposition RatesClosed-Loop Control “Metal Mode” Al2O3Deposition for High Deposition Rates
Closed-loop control with High Speed MFC partial pressure controller– Target voltage continuously monitored– High speed piezo MFC: oxygen flow adjusted to maintain target voltage– Can now operate in “forbidden” range of voltages– Rate ~ 50% of Al deposition rate achievable
150
200
250
300
350
400
450
0 5 10 15 20 25 30 35 40
O2 Flow (sccm)
V (V
olts
)
O2 incr
O2 decr
Characteristic Hysteresis Loop of Target V vs. O2
Target surface metallic
Target surface dielectric
Ideal operating point• High rate• Good film properties• Difficult to control due
to target poisoning
©2010 Veeco Instruments Inc22
Al2O3: Optical LossAl2O3: Optical Loss
Strong dependence of optical loss on wafer deposition temperature No significant changes observed using ellipsometry measurement
– Measure extinction coefficient zero for all above films.– Loss not correlated to measured refractive index
• Within 1.63-1.66 range
Optical Loss vs. Wafer Temperature
0
5
10
15
20
25
30
200 210 220 230 240 250 260 270 280 290Temperature
Opt
ical
Los
s [d
b/cm
]Optical Loss vs. Deposition Temperature
©2010 Veeco Instruments Inc23
0
1
2
3
4
5
6
7
8
0 2 4 6 8 10
Annealing Time [hrs]
Leng
th o
f Lig
ht S
treak
[cm
]
175C dep + 200C anneal (Ar)175C dep + 175C anneal (Ar)175C dep + 175C anneal (Ar) (cummulative anneal time)175C dep + 175 C anneal (Ar/O2)
reference:3.3 dB/cm
reference: 8.0 dB/cm
Al2O3 Post Deposition AnnealingAl2O3 Post Deposition Annealing
Appropriate post-deposition annealing (<2hrs) results in low loss Al2O3
Optimized Annealing Process
Reference data (no anneal)
0 dB
5 dB
10 dB
15 dB
20 dB
3.33 db/cm (7cm light streak)270°C deposition
8 db/cm (4-5 cm light streak)250°C deposition
waferLight Streak waferLight Streak
waferLight Streak
Prism Coupler
waferLight Streak
Prism Coupler
Optical Loss vs. Annealing Time and Temperature
Veeco has developed optimized annealing conditions that enable low temperature deposition of alumina with low-loss optical properties
©2010 Veeco Instruments Inc24
Ta2O5 Thickness and Index RepeatabilityTa2O5 Thickness and Index Repeatability
<2dB/cmOptical Loss
-147Stress [MPa]
2.13Index
0.24%WIW Index (1)
0.08%WTW Index (1)
0.75%WIW Thickness (1)
0.75% (300 A/min)WTW Thickness (1)
DataTa2O5 – 2000 ÅTa2O5 Thickness Repeatability
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
0 5 10 15 20 25 30 35 40
Run #
Thic
knes
s [A
]
Ta2O5 Index Repeatability
2.0500
2.0700
2.0900
2.1100
2.1300
2.1500
2.1700
2.1900
0 5 10 15 20 25 30 35 40
Run #
Inde
x
©2010 Veeco Instruments Inc25
Al2O2 Thickness and Index RepeatabilityAl2O2 Thickness and Index Repeatability
<5dB/cmOptical Loss
-126Stress [MPa]
1.67Index
0.1%WIW Index (1)
0.03%WTW Index (1)
0.4%WIW Thickness (1)
0.3% (1300 A/min)WTW Thickness (1)
DataAlumina – 1m
Ta2O5 Index Repeatability
1.65
1.655
1.66
1.665
1.67
1.675
1.68
0 5 10 15 20 25 30 35 40
Run #
Inde
x
Al2O3 Thickness Repeatability
9000
9200
9400
9600
9800
10000
10200
10400
10600
10800
11000
0 5 10 15 20 25 30 35 40
Run #
Thic
knes
s [A
]
©2010 Veeco Instruments Inc26
Bi-layer Optical Loss Repeatability Al2O3 1m / Ta2O5 2000ÅBi-layer Optical Loss Repeatability Al2O3 1m / Ta2O5 2000Å
Alumina/Tantala Bilayer Loss
0
1
2
3
4
5
6
7
8
0 10 20 30 40 50Run #
Loss
[dB
/cm
]
Cassette 1 Cassette 2 Cassette 3 Cassette 4
Repeatable optical loss <5dB/cm obtained with Al2O3 1m / Ta2O5 2000Å– 12 wafers measured (total of 48 wafers run)
Four cassettes run over two days
©2010 Veeco Instruments Inc27
HAMR Data SummaryHAMR Data Summary
Ta2O5 process developed for low loss core layers– Loss: <2 dB/cm– Rate: 334 Å/min– U%: 0.7% 1– WTW% 0.7% 1
Al2O3 for cladding layers– Loss: <2dB/cm (with post deposition annealing – max temp 200°C)– Rate: >1000 Å/min– U%: 0.4% 1– WTW%: 0.3% 1
Bi-layer Al2O3/Ta2O5 films show <5dB/cm loss SiO2 films in development