Answers are grounded in this paper and cite their sources.
Strong red low-temperature photoluminescence results from Li diffusion of as-grown GaP crystals prepared from the vapor by the wet H₂ method, or from Ga solution with O doping. The luminescence spectrum contains sharp strong no-phonon lines and many well-resolved phonon replicas. Four no-phonon lines can be seen at zero field. The spacings between these lines are approximately consistent with a model in which the luminescence arises from the decay of an exciton bound by {~} 0.24 eV to an axial (C₃ᵥ) center. The J-J and crystal-field splittings are 1.1 and 3.4 meV. The symmetry axis is shown to be $〈111〉$ from detailed magneto-optical studies of the no-phonon lines, and the electron and hole g values are determined from an analysis for strong crystal field, gₑ=1.76±0.14, gₕ=1.04±0.05. Use of this form of analysis is suggested by the degree of mixing of the J-J-split states by the crystal field, indicated by the relative oscillator strengths of the no-phonon lines. The chemical identity of components of this axial center has been determined from isotope experiments. The substitution O¹⁶{→}O¹⁸ increases the no-phonon line energies by 0.8 meV, and shifts some of the many local-mode lines resolved in the phonon wing of the luminescence spectrum. Only changes in local-mode energies occur for the substitution Li⁷{→}Li⁶, but these changes are generally larger than for O. The form of the true local-mode replicas for crystals containing roughly equal amounts of Li⁶ and Li⁷ proves that the center contains at least two inequivalent Li atoms. The simplest model for the center consistent with all the detailed experimental evidence is LiI-LiGa-OP (I=interstitial). This complex is isoelectronic with the Ga-P atom pair it replaces, can produce efficient low-temperature bound-exciton luminescence with decay time consistent with experiment ({~} 200 nsec for decay allowed by electric dipole selection rules), and has no free spin in the final state. The temperature quenching rate of the Li-Li-O luminescence is large compared with the familiar red Zn-O luminescence because of the high degree of compensation produced by Li diffusion, and the Li-Li-O photo- and electroluminescence efficiencies are negligible near 300 ^∘{}K. The associate VGa-OP (V=vacancy) is necessary for the formation of the Li-Li-O centers. Evidence is presented that a significant proportion of the substitutional O exists in these associates before Li diffusion for GaP crystals grown or annealed below 1100 ^∘{}C. Apparently, the VGa-OP associate is not stable significantly above 1100 ^∘{}C, while association is essentially quenched at T700 ^∘{}C. The behavior of VGa in GaP can be studied very conveniently from the optical properties of the Li-Li-O associate. In addition, the efficiency of O-doping techniques for GaP crystals grown by different methods can be assessed easily and with high accuracy and sensitivity by purely optical measurements, using appropriate annealing techniques together with O¹⁸ doping.
No takes yet. Share an insight, caveat, or question.
P. J. Dean (1971) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: