Identification of substitutional Li inn-type ZnO and its role as an acceptor

Abstract

Monocrystalline n-type zinc oxide (ZnO) samples prepared by different techniques and containing various amounts of lithium (Li) have been studied by positron annihilation spectroscopy (PAS) and secondary ion mass spectrometry. A distinct PAS signature of negatively charged Li atoms occupying a Zn-site (Li − Zn), socalled substitutional Li, is identified and thus enables a quantitative determination of the content of Li Zn. In hydrothermally grown samples with a total Li concentration of ∼2 × 10 17 cm −3 , Li Zn is found to prevail strongly, with only minor influence, by other possible configurations of Li. Also in melt grown samples doped with Li to a total concentration as high as 1.5 × 10 19 cm −3 , a considerable fraction of the Li atoms (at least 20%) is shown to reside on the Zn-site, but despite the corresponding absolute acceptor concentration of (2-3) × 10 18 cm −3 , the samples did not exhibit any detectable p-type conductivity. The presence of Li Zn is demonstrated to account for the systematic difference in positron lifetime of 10-15 ps between Li-rich and Li-lean ZnO materials as found in the literature, but further work is needed to fully elucidate the role of residual hydrogen impurities and intrinsic open volume defects.

Defect evolution and impurity migration in Na-implanted ZnO

“…3(a) and 3(b) shows that the data from wafer B are close to the ZnO-bulk point, while the data from wafer A are close to the HT-ZnO point. 20 Altogether, these results provide strong evidence for a reduction in the open volume of defects due to the presence of Na and more specifically due to the substitution of Li Zn by Na Zn at temperatures 600 • C. It can be noted that even if the Li-related positron signal is clearly stronger in the A than in the B samples, sample B 4 seems to exhibit a noticeable Li signal. This may be due to the subtle balance between Na and Li concentrations in the region probed by the positrons: For wafer A, the V Zn concentration cannot be deduced from the PAS data because of the strong contribution from Li Zn .…”

“…[31][32][33] If V Zn is the dominating trap for positrons, all data points follow the V Zn line and the position represents the actual V Zn concentration. However, this is not the case in Li-rich samples, where the W-S data converge below the V Zn line due to positron trapping at Li Zn as shown recently by Johansen et al 20 A similar conclusion can also be drawn from Fig. 3(a) identifying the region just below the ZnO-bulk point, where our data converge (S ≈ 1.005, W ≈ 0.95), as the "HT-ZnO" point.…”

“…This may be due to the subtle balance between Na and Li concentrations in the region probed by the positrons: For wafer A, the V Zn concentration cannot be deduced from the PAS data because of the strong contribution from Li Zn . 20,31 However, the data from wafers B (B 3), C (C 4), and D (D 3) can be used and the corresponding V Zn contents are 7 × 10 15 , 1.4 × 10 16 , and 7 × 10 15 cm −3 , respectively, where the influence by negatively charged impurities is neglected, as discussed previously. 34 Assuming that equilibrium conditions apply during the annealing and that the V Zn 's are stable during cooling down, the following formation energies of V Zn are obtained in these three samples: 1.44, 1.57, and 1.44 eV, respectively.…”

“…17,18 Thus, in n-type hydrothermally grown (HT) ZnO, which contains 10 17 Li/cm 3 , the predominant configuration is Li Zn , often resulting in highly resistive material. 19,20 In this study, the interaction between Li and Na has been investigated by implanting Na into (i) HT samples with a Li content of ∼4 × 10 17 cm −3 and (ii) HT samples subjected to postgrowth anneals reducing the Li content to the 10 15 cm −3 range. Especially, secondary ion mass spectrometry (SIMS) and positron annihilation spectroscopy (PAS) have been combined to reveal the role of zinc vacancies (V Zn ) in the interaction between Li and Na.…”

Nitrogen and vacancy clusters in ZnO

“…The values characterizing the layers are highlighted in the figure. Figure 2 also shows the values of the ZnO lattice, 11 the isolated Zn vacancy, 20,21 and those typical of HT-ZnO, 22,23 adjusted for the detector resolution, and geometry applied in these experiments. It is clearly seen that at high positron implantation energies, the measured data converge toward the HT-ZnO point as expected.…”