Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
APPLIED PHYSICS LETTERS VOLUME 84, NUMBER 20 17 MAY 2004 Strong quantum-confinement effects in the conduction band of germanium nanocrystals C. Bostedta) Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, California 94550 and Institut für Experimentalphysik, Universität Hamburg, Germany T. van Buuren, T. M. Willey, N. Franco, and L. J. Terminello Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, California 94550 C. Heske Department of Chemistry, University of Nevada, Las Vegas, Nevada 89154 T. Möller Hamburger Synchrotronstrahlungslabor Hasylab at DESY, Hamburg, Germany 共Received 26 January 2004; accepted 24 March 2004; published online 5 May 2004兲 Quantum-confinement effects in the conduction band of deposited germanium nanocrystals are measured to be greater than in similar-sized silicon nanocrystals. The germanium particles are condensed out of the gas phase and their electronic properties are determined with x-ray absorption spectroscopy. The conduction band edge shifts range from 0.2 eV for 2.7 nm particles up to 1.1 eV for 1.2 nm particles. © 2004 American Institute of Physics. 关DOI: 10.1063/1.1751616兴 Over the last decade nanometer-sized structures have attracted tremendous interest due to their size-dependent properties.1 Germanium nanocrystals embedded in SiO2 matrixes have drawn significant attention since their blue luminescence was first reported.2 However, the source of the luminescence is not fully understood and it has been attributed to surface species3 as well as quantum confinement effects.4 Little is known about the electronic structure of germanium nanocrystals upon size reduction. Various calculations suggest strong confinement in Ge and thus a critical particle size, below which the band gap of germanium becomes larger than that of silicon.5–7 These predictions have been questioned in recent calculations, which suggest that germanium and silicon nanoparticles should exhibit a similar band gap due to structural changes in the Ge conduction band.8 To date, no experimental investigations exist which can unambiguously identify quantum-size effects at the conduction band edge of germanium nanocrystals. Moreover, the theoretical approaches to this matter are contradictory. The present experiments focus on probing the size-dependent electronic structure of germanium nanocrystals with synchrotron radiation based x-ray absorption 共XAS兲. This technique has been previously shown to be very powerful to study the conduction band of reduced-dimensional structures.9 It allows the independent investigation of the conduction band edges, not possible with optical spectroscopy techniques. Consequently it is possible to unambiguously identify the origin of changes in the electronic structure, and to answer the questions concerning the extent of quantum confinement in germanium nanocrystals and its comparison to silicon. The germanium nanoparticles are condensed out of the gas-phase and subsequently deposited in situ at the synchrotron facility.10 Their average size can be tuned from 1 to 5 a兲 Electronic mail: [email protected] nm and they exhibit a narrow size-distribution of 20% full width at half maximum 共FWHM兲 as determined by atomic force microscopy.10 As substrates, HF-etch cleaned silicon wafers are used for the XAS measurements. The produced films consist of randomly scattered, individual, and nontouching particles. Structural analysis of nanoparticles reveals the bulk-like cubic 共diamond兲 crystal structure.10 The electronic structure of the deposited nanocrystals is probed with XAS at the high resolution and high flux undulator beamline 8.0 of the Advanced Light Source at the Lawrence Berkeley National Laboratory.11 In XAS a Ge 2p core electron is excited into the empty states of the conduction band. The conduction band minimum has both, s- and p-like contributions12 and hence the excitation of a p electron into the bottom of the conduction band 共CB兲 is dipole allowed. If the conduction band minimum of the clusters shifts to higher energies by ⌬E CB due to confinement effects in the nanoclusters, the absorption threshold will also shift by ⌬E CB to higher energies. The experimental resolution of the absorption measurements at the Ge L edge 共around 1219 eV兲 is approximately 0.3 eV,11 small enough to permit observation of small CB shifts in our clusters. It should be noted that this resolution describes the experimental, i.e., gaussian broadening of the spectra. Peak positions can be fit and compared with higher accuracy. In Fig. 1 the XAS data of a Ge bulk crystal reference and three nanocrystal samples with decreasing average sizes are shown. The absorption edge is blueshifted with decreasing particle size, consistent with the predictions of the quantum-confinement model. The nanocrystal edges are slightly broadened with respect to the bulk and their density-of-states features are washed out. This effect was observed in earlier experiments on Si nanocrystals, where it had been attributed to the particle size distribution.9 Each individual cluster size contributes with its distinct quantum shift to the overall absorption edge, resulting in an edge broadening. None of the reported fingerprints for oxide 0003-6951/2004/84(20)/4056/3/$22.00 4056 © 2004 American Institute of Physics Downloaded 24 Sep 2006 to 131.169.252.49. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp Appl. Phys. Lett., Vol. 84, No. 20, 17 May 2004 Bostedt et al. 4057 ment effects in Ge are increasingly stronger than in Si. The difference between the Ge and Si conduction band edge shifts grows from 0.1 eV for 2.7 nm particles up to 0.7 eV for 1.2 nm particles. In a previous optical study on chemically prepared germanium and silicon nanocrystals, stronger quantum confinement for specific optical transitions from the upper split-off conduction band to the valence band at ⌫ in germanium with respect to silicon nanocrystals were observed.4 However, no conclusions could be drawn about the extent of quantum confinement at the band edges. In other studies about the optical properties of germanium nanocrystals embedded in a SiO2 matrix ambiguous results were obtained.3,14 It was later concluded that the optical properties of these systems were dominated by interface and defect states.7 In the present FIG. 1. X-ray absorption spectra of a Ge bulk crystal and three nanocrystal samples. The XAS edges of the nanoparticles are blueshifted with respect to study, relying on synchrotron radiation based spectroscopy the bulk edge. The arrows indicate the inflection point of each absorption techniques, the conduction band edges are investigated indeedge. pendently and the whole particle electronic structure is probed. Thus, we were able to conclusively show that gercontamination or Ge–Si alloying, appearing as distinct feamanium exhibits stronger quantum confinement at the contures between 3 and 6 eV above the absorption threshold,13 duction band edge than silicon. are evident in the data. Therefore, it can be concluded that From a theoretical point of view there is an ongoing and the nanocrystals neither exhibit significant particle— vibrant discussion about quantum confinement in germanium substrate interactions nor do they display oxide contaminananocrystals.5– 8 Two independent empirical tight binding tions. calculations,6,7 which have been able to describe silicon sysIn Fig. 2 the data points from the XAS experiments for tems well, are in good agreement with the obtained experithe Ge conduction band shifts are summarized. For comparimental results on germanium. They both predict stronger son previously published silicon data points9 are added to the quantum confinement effects in Ge compared to Si. Howgraph. In case of Ge the conduction band shifts range from ever, these calculations are tailored to describe the optical 0.2⫾0.1 eV for 2.7⫾0.3 nm particles up to 1.1⫾0.2 eV for properties of nanocrystals involving valence excitons. There1.2⫾0.4 nm particles. The error bars in x indicate the fore, they are not able to predict the behavior of the conducFWHM of the particle size distributions. The larger error tion band edge for decreasing particle sizes. In contrast to bars for the 1.2, 1.6, and 1.8 nm XAS data are due to a wider these tight binding calculations, another theoretical approach FWHM of the size distribution for these samples which were utilizing the empirical pseudopotential method predicts that condensed in an Ar buffer.10 The error bars in y indicate the the band gap of germanium and silicon should be similar uncertainty in determining the edge shift. The Si conduction upon size reduction.8 The predicted similarities in the elecband shifts range from 0.1⫾0.1 eV for 2.8⫾0.3 nm particles tronic structure of Si and Ge are explained with sizeup to 0.4⫾0.1 eV for 1.1⫾0.2 nm particles. The observed dependent structural changes in the conduction band of gershift of the Ge conduction band edge is larger than the shift manium: the conduction band minimum is found to move of the Si conduction band edge over the complete range of from the L point to the X point for reduced particle sizes and investigated particle sizes. Additionally the quantum confinethus the germanium conduction band minimum becomes silicon-like for small sizes.8 In our experiments we always find stronger quantum confinement effects for germanium compared to silicon upon size reduction 共Fig. 2兲 what is in contrast to these EPM calculations. The conduction band minimum of germanium shifts from 0.2 eV for 2.7 nm particles up to 1.1 eV for 1.2 nm particles 共Fig. 2兲, compared to 0.1 up to 0.4 eV for similar particle sizes in the case of silicon.9 However, the experimental data allows only statements about the absolute conduction band shift and none about the possible existence of the L-to-X crossover in Ge. In this context it should also be mentioned that both the cubic and tetragonal phase have been suggested for Ge nanoparticles.15 The creation of either phase has been explained with the employed synthesis routes.10 The presently investigated samples have been shown with electron diffraction to be in the cubic phase for particles above 3 nm and for smaller sizes no indications for phase transitions, monitored FIG. 2. Measured conduction band edge shifts for Ge nanocrystals 共solid by Ge 3d core level photoemission, have been found. squares兲 and its comparison to similarly sized Si particles 共see Ref. 9兲 共open circles兲. The data show much stronger quantum effects in Ge compared On a final note an estimate for the overall band gap to Si. behavior shall be done. It has been reported experimentally9 Downloaded 24 Sep 2006 to 131.169.252.49. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp 4058 Bostedt et al. Appl. Phys. Lett., Vol. 84, No. 20, 17 May 2004 and theoretically16 that the band gap in semiconductor nanocrystals opens with a fixed ratio of the valence to the conduction band edge shift ⌬E VBM :⌬E CBM . This ratio was determined experimentally for Si nanocrystals to Si Si :⌬E CBM ⬇2 and for Ge theoretically to be ⌬E VBM Ge Ge ⬇1. The overall nanocrystal band gap can be ⌬E VBM :⌬E CBM extrapolated from this ratio and the well-known bulk band NC bulk gap according to E gap ⫽E gap ⫹⌬E CBM⫹R * ⌬E CBM9 with R⫽2 for Si and R⫽1 for Ge. Using the quantum shift data from Fig. 2 for 1.2 nm particles a nanoparticle gap of 2.4 eV results for Si and a gap of 3.0 eV results for Ge. This estimate shows that the band gap of Ge may become larger than that of Si upon size reduction what is expected to have a significant impact on the importance of Ge as a material for reduced-dimensional structures. In conclusion strong quantum confinement effects in the conduction band of deposited germanium nanocrystals have been observed. With respect to the bulk value the conduction band edge of germanium has been determined to shift from 0.2 eV for 2.7 nm particles up to 1.1 eV for 1.2 nm particles. In the range of investigated particle sizes the quantum size effects of Ge nanocrystals are always stronger compared to Si particles of similar size. These results reveal that germanium nanocrystals exhibit a high tunability of their electronic properties making them a promising material for potential optoelectronic applications. This work was performed under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under Contract No. W-7405-Eng-48. The Advanced Light Source is supported by the U.S. DOE under Contract No. DE-AC0376SF00098, LBNL. C.B. acknowledges partial financial support from the Deutscher Akademischer Austauschdienst DAAD in the HSP-III program under Contract No. D/97/ 18881. A. P. Alivisatos, J. Phys. Chem. 100, 13226 共1996兲. M. Zacharias and P. M. Fauchet, Appl. Phys. Lett. 71, 380 共1997兲. 3 K. S. Min, K. V. Shcheglov, C. M. Yang, H. A. Atwater, M. L. Brongersma, and A. Polman, Appl. Phys. Lett. 68, 2511 共1996兲. 4 J. P. Wilcoxon, P. P. Provencio, and G. A. Samara, Phys. Rev. B 64, 035417 共2001兲. 5 T. Takagahara and K. Takeda, Phys. Rev. B 46, 15578 共1992兲. 6 N. A. Hill, S. Pokrant, and A. J. Hill, J. Phys. Chem. B 103, 3156 共1999兲. 7 Y. M. Niquet, G. Allan, C. Delerue, and M. Lanoo, Appl. Phys. Lett. 77, 1182 共2000兲. 8 F. A. Reboredo and A. Zunger, Phys. Rev. B 62, 2275 共2000兲. 9 T. van Buuren, L. N. Dinh, L. L. Chase, W. J. Siekhaus, and L. J. Terminello, Phys. Rev. Lett. 80, 3803 共1998兲. 10 C. Bostedt, T. van Buuren, J. M. Plitzko, T. Möller, and L. J. Terminello, J. Phys.: Condens. Matter 15, 1017 共2003兲. 11 J. J. Jia, T. A. Calcott, J. Yurkas, A. W. Ellis, F. J. Himpsel, M. G. Samant, J. Stohr, D. L. Ederer, J. A. Carlisle, E. A. Hudson, L. J. Terminello, D.K. Shuh, and R.C. C. Perera, Rev. Sci. Instrum. 66, 1394 共1995兲. 12 D. A. Papaconstantopoulos, Phys. Rev. B 27, 2569 共1983兲. 13 P. Castrucci, R. Gunnella, M. De Crescenzi, M. Sacchi, G. Dufour, and F. Rochet, J. Vac. Sci. Technol. B 16, 1616 共1998兲. 14 S. Takeoka, M. Fujii, S. Hayashi, and K. Yamamoto, Phys. Rev. B 58, 7921 共1998兲. 15 S. Sato, S. Nozaki, H. Morisaki, and M. Iwase, Appl. Phys. Lett. 66, 3176 共1995兲. 16 F. A. Reboredo, A. Franceschetti, and A. Zunger, Phys. Rev. B 61, 13073 共2000兲. 1 2 Downloaded 24 Sep 2006 to 131.169.252.49. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp