**4. Conclusions**

Atomic layer deposition presents the opportunity to apply nanometer-scale corrosion barrier coatings to very high aspect ratio structures and component interiors, which is not possible with many other coating techniques. The results presented here show the promise of ultra-thin ALD barrier coatings in chloride environments. DC voltammetry measurements produced 10–15x increases in polarization resistance, up to 3.4 MΩ cm2, with 50 nm nanolaminate films of Al2O3 and TiO2, two materials that are readily available and widely deposited by ALD. With longer equilibration time than for the DC measurements, the nanolaminate films saw nearly a 100x increase in the polarization resistance determined from equivalent circuit fitting of impedance spectra, up to 12 MΩ cm2. This was even higher, well above 100 MΩ cm2, when the initial electroreduction step was removed and could be further increased using thicker films or different materials.

The stability of single-layer Al2O3 ALD films on copper was found to be poor during immersion in neutral 0.1 M NaCl, with severe degradation in measured impedance response occurring within 24 h. This agrees with previous reports of the dissolution of ALD alumina on active substrates [76,90] and has been attributed to hydroxide generation during oxygen reduction at the substrate surface exposed through pores or defects in the ALD film [90]. TiO2 was found to be very stable, although it showed the

smallest initial increase in corrosion resistance because of its inferior nucleation and sealing capabilities compared to alumina. Nanolaminate film structures, including ATx1 [10 nm Al2O3 + 40 nm TiO2], ATx5 [(5 nm Al2O3 + 5 nm TiO2)x5], and ATx10 [(2.5 nm Al2O3 + 2.5 nm TiO2)x10], were successful at combining the superior sealing properties of alumina with the excellent chemical stability of titania. ATx5 exhibited the best overall performance according to impedance spectroscopy over 72 h. We attribute this to the beneficial effects of multiple thin film layers without individual layer quality suffering from being too thin to form a dense and effective barrier. It remains to be seen how these samples perform over longer immersion times and in flowing electrolyte conditions.

The effect of dissolved oxygen in the electrolyte was probed through various electrochemical measurements. As expected, the presence of oxygen enhances corrosion in chloride-containing media. However, perhaps unexpectedly, dissolved oxygen was found to have little effect on the stability of ALD films. The ability of copper exposed through coating pores to passivate caused there to be minimal variation in impedance over 72 h, comparable to deoxygenated conditions. Laser microscope imaging revealed little difference in surface morphology between nanolaminate samples exposed to oxygenated or deoxygenated electrolyte.

Also explored was the use of different deposition temperatures to determine its effect on film quality and corrosion barrier efficacy. An obviously higher initial film quality existed for Al2O3 single layers deposited at 250 ◦C, though degradation over time was found to be at least as, if not more, severe than for the films deposited at 150 ◦C, according to impedance spectra. Substrate temperature during deposition of TiO2 is more complicated in that it impacts the crystallinity as well as the dielectric quality and contaminant concentrations. As with alumina, the initial dielectric quality of the titania films was found to increase with deposition temperature, although crystallization and increased porosity at 250 ◦C resulted in poor performance over the 72-h immersion. The stability of TiO2 at 100 ◦C was good, but it was still outperformed by the 150 ◦C deposition, which appears to be in the range of an optimal temperature for titania ALD for corrosion protection. ATx10 was also deposited at 250 ◦C and showed good stability over 72 h, unlike both the single-layer alumina and titania deposited at this temperature.

**Author Contributions:** Conceptualization, M.A.F., C.J.O., G.N.P.; methodology, M.A.F.; software, M.A.F.; formal analysis, M.A.F.; investigation, M.A.F.; resources, G.N.P.; data curation, M.A.F.; writing—original draft preparation, M.A.F.; writing—review and editing, C.J.O., G.N.P., M.A.F.; visualization, M.A.F., C.J.O., G.N.P.; supervision, C.J.O. and G.N.P.; project administration, C.J.O. and G.N.P.; funding acquisition, C.J.O. and G.N.P.

**Funding:** This research was funded by the U.S. Navy under contract N00253-16-C-0002.

**Acknowledgments:** This work was performed in part at the Analytical Instrumentation Facility (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation (award number ECCS-1542015). The AIF is a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), a site in the National Nanotechnology Coordinated Infrastructure (NNCI).

**Conflicts of Interest:** The authors declare no conflict of interest.
