Final Report Summary - MOLESOL (All-carbon platforms for highly efficient molecular wire-coupled dye-sensitized solar cells)
Executive Summary:
The MOLESOL project was a 3-years European FP7 project started in October 2010 under Future Emerging Technologies for Energy Applications (FET). The project partners were IMEC/Imomec, Belgium, also project coordinator; Solarprint Limited, Ireland; Ustav Fyzikalni Chemie J. Heyrovskeho AV CR, v.v.i. Czech Republic; Dyesol Italia, Italy; Max Planck Gesellschaft Zur Foerderung der Wissenschaften E.V. Germany; Linkopings Universiteit, Sweden; Ecole Polytechnique Federale de Lausanne, Switzerland and National University of Singapore, Singapore.
The main output and results of MOLESOL at the end of the project are:
1. A highly conductive graphene film using a metal grid exhibiting a sheet resistance of ~ 30 ohm/sq with a transmittance of ~ 80%,
2. A monodispersed graphene ink from G flakes with a mobility up to 233 cm2/Vs and a sheet resistance of 1.3 kOhm/sq. with a transmittance > 80%,
3. A chemical robust and opaque printable graphene counter electrode having a sheet resistance of 5 Ohm/sq
4. An increase of 600% of the photocurrent of covalently bonded graphene-NH2-diamond electrode compared to graphene physisorb to diamond
5. A solar cell made of quantum dots, polymer and a few graphene layers as active layer blend showing an incident photon to current efficiency (IPCE) of >10% at between 300-320 nm.
6. DSC prototypes tailored for low light, showing actual conversion efficiencies under fluorescent light of 7.6% on a 1 cm x 1 cm cell using solvent based electrolytes, and 9.7% on a 6 cm x 6 cm area using a novel iodine-based ionic liquid.
7. A dye stability in DSC device of 90% after 1000 h in accelerated tests under full sunlight soaking at 60 °C
8. A carbon-based DSC prototypes using novel MOLESOL materials achieving nearly 10% efficiency on 1 cm x 1 cm area.
A nice overview of the project achievements is visibale on YOUTUBE:
http://www.youtube.com/watch?v=kMD6uduSfpw&feature=youtu.be(s’ouvre dans une nouvelle fenêtre)
More information on the MOLESOL project can be also found on www.molesol.eu
Project Context and Objectives:
Although crystalline silicon (c-Si) technologies continue to dominate the PV market, thin-film technologies have made significant progress in grid-connected applications. Thin-film based solar cells are potentially cheaper than e.g. c-Si solar cells because of their lower materials costs and larger substrates. They generally show better performance under low light conditions and may offer particular design options for building integrated applications. The most mature thin-film PV technologies use Copper Indium Gallium (di)Selenide, (CIGS), amorphous silicon (a-Si) or Cadnium Telluride (CdTe) as photovoltaic material. Attractive but less mature alternatives are organic solar cells (OPV) and dye sensitized solar cells (DSCs). OPV cell designs have recently made a significant step towards low-cost solar cell technology. But they still need to demonstrate long-term stability and power conversion efficiencies above 15% before they will be considered for large-scale production. Today, the highest power conversion efficiencies for an organic solar cell based on a bulk heterojunction (BHJ) device with [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) and low bandgap conjugated polymers is 9.2% and, 12% using soluble small molecules - but this system seems reaching its limits. Offsets in the energetics lead to large internal energy losses. DSCs on the other hand perform slightly better in terms of conversion efficiency, reaching efficiencies above 14.1% for solid state DSC. Their key component is a dye sensitized semiconductor anode and an electrolyte. However, problems with the stability of the electrolyte hinder widespread deployment.
It is our goal to develop a highly efficient molecular-wire charge transfer platform that can be used in a novel generation of thin-film dye-sensitized solar cells fabricated via organic chemistry routes. By combining the advantages of both OPV and DSC concepts, this novel type of thin-film solar cell will hence be low-cost, easy-to-fabricate, stable and highly efficient. These developments are being carried out in the frame of the MOLESOL project, a collaborative project under the European 7th framework program. In this summary, we describe the visionary approach of the project, its challenges and first realizations.
The MOLESOL approach
To explain our novel ‘MOLESOL’ solar cell concept, we start from considering a generic DSC device, also called Grätzel cell. Such a cell consists of a dye-sensitized TiO2 photoanode (n-type), an electrolyte solution with a redox mediator and the cathode material. The latter is typically a film of Pt nanoparticles on F-doped SnO2 or indium tin oxide (Pt-FTO/ITO) and the former is the I3-/I- redox couple in aprotic electrolyte medium. On photo-excitation, the dye (typically ruthenium-based) injects an electron into the n-type material and the hole is captured by the electrolyte. The electrons then travel through the nanostructure to be collected as current at the external contact, while the holes are transported to the cathode by the redox shuttle in the electrolyte solution. This solar cell is an attractive alternative to solid state OPV due to its high efficiency, low cost and ease of fabrication. In view of large-scale production, the most important drawback of such a system is the instability of the electrolyte solution.
The technology proposed within the MOLESOL project makes use of an assembled dye monolayers linked through organic molecular wires to a semiconducting thin film that is deposited on an optically transparent substrate. Short molecular wires will be used that are compatible with the exciton diffusion length. This way, the critical length for charge collection generated in the dye monolayer by the inorganic bottom electrode will be significantly reduced. In addition, the inorganic ITO/FTO n-type layer as used in a traditional DSC will be replaced by a novel transparent wideband p-type semiconductor that enables engineering of the surface work function, leading to a perfect match between the highest occupied molecular orbital (HOMO) of the dye layer and the valence band of the semiconductors. This opens routes to an increased open circuit voltage (Voc). For this purpose, the use of graphene and screen printing solutions has been investigated. The aim of this concept is to establish cost-effective, stable cells with enhanced conversion efficiency, scalable above 11%.
Project Results:
WP1. Developement of conductive nano-diamond & graphene layers for efficient charge collection
Task 1.1 Boron-doped nanodiamond films (BDD) for replacement of ITO/FTO
MOLESOL has developed a simple means to achieve a stable and low cost anode in organic photovoltaic cells (OPVs) by integrating the chemically inert boron-doped nanocrystalline diamond (BBNCD) with the highly conducting metal grid (MG). This hybrid electrode (BBNCD-MG) is a promising candidate to replace conventional indium tin oxide due to its several superior properties. Contrary to plane ITO anode, more efficient hole collection can occur along the three-dimensional metal grids with the overlaying electron blocking BNCD film. High work function of the BNCD-MG anode provides an ohmic contact for hole extraction in OPVs. Spectral response measurements suggest the photo-response of the P3HT:PCBM device with BNCD-MG anode can cover the whole range of UV-visible wavelength, which is advantage compare to ITO. In addition these BNCD electrodes can be used in DSSC solar cells when using BNCD as hole-collecting layer. Such layers for DSSC are not available on the market and could be potentially used in tandem cells where first electrode is working with a standard electron collection transparent layer and second electrode, functionalized with a dye would work for hole collection. Such approach has been developed in the MOLESOL project. The boron doping as well as the layer transparency has been optimized. As a results, BNCD electrodes with a resistivity of about 20 Ohms/sq and a transparency of 70% have been prepared and is an important achievement of this project.
Task 1.2. Preparation of highly conductive graphene (G) film from nanographene (NG) molecules and by Chemical Vapor Deposition (CVD)
We have developed hybrid film in which gold (Au) metal grids were deposited on thermally reduced graphene oxide (thickness ~ 5 nm) by photolithography. The distance between one Au grid to another was 200 μm. Such a graphene-metal grid hybrid film exhibited a sheet resistance of ~ 30 ohm/sq with transmittance of ~ 80%. The performance of the hybrid electrode is comparable to that of ITO and fulfilled our target.
We further developed other approaches such as plasma treatment of GO film for quick and low temperature reduction, electrochemical exfoliation of graphite for solution process (graphene ink) and optimized the CVD process to produce highly conductive graphene thin film.
Monodispersed G ink (from G flakes) with Mobility > 5000 cm2/Vs was a very challenging target and the best we could obtained at the end of the project is a mobility measured to be up to 233 cm2/Vs with a sheet resistance of 1.3 kOhm/sq. with > 80% transmittance
Highly conductive graphene film were also developed with the CVD process to produce single layer graphene doped with a fluoropolymer and having a sheet resistance of 256 Ohm/sq with transmittance of ~96% (at 550 nm).
Task 1.3 Chemical routes for the preparation of solution processable graphene (G) and graphene oxide (GO) layers
A chemical route based on electrochemical exfoliation of graphite has been developed to chemically expand graphite and exfoliate them into thin layer graphene sheets. These exfoliated graphene sheets can be solution-processed into conductive films and inks. We have attempted to use these sheets as carriers for photoactive dyes in solar cells and good performance has been achieved. We can synthesize graphene oxide (GO) by chemically treating the chemically expanded graphite with acid and oxidizing agents. Due to the presence of oxygen groups on GO, it can be dispersed well in water.
Task 1.4 Work function engineering of diamond and graphene surfaces and interfaces
In order to optimize the ability to inject or extract charge at a nanocrystalline diamond (NCD) or grapheme-based electrode, one must adjust the so-called work function (minimum energy needed to extract an electron from the surface of a film) of the NCD/grapheme film to fit the electronic properties of the organic semiconductor (OSC) or dye needed by a particular application/device. As a wide variety of OSC and dyes may be used, the work function should ideally be tuneable over a wide range as well. Several techniques can be used to tune the work function as it depends on both bulk and surface effects. For instance, a surface dipole-induced potential step that increases the work function can be achieved by either exchanging the terminating atomic layer of the film by another element, or by adding an additional (molecular) layer to the surface. Another approach, bulk-based, consist of changing the doping level of the film, which affects the work function by altering the position of the (bulk) Fermi level. In this project, we successfully have used all of these approaches to tune the NCD and grapheme-based electrode materials.
For boron-doped NCD films, we replaced the hydrogen atoms at the surface of the film by atomic species of higher electron affinity, thereby inducing a dipole layer at the surface with the negative part of the pole pointing out, thus increasing the overall work function. By changing the type of replacing atom and the degree of surface affected, we could tune the work function of the boron-doped NCD films over a range of 1 eV, from ~4.2 eV of the pristine films to 5.2 eV when oxygen atoms replaced the hydrogen. This very wide range of accessible work function allows the boron-doped NCD to function as moderate electron injectors (cathodes) similar to titanium up to good hole-injectors (anodes), equivalent roughly to high-grade PEDOT:PSS. To allow further fine tuning of the work function, i.e. enable access to more points between 4.2 to 5.2 eV which can be important for barrier tuning, a second approach was followed where organic molecules were chemically grafted onto the NCD film surfaces. This approach enables tuning of wetting properties as well as more precise work function tuning compared to the first approach. A wide variety of organic molecules were attached and enabled a tuning between 4.2 eV to 5.0 eV with a step size of ~0.1 eV, thus providing a powerful platform for NCD electrode optimization.
Work function tuning of graphene-based films made from reduction of graphene oxide also were carried out and here we used the approach of modifying the doping level of the films. The oxygen/nitrogen concentration in the reduced graphene oxide films were varied, enabling the work function to be tuned from 4.2 to 4.7 eV, the latter being roughly equivalent to ITO as a hole-injecting anode.
In summary, the results of Task 1.4 enables the work function of NCD films to be tuned within the measurement error to match energy levels in OSC and dyes for 4.2 eV to 5.2 eV, whereas a moderate but useful range of work function tuning was developed for graphene-based films: 4.2 eV to 4.7 eV.
WP2. Organic light harvesting molecules and antenna tailored for charge generation
Task 2.1. Synthesis of high molar extinction coefficient metal-free sensitizers compatible with nano-diamond and graphene layers
Dye-sensitized solar cells (DSCs) are a cost effective renewable energy technology. DSCs are currently in a developmental stage industrially for large area photovolatics and are already in early production stages for low-light environment applications, and on flexible substrates. A key component of a functional DSC is the light-harvesting sensitizer, which largely determines the maximum obtainable current and has direct impact on maximum voltage output. The low efficiency of the DSC is due to low molar extinction coefficient of sensitizer and lack of absorption in the entire visible regions. The WP2, task 2.1 is dedicated to synthesize sensitizers with a band gap of 1.5 eV corresponding band gap wavelength of 820 nm. Thus with a band gap of 1.5 eV, short circuit current density (Jsc) 28mA/cm2 is theoretically feasible. Taking into the consideration of 15% reflection losses, 24mA/cm2 of photocurrent density is thermodynamically achievable and by finding a synergy between the sensitizer and the redox mediator the DSC can deliver 20% power conversion efficiency. Towards this goal, we have developed a concise synthetic route to a novel π-bridge organic sensitizers with the common name, Ullazine. Organic sensitizers have undergone considerable study because of their often strong molar absorptivities and possible price advantage when compared to metal-based sensitizers. The dominant structural arrangement for organic sensitizers is the donor-π bridge-acceptor (D-π-A) arrangement. The π-bridge is responsible for the crucial roles of good electronic communication from the donor to the acceptor and providing a chromophore for better light absorption. The fused polycyclic nitrogen containing heterocycle bridge, Ullazine has been characterized and incorporated with an organic sensitizer for DSCs. Ullazine was desirably found to have both strong electron donating properties and weak electron accepting properties. Based on the Ullazine-sensitizer, we have obtained a maximum incident photon-to-electric conversion efficiency of 95%, resulting power conversion efficiency close to 10%.
Task 2.2. Synthesis of light harvesting molecules for “Suzuki” coupling to nano-diamond
The main goal for task 2.2 is to prepare light harvesting molecules which will be attached to the boron doped nano-diamond (=BDD) via covalent bond formation using Suzuki or other similar coupling strategies as described in the project proposal. General structure of the energy harvesting molecules consists of donor part, π-system and acceptor part which are connected to the nano-diamond on the donor part of the molecule in the aforementioned order. Donor and π-system part can be used either as separate entities or as one building block.
Many different dyes incorporating various core as Thiazolothiazoles, Benzodithiophenes, Porphyrin and Cyclopentadithiophene were developed. The most promising dye which was also synthesized in large quantity to be distributed to other partners for coupling to diamond substrate was (E)-2-{4-[2-(6-Bromo-4,4-diethyl-4Hcyclopenta[ 1,2-b:5,4-b']dithiophen-2-yl)vinyl]-3-cyano-5,5-dimethylfuran-2(5H)-ylidene}malononitrile
Task 2.3 Characterization and stability of organic molecular dyes
Power conversion efficiency of the dye sensitized solar (DSC) depends on the spectral response of the sensitizer attached to the nanostructured TiO2 film, and the redox mediator for dye regeneration. The low efficiency of the DSC, besides spectral response is due to the low open-circuit potential dictated by the redox mediators oxidation potential. The WP2, task 2.3 is dedicated to synthesize new redox mediators compatible with organic sensitizers with a band gap of 1.5 eV. Besides developing new redox mediators, stability measurements of the device are explored. In this task, the challenges towards achieving stable dye sensitized solar cells (DSC) based on cobalt redox shuttles are studied. And, we found that under continuous full sun (100 mW cm-2) exposure at 60° C, the concentrations of the oxidizing and reducing cobalt species change, with significant effects on cell degradation. This induced concentration alterations of the cobalt redox couple can be minimized depending on whether the device is being open circuited, short circuited or loaded close to its maximum power point during the above mentioned accelerated durability test. These conditions lead to different endurance results, peaking at 90% power conversion efficiency maintenance after 1000 h cycling as required by the task. The mechanisms occurring in the devices, together with their impact on cell performance degradations were investigated. Conclusively, for the first time ever, stable DSCs based on both Co(II/III)(terpy)2 and Co(II/III)(bpy)3 redox couples have been found after stressing those cells for 1000 h at 60 C under full sun (100 mW cm-2). On the other hand, when ageing the DSCs under short circuit or under a load close to the cell’s maximum power point, this gradual elimination of the reducing cobalt species is circumvented, resulting in cells that maintain up to 90% of the initial PCE employing the Co(bpy)3 redox shuttle.
WP3. Development of molecular wire coupling routes, achieving full charge transfer
Task 3.1. Specific surface functionalization of nano-diamond layers towards coupling reaction
We have successfully functionalized B-NCD surfaces with two molecular wires. The two molecular wires, which we have worked with, 4-bromophenyl and arylboronic esters are functionalized onto B-NCD surfaces with diazonium coupling reaction. Two diazonium coupling techniques are used, namely, (i) spontaneous grafting and, (ii) cyclic voltammetry reduction of in situ generated aryldiazonium salts. Hydrogenated B-NCD is used for both of the diazonium coupling reactions. 4-bromophenyl molecular wires are functionalized with spontaneous grafting which is by simply immersing a hydrogenated B-NCD into an acidic solution containing 4-bromophenyl diazonium salts. Arylboronic ester molecular wires are functionalized with the second technique where cyclic voltammetry scans are applied to the in situ generated arylboronic esters diazonium salts. X-ray photoelectron spectroscopy (XPS) is used to follow the surface composition before and after the functionalization. The B-NCD surfaces functionalized with these types of molecular wires are excellent solid platforms for further Suzuki Coupling with molecular dyes which provide uninterrupted conjugated C-C bond for efficient charge transfer.
Task 3.2. Coupling strategies
Two major coupling reactions employed in this project are Suzuki and EDC-NHS Coupling. The coupling reaction is chosen depending on the molecular structure of the molecular dyes. Three molecular dyes, namely, (i) (Z)-2-(5-((5'-bromo-[2,2'-bithiophen]-5-yl)methylene)-4-oxo-2-thioxothiazolidin-3-yl)acetic acid (Br-2T-Rhodanine), (ii) (E)-2-{4-[2-(6-Bromo-4,4-diethyl-4H-cyclopenta[1,2-b:5,4-b']dithiophen-2-yl)vinyl]-3-cyano-5,5-dimethylfuran-2(5H)-ylidene}malononitrile (Br-CPDT-Acceptor), and (iii) bis(isothiocyanato)bis(2,2'-bipyridyl-4,4'-dicarboxylato)-ruthenium(II) (N3) are coupled to the molecular wires functionalized B-NCD surfaces from Task 3.1.
(i)(Z)-2-(5-((5'-bromo-[2,2'-bithiophen]-5-yl)methylene)-4-oxo-2-thioxothiazolidin-3-yl)acetic acid (Br-2T-Rhodanine) Molecular Dye.
This molecular dye is coupled to the arylboronic ester molecular wires functionalized B-NCD surface through Suzuki Coupling reaction. Three catalytic systems were tested. Highest surface coverage of 0.58 monolayer was obtained when SPhos + Pd(OAc)2 catalytic system was used.
(ii)(E)-2-{4-[2-(6-Bromo-4,4-diethyl-4H-cyclopenta[1,2-b:5,4-b']dithiophen-2-yl)vinyl]-3-cyano-5,5-dimethylfuran-2(5H)-ylidene}malononitrile (Br-CPDT-Acceptor) Molecular Dye.
This molecular dye is coupled to the arylboronic ester molecular wires functionalized B-NCD surface through Suzuki Coupling reaction. Three catalytic systems were tested. Highest surface coverage of 0.56 monolayer was obtained when SPhos + Pd(OAc)2 catalytic system was used.
(iii) bis(isothiocyanato)bis(2,2'-bipyridyl-4,4'-dicarboxylato)-ruthenium(II) (N3).
To couple this molecular dye, 4-bromophenyl functionalized B-NCD was first Suzuki coupled with 4-aminophenyl boronic ester to form a layer of (1,1'-biphenyl)-4-amine group on B-NCD. N3 dye was then EDC-NHS coupled to this layer. Surface coverage as estimated by XPS was 3×1014 Ru/cm2.
X-ray Photoelectron Spectroscopy (XPS) is used to follow the surface compositions before and after each functionalization of the three molecular dyes on B-NCD surfaces. Surface coverage of the molecular dyes on B-NCD surfaces is also estimated by XPS. For molecular dye (i) and (ii), the surface coverage was estimated by the S 2p peak intensity. For molecular dye (iii), Ru 3d3/2 was used to estimate the surface coverage. The surface coverage of the molecular dyes on B-NCD obtained with Suzuki Coupling in this project is the highest compared to the reported value in the literature.
Complete device was fabricated using B-NCD that functionalized with molecular dye (ii). The cell performance was measured to be 0.05 %. Two possible reasons might be attributed for the low cell performance. Firstly, the resistivity of the B-NCD electrode used in the time frame of this project is somehow still high compared to the other transparent electrodes like FTO and ITO. Secondly, the surface coverage of the molecular dye on B-NCD is still low if compared to the common electrode used in classical dye-sensitized solar cells, where the amount of light absorbing dyes that are absorbed on mesoporous titania electrode is higher.
Task 3.3. Coupling of graphene layers via molecular wire
We have developed a process to couple graphene via molecular chains (a long molecule) to diamond substrate. The anchoring of graphene using molecular chain improves the chemical robustness and mechanical strength. Even when we use ultrasound wave to agitate the graphene flake, it can withstand shock and does not fragment. In the absence of the molecular chain, graphene sheet will fragment. This suggests that the molecular chain acts like springs and can cushion the mechanical shock. When light is shone onto this platform, the light can be converted to electricity with good efficiency. Onto this platform, dyes can be coupled to make a solar cell.
Task 3.4. Electrochemical and photoelectrochemical characterization of pure and modified nanodiamond-based electrodes
Electrochemical characterization of boron doped diamond electrodes (BDD) was carried out by using standard techniques, like cyclic voltammetry both in pure supporting electrolyte solution and that with model redox probes to test the interfacial charge transfer. These studies were upgraded by the application of electrochemical impedance spectroscopy (EIS), by the in-situ Raman spectroelectrochemistry and by the atomic forace microscopy (AFM) interfaced to electrochemical treatment of the electrode surface. The obtained results provided selfconsistent information about the electrochemical activity of BDD in relation to the doping level and to the quality of the diamond sample, viz. the presence of defects manifested themselves as sp2 impurities. Mott-Schottky plots from EIS allowed accurate monitoring of the p-doping and the effect of surface functionalization by the dyes. The cleaning of the surface by “electrochemical burning” of sp2 carbon at large anodic potentials was observed and characterized by AFM and Raman spectroelectrochemistry. The fabrication of MWCD (molecular wire-coupled dye-sensitized) solar cells enabled a better understanding of interfacial charge transfer at the diamond/electrolyte solution and also in the dye surface modification of the diamond. To test the surface anchoring, ATR-FTIR spectra were measured. The functionalization of O-terminated BDD manifests itself in the region of C-H stretching vibrations, but the remaining vibrations of the used dye (JD-18) were not resolved. Raman in-situ spectroelectrochemical tests showed the high quality of both polycrystalline and single crystalline BDD films and the protective role of dyes wired to the BDD surface. Raman spectroelectrochemistry points to a positive influence of the dye, which serves as a protective layer sealing the BDD surface against the disruptive influence of intercalating ions during the electrode polarization. Such an effect is of a paramount importance for a long term stability of MWCD solar cells. Photoelectrochemical characterization of dye-sensitized BDD were performed in aqueous electrolyte solution containing methyl viologen (MV2+) as the electron carrier, using a three electrode system: the diamond-based molecular platform as working electrode, a platinum wire counter electrode, and a Ag/AgCl reference electrode, under 1 sun irradiation (100 mW/cm2, AM 1.5G). The corresponding tests confirmed both cathodic and anodic photocurrents in the range of μA/cm2 upon illumination with simulated AM 1.5G solar radiation.
Task 3.5 Characterization of chemical and electronic structure of interfaces between conductive nano-diamond and organic molecular dyes
The aim of this task was to aid in the development of high-quality dye-functionalized boron-doped nanocrystalline diamond (BNCD) surfaces, and it was carried out in an iterative fashion involving the synthetic groups and the surface characterization group. The functionalization of the BNCD surfaces involves several steps where first an organic "anchor“ group is chemically attached to the surface whereupon in a second step the desired dye is coupled to the anchor group. Each step involves complicated chemical reactions, who’s efficacy must be evaluated in order to optimize the whole multi-step synthetic procedure and in extension the dye-coverage of the BNCD surface. Over a hundred samples were hence studied in this task by a combined XPS/UPS approach (occupied electronic structure, elemental composition and chemical bonding), and for those samples that past the quality criteria NEXAFS measurements were added as well, to further characterize the unoccupied electronic structure and the dye molecular order (if any) on the surface.
XPS and NEXAFS results showed that anchor groups can efficently be grafted onto the BNCD surface, subseqently serving as templates for the dye bonding. Of the organic molecular dye systems tested, the so-called N3B4 dye proved to be the most successful in terms of yielding dense films when attached to the functionalized BNCD. UPS and NEXAFS measurements showed that the dye electronic structure of the grafted films correctly reproduced that of the "neat" dye films, with slight modifications to to the bonding to the anchor groups, as expected. XPS measurements were used to determine the density of the dye coverage, which was found to be roughly three N3B4 molecules per nm2. The high-quality functionilized films were not achieved in time for the last synchrotron radiation beam time available to the consortium, hence the so-called core hole clock emasurements to probe the charge transfer dynamics at the dye-BNCD interface could not be carried out before the end of the project.
WP4. Efficient implementation of technological platform for solar cell fabrication
Task 4.1. Development of a chemically robust printable graphene inks to be used as counter electrode with low sheet resistance
An opaque counter electrode graphene material with low sheet resistance has been produced, replacing the platinum catalyst typically employed on the counter electrode of a dye-sensitized solar cell. The graphene was intended to be deposited by printing methods, and therefore was formulated into a printable ink. The objective was then to meet ambitious targets for sheet resistance of the deposited graphene layer.
A graphene ink comprising milled graphene, conductive polymer binder and certain additives was formulated and printed by doctor blade onto glass substrates. After pressing, a uniform layer of less than 50 μm was ultimately achieved. Samples of various deposition areas and patterns were produced, demonstrating the versatility of the process. The formulation and pressing process was then continuously improved through the first two years of the project, culiminating in an electrode with sheet resistance of the carbon layer less than < 5 Ohm/sq.
Task 4.2. Printing processes for the MWDC
Following the findings of task 4.1 doctor blade printing and pressing of the graphene ink formulation was developed as the primary method for producing small runs of sample cells. This process is easily transferable
to screen printing or slot-dye coating as required for volume production, with print areas adjustable to meet each partner’s cell configurations.
The development also included printed layers for other components of the prototype cells, enabling simple, low cost production of samples. Sealing for reliability of DSSC is a particular concern in the field, and fully printable low temperature-cured seals were developed for this purpose.
More specialized deposition methods for graphene nano platelets and graphene oxide were also tested, including spin coating and air driven spray coating; the latter was successfully demonstrated on up to 10cm x 10cm areas.
Task 4.3. Prototyping of printable DSC
The partners in WP4 all proceeded to make prototype DSC using the novel materials developed within this work package and also from other work packages, and performed analysis of the prototypes against benchmarks. The benchmarking protocols were developed as part of the task, considering existing test standards for silicon photovoltaics, as well as suitable acceptable and realizable industry and academic tests that could be used to evaluate performance and reliability.
Cells prepared using novel materials from all the work packages have achieved nearly 10% efficiency on 1cm x 1cm area. This is the best result for carbon based DSC that we are aware of, though it falls slightly short of the original MOLESOL target for a cell this size.
The prototypes prepared by SolarPrint, tailored for low light, show actual conversion efficiencies under fluorescent light of 7.6% on a 1cm x 1cm cell using solvent based electrolytes, and 9.7% on a 6cm x 6cm area when using a novel iodine-based ionic liquid. However these ‘low light’ efficiencies are a non-standard measurement outside the standard benchmarking protocol.
Potential Impact:
There is an urgent necessity for reduction of dangerous environmental impact in the energy production & for maintaining a SECURED and LOW COST energy resource supply. To do so, there is a necessity for the development of cheap and environmentally-friendly energy production: Printed efficient large area OPV is one promising candidate: real breakthroughs will come from dedicated material development and innovative device concepts. However in order to be really competitive with existing technologies worldwide and on European level and see a large market development, further improvement on efficiency, stability and cost are needed. For the outdoor use and a real competition with the inorganic solar cells market for photovoltaic solar energy conversion devices, especially the CE and the stability are the current bottlenecks, which current OPVs are not able to meet due to the limitation of organic polymeric systems. Unless this hurdle is lifted, the replacement of environmentally dangerous technologies will be strongly limited, due to costs and performance criteria. To propose alternative pathways and to meet these challenges we suggest developing the MWCD solar cell platform, which extend the possibilities of DSSC technologies. At the same time development of printable graphene and nanodiamond technologies makes an important step towards the effective replacement of current ITO/FTO electrodes, which have dominant position on the photovoltaic market as whole and which resources are limited. These carbon-based transparent and conducive thin films can be used for many already currently developed solar cells platforms and can lead to a true success story in PV industry.
The potential impacts of MOLESOL results are the following:
- Today, with no large market for organic PV existing, the state-of-the-art market requirements for organic PV can be derived on the basis of the current PV market. The market segments are defined focusing on organic photovoltaics as:
Indoor applications / outdoor consumer applications / off-grid applications
PV as building design element / grid-connected bulk power generation
- OPV is still at an early stage, but due to its intrinsic advantages such as large area coverage, lower price level and device flexibility, it is foreseen that OPV will claim a relevant share in the huge upcoming market of PV devices. With further improvement, the MWCD solar cell which combines the advantages of the OPV
and DSSC concept will impact on development of novel highly efficient, stable and cost effective solar cells.
- In a wider sense the novel materials developed in MOLESOL belong to highly attractive category of nanomaterials with some of them - as printable graphenes being hot candidates for becoming a major player in micro and nanoelectronics. MOLESOL has promoted and pioneered the applications of graphene in photovoltaics.
- Carbon materials prepared in MOLESOL can have a large impact on the utilization of low-cost solar cells. Looking to the price/performance ratio, an estimate for ND coating cost is < 50 Euro/m2 using the current MicroWave-Liner Antenna Technology. The graphene-based printing methods are still of lower cost.
- Additionally, individual components of the proposed technology (as nanodiamond or graphene conductive transparent layers) can be used for both OPVs and DSSCs and even inorganic SCs such as aSi hybrid cells etc.
- The novel solar cells developed in that project are completely carbon-based and can therefore be considered as a ‘green product’ with respect to their production, operation and disposal. Fabrication processes via printing in the project is using as much as possible environmentally-friendly solvents for ink formulations - with therefore a positive impact on environment, and on the reduction of energy consumption: moderate temperature, near-atmospheric pressure, low waste volume.
- The future of traditional industries will largely depend on new knowledge, new materials and on new ways of integrating and exploiting existing and new knowledge. The potential of this project is stimulation of a new prosperous primary and secondary industry for new type of photovoltaics in Europe. In particular new markets are envisaged for organic and nano- materials suppliers that are traditionally not in the photovoltaics business. This is also true for industry involved in flat-glass, display, battery or electrical products. The new technologies developed by the project will have a positive impact on the European society on medium to long term with the creation of high-quality jobs by industries.
The dissemination and use of the MOLESOL results are:
MOLESOL is a FET ‘Future Emerging Technologies for Energy Applications” project under FP7. The aim of such high-risk, high reward research projects is to encourage breakthroughs and radical new ideas, since incremental progress in existing solutions may not be sufficient and is also to push the limits of the future emerging technologies. The research scheme is completely "bottom-up" and scientists are encouraged to think "out-of-the-box". The research proposed in MOLESOL is therefore highly novel and ambitious.
Materials research is likely the most important element for the development of the necessary technologies needed to provide a clean, reliable supply of efficient energy and FET projects is encouraging the materials community to work together and share their findings.
Patents evaluation followed during the all project duration revealed to be more complicated and tricky than expected. This due to the intrinsic complexity and novel multi-disciplinary ideas of the projects outputs.
Anyhow several patents that could be in contrast with MOLESOL outputs have been found and they were subjected to consortium partners attention for review to understand their effectiveness relevancy. The greatest part referred to graphene and only very few to nano-diamonds.
At the end of the project and at this stage of development the relevance under a commercial point of view of the MOLESOL know-how seems to be marginal for companies and no patent filling was done yet. Further development is needed to have patent applications.
However the MOLESOL results were largely disseminated by means of the project webpage, several press releases, a newsletter, a public video of 20 minutes about the project visible on YOUTUBE, an international MOLESOL workshop and presented at several international conferences to R&D and industrial public.
List of Websites:
Project partners:
IMEC/Imomec, Belgium; Solarprint Limited, Ireland; Ustav Fyzikalni Chemie J. Heyrovskeho AV CR, v.v.i. Czech Republic; Dyesol Italia, Italy; Max Planck Gesellschaft Zur Foerderung der Wissenschaften E.V. Germany; Linkopings Universiteit, Sweden; Ecole Polytechnique Federale de Lausanne, Switzerland and National University of Singapore, Singapore.
Imec/Imomec is project coordinator.
Project webpage:
More information on the MOLESOL project can be found on www.molesol.eu.
Project press releases and newsletter:
Nanowork: http://www.nanowerk.com/news/newsid=22498.php(s’ouvre dans une nouvelle fenêtre)
InterPV: http://www.interpv.net/head/head_view.asp?idx=503&part_code=03(s’ouvre dans une nouvelle fenêtre)
IMEC e-news: http://imec.fb.email.addemar.com/c1694/e0/h0916f/t2/s0/index.html(s’ouvre dans une nouvelle fenêtre)
Future PV magazine: http://www.futurepv.com/node/401(s’ouvre dans une nouvelle fenêtre)
Project video:
A video which give an nice overview of the project goals and achievements is visible on YOUTUBE by all public.
http://www.youtube.com/watch?v=kMD6uduSfpw&feature=youtu.be(s’ouvre dans une nouvelle fenêtre)
Contact:
Dr laurence LUTSEN
IMEC-Imomec
Phone : +32 (0)11 26 83 14
Fax : +32 (0)11 26 83 01
e-mail : laurence.lutsen@imec.be
The MOLESOL project was a 3-years European FP7 project started in October 2010 under Future Emerging Technologies for Energy Applications (FET). The project partners were IMEC/Imomec, Belgium, also project coordinator; Solarprint Limited, Ireland; Ustav Fyzikalni Chemie J. Heyrovskeho AV CR, v.v.i. Czech Republic; Dyesol Italia, Italy; Max Planck Gesellschaft Zur Foerderung der Wissenschaften E.V. Germany; Linkopings Universiteit, Sweden; Ecole Polytechnique Federale de Lausanne, Switzerland and National University of Singapore, Singapore.
The main output and results of MOLESOL at the end of the project are:
1. A highly conductive graphene film using a metal grid exhibiting a sheet resistance of ~ 30 ohm/sq with a transmittance of ~ 80%,
2. A monodispersed graphene ink from G flakes with a mobility up to 233 cm2/Vs and a sheet resistance of 1.3 kOhm/sq. with a transmittance > 80%,
3. A chemical robust and opaque printable graphene counter electrode having a sheet resistance of 5 Ohm/sq
4. An increase of 600% of the photocurrent of covalently bonded graphene-NH2-diamond electrode compared to graphene physisorb to diamond
5. A solar cell made of quantum dots, polymer and a few graphene layers as active layer blend showing an incident photon to current efficiency (IPCE) of >10% at between 300-320 nm.
6. DSC prototypes tailored for low light, showing actual conversion efficiencies under fluorescent light of 7.6% on a 1 cm x 1 cm cell using solvent based electrolytes, and 9.7% on a 6 cm x 6 cm area using a novel iodine-based ionic liquid.
7. A dye stability in DSC device of 90% after 1000 h in accelerated tests under full sunlight soaking at 60 °C
8. A carbon-based DSC prototypes using novel MOLESOL materials achieving nearly 10% efficiency on 1 cm x 1 cm area.
A nice overview of the project achievements is visibale on YOUTUBE:
http://www.youtube.com/watch?v=kMD6uduSfpw&feature=youtu.be(s’ouvre dans une nouvelle fenêtre)
More information on the MOLESOL project can be also found on www.molesol.eu
Project Context and Objectives:
Although crystalline silicon (c-Si) technologies continue to dominate the PV market, thin-film technologies have made significant progress in grid-connected applications. Thin-film based solar cells are potentially cheaper than e.g. c-Si solar cells because of their lower materials costs and larger substrates. They generally show better performance under low light conditions and may offer particular design options for building integrated applications. The most mature thin-film PV technologies use Copper Indium Gallium (di)Selenide, (CIGS), amorphous silicon (a-Si) or Cadnium Telluride (CdTe) as photovoltaic material. Attractive but less mature alternatives are organic solar cells (OPV) and dye sensitized solar cells (DSCs). OPV cell designs have recently made a significant step towards low-cost solar cell technology. But they still need to demonstrate long-term stability and power conversion efficiencies above 15% before they will be considered for large-scale production. Today, the highest power conversion efficiencies for an organic solar cell based on a bulk heterojunction (BHJ) device with [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) and low bandgap conjugated polymers is 9.2% and, 12% using soluble small molecules - but this system seems reaching its limits. Offsets in the energetics lead to large internal energy losses. DSCs on the other hand perform slightly better in terms of conversion efficiency, reaching efficiencies above 14.1% for solid state DSC. Their key component is a dye sensitized semiconductor anode and an electrolyte. However, problems with the stability of the electrolyte hinder widespread deployment.
It is our goal to develop a highly efficient molecular-wire charge transfer platform that can be used in a novel generation of thin-film dye-sensitized solar cells fabricated via organic chemistry routes. By combining the advantages of both OPV and DSC concepts, this novel type of thin-film solar cell will hence be low-cost, easy-to-fabricate, stable and highly efficient. These developments are being carried out in the frame of the MOLESOL project, a collaborative project under the European 7th framework program. In this summary, we describe the visionary approach of the project, its challenges and first realizations.
The MOLESOL approach
To explain our novel ‘MOLESOL’ solar cell concept, we start from considering a generic DSC device, also called Grätzel cell. Such a cell consists of a dye-sensitized TiO2 photoanode (n-type), an electrolyte solution with a redox mediator and the cathode material. The latter is typically a film of Pt nanoparticles on F-doped SnO2 or indium tin oxide (Pt-FTO/ITO) and the former is the I3-/I- redox couple in aprotic electrolyte medium. On photo-excitation, the dye (typically ruthenium-based) injects an electron into the n-type material and the hole is captured by the electrolyte. The electrons then travel through the nanostructure to be collected as current at the external contact, while the holes are transported to the cathode by the redox shuttle in the electrolyte solution. This solar cell is an attractive alternative to solid state OPV due to its high efficiency, low cost and ease of fabrication. In view of large-scale production, the most important drawback of such a system is the instability of the electrolyte solution.
The technology proposed within the MOLESOL project makes use of an assembled dye monolayers linked through organic molecular wires to a semiconducting thin film that is deposited on an optically transparent substrate. Short molecular wires will be used that are compatible with the exciton diffusion length. This way, the critical length for charge collection generated in the dye monolayer by the inorganic bottom electrode will be significantly reduced. In addition, the inorganic ITO/FTO n-type layer as used in a traditional DSC will be replaced by a novel transparent wideband p-type semiconductor that enables engineering of the surface work function, leading to a perfect match between the highest occupied molecular orbital (HOMO) of the dye layer and the valence band of the semiconductors. This opens routes to an increased open circuit voltage (Voc). For this purpose, the use of graphene and screen printing solutions has been investigated. The aim of this concept is to establish cost-effective, stable cells with enhanced conversion efficiency, scalable above 11%.
Project Results:
WP1. Developement of conductive nano-diamond & graphene layers for efficient charge collection
Task 1.1 Boron-doped nanodiamond films (BDD) for replacement of ITO/FTO
MOLESOL has developed a simple means to achieve a stable and low cost anode in organic photovoltaic cells (OPVs) by integrating the chemically inert boron-doped nanocrystalline diamond (BBNCD) with the highly conducting metal grid (MG). This hybrid electrode (BBNCD-MG) is a promising candidate to replace conventional indium tin oxide due to its several superior properties. Contrary to plane ITO anode, more efficient hole collection can occur along the three-dimensional metal grids with the overlaying electron blocking BNCD film. High work function of the BNCD-MG anode provides an ohmic contact for hole extraction in OPVs. Spectral response measurements suggest the photo-response of the P3HT:PCBM device with BNCD-MG anode can cover the whole range of UV-visible wavelength, which is advantage compare to ITO. In addition these BNCD electrodes can be used in DSSC solar cells when using BNCD as hole-collecting layer. Such layers for DSSC are not available on the market and could be potentially used in tandem cells where first electrode is working with a standard electron collection transparent layer and second electrode, functionalized with a dye would work for hole collection. Such approach has been developed in the MOLESOL project. The boron doping as well as the layer transparency has been optimized. As a results, BNCD electrodes with a resistivity of about 20 Ohms/sq and a transparency of 70% have been prepared and is an important achievement of this project.
Task 1.2. Preparation of highly conductive graphene (G) film from nanographene (NG) molecules and by Chemical Vapor Deposition (CVD)
We have developed hybrid film in which gold (Au) metal grids were deposited on thermally reduced graphene oxide (thickness ~ 5 nm) by photolithography. The distance between one Au grid to another was 200 μm. Such a graphene-metal grid hybrid film exhibited a sheet resistance of ~ 30 ohm/sq with transmittance of ~ 80%. The performance of the hybrid electrode is comparable to that of ITO and fulfilled our target.
We further developed other approaches such as plasma treatment of GO film for quick and low temperature reduction, electrochemical exfoliation of graphite for solution process (graphene ink) and optimized the CVD process to produce highly conductive graphene thin film.
Monodispersed G ink (from G flakes) with Mobility > 5000 cm2/Vs was a very challenging target and the best we could obtained at the end of the project is a mobility measured to be up to 233 cm2/Vs with a sheet resistance of 1.3 kOhm/sq. with > 80% transmittance
Highly conductive graphene film were also developed with the CVD process to produce single layer graphene doped with a fluoropolymer and having a sheet resistance of 256 Ohm/sq with transmittance of ~96% (at 550 nm).
Task 1.3 Chemical routes for the preparation of solution processable graphene (G) and graphene oxide (GO) layers
A chemical route based on electrochemical exfoliation of graphite has been developed to chemically expand graphite and exfoliate them into thin layer graphene sheets. These exfoliated graphene sheets can be solution-processed into conductive films and inks. We have attempted to use these sheets as carriers for photoactive dyes in solar cells and good performance has been achieved. We can synthesize graphene oxide (GO) by chemically treating the chemically expanded graphite with acid and oxidizing agents. Due to the presence of oxygen groups on GO, it can be dispersed well in water.
Task 1.4 Work function engineering of diamond and graphene surfaces and interfaces
In order to optimize the ability to inject or extract charge at a nanocrystalline diamond (NCD) or grapheme-based electrode, one must adjust the so-called work function (minimum energy needed to extract an electron from the surface of a film) of the NCD/grapheme film to fit the electronic properties of the organic semiconductor (OSC) or dye needed by a particular application/device. As a wide variety of OSC and dyes may be used, the work function should ideally be tuneable over a wide range as well. Several techniques can be used to tune the work function as it depends on both bulk and surface effects. For instance, a surface dipole-induced potential step that increases the work function can be achieved by either exchanging the terminating atomic layer of the film by another element, or by adding an additional (molecular) layer to the surface. Another approach, bulk-based, consist of changing the doping level of the film, which affects the work function by altering the position of the (bulk) Fermi level. In this project, we successfully have used all of these approaches to tune the NCD and grapheme-based electrode materials.
For boron-doped NCD films, we replaced the hydrogen atoms at the surface of the film by atomic species of higher electron affinity, thereby inducing a dipole layer at the surface with the negative part of the pole pointing out, thus increasing the overall work function. By changing the type of replacing atom and the degree of surface affected, we could tune the work function of the boron-doped NCD films over a range of 1 eV, from ~4.2 eV of the pristine films to 5.2 eV when oxygen atoms replaced the hydrogen. This very wide range of accessible work function allows the boron-doped NCD to function as moderate electron injectors (cathodes) similar to titanium up to good hole-injectors (anodes), equivalent roughly to high-grade PEDOT:PSS. To allow further fine tuning of the work function, i.e. enable access to more points between 4.2 to 5.2 eV which can be important for barrier tuning, a second approach was followed where organic molecules were chemically grafted onto the NCD film surfaces. This approach enables tuning of wetting properties as well as more precise work function tuning compared to the first approach. A wide variety of organic molecules were attached and enabled a tuning between 4.2 eV to 5.0 eV with a step size of ~0.1 eV, thus providing a powerful platform for NCD electrode optimization.
Work function tuning of graphene-based films made from reduction of graphene oxide also were carried out and here we used the approach of modifying the doping level of the films. The oxygen/nitrogen concentration in the reduced graphene oxide films were varied, enabling the work function to be tuned from 4.2 to 4.7 eV, the latter being roughly equivalent to ITO as a hole-injecting anode.
In summary, the results of Task 1.4 enables the work function of NCD films to be tuned within the measurement error to match energy levels in OSC and dyes for 4.2 eV to 5.2 eV, whereas a moderate but useful range of work function tuning was developed for graphene-based films: 4.2 eV to 4.7 eV.
WP2. Organic light harvesting molecules and antenna tailored for charge generation
Task 2.1. Synthesis of high molar extinction coefficient metal-free sensitizers compatible with nano-diamond and graphene layers
Dye-sensitized solar cells (DSCs) are a cost effective renewable energy technology. DSCs are currently in a developmental stage industrially for large area photovolatics and are already in early production stages for low-light environment applications, and on flexible substrates. A key component of a functional DSC is the light-harvesting sensitizer, which largely determines the maximum obtainable current and has direct impact on maximum voltage output. The low efficiency of the DSC is due to low molar extinction coefficient of sensitizer and lack of absorption in the entire visible regions. The WP2, task 2.1 is dedicated to synthesize sensitizers with a band gap of 1.5 eV corresponding band gap wavelength of 820 nm. Thus with a band gap of 1.5 eV, short circuit current density (Jsc) 28mA/cm2 is theoretically feasible. Taking into the consideration of 15% reflection losses, 24mA/cm2 of photocurrent density is thermodynamically achievable and by finding a synergy between the sensitizer and the redox mediator the DSC can deliver 20% power conversion efficiency. Towards this goal, we have developed a concise synthetic route to a novel π-bridge organic sensitizers with the common name, Ullazine. Organic sensitizers have undergone considerable study because of their often strong molar absorptivities and possible price advantage when compared to metal-based sensitizers. The dominant structural arrangement for organic sensitizers is the donor-π bridge-acceptor (D-π-A) arrangement. The π-bridge is responsible for the crucial roles of good electronic communication from the donor to the acceptor and providing a chromophore for better light absorption. The fused polycyclic nitrogen containing heterocycle bridge, Ullazine has been characterized and incorporated with an organic sensitizer for DSCs. Ullazine was desirably found to have both strong electron donating properties and weak electron accepting properties. Based on the Ullazine-sensitizer, we have obtained a maximum incident photon-to-electric conversion efficiency of 95%, resulting power conversion efficiency close to 10%.
Task 2.2. Synthesis of light harvesting molecules for “Suzuki” coupling to nano-diamond
The main goal for task 2.2 is to prepare light harvesting molecules which will be attached to the boron doped nano-diamond (=BDD) via covalent bond formation using Suzuki or other similar coupling strategies as described in the project proposal. General structure of the energy harvesting molecules consists of donor part, π-system and acceptor part which are connected to the nano-diamond on the donor part of the molecule in the aforementioned order. Donor and π-system part can be used either as separate entities or as one building block.
Many different dyes incorporating various core as Thiazolothiazoles, Benzodithiophenes, Porphyrin and Cyclopentadithiophene were developed. The most promising dye which was also synthesized in large quantity to be distributed to other partners for coupling to diamond substrate was (E)-2-{4-[2-(6-Bromo-4,4-diethyl-4Hcyclopenta[ 1,2-b:5,4-b']dithiophen-2-yl)vinyl]-3-cyano-5,5-dimethylfuran-2(5H)-ylidene}malononitrile
Task 2.3 Characterization and stability of organic molecular dyes
Power conversion efficiency of the dye sensitized solar (DSC) depends on the spectral response of the sensitizer attached to the nanostructured TiO2 film, and the redox mediator for dye regeneration. The low efficiency of the DSC, besides spectral response is due to the low open-circuit potential dictated by the redox mediators oxidation potential. The WP2, task 2.3 is dedicated to synthesize new redox mediators compatible with organic sensitizers with a band gap of 1.5 eV. Besides developing new redox mediators, stability measurements of the device are explored. In this task, the challenges towards achieving stable dye sensitized solar cells (DSC) based on cobalt redox shuttles are studied. And, we found that under continuous full sun (100 mW cm-2) exposure at 60° C, the concentrations of the oxidizing and reducing cobalt species change, with significant effects on cell degradation. This induced concentration alterations of the cobalt redox couple can be minimized depending on whether the device is being open circuited, short circuited or loaded close to its maximum power point during the above mentioned accelerated durability test. These conditions lead to different endurance results, peaking at 90% power conversion efficiency maintenance after 1000 h cycling as required by the task. The mechanisms occurring in the devices, together with their impact on cell performance degradations were investigated. Conclusively, for the first time ever, stable DSCs based on both Co(II/III)(terpy)2 and Co(II/III)(bpy)3 redox couples have been found after stressing those cells for 1000 h at 60 C under full sun (100 mW cm-2). On the other hand, when ageing the DSCs under short circuit or under a load close to the cell’s maximum power point, this gradual elimination of the reducing cobalt species is circumvented, resulting in cells that maintain up to 90% of the initial PCE employing the Co(bpy)3 redox shuttle.
WP3. Development of molecular wire coupling routes, achieving full charge transfer
Task 3.1. Specific surface functionalization of nano-diamond layers towards coupling reaction
We have successfully functionalized B-NCD surfaces with two molecular wires. The two molecular wires, which we have worked with, 4-bromophenyl and arylboronic esters are functionalized onto B-NCD surfaces with diazonium coupling reaction. Two diazonium coupling techniques are used, namely, (i) spontaneous grafting and, (ii) cyclic voltammetry reduction of in situ generated aryldiazonium salts. Hydrogenated B-NCD is used for both of the diazonium coupling reactions. 4-bromophenyl molecular wires are functionalized with spontaneous grafting which is by simply immersing a hydrogenated B-NCD into an acidic solution containing 4-bromophenyl diazonium salts. Arylboronic ester molecular wires are functionalized with the second technique where cyclic voltammetry scans are applied to the in situ generated arylboronic esters diazonium salts. X-ray photoelectron spectroscopy (XPS) is used to follow the surface composition before and after the functionalization. The B-NCD surfaces functionalized with these types of molecular wires are excellent solid platforms for further Suzuki Coupling with molecular dyes which provide uninterrupted conjugated C-C bond for efficient charge transfer.
Task 3.2. Coupling strategies
Two major coupling reactions employed in this project are Suzuki and EDC-NHS Coupling. The coupling reaction is chosen depending on the molecular structure of the molecular dyes. Three molecular dyes, namely, (i) (Z)-2-(5-((5'-bromo-[2,2'-bithiophen]-5-yl)methylene)-4-oxo-2-thioxothiazolidin-3-yl)acetic acid (Br-2T-Rhodanine), (ii) (E)-2-{4-[2-(6-Bromo-4,4-diethyl-4H-cyclopenta[1,2-b:5,4-b']dithiophen-2-yl)vinyl]-3-cyano-5,5-dimethylfuran-2(5H)-ylidene}malononitrile (Br-CPDT-Acceptor), and (iii) bis(isothiocyanato)bis(2,2'-bipyridyl-4,4'-dicarboxylato)-ruthenium(II) (N3) are coupled to the molecular wires functionalized B-NCD surfaces from Task 3.1.
(i)(Z)-2-(5-((5'-bromo-[2,2'-bithiophen]-5-yl)methylene)-4-oxo-2-thioxothiazolidin-3-yl)acetic acid (Br-2T-Rhodanine) Molecular Dye.
This molecular dye is coupled to the arylboronic ester molecular wires functionalized B-NCD surface through Suzuki Coupling reaction. Three catalytic systems were tested. Highest surface coverage of 0.58 monolayer was obtained when SPhos + Pd(OAc)2 catalytic system was used.
(ii)(E)-2-{4-[2-(6-Bromo-4,4-diethyl-4H-cyclopenta[1,2-b:5,4-b']dithiophen-2-yl)vinyl]-3-cyano-5,5-dimethylfuran-2(5H)-ylidene}malononitrile (Br-CPDT-Acceptor) Molecular Dye.
This molecular dye is coupled to the arylboronic ester molecular wires functionalized B-NCD surface through Suzuki Coupling reaction. Three catalytic systems were tested. Highest surface coverage of 0.56 monolayer was obtained when SPhos + Pd(OAc)2 catalytic system was used.
(iii) bis(isothiocyanato)bis(2,2'-bipyridyl-4,4'-dicarboxylato)-ruthenium(II) (N3).
To couple this molecular dye, 4-bromophenyl functionalized B-NCD was first Suzuki coupled with 4-aminophenyl boronic ester to form a layer of (1,1'-biphenyl)-4-amine group on B-NCD. N3 dye was then EDC-NHS coupled to this layer. Surface coverage as estimated by XPS was 3×1014 Ru/cm2.
X-ray Photoelectron Spectroscopy (XPS) is used to follow the surface compositions before and after each functionalization of the three molecular dyes on B-NCD surfaces. Surface coverage of the molecular dyes on B-NCD surfaces is also estimated by XPS. For molecular dye (i) and (ii), the surface coverage was estimated by the S 2p peak intensity. For molecular dye (iii), Ru 3d3/2 was used to estimate the surface coverage. The surface coverage of the molecular dyes on B-NCD obtained with Suzuki Coupling in this project is the highest compared to the reported value in the literature.
Complete device was fabricated using B-NCD that functionalized with molecular dye (ii). The cell performance was measured to be 0.05 %. Two possible reasons might be attributed for the low cell performance. Firstly, the resistivity of the B-NCD electrode used in the time frame of this project is somehow still high compared to the other transparent electrodes like FTO and ITO. Secondly, the surface coverage of the molecular dye on B-NCD is still low if compared to the common electrode used in classical dye-sensitized solar cells, where the amount of light absorbing dyes that are absorbed on mesoporous titania electrode is higher.
Task 3.3. Coupling of graphene layers via molecular wire
We have developed a process to couple graphene via molecular chains (a long molecule) to diamond substrate. The anchoring of graphene using molecular chain improves the chemical robustness and mechanical strength. Even when we use ultrasound wave to agitate the graphene flake, it can withstand shock and does not fragment. In the absence of the molecular chain, graphene sheet will fragment. This suggests that the molecular chain acts like springs and can cushion the mechanical shock. When light is shone onto this platform, the light can be converted to electricity with good efficiency. Onto this platform, dyes can be coupled to make a solar cell.
Task 3.4. Electrochemical and photoelectrochemical characterization of pure and modified nanodiamond-based electrodes
Electrochemical characterization of boron doped diamond electrodes (BDD) was carried out by using standard techniques, like cyclic voltammetry both in pure supporting electrolyte solution and that with model redox probes to test the interfacial charge transfer. These studies were upgraded by the application of electrochemical impedance spectroscopy (EIS), by the in-situ Raman spectroelectrochemistry and by the atomic forace microscopy (AFM) interfaced to electrochemical treatment of the electrode surface. The obtained results provided selfconsistent information about the electrochemical activity of BDD in relation to the doping level and to the quality of the diamond sample, viz. the presence of defects manifested themselves as sp2 impurities. Mott-Schottky plots from EIS allowed accurate monitoring of the p-doping and the effect of surface functionalization by the dyes. The cleaning of the surface by “electrochemical burning” of sp2 carbon at large anodic potentials was observed and characterized by AFM and Raman spectroelectrochemistry. The fabrication of MWCD (molecular wire-coupled dye-sensitized) solar cells enabled a better understanding of interfacial charge transfer at the diamond/electrolyte solution and also in the dye surface modification of the diamond. To test the surface anchoring, ATR-FTIR spectra were measured. The functionalization of O-terminated BDD manifests itself in the region of C-H stretching vibrations, but the remaining vibrations of the used dye (JD-18) were not resolved. Raman in-situ spectroelectrochemical tests showed the high quality of both polycrystalline and single crystalline BDD films and the protective role of dyes wired to the BDD surface. Raman spectroelectrochemistry points to a positive influence of the dye, which serves as a protective layer sealing the BDD surface against the disruptive influence of intercalating ions during the electrode polarization. Such an effect is of a paramount importance for a long term stability of MWCD solar cells. Photoelectrochemical characterization of dye-sensitized BDD were performed in aqueous electrolyte solution containing methyl viologen (MV2+) as the electron carrier, using a three electrode system: the diamond-based molecular platform as working electrode, a platinum wire counter electrode, and a Ag/AgCl reference electrode, under 1 sun irradiation (100 mW/cm2, AM 1.5G). The corresponding tests confirmed both cathodic and anodic photocurrents in the range of μA/cm2 upon illumination with simulated AM 1.5G solar radiation.
Task 3.5 Characterization of chemical and electronic structure of interfaces between conductive nano-diamond and organic molecular dyes
The aim of this task was to aid in the development of high-quality dye-functionalized boron-doped nanocrystalline diamond (BNCD) surfaces, and it was carried out in an iterative fashion involving the synthetic groups and the surface characterization group. The functionalization of the BNCD surfaces involves several steps where first an organic "anchor“ group is chemically attached to the surface whereupon in a second step the desired dye is coupled to the anchor group. Each step involves complicated chemical reactions, who’s efficacy must be evaluated in order to optimize the whole multi-step synthetic procedure and in extension the dye-coverage of the BNCD surface. Over a hundred samples were hence studied in this task by a combined XPS/UPS approach (occupied electronic structure, elemental composition and chemical bonding), and for those samples that past the quality criteria NEXAFS measurements were added as well, to further characterize the unoccupied electronic structure and the dye molecular order (if any) on the surface.
XPS and NEXAFS results showed that anchor groups can efficently be grafted onto the BNCD surface, subseqently serving as templates for the dye bonding. Of the organic molecular dye systems tested, the so-called N3B4 dye proved to be the most successful in terms of yielding dense films when attached to the functionalized BNCD. UPS and NEXAFS measurements showed that the dye electronic structure of the grafted films correctly reproduced that of the "neat" dye films, with slight modifications to to the bonding to the anchor groups, as expected. XPS measurements were used to determine the density of the dye coverage, which was found to be roughly three N3B4 molecules per nm2. The high-quality functionilized films were not achieved in time for the last synchrotron radiation beam time available to the consortium, hence the so-called core hole clock emasurements to probe the charge transfer dynamics at the dye-BNCD interface could not be carried out before the end of the project.
WP4. Efficient implementation of technological platform for solar cell fabrication
Task 4.1. Development of a chemically robust printable graphene inks to be used as counter electrode with low sheet resistance
An opaque counter electrode graphene material with low sheet resistance has been produced, replacing the platinum catalyst typically employed on the counter electrode of a dye-sensitized solar cell. The graphene was intended to be deposited by printing methods, and therefore was formulated into a printable ink. The objective was then to meet ambitious targets for sheet resistance of the deposited graphene layer.
A graphene ink comprising milled graphene, conductive polymer binder and certain additives was formulated and printed by doctor blade onto glass substrates. After pressing, a uniform layer of less than 50 μm was ultimately achieved. Samples of various deposition areas and patterns were produced, demonstrating the versatility of the process. The formulation and pressing process was then continuously improved through the first two years of the project, culiminating in an electrode with sheet resistance of the carbon layer less than < 5 Ohm/sq.
Task 4.2. Printing processes for the MWDC
Following the findings of task 4.1 doctor blade printing and pressing of the graphene ink formulation was developed as the primary method for producing small runs of sample cells. This process is easily transferable
to screen printing or slot-dye coating as required for volume production, with print areas adjustable to meet each partner’s cell configurations.
The development also included printed layers for other components of the prototype cells, enabling simple, low cost production of samples. Sealing for reliability of DSSC is a particular concern in the field, and fully printable low temperature-cured seals were developed for this purpose.
More specialized deposition methods for graphene nano platelets and graphene oxide were also tested, including spin coating and air driven spray coating; the latter was successfully demonstrated on up to 10cm x 10cm areas.
Task 4.3. Prototyping of printable DSC
The partners in WP4 all proceeded to make prototype DSC using the novel materials developed within this work package and also from other work packages, and performed analysis of the prototypes against benchmarks. The benchmarking protocols were developed as part of the task, considering existing test standards for silicon photovoltaics, as well as suitable acceptable and realizable industry and academic tests that could be used to evaluate performance and reliability.
Cells prepared using novel materials from all the work packages have achieved nearly 10% efficiency on 1cm x 1cm area. This is the best result for carbon based DSC that we are aware of, though it falls slightly short of the original MOLESOL target for a cell this size.
The prototypes prepared by SolarPrint, tailored for low light, show actual conversion efficiencies under fluorescent light of 7.6% on a 1cm x 1cm cell using solvent based electrolytes, and 9.7% on a 6cm x 6cm area when using a novel iodine-based ionic liquid. However these ‘low light’ efficiencies are a non-standard measurement outside the standard benchmarking protocol.
Potential Impact:
There is an urgent necessity for reduction of dangerous environmental impact in the energy production & for maintaining a SECURED and LOW COST energy resource supply. To do so, there is a necessity for the development of cheap and environmentally-friendly energy production: Printed efficient large area OPV is one promising candidate: real breakthroughs will come from dedicated material development and innovative device concepts. However in order to be really competitive with existing technologies worldwide and on European level and see a large market development, further improvement on efficiency, stability and cost are needed. For the outdoor use and a real competition with the inorganic solar cells market for photovoltaic solar energy conversion devices, especially the CE and the stability are the current bottlenecks, which current OPVs are not able to meet due to the limitation of organic polymeric systems. Unless this hurdle is lifted, the replacement of environmentally dangerous technologies will be strongly limited, due to costs and performance criteria. To propose alternative pathways and to meet these challenges we suggest developing the MWCD solar cell platform, which extend the possibilities of DSSC technologies. At the same time development of printable graphene and nanodiamond technologies makes an important step towards the effective replacement of current ITO/FTO electrodes, which have dominant position on the photovoltaic market as whole and which resources are limited. These carbon-based transparent and conducive thin films can be used for many already currently developed solar cells platforms and can lead to a true success story in PV industry.
The potential impacts of MOLESOL results are the following:
- Today, with no large market for organic PV existing, the state-of-the-art market requirements for organic PV can be derived on the basis of the current PV market. The market segments are defined focusing on organic photovoltaics as:
Indoor applications / outdoor consumer applications / off-grid applications
PV as building design element / grid-connected bulk power generation
- OPV is still at an early stage, but due to its intrinsic advantages such as large area coverage, lower price level and device flexibility, it is foreseen that OPV will claim a relevant share in the huge upcoming market of PV devices. With further improvement, the MWCD solar cell which combines the advantages of the OPV
and DSSC concept will impact on development of novel highly efficient, stable and cost effective solar cells.
- In a wider sense the novel materials developed in MOLESOL belong to highly attractive category of nanomaterials with some of them - as printable graphenes being hot candidates for becoming a major player in micro and nanoelectronics. MOLESOL has promoted and pioneered the applications of graphene in photovoltaics.
- Carbon materials prepared in MOLESOL can have a large impact on the utilization of low-cost solar cells. Looking to the price/performance ratio, an estimate for ND coating cost is < 50 Euro/m2 using the current MicroWave-Liner Antenna Technology. The graphene-based printing methods are still of lower cost.
- Additionally, individual components of the proposed technology (as nanodiamond or graphene conductive transparent layers) can be used for both OPVs and DSSCs and even inorganic SCs such as aSi hybrid cells etc.
- The novel solar cells developed in that project are completely carbon-based and can therefore be considered as a ‘green product’ with respect to their production, operation and disposal. Fabrication processes via printing in the project is using as much as possible environmentally-friendly solvents for ink formulations - with therefore a positive impact on environment, and on the reduction of energy consumption: moderate temperature, near-atmospheric pressure, low waste volume.
- The future of traditional industries will largely depend on new knowledge, new materials and on new ways of integrating and exploiting existing and new knowledge. The potential of this project is stimulation of a new prosperous primary and secondary industry for new type of photovoltaics in Europe. In particular new markets are envisaged for organic and nano- materials suppliers that are traditionally not in the photovoltaics business. This is also true for industry involved in flat-glass, display, battery or electrical products. The new technologies developed by the project will have a positive impact on the European society on medium to long term with the creation of high-quality jobs by industries.
The dissemination and use of the MOLESOL results are:
MOLESOL is a FET ‘Future Emerging Technologies for Energy Applications” project under FP7. The aim of such high-risk, high reward research projects is to encourage breakthroughs and radical new ideas, since incremental progress in existing solutions may not be sufficient and is also to push the limits of the future emerging technologies. The research scheme is completely "bottom-up" and scientists are encouraged to think "out-of-the-box". The research proposed in MOLESOL is therefore highly novel and ambitious.
Materials research is likely the most important element for the development of the necessary technologies needed to provide a clean, reliable supply of efficient energy and FET projects is encouraging the materials community to work together and share their findings.
Patents evaluation followed during the all project duration revealed to be more complicated and tricky than expected. This due to the intrinsic complexity and novel multi-disciplinary ideas of the projects outputs.
Anyhow several patents that could be in contrast with MOLESOL outputs have been found and they were subjected to consortium partners attention for review to understand their effectiveness relevancy. The greatest part referred to graphene and only very few to nano-diamonds.
At the end of the project and at this stage of development the relevance under a commercial point of view of the MOLESOL know-how seems to be marginal for companies and no patent filling was done yet. Further development is needed to have patent applications.
However the MOLESOL results were largely disseminated by means of the project webpage, several press releases, a newsletter, a public video of 20 minutes about the project visible on YOUTUBE, an international MOLESOL workshop and presented at several international conferences to R&D and industrial public.
List of Websites:
Project partners:
IMEC/Imomec, Belgium; Solarprint Limited, Ireland; Ustav Fyzikalni Chemie J. Heyrovskeho AV CR, v.v.i. Czech Republic; Dyesol Italia, Italy; Max Planck Gesellschaft Zur Foerderung der Wissenschaften E.V. Germany; Linkopings Universiteit, Sweden; Ecole Polytechnique Federale de Lausanne, Switzerland and National University of Singapore, Singapore.
Imec/Imomec is project coordinator.
Project webpage:
More information on the MOLESOL project can be found on www.molesol.eu.
Project press releases and newsletter:
Nanowork: http://www.nanowerk.com/news/newsid=22498.php(s’ouvre dans une nouvelle fenêtre)
InterPV: http://www.interpv.net/head/head_view.asp?idx=503&part_code=03(s’ouvre dans une nouvelle fenêtre)
IMEC e-news: http://imec.fb.email.addemar.com/c1694/e0/h0916f/t2/s0/index.html(s’ouvre dans une nouvelle fenêtre)
Future PV magazine: http://www.futurepv.com/node/401(s’ouvre dans une nouvelle fenêtre)
Project video:
A video which give an nice overview of the project goals and achievements is visible on YOUTUBE by all public.
http://www.youtube.com/watch?v=kMD6uduSfpw&feature=youtu.be(s’ouvre dans une nouvelle fenêtre)
Contact:
Dr laurence LUTSEN
IMEC-Imomec
Phone : +32 (0)11 26 83 14
Fax : +32 (0)11 26 83 01
e-mail : laurence.lutsen@imec.be