Co-reporter:Fadl H. Saadi, Azhar I. Carim, Walter S. Drisdell, Sheraz Gul, Jack H. Baricuatro, Junko Yano, Manuel P. Soriaga, and Nathan S. Lewis
Journal of the American Chemical Society September 20, 2017 Volume 139(Issue 37) pp:12927-12927
Publication Date(Web):August 28, 2017
DOI:10.1021/jacs.7b07606
Transition metal phosphides exhibit high catalytic activity toward the electrochemical hydrogen-evolution reaction (HER) and resist chemical corrosion in acidic solutions. For example, an electrodeposited CoP catalyst exhibited an overpotential, η, of −η < 100 mV at a current density of −10 mA cm–2 in 0.500 M H2SO4(aq). To obtain a chemical description of the material as-prepared and also while effecting the HER in acidic media, such electrocatalyst films were investigated using Raman spectroscopy and X-ray absorption spectroscopy both ex situ as well as under in situ and operando conditions in 0.500 M H2SO4(aq). Ex situ analysis using the tandem spectroscopies indicated the presence of multiple ordered and disordered phases that contained both near-zerovalent and oxidized Co species, in addition to reduced and oxygenated P species. Operando analysis indicated that the active electrocatalyst was primarily amorphous and predominantly consisted of near-zerovalent Co as well as reduced P.
Co-reporter:David Knapp;Bruce S. Brunschwig
The Journal of Physical Chemistry C July 22, 2010 Volume 114(Issue 28) pp:12300-12307
Publication Date(Web):2017-2-22
DOI:10.1021/jp101375x
The use of Ge in semiconductor electronics has been constrained by the lack of a simple method of passivating the crystal surface. Toward that end, we have explored the utility of chemically bonded hydrocarbon monolayers. Alkylated Ge(111) surfaces have been prepared by addition of 1-alkenes to the H-terminated Ge(111) surface as well as by a two-step halogenation/alkylation procedure. The chemical compositions of the resulting methyl-, ethyl-, and decyl-terminated surfaces have been evaluated using X-ray photoelectron spectroscopy (XPS). Thermal addition of 1-decene produced hydrophobic surfaces with 0.3 ± 0.1 monolayer of Ge oxide detected by XPS, whereas no oxide was observed on the methyl-, ethyl-, or decyl-terminated surfaces that were prepared using the two-step halogenation/alkylation method. Methyl-terminated Ge(111) surfaces prepared by the two-step method displayed a well-resolved C 1s XPS peak at a binding energy of 284 eV, consistent with carbon bonded to a less electronegative element such as Ge. The electronic properties of all of the alkylated surfaces were characterized by measurements of the surface recombination velocity as a function of an externally applied gate voltage. Treatment of HF-etched Ge(111) surfaces with Br2 vapor, followed by reaction with alkylmagnesium or alkyllithium reagents, yielded air-stable surfaces that had surface recombination velocities of 100 cm s−1 or less under flat-band conditions. The field-dependent surface recombination velocity experiments indicated that, in contact with air, methyl-terminated n-type Ge(111) samples had a negative surface potential approaching 300 mV, in contrast to the oxidized Ge(111) surface, which exhibited a strongly positive surface potential under the same conditions. Mercury contacts to n-type methyl-, ethyl-, or decyl-terminated Ge(111) substrates that were alkylated using the two-step method formed rectifying junctions with barrier heights of 0.6 ± 0.1 eV, whereas no measurable rectification was observed for Hg contacts to p-type Ge(111) substrates that were alkylated by the two-step method, to n-type Ge(111) substrates that were alkylated through addition of 1-decene, or to oxidized n-type Ge(111) samples.
Co-reporter:Patrick T. Hurley;Erik Johansson;Bruce S. Brunschwig
The Journal of Physical Chemistry C August 27, 2009 Volume 113(Issue 34) pp:15239-15245
Publication Date(Web):2017-2-22
DOI:10.1021/jp901792y
Fourier transform infrared (FTIR) spectroscopy was used to investigate C2H5−Si(111) surfaces prepared using a chlorination/alkylation method. After alkylation, in addition to ethyl groups, such surfaces showed the presence of hydrogen bonded to atop silicon surface atoms. Systematic isotopic substitution of protic solvents and reagents with their fully or partially deuterated counterparts revealed the origin of the surface-bound hydrogen on the C2H5−Si(111) surfaces. The presence or absence of the Si−H stretch at 2080 cm−1 and the Si−D stretch at ∼1510 cm−1, respectively, indicated that the hydrogen originated from the methyl group of the ethyl Grignard reagent.
Co-reporter:Roc Matheu, Ivan A. Moreno-Hernandez, Xavier Sala, Harry B. Gray, Bruce S. Brunschwig, Antoni Llobet, and Nathan S. Lewis
Journal of the American Chemical Society August 23, 2017 Volume 139(Issue 33) pp:11345-11345
Publication Date(Web):August 7, 2017
DOI:10.1021/jacs.7b06800
A hybrid photoanode based on a molecular water oxidation precatalyst was prepared from TiO2-protected n- or p+-Si coated with multiwalled carbon nanotubes (CNT) and the ruthenium-based water oxidation precatalyst [RuIV(tda)(py-pyr)2(O)], 1(O) (tda2– is [2,2′:6′,2″-terpyridine]-6,6″-dicarboxylato and py-pir is 4-(pyren-1-yl)-N-(pyridin-4-ylmethyl)butanamide). The Ru complex was immobilized by π–π stacking onto CNTs that had been deposited by drop casting onto Si electrodes coated with 60 nm of amorphous TiO2 and 20 nm of a layer of sputtered C. At pH = 7 with 3 Sun illumination, the n-Si/TiO2/C/CNT/[1+1(O)] electrodes exhibited current densities of 1 mA cm–2 at 1.07 V vs NHE. The current density was maintained for >200 min at a constant potential while intermittently collecting voltammograms that indicated that over half of the Ru was still in molecular form after O2 evolution.
Co-reporter:Michael F. Lichterman, Matthias H. Richter, Bruce S. Brunschwig, Nathan S. Lewis, Hans-Joachim Lewerenz
Journal of Electron Spectroscopy and Related Phenomena 2017 Volume 221(Volume 221) pp:
Publication Date(Web):1 November 2017
DOI:10.1016/j.elspec.2017.03.011
•The electrochemical double layer of KOH electrolytes on Ir was investigated by operando XPS.•Debye-Hückel approximations were used to model expected shifts in the XPS spectra.•A general approach for future investigations is described.Tender X-ray operando photoemission spectroscopy has been used to directly analyze the energetics of the double layer at a metal-water interface in a dilute electrolyte having a Debye length of several nanometers. The data are compared to a theoretical evaluation of the potential of the solution near the electrode. Due to its noble nature, Ir was chosen as a working electrode material, and KOH(aq) at varied concentrations and thicknesses constituted the electrolyte. Shifts in peak width and binding energy of the water O 1s core level were analyzed by modeling based on Debye-Hückel approximations. The data are consistent with electrochemical formulations of the double layer that provide a foundation to electrochemistry.
Co-reporter:Sakineh Chabi;Kimberly M. Papadantonakis;Michael S. Freund
Energy & Environmental Science (2008-Present) 2017 vol. 10(Issue 6) pp:1320-1338
Publication Date(Web):2017/06/14
DOI:10.1039/C7EE00294G
Membrane-based architectures enable optimization of charge transport and electrochemical potential gradients in artificial photosynthesis. Spatial integration of the membrane-bound components reduces the impact of charge recombination and can reduce electrical resistances associated with ionic and electronic transport processes. In addition to eliminating the need for external electrical circuits, a membrane-based architecture also ensures separation of energetic products, thereby preventing the formation of potentially dangerous fuel/oxidant mixtures. Membrane-based structures may also be coupled with other devices, such as perovskite-based solar cells, to further benefit solar fuel production. This review discusses the key roles that various different types of membranes play in artificial photosynthetic systems.
Co-reporter:Ivan A. Moreno-Hernandez;Clara A. MacFarland;Carlos G. Read;Kimberly M. Papadantonakis;Bruce S. Brunschwig
Energy & Environmental Science (2008-Present) 2017 vol. 10(Issue 10) pp:2103-2108
Publication Date(Web):2017/10/11
DOI:10.1039/C7EE01486D
Water oxidation is a required half-reaction for electrochemical water splitting. To date, the only well-established active oxygen-evolution catalysts stable under operating conditions and at rest in acidic aqueous media contain Ru or Ir, two of the scarcest non-radioactive elements on Earth. We report herein a nickel-manganese antimonate electrocatalyst with a rutile-type crystal structure that requires an initial voltammetric overpotential of 672 ± 9 mV to catalyze the oxidation of water to O2(g) at a rate corresponding to 10 mA cm−2 of current density when operated in contact with 1.0 M sulfuric acid. Under galvanostatic control, the overpotential initially rose from 670 mV but was then stable at 735 ± 10 mV for 168 h of continuous operation at 10 mA cm−2. We additionally provide an in-depth evaluation of the stability of the nickel-manganese antimonate electrocatalyst, including elemental characterization of the surface, bulk, and electrolyte before and after electrochemical operation.
Co-reporter:Ke Sun;Ivan A. Moreno-Hernandez;William C. Schmidt, Jr.;Xinghao Zhou;J. Chance Crompton;Rui Liu;Fadl H. Saadi;Yikai Chen;Kimberly M. Papadantonakis
Energy & Environmental Science (2008-Present) 2017 vol. 10(Issue 4) pp:987-1002
Publication Date(Web):2017/04/12
DOI:10.1039/C6EE03563A
The in situ optical properties and electrocatalytic performance of representative catalysts for the oxygen-evolution reaction (OER) have been considered together to evaluate system-level effects that accompany the integration of OER catalysts into a solar-fuel device driven by a tandem-junction light absorber with a photoanode top cell, i.e., a design that requires incident light to be transmitted through the OER catalyst before reaching a semiconducting light absorber. The relationship between the overpotential and optical transmission of the catalysts determined the attainable efficiencies for integrated solar-fuel devices as well as the optimal band gaps for the photoanode in such devices. The systems investigated generally showed: (1) the OER catalysts dissolved in acid, and were less stable in buffered near-neutral pH electrolytes than in strongly alkaline electrolytes; (2) higher overpotentials were required to drive the OER at a specified current density when the catalysts were operated in contact with near-neutral pH electrolytes than strong alkaline electrolytes; (3) for some of the OER catalysts, the electrocatalytic activity and in situ absorption spectra depended strongly on the preparation method; (4) increasing the loading of the electrocatalyst reduced the overpotential and the optical transmission; (5) for the catalysts studied, the optical transmission and overpotential were generally correlated, and the trend lines did not cross, indicating that based on these factors alone, the optimal approach is to use lower loadings of highly active catalysts, rather than to use a less active but more transparent catalysts; (6) for a solar-fuel device driven by semiconductors operating at the Shockley–Queisser limit and using a continuous film of a given OER catalyst in the path of incident light, the efficiency decrease due to the reduced optical transmittance that accompanies increased OER catalyst loading can be substantially greater than any efficiency increase that might be gained through the reduction in catalytic overpotential by increasing the catalyst loading; and (7) HER catalysts possessed the same performance trade-off when the light is incident through the HER catalysts as is observed for OER catalysts when the light is incident from the OER side.
Co-reporter:Meenesh R. Singh;Chengxiang Xiang
Sustainable Energy & Fuels (2017-Present) 2017 vol. 1(Issue 3) pp:458-466
Publication Date(Web):2017/05/03
DOI:10.1039/C7SE00062F
The electrochemical performance of three different types of membrane-containing electrolyte-flow schemes for solar-driven water splitting has been studied quantitatively using 1-dimensional and 2-dimensional multi-physics models. The three schemes include a recirculation scheme with a well-mixed bulk electrolyte, a recirculation scheme with laminar flow fields, and a fresh-feed scheme with laminar flow fields. The Nernstian potential loss associated with pH gradients at the electrode surfaces, the resistive loss between the cathode and anode, the product-gas crossovers, and the required pumping energy in all three schemes have been evaluated as a function of the operational current density, the flow rates for the electrolyte, and the physical dimensions of the devices. The trade-offs in the voltage loss, safety considerations, and energy inputs from the balance-of-systems required to produce a practical device have been evaluated and compared to membrane-free devices as well as to devices that operate at extreme pH values.
Co-reporter:Stefan T. OmelchenkoYulia Tolstova, Harry A. Atwater, Nathan S. Lewis
ACS Energy Letters - New in 2016 2017 Volume 2(Issue 2) pp:
Publication Date(Web):January 19, 2017
DOI:10.1021/acsenergylett.6b00704
Excitonic effects account for a fundamental photoconversion and charge transport mechanism in Cu2O; hence, the universally adopted “free carrier” model substantially underestimates the photovoltaic efficiency for such devices. The quasi-equilibrium branching ratio between excitons and free carriers in Cu2O indicates that up to 28% of photogenerated carriers during photovoltaic operation are excitons. These large exciton densities were directly observed in photoluminescence and spectral response measurements. The results of a device physics simulation using a model that includes excitonic effects agree well with experimentally measured current–voltage characteristics of Cu2O-based photovoltaics. In the case of Cu2O, the free carrier model underestimates the efficiency of a Cu2O solar cell by as much as 1.9 absolute percent at room temperature.
Co-reporter:Noah T. Plymale, Mita Dasog, Bruce S. Brunschwig, and Nathan S. Lewis
The Journal of Physical Chemistry C 2017 Volume 121(Issue 8) pp:
Publication Date(Web):January 18, 2017
DOI:10.1021/acs.jpcc.6b11555
H–Si(111) surfaces have been reacted with liquid methanol (CH3OH) in the absence or presence of a series of oxidants and/or illumination. Oxidant-activated methoxylation of H–Si(111) surfaces was observed in the dark after exposure to CH3OH solutions that contained the one-electron oxidants acetylferrocenium, ferrocenium, or 1,1′-dimethylferrocenium. The oxidant-activated reactivity toward CH3OH of intrinsic and n-type H–Si(111) surfaces increased upon exposure to ambient light. The results suggest that oxidant-activated methoxylation requires that two conditions be met: (1) the position of the quasi-Fermi levels must energetically favor oxidation of the H–Si(111) surface and (2) the position of the quasi-Fermi levels must energetically favor reduction of an oxidant in solution. Consistently, illuminated n-type H–Si(111) surfaces underwent methoxylation under applied external bias more rapidly and at more negative potentials than p-type H–Si(111) surfaces. The results under potentiostatic control indicate that only conditions that favor oxidation of the H–Si(111) surface need be met, with charge balance at the surface maintained by current flow at the back of the electrode. The results are described by a mechanistic framework that analyzes the positions of the quasi-Fermi levels relative to the energy levels relevant for each system.
Co-reporter:Nathan S. Lewis
Science 2016 Vol 351(6271) pp:
Publication Date(Web):22 Jan 2016
DOI:10.1126/science.aad1920
Relying more on the Sun
Improved technologies for harnessing solar energy are not limited to creating more efficient solar cells. The associated hardware of delivering power from solar cells to homes and businesses, and storing this intermittent resource on the grid, offer R&D opportunities. Lewis reviews the status of these areas, as well as solar thermal and solar fuels approaches for harnessing solar energy.
Science, this issue p. 10.1126/science.aad1920
Co-reporter:Xinghao Zhou, Rui Liu, Ke Sun, Kimberly M. Papadantonakis, Bruce S. Brunschwig and Nathan S. Lewis
Energy & Environmental Science 2016 vol. 9(Issue 3) pp:892-897
Publication Date(Web):08 Jan 2016
DOI:10.1039/C5EE03655K
Heterojunction photoanodes, consisting of n-type crystalline Si(100) substrates coated with a thin ∼50 nm film of cobalt oxide fabricated using atomic-layer deposition (ALD), exhibited photocurrent-onset potentials of −205 ± 20 mV relative to the formal potential for the oxygen-evolution reaction (OER), ideal regenerative solar-to-O2(g) conversion efficiencies of 1.42 ± 0.20%, and operated continuously for over 100 days (∼2500 h) in 1.0 M KOH(aq) under simulated solar illumination. The ALD CoOx thin film: (i) formed a heterojunction with the n-Si(100) that provided a photovoltage of 575 mV under 1 Sun of simulated solar illumination; (ii) stabilized Si photoanodes that are otherwise unstable when operated in aqueous alkaline electrolytes; and, (iii) catalyzed the oxidation of water, thereby reducing the kinetic overpotential required for the reaction and increasing the overall efficiency relative to electrodes that do not have an inherently electrocatalytic coating. The process provides a simple, effective method for enabling the use of planar n-Si(100) substrates as efficient and durable photoanodes in fully integrated, photovoltaic-biased solar fuels generators.
Co-reporter:Jesus M. Velazquez, Jimmy John, Daniel V. Esposito, Adam Pieterick, Ragip Pala, Guofeng Sun, Xinghao Zhou, Zhuangqun Huang, Shane Ardo, Manuel P. Soriaga, Bruce S. Brunschwig and Nathan S. Lewis
Energy & Environmental Science 2016 vol. 9(Issue 1) pp:164-175
Publication Date(Web):08 Oct 2015
DOI:10.1039/C5EE02530C
The spatial variation in the photoelectrochemical performance for the reduction of an aqueous one-electron redox couple, Ru(NH3)63+/2+, and for the evolution of H2(g) from 0.5 M H2SO4(aq) at the surface of bare or Pt-decorated p-type WSe2 photocathodes has been investigated in situ using scanning photocurrent microscopy (SPCM). The measurements revealed significant differences in the charge-collection performance (quantified by the values of external quantum yields, Φext) on various macroscopic terraces. Local spectral response measurements indicated a variation in the local electronic structure among the terraces, which was consistent with a non-uniform spatial distribution of sub-band-gap states within the crystals. The photoconversion efficiencies of Pt-decorated p-WSe2 photocathodes were greater for the evolution of H2(g) from 0.5 M H2SO4 than for the reduction of Ru(NH3)63+/2+, and terraces that exhibited relatively low values of Φext for the reduction of Ru(NH3)63+/2+ could in some cases yield values of Φext for the evolution of H2(g) comparable to the values of Φext yielded by the highest-performing terraces. Although the spatial resolution of the techniques used in this work frequently did not result in observation of the effect of edge sites on photocurrent efficiency, some edge effects were observed in the measurements; however the observed edge effects differed among edges, and did not appear to determine the performance of the electrodes.
Co-reporter:Ke Sun;Rui Liu;Yikai Chen;Erik Verlage;Chengxiang Xiang
Advanced Energy Materials 2016 Volume 6( Issue 13) pp:
Publication Date(Web):
DOI:10.1002/aenm.201600379
Co-reporter:Mita Dasog, Azhar I. Carim, Sisir Yalamanchili, Harry A. Atwater, and Nathan S. Lewis
Nano Letters 2016 Volume 16(Issue 8) pp:5015-5021
Publication Date(Web):June 20, 2016
DOI:10.1021/acs.nanolett.6b01782
Au was photoelectrochemically deposited onto cylindrical or tapered p-Si microwires on Si substrates to profile the photoinduced charge-carrier generation in individual wires in a photoactive semiconductor wire array. Similar experiments were repeated for otherwise identical Si microwires doped to be n-type. The metal plating profile was conformal for n-type wires, but for p-type wires was a function of distance from the substrate and was dependent on the illumination wavelength. Spatially resolved charge-carrier generation profiles were computed using full-wave electromagnetic simulations, and the localization of the deposition at the p-type wire surfaces observed experimentally correlated well with the regions of enhanced calculated carrier generation in the volumes of the microwires. This technique could potentially be extended to determine the spatially resolved carrier generation profiles in a variety of mesostructured, photoactive semiconductors.Keywords: carrier generation; electrodeposition; microwires; photoelectrochemistry; Semiconductor; silicon;
Co-reporter:Azhar I. Carim, Nicolas A. Batara, Anjali Premkumar, Richard May, Harry A. Atwater, and Nathan S. Lewis
Nano Letters 2016 Volume 16(Issue 5) pp:2963-2968
Publication Date(Web):March 16, 2016
DOI:10.1021/acs.nanolett.5b04999
Highly anisotropic and ordered nanoscale lamellar morphologies can be spontaneously generated over macroscopic areas, without the use of a photomask or any templating agents, via the photoelectrodeposition of Se–Te alloy films. To form such structures, the light source can be a single, linearly polarized light source that need not necessarily be highly coherent. In this work, the variation in the morphologies produced by this deposition process was evaluated in response to differences in the coherence and relative phase between multiple simultaneous linearly polarized illumination inputs. Specifically, the morphologies of photoelectrodeposits were evaluated when two tandem same-wavelength sources with discrete linear polarizations, both either mutually incoherent or mutually coherent (with defined phase differences), were used. Additionally, morphologies were simulated via computer modeling of the interfacial light scattering and absorption during the photoelectrochemical growth process. The morphologies that were generated using two coherent, in-phase sources were equivalent to those generated using only a single source. In contrast, the use of two coherent, out-of-phase sources produced a range of morphological patterns. For small out-of-phase addition of orthogonal polarization components, lamellar-type patterns were observed. When fully out-of-phase orthogonal sources (circular polarization) were used, an isotropic, mesh-type pattern was instead generated, similar to that observed when unpolarized illumination was utilized. In intermediate cases, anisotropic lamellar-type patterns were superimposed on the isotropic mesh-type patterns, and the relative height between the two structures scaled with the amount of out-of-phase addition of the orthogonal polarization components. Similar results were obtained when two incoherent sources were utilized. In every case, the long axis of the lamellar-type morphology component aligned parallel to the intensity-weighted average polarization orientation. The observations consistently agreed with computer simulations, indicating that the observed morphologies were fully determined by the nature of the illumination utilized during the growth process. The collective data thus indicated that the photoelectrodeposition process exhibits sensitivity toward the coherency, relative phase, and polarization orientations of all optical inputs and that this sensitivity is physically expressed in the morphology of the deposit.
Co-reporter:Adam C. Nielander, Annelise C. Thompson, Christopher W. Roske, Jacqueline A. Maslyn, Yufeng Hao, Noah T. Plymale, James Hone, and Nathan S. Lewis
Nano Letters 2016 Volume 16(Issue 7) pp:4082-4086
Publication Date(Web):June 20, 2016
DOI:10.1021/acs.nanolett.6b00773
The behavior of n-Si(111) photoanodes covered by monolayer sheets of fluorinated graphene (F–Gr) was investigated under a range of chemical and electrochemical conditions. The electrochemical behavior of n-Si/F–Gr and np+-Si/F–Gr photoanodes was compared to hydride-terminated n-Si (n-Si−H) and np+-Si−H electrodes in contact with aqueous Fe(CN)63-/4- and Br2/HBr electrolytes as well as in contact with a series of outer-sphere, one-electron redox couples in nonaqueous electrolytes. Illuminated n-Si/F–Gr and np+-Si/F–Gr electrodes in contact with an aqueous K3(Fe(CN)6/K4(Fe(CN)6 solutions exhibited stable short-circuit photocurrent densities of ∼10 mA cm–2 for 100,000 s (>24 h), in comparison to bare Si electrodes, which yielded nearly a complete photocurrent decay over ∼100 s. X-ray photoelectron spectra collected before and after exposure to aqueous anodic conditions showed that oxide formation at the Si surface was significantly inhibited for Si electrodes coated with F–Gr relative to bare Si electrodes exposed to the same conditions. The variation of the open-circuit potential for n-Si/F–Gr in contact with a series of nonaqueous electrolytes of varying reduction potential indicated that the n-Si/F–Gr did not form a buried junction with respect to the solution contact. Further, illuminated n-Si/F−Gr electrodes in contact with Br2/HBr(aq) were significantly more electrochemically stable than n-Si−H electrodes, and n-Si/F−Gr electrodes coupled to a Pt catalyst exhibited ideal regenerative cell efficiencies of up to 5% for the oxidation of Br– to Br2.
Co-reporter:Daniel A. Torelli, Sonja A. Francis, J. Chance Crompton, Alnald Javier, Jonathan R. Thompson, Bruce S. Brunschwig, Manuel P. Soriaga, and Nathan S. Lewis
ACS Catalysis 2016 Volume 6(Issue 3) pp:2100
Publication Date(Web):February 17, 2016
DOI:10.1021/acscatal.5b02888
We report the electrocatalytic reduction of CO2 to the highly reduced C2 products, ethylene and ethane, as well as to the fully reduced C1 product, methane, on three different phases of nickel–gallium (NiGa, Ni3Ga, and Ni5Ga3) films prepared by drop-casting. In aqueous bicarbonate electrolytes at neutral pH, the onset potential for methane, ethylene, and ethane production on all three phases was found to be −0.48 V versus the reversible hydrogen electrode (RHE), among the lowest onset potentials reported to date for the production of C2 products from CO2. Similar product distributions and onset potentials were observed for all three nickel–gallium stoichiometries tested. The onset potential for the reduction of CO2 to C2 products at low current densities catalyzed by nickel–gallium was >250 mV more positive than that of polycrystalline copper, and approximately equal to that of single crystals of copper, which have some of the lowest overpotentials to date for the reduction of CO2 to C2 products and methane. The nickel–gallium films also reduced CO to ethylene, ethane, and methane, consistent with a CO2 reduction mechanism that first involves the reduction of CO2 to CO. Isotopic labeling experiments with 13CO2 confirmed that the detected products were produced exclusively by the reduction of CO2.Keywords: C2 production; CO2 reduction; electrocatalysis; ethane; low overpotential; methane; nickel gallium; NiGa
Co-reporter:Juan F. Callejas, Carlos G. Read, Christopher W. Roske, Nathan S. Lewis, and Raymond E. Schaak
Chemistry of Materials 2016 Volume 28(Issue 17) pp:6017
Publication Date(Web):August 7, 2016
DOI:10.1021/acs.chemmater.6b02148
Hydrogen gas obtained by the electrolysis of water has long been proposed as a clean and sustainable alternative to fossil fuels. Noble metals such as Pt are capable of splitting water at low overpotentials, but the implementation of inexpensive solar-driven water-splitting systems and electrolyzers could benefit from the development of robust, efficient, and abundant alternatives to noble metal catalysts. Transition metal phosphides (MxPy) have recently been identified as a promising family of Earth abundant electrocatalysts for the hydrogen-evolution reaction (HER) and are capable of operating with low overpotentials at operationally relevant current densities while exhibiting stability under strongly acidic conditions. In this review, we highlight the progress that has been made in this field and provide insights into the synthesis, characterization, and electrochemical behavior of transition metal phosphides as HER electrocatalysts. We also discuss strategies for the incorporation of metal phosphides into integrated solar-driven water-splitting systems and highlight key considerations involved in the testing and benchmarking of such devices.
Co-reporter:Chengxiang Xiang, Kimberly M. Papadantonakis and Nathan S. Lewis
Materials Horizons 2016 vol. 3(Issue 3) pp:169-173
Publication Date(Web):12 Feb 2016
DOI:10.1039/C6MH00016A
Efforts to develop renewable sources of carbon-neutral fuels have brought a renewed focus to research and development of sunlight-driven water-splitting systems. Electrolysis of water to produce H2 and O2 gases is the foundation of such systems, is conceptually and practically simple, and has been practiced both in the laboratory and industrially for many decades. In this Focus article, we present the fundamentals of water splitting and describe practices which distinguish commercial water-electrolysis systems from simple laboratory-scale demonstrations.
Co-reporter:Hans-Joachim Lewerenz, Michael F. Lichterman, Matthias H. Richter, Ethan J. Crumlin, Shu Hu, Stephanus Axnanda, Marco Favaro, Walter Drisdell, Zahid Hussain, Bruce S. Brunschwig, Zhi Liu, Anders Nilsson, Alexis T. Bell, Nathan S. Lewis, Daniel Friebel
Electrochimica Acta 2016 Volume 211() pp:711-719
Publication Date(Web):1 September 2016
DOI:10.1016/j.electacta.2016.06.006
Operando synchrotron radiation photoelectron spectroscopy in the tender X-ray energy range has been used to obtain information on the energy-band relations of semiconductor and metal-covered semiconductor surfaces while in direct contact with aqueous electrolytes under potentiostatic control. The system that was investigated consists of highly doped Si substrates that were conformally coated with ∼70 nm titania films produced by atomic-layer deposition. TiO2/electrolyte and Si/TiO2/Ni/electrolyte interfaces were then analyzed by synchrotron radiation photoelectron spectroscopy. The PES data allows for determination of the flat-band position and identification of potential regions in which Fermi level pinning, depletion, or accumulation occurred. Operando X-ray absorption spectroscopy (XAS) techniques were additionally used to investigate the properties of heterogeneous electrocatalysts for the oxygen-evolution reaction. Operando XAS including the pre-edge, edge and EXAFS regions allowed the development of a detailed picture of the catalysts under operating conditions, and elucidated the changes in the physical and electronic structure of the catalyst that accompanied increases in the applied potential. Specifically, XAS data, combined with DFT studies, indicated that the activity of the electrocatalyst correlated with the formation of Fe dopant sites in γ-NiOOH.
Co-reporter:Azhar I. Carim, Nicolas A. Batara, Anjali Premkumar, Harry A. Atwater, and Nathan S. Lewis
ACS Nano 2016 Volume 10(Issue 1) pp:102
Publication Date(Web):November 23, 2015
DOI:10.1021/acsnano.5b05119
The template-free growth of well ordered, highly anisotropic lamellar structures has been demonstrated during the photoelectrodeposition of Se–Te films, wherein the orientation of the pattern can be directed by orienting the linear polarization of the incident light. This control mechanism was investigated further herein by examining the morphologies of films grown photoelectrochemically using light from two simultaneous sources that had mutually different linear polarizations. Photoelectrochemical growth with light from two nonorthogonally polarized same-wavelength sources generated lamellar morphologies in which the long axes of the lamellae were oriented parallel to the intensity-weighted average polarization orientation. Simulations of light scattering at the solution–film interface were consistent with this observation. Computer modeling of these growths using combined full-wave electromagnetic and Monte Carlo growth simulations successfully reproduced the experimental morphologies and quantitatively agreed with the pattern orientations observed experimentally by considering only the fundamental light-material interactions during growth. Deposition with light from two orthogonally polarized same-wavelength as well as different-wavelength sources produced structures that consisted of two intersecting sets of orthogonally oriented lamellae in which the relative heights of the two sets could be varied by adjusting the relative source intensities. Simulations of light absorption were performed in analogous, idealized intersecting lamellar structures and revealed that the lamellae preferentially absorbed light polarized with the electric field vector along their long axes. These data sets cumulatively indicate that anisotropic light scattering and light absorption generated by the light polarization produces the anisotropic morphology and that the resultant morphology is a function of all illumination inputs despite differing polarizations.Keywords: chalcogenide; electrodeposition; maskless; photodeposition; photoelectrochemistry; template-free;
Co-reporter:Yikai Chen, Nathan S. Lewis, and Chengxiang Xiang
ACS Energy Letters 2016 Volume 1(Issue 1) pp:273
Publication Date(Web):June 8, 2016
DOI:10.1021/acsenergylett.6b00134
A multiphysics model that accounts for the performance of electrocatalysts and triple-junction light absorbers, as well as for the transport properties of the electrolyte and dissolved CO2, was used to evaluate the spatial and light-intensity dependence of product distributions in an integrated photoelectrochemical CO2 reduction (CO2R) cell. Different sets of band gap combinations of triple-junction light absorbers were required to accommodate the optimal total operating current density relative to the optimal partial current density for CO2R. The simulated product distribution was highly nonuniform along the width of the electrode and depended on the electrode dimensions as well as the illumination intensity. To achieve the same product selectivity as in a potentiostatic, “half-cell” configuration, the electrocatalyst must retain its selectivity over a range of cathode potentials, and this range is dependent on the transport losses and current–voltage relationship of the light absorbers, the geometric parameters of the cell, and the illumination intensity.
Co-reporter:Xinghao Zhou, Rui Liu, Ke Sun, Yikai Chen, Erik Verlage, Sonja A. Francis, Nathan S. Lewis, and Chengxiang Xiang
ACS Energy Letters 2016 Volume 1(Issue 4) pp:764
Publication Date(Web):September 9, 2016
DOI:10.1021/acsenergylett.6b00317
A solar-driven CO2 reduction (CO2R) cell was constructed, consisting of a tandem GaAs/InGaP/TiO2/Ni photoanode in 1.0 M KOH(aq) (pH = 13.7) to facilitate the oxygen-evolution reaction (OER), a Pd/C nanoparticle-coated Ti mesh cathode in 2.8 M KHCO3(aq) (pH = 8.0) to perform the CO2R reaction, and a bipolar membrane to allow for steady-state operation of the catholyte and anolyte at different bulk pH values. At the operational current density of 8.5 mA cm–2, in 2.8 M KHCO3(aq), the cathode exhibited <100 mV overpotential and >94% Faradaic efficiency for the reduction of 1 atm of CO2(g) to formate. The anode exhibited a 320 ± 7 mV overpotential for the OER in 1.0 M KOH(aq), and the bipolar membrane exhibited ∼480 mV voltage loss with minimal product crossovers and >90 and >95% selectivity for protons and hydroxide ions, respectively. The bipolar membrane facilitated coupling between two electrodes and electrolytes, one for the CO2R reaction and one for the OER, that typically operate at mutually different pH values and produced a lower total cell overvoltage than known single-electrolyte CO2R systems while exhibiting ∼10% solar-to-fuels energy-conversion efficiency.
Co-reporter:Shu Hu
The Journal of Physical Chemistry C 2016 Volume 120(Issue 6) pp:3117-3129
Publication Date(Web):December 15, 2015
DOI:10.1021/acs.jpcc.5b09121
Solid-state electrical, photoelectrochemical, and photoelectron spectroscopic techniques have been used to characterize the behavior and electronic structure of interfaces between n-Si, n+-Si, or p+-Si surfaces and amorphous coatings of TiO2 formed using atomic-layer deposition. Photoelectrochemical measurements of n-Si/TiO2/Ni interfaces in contact with a series of one-electron, electrochemically reversible redox systems indicated that the n-Si/TiO2/Ni structure acted as a buried junction whose photovoltage was independent of the formal potential of the contacting electrolyte. Solid-state current–voltage analysis indicated that the built-in voltage of the n-Si/TiO2 heterojunction was ∼0.7 V, with an effective Richardson constant ∼1/100th of the value of typical Si/metal Schottky barriers. X-ray photoelectron spectroscopic data allowed formulation of energy band-diagrams for the n-Si/TiO2, n+-Si/TiO2, and p+-Si/TiO2 interfaces. The XPS data were consistent with the rectifying behavior observed for amorphous TiO2 interfaces with n-Si and n+-Si surfaces and with an ohmic contact at the interface between amorphous TiO2 and p+-Si.
Co-reporter:Fan Yang
The Journal of Physical Chemistry C 2016 Volume 120(Issue 13) pp:6989-6995
Publication Date(Web):March 10, 2016
DOI:10.1021/acs.jpcc.6b00232
The photoelectrochemical behavior of n-type GaAs(100) electrodes coated with a single layer of graphene was compared with the behavior of bare, freshly etched n-type GaAs(100) electrodes, both for electrodes in contact with an aqueous solution containing K3[Fe(CN)6]/K4[Fe(CN)6] and for electrodes in contact with nonaqueous solutions containing a series of one-electron redox couples selected such that the Nernstian solution potentials spanned a range greater than 1 V. Under simulated 1 Sun illumination, the graphene-coated electrodes produced a short-circuit photocurrent density of 20 mA cm–2 for up to 8 h of continuous operation in nonaqueous electrolytes (H2O concentration 0.1%, v/v), while, under the same conditions, the unprotected n-GaAs electrodes showed a rapid decay of the photocurrent density within ∼400 s. Although the graphene monolayers enhanced the stability of n-GaAs photoanodes in nonaqueous electrolytes, the graphene did not fully protect photoanodes operated in contact with Fe(CN)63–/4–(aq) from corrosion. The dependence of the open-circuit voltage measured for graphene-coated n-GaAs photoanodes on the Nernstian potential of the solution was effectively identical to that of freshly etched n-GaAs photoanodes, indicating that addition of the graphene layer did not introduce significant pinning of the Fermi level of GaAs beyond the Fermi-level pinning attributable to mid-gap and solution-derived charge-carrier trap states previously observed at GaAs/liquid junctions.
Co-reporter:Noah T. Plymale
The Journal of Physical Chemistry C 2016 Volume 120(Issue 26) pp:14157-14169
Publication Date(Web):June 3, 2016
DOI:10.1021/acs.jpcc.6b03824
Functionalization of semiconductor surfaces with organic moieties can change the charge distribution, surface dipole, and electric field at the interface. The modified electric field will shift the semiconductor band-edge positions relative to those of a contacting phase. Achieving chemical control over the energetics at semiconductor surfaces promises to provide a means of tuning the band-edge energetics to form optimized junctions with a desired material. Si(111) surfaces functionalized with 3,4,5-trifluorophenylacetylenyl (TFPA) groups were characterized by transmission infrared spectroscopy, X-ray photoelectron spectroscopy, and surface recombination velocity measurements. Mixed methyl/TFPA-terminated (MMTFPA) n- and p-type Si(111) surfaces were synthesized and characterized by electrochemical methods. Current density versus voltage and Mott–Schottky measurements of Si(111)–MMTFPA electrodes in contact with Hg indicated that the barrier height, Φb, was a function of the fractional monolayer coverage of TFPA (θTFPA) in the alkyl monolayer. Relative to Si(111)–CH3 surfaces, Si(111)–MMTFPA samples with high θTFPA produced shifts in Φb of ≥0.6 V for n-Si/Hg contacts and ≥0.5 V for p-Si/Hg contacts. Consistently, the open-circuit potential (Eoc) of Si(111)–MMTFPA samples in contact with CH3CN solutions that contained the 1-electron redox couples decamethylferrocenium/decamethylferrocene (Cp*2Fe+/0) or methyl viologen (MV2+/+•) shifted relative to Si(111)–CH3 samples by +0.27 V for n-Si and by up to +0.10 V for p-Si. Residual surface recombination limited the Eoc of p-Si samples at high θTFPA despite the favorable shift in the band-edge positions induced by the surface modification process.
Co-reporter:Nathan S. Lewis
Chemical Reviews 2015 Volume 115(Issue 23) pp:12631
Publication Date(Web):December 9, 2015
DOI:10.1021/acs.chemrev.5b00654
Co-reporter:Yikai Chen, Ke Sun, Heather Audesirk, Chengxiang Xiang and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 6) pp:1736-1747
Publication Date(Web):23 Mar 2015
DOI:10.1039/C5EE00311C
The trade-off between the optical obscuration and kinetic overpotentials of electrocatalyst films patterned onto the surface of tandem light-absorber structures in model photoelectrosynthetic water-splitting systems was investigated using a 0-dimensional load-line analysis and experimental measurements. The electrocatalytic performance of the catalyst at high current densities, normalized to the electrocatalyst surface area, is an important factor in the dependence of the optimal solar-to-hydrogen (STH) conversion efficiency, ηSTH,opt, on the filling fraction (fc) of the patterned catalysts, because even under conditions that produce minority-carrier current densities of ∼10 mA cm−2 at the solid/liquid interface, the current density at catalyst-bearing sites can be >1–2 A cm−2 in low filling-fraction films. A universal current-density versus potential relationship, up to current densities of 10 A cm−2, was obtained experimentally for the hydrogen-evolution reaction (HER) using patterned Pt ultramicroelectrode (UME) arrays with a range of filling fractions and disc diameters. The ηSTH,opt of system designs that utilize patterned electrocatalysts located on the illuminated side of tandem photoabsorbers was then evaluated systematically. The maximum STH conversion efficiency, ηSTH,max, using a hypothetical electrocatalyst that was optically transparent but which nevertheless exhibited a current-density versus potential behavior that is characteristic of the most active Pt films measured experimentally regardless of their optical obscuration, was 26.7%. By comparison, the maximum ηSTH,opt of 24.9% for real patterned Pt electrocatalyst films closely approached this ideal-case limit. The performance and materials utilization of the patterned electrocatalysts and of the uniformly coated electrocatalysts on tandem photoabsorbers were also compared in this study. Hence, patterned electrocatalysts with very low filling fractions can provide a potentially promising path to the realization of efficient large-scale photoelectrolysis systems while minimizing the use of scarce noble metals.
Co-reporter:Erik Verlage, Shu Hu, Rui Liu, Ryan J. R. Jones, Ke Sun, Chengxiang Xiang, Nathan S. Lewis and Harry A. Atwater
Energy & Environmental Science 2015 vol. 8(Issue 11) pp:3166-3172
Publication Date(Web):18 Aug 2015
DOI:10.1039/C5EE01786F
A monolithically integrated device consisting of a tandem-junction GaAs/InGaP photoanode coated by an amorphous TiO2 stabilization layer, in conjunction with Ni-based, earth-abundant active electrocatalysts for the hydrogen-evolution and oxygen-evolution reactions, was used to effect unassisted, solar-driven water splitting in 1.0 M KOH(aq). When connected to a Ni–Mo-coated counterelectrode in a two-electrode cell configuration, the TiO2-protected III–V tandem device exhibited a solar-to-hydrogen conversion efficiency, ηSTH, of 10.5% under 1 sun illumination, with stable performance for >40 h of continuous operation at an efficiency of ηSTH > 10%. The protected tandem device also formed the basis for a monolithically integrated, intrinsically safe solar-hydrogen prototype system (1 cm2) driven by a NiMo/GaAs/InGaP/TiO2/Ni structure. The intrinsically safe system exhibited a hydrogen production rate of 0.81 μL s−1 and a solar-to-hydrogen conversion efficiency of 8.6% under 1 sun illumination in 1.0 M KOH(aq), with minimal product gas crossover while allowing for beneficial collection of separate streams of H2(g) and O2(g).
Co-reporter:Matthew R. Shaner, James R. McKone, Harry B. Gray and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 10) pp:2977-2984
Publication Date(Web):07 Aug 2015
DOI:10.1039/C5EE01076D
An n+p-Si microwire array coupled with a two-layer catalyst film consisting of Ni–Mo nanopowder and TiO2 light-scattering nanoparticles has been used to simultaneously achieve high fill factors and light-limited photocurrent densities from photocathodes that produce H2(g) directly from sunlight and water. The TiO2 layer scattered light back into the Si microwire array, while optically obscuring the underlying Ni–Mo catalyst film. In turn, the Ni–Mo film had a mass loading sufficient to produce high catalytic activity, on a geometric area basis, for the hydrogen-evolution reaction. The best-performing microwire array devices prepared in this work exhibited short-circuit photocurrent densities of −14.3 mA cm−2, photovoltages of 420 mV, and a fill factor of 0.48 under 1 Sun of simulated solar illumination, whereas the equivalent planar Ni–Mo-coated Si device, without TiO2 scatterers, exhibited negligible photocurrent due to complete light blocking by the Ni–Mo catalyst layer.
Co-reporter:Robert H. Coridan, Adam C. Nielander, Sonja A. Francis, Matthew T. McDowell, Victoria Dix, Shawn M. Chatman and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 10) pp:2886-2901
Publication Date(Web):13 Apr 2015
DOI:10.1039/C5EE00777A
The energy-conversion efficiency is a key metric that facilitates comparison of the performance of various approaches to solar energy conversion. However, a suite of disparate methodologies has been proposed and used historically to evaluate the efficiency of systems that produce fuels, either directly or indirectly, with sunlight and/or electrical power as the system inputs. A general expression for the system efficiency is given as the ratio of the total output power (electrical plus chemical) divided by the total input power (electrical plus solar). The solar-to-hydrogen (STH) efficiency follows from this globally applicable system efficiency but only is applicable in the special case for systems in which the only input power is sunlight and the only output power is in the form of hydrogen fuel derived from solar-driven water splitting. Herein, system-level efficiencies, beyond the STH efficiency, as well as component-level figures of merit are defined and discussed to describe the relative energy-conversion performance of key photoactive components of complete systems. These figures of merit facilitate the comparison of electrode materials and interfaces without conflating their fundamental properties with the engineering of the cell setup. The resulting information about the components can then be used in conjunction with a graphical circuit analysis formalism to obtain “optimal” system efficiencies that can be compared between various approaches. The approach provides a consistent method for comparison of the performance at the system and component levels of various technologies that produce fuels and/or electricity from sunlight.
Co-reporter:Meenesh R. Singh, Kimberly Papadantonakis, Chengxiang Xiang and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 9) pp:2760-2767
Publication Date(Web):30 Jun 2015
DOI:10.1039/C5EE01721A
The solution transport losses in a one-dimensional solar-driven water-splitting cell that operates in either concentrated acid, dilute acid, or buffered near-neutral pH electrolytes have been evaluated using a mathematical model that accounts for diffusion, migration and convective transport, as well as for bulk electrochemical reactions in the electrolyte. The Ohmic resistance loss, the Nernstian potential loss associated with pH gradients at the surface of the electrode, and electrodialysis in different electrolytes were assessed quantitatively in a stagnant cell as well as in a bubble-convected cell, in which convective mixing occurred due to product-gas evolution. In a stagnant cell that did not have convective mixing, small limiting current densities (<3 mA cm−2) and significant polarization losses derived from pH gradients were present in dilute acid as well as in near-neutral pH buffered electrolytes. In contrast, bubble-convected cells exhibited a significant increase in the limiting current density, and a significant reduction of the concentration overpotentials. In a bubble-convected cell, minimal solution transport losses were present in membrane-free cells, in either buffered electrolytes or in unbuffered solutions with pH ≤ 1. However, membrane-free cells lack a mechanism for product-gas separation, presenting significant practical and engineering impediments to the deployment of such systems. To produce an intrinsically safe cell, an ion-exchange membrane was incorporated into the cell. The accompanying solution losses, especially the pH gradients at the electrode surfaces, were modeled and simulated for such a system. Hence this work describes the general conditions under which intrinsically safe, efficient solar-driven water-splitting cells can be operated.
Co-reporter:Xinghao Zhou, Rui Liu, Ke Sun, Dennis Friedrich, Matthew T. McDowell, Fan Yang, Stefan T. Omelchenko, Fadl H. Saadi, Adam C. Nielander, Sisir Yalamanchili, Kimberly M. Papadantonakis, Bruce S. Brunschwig and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 9) pp:2644-2649
Publication Date(Web):15 Jul 2015
DOI:10.1039/C5EE01687H
Introduction of an ultrathin (2 nm) film of cobalt oxide (CoOx) onto n-Si photoanodes prior to sputter-deposition of a thick multifunctional NiOx coating yields stable photoelectrodes with photocurrent-onset potentials of ∼−240 mV relative to the equilibrium potential for O2(g) evolution and current densities of ∼28 mA cm−2 at the equilibrium potential for water oxidation when in contact with 1.0 M KOH(aq) under 1 sun of simulated solar illumination. The photoelectrochemical performance of these electrodes was very close to the Shockley diode limit for moderately doped n-Si(100) photoelectrodes, and was comparable to that of typical protected Si photoanodes that contained np+ buried homojunctions.
Co-reporter:Michael F. Lichterman, Shu Hu, Matthias H. Richter, Ethan J. Crumlin, Stephanus Axnanda, Marco Favaro, Walter Drisdell, Zahid Hussain, Thomas Mayer, Bruce S. Brunschwig, Nathan S. Lewis, Zhi Liu and Hans-Joachim Lewerenz
Energy & Environmental Science 2015 vol. 8(Issue 8) pp:2409-2416
Publication Date(Web):29 May 2015
DOI:10.1039/C5EE01014D
Photoelectrochemical (PEC) cells based on semiconductor/liquid interfaces provide a method of converting solar energy to electricity or fuels. Currently, the understanding of semiconductor/liquid interfaces is inferred from experiments and models. Operando ambient-pressure X-ray photoelectron spectroscopy (AP-XPS) has been used herein to directly characterize the semiconductor/liquid junction at room temperature under real-time electrochemical control. X-ray synchrotron radiation in conjunction with AP-XPS has enabled simultaneous monitoring of the solid surface, the solid/electrolyte interface, and the bulk electrolyte of a PEC cell as a function of the applied potential, U. The observed shifts in binding energy with respect to the applied potential have directly revealed ohmic and rectifying junction behavior on metallized and semiconducting samples, respectively. Additionally, the non-linear response of the core level binding energies to changes in the applied electrode potential has revealed the influence of defect-derived electronic states on the Galvani potential across the complete cell.
Co-reporter:Shane Ardo, Sang Hee Park, Emily L. Warren and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 5) pp:1484-1492
Publication Date(Web):01 Apr 2015
DOI:10.1039/C5EE00227C
Free-standing, membrane-embedded, Si microwire arrays have been used to affect the solar-driven, unassisted splitting of HI into H2 and I3−. The Si microwire arrays were grown by a chemical-vapor-deposition vapor–liquid–solid growth process using Cu growth catalysts, with a radial n+p junction then formed on each microwire. A Nafion proton-exchange membrane was introduced between the microwires and Pt electrocatalysts were then photoelectrochemically deposited on the microwires. The composite Si/Pt–Nafion membrane was mechanically removed from the growth substrate, and Pt electrocatalysts were then also deposited on the back side of the structure. The resulting membrane-bound Si microwire arrays spontaneously split concentrated HI into H2(g) and I3− under 1 Sun of simulated solar illumination. The reaction products (i.e. H2 and I3−) were confirmed by mass spectrometry and ultraviolet–visible electronic absorption spectroscopy.
Co-reporter:Yikai Chen, Shu Hu, Chengxiang Xiang and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 3) pp:876-886
Publication Date(Web):24 Dec 2014
DOI:10.1039/C4EE02314E
A sensitivity analysis has been performed for a variety of generic designs for solar-fuels generators. The analysis has revealed the relative importance of reductions in the overpotentials of electrocatalysts, of improvements in the materials properties of light absorbers, and of optimization in the system geometry for various different types of solar-fuels generators, while considering operation at a range of temperatures as well as under a variety of illumination intensities including up to 10-fold optical concentration.
Co-reporter:Matthew R. Shaner, Shu Hu, Ke Sun and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 1) pp:203-207
Publication Date(Web):05 Nov 2014
DOI:10.1039/C4EE03012E
Conductive, amorphous TiO2 coatings deposited by atomic-layer deposition, in combination with a sputter deposited NiCrOx oxygen-evolution catalyst, have been used to protect Si microwire arrays from passivation or corrosion in contact with aqueous electrolytes. Coated np+-Si/TiO2/NiCrOx as well as heterojunction n-Si/TiO2/NiCrOx Si microwire-array photoanodes exhibited stable photoelectrochemical operation in aqueous ferri-/ferro-cyanide solutions. The coatings also allowed for photoanodic water oxidation in 1.0 M KOH(aq) solutions for >2200 h of continuous operation under simulated 1 Sun conditions with ∼100% Faradaic efficiency for the evolution of O2(g).
Co-reporter:Adam C. Nielander, Matthew R. Shaner, Kimberly M. Papadantonakis, Sonja A. Francis and Nathan S. Lewis
Energy & Environmental Science 2015 vol. 8(Issue 1) pp:16-25
Publication Date(Web):19 Sep 2014
DOI:10.1039/C4EE02251C
A number of approaches to solar fuels generation are being developed, each of which has associated advantages and challenges. Many of these solar fuels generators are identified as “photoelectrochemical cells” even though these systems collectively operate based on a suite of fundamentally different physical principles. To facilitate appropriate comparisons between solar fuels generators, as well as to enable concise and consistent identification of the state-of-the-art for designs based on comparable operating principles, we have developed a taxonomy and nomenclature for solar fuels generators based on the source of the asymmetry that separates photogenerated electrons and holes. Three basic device types have been identified: photovoltaic cells, photoelectrochemical cells, and particulate/molecular photocatalysts. We outline the advantages and technological challenges associated with each type, and provide illustrative examples for each approach as well as for hybrid approaches.
Co-reporter:Ke Sun;Yanjin Kuang;Erik Verlage;Bruce S. Brunschwig;Charles W. Tu
Advanced Energy Materials 2015 Volume 5( Issue 11) pp:
Publication Date(Web):
DOI:10.1002/aenm.201402276
Co-reporter:Azhar I. Carim, Nicolas A. Batara, Anjali Premkumar, Harry A. Atwater, and Nathan S. Lewis
Nano Letters 2015 Volume 15(Issue 10) pp:7071-7076
Publication Date(Web):September 21, 2015
DOI:10.1021/acs.nanolett.5b03137
Photoelectrochemical growth of Se–Te films spontaneously produces highly ordered, nanoscale lamellar morphologies with periodicities that can be tuned by varying the illumination wavelength during deposition. This phenomenon has been characterized further herein by determining the morphologies of photoelectrodeposited Se–Te films in response to tailored spectral illumination profiles. Se–Te films grown under illumination from four different sources, having similar average wavelengths but having spectral bandwidths that spanned several orders of magnitude, all nevertheless produced similar structures which had a single, common periodicity as quantitatively identified via Fourier analysis. Film deposition using simultaneous illumination from two narrowband sources, which differed in average wavelength by several hundred nanometers, resulted in a structure with only a single periodicity intermediate between the periods observed when either source alone was used. This single periodicity could be varied by manipulating the relative intensity of the two sources. An iterative model that combined full-wave electromagnetic effects with Monte Carlo growth simulations, and that considered only the fundamental light-material interactions during deposition, was in accord with the morphologies observed experimentally. Simulations of light absorption and concentration in idealized lamellar arrays, in conjunction with all of the available data, additionally indicated that a self-optimization of the periodicity of the nanoscale pattern, resulting in the maximization of the anisotropy of interfacial light absorption in the three-dimensional structure, is consistent with the observed growth process of such films.
Co-reporter:Keith T. Wong; Youn-Geun Kim; Manuel P. Soriaga; Bruce S. Brunschwig
Journal of the American Chemical Society 2015 Volume 137(Issue 28) pp:9006-9014
Publication Date(Web):July 8, 2015
DOI:10.1021/jacs.5b03339
Atomically flat, terraced H–Ge(111) was prepared by annealing in H2(g) at 850 °C. The formation of monohydride Ge–H bonds oriented normal to the surface was indicated by angle-dependent Fourier-transform infrared (FTIR) spectroscopy. Subsequent reaction in CCl3Br(l) formed Br-terminated Ge(111), as shown by the disappearance of the Ge–H absorption in the FTIR spectra concomitant with the appearance of Br photoelectron peaks in X-ray photoelectron (XP) spectra. The Br–Ge(111) surface was methylated by reaction with (CH3)2Mg. These surfaces exhibited a peak at 568 cm–1 in the high-resolution electron energy loss spectrum, consistent with the formation of a Ge–C bond. The absorption peaks in the FTIR spectra assigned to methyl “umbrella” and rocking modes were dependent on the angle of the incident light, indicating that the methyl groups were bonded directly atop surface Ge atoms. Atomic-force micrographs of CH3–Ge(111) surfaces indicated that the surface remained atomically flat after methylation. Electrochemical scanning–tunneling microscopy showed well-ordered methyl groups that covered nearly all of the surface. Low-energy electron diffraction images showed sharp, bright diffraction spots with a 3-fold symmetry, indicating a high degree of order with no evidence of surface reconstruction. A C 1s peak at 284.1 eV was observed in the XP spectra, consistent with the formation of a C–Ge bond. Annealing in ultrahigh vacuum revealed a thermal stability limit of ∼400 °C of the surficial CH3–Ge(111) groups. CH3–Ge(111) surfaces showed significantly greater resistance to oxidation in air than H–Ge(111) surfaces.
Co-reporter:Eric J. Popczun, Christopher W. Roske, Carlos G. Read, J. Chance Crompton, Joshua M. McEnaney, Juan F. Callejas, Nathan S. Lewis and Raymond E. Schaak
Journal of Materials Chemistry A 2015 vol. 3(Issue 10) pp:5420-5425
Publication Date(Web):16 Jan 2015
DOI:10.1039/C4TA06642A
CoP nanostructures that exposed predominantly (111) crystal facets were synthesized and evaluated for performance as electrocatalysts for the hydrogen-evolution reaction (HER). The branched CoP nanostructures were synthesized by reacting cobalt(II) acetylacetonate with trioctylphosphine in the presence of trioctylphosphine oxide. Electrodes comprised of the branched CoP nanostructures deposited at a loading density of ∼1 mg cm−2 on Ti electrodes required an overpotential of −117 mV to produce a current density of −20 mA cm−2 in 0.50 M H2SO4. Hence the branched CoP nanostructures belong to the growing family of highly active non-noble-metal HER electrocatalysts. Comparisons with related CoP systems have provided insights into the impact that shape-controlled nanoparticles and nanoparticle–electrode interactions have on the activity and stability of nanostructured HER electrocatalysts.
Co-reporter:Matthew T. McDowell, Michael F. Lichterman, Azhar I. Carim, Rui Liu, Shu Hu, Bruce S. Brunschwig, and Nathan S. Lewis
ACS Applied Materials & Interfaces 2015 Volume 7(Issue 28) pp:15189
Publication Date(Web):June 17, 2015
DOI:10.1021/acsami.5b00379
Light absorbers with moderate band gaps (1–2 eV) are required for high-efficiency solar fuels devices, but most semiconducting photoanodes undergo photocorrosion or passivation in aqueous solution. Amorphous TiO2 deposited by atomic-layer deposition (ALD) onto various n-type semiconductors (Si, GaAs, GaP, and CdTe) and coated with thin films or islands of Ni produces efficient, stable photoanodes for water oxidation, with the TiO2 films protecting the underlying semiconductor from photocorrosion in pH = 14 KOH(aq). The links between the electronic properties of the TiO2 in these electrodes and the structure and energetic defect states of the material are not yet well-elucidated. We show herein that TiO2 films with a variety of crystal structures and midgap defect state distributions, deposited using both ALD and sputtering, form rectifying junctions with n-Si and are highly conductive toward photogenerated carriers in n-Si/TiO2/Ni photoanodes. Moreover, the photovoltage of these electrodes can be modified by annealing the TiO2 in reducing or oxidizing environments. All of the polycrystalline TiO2 films with compact grain boundaries investigated herein protected the n-Si photoanodes against photocorrosion in pH = 14 KOH(aq). Hence, in these devices, conduction through the TiO2 layer is neither specific to a particular amorphous or crystalline structure nor determined wholly by a particular extrinsic dopant impurity. The coupled structural and energetic properties of TiO2, and potentially other protective oxides, can therefore be controlled to yield optimized photoelectrode performance.Keywords: heterojunction interfaces; oxygen evolution reaction; photocorrosion; photoelectrochemistry; solar fuels; water splitting;
Co-reporter:Heather A. Audesirk, Emily L. Warren, Jessie Ku, and Nathan S. Lewis
ACS Applied Materials & Interfaces 2015 Volume 7(Issue 3) pp:1396
Publication Date(Web):January 6, 2015
DOI:10.1021/am507200j
Silicon microwires grown by the vapor–liquid–solid process have attracted a great deal of interest as potential light absorbers for solar energy conversion. However, the research-scale techniques that have been demonstrated to produce ordered arrays of micro and nanowires may not be optimal for use as high-throughput processes needed for large-scale manufacturing. Herein we demonstrate the use of microimprint lithography to fabricate patterned templates for the confinement of an electrodeposited Cu catalyst for the vapor–liquid–solid (VLS) growth of Si microwires. A reusable polydimethylsiloxane stamp was used to pattern holes in silica sol–gels on silicon substrates, and the Cu catalyst was electrodeposited into the holes. Ordered arrays of crystalline p-type Si microwires were grown across the sol–gel-patterned substrates with materials quality and performance comparable to microwires fabricated with high-purity metal catalysts and cleanroom processing.Keywords: electrodeposition; imprint lithography; photoelectrochemistry; silicon microwire; VLS
Co-reporter:Betar M. Gallant, X. Wendy Gu, David Z. Chen, Julia R. Greer, and Nathan S. Lewis
ACS Nano 2015 Volume 9(Issue 5) pp:5143
Publication Date(Web):April 14, 2015
DOI:10.1021/acsnano.5b00468
The interfacial shear strength between Si microwires and a Nafion membrane has been tailored through surface functionalization of the Si. Acidic (−COOH-terminated) or basic (−NH2-terminated) surface-bound functionality was introduced by hydrosilylation reactions to probe the interactions between the functionalized Si microwires and hydrophilic ionically charged sites in the Nafion polymeric side chains. Surfaces functionalized with SiOx, Si–H, or Si–CH3 were also synthesized and investigated. The interfacial shear strength between the functionalized Si microwire surfaces and the Nafion matrix was quantified by uniaxial wire pull-out experiments in an in situ nanomechanical instrument that allowed simultaneous collection of mechanical data and visualization of the deformation process. In this process, an axial load was applied to the custom-shaped top portions of individual wires until debonding occurred from the Nafion matrix. The shear strength obtained from the nanomechanical measurements correlated with the chemical bond strength and the functionalization density of the molecular layer, with values ranging from 7 MPa for Si–CH3 surfaces to ∼16–20 MPa for oxygen-containing surface functionalities. Hence surface chemical control can be used to influence the mechanical adhesion forces at a Si–Nafion interface.Keywords: in situ tension; Nafion; nanomechanical; Si microwires; surface functionalization;
Co-reporter:Christopher W. Roske; Eric J. Popczun; Brian Seger; Carlos G. Read; Thomas Pedersen; Ole Hansen; Peter C. K. Vesborg; Bruce S. Brunschwig; Raymond E. Schaak; Ib Chorkendorff; Harry B. Gray
The Journal of Physical Chemistry Letters 2015 Volume 6(Issue 9) pp:1679-1683
Publication Date(Web):April 20, 2015
DOI:10.1021/acs.jpclett.5b00495
The electrocatalytic performance for hydrogen evolution has been evaluated for radial-junction n+p-Si microwire (MW) arrays with Pt or cobalt phosphide, CoP, nanoparticulate catalysts in contact with 0.50 M H2SO4(aq). The CoP-coated (2.0 mg cm–2) n+p-Si MW photocathodes were stable for over 12 h of continuous operation and produced an open-circuit photovoltage (Voc) of 0.48 V, a light-limited photocurrent density (Jph) of 17 mA cm–2, a fill factor (ff) of 0.24, and an ideal regenerative cell efficiency (ηIRC) of 1.9% under simulated 1 Sun illumination. Pt-coated (0.5 mg cm–2) n+p-Si MW-array photocathodes produced Voc = 0.44 V, Jph = 14 mA cm–2, ff = 0.46, and η = 2.9% under identical conditions. Thus, the MW geometry allows the fabrication of photocathodes entirely comprised of earth-abundant materials that exhibit performance comparable to that of devices that contain Pt.
Co-reporter:Leslie E. O’Leary; Nicholas C. Strandwitz; Christopher W. Roske; Suyeon Pyo; Bruce S. Brunschwig
The Journal of Physical Chemistry Letters 2015 Volume 6(Issue 4) pp:722-726
Publication Date(Web):January 26, 2015
DOI:10.1021/jz502542a
Silicon surfaces terminated with a mixed monolayer containing both a propyl aldehyde functionality and methyl groups were prepared and used to control the interfacial chemical and electronic properties of Si(111) surfaces during atomic-layer deposition (ALD) of Al2O3 or MnO. Si(111) surfaces functionalized only with the aldehyde moiety exhibited surface recombination velocities, S, of 2500 ± 600 cm s–1 whereas the mixed CH3–/HC(O)CH2CH2–Si(111) surfaces displayed S = 25 ± 7 cm s–1. During the ALD growth of either Al2O3 or MnO, both the HC(O)CH2CH2–Si(111) and CH3–/HC(O)CH2CH2–Si(111) surfaces produced increased metal oxide deposition at low cycle number, relative to H–Si(111) or CH3–Si(111) surfaces. As detected by X-ray photoelectron spectroscopy after the ALD process, the CH3– and mixed CH3–/HC(O)CH2CH2– functionalized Si(111) surfaces exhibited less interfacial SiOx than was observed for ALD of metal oxides on H–Si(111) substrates.
Co-reporter:Ke Sun; Matthew T. McDowell; Adam C. Nielander; Shu Hu; Matthew R. Shaner; Fan Yang; Bruce S. Brunschwig
The Journal of Physical Chemistry Letters 2015 Volume 6(Issue 4) pp:592-598
Publication Date(Web):January 19, 2015
DOI:10.1021/jz5026195
Semiconductors with small band gaps (<2 eV) must be stabilized against corrosion or passivation in aqueous electrolytes before such materials can be used as photoelectrodes to directly produce fuels from sunlight. In addition, incorporation of electrocatalysts on the surface of photoelectrodes is required for efficient oxidation of H2O to O2(g) and reduction of H2O or H2O and CO2 to fuels. We report herein the stabilization of np+-Si(100) and n-Si(111) photoanodes for over 1200 h of continuous light-driven evolution of O2(g) in 1.0 M KOH(aq) by an earth-abundant, optically transparent, electrocatalytic, stable, conducting nickel oxide layer. Under simulated solar illumination and with optimized index-matching for proper antireflection, NiOx-coated np+-Si(100) photoanodes produced photocurrent-onset potentials of −180 ± 20 mV referenced to the equilibrium potential for evolution of O2(g), photocurrent densities of 29 ± 1.8 mA cm–2 at the equilibrium potential for evolution of O2(g), and a solar-to-O2(g) conversion figure-of-merit of 2.1%.
Co-reporter:Rui Liu, Lihao Han, Zhuangqun Huang, Ivonne M. Ferrer, Arno H.M. Smets, Miro Zeman, Bruce S. Brunschwig, Nathan S. Lewis
Thin Solid Films 2015 Volume 586() pp:28-34
Publication Date(Web):1 July 2015
DOI:10.1016/j.tsf.2015.04.018
•Pure Pt films were grown by atomic layer deposition (ALD) using MeCpPtMe3 and ozone.•ALD-grown Pt thin films had high growth rates of 110 pm/cycle.•ALD-grown Pt films were electrocatalytic for hydrogen evolution from water.•Electrocatalytic activity of the ALD Pt films was equivalent to e-beam deposited Pt.•No carbon species were detected in the ALD-grown Pt films.Atomic layer deposition (ALD) was used to deposit nanoparticles and thin films of Pt onto etched p-type Si(111) wafers and glassy carbon discs. Using precursors of MeCpPtMe3 and ozone and a temperature window of 200–300 °C, the growth rate was 80–110 pm/cycle. X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), and scanning electron microscopy (SEM) were used to analyze the composition, structure, morphology, and thickness of the ALD-grown Pt nanoparticle films. The catalytic activity of the ALD-grown Pt for the hydrogen evolution reaction was shown to be equivalent to that of e-beam evaporated Pt on glassy carbon electrode.
Co-reporter:Ke Sun;Fadl H. Saadi;Michael F. Lichterman;Hsin-Ping Wang;William G. Hale;Noah T. Plymale;Xinghao Zhou;Jr-Hau He;Kimberly M. Papadantonakis;Stefan T. Omelchenko;Bruce S. Brunschwig
PNAS 2015 Volume 112 (Issue 12 ) pp:3612-3617
Publication Date(Web):2015-03-24
DOI:10.1073/pnas.1423034112
Reactively sputtered nickel oxide (NiOx) films provide transparent, antireflective, electrically conductive, chemically stable coatings that also are highly active
electrocatalysts for the oxidation of water to O2(g). These NiOx coatings provide protective layers on a variety of technologically important semiconducting photoanodes, including textured
crystalline Si passivated by amorphous silicon, crystalline n-type cadmium telluride, and hydrogenated amorphous silicon.
Under anodic operation in 1.0 M aqueous potassium hydroxide (pH 14) in the presence of simulated sunlight, the NiOx films stabilized all of these self-passivating, high-efficiency semiconducting photoelectrodes for >100 h of sustained, quantitative
solar-driven oxidation of water to O2(g).
Co-reporter:Noah T. Plymale
The Journal of Physical Chemistry C 2015 Volume 119(Issue 34) pp:19847-19862
Publication Date(Web):July 10, 2015
DOI:10.1021/acs.jpcc.5b05028
Ethynyl- and propynyl-terminated Si(111) surfaces synthesized using a two-step halogenation/alkylation method have been characterized by transmission infrared spectroscopy (TIRS), high-resolution electron energy-loss spectroscopy (HREELS), X-ray photoelectron spectroscopy (XPS), low-energy electron diffraction (LEED), atomic-force microscopy (AFM), electrochemical scanning–tunneling microscopy (EC-STM) and measurements of surface recombination velocities (S). For the ethynyl-terminated Si(111) surface, TIRS revealed signals corresponding to ethynyl ≡C–H and C≡C stretching oriented perpendicular to the surface, HREELS revealed a Si–C stretching signal, and XPS data showed the presence of C bound to Si with a fractional monolayer (ML) coverage (Φ) of ΦSi–CCH = 0.63 ± 0.08 ML. The ethynyl-terminated surfaces were also partially terminated by Si–OH groups (ΦSi–OH = 0.35 ± 0.03 ML) with limited formation of Si3+ and Si4+ oxides. For the propynyl-terminated Si(111) surface, TIRS revealed the presence of a (C–H)CH3 symmetric bending, or “umbrella,” peak oriented perpendicular to the surface, while HREELS revealed signals corresponding to Si–C and C≡C stretching, and XPS showed C bound to Si with ΦSi–CCCH3 = 1.05 ± 0.06 ML. The LEED patterns were consistent with a (1 × 1) surface unit cell for both surfaces, but room-temperature EC-STM indicated that the surfaces did not exhibit long-range ordering. HCC–Si(111) and CH3CC–Si(111) surfaces yielded S values of (3.5 ± 0.1) × 103 and (5 ± 1) × 102 cm s–1, respectively, after 581 h exposure to air. These observations are consistent with the covalent binding of ethynyl and propynyl groups, respectively, to the Si(111) surface.
Co-reporter:Karl Walczak;Yikai Chen;Christoph Karp;Jeffrey W. Beeman;Matthew Shaner;Joshua Spurgeon;Ian D. Sharp;Xenia Amashukeli;William West;Jian Jin;Chengxiang Xiang
ChemSusChem 2015 Volume 8( Issue 3) pp:544-551
Publication Date(Web):
DOI:10.1002/cssc.201402896
Abstract
A fully integrated solar-driven water-splitting system comprised of WO3/FTO/p+n Si as the photoanode, Pt/TiO2/Ti/n+p Si as the photocathode, and Nafion as the membrane separator, was simulated, assembled, operated in 1.0 M HClO4, and evaluated for performance and safety characteristics under dual side illumination. A multi-physics model that accounted for the performance of the photoabsorbers and electrocatalysts, ion transport in the solution electrolyte, and gaseous product crossover was first used to define the optimal geometric design space for the system. The photoelectrodes and the membrane separators were then interconnected in a louvered design system configuration, for which the light-absorbing area and the solution-transport pathways were simultaneously optimized. The performance of the photocathode and the photoanode were separately evaluated in a traditional three-electrode photoelectrochemical cell configuration. The photocathode and photoanode were then assembled back-to-back in a tandem configuration to provide sufficient photovoltage to sustain solar-driven unassisted water-splitting. The current–voltage characteristics of the photoelectrodes showed that the low photocurrent density of the photoanode limited the overall solar-to-hydrogen (STH) conversion efficiency due to the large band gap of WO3. A hydrogen-production rate of 0.17 mL hr−1 and a STH conversion efficiency of 0.24 % was observed in a full cell configuration for >20 h with minimal product crossover in the fully operational, intrinsically safe, solar-driven water-splitting system. The solar-to-hydrogen conversion efficiency, ηSTH, calculated using the multiphysics numerical simulation was in excellent agreement with the experimental behavior of the system. The value of ηSTH was entirely limited by the performance of the photoelectrochemical assemblies employed in this study. The louvered design provides a robust platform for implementation of various types of photoelectrochemical assemblies, and can provide an approach to significantly higher solar conversion efficiencies as new and improved materials become available.
Co-reporter:Ragip A. Pala, Andrew J. Leenheer, Michael Lichterman, Harry A. Atwater and Nathan S. Lewis
Energy & Environmental Science 2014 vol. 7(Issue 10) pp:3424-3430
Publication Date(Web):29 Jul 2014
DOI:10.1039/C4EE01580K
Measurement of the photocurrent as a function of the thickness of a light absorber has been shown herein both theoretically and experimentally to provide a method for determination of the minority-carrier diffusion length of a sample. To perform the measurement, an illuminated spot of photons with an energy well above the band gap of the material was scanned along the thickness gradient of a wedge-shaped, rear-illuminated semiconducting light absorber. Photogenerated majority carriers were collected through a back-side transparent ohmic contact, and a front-side liquid or Schottky junction collected the photogenerated minority carriers. Calculations showed that the diffusion length could be evaluated from the exponential variation in photocurrent as a function of the thickness of the sample. Good agreement was observed between experiment and theory for a solid-state silicon Schottky junction measured using this method. As an example for the application of the technique to semiconductor/liquid-junction photoelectrodes, the minority-carrier diffusion length was determined for graded thickness, sputtered tungsten trioxide and polished bismuth vanadate films under back-illumination in contact with an aqueous electrolyte. This wedge technique does not require knowledge of the spectral absorption coefficient, doping, or surface recombination velocity of the sample.
Co-reporter:Michael F. Lichterman, Azhar I. Carim, Matthew T. McDowell, Shu Hu, Harry B. Gray, Bruce S. Brunschwig and Nathan S. Lewis
Energy & Environmental Science 2014 vol. 7(Issue 10) pp:3334-3337
Publication Date(Web):22 Aug 2014
DOI:10.1039/C4EE01914H
Although II–VI semiconductors such as CdS, CdTe, CdSe, ZnTe, and alloys thereof can have nearly ideal band gaps and band-edge positions for the production of solar fuels, II–VI photoanodes are well-known to be unstable towards photocorrosion or photopassivation when in contact with aqueous electrolytes. Atomic-layer deposition (ALD) of amorphous, “leaky” TiO2 films coated with thin films or islands of Ni oxide has been shown to robustly protect Si, GaAs, and other III–V materials from photocorrosion and therefore to facilitate the robust, solar-driven photoelectrochemical oxidation of H2O to O2(g). We demonstrate herein that ALD-deposited 140 nm thick amorphous TiO2 films also effectively protect single crystalline n-CdTe photoanodes from corrosion or passivation. An n-CdTe/TiO2 electrode with a thin overlayer of a Ni-oxide based oxygen-evolution electrocatalyst produced 435 ± 15 mV of photovoltage with a light-limited current density of 21 ± 1 mA cm−2 under 100 mW cm−2 of simulated Air Mass 1.5 illumination. The ALD-deposited TiO2 films are highly optically transparent and electrically conductive. We show that an n-CdTe/TiO2/Ni oxide electrode enables the stable solar-driven oxidation of H2O to O2(g) in strongly alkaline aqueous solutions, where passive, intrinsically safe, efficient systems for solar-driven water splitting can be operated.
Co-reporter:Elizabeth A. Santori, Nicholas C. Strandwitz, Ronald L. Grimm, Bruce S. Brunschwig, Harry A. Atwater and Nathan S. Lewis
Energy & Environmental Science 2014 vol. 7(Issue 7) pp:2329-2338
Publication Date(Web):20 Mar 2014
DOI:10.1039/C4EE00202D
The operation of lightly doped Si microwire arrays under high-level injection conditions was investigated by measurement of the current-potential behavior and carrier-collection efficiency of the wires in contact with non-aqueous electrolytes, and through complementary device physics simulations. The current-potential behavior of the lightly doped Si wire array photoelectrodes was dictated by both the radial contact and the carrier-selective back contact. For example, the Si microwire arrays exhibited n-type behavior when grown on a n+-doped substrate and placed in contact with the 1,1′-dimethylferrocene+/0–CH3OH redox system. The microwire arrays exhibited p-type behavior when grown on a p+-doped substrate and measured in contact with a redox system with a sufficiently negative Nernstian potential. The wire array photoelectrodes exhibited internal quantum yields of ∼0.8, deviating from unity for these radial devices. Device physics simulations of lightly doped n-Si wires in radial contact with the 1,1′-dimethylferrocene+/0–CH3OH redox system showed that the carrier-collection efficiency should be a strong function of the wire diameter and the carrier lifetime within the wire. Small diameter (d < 200 nm) wires exhibited low quantum yields for carrier collection, due to the strong inversion of the wires throughout the wire volume. In contrast, larger diameter wires (d > 400 nm) exhibited higher carrier collection efficiencies that were strongly dependent on the carrier lifetime in the wire, and wires with carrier lifetimes exceeding 5 μs were predicted to have near-unity quantum yields. The simulations and experimental measurements collectively indicated that the Si microwires possessed carrier lifetimes greater than 1 μs, and showed that radial structures with micron dimensions and high material quality can result in excellent device performance with lightly doped, structured semiconductors.
Co-reporter:Zhuangqun Huang, Chengxiang Xiang, Hans-Joachim Lewerenz and Nathan S. Lewis
Energy & Environmental Science 2014 vol. 7(Issue 4) pp:1207-1211
Publication Date(Web):16 Jan 2014
DOI:10.1039/C3EE90043F
Co-reporter:Matthew R. Shaner, Katherine T. Fountaine, Shane Ardo, Rob H. Coridan, Harry A. Atwater and Nathan S. Lewis
Energy & Environmental Science 2014 vol. 7(Issue 2) pp:779-790
Publication Date(Web):16 Dec 2013
DOI:10.1039/C3EE43048K
Tandem junction (n–p+-Si/ITO/WO3/liquid) core–shell microwire devices for solar-driven water splitting have been designed, fabricated and investigated photoelectrochemically. The tandem devices exhibited open-circuit potentials of Eoc = −1.21 V versus E0′(O2/H2O), demonstrating additive voltages across the individual junctions (n–p+-Si Eoc = −0.5 V versus solution; WO3/liquid Eoc = −0.73 V versus E0′(O2/H2O)). Optical concentration (12×, AM1.5D) shifted the open-circuit potential to Eoc = −1.27 V versus E0′(O2/H2O) and resulted in unassisted H2 production during two-electrode measurements (anode: tandem device, cathode: Pt disc). The solar energy-conversion efficiencies were very low, 0.0068% and 0.0019% when the cathode compartment was saturated with Ar or H2, respectively, due to the non-optimal photovoltage and band-gap of the WO3 that was used in the demonstration system to obtain stability of all of the system components under common operating conditions while also insuring product separation for safety purposes.
Co-reporter:Jian Jin, Karl Walczak, Meenesh R. Singh, Chris Karp, Nathan S. Lewis and Chengxiang Xiang
Energy & Environmental Science 2014 vol. 7(Issue 10) pp:3371-3380
Publication Date(Web):25 Jul 2014
DOI:10.1039/C4EE01824A
The efficiency limits, gas-crossover behavior, formation of local pH gradients near the electrode surfaces, and safety characteristics have been evaluated experimentally as well as by use of multi-physics modeling and simulation methods for an integrated solar-driven water-splitting system that operates with bulk electrolyte solutions buffered at near-neutral pH. The integrated membrane-free system utilized a triple-junction amorphous hydrogenated Si (a-Si:H) cell as the light absorber, Pt and cobalt phosphate (Co–Pi) as electrocatalysts for the hydrogen-evolution reaction (HER) and oxygen-evolution reaction (OER), respectively, and a bulk aqueous solution buffered at pH = 9.2 by 1.0 M of boric acid/borate as an electrolyte. Although the solar-to-electrical efficiency of the stand-alone triple-junction a-Si:H photovoltaic cell was 7.7%, the solar-to-hydrogen (STH) conversion efficiency for the integrated membrane-free water-splitting system was limited under steady-state operation to 3.2%, and the formation of pH gradients near the electrode surfaces accounted for the largest voltage loss. The membrane-free system exhibited negligible product-recombination loss while operating at current densities near 3.0 mA cm−2, but exhibited significant crossover of products (up to 40% H2 in the O2 chamber), indicating that the system was not intrinsically safe. A system that contained a membrane to minimize the gas crossover, but which was otherwise identical to the membrane-free system, yielded very low energy-conversion efficiencies at steady state, due to low transference numbers for protons across the membranes resulting in electrodialysis of the solution and the consequent formation of large concentration gradients of both protons and buffer counterions near the electrode surfaces. The modeling and simulation results showed that despite the addition of 1.0 M of buffering agent to the bulk of the solution, during operation significant pH gradients developed near the surfaces of the electrodes. Hence, although the bulk electrolyte was buffered to near-neutral pH, the electrode surfaces and electrocatalysts experienced local environments under steady-state operation that were either highly acidic or highly alkaline in nature, changing the chemical form of the electrocatalysts and exposing the electrodes to potentially corrosive local pH conditions. In addition to significant pH gradients, the STH conversion efficiency of both types of systems was limited by the mass transport of ionic species to the electrode surfaces. Even at operating current densities of <3 mA cm−2, the voltage drops due to these pH gradients exceeded the combined electrocatalyst overpotentials for the hydrogen- and oxygen-evolution reactions at current densities of 10 mA cm−2. Hence, such near-neutral pH solar-driven water-splitting systems were both fundamentally limited in efficiency and/or co-evolved explosive mixtures of H2(g) and O2(g) in the presence of active catalysts for the recombination of H2(g) and O2(g).
Co-reporter:Keith T. Wong and Nathan S. Lewis
Accounts of Chemical Research 2014 Volume 47(Issue 10) pp:3037
Publication Date(Web):September 5, 2014
DOI:10.1021/ar500207y
The chemical, electronic, and structural properties of surfaces are affected by the chemical termination of the surface. Two-step halogenation/alkylation of silicon provides a scalable, wet-chemical method for grafting molecules onto the silicon surface. Unlike other commonly studied wet-chemical methods of surface modification, such as self-assembly of monolayers on metals or hydrosilylation on silicon, the two-step method enables attachment of small alkyl chains, even methyl groups, to a silicon surface with high surface coverage and homogeneity. The methyl-terminated Si(111) surface, by comparison to hydrogen-terminated Si(111), offers a unique opportunity to study the effects of the first surface bond connecting the overlayer to the surface. This Account describes studies of methyl-terminated Si(111), which have shown that the H–Si(111) and CH3–Si(111) surfaces are structurally nearly identical, yet impart significantly different chemical and electronic properties to the resulting Si surface.The structure of methyl-terminated Si(111) formed by a two-step halogenation/methylation process has been studied by a variety of spectroscopic methods. A covalent Si–C bond is oriented normal to the surface, with the methyl group situated directly atop a surface Si atom. Multiple spectroscopic methods have shown that methyl groups achieve essentially complete coverage of the surface atoms while maintaining the atomically flat, terraced structure of the original H–Si(111) surface. Thus, the H–Si(111) and CH3–Si(111) surface share essentially identical structures aside from the replacement of a Si–H bond with a Si–C bond.Despite their structural similarity, hydrogen and methyl termination exhibit markedly different chemical passivation. Specifically, CH3–Si(111) exhibits significantly greater oxidation resistance than H–Si(111) in air and in aqueous electrolyte under photoanodic current flow. Both surfaces exhibit similar thermal stability in vacuum, and the Si–H and Si–C bond strengths are expected to be very similar, so the results suggest that methyl termination presents a greater kinetic barrier to oxidation of the underlying Si surface. Hydrogen termination of Si(111) provides nearly perfect electronic passivation of surface states (i.e., less than 1 electronic defect per 40 million surface atoms), but this electronic passivation is rapidly degraded by oxidation in air or under electrochemical conditions. In contrast, methyl termination provides excellent electronic passivation that resists degradation due to oxidation. Moreover, alkylation modifies the surface electronic structure by creating a surface dipole that effectively changes the electron affinity of the Si surface, facilitating modification of the charge-transfer kinetics across Si/metal or Si/electrolyte junctions.This Account also briefly describes recent studies of mixed monolayers formed by the halogenation/alkylation of silicon. Mixed monolayers allow attachment of bulkier groups that enable secondary chemistry at the surface (e.g., attachment of molecular catalysts or seeding of atomic layer deposition) to be combined with methyl termination of remaining unreacted surface sites. Thus, secondary chemistry can be enabled while maintaining the chemical and electronic passivation provided by complete termination of surface atoms with Si–C bonds. Such studies of mixed monolayers demonstrate the potential use of a wet-chemical surface modification scheme that combines both chemical and electronic passivation.
Co-reporter:Robert H. Coridan, Kevin A. Arpin, Bruce S. Brunschwig, Paul V. Braun, and Nathan S. Lewis
Nano Letters 2014 Volume 14(Issue 5) pp:2310-2317
Publication Date(Web):March 28, 2014
DOI:10.1021/nl404623t
WO3 thin films have been deposited in a hierarchically structured core–shell morphology, with the cores consisting of an array of Si microwires and the shells consisting of a controlled morphology WO3 layer. Porosity was introduced into the WO3 outer shell by using a self-assembled microsphere colloidal crystal as a mask during the deposition of the WO3 shell. Compared to conformal, unstructured WO3 shells on Si microwires, the hierarchically structured core–shell photoanodes exhibited enhanced near-visible spectral response behavior, due to increased light absorption and reduced distances over which photogenerated carriers were collected. The use of structured substrates also improved the growth rate of microsphere-based colloidal crystals and suggests strategies for the use of colloidal materials in large-scale applications.
Co-reporter:Heather C. McCaig, Ed Myers, Nathan S. Lewis, and Michael L. Roukes
Nano Letters 2014 Volume 14(Issue 7) pp:3728-3732
Publication Date(Web):June 12, 2014
DOI:10.1021/nl500475b
Surface-initiated polymerization has been used to grow thick, uniform poly(methyl methacrylate) films on nanocantilever sensors. Cantilevers with these coatings yielded significantly greater sensitivity relative to bare devices as well as relative to devices that had been coated with drop-cast polymer films. The devices with surface-initiated polymer films also demonstrated high selectivity toward polar analytes. Surface-initiated polymerization can therefore provide a straightforward, reproducible method for large-scale functionalization of nanosensors.
Co-reporter:Fadl H. Saadi, Azhar I. Carim, Jesus M. Velazquez, Jack H. Baricuatro, Charles C. L. McCrory, Manuel P. Soriaga, and Nathan S. Lewis
ACS Catalysis 2014 Volume 4(Issue 9) pp:2866
Publication Date(Web):July 17, 2014
DOI:10.1021/cs500412u
The catalytically inactive components of a film have been converted, through an operando method of synthesis, to produce a catalyst for the reaction that the film is catalyzing. Specifically, thin films of molybdenum diselenide have been synthesized using a two-step wet-chemical method, in which excess sodium selenide was first added to a solution of ammonium heptamolydbate in aqueous sulfuric acid, resulting in the spontaneous formation of a black precipitate that contained molybdenum triselenide (MoSe3), molybdenum trioxide (MoO3), and elemental selenium. After purification and after the film had been drop cast onto a glassy carbon electrode, a reductive potential was applied to the precipitate-coated electrode. Hydrogen evolution occurred within the range of potentials applied to the electrode, but during the initial voltammetric cycle, an overpotential of ∼400 mV was required to drive the hydrogen-evolution reaction at a benchmark current density of −10 mA cm–2. The overpotential required to evolve hydrogen at the benchmark rate progressively decreased with subsequent voltammetry cycles, until a steady state was reached at which only ∼250 mV of overpotential was required to pass −10 mA cm–2 of current density. During the electrocatalysis, the catalytically inactive components in the as-prepared film were (reductively) converted to MoSe2 through an operando method of synthesis of the hydrogen-evolution catalyst. The initial film prepared from the precipitate was smooth, but the converted film was completely covered with pores ∼200 nm in diameter. The porous MoSe2 film was stable while being assessed by cyclic voltammetry for 48 h, and the overpotential required to sustain 10 mA cm–2 of hydrogen evolution increased by <50 mV over this period of operation.Keywords: hydrogen-evolution reaction; mesoporous catalysts; synthesis of group VI dichalcogenides; synthesis of molybdenum diselenide; wet-chemical synthesis of layered electrocatalysts
Co-reporter:Joshua M. McEnaney, J. Chance Crompton, Juan F. Callejas, Eric J. Popczun, Adam J. Biacchi, Nathan S. Lewis, and Raymond E. Schaak
Chemistry of Materials 2014 Volume 26(Issue 16) pp:4826
Publication Date(Web):July 17, 2014
DOI:10.1021/cm502035s
Amorphous molybdenum phosphide (MoP) nanoparticles have been synthesized and characterized as electrocatalysts for the hydrogen-evolution reaction (HER) in 0.50 M H2SO4 (pH 0.3). Amorphous MoP nanoparticles (having diameters of 4.2 ± 0.5 nm) formed upon heating Mo(CO)6 and trioctylphosphine in squalane at 320 °C, and the nanoparticles remained amorphous after heating at 450 °C in H2(5%)/Ar(95%) to remove the surface ligands. At mass loadings of 1 mg cm–2, MoP/Ti electrodes exhibited overpotentials of −90 and −105 mV (−110 and −140 mV without iR correction) at current densities of −10 and −20 mA cm–2, respectively. These HER overpotentials remained nearly constant over 500 cyclic voltammetric sweeps and 18 h of galvanostatic testing, indicating stability in acidic media under operating conditions. Amorphous MoP nanoparticles are therefore among the most active known molybdenum-based HER systems and are part of a growing family of active, acid-stable, non-noble-metal HER catalysts.
Co-reporter:Priscilla D. Antunez, Daniel A. Torelli, Fan Yang, Federico A. Rabuffetti, Nathan S. Lewis, and Richard L. Brutchey
Chemistry of Materials 2014 Volume 26(Issue 19) pp:5444
Publication Date(Web):September 22, 2014
DOI:10.1021/cm503124u
Co-reporter:Azhar I. Carim, Fadl H. Saadi, Manuel P. Soriaga and Nathan S. Lewis
Journal of Materials Chemistry A 2014 vol. 2(Issue 34) pp:13835-13839
Publication Date(Web):24 Jul 2014
DOI:10.1039/C4TA02611J
Using an electrochemical method under ambient conditions, crystallographically amorphous films of cobalt selenide have been deposited from aqueous solution onto planar Ti supports. These films have been evaluated as electrocatalysts for the hydrogen-evolution reaction. In 0.500 M H2SO4, the cobalt selenide films required an overpotential of ∼135 mV to drive the hydrogen-evolution reaction at a benchmark current density of −10 mA cm−2. Galvanostatic measurements indicated stability of the electrocatalytic films for >16 h of continuous operation at −10 mA cm−2. The facile preparation method, and the activity of the cobalt selenide films, suggest that electrodeposited metal chalcogenides are potentially attractive earth-abundant electrocatalysts for the hydrogen-evolution reaction.
Co-reporter:Joshua M. McEnaney, J. Chance Crompton, Juan F. Callejas, Eric J. Popczun, Carlos G. Read, Nathan S. Lewis and Raymond E. Schaak
Chemical Communications 2014 vol. 50(Issue 75) pp:11026-11028
Publication Date(Web):30 Jul 2014
DOI:10.1039/C4CC04709E
Amorphous tungsten phosphide (WP), which has been synthesized as colloidal nanoparticles with an average diameter of 3 nm, has been identified as a new electrocatalyst for the hydrogen-evolution reaction (HER) in acidic aqueous solutions. WP/Ti electrodes produced current densities of −10 mA cm−2 and −20 mA cm−2 at overpotentials of only −120 mV and −140 mV, respectively, in 0.50 M H2SO4(aq).
Co-reporter:Judith R. C. Lattimer, James D. Blakemore, Wesley Sattler, Sheraz Gul, Ruchira Chatterjee, Vittal K. Yachandra, Junko Yano, Bruce S. Brunschwig, Nathan S. Lewis and Harry B. Gray
Dalton Transactions 2014 vol. 43(Issue 40) pp:15004-15012
Publication Date(Web):17 Jul 2014
DOI:10.1039/C4DT01149J
Silicon(111) surfaces have been functionalized with mixed monolayers consisting of submonolayer coverages of immobilized 4-vinyl-2,2′-bipyridyl (1, vbpy) moieties, with the remaining atop sites of the silicon surface passivated by methyl groups. As the immobilized bipyridyl ligands bind transition metal ions, metal complexes can be assembled on the silicon surface. X-ray photoelectron spectroscopy (XPS) demonstrates that bipyridyl complexes of [Cp*Rh], [Cp*Ir], and [Ru(acac)2] were formed on the surface (Cp* is pentamethylcyclopentadienyl, acac is acetylacetonate). For the surface prepared with Ir, X-ray absorption spectroscopy at the Ir LIII edge showed an edge energy as well as post-edge features that were essentially identical with those observed on a powder sample of [Cp*Ir(bpy)Cl]Cl (bpy is 2,2′-bipyridyl). Charge-carrier lifetime measurements confirmed that the silicon surfaces retain their highly favorable photoelectronic properties upon assembly of the metal complexes. Electrochemical data for surfaces prepared on highly doped, n-type Si(111) electrodes showed that the assembled molecular complexes were redox active. However the stability of the molecular complexes on the surfaces was limited to several cycles of voltammetry.
Co-reporter:Eric J. Popczun;Carlos G. Read;Christopher W. Roske; Nathan S. Lewis; Raymond E. Schaak
Angewandte Chemie International Edition 2014 Volume 53( Issue 21) pp:5427-5430
Publication Date(Web):
DOI:10.1002/anie.201402646
Abstract
Nanoparticles of cobalt phosphide, CoP, have been prepared and evaluated as electrocatalysts for the hydrogen evolution reaction (HER) under strongly acidic conditions (0.50 M H2SO4, pH 0.3). Uniform, multi-faceted CoP nanoparticles were synthesized by reacting Co nanoparticles with trioctylphosphine. Electrodes comprised of CoP nanoparticles on a Ti support (2 mg cm−2 mass loading) produced a cathodic current density of 20 mA cm−2 at an overpotential of −85 mV. The CoP/Ti electrodes were stable over 24 h of sustained hydrogen production in 0.50 M H2SO4. The activity was essentially unchanged after 400 cyclic voltammetric sweeps, suggesting long-term viability under operating conditions. CoP is therefore amongst the most active, acid-stable, earth-abundant HER electrocatalysts reported to date.
Co-reporter:Eric J. Popczun;Carlos G. Read;Christopher W. Roske; Nathan S. Lewis; Raymond E. Schaak
Angewandte Chemie 2014 Volume 126( Issue 21) pp:5531-5534
Publication Date(Web):
DOI:10.1002/ange.201402646
Abstract
Nanoparticles of cobalt phosphide, CoP, have been prepared and evaluated as electrocatalysts for the hydrogen evolution reaction (HER) under strongly acidic conditions (0.50 M H2SO4, pH 0.3). Uniform, multi-faceted CoP nanoparticles were synthesized by reacting Co nanoparticles with trioctylphosphine. Electrodes comprised of CoP nanoparticles on a Ti support (2 mg cm−2 mass loading) produced a cathodic current density of 20 mA cm−2 at an overpotential of −85 mV. The CoP/Ti electrodes were stable over 24 h of sustained hydrogen production in 0.50 M H2SO4. The activity was essentially unchanged after 400 cyclic voltammetric sweeps, suggesting long-term viability under operating conditions. CoP is therefore amongst the most active, acid-stable, earth-abundant HER electrocatalysts reported to date.
Co-reporter:Emily L. Warren, Harry A. Atwater, and Nathan S. Lewis
The Journal of Physical Chemistry C 2014 Volume 118(Issue 2) pp:747-759
Publication Date(Web):December 9, 2013
DOI:10.1021/jp406280x
Co-reporter:Michael B. McDonald; Shane Ardo; Nathan S. Lewis; Michael S. Freund
ChemSusChem 2014 Volume 7( Issue 11) pp:3021-3027
Publication Date(Web):
DOI:10.1002/cssc.201402288
Abstract
A bipolar membrane can maintain a steady-state pH difference between the sites of oxidation and reduction in membrane-supported, solar-driven water-splitting systems without changing the overall thermodynamics required to split water. A commercially available bipolar membrane that can serve this purpose has been identified, its performance has been evaluated quantitatively, and is demonstrated to meet the requirements for this application. For effective utilization in integrated solar-driven water-splitting systems, such bipolar membranes must, however, be modified to simultaneously optimize their physical properties such as optical transparency, electronic conductivity and kinetics of water dissociation.
Co-reporter:Jesus M. Velazquez, Fadl H. Saadi, Adam P. Pieterick, Joshua M. Spurgeon, Manuel P. Soriaga, Bruce S. Brunschwig, Nathan S. Lewis
Journal of Electroanalytical Chemistry 2014 716() pp: 45-48
Publication Date(Web):
DOI:10.1016/j.jelechem.2013.11.030
Co-reporter:Juan F. Callejas, Joshua M. McEnaney, Carlos G. Read, J. Chance Crompton, Adam J. Biacchi, Eric J. Popczun, Thomas R. Gordon, Nathan S. Lewis, and Raymond E. Schaak
ACS Nano 2014 Volume 8(Issue 11) pp:11101
Publication Date(Web):September 24, 2014
DOI:10.1021/nn5048553
Nanostructured transition-metal phosphides have recently emerged as Earth-abundant alternatives to platinum for catalyzing the hydrogen-evolution reaction (HER), which is central to several clean energy technologies because it produces molecular hydrogen through the electrochemical reduction of water. Iron-based catalysts are very attractive targets because iron is the most abundant and least expensive transition metal. We report herein that iron phosphide (FeP), synthesized as nanoparticles having a uniform, hollow morphology, exhibits among the highest HER activities reported to date in both acidic and neutral-pH aqueous solutions. As an electrocatalyst operating at a current density of −10 mA cm–2, FeP nanoparticles deposited at a mass loading of ∼1 mg cm–2 on Ti substrates exhibited overpotentials of −50 mV in 0.50 M H2SO4 and −102 mV in 1.0 M phosphate buffered saline. The FeP nanoparticles supported sustained hydrogen production with essentially quantitative faradaic yields for extended time periods under galvanostatic control. Under UV illumination in both acidic and neutral-pH solutions, FeP nanoparticles deposited on TiO2 produced H2 at rates and amounts that begin to approach those of Pt/TiO2. FeP therefore is a highly Earth-abundant material for efficiently facilitating the HER both electrocatalytically and photocatalytically.Keywords: electrocatalysis; hydrogen evolution reaction; metal phosphides; nanoparticles; photocatalysis;
Co-reporter:Matthew T. McDowell ; Michael F. Lichterman ; Joshua M. Spurgeon ; Shu Hu ; Ian D. Sharp ; Bruce S. Brunschwig
The Journal of Physical Chemistry C 2014 Volume 118(Issue 34) pp:19618-19624
Publication Date(Web):August 7, 2014
DOI:10.1021/jp506133y
Ultrathin dual layers of TiO2 and Ni have been used to stabilize polycrystalline BiVO4 photoanodes against photocorrosion in an aqueous alkaline (pH = 13) electrolyte. Conformal, amorphous TiO2 layers were deposited on BiVO4 thin films by atomic-layer deposition, with Ni deposited onto the TiO2 by sputtering. Under simulated air mass 1.5 illumination, the dual-layer coating extended the lifetime of the BiVO4 photoanodes during photoelectrochemical water oxidation from minutes, for bare BiVO4, to hours, for the modified electrodes. X-ray photoelectron spectroscopy showed that these layers imparted chemical stability to the semiconductor/electrolyte interface. Transmission electron microscopy revealed the structure and morphology of the polycrystalline BiVO4 film as well as of the thin coating layers. This work demonstrates that protection schemes based on ultrathin corrosion-resistant overlayers can be applied beneficially to polycrystalline photoanode materials under conditions relevant to efficient solar-driven water-splitting systems.
Co-reporter:Fadl H. Saadi ; Azhar I. Carim ; Erik Verlage ; John C. Hemminger ; Nathan S. Lewis ;Manuel P. Soriaga
The Journal of Physical Chemistry C 2014 Volume 118(Issue 50) pp:29294-29300
Publication Date(Web):September 27, 2014
DOI:10.1021/jp5054452
Films of CoP have been electrochemically synthesized, characterized, and evaluated for performance as a catalyst for the hydrogen-evolution reaction (HER). The film was synthesized by cathodic deposition from a boric acid solution of Co2+ and H2PO2– on copper substrates followed by operando remediation of exogenous contaminants. The films were characterized structurally and compositionally by scanning-electron microscopy, energy-dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy, and Raman spectrophotometry. The catalytic activity was evaluated by cyclic voltammetry and chronopotentiometry. Surface characterization prior to electrocatalysis indicated that the film consisted of micrometer-sized spherical clusters located randomly and loosely on a slightly roughened surface. The composition of both the clusters and surface consisted of cobalt in the metallic, phosphide, and amorphous-oxide forms (CoO·Co2O3) and of phosphorus as phosphide and orthophosphate. The orthophosphate species, produced by air-oxidation, were eliminated upon HER electrocatalysis in sulfuric acid. The operando film purification yielded a functional electrocatalyst with a Co:P stoichiometric ratio of 1:1. After the HER, the surface was densely packed with micrometer-sized, mesa-like particles whose tops were flat and smooth. The CoP eletrodeposit exhibited an 85 mV overvoltage (η) for the HER at a current density of 10 mA cm–2 and was stable under operation in highly acidic solution, with an increase in η of 18 mV after 24 h of continuous operation. The comparative HER catalytic performance of CoP, film or nanoparticles, is as follows: ηPt < ηCoP film = ηCoP NP, ηNi2P < ηCoSe2 < ηMoS2 < ηMoSe2.
Co-reporter:Qixi Mi, Robert H. Coridan, Bruce S. Brunschwig, Harry B. Gray and Nathan S. Lewis
Energy & Environmental Science 2013 vol. 6(Issue 9) pp:2646-2653
Publication Date(Web):18 Jun 2013
DOI:10.1039/C3EE40712H
The behavior of WO3 photoanodes has been investigated in contact with a combination of four anions (Cl−, CH3SO3−, HSO4−, and ClO4−) and three solvents (water, acetonitrile, and propylene carbonate), to elucidate the role of the semiconductor surface, the electrolyte, and redox kinetics on the current density vs. potential properties of n-type WO3. In 1.0 M aqueous strong acids, although the flat-band potential (Efb) of WO3 was dominated by electrochemical intercalation of protons into WO3, the nature of the electrolyte influenced the onset potential (Eon) of the anodic photocurrent. In aprotic solvents, the electrolyte anion shifted both Efb and Eon, but did not significantly alter the overall profile of the voltammetric data. For 0.50 M tetra(n-butyl)ammonium perchlorate in propylene carbonate, the internal quantum yield exceeded unity at excitation wavelengths of 300–390 nm, indicative of current doubling. A regenerative photoelectrochemical cell based on the reversible redox couple B10Br10˙−/2− in acetonitrile, with a solution potential of ∼1.7 V vs. the normal hydrogen electrode, exhibited an open-circuit photovoltage of 1.32 V under 100 mW cm−2 of simulated Air Mass 1.5 global illumination.
Co-reporter:Shu Hu, Chengxiang Xiang, Sophia Haussener, Alan D. Berger and Nathan S. Lewis
Energy & Environmental Science 2013 vol. 6(Issue 10) pp:2984-2993
Publication Date(Web):10 May 2013
DOI:10.1039/C3EE40453F
The solar-to-hydrogen (STH) efficiency limits, along with the maximum efficiency values and the corresponding optimal band gap combinations, have been evaluated for various combinations of light absorbers arranged in a tandem configuration in realistic, operational water-splitting prototypes. To perform the evaluation, a current–voltage model was employed, with the light absorbers, electrocatalysts, solution electrolyte, and membranes coupled in series, and with the directions of optical absorption, carrier transport, electron transfer and ionic transport in parallel. The current density vs. voltage characteristics of the light absorbers were determined by detailed-balance calculations that accounted for the Shockley–Queisser limit on the photovoltage of each absorber. The maximum STH efficiency for an integrated photoelectrochemical system was found to be ∼31.1% at 1 Sun (=1 kW m−2, air mass 1.5), fundamentally limited by a matching photocurrent density of 25.3 mA cm−2 produced by the light absorbers. Choices of electrocatalysts, as well as the fill factors of the light absorbers and the Ohmic resistance of the solution electrolyte also play key roles in determining the maximum STH efficiency and the corresponding optimal tandem band gap combination. Pairing 1.6–1.8 eV band gap semiconductors with Si in a tandem structure produces promising light absorbers for water splitting, with theoretical STH efficiency limits of >25%.
Co-reporter:Chengxiang Xiang, Yikai Chen and Nathan S. Lewis
Energy & Environmental Science 2013 vol. 6(Issue 12) pp:3713-3721
Publication Date(Web):23 Sep 2013
DOI:10.1039/C3EE42143K
Two designs for an integrated photoelectrolysis system sustained by water vapor have been investigated using a multi-physics numerical model that accounts for charge and species conservation, electron and ion transport, and electrochemical processes. Both designs leverage the use of a proton-exchange membrane that provides conductive pathways for reactant/product transport and prevents product crossover. The resistive losses, product gas transport, and gas crossovers as a function of the geometric parameters of the two designs have been evaluated systematically. In these designs, minimization of pathways in the membrane that can support the diffusive transport of product gases from the catalyst to the gas-collecting chamber was required to prevent supersaturation of hydrogen or oxygen gases at the Nafion/catalyst interface. Due to the small, thin membrane layer that was required, a small electrode width (<300 μm) was also required to produce low resistive losses in the system. Alternatively, incorporation of a structured membrane that balances the gas transport and ionic transport allows the maximum electrode width to be increased to dimensions as large as a few millimeters. Diffusive gas transport between the cathode and anode was the dominant source for crossover of the product gases under such circumstances. The critical dimension of the electrode required to produce acceptably low rates of product crossover was also investigated through the numerical modeling and device simulations.
Co-reporter:Sophia Haussener, Shu Hu, Chengxiang Xiang, Adam Z. Weber and Nathan S. Lewis
Energy & Environmental Science 2013 vol. 6(Issue 12) pp:3605-3618
Publication Date(Web):17 Jun 2013
DOI:10.1039/C3EE41302K
The instantaneous efficiency of an operating photoelectrochemical solar-fuel-generator system is a complicated function of the tradeoffs between the light intensity and temperature-dependence of the photovoltage and photocurrent, as well as the losses associated with factors that include ohmic resistances, concentration overpotentials, kinetic overpotentials, and mass transport. These tradeoffs were evaluated quantitatively using an advanced photoelectrochemical device model comprised of an analytical device physics model for the semiconducting light absorbers in combination with a multi-physics device model that solved for the governing conservation equations in the various other parts of the system. The model was used to evaluate the variation in system efficiency due to hourly and seasonal variations in solar irradiation as well as due to variation in the isothermal system temperature. The system performance characteristics were also evaluated as a function of the band gaps of the dual-absorber tandem component and its properties, as well as the device dimensions and the electrolyte conductivity. The modeling indicated that the system efficiency varied significantly during the day and over a year, exhibiting local minima at midday and a global minimum at midyear when the solar irradiation is most intense. These variations can be reduced by a favorable choice of the system dimensions, by a reduction in the electrolyte ohmic resistances, and/or by utilization of very active electrocatalysts for the fuel-producing reactions. An increase in the system temperature decreased the annual average efficiency and led to less rapid ramp-up and ramp-down phases of the system, but reduced midday and midyear instantaneous efficiency variations. Careful choice of the system dimensions resulted in minimal change in the system efficiency in response to degradation in the quality of the light absorbing materials. The daily and annually averaged mass of hydrogen production for the optimized integrated system compared favorably to the daily and annually averaged mass of hydrogen that was produced by an optimized stand-alone tandem photovoltaic array connected electrically to a stand-alone electrolyzer system. The model can be used to predict the performance of the system, to optimize the design of solar-driven water splitting devices, and to guide the development of components of the devices as well as of the system as a whole.
Co-reporter:Shu Hu, Chun-Yung Chi, Katherine T. Fountaine, Maoqing Yao, Harry A. Atwater, P. Daniel Dapkus, Nathan S. Lewis and Chongwu Zhou
Energy & Environmental Science 2013 vol. 6(Issue 6) pp:1879-1890
Publication Date(Web):17 Apr 2013
DOI:10.1039/C3EE40243F
Periodic arrays of n-GaAs nanowires have been grown by selective-area metal–organic chemical-vapor deposition on Si and GaAs substrates. The optical absorption characteristics of the nanowire-arrays were investigated experimentally and theoretically, and the photoelectrochemical energy-conversion properties of GaAs nanowire arrays were evaluated in contact with one-electron, reversible, redox species in non-aqueous solvents. The radial semiconductor/liquid junction in the nanowires produced near-unity external carrier-collection efficiencies for nanowire-array photoanodes in contact with non-aqueous electrolytes. These anodes exhibited overall inherent photoelectrode energy-conversion efficiencies of ∼8.1% under 100 mW cm−2 simulated Air Mass 1.5 illumination, with open-circuit photovoltages of 590 ± 15 mV and short-circuit current densities of 24.6 ± 2.0 mA cm−2. The high optical absorption, and minimal reflection, at both normal and off-normal incidence of the GaAs nanowire arrays that occupy <5% of the fractional area of the electrode can be attributed to efficient incoupling into radial nanowire guided and leaky waveguide modes.
Co-reporter:Eric J. Popczun ; James R. McKone ; Carlos G. Read ; Adam J. Biacchi ; Alex M. Wiltrout ; Nathan S. Lewis ;Raymond E. Schaak
Journal of the American Chemical Society 2013 Volume 135(Issue 25) pp:9267-9270
Publication Date(Web):June 13, 2013
DOI:10.1021/ja403440e
Nanoparticles of nickel phosphide (Ni2P) have been investigated for electrocatalytic activity and stability for the hydrogen evolution reaction (HER) in acidic solutions, under which proton exchange membrane-based electrolysis is operational. The catalytically active Ni2P nanoparticles were hollow and faceted to expose a high density of the Ni2P(001) surface, which has previously been predicted based on theory to be an active HER catalyst. The Ni2P nanoparticles had among the highest HER activity of any non-noble metal electrocatalyst reported to date, producing H2(g) with nearly quantitative faradaic yield, while also affording stability in aqueous acidic media.
Co-reporter:Leslie E. O’Leary ; Michael J. Rose ; Tina X. Ding ; Erik Johansson ; Bruce S. Brunschwig
Journal of the American Chemical Society 2013 Volume 135(Issue 27) pp:10081-10090
Publication Date(Web):June 26, 2013
DOI:10.1021/ja402495e
The Heck reaction has been used to couple olefins to a Si(111) surface that was functionalized with a mixed monolayer comprised of methyl and thienyl groups. The coupling method maintained a conjugated linkage between the surface and the olefinic surface functionality, to allow for facile charge transfer from the silicon surface. While a Si(111) surface terminated only with thienyl groups displayed a surface recombination velocity, S, of 670 ± 190 cm s–1, the mixed CH3/SC4H3–Si(111) surfaces with a coverage of θSC4H3 = 0.15 ± 0.02 displayed a substantially lower value of S = 27 ± 9 cm s–1. Accordingly, CH3/SC4H3–Si(111) surfaces were brominated with N-bromosuccinimide, to produce mixed CH3/SC4H2Br–Si(111) surfaces with coverages of θBr–Si < 0.05. The resulting aryl halide surfaces were activated using [Pd(PPh3)4] as a catalyst. After activation, Pd(II) was selectively coordinated by oxidative addition to the surface-bound aryl halide. The olefinic substrates 4-fluorostyrene, vinylferrocene, and protoporphyrin IX dimethyl ester were then coupled (in dimethylformamide at 100 °C) to the Pd-containing functionalized Si surfaces. The porphyrin-modified surface was then metalated with Co, Cu, or Zn. The vinylferrocene-modified Si(111) surface showed a linear dependence of the peak current on scan rate in cyclic voltammetry, indicating that facile electron transfer had been maintained and providing evidence of a robust linkage between the Si surface and the tethered ferrocene. The final Heck-coupled surface exhibited S = 70 cm s–1, indicating that high-quality surfaces could be produced by this multistep synthetic approach for tethering small molecules to silicon photoelectrodes.
Co-reporter:Adam C. Nielander ; Matthew J. Bierman ; Nicholas Petrone ; Nicholas C. Strandwitz ; Shane Ardo ; Fan Yang ; James Hone
Journal of the American Chemical Society 2013 Volume 135(Issue 46) pp:17246-17249
Publication Date(Web):October 14, 2013
DOI:10.1021/ja407462g
The behavior of graphene-coated n-type Si(111) photoanodes was compared to the behavior of H-terminated n-type Si(111) photoanodes in contact with aqueous K3[Fe(CN)6]/K4[Fe(CN)6] as well as in contact with a series of outer-sphere, one-electron redox couples in nonaqueous electrolytes. The n-Si/Graphene electrodes exhibited stable short-circuit photocurrent densities of over 10 mA cm–2 for >1000 s of continuous operation in aqueous electrolytes, whereas n-Si–H electrodes yielded a nearly complete decay of the current density within ∼100 s. The values of the open-circuit photovoltages and the flat-band potentials of the Si were a function of both the Fermi level of the graphene and the electrochemical potential of the electrolyte solution, indicating that the n-Si/Graphene did not form a buried junction with respect to the solution contact.
Co-reporter:Karla R. Reyes-Gil, Craig Wiggenhorn, Bruce S. Brunschwig, and Nathan S. Lewis
The Journal of Physical Chemistry C 2013 Volume 117(Issue 29) pp:14947-14957
Publication Date(Web):June 17, 2013
DOI:10.1021/jp4025624
Ordered structures offer the potential for producing photoanodes with enhanced minority-carrier collection. To evaluate this approach to visible-light-driven oxidation in aqueous electrolytes, porous WO3 structures were synthesized by the potentiostatic anodization of W foil. The photoelectrochemical behavior of the porous WO3 photoanodes was compared to that of compact WO3 films. Relative to planar electrodes, the porous WO3 electrodes exhibited a 6-fold increase in photocurrent density, from 0.12 to 0.75 mA cm–2, under 100 mW cm–2 of simulated solar illumination. Spectral response measurements indicated that the porous electrodes exhibited internal quantum yields of ∼0.5 throughout most of the region of WO3 absorption. The external quantum yield of the porous WO3 films was a function of the angle of incidence of the light, increasing from 0.25 at normal incidence to 0.50 at 65° off normal. The porous WO3 films showed excellent stability against photodegradation. This work demonstrates that morphological control can improve the internal quantum yield of photoanodes in contact with aqueous electrolytes.
Co-reporter:Michael G. Walter, Xueliang Liu, Leslie E. O’Leary, Bruce S. Brunschwig, and Nathan S. Lewis
The Journal of Physical Chemistry C 2013 Volume 117(Issue 28) pp:14485-14492
Publication Date(Web):May 8, 2013
DOI:10.1021/jp4018162
The electronic and photovoltaic properties of junctions between the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT) and Si(111) surfaces have been investigated for a range of doping types, doping levels, and surface functionalization of the Si. PEDOT–poly(styrenesulfonate) (PSS) formed ohmic, low resistance contacts to H-terminated and CH3-terminated p-type Si(111) surfaces. In contrast, PEDOT formed high barrier height (0.8–1.0 V) contacts to n-Si(111) surfaces, with CH3-terminated n-Si(111)/PEDOT contacts showing slightly higher barrier heights (1.01 eV) than H-terminated n-Si(111)/PEDOT contacts (0.89 V). PEDOT contacts to CH3-terminated and H-terminated n-Si(111) surfaces both produced photovoltages under illumination in accord with the Shockley diode limit based on bulk/recombination diffusion in the semiconductor. Such devices produced solar energy-conversion efficiencies of 5.7% under 100 mW cm–2 of simulated air mass 1.5 illumination. The electrical properties of PEDOT contacts to CH3-terminated Si surfaces were significantly more stable in an air ambient than the electrical properties of PEDOT contacts to H-terminated Si surfaces. PEDOT films produced a low resistance, tunnel-barrier type of ohmic contact to n+-Si(111) surfaces. Hence, through various combinations of doping type, doping level, and surface functionalization, the PEDOT/Si contact system offers a wide range of opportunities for integration into monolithic photovoltaic and/or artificial photosynthetic systems.
Co-reporter:Nicholas C. Strandwitz, David J. Comstock, Ronald L. Grimm, Adam C. Nichols-Nielander, Jeffrey Elam, and Nathan S. Lewis
The Journal of Physical Chemistry C 2013 Volume 117(Issue 10) pp:4931-4936
Publication Date(Web):February 6, 2013
DOI:10.1021/jp311207x
Thin (10 nm) films of manganese oxide have been deposited by atomic layer deposition (ALD) onto n-type silicon and onto degenerately doped p-type silicon. The photoelectrochemical properties of the resulting semiconductor/metal-oxide structures were evaluated in contact with aqueous 0.35 M K4Fe(CN)6–0.05 M K3Fe(CN)6, 1.0 M KOH(aq), as well as in contact with a series of nonaqueous one-electron, reversible, outer-sphere redox systems. Under simulated air mass (AM) 1.5 illumination in contact with 0.35 M K4Fe(CN)6–0.05 M K3Fe(CN)6(aq), MnO-coated n-Si photoanodes displayed open-circuit voltages of up to 550 mV and stable anodic currents for periods of hours at 0.0 V versus the solution potential. In contact with 1.0 M KOH(aq), at current densities of ∼25 mA cm–2, MnO|Si photoanodes under 100 mW cm–2 of simulated AM 1.5 illumination yielded stable oxygen evolution for 10–30 min. Variation in the thickness of the MnO films from 4 to 20 nm indicated the presence of a series resistance in the MnO film that limited the fill factor and thus the solar energy-conversion efficiency of the photoelectrodes. Open-circuit photovoltages of 30 and 450 mV, respectively, were observed in contact with cobaltocene+/0 or ferrocene+/0 in CH3CN, indicating that the energetics of the MnO-coated Si surfaces were a function of the electrochemical potential of the contacting electrolyte solution.
Co-reporter:Bryce Sadtler;Stanley P. Burgos;Nicolas A. Batara;Joseph A. Beardslee;Harry A. Atwater;
Proceedings of the National Academy of Sciences 2013 110(49) pp:19707-19712
Publication Date(Web):November 11, 2013
DOI:10.1073/pnas.1315539110
Photoresponsive materials that adapt their morphologies, growth directions, and growth rates dynamically in response to the
local incident electromagnetic field would provide a remarkable route to the synthesis of complex 3D mesostructures via feedback
between illumination and the structure that develops under optical excitation. We report the spontaneous development of ordered,
nanoscale lamellar patterns in electrodeposited selenium–tellurium (Se–Te) alloy films grown under noncoherent, uniform illumination
on unpatterned substrates in an isotropic electrolyte solution. These inorganic nanostructures exhibited phototropic growth
in which lamellar stripes grew toward the incident light source, adopted an orientation parallel to the light polarization
direction with a period controlled by the illumination wavelength, and showed an increased growth rate with increasing light
intensity. Furthermore, the patterns responded dynamically to changes during growth in the polarization, wavelength, and angle
of the incident light, enabling the template-free and pattern-free synthesis, on a variety of substrates, of woodpile, spiral,
branched, or zigzag structures, along with dynamically directed growth toward a noncoherent, uniform intensity light source.
Full-wave electromagnetic simulations in combination with Monte Carlo growth simulations were used to model light–matter interactions
in the Se–Te films and produced a model for the morphological evolution of the lamellar structures under phototropic growth
conditions. The experiments and simulations are consistent with a phototropic growth mechanism in which the optical near-field
intensity profile selects and reinforces the dominant morphological mode in the emergent nanoscale patterns.
Co-reporter:Robert H. Coridan, Matthew Shaner, Craig Wiggenhorn, Bruce S. Brunschwig, and Nathan S. Lewis
The Journal of Physical Chemistry C 2013 Volume 117(Issue 14) pp:6949-6957
Publication Date(Web):March 15, 2013
DOI:10.1021/jp311947x
Co-reporter:Elizabeth A. Santori, James R. Maiolo III, Matthew J. Bierman, Nicholas C. Strandwitz, Michael D. Kelzenberg, Bruce S. Brunschwig, Harry A. Atwater and Nathan S. Lewis
Energy & Environmental Science 2012 vol. 5(Issue 5) pp:6867-6871
Publication Date(Web):20 Feb 2012
DOI:10.1039/C2EE03468A
Arrays of n-Si microwires have to date exhibited low efficiencies when measured as photoanodes in contact with a 1-1′-dimethylferrocene (Me2Fc+/0)–CH3OH solution. Using high-purity Au or Cu catalysts, arrays of crystalline Si microwires were grown by a vapor-liquid-solid process without dopants, which produced wires with electronically active dopant concentrations of 1 × 1013 cm−3. When measured as photoanodes in contact with a Me2Fc+/0–CH3OH solution, the lightly doped Si microwire arrays exhibited greatly increased fill factors and efficiencies as compared to n-Si microwires grown previously with a lower purity Au catalyst. In particular, the Cu-catalyzed Si microwire array photoanodes exhibited open-circuit voltages of ∼0.44 V, carrier-collection efficiencies exceeding ∼0.75, and an energy-conversion efficiency of 1.4% under simulated air mass 1.5 G illumination. Lightly doped Cu-catalyzed Si microwire array photoanodes have thus demonstrated performance that is comparable to that of optimally doped p-type Si microwire array photocathodes in photoelectrochemical cells.
Co-reporter:Qixi Mi, Almagul Zhanaidarova, Bruce S. Brunschwig, Harry B. Gray and Nathan S. Lewis
Energy & Environmental Science 2012 vol. 5(Issue 2) pp:5694-5700
Publication Date(Web):03 Jan 2012
DOI:10.1039/C2EE02929D
The faradaic efficiency for O2(g) evolution at thin-film WO3 photoanodes has been evaluated in a series of acidic aqueous electrolytes. In 1.0 M H2SO4, persulfate was the predominant photoelectrochemical oxidation product, and no O2 was detected unless catalytic quantities of Ag+(aq) were added to the electrolyte. In contact with 1.0 M HClO4, dissolved O2 was observed with nearly unity faradaic efficiency, but addition of a hole scavenger, 4-cyanopyridine N-oxide, completely suppressed O2 formation. In 1.0 M HCl, Cl2(g) was the primary oxidation product. These results indicate that at WO3 photoanodes, water oxidation is dominated by oxidation of the acid anions in 1.0 M HCl, H2SO4, and HClO4, respectively.
Co-reporter:Qixi Mi ; Yuan Ping ; Yan Li ; Bingfei Cao ; Bruce S. Brunschwig ; Peter G. Khalifah ; Giulia A. Galli ; Harry B. Gray
Journal of the American Chemical Society 2012 Volume 134(Issue 44) pp:18318-18324
Publication Date(Web):September 28, 2012
DOI:10.1021/ja3067622
We describe stable intercalation compounds of the composition xN2·WO3 (x = 0.034–0.039), formed by trapping N2 in WO3. The incorporation of N2 significantly reduced the absorption threshold of WO3; notably, 0.039N2·WO3 anodes exhibited photocurrent under illumination at wavelengths ≤640 nm with a faradaic efficiency for O2 evolution in 1.0 M HClO4(aq) of nearly unity. Spectroscopic and computational results indicated that deformation of the WO3 host lattice, as well as weak electronic interactions between trapped N2 and the WO3 matrix, contributed to the observed red shift in optical absorption. Noble-gas-intercalated WO3 materials similar to xN2·WO3 are predicted to function as photoanodes that are responsive to visible light.
Co-reporter:James R. McKone ; Adam P. Pieterick ; Harry B. Gray
Journal of the American Chemical Society 2012 Volume 135(Issue 1) pp:223-231
Publication Date(Web):November 30, 2012
DOI:10.1021/ja308581g
Crystalline p-type WSe2 has been grown by a chemical vapor transport method. After deposition of noble metal catalysts, p-WSe2 photocathodes exhibited thermodynamically based photoelectrode energy-conversion efficiencies of >7% for the hydrogen evolution reaction under mildly acidic conditions, and were stable under cathodic conditions for at least 2 h in acidic as well as in alkaline electrolytes. The open circuit potentials of the photoelectrodes in contact with the H+/H2 redox couple were very close to the bulk recombination/diffusion limit predicted from the Shockley diode equation. Only crystals with a prevalence of surface step edges exhibited a shift in flat-band potential as the pH was varied. Spectral response data indicated effective minority-carrier diffusion lengths of ∼1 μm, which limited the attainable photocurrent densities in the samples to ∼15 mA cm–2 under 100 mW cm–2 of Air Mass 1.5G illumination.
Co-reporter:Joseph A. Beardslee, Bryce Sadtler, and Nathan S. Lewis
ACS Nano 2012 Volume 6(Issue 11) pp:10303
Publication Date(Web):October 20, 2012
DOI:10.1021/nn304180k
External magnetic fields have been used to vertically align ensembles of silicon microwires coated with ferromagnetic nickel films. X-ray diffraction and image analysis techniques were used to quantify the degree of vertical orientation of the microwires. The degree of vertical alignment and the minimum field strength required for alignment were evaluated as a function of the wire length, coating thickness, magnetic history, and substrate surface properties. Nearly 100% of 100 μm long, 2 μm diameter, Si microwires that had been coated with 300 nm of Ni could be vertically aligned by a 300 G magnetic field. For wires ranging from 40 to 60 μm in length, as the length of the wire increased, a higher degree of alignment was observed at lower field strengths, consistent with an increase in the available magnetic torque. Microwires that had been exposed to a magnetic sweep up to 300 G remained magnetized and, therefore, aligned more readily during subsequent magnetic field alignment sweeps. Alignment of the Ni-coated Si microwires occurred at lower field strengths on hydrophilic Si substrates than on hydrophobic Si substrates. The magnetic field alignment approach provides a pathway for the directed assembly of solution-grown semiconductor wires into vertical arrays, with potential applications in solar cells as well as in other electronic devices that utilize nano- and microscale components as active elements.Keywords: image analysis; magnetic coatings; Ni electrodeposition; vapor−liquid−solid growth; X-ray diffraction
Co-reporter:Chengxiang Xiang;Andrew C. Meng
PNAS 2012 Volume 109 (Issue 39 ) pp:
Publication Date(Web):2012-09-25
DOI:10.1073/pnas.1118338109
Physical integration of a Ag electrical contact internally into a metal/substrate/microstructured Si wire array/oxide/Ag/electrolyte
photoelectrochemical solar cell has produced structures that display relatively low ohmic resistance losses, as well as highly
efficient mass transport of redox species in the absence of forced convection. Even with front-side illumination, such wire-array
based photoelectrochemical solar cells do not require a transparent conducting oxide top contact. In contact with a test electrolyte
that contained 50 mM/5.0 mM of the cobaltocenium+/0 redox species in CH3CN–1.0 M LiClO4, when the counterelectrode was placed in the solution and separated from the photoelectrode, mass transport restrictions
of redox species in the internal volume of the Si wire array photoelectrode produced low fill factors and limited the obtainable
current densities to 17.6 mA cm-2 even under high illumination. In contrast, when the physically integrated internal Ag film served as the counter electrode,
the redox couple species were regenerated inside the internal volume of the photoelectrode, especially in regions where depletion
of the redox species due to mass transport limitations would have otherwise occurred. This behavior allowed the integrated
assembly to operate as a two-terminal, stand-alone, photoelectrochemical solar cell. The current density vs. voltage behavior
of the integrated photoelectrochemical solar cell produced short-circuit current densities in excess of 80 mA cm-2 at high light intensities, and resulted in relatively low losses due to concentration overpotentials at 1 Sun illumination.
The integrated wire array-based device architecture also provides design guidance for tandem photoelectrochemical cells for
solar-driven water splitting.
Co-reporter:Ronald L. Grimm, Matthew J. Bierman, Leslie E. O’Leary, Nicholas C. Strandwitz, Bruce S. Brunschwig, and Nathan S. Lewis
The Journal of Physical Chemistry C 2012 Volume 116(Issue 44) pp:23569-23576
Publication Date(Web):October 13, 2012
DOI:10.1021/jp308461q
The photoelectrochemical behavior of methyl-terminated p-type and n-type Si(111) surfaces was determined in contact with a series of one-electron, outer-sphere, redox couples that span >1 V in the Nernstian redox potential, E(A/A–), of the solution. The dependence of the current vs potential data, as well as of the open-circuit photovoltage, Voc, on E(A/A–) was compared to the behavior of H-terminated p-type and n-type Si(111) surfaces in contact with these same electrolytes. For a particular E(A/A–) value, CH3-terminated p-Si(111) electrodes showed lower Voc values than H-terminated p-Si(111) electrodes, whereas CH3-terminated n-Si(111) electrodes showed higher Voc values than H-terminated n-Si(111) electrodes. Under 100 mW cm–2 of ELH-simulated Air Mass 1.5 illumination, n-type H–Si(111) and CH3–Si(111) electrodes both demonstrated nonrectifying behavior with no photovoltage at very negative values of E(A/A–) and produced limiting Voc values of >0.5 V at very positive values of E(A/A–). Illuminated p-type H–Si(111) and CH3–Si(111) electrodes produced no photovoltage at positive values of E(A/A–) and produced limiting Voc values in excess of 0.5 V at very negative values of E(A/A–). In contact with CH3CN-octamethylferrocene+/0, differential capacitance vs potential experiments yielded a −0.40 V shift in flat-band potential for CH3-terminated n-Si(111) surfaces relative to H-terminated n-Si(111) surfaces. Similarly, in contact with CH3CN-1,1′-dicarbomethoxycobaltocene+/0, the differential capacitance vs potential data indicated a −0.25 V shift in the flat-band potential for CH3-terminated p-Si(111) electrodes relative to H-terminated p-Si(111) electrodes. The observed trends in Voc vs E(A/A–), and the trends in the differential capacitance vs potential data are consistent with a negative shift in the interfacial dipole as a result of methylation of the Si(111) surface. The negative dipole shift is consistent with a body of theoretical and experimental comparisons of the behavior of CH3–Si(111) surfaces vs H–Si(111) surfaces, including density functional theory of the sign and magnitude of the surface dipole, photoemission spectroscopy in ultrahigh vacuum, the electrical behavior of Hg/Si contacts, and the pH dependence of the current–potential behavior of Si electrodes in contact with aqueous electrolytes.
Co-reporter:Chengxiang Xiang, Gregory M. Kimball, Ronald L. Grimm, Bruce S. Brunschwig, Harry A. Atwater and Nathan S. Lewis
Energy & Environmental Science 2011 vol. 4(Issue 4) pp:1311-1318
Publication Date(Web):02 Feb 2011
DOI:10.1039/C0EE00554A
P-Type cuprous oxide (Cu2O) photoelectrodes prepared by the thermal oxidation of Cu foils exhibited open-circuit voltages in excess of 800 mV in nonaqueous regenerative photoelectrochemical cells. In contact with the decamethylcobaltocene+/0 (Me10CoCp2+/0) redox couple, cuprous oxide yielded open-circuit voltage, Voc, values of 820 mV and short-circuit current density, Jsc, values of 3.1 mA cm−2 under simulated air mass 1.5 illumination. The energy-conversion efficiency of 1.5% was limited by solution absorption and optical reflection losses that reduced the short-circuit photocurrent density. Spectral response measurements demonstrated that the internal quantum yield approached unity in the 400–500 nm spectral range, but poor red response, attributable to bulk recombination, lowered the overall efficiency of the cell. X-Ray photoelectron spectroscopy and Auger electron spectroscopy indicated that the photoelectrodes had a high-quality cuprous oxide surface, and revealed no observable photocorrosion during operation in the nonaqueous electrolyte. The semiconductor/liquid junctions thus provide a noninvasive method to investigate the energy-conversion properties of cuprous oxide without the confounding factors of deleterious surface reactions.
Co-reporter:James R. McKone, Emily L. Warren, Matthew J. Bierman, Shannon W. Boettcher, Bruce S. Brunschwig, Nathan S. Lewis and Harry B. Gray
Energy & Environmental Science 2011 vol. 4(Issue 9) pp:3573-3583
Publication Date(Web):01 Aug 2011
DOI:10.1039/C1EE01488A
The dark electrocatalytic and light photocathodic hydrogen evolution properties of Ni, Ni–Mo alloys, and Pt on Si electrodes have been measured, to assess the viability of earth-abundant electrocatalysts for integrated, semiconductor coupled fuel formation. In the dark, the activities of these catalysts deposited on degenerately doped p+-Si electrodes increased in the order Ni < Ni–Mo ≤ Pt. Ni–Mo deposited on degenerately doped Si microwires exhibited activity that was very similar to that of Pt deposited by metal evaporation on planar Si electrodes. Under 100 mW cm−2 of Air Mass 1.5 solar simulation, the energy conversion efficiencies of p-type Si/catalyst photoelectrodes ranged from 0.2–1%, and increased in the order Ni ≈ Ni–Mo < Pt, due to somewhat lower photovoltages and photocurrents for p-Si/Ni–Mo relative to p-Si/Ni and p-Si/Pt photoelectrodes. Deposition of the catalysts onto microwire arrays resulted in higher apparent catalytic activities and similar photoelectrode efficiencies than were observed on planar p-Si photocathodes, despite lower light absorption by p-Si in the microwire structures.
Co-reporter:Joshua M. Spurgeon and Nathan S. Lewis
Energy & Environmental Science 2011 vol. 4(Issue 8) pp:2993-2998
Publication Date(Web):02 Jun 2011
DOI:10.1039/C1EE01203G
The current–voltage characteristics of a proton exchange membrane (PEM) electrolyzer constructed with an IrRuOx water oxidation catalyst and a Pt black water reduction catalyst, under operation with water vapor from a humidified carrier gas, have been investigated as a function of the gas flow rate, the relative humidity, and the presence of oxygen. The performance of the system with water vapor was also compared to the performance when the device was immersed in liquid water. With a humidified Ar(g) input stream at 20 °C, an electrolysis current density of 10 mA cm−2 was sustained at an applied voltage of ∼1.6 V, with a current density of 20 mA cm−2 observed at ∼1.7 V. In the system evaluated, at current densities >40 mA cm−2 the electrolysis of water vapor was limited by the mass flux of water to the PEM. At <40 mA cm−2, the electrolysis of water vapor supported a given current density at a lower applied bias than did the electrolysis of liquid water. The relative humidity of the input carrier gas strongly affected the current–voltage behavior, with lower electrolysis current density attributed to dehydration of the PEM at reduced humidity values. The results provide a proof-of-concept that, with sufficiently active catalysts, an efficient solar photoelectrolyzer could be operated only with water vapor as the feedstock, even at the low operating temperatures that may result in the absence of active heating. This approach therefore offers a route to avoid the light attenuation and mass transport limitations that are associated with bubble formation in these systems.
Co-reporter:Shaune L. McFarlane, Brittney A. Day, Kevin McEleney, Michael S. Freund and Nathan S. Lewis
Energy & Environmental Science 2011 vol. 4(Issue 5) pp:1700-1703
Publication Date(Web):10 Jan 2011
DOI:10.1039/C0EE00384K
We discuss the figures of merit for conducting membranes in artificial photosynthetic systems and describe an electronically and ionically conducting polymer composite with attractive performance characteristics.
Co-reporter:Shannon W. Boettcher ; Emily L. Warren ; Morgan C. Putnam ; Elizabeth A. Santori ; Daniel Turner-Evans ; Michael D. Kelzenberg ; Michael G. Walter ; James R. McKone ; Bruce S. Brunschwig ; Harry A. Atwater
Journal of the American Chemical Society 2011 Volume 133(Issue 5) pp:1216-1219
Publication Date(Web):January 7, 2011
DOI:10.1021/ja108801m
Arrays of B-doped p-Si microwires, diffusion-doped with P to form a radial n+ emitter and subsequently coated with a 1.5-nm-thick discontinuous film of evaporated Pt, were used as photocathodes for H2 evolution from water. These electrodes yielded thermodynamically based energy-conversion efficiencies >5% under 1 sun solar simulation, despite absorbing less than 50% of the above-band-gap incident photons. Analogous p-Si wire-array electrodes yielded efficiencies <0.2%, largely limited by the low photovoltage generated at the p-Si/H2O junction.
Co-reporter:Iman Yahyaie, Kevin McEleney, Michael Walter, Derek R. Oliver, Douglas J. Thomson, Michael S. Freund, and Nathan S. Lewis
The Journal of Physical Chemistry Letters 2011 Volume 2(Issue 6) pp:675-680
Publication Date(Web):March 4, 2011
DOI:10.1021/jz2001375
The electrical (DC) behavior of single silicon microwires has been determined by the use of tungsten probes to make ohmic contact to the silicon microwires. The basic electrical properties of the microwires, such as their DC resistivity and the doping distribution along the length of the microwires, were investigated using this approach. The technique was also used to characterize the junction between silicon microwires and conducting polymers to assess the suitability of such contacts for use in a proposed artificial photosynthesis system.Keywords: artificial photosynthesis system; conducting polymer; junction behavior; silicon microwire; soft contact formation;
Co-reporter:Edgardo García-Berríos, Ting Gao, Don Walker, Bruce S. Brunschwig, Nathan S. Lewis
Sensors and Actuators B: Chemical 2011 Volume 158(Issue 1) pp:17-22
Publication Date(Web):15 November 2011
DOI:10.1016/j.snb.2011.04.022
Titanium (IV) dioxide (TiO2) nanoparticles (NPs) with a 1–5 nm diameter were synthesized by a sol–gel method, functionalized with carboxylate ligands, and combined with carbon black (CB) to produce chemiresistive chemical vapor sensor films. The TiO2 acted as an inorganic support phase for the swellable, organic capping groups of the NPs, and the CB imparted electrical conductivity to the film. Such sensor composite films exhibited a reproducible, reversible change in relative differential resistance upon exposure to a series of organic test vapors. The response of such chemiresistive composites was comparable to, but generally somewhat smaller than, that of thiol-capped Au NPs. For a given analyte, the resistance response and signal-to-noise ratio of the capped TiO2-NP/CB composites varied with the identity of the capping ligand. Hence, an array of TiO2-NP/CB composites, with each film having a compositionally different carboxylate capping ligand, provided good vapor discrimination and quantification when exposed to a series of organic vapors. Principal components analysis of the relative differential resistance response of the sensor array revealed a clear clustering of the response for each analyte tested. This approach expands the options for composite-based chemiresistive vapor sensing, from use of organic monomeric or polymeric sorbent phases, to use of electrically insulating capped inorganic NPs as the nonconductive phase of chemiresistive composite vapor sensors.
Co-reporter:Erik Johansson ; Shannon W. Boettcher ; Leslie E. O’Leary ; Andrey D. Poletayev ; Stephen Maldonado ; Bruce S. Brunschwig
The Journal of Physical Chemistry C 2011 Volume 115(Issue 17) pp:8594-8601
Publication Date(Web):April 8, 2011
DOI:10.1021/jp109799e
The open-circuit potentials of p-Si/((MV2+/MV+)(aq)) junctions with Si(111) surfaces functionalized with H−, CH3−, CH2CHCH2−, or mixed CH3−/CH2CHCH2− monolayers have been investigated as the solution pH was changed from 2.5 to 11. The pH sensitivity of the open-circuit potentials, and therefore the band-edge positions, was anticorrelated with the total fraction of Si atop sites that were terminated by Si−C bonds. This behavior is consistent with the hypothesis that the non Si−C terminated atop sites were initially H-terminated and were unstable to oxide growth under aqueous conditions with the oxidation-product inducing a pH-dependent dipole. Metal-semiconductor junctions between Hg and CH3-, CH2CHCH2-, or mixed CH3-/CH2CHCH2-terminated n-Si(111) surfaces formed rectifying Hg/Si Schottky junctions and exhibited mutually similar barrier-heights (∼0.9 V), suggesting similar magnitudes and direction of the surface dipoles on all of these functionalized surfaces.
Co-reporter:David Knapp ; Bruce S. Brunschwig
The Journal of Physical Chemistry C 2011 Volume 115(Issue 33) pp:16389-16397
Publication Date(Web):June 16, 2011
DOI:10.1021/jp110550t
The surface chemistry of CH3–, CD3–, and C10H21–Ge(111) surfaces prepared through a bromination/alkylation method have been investigated by infrared spectroscopy. Well-ordered CH3–Ge(111) surfaces could be prepared only if, prior to bromination, the surface was etched with 6.0 M HCl or with a two-step etch of H2O2 (1.5 M)/HF (5.1 M) followed by a short HF (6.0 M) etch. The etching method used to make the Ge precursor surface, and the formation of a bromine-terminated intermediate Ge surface, were of critical importance to obtain clear, unambiguous infrared absorption peaks on the methyl-terminated Ge surfaces. Polarization-dependent absorption peaks observed at 1232 cm–1 for CH3–Ge(111) surfaces and at 951 cm–1 for CD3–Ge(111) surfaces were assigned to the methyl “umbrella” vibrational mode. A polarization-dependent peak at 2121 cm–1 for CD3–Ge(111) surfaces was assigned to the symmetric methyl stretching mode. Polarization-independent absorption peaks at 755 cm–1 for CH3–Ge(111) and at 577 cm–1 for CD3–Ge(111) were assigned to the methyl rocking mode. These findings provide spectroscopic evidence that the methyl monolayer structure on the alkylated Ge is well-ordered and similar to that on analogous Si(111) surfaces, despite differences in the composition of the precursor surfaces. The X-ray photoelectron spectra of CH3–Ge(111) surfaces, however, were not highly dependent upon the etching method and showed a constant C 1s:Ge 3d ratio, independent of the etching method. The infrared spectra of C10H21–Ge(111) surfaces were also not sensitive to the initial etching method. Hence, while the final packing density of the alkyl groups on the surface was similar for all etch methods studied, not all methods yielded a well-ordered Ge(111)/overlayer interface.
Co-reporter:Emily L. Warren ; Shannon W. Boettcher ; Michael G. Walter ; Harry A. Atwater
The Journal of Physical Chemistry C 2011 Volume 115(Issue 2) pp:594-598
Publication Date(Web):December 23, 2010
DOI:10.1021/jp109147p
The effects of introducing an n+-doped emitter layer have been evaluated for both planar Si photoelectrodes and for radial junction Si microwire-array photoelectrodes. In contact with the pH-independent, one-electron, outer-sphere, methyl viologen redox system (denoted MV2+/+), both planar and wire array p-Si photoelectrodes yielded open-circuit voltages, Voc, that varied with the pH of the solution. The highest Voc values were obtained at pH = 2.9, with Voc = 0.53 V for planar p-Si electrodes and Voc = 0.42 V for vapor−liquid−solid catalyzed p-Si microwire array samples, under 60 mW cm−2 of 808 nm illumination. Increases in the pH of the electrolyte produced a decrease in Voc by approximately −44 mV/pH unit for planar electrodes, with similar trends observed for the Si microwire array electrodes. In contrast, introduction of a highly doped, n+ emitter layer produced Voc = 0.56 V for planar Si electrodes and Voc = 0.52 V for Si microwire array electrodes, with the photoelectrode properties in each system being essentially independent of pH over six pH units (3 < pH < 9). Hence, formation of an n+ emitter layer not only produced nearly identical photovoltages for planar and Si microwire array photoelectrodes, but decoupled the band energetics of the semiconductor (and hence the obtainable photovoltage) from the value of the redox potential of the solution. The formation of radial junctions on Si microwire arrays thus provides an approach to obtaining Si-based photoelectrodes with high-photovoltages that can be used for a variety of photoelectrochemical processes, including potentially the hydrogen evolution reaction, under various pH conditions, regardless of the intrinsic barrier height and flat-band properties of the Si/liquid contact.
Co-reporter:Edgardo García-Berríos ; Ting Gao ; Jordan C. Theriot ; Marc D. Woodka ; Bruce S. Brunschwig
The Journal of Physical Chemistry C 2011 Volume 115(Issue 14) pp:6208-6217
Publication Date(Web):March 15, 2011
DOI:10.1021/jp110793h
The response and discrimination performance of an array that consisted of 20 different organothiol-capped Au nanoparticle chemiresistive vapor sensors was evaluated during exposure to 13 different organic vapors. The passivating organothiol ligand library consisted of collections of straight-chain alkanethiols, branched alkanethiols, and aromatic thiols. A fourth collection of sensors was formed from composites of 2-phenylethanethiol-capped Au nanoparticles and nonpolymeric aromatic materials that were coembedded in a sensor film. The organic vapors consisted of six hydrocarbons (n-hexane, n-heptane, n-octane, isooctane, cyclohexane, and toluene), three polar aprotic vapors (chloroform, tetrahydrofuran, and ethyl acetate), and four alcohols (methanol, ethanol, isopropanol, and 1-butanol). Trends in the resistance response of the sensors were consistent with expected trends in sorption due to the properties of the test vapor and the molecular structure of the passivating ligands in the sensor films. Classification algorithms including principal components analysis and Fisher’s linear discriminant were used to evaluate the discrimination performance of an array of such sensors. Each collection of sensors produced accurate classification of most vapors, with misclassification occurring primarily for vapors that had mutually similar polarity. The classification performance for an array that contained all of the sensor collections produced nearly perfect discrimination for all vapors studied. The dependence of the array size (i.e., the number of sensors) and the array chemical diversity on the discrimination performance indicated that, for an array of 20 sensors, an array size of 13 sensors or more produced the maximum discrimination performance.
Co-reporter:Michael G. Walter, Emily L. Warren, James R. McKone, Shannon W. Boettcher, Qixi Mi, Elizabeth A. Santori, and Nathan S. Lewis
Chemical Reviews 2010 Volume 110(Issue 11) pp:6446-6473
Publication Date(Web):November 10, 2010
DOI:10.1021/cr1002326
Co-reporter:Edgardo García-Berríos ; Ting Gao ; Marc D. Woodka ; Stephen Maldonado ; Bruce S. Brunschwig ; Mark W. Ellsworth
The Journal of Physical Chemistry C 2010 Volume 114(Issue 50) pp:21914-21920
Publication Date(Web):August 25, 2010
DOI:10.1021/jp101331g
Au nanoparticles capped with a homologous series of straight chain alkanethiols (containing 4−11 carbons in length) have been investigated as chemiresistive organic vapor sensors. The series of alkanethiols was used to elucidate the mechanisms of vapor detection by such capped nanoparticle chemiresistive films and to highlight the molecular design principles that govern enhanced detection. The thiolated Au nanoparticle chemiresistors demonstrated rapid and reversible responses to a set of test vapors (n-hexane, n-heptane, n-octane, iso-octane, cyclohexane, toluene, ethyl acetate, methanol, ethanol, isopropanol, and 1-butanol) that possessed a variety of analyte physicochemical properties. The resistance sensitivity to nonpolar and aprotic polar vapors systematically increased as the chain length of the capping reagent increased. Decreases in the nanoparticle film resistances, which produced negative values of the differential resistance response, were observed upon exposure of the sensor films to alcohol vapors. The response signals became more negative with higher alcohol vapor concentrations, producing negative values of the sensor sensitivity. Sorption data measured on Au nanoparticle chemiresistor films using a quartz crystal microbalance allowed for the measurement of the partition coefficients of test vapors in the Au nanoparticle films. This measurement assumed that analyte sorption only occurred at the organic interface and not the surface of the Au core. Such an assumption produced partition coefficient values that were independent of the length of the ligand. Furthermore, the value of the partition coefficient was used to obtain the particle-to-particle interfacial effective dielectric constant of films upon exposure to analyte vapors. The values of the dielectric constant upon exposure to alcohol vapors suggested that the observed resistance response changes observed were not significantly influenced by this dielectric change, but rather were primarily influenced by morphological changes and by changes in the interparticle spacing.
Co-reporter:Morgan C. Putnam;James R. Maiolo;Shannon W. Boettcher;Daniel B. Turner-Evans;Harry A. Atwater;Joshua M. Spurgeon;Michael D. Kelzenberg;Emily L. Warren
Science 2010 Volume 327(Issue 5962) pp:185-187
Publication Date(Web):08 Jan 2010
DOI:10.1126/science.1180783
Co-reporter:Leslie E. O’Leary, Erik Johansson, Bruce S. Brunschwig, and Nathan S. Lewis
The Journal of Physical Chemistry B 2010 Volume 114(Issue 45) pp:14298-14302
Publication Date(Web):August 19, 2010
DOI:10.1021/jp911379c
The formation of mixed methyl/allyl monolayers has been accomplished through a two-step halogenation/alkylation reaction on Si(111) surfaces. The total coverage of alkylated Si, the surface recombination velocities, and the degree of surface oxidation as a function of time have been investigated using X-ray photoelectron spectroscopy, Fourier-transform infrared spectroscopy, and microwave conductivity measurements. The total coverage of alkyl groups, the rate of oxidation, and the surface recombination velocities of Si(111) terminated by mixed monolayers were found to be close to those observed for CH3−Si(111) surfaces. Hence, the mixed-monolayer surfaces retained the beneficial properties of CH3−Si(111) surfaces while allowing for convenient secondary surface functionalization.
Co-reporter:Amanda L. Smeigh ; Jordan E. Katz ; Bruce S. Brunschwig ; Nathan S. Lewis ;James K. McCusker
The Journal of Physical Chemistry C 2008 Volume 112(Issue 32) pp:12065-12068
Publication Date(Web):July 23, 2008
DOI:10.1021/jp803402x
The electron injection dynamics of dye-sensitized TiO2-based solar cells have been investigated to determine the effects of replacing the I3−/I− redox system by non-redox-active supporting electrolytes. TiO2 films were sensitized with Ru(dcbpy)2(NCS)2, where dcbpy = 4,4′-dicarboxylic acid-2,2′-bipyridine (the “N3” dye), and placed in contact with either M(ClO4) or M(I3−/I−) solutions (M = Li+ or (n-C4H9)4N+); cells that contained I3−/I− were fully functional solar cells whose steady-state photocurrents were directly measured. In (n-C4H9)4N+-containing solutions, significant differences were observed between the measured kinetics when ClO4− was replaced by the redox-active I3−/I− system. In particular, a ps time scale loss of the metal-to-ligand charge-transfer excited-state of the N3 dye, associated with electron injection, that was observed in cells containing either LiClO4 or [(n-C4H9)4N]ClO4 was absent in fully functional solar cells that contained [(n-C4H9)4N]I/I2. These results underscore the importance of performing kinetics measurements on this class of solar cells under operational conditions if one is to obtain reliable correlations between the dynamics data and the steady-state performance metrics of the solar cell devices.
Co-reporter:Matthew C. Traub, Julie S. Biteen, David J. Michalak, Lauren J. Webb, Bruce S. Brunschwig and Nathan S. Lewis
The Journal of Physical Chemistry C 2008 Volume 112(Issue 47) pp:18467-18473
Publication Date(Web):2017-2-22
DOI:10.1021/jp803992h
Phosphorus-functionalized GaAs surfaces have been prepared by exposure of Cl-terminated GaAs(111)A surfaces to triethylphosphine (PEt3) or trichlorophosphine (PCl3), or by the direct functionalization of the native-oxide terminated GaAs(111)A surface with PCl3. The presence of phosphorus on each functionalized surface was confirmed by X-ray photoelectron spectroscopy. High-resolution, soft X-ray photoelectron spectroscopy was used to evaluate the As and Ga 3d regions of such surfaces. On PEt3 treated surfaces, the Ga 3d spectra exhibited a bulk Ga peak as well as peaks that were shifted to 0.35, 0.92 and 1.86 eV higher binding energy. These peaks were assigned to residual Cl-terminated Ga surface sites, surficial Ga2O and surficial Ga2O3, respectively. For PCl3-treated surfaces, the Ga 3d spectra displayed peaks ascribable to bulk Ga(As), Ga2O, and Ga2O3, as well as a peak shifted 0.30 eV to higher binding energy relative to the bulk signal. A peak corresponding to Ga(OH)3, observed on the Cl-terminated surface, was absent from all of the phosphine-functionalized surfaces. After reaction of the Cl-terminated GaAs(111)A surface with PCl3 or PEt3, the As 3d spectral region was free of As oxides and As0. Although native oxide-terminated GaAs surfaces were free of As oxides after reaction with PCl3, such surfaces contained detectable amounts of As0. Photoluminescence measurements indicted that phosphine-functionalized surfaces prepared from Cl-terminated GaAs(111)A surfaces had better electrical properties than the native-oxide capped GaAs(111)A surface, while the native-oxide covered surface treated with PCl3 showed no enhancement in PL intensity.
Co-reporter:Kimberly M. Papadantonakis, Bruce S. Brunschwig and Nathan S. Lewis
Langmuir 2008 Volume 24(Issue 19) pp:10543-10548
Publication Date(Web):September 3, 2008
DOI:10.1021/la801285u
Scanning tunneling microscopy has been used to determine the molecular ordering in stable, ordered monolayers formed from long-chain normal and substituted alkanes in solution on highly oriented pyrolytic graphite surfaces. Monolayers were initially formed using an overlying solution of either a symmetrical dialkylthioether or a symmetrical dialkylether. Initially pure thioether solutions were then changed to nearly pure solutions of the identical chain-length ether, and vice versa. The direct application of a pure solution of long-chain symmetrical ethers onto graphite produced a lamellate monolayer within which the individual molecular axes were oriented at an angle of ∼65° to the lamellar axes. In contrast, a pure solution of long-chain symmetrical thioethers on graphite produced a monolayer within which the molecular axes were oriented perpendicular to the lamellar axes. When ethers were gradually added to solutions overlying pure thioether monolayers, the ethers substituted into the existing monolayer structure. Thus, the ether molecules could be forced to orient in the perpendicular thioether-like manner through the use of a thioether template monolayer. Continued addition of ethers to the solution ultimately produced a nearly pure ether monolayer that retained the orientation of the thioether monolayer template. However, a monolayer of thioether molecules formed by gradual substitution into an ether monolayer did not retain the 65° orientation typical of dialkylethers, but exhibited the 90° orientation typical of dialkylthioether monolayers. The thioethers and ethers were easily distinguished in images of mixed monolayers, allowing both an analysis of the distribution of the molecules within the mixed monolayers and a comparison of the monolayer compositions with those of the overlying solutions. Substitution of molecules into the template monolayer did not proceed randomly; instead, a molecule within a monolayer was more likely to be replaced by a molecule in the overlying solution if it was located next to a molecule that had already been replaced.
Co-reporter:Ralf Hunger, Rainer Fritsche, Bengt Jaeckel, Lauren J. Webb, Wolfram Jaegermann, Nathan S. Lewis
Surface Science 2007 Volume 601(Issue 14) pp:2896-2907
Publication Date(Web):15 July 2007
DOI:10.1016/j.susc.2007.04.249
Co-reporter:Matteo Pardo, Brian C. Sisk, Giorgio Sberveglieri, Nathan S. Lewis
Sensors and Actuators B: Chemical 2006 Volume 115(Issue 2) pp:647-655
Publication Date(Web):26 June 2006
DOI:10.1016/j.snb.2005.10.033
Two different classification methods, Fisher's linear discriminant (FLD) and a multilayer perceptron neural network (MLP), were directly compared with respect to their abilities to differentiate response patterns arising from arrays of chemical vapor detectors. The algorithms were compared in five different types of tasks that had been selected because they produced classification problems of varying character and difficulty. In one task, an array of 20 compositionally distinct carbon black–polymer composite vapor detectors was exposed to P/P0 = 0.0075 1-propanol and P/P0 = 0.0083 2-propanol, where P and P0 are the partial pressure and standard vapor pressure, respectively, of a given analyte. The second task consisted of classification of a mixture of P/P0 = 0.011 1-propanol and P/P0 = 0.0090 2-propanol versus a mixture of P/P0 = 0.0090 1-propanol and P/P0 = 0.011 2-propanol. A third task consisted of multiple concentrations of three hydrocarbons, and a fourth task involved clustering two hydrocarbons in the presence of a variable background composition. An additional dataset was generated by exposing an array of five thin-film metal-oxide sensors to the headspace of seven different coffee blends. In each case, the MLP and FLD techniques were compared using the 5-sensor subset of the 20 available sensors that proved optimal for that dataset. The FLD and MLP algorithms yielded comparable performance on straightforward classification tasks, whereas the MLP technique yielded better performance on tasks that involved non-linear classification boundaries. In addition, for the four datasets produced by the carbon black–polymer composite detector array, the performance of each possible 5-sensor subset was evaluated using both signal processing approaches. The performance of the best 5-sensor subset selected with MLP was found to be slightly better than the performance of the FLD-selected subsets, and the performance of the median 5-sensor subset using MLP was nearer to that of the optimal subset than the median sensor array selected by FLD. In one case, the optimal test set performance distribution was found to be significantly better with MLP than with FLD: MLP had a clear advantage (86% versus 57% correct classification rate) when applied to the “coffees” dataset, and this trend is likely applicable to other multi-cluster classification tasks that consisted of non-Gaussian shaped data in lower-dimensional spaces.
Co-reporter:Brian C. Sisk, Nathan S. Lewis
Sensors and Actuators B: Chemical 2005 Volume 104(Issue 2) pp:249-268
Publication Date(Web):24 January 2005
DOI:10.1016/j.snb.2004.05.010
The responses of 15 carbon black-polymer composite chemiresistors have been analyzed during exposure to eight different analytes (n-hexane, tetrahydrofuran, ethanol, ethyl acetate, cyclohexane, n-heptane, n-octane, and isooctane) in random order at low concentration (0.5% of the vapor pressure of analyte at room temperature) over 4 months (8000 total analyte exposures) of data collection. Data were collected for periods during which the array was continuously exposed periodically to analytes and after long periods during which no analyte exposures had been performed. All but the most difficult separation tasks (for example, discrimination between low concentrations of straight-chain hydrocarbons) could be performed robustly over the entire 4 month time period based only on the use of a decision boundary formulated from an initial training set of 200 exposures, indicating the sensor drift had minimal effect on system performance in such classification tasks. For the remaining classification tasks, modeling the dynamics of sensor drift either through a linear regression or Fourier transform decomposition of the individual relative differential resistance responses versus time of each sensor yielded little improvement in classification performance, indicating that external events were largely responsible for changes in sensor response versus time. Six analytes that were not treated as unknowns for a binary separation task were individually treated as calibrants whose response was intermittantly used to renormalize the response of the sensor array. A simple linear sensor-by-sensor calibration scheme proved effective at restoring the classification performance of difficult binary separation tasks to the performance that was observed in the initial training set period. Calibrants that were mutually similar to the analytes being differentiated tended to be more effective than calibrants that were very chemically different from the analytes of interest. Evaluation of various calibration protocols indicated that an optimal tradeoff existed between the number of calibration exposures and the frequency of calibration periods. Condition-based calibration, in which calibration was only performed when the classification model exhibited a decline in classification performance below a predetermined threshold value, was observed to be superior to a time-based calibration approach or to interval-based, cyclic calibration protocols for this set of analytes exposed under the chosen analysis conditions.
Co-reporter:Brian C. Sisk, Nathan S. Lewis
Sensors and Actuators B: Chemical 2003 Volume 96(1–2) pp:268-282
Publication Date(Web):15 November 2003
DOI:10.1016/S0925-4005(03)00543-4
Analysis of the signals produced by a collection of organic polymer-carbon black composite vapor detectors has been performed to assess the ability to estimate various chemical and physical properties of analyte vapors based on information contained in the response patterns of the detector array. A diverse array of composite chemiresistive vapor detectors was exposed to a series of 75 test analytes that had been selected from among five different chemical classes: alcohols, halogenated hydrocarbons, aromatics, unsubstituted hydrocarbons, and esters. The algorithmic task of interest was to use the resulting array of response data to assign one of the five chemical class labels to a test analyte, despite having left that analyte out of the model used to generate the class labels. Algorithms evaluated for this purpose included principal components analysis (PCA) and k-nearest neighbor (k-NN) analysis employing either Euclidean or Mahalanobis distance calculations. Each data cluster that was produced by replicate exposures to an individual analyte was well resolved from all of the other 74 analyte clusters. Furthermore, with the exception of the halide cluster, the analyte response clusters could be robustly grouped into supersets such that each of the five individual chemical classes was well-separated from every other class of analytes in principal component space. Accordingly, using either of the k-nearest neighbor algorithms, in excess of 85% of the non-halide test analyte exposures were correctly assigned to their chemical classes, and halides were only routinely confused with aromatics or esters but not with alcohols or hydrocarbons. The detector array response data also was found to contain semi-quantitative information regarding physicochemical properties of the members of the test analyte series, such as the degree of unsaturation of the carbon chain, the dipole moment, the molecular weight, the number of halogen atoms, and type of aromatic ring in the test analytes. The performance in these types of tasks is relevant for applications of a semi-selective array of vapor detectors in situations when no prior knowledge of the analyte identity is available and when there is no assurance that the test analyte will have been contained in the training set database produced by a compiling a library of responses from the detector array.
Co-reporter:Eric S. Tillman, Nathan S. Lewis
Sensors and Actuators B: Chemical 2003 Volume 96(1–2) pp:329-342
Publication Date(Web):15 November 2003
DOI:10.1016/S0925-4005(03)00567-7
Enhanced sensitivity towards volatile carboxylic acid vapors is obtained when the basic, amine-containing polymer, linear poly(ethylenimine), l-PEI, is used as the insulating component in a carbon black-polymer composite vapor detector. Specifically, at a partial pressure of analyte corresponding to 1% of its vapor pressure at room temperature, the signal-to-noise ratio for detection of acetic acid is 103 times larger than that for non-acidic organic vapors. Measurements of the mass uptake, thickness change, and electrical conductivity of such composites have been performed to elucidate the mechanism of this sensitivity enhancement towards volatile carboxylic acid vapors. These data have allowed quantification of the relative contributions of electrical percolation effects, increases in analyte sorption, and charge-induced swelling effects in determining the response characteristics of l-PEI composites.
Co-reporter:Shawn M Briglin, Michael S Freund, Phil Tokumaru, Nathan S Lewis
Sensors and Actuators B: Chemical 2002 Volume 82(Issue 1) pp:54-74
Publication Date(Web):1 February 2002
DOI:10.1016/S0925-4005(01)00991-1
Co-reporter:Michael C Burl, Brian C Sisk, Thomas P Vaid, Nathan S Lewis
Sensors and Actuators B: Chemical 2002 Volume 87(Issue 1) pp:130-149
Publication Date(Web):15 November 2002
DOI:10.1016/S0925-4005(02)00229-0
The vapor classification performance of arrays of conducting polymer composite vapor detectors has been evaluated as a function of the number and type of detectors in an array. Quantitative performance comparisons were facilitated by challenging a collection of detector arrays with vapor discrimination tasks that were sufficiently difficult that at least some of the arrays did not exhibit perfect classification ability for all of the tasks of interest. Specific discrimination tasks involved differentiating between low concentration (<1% of the vapor pressure) exposures to 1-propanol versus 2-propanol, low concentration exposures to n-hexane versus n-heptane, and differentiating between compositionally similar mixtures of closely related analytes, such as 9.37 ppm m-xylene with 10.2 ppm p-xylene versus 7.67 ppm m-xylene with 12.4 ppm p-xylene. A decision boundary was developed using a cross-validated Fisher linear discriminant algorithm on a training set of analyte presentations and the resulting chemometric model was then used to classify a subsequent collection of test analyte presentations to the array being evaluated. In other cases, classification performance was evaluated using the Fisher linear discriminant and a leave-one-out (LOO) cross-validation procedure. For nearly all of the discrimination tasks investigated in this work, classification performance either increased or did not significantly decrease as the number of chemically different detectors in the array increased. Any given subset of the full array of detectors, selected because it yielded the best classification performance at a given array size for one particular task, was invariably outperformed by a different subset of detectors, and by the entire array of 20 chemically diverse detectors when used in at least one other vapor discrimination task. Arrays of detectors were nevertheless identified that yielded robust discrimination performance between compositionally close mixtures of 1-propanol and 2-propanol, n-hexane and n-heptane, and m-xylene and p-xylene, attesting to the excellent analyte classification performance that can be obtained through the use of such semi-selective vapor detector arrays.
Co-reporter:Brett J Doleman, Nathan S Lewis
Sensors and Actuators B: Chemical 2001 Volume 72(Issue 1) pp:41-50
Publication Date(Web):5 January 2001
DOI:10.1016/S0925-4005(00)00635-3
Response data from an array of conducting polymer composite vapor detectors that form an electronic nose were collected for the purpose of comparing selected, quantitatively measurable, phenomena in odor detection and classification to the olfactory characteristics of monkeys and humans. Odor detection thresholds and discriminability between structurally similar pairs of odorants were the two primary quantities evaluated for this comparison. Comparisons were only made for volatile organic vapors as opposed to aroma active odorant vapors. Electronic nose detection thresholds for a homologous series of n-alkane and 1-alcohol odorants were determined and the results were compared to literature values for the mean olfactory detection thresholds observed in psychophysical experiments on humans exposed to these same vapors. The trends in odor detection thresholds of the electronic nose towards the tested analytes were very similar to those exhibited by humans. The discrimination performance of the electronic nose for distinguishing between pairs of odorants within incrementally varying series of esters, carboxylic acids and alcohols were also compared to the published data of Laska and co-workers on the psychophysical performance of humans and monkeys for these same odorant pairs. Similar trends were generally observed between the humans, monkeys, and the electronic nose in that discrimination performance increased as the compounds of an odorant pair became more structurally dissimilar. With use of the Fisher linear discriminant algorithm for classification of these test pairs of odorants, the electronic nose exhibited significantly better discriminability than humans or monkeys for the odorant pairs evaluated in this work under the test conditions for which the discriminability was evaluated.
Co-reporter:Michael C. Burl, Brett J. Doleman, Amanda Schaffer, Nathan S. Lewis
Sensors and Actuators B: Chemical 2001 Volume 72(Issue 2) pp:149-159
Publication Date(Web):25 January 2001
DOI:10.1016/S0925-4005(00)00645-6
The responses of a conducting polymer composite “electronic nose” detector array were used to predict human perceptual descriptors of odor quality for a selected test set of analytes. The single-component odorants investigated in this work included molecules that are chemically quite distinct from each other, as well as molecules that are chemically similar to each other but which are perceived as having distinct odor qualities by humans. Each analyte produced a different, characteristic response pattern on the electronic nose array, with the signal strength on each detector reflecting the relative binding of the odorant into the various conducting polymer composites of the detector array. A “human perceptual space” was defined by reference to English language descriptors that are frequently used to describe odors. Data analysis techniques, including standard regression, nearest-neighbor prediction, principal components regression, partial least squares regression, and feature subset selection, were then used to determine mappings from electronic nose measurements to this human perceptual space. The effectiveness of the derived mappings was evaluated by comparison with average human perceptual data published by Dravnieks. For specific descriptors, some models provided cross-validated predictions that correlated well with the human data (above the 0.60 level), but none of the models could accurately predict the human values for more than a few descriptors.
Co-reporter:Azhar I. Carim, Fadl H. Saadi, Manuel P. Soriaga and Nathan S. Lewis
Journal of Materials Chemistry A 2014 - vol. 2(Issue 34) pp:NaN13839-13839
Publication Date(Web):2014/07/24
DOI:10.1039/C4TA02611J
Using an electrochemical method under ambient conditions, crystallographically amorphous films of cobalt selenide have been deposited from aqueous solution onto planar Ti supports. These films have been evaluated as electrocatalysts for the hydrogen-evolution reaction. In 0.500 M H2SO4, the cobalt selenide films required an overpotential of ∼135 mV to drive the hydrogen-evolution reaction at a benchmark current density of −10 mA cm−2. Galvanostatic measurements indicated stability of the electrocatalytic films for >16 h of continuous operation at −10 mA cm−2. The facile preparation method, and the activity of the cobalt selenide films, suggest that electrodeposited metal chalcogenides are potentially attractive earth-abundant electrocatalysts for the hydrogen-evolution reaction.
Co-reporter:Eric J. Popczun, Christopher W. Roske, Carlos G. Read, J. Chance Crompton, Joshua M. McEnaney, Juan F. Callejas, Nathan S. Lewis and Raymond E. Schaak
Journal of Materials Chemistry A 2015 - vol. 3(Issue 10) pp:NaN5425-5425
Publication Date(Web):2015/01/16
DOI:10.1039/C4TA06642A
CoP nanostructures that exposed predominantly (111) crystal facets were synthesized and evaluated for performance as electrocatalysts for the hydrogen-evolution reaction (HER). The branched CoP nanostructures were synthesized by reacting cobalt(II) acetylacetonate with trioctylphosphine in the presence of trioctylphosphine oxide. Electrodes comprised of the branched CoP nanostructures deposited at a loading density of ∼1 mg cm−2 on Ti electrodes required an overpotential of −117 mV to produce a current density of −20 mA cm−2 in 0.50 M H2SO4. Hence the branched CoP nanostructures belong to the growing family of highly active non-noble-metal HER electrocatalysts. Comparisons with related CoP systems have provided insights into the impact that shape-controlled nanoparticles and nanoparticle–electrode interactions have on the activity and stability of nanostructured HER electrocatalysts.
Co-reporter:Judith R. C. Lattimer, James D. Blakemore, Wesley Sattler, Sheraz Gul, Ruchira Chatterjee, Vittal K. Yachandra, Junko Yano, Bruce S. Brunschwig, Nathan S. Lewis and Harry B. Gray
Dalton Transactions 2014 - vol. 43(Issue 40) pp:NaN15012-15012
Publication Date(Web):2014/07/17
DOI:10.1039/C4DT01149J
Silicon(111) surfaces have been functionalized with mixed monolayers consisting of submonolayer coverages of immobilized 4-vinyl-2,2′-bipyridyl (1, vbpy) moieties, with the remaining atop sites of the silicon surface passivated by methyl groups. As the immobilized bipyridyl ligands bind transition metal ions, metal complexes can be assembled on the silicon surface. X-ray photoelectron spectroscopy (XPS) demonstrates that bipyridyl complexes of [Cp*Rh], [Cp*Ir], and [Ru(acac)2] were formed on the surface (Cp* is pentamethylcyclopentadienyl, acac is acetylacetonate). For the surface prepared with Ir, X-ray absorption spectroscopy at the Ir LIII edge showed an edge energy as well as post-edge features that were essentially identical with those observed on a powder sample of [Cp*Ir(bpy)Cl]Cl (bpy is 2,2′-bipyridyl). Charge-carrier lifetime measurements confirmed that the silicon surfaces retain their highly favorable photoelectronic properties upon assembly of the metal complexes. Electrochemical data for surfaces prepared on highly doped, n-type Si(111) electrodes showed that the assembled molecular complexes were redox active. However the stability of the molecular complexes on the surfaces was limited to several cycles of voltammetry.
Co-reporter:Joshua M. McEnaney, J. Chance Crompton, Juan F. Callejas, Eric J. Popczun, Carlos G. Read, Nathan S. Lewis and Raymond E. Schaak
Chemical Communications 2014 - vol. 50(Issue 75) pp:NaN11028-11028
Publication Date(Web):2014/07/30
DOI:10.1039/C4CC04709E
Amorphous tungsten phosphide (WP), which has been synthesized as colloidal nanoparticles with an average diameter of 3 nm, has been identified as a new electrocatalyst for the hydrogen-evolution reaction (HER) in acidic aqueous solutions. WP/Ti electrodes produced current densities of −10 mA cm−2 and −20 mA cm−2 at overpotentials of only −120 mV and −140 mV, respectively, in 0.50 M H2SO4(aq).
.jpg)
![[(sulfonatoperoxy)sulfonyl]oxidanide](http://img.cochemist.com/ccimg/15100/15092-81-6.png)
![[(sulfonatoperoxy)sulfonyl]oxidanide](http://img.cochemist.com/ccimg/15100/15092-81-6_b.png)
![Bis[(pentamethylcyclopentadienyl)dichloro-rhodium]](http://img.cochemist.com/ccimg/12400/12354-85-7.png)
![Bis[(pentamethylcyclopentadienyl)dichloro-rhodium]](http://img.cochemist.com/ccimg/12400/12354-85-7_b.png)
