A Paradigm Shiftee290h/fa05... · Lecture 20: On-Wafer Sensors EE290H F05 Spanos 3 Thermocouples • operating principle Peltier-Seebeck effect, up to 3000o C T gradient along wires

Lecture 20: On-Wafer Sensors

SpanosEE290H F05

1

In-chamber and on-wafer sensors

A Paradigm Shift


SpanosEE290H F05

2

Overview• Exact chamber environment control is relatively new• Various sensors (pressure, gas flow, gas composition,

temperature) are needed to accomplish it.• An interesting transition to “on-wafer” sensors holds

much promise...


SpanosEE290H F05

3

Thermocouples• operating principle

Peltier-Seebeck effect, up to 3000o CT gradient along wires of different materials develop different emfemf measures junction Tplatinum rhodium alloy, or silicon based sensitivity 100-200μV /oK

• problemsbig problems with shield designradiative effectslow signal -- need amplifiers or use thermopileinvasivegas T measurement is very hard, especially < 10-4 torr

• commentsinexpensive, low drift low bandwidth accuracy ~+/- 5oC at 800oC where do you want to measure T ?


SpanosEE290H F05

4

Acoustic Wave sensors

• operating principle – acoustic wave is transmitted through body– surface and internal waves propagate through body at

T dependent speed– interference with source gives beats – beat frequency determines T

• issues– implementation difficulty – invasive– calibration


SpanosEE290H F05

5

Pyrometry• operating principle

– hot objects radiate– radiation is wavelength dependent– radiation model for black bodies (Planck's Law)

λ in microns, T in oK, Rλ

– for non-black bodies need to account for emissivity• issues

– surface properties affect radiation– multiple internal reflections– emissivity is wavelength and geometry dependent– can change during processing– calibrations via thermocouples, difficult

)1(37418

/143885 −= Te

R λλ λ


SpanosEE290H F05

6

Pressure Sensors

• direct gauges– displacement of a solid or liquid surface– capacitance manometer, McLeod pressure transducer

• indirect gauges– measurement of a gas related property– momentum transfer, charge generation

• huge range of available sensors– cost– sensitivity– range


SpanosEE290H F05

7

Capacitance manometer

• basic idea– pressure differential causes

displacement of diaphragm– sense capacitance change

between diaphragm and fixed electrode

– resolution 10-2 %at 2 hertz and 10-3 torr


SpanosEE290H F05

8

Gas flow meters• differential pressure meters• thermal mass flow meters

– mass flow = K / (T1 - T2)– K depends on specific heat of gas etc.– must be calibrated for different gases– accuracy ~ 1 sccm at flows of 40 sccm– low bandwidth because of thermal inertia


SpanosEE290H F05

9

Mass Spectrometers

• two types– flux analyzers : sample gas through aperture– partial pressure sensors : analysis in exhaust stack

• issues– recombination in mass spec tube changes – indistinguishable species : (ex: CO, N2 and Si have

same amu (28))– pressure measurements are removed from

processing chamber


SpanosEE290H F05

10

RGA• basic idea

special kind of mass spectrometermeasures gas compositions works at low vacuum < 10-5 torrion beam is produced from gas sample by e-bombardmentbeam is collimated by electric fieldsq/m ratio of ions determines bending in B fielddetection of ions via a Faraday cup

• issuesquadrupole (magnetless design)very noisy !!good for diagnosticscan withstand 500 oCcan also be used at higher pressures with differential pumpsmass range 50 amu, resolution 2 amu,


SpanosEE290H F05

11

How about placing sensors on the wafer???

Sensarray products


SpanosEE290H F05

12

Calibration is an issue...


SpanosEE290H F05

13

Long Term Reliability also an Issue...


SpanosEE290H F05

14

On-Wafer Etch Rate by Resonant Structure

IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, VOL. 11, NO. 2, MAY 1998A Novel In Situ Monitoring Technique for Reactive Ion Etching Using a Surface Micromachined SensorMichael D. Baker, Frances R. Williams, Student Member, IEEE, and Gary S. May, Senior Member, IEEE


SpanosEE290H F05

15

Remote reading of resonant sensor


SpanosEE290H F05

16

Noise is the biggest problem...

On the bench... In the chamber...

When plasma is on...


SpanosEE290H F05

17

But it works! (almost)

Innovative

noisy

intrusive

may contaminate...


SpanosEE290H F05

18

Our VisionIn-situ sensor array, with integrated power and telemetry

Applications:process control, calibration, diagnostics & monitoring,process design


SpanosEE290H F05

19

Issues

• Sensor arrays– inexpensive, modular– environmentally isolated– transparent to wafer handling robotics– on-board power & communications

• Operating mode– no equipment modifications !!– Smart “dummy” wafer for in-situ metrology


SpanosEE290H F05

20

Test Case: Etch Rate

• Onboard etch-rate sensor for plasma etch– many sensor points on a wafer– accurate film thickness measurement– real-time data available– etch-friendly materials– wired power and communications (for now)


SpanosEE290H F05

21

Transduction Scheme - Etch Rate

Van der Pauw structure: ρπ

⎟⎠⎞

⎜⎝⎛⎟⎠⎞

⎜⎝⎛=

VIt 2ln


SpanosEE290H F05

22

Current Design

• Integrated Sensor Wafer Test Design• 57 etch-rate sensors on a 4” wafer• Full-wafer addressing of each sensor from a single die• Redundant interconnect to enhance yield• Four styles of sensor, selectable from a single die• On-board current-sourcing• Wired power and communications (at first)• Expandable to allow wireless power and communication


SpanosEE290H F05

24

Experimental Procedure

• Bond wires to wafer– solder wires to “strip header”– glue header to wafer edge– wire bond from header to wafer’s bond pads

• Verify operation on bench• Place wafer in XeF2 Chamber

– Measure film-thickness / etch-rate in real time– Calibrate using Nanospec thickness measurements


SpanosEE290H F05

25

Pictures


SpanosEE290H F05

26

Results

• 8 sensors (in a row) wired together in series• Everything works perfectly!• In-Situ XeF2 test performed

– XeF2 etch rate much too fast (~0.2μm/sec)– Sensor structure only 0.45 μm thick, gone in 2 sec– Sensors wired in series so when one etches through,

all measurements stop⇒ Data collected during etch, but no calibration

available


SpanosEE290H F05

27

Data - Etch #3

0 1 2 3 4 5 6 7 8 9 100

100 0

200 0

300 0

400 0

500 0P o ly s ilic on Th ic k ne s s vs . Tim e fo r E x pe rim en t #3

Tim e (s ec )

Mea

sure

d Th

ickn

ess

(A)

0 1 2 3 4 5 6 7 8 9 100

50 0

100 0

150 0

200 0P o ly s ilic on E tc h -R a te vs . Tim e fo r E x pe rim e n t #3

Tim e (s ec )

Filt

ered

Etc

h-R

ate

(A/s

ec)


SpanosEE290H F05

28

How about completely wireless???


SpanosEE290H F05

29


SpanosEE290H F05

30

“Smart dummy” developed in 1998

• Developed and tested at the UC Berkeley Microfabrication Laboratory.

• 4 sensors, wafer covered with layer of epoxy• LED used for real-time, one-way transmission


SpanosEE290H F05

31

First Test results in plasma, 1999

13.56MHz, 100W, 0.76 Torr, O2


SpanosEE290H F05

32

An Update on OnWafer Sensors

• OnWafer technologies Inc, a Berkeley startup, was founded in 2000.

• Today OnWafer products are in use in all of the major fabs around the world, and by all the major tool makers (LAM, Applied, TEL, Nikon).


SpanosEE290H F05

33

Basic OnWafer System

IR DongleIR Dongle

Base StationBase Station

SensorWaferSensorWafer

ShipperShipper


SpanosEE290H F05

34

feedbackprocess control

processingequipment

data

OnWafer

base station

The Approach

wafers to be processed


SpanosEE290H F05

35

• 42 sensors/wafer, 1Hz• 0.5 °C accuracy• Rechargeable. • Functional up to 140 °C, several kW RF• Suitable for oxide/poly plasma etch• Non-contaminating, cleanable and reusable

PlasmaTemp SensorWafer


SpanosEE290H F05

36

Example - Process Monitoringof 200mm Poly Etching

Increased HeLow He

pre-etch

main etch

de-chuckover etch


SpanosEE290H F05

37

Cool chuck - 200mm Poly Etching

main etch

Temperature fluctuations during main etch


SpanosEE290H F05

38

Can see rotating magnetic field !

phase delay in temp fluctuationCan calculate B-field periodCan see rotation is clockwise


SpanosEE290H F05

39

Example - Gas flow trouble in TEL DRM Etcher

“before” data is hotter, further, the pre-etch step is significantly less uniform…“before” data is hotter, further, the pre-etch step is significantly less uniform…


SpanosEE290H F05

40

Example - Comparison between 8 PEB plates on a 193nm wafer track (+/- 0.1C accuracy)

11 12 13 14 21 22 23 24Best!

Worst!

Data collection in two 10-minute cassette-to-cassette missionsData collection in two 10-minute cassette-to-cassette missions


SpanosEE290H F05

41

On-Wafer PEB / CD Analysis

P6P6P5P5P4P4P3P3P2P2P1P1

Six TEL ACT 8 plates used for 90nm CD lines (193nm Lithography)Six TEL ACT 8 plates used for 90nm CD lines (193nm Lithography)


SpanosEE290H F05

42

Much more than you ever wanted to know about Post Exposure Bake

Overshoot

Cooling

Steady

Heating

Chill

200mm ArF90nm

130oC 60sec

Courtesy OnWafer Technologies


SpanosEE290H F05

43

PEB Evolved from a Single Zone to Multi-Zone Control System Why?

Multi-Zone Control

Single Zone Control

10 Years of Product

Evolution

Post Exposure Bake Track Equipment Complexity is Increasing


SpanosEE290H F05

44

PEB Hotplate Thermal Profile Optimization System

Baseline Thermal Profile

Condition

Offset Generator

Engine

Plate Type Specific Thermal Profile Modeling

Engine

Offset ValuesOptimized for Both Within-Plate and

Plate-to-Plate Thermal Profile

Uniformities

Output

Input

BakeTemp™& OnView™

AutoCal™

Input



SpanosEE290H F05

45

PEB Temp Control

16 plates, 120 ºC Target

2.700oC

Target = 120oC

0.175oC

Before After



SpanosEE290H F05

46

Spatial PEB/CD Distribution Correlation

• Plotting both the bake plate temperature trajectory and R2

from temperature-CD correlation against bake time:

Max R2 during the transient heating period

Continued high R2

during steady state due to poor temperature control in single-zone plate design

• Plotting both the bake plate temperature trajectory and R2

from temperature-CD correlation against bake time:

Max R2 during the transient heating period

Continued high R2

during steady state due to poor temperature control in single-zone plate design


SpanosEE290H F05

47

PEB Hotplate Critical Dimension Optimization System

Baseline Thermal Profile

Condition

Offset Generator

Engine

Plate Type Specific Thermal Profile Modeling

Engine

Offset ValuesOptimized for Both Within-Plate and

Plate-to-Plate Critical Dimensions

Uniformities

Output

Input

BakeTemp™& OnView™

AutoCal™

Input

Input

Resist & Litho Cell Specific CD

Modeling Engine

AutoCD™

Baseline CD Profiles per Plate Customer

Provided

Input



SpanosEE290H F05

48

CDU Improvement

AcrossPlate Plate to

Plate

AutoCD

AutoCalPOR

0

0.5

1

1.5

2

2.5

3

3.5

AutoCDAutoCalPOR



SpanosEE290H F05

49

The New Problem

How can we improve the across-wafer CDU? How much can we improve CDU?

Poor Across-Wafer CD Uniformity

Processing Tool

EtchEtch

Wafer

LithoLitho


SpanosEE290H F05

50

Supervisory Control with Wireless Metrology

Compensate for systematic spatial non-uniformities across the litho-etch sequence using all available control authority:

– Exposure step: die to die dose

– PEB step: temperature of multi-zone bake plate

– Etch: backside pressure of dual-zone He chuck

Exposure PEB /Develop Etch

Wafer-levelCD MetrologyOptimizer

Scatterometry/CDSEM

dose temperature He pressure


SpanosEE290H F05

51

Next Step: Zero-Footprint Metrology Wafer

Prototyping a zero-footprint metrology wafer with optical detection unit and encapsulated power source.

Data Transmission

Photo-/RF Transmitter

Data Processing, Storage Unit

Dielectric Layer

Si

BatteryData Acquisition Unit

500μm


SpanosEE290H F05

52

Self-contained wireless transmitter

+ -Power

+ -

Power Management & RF Transmission Unit

Measurement Units

Integrated excitation/detection Unit

Power Unit Thin film Battery

Proposed Architecture


SpanosEE290H F05

53

3 x 3 Pixels Optical Metrology Prototype

Bottom Wafer with LEDPhotodetector integrated

Top Wafer


SpanosEE290H F05

54

Feasibility Test: Thickness Measurement

LEDPD

Si

Packaging Substrate

Glass Slide

Test Coating

Filter

5mm

Green LED

Blue LED

1500 1600 1700 1800 1900 2000 2100 2200-16

-14

-12

-10

-8

-6

-4

-2

Ref

lect

ion

Inte

nsity

(a.u

.)

PR Thickness (nm)

Excitation λ=525nm

Theoretical Curve

1500 1600 1700 1800 1900 2000 2100 2200-16

-14

-12

-10

-8

-6

-4

-2

Ref

lect

ion

Inte

nsity

(a.u

.)

PR Thickness (nm)

Excitation λ=525nm

Theoretical Curve

Shipley S1818 PRon Glass Slide


SpanosEE290H F05

55

Wireless Aerial Image Metrology

Mask

light

Image system

Wafer

NAPartial coherenceIllumination aberrationsDefocusmagnification

Aerial image

Latent image

Resist image

Mask image


SpanosEE290H F05

56

An Integrated Aerial Image Sensor

How can a μm detector retrieve nanometer-scale resolution of the aerial image?

How can a μm detector retrieve nanometer-scale resolution of the aerial image?

Dark contact mask forms a “moving” aperture to capture incident electromagnetic field.

Poly-silicon mask

Substrate

Photo-detector

Mask aperture

Φ1 Φ2 Φ3 Φ1 Φ2 Φ3

p-Si


SpanosEE290H F05

57

Moiré Patterns for Spatial Frequency Shift

Patterns Overlap Pattern rotates 4o Pattern rotates 8o Pattern rotates 16o

Nar

row

CD

Wid

e C

D


SpanosEE290H F05

58

Aperture pattern shift testing

Image pattern

Detect pattern

4.4

2.2

2

4.4

d

Detector mask layout design

Mask layout

-40 -20 0 20 40 60 80 100 120 140 16035000

40000

45000

50000

55000

60000

-20 0 20 40 60 80 100 120 140 160

60000

70000

80000

90000

100000

110000

120000

image pattern

inte

nsity

x(pixel)

Measurement result


SpanosEE290H F05

59

Near-Field Optical Simulation

Intensity at the center of the simulation domainm

ask


SpanosEE290H F05

60

Spin

HMDS

PA Bake

Exposure

PEB

Develop

PD Bake

Photoresist Removal

ELMELM

Poly Etch System

ADIADI AEIAEI

Etch

Etch

Etch

Present Status of “Active” CD Control


SpanosEE290H F05

61

On-wafer and in-line metrology in pattern transfer

Spin

HMDS

PA Bake

Exposure

PEB

Develop

PD Bake

Photoresist Removal

ELMELM

Poly Etch SystemEtch

Etch

Etch

T (t, x, y)T (t, x, y)

T (t, x, y)V (t, x, y)E (t, x, y)…

T (t, x, y)V (t, x, y)E (t, x, y)…

I (x, y)I (x, y)

OCDOCD

Thin FilmThin Film

OCDOCD OCDOCD


SpanosEE290H F05

62

CDU control has to incorporate many strategies

Spin

HMDS

PA Bake

Exposure

PEB

Develop

PD Bake

Photoresist Removal

ELMELM

Poly Etch SystemEtch

Etch

Etch

T (t, x, y)FF control

T (t, x, y)FF control

T (t, x, y)V (t, x, y)E (t, x, y)FF/FB Control, chuck diagnostics

T (t, x, y)V (t, x, y)E (t, x, y)FF/FB Control, chuck diagnostics

I (x, y)Optimal Pattern Design

I (x, y)Optimal Pattern Design

OCDProfile Inversion

FB Control

OCDProfile Inversion

FB Control

Thin FilmFB/FF Control

Thin FilmFB/FF Control

OCDFB Control

OCDFB Control

OCDFB/FF Control

OCDFB/FF Control

Documents

A Paradigm Shiftee290h/fa05... · Lecture 20: On-Wafer Sensors EE290H F05 Spanos 3 Thermocouples • operating principle Peltier-Seebeck effect, up to 3000o C T gradient along wires