Motivation

1
Motivation

• IC business requires a sub 100 nm Next
  Generation Lithography tool.
• (100 nm for 16GDRAM)
• Any of the following 4 major candidates are not
  prevailing.
• EUV(Extreme UV)
• SCALPEL(SCattering with Angular Limitation in
  Projection Electron beam Lithography)
• X-ray with Synchrotron
• IPL(Ion Projection Lithography)
• Generally, it is assumed that due to the large
  lateral straggling of ions in the membrane mask,
  it is not possible to get high resolution with
  ion beam - which is not necessarily so.
• As a first step towards Ion beam lithography
  (IBL) using membrane mask, it is necessary
  to demonstrate the good spatial resolution

2
Advantage and Disadvantage of IBL

• Advantage
• Good sensitivity for 0.1 um pattern
• X-ray 375 mJ/cm2
• e-beam 100 uC/cm2
• IBL 4.5 uC/cm2 (720mJ)
• Good intrinsic resolution
• 10 nm limitation not from the wavelength but
  from PR

• Disadvantage
• vacuum treatment
• 11 membrane mask
• lateral straggling
• non familiar method - no extensive study

3
Current Status of Ion Beam Lithography

• IPL
• IMS (Ionen Mikrofabikations System) and Vienna
  University since 1986
• ALG consortium in USA
• Siemens, ASM lithography, Leica and IMS-Stuttgart
  formulated 36M 3-year research program in 2000
• 0.1 um pre-production type stepper in 1999

• IBL with membrane mask
• No dominant study after the proximity IBL by
  Hughes Research Laboratory

4
Experiment
5
KIGAM Implantation System
6
Simulation of Dose distribution at PR

• Purpose To see and understand the dose
  distribution at pattern edges which is directly
  responsible for the edge definition in the
  development process
• Simulation tool TRIM (SRIM2000)
• Simulation Geometry simple infinite slit

7
Factors affecting the line definition

• Mask Quality
• mask production by e-beam writer
• problem approx. 1 mm thick PMMA should be used
  - Resolution worsening
• distortion during irradiation

• Ion Beam quality
• Parellelity and homogeniety
• dose measurement

• Development
• precise temperature control - find the
  temperature at which until the midde irradiated
  point is developed
• not controllable by develop time because of the
  statistical character of melting process

8
Change of molecular weight by proton irradiation

• Molecular weight of PMMA changes drastically by
  proton irradiation which enables the very well
  defined structure reproduction

9
Result of simulation - 1mm slit
10
Result of simulation - 1mm slit

• Small conclusion
• Theoretically, the edge definition can be
  controlled within 20 nm if the development
  process can be performed very precisely
• Even taking into account the 14 - 86 dose
  width, edge definition can be controlled at least
  within 50nm with rather rough develop condition

11
Comparison of Simulation and Experiment- for the
case of large mask to PR distance
Depth profile of PMMA after development Proton
Energy 500keV Membrane 2mm Si3N4 shadow width
100mm Mask to PR distance 35mm
12
Extreme Cases

• Depth profile of PMMA after development
• Proton Energy 500keV
• Membrane 2mm Si3N4
• shadow width 100mm

Mask to PR distance 0
Mask to PR distance 530mm
13
AFM results
Edge configuration 500keV proton Au wire mask w/o
membrane
Edge configuration 800 keV proton Au wire mask
with 10 mm mylar membrane
14
SEM observations
400keV with membrane mask to sample
contact tilt angle 50o
450keV with membrane mask to sample 25mm tilt
angle 50o

• 500keV w/o membrane
• tilt angle 50o

15
Conclusion

• Simulation results show the good possibility of
  employing IBL using membrane mask as the NGL
  tool.
• Well below 100nm pattern definition can be
  obtained if develop condition can be found at
  which only until the middle dose position at the
  pattern edge is developed.
• There are still, however, many basic works to be
  performed before real launch. They are
• 1. The relationship between proton dose, develop
  condition (Temperature, time) and pattern edge
  (the position until which PR is developed)
• 2. Mask quality (e-beam writing)
• 3. Understanding the deviation of simulation
  result and the real measurement

16
?? keV ???? ??? ??? ?????? ??? ??

• ???, ??, ???, ???
• ??????? ?????????