Chemical Physics Letters 394 (2004) 400–404 www.elsevier.com/locate/cplett
Luminescence quenching in PbS nanoparticles Steven W. Buckner *, Robert L. Konold, Paul A. Jelliss
*
Department of Chemistry, Saint Louis University, St. Louis, MO 63103, USA Received 23 April 2004; in final form 19 May 2004 Available online 31 July 2004
Abstract PbS and ZnS- and Pb(OH)2-capped PbS nanoparticles produced in inverted micelles luminesce extremely weakly. Removal of water present during synthesis results in a significant increase in the luminescence quantum yield of the nanoparticles. Water addition to the dried nanoparticles induces efficient luminescence quenching for both uncapped and capped particles. H2O and D2O quench the luminescence of the dried particles with similar efficiencies. We propose a quenching model where water penetrates into the nanoparticle crystal lattice. Ó 2004 Elsevier B.V. All rights reserved.
1. Introduction Much of the work in nanoscale materials has centered on the interesting and useful size-dependent optical properties of these structures [1–3]. Lead salt nanoparticles [4–10] are particularly interesting because their large Bohr exciton radii and small bulk band gaps result in strong quantum confined behavior. Applications of PbS nanoparticles have been limited, however, by their generally weak to undetectable photoluminescence [8,9]. There have been a few recent studies reporting luminescence from PbS nanoparticles [11–17]. In each of the cases the nanoparticle surface was passivated using a different method and there are some significant variations in the emission spectra reported in these studies. It is generally believed that the presence of surface trap states in PbS nanoparticles results in very efficient non-radiative decay. It is also believed that these traps sites must be eliminated by surface passivation in order to allow for observable luminescence. In this letter we report on the quenching of PbS nanoparticle luminescence by water. Luminescence from both capped and uncapped PbS is efficiently quenched by the presence *
Corresponding authors. Fax: +1 314 977 2521. E-mail address:
[email protected] (S.W. Buckner).
0009-2614/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2004.06.138
of water. By removal of the water that is present during synthesis of the nanoparticles, luminescent nanoparticles are produced. We present a model for the nanoparticle luminescence quenching process by water.
2. Experimental For all experiments, spectrophotometric grade solvents were used as supplied except hexane, which was dried over molecular sieves. AOT (AOT = sodium dioctyl sulfosuccinate), NaOH, Na2S and inorganic salts were used as supplied by Fisher Scientific. Pb(NO3)2 and Zn(NO3)2 were used as supplied by Aldrich chemical company. Nanopure water was used to prepare all salt solutions and for the quenching studies. All the nanoparticles in this study were synthesized using the reverse-micelle method [18,19]. 72 ll of a soluble salt solution was added to 5 ml of a 0.20 M AOT in heptane solution and the resulting solution was sonicated for 5 min and then stored in the dark for 1 h to allow formation of the reverse micelles (w0 = 4). The concentrations of the soluble salt solutions that were used are listed in Table 1. After formation of the Pb2+ micellar solution and the S2 micellar solution, the two micellar solutions were mixed and allowed to react for 1 h. For the preparation of the
S.W. Buckner et al. / Chemical Physics Letters 394 (2004) 400–404
401
Table 1 Ion concentrations used for synthesis of quantum dots Nanoparticlesa
[Pb2+] (mol/liter)
[S2 ] (mol/liter)
[cap] (mol/liter)
Bare PbS ZnS-capped PbSb Pb(OH)2-capped PbS
0.20 0.16 0.32
0.20 0.32 0.16
0 0.16 0.32
a b
All volumes of salt solutions used to create the reverse micelles were the same to ensure a constant w0 for all micelle solutions. The Zn2+ solutions periodically showed traces of hydrolysis products.
capped systems, the PbS nanoparticles were first allowed to form. These nascent nanoparticle solutions contained either excess Pb2+ (to allow formation of the Pb(OH)2 coating) or excess S2 (to allow formation of the ZnS coating). After the PbS nanoparticles were allowed to form, a micellar solution that contained the capping reagent (NaOH for the Pb(OH)2 cap and Zn(NO3)2 for the ZnS cap) was mixed with the nascent nanoparticle solution. TEM images of Pb(OH)2-capped nanoparticles show spherical particles with a mean nanoparticle diameter of 3.4 nm. The nanoparticles formed periodic arrays with a hexagonal packing pattern on the TEM grids suggesting the particles were uniform in size. The Pb(OH)2 and ZnS overlayers are 1.05 and 1.1 monolayers thick, respectively. The freshly prepared nanoparticles were characterized with UV–Vis and luminescence emission spectroscopy. The solution was then heated in a heated water bath (T = 85–90 °C) under vacuum (P < 0.1 Torr) to remove all of the heptane and water. The nanoparticles were allowed to dry under heat and vacuum for 0.5 hours after the solvents were removed. A portion of the remaining AOT–PbS nanoparticle matrix was redissolved in dry hexane. After drying and re-dissolution the nanoparticleÕs spectra were obtained. All solutions were analyzed by UV–Vis spectroscopy and luminescence emission and excitation spectroscopy. UV–Vis spectra were obtained on a Shimadzu 2530 UV– Vis absorption spectrometer. Luminescence spectra were obtained on a Photon Technologies QM 4-2003 fluorescence spectrophotometer. Luminescence measurements were performed on samples with their absorbencies adjusted to 0.10 at the excitation wavelength. Quantum yields were performed by measuring integrated emission intensities relative to the integrated emission intensity of a solution of 9-cyanoanthracene in hexane (UL = 1.0). Luminescence emission spectra were obtained out to 1000 nm and corrected for detector response. Appropriate cut-off filters were placed before the emission monochromator to eliminate Raman scattering and harmonics. All luminescence spectra were corrected for instrument response and residual harmonic peaks. 3. Results Fig. 1a presents luminescence emission spectra for PbS (uncapped) before drying and after drying along
with a luminescence excitation spectrum for dried PbS (uncapped). Fig. 1b presents a UV–Vis absorption spectrum for the nanoparticles immediately after drying. The emission maximum for the PbS nanoparticles is near 800 nm. This emission maximum is a lower limit due to the lower detector sensitivity out in the near infrared. The luminescence emission maximum does not shift with excitation from 310 out to 500 nm. The emission spectra we observe here are in good agreement with those obtained by Chen and coworkers for alkanethiolatecapped PbS [12] and by Fernee et al. [16] (inorganically capped PbS nanoparticles in aqueous colloidal solution). There are significant differences with the spectra
Fig. 1. (a) Luminescence emission spectra (kexc = 330 nm) of PbS nanoparticles (uncapped) before drying (right, lower curve) and after drying(right, upper curve), and luminescence excitation (kem = 800 nm) spectrum of PbS nanoparticles (uncapped) immediately after drying (left curve). (b) UV–Vis absorption spectrum of PbS nanoparticles (uncapped) immediately after drying.
402
S.W. Buckner et al. / Chemical Physics Letters 394 (2004) 400–404
that were reported by Yang et al. [13] (PbS nanoparticles in silica glass) and Lifshitz et al. [14] (PbS nanoparticles in polyacrylamide films). Some of these differences may be due to the environment of the PbS quantum dots. The luminescence intensity of our PbS nanoparticles increases by much more than a factor of 10 after drying. The luminescence intensity of the PbS nanoparticles in the AOT/hexane (dry) solution continues to increase over the period of 1 week, which may be due to slow oxidation of surface trap sites resulting in surface passivation. PbS nanoparticles capped with >1 monolayer of ZnS or Pb(OH)2 also show a strong luminescence increase upon drying under vacuum. Like the bare nanoparticles, removal of water gives a significant increase in luminescence intensity. The quantum yield for Pb(OH)2-capped PbS nanoparticles is 3.5% after drying. The quantum yield for emission from the capped and dried PbS is significantly higher than the uncapped, dried PbS. The enhanced luminescence upon capping with ZnS or Pb(OH)2 is presumably due to passivation of surface trap states. Fig. 2 presents luminescence emission spectra for the ZnS-capped PbS after drying along with the luminescence emission spectrum for bare, dried PbS for comparison. Further demonstration of the water quenching effect is given in Fig. 3a. Here, a series of spectra are presented for samples where increasing amounts of water have been added to the ZnS-capped PbS after drying and re-dissolution in dry hexane. Water addition efficiently quenches the luminescence intensity. A Stern–Volmer plot for bare PbS is presented in Fig. 3b. For the bare PbS a quenching constant of 140 M 1 is obtained. This compares with a reported value of 26.7 M 1 obtained for bare CdS nanoparticles [20]. For comparison, the initial reaction solutions in which the PbS nanoparticles were prepared contained 0.8 M water. Addition of water quenches the luminescence of the dried and capped nanoparticles, but with lower efficiency. The Stern–Volmer plots for the capped systems show some curvature.
Relative Intensity
1.00 0.80 0.60 0.40 0.20 0.00 550
650
750 Wavelength (nm)
850
Fig. 2. Comparison of luminescence emission spectra (kexc = 330 nm) of PbS nanoparticles (lower curve) and ZnS-capped PbS nanoparticles (upper curve). Both samples were dried and re-dissolved with dry hexane.
Modified Stern–Volmer plots for the ZnS- and Pb(OH)2-capped systems are presented in Figs. 3c,d. Here, the quenching data are plotted as I0/(I0 I) vs. 1/ [H2O]. From these plots, quenching constants of 91.7 and 47.0 M 1 are derived for the ZnS- and Pb(OH)2capped nanoparticles, respectively.
4. Discussion John and coworkers [20] were the first to observe reversible water quenching of nanoparticle luminescence. They observed luminescence quenching of CdS nanoparticles formed in reverse micelles. Bunker and coworkers have recently produced CdS nanoparticles with extremely high quantum yields upon thermal treatment [21,22]. The heating process results in removal of water. We observe water quenching of Cd(OH)2-capped CdS nanoparticles. In our original observation, Cd(OH)2-capped CdS particles were precipitated from solution, centrifuged, dried and re-dissolved in a 0.2 M heptane–AOT solution (no water present). After this process, we observed that the luminescence intensity of the particles had increased substantially. We observe a similar effect when the Cd(OH)2-capped CdS nanoparticles are treated with the same drying process used here for PbS. In each of these cases, addition of water to the dried nanoparticles results in significant luminescence intensity quenching. We have also found that HgS nanoparticles luminesce more intensely after drying [23]. The origin of the water luminescence quenching is unclear. H2O luminescence quenching of lanthanide ion luminescence is well known [24,25]. Quenching is due to energy transfer from the excited state to high-energy O–H vibrations [25]. The lower O–D stretching frequency in D2O makes it a much less effective lanthanide ion luminescence quencher. This process does not appear to operate here. The Pb(OH)2-capped particles that have been dried are found to luminesce more strongly than do the bare PbS particles. Thus, the presence of an O–H vibration is clearly not quenching the luminescence. Further evidence against the vibrational-energy transfer process is provided by D2O quenching studies. We observe D2O to quench the PbS nanoparticle luminescence with 96% of the efficiency of H2O. The small difference observed is comparable to that expected based on the relative differences in the diffusion coefficients for D2O and H2O in hexane [D(D2O)/D(H2O) = 0.96] [26]. Ostwald ripening was first suggested as the mechanism of water luminescence quenching in CdS nanoparticles [27]. A later study disproved this mechanism [20]. Our results also indicate Ostwald ripening is not the source of water quenching. First, the quenching process is reversible and increased luminescence can be re-attained by removing the water. Second, the quenching occurs
S.W. Buckner et al. / Chemical Physics Letters 394 (2004) 400–404
403
Fig. 3. (a) Luminescence emission spectra (kexc = 330 nm) for samples with increasing amounts of water added to the dried PbS/AOT/hexane solutions. Water concentrations are 0 , 0.0056, 0.0167, 0.0389 and 0.111 M. Water addition quenches the luminescence intensity. (b) Stern–Volmer plots for H2O quenching of bare PbS nanoparticle luminescence. (c) Modified Stern–Volmer plots for H2O quenching of ZnS-capped PbS nanoparticle luminescence. Data plotted as I0/(I0 I) vs. 1/[Q]. (d) Modified Stern-Volmer plots for H2O quenching of Pb(OH)2-capped PbS nanoparticle luminescence. Data plotted as Io/(Io – I) vs. 1/[Q].
immediately after water addition. There is no time for the ripening process to occur. Another potential mechanism for water quenching is interference with the AOT capping of the nanoparticle surface. This was suggested as a potential source for the water quenching of CdS luminescence [20]. For the bare clusters, the surfactant AOT provides surface passivation of the nanoparticles. Addition of water may disrupt the surfactant–nanoparticle interaction and open up surface trap states that favor nonradiative decay. Results from the capped PbS suggest this is not the sole mechanism, however. The ZnS- and Pb(OH)2-capped nanoparticles have significantly higher luminescence intensities than the bare PbS nanoparticles which indicates their surface trap states have been effectively passivated. But their luminescence is quenched by water, suggesting that their luminescence quenching is not due solely to AOT removal from the surface trap states. Ultrafine PbS particles contained in a reverse micellar system have been observed to act as a photocatalyst for the reduction of water [28]. The PbS nanoparticles here could be acting as a photocatalyst for water reduction to H2. This would result in a decrease in the luminescence
efficiency, as energy from the excited state is diverted toward water reduction. However, the capped systems also show a significant luminescence quenching effect. Capping the surface of the nanoparticles with Pb(OH)2 or ZnS should prevent the nanoparticles from photocatalyzing water reduction. An alternative mechanism is that the water may be able to partially penetrate into the nanoparticle crystal lattice and create strain or defects in the crystal. Many sulfides are hygroscopic or crystallize as hydrates. Absorption of water may be particularly important for nanocrystals, where the majority of the atoms are at or near the surface. Penetration of water into the nanocrystal structure can create strain in the crystal, which acts as a trap site for the exciton. The trap state can then deactivate the excited state by non-radiative decay. The Stern–Volmer constants for the ZnS- and Pb(OH)2capped nanoparticles are lower than for the bare PbS nanoparticles. This reduced efficiency may be due to the inorganic capping agent serving to reduce the water penetration into the PbS core of the nanocrystals. All prior studies of PbS nanoparticle luminescence have reported systems with strong capping agents or
404
S.W. Buckner et al. / Chemical Physics Letters 394 (2004) 400–404
environments where water was excluded. Further, efficient metal sulfide nanoparticle luminescence quenching by water may be a general phenomenon. This suggests that protection of metal sulfide nanocrystals from water can significantly enhance luminescence quantum yields. We are currently investigating the nature of this quenching process and other small molecules that may induce similar effects.
Acknowledgements We gratefully acknowledge helpful discussions with Professor Bruce Kowert (Saint Louis University) and Dr. Chris Bunker (AFRL, Wright-Patterson AFB, Dayton, OH).
References [1] [2] [3] [4] [5]
S. Empedocles, M. Bawendi, Acc. Chem. Res. 32 (1999) 389. M. Nirmal, L. Brus, Acc. Chem. Res. 32 (1999) 407. A.P. Alivisatos, J. Phys. Chem. 100 (1996) 13226. I. Kang, F.W. Wise, J. Opt. Soc. Am. B 14 (1997) 1632. R. Rossetti, R. Hull, J.M. Gibson, L.E. Brus, J. Chem. Phys. 83 (1985) 1406. [6] F. Wise, Acc. Chem. Res. 33 (2000) 773. [7] P. Jiang, Z.-F. Liu, S.-M. Cai, Langmuir 18 (2002) 4495. [8] A.A. Patel, F. Wu, J.Z. Zhang, C.L. Torres-Martinez, R.K. Mehra, Y. Yang, S.H. Risbud, J. Phys. Chem. B 104 (2000) 11598.
[9] Y.-P. Sun, R. Guduru, F. Lin, T. Whiteside, Ind. Eng. Chem. Res. 39 (2000) 4663. [10] R.D. Schaller, M.A. Petruska, V.I. Klimov, J. Phys. Chem. 107 (2003) 13765. [11] S. Gallardo, M. Gutierrez, A. Henglein, E. Janata, Ber. BunsenGes. Phys. Chem. 93 (1989) 1080. [12] S. Chen, L.A. Truax, J.M. Sommers, Chem. Mater. 12 (2000) 3864. [13] P. Yang, C.F. Song, M.K. Lu, X. Yin, G.J. Zhou, D. Xu, D.R. Yuan, Chem. Phys. Lett. 345 (2001) 429. [14] E. Lifshitz, M. Sirota, H. Porteanu, J. Cryst. Growth 196 (1999) 126. [15] M. Sirota, E. Minkin, E. Lifshitz, V. Hensel, M. Lahav, J. Phys. Chem. B 105 (2001) 6792. [16] M.J. Fernee, A. Watt, J. Warner, S. Cooper, N. Heckenberg, H. Rubinsztein-Dunlop, Nanotechnology 14 (2003) 991. [17] M.J. Fernee, J. Warner, A. Watt, S. Cooper, N. Heckenberg, H. Rubinsztein-Dunlop, Nanotechnology 15 (2004) 16. [18] P. Llianos, Thomas, Chem. Phys. Lett. 125 (1986) 299. [19] B.S. Zou, R.B. Little, J.P. Wang, M.A. El-Sayed, Int. J. Quantum Chem. 72 (1999) 439. [20] M. Tata, S. Banerjee, V.T. John, Y. Waguespack, G.L. McPherson, Colloids Surf. A 127 (1997) 39. [21] B.A. Harruff, C.E. Bunker, Langmuir 19 (2003) 893. [22] C.E. Bunker, Air Force Research Laboratory, WPAFB, OH, Personal communication, 2003. [23] S.W. Buckner, R.L. Konold, Unpublished results, 2003. [24] R.E. Ballard, Spectrochim. Acta A 24 (1968) 65. [25] Y. Haas, G. Stein, Chem. Phys. Lett. 15 (1972) 12. [26] K. Rah, S. Kwak, B.-Y. Eu, M. Lafleur, J. Phys. Chem. A 106 (2002) 11841. [27] B.H. Robinson, T.F. Towey, S. Zourab, A.J.W.B. Visser, A. Van Hoek, Colloids Surf. 61 (1991) 175. [28] T. Hirai, Y. Tsubaki, H. Sato, I. Komasawa, J. Chem. Eng. Japan 28 (1995) 468.
