Response of Semiconductor Nanocrystals to Extremely Energetic

Feb 1, 2013 - First, we investigate a drop-cast CdSe/ZnS NC film using 3.1 eV photoexcitation to characterize spectral and dynamical signatures of the...
0 downloads 0 Views 829KB Size
Letter pubs.acs.org/NanoLett

Response of Semiconductor Nanocrystals to Extremely Energetic Excitation Lazaro A. Padilha,† Wan K. Bae,† Victor I. Klimov,† Jeffrey M. Pietryga,*,† and Richard D. Schaller*,‡,§ †

Chemistry Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois 60439, United States § Department of Chemistry, Northwestern University, Evanston, Illinois 60208, United States ‡

S Supporting Information *

ABSTRACT: Using a combination of transient photoluminescence and transient cathodoluminescence (trCL) we, for the first time, identify and quantify the distribution of electronic excitations in colloidal semiconductor nanocrystals (NCs) under high-energy excitation. Specifically, we compare the temporally and spectrally resolved radiative recombination produced following excitation with 3.1 eV, subpicosecond photon pulses, or with ionizing radiation in the form of 20 keV picosecond electron pulses. Using this approach, we derive excitation branching ratios produced in the scenario of energetic excitation of NCs typical of X-ray, neutron, or gamma-ray detectors. Resultant trCL spectra and dynamics for CdSe NCs indicate that all observable emission can be attributed to recombination between states within the quantum-confined nanostructure with particularly significant yields of trions and multiexcitons produced by carrier multiplication. Our observations offer direct insight into the transduction of atomic excitation into quantum-confined states within NCs, explain that the root cause of poor performance in previous scintillation studies arises from efficient nonradiative Auger recombination, and suggest routes for improved detector materials. KEYWORDS: Nanocrystal, quantum dot, multiexciton, charged nanocrystal, radiation detection, cathodoluminescence

Q

remain unclear, as several issues may be responsible. The few studies of NC-based radiation detectors available all examined CdSe/ZnS core/shell NC-based composite materials in a scintillator modality, which raises questions including whether the relatively low volume loadings led to reduced radiation stopping power, or whether the relatively small emission Stokes shift (25 to 80 meV)19 relative to the ensemble-broadened emission band (typically full-width-at-half-maximum of 80−120 meV) in CdSe/ZnS NCs caused substantial photon reabsorption issues. However, the influences of these factors are largely calculable and lend themselves to obvious redress (e.g., use higher loadings and/or engineered NCs with larger Stokes shifts3,20−23). We instead suggest that advances in this area await a better understanding of the underlying and, to date, impenetrable physics of high energy (keV-MeV) relaxation in these complex materials. Motivated by their potential for optoelectronic application, over the past two decades researchers worldwide have shed substantial light on the behavior of quantum-confined colloidal NCs under optical (infrared through ultraviolet) excitation.1 The result has been a fairly refined fundamental understanding

uantum-confined, colloidally synthesized semiconductor nanocrystals (NCs) present size-tunable energy gaps and amenability to solution processing.1 These qualities offer potential utility in a wide range of applications including lightemitting diodes, optical amplifiers, solar cells, and photodetectors.2−5 NCs can also potentially impact radiation detector technologies, such as energy-resolving gamma-ray scintillators and diodes, X-ray imaging arrays, and solid-state neutron detectors.6−9 In fact, high-fluorescence quantum yields,10,11 low optical scatter due to nanometer-scale particle diameters, and demonstrated means for large-scale production of materials with narrow size distributions12 make NCs and NC-based composites extremely attractive as potential replacement materials for energy-resolving gamma-ray detector applications that currently rely on large bulk-phase single crystals. Moreover, because energy resolution in such devices directly depends on the number of excitations produced per unit of incident energy,13 NC phenomena such as carrier multiplication (CM)14,15 that can increase the efficiency of photovoltaic devices16 can have a much more striking impact on radiation detectors.17 Despite this apparent promise, to date semiconductor NC gamma-ray detection efforts have yielded only proof-ofprinciple performance with unremarkable sensitivity and energy resolution.6,8,18 The intrinsic reasons for poor performance © 2013 American Chemical Society

Received: October 5, 2012 Published: February 1, 2013 925

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932

Nano Letters

Letter

Figure 1. Cathodoluminescence apparatus and signatures of typical electronic excitations in semiconductor nanocrystals. (a) The transient cathodoluminescence apparatus converts femtosecond laser pulses at 4.66 eV into photoelectrons via excitation of a gold cathode. Electrons are accelerated to 20 keV using an extraction pinhole and focused into the sample using a three-element Einzel lens and grounded exit pinhole. trCL signal is collimated with an off-axis collection lens and steered to a single-photon-sensitive streak camera. (b) A single exciton (X), consisting of one electron (filled circle) and one hole (empty circle) is highly emissive with a radiative lifetime of ∼20 ns in CdSe at room temperature. (c) A biexciton (2X) radiates more rapidly than a single exciton but annihilates on a subnanosecond time scale into a single X due to nonradiative Auger recombination. (d) An exciton plus an additional charge or “trion” (X*) also radiates more rapidly than a single X, but undergoes Auger recombination at a slower rate than the 2X leaving a nonemissive singly charged NC at long time.

incapable of providing the type of insight generated by the common ultrafast time-resolved spectroscopic techniques used to such tremendous impact by those in fields of, for example, colloidal NC photovoltaics and LEDs.6,8,18 Thus, our conception of high-energy processes within NCs severely lags that of lower-energy optical processes, and so both fundamental and technological advances have been equally limited. Here, we present ultrafast trCL studies as a means to probe for the first time high-energy relaxation in the well-known colloidal CdSe/ ZnS core/shell NC material system. We find that energetic excitation of such materials produces high yields of weakly emissive multiexcitons and trions, suggesting significant roles of processes such as CM, and scattering-type or Auger-assisted ionization in energy relaxation. Further, we conclude that these processes result in an excited-state population that presents a heretofore under-appreciated and in fact dominant channel for loss of absorbed energy in previously reported NC-based gamma-ray scintillation experiments. Finally, we conclude that the performance of future NC-based detectors may be improved not by use of materials with large Stokes shift (as is the prevailing view), but rather through use of a specific, emerging class of engineered nanomaterials exhibiting suppressed Auger recombination. As described above, the interaction of high-energy radiation, such as gamma-rays or X-rays, with essentially any material results in a great deal of energy released in the form of fast electrons with a spectrum of energies that are fairly well-known. The unexamined steps in relaxation down to semiconductor band states within a NC excited with such high energy comes after this initial electron shower. Thus, we can simulate such absorption events with a pulse of energetic electrons. The key

of the dependence of excited state populations and dynamics on excitation energy and intensity for an ever increasing number of NC materials, particularly when the excitation energy is within ∼1 order of magnitude of the principal material band gap. Within this region, several important phenomena unique to colloidal NCs and of tremendous importance to optoelectronic application have been characterized, including efficient CM14,15 and photoionization.24,25 However, carrier behavior within this ultraviolet-to-visible optical energy regime can still generally be understood in terms of the established NC band structure. Left entirely unexamined in these materials is the question of what happens at still higher energies, well in excess of bulk or quantum-confined semiconductor states, when excitations become atomic in nature (e.g., 1−100 keV). The initial interaction of, for example, X-rays or gamma-rays with matter is a well-established, atom-level process that yields a predictable cascade of energetic electrons and secondary X-rays (that either escape or subsequently produce additional energetic electrons).26,27 What remains unknown in colloidal NC materials, however, is the link between these two excitation regimes, that is, how a collection of highly excited atoms within a NC ultimately relaxes to produce a collection of excited states within the quantumconfined electronic structure. Given the uncertain scaling of important processes such as CM and photoionization to such high energies, it is not at all clear that the distribution of states produced by such relaxation can be straightforwardly extrapolated from studies utilizing optical excitation. Previous studies of colloidal NC response to energetic excitation have largely been limited to static radioluminescence (X-ray excited luminescence) measurements, which are 926

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932

Nano Letters

Letter

Figure 2. Pump-intensity-dependent transient photoluminescence of a CdSe/ZnS NC film. Photoexcitation at 3.1 eV with a controlled fluence yields photon-counting streak camera images with the indicated average number of excitons per NC. (a) For ⟨N⟩ = 0.02 (estimated from the measured laser intensity and absorption cross-section listed in Table 1), photoexcited NCs contain only slowly emitting single Xs that evolve little over the displayed time window. (b) At ⟨N⟩ = 0.49, some NCs contain 2Xs, which emit at roughly the same energy as single excitons, but undergo Auger recombination on a subnanosecond time scale. (c) At the highest intensity of ⟨N⟩ = 2.93, in addition to the single and biexciton feature some NCs contain a 3X or higher-order multiexciton that emits at higher energy and will also undergo faster Auger recombination. (d) Normalized spectral slices produced by binning the first 0 to 50 ps of streak camera data as a function of ⟨N⟩ show the single exciton spectrum at ⟨N⟩ = 0.03 and highlight the development of blue-shifted PL associated with 3X and higher order excitons. (e) PL decay dynamics, derived from the streak camera data at 1.94 eV for indicated values of ⟨N⟩ and normalized at long pump-delay time, highlight the development of a rapidly decaying biexciton population in NCs, which contrasts the far slower-decaying single exciton dynamics observed for the lowest value of ⟨N⟩. The maximum PL amplitude near 0 ps divided by the long-delay amplitude near 1200 ps, shown in the inset, demonstrates that this ratio follows the calculated Poisson distribution (magenta line) for 2X populations. (f) A composite of extracted decay dynamics (symbols) shows the decay profiles of 2X and 3X with exponential-fitted Auger recombination lifetimes (solid fit lines) in addition to the slowly evolving single X dynamics (solid black line data).

under certain excitation conditions. Briefly, absorption of photons with energy at or modestly above the band gap of a NC produces charge-neutral electron−hole pairs according to Poisson statistics with the average NC occupancy defined by the product of the NC absorption cross-section and the incident photon intensity.32 Single electron−hole pairs (excitons) in CdSe NCs efficiently radiate with a ca. 20 ns lifetime at room temperature (Figure 1b).33 Biexcitons radiate roughly four times faster than single excitons,24,34 but they emit with low efficiency owing to subnanosecond annihilation via nonradiative Auger recombination (Figure 1c).32 At slightly higher photon energies (i.e., > 3× the energy gap), single photons can directly produce biexcitons through CM14,15 or, less frequently, a long-lived charged NC that when re-excited yields trions that also undergo Auger recombination (Figure 1d).24,25,35 First, we investigate a drop-cast CdSe/ZnS NC film using 3.1 eV photoexcitation to characterize spectral and dynamical signatures of the resultant excitations. Figure 2a−c shows three streak camera images recorded using progressively higher excitation intensities (and correspondingly larger number of photons absorbed per NC on average, ⟨N⟩) from a sample emitting with a PL maximum of 1.94 eV (630 nm). The lowest fluence (1.6 μJ/cm2, ⟨N⟩ = 0.02, Figure 2a) produces a single emission feature that does not evolve appreciably in either amplitude or spectral profile over the displayed 1.3 ns

advantage of this approach is that laser sources can produce short electron pulses with controllable repetition rate, fluence, and temporal synchronization, allowing for time-resolved collection of resultant emission. While time-integrated CL studies6,28 lack sensitivity to excitations with negligible quantum yield (such as multiexcitons and trions), transient detection permits identification and quantification of these short-lived states commonly observed in NC materials. To construct a suitable trCL apparatus, we coupled electron pulse generation techniques developed for transient electron diffraction and cathodoluminescence microscopy with ultrafast streak-camera detection (Figure 1a).29−31 The sample is mounted on a two-dimensional scanning stage within a vacuum chamber featuring large quartz windows on three sides. The apparatus allows for easy sample translation during the experiment, light collection at either orthogonal or transmissive angles, and the introduction of additional pulsed or continuous light sources to allow photoluminescence (PL) experiments to be performed in situ. In order to observe and quantify the types of excitations produced in NCs under high-energy excitation, we compare time-resolved photoluminescence (trPL) and trCL for closepacked films of a size-series of CdSe/ZnS core/shell NCs. trPL spectra of this material system should exhibit distinct features separable by their unique spectral and temporal properties and attributable to specific excited states that are known to arise 927

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932

Nano Letters

Letter

Figure 3. Transient cathodoluminescence. (a) Streak camera images produced for the same CdSe/ZnS NC film examined in Figure 2 but using 20 keV, 100 pA electron-pulse-excitation. (b) Normalized spectral slices for the indicated excitation-delay times exhibit a rapidly decaying feature at 2.16 eV, similar to high photoexcitation intensity data. Lower electron beam current does not alter the early spectral profile (gray open circle symbols). The static PL spectrum of the film is shown for reference (dotted black line). (c) Normalized transient dynamics at the indicated CL spectral energies (symbols) initially decay with time constants that closely match biexciton and triexciton decay time scales, as noted by the early decay exponential fits (solid lines). However, an additional, more slowly decaying feature with a 770 ps time constant persists following 2X and 3X decay dynamics. This slower decay at 1.94 eV arises due to trions (X*) and the slower decay at 2.25 eV is attributed to decay of charged biexcitons (2X*).

such as trap-mediated photoionization24,25 or Auger-assisted ionization.39,40 The trCL feature at 2.16 eV initially shows much of the character seen in trPL, dominated at early times by 3X decay with some contribution from higher-order excitons. Here, we again observe a more slowly decaying contribution to trCL signals than in trPL data, consistent with formation of charged biexcitons (2X*) that, like 3X states, emit at higher energy due to state filling.36,37 Because the trCL measurements require repetitive excitation of the NC film over several minutes, we performed two control studies to establish the robustness of presented data. In addition to the potential for NC degradation over time, a concern in the trCL studies was the potential development of a significant population of charged NCs that persist between pulses that might give rise to surreptitious populations of charged exciton species such as trions. First, we investigated the electron beam intensity dependence of the trCL signals. Lower beam currents did, as expected, decrease the rise times of the trCL signal owing to reduced electron-pulse self-broadening arising from space-charge effects (see Supporting Information).31 However, as can be seen in Figure 3b, a trCL spectrum at early time using 10 pA appears unchanged relative to the spectrum resulting from 100 pA excitation, which suggests higher currents do not cause any additional sample degradation. Second, we verified that traces do not visibly change on repeated or prolonged exposure to the electron beam, either over several hours of nearly continuous experiments, or over months of periodic measurements on the same drop-cast films. Shorter term effects were also ruled out, as continuous lateral translation of the NC film (to constantly provide a fresh sample spot) also did not produce any discernible changes in trCL data, which suggests negligible effects of either sample degradation or charge-build up in the film for the utilized electron beam currents. We performed similar trPL and trCL measurements on smaller core CdSe/ZnS NCs films with static PL maxima of 2.07 and 2.34 eV (see Supporting Information). Comparative analyses of such studies resulted in similar conclusions. Overall, under conditions of energetic excitation multiexciton states are formed in abundance, which is in line with expected CM efficiencies for our estimated average deposited energy of ∼12 eV per NC (see Supporting Information). There is also a

observation window, which is due to single exciton decay from the lowest-energy 1S exciton state. The Gaussian spectral profile (see low fluence data in Figure 2d) exhibits a line width that reflects the size dispersity within the NC ensemble. Medium excitation fluence (40 μJ/cm2, ⟨N⟩ = 0.49, Figure 2b) yields a faster, pump-power-dependent decay component at roughly the same energy that arises from NCs containing a biexciton (2X). Biexcitons decay with a rate that is inversely proportional to NC volume32 and emit at energies ∼10 meV lower than the single exciton34,36 (unresolved here). Analysis of late-time-normalized dynamics (when only single excitons remain) at several pump intensities (Figure 2e) indicate a 2X lifetime of ∼196 ps, consistent with literature values.32,34,37,38 Finally, at high pump fluences (e.g., at 230 μJ/cm2, ⟨N⟩ = 2.93, Figure 2c) a short-lived, spectrally distinct PL feature appears at higher energy (2.16 eV, 580 nm). The intensity of this bluer peak is also pump-dependent (Figure 2d) and consistent with emission from NCs containing at least three excitons (a triexciton, 3X).37 The 3X and other higher-order states emit at higher energies due to state-filling of the two-fold degenerate 1S state, which causes additional excitons to populate the 1P level.32,34,37 The higher number of carriers in a 3X, relative to a 2X, leads to a predictably shorter Auger lifetime, which we determine to be ∼69 ps (Figure 2f). At this highest intensity, we also observe some contribution from even shorter-lived, higher-order excitonic states that are challenging to characterize owing to Poisson distribution spreading and instrumental response function convolution. We next investigate the same NC film using 20 keV electron excitation (trCL). Importantly, the trCL features shown in Figure 3a appear familiar. As in the high-pump-fluence trPL data, trCL spectra (Figure 3b) reveal a rapidly decaying bandedge emission feature as well as a distinct, short-lived feature at 2.16 eV indicating multiexcitons. Initial decay dynamics at the 1.94 eV band-edge in fact closely match the 2X decay constant observed in trPL studies (Figure 3c). However, we see an additional, much slower decay process at this energy characterized by a 770 ps exponential decay time constant that we attribute to NCs containing charged excitons or trions (X*, Figure 1c). Such excitations are known to occur in optically excited NCs at high photon energies or for large excitation densities and are thought to be the result of events 928

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932

Nano Letters

Letter

Table 1. Fitted Decay Constants for CdSe/ZnS NC Samples CdSe/ZnS Eg (eV)

tau X* (ps)

tau 2X (ps)

tau 2X* (ps)

tau 3X (ps)

tau 4X (ps)

cross section @3.1 eV (cm2)

2.34 2.07 1.94

254 ± 20 510 ± 40 770 ± 50

42 ± 7 98 ± 10 196 ± 20

(unresolved) (unresolved) 120 ± 20

18 ± 6 30 ± 10 69 ± 15

2 ns in our experiments and labeled “single exciton”) is roughly 4, attributing a nearly quantitative origin of the “single exciton” signal to NCs initially containing (at least) biexcitons. Similar relations used to calculate excitation branching ratios are detailed in Supporting Information for all discernible species. Using such detailed characterization of the dynamic and spectroscopic responses of the different electronic excitations formed in the NCs in addition to knowledge of the instrument response function, in Figure 4c,d we show that we can successfully reconstruct measured trCL data. Table 2 summarizes the excitation fractions produced via analysis of trCL data collected for the multiple samples. Here, it can be seen that 20 keV excitation yields predominantly charged exciton species (55−60%) and multiexciton states (30−40%) and remarkably few single excitons. Upon the basis of the control experiments described earlier that indicate a lack of dependence on the experimentally utilized range of beam currents, it seems that such high-order electronic species arise due to efficient CM wherein a single energetic excitation event yields multiple excited carriers in most NCs. It is important to consider that because our measurement relies upon optical emission, it is insensitive to any NCs that contain only a single charge following excitation, so they are not accounted for in our excitation branching ratios. However, given the very small number of single-exciton-excited NCs, it seems unlikely that this is a substantial fraction of the total excited population. Furthermore, the data in Table 2 offers little evidence of any size-dependent trends within the studied range of sizes, likely as a result of the large amount of energy deposited per NC on average. At first glance, the large fraction of excited states that emit at energies greater than the NC band gap (3X, 2X*, etc.) would seem to suggest that the steady-state emission under electron excitation should be substantially different than that

observed under optical excitation. However, despite the negligible amount of NCs initially populated with a single exciton, the X states are still responsible for the vast majority of photons emitted under electron excitation, because of the very low effective quantum yield of the higher states due to Auger recombination. A simple calculation of expected photon yields (Table S1 in the Supporting Information) reveals that total emission is still nearly entirely at or near the band edge, which is in agreement with previously reported steady-state CL of similar NCs.28 In conclusion, these studies demonstrate that trCL offers a useful means to determine excitation branching ratios under energetic excitation scenarios such as radiation detection, which permits in-depth characterization and insight regarding novel radiation detection materials. Specifically, we showed that 20 keV excitation of colloidal NC films ultimately produces an identifiable population of well-understood excitations within the quantum-confined band structure, despite nanoscale granularity. Our analysis directly shows that energetic, atomic excitation yields a highly non-Poissonian population of chargedand multiexcitonic states, consistent with large contributions from CM. This finding has direct consequences for NC-based gammaand X-ray detectors, and for scintillators in particular, because it means that a great deal of the incident energy contained in such high-energy excitations goes toward creation of states featuring more than two excited carriers. Decay of such states in these and essentially all colloidal NCs is dominated by nonradiative Auger recombination processes with efficiencies that scale with the number of total carriers, meaning that each additional excited carrier adds very little to the total light emission. Because we find that most excited NCs do absorb more than one band gap of energy, total time-integrated signal can be expected to be largely insensitive to the energy of the incident photon. These findings account for the relatively modest performance of NC-based composites in previous energyresolving radiation detector studies without invoking extraneous effects, such as self-reabsorption in composites due to insufficient Stokes shift. As such, they suggest that future efforts should focus on using and further refining the emergent class of engineered NCs in which Auger recombination is highly suppressed,44−48 instead of on use of, for example, type-II or doped NCs with large effective Stokes shifts.3,20−23 Experimental Methods. Three sizes of CdSe/ZnS NC materials with high-photoluminescence quantum yields were produced using literature methods.49 After twice-precipitating the NCs, close-packed films up to 1 mm thick were produced by drop-casting on either glass microscope slides or commercially acquired conductive ITO substrates, which were electrically grounded to the chamber wall. Use of glass or ITO did not affect presented results. NC films were placed in a 10−6 Torr vacuum chamber on a translation stage with motion normal to the excitation axis. trPL experiments were conducted at 1 kHz with 50 fs pulses from an amplified Ti-sapphire laser at 930

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932

Nano Letters

Letter

(8) Letant, S. E.; Wang, T. F. Nano Lett. 2006, 6, 2877. (9) Wang, C. L.; Gou, L.; Zaleski, J. M.; Friesel, D. L. Nucl. Instrum. Methods Phys. Res., Sect. A 2010, 622, 186. (10) Dabbousi, B. O.; RodriguezViejo, J.; Mikulec, F. V.; Heine, J. R.; Mattoussi, H.; Ober, R.; Jensen, K. F.; Bawendi, M. G. J. Phys. Chem. B 1997, 101, 9463. (11) Hines, M. A.; Guyot-Sionnest, P. J. Phys. Chem. 1996, 100, 468. (12) Chan, E. M.; Mathies, R. A.; Alivisatos, A. P. Nano Lett. 2003, 3, 199. (13) Klein, C. A. J. Appl. Phys. 1968, 39, 2029. (14) Schaller, R. D.; Klimov, V. I. Phys. Rev. Lett. 2004, 92, 186601. (15) McGuire, J. A.; Joo, J.; Pietryga, J. M.; Schaller, R. D.; Klimov, V. I. Acc. Chem. Res. 2008, 41, 1810. (16) Semonin, O. E.; Luther, J. M.; Choi, S.; Chen, H.-Y.; Gao, J.; Nozik, A. J.; Beard, M. C. Science 2011, 334, 1530. (17) Geehyun, K.; Huang, J.; Hammig, M. D. IEEE Trans. Nucl. Sci. 2009, 56, 841. (18) Letant, S. E.; Wang, T.-F. Appl. Phys. Lett. 2006, 88, 103110. (19) Liptay, T. J.; Marshall, L. F.; Rao, P. S.; Ram, R. J.; Bawendi, M. G. Phys. Rev. B 2007, 76, 155314. (20) Kim, S.; Fisher, B.; Eisler, H.-J.; Bawendi, M. J. Am. Chem. Soc. 2003, 125, 11466. (21) Li, L.; Pandey, A.; Werder, D. J.; Khanal, B. P.; Pietryga, J. M.; Klimov, V. I. J. Am. Chem. Soc. 2011, 133, 1176. (22) Pradhan, N.; Peng, X. J. Am. Chem. Soc. 2007, 129, 3339. (23) Viswanatha, R.; Brovelli, S.; Pandey, A.; Crooker, S. A.; Klimov, V. I. Nano Lett. 2011, 11, 4753. (24) McGuire, J. A.; Sykora, M.; Robel, I.; Padilha, L. A.; Joo, J.; Pietryga, J. M.; Klimov, V. I. ACS Nano 2010, 4, 6087. (25) Padilha, L. A.; Robel, I.; Lee, D. C.; Nagpal, P.; Pietryga, J. M.; Klimov, V. I. ACS Nano 2011, 5, 5045. (26) Knoll, G. F. Radiation detection and measurement, 4th ed.; Wiley: Hoboken, NJ, 2010; p 830. (27) Milbrath, B. D.; Peurrung, A. J.; Bliss, M.; Weber, W. J. J. Mater. Res. 2008, 23, 2561. (28) Rodriguez-Viejo, J.; Jensen, K. F.; Mattoussi, H.; Michel, J.; Dabbousi, B. O.; Bawendi, M. G. Appl. Phys. Lett. 1997, 70, 2132. (29) Merano, M.; Sonderegger, S.; Crottini, A.; Collin, S.; Renucci, P.; Pelucchi, E.; Malko, A.; Baier, M. H.; Kapon, E.; Deveaud, B.; Ganiere, J. D. Nature 2005, 438, 479. (30) Williamson, J. C.; Cao, J.; Ihee, H.; Frey, H.; Zewail, A. H. Nature 1997, 386, 159. (31) Janzen, A.; Krenzer, B.; Heinz, O.; Zhou, P.; Thien, D.; Hanisch, A.; Heringdorf, F.; von der Linde, D.; von Hoegen, M. H. Rev. Sci. Instrum. 2007, 78, 7. (32) Klimov, V. I.; Mikhailovsky, A. A.; McBranch, D. W.; Leatherdale, C. A.; Bawendi, M. G. Science 2000, 287, 1011. (33) Crooker, S. A.; Barrick, T.; Hollingsworth, J. A.; Klimov, V. I. Appl. Phys. Lett. 2003, 82, 2793. (34) Fisher, B.; Caruge, J.-M.; Chan, Y.-T.; Halpert, J.; Bawendi, M. G. Chem. Phys. 2005, 318, 71. (35) Jha, P. P.; Guyot-Sionnest, P. ACS Nano 2009, 3, 1011. (36) Achermann, M.; Hollingsworth, J. A.; Klimov, V. I. Phys. Rev. B 2003, 68, 245302. (37) Caruge, J. M.; Chan, Y.; Sundar, V.; Eisler, H.-J.; Bawendi, M. G. Phys. Rev. B 2004, 70, 085316. (38) Klimov, V. I.; McGuire, J. A.; Schaller, R. D.; Rupasov, V. I. Phys. Rev. B 2008, 77, 195324. (39) Chepic, D. I.; Efros, A. L.; Ekimov, A. I.; Ivanov, M. G.; Kharchenko, V. A.; Kudriavtsev, I. A.; Yazeva, T. V. J. Lumin. 1990, 47, 113. (40) Klimov, V. I.; McBranch, D. W. Phys. Rev. B 1997, 55, 13173. (41) Schaller, R. D.; Klimov, V. I. Phys. Rev. Lett. 2006, 96, 097402. (42) Klimov, V. I. J. Phys. Chem. B 2006, 110, 16827. (43) Robel, I.; Gresback, R.; Kortshagen, U.; Schaller, R. D.; Klimov, V. I. Phys. Rev. Lett. 2009, 102, 177404. (44) Chen, Y.; Vela, J.; Htoon, H.; Casson, J. L.; Werder, D. J.; Bussian, D. A.; Klimov, V. I.; Hollingsworth, J. A. J. Am. Chem. Soc. 2008, 130, 5026.

3.1 eV. PL photons were collected at a small angle from the excitation and directed to a 150 mm spectrograph and photoncounting streak camera. Detector regions were binned appropriately to produce time-resolved spectra or spectrally resolved dynamics. Photoexcitation fluence was controllably varied in order to produce single- and multiexciton states for comparison with trCL measurements. For trCL measurements, femtosecond laser pulses at 4.66 eV with a 1 kHz repetition rate were focused (slightly off-axis with a 25 mm focal length lens) onto the back of a 2.5 nm thick gold photocathode biased to −20 keV in order to produce picosecond electron pulses.29−31 Photoejected electrons were accelerated toward the sample using a grounded gold-coated extraction pinhole and focused to a ∼1 mm diameter spot using a three-element electrostatic Einzel lens. The electron gun produced between 1 and 200 pA electron current at 1 kHz as controlled by the 4.66 eV laser power and measured with a retractable anode and electrometer. Presented trCL streak images utilized 100 pA beam current (600 000 electrons/pulse) at 1 kHz for high quality images. trCL dynamics utilized a range of currents from 10 pA (60 000 electrons) to as high as 150 pA (∼900 000 electrons).



ASSOCIATED CONTENT

* Supporting Information S

trCL instrument response function characterization, estimate of the charge and energy deposited by electron pulses, measurements of trCL for additional NC samples, identification of different excited species, quantification of excitation branching ratios, Figures S1−S6, and Table S1. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: (J.M.P.) [email protected]; (R.D.S.)[email protected]. Notes

The authors declare no competing financial interests.



ACKNOWLEDGMENTS L.A.P., W.K.B., and J.M.P. were supported by the Los Alamos National Laboratory LDRD program. V.I.K. was supported by Chemical Sciences, Bioscience and Geosciences Division of the Office of Basic Energy Sciences (BES), Office of Science, U.S. Department of Energy. This work was performed, in part, at the Center for Nanoscale Materials, a U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences User Facility under Contract No. DE-AC02-06CH11357.



REFERENCES

(1) Klimov, V. I. Semiconductor and metal nanocrystals: Synthesis and electronic and optical properties, 2nd ed.; Marcel Dekker: New York, 2010. (2) Nozik, A. J.; Beard, M. C.; Luther, J. M.; Law, M.; Ellingson, R. J.; Johnson, J. C. Chem. Rev. 2010, 110, 6873. (3) Klimov, V. I.; Ivanov, S. A.; Nanda, J.; Achermann, M.; Bezel, I.; McGuire, J. A.; Piryatinski, A. Nature 2007, 447, 441. (4) Pal, B. N.; Robel, I.; Mohite, A.; Laocharoensuk, R.; Werder, D. J.; Klimov, V. I. Adv. Mater. 2012, 22, 1741. (5) Kwak, J.; Bae, W. K.; Lee, D.; Park, I.; Lim, J.; Park, M.; Cho, H.; Woo, H.; Yoon, D. Y.; Char, K.; Lee, S.; Lee, C. Nano Lett. 2012, 12, 2362. (6) Campbell, I. H.; Crone, B. K. Adv. Mater. 2006, 18, 77. (7) Kang, Z.; Zhang, Y.; Menkara, H.; Wagner, B. K.; Summers, C. J.; Lawrence, W.; Nagarkar, V. Appl. Phys. Lett. 2011, 98, 181914. 931

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932

Nano Letters

Letter

(45) García-Santamaría, F.; Chen, Y.; Vela, J.; Schaller, R. D.; Hollingsworth, J. A.; Klimov, V. I. Nano Lett. 2009, 9, 3482. (46) Mahler, B.; Spinicelli, P.; Buil, S.; Quelin, X.; Hermier, J.-P.; Dubertret, B. Nat. Mater. 2008, 7, 659. (47) Spinicelli, P.; Buil, S.; Quélin, X.; Mahler, B.; Dubertret, B.; Hermier, J. P. Phys. Rev. Lett. 2009, 102, 136801. (48) Wang, X.; Ren, X.; Kahen, K.; Hahn, M. A.; Rajeswaran, M.; Maccagnano-Zacher, S.; Silcox, J.; Cragg, G. E.; Efros, A. L.; Krauss, T. D. Nature 2009, 459, 686. (49) Bae, W. K.; Char, K.; Hur, H.; Lee, S. Chem. Mater. 2008, 20, 531.

932

dx.doi.org/10.1021/nl400141w | Nano Lett. 2013, 13, 925−932