| 1. |
|
|
| 2. |
- Strokan, N.B., et al.
(författare)
-
Measurement of micrometer diffusion lengths by nuclear spectrometry
- 2005
-
Ingår i: Semiconductors (Woodbury, N.Y.). - : Pleiades Publishing Ltd. - 1063-7826 .- 1090-6479. ; 39:12, s. 1394-1398
-
Tidskriftsartikel (refereegranskat)abstract
- A method for determination of diffusion lengths in the range 0.5-50 µm, which corresponds to carrier lifetimes in the nanosecond range, is suggested A calibrated nonequilibrium charge is injected into the base of the reverse-biased diode structure. The injection is provided by alpha particles generated by natural decay in the single-particle counting mode. The nuclear spectrometry technique is used to measure the amount of charge that diffused across the base to the boundary of the electric-field region. The loss of charge during the diffusion is calculated as a function of the depth of alpha particle penetration beyond the electric-field region. The derived power-law functions make it possible to relate the diffusion length with the exponent and numerical factor that describes the loss of charge. The experiment is performed with lightly doped 4H-SiC epitaxial films. © 2005 Pleiades Publishing, Inc.
|
|
| 3. |
- Strokan, N.B., et al.
(författare)
-
The limiting energy resolution of SiC detectors in ion spectrometry
- 2005
-
Ingår i: Semiconductors (Woodbury, N.Y.). - : Pleiades Publishing Ltd. - 1063-7826 .- 1090-6479. ; 39:12, s. 1420-1425
-
Tidskriftsartikel (refereegranskat)abstract
- The Monte Carlo method is used to simulate the complete stopping of a particles in SiC. A histogram of energy losses in nuclear-scattering events is obtained. The energy-loss spectrum has the characteristic asymmetric shape with the line full width at the half-maximum FWHMnucl ˜ 4. 22 keV. The final shape of the spectral line is obtained by a convolution with the Gaussian function that describes the contribution of the ionization and noise fluctuations (originated in the detector and instrumentation) to the signal. The resulting value of FWHM for the line is equal to 8.75 keV (at a noise variance of 1.7 keV). The experimental energy resolution of the detectors was found to be poorer than the calculated value by a factor of 2. It is established that the losses of charge during its transport in the detector bulk are insignificant, so that the discrepancy between the calculated and experimental values of the resolution should be attributed to the nonoptimal design of the detector window. © 2005 Pleiades Publishing, Inc.
|
|
