Materials Science and Engineering
Julian Minuzzo

A Self Assembled Organic/Inorganic Nanostructure for Photovoltaic Applications

  • Faculty Advisor

    Samuel I. Stupp and David Herman

Published On

May 2014

Originally Published

NURJ 2013-14
Honors Thesis

Abstract

Low-cost, scalable photovoltaics are of particular importance because they may allow for the widespread implementation of solar energy. Herein, a low-cost self assembly process is used to fabricate ordered heterojunction solar cells in the form of a lamellar structure of alternating organic/inorganic domains. The domains grow as high-aspect ratio wires on a transparent PEDOT:PSS-coated indium tin oxide (ITO) substrate. The lamellar structure is oriented such that domains of a small, light-absorbing organic molecule and inorganic zinc oxide provide perpendicularly aligned pathways for holes and electrons to reach their respective electrodes after exciton splitting occurs at the interface between the two materials. SEM, TEM, and small angle X-ray scattering are used to probe film morphology, structure of the active layer, and perpendicular orientation of the high aspect ratio wires, respectively. The first ever lamellar organic-inorganic solar cell was fabricated; the best device obtained an overall conversion efficiency of 0.04% along with an open circuit voltage of 0.53 V and a short circuit current of 0.25mA/cm^2.

Introduction

The world consumes the equivalent of 88 billion barrels of oil each year.1 With energy usage only projected to increase, talk of an energy crisis has become commonplace. Greenhouse gas-emitting fossil fuels currently comprise the vast majority of energy sources, and while they are abundant, they are not unlimited and will eventually be depleted. Border wars and national security concerns that are closely tied to the control of fossil fuels pose another complication. Because of the inherent problems associated with fossil fuels and crude oil, exploring new methods of energy production must be a strong focus for researchers in the years to come. While currently still in their infancy, renewable energy sources like solar and wind energy have the potential to provide enough energy to meet the world’s current and future energy needs.

Solar energy is the world’s fastest growing renewable energy technology at an annual growth rate of 30 percent.2 Nonetheless, renewable energy (including solar, wind, and geothermal), currently comprises less than one percent of the worldwide energy supply.3 Although renewable energy production is not a large portion of the world’s overall current energy supply, it may have the capability to avert a worldwide energy crisis. Solar energy is a renewable energy that has enormous potential. Figure 1 manifests this potential; if solar cells with an efficiency of just five percent were placed in the areas represented by black dots, they would produce enough energy to power the world.4

Figure 1. Diagram of area required by five-percent efficient solar cells (represented by six black dots) to power the planet.4

Solar photovoltaics—which convert the sun’s photons directly to electricity—come in many forms. Commercially available cells are usually made of silicon and achieve moderate efficiencies of 10-20 percent.5 However, the high purity of silicon required to achieve these efficiencies, the high energy processes used to make silicon cells (including expensive vapour deposition techniques), and their inherent bulkiness are burdensome. Other types of solar photovoltaics include multi-junction, dye sensitized, organic, and inorganic/organic hybrid solar cells. Each of these technologies uses different materials to achieve the same goal: electricity from sunlight. However, the chief characteristic that will determine which solar technology becomes the most economically feasible is the total cost per watt of electricity generated. This cost includes raw materials, processing, and maintenance costs. Thus, simple, low-cost processes that create moderate to high performing solar cells may be able to provide the low cost solar power that we desperately need.

Photovoltaic Electricity Generation

One method to improve performance without incurring higher materials or processing costs is to improve solar device geometry at the nano-level. Excitons are electron-hole pairs that are created when a photon strikes the active, light absorbing layer in a solar cell. Upon formation, excitons must diffuse to the interface between the electron acceptor and electron donor material of the cell before they can be split into electrons and holes and harvested for energy at the electrodes of the cell. Because excitons can only diffuse between 5-20 nm before they recombine and their energy is lost, solar geometries that have the electron donating and electron accepting material apart by this length scale will reach the highest efficiency.6

Types of Solar Cell Geometries

Three types of solar geometries, along with explanations of each, are shown in Figure 2. As discussed in Figure 2, the most desirable geometry is that of the ordered heterojunction (Figure 2, bottom image). The ordered heterojunction avoids both the low surface of the planar architecture as well as the charge entrapment of the bulk heterojunction. Because ordered heterojunctions can hypothetically be tuned such that no photoactive section of the cell is outside of the 20 nm exciton diffusion length, more excitons are harvested and more energy is produced. Furthermore, if the ordered regimes of a heterojunction are arranged perpendicular to the top and bottom electrodes, holes and electrons will reach their respective electrodes as fast as possible due to the straight path that is available—lowering the chances of charge recombination considerably. Previous examples of ordered heterojunctions include nanowire and nanotube arrays,7,8 nano-porous films filled with absorbing material,9,10 and block copolymer assemblies,11 among others.

Figure 2. Three different solar cell geometries. While the planar geometry (top) is the most common, it suffers from losses in efficiency because excitons can be created too far away from the acceptor/donor interface to be harvested for energy and it also has a low interfacial area between the donor and acceptor. While the bulk heterojunction geometry (middle) has a higher interfacial area between the donor and acceptor, it suffers from charge entrapment in “islands” of donor or acceptor material that are not in contact with either the cathode or anode—leading to a loss in efficiency. The ordered heterojunction (bottom) maintains a high interfacial surface area while avoiding charge entrapment—leading to the highest theoretical efficiency.

Self Assembly to Achieve an Ordered Heterojunction

Self assembly is defined as the organization of the components in a disordered system into a hierarchical pattern or structure via inter-component interactions and without external interference. A schematic diagram of a self assembled process is shown in Figure 3. Self assembly can take place on any length scale, from atoms to molecules to full-size robots. Examples of self assembly in nature include phospholipid bilayers, phase separation of materials, and DNA transcription. It is easy to see how self assembly on all length scales can lead to new shapes and functionalities that were not initially present in the disordered system. Because they involve no external interference, self assembled processes are also highly scalable. On the molecular level, self assembly occurs via intermolecular forces. This type of self assembly, known as supramolecular self assembly, is facilitated by noncovalent intermolecular interactions including hydrogen bonds, pi-pi stacking, and dipole interactions. Through these forces, hierarchical structures with novel functionalities can be formed spontaneously. Thus, if donor and acceptor materials can be arranged on the nano-level via self-assembly, the benefits of an ordered structure can be attained without a complicated multi-step procedure.

Figure 3. Schematic diagram of a self assembly process. The disordered system on the left achieves an order structure without any external interference due to the interactions between the red spheres in the middle of each structural unit.

Layered Double Hydroxides

Layered double hydroxides, or LDHs, provide a promising structure with which to achieve the bulk heterojunction nanostructure. LDHs, schematically represented in Figure 4, are metal hydroxides that form large, positively charged two-dimensional sheets approximately 5-20 nm apart over large length scales.12 Because of their positive charge, LDHs can store many different types of negatively charged compounds between sheets, including atoms, molecules, and even larger compounds such as DNA.13 With regards to photovoltaic technology, layered double hydroxides can template the growth of light-absorbing negatively charged small molecules between positively charged LDH sheets. Once light-absorbing materials are incorporated into the voids between LDH sheets, however, they must be able to donate photo-excited electrons to a good electronic conductor. While metal hydroxides are poor conductors of electrons, metal oxides are often excellent electron conductors. Because it can readily form the LDH structure and be converted to a metal oxide by simply annealing at 150° C, Zn(OH)2 is a good electron-acceptor material for use in bulk heterojunction photovoltaic devices.14

Figure 4. Layered Double Hydroxide (LDH) structure. Imperfections in the metal hydroxide structure lead to a positive charge on the surface of the LDH sheets, allowing negatively charged anions (indicated by X- in the figure) to fill the area between sheets.

Small Organic Molecules for Light Harvesting

The next step in creating a nanoscale heterojunction is to incorporate a light-absorbing electron donating material between the electron accepting ZnO LDH sheets. Because of their relatively low cost of production, exciton diffusion length of 5-40 nm, and high absorption coefficients, organic molecules are a good candidate for an embedded light harvesting component between Zn(OH)2 LDH layers.15 Hybrid (i.e. inorganic and organic) composite films composed of small organic molecules and inorganic Zn(OH)2 have been previously achieved via electrochemical self assembly. These structures are formed by utilizing negatively charged side-groups on the organic molecules to insert them between positively charged Zn(OH)2 sheets, but no working solar devices have yet been achieved with the hybrid films.16 One of the challenges in forming a solar device from a ZnO/small organic molecule hybrid lamellar structure is to ensure that the small organic molecule has good optical properties and can absorb well in the visible regime. As a general rule, the larger the organic molecule, the better its absorption properties due to delocalization of electrons (and thus a smaller bandgap). Conversely, the larger the organic molecule, the more difficult it is to insert in the LDH structure of the ZnO electron acceptor. Thus, the goal is to introduce the organic molecule with the most delocalization of electrons possible into the Zn(OH)2 LDH voids while ensuring that the molecule is mobile enough to pack densely within the voids. The organic electron donating material chosen for this work is dicarboxylic acid 3- methylquinquethiophene (5TmDCA), shown in Figure 5. 5TmDCA absorbs light in the near-UV range, with an absorption peak around 380 nm, which is where a large portion of sun’s emitted power lies in the electromagnetic spectrum.

Figure 5. Structure of dicarboxylic acid 3- methylquinquethiophene (5TmDCA). 5TmDCA is the organic molecule used as the light absorber and electron donating material in the ordered heterojunction solar cells reported in this work.

Perpendicular Orientation for Efficient Charge Transfer

In order to fully reap the benefits of the nanoscale lamellar structure of inorganic ZnO and organic light-harvesting small molecules, the alternating layers must be oriented with their faces perpendicular to the electrodes of the solar cell instead of parallel or randomly oriented (see Figure 6). This orientation allows for the straightest and quickest path for charge carriers to reach their respective electrodes, thus lowering the probability of charge recombination. Previous experiments demonstrated that electrodeposited Zn(OH)2 and organic molecules assemble themselves on an indium tin oxide (ITO) substrate in a random manner, without perpendicular orientation.17 However, by coating the ITO surface (which acts as the bottom electrode in a solar cell) with poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS, a commonly used polymer in organic photovoltaics), the surface energy is altered and perpendicular orientation of the lamellar structure is more readily obtained as a result.18 Along with increasing overall efficiency due to the previously mentioned charge separation properties, a perpendicularly-oriented lamellar structure can also lead to an increased film smoothness. A smooth film is important when applying the top electrode of a solar cell, because if the top of the film is very rough the top electrode will not contact the top of every active material domain, leading to a drop in efficiency. Through the implementation of small organic molecules between Zn(OH)2¬ LDH sheets via electrodeposition on PEDOT:PSS modified ITO, a relatively perpendicular lamellar array can form that has the ideal geometry for solar photovoltaic devices.

Figure 6. Schematic diagram of hybrid lamellar domains (each domain containing a layer of light absorbing small organic molecules [white] between inorganic electron accepting ZnO [blue]) oriented randomly (a) and perpendicular to the electrode surface, facilitating the best charge transport (b). In reality, these domains are much more densely packed to form a 500-1000 nm thick film atop the substrate (shown in grey). Image, courtesy of Dr. Dave Herman.

Figure 7. Experimental set up used to facilitate the self-assembly of 5TmDCA and Zn(OH)2. Zn(OH)2 is converted to ZnO upon annealing at 150°C.

Results and Discussion

The first ever working solar cell made of a lamellar ordered heterojunction containing 5TmDCA and ZnO was created by the following self-assembly process. Shown in Figure 7, the self-assembly of 5TmDCA and ZnO takes place through a simple electrodeposition experiment. Initially, a solution containing 5TmDCA molecules, Zn2+ ions, NO3- ions, water, and dimethyl sulfoxide (DMSO, an organic solvent) is heated to 80°C. Addition of a base to the solution then deprotonates the H atoms from either side of the 5TmDCA molecules, leaving the molecules with a negative charge. A negative voltage bias is then applied to the poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS) coated indium tin oxide (ITO) working electrode relative to the zinc counter electrode. The negative bias reduces NO3- ions in solution to NO2- ions, producing base (OH-) in the process (Equation 1).

〖〖NO〗_3〗^-+H_2 O+2e^-→〖〖NO〗_2〗^-+2〖OH〗^- (1)

The basic OH- ions then combine with the Zn2+ ions in solution to form Zn(OH)2. The solid Zn(OH)2 that forms on the surface of the working electrode grows as positively charged sheets (the positive charge arises due to vacancies and charge imbalance during growth) with their normal vectors parallel to the ITO substrate, shown in Figure 8. Because the sheets are positively charged, the negatively charged deprotonated 5TmDCA molecules self-assemble themselves such that their negative ends link two adjacent Zn(OH)2 sheets, forming a lamellar structure. Films are grown for two hours and reach a final thickness of 500-600 nm. Finally, Zn(OH)2 is converted to ZnO (a much better electronic conductor) upon annealing films at 150°C for 12 hours.

Figure 8. Schematic diagram positively charged of Zn(OH)2 sheets that form during the application of a negative bias to the ITO electrode. Negatively charged 5TmDCA molecules self-assemble inside of the Zn(OH)2 sheets such that their negatively charged ends link to two adjacent Zn(OH)2 sheets.

Scanning electron microscopy (SEM) was used to determine general film morphology, transmission electron microscopy (TEM) was used to observe the nanoscale lamellar structure of 5TmDCA and ZnO, and grazing incidence small angle X-ray scattering (GISAXS) was used to determine the angle that the lamellar structure formed in relation to the ITO substrate. Shown in Figure 9, GISAXS measures the angle of reflection of a very small angle incoming X-ray beam to determine which orientation the lamellar structure possesses. SEM and TEM images of films are shown in Figure 10. SEM, TEM, and GISAXS revealed that films grown via the electrodeposition method previously described are composed of perpendicularly aligned (normal parallel to the ITO substrate) nanoscale “wires.” Inside of each wire is the lamellar, ordered heterojunction structure that is highly desirable for solar cells.

Figure 9. Schematic diagram showing how GISAXS provides information about the orientation of films composed of a lamellar structure. Isotropic orientations (a) produce rings on the X-ray film, parallel orientations (b, undesirable for solar cells) produce images along the y-axis of the X-ray film, and perpendicularly aligned (c, desirable for solar cell) films show images along the x-axis of the film.

Figure 10. SEM (left) and TEM (inset, right) images of electrodeposition-grown films of 5TmDCA and ZnO (after annealing). The SEM image shows that films are composed of nanometer-scale “wires,” while TEM images reveal that each wire is composed of the desired lamellar structure. In the TEM image of a single wire, the organic 5TmDCA layers are lighter while inorganic ZnO layers are darker.

Solar cells were fabricated from the ordered heterojunction films by first depositing a silver electrode on top of the active 5TmDCA/ZnO active layer. Cells were then illuminated with a 1.5 air-mass solar simulator. Measurement of the current-voltage characteristics of fabricated devices upon illumination verified the first ever working organic/inorganic ordered heterojunction solar cell with a nanoscale lamellar structure. The best solar cell achieved an efficiency of 0.040 percent, an open circuit voltage (VOC) of 0.53V, and a short-circuit current (JSC) of 0.25mA.

Conclusion

By subsequently optimizing electrodeposition growth conditions and device fabrication, the first ordered organic/inorganic grown using one-step electrodeposition was fabricated. This result is extremely exciting and should spur much more work in the field of organic/inorganic ordered heterojunction solar cells. Future work will include further optimizing of growth conditions and introducing a new organic molecule with a peak absorption more in the visible light range into the lamellar structure.

ABOUT THE AUTHOR

Julian Minuzzo

BIBLIOGRAPHY

1. Lior, N. The ECOS 2009 World Energy Panel: An introduction to the Panel and to the present (2009) situation in sustainable energy development. Energy 36, 3620–3628 (2011).
2. Lewis, N. Basic Research Needs for Solar Energy Utilization. (U.S. Department of Energy). at <http://www.sc.doe.gov/bes/reports/files/SEU_rpt.pdf>
3. Key World Energy Statistics 2010. (International Energy Agency, 2010). at <http://www.iea.org/textbase/nppdf/free/2010/key_stats_2010.pdf>
4. Loster, M. Total Primary Energy Supply From Sunlight. (2010).
5. Green, M. A., Emery, K., Hishikawa, Y., Warta, W. & Dunlop, E. D. Solar cell efficiency tables (version 39). Progress in Photovoltaics: Research and Applications 20, 12–20 (2012).
6. Scully, S. R. & McGehee, M. D. Effects of optical interference and energy transfer on exciton diffusion length measurements in organic semiconductors. Journal of Applied Physics 100, 034907 –034907–5 (2006).
7. Tsakalakos, L. et al. Silicon nanowire solar cells. Applied Physics Letters 91, 233117 –233117–3 (2007).
8. Paulose, M., Shankar, K., Varghese, O. K., Mor, G. K. & Grimes, C. A. Application of highly-ordered TiO 2 nanotube-arrays in heterojunction dye-sensitized solar cells. Journal of Physics D: Applied Physics 39, 2498–2503 (2006).
9. Zhu, R., Jiang, C.-Y., Liu, B. & Ramakrishna, S. Highly Efficient Nanoporous TiO2-Polythiophene Hybrid Solar Cells Based on Interfacial Modification Using a Metal-Free Organic Dye. Advanced Materials 21, 994–1000 (2009).
10. Coakley, K. M., Liu, Y., Goh, C. & McGehee, M. D. Ordered organic-inorganic bulk heterojunction photovoltaic cells. MRS Bulletin 30, 37–40 (2005).
11. Hadziioannou, G. Semiconducting Block Copolymers for Self-Assembled Photovoltaic Devices. MRS Bulletin 27, 456–460 (2002).
12. Evans, D. G. & Slade, R. C. T. in Layered Double Hydroxides (Duan, X. & Evans, D. G.) 1–87 (Springer Berlin Heidelberg, 2006). at <http://link.springer.com/chapter/10.1007/430_005>
13. Meyn, M., Beneke, K. & Lagaly, G. Anion-exchange reactions of layered double hydroxides. Inorg. Chem. 29, 5201–5207 (1990).
14. Sofos, M. et al. A synergistic assembly of nanoscale lamellar photoconductor hybrids. Nature Materials 8, 68–75 (2009).
15. Rand, B. P., Genoe, J., Heremans, P. & Poortmans, J. Solar cells utilizing small molecular weight organic semiconductors. Progress in Photovoltaics: Research and Applications 15, 659–676 (2007).
16. Choi, K.-S. & Steinmiller, E. M. P. Electrochemical synthesis of lamellar structured ZnO films via electrochemical interfacial surfactant templating. Electrochimica Acta 53, 6953–6960 (2008).
17. Gan, X., Gao, X., Qiu, J. & Li, X. Growth and characterization of ZnO-SDS hybrid thin films prepared by electrochemical self-assembly method. Applied Surface Science 254, 3839–3844 (2008).
18. Herman, D. J., Goldberger, J. E., Chao, S., Martin, D. T. & Stupp, S. I. Orienting Periodic Organic−Inorganic Nanoscale Domains Through One-Step Electrodeposition. ACS Nano 5, 565–573 (2010).