ZnO Colloidal

Feb 2, 2012 - cally with no depletion region in the PbS layer such that only ... depletion region is found to lie within the PbS layer increasing char...
0 downloads 0 Views 1MB Size
Letter pubs.acs.org/NanoLett

The Transitional Heterojunction Behavior of PbS/ZnO Colloidal Quantum Dot Solar Cells Shawn M. Willis, Cheng Cheng, Hazel E. Assender, and Andrew A. R. Watt* Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, United Kingdom S Supporting Information *

ABSTRACT: The nature of charge separation at the heterojunction interface of solution processed lead sulphidezinc oxide colloidal quantum dot solar cells is investigated using impedance spectroscopy and external quantum efficiency measurements to examine the effect of varying the zinc oxide doping density. Without doping, the device behaves excitonically with no depletion region in the PbS layer such that only charge carriers generated within a diffusion length of the PbS/ ZnO interface have a good probability of being harvested. After the ZnO is photodoped such that the doping density is near or greater than that of the PbS, a significant portion of the depletion region is found to lie within the PbS layer increasing charge extraction (p−n operation). KEYWORDS: Colloidal quantum dot, photovoltaics, solar cell, nanocrystal doping, lead chalcogenide

T

report a transition from excitonic behavior to p−n behavior of the PbS/ZnO heterojunction following exposure of the cell to UV light. UV exposure has been shown to enhance the power conversion efficiency of devices by increasing the conductivity of the ZnO layer.15,16 ZnO is inherently n-type,17,18 but mobile electrons can become trapped when gas molecules such as O2, NO2, and CO are adsorbed on the ZnO surface.19 In order to free the electrons, oxygen desorption from the nanocrystal surface can be induced by UV illumination in an O2-free environment.20,21 Therefore before UV illumination, the ZnO nanocrystals can be considered an intrinsic, wide-bandgap semiconductor.17 After UV illumination, the ZnO doping density has been reported to reach up to ∼1019 cm−3.17 Using a combination of capacitance and quantum efficiency measurements, we investigate how the ZnO doping level modifies the interfacial charge separation mechanism. Figure 1 shows a comparison of the capacitance−voltage (C−V) and C−2 versus bias curves before and after UV exposure. Exposure to UV radiation substantially changes the capacitance profile from the nearly constant value seen before exposure, to showing an initial increase followed by a substantial drop in capacitance after exposure (Figure 1a). Mott−Schottky analysis is commonly used to determine a semiconductor’s doping density and the built-in bias at a metalsemiconductor junction.22 It can also be applied in a simple manner to semiconductor-semiconductor p−n junctions when one semiconductor is doped to a much greater extent than the other.23 For these cases, the capacitance is determined by the

he efficiency of solution processed lead chalcogenide colloidal quantum dot (CQD) solar cells has increased from less than 1 to over 5% in the last 4 years.1 They have proven to be air-stable2 and do not require high temperature processing,3 which are major drawbacks for competing thin film, organic, and dye sensitized technologies. To further increase efficiencies and develop a commercial technology there is a need to understand exactly what happens between photon absorption and charge extraction as current and most importantly how this is impacted by the choice of constituent materials. This paper probes the nature of charge separation at the heterojunction interface. The operating mechanism at CQD heterojunction interfaces is an area of considerable discussion. Solar cells fabricated from PbS and PbSe QDs have both been shown to form Schottky barriers that separate electron and hole charge carriers when in contact with a metal electrode,2,4−7 indicating there is no fundamental limitation for the formation of a depletion region within these materials. However, when forming a heterojunction with an n-type material there has been much debate as to whether the operating mechanism of the junction is depleted p−n8−10 or excitonic.11−13 This is important because thick PbS layers would improve absorption of photons with energy greater than the quantum confined bandgap, but thicker layers reduce the charge separation efficiency due to the extra distance they must travel to the heterojunction.3 In an excitonic junction, the charge harvesting ability is determined by the exciton diffusion length.14 However, a p−n junction creates a depletion region that extends out from the junction increasing the charge collection distance within the material. Minority carriers only need to diffuse to the depletion region in order to be swept into the other material due to the electric field in the depletion region, becoming majority carriers. In this paper, we © 2012 American Chemical Society

Received: December 8, 2011 Revised: January 30, 2012 Published: February 2, 2012 1522

dx.doi.org/10.1021/nl204323j | Nano Lett. 2012, 12, 1522−1526

Nano Letters

Letter

Figure 1. (a) C−V and (b) C−2 curves measured at room temperature before and after UV exposure.

Figure 2. External quantum efficiency spectra as a function of UV exposure. Initially, only the peak near 600 nm is evident. After sufficient doping, the peak near 450 nm becomes visible. The peak near 920 nm reflects the 1.35 eV bandgap of the PbS QDs.

width of the depletion region, which is bias dependent. Hence the depletion capacitance, C, is also bias dependent and can be expressed as4,24 1 C2

=

excited charge carriers have a good probability of being collected or they are absorbed in a region where there is little probability for charge collection.26 In the p−n model, the region of high charge collection probability consists of two pieces. First is the depletion region that sweeps minority carriers across the junction, separating the electron−hole pair. Outside the depletion region, up to a distance given by the exciton diffusion length, is an area where excitons have a good probability of diffusing to the depletion region and dissociating. Beyond this region, excitons will most likely recombine before they are separated. For the discussion here, this is termed the dead zone.25 In the excitonic model, excitons must diffuse to an interface to be separated. Therefore, the region of high charge collection probability in the excitonic model consists only of a region defined by the exciton diffusion length (no depletion region). Outside of this region is the dead zone. The regions are illustrated in Figure 3. Photon absorption also needs to be considered in conjunction with the two PbS regions. High-energy photons have shorter absorption lengths than lower-energy photons.10,27,28 This means short-wavelength photons are absorbed near the front (PEDOT) surface of the PbS layer while longerwavelength photons propagate further into the material before being absorbed. For a thin device (Figure 3a), short-wavelength photons create excitons in the extraction region (either the depletion region under the p−n model or within an exciton diffusion length of the junction for the excitonic model) while longer-wavelength photons propagate further and possibly through the PbS without being absorbed. In a thick device, short wavelength photons create excitons in the dead zone and longer wavelength photons create excitons in the extraction region. Hence there is a peak in the EQE spectrum: photons of lower energy than the peak are not absorbed, and photons of higher energy than the peak are absorbed too far from the interface for their charges to be extracted. As the PbS layer is made thicker, the longer wavelength photons have a better chance of being absorbed, but absorption of short wavelength photons moves into the dead zone. If the PbS layer is made thicker, the peak in the EQE spectrum shifts to longer wavelengths.6 The Supporting Information demonstrates this effect in devices of different PbS-layer thicknesses. Under the p−n model, a depletion region should form at the PbS/ZnO junction. As discussed, UV exposure transitions the ZnO from an intrinsic semiconductor (or nearly intrinsic) to an

2(Vbi − V ) A2qεε0Na

(1)

where Vbi is the built-in bias, V is the applied bias, A is the device area, q is the elementary charge, ε is the material’s dielectric constant, ε0 is the permittivity of free space, and Na is the doping density of the lower doped material. Therefore a bias-dependent capacitance that follows eq 1 demonstrates the presence of a p−n junction. The built-in bias and doping density are then found by fitting eq 1 to the linear portion of the C−2 versus bias voltage plot.22 After UV exposure, a linear region in the C−2 curve exists which can be fitted with eq 1, demonstrating the change to p−n behavior. The fit yields Vbi = 0.6 V and Na = 2.5 × 1017 cm−3. The Vbi is considerably higher than expected from reported Fermi level differences.8,10 Likewise, Na is higher than expected based on Mott−Schottky analysis of PbS CQD/Al Schottky cells we fabricated, which gave Na values near 4 × 1016 cm−3 (included in the Supporting Information). The high values for the two parameters can be explained by barriers to charge injection at the electrode interfaces that creates a constant capacitance that must be taken into account with the depletion capacitance.25 The same method was attempted on the capacitance data taken before UV exposure. In this case the expected Mott−Schottky parameters could not be attained and the curve became distorted (see Supporting Information). Further evidence for the transition from excitonic to p−n behavior with UV exposure is seen in the evolution of the external quantum efficiency (EQE) spectrum during UV exposure shown in Figure 2. Before any UV exposure and at low exposure times, only the peak near 600 nm is evident. With increased UV exposure the shorter wavelength response is seen to increase, yielding a significant EQE response down to ∼450 nm. This behavior can be explained by the growth of a depletion region within the PbS material during UV illumination. The peak near 920 nm reflects the bandgap of the PbS QDs (1.35 eV), associated with the peak in absorptivity at this wavelength. The overall increase in EQE response at all wavelengths is due to the increased conductivity of the ZnO layer due to UV exposure. In a general sense, it is possible for two regions to exist in the PbS layer. Photons are either absorbed in a region where the 1523

dx.doi.org/10.1021/nl204323j | Nano Lett. 2012, 12, 1522−1526

Nano Letters

Letter

Using eq 2, Figure 4 shows the width of the depletion region extending into the PbS layer for ZnO doping levels between

Figure 4. Depletion width thickness in the PbS layer as a function of ZnO doping density for three different built-in bias values.

1013 and 1019 cm−3. The PbS doping density used in the calculations was 4 × 1016 cm−3 based on Mott−Schottky analysis of the PbS/Al Schottky devices we fabricated. The growing depletion region within the PbS layer causes the EQE spectrum to initially show a peak at longer wavelengths. Then with increased UV exposure, the response at shorter wavelengths increases as seen in Figure 2. The longer wavelength peak position does not shift though, since the PbS layer thickness does not change and the longer wavelength photons are still absorbed. Therefore initially, with basically no depletion region in the PbS layer (intrinsic ZnO), only charge carriers generated within a diffusion length of the PbS/ZnO interface have a good probability of being harvested. The junction behavior is then initially excitonic. After the ZnO is photodoped such that the doping density is near or greater than that of the PbS, a significant portion of the depletion region lies within the PbS layer which increases the charge extraction region within the PbS layer (p−n operation). The results suggest an initial excitonic junction behavior before UV exposure that transitions to a p−n junction mechanism as the ZnO layer becomes doped. C−V measurements on the PbS/ZnO heterojunction cells taken before UV exposure display a near constant capacitance in reverse bias showing no evidence for a bias-dependent depletion region, indicating excitonic junction behavior. In other words, excitons created within a diffusion length of the heterojunction have a high probability of diffusing to the junction and dissociating without the presence of a depletion region. Exposing the cell to UV radiation changes the behavior and evidence of a depletion region becomes visible. This is confirmed by EQE data where the evolution of the response at short wavelengths during UV exposure indicates a depletion region within the PbS layer is formed. Therefore before UV exposure, the junction initially lacks a depletion region requiring electron−hole pairs to be created within a diffusion length of the junction to harvest the charge carriers. After UV exposure the junction behaves as a depleted p−n junction enhancing charge collection. The underlying physics of the system is complex and there are a number of other variables that influence the nature of the heterojunction for example nanocrystal capping molecule, surface trap states and nanocrystal size which warrant further detailed examination.

Figure 3. (a) Purely excitonic junction. Short wavelength photons are absorbed in the dead zone, medium wavelength photons are absorbed in the exciton diffusion region, and long wavelength photons pass though without being absorbed. (b) Depleted p−n junction. Short and medium wavelength photons are absorbed in either the depletion region or within the exciton diffusion length from the depletion region. Long wavelength photons pass through without being absorbed. (c) Depleted p−n junction with a thick PbS layer. Short wavelength photons are absorbed in the dead zone. Medium and long wavelength photons are absorbed in either the depletion region or within the exciton diffusion length from the depletion region.

n-type semiconductor.17 Since the ZnO nanocrystals were synthesized in air, they are expected to be very nearly intrinsic due to oxygen adsorption. This means the depletion region transitions from being located mainly in the ZnO layer when the ZnO doping density is much lower than the PbS doping density to being located mainly in the PbS when the ZnO doping density is much higher than the PbS doping density.26 Using the discrete boundary approximation, the depletion region, wp, extending into the PbS layer is given by26 wp =

1 Na

2εε0Vbi

⎛1 1 ⎞ q⎜ + ⎟ Nd ⎠ ⎝ Na

(2)

where Nd is the ZnO doping level. The depletion region in the PbS layer is of interest since the ZnO layer is not expected to contribute much to photon absorption due to its large bandgap. 1524

dx.doi.org/10.1021/nl204323j | Nano Lett. 2012, 12, 1522−1526

Nano Letters

Letter

Methods. Materials. Oleic acid (technical grade 90%), 1octadecene (technical grade 90%), hexamethyldisilathiane (technical grade 90%), butylamine (99.5%), lead oxide yellow (PbO, ≥ 99%), and 1,2-ethanedithiol (≥98%) were purchased from Sigma-Aldrich. HPLC grade hexane, methanol, and isopropanol were purchased from Fisher Scientific. Ethyl acetate was purchased from Romil. Zinc acetate dihydrate (technical grade) was purchased from Riedel-de Haen. Potassium hydroxide (reagent grade) was purchased from Fisher Scientific. Poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT/PSS) was purchased from H.C. Stark. Synthesis of PbS Nanocrystals. Synthesis of 1.3−1.4 eV bandgap PbS nanocrystals was performed based on the method described by Luther et al.29 A mixture of 0.47 g of PbO, 10 mL of 1-octadecene, and 2 mL of oleic acid was mixed and heated in a flask to 120 °C under N2 for 2 h. During this process the PbO dissolved and produced a clear solution. In a separate vial, 180 μL of hexametyldisilathiane (TMS2S) was mixed with 3 mL of 1-octadecene and degassed under N2 flow. The TMS2S solution was loaded into a syringe and swiftly injected into the PbO solution at 100 °C. The resulting mixture turned from clear to dark, indicating rapid nucleation. The nanocrystals were allowed to grow for 1 min at which point the growth was halted by quenching the flask in an ice water bath. The solution temperature dropped to room temperature in less than 1 min. The solution was then transferred to a 50 mL centrifuge tube and 35 mL of 1:1 ethyl aceate/methanol was added to precipitate the nanocrystals. The solids were separated by centrifuging at 4000 rpm for 5 min. The supernatant was decanted and 5 mL of hexane was used to redisperse the nanocrystals. The nanocrystals were washed with methane twice, and then redispersed in 5 mL of hexane. Finally, the nanocrystals (in hexane) were driven through a 0.45 μm PTFE filter. Butylamine Ligand Exchange. Butylamine ligand exchange was performed by adding 5 mL of butylamine to dissolve the precipitated nanocrystals. This process was assisted by ultrasonication for 30 min. The nanocrystals were precipitated again by adding isopropanol and redispersed in 5 mL of butylamine. After another 30 min of ultrasonication, the nanocrystals were transferred to a N2-filled drybox. The next day the nanocrystals were precipitated by isopropanol, centrifuged, and dried under N2 flow. The dry weight was then calculated. Hexane was used to disperse the nanocrystals to 25 mg/mL for device fabrication. The solution was stored in a nitrogen drybox. ZnO Nanoparticle Synthesis. ZnO nanoparticles were synthesized using the method reported by Pacholski et al.30 A zinc precursor was prepared by dissolving 0.002 mol zinc acetate dihydrate in 100 mL of methanol. Potassium hydroxide in 50 mL of methanol was added by drops to the zinc precursor at 60 °C while stirring. The solution turned cloudy at first and then became clear. After 2 h, the ZnO nanoparticles started to precipitate from the solution. The precipitates were collected by centrifugation, and redispersed in methanol. The nanoparticles were washed with methanol 3 times, dried, and then redispersed in chloroform to 40 mg/mL. The nanoparticle solution was then transferred to a N2-filled drybox. Device Fabrication. Prepatterned ITO substrates were cleaned by the sequence of 10 min ultrasonication in deionized water containing Decon, 10 min ultrasonication in deionized water, 5 min in 50 °C acetone, and 5 min in 50 °C isopropanol.

The substrates were then treated with plasma for 1 min. PEDOT/PSS was spin coated on the substrate at 5000 rpm for 30 s. The substrate was then annealed in air for 5 min at 150 °C. Layer-by-layer spin coating was used to fabricate crack-free PbS nanocrystal thin films.12 During each iteration, 6 drops of PbS were spin coated onto the substrate at 2000 rpm followed by substrate immersion into a 0.01 vol % 1,2-ethanedithiol (EDT) in acetonitrile solution. The residual EDT solution was spun off at 2000 rpm for 1 min. While spinning, 10 drops of methanol and 10 drops of hexane were dispensed onto the substrate to wash off the residual EDT and oleic acid. The spin coating iteration was repeated to reach thicknesses of 50, 125, 190, and 230 nm as measured by DEKTAK. The PbS thickness of the cell used in this paper was 50 nm for C−V measurements and 190 nm for EQE measurements. Next, a 100 nm thick ZnO layer was spin coated on top of the PbS. Al electrodes were deposited by thermal evaporation in an Edwards E04 evaporator at 10−6 Torr to a thickness of 100 nm. The evaporation rate was 0.1 nm/s. A shadow mask was used to pattern the electrodes. Device Characterization. Each cell was mounted in a purpose-built chamber that was continually flushed with N2. C−V measurements were performed with an Agilent E4980A opt001 LCR meter. C−V sweeps were performed at biases from −2 to 2 V at 0.02 V intervals. Each measurement was taken in the R-X mode and the real portion of the complex capacitance was calculated. The AC signal was set to 25 mV and 500 Hz. All C−V measurements were performed in the dark. The purposebuilt chamber provided electrical shielding. Illumination for the EQE measurements was provided by a halogen lamp and Oriel Conerstone 130 monochromator. Light intensity at different wavelengths was calibrated using a Newport 818 UV enhanced silicon photodetector and a Newport 918 IR germanium photodetector. The current signal was measured with a Keithley 6845 picoammeter. After 10 min of UV exposure no further improvement was seen in the EQE, C−V, and IV characteristics. After 30 min the device had stabilized and slightly degraded, possibly due to the presence of oxygen in the testing chamber.



ASSOCIATED CONTENT

S Supporting Information *

Mott−Schottky analysis of PbS/Al Schottky junction; injection barrier capacitance analysis of non UV-doped PbS/ZnO heterojunction device; and EQE measurements of PbS/ZnO solar cells with multiple PbS layer thicknesses. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



REFERENCES

(1) Tang, J.; Sargent, H. Infrared Colloidal Quantum Dots for Photovoltaics: Fundamentals and Recent Progress. Adv. Mater. 2011, 23, 12−29. (2) Tang, J.; Wang, X.; Brzozowski, L; Barkhouse, D. A. R.; Debnath, R; Levina, L.; Sargent, E. H. Schottky Quantum Dot Solar Cells Stable in Air under Solar Illumination. Adv. Mater. 2010, 22, 1398−1402. (3) Sargent, E. H. Infrared Photovoltaics Made by Solution Processing. Nat. Photonics. 2009, 3, 325−331.

1525

dx.doi.org/10.1021/nl204323j | Nano Lett. 2012, 12, 1522−1526

Nano Letters

Letter

(4) Clifford, J. P.; Johnston, K. W.; Levina, L.; Sargent, E. H. Schottky Barriers to Colloidal Quantum Dot Films. Appl. Phys. Lett. 2007, 91, 253117. (5) Johnston, K. W.; Pattantyus-Abraham, A. G.; Clifford, J. P.; Myrskog, S. H.; Hoogland, S.; Shukla, H.; Klem, E. J. D.; Sargent, E. H. Efficient Schottky-quantum-dot photovoltaics: The Roles of Depletion, Drift, and Diffusion. Appl. Phys. Lett. 2008, 92, 122111. (6) Tang, J.; Brzozowski, L.; Barkhouse, D. A. R.; Wang, X.; Debnath, R.; Wolowiec, R.; Palmiano, E.; Levina, L.; Pattantyus-Abraham, A. G.; Jamakosmanovic, D.; Sargent, E. H. Quantum Dot Photovoltaics in the Extreme Quantum Confinement Regime: The Surface-Chemical Origins of Exceptional Air- and Light-Stability. ACS Nano 2010, 4, 869−878. (7) Luther, J. M.; Law, M.; Beard, M. C.; Song, Q.; Reese, M. O.; Ellingson, R. J.; Nozik, A. J. Schottky Solar Cells Based on Colloidal Nanocrystal Films. Nano Lett. 2008, 10, 3488−3492. (8) Pattantyus-Abraham, A. G; Kramer, I. J.; Barkhouse, A. R.; Wang, X.; Konstantatos, G.; Debnath, R.; Levina, L.; Raabe, I.; Nazeeruddin, M.; Gratzel, M.; Sargent, E. H. Depleted-Heterojunction Colloidal Quantum Dot Solar Cells. ACS Nano 2010, 4, 3374−3380. (9) Barkhouse, D. A. R.; Debnath, R.; Kramer, I. J.; Zhitomirsky, D.; Pattantyus-Abraham, A. G.; Levina, L.; Etgar, L.; Gratzel, M.; Sargent, E. H. Depleted Bulk Heterojunction Colloidal Quantum Dot Photovoltaics. Adv. Mater. 2011, 23, 3134−3138. (10) Gao, J.; Luther, J. M.; Semonin, O. E.; Ellingson, R. J.; Nozik, A. J.; Beard, M. C. Quantum Dot Size Dependent J-V Characteristics in Heterojunction ZnO/PbS Quantum Dot Solar Cells. Nano Lett. 2011, 11, 1002−1008. (11) Leschkies, K. S.; Beatty, T. J.; Kang, M. S.; Norris, D. J.; Aydill, E. S. Solar Cells Based on Junctions between Colloidal PbSe Nanocrystals and Thin ZnO Films. ACS Nano 2009, 11, 3683−3648. (12) Choi, J. J.; Lim, Y.; Santiago-Berrios, M. B; Oh, M.; Hyun, B.; Sun, L.; Bartnik, A. C.; Goedhart, A.; Malliaras, G. G.; Abruna, H. D.; Wise, F. W.; Hanrath, T. PbSe Nanocrystal Excitonic Solar Cells. Nano Lett. 2009, 11, 3749−3755. (13) Choi, J. J.; Luria, J.; Hyun, B.; Bartnik, A. C.; Sun, L.; Lim, Y.; Marohn, J. A.; Wise, F. W.; Hanrath, T. Photogenerated Exciton Dissociation in Highly Coupled Lead Salt Nanocrystal Assemblies. Nano Lett. 2010, 10, 1805−1811. (14) Gregg, B. A. Excitonic Solar Cells. J. Phys. Chem. B. 2003, 107, 4688−4698. (15) Verbakel, F.; Meskers, S. C. J.; Janssen, R. A. J. Electronic Memory Effects in Diodes from a Zinc Oxide NanoparticlePolystyrene Hybrid Material. Appl. Phys. Lett. 2006, 89, 102103. (16) Gilot, J.; Wienk, M. M.; Janssen, R. A. J. Double and Triple Junction Polymer Solar Cells Processed from Solution. Appl. Phys. Lett. 2007, 90, 143512. (17) Lakhwani, G.; Roijmans, R. F. H.; Kronemeijer, A. J.; Gilot, J.; Janssen, R. A. J.; Meskers, S. C. J. Probing Charge Carrier density in a Layer of Photodoped ZnO Nanoparticles by Spectroscopic Ellipsometry. J. Phys. Chem. C. 2010, 114, 14804−14810. (18) Shim, M.; Guyot-Sionnest, P. Organic-Capped ZnO Nanocrystals: Synthesis and n-Type Character. J. Am. Chem. Soc. 2001, 123, 11651−11654. (19) Collins, R. J.; Thomas, D. G Photoconduction and Surface Effects with Zinc Oxide Crystals. Phys. Rev. 1958, 112, 388−395. (20) Verbakel, F.; Meskers, S. C. J.; Janssen, R. A. J. Electronic Memory Effects in Diodes of Zinc Oxide Nanoparticles in a Matrix of Polystyrene or Poly(3-hexythiophene). J. Appl. Phys. 2007, 102, 083701. (21) Liu, W. K.; Whitaker, K. M; Kittilstved, K. R.; Gamelin, D. R. Stable Photogenerated Carriers in magnetic Semiconductor Nanocrystals. J. Am. Chem. Soc. 2006, 128, 3910−3911. (22) Schroder, D. K. Semiconductor Material and Device Characterization, 3rd ed.; Wiley-IEEE Press: New York, 2006. (23) Mora-Sero, I.; Garcia-Belmonte, G.; Boix, P. P.; Vazquez, M. A; Bisquert, J. Impedance Spectroscopy Characterisation of Highly Efficient Silicon Solar Cells under Different Light Illumination Intensities. Energy Environ. Sci. 2009, 2, 678−686.

(24) Garcia-Belmonte, G.; Munar, A.; Barea, E. M.; Bisquert, J.; Ugarte, I.; Pacios, R. Charge Carrier Mobility and Lifetime of Organic Bulk Heterojunctions Analyzed by Impedance Spectroscopy. Org. Electron. 2008, 9, 847−851. (25) Willis, S. M.; Cheng, C.; Assender, H. E.; Watt, A. A. R. Modified Mott-Schottky Analysis of Nanocrystal Solar Cells, http://arxiv.org/abs/ 1112.1623. Accessed December 7, 2011. (26) Barkhouse, D. A. R.; Kramer, I. J.; Wang, X.; Sargent, E. H. Dead Zones in Colloidal Quantum Dot Photovoltaics: Evidence and Implications. Opt. Express. 2010, 18, A451−A457. (27) Nelson, J. The Physics of Solar Cells; Imperial College Press: London, 2003. (28) Law, M.; Beard, M. C.; Choi, S.; Luther, J. M.; Hanna, M. C.; Nozik, A. J. Determining the Internal Quantum Efficiency of PbSe Nanocrystal Solar Cells with the Aid of an Optical Model. Nano Lett. 2008, 8, 3904−3910. (29) Luther, J. M.; Gao, J.; Lloyd, M. T.; Semonin, O. E.; Beard, M. C.; Nozik, A. J. Stability Assessment on a 3% Bilayer PbS/ZnO Quantum Dot Heterojunction Solar Cell. Adv. Mater. 2010, 22, 3704− 3707. (30) Pacholski, C.; Kornowski, A.; Weller, H. Self-Assembly of ZnO: From Nanodots to Nanorods. Angew. Chem., Int. Ed. 2002, 47, 1188− 1191.

1526

dx.doi.org/10.1021/nl204323j | Nano Lett. 2012, 12, 1522−1526