Using our STM we have developed a method to analyze the spatially resolved electric properties of a semiconductor sample using the photocurrent as a sensitive tool. With the experimental setup we are able to measure the local photocurrent together with the topography on semiconductor samples with a resolution less than 1 nm. The method used here is especially able to measure the short circuit photocurrent in order to distinguish it from pure photoconductivity but can also record the photocurrent under applied bias. Most measurements have been taken in air, but the method is also working in an electrochemical cell under potentiostatic control. WSe₂ is used as a model semiconductor. This is a layered compound with a nonreactive van der Waals surface providing large atomically flat terraces after cleaving. Crystals of a n- and p-type doping level of 4 x10¹⁶ are used.
The measurements in ambient environment on a tungsten diselenide model semiconductor have already shown that the system consisting of the tunneling tip, the tunnel gap and the semiconductor behaves like a nanosized MIS solar cell [1, 2]. Here was also shown that the magnitude of the short circuit photocurrent does within certain limits not depend on the distance between the tip and the semiconductor sample.
Here, we show that the spatially resolved photocurrent is a quite sensitive tool for analyzing the electronic variations on a semiconductor surface. Variations in flat band potential as well as changes in the recombination rate can be detected and linked to structures of the surface, below the surface, or on metal particle modified surfaces. It is possible to visualize the space charge zones along steps as well as around catalyst particles on the surface.
Measuring the photocurrent makes it possible to directly observe effective barriers of copper particles and of size effects influencing the width of the space charge layer at the surface. An example of the influence of the size of copper particles on the electric fields on n-WSe₂2 is presented in Fig. 1 and 2. In Fig. 1 the 70 nm wide agglomerate of three copper clusters influences the short circuit photocurrent in an area of 160 nm in diameter, more than double its geometric size. In contrast in Fig. 2 the single copper cluster of 40 nm influences the photocurrent only in an area having the same size as its geometric diameter.
These results fit to a model suggested already in 1984 [3, 4], which has yet not been proven experimentally but is now finally confirmed by theses measurements.
Further experiments will be performed especially in order to study the contact properties of the device in the electrolyte environment.

REFERENCES

[1]	R. Hiesgen, D. Meissner, Electrochimica Acta 1997, 42, 2881.
[2]	R. Hiesgen, D. Meissner, Fres. J. Anal. Chem. 1997, 358, 54.
[3]	Y. Nosaka, K. Norimatsu, H. Miyama, Chem. Phys. Lett, 106, 128 (1984)
[4]	Y. Nakato, K. Ueda, H. Yano, H. Tsubomura, J. Phys. Chem. 92, 2316 (1988) and references cited therein.

Fig. 1 Topography (left side) and short circuit photocurrent image (right side) from an agglomerate of three copper clusters on n-WSe₂2 in air.

Fig. 2 Topography (left side) and short circuit photocurrent image (right side) from a single copper cluster on n-WSe₂2 in air.

[BACK] [MAIN] [FURTHER]