His PhD Thesis dealt with Surface Studies on Semiconductor Nanostructures. The aim of his Thesis work was to generate knowledge relevant to the understanding of physicochemical phenomena affecting surface processes in semiconductor nanostructures of importance for the determination of their electronic properties. Inparticular, it was proposed to study the electronic surface states in ZnO nanowires, through physicochemical analytical methods sensitive to the surface.
In a first stage, efforts were dedicated to prepare ZnO nanowire samples suitable for this kind of experiments. The growth of NWs of ZnO on compacted graphite was studied in a book-type tubular furnace with two heating zones, by the vapor phase transport method under the flow of Ar and O2. In this stage, the growth parameters and NW harvest technique were optimized. Subsequently, NW growth was studied in a single-zone tubular furnace, obtaining a self-sustained sheet of high density of randomly-oriented NWs, with 60nm of average diameter and lengths between 2 and 6 μm. Then, arrays of carbon fibers were used as growth substrates, obtaining NWs of 40nm in diameter and 700nm in length around the entire circumference of the most exposed fibers and some of the underlying fibers. These NWs grew oriented with respect to the radial direction at angles of about ±30º and directly on the carbon fiber surface. The NWs grown on carbonaceous substrates were studied by X-ray diffraction. For the NW sheet the observed diffraction pattern agreed well with that
for a polycrystalline ZnO standard, and for the NWs on carbon fibers evidence of preferential orientation, associated with the fact that NWs are strongly aligned in the −30º and +30º directions, was observed.
Finally, two different methods for the transfer of NWs from the growth substrate to other substrates of interest were studied. First, a drop cast method was implemented and characterized, from a suspension of high NW density in isopropyl alcohol that was deposited on Si substrates. The NWs remained fixed on the substrates after evaporation of the solvent, and high density of NWs could be transferred, however with no proper NW orientation control. Then, a “dry” transfer mechanism from the NWs on carbon fibers was developed, which allowed controlled and semi-oriented transfer of NWs on Si substrates. This fact represents an important breakthrough since it allowed setting up a system of individual NWs with their lateral faces exposed for different surface exploration techniques.
With X- ray photoelectron spectroscopy (XPS), the surface chemical composition and the valence band near the Fermi level for ZnO NWs on compacted graphite, on carbon fibers and transferred on Si, as well as for a (0001) oriented ZnO crystal, were studied. In the first two samples the valence band maximum was determined to be 3 eV, which allowed to estimate the conduction band minimum at 0.3 - 0.4 eV above the Fermi level, in agreement with the n-type nature of ZnO. For the NWs
transferred on Si sample, a photoelectron spectrum corresponding to a combination of the XPS spectra from clean Si and from ZnO was obtained. 5% of NW coverage on Si was deduced, which agrees well with observations by scanning electron microscopy.
The room temperature photoluminescence (PL) for ZnO NWs grown on compacted graphite (on both the bottom of the crucible and in pellets), and for NWs on carbon fibers, were compared with PL for a ZnO(0001) crystal and for conventional NWs grown on Si substrates precatalyzed with Au. In the NWs on carbonaceous substrates, a very high intensity of ultraviolet emission and a very low intensity of visible emission were measured, in contrast to what happens with the NWs grown on Si. It was found that the UV emission is almost 85 times greater than the green emission for the NW sheet on compacted graphite at the bottom of the crucible, which leads to a ratio of intensities up
to 103 times larger than that obtained for ZnO NWs grown on Si by the conventional vapor phase transport method. The excitation power dependence of the PL from NWs on compacted graphite at the bottom of the crucible was compared with the results obtained for NWs grown on Si and for the ZnO(0001) crystal. The PL intensity dependence on excitation power was found to be approximately linear for the visible band and super-linear for the UV emission. Contrary to what was expected, the UV PL for ZnO NWs on compacted graphite followed the same dependence on excitation power as for the other samples studied, indicating a similar UV luminescence generation mechanism for the different samples, despite the large difference in intensities. However, the position of the UV PL peak was found to redshift with increasing excitation power in the case of NWs on compacted graphite, an effect that was not observed for any of the other samples studied. A comparison showed that while the UV emission peaks for different samples overlap without appreciable differences in shape, the weak green emission band from ZnO NWs on compacted graphite presents significant differences.
From this results, it was suggested that observed differences stem from differences in the NW growth process. It is proposed that the NWs grown on compacted graphite inside the substrate holder crucible have a much lower density of defects than the other samples, both of luminescent type and nonradiative recombination centers, giving rise to the high emission in the UV.
The dynamics of transport gases and the trajectory of the Zn vapor atoms within the tubular furnace during vapor transport NW growth were simulated for two different crucible orientations and substrate positions. The Zn vapor atoms were found to travel up to 10 times slower on the growth substrate when the crucible is in its upside up position (as used to grow NWs on compacted graphite), with respect to the upside down position (used to grow NWs on graphite pellets and Au-catalyzed Si substrates). In this way, Zn atoms contributing to NW growth in the first configuration arrive at a lower speed and hence settle into their corresponding positions in the crystal lattice in an ordered manner, giving rise to NWs with a much lower density of defects than obtained for the second configuration. Therefore, the density of surface defects that act as luminescent centers in the visible is reduced, as well as of those acting as nonradiative recombination centers. As a combined result, the emission in the visible is reduced and the intensity of UV emission due to excitonic recombinations increases considerably.
With C-AFM microscopy, the topography and three-dimensional profile of ZnO NWs transferred on Si and ZnO NWs grown by hydrothermal synthesis at low temperature on Si substrates were analyzed. In this way it was possible to access information from the NW lateral faces (10¯10) and NWs tips (0001) separately. With this technique it was not only possible to study groups of nanowires but also to analyze individual nanowires. The height profiles in the transferred NWs present an abrupt step in the position of the same, with an average height between 40 and 80 nm, while for the vertical NWs a slope profile is obtained due to artifacts produced as the AFM tip slides on the side of the nanowire as it rises. Using a special type of super fine conical tip and the high resolution QNM operation mode, the correct topographic profile of these NWs was measured, eliminating the artifacts of tip-sample interaction. In addition to topography, two-dimensional electrical current maps for both horizontal and vertical NWs with respect to the plane of the substrate were measured with C-AFM.
Finally, the I-V behavior was characterized in individual horizontal and vertical NWs. In the horizontal NWs, diode characteristics were obtained with a threshold voltage of −0.2 V. The same results were obtained for the IV curves on the (0001) surface of vertical NWs, but with a threshold voltage between −1V and −3 V. The current densities were calculated and the analysis of the IV characteristics was carried out in terms of the formation of a Schottky barrier between the AFM probe and the NW surfaces.
XPS was used to study the surface band bending on the clean ZnO(0001) crystal surface under ultra high vacuum conditions, and changes that occur during the adsorption of electropositive (Sn and Mg) and electronegative (Se and TCNQ) species. Changes of all internal levels accessible in the experiment, as well as of the valence band were recorded during the adsorption. This allowed to determine that the band bending affected in the same way the localized and extended levels, as well as the core levels of Zn and O atoms. It was found that the adsorption of electronegative species,
which charge the surface negatively with respect to the bulk, always produces upward band bending, and that of the electropositive species, which positively charge the surface, always produces downward band bending. With respect to the magnitude of the bending, it was found that, while the downward band bendings were always close to the limit given by the distances (in energy) between the VBM or the CBM and the Fermi level in the bulk, the upward band bendings were always much smaller. This was explained in terms of the incomplete filling of the acceptor level of the adsorbate, which acts as a regulator that pins the band bending at a certain value. Although very low adsorption rates were used, it was never possible to observe the gradual development of band bendings; this means that the bending is immediately formed with a very small density of adsorbates, and saturates very rapidly. Finally, band bendings were analyzed in two other cases of interest: when the ZnO(0001) surface is exposed to atmospheric conditions and when it is altered by Ar+ bombardment. In both cases, the modifications of the surface produced downward band bending. The downward displacement in the case of Ar+ bombardment was attributed to preferential sputtering of O atoms;
this produces a slight metallization of the surface region (by enrichment with Zn) and, therefore, conditions similar to those obtained by adsorbing Sn or Mg. The modification by exposure to atmospheric conditions can be explained by taking into account the low reactivity of the ZnO (0001) (Zn terminated surface) with the more electronegative adsorbates (O2 and OH) and the dominant effect of contamination associated with C.
Dr. Tosi is now in Trieste, Italy, carrying out posdoctoral work
https://www.lagaceta.com.ar/nota/809025/actualidad/joven-fisico-tucumano-fue-contratado-centro-investigacion-internacional.html
No hay comentarios:
Publicar un comentario