Nano-Scale Structures Fabricated using Anodic

Aluminum Oxide Templates

Outline

I Introduction and Motivation

II Porous Alumina Masks
III Results
IV Conclusions
V NanoLab Experiments

OU NanoLab/NSF NUE/Bumm & Johnson

 Objective:
Introduction
Fabricate ordered arrays of structures on the nanometer scale using porous alumina
templates.

OU NanoLab/NSF NUE/Bumm & Johnson

Integrated Circuits
Moores Law:
 Dr. Gordon E. Moore, founder of Intel, predicted in 1965 that the
number of transistors per IC doubles every 18 months.

http://www.intel.com/research/silicon/mooreslaw.htm
OU NanoLab/NSF NUE/Bumm & Johnson
Semiconductor Roadmap
Year 1999 2002 2005 2008 2011 2014
Technology 180 130 100 70 50 35
Generation (nm)
DRAM Half Pitch 180 130 100 70 50 35
(nm)
MPU Gate Length 140 85-90 65 45 30-32 20-22
(nm)
Gate Oxide 1.9-2.5 1.5-1.9 1-1.5 0.8-1.2 0.6-0.8 0.5-0.6
Thickness (nm)
MPU transistor 6.6 M 18 M 44 M 109 M 269 M 664 M
-2
density (cm )
MPU Speed 1250 2100 3500 6000 10000 16903
(MHz)
Cost/transistor 1735 580 255 110 49 22
(microcents)
Important characteristics of “The 1999 National Technology Roadmap
for Semiconductors” published by the SIA.

 Current technology hits a roadblock in about 2012 in terms of fabrication and

device operation.
 Alternative patterning techniques and computing schemes are needed (e.g.
Quantum, Molecular, Optical Computers, Carbon Nanotubes based devices, etc.).
OU NanoLab/NSF NUE/Bumm & Johnson
Motivation: General
What is Anodic Porous Alumina?
 Aluminum oxide grown on an Al substrate in an electrolytic cell. The
resulting structure consists of an array of tunable nanometer-sized pores
surrounded by an alumina backbone.

Purpose:
 To understand the mechanisms involved in the growth and ordering of
anodic porous alumina.
Motivation:
 Interest in using anodic porous alumina as a nano- template to
fabricate nanometer-sized structures (e.g. nanofabrication of quantum
dots).
Why do we want to fabricate nanostructures?
1. Fundamental physical interest in the nanometer size regime. Properties of
nano-sized structures are different from their bulk and molecular counterparts.
2. Technological applications as electronic and optical devices.
OU NanoLab/NSF NUE/Bumm & Johnson
Motivation: Applications
1. Physics:
 Explore optical, electrical, and
Commercially available
Anopore filter.
http://www.2spi.com/catalog/s
pec_prep/filter2.html

magnetic quantum confinement.

2. Engineering:
 Microfiltration. Porous Alumina used as optical
waveguide.
 Optical waveguides and H. Masuda, et. al., Jpn. J. Appl. Phys.
photonic crystals for 38, L1403 (1999).

optical circuits.
 Template for carbon
nanotube growth for
electronic, mechanical
applications.
 Ordered arrays of Ordered arrays of carbon
quantum dots for lasers, nanotubes fabricated using a
photodetectors. porous alumina template.
J. Li, et al., Appl. Phys. Lett.
 ULSI memory devices 75(3), 367 (1999).

and ICs.
OU NanoLab/NSF NUE/Bumm & Johnson
Overview of Anodic Oxide Films
Fabrication
 Anodize aluminum in electrolyte
(e.g. Oxalic Acid)

Two main types of anodic oxide films can be grown

depending on the nature of the electrolyte:
1. Barrier-Type Films:
 Grown Oxide Insoluble in Electrolyte
 Nearly Neutral Electrolytes (pH 5-7)
2. Porous-Type Films:
 Grown Oxide Slightly Soluble in Electrolyte
 Aqueous Sulfuric, Oxalic, and Phosphoric Acid Electrolytes
OU NanoLab/NSF NUE/Bumm & Johnson
Historical Timeline
 1920’s Porous alumina starts to be used commercially to protect and finish
bulk Al surfaces.
 1940’s-1960’s With advent of electron microscopes, first characterization
of structure of porous alumina, but growth theories are experimentally
unsubstantiated.
 1970 Manchester group does first real experimental work showing pore
radius dependence on applied voltage,etc.
 1992 First “quantitative” theoretical attempt to explain pore growth from
first principles by Belorus group.
 1995 Japanese group discovers pores will self-order into close packed
array under the right anodization conditions.
 1996-Present Use of porous alumina for nano-applications abound.
 1998 Although mechanism for ordering still not clear, German group
proposes one possible mechanism.

OU NanoLab/NSF NUE/Bumm & Johnson

Porous Alumina
 Anodize aluminum in electrolyte (e.g. Oxalic Acid).
Apparatus
 Oxide grows at the metal/oxide and oxide/electrolyte
interfaces, pores initiate at random positions by field-assisted
dissolution at the oxide/electrolyte interface.
 Ordering requires appropriate potentials and long anodization
times.
 Ordering results from repulsion between neighboring pores
due to mechanical stress at the metal/oxide interface.

Resulting Structure

H. Masuda and K. Fukuda, Science 268, 1466 (1995).

OU NanoLab/NSF NUE/Bumm & Johnson
Barrier-Type Anodic Oxide Films
Growth Mechanism
 Oxide growth proceeds at the Aluminum
anode (+).
 Hydrogen gas is evolved at the Platinum
cathode (-).
 The current between the cathode and
anode is carried by the electrolyte.

 Oxidation reactions at theAl anode

2 Al  3OH  Al2 O3  3H   6e 
{ 2 Al  3O 2   Al 2O3  6e  }
 Electrolysis of water at aluminum oxide/ electrolyte interface
H 2 O  OH   H 
( aq )

{ 2 H 2 O  2O 2   4 H 
( aq ) }

 Reduction reaction at the cathode:

2 H   2e   H 2 
 The overall electrochemical reaction occurring is:
2 Al  3 H 2 O  Al2 O3  3 H 2
OU NanoLab/NSF NUE/Bumm & Johnson
Barrier-Type Anodic Oxide Films
Growth Mechanism
 Oxide growth proceeds at the metal/oxide and the oxide/electrolyte
interface.
 Growth proceeds due to the motion of ions under the applied field.

 Growth at the metal/oxide interface is due to oxygen containing anions

(mainly OH- and O2-) moving through interstitial/vacancy sites.
 Growth at the oxide/electrolyte interface is due to Al3+ cations moving
through interstitial/place exchange mechanisms.

OU NanoLab/NSF NUE/Bumm & Johnson

Overview of Film Anodization
 Oxide growth proceeds via ionic
conduction and reaction of Al
cations and oxygen containing
anions under the influence of an
applied field.
(e.g. 2Al+ + 3OH-  Al2O3+3H+
+6e-)
 Pores initiate at random positions
through field-assisted dissolution of
the oxide at the oxide/electrolyte
interface.
 Initially oxide growth
dominates. (I)
 Dissolution becomes
competitive, barrier layer
thins, and pores initiate. (II)
 Approaches steady state
where both mechanisms occur
at roughly the same rate. (III
and IV)
V.P. Parkhutik, and V.I. Shershulsky, J. Phys. D:Appl. Phys. 25, 1258
OU NanoLab/NSF NUE/Bumm & Johnson (1992).
Porous-Type Anodic Oxide Films
Field-Assisted Dissolution
 Application of a field across the oxide polarizes the oxide bonds.

 This polarization effectively lowers the activation energy for

dissolution of the oxide.
 This promotes solvation of Al3+ ions by water molecules and the
removal of O2- ions by H+ ions.
 This NUE/Bumm
OU NanoLab/NSF processes&is strongly dependent on the E-field strength.
Johnson
Ordered Growth of Porous Alumina

 In 1995, Japanese group found that

pores will self-order under the right
anodization conditions.
 The two most important conditions
are narrow voltage ranges and long
anodization times.

OU NanoLab/NSF NUE/Bumm & Johnson

 Ordered Nano-Templates
 Tunable diameters and spacings from
20 nm to 500 nm.
 Polycrystalline structure: ordered
micron-sized domains, defects at
grain boundaries.
 Low temperature growth produces
unordered 4-10 nm arrays.

Ordered Oxalic

Near-Ordered Sulfuric

OU NanoLab/NSF NUE/Bumm & Johnson

 Ordered Growth of Porous Alumina
 Ordered pore arrays obtained in three
different electrolytes for long anodization
times and appropriate voltages (specific for
each electrolyte).
 Polycrystalline structure with perfectly
ordered domains a few microns in size.
Defects occur at grain boundaries.

OU NanoLab/NSF NUE/Bumm & Johnson

 To create an ordered Mask Processing
through-hole mask:
1. Anodize for a long time
allowing pores to order. 1.
2. Chemically remove the AFM of Unopened
alumina in a mixture of
phosphoric and chromic acid. 2. Barrier Layer
(1 m x 1 m)
3. Anodize for a short time
(now pores are ordered). 3.
4. Coat top surface of alumina
with a polymer (collodion)
to protect it from further
processing.
4.

5. Remove Al Substrate in a
saturated HgCl2 solution. 5.
6. Remove the barrier layer in
5 wt.% Phosphoric Acid. 6.
7. Remove collodion and place
alumina on desired 7.
substrate.

OU NanoLab/NSF NUE/Bumm & Johnson H. Masuda et al. , Jpn. J. Appl. Phys. 35, L126 (1996).
Pattern Transfer Techniques: Results

1. Etching Processes
Fluorine Beam Ion Beam
Transfer mask pattern via etching into Transfer mask pattern via ion etching
substrate for ordered arrays of into substrate for ordered arrays of
trenches. trenches or pillars.

2. Growth Processes
Sputtering and Thermal
Deposition
Transfer mask pattern via deposition
onto substrate for ordered arrays of
dots.

OU NanoLab/NSF NUE/Bumm & Johnson

F-Etched Array of Si(001) Nano-Holes
SAMPLE: ~500nm thick Free-Standing AAO/Si(001)
F-ETCH: 1 min. 20 sec.
TSUB = 250oC
PORES: Width 70 nm, Depth 100-120 nm
TOP DOWN VIEW X-SECT. VIEW

50 nm

200 nm
 Walls are ~30 nm thick (near top).
OU NanoLab/NSF NUE/Bumm & Johnson
Ion Etched Array of GaAs Nano-Holes
SAMPLE: ~500nm thick Free-Standing
AAO/GaAs(100)
ION BEAM: 500 eV Ar+, 0.05 mA/cm2
Time = 2hrs. 12min.
PORES: Width 50 nm, Depth 50-60 nm

OBLIQUE VIEW ~TOP DOWN VIEW X-SECT. VIEW

OU NanoLab/NSF NUE/Bumm & Johnson

Thermally Evaporated Nano-Dots: MgF2
SEM Top Views

MgF2 dots/Si Au dots/SiO2

AFM Views 200 nm

Line Scan
18
16
14

Height (nm)
12
10 ~60 nm
8
6
4

3-D 2
Rendered Height: 12 nm 11%0
Diameter: 60 nm 9% 0 100 200 300 400 500

Spacing: 110 nm 5% Distance (nm)

OU NanoLab/NSF NUE/Bumm & Johnson
Thermally Evaporated Nano-Dots: Gold
 Porous alumina used as an evaporation
mask to grow quantum dots.

H. Masuda et al. , Jpn. J. Appl. Phys. 35,

L126 (1996).

OU NanoLab/NSF NUE/Bumm & Johnson

Ion Etched Array of GaAs Nano-Pillars
SAMPLE: ~20nm thick Fe dots on GaAs(100).
ION BEAM: 500 eV Ar+, 0.05 mA/cm2
Time = 17 min.
PILLARS: Width 50 nm, Height 50 nm

OBLIQUE VIEW TOP DOWN VIEW X-SECT. VIEW

Note: No Fe
remaining.

OU NanoLab/NSF NUE/Bumm & Johnson

Evaporated Catalyst Dots For Carbon
Nanotube Growth
SAMPLE: ~20nm thick Fe catalyst dots on 100nm Ti/Si
GROWTH: CVD using Methane gas at 500 Torr, 800oC
NANOTUBES: Multi-walled tubes, ~10s of microns long

TOP DOWN VIEW

 Collaboration with
Dr. Shen Zhu of
Marshall Space
Flight Center.

OU NanoLab/NSF NUE/Bumm & Johnson

 Conclusions
Fabricated ordered, arrays of nanostructures using porous alumina templates as
masks:

 Arrays of 50 nm wide trenches in Si and GaAs by atom-beam and sputter

etching.
 Arrays of 50 nm dots of various materials onto substrates by evaporation
and sputtering.
 Arrays of nano-pillars in Si and GaAs by etching nano-dot arrays.

Future
 Make pores smaller (to 5 nm) using sulfuric acid electrolytes and low temp.
anodization.
 Seed for carbon nanotube growth.
 Explore optical, electrical, and magnetic properties of nanostructures.
 Explore ways to transfer single or arbitrary dot/trench patterns.
 Fabricate such nanostructures in situ in multichamber MBE system.
OU NanoLab/NSF NUE/Bumm & Johnson
NanoLab Class: AAO Templated Structures

• Fabricate AAO Masks:

– Ordered and Disordered Oxalic Masks (50 nm/100 nm).
– Ordered film: 15 hr first anodization.
– Disordered film: 1 hr first anodization.

• Lift-Off onto Silicon and Quartz substrates.

– Silicon substrates for SEM characterization.
– Quartz substrates for UV-Vis characterization.

• Thermally Evaporate Gold onto all Samples

– Must be done one sample at a time, because alignment is critical.

• Characterize Samples
– AFM -both samples
– SEM of Au dots on Silicon.
– UV-Vis of Au dots on Quartz.
OU NanoLab/NSF NUE/Bumm & Johnson