SIMS profiles for each of the different junction specimens examined here.
(a)–(c) show scanning electron microscopy images of a specimen prepared for examination using conventional FIB milling. (d) shows a TEM image of a sidewall of a TEM membrane prepared using 30 kV Ga ions. Care has been taken to limit redeposition during milling. The image reveals the presence of an amorphous surface layer and damaged crystalline regions.
(a) shows a reconstructed phase image of 520-nm-thick specimen containing a symmetrically doped junction. The top surface of the specimen is in the top-right direction and the differently doped regions of the specimen are indicated. The contrast arising from the presence of the dopants can be clearly seen. (b) shows a reconstructed amplitude image of the specimen; no contrast difference can be seen from the differently doped regions. (c) shows suggesting that the surface of the specimen is an iso-potential. (d) shows a thickness map of the specimen suggesting that the specimen is parallel-sided. (e) shows “unflattened” phase profiles extracted perpendicular to the junction acquired different distances, , into the specimen [indicated in (a)]. (f) shows a profile extracted from across the thickness map indicated in (d) revealing that the specimen is indeed flat.
The phase measured across (a) a 400-nm-thick specimen containing a junction and (b) a 600-nm-thick specimen containing a junction each with a symmetrical dopant concentration of as a function of the electron beam intensity (spot size). The phase profiles were extracted perpendicular to the junctions and have not been flattened to remove any artifacts from thickness variations across the specimens and the gradients in the phase profiles are thought to arise from changes in the electrostatic potential in the specimens. The poor signal-to-noise ratio in (b) compared to (a) is due to the relatively large thickness of the specimen.
The step in phase measured across the six different or junction specimens with dopant concentrations of as a function of crystalline specimen thickness measured using CBED. The predicted values for specimens with bulklike properties are indicated by the dashed line.
Measured phase and simulated phase for a specimen with bulk properties for -thick junction specimens with different dopant concentrations.
(a) shows step in phase measured across differently doped junctions as a function of crystalline specimen thickness measured using CBED. The closed symbols show data acquired from junctions in the middle of the specimen and open symbols show data acquired from the edges. (b) shows the theoretical and experimentally determined values of as a function of the dopant concentration in the junctions measured using SIMS. (c) shows the electrically inactive thickness plotted as a function of the dopant concentration in the junctions measured using SIMS.
Phase images and cosine of each phase image for the -thick specimens containing the differently doped junctions.
Phase measured across junctions with dopant concentrations of (a) and (b) using different intensity electron beams.
Table showing and measured for the membranes in the mid (electrically isolated) positions for the differently doped junctions. The predicted values of indicate calculations for specimens with bulk properties.
Article metrics loading...
Full text loading...