Direct Imaging of Surface Plasmon-Driven Hot Electron Flux on Au

3 days ago - Direct measurement of hot electron flux from a plasmonic Schottky nanodiode is important for obtaining fundamental insights explaining th...
0 downloads 0 Views 659KB Size
Subscriber access provided by Iowa State University | Library

Communication

Direct Imaging of Surface Plasmon-Driven Hot Electron Flux on Au Nanoprism/TiO2 Hyunhwa Lee, Hyunsoo Lee, and Jeong Young Park Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b04119 • Publication Date (Web): 04 Jan 2019 Downloaded from http://pubs.acs.org on January 5, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Direct Imaging of Surface Plasmon-Driven Hot Electron Flux on Au Nanoprism/TiO2 Hyunhwa Lee1,2, Hyunsoo Lee2, and Jeong Young Park1,2*

1Graduate

School of EEWS and Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 34141, Republic of Korea

2Center

for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Daejeon 34141, Republic of Korea Abstract

Direct measurement of hot electron flux from a plasmonic Schottky nanodiode is important for obtaining fundamental insights explaining the mechanism for electronic excitation on a surface. Here, we report the measurement of photo-induced hot electrons on a triangular Au nanoprism on TiO2 under incident light with photoconductive atomic force microscopy (pc-AFM), which is direct proof of the intrinsic relation between hot electrons and localized surface plasmon resonance. We find that the local photocurrent measured on the boundary of the Au nanoprism is higher than that inside the Au nanoprism, indicating that field confinement at the boundary of the Au nanoprism acts as a hot spot, leading to the enhancement of hot electron flow at the boundary. Under incident illumination with a wavelength near the absorption peak (645 nm) of a single Au nanoprism, localized surface plasmon resonance resulted in the generation of a higher photo-induced hot electron flow for the Au nanoprism/TiO2, compared with that at a wavelength of 532 nm. We show that the application of a reverse bias results in a higher photocurrent for the Au nanoprism/TiO2, which is associated with a lowering of the Schottky barrier height caused by the image force. These nanoscale measurements of hot electron flux with pc-AFM indicate efficient photon energy transfer mediated by surface plasmons in hot electron-based energy conversion.

*To whom correspondence should be addressed. E-mail: [email protected]

Keywords: Photoconductive atomic force microscopy, Schottky diode, hot electron, Au nanoprism, field confinement, localized surface plasmon resonance

1

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 17

Plasmonic metal–semiconductor diodes are an effective way to convert incident photons and chemical energy to electrical flow.1–9 During light absorption and exothermic chemical reactions on the metal surface, excited elections are generated with high kinetic energy (1–3 eV); these excited electrons are called hot electrons.10–16 Hot electrons, which have a short lifetime on the order of femtoseconds and are not in thermal equilibrium,1, 17–19 can be collected when the energy of the excited electrons is sufficient to surmount the Schottky barrier height and when the thickness of the metal is less than the mean free path of the electrons (i.e., a few nanometers).20, 21 Diverse approaches to enhance hot electron generation have evolved that utilize localized surface plasmon resonance, which is the collective oscillation of electrons in metal nanostructures at the same frequency as the incident light.9, 22–31 Field confinement (e.g., enhancement) at the boundaries of a nanostructured metal is a distinct phenomenon in nanophotonics.32–34 Upon surface plasmon resonance excitation, efficient surface plasmon field confinement in the plasmonic nanodiodes increases the light absorption of the metal nanostructure and transfers the photon energy to the electrons in the metal nanostructure, resulting in the increased generation of hot electrons.2, 3, 5–7 Amplification of the hot electron flux within plasmonic Schottky nanodiodes is a promising strategy to achieve high-efficiency solar energy conversion.22–30, 35 The concept of solar energy conversion into hot electron flow was demonstrated by McFarland and Tang who showed that hot electrons can be converted into photo-induced current.36 Sundararaman et al. provided theoretical descriptions to address hot carrier generation induced by surface plasmon decay that show that carrier distributions are important for the metal band structure. The geometry of the plasmonic metals also plays a significant role in achieving high efficiencies for hot carrier collection, depending on the plasmonic metals (e.g., copper, gold, aluminum, and silver).37 As the applications for surface plasmon-driven hot electrons have become increasingly important, there have been various studies to characterize the properties of hot electrons by using conventional spectroscopy methods (e.g., time-resolved pump probe spectroscopy, incident photon-to-electron conversion efficiency measurements).5,

6, 38

Ratchfold et al.

reported quantitative experiments of hot electron injection in Au nanoparticles/TiO2 with transient absorption spectroscopy, revealing injection efficiencies from Au nanoparticles to TiO2 reaching at least 25%.8 Also, Tagliabue et al. demonstrated practical experiments with internal quantum efficiency spectra to address hot carrier transport in Au/n-type GaN via 2

ACS Paragon Plus Environment

Page 3 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

surface plasmon excitation, indicating that the electronic band structure of the plasmonic metal affects the photoexcited hot carrier distribution.9 However, because of limitations in lateral resolution, these results can be employed to measure the average electrical properties within the target area, but they have difficulties in characterizing the local properties of each Schottky nanodiode. To address the intrinsic relation between hot electron flux and surface plasmon resonance on Au nanostructures, we employed photoconductive atomic force microscopy (pc-AFM) to simultaneously provide both the morphological variation and local photon-induced current for individual Schottky nanodiodes.39 We suggest that direct detection with pc-AFM is a more effective way to discuss amplification of hot electrons by field confinement at the boundary of the Au nanoprism upon SPR excitation,24 showing that the boundary of a Au nanoprism acts as a hot spot to increase the number of hot electrons generated. Model systems for the plasmonic nanostructures are Schottky nanodiodes formed by triangular Au nanoprisms on n-type TiO2. We used these nanoprisms to collect hot electrons, which are measured as the photocurrent in the pc-AFM system. Through current mapping, we demonstrated that hot electron generation was enhanced at the boundary of the Au nanoprisms, which was caused by field enhancement at the edge where the incident photon wavelength matched the surface plasmon resonance. The Au nanoprism/n-type TiO2 diodes were fabricated as follows: First, a 150-nm thick Ti film was deposited on the quartz substrate by e-beam evaporation. The Ti film deposited on the substrate was annealed in air at 470 °C for 2 h 45 min to oxidize the Ti into n-type TiO2. To observe the current in the pc-AFM system, a 50-nm Ti film and a 100-nm Au film (i.e., ohmic contact) were sequentially deposited on the TiO2 substrate. The hexagonal close-packed nanopattern on the TiO2 substrate was formed via the self-assembly-based nanosphere lithography technique using polystyrene40 (PS) nanospheres with a diameter of 460 nm (from Sigma-Aldrich, see Figure S1b). The monodispersed PS nanospheres were prepared by creating a solution with 10 wt.% water diluted by adding an equal amount of ethanol. After adding 150 μl of a 2% dodecyl sodium sulfate solution onto the surface of the water in a vessel, about 10 μl of the prepared PS solution was dropped onto the surface of a slide glass immersed in the water. The TiO2 substrate was slowly immersed in the water in the vessel. After consolidation of the PS monolayer on the water surface, the PS monolayer was lifted off the water surface. After deposition of a 15-nm thick gold film on the 3

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TiO2 substrate (as shown in Figure 1a), the PS monolayer was removed via ultrasonic treatment. The hexagonal close-packed Au nanoprisms on the TiO2 are shown in Figure 1c; the length of one side of the nanoprism was 132 nm, which corresponds to the SEM image in Figure 1b. The absorbance of the Au nanoprisms measured using UV-Vis spectroscopy is shown in Figure 1d, which exhibits a broad spectrum that results from the combination of several peaks. To find the resonance peak of a Au nanoprism, we calculated the absorbance of a single Au nanoprism via final differential time domain (FDTD) analysis where the E-field is vertical at the base of the triangle. This shows that the simulated absorbance of a single Au nanoprism with a peak position at 645 nm and hexagonal close-packed Au nanoprism arrays with peak positions at 630 and 691 nm are comparable to the measured UV-Vis absorption spectrum, which is shown as having well-matched absorption peaks. Under illumination near a wavelength of 645 nm, we expected the single Au nanoprism to couple with the surface plasmon resonance to increase light absorption, revealing enhanced hot electron generation. The measured absorption spectrum is broader than the calculated one, which is associated with a non-uniform array of the Au prism that is caused by imperfections during the self-assembly of the polystyrene monolayer produced using nanosphere lithography. For the pc-AFM measurements, we modified the sample stage and adopted a fused silica prism as the waveguide for the backside-illuminated laser (OBIS series, Coherent).39 Passing through the prism, the focused laser entered perpendicular to the back of the sample. After the AFM probe approached the surface of sample, we could simultaneously measure both the topography and the current. An external bias was applied to the metal-coated AFM probe and the current was collected by the ohmic contact between the conductive probe and the TiO2 substrate. Figure 2a shows a scheme of the experimental AFM system for probing the photocurrent in the Au nanoprism/TiO2 during light illumination. A Schottky barrier was created between the Au nanoprisms and the TiO2 substrate. Upon light excitation, the generated hot electrons can transfer from the Au nanoprisms to the TiO2 substrate and are captured through the ohmic contact (Au/Ti) between the nanodiode and the pc-AFM system. To evaluate the Schottky barrier height of the Au nanoprism/TiO2 using pc-AFM, a current–voltage (I–V) curve was measured and fitted to the thermionic emission equation41 (Figure 2c), resulting in a barrier height of 0.9 eV and an ideal factor of 2.2 for the 15-nm gold layer deposited on TiO2 (see Figure 2b). Given by the work function of Au (5 eV) and the electron affinity of TiO2 (3.9 4

ACS Paragon Plus Environment

Page 4 of 17

Page 5 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

eV), the experimentally measured Schottky barrier height is lower than the ideal barrier height (1.1 eV) because of the image-force-induced lowering of the barrier height.41–43 Figure 3b–d shows the photocurrent mapping of a Au nanoprism on TiO2 corresponding to the topography shown in Figure 3a, which was performed using a TiN-coated probe (CSG10, NT-MDT) with a normal load of 2 nN under laser illumination. In the dark (Figure 3b), there was no distinguishable current on the Au nanoprism. During light illumination with an intensity of 10 kW/m2 at wavelengths of 532 (Figure 3c) and 640 nm (Figure 3d), the photocurrent was higher at the boundary of the Au nanoprism than at the inner area, which revealed that the photocurrent at the boundary was enhanced under the incident wavelength of 640 nm. To determine the origin of the increased number of hot electrons at the boundary, we observed a rise in the photocurrent in response to the intensity of the light (Figure S2), revealing that the increase in hot electrons at the boundary of the Au nanoprism was caused by field confinement upon localized surface plasmon resonance excitation that can be directly measured in the pc-AFM system, despite the increase in the contact area between the tip and the Au nanoprism at the boundary. Corresponding to the region of localized surface plasmon resonance excitation, the generated hot electron flow was amplified at the boundary of the Au nanoprism because of field enhancement. Figure S3 shows I–V curves at the inner Au nanoprism and at the boundary of the Au nanoprism during illumination, which is consistent with the photocurrent mapping in Figure 3c,d. Figure 3e shows a plot of the short-circuit photocurrent measured on the Au nanoprism/TiO2 diode in the dark and during laser illumination. The photocurrent measured at the boundary when illuminated by a 640-nm laser was 2.7 times higher than that illuminated by a 532-nm laser. These results imply a correlation between hot electron generation and field enhancement at the boundary of a metal nanostructure, which is associated with the generation of more hot electrons induced by field enhancement during localized surface plasmon resonance excitation. Furthermore, we performed current mapping by applying a tip bias. Application of a reverse bias can change the Schottky barrier height, which affects the efficiency of hot electron collection. Figure 4 shows photocurrent maps while applying a reverse bias to the AFM tip. The induced image charges in the Au nanoprism lead to an electric field in the TiO2. The image force induced by coulomb attraction between the image charge and the electrons modifies the total potential barrier in the electric field. Thus, the Schottky barrier height was reduced by 5

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ΔΦB upon applying the reverse bias to the Au nanoprism/TiO2 diode. When applying a bias of −0.3 Vtip in the dark, the current is not a distinguishable feature at the boundary of the Au nanoprism (Figure 4b). However, the photocurrent mapping during illumination at wavelengths of 532 (Figure 4c,d) and 640 nm (Figure 4e,f) with an applied reverse bias of −0.3 V revealed a significant enhancement of the photocurrent at the boundary compared with the inside of the Au island. Because the Schottky barrier height is lower, the number of collected hot electrons increased during illumination, which led to the enhanced photocurrent at the boundary of the Au nanoprism. With the influence of the lower image force barrier, the results imply that an applied reverse bias can promote the capture of more hot electrons transferring from the Au into the TiO2 through the lower barrier at the interface between the Au nanoprism and the TiO2. These bias-dependent experiments demonstrate that the hot electrons generated via surface plasmon resonance dominate the measured photocurrent in the AFM system. In Figure 5, FDTD-simulation models for a 132-nm long Au nanoprism show field enhancement at the boundary that is dependent on incident wavelengths of 500–800 nm. The |E| distributions were performed using a vertically polarized incident E-field. This revealed that field enhancement near a wavelength of 650 nm was strongly confined at the boundary of the Au nanoprism; this is associated with the calculated spectrum of a single Au nanoprism (see Figure 1d) and is related to an increase in photon absorption via localized surface plasmon resonance, which corresponds to enhanced photocurrent at the boundary of the Au nanoprism (Figure 4g). Figure 5i shows the plot of averaged |E| values at the boundary of a Au nanoprism from 500 to 800 nm through the spatial E-field distribution of the FDTD simulations (Figure 5b–h), which indicates that the magnitude of E-field confinement at the boundary under illumination by 640nm light increased by 2.6 times that illuminated at 532 nm, in accordance with the ratio of the photocurrent increase at the boundary (2.7 times, see Figure 3e). In Figure S4, we carried out detailed FDTD-simulations for a Au nanoprism with identical polarization as in Figure 5, which shows a distinguishable field distribution at λ= 800 nm with strong amplification at the edges that differs from the E-field distributions at λLSPR= 640 nm and λ= 532 nm. Since we focused on studying the effects of resonant E-field amplification on generated hot electrons and compared the generation of hot electrons according to the intensity of the field confinement, we used 640- and 532-nm lasers to conduct the wavelength-dependent experiments (Figure S4 in the Supporting Information). 6

ACS Paragon Plus Environment

Page 6 of 17

Page 7 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

To show the correlation between the absorption spectrum and the pc-AFM data, we calculated the external quantum efficiency (EQE), which is the ratio of the number of generated hot electrons to the number of absorbed photons, following9 Number of collected electrons/sec Current × 6.22 × 1018 EQE = = Number of incident photons/sec Total energy/Energy of 1 photon where we used a total energy of 1.97 × 10−11 J∙s−1 and single-photon energies of 3.10 × 10−19 and 3.73 × 10−19 J at wavelengths of 640 and 532 nm, respectively. We calculated that the conversion efficiency at the boundary upon excitation from the 640-nm laser was 2.3 times higher than that upon excitation from the 532-nm laser (as shown in Figure 5j). These results indicate that field confinement upon localized surface plasmon resonance excitation affects the amplification of the hot electron flux. By tailoring the nanostructured Schottky diode, we suggest that studies on devices with high solar energy conversion efficiency will show progress in the future. Internal photoemission of plasmonic nanodiodes via localized surface plasmon resonance can be enhanced, which improves applications for light-harvesting devices. Upon applying a reverse bias to Schottky nanodiodes, the number of hot electrons collected increases because of the lower Schottky barrier height. The increase in the number of generated hot electrons via localized surface plasmon resonance was directly measured at nanoscale using a modified pc-AFM (Agilent 5500), which demonstrates that field confinement induced by the geometry of the triangular Au nanodisk leads to an increase in hot electron flux upon the excitement of localized surface plasmon resonance. We suggest that nanoscale measurements with pc-AFM present a fundamental insight into increasing the hot electron flux and efficiently transferring photon energy with hot-electron-based photodetectors. Future study involves the direct measurement of hot electron flux on Schottky nanostructures with different metal or oxide materials that can reveal the relation between the hot electron flux, Schottky barrier height, and the modes of surface plasmon resonance. In summary, the amplification of hot electron flux at the boundary of a Au nanoprism was directly probed with Schottky nanodiodes using photoconductive AFM. Corresponding to the absorption spectrum, FDTD simulations indicate field enhancement at the boundary of a triangular Au nanoprism during the excitation of localized surface plasmon resonance at a 7

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

wavelength of 645 nm. The measured photocurrent at the boundary was greater using a 640nm laser than with a 532-nm laser, which demonstrates that the boundary of the Au nanoprism acted as an active site for amplifying the generation of hot electrons induced by field enhancement via localized surface plasmon resonance. Bias-dependent experiments showed that hot electrons are dominant in the measured photocurrent because of the lower Schottky barrier height. Furthermore, quantitative calculations of the conversion efficiency showed that the generation of hot electrons increased significantly at the boundary, revealing that field confinement during localized surface plasmon resonance excitation directly affects the amplification of hot electron flux.

ASSOCIATED CONTENT Supporting information. Experimental details, additional current mapping, I–V curves, and FDTD simulations are provided in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]

ACKNOWLEDGMENTS This work was supported by the Institute for Basic Science (IBS) [IBS-R004].

REFERENCES (1) Furube, A.; Du, L.; Hara, K.; Katoh, R.; Tachiya, M. Journal of the American Chemical Society 2007, 129, (48), 14852–14853. (2) Tian, Y.; Tatsuma, T. Chemical Communications 2004, (16), 1810–1811. (3) Tian, Y.; Tatsuma, T. Journal of the American Chemical Society 2005, 127, (20), 7632– 7637. (4) Wang, F.; Melosh, N. A. Nano Letters 2011, 11, (12), 5426–5430. 8

ACS Paragon Plus Environment

Page 8 of 17

Page 9 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(5) Lee, Y. K.; Jung, C. H.; Park, J.; Seo, H.; Somorjai, G. A.; Park, J. Y. Nano Letters 2011, 11, (10), 4251–4255. (6) Lee, H.; Lee, Y. K.; Hwang, E.; Park, J. Y. The Journal of Physical Chemistry C 2014, 118, (11), 5650–5656. (7) Lee, S. W.; Hong, J. W.; Lee, H.; Wi, D. H.; Kim, S. M.; Han, S. W.; Park, J. Y. Nanoscale 2018, 10, (23), 10835–10843. (8) Ratchford, D. C.; Dunkelberger, A. D.; Vurgaftman, I.; Owrutsky, J. C.; Pehrsson, P. E. Nano Letters 2017, 17, (10), 6047–6055. (9) Tagliabue, G.; Jermyn, A. S.; Sundararaman, R.; Welch, A. J.; DuChene, J. S.; Pala, R.; Davoyan, A. R.; Narang, P.; Atwater, H. A. Nature Communications 2018, 9, (1), 3394. (10) Park, J. Y.; Somorjai, G. A. Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures Processing, Measurement, and Phenomena 2006, 24, (4), 1967– 1971. (11) Somorjai, G. A.; Frei, H.; Park, J. Y. Journal of the American Chemical Society 2009, 131, (46), 16589–16605. (12) Park, J. Y.; Lee, H.; Renzas, J. R.; Zhang, Y.; Somorjai, G. A. Nano Letters 2008, 8, (8), 2388–2392. (13) Nienhaus, H. Surface Science Reports 2002, 45, (1), 1–78. (14) Park, J. Y.; Somorjai, G. A. ChemPhysChem 2006, 7, (7), 1409–1413. (15) Gadzuk, J. W. The Journal of Physical Chemistry B 2002, 106, (33), 8265–8270. (16) Karpov, E. G.; Nedrygailov, I. I. Applied Physics Letters 2009, 94, (21), 214101. (17) Brongersma, M. L.; Halas, N. J.; Nordlander, P. Nature Nanotechnology 2015, 10, 25. (18) Wu, K.; Chen, J.; McBride, J. R.; Lian, T. Science 2015, 349, (6248), 632. (19) Landau, L. D. Jounal of Physics 1946. (20) Inouye, H.; Tanaka, K.; Tanahashi, I.; Hirao, K. Physical Review B 1998, 57, (18), 11334– 11340. (21) Hohlfeld, J.; Wellershoff, S. S.; Güdde, J.; Conrad, U.; Jähnke, V.; Matthias, E. Chemical Physics 2000, 251, (1), 237–258. (22) Giannini, V.; Fernández-Domínguez, A. I.; Heck, S. C.; Maier, S. A. Chemical Reviews 2011, 111, (6), 3888–3912. (23) Hartland, G. V. Chemical Reviews 2011, 111, (6), 3858–3887. (24) Swearer, D. F.; Zhao, H.; Zhou, L.; Zhang, C.; Robatjazi, H.; Martirez, J. M. P.; Krauter, 9

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

C. M.; Yazdi, S.; McClain, M. J.; Ringe, E.; Carter, E. A.; Nordlander, P.; Halas, N. J. Proceedings of the National Academy of Sciences 2016, 113, (32), 8916–8920. (25) Linic, S.; Aslam, U.; Boerigter, C.; Morabito, M. Nature Materials 2015, 14, 567. (26) Scarabelli, L.; Hamon, C.; Liz-Marzán, L. M. Chemistry of Materials 2017, 29, (1), 15– 25. (27) Jain, P. K.; Lee, K. S.; El-Sayed, I. H.; El-Sayed, M. A. The Journal of Physical Chemistry B 2006, 110, (14), 7238–7248. (28) Xiao, F.-X.; Zeng, Z.; Liu, B. Journal of the American Chemical Society 2015, 137, (33), 10735–10744. (29) Liu, L.; Ouyang, S.; Ye, J. Angewandte Chemie International Edition 2013, 52, (26), 6689–6693. (30) Lee, C.; Choi, H.; Nedrygailov, I. I.; Lee, Y. K.; Jeong, S.; Park, J. Y. ACS Applied Materials & Interfaces 2018, 10, (5), 5081–5089. (31) Brown, A. M.; Sundararaman, R.; Narang, P.; Goddard, W. A.; Atwater, H. A. ACS Nano 2016, 10, (1), 957–966. (32) des Francs, G. C.; Molenda, D.; Fischer, U. C.; Naber, A. Physical Review B 2005, 72, (16), 165111. (33) Naber, A.; Molenda, D.; Fischer, U. C.; Maas, H. J.; Höppener, C.; Lu, N.; Fuchs, H. Physical Review Letters 2002, 89, (21), 210801. (34) Tumkur, T.; Yang, X.; Zhang, C.; Yang, J.; Zhang, Y.; Naik, G. V.; Nordlander, P.; Halas, N. J. Nano Letters 2018, 18, (3), 2040–2046. (35) Raether, H., Surface Plasmons on Smooth Surfaces. In Surface Plasmons on Smooth and Rough Surfaces and on Gratings. Springer: Berlin, 1988. (36) McFarland, E. W.; Tang, J. Nature 2003, 421, 616. (37) Sundararaman, R.; Narang, P.; Jermyn, A. S.; Goddard Iii, W. A.; Atwater, H. A. Nature Communications 2014, 5, 5788. (38) Heilpern, T.; Manjare, M.; Govorov, A. O.; Wiederrecht, G. P.; Gray, S. K.; Harutyunyan, H. Nature Communications 2018, 9, (1), 1853. (39) Eichhorn, J.; Kastl, C.; Cooper, J. K.; Ziegler, D.; Schwartzberg, A. M.; Sharp, I. D.; Toma, F. M. Nature Communications 2018, 9, (1), 2597. (40) Hulteen, J. C.; Van Duyne, R. P. Journal of Vacuum Science & Technology A 1995, 13, (3), 1553–1558. 10

ACS Paragon Plus Environment

Page 10 of 17

Page 11 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(41) Ng, S. M. S. K., Physics of Semiconductor Devices. John Wiley & Sons: 2006. (42) Sze, S. M.; Crowell, C. R.; Kahng, D. Journal of Applied Physics 1964, 35, (8), 2534– 2536. (43) Lee, H.; Lee, Y. K.; Van, T. N.; Park, J. Y. Applied Physics Letters 2013, 103, (17), 173103.

11

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. (a) Scheme of the sample preparation steps to build triangle-shaped Au nanoprisms on TiO2 substrate using the self-assembly nanosphere lithography technique. A monolayer was formed by polystyrene(PS) nanospheres with a diameter of 460 nm. A thin gold film was then deposited on the PS monolayer using e-beam evaporation. After ultrasonic treatment, hexagonal close-packed Au nanoprisms were fabricated. (b) SEM image and (c) AFM topography showing the hexagonal closepacked Au nanoprisms on TiO2. (d) Absorption spectra of the measured absorbance (black) compared with a Au nanoprism array (blue) and a single Au nanoprism (red) calculated using FDTD. The single Au nanoprism has a resonant peak at 645 nm.

12

ACS Paragon Plus Environment

Page 12 of 17

Page 13 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2. (a) Scheme for probing enhanced hot electron flux at the boundaries of a triangular Au nanoprism/n-typeTiO2 diode under light illumination. The energy band diagram shows photo-induced hot electrons transferred from the surface of the metal to the TiO2 substrate across the Schottky barrier. (b) A cross-sectional profile of the Au nanoprism/TiO2 revealing a Au thickness of 15 nm, which is within the mean free path of the electrons in the metal. The inset shows the topography of the Au nanoprism. (c) Current–voltage (I–V) curves of the Schottky nanodiode in log scale that were measured by sweeping the applied tip bias in the pc-AFM system. Fitting I–V curves for the Au/TiO2 nanodiode to the thermionic emission equation shows a Schottky barrier height of 0.9 eV and ideality factor of 2.2.

13

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Spatial mapping of the enhanced photocurrent at the boundary of a Au nanoprism/TiO2 nanodiode. (a) Topography of the Au nanoprism/TiO2, revealing a Au thickness of 15 nm, which is within the mean free path of the electrons in the metal. (c) In the dark, there was no distinguishable current signal on the Au nanoprism. Under light illumination, the enhanced photocurrent at the boundary of a Au nanoprism was measured at photon wavelengths of (c) 532 and (d) 640 nm with a light intensity of 10 kW/m2. (e) Depending on the measurement site, the photocurrents showed different behaviors that resulted in an increase in the number of hot electrons generated at the boundary of the Au nanoprism. Furthermore, the overall photocurrent measured when illuminated by a 640-nm laser was higher than that with a 532-nm laser regardless of the active sites.

14

ACS Paragon Plus Environment

Page 14 of 17

Page 15 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 4. Spatial mapping of the enhanced photocurrent at the boundary of a Au nanoprism/TiO2 nanodiode upon applying a reverse bias. (a) Current mapping in the dark and (c–f) photocurrent mapping of Au/TiO2 diode under light illumination with a reverse bias of −0.3 Vtip. The enhanced photocurrent at the boundaries of the Au nanoprism was measured, depending on the photon energies and different intensities of (c) 5 and (d) 10 kW/m2 at 532 nm and (e) 5 and (f) 10 kW/m2 at 640 nm with −0.3 Vtip. (g) Deviation between the currents at the Au edge and at the inner Au, revealing an enhancement of the hot electron flux at the boundary depending on the incident wavelengths and intensities. In contrast to having no external bias, the reverse bias contributed to a greater enhancement of the photocurrent at the boundaries that led to a higher hot electron flux at the boundaries because the reverse bias can adjust the energy band bending of the Au/TiO2 interface and lower the Schottky barrier height, which resulted in the collection of generated hot electrons from the gold to the TiO2.

15

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. FDTD-simulation models for calculating the |E| distributions of the 132-nm Au nanoprism on TiO2 in the direction of (a) linear polarization parallel to the centerline of the Au triangle and |E| distributions under (b) 800-, (c) 750-, (d) 700-, (e) 650-, (f) 600-, (g) 550-, and (h) 500-nm incident light, revealing a strong field confinement near 650 nm that leads to an increase in absorption at the localized surface plasmon resonance peak and enhances the photocurrent at the boundaries of the Au nanoprism. (i) Averaged |E|/|E0| at the boundary of a Au nanoprism from FDTD simulations depending on incident wavelengths of 500–800 nm. (j) External quantum efficiency of the measured photocurrent, resulting in a 2.3-fold amplification of the conversion at the boundary when excited by 640-nm light, revealing that field confinement upon localized surface plasmon resonance excitation directly affects amplification of the hot electron flux. 16

ACS Paragon Plus Environment

Page 16 of 17

Page 17 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Table of Contents

17

ACS Paragon Plus Environment