some things to think about when doing evaporation liftoffof nanometer scale patterns

1/30/09

review fundamentals

sample

crucible

sample holder

revap = evaporation rate (mass/sec)

rdep = deposition rate (thickness/sec)

Evaporation Rate

eevap PkT

Mrπ2

=

revap = evaporation rateM = atomic massk = Boltzman’s constantT = temperaturePe = vapor pressure

Deposition Rate

• deposition rate depends on the location and orientation of the wafer in the chamber

Deposition Rate

θρ

cos2dr

r evapdep Ω

=

rdep = deposition rate (thickness/sec)revap = evaporation rate (mass/sec)Ω

= solid angle over which source emits (unit less steradians)d = source to substrate distanceρ

= material densityθ

= inclination of substrate away from direction to source

dθ

Ω

Uniformity

d1 d2

d1 < d2

rdep (d1 ) > rdep (d2 )

d1 d2

d1 ≈

d2

rdep (d1 ) ≈

rdep (d2 )

x

x

Case 1: d ≤

x Case 2: d >> x

2)(1d

rdep Δ=Δ

2=Δdif41

=Δ depr, then

example

Detecting DNA with Carbon Nanotube ArraysPrabhu Arumugam, NASA Ames Research Center

Devin Brown, Georgia Tech Nanotechnology Research CenterBruce Gale, University of Utah Neil Gordon, Early Warning Inc.

At left is an artist’sconception of anultrasensitive multiplexelectronics biosensorbased on a carbonnanotube nanoelectrodearray. The insets on theright represent applicationsin DNA (top) and antigendetection (bottom).

Below is a 4 inch wafer with 30 chips.

Each chip has a 3 x 3 array of 200 micron pads shown above.

Each pad has 100nm nickel dots spaced at 1 micron, shown below.

Multi-walled carbon nanotubes are grown on each nickel dot.

Carbon nanotubes offer a wide electrochemicalwindow, flexible surface chemistry,and biocompatibility. By placing athousand nanotube probes in the space ofone of today’s metal electrodes, DNA sequences can be detected from less than athousand strands. This is sensitive enoughto directly measure mRNAs in a drop ofblood or a piece of tiny tissue sample. Itmatches the upper limit of sensitivity ofconventional laser-based fluorescence techniques,but doesn’t require time-consumingsample preparation and expensive and bulkyanalytical equipment.

Carbon nanotube catalyst pattern130nm diameter on 1um pitch

100A Cr + 300A Ni

step description equipment

1spincoat PMMA A4 at 2000RPM, 1000RPM/s, 60sec CEE Brewer 100CB spincoater

2 hotplate bake 180C, 90sec CEE Brewer 100CB spincoater

3 resist thickness measurementWoolam ellipsometer (180nm), Tencor P15 profilometer (230nm)

3

EBL expose requires prealignment 100kV, 2nA, 1950uC/cm2, shot pitch = 4nm JEOL JBX-9300FS EBL system

4develop 1:1 MIBK:IPA 2min immersion, IPA immersion 30sec wet bench

5 optical microscope inspection Leitz Ergolux

6e-beam evaporate 10nm Cr @ 1A/s, 30nm Ni @ 2A/s CVC E-beam evaporator

7 acetone liftoff (2 to 3 hrs) wet bench

8SEM inspection (Hitachi full die inspection) Zeiss Ultra 60 FESEM, or Hitachi 3500 Thermionic SEM

Process Flow

Problem

want this but often get this

8

7

6

5

4

3

2

1

0

row

100 97

107 102 106 112

125 111 110 112 120 124

112 126 140 128 124 119

122 125 127 136 127 115

124 132 135 125

134 125

0 1 2 3 4 5 6 7

column

Legend

diameter

<= 100

<= 105

<= 110

<= 115

<= 120

<= 125

<= 130

<= 135

> 135

all die measured

nanodot diameters for uu27 / slot 8

229mm

50mm 25mm

12°

6°

100mm wafer

crucible

wafer holder

evaporator geometry

130nm

y = 235nm

x1

PMMA

siliconx2

yx2tan =θ

θ

yx ×= θtan2

wafer point of view to incoming metal evaporation

130nm

235nm

12°

80nm

PMMA

silicon

wafer point of view to incoming metal evaporation

angle (degrees) nanodot size (nm) top down shape side view shape

0 130

1 128

3 123

6 116

9 109

12 80

954mm

50mm

3°

100mm wafer

crucible

wafer holder

effect of increasing sample distance from crucible

in order to limit incoming angle to 3 degrees or less across entire wafer, the sample would have to be placed almost 1m away from crucible, however this would decrease evaporation rate by 1/16.

wafer flat up

33,000

11,000

22,000

11,000

6” wafer

-60000

-40000

-20000

0

20000

40000

60000y

(um

)

1018 976 851

922

953

967

971

9591109

1120

1110

1092

10641084

1118

1159

1159

11471175

1186

1183

1155

1121

1030

-60000 -40000 -20000 0 20000 40000 60000

x (um)

Cr + Ni thickness

measured by contact profilometry on the left alignment mark for each chip

wafer flat up

Cr+Ni thickness measured on this mark of each chip

6.5”

6.5”

2.75”

1. 5”

CVC1 evaporator

crucible not centered relative to sample holder

AFM measurement of nanodots

803A

1393A1092A

-60000

-40000

-20000

0

20000

40000

60000

y (u

m)

1018 976 851

922

953

967

971

9591109

1120

1110

1092

10641084

1118

1159

1159

11471175

1186

1183

1155

1121

1030

-60000 -40000 -20000 0 20000 40000 60000

x (um)

AFM measurement of nanodots

183A

371A

follow up points for next time

thicker resist is worse

characterize your resist (dose vs. feature size)