Imaging Oxide-Covered Doped Silicon Structures Using Low-Energy Electron Microscopy
- PDF / 301,707 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 90 Downloads / 169 Views
1026-C15-03
Imaging Oxide-Covered Doped Silicon Structures Using Low-Energy Electron Microscopy M.L. Anderson, C.Y. Nakakura, K.F. Saiz, and G.L. Kellogg Sandia National Laboratories, Albuquerque, NM, 87123 ABSTRACT Low energy electron microscopy (LEEM) operated at incident electron energies just above the “mirror” mode is used to image Si-based test devices. Significant p- vs. n- doping contrast is observed, even when the structures are covered with a ~3.5 nm thermal oxide. The contrast arises from a difference in surface potential between the two regions and is related to both p-n work function differences and electron-beam-induced charging of the oxide. The results show that the LEEM is capable of characterizing pn junctions without complicated sample preparation and that it is a promising technique for rapid, non-destructive imaging of microelectronic devices. INTRODUCTION With the continuing decrease in microelectronic feature sizes, and resulting increase in complexity of IC’s, there is a strong incentive to develop device imaging techniques that offer both high spatial resolution and high sample throughput. The low energy electron microscope (LEEM), with a spatial resolution of 7-8 nm, a parallel imaging scheme and multiple contrast mechanisms is potentially such an instrument. Although the related technique of photoemission electron microscopy (PEEM) has been used successfully to image doping variations in various device structures [1-6], comparatively little work has been reported to assess the capabilities of the LEEM. Here, we report a LEEM investigation to image pn test structures buried under a ~3.5 nm thermal oxide. We find surprisingly good contrast between doping type when imaging with electron energies of just a few volts above the mirror mode (electron reflection). We also observe time-dependent changes in the image intensity indicating significant electron beam charging of the oxide during the measurements. The charge on the oxide persists for periods of hours to days and has a marked effect on the optimum electron energy used for imaging.
EXPERIMENT The experiments detailed in this paper were performed using a commercial LEEM [7]. In this type of microscope, high-energy electrons (20 kV) from an electron gun (a LaB6 source in our system) are focused on the back focal plane of an objective lens producing a parallel beam at the sample surface. Low electron energies at the sample are achieved by biasing it at a negative potential near to that of the source. The difference between the source and sample potential is called the start voltage, Vs. Depending on the value of Vs (-2 to +3 V in this study), electrons are either reflected (mirror electron microscopy, MEM) or scattered (LEEM) from the surface. The image formed by the objective lens using these electrons are re-accelerated to 20 kV, transferred by a magnetic prism, magnified by a series of lenses and projected on to a viewing screen.
Figure 1 shows a plot of how the reflected or scattered electron intensity varies as a function of Vs at lo
Data Loading...
