Procedures to grow porous anodic alumina films with ordered and controllable arrangements of nanoscale pores have led to extensive research on PAA films as functional materials. Experiments and calculations show that transport in the oxide is assisted by plastic flow driven by mechanical stress (1-3). Pores initiate as the result of a morphological instability at the oxide-solution interface of the initially formed barrier oxide (4). Stress distribution measurements revealed high near-surface compressive stress levels, which may be associated with locally elevated concentrations of incorporated electrolyte anions. The instability leading to pore formation may be influenced by flow driven by anion incorporation-induced stress (5). Direct evidence for anion-induced stress in aluminum oxide was sought through stress measurements during open-circuit immersion of Al in phosphoric and sulfuric acids. The stress is compared with elastic anion incorporation-induced stress calculated from X-ray photoelectron spectroscopy (XPS) measurements of film composition.

Al sheet samples of 99.998% purity were exposed at open circuit in either 0.4 M H₃PO₄ or 0.4 M H₂SO₄ solutions. In situ stress measurements during acid dissolution were carried out by phase-shifting curvature interferometry (6). According to the thin-film Stoney approximation, the detected sample curvature changes are directly related to the surface force change per sample width. The force per width is the biaxial in-plane stress integrated through the sample thickness. XPS was performed with a K-alpha instrument (Thermo Scientific) (7). The relative concentrations of P, S, Al, and O in the oxide were determined from areas of P2s, S2s, Al2p, O1s peaks. The oxide thickness and the anion concentration per area were determined from the relative areas of the Al2p peak contributions from oxide and metal.

Fig. 1 shows that upon immersion of Al in both acids, force per width increased abruptly in the compressive direction to 2 min, and then relaxed slowly to constant levels. The initial compressive increases are not due to mechanical disturbances, as they significantly exceed the stress change in distilled water. The hypothetical anion incorporation-induced stress was calculated from the anion concentration per area by assuming that all the detected anions are absorbed within the oxide. The elastic force per width accompanying incorporation is then

F_wA = -(2/9) W_AE_oxN_A/(1 - n_ox)

where W_A is the anion molar volume, E_ox and n_ox the elastic modulus and Poisson's ratio of alumina, and N_A the per-area anion concentration from XPS. Fig. 1 shows that for dissolution in H₂SO₄, F_wA is in excellent agreement with the measured force at all times. For H₃PO₄, agreement during the initial compressive increase is found but thereafter phosphate adsorption causes the calculation to diverge. Fig. 1 is strong evidence that anion incorporation in alumina films induces compressive stress. The stress of ~ 2 GPa inferred from the figure are comparable to near-surface stress levels measured earlier in anodic films (5).

Additionally, we discuss evidence for anion-incorporated surface layers in anodic alumina from high-voltage cyclic voltammetry (CV) experiments (8). At high scan rates, the current-voltage dependence in the cathodic-direction scan reveals the conduction characteristics of the anodic film. By varying the anodic potential limit of CV, the dependences of conduction on both film thickness and electric field are assessed. High-resistivity surface layers are revealed in anodic films formed in phosphoric, sulfuric and oxalic acids. An instability mechanism for anodic alumina is described in which both the high resistivity and elevated stress level of the anion-containing near-surface layer play important roles.

ACKNOWLEDGMENT

Financial support was provided by JFE Steel Corporation and China Scholarship Council.

REFERENCES

1. S. J. Garcia-Vergara, et al., Electrochim. Acta. 52, 681 (2006).

2. J. E. Houser, K. R. Hebert, Nature Mater. 8, 415 (2009).

3. J. Oh, C. V. Thompson, Electrochim. Acta, 56, 4044 (2011).

4. P. Mishra, K. R. Hebert, Electrochim. Acta, 222, 1186-1190 (2016).

5. Ö. Ö. Çapraz, et al., Electrochim. Acta 167, 404-411 (2015).

6. Ö. Ö. Çapraz, K. R. Hebert, P. Shrotriya, J. Electrochem. Soc., 160, D501-516 (2013).

7. Ö. Ö. Çapraz et al., Electrochim. Acta, 238, 368-374 (2017).

8. S. Ide, P. Mishra and K. R. Hebert, Electrochim. Acta, 221, 1-7 (2016).