10

20

30

40

50

60

70

80

-4 -2 0 2 4 6 8 10 12

Current Density Effects on Elastic Modulus

Elas

tic M

odul

us G

PA

Depth µm

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

0

1

2

3

4

5

6

-4 -2 0 2 4 6 8 10 12

Effects of Current Density on Hardness

Unsealed Dyed Selaed 5 min @ 65 C

Dyed Sealed 5 min. @ 65 C .5 amps

Har

dnes

s G

PA

Depth µm

Bulk

Al

umin

um Porous Layer

Barrier ____Layer

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

0

1

2

3

4

5

6

-4 -2 0 2 4 6 8 10 12

Effects of Dyeing on Hardness

Unsealed Sealed 5 min. @ 65 C

Dyed Selaed 5 min @ 65 C

Har

dnes

s G

PA

Depth µm

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

10

20

30

40

50

60

70

80

-4 -2 0 2 4 6 8 10 12

Effects of Dyeing on Elastic Modulus

Elas

tic M

odul

us G

PA

Depth µm

ConclusionsThe mechanical properties as a function of sealing procedures used to prepare decorative anodized aluminum in this work showed no time or temperature dependence. However, the process of sealing caused an overall decrease in hardness of the films. This decrease in hardness was more extreme as the distance from the substrate increased and resulted in poor edge retention during polishing. There was little difference under these sealing conditions between the any of sealed samples with one exception.

The sample anodized at a higher current density showed an increase in hardness across the thickness of the porous layer while the barrier layer and bulk aluminum remained unchanged. This film proved thicker for the same growth time and is suspected to be less porous, leading to the increase in hardness and modulus thorough the porous layer.

The films grown in this study are not uniform as a function of position through the film. The properties of anodic aluminum films can be controlled with anodizing procedures but sealing time, temperature, and dying have no apparent effect on the films grown here, while increased current density increases the hardness and stiffness of the films.

AcknowledgementsThis work was supported through the National Science Foundation: Division

of Materials Research REU site program under grant number 0139125.

Characterization of Anodic Aluminum Thin Films Through Nanoindentation

Wayne Burnett, David F. BahrResearch Experience for Undergraduates; Washington State University, Pullman

Summer 2004

Effects of Current DensityIncreased current density during film growth has shown an increase in film properties in the porous layer. When current is doubled during film growth the hardness and modulus of the barrier layer and bulk aluminum remain unchanged when compared with the dyed and sealed and the unsealed sample. The properties of the porous layer still decrease across the layer as the distance from the barrier layer increase but the change is less negative. Sample seven was measured to be approximately 12 microns thick; the other six samples were around 10 microns. This indicates that for the same growth time higher current densities grow thicker films. This film also does not show the characteristics of poor edge retention seen in the other sealed samples nor does it show the dramatic softening at the extreme edge of the sample that was noted in the industrial grade film. The changes in the properties are expected to be due to a change in porosity in the film due to the increased current density and will be investigated in future work.

Effects of DyeingThe effects of dyeing were illustrated by comparing the unsealed sample with a sealed sample and a sample that was dyed and sealed. The dye was shown to have a similar effect on the mechanical properties as sealing time and temperature with the effects of the sealing process out weighing that of the dye.

MethodsAll the films that were grown and investigated in this project were grown using industrially accepted procedures on 6061 aluminum in an electrolyte of sulfuric acid. The Elastic modulus and hardness of the cross-section of seven locally grown samples and one industrially grown sample were measured using the DSM technique. The first film was not dyed and remained unsealed; the second sample was sealed for five minutes at a temperature of sixty-five degrees Celsius. The third sample was sealed for fifteen minutes at sixty-five degrees Celsius. The forth and fifth sample were both sealed for five minutes but at a temperature of fifty degrees and seventy-five degrees Celsius respectively. Sample six was the same as sample two with the exception of an industrial grade organic dye that was added prior to sealing. The seventh sample wassealed under identical conditions to sample six but the film was grown at .68 amps rather than the .34 amps of all the other samples.

Following the initial film preparations the samples were sectioned and mounted in Epomet. The samples were mechanically ground and polished with particular care to allow for maximum edge retention. Even small amounts of round off excluded large portions of the films from testing.

All seven samples were indented in a grid of evenly spaced indents, such as the one imaged on the right, to allow the properties at a given film depth to be accurately mapped for each sample.

10 µmCross Section of Anodic Aluminum Film following Indentation

Procedure

Bulk Aluminum

Al2O3 Film

Film Edge

DSM Nanoindentation Method

Why nanoindentation?Traditional indentation methods of using Vickers indenter failed due to the brittle nature of the film. At a load of 50 grams both radial and circumferential cracks form, while a load of 25 grams caused the film to delaminate from the substrate. Testing conducted at lower loads produced indents that where inadequate in size to be measured accurately. Even if the films did not undergo failure the thickness of the films that were to be measured were of a similar thickness as the resulting Vickers indents.

SEM Micrographs of Film Failure Under 50 g and 25 g Load Using Vickers Indenter.

IntroductionThe Structure of Anodic Aluminum Films

Aluminum is anodized to provide increased corrosion protection, Increased wear resistance, and for decorative purposes. The typical structure of a decorative anodized aluminum film is composed of three layers. The first layer lies in contact with the aluminum and is composed of a dense compact layer of AL2O3. This layer is commonly referred to as the barrier layer and can be seen in the micrograph below as the thin stripe running through the lower portion of the sample. The next layer is composed of porous AL2O3. This layer makes up the largest portion of the film. The structure in this layer istaken to be hexagonal with tubular pores in the center that run orthogonal to the surface of the aluminum. These tubes are closed on one end by the barrier layer and the sealing layer on the other. The pores are commonly filled with dye or other performance enhancing compounds. The sealing layer is composed of böhmite and seals the additives inside the pores. Aluminum Substrate

Dense Al2O3 Layer

Porous Al2O3

Böhmite Sealing LayerEpoxy Mount

Load

mN

Displacement nm

Since the film’s properties continuously change through the thickness of the film it was necessary to extract an elastic modulus and hardness value from each indent. The Dynamic Stiffness Measurement allows for accurate measure of mechanical properties of films using a single indent. This is achieved by imposing a sinusoidal force signal to the loading portion of the loading-unloading curve providing stiffness and displacement data continuously up to the maximum load.

Results

20

30

40

50

60

70

80

90

-4 -2 0 2 4 6 8 10 12

Sealing Time Effects on Elastic Modulus

Elas

tic m

odul

us G

PA

Depth µm

Porous Layer

Bulk

Al

umin

um

Barrier ____Layer

0

1

2

3

4

5

6

7

-4 -2 0 2 4 6 8 10 12

Sealing Time Effects on Hardness

Unsealed Sealed 5 min. @ 65 C

Sealed 15 min. @ 65 C

Har

dnes

s G

PA

Depth µm

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

1

2

3

4

5

6

7

8

-2 0 2 4 6 8 10 12

Hardness of Industrial Grade Film

Har

dnes

s G

PA

Depth µm

30

40

50

60

70

80

90

100

110

-2 0 2 4 6 8 10 12

Elastic Modulus of Industrial Grade Film

Elas

tic M

odul

us G

PA

Depth µm

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

20

30

40

50

60

70

80

-4 -2 0 2 4 6 8 10 12

Sealing Temperature Effects on Elastic Modulus

Elas

tic M

odul

us G

PA

Depth µm

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

0

1

2

3

4

5

6

-4 -2 0 2 4 6 8 10 12

Sealing Temperature Effects on Hardness

Unsealed Sealed 5 min. @ 65 C

Sealed 5 min @ 50 CSealed 5 min. @ 75 C

Har

dnes

s G

PA

Depth µm

Bulk

Al

umin

um

Porous Layer

Barrier ____Layer

Time EffectsSealing time also showed to have little effect on the propertiesof the film. Sealing times of 5 and 15 minutes were compared to an unsealed sample. The 15 minute sample showed a higher hardness and elastic modulus through the barrier layer than any of the other samples. Variations in laboratory conditions are expected to be the cause of this as no other sample showed any changes in barrier layer properties as the result of sealing. The same condition of poor edge retention shows that sealing has a greater effect than time variation on the properties of the porous layer.

Temperature EffectsThe effects of sealing temperature on hardness and elastic modulus proved to be minimal as sealing temperatures of 50 C, 65 C, and 75C were compared to an unsealed sample. Due to poor edge retention a large amount of the film was excluded from testing. This poor edge retention and the slight deviation of the sealed samples from the unsealed sample, seen in the graphs to the right, imply that sealing affects the film far greater than the temperature at which the films were sealed. Sealing, however, does not seem to affect the properties of the barrier layer or bulk aluminum.

To verify the accuracy of the nanoindenter to the cross-section of the anodic aluminum system an industrially grown standard was indented and compared to the results of Alcalá et al. This work studied the mechanical properties of anodic aluminum barrier films grown in an electrolyte of ammonium pentaborate and shows values of 140 GPa for the elastic modulus and 8.5 GPa for hardness. The maximum values for the industrial grown film were 7.34 GPa for hardness and 100 GPa for the elastic modulus of the barrier layer portion of the film. Variations are related to differences in anodizing conditions. The values of hardness and modulus in the aluminum portion of the sample were comparable to that of the accepted values of aluminum. The porous portion of the film showed a linearly decreasing trend until the extreme outer portion of the film. It is expected that this decrease is the result of a change in porosity as the film thickens and is to be the subject of future work. The sharp change in properties seen after the ten micron mark is likely due to the böhmite sealing layer.

Nanoindentation of the films grown in the lab follow the expectations associated with the accepted structure of anodic aluminum films and that demonstrated by the industrial standard. The measured value for the elastic modulus of the aluminum substrate was centered on the accepted value of 69 GPa and hardness values were consistently between 1 and 2 GPa. Variations in these values were the result of poor surface finish and inherent errors associated with nanoindentation techniques. The effects of sealing temperature, sealing time, dye and current density during growth were investigated here.