Implant Process Characterization With Modern In-line Metrologies

July 16, 2009

Junction Technology Group, Semicon West 2009 Program

Outline
Introduction to Semilab Implant metrology
Dose Implant depth Junction depth Sheet resistance Doping profiles Activated surface dopant density

Will illustrate techniques that provide above parameters Summary

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

Semilab Background
We have brought several powerful techniques into the Semilab

family We offer multiple products for monitoring the implant/anneal processes The specific customer needs will determine the best fit.

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

The Ion Implant Process

Deposit photo-resist Expose / develop Perform blanket implant on product wafers and often monitor

wafers also
Measure implant dose, depth Anneal, to activate dopant Measure sheet resistance, junction depth, activation, profiles

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

Monitoring Before Anneal

Provides SPC monitoring of implanters and implantation process
Real-time monitoring, immediate feedback

Measures
Dose Implant Depth

Assumes damage = f(dose)

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

Monitoring After Anneal

The dopant species is activated by the anneal The activated dopant affects device performance Measure
Sheet resistance Junction depth Profiles

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

Obsolete or Timeless?
Before anneal
Therma-Wave (traditional approach) Measures in TW units, dependent on energy and dose Actually measures the damage caused by the implant process

After anneal
4-Point Probe Measures sheet resistance (ohms/square)

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

Semilab Offerings
Dose Implant depth BX CI QCS SPV Rs JPV Blanket Activation FastGate Nsurf Product Activation SDi NSD Blanket

Ion Implant

Anneal

Profiles SSM SRP Hg CV Junction depth BX CI

Dose

Leakage JPV FastGate JL

8

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

QCS ICT-300
Non-Contact Fast Mapping Metrology
Based on ac-SPV Method
h
Re h Rh
Illuminated Dark

e
Re

Bulk p-type Implant defect levels

Both as-implanted and annealed wafers Energy range: 0.5keV to 3.0MeV Dose range: 1E10cm-2 to 5E15cm-2 All common species: B, P, As, BF2,F, He, In, etc. Repeatability: < 1% (for low/medium dose < 0.5%)

Rh Wd VSPV Wd

EF

What is Measured?

Implanted Silicon
Implant dose, energy, angle
VSPV = I eh kT 1
V SPV

Annealed Silicon
Average doping density
kT ln ( N sc / n i ) I eh qN sc 1

q 2 ni N d R

1/ 2

Nd implant induced defect density R -- implant region width -- capture probability

NSC doping concentration Ieh light intensity -- light modulation frequency

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

ICT-300 Implant Monitoring Capabilities

Implanter Micro-Uniformity Detection

Implant system A B 5keV 1E15

Implant system B P 45keV 8E14

Implant system C As 3keV 4E15

Implant system D B 5keV 1E15

Implant system E B 2keV 3E15

Implant system F B 0.65keV 1E15

Multi-Implanter SPC: B 31keV 8e12

Implanter_A Streaming Wafer
Av 17116 St dev 1.42%
2 1 3 4 5 6 7 8 9 10 11 12

Correlation to Final Electric Test

Implant: As75 2e12 cm-2 300keV

Implanter_C Streaming Wafer

Av 17888 St dev 1.94%
2 3

10

11

12

13

14

15

16

Implanter_D Streaming Wafer

Av 18472 St dev 1.42% 2 3 4

10

11

12

13

14

15

16

17

18

Implanter_E Streaming Wafer

Av 18160 St dev 2.32% 1 2

QCS
as-implanted
8 9 10

ET

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

10

SDI measurement technique and Vsb corrections

Implant profile - concentration dependent on depth concentration

SDI independently measures 2 parameters: surface barrier (Vsb) and ac-SPV signal (VSPV)
2 (Vsb NSD = N A = q kT ) q

SCR depth at Vsb = 250mV

CD

COCOS (ref. Wilson [1])

VCPD(dark)=VOX+Vsb(+const) VCPD(light)=VOX(+const) Vsb= VCPD(dark)- VCPD(light)

SCR depth at Vsb = 400mV depth Measurement depth = Space Charge Region depth W Calibration: NSD versus surface barrier, VSB, dependence is measured and implant specific calibration is introduced. Measured data corrected for variations in surface barrier, VSB.
C alib ratio n 40keV
2 .8 0 E+1 7 2 .6 0 E+1 7 2 .0 0 e 1 2 cm-2 2 .2 5 e 1 2 cm-2 2 .5 0 e 1 2 cm-2 y = 6 .3 E+1 7 x + 3 .8 E+1 6 R = 9 .9 E-0 1 y = 5 .4 E+1 7 x + 2 .5 E+1 6 R = 9 .9 E-0 1 y = 4 .3 E+1 7 x + 2 .6 E+1 6 2 R = 9 .9 E-0 1 0 .1 0 .2 0 .3 0 .4 0 .5
2 2

ac-SPV (ref. Nakhmanson [2])

Concentration, [cm-3]

2 .4 0 E+1 7 2 .2 0 E+1 7 2 .0 0 E+1 7 1 .8 0 E+1 7 1 .6 0 E+1 7 1 .4 0 E+1 7 1 .2 0 E+1 7 1 .0 0 E+1 7 8 .0 0 E+1 6

const Ieff CD = VSPV

1. M. Wilson et al., ASTM STP 1382, (1999) 2. R. Nakhmanson, Solid State Electron. 18, 617 (1975)
Slide 090716001

Su r 2009 Semilab ALL RIGHTS RESERVED 2009 Semilab ALL RIGHTS RESERVEDfa c e B a r r ie r , Vs b , [V]

11

NSD Ion implant measurement data

Outstanding measurement Precision (P/T)
3.0E+17

25keV P implant correlation vs dose

NSD repeatability
2.5E+17

6.000E+14 Doping Concentration 5.600E+14 5.200E+14 4.800E+14 4.400E+14

NSD (at/cm3)

Average: 5.0843 e14 cm StdDev: 3.3374 e11 cm-3 StdDev%: 0.07%

-3
2.0E+17

y = 35989x + 1E+16 R2 = 0.9989

1.5E+17

1.0E+17

5.0E+16

4.000E+14 0 2 4 6 8 10 12 14 Run Number

0.0E+00 0.E+00 1.E+12 2.E+12 3.E+12 4.E+12 5.E+12
2

6.E+12

7.E+12

8.E+12

Implanted dose (at/cm )

Includes full wafer mapping capability

500keV P implant Dose: 5E12 (0, 0) angle

NSD provides excellent day-to-day stability and P/T capability, accounting for variations of wafer surface state. Dose measurement range : 1x1010 to 1x1014 cm-2; plus higher doses with junction Software automatically corrects for light reflectivity due to oxide films.
12

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

MBIR USJ Data Analysis

Refractive index of the doped layer was calculated using

Drude model, and rectangular profile of the concentration.

Model fit was performed with three varied parameters: doped

layer thickness, activated carrier concentration and carrier mobility.

Wavenumber range used : 600-7000 cm-1.

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

13

MBIR Correlation With Reference Methods Sheet Resistance

1/sheet resistance, MBIR (1/ohm) 0.008
MBIR thickness (nm)

Doped Layer Thickness

140 120
y = 1.1294x 2 R = 0.9923

y = 0.7409x 0.006 R = 0.9938

2

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

0.004

0.002

samples 1-7 samples 8-9

0 0 0.002 0.004 0.006 0.008 0.01 1/sheet resistance, 4pt. probe (1/ohm)

SIMS junction depth (nm)

Activated Dopant Dose

1.6E+15 1.4E+15

product of the mobility and activated dopant dose (times the electron charge): 1/Rs=eDose

MBIR dose (1/cm )

MBIR 1/Rs is calculated as the

1.2E+15 1E+15 8E+14 6E+14 4E+14 2E+14 0 0 0.002 0.004 0.006 0.008 0.01

samples 1-7 samples 8-9

1/sheet resistance, 4pt probe (1/ohm)

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

14

Principle of JPV Sheet Resistance Measurements

e- and hole generation by chopped LED

light This causes change in junction voltage Change spreads laterally, and the attenuation depends on sheet resistance Change of the potential is picked up by capacitive sensors Signal depends strongly on LED chopping frequency (f) Rs, Cd and Rd (Jleak) are calculated by fitting the theoretical JPV signal

Current spreading

surface potential

Junction Photovoltage-based sheet resistance measurement: a method for non-contact implant monitoring with high-resolution mapping capability. Basic principle:

Potential spreading
R U = IdRs = I s r 2r

pulsed LED

2 concentric capacitive electrodes

1 I = 2r J(r0 r ) + iCd U R r d

position on wafer relative to excitation LED (r)

Equation to solve to
R 2U U +r r 2 s + iRs Cd U + J(r0 r ) = 0 r R r 2 r d
2

obtain Rs

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

15

JPV Sheet Resistance Measurements

Applications and specifications

Sheet Resistance and Leakage Current Mapping

Implant process control (Rs depends on dose and energy) on various implant types: USJ, deep implant, pocket implant, plasma immersion implant, etc. Dose range: >5E11cm-2 Species: B, P, As, BF2, etc.

Visualization of non-uniformities, implanter errors (striping, etc.) which are not detectable by low resolution methods Detects variations and inhomogeneities smaller than 1 % of the wafer average Works on oxidized and non-oxidized wafers

Repeatability test on 3 different wafers Repeatability < 0.2 %

3700 3600 3500 3400 3300 3200 3100 1 2 3 4 5 6 7 8 9 10 Slot 1 Slot 2 Slot 3

JPV 4PP correlation: Same wafer. Top half is 4PP map (625 points) Bottom half is JPV map (17,000 points) Both maps take the same time to make.
16

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

Principle of Carrier Illumination Technology

Probe laser
(980 nm IR)

Generation laser
(830 nm red)

Generation laser creates excess carriers and, where significant damage is present, heat. Excess carrier gradient forms index of refraction gradient Probe laser uses index of refraction gradient or surface heat to determine junction depth, dose level or PAI depth. Generation laser is modulated (2kHz) to enable high signal/noise ratio.

Detector Cognex PatMax Vision system

Beam splitter

Objective lens

Beam
Excess carriers

2-um spot size

Junction

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

17

Carrier Illumination Technology

Implant monitoring

Measurements on Boron implanted wafers

Dose response on BF2 implant

Dose range: 1x1010 cm-2 to 1x1016 cm-2 Energy range: 100 eV to 3 MeV Species: As, B, P, BF2, In, Sb Depth range10 nm to 100 nm Depth range:10 to 70 nm Measurement on SOI substrates Cu (metal) via structures measurement Integration with Semilab JPV method

PAI depth Junction depth measurement Scope applications for development:

-12340 -12380

BX Signal(uV)

-12420 -12460 -12500 -12540 1.2E+15 1.3E+15 1.4E+15

-2

1.5E+15

1.6E+15

1.7E+15

Dose (cm )
Slide 090716001
2009 Semilab ALL RIGHTS RESERVED

18

Contact Probing Metrology

SSM-Hg CV SSM FastGate CV SRP SSM 2000

......... .........

97-7rgm201

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

19

Contact Probing Metrology

SOI
1.E+18
n

1.E+17 1.E+16 Carrier Density (cm-3)

p

1.E+15 1.E+14
n n

1.E+13 1.E+12
oxide

1.E+11 1.E+10 0 2 4 6 8 10 12 14 Depth (mm) SOI 1 SOI 2

Freescale ISPSD06

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

20

Ion Implant Matrix-Sensitivity

Ion Implant Metrology Sensitivity
1E+7 CV surface
Super deep w ell (CCD)

1.0

JPV ac-SPV
0.8
Deep w ell (latch up)

1E+6 4PP

Ion Energy (eV)

1E+5

Mid w ell (punchthru)

SIMOX

0.6

ThermaWave

Retro grade Channel

1E+4
Halo S-D contact pSi gate
S-D extensio n

0.4

Channel

1E+3

0.2

1E+2 1E+9 1E+10 1E+12 1E+13 1E+14 1E+15 1E+16 1E+17 1E+18 1E+11

0.0

Ion Dose (cm )

-2

Slide 090716001

1E+19

Sensitivity ( signal/ dose)

2009 Semilab ALL RIGHTS RESERVED

21

Ion Implant Matrix-Range

Ion Implant Metrology Range
1E+7

Super deep well (CCD)

SRP

1E+6
Deep well (latch up)

Ion Energy (eV)

1E+5

Mid well (punchthru)

Retrograde Channel

SIMOX

1E+4
Halo CV surface Channel S-D contact pSi gate
S-D extension

1E+3

JPV ac-SPV
1E+2 1E+10 1E+12 1E+13 1E+14 1E+15 1E+16 1E+17 1E+18 1E+19 1E+9 1E+11

Ion Dose (cm )

Slide 090716001
2009 Semilab ALL RIGHTS RESERVED

-2

22

Case Study-AVS Insight 09

Former QCS Former SSM

Former competitors learning that their techniques are complementary in many areas. The winner in this case study is the metrology end user.

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

23

Summary
We have brought several powerful techniques into the Semilab

family We offer multiple products for monitoring the implant/anneal processes We look forward to discussion of your needs to find the best fit.

Slide 090716001

2009 Semilab ALL RIGHTS RESERVED

24