No job name

J. Phys. Chem. B 1999, 103, 10109-10113
Luminescence Properties of Thiol-Stabilized CdTe Nanocrystals
A. M. Kapitonov,† A. P. Stupak,† S. V. Gaponenko,*,† E. P. Petrov,‡ A. L. Rogach,§,| and
A. Eychmu
1 ller*,§
Institute of Molecular and Atomic Physics and B. I. StepanoV Institute of Physics, National Academy ofSciences of Belarus, F. Skaryna AVe. 70, 220072 Minsk, Belarus, and Institute of Physical Chemistry,UniVersity of Hamburg, Bundesstrasse 45, 20146 Hamburg, Germany Thiol-capped CdTe nanocrystals with cubic zinc blende structure are synthesized in aqueous solution. Theirsteady-state and time-resolved luminescence characteristics are studied at room and liquid nitrogen temperatures.
A strong exciton luminescence peak at 2.3 eV dominates the emission spectrum of CdTe nanocrystals atroom temperature, whereas the trap band centered at 2.0 eV undergoes substantial temperature quenching.
Luminescence excitation spectra reveal different channels leading to radiative recombination via either excitonsor traps. The mean luminescence decay time of CdTe nanocrystals at room temperature decreases from 120ns at 1.94 eV to 20 ns at 2.43 eV. Luminescence decay kinetics of CdTe nanocrystals are stronglynonexponential and are described by extremely broad lifetime distributions lying within the range from a fewhundred picoseconds to a few hundred nanoseconds.
Introduction
nanocrystals in aqueous solution. Unlike nanocrystals of CdSand CdSe whose structural and luminescence properties have Semiconductor nanocrystals are currently being extensively been thoroughly examined and reviewed,1,4,5,16,17 cadmium studied in the context of their size-dependent photophysical and telluride nanocrystals are much less studied. Although several photochemical properties. The size of the nanocrystals in all groups have performed the synthesis of CdTe nanocrystals in three dimensions is less than the de Broglie wavelength of the various environments,18-25 studies of their luminescence proper- electronic subsystem excitations. Therefore, many properties of ties are rather fragmentary and preliminary.
these materials can be systematically described and understoodin terms of quantum confinement effects.1-5 From the viewpoint Experimental Section
of basic science, a nanocrystal is in an intermediate state ofmatter between molecule-like clusters and bulk crystals, and A. Sample Preparation and Characterization. The prepara-
therefore offers a possibility to trace an evolution of electronic tion and characterization of 1-thioglycerol-capped CdTe particles and optical properties of the matter from small atomic clusters have been described in detail elsewhere (sample c in ref 24).
to bulk solids. In the context of applications, nanocrystals can Briefly, 22 mL of freshly prepared oxygen-free 0.05 M NaHTe be efficiently used in novel light-emitting devices, optical solution was added to 125 mL of a 0.013 M nitrogen-saturated transformers, photonic switches, nanoelectronic circuitry, and 4)2 6H2O aqueous solution at pH 11.2 in the presence of 0.5 mL of 1-thioglycerol as a stabilizing agent. The solution Though the basic properties of II-VI nanocrystals including was heated to 96 °C and refluxed for 8 h to promote the growth electron-hole energy states and optical transition probabilities of CdTe nanoparticles. The method of size-selective precipita- are well established and understood,12-14 there are a number of tion26 was used to separate a CdTe nanoparticle fraction with important issues relevant to primary photophysical and photo- relatively narrow (<10% as confirmed by transmission electron chemical processes that are still under investigation. Photolu- microscopy) size distribution. CdTe nanoparticles precipitated minescence properties of nanocrystals typically depend on a by 2-propanol were redissolved in water and used for spectro- variety of parameters that do not affect the absorption proper- scopic measurements. Powder X-ray diffractometry was per- ties.15 These parameters are surface structure, chemical environ- formed using a powdered CdTe sample. A predominantly cubic ment, migration of carriers, and a number of guest-host effects.
zinc blende phase was derived from the diffractograms. The An understanding of nanocrystal properties beyond simplified existence of a well-pronounced diffraction peak in the small- quantum dot considerations is necessary for establishing pho- angle region further confirms the monodispersity of the sample.
toluminescence mechanisms in real nanocrystals and the de- High-resolution transmission electron microscopy (HRTEM) velopment of highly efficient light-emitting nanocrystalline indicates good crystallinity of the CdTe particles; Fourier analysis of high-resolution micrographs of single CdTe particles In the present paper we report on steady-state and time- also revealed the cubic structure of the nanocrystallites.24 The resolved photoluminescence properties of thiol-capped CdTe mean particle size estimated from the HRTEM images and fromthe X-ray diffraction patterns of CdTe particles yields a value † Institute of Molecular and Atomic Physics; http://imaph.bas-net.by.
‡ B.I. Stepanov Institute of Physics.
B. Spectroscopic Measurements and Data Processing.
Absorption spectra were recorded at room temperature on a Permanent address: Physico-Chemical Research Institute, Belarusian State University, 220080 Minsk, Belarus.
Perkin-Elmer Lambda 14 UV-vis spectrophotometer. Lumi- 10110 J. Phys. Chem. B, Vol. 103, No. 46, 1999
nescence emission and excitation spectra and luminescencedecay kinetics were measured at room and liquid nitrogentemperatures. Steady-state luminescence emission and excitationspectra were recorded on an SFL-1211A (SOLAR, Minsk,Belarus) spectrofluorometer with a cooled photomultiplier as adetector. The typical spectral resolution for both the excitationand emission monochromators was 3-4 nm.
Time-resolved luminescence measurements were carried out on a PRA-3000 (Photochemical Research Associates, London,Ontario) spectrofluorometer in the time-correlated single photoncounting mode with a wide-band cooled photomultiplier tube(PMT) used as a detector. An electric discharge in air (fwhm ≈ 2 ns) at 20 kHz repetition rate was used as an excitationsource. Slit widths of monochromators of both excitation andemission channels were set to 16 nm. Decay kinetics wererecorded in m ) 511 time channels. Typically, 5 × 103-2 × Figure 1. Normalized luminescence and absorption spectra of CdTe
105 counts in a peak channel (CPC) of the decay curve were nanocrystals. Absorption was measured at room temperature and luminescence at both room (excitation at 480 nm ) 2.58 eV) and liquid C. Luminescence Decay Data Analysis. In the case when
nitrogen (excitation at 460 nm ) 2.69 eV) temperatures. The insetpresents the spectral dependence of the mean luminescence lifetime at the finite duration of the impulse response function of a room (solid circles) and liquid nitrogen (open circles) temperatures.
fluorometer cannot be neglected, an experimentally detectedluminescence decay kinetics d(t) is represented by a convolution The system of equations (3) approximates an integral equation of a “true” decay law D(t) with the apparatus response function of the first kind and is therefore severely ill-conditioned (see, e.g., ref 31). Therefore, a regularization method should be usedfor the recovery of a decay time distribution from luminescence tR(t - t′) D(t′) dt′ ) d(t) decay kinetics. In the present work, luminescence decay data were analyzed using a Tikhonov-regularization-based routine,32in which the regularization parameter is evaluated on the basis Two different methods of decay data analysis were used in of a combination of a partial singular-value decomposition this work. First, luminescence decays were analyzed using the computed using the NIPALS algorithm33 and the generalized model-based reconvolution using the conventional one- to three- cross-validation approach.34 Decay time distributions were exponential decay models and several nonexponential decay computed using n ) 100 exponential terms with decay times kinetics previously successfully applied to an analysis of uniformly spaced on the logarithmic scale, which provides luminescence decays of semiconductor nanocrystals (vide infra).
optimum resolution for inverting the Laplace transform.35 The second-order local polynomial approximation of the ap-paratus response function29 was used in reconvolution routines.
Results and Discussion
The quality of data fitting and the consistency of a decay modelwere judged on the basis of the visual inspection of plots of A. Steady-State Luminescence. Room-temperature absorp-
weighted residuals and their autocorrelation function, as well tion and luminescence spectra of CdTe nanocrystals and the as on the basis of the values of the reduced chi squared 2 and luminescence spectrum recorded at liquid nitrogen temperature (77 K) are presented in Figure 1. The well-pronounced absorp- Second, luminescence kinetics were analyzed using the tion at about 2.5 eV is indicative of a rather narrow size approach based on the recovery of decay time distributions (see, distribution of CdTe nanocrystals. In the case of a 10% size e.g., ref 30) with minimum a priori assumptions on the distribution inherent in the samples investigated, the variation luminescence decay law. In this case, the “true” luminescence in the energy E1S1S of the lowest excitonic transition in decay is presented as originating from an (unknown) distribution nanocrystals should lie in the range of 150-200 meV. However, the spectral width of the first absorption feature is about 500meV. Therefore, this feature should consist of a series of unresolved electronic transitions of different kinds and cannot D(t) ) ∫ f(τ) exp(-t/τ) dτ be associated with the lowest energy of the E1S1S transition alone.
This agrees with the results of photophysical and photochemical and the decay time distribution analysis reduces to the inversion spectral hole burning25,36,37 in an ensemble of II-VI crystallites, of the quasi-Laplace transform (2). In the numerical treatment where a complex structure of the first absorption feature was of decay data, the integral (2) is approximated by the exponential revealed for nanocrystals in the strong confinement regime.
The luminescence spectra of CdTe nanocrystals consist of a i)1 i exp(-t/τi) with fixed decay times τi spanning the range (τmin,τmax) and unknown preexponential factors fi. Thus, sharp high-energy peak of excitonic luminescence with a wide discretization of eq 1 with regard for eq 2 leads to a system of low-energy band of emission from trap states. The contribution from trap luminescence decreases with increasing temperature.
At 77 K, the exciton luminescence band is centered at about 2.34 eV. This band is inhomogeneously broadened due to thesize distribution of CdTe crystallites and fluctuations in their where K is an m × n-matrix combining the discretized
microenvironment. Its fwhm exceeds 100 meV and the full convolution and Laplace transform kernels, d is an m-vector of
width at the base of the band is in the range of 150-200 meV, raw decay data, and the n-vector f is the sought distribution of
which agrees well with the above-estimated variation in the lowest transition energy. The trap luminescence band spreads Luminescence Properties of CdTe Nanocrystals J. Phys. Chem. B, Vol. 103, No. 46, 1999 10111
The latter value was chosen on the basis of the low-temperaturedata, where the trap band is more pronounced. Both excitationcurves exhibit similar features, namely, a low-energy maximum.
The fact that they do not coincide means that the absorbedphoton should possess a higher energy to excite the trap emissionas compared to the exciton luminescence. There are threepossible explanations for this phenomenon.
First, smaller crystallites whose absorption edge lies at relatively higher energies due to stronger confinement couldbe responsible for virtually all emission from trap states, whichwas established for Cd3P2 nanoparticles.38 In fact, smallernanocrystals have larger surface-to-volume ratio, and hence theprobability of the appearance of trap states due to surface defectsis also higher. Alternatively, if a nanoparticle with dimensionsof several crystal lattice periods (0.648 nm for zinc blende-type CdTe39) has a few defects (on the surface or in the volume), Figure 2. Luminescence of CdTe nanocrystals recorded for a set of
then the effective radius a of this particle can decrease because of a decrease in the volume available for propagation of Bloch’swaves. Such a scenario should be valid if defects distort theperiodic lattice potential at least to the extent of one crystalunit cell. A rather small integral quantum yield of luminescenceat room temperature (∼3%) agrees with the assumption that aconsiderable part of the crystallites is defective.
Second, nanocrystals that differ in the quantum yield of intrinsic and trap emission may differ not only in size. Forexample, we could have two different kinds of particles in thesolution: one exhibiting exclusively the exciton luminescenceand the other having traps at the surface because of incompletecapping giving rise to the red-shifted luminescence. Directobservation of the single-particle luminescence should verifythe possibility of coexistence of different kinds of nanoparticles.
This work is presently in progress.
Third, as has been shown above, the inhomogeneous broad- ening due to the size distribution of nanocrystals does not exceed Figure 3. Room-temperature excitation spectra of trap and exciton
200 meV. This value is a reasonable estimate for the peak width emission in CdTe nanocrystals. The absorption spectrum is also of the excitation spectrum for exciton luminescence. Conse- quently, it can be assumed that this peak is nothing but an from the exciton peak down to ∼1.7 eV, and its maximum at envelope of the lowest E1S1S-type transitions for different 77 K is located near 1.95-2.00 eV. Such a wide emission band nanocrystals. The wider band in the excitation spectrum detected indicates a wide energy distribution in the trap states, implying within the trap emission can then be associated with an envelope a variety of defects involved in light emission. The room- of transitions to higher lying energy levels. It is possible that temperature luminescence spectrum possesses the same features, carrier trapping is more effective from some excited electron namely, the sharp exciton luminescence peak and the wide trap- or hole states with some special configurations of eigenfunctions.
related band, which is less pronounced than at low temperature.
Inhibited relaxation between neighboring electron (hole) states The evolution of the normalized luminescence spectra at 77 due to acoustic phonon quantization40 promotes the capture of K of thiol-stabilized CdTe nanoparticles with the excitation a nonequilibrium carrier. The smaller a nanoparticle is, the energy is shown in Figure 2. The excitation wavelength was higher is the probability of this process. The observation can varied within the range of 460-510 nm (2.69-2.43 eV). No also be explained by exciton self-trapping in sufficiently small changes in the luminescence spectrum were detected in the semiconductor nanoparticles;41 a critical factor is the possibility vicinity of the exciton luminescence peak, even with the of exciton localization on a particular covalent bond which then excitation within the high-energy emission tail. The low-energy leads to deformation of the capped nanocrystal. In this case, part of the luminescence spectrum (which is characteristic of the difference in the excitation spectra, namely a shift by 50- radiative recombination via defects) undergoes a dramatic 100 meV, can result from an increase in the first excited energy change under the same conditions. When the excitation ap- level with a deformation of the nanoparticle. The exciton could proaches the absorption edge, the relative contribution of possibly be localized on a bond between the crystallite core emission from trap states decreases abruptly. A small deviation and a capping thiol group.42 This assumption is supported by from this behavior occurs only in the vicinity of 2.55 eV.
the observation that degradation processes develop much faster Figure 3 shows the room-temperature excitation spectra for when nanoparticles are in liquid solutions and when they are trap emission (dashed line) and excitonic emission (solid line) along with the absorption spectrum (dotted line). The excitation B. Time-Resolved Luminescence. Luminescence decay
spectra have been normalized to provide the same short- kinetics provide additional important information on the re- wavelength intensity values. Detection energies were chosen combination of photoinduced carriers in CdTe nanocrystals. We at the maxima of the corresponding bands, i.e., 2.294 eV for have found that luminescence decays of CdTe nanocrystals are excitonic emission and 1.936 eV for emission from trap states.
wavelength-dependent, with the mean luminescence decay time 10112 J. Phys. Chem. B, Vol. 103, No. 46, 1999
Figure 4. Room-temperature (solid symbols) and liquid-nitrogen-
temperature (open symbols) luminescence decay kinetics of CdTe
nanocrystals detected at 1.94 (squares), 2.10 (circles), 2.21 (up triangles),
2.29 (down triangles), and 2.43 eV (diamonds). The excitation pulse
is shown for reference.
decreasing with increasing energy of the luminescence quantum(see inset in Figure 1 and Figure 4). Note the differenttemperature behavior of luminescence kinetics detected withinthe exciton and trap luminescence bands. All measured lumi-nescence decays were found to be strongly nonexponential: bothtwo- and three-exponential decay models failed to provide asatisfactory fit to the decay data yielding systematic plots ofweighted residuals and 2 values in the range of 1.25-1.61.
Earlier, several luminescence models implying nonexponential Figure 5. Decay time distributions recovered from the luminescence
decay kinetics were successfully used in an analysis of photo- kinetics (T ) 290 K) of the exciton band detected at 2.43 (a), 2.29 (b), luminescence decay curves of semiconductor nanocrystals. The 2.21 (c), 2.10 (d), and 1.94 eV (e). Insets show the corresponding plots micellar kinetics43 originally developed to describe luminescence of weighted residuals (WR) and their autocorrelation functions (AC),and values of the 2 and DW parameters. 100 exponential terms with quenching in organic micelles implies a discrete number of traps lifetimes uniformly distributed on the log τ axis within the range of per crystallite with a random distribution of traps over an 0.32-640.0 ns were included in the decay analysis. 511 channels, 0.64 ensemble of crystallites and is widely accepted as the relevant model for nonradiative dynamics in nanocrystals.44-46 Stretched-exponential decay laws imply migration of recombining com-ponents in a fractal space and can account for coexistence of mally determined by nonradiative deactivation of defect states local and migration lifetimes. These laws are inherent in a vast show a strong temperature dependence becoming substantially number of processes,47,48 including the photoluminescence of faster at room temperature (see inset in Figure 1 and Figure 4).
nanoparticles.49,50 However, our attempts to apply these models The thermal quenching of trap luminescence observed in the to the analysis of CdTe luminescence decays failed to provide decay kinetics agrees with a decrease in the trap luminescence both reasonably good fits to experimental decay kinetics and intensity with increasing temperature (Figure 1). The activation consistent spectral behavior of the parameters recovered.
energy of the nonradiative recombination via trap states cannot Therefore, to provide quantitative information on the character exceed a few tens of millielectronvolts because this process of the photoluminescence decays of CdTe nanocrystals, we completely dominates at kBT (T ) 290 K) = 25 meV and starts recovered decay time distributions from the luminescence decay to compete with the radiative processes only at kBT (T ) 77 K) curves. The decay time distribution analysis shows that CdTe = 6.6 meV. The recombination kinetics via the exciton channel luminescence decay kinetics involve processes with lifetimes is found to be identical at 77 and 290 K. Whether this is ranging from a few hundred picoseconds to a few hundred indicative of the direct character of the recombination process nanoseconds (Figure 5). At present, we have no reasonable or this finding is the result of competing processes involved in explanation for these extremely broad distributions of decay the deactivation scheme17 has still to be unraveled. A thorough times describing photoluminescence decay kinetics of CdTe temperature-dependent study of both the static and the time- nanocrystals. This will be the subject of further investigations.
resolved emission will help to answer this question. At least, As has been pointed out above, the luminescence kinetics of we can state that the energy gap between the lowest allowed the exciton and trap bands of CdTe nanocrystals show different and forbidden excition states16,51 does not exceed the value of temperature behaviors. Trap luminescence decay kinetics nor- Luminescence Properties of CdTe Nanocrystals J. Phys. Chem. B, Vol. 103, No. 46, 1999 10113
Conclusions
(19) Liu, L.-C.; Kim, M. J.; Risbud, S. H.; Carpenter, R. W. Philos. Mag. B 1991, 63, 769.
In summary, steady-state and time-resolved luminescence (20) Potter, B. G., Jr.; Simmons, J. H. J. Appl. Phys. 1990, 68, 1218.
studies have been reported for thiol-stabilized CdTe nanopar- Ochoa, O. R.; Colajacomo, C.; Witkowski, E. J., III; Simmons, J. H.; Potter, ticles in aqueous solution. They possess the zinc blende crystal B. G., Jr.; Solid State Commun. 1996, 98, 717.
structure, with a mean size of the nanocrystals of a ∼ 2.4 nm (21) Neto, J. A. M.; Barbosa, L. C.; Cesar, C. L.; Alves, O. L.; Galembeck, F. Appl. Phys. Lett. 1991, 59, 2715. de Oliveira, R. C. M.; de
which is substantially smaller than the exciton Bohr radius of Paula, A. M.; Plentz Fiho, F. O.; Neto, J. A. M.; Barbosa, L. C.; Alves, O.
7.3 nm. The luminescence of the CdTe nanocrystals shows L.; Menezes, E. A.; Rios, J. M. M.; Fragnito, H. L.; Brito Cruz, C. H.; both a sharp exciton band and a wide band of emission via trap Cesar, C. L. Appl. Phys. Lett. 1995, 66, 439.
(22) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. states. The trap luminescence band is substantially quenched 1993, 115, 8706.
when the temperature is increased from 77 to 290 K, which (23) Bandaranayake, R. J.; Wen, G. W.; Lin, J. Y.; Jiang, H. X.; manifests itself in both steady-state emission spectra and Sorensen, C. M. Appl. Phys. Lett. 1995, 67, 831.
luminescence decay kinetics. Both the spectrum and lumines- (24) Rogach, A. L.; Katsikas, L.; Kornowski, A.; Su, D.; Eychmu¨ller, A.; Weller, H. Ber. Bunsen-Ges. Phys. Chem. 1996, 100, 1772.
cence decays of the exciton band are similar at 77 and 290 K.
(25) Masumoto, Y.; Sonobe, K. Phys. ReV. B 1997, 56, 9734.
Luminescence excitation spectra of CdTe nanocrystals reveal (26) Chemseddine, A.; Weller, H. Ber. Bunsen-Ges. Phys. Chem. 1993,
different maxima within the first absorption feature for detection (27) Evaluation of the “size” of the extremely small nanoparticles is within the exciton and trap emission bands. Higher energies of not obvious: depending on the method used values of 1.9 nm (broadening the exciting radiation quantum may promote a more efficient of the diffraction peaks, Scherrer equation), 2.4 nm (position of the capture of the charge carriers by trap states. The luminescence diffraction maximum in the small-angle region, Bragg equation), and 2.5nm (HRTEM) were obtained. Generally, XRD underestimates and HRTEM decay kinetics of the CdTe nanocrystals depend on the emission overestimates the particle size. The nonspherical shape of CdTe nanoparticles wavelength, with the mean lifetime decreasing from ∼120 ns also leads to difficulties in the size determination.
at 1.94 eV to ∼20 ns at 2.43 eV. CdTe luminescence decays (28) O’Connor, D. V.; Phillips, D. Time-Correlated Single Photon are strongly nonexponential and are described by extremely wide Counting; Academic Press: London, 1984.
(29) Vecer, J.; Kowalczyk, A. A.; Dale, R. E. ReV. Sci. Instrum. 1993,
distributions of decay times spreading from a few hundred picoseconds to a few hundred nanoseconds.
(30) Siemiarczuk, A.; Wagner, B. D.; Ware, W. R. J. Phys. Chem. 1990,
94, 1661. Gakamsky, D. M.; Goldin, A. A.; Petrov, E. P.; Rubinov, A. N.
Acknowledgment. This work was partially supported by the
Biophys. Chem. 1992, 44, 47.
Volkswagen Foundation, Hannover. Thanks are due to Dr. M.
(31) Tikhonov, A. N.; Arsenin, V. Ya. Methods for SolVing Ill-Posed Harrison for carefully reading the manuscript and for fruitful Problems [in Russian], 3rd ed.; Nauka: Moscow, 1986.
(32) Petrov, E. P.; Kruchenok, J. V.; Rubinov, A. N. J. Fluoresc. 1999,
(33) Geladi, P.; Kowalski, B. R. Anal. Chim. Acta 1986, 185, 1.
References and Notes
(34) Golub, G. H.; Heath, M.; Wahba, G. Technometrics 1979, 21,
(1) Brus, L. E. Appl. Phys. A 1991, 53, 465.
(2) Banyai, L.; Koch, S. W. Semiconductor Quantum Dots; World (35) Ostrowsky, N.; Sornette, D.; Parker, P.; Pike, E. R. Opt. Acta 1981,
(3) Spectroscopy of Isolated and Assembled Semiconductor Nano- (36) Bawendi, M. G.; Wilson, W. L.; Rothberg, L.; Carroll, P. J.; Jedju, crystals. Special Issue of J. Lumin. 1996, 70.
T. M.; Steigerwald, M. L.; Brus, L. E. Phys. ReV. Lett. 1990, 65, 1623.
(4) Gaponenko, S. V. Optical Properties of Semiconductor Nanocrys- (37) Woggon, U.; Gaponenko, S. V.; Langbein, W.; Klingshirn, C. Phys. tals; Cambridge University Press: Cambridge, 1998.
ReV. B 1993, 47, 3684.
(5) Woggon, U. Optical Properties of Semiconductor Quantum Dots; (38) Kornowski, A.; Eichberger, R.; Giersig, M.; Weller, H.; Eychmu¨ller, A. J. Phys. Chem. 1996, 100, 12467.
(6) Klein, D. L.; Roth, R.; Lim, A. K. L.; Alivisatos, A. P.; McEuen, (39) Landolt-Bo¨rnstein, Numerical Data and Functional Relationships P. L. Nature 1997, 389, 699.
in Science and Technology; New Series, Group III, Vol. 22, Part a, (7) Feldheim, D. L.; Keating, C. D. Chem. Soc. ReV. 1998, 28, 1.
Madelung, O., Ed.; Springer-Verlag: Berlin, 1987.
(8) Colvin, V. L.; Schlamp, M. C.; Alivisatos, A. P. Nature 1994, 370,
(40) Woggon, U.; Gaponenko, S. V.; Uhrig, A.; Langbein, W.; Kling- shirn, C. AdV. Mater. Opt. Electron. 1994, 3, 141.
(9) Dabbousi, B. O.; Bawendi, M. G.; Onitsuka, O.; Rubner, M. F.
(41) Allan, G.; Delerue, C.; Lannoo, M. Phys. ReV. Lett. 1996, 76, 2961.
Appl. Phys. Lett. 1995, 66, 1316.
(42) Similar thiol-capped CdS nanoparticles were found to show certain (10) Gao, M.; Richter, B.; Kirstein, S.; Mo¨hwald, H. J. Phys. Chem. B exchange kinetics as revealed by NMR; i.e., thiol groups occasionally 1998, 102, 4096.
“leave” the semiconductor nanoparticles (unpublished).
(11) Gaponenko, S. V.; Kapitonov, A. M.; Bogomolov, V. N.; Prokofiev, (43) Gehlen, M. H.; De Schryver, F. C. Chem. ReV. 1993, 93, 199.
A. V.; Eychmu¨ller, A.; Rogach, A. L. Pis’ma Zh. EÄksp. Teor. Fiz. 1998,
(44) Gaponenko, S. V.; Germanenko, I. N.; Petrov, E. P.; Stupak, A.
68, 131 [JETP Lett. 1998, 68, 142].
P.; Bondarenko, V. P.; Dorofeev, A. M. Appl. Phys. Lett. 1994, 64, 85.
(12) Henglein, A. Chem. ReV. 1989, 89, 1861.
(45) Masumoto, Y.; Kawamura, T.; Ohzeki, T.; Urabe, S. Phys. ReV. B (13) Weller, H. AdV. Mater. 1993, 5, 88.
1992, 46, 1827.
(14) Kamat, P. V. Chem. ReV. 1993, 93, 267.
(46) Brus, L. E. J. Phys. Chem. 1994, 98, 3573.
(15) Gao, M.; Kirstein, S.; Mo¨hwald, H.; Rogach, A. L.; Kornowski, (47) Kopelman, R. J. Stat. Phys. 1986, 42, 185; Science 1988, 241, 1620.
A.; Eychmu¨ller, A.; Weller, H. J. Phys. Chem. B 1998, 102, 8360.
(48) Lianos, P. J. Chem. Phys. 1988, 89, 5237.
(16) Efros, Al. L.; Rosen, M.; Kuno, M.; Nirmal, M.; Norris, D. J.; (49) O’Neil, M.; Marohn, J.; McLendon, G. J. Phys. Chem. 1990, 94,
Bawendi, M. Phys. ReV. B 1996, 54, 4843.
(17) Eychmu¨ller, A.; Ha¨sselbarth, A.; Katsikas, L.; Weller, H. Ber. (50) Gaponenko, S. V.; Petrov, E. P.; Woggon, U.; Wind, O.; Klingshirn, Bunsen-Ges. Phys. Chem. 1991, 95, 79.
C.; Xie, Y. H.; Germanenko, I. N.; Stupak, A. P. J. Luminesc. 1996, 70,
(18) Esch, V.; Flugel, B.; Khitrova, G.; Gibbs, H. M.; Jiajin, X.; Kang, K.; Koch, S. W.; Liu, L.-C.; Risbud, S. B.; Peyghambarian, N. Phys. ReV. (51) Lefebvre, P.; Richard, T.; Mathieu, H.; Alle´gre, G. Solid State B 1990, 42, 7450.
Commun. 1996, 98, 303.

Source: http://imaph.bas-net.by/imaph/lno/papers/pdf/article4_99.pdf