The total energy flux at the substrate: is it the sole parameter governing the phase formation of reactively sputtered metal oxide thin film ?

In the present study, our goal is to contribute towards a better understanding of the phase formation of transition metal oxide thin films synthesized by reactive magnetron sputter deposition. In general, one would invoke the variation of the energy flux at the substrate surface (e.g. through ion bombardment during growth) as a way to manipulate the phase constitution of magnetron sputtered functional thin films [1-3].
In the first part of this work, a titanium target is sputtered in an Ar/O2 atmosphere by systematically varying the working conditions (magnetic field architecture of the cathode, sputter power, and sputtering regime). The normalized energy flux (φnorm) was calculated from in situ energy flux measurements carried out with a heat flux sensor [4] located at the sample position. Our data emphasize that the flux of IR photons emitted by the hot Ti target surface should be taken into account when calculating the total energy flux. It also appears that, regardless of the working conditions, the film phase constitution evolves from phase pure anatase to rutile-rich anatase/rutile phase mixtures as φnorm increases. Anatase is the low-temperature stable phase of TiO2 and rutile the high-temperature phase.
In the second part, the phase formation of ZrO2 films grown by DC-MS is investigated. A Zr target is sputtered in an Ar/O2 ambient, in the so-called poisoned mode or in the transition zone with the help of voltage feedback control loop [5]. The films grown in the poisoned regime always exhibit the low-temperature stable monoclinic phase of ZrO2. By working in the transition zone, the films are oxygen deficient and only the high-temperature cubic phase is detected. DFT calculations support the notion that the cubic phase is promoted solely because O vacancies are incorporated in the zirconia lattice. These last results demonstrate that the energy flux provided to the growing film is actually not the only not 'knob' available to tailor the phase constitution of functional transition metal oxide thin films.
[1] I. Petrov et al, J. Vac. Sci. Technol. A21, S117 (2003).
[2] P. Löbl et al, Thin Solid Films 6090, (1994).
[3] S. Mráz and J.M. Schneider, J. Appl. Phys. 109, 023512 (2011).
[4] A.L. Thomann et al, Rev. Sci. Instrum. 77, 033501, 2006
[5] SpeedfloTM, Gencoa, UK