Mid-Infrared Photoluminescence of CdS and CdSe Colloidal Quantum

Mid-Infrared Photoluminescence of CdS and CdSe Colloidal Quantum Dots Kwang Seob Jeong† and Philippe Guyot-Sionnest* James Franck Institute, The University of Chicago, 929 East 57th Street, Chicago, Illinois 60637, United States S Supporting Information *

ABSTRACT: Mid-infrared intraband photoluminescence is observed from CdSe and CdS colloidal quantum dots (CQDs) and core/shell systems when excited by a visible laser. The CQDs show more intraband photoluminescence with dodecanethiol than with other ligands. Core/shells show an increase of the intraband photoluminescence with increasing shell thickness. The detected emission is restricted to below 2900 cm−1, bounded by the C−H vibrational modes of the organic ligands. Upon photoexcitation in air for all dodecanethiol ligands capped CQD systems studied, the intraband photoluminescence is quenched over time, and emission at lower frequency is observed, which is assigned to laser heating and thermal emission from oxides.

KEYWORDS: intraband photoluminescence, II−VI colloidal quantum dot, CdSe, CdSe/ZnS, CdSe/ZnSe, CdSe/ZnTe, CdS/ZnSe

F

vibrational absorption, it was proposed that direct near-field energy transfer to the ligands vibrations led to very fast intraband relaxation.23 Removing the hole and reducing the coupling to ligand vibrations guided the strategy to slow down the electron cooling to ∼1 ns in a subsequent transient absorption investigation.24 The most efficient intraband PL should therefore arise when the holes can be extracted and the ligands vibrations are weak or far away from the intraband transitions. Lifetimes of nanosecond duration should lead to PL quantum yields in excess of 10−3 and therefore be measurable. This work reports the first observations of intraband photoluminescence of visible band gap CdSe, CdSe/ZnS, CdSe/ ZnSe, CdSe/ZnTe, and CdS/ZnSe CQDs, focusing on the electron intraband transition occurring between 1Pe and 1Se states. The intraband PL is measured in the mid-infrared as a function of nanoparticle size, ligand and shell.

or the past 20 years, the tunable bandgap transitions of colloidal quantum dots (CQDs) have been the sole focus to develop the optical applications of these materials, whereas the intraband transitions in epitaxial quantum dots and inter-subband transitions in quantum wells have been used for just as long for optoelectronic applications.1,2 Recently, mid-infrared photoluminescence (PL) and photoconduction directly from intraband transitions indicate that the intraband transitions provide a new opportunity for colloidal quantum dots in the mid-infrared.3,4 Studying intraband PL and hot electron cooling is also relevant to hot carrier concepts for water splitting, hydrogen gas generation/dissociation, and photovoltaic devices.5−18 Although intraband PL has not been previously reported in cadmium chalcogenides CQDs, it should be related to the nonradiative intraband relaxation which has been extensively studied. Previous transient absorption (TA) studies showed that relaxation of high energy excitonic transitions such as 1P is sub-picoseconds in CdSe CQDs,19 and it is believed to be driven by the Coulomb interaction between electrons and holes.20 Further transient absorption measurements of the intraband relaxation showed that removing the hole led to a small effect on the intraband relaxation where lifetimes remained only a few picoseconds with hole trapping surface ligands, such as thiophenol or pyridine, or n-type dots.21,22 Later, significantly longer intraband relaxation lifetimes of ∼30 ps were observed with dodecanethiol capped CdSe CQDs.23 Such slow intraband relaxation was however only observed in the 4−5 μm range which is rather void of molecular vibrational absorption. After a comparison of several ligands and their © 2016 American Chemical Society

RESULTS AND DISCUSSION To photoexcite the CQDs, a pulsed laser at 527 nm is used (1 kHz, ∼250 ns). This wavelength can excite the 1S exciton of all the CdSe dots studied. The low energy tail of the 1P exciton can also be excited for the larger size CdSe CQDs as shown in Figure S1. Figure 1a shows mid-IR TA and PL spectra of dodecanethiol capped CdSe CQDs. The samples are films on a ZnS infrared transparent window, sealed in an airtight cell inside a glovebox before being taken out for optical studies. The Received: October 31, 2015 Accepted: January 22, 2016 Published: January 22, 2016 2225

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231

Article

www.acsnano.org

Article

ACS Nano

Figure 1. (a) TA (black) and PL (red) spectra of CdSe-dodecanethiol CQD films. The raw PL intensities are in the ratios of 1, 0.5, and 0.2 for the samples with 591, 588, and 579 nm lowest exciton peak, respectively. No intraband PL could be detected for the smallest sample with 567 nm exciton peak. (b) The schematic diagram represents the 1Pe−1Se intraband absorption (black) and emission (red).

Figure 2. (a) Pump power dependence of the TA signal for a film of dodecanethiol-capped CdSe. The line is a fit as described in the text with a saturation power of 41 mW. (b) Pump power dependence of the intraband PL. The lines are fits as described in the text. The dotted line is a two parameter fit giving σ0 IR/σ1 IR ∼ 0.06. The solid line is a fit setting σ0 IR = 0. (c) PL spectra as a function of power. (d) TA spectra as a function of power.

observed in the region between the CH stretching (2900 cm−1) and bending modes (1400 cm−1) of hydrocarbons ligands. Previous studies could also not measure intraband relaxation in the CH stretch region presumably because it was too fast,23 and this is consistent with the absence of intraband PL for the smallest CdSe sample in Figure 1a with a TA peak around 3000 cm−1. We propose that the Stokes shift and the narrower PL compared to the TA in Figure 1a are in large part due to the ligands shaping the emission within the inhomogeneous CQD distribution, another possible contribution being the splitting of the 1Pe states. The ligands strongly affect the PL intensity, where octadeylamine, stearic acid, and TOP/TOPO ligands led to

mid-IR TA results from occupation of the 1Se state and the resulting absorption to the 1Pe state shown in the schematic in Figure 1b. With a decrease of nanocrystal size from 4.6 nm (excitonic absorption peak at 591 nm) to 3.7 nm (567 nm), the TA peak blue-shifts by 600 cm−1 (74 meV) consistent with earlier studies of size dependent intraband spectra.25 The intraband PL spectra follow the TA spectra, but they are narrower and the Stokes shift decreases with increasing nanocrystal size. Qualitatively similar results were obtained for dodecanethiol capped CQDs in solutions. The characteristics of the intraband PL spectra obtained here are rather consistent with the previous understanding on the intraband relaxation, i.e., strongest PL 2226

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231

Article

ACS Nano

Figure 3. (a) Evolution of the IR emission as a function of time for a sample exposed to air; (b) as a function of gate delay. (c) Intraband PL for a sample kept in inert conditions (filled circles), and thermal emission after prolonged photoexcitation in air (open squares) at two different temperature 300 K (red) and 80 K (blue). (d) Increased IR absorption of a film of CdSe-dodecanethiol, upon exposure to air and irradiation at 450 nm.

shows an delayed increase with power as 1Se first needs to be populated, but it is then linear with power. To model the power dependence, we consider that the dots can be photocharged with one electron but that the negative trion lifetime is sufficiently short to prevent two electrons charging. At quasi-steady state, the number of dots with one electron will be n1 ∝ I/(I/I0 + 1) where I0 is a saturation photon flux (cm−2 s−1) which is the product of the inverses of the interband excitation cross-section of the dots and the lifetime of the electron. The power at saturation Po = 41 mW is obtained by the fit in Figure 2. The peak photon flux is I0 = 2P0/(hνfτpπw2), where w is the spot size, 0.48 mm, f is the repetition rate of 1 kHz, τp is the pulse width of 250 ns, h is the Plank’s constant and ν is the photon frequency. Saturation occurs at a photon flux of 1.2 × 1023 photons/cm2s. with the use of an absorption cross section of ∼2 × 10−15 cm2,27 this gives a recombination/trapping time of the electrons of 4 ns, which is reasonable at room temperature.21,23,24 The intraband PL of CdSe nanocrystals as a function of laser excitation is then modeled with a direct intraband PL from interband excitation of ground state nanocrystals, as well as an intraband reexcitation of the excited nanocrystals, such that the intraband PL is proportional to (1 − n1)σo IRP + n1σ1 IRP where n1 is determined from the fit of the TA data. The fit in Figure 2b shows that the cross section for intraband PL arising from direct interband excitation, σo IR, is at least 10 times weaker than σ1 IR. These cross sections are products of absorption and emission cross sections. A small value of σo IR can arise from a small 1P exciton cross section, too blue compared to the laser, as well as weaker intraband PL due to the fast Auger relaxation

weak or undetectable intraband PL, even though the TA were similarly strong. This is consistent with previous reports of the effect of the ligands on the intraband relaxation.23 This is attributed to the larger vibrational absorption of these ligands in the 4−5 μm range.23 In addition, the thiol is a hole acceptor for CdSe21,26 and suppresses more effectively the electron−hole Auger relaxation channel. Power Dependence. The intraband emission could possibly occur from direct excitation of the 1Pe state or by a two-step process, where an interband absorption leads to occupation of 1Se, followed by a high energy intraband transition. Qualitative evidence for the two-step process is provided by the observation of intraband PL for the CdSe dots where 527 nm cannot directly excite the 1Pe state. Quantitative evidence is provided by the analysis of the power dependence of the TA and intraband PL. Figure 2 shows data for a film of dodecanethiol capped CdSe CQDs. In Figure 2a, the TA shows a fast increase with laser power followed by a saturation, while the PL in Figure 2b shows a slow rise followed by a linear response. As the pulse promotes electron to 1Se, the TA signal increases linearly at first. However, increasing the electron occupation in 1Se depends on the competition between excitation, Auger recombination and hole trapping, while the transient absorption is maximum at 2 electrons per dot. Therefore, the TA signal saturates at increasing power. For the intraband PL, the electrons need to be in the 1Pe state. Relaxation from 1Pe to 1Se is ultrafast in the presence of a hole, but if the dots is negatively charged, the emission from 1Pe to 1Se can be more efficient. This explains why the intraband PL 2227

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231

Article

ACS Nano

Figure 4. (a) TA (black) and PL (red) of films of CdS/ZnSe-dodecanethiol with different ZnSe shell thickness. (b) Evolution of the relative emission intensity of the CdS/ZnSe core−shell samples as a function of shell thickness, normalized by the absorbance at 527 nm. No intraband PL was detected for zero shell thickness. (c) PL of CdSe/ZnSe (red line) and CdSe/ZnTe (black line). (d) Emission of a CdSe/ ZnSe dodecanethiol film before (blue, intraband PL) and after annealing at 150 °C (red, thermal emission) under N2.

quenched while the long wave IR emission appears. Furthermore, as shown in Figure 3c, this emission is completely quenched at 80 K. Assuming that the sample temperature rises by the same temperature ΔT under the pulsed laser, for a constant heat capacity, the thermal radiation increase starting from 80 K is expected to be 200 times weaker than from 300 K due to the fourth power of the Stefan−Boltzmann law, and the disappearance of the long wave emission at low substrate temperature confirms its assignment to a thermal contribution. The long wavelength emission observed is therefore the ∼300 K blackbody modulated by laser heating and the emissivity of the films. An estimate of the laser heating and resulting thermal signal is provided in the Supporting Information. The increased IR emissivity of the samples under irradiation implies an increased IR absorption, which can arise from a photoinduced chemical modification. This was verified directly with FTIR spectra in Figure 3d showing that laser irradiation of a CdSethiol film in air leads to an increased absorption around 1100 cm−1, consistent with sulfonates, and that could account for the increased thermal emissivity. The formation of sulfonates by photooxidation of self-assembled monolayers of thiols on gold surfaces and in air is documented in the literature,30 and it is proposed that the photogenerated holes of CdS and CdSe facilitate thiol oxidation. The quenching of the intraband PL upon irradiation in air is therefore assigned to increased non radiative energy transfer to the new vibrations, noting that electron trapping/recombination is not strongly affected since the TA signal remains strong (Figure S3). Core/Shells. To reduce the coupling with surface ligands and help extract the hole, the type-II CdS/ZnSe, CdSe/ZnSe, and CdSe/ZnTe core/shell were synthesized. For CdS cores without shell, there is already a TA signal even though 527 nm is below the gap. This likely arises because the 527 nm laser can excite electrons from occupied surface states

by the hole created in the interband excitation. Conversely, σ1 IR can be larger because the hole is trapped, leading to more efficient intraband PL although the intraband absorption cross section from 527 nm should be small. Figure 2a,b shows that the model accounts well for the power dependence, confirming that the intraband PL arises from a two-step process. We note that the need to photoexcite the system and the use of a single color prevent a PL quantum yield measurement. Future experiments should focus on using sub-band gap excitation and doping to directly measure the intraband PL quantum yield. Preliminary experiments using a hole scavenger such as Li(Et)3BH,28 or electrochemical charging indicated that intraband PL could be improved using the same excitation wavelength of 527 nm (Figure S2). Photooxidation. While the PL is stable for samples under nitrogen, we observed that when the samples are in air, the laser excitation leads to an irreversible decrease of the intraband PL. Figure 3a shows the time evolution of the intraband PL spectrum under irradiation in air. The PL is quenched while a new signal appears at lower energy, down to the cutoff of the detector, around 900 cm−1. This long wavelength emission is not PL but rather thermal emission due to laser heating.29 One evidence for the thermal origin of the signal is the delay compared to the PL signal shown in Figure 3b. The temperature also affects differently the PL and thermal emission. Figure 3c shows the PL spectra of a dodecanethiol capped CdSe film kept under inert conditions at 300 K (red) to 80 K (blue). The intraband PL peak red-shifts by ∼10 meV and the fwhm decreases from 65 to 48 meV. The changes of the line shape are tentatively attributed to the electron−phonon coupling. The intensity increase at 80 K may be convoluted with the spectral shift decreasing the role of energy transfer to the CH vibrations, and changes of the photocharging kinetics. After the sample has been irradiated in air, the intraband PL is 2228

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231

Article

ACS Nano

nm under inert atmosphere and the intraband emission was successfully detected below ∼2900 cm−1, and above 1400 cm−1, windowed by the ligand CH vibrations. Intraband emission with dots capped by other ligands was weaker or not detectable. The analysis of the transient IR absorption and intraband PL as a function of pump power showed that the intraband PL arises mostly from intraband re-excitation of the CQDs, photocharged during the visible laser pulse, consistent with the need to reduce the electron−hole Auger relaxation. Upon photoexcitation in air, the samples show a quench of the intraband PL and the rise of thermal emission at longer wavelength. This was attributed to photooxidation of the thiol, giving rise to new vibrational absorption bands at ∼1100 cm−1. Attempts to improve the PL with inorganic ligands were not successful, but core/shell nanocrystals showed improved intraband PL. Finding appropriate matrices and processing to improve the infared intraband emission will be important for the future applications of CQDs in the mid-infrared.

to conduction states. However, the intraband PL is too weak to detect. By growing the ZnSe shell, we find that the indirect absorption at 527 nm increases strongly. Figure S4 shows the increased absorption and TEM images as a function of shell growth. Figure 4a shows that increasing the shell thickness leads to a red shift of the TA, as well as an increase of the PL intensity when the PL intensity is normalized by the laser absorption in Figure 4b. The increase of the intraband PL intensity is expected by the reduced coupling with the ligand vibrations through the shell, but the spectral shift and different photocharging kinetics will also play a role. Although the direct excitation of 1Pe is facilitated in these core/shells, the power dependence of TA and PL is similar to the CdSe cores previously described, therefore the intraband PL also arises from re-excitation of the quantum dots, and this prevents a more quantitative comparison of the different systems. CdSe/ZnSe and CdSe/ZnTe core/shell also exhibit the intraband PL when recapped with DDT as shown in Figure 4c. Among the core/shells studied, the CdSe/ZnSe gave the best signal-to-noise for samples emitting at 1900 cm−1. However, quantitative comparisons of the signal strengths were not pursued given the two-step nature of the excitation. Although the shells allow to retain good TA signals in films that are annealed at moderate temperature, they did not protect the intraband PL against photooxidation in air. Figure 4d shows that the intraband PL of dodecanethiol capped CdSe/ZnSe CQD is also completely replaced by a much stronger thermal emission at ∼1100 cm−1 after a moderate 150 °C anneal in a glovebox with less that 10 ppm of O2. This is so even as the intraband TA is not significantly modified. The disappearance of the PL after such annealing in inert conditions indicates the creation of nonradiative channels, and the increase of the IR emissivity suggests that they are due to the creation of strong IR optical modes, still possibly oxides created by imperfect exchange of the stearic acid to thiols. Inorganic Ligands. Since a major cause of the nonradiative relaxation is assigned to the energy transfer to ligands vibrations, it is natural to explore ligands and matrices with low vibrational frequencies. Much effort was therefore spent on trying various inorganic ligands and matrices following various literature reports.31,32 Inorganic ligands such as (NH4)2S, KH2PO3, Na2S2O3, As2S3, and NaBH2S3 were tested, but none of the CQDs with inorganic ligands exhibited intraband PL. In all cases, as shown in Figure S7, thermal emission similar to Figure 4d, dominated the emission even though the samples were processed in the glovebox and kept under N2 during the measurements. There is a good match between the vibrational absorption of the films and the IR emission (Figure S8). It is possible that the inorganic ligands exchanges were defective, leading to additional decay channels, either through electronic relaxation via subgap density of states or through coupling to surface vibrations, and it is still expected that an improved synthetic protocol could be eventually successful.

METHODS Transient Infrared Absorption Spectrometer. For the TA measurement, the light from a globar is focused on the sample, at the same spot as the 527 nm laser, and after recollimation, it passes through a step-scan interferometer and is detected by a liquid nitrogen cooled photovoltaic HgCdTe detector (MCT) with a ∼20 MHz frequency response. The 527 nm laser (Empower, Spectra-Physics) provides ∼250 ns pulses at a repetition rate of 1 kHz. The transient changes of the MCT signal are integrated during the pulse width with a gated integrator, and accumulated over a number of pulses, providing the pump component of the signal as a function of interferometer mirror position. The dc signal of the MCT is measured as well. The Fourier transform of the pump interferogram (ΔT) is then divided by the Fourier transform of the dc component (T) to produce the transient IR spectrum (ΔT/T). Intraband Photoluminescence Measurement. The infrared emission is recorded with the same instrument as the TA by simply turning off the globar, under the same excitation conditions. The IR spectrum is corrected for the instrument spectral response, obtained by comparing a calculated blackbody spectrum at the globar temperature and the measured globar spectrum. For intensity dependent measurements, the Gaussian pulse beam size at the sample is measured and the average power is measured with a power meter. Synthesis. CdSe. CdSe CQD (wurtzite) was synthesized by injecting a Se solution comprising of octadecene (ODE, 1 mL), Se (powder, 70 mg) and trioctylphosphine (TOP, 0.57 mL) into a mixture of cadmium stearate solution, trioctylphosphineoxide (TOPO, 0.5 g) and octadecylamine (ODA, 2.5 g) at 280 °C under an argon atmosphere. The cadmium stearate solution was prepared by heating the mixture of cadmium oxide (25.6 mg), ODE (2 mL) and strearic acid (225 mg) at 200 °C for 30 min under an argon atmosphere. The reaction time is from 30 s to 2 min depending upon the desired size of the CdSe colloidal quantum dot. To increase the size of the CdSe quantum dots, 0.1 M ODE solution of cadmium oleate and 0.1 M SeODE solution were alternatively added into the 3 mL of solution of CdSe CQD, at 230 °C under argon atmosphere. Four drops (∼80 μL) of each solution were added into the CdSe CQD solution every minute to avoid independent nucleation of CdSe nanocrystals. CdSe/ZnS. For the ZnS shell capping, successive ionic layer adsorption and reaction (SILAR) were performed by addition of 0.1 M octadecene solution of zinc stereate and 0.1 M octadecene solution of sulfur. Reaction time of each step is 10 min at 230 °C. The addition was done at room temperature. CdSe/ZnSe. For the ZnSe shell capping, similar to that of ZnS shell capping, SILAR method was performed. Briefly, 0.1 M octadecene solution of zinc stereate and 0.1 M octadecene solution of selenium were added alternatively. Reaction time of each step is 10 min at 230 °C.

CONCLUSION In conclusion, this is a first study of the potential for midinfrared photoluminescence from cadmium chalcogenides colloidal quantum dots. Although these materials have wide band gaps, the intraband transitions provides a novel possibility for infrared emission, if the intraband relaxation is long enough. Dodecanethiol capped CdSe, CdSe/ZnS, CdSe/ZnSe and CdS/ZnSe, CdSe/ZnTe colloidal quantum dots in films and solutions were illuminated with pulsed photoexcitation at 527 2229

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231

Article

ACS Nano CdS. Synthesis of CdS CQD is identical to that of CdSe CQD, but the ODE solution of sulfur was used instead of the ODE solution of selenium. CdS/ZnSe. For the CdS/ZnSe core−shell synthesis, the synthetic procedure is identical to that of CdSe/ZnSe except the core is CdS instead of CdSe. Sample Preparation. Inorganic Ligand Exchange. Inorganic ligand exchange were performed inside the glovebox. All the inorganic ligand solutions were prepared in the glovebox. The polar inorganic ligand solution was mixed with organic ligands capped colloidal quantum dot solution. During the inorganic ligand exchange under stirring, the colloidal quantum dots transfer from the nonpolar to the polar solvent as the nonpolar ligands are replaced by the polar inorganic ligands. The reaction time depends on the inorganic ligands. NaBH2S3: Sulfurated sodium borohydride was synthesized by reaction between sulfur power and sodium borohydride in tetrahydrofuran. One millimole of sodium borohydride and 3 mmol of sulfur powder were added into 5 mL of tetrahydrofuran in 250 mL of round-bottom flask immersed in cold water. The NaBH2S3 THF solution was mixed with CdSe CQDs in chloroform for ligand exchange. (NH4)2S: 100 μL of (NH4)2S (40−44%, water) was added into 10 mL of methanol solvent. The (NH4)2S methanol solution was mixed with the colloidal quantum dot in hexane and the mixture was stirred for 10 min. The nonpolar layer was decanted and the inorganic ligand passivated CQDs were cleaned once or twice more. The inorganic ligand passivated CQD was taken by using a syringe and dropped on an infrared transparent window. Na2S2O3: 3.16 g (20 mmol) of sodiumthiosulfate was dissolved in 10 mL of methanol. The sodiumthiosulfate methanol solution was mixed with CdSe/ZnS CQDs in chloroform. The mixture was centrifuged for 1−2 min and redissolved in chloroform. KH2PO3: 2.70 g (20 mmol) of potassium phosphate was dissolved in 10 mL of methanol. The CdSe/ZnS CQDs in chloroform was added into the potassium phosphate methanol solution. The mixture was centrifuged for 1−2 min and the precipitate was redissolved in chloroform. As2S3: 2.0 g (8 mmol) of arsenic sulfide was dissolved in 1.4 mL of propylamine with a 20:1 mass ratio. The arsenic sulfide propylamine solution was stirred for 2 days inside glovebox. The solution was filtered by using 0.45 μm pore syringe filter before use.

(2) Faist, J.; Capasso, F.; Sivco, D. L.; Sirtori, C.; Hutchinson, A. L.; Cho, A. Y. Quantum Cascade Laser. Science 1994, 264, 553−556. (3) Jeong, K. S.; Deng, Z. Y.; Keuleyan, S.; Liu, H.; Guyot-Sionnest, P. Air-Stable n-Doped Colloidal HgS Quantum Dots. J. Phys. Chem. Lett. 2014, 5, 1139−1143. (4) Deng, Z.; Jeong, K. S.; Guyot-Sionnest, P. Colloidal Quantum Dots Intraband Photodetectors. ACS Nano 2014, 8, 11707−11714. (5) Nozik, A. J. Quantum Dot Solar Cells. Phys. E 2002, 14, 115− 120. (6) Yu, S.; Kim, Y. H.; Lee, S. Y.; Song, H. D.; Yi, J. Hot-ElectronTransfer Enhancement for the Efficient Energy Conversion of Visible Light. Angew. Chem. 2014, 126, 11385−11389. (7) Chen, H. M.; Chen, C. K.; Chen, C.-J.; Cheng, L.-C.; Wu, P. C.; Cheng, B. H.; Ho, Y. Z.; Tseng, M. L.; Hsu, Y.-Y.; Chan, T.-S.; Lee, J.F.; Liu, R.-S.; Tsai, D. P. Plasmon Inducing Effects for Enhanced Photoelectrochemical Water Splitting: X-ray Absorption Approach to Electronic Structures. ACS Nano 2012, 6, 7362−7372. (8) DuChene, J. S.; Sweeny, B. C.; Johnston-Peck, A. C.; Su, D.; Stach, E. A.; Wei, W. D. Prolonged Hot Electron Dynamics in Plasmonic-Metal/Semiconductor Heterostructures with Implications for Solar Photocatalysis. Angew. Chem., Int. Ed. 2014, 53, 7887−7891. (9) Kang, Y.; Gong, Y.; Hu, Z.; Li, Z.; Qiu, Z.; Zhu, X.; Ajayan, P. M.; Fang, Z. Plasmonic Hot Electron Enhanced MoS2 Photocatalysis in Hydrogen Evolution. Nanoscale 2015, 7, 4482−4488. (10) Warren, S. C.; Thimsen, E. Plasmonic Solar Water Splitting. Energy Environ. Sci. 2012, 5, 5133. (11) Ingram, D. B.; Linic, S. Water Splitting on Composite Plasmonic-Metal/Semiconductor Photoelectrodes: Evidence for Selective Plasmon-Induced Formation of Charge Carriers near the Semiconductor Surface. J. Am. Chem. Soc. 2011, 133, 5202−5205. (12) Mubeen, S.; Lee, J.; Singh, N.; Kramer, S.; Stucky, G. D.; Moskovits, M. An Autonomous Photosynthetic Device in which All Charge Carriers Derive from Surface Plasmons. Nat. Nanotechnol. 2013, 8, 247−251. (13) Sil, D.; Gilroy, K. D.; Niaux, A.; Boulesbaa, A.; Neretina, S.; Borguet, E. Seeing Is Believing: Hot Electron Based Gold Nanoplasmonic Optical Hydrogen Sensor. ACS Nano 2014, 8, 7755−7762. (14) Atar, F. B.; Battal, E.; Aygun, L. E.; Daglar, B.; Bayindir, M.; Okyay, A. K. Plasmonically Enhanced Hot Electron Based Photovoltaic Device. Opt. Express 2013, 21, 7196−7201. (15) Mubeen, S.; Hernandez-Sosa, G.; Moses, D.; Lee, J.; Moskovits, M. Nano Lett. 2011, 11, 5548. (16) Wang, F. M.; Melosh, N. A. Plasmonic Energy Collection through Hot Carrier Extraction. Nano Lett. 2011, 11, 5426−5430. (17) Dong, Y.; Choi, J.; Jeong, H.; Son, D. H. Hot Electrons Generated from Doped Quantum Dots via Upconversion of Excitons to Hot Charge Carriers for Enhanced Photocatalysis. J. Am. Chem. Soc. 2015, 137, 5549−5554. (18) Murdoch, M.; Waterhouse, G. I. N.; Nadeem, M. A.; Metson, J. B.; Keane, M. A.; Howe, R. F.; Llorca, J.; Idriss, H. The Effect of Gold Loading and Particle Size on Photocatalytic Hydrogen Production from Ethanol over Au/TiO2 Nanoparticles. Nat. Chem. 2011, 3, 489− 492. (19) Klimov, V. I.; McBranch, D. W. Femtosecond 1P-to-1S Electron Relaxation in Strongly Confined Semiconductor Nanocrystals. Phys. Rev. Lett. 1998, 80, 4028−4031. (20) Efros, A. L.; Kharchenko, V. A.; Rosen, M. Breaking The Phonon Bottleneck in Nanometer Quantum Dots - Role of Auger-Like Processes. Solid State Commun. 1995, 93, 281−284. (21) Guyot-Sionnest, P.; Shim, M.; Matranga, C.; Hines, M. Intraband relaxation in CdSe Quantum Dots. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 60, R2181−R2184. (22) Klimov, V. I.; Mikhailovsky, A. A.; McBranch, D. W.; Leatherdale, C. A.; Bawendi, M. G. Mechanisms for Intraband Energy Relaxation in Semiconductor Quantum Dots: The Role of Electronhole Interactions. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, R13349−R13352.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.5b06882. Details of CQD synthesis, sample preparation, TEM analysis, TA, mid-IR PL emission measurement; estimate of laser heating and thermal emission (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Present Address †

(K.S.J.) Department of Chemistry, Korea University, Seoul 136-701, Korea. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported by NSF-DMR 1104755. REFERENCES (1) Levine, B. F. Quantum-Well Infrared Photodetectors. J. Appl. Phys. 1993, 74, R1−R81. 2230

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231

Article

ACS Nano (23) Guyot-Sionnest, P.; Wehrenberg, B.; Yu, D. Intraband Relaxation in CdSe Nanocrystals and the Strong Influence of the Surface Ligands. J. Chem. Phys. 2005, 123, 074709. (24) Pandey, A.; Guyot-Sionnest, P. Slow Electron Cooling in Colloidal Quantum Dots. Science 2008, 322, 929−932. (25) Pandey, A.; Guyot-Sionnest, P. Intraband Spectroscopy and Band Offsets of Colloidal II-VI Core/Shell Structures. J. Chem. Phys. 2007, 127, 104710. (26) Wuister, S. F.; Donegá, C. M.; Meijerink, A. Influence of Thiol Capping on the Exciton Luminescence and Decay Kinetics of CdTe and CdSe Quantum Dots. J. Phys. Chem. B 2004, 108, 17393−17397. (27) Leatherdale, C. A.; Woo, W. K.; Mikulec, F. V.; Bawendi, M. G. On the Absorption Cross Section of CdSe Nanocrystal Quantum Dots. J. Phys. Chem. B 2002, 106, 7619−7622. (28) Rinehart, J. D.; Schimpf, A. M.; Weaver, A. L.; Cohn, A. W.; Gamelin, D.l R. Photochemical electronic doping of colloidal CdSe nanocrystals. J. Am. Chem. Soc. 2013, 135, 18782−18785. (29) Sean Keuleyan, Mid-Infrared Mercury Chalcogenides Colloidal Quantum Dots, Ph.D. Thesis, The University of Chicago, 2013. (30) Huang, J. Y.; Dahlgren, D. A.; Hemminger, J. C. Photopatterning of Self-Assembled Alkanethiolate Monolayers on Gold - A Simple Monolayer Photoresist Utilizing Aqueous Chemistry. Langmuir 1994, 10, 626−628. (31) Kovalenko, M. V.; Scheele, M.; Talapin, D. V. Colloidal Nanocrystals with Molecular Metal Chalcogenide Surface Ligands. Science 2009, 324, 1417−1420. (32) Nag, A.; Kovalenko, M. V.; Lee, J. S.; Liu, W. Y.; Spokoyny, B.; Talapin, D. V. Metal-free Inorganic Ligands for Colloidal Nanocrystals: S2-, HS-, Se2-, HSe-, Te2-, HTe-, TeS32-, OH-, and NH2- as Surface Ligands. J. Am. Chem. Soc. 2011, 133, 10612−10620.

2231

DOI: 10.1021/acsnano.5b06882 ACS Nano 2016, 10, 2225−2231