NANO LETTERS
Highly Luminescent CdSe/ZnSe Core/Shell Nanocrystals of Low Size Dispersion
2002 Vol. 2, No. 7 781-784
Peter Reiss,* Joe1 l Bleuse, and Adam Pron CEA Grenoble, De´ partement de Recherche Fondamentale sur la Matie` re Condense´ e, 17 rue des Martyrs, 38054 Grenoble CEDEX 9, France Received April 25, 2002; Revised Manuscript Received May 17, 2002
ABSTRACT A simple synthetic route for the preparation of high-quality CdSe/ZnSe core/shell nanocrystals without the use of any pyrophoric organometallic precursors is presented. Effective surface passivation of monodisperse CdSe nanocrystals is achieved by overcoating them with a ZnSe shell, applying zinc stearate as a zinc source. The resulting core/shell nanocrystals exhibit high room temperature photoluminescence efficiencies (60−85%) in organic solvents as well as in water after functionalization with mercaptoundecanoic acid (MUA).
Cadmium chalcogenide nanocrystals exhibit an efficient photoluminescence, whose energy can be controlled by adjusting their size.1-4 In addition, the surface of these nanocrystals can be functionalized relatively easily to enable not only their grafting to other molecules but also their colloidal dispersion in various solvents. Among numerous possible applications of CdSe nanocrystals, biological labeling can be pointed out, as the CdSe nanocrystals exhibit much larger photostability than the organic dyes used routinely and they facilitate multicolor experiments.5-8 For this application, both the efficiency and the line width of the photoluminescence are of crucial importance, with the line width depending strongly on the size dispersion of the prepared nanocrystals. The luminescence efficiency can, in turn, be significantly improved if the nanocrystals are passivated on their surface by a shell of a larger band gap semiconductor. Among these “core/shell” systems, CdSe/ ZnS and CdSe/CdS are the most intensively studied,9-11 whereas almost no research has been carried out on the CdSe/ ZnSe core/shell system, except for one paper that reports a very low fluorescence quantum yield (below 0.4%). The authors explain this low value by structural defects in the ZnSe overlayer, as well as at the CdSe/ZnSe interface.12 ZnSe should in principle be a very good candidate as a shell material. First, its band gap is wider (2.72 eV) than that of CdSe (1.76 eV), and the band alignment is of type I, i.e., both holes and electrons are confined in CdSe.13 Second, its lattice parameter mismatch relative to the CdSe core (6.3%), although being larger than that of CdS (3.9%), is significantly lower than that of the most commonly used shell * Corresponding author. E-mail:
[email protected] 10.1021/nl025596y CCC: $22.00 Published on Web 06/06/2002
© 2002 American Chemical Society
material ZnS (10.6%). Third, keeping the same anion at the interface leads to a larger band offset in the conduction band,14 therefore to a better compromise in the relative confinements of the (light) electrons and (heavy) holes. We show here that these arguments apply by proposing a facile synthetic approach to high-quality CdSe/ZnSe core/shell nanocrystals that exhibit luminescence quantum yields in the range of 60-85%. The methods developed initially for the preparation of CdSe core nanocrystals involve the use of highly pyrophoric substrates such as dialkyl cadmium.15 More recently, safer reagents were proposed, such as cadmium oxide,16 cadmium carbonate, or cadmium acetate.17 On the other hand, the influence of the reaction medium has been investigated and it has been shown that the replacement of pure trioctylphosphine oxide (TOPO), used in the original organometallic method,15 by a mixture of TOPO and hexadecylamine (HDA) leads to nanocrystals with a much lower size dispersion.18 Combining these approaches, we synthesized CdSe core nanocrystals in a mixed TOPO/HDA solvent with a molar ratio of 60-80% HDA and using CdO, complexed by dodecylphosphonic acid, as cadmium precursor.19,20 As demonstrated in Figure 1 the mixed solvent gives rise to nanocrystals with constant emission line widths (and size dispersions), whereas in the case of pure TOPO solvent a nearly linear increase is observed with reaction time. In contrast to a similar method recently proposed by Qu et al. where stearic acid is applied as complexing agent for CdO,21 the use of dodecylphosphonic acid facilitates the preparation of nanocrystals of a smaller size range (2.5 to 6 nm diameter). Their narrow size distribution allows their self-assembly in
Figure 1. Evolution of the photoluminescence spectra during the synthesis of CdSe nanocrystals at 250 °C in (a) TOPO solvent and (b) TOPO/HDA (molar ratio 1:4) mixed solvent.
Figure 2. TEM images of a three-dimensional superlattice of CdSe nanocrystals (diameter 3.6 nm), prepared by solvent evaporation of a colloidal solution in toluene.
three-dimensional superlattices, as demonstrated in Figure 2. The formation of such “colloidal crystals” requires a size dispersion not exceeding 5%.4,22,23 This is achieved here for as-grown nanocrystals without any fractionation or size sorting. Contrary to the core preparation, the shell growth method that was initially developed9-11 has remained until now nearly unchanged and still involves highly pyrophoric dilalkyl cadmium or zinc precursors. With the goal of avoiding the use of these compounds, we transposed the above core preparation method in our attempts to grow a ZnSe shell on the CdSe nanocrystals. Consequently, ZnO was complexed with dodecylphosphonic acid and slowly injected together with TOPSe into a mixture of HDA/TOPO containing CdSe core nanocrystals. One disadvantage of this method is that high temperatures approaching 360 °C are needed for the complexation of ZnO with dodecylphosphonic acid. Furthermore, the resulting complex shows such a low reactivity toward TOPSe such that temperatures significantly above 200 °C were needed during the shell growth. These conditions caused restart of the growth of the CdSe nanocrystals via Ostwald ripening, which deteriorates their size distribution. In subsequent attempts, zinc stearate turned out to be an excellent zinc source for the preparation of the ZnSe shell. This easy to handle, commercially available compound lowers the shell growth temperature down to 190-200 °C.24 In addition, the complexation with a phosphonic acid is not required anymore. As shown in Figure 3a, the growth of the ZnSe shell causes a small bathochromic shift of a few nanometers in the UV782
Figure 3. (a) Evolution of the absorption spectra of CdSe/ZnSe nanocrystals during the shell growth (core diameter 3.6 nm). (b) Evolution of the photoluminescence spectra and comparison to rhodamine 6G.
visible absorption spectrum. Similar shifts, originating from the partial leakage of the exciton into the shell, have previously been reported for CdSe/CdS and CdSe/ZnS systems.9-11 The evolution of the fluorescence quantum yield of the CdSe/ZnSe core/shell nanocrystals is depicted in Figure 3b. Typically, it first rises progressively because of the increasing passivation of the core surface, then decreases, presumably as a result of the higher concentration of structural defects created within the thicker deposited shell. Annealing of the freshly prepared shell, in the absence of the shell forming reagents, leads to a significant improvement in the fluorescence quantum yield, which can now reach 85% (Figure 3b).25 At the same time, the line width of the photoluminescence peak remains essentially unchanged (28 nm at fwhm) during the shell growth. In Figure 4 TEM micrographs of the core/shell system are presented. It is clear that the growth of the shell layers resulted in an increase of the average size of the nanocrystals as well as in a broadening of their dispersion. No superlattice is formed in this case (compare Figures 2 and 4). Finally, we considered ligand exchange with mercaptocarboxylic acids. These are widely used to enable the dispersion of nanocrystals in polar media and/or to link them to other molecules via amidation reactions.6,26,27 The TOPO, TOPSe, and HDA ligands at the surface of the above core/ shell nanocrystals were replaced by mercaptoundecanoic acid (MUA).28 The presence of the functional groups originating from MUA could be clearly seen in the IR spectra taken after the ligand exchange. In particular, two strong bands characteristic of the carboxylic group at 3440 cm-1 (OH stretchings) and at 1730 cm-1 (CdO stretchings) dominated the spectrum of the MUA functionalized nanocrystals. A band of weak intensity at 1180 cm-1 assigned to the P ) 0 group of TOPO was, however, still visible and indicated that the ligand exchange was not complete. The nanocrystals were then dispersed in water after deprotonation of the carboxylic groups by a base. As demonstrated in Figure 5a, for both CdSe core and CdSe/ZnSe core/shell nanocrystals the absorption spectra were not affected by the ligand exchange, with the exception of a slight bathochromic shift associated with the solvent change. On the contrary, the photolumiNano Lett., Vol. 2, No. 7, 2002
property is of crucial importance for the nanocrystals’ functionalization preceding their dispersion in polar media or their grafting to other molecules. Acknowledgment. We thank Jany Thibault for valuable assistance with the transmission electron microscopy. References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Figure 4. HRTEM images of a CdSe nanocrystal (a) before and (b) after growth of the ZnSe shell (the length bars indicate 5 nm). (c) TEM image of CdSe/ZnSe core/shell nanocrystals. The poor contrast in these micrographs is due to the background signal from the amorphous carbon substrate.
(11) (12) (13)
(14) (15) (16) (17) (18) (19)
(20)
Figure 5. Absorption (a) and photoluminescence (b) spectra of CdSe core and CdSe/ZnSe core/shell nanocrystals before (dashed lines, toluene solutions) and after (solid lines, aqueous solutions) functionalization with MUA. All colloidal solutions exhibit identical optical densities at the first exciton absorption peak.
nescence efficiency of the core nanocrystals was dramatically reduced in aqueous media because of a decreased surface passivation (Figure 5b). This was not the case for the core/ shell nanocrystals whose luminescence spectrum remained unchanged with respect to its peak position, shape and intensity, indicating an efficient passivation of the CdSe core by the ZnSe overlayer. To summarize, we have developed a new, simple, and very reproducible method for the preparation of CdSe/ZnSe core/ shell nanocrystals. This system, previously believed to be very inefficient, exhibits fluorescence quantum yields exceeding 80%. Moreover, the luminescence properties are much more stable and less sensitive to surface ligand exchange as compared to “bare” core nanocrystals. The latter Nano Lett., Vol. 2, No. 7, 2002
(21) (22) (23) (24)
Weller H. Angew. Chem., Int. Ed. Engl. 1993, 32, 41-53. Alivisatos, A. P. J. Phys. Chem. 1996, 100, 13226-13239. Heath, J. R., Ed. Acc. Chem. Res. 1999, 32. Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Annu. ReV. Mater. Sci. 2000, 30, 545-610. Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science 1998, 281, 2013-2016. Chan, W. C. W.; Nie, S. M. Science 1998, 281, 2016-2018. Dahan, M.; Laurence, T.; Pinaud, F.; Chemla, D. S.; Alivisatos, A. P.; Sauer, M.; Weiss, S. Opt. Lett. 2001, 26, 825-827. Pathak, S.; Choi, S.-K.; Arnheim, N.; Thompson, M. E. J. Am. Chem. Soc. 2001, 123, 4103-4014. Hines, M. A.; Guyot-Sionnest, P. J. Phys. Chem. 1996, 100, 468471. Dabboussi, B. O.; Rodriguez-Viejo, J.; Mikulec, F. V.; Heine, J. R.; Mattoussi, H.; Ober, R.; Jensen, K. F.; Bawendi, M. G. J. Phys. Chem. B 1997, 101, 9463-9475. Peng, X.; Schlamp, M. C.; Kadavanich, A. V.; Alivisatos, A. P. J. Am. Chem. Soc. 1997, 119, 7019-7029. Danek, M.; Jensen, K. F.; Murray, C. B.; Bawendi, M. G. Chem. Mater. 1996, 8, 173-180. Gue´naud, C.; Deleporte, E.; Filoramo, A.; Lelong, P.; Delalande, C.; Morhain, C.; Tournie´, E.; Faurie, J.-P. J. Appl. Phys. 2000, 87, 1863-1868. Wei, S.-H.; Zhang, S. B.; Zunger, A. J. Appl. Phys. 2000, 87, 13041311. Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993, 115, 8706-8715. Peng, Z. A.; Peng, X. J. Am. Chem. Soc. 2001, 123, 183-184. Qu, L.; Peng, Z. A.; Peng, X. Nano Lett. 2001, 1, 333-337. Talapin, D. V.; Rogach, A. L.; Kornowski, A.; Haase, M.; Weller, H. Nano Lett. 2001, 1, 207-211. Dodecylphosphonic acid was synthesised following a literature method: Sasse K. in Methoden der organischen Chemie; Mu¨ller, E., Ed.; G. Thieme Verlag: Stuttgart, 1963; vol. 12/1, p 435, 352353. Trioctylphosphine oxide (TOPO), trioctylphosphine (TOP), and hexadecylamine (HDA) were of the highest purity available and additionally purified by distillation. A typical synthesis of CdSe nanocrystals: 51.4 mg (0.4 mmol) of CdO is placed into a flask containing 1.15 mL of TOPO and 2.85 mL of HDA. The mixture is heated to ca. 270 °C under argon flow, then 230 µL (0.8 mmol) of dodecylphosphonic acid is added. After stabilizing the temperature of the resulting colorless solution at 250 °C, 2.5 mL of a 0.2 M solution of Se powder in TOP is quickly injected. The nanocrystal size depends on the growth time. Qu, L.; Peng, X. J. Am. Chem. Soc. 2002, 124, 2049-2055. Murray, C. B.; Sun S.; Doyle H.; Betley, T. MRS Bull. 2001, 26, 985-991. Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Science 1995, 270, 1335-1338. A colloidal solution of ca. 20 mg of CdSe nanocrystals with an average diameter of 3.6 nm in 4 mL of heptane is placed in a threeneck flask under purified argon flow. After addition of 2.5 mL of TOPO and 1.5 mL of HDA, the mixture is heated to 190 °C and then kept at this temperature with the goal of a complete heptane evaporation. Zinc stearate (316 mg) is dissolved in 2.5 mL of toluene upon gentle heating (ca. 60 °C). After cooling to room temperature, the resulting 0.2 M solution is mixed with 2.5 mL of a 0.2 M solution of Se in TOP. By means of a syringe pump this mixture is injected within 1 h into the reaction flask containing the core nanocrystals at 190-200 °C. Periodically small aliquots are removed in order to monitor the shell growth. After the addition is completed the crystals are annealed at 190 °C for an additional 1-1.5 h. 783
(25) For the determination of room temperature fluorescence quantum yields, the spectrally integrated emission of a nanocrystal dispersion in toluene was compared to the emission of an ethanol solution of rhodamine 6G of identical optical density (
