Citation Export
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Singh, Ranveer | - |
dc.contributor.author | Sial, Qadeer Akbar | - |
dc.contributor.author | Kim, Unjeong | - |
dc.contributor.author | Nah, Sanghee | - |
dc.contributor.author | Seo, Hyungtak | - |
dc.date.issued | 2022-12-01 | - |
dc.identifier.issn | 1369-8001 | - |
dc.identifier.uri | https://dspace.ajou.ac.kr/dev/handle/2018.oak/32953 | - |
dc.description.abstract | The concept of hot carrier-based solar cell is suggested for high-efficiency photovoltaics by overcoming thermodynamic limit of conversion efficiency. However, efficiency is largely affected by the quality of absorber materials and dynamics of hot carriers (HCs) in the same. Here, we demonstrate the generation of HCs and their transfer dynamics to the metal oxides by performing the femtosecond transient absorption spectroscopy measurements. The role of oxygen composition/content is highlighted on the different physical properties of hafnium nitride (HfN) thin films because the presence of oxygen in HfN deteriorates/quenches the HC generation. All HfN thin films exhibit the polycrystalline nature and the degree of crystallinity decreases with increasing the oxygen concentration as confirmed by X-ray diffraction and high-resolution transmission electron microscopic studies. In addition, the absorption measurements exhibit the broadband light absorption efficiency of the HfN films. The transient absorption maps show a strong photobleaching signal over a wide spectral range. The transient bleach dynamics probed at 650 nm show the faster decay rate (0.77 ps) for the HfN/TiOx in comparison to HfN (1.5 ps) and HfN/MoOx (3.7 ps), revealing the efficient electron transfer from HfN to TiOx layer. These results advance our understanding of hot-electron dynamics in HfN-oxide heterostructures and offers to design an ultra-fast optoelectronics device of HCs. | - |
dc.description.sponsorship | The surface of the films was characterized by scanning electron microscopy (SEM) technique for morphological analyses. The planar-view SEM image of the HfN thin films which are fabricated at working pressure of 1.0 and 3.0 mTorr (whereas RF power was fixed at 100 W) are shown in Fig. 1(c) and (d), respectively. These SEM images exhibit the granular structures and uniform distribution of the grains over the surface. The magnified SEM images of Fig. 1(c) and (d) are shown in Figs. S3(a) and (b) (Supporting Information), respectively. In addition, it is observed that there is no significant change in the surface morphology of the HfN films by varying the working pressure. Further, for every deposition, the thickness of films was fixed at 240 ± 10 nm for the comparison of structural and optical properties. For instance, the cross-sectional SEM (XSEM) images of films grown at 1.0 and 3.0 mTorr working pressure (at a fixed RF power of 100 W) are shown in Fig. 1(e) and (f), respectively. The XSEM images confirm not only confirm the uniformity in the film thickness but also the formation of closely packed and columnar nanostructures of the films. In addition, these images reveal a smooth interface between the film and the substrate as well as the absence of any microscopic cracks in the films. Further, for in-depth microstructural analysis, the cross-sectional transmission electron microscopy (XTEM) measurements were carried out, as presented in Fig. 2(a) and S4(a) (Supporting Information) corresponding to HfN films grown at 1.0 and 3.0 mTorr working pressure, respectively. These XTEM images clearly depict the formation of closely packed nanostructures in the HfN thin films which are aligned normal to substrate as well as the thickness (240 ± 10 nm) of the films. Further, the high-resolution TEM (HRTEM) measurements were performed to confirm the nature of crystallinity. Fig. 2(b) and S4(b) (Supporting Information) show the HRTEM of the HfN films deposited at 1.0 and 3.0 mTorr working pressure (whereas sputtering power is fixed at 100 W), respectively, confirming the formation of poly-crystalline films. Moreover, a line profile over the HfN films has been drown and corresponding data is plotted in the inset of Fig. 2(b) which also confirms the crystalline nature of the films. The average interplanar spacing matches well to the (111) plane of the cubic phase of HfN [35,36]. The observed results of HRTEM are in good agreement well with the XRD results, as discussed above.Further, the electron energy loss spectroscopy (EELS) spectra were measured from top to bottom [as illustrated by several spots in Figs. S5(a) and (b), Supporting Information] to understand the depth-dependent oxygen content in the films and the effect of working pressure on quality of the HfN films. The corresponding scanning transmission electron microscopy (STEM) images of 1.0 and 3.0 mTorr working pressure are shown in Figs. S5(a) and (b) (Supporting Information), respectively. The EELS spectra obtained from both films exhibit the presence of oxygen [see Fig. 2(c) and (d)]. For instance, the observed O–K onset energy of the sample with 1.0 mTorr working pressure is found to be 521.5 eV, confirming the formation of partially hafnium oxide in the film [Fig. 2(c)] [37]. In addition, Fig. 2(c) reveals that the top surface of the film is relatively more oxidized than the bottom part which has uniform oxygen concentration. Further, the O–K edge peak shifts towards the high binding energy of 525.5 eV with increasing the working pressure to 3.0 mTorr [Fig. 2(d)] and the intensity of the EELS spectra decreases from top to bottom, revealing the different content of the oxygen in the film [37]. The EELS spectra confirm that the amount of oxygen in the HfN film increases with increasing the working pressure. However, the observed O–K onset energy for both films is found to be different from the HfO2 (528.4 eV), which confirms the partially oxidation of the HfN films [38,39].Further, to probe the HC generation in HfN layer and HC injection at the metal-nitride/metal-oxide interface experimentally, we employed ultrafast TA spectroscopy measurements on the HfN, MoOx/Al2O3/HfN, and TiOx/Al2O3/HfN samples (see Experimental section). To generate the HCs, HfN films was pumped with a Yb:KGW-based femtosecond amplifier system at 2.75 eV (450 nm) which is below the band gap of MoOx or TiOx and probed with broadband white light continuum spectrum. In addition, we compared the TA dynamics as a function of delay time from −2 to 5000 ps at probe wavelength of 650 and 800 nm. Fig. 5(a)-(c) show the ultrafast TA surface maps (corresponding to a pump wavelength of 450 nm) as a function of probe energy and pump-probe delay time (t) of HfN, MoOx/Al2O3/HfN, and TiOx/Al2O3/HfN, respectively. It is interesting to note that the HfN, MoOx/Al2O3/HfN, and TiOx/Al2O3/HfN samples exhibit photobleaching (ΔA < 0) and photo-induced absorption (ΔA > 0), as shown in Fig. 5(a)–(c), respectively. We have also measured the ultrafast TA surface map of TiOx/Al2O3/HfN at a pump wavelength of 370 nm (close to the band gap of TiOx), as shown in Fig. S9(a) (Supporting Information). In addition, the bleaching features in the HfN, MoOx/Al2O3/HfN, and TiOx/Al2O3/HfN samples are found to be dominate around the probe wavelength of 650 nm within a time delay of t = 50 ps [Fig. 5(d)]. Consequently, in case of MoOx/Al2O3/HfN or TiOx/Al2O3/HfN sample, the HCs (generated from the HfN) having energy higher than the energy barrier at the heterostructure interface can overcome this barrier height and occupy energy states in the conduction band of metal oxide, leading to prominent change in the TA spectra and surface map [Fig. 5(c) and (d)]. Further, the changes in the decay dynamics of HCs will be manifest by the electron-phonon coupling time. Therefore, the decay rate (τ) of the HfN, MoOx/Al2O3/HfN, and TiOx/Al2O3/HfN samples was determined by fitting the decay dynamics [Fig. 5(d)] within t of 50 ps at probe wavelength of 650 nm. In the addition, Fig. 5(e) shows the TA decay dynamics for the longer time delay (t = 5 ns) span. The TA dynamics of the HfN sample indicate that HCs were relaxed from the excited states by two decay rates (τ1 = 1.5 ps and τ2 = 10.5 ps) through electron-electron and electron-phonon interactions. On the other hand, the bi-exponential fitting of MoOx/Al2O3/HfN (TiOx/Al2O3/HfN) samples reveals that the injected HCs dissipate their excess kinetic energy to lattice with the carrier decay rate of τ1 = 3.7 (0.77) ps and τ2 = 15.1 (9.6) ps via carrier-carrier and carrier-phonon interactions. All the fitting TA parameters are summarized in the Table- S4 (Supporting Information). The faster decay rate of HCs in TiOx/Al2O3/HfN than the MoOx/Al2O3/HfN and changes in femtosecond TA dynamics with delay time reveal that the HCs are injected from HfN to TiOx. In addition, it is observed that the HC injection is more effective and favourable in the case of TiOx instead of MoOx because of smaller barrier height at the interface of HfN and TiOx [Fig. 4(c) and (d)]. In addition, the work function difference between HfN-TiOx heterostructure is much smaller than HfN-MoOx heterostructure, providing less barrier height to HCs and good overlap between the conduction band of TiOx and the Fermi level of HfN. In addition, the TA dynamics of the HfN, MoOx/Al2O3/HfN, and TiOx/Al2O3/HfN samples at a probe wavelength of 800 nm also indicate the transfer of HCs from HfN to metal oxides [Fig. 5(f) and Fig. S9(b), Supporting Information]. Thus, the TA results clearly indicates the generation of HCs and their injection to the adjacent metal-oxide layer before recombination which is a key factor for increasing the efficiency of the ultra-fast optoelectronics devices.This work was supported through the National Research Foundation of Korea (NRF) [NRF-2019H1D3A1A01102524 and NRF-2019R1A2C2003804] funded by the Ministry of Science and ICT, South Korea. This work was also supported by Ajou University. | - |
dc.description.sponsorship | This work was supported through the National Research Foundation of Korea (NRF) [ NRF-2019H1D3A1A01102524 and NRF-2019R1A2C2003804 ] funded by the Ministry of Science and ICT, South Korea . This work was also supported by Ajou University . | - |
dc.language.iso | eng | - |
dc.publisher | Elsevier Ltd | - |
dc.subject.mesh | Barrier heights | - |
dc.subject.mesh | Hafnium nitride films | - |
dc.subject.mesh | Hafnium nitrides | - |
dc.subject.mesh | Hot electron injection | - |
dc.subject.mesh | Hot-carriers | - |
dc.subject.mesh | Kelvin probe force microscopy | - |
dc.subject.mesh | Metal-oxide | - |
dc.subject.mesh | TiO | - |
dc.subject.mesh | Transition metal nitrides | - |
dc.subject.mesh | Ultra-fast | - |
dc.title | Ultrafast hot-electron injection at HfN-metal oxide heterojunctions: Role of barrier height | - |
dc.type | Article | - |
dc.citation.title | Materials Science in Semiconductor Processing | - |
dc.citation.volume | 152 | - |
dc.identifier.bibliographicCitation | Materials Science in Semiconductor Processing, Vol.152 | - |
dc.identifier.doi | 10.1016/j.mssp.2022.107117 | - |
dc.identifier.scopusid | 2-s2.0-85138767179 | - |
dc.identifier.url | https://www.journals.elsevier.com/materials-science-in-semiconductor-processing | - |
dc.subject.keyword | HfN films | - |
dc.subject.keyword | Hot carriers | - |
dc.subject.keyword | Kelvin probe force microscopy | - |
dc.subject.keyword | Optoelectronic | - |
dc.subject.keyword | Transition metal nitrides | - |
dc.subject.keyword | Work function | - |
dc.description.isoa | false | - |
dc.subject.subarea | Materials Science (all) | - |
dc.subject.subarea | Condensed Matter Physics | - |
dc.subject.subarea | Mechanics of Materials | - |
dc.subject.subarea | Mechanical Engineering | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.