We describe the development of solar water-splitting cells comprising earth-abundant elements that operate in near-neutral pH conditions, both with and without connecting wires. The cells consist of a triple junction, amorphous silicon photovoltaic interfaced to hydrogen- and oxygen-evolving catalysts made from an alloy of earth-abundant metals and a cobalt|borate catalyst, respectively. The devices described here carry out the solar-driven water-splitting reaction at efficiencies of 4.7% for a wired configuration and 2.5% for a wireless configuration when illuminated with 1 sun (100 milliwatts per square centimeter) of air mass 1.5 simulated sunlight. Fuel-forming catalysts interfaced with light-harvesting semiconductors afford a pathway to direct solar-to-fuels conversion that captures many of the basic functional elements of a leaf.

One of the important factors limiting solar-cell efficiency is that incident photons generate one electron-hole pair, irrespective of the photon energy. Any excess photon energy is lost as heat. The possible generation of multiple charge carriers per photon (carrier multiplication) is therefore of great interest for future solar cells 1 . Carrier multiplication is known to occur in bulk semiconductors, but has been thought to be enhanced significantly in nanocrystalline materials such as quantum dots, owing to their discrete energy levels and enhanced Coulomb interactions 1-3 . Contrary to this expectation, we demonstrate here that, for a given photon energy, carrier multiplication occurs more efficiently in bulk PbS and PbSe than in quantum dots of the same materials. Measured carriermultiplication efficiencies in bulk materials are reproduced quantitatively using tight-binding calculations, which indicate that the reduced carrier-multiplication efficiency in quantum dots can be ascribed to the reduced density of states in these structures.Carrier multiplication is the process in which the absorption of a single, high-energy photon results in the generation of two or more electron-hole pairs. The excess energy of the initially excited electron is used to excite a second electron over the bandgap, rather than being converted into heat through sequential phonon emission. Carrier multiplication is important for the operation for high-speed electronic devices 4 , but is especially relevant for solar cells 1 , because relaxation of hot carriers through phonon emission is a common loss mechanism in bulk semiconductor solar cells. In this context, semiconductor quantum dots are promising building blocks for future solar cells 1 . In addition to the size-tunability of the quantum-dot optical properties, the carrier-multiplication efficiency in quantum dots was reported to be much higher than in bulk materials, where the process is generally referred to as impact ionization. It has been argued that carrier multiplication is more efficient in nanostructured semiconductors owing to quantum-confinement effects causing (1) a slowing of the phonon-mediated relaxation channel 1 and (2) enhanced Coulomb interactions 2 , resulting from forced overlap between wavefunctions and reduced dielectric screening at the quantum-dot surface 3 . In recent years, several femtosecond spectroscopy studies have revealed highly efficient carrier multiplication in PbSe and PbS (refs 2, 5-9) (refs 18, 19) quantum dots. In initial studies, carrier-multiplication efficiencies may have been overestimated owing to several experimental complications, including too high excitation fluences (generating multiple carriers by sequential absorption of multiple photons), lack of stirring of quantum-dot suspensions (causing photo-induced charging) and sample-to-sample variability 19 . Furthermore, recent tight-binding calculations 20 suggest carrier multiplication in quantum dots is not only not enhanced relative to bulk, but is actually lower. Answering the...

Integrating a silicon solar cell with a recently developed cobaltbased water-splitting catalyst (Co-Pi) yields a robust, monolithic, photo-assisted anode for the solar fuels process of water splitting to O 2 at neutral pH. Deposition of the Co-Pi catalyst on the Indium Tin Oxide (ITO)-passivated p-side of a np-Si junction enables the majority of the voltage generated by the solar cell to be utilized for driving the water-splitting reaction. Operation under neutral pH conditions fosters enhanced stability of the anode as compared to operation under alkaline conditions (pH 14) for which long-term stability is much more problematic. This demonstration of a simple, robust construct for photo-assisted water splitting is an important step towards the development of inexpensive direct solar-to-fuel energy conversion technologies.photoelectrochemical | hydrogen | solar energy | storage P hotosynthetic organisms convert the energy of sunlight into chemical energy by splitting water, producing molecular oxygen and hydrogen equivalents in the highly conserved enzyme complex photosystem II (PSII) (1). Absorbed photons are transferred to the reaction center of PSII, where a single electron/hole charge separation occurs. The oxidative power of the photo-produced hole in PSII is transferred to the oxygen evolving complex (OEC) where water splitting occurs. The electron is transferred to the adjacent photosystem I (PSI), where it participates in the reduction reaction of NAD þ into NADH, which is ultimately used to fix CO 2 . Crucial in the above configuration is the separation of the functions of light collection and conversion from catalysis. Whereas light collection/conversion generates electron/ hole pairs one at a time, water splitting is a four-electron/hole process (2, 3). Hence, the multielectron catalysts of PSII and PSI, positioned at the terminus of the photosynthetic charge-separating network, are compulsory so that the one photon-one-electron/ hole "wireless current" can be bridged to the four-electron/hole chemistry of water splitting.An artificial photosynthesis can be designed if the one-electron/hole wireless current of a semiconductor can be integrated directly with catalysts to perform the four-electron-four proton catalysis of water splitting. To this end, an important recent advance has been the creation of a cobalt-phosphate (Co-Pi) catalyst (4, 5) that captures the functional elements of the OEC of PSII (6). As in PSII OEC, the Co-Pi catalyst self-assembles upon oxidation of an earth-abundant metal [Co 2þ for Co-Pi vs. Mn 2þ for OEC (7-9)] in phosphate-buffered solutions at neutral pH (4, 10), exhibits high activity in natural water and sea water at room temperature (11), activates water by proton-coupled electron transfer (3) [as does the OEC of PSII (12, 13)], and is self-healing (14) [as is PSII (15-18)]. Moreover, X-ray Absorption Spectroscopy (XAS) studies (19,20) have established that the Co-Pi catalyst is a structural relative of PSII OEC. PSII OEC is a Mn 3 CaO 4 -Mn cubane (21) where the fourth M...

Carrier (exciton) multiplication in colloidal InAs/CdSe/ZnSe core-shell quantum dots (QDs) is investigated using terahertz time-domain spectroscopy, time-resolved transient absorption, and quasi-continuous wave excitation spectroscopy. For excitation by high-energy photons (∼2.7 times the band gap energy), highly efficient carrier multiplication (CM) results in the appearance of multi-excitons, amounting to ∼1.6 excitons per absorbed photon. Multi-exciton recombination occurs within tens of picoseconds via Auger-type processes. Photodoping (i.e., photoinjection of an exciton) of the QDs prior to excitation results in a reduction of the CM efficiency to ∼1.3. This exciton-induced reduction of CM efficiency can be explained by the twofold degeneracy of the lowest conduction band energy level. We discuss the implications of our findings for the potential application of InAs QDs as light absorbers in solar cells.

No abstract

In our earlier paper, 1 we reported carrier multiplication (CM) in colloidal InAs quantum dots (QDs) and CM in dots with a pre-population of one exciton. The occurrence of CM in relaxed InAs QDs was concluded from the results of time-resolved TeraHertz (THz). Both THz and quasi-continuous wave (quasi-CW) experiments were performed to study the CM in preexcited dots. Recent attempts to reproduce the observations of CM using THz spectroscopy on InAs based QDs of two sizes were unsuccessful. These attempts followed transient absorption measurements in Jerusalem by one of us (S. R.) and co-workers reported elsewhere, 2 which have not yielded evidence for CM in these QDs.Time-resolved THz measurements were repeated for InAs/ CdSe/ZnSe core/shell-1/shell-2, synthesized as reported elsewhere. 3 The investigated QDs had an InAs core of 4.4 nm diameter (E g ) 1.1 eV) onto which one atomic layer of CdSe and four layers of ZnSe were deposited. As in our original paper, the presence or absence of CM was investigated by comparing two excitation wavelengths above and below the theoretical onset for CM (for InAs: 2.05 times E g ). 4 CM is characterized by the presence of relatively short-lived biexcitons (lifetime tens of picoseconds), 1,4,5 which are created by the absorption of one photon. However, biexcitons are also readily created by sequential multiphoton absorption. Hence, the relative yield of biversus single excitons has to be determined for fluences where multiexciton generation by multiphoton absorption is negligible. To disentangle effects of multiphoton excitation versus CM, we use various excitation fluences. Single excitons are longlived (∼100 ns) as compared to the time frame of our experiment.For the 4.4 nm particles, the scaled signals for excitation wavelengths of 400 and 800 nm, corresponding to 2.74 and 1.35 times the gap are shown in Figure 1 (offset for clarity).The data in Figure 1 correspond to 400 and 800 nm excitation fluences, which result in approximately the same average number of absorbed photons per particle (considering the optical density of the sample at the two wavelengths and the ratio of absorption cross sections σ at 400 and 800 nm, σ 400 nm/σ 800 nm ) 10.0). It is apparent from the data that there is no significant bi-exciton decay visible at low fluence. This points directly to the absence of CM. A conservative estimate, considering both the fluence dependence of the 400 nm signal, and a comparison of the signals at 400 and 800 nm at roughly the same excitation densities, provides an upper limit for CM of 10%, well below the factor of 1.6 concluded previously, 1 and also lower than the factor of 1.2 concluded in ref 6 under similar conditions. Summarizing, we could not reproduce our earlier results and the conclusions regarding the presence of CM. One or a combination of the following effects may explain these contradictory observations:•The QDs in previous and recent measurements were synthesized at different times. We cannot exclude the possibility that QDs from different synthesis ba...

Optimization of interfacial charge transfer in quantum dot (QD) sensitized mesoporous oxide films is crucial for the efficient design of QD sensitized solar cells (QDSSC). We employ TeraHertz time-domain spectroscopy (THz-TDS), transient absorption (TA) and time-resolved luminescence measurements, combined with transmission electron microscopy (TEM) and current-voltage measurements to study injection of electrons from PbSe QDs that are chemically linked to mesoporous oxide films. We illustrate that the interpretation of injection experiments is ambiguous for time-resolved optical measurements (TA and time-resolved luminescence) because nonradiative recombination processes at the oxide-QD interface have the same spectroscopic signature as electron injection. Complementary THz-TDS and current-voltage measurements demonstrate electron injection from the 1S e level of PbSe QDs into SnO 2 mesoporous films, but injection into TiO 2 films is not observed, although the time-resolved optical measurements could be interpreted as indicating injection. That injection takes place into SnO 2 but not into TiO 2 can be explained by the energy alignment of the 1S e level of the PbSe QD, which is energetically favorable and unfavorable respectively with respect to the oxide conduction band edges of SnO 2 and TiO 2 . THz-TDS experiments demonstrate that electron injection from 5.5 nm PbSe QDs into SnO 2 occurs on a time scale of 125 ( 40 ps. THz-TDS experiments further reveal a subensemble of photogenerated carriers in these QD-oxide systems that recombine within 10 ps after photoexcitation. These carriers are likely located in QD clusters formed on the oxide surface that are apparent from TEM images and where efficient nonradiative recombination processes occur. Control of QD cluster formation is therefore essential in the optimization of QDSSC devices because the time scale of carrier recombination in the QD clusters is shorter than the time scale of electron injection into the mesoporous oxide film.

We employ time-resolved terahertz (THz) spectroscopy (TRTS) to directly monitor the picosecond dynamics of electron transfer in dye-sensitized oxides in the presence of an electrolyte phase. Understanding the time scale on which electrons are injected from the dye into the oxide phase in the presence of electrolyte is important for optimization of the solar cell efficiency. We quantify injection dynamics from two different dyes into both mesoporous TiO2 and SnO2 films. Measurements are performed in inert media (air, acetonitrile), in the presence of two different electrolytes (the conventional iodine/iodide couple and the recently reported disulfide/thiolate redox couple), and in the presence of two different electrolyte additives (Li+ ions and tert-butyl pyridine). Electron injection dynamics in TiO2 is found to occur on two time scales: sub-150 fs and ∼10 ps, attributed to injection from the singlet and lower-lying triplet state, respectively. For SnO2, injection is slower, despite the lower energy of the band edge. The slow injection observed for SnO2 is attributed to the reduced density of electronic states in the material. We observe that for both oxides electron injection can be strongly retarded by changing the composition of the medium in which the sensitized oxide film is immersed. In particular, our results indicate that injection dynamics can be significantly slowed down in the presence of the disulfide/thiolate redox couple and/or tert-butyl pyridine.