Components Division
Ultimate in Vacuum since 1952
IntegratedIntegrated Metrology System for Metallization ProcessMetrology System for Metallization Process
MESEC SeriesMESEC Series
Components DivisionComponents DivisionULVAC, Inc.ULVAC, Inc.
121.3
124.5
127.8
131.0
134.3
137.5
140.8
-100 -80 -60 -40 -20 0 20 40 60 80 100
-100
-80
-60
-40
-20
0
20
40
60
80
100
6.29675DEV:132.83754Average:120.55Min:143.64999Max:
f=10MHzLiftOff=300μmt=0.15μmlong,Mode2,1sΦ175mm,Δ20mm92.9,147.8,202.8nm
Comment:naganoOperator:000114-6Sample Name:
DData:
2000/01/14Date:
ULVAC 2D Graph
X Axis
Y Axis
-100
-80
-60
-40
-20
0
20
40
60
80
100
880.00
900.00
920.00
940.00
960.00
-100-80-60-40-20020406080100
f=4MHzLiftOff=300μmt=150nm+800nmlong,Mode2,1sΦ175mm,Δ20mm
ULVAC 3D Graph
19.76964DEV:921.10804Average:883.79999Min:965.03003Max:D(nm)Data:
ULVAC950nmP
1999/12/21
Comment:Operator:Sample Name:Date:
X (mm)D(nm)
Y(mm)
2004/07/28
Components Division
Ultimate in Vacuum since 1952
IntroductionIntroduction
■■ To measure the conductive film thickness on silicon wafer by non-contact and non-destructive method, ULVAC developed the measurement system based on eddy current detecting technology (MESEC).
■■ Usually, 4-point probe measurement system is used to measure the conductive film thickness on wafer. The probe contacts the film, and cause damage on the surface. This kind of system is only used for indirect evaluation of semiconductor process on monitor wafers. But MESEC can evaluate directly on product wafers. So it is possible to obtain more reliable data and to reduce the cost on monitor wafers.
■■ ULVAC developed following three kinds of systems,
Built-in Type (MESEC-BIT) : Integrated onto the deposition systems, such as PVD, CVD and ECD
Stand-alone Type (MESEC-SAT) : Used for in-line or off-line measurement.
In-situ Type (MESEC-IST) : Under evaluation on CMP equipment
Components Division
Ultimate in Vacuum since 1952
Measurement PrincipleMeasurement Principle
The eddy current technique is based on the principle of magnetic inductance effect. If conductive material, like metal film, is placed in the presence of a changing magnetic field, eddy current is induced on the film. The inductance of detective coil changes with the eddy current. The value of inductance is corresponded with thickness of the conductive film. Therefore, the thickness of the film is possible to be measured by measuring inductance of the coil.
Components Division
Ultimate in Vacuum since 1952
FeaturesFeatures ■ It is possible to measure the thickness of thin metal films on silicon device wafers by
using this non-contact and non-destructive technique which enables product wafers to be measured directly.
■ Small-size eddy current probe and speedy measuring time makes it possibleto built this system into PVD, CVD, ECD(plating), and CMP tools, etc.
■ A capacitance displacement sensor measures the distance between eddy current probe and the wafer surface, ensuring precision and accuracy of film thickness measurements.
■ Auto dynamic calibration prevents long term drift in the measuring system, helping to ensure excellent repeatability of film thickness measurements.
■ Users can easily construct a database to calculate thickness of metal alloy films, make it possible to measure thickness of most metal alloy films. (Stand-alone type only)
■ Powerful data analysis software and flexible graphic functions are available with graphical user interface. (Stand-alone type only)
Components Division
Ultimate in Vacuum since 1952
OutsideOutside((StandStand--alone alone TypeType) )
MESEC SAT (Open cassette)MESEC SAT (Open cassette)
Components Division
Ultimate in Vacuum since 1952OutsideOutside((StandStand--alone Typealone Type) )
MESECMESEC--SATSAT((FOUPFOUP) )
Components Division
Ultimate in Vacuum since 1952Possible MESEC Integration forDual -damascene Copper Process
Components Division
Ultimate in Vacuum since 1952
OutsideOutside((BuiltBuilt--in Typein Type) )
MESECMESEC--BIT2000ABIT2000A MESECMESEC--BIT2000BBIT2000B MESECMESEC--BIT3000SBIT3000S
Components Division
Ultimate in Vacuum since 1952MESEC- Evaluation for Cu-seed process -
MESEC Integrated into ULVAC ENTRON W-300 System
BIT-MESEC
F3
TaN/TaPre-clean
F1 R1
R2
R3
R4
RXFXLA
LB
MArm A
Arm B
Cu
Arm A
Arm B
Aligner
Port1
Port3
Port2
BIT-MESEC
Components Division
Ultimate in Vacuum since 1952
Benefits for ECD Suppliers
Integrated metrology is very important for advanced 200mm and all 300mm ECD.
Several ECD suppliers have integrated metrology units in ECD plating tools.
Multiple ULVAC MESEC Integrated Metrology Modules are running in field testing with production ECD copper wafers.
Components Division
Ultimate in Vacuum since 1952
Standard specificationStandard specification
Film thickness measurement range: 0.03μm~5μm
Measurable material: Thin conductive film on silicon wafer
(Cu、AL、AlCu,etc)
Film thickness measurement precision: +/- 1% (for calibrated reading)
Film thickness measurement repeatability : +/- 0.5% (1σ)
(In the case of measuring same point 10 times repeatability)
Spatial resolution: 3mm(Sensor diameter:2.2mm)
Measuring speed: 1second/point or less
(except movement time of stage)
Stage size: 200mm、300mm
Measurement mode: R-θ mapping, X-Y grid
Software operational environment: Windows NT4
Components Division
Ultimate in Vacuum since 1952
Correlation between MESEC and 4Correlation between MESEC and 4--Points ProbePoints Probe
1800
1850
1900
1950
2000
2050
2100
-100 -80 -60 -40 -20 0 20 40 60 80 100
X(mm)
Thi
ckne
ss(n
m)
MESEC
Measured Sample: Cu plating film on 200mm wafer substrateScan Mode: Diameter scan on wafer, total 37 points with 5mm pitch
4-point Probe
200mm wafer Correlation Correlation coefficient coefficient is 99%!!is 99%!!
Components Division
Ultimate in Vacuum since 1952
Correlation between MESEC, MetaPulse and 4Correlation between MESEC, MetaPulse and 4--Points ProbePoints Probe
Post ECD Film Measurement
80%
85%
90%
95%
100%
105%
110%
115%
120%
-100 -50 0 50 100
Radial position (mm)
Thi
ckne
ss (n
orm
aliz
ed)
ULVAC MESEC 4 pt probe Rudolph Metapulse
Components Division
Ultimate in Vacuum since 1952Performance Comparison for4-Points Probe and MESEC
0 100 200 300 400 500 600 700 8000
100
200
300
400
500
600
700
800Thickness Measured by Two Kinds of Methods (nm)
Thickness of Cu Film by Stylus Profiler (nm)
4-Points Probe MESEC
40 60 80 100
20
40
60
80
100
Thickness Measured by Two Kinds of Methods (nm)
Thickness of Cu Film by Stylus Profiler (nm)
FourPoints MESEC
Components Division
Ultimate in Vacuum since 1952
30 60 90
860
880
900
920
940
Cu Plating Film (Target Thickness: 900nm) Cu Plating Film (Target Thickness: 2000nm)
Measurement Repeat Times
Thickness Measured by MESEC(nm)
1800
1850
1900
1950
2000
2050
2100
2150
2200
2250
Thickness Measured by MESEC(nm)
Average = 2139.2Std. Dev.= 3.91Std. Dev./Average = 0.18%
Average = 904.0Std. Dev.= 1.30Std. Dev./Average = 0.14%
Evaluation onLong Term Stability (Several Days Interval, 30 times/day)
Components Division
Ultimate in Vacuum since 1952Evaluation onLong Term Stability (Several Days Interval, 30 times/day)
30 60 9050
52
54
56
58
60Thickness Measured by MESEC(nm)
Measurement Repeat Times
Cu Thin Film by PVD Process (Target Thickness: 50nm)
Average = 54.4nmStd. Dev.= 0.10nmStd. Dev./Average = 0.18%
Components Division
Ultimate in Vacuum since 1952Measurement on Multi-layer Patterned Waferafter ECD Process
Top Layer
Under Layer
Plating Layer (Cu)
Seed Layer (Cu)
Barrier Layer (Ta/TaN)
The top film is unpatterned and can support eddy current formation. Underlying metal are patterned which breaks up the eddy currents, unless there are large areas of unpatterned metal. These large areas of unpatterned underlying metal may occur on test die or in alignment areas but would not normally be found within an active die area.
Components Division
Ultimate in Vacuum since 1952Measurement on Multi-layer Patterned Waferafter ECD Process
Diameter Scan on Patterned Wafer (After ECD Process)
0
200
400
600
800
1000
1200
-100 -80 -60 -40 -20 0 20 40 60 80 100
X(mm)
Thic
knes
s(nm
) Underlying multi-layer does not affects the measurement results of top plating layer.
Components Division
Ultimate in Vacuum since 1952
Why our customers selected Why our customers selected the MESEC ? the MESEC ?
-- -- -- Real value of the MESEC Real value of the MESEC
Process OptimizationProcess Optimization
Improved ReliabilityImproved Reliability
Cost of Ownership SavingCost of Ownership Saving
Components Division
Ultimate in Vacuum since 1952
Conventional line Conventional line ------
CMPPVD ECD Inspection
ECD Metrology
The wafers are inspected after all processes. Or even if after each process,
they are evaluated all at once.
Components Division
Ultimate in Vacuum since 1952
As a result As a result -- -- --
When defects are found during metrology measurement, feed back information to the deposition tool is too late.
ECD Metrology
Feed backFeed back
Components Division
Ultimate in Vacuum since 1952
ULVAC SolutionULVAC Solution
Integrate the MESEC onto each Cu process line.Integrate the MESEC onto each Cu process line.
NonNon--contact contact
Speedy measurementSpeedy measurement
CompactCompact
StabilityStability
Accuracy Accuracy
Cost savingCost saving
High reliabilityHigh reliability
Measurement of all production wafersMeasurement of all production wafers
Easy to integrateEasy to integrate
Worth to integrate the MESECWorth to integrate the MESEC
QC wafers QC wafers
Indirect inspection time Indirect inspection time
The waste wafer The waste wafer
Reduce Reduce
NecessaryNecessary
Components Division
Ultimate in Vacuum since 1952
Measurement applicationMeasurement application
Diameter ScanDiameter Scan
For surface uniformity analysis.
To see the result of deposition.
Overall ScanOverall ScanFor surface uniformity analysis.
For control of the exchange cycle of consumable parts for deposition tool, such as target, bath and pad, etc.
For Quality Control.
Components Division
Ultimate in Vacuum since 1952
Measuring inMeasuring in--coming wafercoming wafer
Feed forward the result to select an optimum recipe.
Profile OptimizerProfile OptimizerDiameter Scan on Patterned Wafer (Before ECD Process)
0
20
40
60
80
100
120
140
160
-100 -80 -60 -40 -20 0 20 40 60 80 100
X(mm)
Thickness(nm)
MESEC
Components Division
Ultimate in Vacuum since 1952
Measuring outgoing waferMeasuring outgoing wafer
Check results and provide feed back for ECD improvement. Stop producing waste wafers at once !
MESEC
Diameter Scan on Patterned Wafer (After ECD Process)
0
200
400
600
800
1000
1200
-100 -80 -60 -40 -20 0 20 40 60 80 100
X(mm)
Thickness(nm)
Profile KeeperProfile Keeper
Components Division
Ultimate in Vacuum since 1952
Why integrated the MESEC
Advantage for the PVD, ECD and CMP supplier Advantage for the PVD, ECD and CMP supplier
Optimization of film thickness profile (Feed-forward)Dynamic deposition control (Feed-back)Yes/No decision on the production wafer
Advantage for the end user Advantage for the end user
Reduction of QC/monitor wafersOn the fly tool qualificationsHigher Yield Ratio
Components Division
Ultimate in Vacuum since 1952
Trial calculationTrial calculation
Reduction effect on test wafer expense.Reduction effect on test wafer expense.
0.81.6
3.24
6.4
2
16
8
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
0 10 20 30 40 50 60 70 80 90
Wafer throughput capacity (10 thousand wafer/ year)
Annual costreduction(million $)
200mm
300mm-- Condition Condition --
--5 5 wiring processes / waferwiring processes / wafer--Unit price of 200mm wafer: $ 80.Unit price of 200mm wafer: $ 80.--Unit price of 300mm wafer: $ 200. Unit price of 300mm wafer: $ 200. --Test wafer / Production wafer ratio: 1/50 Test wafer / Production wafer ratio: 1/50
16
Components Division
Ultimate in Vacuum since 1952Performance Comparison with Conductive Film Thickness Measuring Systems
Eddy Current Type (MESEC)
4-Point Probe Type
Fluorescent X-ray Type
X-ray Interference
Type
Pulse Laser Type
Destructive Dangerous
◎
Non-destructive
×
Contact
× Radioactive
Source
× Radioactive
Source
× High Power
Laser
Measuring Range
◎
30nm~5μm
○
100nm~10μm
◎
10nm~15μm
×
10nm~400nm
○
10nm~3.5μm
Measuring Time
(Per Point)
◎
Less than 1 second
○
A few seconds
×
A few minutes
×
A few minutes
○
A few seconds
Spatial Resolution
○
A few mm
○
A few mm
×
A few cm
○
A few mm
◎
A few μm
◎Excellent ○Good ×Poor
Components Division
Ultimate in Vacuum since 1952
MESECMESEC PerformancePerformance
Non-contact, no wafer damage !Excellent correlation with other measurement methods.Capable of measuring film on top of multi-stacked damascene layers.Minimal loss of production wafersEarly detection of deposition tool problem
Advantages:Advantages:
User friendly:User friendly:Easy to operate.
No consumable parts.No adjustments.
Maintenance free.Reasonable price.
Low CoO and fast amortization can be realized with continuous use
of the MESEC.
Components Division
Ultimate in Vacuum since 1952Application DataApplication Data
-80-60-40-20
020406080
120140
160
-80-60-40-20020406080
MESEC 3D Graph
3.75nmDEV:159.4nmAverage:109.2nmMin:164.3nmMax:ThicknessData:
Sputtering Cu Film 160nm2002/03/07
Sample Name:Date:
X (mm)
Thickness(nm)Y (mm)
Cu Seed Film on 200mm Silicon Wafer
-100
-80
-60
-40
-20
0
20
40
60
80
100
-100
-80
-60
-40
-20
0
20
4060
80100
120.00
125.00
130.00
135.00
140.00
145.00
MESEC 3D Graph
6.71nmDEV:134.13nmAverage:120.69nmMin:145.64nmMax:ThicknessData:
ULVACCu Seed Film
Operator:Sample Name:
X (mm)
D(nm)
Y (mm)
-100
-80
-60
-40
-20
020406080100
-100
-80
-60
-40
-20
0
20
40
60
80
100
1600
1700
1800
1900
2000
3D Thickness Graph
57.83nmSTDDEV:
1804.06nmAverage:1708.3nmMin:1956.7nmMax:ThicknessData:
UJ25830
Cu ECD Film
Operator:Sample:
X (mm)
Thick
ness(
nm)
Y (mm
)
Cu Plating Film on 200mm Silicon Wafer
Detecting Scratch on Wafer
Components Division
Ultimate in Vacuum since 1952
Why select the MESEC Why select the MESEC
NonNon--contact, Speedy measurement,contact, Speedy measurement,Excellent repeatability, High accuracyExcellent repeatability, High accuracy
Optimization for processOptimization for process
Test all Test all production wafersproduction wafers Cut costCut cost
(Initial, (Initial, Running, Running,
Monitor wafers)Monitor wafers)
Cost EffectiveCost Effective
Integrating Integrating into depositioninto deposition
tooltool
Improve reliabilityImprove reliability Cost downCost down
Components Division
Ultimate in Vacuum since 1952
Global Network
ULVAC Technologies,Inc.ULVAC Ningbo Co., Ltd.
ULVAC,IncULVAC,Inc
UTECH NewYork
UTECHArizona
UTECHTexas
UTECHCalifornia
ULVAC Taiwan,Inc
ULVAC System Control, Taiwan., Ltd.
ULVAC KOREA, Ltd.ULVAC G.m.b.H
Hong Kong ULVAC Co., Ltd.
UJ: ULVAC,Inc.UTECH: ULVAC Technologies,Inc
India officePenang CSC
UJUJ
Guangzhou officeUJ
Beijing office UJ
ULVAC Shanghai Co., Ltd.
ULVAC Singapore PTE LTD( Jan. 2002 open)