Skip to main content

Archived Comments for: Transition from ripples to faceted structures under low-energy argon ion bombardment of silicon: understanding the role of shadowing and sputtering

Back to article

  1. Comment on "Transition from ripples to faceted structures under low-energy argon ion bombardment of silicon: understanding the role of shadowing and sputtering" by T. Basu et al.

    Frank Frost, IOM Leipzig

    7 September 2014

    In their article T. Basu et al. have examined and analyzed the temporal evolution of silicon topography under 500 eV Ar ion bombardment at incidence angles of 70° and 72.5°. At higher ion fluences a distinct transition to faceted surface structures has been observed. The authors conclude that this transition is caused by a shadowing effect. In view of the presented experiments this conclusion is not justified, shadowing is not relevant for the experimental conditions mentioned above. In the following I will explain why shadowing is irrelevant and point out a misinterpretation of Eq. (2) in the article.
    Eq. (2) gives the limiting condition for non-shadowing of a sine wave surface profile (derived by G. Carter (Ref. [26]). As correct indicated in Fig. 1, h0 is the amplitude of the sinusoidal wave. However, in the analysis of the surface profiles (Figs. 5/6) the authors extracted the average feature height h. This average height h is, for a sinusoidal profile approach, tantamount with the peak-to-peak or peak-to-valley value of the sine wave, i. e. h = 2 h0 or h0 = h / 2. Consequently, the numerical values given on page 5 of the article change considerably. For example for an incidence angle of 70° and a fluence of 2 x 10^17 cm^-2, the value 0.551 changes to 0.275 and therefore no shadowing occurs.
    In addition, from the analysis of the presented data there is some more evidence that shadowing is not important, not even at higher fluences.
    From the profiles (fluences 20 x 10^17 cm^-2) in Figs. 5 and 6 for ion incidence angles of 70° and 72.5° downstream facet angles of 12° and 9° - 10° can be estimated (with respect to the global surface plane), respectively. Therefore, the local incidence angle to these facets is 82° and 81.5° ¿ 82.5°, respectively. That means that also for the largest surface features shadowing can be ruled out.
    As already mentioned by Carter (Ref. [26]) the onset of shadowing is accompanied by preferred erosion of the peaks whereas the valleys will no longer be eroded and the surface roughness decreases. From Fig. 7 it is seen that this not the case for the presented experiments, instead the surface roughness increases with increasing erosion time.
    Finally it should be mentioned that structure coarsening via the Hauffe-mechanism (Ref. [32]) can occur only if large parts of the downstream facets are not shadowed.
    Nevertheless, the observed faceting resulting in local incidence angles of ~ 82° is very noteworthy because exactly the same angle are found in a former study for 2000 eV Kr ion erosion of Si (S. Macko et al., Nanotechnology 21, 085301 (2010)). Faceting of ripple pattern in the low-energy was also reported for SiO2 (J. Völlner et al., J. Appl. Phys. 109, 043501 (2011)). In both cases the observed faceting is related to the angular dependence of the sputtering yield as proposed and evaluated in detail by G. Carter some decades ago (for a review see e. g. G. Carter, J. Phys. D: Appl. Phys. 34, R1-R22 (2001)).

    Competing interests

    The author declare that he has no competing interests.