Subscriber access provided by George Mason University Libraries & VIVA (Virtual Library of Virginia)
Communication
Aluminum Plasmonics for Enhanced Visible Light Absorption and High Efficiency Water Splitting in Core-Multishell Nanowire Photoelectrodes with Ultrathin Hematite Shells Sarath Ramadurgam, Tzu-ging Lin, and Chen Yang Nano Lett., Just Accepted Manuscript • DOI: 10.1021/nl501541s • Publication Date (Web): 27 Jun 2014 Downloaded from http://pubs.acs.org on July 1, 2014
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 free 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 accessible to all readers and 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.
Nano Letters 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 21
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
Aluminum Plasmonics for Enhanced Visible Light Absorption and High Efficiency Water Splitting in Core-Multishell Nanowire Photoelectrodes with Ultrathin Hematite Shells Sarath Ramadurgam1, Tzu-Ging Lin1 and Chen Yang1, 2 1
Department of Physics and Astronomy, Purdue University, West Lafayette, Indiana 47907, United States
2
Department of Chemistry, Purdue University, West Lafayette, Indiana 47907, United States
KEYWORDS: solar hydrogen, aluminum plasmonics, hematite, core-shell, core-multishell, nanowires, plasmonic light harvesting
ABSTRACT
The poor internal quantum efficiency (IQE) arising from high recombination and insufficient absorption is one of the critical challenges towards achieving high efficiency water splitting in hematite (α-Fe2O3) photoelectrodes. By combining the nanowire (NW) geometry with the localized surface plasmon resonance (LSPR) in semiconductor-metal-metal oxide core-multishell (CMS) NWs, we theoretically demonstrate an effective route to strongly improve absorption
ACS Paragon Plus Environment
1
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 21
within ultrathin (sub-50 nm) hematite layers. We show that Si-Al-Fe2O3 CMS NWs exhibit photocurrent densities comparable to Si-Ag-Fe2O3 CMS and outperform Fe2O3, Si-Fe2O3 CS and Si-Au-Fe2O3 CMS NWs. Specifically; Si-Al-Fe2O3 CMS NWs reach photocurrent densities of ~11.81 mA/cm2 within a 40 nm thick hematite shell which corresponding to a solar to hydrogen (STH) efficiency of 14.5%. This corresponds to about 93% of the theoretical maximum for bulk hematite. Therefore, we establish Al as an excellent alternative plasmonic material compared to precious metals in CMS structures. Further, the absorbed photon flux is close to the NW surface in the CMS NWs, which ensures the charges generated can reach the reaction site with minimal recombining. Although the NW geometry is anisotropic, the CMS NWs exhibit polarization independent absorption over a large range of incidence angles. Finally, we show that Si-AlFe2O3 CMS NWs demonstrate photocurrent densities greater than ~8.2 mA/cm2 (STH efficiency of 10%) for incidence angles as large as 45°. These theoretical results strongly establish the effectiveness of the Al based CMS NWs for achieving scalable and cost-effective photoelectrodes with improved IQE, enabling a novel route towards high efficiency water splitting.
Photoelectrochemical cells employing photoelectrodes made of earth abundant materials offer an elegant, inexpensive and clean route for the production of hydrogen from water. To this end, hematite (α-Fe2O3) as a photoanode, has drawn great interest due to its band-gap of 2.1 eV ensuring good absorption in the visible range (250-600 nm) and its valence band alignment ideal to catalyze the oxygen evolution reaction1-6. In addition, hematite is inexpensive, abundant and stable in aqueous medium. The theoretical maximum photocurrent density for hematite under AM 1.5G (1-sun) illumination is 12.6 mA/cm2, which corresponds to a maximum solar to
ACS Paragon Plus Environment
2
Page 3 of 21
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
hydrogen (STH) efficiency of 15.5%, making it a promising material for photocatalysis4, 7. However, the minority carrier (hole) diffusion length in the bulk hematite is about 20 nm, which limits the effective thickness of hematite contributing towards the photocurrent density7-9. In addition, the conduction band edge of hematite is lower than the reduction potential of water; hence a large overpotential is necessary to drive the hydrogen evolution. Over the past decade, several strategies, such as the use of nanostructures, doping, heterojunctions and co-catalysts, have been developed for improving the absorption, conductivity, charge collection, overpotential and reaction kinetics of hematite photoelectrodes. The best STH efficiency achieved experimentally so far employing these strategies is only about 5.3%10. The main challenge towards achieving the coveted 10% efficiency required for commercial applications is the poor internal quantum efficiency (IQE) arising from the recombination in bulk. The challenge of low IQE can be partially addressed by using nanostructures, especially nanowires (NWs)2, 3. NWs exhibit size dependant optical (Mie like) resonances which can be tuned to enhance absorption in the visible range. Such optical resonances and absorption enhancement have been reported for semiconductor NW11, 12 and dielectric-dielectric core-shell (CS) NW13. More recently, the use of plasmonic nanostructures14-18, opal scaffolds19 and resonant light trapping in thin films7 and photonic nanostructures20 has been demonstrated as another strategy to improve IQE. Plasmonic nanostructures can be designed to effectively concentrate light into small spaces21, enhance absorption in the surrounding semiconductor22, provide hot electrons and drive the electrolysis of water23-26. For instance, Si-Fe2O3 CS NW photoelectrodes with Au nanoparticles embedded in the hematite shell have demonstrated a STH efficiency of up to 6%18. An alternative approach to utilize plasmonics is the metal-semiconductor CS structures, such as uniform27, 28 and tapered29 Ag-Si CS NW, which combine the NW geometry and the localized
ACS Paragon Plus Environment
3
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 4 of 21
surface plasmon resonance (LSPR) in the metal core to strongly confine light within an ultrathin semiconductor shell. All these designs utilize precious metals as the plasmonic materials, which are expensive and unsustainable to scale up. It is highly beneficial to design and develop a scalable plasmonic photoelectrode that can fundamentally boost the IQE irrespective of the photocatalyst material. In this letter, we present semiconductor-metal-metal oxide core-multishell (CMS) nanowires employing aluminum (Al) as a novel class of plasmonic photoelectrodes for high efficiency water splitting. Al is an intriguing plasmonic material as it potentially offers the following features. First, the LSPR in Al nanorods and nanoparticles is tunable through the UV and visible spectrum30, 31. Second, it can strongly enhance fields due to nature of its dielectric function and high electron density. Third, it is inexpensive, abundant and compatible with CMOS fabrication technology. In CMS NW structures proposed, such as Si-Al-Fe2O3 CMS NWs, the visible range plasmon resonances in Al are utilized to strongly enhance fields in sub-50 nm hematite shells and thereby drastically increase absorption. In addition, we pick Si as the semiconductor core, as Si in the Si-hematite heterostructures has been reported to substantially reduce the requisite overpotential18,
32-34
. Figure 1(a) shows the schematic of the Si-Al-Fe2O3 CMS NWs and the
band-structure of these heterostructures along the radius. As compared to previous plasmonic nanoparticles based CS designs, the use of Si NWs as a scaffold and Al thin films as shells in the proposed structure offers greater control over the material quality and thickness. The CMS structure also ensures that the plasmonic field enhancement is uniform over the entire length of the nanowire without any additional design constraints due to size, density and position variability often encountered with nanoparticles. More significantly, the use of Al as the
ACS Paragon Plus Environment
4
Page 5 of 21
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
plasmonic material, unlike precious metals such as Ag, Au or Pt, enables potentially costeffective and scalable plasmonic light harvesting. In order to evaluate the effectiveness of the proposed structures, we calculated and studied the absorption within each layer of the CMS structures with a focus on the hematite shell. We obtained rigorous solutions to the Maxwell’s equations using Mie-formalism for a single CMS NW in vacuum under arbitrary polarized incidence and computed the absorption efficiency (ηi where i=0, 1, 2 or 3 corresponds to vacuum, the outer shell, intermediate shell and core respectively) as a function of wavelength in each layer of the CMS NW (see Supporting Information for a detailed derivation)35-38. The incident light can be resolved into two cases: (I) uur ur magnetic field ( H ) perpendicular to the nanowire axis and (II) electric field ( E ) perpendicular
to the nanowire axis. When the incident light is normal to the NW axis, case (I) corresponds to the transverse magnetic (TM) illumination and case (II) corresponds to the transverse electric (TE) illumination. For case (I), the absorption efficiency within each layer is given as follows28:
ηi( I ) =
( ) + ( E( ) ) + ( E( ) )
k0 I Im ε r Er( ) 2 ∬ { ( )} 2r0 E0
2
I
2
ϕ
I
z
2
r dr dϕ
where, r0 is the total radius of the nanowire, k0 is the wave-number of the incident light, ε ( r ) is the dielectric function, E0 is the amplitude of the incident light and E ( I ) is the electric field obtained from the solution to Maxwell’s equations for case (I) illumination. The limits to the integral are set accordingly to obtain absorption within individual layers. The absorption efficiency ηi( ) for case (II) can be obtained similarly. The absorption efficiency for unpolarized II
illumination is given as ηi =
(
)
1 (I ) ηi + ηi( II ) . In order to obtain the photon flux absorbed within 2
the hematite layer and to estimate the ideal photocurrent density achievable in these structures,
ACS Paragon Plus Environment
5
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 6 of 21
we assume air mass 1.5 global (AM 1.5G; 1-sun) illumination incident perpendicular to the NW axis. The photon flux absorbed is integrated over the range between 300 and 590 nm which corresponds to the overlap between AM 1.5G illumination and the absorption spectrum of hematite7. The integrated photon flux absorbed is given as follows:
φabs ,i =
λ2 =590 nm
∫
λ1 =300 nm
λ hPlanck clight
I AM 1.5G ( λ )ηi d λ
where, I AM 1.5G ( λ ) is the intensity of AM 1.5G illumination, hPlanck is the Plank’s constant and clight is the speed of light. The photon flux is obtained through dividing the intensity I AM 1.5G ( λ )
by the energy of the photon hPlanck clight λ . The integrated photon flux absorbed calculated in this manner provides a good basis for comparison between different NW structures. Further, the photocurrent density assuming ideal generation of charges and forward injection to drive the reaction can be estimated as J = qφabs ,i . For all subsequent calculations presented, we utilize the dielectric functions interpolated from experimental data for thin films7, 39. In all calculation, NWs are assumed to be in vacuum. Results obtained are applicable for non-absorbing medium with a refractive index of 1. However, for typical aqueous electrolytes, the spectral position and characteristics of the optical resonances change depending on the refractive index. Further, the absorption within the nanowire is also scaled down as a function of the refractive index (see Equation 14 in Supporting Information). Therefore, for specific electrolytes, the size of the CMS NWs has to be further optimized to obtain maximum absorption. In this work, to compare absorption occurring in various NWs and evaluating these different NW structures, we consider vacuum as the medium and refer to the hematite-vacuum interface as the hematite-electrolyte interface in the photoelectrodes. The absorption and photocurrent density calculated in this work is only for single NWs, which provides insights and a guide for designing high performance NW
ACS Paragon Plus Environment
6
Page 7 of 21
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
photoelectrode arrays. The optical and photoelectrical performance of NW arrays needs further investigations.
Figure 1. (a) Schematic and band-structure of a Si-Al-Fe2O3 CMS NW photoelectrode. In the band diagram both Si and Fe2O3 have been assumed to be n-doped. (b) Integrated absorbed photon flux ( φabs , Fe2O3 ) and ideal photocurrent density ( J = qφabs , Fe2O3 ) within the hematite shell for various NW structures under unpolarized 1-sun illumination incident normal to the NW axis. The grey dashes indicate the theoretical maximum absorption in bulk hematite. Figure 1b plots the integrated photon flux absorbed in the hematite shell of a Si-Al-Fe2O3 CMS NW, compared with the results from Fe2O3, Si-Fe2O3 CS, Ag-Fe2O3 CS, Si-Ag-Fe2O3 CMS and Si-Au-Fe2O3 CMS NWs for unpolarized light incident perpendicular to the NW axis. The Si core
ACS Paragon Plus Environment
7
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 8 of 21
radius and the metal shell thickness in CS and CMS nanowires are fixed while the hematite shell thickness is varied. A detailed optimization of the core and metal shell radii has been performed to achieve highest absorption, and the optimal core and the shell dimension identified are used in Figure 1b (see Supporting Information Figure S2). Considering the carrier diffusion length in hematite as 20 nm7-9, we chose 50 nm as the upper limit to hematite feature size in the CS and CMS structures. A simple exponential model for the photo-generated charges that reach the reaction surface as a function of the hematite thickness can be written as e−t / Ld where t is the thickness and Ld is the minority carrier diffusion length. A thickness of about 46 nm results in only about 10% of the charges generated at one end reaching the other end. Hence we pick 50 nm to roughly be the upper limit for the hematite thickness. Further, the first absorption peak (Figure 1b) in various NWs occurs within 50 nm thick hematite layers. This ensures that the structures considered have charges generated close to the electrolyte-hematite interface and hence can effectively participate in the reaction. All NWs examined demonstrate a general trend of an increase in the integrated absorbed photon flux with increasing hematite feature size until the bulk limit is reached. This trend is expected to be fairly monotonic in the case of thin films. In NWs, the optical resonances (in TM illumination) result in multiple local maxima in the absorption. Due to their anisotropy, solid hematite NWs exhibit substantially higher absorption under TM illumination as compared to TE. In the case of CS and CMS nanowires, utilizing the additional plasmon resonance (in TE illumination) the first local maxima was found to be further enhanced, opening up for potential for enhanced absorption within ultrathin hematite layers. Significantly, the integrated photon flux absorbed within sub-50 nm hematite shells of Si-AlFe2O3 CMS NW is comparable to that of Si-Ag-Fe2O3 CMS NW and outperforms Fe2O3, SiFe2O3 CS and Si-Au-Fe2O3 CMS NWs. Ag-Fe2O3 CS NW exhibits absorption beyond the bulk
ACS Paragon Plus Environment
8
Page 9 of 21
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
limit, however the Al based nanowires come remarkably close to this high absorption. The high absorption enhancement in Al based CMS NWs is indeed comparable to Ag CMS counterparts, which is attributed to the following two factors. First, the outer hematite shell thickness has been chosen such that the LSPR in Al occurs at around 550 nm, similar to that in Ag. This is ideal for enhancing absorption in hematite as will be shown in the subsequent section. Second, although the imaginary component of the dielectric constant for Al is large which would lead to a large absorption within Al, the corresponding real part is large too which leads to reduced electric fields within the Al shell. Since the absorption is proportional to the square of the electric field, the effective absorption within the Al layer is mitigated. Notably, in Si-Au-Fe2O3 CMS NW, the plasmon resonance in Au occurs at a higher wavelength (>550 nm) as compared to Al and Ag based NW of similar diameters. For CMS nanowires with hematite as the photocatalyst, an LSPR greater than 550 nm is unfavorable as it is beyond the hematite absorption range and hence results in poorer absorption enhancement. This is evident from Figure S3 (Supporting Information) plotting the absorbed photon flux as a function of Au layer thickness, in which thinner Au layers with a LSPR occurring closer to 550 nm result in higher performance. However, as the total NW diameter reduces, the absorption under TE polarization reduces due to poorer coupling with light. Together, these effects limit the performance of Au based CMS nanowires with hematite as the photocatalyst. However, for photocatalysts which absorb well beyond 600 nm, Au may be preferable over Ag due to its LSPR in these wavelengths. Further, similar CS and CMS NWs based on Cu (see Supporting Information Figure S3) exhibit a performance remarkably close to Au based counterparts. Therefore, we expect Cu to be a good scalable plasmonic alternative for photoelectrode designs that require Au.
ACS Paragon Plus Environment
9
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 10 of 21
Under ideal conditions, the maximum photocurrent density within the hematite shell of a Si (40 nm dia.)-Al (50 nm)-Fe2O3 (40 nm) CMS NW reaches approximately 11.81 mA/cm2, corresponding to a STH efficiency of 14.5%. This value is about 93% of the theoretical maximum for bulk hematite, indicated by grey dashes in Figure 1(b), and over 20% higher than the best predicted value for ultrathin film hematite resonant light traps with comparable thickness7. We note here that this absorption beyond the bulk limit (grey dashes) is only for single NWs and not for large scale devices. The effect of NW arrays has not been considered in this work and hence further studies are necessary to predict the large scale performance of CMS NW photoelectrodes. Collectively, we have demonstrated that the Al based CMS structures are a promising design to match the high performance expected from the precious metal counterparts and a remarkable improvement over the hematite thin film designs.
ACS Paragon Plus Environment
10
Page 11 of 21
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. The absorption efficiency under TM (dashed lines), TE (dotted lines) and unpolarized (solid lines) illumination as a function of wavelength plotted for individual core and shell layers of (a) a Fe2O3(100 nm dia.), (b) a Al(140 nm dia.) - Fe2O3(40 nm) CS, (c) a Si(40 nm dia.) Al(50 nm) - Fe2O3(40 nm) CMS, (d) a Si(60 nm dia.) - Fe2O3(30 nm) CS, (e) a Ag(80 nm dia.) Fe2O3(30 nm) CS and (f) a Si(40 nm dia.) - Ag(50 nm) - Fe2O3(30 nm) CMS NW. A more detailed analysis on the absorption of each individual layer of the NWs has been performed to understand the contribution of plasmon based absorption enhancement. Figure 2 plots the absorption efficiency within the cores and shells of various NWs as a function of wavelength for transverse electric (TE), transverse magnetic (TM) and unpolarized incidence. For NWs examined here, we used hematite shell thicknesses identified for maximum absorbed photon flux from Figure 1b. Due to their anisotropy, the NWs couple strongly to TM
ACS Paragon Plus Environment
11
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 12 of 21
illumination, resulting in enhanced absorption and optical resonance. In the case of Fe2O3 (100 nm dia.) NW (Figure 2a), the strong optical resonance in TM illumination was found to be around 450 nm, a main contribution to the overall absorption. In the Si(60 nm dia.) - Fe2O3(30 nm) CS NW (Figure 2d), although the thickness of hematite is smaller in comparison to the Fe2O3 (100 nm dia.) NW, the NW has a larger total diameter which results in a shift of the optical resonance under TM illumination to 500 nm. This shift in the resonance leads to stronger absorption for the Si-Fe2O3 CS NW observed in Figure 1b. These results essentially illustrated the advantage of CS structures vs. Fe2O3 NWs that the CS structures offer larger tunability of the optical resonance. For a Al(140 nm dia.) - Fe2O3 (40 nm) CS NW (Figure 2b) and a Ag(80 nm dia.) - Fe2O3(30 nm) CS NW (Figure 2e), in addition to the optical resonance in TM, the metal surface offers an LSPR in TE illumination. The LSPR is around 550 nm in both structures, indicated by the absorption peaks within the metal core for TE illumination. Consequently, the absorption efficiencies observed under TE and TM illuminations are similar. Previous reports on similar metal-dielectric CS structures have demonstrated that such degeneracy under TE and TM illuminations is indicative of Fabry-Pérot like resonance28,
29
. In the case of the Si-Ag-Fe2O3
CMS NW, two absorption peaks in the Ag shell at around 570 and 760 nm under TE illumination are observed, indicating a hybridization of the cavity and the surface plasmon (Figure 2f)40. In contrast, we do not observe two distinct absorption peaks in the Al shell of Si-Al-Fe2O3 CMS NW. Additionally, the absorption in the Al core of Al-Fe2O3 CS NW and Al shell of Si-Al-Fe2O3 CMS NW are identical. These observations indicate the absence of clear plasmon hybridization in Al based CMS NWs. We attribute this to the screening effect due to the large dielectric
ACS Paragon Plus Environment
12
Page 13 of 21
Nano Letters
1 2 3 constant of Al. Similar to the CS NWs, the absorption efficiencies of CMS NWs under TE and 4 5 TM illumination indicate Fabry-Pérot like resonances. 6 7 8 We have demonstrated that new photoelectrode design proposed exhibit enhanced 9 10 absorption. To improve the IQE through reducing recombination, the photon flux absorbed and 11 12 charges generated should be close to the hematite-electrolyte interface. Assuming an unpolarized 13 14 15 AM 1.5G illumination incident perpendicular to the NW axis and using the spatial distribution of 16 17 the absorption efficiency, we calculate the spatial distribution of the photon flux absorbed and 18 19 20 integrate over 300 – 590 nm as follows: 21 22 λ2 =590 nm Im {ε ( r )} ( I ) 2 k λ (I ) 2 (I) 2 ( II ) 2 ( II ) 2 ( II ) 23 χ ϕ r I AM 1.5G ( λ ) 0 2 Er , = + Eϕ + Ez + Er + Eϕ + Ez ( ) abs ,i 24 2r0 E0 2 λ1 =300 nm hPlanck clight 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 Figure 3. The spatial distribution of the absorbed photon flux under unpolarized illumination 56 57 plotted for (a) Fe2O3(100 nm dia.), (b) Si(60 nm dia.) - Fe2O3(30 nm) CS, (c) Ag(80 nm dia.) 58 59 60
∫
( ) ( ) ( )
(
) (
) (
ACS Paragon Plus Environment
13
)
2
dλ
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 14 of 21
Fe2O3(30 nm) CS, (d) Si(40 nm dia.) - Ag(50 nm) - Fe2O3(30 nm) CMS, (e) Al(140 nm dia.) Fe2O3(40 nm) CS and (f) Si(40 nm dia.) - Al(50 nm) - Fe2O3(40 nm) CMS NWs. Figure 3 plots the spatial distribution of the photon flux absorbed for the NW structures evaluated in Figure 2. The color scale has been kept constant for easy comparison. The absorption within hematite shells in the metal based CS and CMS structures all demonstrate a localized field close to the electrolyte interface in comparison with the Fe2O3 NW (Figure 3c-f). These contour plots clearly demonstrate that the absorption is preferentially close to the hematite-electrolyte interface in the CMS design; hence a substantial improvement in the IQE is expected.
Figure 4. The ideal photocurrent density (Blue) and the ratio of the photocurrent density acquired under TM vs. TE illumination (Green) as a function of the incident angle within the hematite shell of a Si(40 nm dia.) - Al(50 nm) - Fe2O3(40 nm) CMS NW. Inset: Schematic of CMS nanowire under oblique incidence. The discussion above has been focusing on the response of CMS NWs under unpolarized illumination normal to the NW axis. The photoelectrode device performance is critically
ACS Paragon Plus Environment
14
Page 15 of 21
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
dependant on the absorption under unpolarized oblique incidence. Figure 4 plots the expected ideal photocurrent density (blue curve) as a function of the angle of incidence α for a Si (40 nm dia.)-Al(50 nm)-Fe2O3(40 nm) CMS NW under unpolarized AM 1.5 G illumination. Remarkably, the ideal photocurrent density is found to be larger than 8.2 mA/cm2, which corresponds to a STH efficiency higher than 10%, for incidence angles as large as 45°. This finding relaxes the requirements on the orientation of the NW arrays with respect to the illumination during operation. The ratio of the ideal photocurrent density under TM illumination to TE illumination (green curve) is plotted in Figure 4 as a function of the incidence angle. As indicated by Fabry-Pérot like resonances in CS and CMS NWs, absorption resulting from the optical resonance in TM and plasmon resonance in TE over 300 nm to 590 nm were found to be similar. This implies that the ratio of the photocurrent density under TM and TE is nearly 1 (i.e. JTM / J TE ≈ 1 ). A small variation of 6% in this ratio was found over the plotted range of illumination angles. Such polarization independence over such a large range of incident angles implies that the CMS nanowires are highly isotropic despite the anisotropy of the NW geometry. This overcomes the critical challenge faced by non-plasmonic NW designs where the anisotropy of the NW structure results in polarization dependant absorption.
In conclusion, we proposed novel Si-Al-Fe2O3 CMS NW structures as photoelectrodes with improved IQE expected from enhanced absorption and reduced recombination. Al has been shown to be an excellent alternative plasmonic material to precious metals in CMS structures and Si-Al-Fe2O3 CMS NWs exhibit strong absorption in sub-50 nm hematite shells due to the Al plasmon resonance. The maximum photocurrent density calculated assuming ideal conditions
ACS Paragon Plus Environment
15
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 16 of 21
within the hematite shell of Si (40 nm dia.)-Al (50 nm)-Fe2O3 (40 nm) CMS NW reaches approximately 11.81 mA/cm2 (i.e. STH efficiency of 14.5%). The absorption efficiency within the 40 nm hematite shell is about 93% of the theoretical maximum for bulk hematite. The spatial distribution of the absorbed photon flux is highly localized within the hematite shell so that the charges are generated close to electrolyte interface and expected to exhibit minimal recombination before reaching the interface. Although the NW geometry is highly anisotropic, the CMS NWs exhibit polarization independent absorption over all incidence angles. Further, a high photocurrent density, under ideal conditions, is predicted for incidence angles as large as 45° which relaxes the requirements on the orientation of NW arrays with respect to the incident light. Therefore, we believe that the CMS NW photoelectrodes integrate plasmonic absorption enhancement and efficient charge collection into a single device while ensuring cost effectiveness and scalability. Significantly, the Al based CMS design is general and expected to provide similar absorption enhancement in other photocatalyst materials beyond hematite. Notably, the Si-Hematite ‘dual absorber’ system has been shown to be an effective strategy to reduce the overpotential required for hydrogen evolution32,
33
. The results presented here for
individual NWs indicate that the absorption within the Si core in the individual CMS NWs is negligible. By varying the Al thickness, we note that there exists a minimum thickness for the Al shell (about 25 nm) for obtaining the high absorption in the hematite shell but the Si core still does not show appreciable absorption (see Supporting Information Figure 4S). This is mainly due to the screening of electric field by the metal intermediate shell. In a practical device consisting of CMS NW arrays on a Si wafer, the absorption within Si cores and the wafer itself may be substantial. Therefore, we expect the photoelectrodes based on CMS arrays to potentially offer the additional advantage of reduced overpotential.
ACS Paragon Plus Environment
16
Page 17 of 21
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
ASSOCIATED CONTENT
Supporting Information. Rigorous solution to Maxwell’s Equations derived for CMS NW under oblique illumination. Identifying ideal NW core diameters and shell thicknesses for highest integrated photon flux absorption. Effect of metal thickness on the ideal photocurrent density studied in CS and CMS NWs. This material is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author *Corresponding Author:
[email protected] Author Contributions C.Y and S.R conceived the idea and C.Y provided guidance throughout the project. S.R did the Mie scattering analysis and computed the integrated photon flux absorbed its spatial distribution in various CS and CMS NWs under normal incidence. S.R and T.L extended this analysis to CMS NWs under oblique incidence. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. The authors declare no competing financial interests.
ACKNOWLEDGMENT S. R. thanks the support from National Science Foundation ECCS 1118934.
ACS Paragon Plus Environment
17
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 18 of 21
ABBREVIATIONS NW, nanowire; CS, core-shell; CMS, core-multishell; TE, transverse electric; TM, transverse magnetic; AM 1.5G, air mass 1.5 global
REFERENCES 1.
Bak, T.; Nowotny, J.; Rekas, M.; Sorrell, C. Int. J. Hydrogen Energy 2002, 27, 991-1022.
2.
van de Krol, R.; Liang, Y.; Schoonman, J. J. Mater. Chem. 2008, 18, 2311.
3.
Sun, J.; Zhong, D. K.; Gamelin, D. R. Energy Environ. Sci. 2010, 3, 1252.
4.
Chen, Z.; Jaramillo, T. F.; Deutsch, T. G.; Kleiman-Shwarsctein, A.; Forman, A. J.;
Gaillard, N.; Garland, R.; Takanabe, K.; Heske, C.; Sunkara, M.; McFarland, E. W.; Domen, K.; Miller, E. L.; Turner, J. a.; Dinh, H. N. J. Mater. Res. 2011, 25, 3-16. 5.
Sivula, K.; Le Formal, F.; Grätzel, M. ChemSusChem 2011, 4, 432-49.
6.
Katz, M. J.; Riha, S. C.; Jeong, N. C.; Martinson, A. B. F.; Farha, O. K.; Hupp, J. T.
Coord. Chem. Rev. 2012, 256, 2521-2529. 7.
Dotan, H.; Kfir, O.; Sharlin, E.; Blank, O.; Gross, M.; Dumchin, I.; Ankonina, G.;
Rothschild, A. Nat. Mater. 2013, 12, 158-64. 8.
Dare-Edwards, M.; Goodenough, J.; Hamnett, A.; Trevellick, P. J. Chem. Soc. Faraday
Trans. 1983, 79, 2027-2041. 9.
Dotan, H.; Sivula, K.; Grätzel, M.; Rothschild, A.; Warren, S. C. Energy Environ. Sci.
2011, 4, 958. 10.
Kim, J. Y.; Magesh, G.; Youn, D. H.; Jang, J.-W.; Kubota, J.; Domen, K.; Lee, J. S. Sci.
Rep. 2013, 3, 2681.
ACS Paragon Plus Environment
18
Page 19 of 21
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
11.
Cao, L.; White, J. S.; Park, J.-S.; Schuller, J. a.; Clemens, B. M.; Brongersma, M. L. Nat.
Mater. 2009, 8, 643-7. 12.
Cao, L.; Fan, P.; Vasudev, A. P.; White, J. S.; Yu, Z.; Cai, W.; Schuller, J. a.; Fan, S.;
Brongersma, M. L. Nano Lett. 2010, 10, 439-45. 13.
Yu, Y.; Ferry, V. E.; Alivisatos, a. P.; Cao, L. Nano Lett. 2012, 12, 3674-81.
14.
Thimsen, E.; Le Formal, F.; Grätzel, M.; Warren, S. C. Nano Lett. 2011, 11, 35-43.
15.
Thomann, I.; Pinaud, B. a.; Chen, Z.; Clemens, B. M.; Jaramillo, T. F.; Brongersma, M.
L. Nano Lett. 2011, 11, 3440-6. 16.
Gao, H.; Liu, C.; Jeong, H. E.; Yang, P. ACS Nano 2012, 6, 234-40.
17.
Li, J.; Cushing, S. K.; Zheng, P.; Meng, F.; Chu, D.; Wu, N. Nat. Commun. 2013, 4,
2651. 18.
Wang, X.; Peng, K.; Hu, Y.; Zhang, F.; Hu, B. Nano Lett. 2013, 1-8.
19.
Riha, S. C.; Klahr, B. M.; Tyo, E. C.; Seifert, S.; Vajda, S.; Pellin, M. J.; Hamann, T. W.;
Martinson, A. B. F. ACS Nano 2013, 7, 2396-405. 20.
Qiu, Y.; Leung, S.-F.; Zhang, Q.; Hua, B.; Lin, Q.; Wei, Z.; Tsui, K.-H.; Zhang, Y.;
Yang, S.; Fan, Z. Nano Lett. 2014. 21.
Schuller, J. a.; Barnard, E. S.; Cai, W.; Jun, Y. C.; White, J. S.; Brongersma, M. L. Nat.
Mater. 2010, 9, 193-204. 22.
Cushing, S.; Li, J.; Meng, F. J. Am. Chem. Soc. 2012, 134, 15033-15041.
23.
Linic, S.; Christopher, P.; Ingram, D. B. Nat. Mater. 2011, 10, 911-21.
24.
Warren, S. C.; Thimsen, E. Energy Environ. Sci. 2012, 5, 5133.
25.
Mubeen, S.; Lee, J.; Singh, N.; Krämer, S.; Stucky, G. D.; Moskovits, M. Nat.
Nanotechnol. 2013, 8, 247-51.
ACS Paragon Plus Environment
19
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 20 of 21
26.
Mukherjee, S.; Libisch, F.; Large, N. Nano Lett. 2013, 13, 240-247.
27.
Zhan, Y.; Zhao, J.; Zhou, C.; Alemayehu, M.; Li, Y.; Li, Y. Opt. Express 2012, 20,
11506-16. 28.
Mann, S. A.; Garnett, E. C. Nano Lett. 2013, 13, 3173-3178.
29.
Zhou, L.; Yu, X.; Zhu, J. Nano Lett. 2014.
30.
Knight, M. W.; Liu, L.; Wang, Y.; Brown, L.; Mukherjee, S.; King, N. S.; Everitt, H. O.;
Nordlander, P.; Halas, N. J. Nano Lett. 2012, 12, 6000-4. 31.
Knight, M. W.; King, N. S.; Liu, L.; Everitt, H. O.; Nordlander, P.; Halas, N. J. ACS
Nano 2013. 32.
Mayer, M. T.; Du, C.; Wang, D. J. Am. Chem. Soc. 2012, 134, 12406-9.
33.
Du, C.; Yang, X.; Mayer, M. T.; Hoyt, H.; Xie, J.; McMahon, G.; Bischoping, G.; Wang,
D. Angew. Chem. Int. Ed. 2013, 52, 12692-12695. 34.
van de Krol, R.; Liang, Y. Chimia 2013, 67, 168-71.
35.
Kerker, M.; Matijevic, E. J. Opt. Soc. Am. 1961, 51, 506-508.
36.
Lind, A. C. J. Appl. Phys. 1966, 37, 3195.
37.
Shah, G. Mon. Not. R. Astr. Soc. 1970, 148, 93-102.
38.
Bohren, C.; Huffman, D., Absorption and scattering of light by small particles. Wiley-
VCH: 2004. 39.
Ni, X.; Liu, Z.; Kildishev, A., PhotonicsDB: Optical Constants. 2010.
40.
Ramadurgam, S.; Yang, C. Sci. Rep. 2014, 4, 1-7.
ACS Paragon Plus Environment
20
Page 21 of 21
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
TOC
ACS Paragon Plus Environment
21