Impact Excitation And Auger Quenching Processes In Er Doped Light Emitting Si Devices
- PDF / 1,047,070 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 77 Downloads / 172 Views
Mat. Res. Soc. Symp. Proc. Vol. 486 01998 Materials Research Society
processes that are in competition with the radiative de-excitation of Er. In the second section it will be shown how, from an understanding of these processes, it is possible to fabricate Er doped Si LEDs having simultaneously high efficiency and fast modulation properties. AUGER DE-EXCITATION PROCESSES The Auger impurity de-excitation with the energy released to free carriers is one of the main nonradiative de-excitation processes which compete with the radiative emission of the impurity. This process has been studied theoretically by Langer et al. [11,12] for different impurity semiconductor systems, but it should be general and it might apply also to the Er - Si system. In particular they obtained that the reciprocal of the luminescence decay lifetime t is proportional to the free carrier concentration n according to the following formula: 1
=CAn, being
11 CA
T'Irad no
and n. =41n.
137a,
X
"tradis the radiative lifetime, nr is the refractive index, m. and m* are the electron mass and the effective mass respectively, a. is the Bohr radius and X0 is the wavelength of emission. If we apply this formula to the Er - Si system, we obtain an Auger coefficient CA of lx 10-2 cm /s. In order to experimentally determine the Auger coefficient we have measured at a fixed temperature the Er decay lifetime as a function of free carrier concentration in samples prepared according to the following procedure. Several P (B) implants have been performed in n-type (ptype) CZ Si in order to produce an almost constant P (B) concentration (ranging between 4x1016 and 1.2x1018/cm 3) at depths between 0.5 and 2.5 g1m. After implantation the samples were annealed at 1000 'C - 30 s in N 2 flux to remove the implantation damage and to activate3 the2 dopants. All of these samples have then been implanted with 4 MeV Er to a dose of 3.3xl01 /cm in order to locate the Er peak in the region where the dopant profile is flat. Finally a thermal treatment at 900 'C - 30 min in N2 flux was performed. All of these samples have been studied in photoluminescence and, as an example, in Fig. 1 we report the luminescence decay curves for the p-type samples obtained by shutting off the pump laser at t = 0 and monitoring the time evolution of the Er signal. The pump power of the laser beam was 1 mW in order to have a negligible contribution of the free carriers generated by the laser itself. In particular in Fig. 1 the time decay curves measured at 77 K in samples having different B concentrations are reported. The results show that the luminescence decay lifetime at 77 K decreases with increasing B (and hence free holes) concentration and approaches the system response at a B concentration of 1.2x10 8 /cm 3 . In order to obtain a quantitative estimate of CA we have calculated, for each dopant concentration, the density of free carriers at the measurement temperature. The results are summarized in Fig. 2 where we report the reciprocal of the luminescence decay lifetime as a function
Data Loading...
