1

Simultaneous and Non-Destructive Measurements of

• Depth

• Top and Bottom CD

• Residual Layer Thickness, RLT

• DLC Thickness

• Side Wall Angles

Complete Metrology Solutions Imprint Technology

Templates, DTR and BPM Media

Iris Bloomern&k Technology

March 2009

2

OutlineOutline

The Challenge of Measuring Imprint Templates, DTR and BPM Media

Surmounting this Challenge

The n&k Gemini

Some Measurement Examples− E-Beam Template− Template and Corresponding DTR Sample− BPM Media Sample

Summary

3

The ChallengeThe Challenge

Next Generation Template

40nm Depth

30nm MCD

45nm Pitch

70nm Depth

60nm MCD 100nm

Pitch

Current Generation Imprint

Current Generation Template

120nm Depth

50nm MCD

80nm Pitch

Next Generation Imprint

40nm Depth

30nm MCD

50nm Pitch

4

SensitivitySensitivity

Changes in parameters of interest produce only a very small change in the measured data

Changes in measured data caused by variations of parameters of interest can occur anywhere in the measured spectrum, from the deep UV to the NIR

The challenge corresponds to adequate sensitivity of the metrology tool

5

Change in Template Depth by a Few nm, Change in Template Depth by a Few nm, ⇒⇒ Change in Measured Data by a Fraction of a Per CentChange in Measured Data by a Fraction of a Per Cent

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

200 300 400 500 600 700 800 900

Rs

& R

p (%

)Wavelength (nm)

Rs-cal Rp-cal

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

200 300 400 500 600 700 800 900

Rs

& R

p (%

)

Wavelength (nm)

Rs-cal Rp-cal

Rs: 117nm, 120nm, 123nmRp: 117nm, 120nm, 123nm

Quartz Depth Simulations for Nominal Value +/- 3nm

Rs: 37nm, 40nm, 43nmRp: 37nm, 40nm, 43nm

91

92

93

94

95

96

200 300 400 500 600 700 800 900

Ts &

Tp

(%)

Wavelength (nm)

Ts-cal Tp-cal

91

92

93

94

95

96

200 300 400 500 600 700 800 900

Ts

& T

p (%

)

Wavelength (nm)

Ts-cal Tp-cal

Next Generation TypeCurrent Template Type

Ts: 37nm, 40nm, 43nmTp: 37nm, 40nm, 43nm

Ts: 117nm, 120nm, 123nmTp: 117nm, 120nm, 123nm

6

91

92

93

94

95

96

200 300 400 500 600 700 800 900

Ts &

Tp

(%)

Wavelength (nm)

Ts-cal Tp-cal

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

200 300 400 500 600 700 800 900

Rs

& R

p (%

)

Wavelength (nm)

Rs-cal Rp-cal

91

92

93

94

95

96

200 300 400 500 600 700 800 900

Ts

& T

p (%

)

Wavelength (nm)

Ts-cal Tp-cal

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

200 300 400 500 600 700 800 900

Rs

& R

p (%

)Wavelength (nm)

Rs-cal Rp-cal

Change in Template MCD by a Few nm, Change in Template MCD by a Few nm, ⇒⇒ Change in Measured Data by a Fraction of a Per CentChange in Measured Data by a Fraction of a Per Cent

Rs: 53nm, 55nm, 57nmRp: 53nm, 55nm, 57nm

Rs: 27nm, 29nm, 31nmRp: 27nm, 29nm, 31nm

Next Generation TypeCurrent Template Type

Middle CD Simulations for Nominal Value +/- 2nm

Ts: 53nm, 55nm, 57nmTp: 53nm, 55nm, 57nm

Ts: 27nm, 29nm, 31nmTp: 27nm, 29nm, 31nm

7

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

R

p (%

)

Wavelength (nm)

Rp-cal

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rs

(%)

Wavelength (nm)

Rs-cal

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rp

(%)

Wavelength (nm)

Rp-cal

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rs

(%)

Wavelength (nm)

Rs-cal

Rs: 65nm, 70nm, 75nm

Change in Imprint Depth by a Few nm, Change in Imprint Depth by a Few nm, ⇒⇒ Change in Measured Data by a Fraction of a Per CentChange in Measured Data by a Fraction of a Per Cent

Grating Height Simulations for Nominal Value +/- 5nm

Next Generation TypeCurrent Imprint Type

Rp: 65nm, 70nm, 75nm

Rs: 35nm, 40nm, 45nm

Rp: 35nm, 40nm, 45nm

8

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

R

p (%

)

Wavelength (nm)

Rp-cal

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rs

(%)

Wavelength (nm)

Rs-cal

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rp

(%)

Wavelength (nm)

Rp-cal

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rs

(%)

Wavelength (nm)

Rs-cal

Change in Imprint MCD by a Few nm, Change in Imprint MCD by a Few nm, ⇒⇒ Change in Measured Data by a Fraction of a Per CentChange in Measured Data by a Fraction of a Per Cent

Middle CD Simulations for Nominal Value +/- 3nm

Rs: 57nm, 60nm, 63nm

Next Generation TypeCurrent Imprint Type

Rs: 27nm, 30nm, 33nm

Rp: 57nm, 60nm, 63nm Rp: 27nm, 30nm, 33nm

9

Achieving Measurement SensitivityAchieving Measurement SensitivityOptimized Signal to Noise− Necessary in order to distinguish changes in measured data due to changes in parameters

(e.g. depth or CD) from noise in measured data

Wavelength Range− Large wavelength range that includes DUV data is critical− Changes in parameters can cause changes anywhere in the DUV to NIR wavelength range

Transmittance Measurements + Reflectance Needed for Templates− T provides greater sensitivity than R for Templates− Measured reflectance ≈ A few % for Templates− Measured transmittance > 90 % for Templates

∗ Intensity of T >> Intensity of R, ∗ ∴ more photons reach detector∗ The greater # of photons detected, the greater the signal

Valid Physical Model for Analysis of Raw Data− Software must converge and overcome ambiguities in results

Sample Handling for Disks− X-Y Stage not adequate for aligning gratings− X-Y-Z-θ is necessary

10

Pinhole

Polarizer

Polarizer

UV

Vis

Detector - R

Detector - T

Source

• Patented all-reflective optical design, without beam splitters and refractive lenses

• Optimized polarized R and T data over the entire wavelength range DUV-VIS-NIR

• 190 – 1000 nm spectrum in 1 nm intervals

• Measurement Spot

• Reflectance: 50 µm

• Transmittance: 50 µm

• Measurement Time: 1 s/pt

The n&k Gemini Optical ConfigurationThe n&k Gemini Optical ConfigurationResults in Optimized Signal to NoiseResults in Optimized Signal to Noise

11

The n&k Gemini Valid Physical ModelThe n&k Gemini Valid Physical Model

Forouhi-Bloomer Model Combined with RCWA is Used for Analysis

12

The n&k GeminiThe n&k GeminiProvides Sufficient Sensitivity for Measurements of Templates, DProvides Sufficient Sensitivity for Measurements of Templates, DTR and BPM MediaTR and BPM Media

• Incorporates optical design for optimized signal to noise and wide wavelength range (190 – 1000 nm)

• Analysis based on Hybrid RCWA• Fully Automated• Polarized Reflectance (Rs and Rp)

−Polarized Transmittance (Ts and Tp)Rs, Rp, Ts, Tp simultaneously measured at the same point

−Wavelength Range from 190 -1000 nm

• Spot Size: 50 µm for Reflectance, 50 µm for Transmittance

• Automated X-Y-Z-θ Stage and Robot are designed to accommodate various size disks−Automatically aligns gratings to increase speed and

reduce user error.

Engine of the n&k Gemini

13

EE--Beam Template MeasurementBeam Template Measurement

65mm square Sample, 0.25”thickGrating pattern at sample centerTrenches in quartz, 80nm pitch

Quartz

14

n&k Model and Typical Fitn&k Model and Typical Fit

n&k Model

Quartz Depth

Top CD

Bottom CD SWAExperimental and Calculated Reflectance (Rs, Rp)

Experimental and Calculated Transmittance (Ts, Tp)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

200 300 400 500 600 700 800 900

Rs

& R

p (%

)

Wavelength (nm)

Rs-exp Rs-cal Rp-exp Rp-cal

86

88

90

92

94

96

98

100

200 300 400 500 600 700 800 900

Ts &

Tp

(%)

Wavelength (nm)

Ts-exp Ts-cal Tp-exp Tp-cal

15

Parameters Simultaneously Measured and Mapped

• Quartz DepthMean = 1226.9Å, STD = 1.72

• Top CDMean 67.20nm, STD = 0.19

• Bottom CDMean 45.32nm, STD = 0.24

• Sidewall AngleMean 86.3º, STD = 0.025º

Quartz Depth

Top CD

Bottom CD SWA

25 Point Map

16

Repeatability ResultsRepeatability Results

0.146nm45.02nmBottom CD

0.074nm67.41nmTop CD

0.021°

0.43Å

Std Dev (1σ)

86.2°Sidewall Angle

1226.8ÅQuartz Depth

Mean Value

Center point measured 30 times consecutively

Quartz Depth

Top CD

Bottom CD SWA

17

Template and Corresponding Imprint MeasurementTemplate and Corresponding Imprint Measurement

Measurements of Template and DTR disk made using the same Template

Depth, and CD results of the Template and the DTR disk are compared

Template

Imprinted Disk

18

Template and Corresponding Imprint MeasurementTemplate and Corresponding Imprint MeasurementTemplate Model and Data FitTemplate Model and Data Fit

Cross Section Top-Down, Pitch=120nm

Typical Analysis Fit

Sidewall Angle

Bottom CD

Etch Depth

Fused Silica

Top CD

0

20

40

60

80

100

200 300 400 500 600 700 800 900

Rs,

Rp,

Ts,

Tp

(%)

Wavelength (nm)

Rs-exp Rs-cal Rp-exp Rp-cal Ts-exp Ts-calTp-exp Tp-cal

19

Template and Corresponding Imprint MeasurementTemplate and Corresponding Imprint MeasurementDTR Imprint Sample Model and Data Fit DTR Imprint Sample Model and Data Fit

Cross Section Top-Down, Pitch=120nm

Typical Analysis Fit

Sidewall Angle

Top CD

Bottom CD

Polymer Height

RLT

Ta

Polymer

Glass

0

10

20

30

40

50

60

200 300 400 500 600 700 800 900

Rs

& R

p (%

)

Wavelength (nm)

Rs-exp Rs-cal Rp-exp Rp-cal

20

Template & Corresponding DTR DiskParameters Simultaneously Measured and Mapped

Corresponding DTR Imprint Disk• Polymer Height

Mean = 84.1nm, STD = 2.3nm• Residual Layer Thickness

Mean = 27.8nm, STD = 6.9nm• Top CD

Mean = 83.3nm, STD = 5.2nm• Bottom CD

Mean = 74.7nm, STD = 6.3nm• Sidewall Angle

Mean = 85.5°, STD = 0.7°

Results for a 60 Point Radial Map

Template• Quartz Etch Depth

Mean = 85.2nm, STD = 1.1nm• Top CD

Mean = 51.9nm, STD = 6.0nm• Bottom CD

Mean = 38.4nm, STD = 5.0nm• Sidewall Angle

Mean = 82.8°, STD = 0.2°

21

CD Correspondence Between CD Correspondence Between Template and DTR Imprint Disk Template and DTR Imprint Disk

Template Pitch ⎯ Bottom CDMean = 120nm ⎯ 38.4nm = 81.6nm

Std Dev. = 5.0nm

Good Correlation between the Template and Imprint Sample – Same Uniformity Pattern

Imprint Disk Top CDMean = 83.3nm

Std Dev. = 5.2nm

Template Pitch ⎯ Top CDMean = 120nm ⎯ 51.9nm = 68.1nm

Std Dev. = 6.0nm

Imprint Disk Bottom CDMean = 74.7nm

Std Dev. = 6.3nm

Uniformity pattern between Template and DTR Disk were similar in both cases

22

BPM MeasurementsBPM Measurements

Cross Section Top-Down, Pitch=36nm

Typical Analysis Fit

Sidewall Angle

Top CD

Bottom CD

Polymer Height

RLT

Ta

Polymer

Glass

0

10

20

30

40

50

60

70

200 300 400 500 600 700 800 900

Rs

& R

p (%

)

Wavelength (nm)

Rs-exp Rs-cal Rp-exp Rp-cal

23

BPM DiskParameters Simultaneously Measured and Mapped

Results for a 60 Point Radial Map

• Polymer HeightMean = 74.1nm, STD = 1.9nm

• Residual Layer ThicknessMean = 29.8nm, STD = 2.5nm

• Top CDMean = 26.8nm, STD = 0.6nm

• Bottom CDMean = 17.2nm, STD = 0.9nm

• Sidewall AngleMean = 85.6°, STD = 0.5°

24

BPM Results BPM Results Static & Dynamic Static & Dynamic RepeatabilityRepeatability

0.5386.00.0685.7Sidewall Angle (°)

0.8217.70.1016.8Bottom CD (nm)

0.6926.70.0926.5Top CD (nm)

1.1633.20.1529.3Residual Layer Thickness (nm)

2.2472.40.3374.2Polymer Height (nm)

Std Dev (1σ)MeanStd Dev (1σ)Mean

Dynamic RepeatabilityStatic RepeatabilityParameter

• Repeatability Measurements performed at 1 location• For Static Repeatability, the location is measured 10 times without stage

movement• For Dynamic Repeatability, the location is measured 10 times with sample

load/unload between measurements

25

SummarySummarySensitivity of optical metrology is the key challenge in measuring Imprint Templates, DTR and BPM DisksTo surmount this challenge, an optical metrology system must include the following features:− A high signal to noise ratio over the entire measured WL range, − Cover a wide range of wavelengths,

∗ Since changes in measured spectrum, caused by variations of parameters, can affect various regions of the spectrum

− Include transmittance, not just reflectance, for measuring templates− A valid physical model for analysis− Include proper hardware for handling disks

It is shown that the n&k Gemini includes the above mentioned features

26

Thank You