Ultrafast stimulated emission microscopy of single nanocrystals

See allHide authors and affiliations

Science  06 Dec 2019:
Vol. 366, Issue 6470, pp. 1240-1243
DOI: 10.1126/science.aay1821

Tracking excitations

Illumination can be used to excite a sample from its ground state to a number of excited states. Typically, however, the details of the excitation dynamics are hidden from view because they decay so fast. Piatkowski et al. combined pump-probe transient absorption and two-pulse photoluminescence correlation spectroscopy, allowing them to assess stimulated emission and ground-state bleaching contributions to the transient absorption signal. This approach provides a window on the excitation dynamics within single nanocrystals and should also be useful for ultrafast nanocharacterization of complex samples.

Science, this issue p. 1240


Single-molecule detection is a powerful method used to distinguish different species and follow time trajectories within the ensemble average. However, such detection capability requires efficient emitters and is prone to photobleaching, and the slow, nanosecond spontaneous emission process only reports on the lowest excited state. We demonstrate direct detection of stimulated emission from individual colloidal nanocrystals at room temperature while simultaneously recording the depleted spontaneous emission, enabling us to trace the carrier population through the entire photocycle. By capturing the femtosecond evolution of the stimulated emission signal, together with the nanosecond fluorescence, we can disentangle the ultrafast charge trajectories in the excited state and determine the populations that experience stimulated emission, spontaneous emission, and excited-state absorption processes.

Complex physical, chemical, and biological processes are determined by fundamental spatial and temporal interaction trajectories. Only ultrafast techniques with single-emitter sensitivity can unveil their inherent transient intermediates and allow exploration of processes such as molecular vibrations and energy transfer (13) and of nanoscale dynamics in plasmonic or two-dimensional materials (4, 5). The small interaction cross sections of individual nanoparticles make it difficult to rely on the conventional ultrafast approaches, such as transient absorption and nonlinear four-wave mixing. Consequently, single molecules and nanoparticles are almost exclusively detected through Stokes-shifted spontaneous emission [fluorescence or photoluminescence (PL)], which is background-free, allowing for photon counting sensitivity and detection of weakly fluorescent emitters. The use of fluorescence detection, however, is hampered by a number of limitations: It is restricted to luminescent samples, is sensitive to bleaching, and, in the linear regime, is slow (nanoseconds), reporting only on the population of the final emitting state and missing out on femtosecond dynamics. Despite the exploration of several alternative detection schemes, such as photothermal (6), linear absorption (7, 8), and enhanced Raman (9), ultrafast detection of single entities beyond fluorescence has remained challenging.

Here, we demonstrate a highly sensitive experimental scheme based on the direct detection of stimulated emission (SE) for studying the excited-state dynamics in nanoscopic samples with femtosecond temporal resolution. SE microscopy involves one laser pulse for promotion to the excited state and a second, delayed pulse for stimulation back to the ground state, generating a new SE photon (10). SE forms the basis of stimulated emission depletion (STED) microscopy; however, in a typical STED experiment, the stimulated photons are discarded and only PL is recorded. Yet, the instantaneous SE photons contain information on the excited-state population and its dynamics and relaxation pattern, which is otherwise inaccessible from the slow PL. To its advantage, SE is not dependent on the quantum efficiency of the sample, has femtosecond temporal resolution, is coherent, and is capable of mapping the dynamics of an arbitrary excited state.

We present direct stimulated emission detection and imaging of individual nanocrystals (NCs) and trace the excited-state dynamics of single colloidal CdSe/CdS rod-in-rod NCs (11) with femtosecond temporal resolution at ambient conditions. The PL is detected simultaneously with the SE, generating two independent, complementary images. It is important to understand the dynamic interplay between various charge relaxation pathways—such as charge injection, extraction, transfer and delocalization, and excited-state relaxation—both ultrafast and with nanoscopic sensitivity (1214). Our femtosecond SE experiment on single NCs shows the excited-state relaxation dynamics of individual charges, the dynamical heterogeneity of NCs, and the relative contributions of the various stimulated processes, all with single-NC sensitivity.

A pump beam excites the NC through two-photon absorption to a highly excited state in the conduction band [Fig. 1; for details, see materials and methods (15)]. The excited hot electrons and holes, initially localized in the shell, decay through the excited-state progression and eventually localize in the lowest excited state (band edge) in the core (Fig. 1C). The probe (stimulation) beam, resonant with the core band-edge transition, leads to charge recombination, stimulates the NC back to the ground state, and induces emission of a stimulated photon. Therefore, any information on the excited charges imprinted by the pump beam in the shell is “read out” by the stimulating probe beam, when one of the excited charges reaches the core band-edge states. The pump beam is modulated, and the SE signal is retrieved by lock-in detection.

Fig. 1 Concept of the ultrafast stimulated emission nanoscopy.

(A) Schematic of the experimental setup. PD, photodiode; LIA, lock-in amplifier; APD, avalanche photodiode; AOM, acousto-optic modulator. (B) Spectral characteristics of the broadband laser pulse (pump pulse, brown; probe pulse, red) and CdSe/CdS NCs. Gray- and blue-shaded areas represent the absorption and emission spectra of the NCs, respectively. The black area indicates the spectral range of the two-photon absorption (2PA). (C) Energy-level sketch of a core/shell CdSe/CdS NC. CB, conduction band; VB, valence band.

As a first step, we raster-scanned the sample while simultaneously detecting both modulated signal (Smod) and PL (Fig. 2, A and B). The PL image clearly reveals the presence of the NCs, which we verified through their emission spectra (fig. S1). The corresponding Smod image shows contrast at the same sample positions where the PL signal appears. Moreover, the measured Smod signal was always positive, meaning we detected extra photons in our stimulation beam (supplementary text 1). Two effects can, in principle, lead to an increase in the transmitted probe-beam intensity when the NC is excited: stimulated emission and ground-state depletion (GSD). In the first case, the SE process following electron-hole recombination gives a net increase in the probe-beam intensity. In the second case, the absorption of the probe beam is lower because of the depletion of the ground state, owing to the presence of either a hole or electron in its respective energy level. The two contributions can be readily distinguished by time-resolved experiments, as shown later. For most NCs, we found a perfect correspondence between PL and Smod images and observed Smod wherever PL appeared (Fig. 2C). Interestingly, in some cases, we detected PL but no measurable Smod (Fig. 2E). We assigned this signal to core-free CdS shell nanoparticles that conucleated during synthesis. Finally, on rare occasions, we observed Smod contrast but no PL (Fig. 2D). The signal likely originated from highly quenched NCs, because it is improbable that we would have observed other particles with the exact same spectral signature. Clearly, the spectral dependence of Smod correlated with the probe beam, and the ability to simultaneously detect PL and Smod gives us extra insight as to the nature of the detected NCs.

Fig. 2 Stimulated emission imaging.

(A and B) Confocal images of the same sample area showing PL and the lock-in signal (Smod), respectively. The stimulation beam was set to arrive 7 ps after the pump beam (supplementary text 2). (C to E) Comparison between the PL and Smod images for the three regions of interest indicated in (A) and (B).

Ultrafast coherent response is the main advantage of SE detection. In Fig. 3A, we show a series of PL and Smod images for different interpulse delay times (see fig. S2 for more images). Although the PL signal is detected at all delay times ∆t, the Smod signal appears only when the pump pulse overlaps or precedes the stimulation pulse. At negative delay times, when the stimulation pulse arrives before the pump pulse, the NC is in its ground state and there is no excited-state population for the probe pulse to interact with. For the NC marked with an “x” in Fig. 3A, the second-order autocorrelation trace exhibits a dip with degree of coherence g(2)(0) ≲ 0.5, indicating the nonclassical emission of a single NC (fig. S3). The time-resolved traces revealed that when Smod (blue) increases in time, the PL (red) decreases (Fig. 3B). This is intuitive: The excited-state population, which is stimulated down back to the ground state, does not contribute to the spontaneous emission, leading to PL depletion. The fact that Smod and PL signals are anticorrelated unambiguously indicates that Smod contains a substantial contribution from the SE process. Furthermore, we found that the changes in both signals, Smod ingrowth (∆Smod) and PL depletion (∆PL), occur on specific time scales. Interestingly, the ∆PL depletion occurs with a single time constant, whereas ∆Smod grows in with two time constants. The slower time constant of ~400 to 700 fs is identical to the time constant with which ∆PL decreases. However, a considerable part of the Smod grows on a faster time scale and cannot be observed within the cross-correlation time of the pump and probe pulses (<200 fs). To understand this, one needs to consider that the NCs are initially pumped to a highly excited state in the shell (supplementary text 3), whereas the stimulation pulse probes the lowest excited state in the core. GSD occurs when charges are present in the excited state of the transition resonant with the probe energy. As soon as the faster of the two charges reaches the lowest excited state of the core (1619), the probe-beam absorption will decrease. This means that GSD reports on the relaxation rate of the fastest charge, either the electron or the hole. By contrast, the probe beam can induce charge recombination and SE only when both electron and hole localize into the core. Consequently, SE is sensitive to relaxation of the slower of the two charges. In the PL, we see only the slower component because PL is a time-averaged signal, which is mostly sensitive to the population decay of the lowest excited state (supplementary text 4).

Fig. 3 Time-resolved stimulated emission microscopy.

(A) A series of images acquired by detecting PL and the Smod signal for different excitation and stimulation interpulse delays ∆t. (B) Simultaneously detected Smod (blue) and PL (red) time traces for a CdSe/CdS NC. (C) Histogram of the exciton relaxation times. Red, blue, and green histograms correspond to relaxation times extracted from the fits to individual time traces of three different, single NCs. The black histogram shows occurrences of relaxation times extracted from averaged traces from NC clusters. (D) Histograms showing the relative contributions of the SE (blue) and the ∆PL (red) to the total detected signal change Smod and PLt− − PLt+, respectively.

We quantified the observed dynamics by simultaneously fitting the PL and Smod traces (supplementary text 5). PL and Smod traces acquired on small NC clusters revealed that the average slower charge relaxation time is 550 fs (black histogram in Fig. 3C). The time-delay traces recorded repeatedly on the same individual NCs (for more traces, see fig. S5) revealed the relaxation heterogeneity among individual NCs (Fig. 3C). From the difference in the dynamics between SE and GSD, we determined the relative contribution of the two signals to the total measured signal Smod by performing simple, qualitative kinetic rate equation calculations (supplementary text 6). The experimental ratio of SE/Smod extracted from individual time traces for a large number of NCs centers around a value of ~0.17 (Fig. 3D). The observation of a ratio SE/Smod < 0.2 strongly suggests that the cross sections for absorption and SE might be somewhat different, given the large asymmetry between the shape of the absorption and emission bands.

The lower SE signal with respect to GSD signal might also be caused by an excited-state absorption (ESA) process. In ESA, the probe beam promotes the excited charges to higher excited states at the cost of absorbing a probe-beam photon, leading to a reduction of the apparent SE contrast and enhanced bleaching (14, 20) and quenching (21). To uncover the role of ESA in our NC dynamics, we varied the duration of the probe pulse, because the ESA timing should be sensitive to the observed 550-fs relaxation time of the hot state. Once the charges have again returned to the emitting state, the probe pulse should stimulate the NC down. The concept, depicted in Fig. 4B, is analogous to STED experiments, where the STED pulse is stretched to prevent reexcitation (22). We measured the Smod and ∆PL contrast for increasing probe-pulse duration (Γ), stretched up to 2.5 ps, at ∆t = 7 ps delay. In Fig. 4C, both Smod and ∆PL show increased contrast with the probe-pulse duration. Interestingly, the ingrowth matches very well the 550-fs excited-state charge relaxation time determined from the pump-probe traces. A simulation using the kinetic rate equation model expanded with the ESA process (supplementary text 7) reproduces the experimental data well and confirms our hypothesis that stretching the stimulating probe pulse allows stimulation down of charges that otherwise undergo ESA.

Fig. 4 Higher stimulated emission and photoluminescence contrast with longer probe pulse.

(A) PL and Smod signals recorded in time while repeatedly scanning the interpulse delay time ∆t from negative to positive values. a.u., arbitrary units. (B) Concept of the varying probe-pulse duration experiment. (C) Normalized Smod and ∆PL as a function of probe-pulse duration. The traces were averaged from seven separate measurements (four positively and three negatively chirped probe traces) on different NC clusters. Error bars indicate the standard deviation. The black dashed line is the result of solving the set of kinetic rate equations described in supplementary text 7.

Interestingly, the simultaneous detection of stimulated and spontaneous emission of a single NC allows us to correlate the decays in a quantitative manner. The number of photons detected in SE should be equal to the number of photons missing in PL, that is, PL depletion. For the data shown in Fig. 3B, we determined an effective number of photons depleted from PL, ΔPLeff = 1.6 × 107 photons/s, and an effective number of photons gained in the stimulation beam, ΔSEeff = 1.3 × 107 photons/s per NC (supplementary text 8). The two values are in good agreement, given that the detection occurs in two independent channels, using photon counting versus analog detectors.

The high sensitivity of the presented SE detection opens up new imaging possibilities for weakly fluorescing or quenched systems, and the time-resolved experiment provides information on the excited-state relaxation dynamics and its mechanism, all with femtosecond time resolution and single-emitter sensitivity. The unconventional, simultaneous detection of the spontaneous and stimulated emission provides large imaging specificity: The fact that SE depends on two distinct frequencies, in combination with the interpulse time delay, makes the method extremely sensitive to different species within a dense ensemble.

The time-resolved femtosecond SE experiment allowed to us to provide a comprehensive picture of the excited charges, which are either stimulated down or promoted to higher excited states or recombine spontaneously. The SE and GSD contributions comprise <20 and >80% of the total induced ground- and excited-state population difference, respectively. This was aided by the fact that the two excited charges—electrons and holes—exhibit different relaxation times (supplementary text 9). The rod-in-rod CdSe/CdS NC excited holes localize at the core band edge within 200 fs, whereas the excited electrons relax to the core band edge on a time scale of 550 fs. We found that the electron relaxation time differs by nearly a factor of two between individual NCs. Finally, the single-emitter sensitivity of our experiment allowed us to compare the number of photons lost in PL and gained through SE in absolute terms, which is difficult to achieve for ensembles (23). Stretching the stimulation pulse allowed us to elucidate the presence of ESA and increase the SE efficiency by 40 to 50%, that is, a substantial portion of the excited charges undergo ESA and relax back to the core band-edge states.

The ultrafast SE microscopy opens up a spectrum of experiments for exploration (supplementary text 10). Scanning the stimulation pulse energy would allow for state selectivity and enable the study of excited state-to-state dynamics (16). Given its coherent nature, SE microscopy could be expanded to accommodate heterodyne detection of the stimulation beam and could provide easy access to investigating coherent effects such as coherent energy transfer (3, 24). The absorption cross section of our NCs at the stimulation wavelength is approximately an order of magnitude larger than the absorption cross section of a typical fluorescent dye (3 × 10−16 versus 10−17 cm2) (25). Therefore, even single molecules could be detected in stimulated emission.

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S12

References (2635)

References and Notes

  1. See the supplementary materials.
Acknowledgments: L.P. acknowledges the Marie Skłodowska-Curie COFUND and the ICFOnest programs. Funding: This project received funding from the National Science Centre, Poland, grant 2015/19/P/ST4/03635, POLONEZ 1, and from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement no. 665778. This research was funded by the European Commission (ERC Advanced Grant 670949-LightNet), the Spanish Ministry of Economy MINECO (FIS2012-35527, FIS2015-72409-EXP, FIS2015-69258-P, Network FIS2016-81740-REDC “NanoLight,” and Severo Ochoa Grant SEV2015-0522), the Catalan AGAUR (no. 2017SGR1369), Fundació Privada Cellex, Fundació Privada Mir-Puig, and Generalitat de Catalunya through the CERCA Program. Author contributions: L.P. and N.F.v.H. designed the experiment. L.P., N.A., and G.C. performed the experiments and data analysis. S.C. and I.M. provided the samples. L.P. and N.F.v.H. wrote the manuscript. All authors discussed the results and commented on the manuscript. Competing interests: The authors declare no competing financial interests. Data and materials availability: All data are available in the main text or the supplementary materials.

Stay Connected to Science

Navigate This Article