Control of Electronic Structures and Phonon Dynamics in Quantum

Jul 6, 2016 - Quantum dot (QD) superlattices, periodically ordered array structures of QDs, are expected to provide novel photo-optical functions due ...
3 downloads 13 Views 3MB Size
Download PDF
Research Article www.acsami.org

Control of Electronic Structures and Phonon Dynamics in Quantum Dot Superlattices by Manipulation of Interior Nanospace I-Ya Chang,†,‡ DaeGwi Kim,¶ and Kim Hyeon-Deuk*,†,‡ †

Department of Chemistry, Kyoto University, Kyoto 606-8502, Japan PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan ¶ Department of Applied Physics, Osaka City University, Osaka 558-8585, Japan ‡

S Supporting Information *

ABSTRACT: Quantum dot (QD) superlattices, periodically ordered array structures of QDs, are expected to provide novel photo-optical functions due to their resonant couplings between adjacent QDs. Here, we computationally demonstrated that electronic structures and phonon dynamics of a QD superlattice can be effectively and selectively controlled by manipulating its interior nanospace, where quantum resonance between neighboring QDs appears, rather than by changing component QD size, shape, compositions, etc. A simple Hpassivated Si QD was examined to constitute one-, two-, and three-dimensional QD superlattices, and thermally fluctuating band energies and phonon modes were simulated by finite-temperature ab initio molecular dynamics (MD) simulations. The QD superlattice exhibited a decrease in the band gap energy enhanced by thermal modulations and also exhibited selective extraction of charge carriers out of the component QD, indicating its advantage as a promising platform for implementation in solar cells. Our dynamical phonon analyses based on the ab initio MD simulations revealed that THz-frequency phonon modes were created by an inter-QD crystalline lattice formed in the QD superlattice, which can contribute to low energy thermoelectric conversion and will be useful for direct observation of the dimension-dependent superlattice. Further, we found that crystalline and ligand-originated phonon modes inside each component QD can be independently controlled by asymmetry of the superlattice and by restriction of the interior nanospace, respectively. Taking into account the thermal effects at the finite temperature, we proposed guiding principles for designing efficient and space-saving QD superlattices to develop functional photovoltaic and thermoelectric devices. KEYWORDS: quantum dot superlattice, carrier extraction, thermoelectric conversion, THz phonons, quantum resonance, thermal effects, time-dependent density functional theory

■

solution cases, where each QD was almost isolated.22 The red shift of the absorption peak will accelerate multiple exciton generations by reducing the absolute threshold energy.3,23−26 An intermediate band, formed within the band gap of the colloidal QD superlattice, was also proposed for efficient upconversion processes by intermediate-band mediated absorptions.27 Even carrier dynamics such as energy transfer and photoconductivity were observed on the closely packed QD assemblies.28−30 While QDs can be made of a variety of chemical compositions, silicon has advantages for photovoltaic and thermoelectric applications. Silicon is a photostable, nontoxic, abundant, and technologically well-established material. In fact, silicon QDs have been examined in onedimensional (1D) crystalline and amorphous nanowires,24,31 in

INTRODUCTION

Since the early 1980s, much progress has been made in elucidating physical and chemical properties of isolated semiconductor quantum dots (QDs) in colloidal solutions.1−3 After it was understood that colloidal QDs can form not only random arrays but also perfectly organized assemblies,4−6 considerable attention began to be paid to understand changes of the properties in hyperstructure QD materials.5,7−12 Twodimensional (2D) and three-dimensional (3D) QD superlattices have attracted the expectation that they can realize novel optoelectronic and photovoltaic functions based on resonant couplings between adjacent QDs.13−19 QD superlattices fill a large fraction of a solar cell absorber because of their assembly, making the absorbing volume fraction substantially higher than that of isolated QDs.7 The absorption cross section was enhanced 4−5 times due to the dipole− dipole couplings between neighboring QDs,20,21 and the absorption peaks of the closely packed 2D QD films were shifted toward lower energies compared with those of the dilute © XXXX American Chemical Society

Received: March 16, 2016 Accepted: June 27, 2016

A

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces the energy transfer between two QDs with dopants and adsorbates,32 and in luminescent QD aggregates.33 The resonant couplings between adjoining QDs are classified into two typical interactions: the long-range dipole−dipole Coulomb coupling and short-range quantum resonance.21,28−30,34 In most of the experimentally prepared superlattices of face-centered cubic or body-centered cubic symmetry, spaces between adjacent QDs and component QD size were relatively large, and thus the long-range dipole−dipole couplings inversely proportional to the sixth power of the separation distance were mainly observed or assumed.9,12,20,21,28,29,35,36 Further, because most of the previous QD superlattices included partial areas of 2D and 3D assemblies and inevitably involved various kinds of the resonant couplings, it has been difficult to identify the effects of the pure quantum resonance.8,30,34 Precise and systematic manipulation of the nanospace in QD superlattices is indispensable for controlling the resonant coupling and for purely inducing the short-range quantum resonance.30,34 Recently, we successfully prepared the 3D CdTe QD superlattice deposited by a layer-by-layer assembly of positively charged polyelectrolytes and negatively charged CdTe QDs.34,37 Nanospace between the homogeneous QD monolayers was precisely manipulated by the spacer-layer thickness with subnanometer accuracy, demonstrating the existence of the quantum resonance which exponentially depended on the nanospace in the rectangular cubic QD superlattice. In this article, we computationally demonstrate that electronic structures and phonon dynamics of QD superlattices can be effectively and selectively controlled by their interior nanospace to induce the quantum resonance rather than by changing component QD size, shape, compositions, etc. Accounting for thermal fluctuations at finite temperature, we will propose guiding principles for designing functional nanospace in QD superlattices which can be adjusted by ligand attaching down to the subnanometer scale.3,7,10,34,38 The possibility of precisely manipulating nanospace in QD superlattices opens new opportunities for developing advanced photovoltaic and thermoelectric devices.

■

Figure 1. (a) 3D Si29H24 superlattice with nanospace (A, B). The 3D superlattice became a 2D sheet with a small A and sufficiently large B, while it became comprised of 1D beads with a small B and substantially large A. (b) Band-gap energy shifts of the Si29H24 superlattice as a function of nanospace (A, B). The quantum resonance caused in the superlattice significantly decreased the band gap energy Eg. (c) LUMO and (d) HOMO in the Si29H24 superlattice with the tabularly indicated nanospace (A, B). The LUMO was substantially delocalized to adjoining QDs with the small nanospace, while the HOMO was always localized to Si29H24 itself at each lattice point. density functional and the projector-augmented-wave (PAW) pseudopotentials were used. The cutoff for the plan-wave basis set was 312.5 eV for the Si29H24 superlattice and the Γ point was adopted for the k-point sampling. For the validation of our QD superlattice model, a 3D superlattice composed of a Si29H24 QD dimer was also examined; each simulation cell contains two Si29H24 QDs between which the nearest-surface distance was set as A. All computations of the QD superlattice were executed without any geometry constraints. The QD superlattice optimized at zero temperature was finally brought up to ambient temperature by repeated velocity rescaling. Microcanonical trajectories of 5 ps on the ground electronic state were computed for the Si29H24 QD superlattice with various nanospace (A, B). The Verlet algorithm with a 1 fs time step and the cutoff for the plane-wave basis of 200 eV were used. The average temperature was almost 300 K, and stable temperature fluctuations were achieved (Figure S2). We found that the temperature fluctuations were suppressed with the smallest nanospace; the deviations were 21.7 K for (A, B) = (0.1 nm, 0.1 nm) and 26.3 K for (A, B) = (0.8 nm, 0.8 nm). All nuclei in the Si29H24 QD with the largest nanospace can move more freely owing to the absence of the additional inter-QD interactions, while the nuclear dynamics of the QD were restricted with the small nanospace as will be discussed later; temperature fluctuations in the superlattice can be controlled by changing the nanospace (A, B).

COMPUTATIONAL METHODS AND MATERIAL DESIGN

We will focus on a QD superlattice composed of H-passivated Si QDs to discuss novel properties appearing in the QD superlattice, to investigate a role of dimensionality in 1D, 2D, and 3D superlattices, and also to reveal finite temperature effects, such as thermally fluctuating band energies, and the influence of phonon dynamics on the functional properties of the QD superlattice. A primitive structure of isolated Si29 was cut off from a bulk Si crystal, and the outer-layer Si atoms were passivated with H atoms. The Si29H24 QD with a 1.1 nm diameter was then fully optimized in the PBE/def2-SVP level of theory at zero temperature in vacuum by the Gaussian09 package.39 The QD superlattice was made by periodically replicating the cubic cell in the xyz directions. Figure 1a shows 3D QD superlattice in which the Si29H24 QD is located at each cubic lattice point. A was defined as nanospace between two closest H atoms on the nearest-neighbor QD surfaces in each 2D xy-layer, while B was introduced as nanospace between two closest H atoms of the nearest-neighbor 2D layers along the z-axis. The 3D superlattice geometry was thus characterized by the two nanospaces (A, B) as graphically indicated in Figure 1a. The optimized QD structures in the Si29H24 superlattice are listed in Figure S1. All calculations related to the QD superlattice were performed with Vienna Ab-initio Simulation Package (VASP)40 using a converged plane-wave basis in the simulation cell built above. The generalized gradient approximation of the Perdew, Burke, and Ernzerhof (PBE)

■

RESULTS AND DISCUSSIONS Band-gap energy shifts of the Si29H24 superlattice at 0 K are shown in Figure 1b as a function of nanospace (A, B). The quantum resonance caused in the superlattice significantly decreased the band gap energy Eg. Eg approached its original value, 2.6 eV, for the isolated Si29H24 QD with increasing nanospace (A, B). Because each QD was symmetrically compressed due to the interaction from adjoining QDs, the B

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 2. (a) Band-gap energy shifts of the Si29H24 superlattice at 300 K. The band gap energy Eg at 300 K decreased compared to Eg at 0 K. Heating to 300 K also induced the larger Eg shift in the case of the small nanospace. (b) HOMO and (c) LUMO at 300 K. The HOMO was sometimes connected over adjacent QDs due to Si−H stretching modes, while the LUMO was always substantially delocalized to neighboring QDs even at 300 K.

converged values of (A, B) slightly differed from the initially input (A, B), especially in the cases of small nanospace (Figure S3). All nanospace values (A, B) at 0 K shown in this article were obtained by rounding off the second decimal place of the optimized real (A, B) values, while the nanospace (A, B) for 300 K, which will be displayed later, is the initially input (A, B). The threshold nanospace (A, B) with which Eg saturated was smaller than the corresponding threshold nanospace obtained for the Cd26Te26 QD superlattice;34 each Cd26Te26 QD needed a larger nanospace to be free from the influence of adjacent QDs. The Cd26Te26 QD is softer than the Si29H24 QD, and the former is deformed more than the latter with the same nanospace. Figure 1c demonstrates that the Eg shifts originated from the quantum resonance of the LUMO between the neighboring QDs. The LUMO was notably delocalized over the inter-QD nanospace with the small (A, B), while the LUMO became localized to each Si29H24 with the larger (A, B). The experimental observation that the absorption peak shift of the QD superlattice became larger with a decrease in the QD diameter supports this quantum resonance picture because penetration of a wave function outside a QD can be enhanced by a larger ratio of a QD surface over volume.34 In the 2D sheet with small A and sufficiently large B, the LUMO was delocalized over the 2D xy-layer, as shown in the bottom left panel. The 2D sheet with A = 0.1 nm exhibited much smaller Eg over the examined B, and the Eg shift along B was quite small, indicating that the 2D sheet already achieved the sufficient quantum resonance without the quantum resonance along the z-axis. In the 1D beads with small B and sufficiently large A, the LUMO was delocalized along the 1D beads, as shown in the upper right panel. The large A in the 1D beads led to the wider LUMO delocalization along the z-axis compared to the that of the 3D superlattice case, displayed in the upper left panel; the LUMO tended to compensate the lack of quantum resonance over the xy-plane by being substantially delocalized along the zaxis with small B. This tendency rationalized the results in Figure 1b that Eg changed more significantly with larger A, and that the threshold B with which Eg saturated elongated with increasing A. Interestingly, the change of the HOMO depending on (A, B) was not as substantial as that of the LUMO, as Figure 1d indicates; the HOMO was always localized to Si29H24 itself located at each lattice point, even in the 3D superlattice with the small (A, B), and did not contribute to the Eg shift due to the absence of the quantum resonance. This qualitatively different behavior of the LUMO and HOMO stemmed from

the larger electronegativity of passivated H atoms than that of the core Si atoms. Electrons can be effectively and separately extracted from the 3D Si29H24 superlattice because of the delocalized LUMO and localized HOMO. In other words, it is possible to control charge separation by preparing surface ligands of different electronegativity from the core regions with suitable nanospace. It should be noted here that almost the same results were reproduced even when we performed the same calculations for superlattices with the QD dimer unit cell, validating the current periodic unit cell of the single QD for the superlattice (Figures S4, S5, S6, and S7) To discuss finite temperature effects, Figure 2a compares the band gap energy Eg at 0 and 300 K as a function of the nanospace (A, B). Eg decreased with the smaller nanospace (A, B) regardless of the temperature. Eg at 300 K was smaller than Eg at 0 K over all the nanospace (A, B) because the thermal modulations and expansions of the QD reduced the band gap energy; it was reported that Eg was inversely proportional to temperature in the isolated Si29H24 case, i.e., with (A, B) = (0.8 nm, 0.8 nm).23 What we found here is that heating to 300 K induced the larger Eg shift because the thermal modulation enhanced the quantum resonance by structural fluctuations of the Si29H24 QD, such as Si−H stretching. As reflected in the larger decrease of Eg at 300 K with (A, B) = (0.1 nm, 0.1 nm) than the decrease at 300 K with (A, B) = (0.8 nm, 0.8 nm), such an enhancement of the quantum resonance was more apparent with the small nanospace because the effect of the structural fluctuations was relatively larger with the smaller nanospace. As evidenced in Figure 2b, there were some instances when the HOMO, which was always localized at 0 K regardless of the nanospace, was delocalized over neighboring QDs with the small nanospace due to the Si−H stretching modes. It should be noted, however, that the delocalization of the LUMO at 300 K shown in Figure 2c was still more significant and thus more important for the Eg shift than the intermittent delocalization of the HOMO. All structural and orbital information on the Si29H24 superlattice at 300 K is given in Figure S8. Figure 3 compares the (a) valence band (VB) and (b) conduction band (CB) density of state (DOS) of the Si29H24 superlattice at 0 and 300 K for various nanospace (A, B). The DOS at 300 K was obtained from the corresponding timeaveraged data of the band energy along the 5 ps microcanonical trajectories at 300 K. The smaller nanospace in the Si29H24 superlattice induced the smoother VB DOS than that in the isolated Si29H24 case with (A, B) = (0.8 nm, 0.8 nm). This tendency was kept even at 300 K, as shown in the blue and pink C

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

superlattice with the small nanospace (A, B) = (0.1 nm, 0.1 nm) and (A, B) = (0.2 nm, 0.1 nm) exhibited the phonon peaks in the THz region, as the red lines in Figures4b and c show. The phonon spectra of the Si29H24 superlattice with the larger nanospace did not have any peaks because of the weaker interaction between the nearest QDs. Note that the COM distance mixed all of the x, y, and z coordinates of the COM and made the phonon peaks insensitive and absent in the 1D and 2D superlattices. The anisotropic phonon spectra shown in Figure S9, which were obtained by Fourier transforming timedependent x, y, and z coordinates of the COM, exhibited broad shoulders from 40 cm−1 through 100 cm−1, not only when the symmetric 3D Si29H24 superlattice formed but also when the asymmetric 1D beads and 2D sheet were well-prepared. Phonon dynamics are closely related to an electron−phonon coupling and thus essentially influence photoexcited dynamics.23 Figures 5 and 6 display low- and high-frequency phonon

Figure 3. (a) VB DOS and (b) CB DOS of the Si29H24 superlattice for different nanospace and temperatures. The VB DOS became smoother at the higher temperature and also with the smaller nanospace. The CB DOS with the small nanospace had an island at the band edge regardless of the temperature, while the CB DOS at 300 K with the large nanospace became more continuous.

lines, although the VB DOS became smoother due to the thermal modulations at 300 K. The Si29H24 deformation as well as the orbital delocalization in the smaller nanospace made the VB DOS of the Si29H24 superlattice close to the continuous DOS of a bulk crystal. The CB DOS also exhibited the difference: the CB DOS with (A, B) = (0.1 nm, 0.1 nm) had an island around the CB edge both at 0 and 300 K, while the CB DOS with (A, B) = (0.8 nm, 0.8 nm) became continuous at 300 K (drawn by the pink line) because the original CB DOS at 0 K had the evenly placed three peaks shown by the green line, which were broadened by the thermal modulations at 300 K. The present Si29H24 superlattice should be a new kind of crystal as long as the QD superlattice formed a periodic lattice at which each QD oscillated. Figure 4a provides a sketch of the Figure 5. Phonon spectra for low-frequency Si−Si crystalline phonon modes. The Si−Si phonon modes were suppressed due to only asymmetric deformation of the Si29H24 QD in the asymmetric 1D beads and 2D sheet.

modes corresponding to Si−Si crystalline and Si−H ligand vibrations inside Si29H24 of the superlattice with various nanospace (A, B), respectively. All examined nanospace is listed in Figures S10 and S11. The phonon intensity was obtained by Fourier transforming time-dependent energy Figure 4. (a) Schematic origin of THz phonon modes caused by interactions between neighboring Si29H24 QDs in the superlattice. (b− d) Phonon spectra of the COM of each Si29H24 QD in the superlattice. The THz phonon modes were evidenced by the appearance of the spectral peaks only with the small nanospace (A, B).

periodic lattice whose potential was formed by the short-range interaction between the adjacent Si29H24 QDs. Figures 4b−d show phonon spectra of the center of mass (COM) of Si29H24 in the superlattice by Fourier transforming time-dependent COM distance in the space-fixed flame at 300 K. All power spectra were scaled to 1 at 13.3 cm−1 in each panel to clearly compare each case. These COM phonon spectra demonstrated that the Si29H24 superlattice formed the THz lattice phonon modes by the inter-QD interactions through its interior nanospace. The current inter-QD interactions are not a longrange dipole−dipole coupling but are the short-range quantum resonance as discussed above. Therefore, only the Si29H24

Figure 6. Phonon spectra for high-frequency Si−H ligand phonon modes. The Si−H phonon modes were simply more suppressed, as the interior nanospace was more restricted. D

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces fluctuations of the band states indicated at the lower right part of Figures 5 and 6. The Si−Si crystalline phonon peaks appeared between 200 and 400 cm−1 and around 600 cm−1 in all cases. The Si−Si phonon modes of the isolated Si29H24 case with (A, B) = (0.8 nm, 0.8 nm) exhibited strong peak intensity because they could freely oscillate without any inter-QD interactions. It is interesting that the Si−Si phonon modes of the symmetric 3D Si29H24 superlattice with the small nanospace (A, B) = (0.1 nm, 0.1 nm) also had larger peak intensity as well as the isolated Si29H24 case, while the Si−Si phonon intensity of the asymmetric 1D beads and 2D sheet became weaker than those of the above two symmetric cases, as can be seen in Figure 5 and Figure S10. The suppressed Si−Si phonon intensity in the asymmetric 1D and 2D superlattices reflected that the low-frequency Si−Si phonons partially collapsed and became weakened by the asymmetric distortion of Si29H24 caused by the asymmetric nanospace (A, B). We propose that crystalline phonon modes inside each QD in QD superlattices can be controlled by symmetricity of the nanospace in superlattices. In Figure 6, the high-frequency phonon modes corresponding to the Si−H ligand vibrations had the largest intensities with the nanospace (A, B) = (0.8 nm, 0.8 nm), reflecting that the Si−H surface ligands can oscillate more freely in the isolated case owing to the sufficiently large nanospace. The phonon spectra had two clear peaks: the lower-frequency peak originated from Si−H bonds whose Si attached to three Si atoms and had no dangling bond, and the other higherfrequency peak stemmed from Si−H bonds whose Si bonded to only two Si atoms and included a dangling bond. In the other cases where at least one of the nanospaces A or B is substantially small, the Si−H vibrations were partially suppressed due to the nonzero interaction from the neighboring QDs (Figure S11). The phonon intensity was actually smallest with the most restricted nanospace, (A, B) = (0.1 nm, 0.1 nm). Figures 5 and 6 suggest that we can independently and separately control crystalline and ligand-originated phonon modes by operating the interior nanospace of QD superlattices.

direct observation of the asymmetric 1D and 2D superlattices as well as the symmetric 3D superlattice and will contribute to developments of a powerful material for thermoelectric energy conversion in the THz frequency region.35 Similar kinds of lowfrequency phonon modes will appear in other QD superlattices as long as the crystalline lattice is well-formed by its interior nanospace, although the phonon frequency depends on the inter-QD interaction. The crystalline and ligand-originated phonon modes inside each component QD also highly depend on materials of component QDs. The independent and separate controls of the low-frequency crystalline phonon modes and the high-frequency ligand-originated phonon modes by the asymmetric and restricted nanospace will lead to a new way to control photoexcited dynamics, which is strongly influenced by electron−phonon couplings. Sparse nanospace between component QDs as well as phonon dispersion at QD interfaces can selectively suppress thermal energy transfer by phonons, whose mean free paths are longer than that of the interior nanospace and thus provide a new option to control thermal conductivity. All of the guiding principles we suggested here will contribute to designing efficient and space-saving photovoltaic and thermoelectric devices using QD superlattices.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b03219. Optimized QD structures, temperature fluctuations, LUMOs, HOMOs, reproduction of the same results by a QD superlattice with a QD dimer unit cell, anisotropic THz phonon spectra, and additional crystalline and ligand phonon spectra (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail:[email protected].

■

Notes

The authors declare no competing financial interest.

■

CONCLUSIONS In summary, we computationally demonstrated that the electronic structures and phonon dynamics of the QD superlattice can be effectively and separately controlled simply by manipulating its interior nanospace. The decrease in the band gap energy and the selective extraction of electrons, which strongly depended on the dimensionality of the Si29H24 superlattice, indicated that nonspherical and nonsymmetric QD superlattices can further enhance such dimension-dependent functional properties to increase the photovoltaic efficiency. A large difference in electronegativity between the core and outer regions of component QDs will additionally help the effective carrier separations. Passivating QDs by various kinds of ligands, especially by charged ligands, will significantly elongate the threshold nanospace with which the quantum resonance disappears. Preparing QD superlattices with smaller QDs will enhance the quantum resonance and its effects, such as the band gap energy shift because a larger ratio of a QD surface over volume causes the larger penetration of a wave function into nanospace from each QD. The obtained THzfrequency phonon modes demonstrated that the QD superlattice provided the new crystalline lattice through its shortrange interior nanospace between neighboring QDs. These THz-frequency phonon peaks and shoulders will be useful for

ACKNOWLEDGMENTS K.H.-D. acknowledges financial support from JST (PRESTO) and Grant-in-Aids for Scientific Research from the Japan Society for the Promotion of Science (KAKENHI), Grant 15K05386. D.K. acknowledges the financial support by KAKENHI, Grant 24560015.

■

REFERENCES

(1) Henglein, A. Photo-Degradation and Fluorescence of ColloidalCadmium Sulfide in Aquous-Solution. Ber. Bunsen-Ges. Phys. Chem. 1982, 86, 301−305. (2) Hyeon-Deuk, K.; Kim, J.; Prezhdo, O. V. Ab Initio Analysis of Auger-Assisted Electron Transfer. J. Phys. Chem. Lett. 2015, 6, 244− 249. (3) Kobayashi, Y.; Nishimura, T.; Yamaguchi, H.; Tamai, N. Effect of Surface Defects on Auger Recombination in Colloidal CdS Quantum Dots. J. Phys. Chem. Lett. 2011, 2, 1051−1055. (4) Calvo, M. E.; Hidalgo, N.; Schierholz, R.; Kovacs, A.; Fernandez, A.; Bellino, M. G.; Soler-Illia, G. J. A. A.; Miguez, H. Full Solution Processed Mesostructured Optical Resonators Integrating Colloidal Semiconductor Quantum Dots. Nanoscale 2015, 7, 16583−16589. (5) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Self-Organization of CdSe Nanocrystallites into Three-Dimensional Quantum Dot Superlattices. Science 1995, 270, 1335−1338.

E

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (6) Ponsinet, V.; Barois, P.; Gali, S. M.; Richetti, P.; Salmon, J. B.; Vallecchi, A.; Albani, M.; Le Beulze, A.; Gomez-Grana, S.; Duguet, E.; Mornet, S.; Treguer-Delapierre, M. Resonant Isotropic Optical Magnetism of Plasmonic Nanoclusters in Visible Light. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 92, 220414. (7) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Synthesis and Characterization of Monodisperse Nanocrystals and Close-Packed Nanocrystal Assemblies. Annu. Rev. Mater. Sci. 2000, 30, 545−610. (8) Taleb, A.; Petit, C.; Pileni, M. P. Optical Properties of SelfAssembled 2D and 3D Superlattices of Silver Nanoparticles. J. Phys. Chem. B 1998, 102, 2214−2220. (9) Wang, Z. L.; Harfenist, S. A.; Whetten, R. L.; Bentley, J.; Evans, N. D. Bundling and Interdigitation of Adsorbed Thiolate Groups in Self-Assembled Nanocrystal Superlattices. J. Phys. Chem. B 1998, 102, 3068−3072. (10) Zaitseva, N.; Dai, Z. R.; Leon, F. R.; Krol, D. Optical Properties of CdSe Superlattices. J. Am. Chem. Soc. 2005, 127, 10221−10226. (11) Lee, K. D.; Park, M. J.; Kim, D.-Y.; Kim, S. M.; Kang, B.; Kim, S.; Kim, H.; Lee, H.-S.; Kang, Y.; Yoon, S. S.; Hong, B. H.; Kim, D. Graphene Quantum Dot Layers with Energy-Down-Shift Effect on Crystalline-Silicon Solar Cells. ACS Appl. Mater. Interfaces 2015, 7, 19043−19049. (12) Carey, G. H.; Yuan, M.; Comin, R.; Voznyy, O.; Sargent, E. H. Cleavable Ligands Enable Uniform Close Packing in Colloidal Quantum Dot Solids. ACS Appl. Mater. Interfaces 2015, 7, 21995− 22000. (13) Artemyev, M. V.; Woggon, U.; Jaschinski, H.; Gurinovich, L. I.; Gaponenko, S. V. Spectroscopic Study of Electronic States in an Ensemble of Close-Packed CdSe Nanocrystals. J. Phys. Chem. B 2000, 104, 11617−11621. (14) Choi, J.-H.; Fafarman, A. T.; Oh, S. J.; Ko, D.-K.; Kim, D. K.; Diroll, B. T.; Muramoto, S.; Gillen, J. G.; Murray, C. B.; Kagan, C. R. Bandlike Transport in Strongly Coupled and Doped Quantum Dot Solids: A Route to High-Performance Thin-Film Electronics. Nano Lett. 2012, 12, 2631−2638. (15) Dollefeld, H.; Weller, H.; Eychmuller, A. Semiconductor Nanocrystal Assemblies: Experimental Pitfalls and a Simple Model of Particle-Particle Interaction. J. Phys. Chem. B 2002, 106, 5604− 5608. (16) Hanrath, T. Colloidal Nanocrystal Quantum Dot Assemblies as Artificial Solids. J. Vac. Sci. Technol., A 2012, 30, 030802. (17) Micic, O. I.; Ahrenkiel, S. P.; Nozik, A. J. Synthesis of Extremely Small InP Quantum Dots and Electronic Coupling in Their Disordered Solid Films. Appl. Phys. Lett. 2001, 78, 4022−4024. (18) Park, D. J.; Zhang, C.; Ku, J. C.; Zhou, Y.; Schatz, G. C.; Mirkin, C. A. Plasmonic Photonic Crystals Realized through DNA-Programmable Assembly. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 977−981. (19) Talgorn, E.; Gao, Y.; Aerts, M.; Kunneman, L. T.; Schins, J. M.; Savenije, T. J.; van Huis, M. A.; van der Zant, H. S. J.; Houtepen, A. J.; Siebbeles, L. D. A. Unity Quantum Yield of Photogenerated Charges and Band-Like Transport in Quantum-Dot Solids. Nat. Nanotechnol. 2011, 6, 733−739. (20) Zheng, Z.; Yim, K.-H.; Saifullah, M. S. M.; Welland, M. E.; Friend, R. H.; Kim, J.-S.; Huck, W. T. S. Uniaxial Alignment of LiquidCrystalline Conjugated Polymers by Nanoconfinement. Nano Lett. 2007, 7, 987−992. (21) Geiregat, P.; Justo, Y.; Abe, S.; Flamee, S.; Hens, Z. Giant and Broad-Band Absorption Enhancement in Colloidal Quantum Dot Mono layers through Dipolar Coupling. ACS Nano 2013, 7, 987−993. (22) Hodes, G.; Grunbaum, E.; Feldman, Y.; Bastide, S.; LevyClement, C. Variable Optical Properties and Effective Porosity of CdSe Nanocrystalline Films Electrodeposited from Selenosulfate Solutions. J. Electrochem. Soc. 2005, 152, G917−G923. (23) Hyeon-Deuk, K.; Kobayashi, Y.; Tamai, N. Evidence of PhononAssisted Auger Recombination and Multiple Exciton Generation in Semiconductor Quantum Dots Revealed by Temperature-Dependent Phonon Dynamics. J. Phys. Chem. Lett. 2014, 5, 99−105.

(24) Kryjevski, A.; Kilin, D. Multiple Exciton Generation in Silicon Quantum Dot Arrays: Density Functional Perturbation Theory Computation. Mol. Phys. 2014, 112, 430−440. (25) Hyeon-Deuk, K.; Prezhdo, O. V. Time-Domain Ab Initio Study of Auger and Phonon-Assisted Auger Processes in a Semiconductor Quantum Dot. Nano Lett. 2011, 11, 1845−1850. (26) Hyeon-Deuk, K.; Prezhdo, O. V. Multiple Exciton Generation and Recombination Dynamics in Small Si and CdSe Quantum Dots: An Ab Initio Time-Domain Study. ACS Nano 2012, 6, 1239−1250. (27) Voros, M.; Galli, G.; Zimanyi, G. T. Colloidal Nanoparticles for Intermediate Band Solar Cells. ACS Nano 2015, 9, 6882−6890. (28) Crooker, S. A.; Hollingsworth, J. A.; Tretiak, S.; Klimov, V. I. Spectrally Resolved Dynamics of Energy Transfer in Quantum-Dot Assemblies: Towards Engineered Energy Flows in Artificial Materials. Phys. Rev. Lett. 2002, 89, 186802. (29) Kagan, C. R.; Murray, C. B.; Bawendi, M. G. Long-Range Resonance Transfer of Electronic Excitations in Close-Packed CdSe Quantum-Dot Solids. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54, 8633−8643. (30) Leatherdale, C. A.; Kagan, C. R.; Morgan, N. Y.; Empedocles, S. A.; Kastner, M. A.; Bawendi, M. G. Photoconductivity in CdSe Quantum Dot Solids. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 62, 2669−2680. (31) Kryjevski, A.; Kilin, D.; Kilina, S. Amorphous Silicon Nanomaterials: Quantum Dots versus Nanowires. J. Renewable Sustainable Energy 2013, 5, 043120. (32) Vinson, N.; Freitag, H.; Micha, D. A. Structural and Orientation Effects on Electronic Energy Transfer between Silicon Quantum Dots with Dopants and with Silver Adsorbates. J. Chem. Phys. 2014, 140, 244709. (33) Miller, J. B.; Dandu, N.; Velizhanin, K. A.; Anthony, R. J.; Kortshagen, U. R.; Kroll, D. M.; Kilina, S.; Hobbie, E. K. Enhanced Luminescent Stability through Particle Interactions in Silicon Nanocrystal Aggregates. ACS Nano 2015, 9, 9772−9782. (34) Kim, D. G.; Tomita, S.; Ohshiro, K.; Watanabe, T.; Sakai, T.; Chang, I.-Y.; Hyeon-Deuk, K. Evidence of Quantum Resonance in Periodically-Ordered Three-Dimensional Superlattice of CdTe Quantum Dots. Nano Lett. 2015, 15, 4343−4347. (35) Georgiou, G.; Tserkezis, C.; Schaafsma, M. C.; Aizpurua, J.; Rivas, J. G. Active Loaded Plasmonic Antennas at Terahertz Frequencies: Optical Control of Their Capacitive-Inductive Coupling. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91, 125443. (36) Yang, Y.; Liu, Z.; Lian, T. Bulk Transport and Interfacial Transfer Dynamics of Photogenerated Carriers in CdSe Quantum Dot Solid Electrodes. Nano Lett. 2013, 13, 3678−3683. (37) Kim, D. G.; Okahara, S.; Nakayama, M.; Shim, Y. G. Experimental Verification of Forster Energy Transfer between Semiconductor Quantum Dots. Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 78, 153301. (38) Liu, Y.; Gibbs, M.; Puthussery, J.; Gaik, S.; Ihly, R.; Hillhouse, H. W.; Law, M. Dependence of Carrier Mobility on Nanocrystal Size and Ligand Length in PbSe Nanocrystal Solids. Nano Lett. 2010, 10, 1960−1969. (39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.02; Gaussian, Inc.: Wallingford, CT, 2009. F

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (40) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy Calculations Using a Plane-Wave Basis Set. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54, 11169−11186.

G

DOI: 10.1021/acsami.6b03219 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX