Laser spot scanning in photoelectrochemical systems, relation between spot size and spatial resolution of the photocurrent

Abstract

Laser spot scanning studies of single-crystallinep-InSe in contact with a neutral aqueous solution reveal a dramatic difference in lateral resolution between the material "as cleaved" and after platinization by brief dipping in a dilute H,PtCl, solution. A model is developed to explain these observations, and the resolution is calculated as a function of the diffusion coefficient and the life time of minority carriers, and of the charge transfer rate. The improvement of the resolution is found to be due to the increase of the rate of hydrogen evolution at the illuminated semiconductor by Pt catalyst. The model also provides numerical values for the charge transfer rate in the noncatalyzed and the catalyzed cases. INTRODUCTlONDuring the last decade a new technique to study the photoelectrochemical (PEC) properties of semiconductor surfaces has been developed, utilizing a spot of highly focused laser light scanned across the surface.'-' This technique, known as laser spot scanning (LSS), has opened new possibilities for laterally resolved studies of the effects of illumination at semiconductor-liquid junctions.In this technique, a laser spot is scanned over the semiconductor surface and the photocurrent is recorded. Photocurrent images can be directly related to surface structures of polycrystalline materials, such as grain boundaries and surface defects, as shown for TiO, (Refs. 1 and 5) and for nSi.3 Studying the effect of surface treatments is greatly facilitated, as different treatments can be made on the same material and then directly examined with the LSS apparatus. The correlation of surface treatment or structural features of the surface with PEC properties has been studied with the LSS technique, for n = GaAs, n-WSe,,4 and for p-InSe.' PEC etching of pits on CdSe with focused laser light of high intensity might have applications in data storage.* We took a novel approach to the laser scanning microscopy by introducing fiber optics.7 We let the laser beam focus onto the end of a single mode glass fiber. The other end is fitted with a SelfocO lens that focuses the outcoming laser light to a small spot. This approach gives a simpler apparatus for the x/y scanning of the focused light spot across the semiconductor surface. Another benefit of this approach is evident in photoelectrochemical studies. The apparent light source (the end of the fiber) can be moved very close to the semiconductor surface and thus avoiding blurring of the image or damping of the light as the light passes through the electrolyte.The resolution of LSS is limited by the size of the focused spot of laser light; values down to a few microns are reported. IV3 However, we found that the spatial resolution was much lower than the spot size.In this paper we develop a model to simulate the spatial resolution as a function of properties of the semiconductor "'Authors to whom correspondence should be addressed. material (mobility and life time of minority carriers) and of the chemical system (rate of charge transfer from the semiconductor to t...

Photoinduced electrochemical deposition of Cu onp-type Si substrates

“…First, the electron consumption due to the plating process must be taken into account. Eriksson et al 16 have shown that this effect can significantly decrease the deposited spot diameter. Second, band bending effects ͑schematically shown in Fig.…”

“…The extent of the effect depends on the resistivity of Si and the minority carrier lifetime. 16 However, the comparison between n-and p-type Si is only qualitative in their opposite signs of OCP upon illumination since specific doping concentration affects the quantitative values. Upon removal of the substrate from the solution, Cu deposition in the region illuminated by the laser spot is observed on the p-Si, while no Cu is visible by the naked eye on the n-Si.…”

“…16 D is defined as k B T/e, where is the charge-carrier mobility, k B is the Boltzmann's constant, T is the temperature, and e is the electron charge. 16 Since for p-Si with the resistivity range used in this work is around 450 cm 2 V Ϫ1 s Ϫ1 , 19 TϷ300°K and is in the millisecond range, say 0.4 ms, 16 D and l are thus estimated to be 11.6 cm 2 s Ϫ1 and 680 m, respectively. The observed dot diameters are much smaller than the diffusion length thus determined; consequently, other contributions tend to limit electron diffusion.…”

“…19 Therefore, the penetration depth can be characterized by 1/␣. 16 For the red͑green͒ laser light used here, the photon energy E p is 1.88 eV ͑2.33 eV͒ yielding an absorption coefficient of 3000 cm Ϫ1 (11 000 cm Ϫ1 ). 19,20 The resulting penetration depth is then estimated to be 3.3 m for the red laser beam and 0.9 m for the green laser beam.…”

“…15͒ structures of down to 250 nm on p-Si, the sizes of the deposited features were reported to be much larger than the laser beam diameter. Light-induced charge-carrier diffusion, 16 localized thermal profile, 7,17 and laser beam intensity profile 16 are the main possible factors limiting the miniaturization of these structures. The structure sizes obtained by far field laser-induced Cu deposition were reported to be as small as 1 m with dilute HF present in the solution on n-GaAs substrates but they were not as small on p-Si or p-GaAs using a relative motion between the substrate and the laser beam.…”