Effective control of organic-metal interfaces is critical for achieving high-performance polymer solar cells (PSCs). Ideally, the work-function (F) of the cathode and anode should be aligned with the energy of the photo-excited quasi-Fermi levels (EF) of organic semiconductors to create Ohmic contact for maxing achievable open-circuit voltage (Voc) and minimized energy barrier for charge-extraction. Although low Φ metal such as Ca (Φ=2.9 eV) has been proved to form good contact with bulk heterojunction (BHJ) layer as cathode, its vulnerability to environmental conditions undermines its use for practical applications. More stable metals like Al (Φ=4.28 eV) and Ag (Φ=4.57 eV) have been used as cathode, but their relatively high Φ often cause energy mismatch between BHJ blends and themselves, which results in lower Voc and device performance.
To alleviate this problem, proper interfacial engineering by inserting a thin layer between cathode and active layer has been vigorously explored. For example, inorganic materials such as LiF and Cs2CO3 and metal oxides (TiOx, ZnOx), and organic materials such as insulating poly(ethylene oxide) (PEO) and conjugated polyelectrolyte (CPE) have also been proved to be effective in improving Al cathode based device performance. In a recent study, 8.37% of PCE was reported by inserting polyfluorene derivative (PFN) between the high performance PTB7:PC71BM BHJ and Ca/Al. In addition, self-assembled fullerenes (e.g., PCBM capped PEG and fluorocarbon modified PCBM (F-PCBM)) have also been reported to increase P3HT:PCBM based device performance.
Despite that interface engineering has been performed for conventional PSCs, the performances obtained from Ag-based devices were usually lower than those using Ca/Al and Al cathode. This significantly limits the utilization of stable and reflective Ag as cathode for improving performance and stability of devices, though it is well-known Ag anode can be advantageous in inverted PSCs to facilitate the printing process.
On the other hand, fullerene-based materials not only can match well with the energy level of the lowest unoccupied molecular orbital (LUMO) of commonly used acceptor (e.g., PCBM), but also possess sufficiently deep highest occupied molecular orbital (HOMO) energy level, which make them as energetically ideal candidates for electron transport layer (ETL) to facilitate electron-selecting and hole-blocking in PSCs.
Despite the advances in the development of materials to enhance solar cell performance, a need exists to provide effective interfacial materials that are capable of adjusting the Φ of cathode to improve the contact with the BHJ layer, possess reasonable electron mobility to minimize electrical resistance across the interfacial layer, and have sufficient orthogonal solvent-processability and film forming properties to avoid eroding into the BHJ layer. The present invention seeks to fulfill this need and provides further related advantages.
The present invention provides fullerene surfactant compounds that can be incorporated into polymer solar cells as an interfacial layer intermediate the cells' active layer and cathode to enhance solar cell efficiency.
In one aspect the invention provides a fullerene compound, comprising:
(a) a fullerene group;
(b) one or more cationic nitrogen centers covalently coupled to the fullerene group;
(c) one or more hydrophilic groups covalently coupled to the fullerene group; and
(d) one or more counter ions associated with the cationic nitrogen center.
Representative fullerene groups include C60, C70, C76, C78, C82, C84, and C92 fullerene groups. In one embodiment, the fullerene group is a C60 fullerene group. In one embodiment, the cationic nitrogen center is a quaternary amine group. Suitable hydrophilic groups include polyether and polyol groups. In certain embodiments, the polyether group is a polyalkene oxide group such as a polyethylene oxide group having the formula —(CH2CH2O)n—, where n is from 1 to about 20. In certain embodiments, the fullerene compound further includes comprising an anionic center. Representative anionic centers include sulfonate (SO32−) and carboxylate (—CO2−) groups. In one embodiments, the fullerene compound is a mono-fulleropyrrolidium. In other embodiment, the fullerene compound is a bis-fulleropyrrolidium.
In one embodiment, the compound has the structure:
In another embodiment, the compound has the structure:
In these embodiments, F is a fullerene group; B is a N-containing ring having from 5-7 ring atoms; R1 and R2 are independently selected from the group consisting of a polyalkylene oxide and a C1-C20 alkyl optionally substituted with an anionic center; Ar is —C6H5-PEO, wherein —C6H5-PEO is selected from the group consisting of mono-, di-, tri-, and tetra-PEO substituted phenyl; and A− is a counter ion associated with the cationic nitrogen center.
In another aspect of the invention, photovoltaic devices are provided. In certain embodiments, the photovoltaic device includes an interfacial layer intermediate the cathode and active layer, wherein the interfacial layer includes one or more fullerene surfactant compounds of the invention. In one embodiment, the photovoltaic device includes:
(a) a first electrode;
(b) an active layer disposed on a surface of the first electrode;
(c) a layer comprising a fullerene compound of the invention disposed on a surface of the active layer opposite the first electrode; and
(d) a second electrode disposed on a surface of the layer comprising the fullerene compound of the invention opposite the active layer.
In another embodiment, the device further includes a hole transport layer intermediate the first electrode and the active layer.
The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings.
The present invention provides fullerene surfactants and their use to modify the interface of the cathode and bulk heterojunction layer in organic solar cells. The incorporation of an interfacial layer including a fullerene surfactant of the invention in a conventional polymer solar cell enhances the efficiency of the solar cell.
In one aspect, the invention provides a fullerene surfactant. As used herein, the term “fullerene surfactant” refers to a fullerene that includes hydrophilic group sufficient to render the fullerene solution processable in the fabrication of polymer solar cells. The fullerene surfactant includes a fullerene group, one or more cationic amine centers, one or more hydrophilic groups, and one or more counter ions. The cationic amine group is covalently coupled to the fullerene group. The hydrophilic group is covalently the fullerene group. In certain embodiments, the hydrophilic group is covalently coupled to the fullerene group through the cationic amine group. In certain embodiments, the fullerene surfactant further includes an anionic center.
Representative fullerene groups include C60, C70, C76, C78, C82, C84, and C92 fullerene groups. In one embodiment, the fullerene surfactant of the invention includes a C60 group.
The cationic amine group is a positively-charged amine center. Suitable cationic amine groups include quaternary amine groups prepared by quaternizing amine precursor compounds. In certain embodiments, the fullerene surfactants of the invention are prepared by quaternization of precursor fullerene amine compounds.
Representative hydrophilic groups include one or more hydrophilic substituents such as ether and alcohol groups. In certain embodiments, the hydrophilic group is a polyether group. Representative polyether groups include polyalkylene oxides with as polyethylene oxide (PEO) groups, polypropylene oxide (PPO) groups, and groups that include ethylene oxide and propylene oxide groups. Suitable polyethylene oxide groups have the formula —(CH2CH2O)n—, where n is from 1 to about 20, and —(CH(CH3)CH2O)n—, where n is from 1 to about 20. In other embodiments, the hydrophilic group is a polyol.
In embodiments of the fullerene surfactants that include anionic centers, representative anionic centers include sulfonate (SO32−) and carboxylate (—CO2−) groups. The anionic centers are covalently coupled to the fullerene group.
Representative fullerene surfactants of the invention are illustrated in
In one embodiment, the fullerene surfactant compounds of the invention have formula (IA):
In another embodiment, the fullerene surfactant compounds of the invention have formula (IB):
In a further embodiment, the fullerene surfactant compounds of the invention have formula (IIA):
In another embodiment, the fullerene surfactant compounds of the invention have formula (IIB):
In one embodiment, the fullerene surfactant compounds of the invention have formula (III):
In another embodiment, the fullerene surfactant compounds of the invention have formula (IV):
In one embodiment, the fullerene surfactant compounds of the invention have formula (V):
In another embodiment, the fullerene surfactant compounds of the invention have formula (VI):
For fullerene surfactant compounds noted above (i.e., compounds of formula (IA)-(VI)), F is a fullerene group (e.g., C60, C70, C76, C78, C82, C84, and C92); B is a N-containing ring fused to the fullerene group and having from 5-7 ring atoms (e.g., pyrrolidine, a 5-membered ring); R1 and R2 are independently selected from the group consisting of a polyalkylene oxide (e.g., PEO or PPO), as described above, and a C1-C20 alkyl optionally substituted with an anionic center (e.g., sulfonyl or carboxyl); Ar, Ar1, and Ar2 are independently selected from the group consisting of —C6H5-PEO, —C6H5—N+(PEO)2R1|A−; and —C5H4N+—R1|A−, wherein —C6H5-PEO is selected from the group consisting of mono-, di-, tri-, and tetra-PEO substituted phenyl, wherein —C6H5—N+(PEO)2R1 is a substituted aniline, and wherein —C5H4N+—R1 is a substituted pyridinium; L1 and L2 are linkers having from 1 to 20 carbon atoms (e.g., C1-C20 alkylene) optionally including one or more heteroatoms (e.g., O, N, or S) and/or one or more functionalized carbon atoms (e.g., C═O); and A− is a counter ion associated with the cationic nitrogen center.
The preparation of two representative fullerene surfactants of the invention, ETL-1 and ETL-2, is illustrated schematically in
By virtue of its component groups, the fullerene surfactant of the invention is advantageously soluble in a solvent orthogonal to the device active layer. In practice of the method of the invention, device fabrication includes forming a layer intermediate the active layer and cathode. Application of the fullerene surfactant to the active layer provides a fullerene surfactant layer onto which the cathode is formed.
In the devices of the invention, the hole transport and electron transport layers define the charge collection properties in the devices. The best devices reported to date are composed of a layer of polymer donor and fullerene acceptor bulk-heterojunction (BHJ) film sandwiched between a transparent electrode, such as indium tin oxide (ITO), and a metal electrode. Under illumination, photo-generated excitons will dissociate at the donor-acceptor interface, driven by the difference in energy levels between the two semiconductors. The separated charges will then drift under the inherent electric field created by the work-function difference between the asymmetric electrodes and ultimately, will be collected by the corresponding electrodes. The PCE is defined by the product of three parameters including short-circuit current density (Jsc), open-circuit voltage (Voc), and fill factor (FF).
The nature of electrical contact between the active BHJ layer and the electrodes can significantly affect all three device-related parameters and modification of those interfaces by inserting appropriate interfacial layers can significantly alter the contact properties to improve the PCE of OPVs. The interfacial layer of the invention serves multiple functions that include:
(a) tuning the energy level alignment at the electrode/active layer interface;
(b) defining polarity of electrodes and improving charge selectivity;
(c) controlling surface properties to alter the morphology of the active layer;
(d) introducing optical spacer and plasmonic effects to modulate light absorption in the active layer; and
(e) improving interfacial stability between the active layer and electrodes.
The photovoltaic layer (or active layer) can include any one of a variety of materials and mixtures of materials as known in the art. Representative useful materials include P3HT, PIDT-PhanQ, PECz-DTQx, PCDTBT, PDTSTPD, PDTGTPD, PTB7. Representative active fullerene materials include PCBM and ICBA. Other representative active fullerene materials suitable for inclusion in a photoactive layer include those described in U.S. Patent Application Publication No. US 2011/0132439, incorporated herein by reference in its entirety.
The following is a description of representative fullerene surfactants of the invention and their use in interfacial layers to enhance the efficiency of polymer solar cells.
The present invention provides fullerene surfactants, ETL-1 and ETL-2, that can be readily dissolved in alcoholic solvents and applied as interfacial layer for cathode (see
ETL-1 and ETL-2 having compact integration of both ionic moieties and polar ethylene oxide chains onto a C60 core were prepared by quaternizing the tertiary nitrogen of fulleropyrrolidines with methyl iodide (
The energy levels of ETL-1 and ETL-2 were estimated by cyclic voltammogram measurements. As shown in Table 1, the LUMOs of ETLs _exhibit small energy-gradient compared to that of PCBM due to that the electron-deficient cationic nitrogen is in close vicinity of the fullerene core, which made this interfacial material energetically favor electron collection and transport from PCBM to cathode.
aPotential in volt vs. a ferrocene/ferrocenium couple.
bThe LUMO levels were estimated using the following equation: LUMO level = −(4.8 + E1/2red1) eV.
ccorrelated LUMOs according to PCBM standard (LUMO = −4.30 eV).
Both ETL surfactants possess reasonable electron motilities (2.18×10−4 cm2 V−1 s−1 for ETL-1 and 4.91×10−6 cm2 V−1 s−1 for ETL-2), and show negligible absorbance to visible light, which qualify them as proper electron-transporting layer (ca. 10 nm). ETL-1 and ETL-2 bearing cationic nitrogen and PEO linkage effectively up-shifted the Φ of Al and Ag, around 0.8 eV by X-ray photoelectron spectroscopy (XPS) studies. It may be due to the polar interaction between fullerene surfactants and metal facilitate pinning of the metal EF to that of the ETLs upon equilibration, which reduced energy barrier between BHJ layer and cathodes. This, in turn, increases Voc and charge extraction efficiency.
The presence of these fullerene layers creates only minimal energy barrier height for electron extraction from PCBM (due to matched ETL LUMOs to that of PCBM). This is different from using the insulating PEO and p-type CPE process that have unfavored energy level and charge-transporting properties. Moreover, the n-type nature of fullerene surfactant layer creates an extra acceptor-donor junction that can potentially enhance exciton dissociation and prevent cathode from forming direct contact with active layer to quench excitons. These rationale are supported, vide infra, by the enhanced performance of PSCs with spun interfacial layers.
PSCs with fullerene surfactant-modified Al were studied. Device configuration of ITO/PEDOT:PSS/PIDT-PhanQ:PC71BM/ETL/Al (
The values in parentheses were calculated from EQE spectrum.
To further understand the effect on surfactant-modified cathodes, the commonly adopted Ca/Al cathode based device were also studied, in the device configuration of ITO/PEDOT:PSS/PIDT-PhanQ:PC71BM/ETL/Ca/Al. Good contact between Ca/Al cathode and BHJ can give essentially high-performance, 6% PCE of device D (Voc=0.86 V, Jsc=11.08 mA cm−2, and FF=0.63). Slight improvements of device characteristics could be observed when the ETL layer was applied (
The external quantum efficiency (EQE) spectra of devices A-I (
In all devices, ETL-1 show slightly lower PCE and relevant parameters (Jsc and FF) than those of ETL-2, which may be due to the difference of film quality of these two ETLs on top of the BHJ layer. The topography and surface profile of devices with and without ETL layer were characterized by atomic force microscopy (AFM) and is shown in
In one aspect, the invention provides representative fullerene surfactants, ETL-1 and ETL-2, which can be readily processed in orthogonal solvents (e.g., methanol) on a BHJ layer in PSCs. These materials possess proper electron mobility and the capability of tuning cathode Φ to improve electron extraction and photocurrent generation. Upon the insertion of a thin ETL-1 or ETL-2 between various metal cathodes and BHJ layer (device A-I), simultaneously improved Voc, Jsc, and FF could be achieved for these devices compared to those without using surfactant. The performance of PSCs is significantly improved (70% for Al cathode and 40% for Ag cathode) when surfactant-modified cathode was applied. High performance PSCs using fullerene ETL modified Ag cathode were realized (as high as 6.63%) which is superior to those of Ca/Al and Al based devices.
The following is a description of the use of a representative fullerene surfactant of the invention, ETL-2 (“C60-bis”) (
Devices were fabricated with higher WF metals less prone to oxidation, which are shown to perform better than Al devices over time. Remarkably, the VOC appears to be independent of the choice of cathode metal when C60-bis is used as a buffer layer.
To investigate the improvement in JSC, external quantum efficiency (EQE) spectra (
To further demonstrate the utility of C60-bis as an interfacial layer, the PCE of devices with different cathode metals were tracked over a period of time under exposure to ambient conditions.
By far the most obvious benefit of C60-bis is a strongly enhanced VOC. To further investigate the dramatic increase in VOC when C60-bis is used, capacitance-voltage characteristics (C-V) were obtained and devices were analyzed via Mott-Schottky (MS) analysis. It has previously been shown that, due to the intrinsic p-doped nature of semiconducting polymers, a Schottky contact is formed upon deposition of the cathode onto the photoactive layer. The depletion zone formed at this interface is modulated by the applied voltage under reverse and low (<1.5V) forward bias. Band-bending has been shown to result in the vicinity of the cathode, allowing extraction of the built-in potential (VBI) and impurity concentration (N) of the region by application of C−2=(2/qN)(VBI−V) to the appropriate bias voltage range.
The depletion width extracted from the capacitance-voltage data extends over almost the entire thickness of the active layer. When taken with the N values obtained from the same data, this indicates a consistent doping profile across the entire layer that changes negligibly by inclusion of C60-bis. Because the change in the Fermi level of the active layer (EFp) can be approximated by ΔEFp=kbT ln(Nb/Na), where Nb and Na are the dopant concentrations of the device with and without C60-bis, respectively, it is reasonable to conclude that EFp does not change more than ca. 10 meV. When a semiconductor is placed in intimate contact with a metal, their respective EF come into equilibrium by electrons being transferred “downhill” in energy. Referencing VBI to EFp by VBI=(EFp−Φcathode), where Φcathode is the cathode WF, then the difference in VBI with and without C60-bis can be attributed to a modification of Φcathode by the surfactant. Furthermore, because the relative shifts in VBI closely follow those of VOC for all three metals we can conclude that the observed increase in VOC upon inclusion of C60-bis is due to a dipole-induced shift in Φcathode at the interface.
To further investigate the energetics at the interface, WFs were obtained for Al, Ag, and Cu with and without C60-bis spin-coated on top and are summarized in Table 4. WFs of in-situ, sputter-cleaned Al, Ag, and Cu films were measured to be 4.25 eV, 4.57 eV and 4.70 eV, respectively (
It should be stressed that these conditions do not prevail for regular device fabrication since the cathode is deposited under high vacuum after spin-coating the C60-bis layer outside the glovebox. Regardless, at a distance sufficiently far into the bulk of the photoactive layer only the effective WF of the C60-bis modified cathode can be seen by the rest of the device. This ensures a constant difference between EFp and Φcathode, and explains why VBI, and consequently VOC, is nearly the same for all three metals when C60-bis is employed.
A C60 bis-adduct surfactant was used to modify the energy level alignment at the organic/cathode interface in conventional structure, bulk-heterojunction OSC devices. A well-defined interface between the photoactive layer and the surfactant was ensured by virtue of process solvent orthogonality. The large increase in device VOC is independent of the choice of cathode metal due to pinning of the metal EF to that of the C60-bis upon equilibration. Mott-Schottky analysis of the interface formed between the photoactive layer and the cathode yields a built-in potential defined by the difference between the Fermi level of the bulk-heterojunction EF and the effective cathode work function Φcathode. The observed changes in VBI are reflected in the magnitude of the change in VOC. Further, EQE data reveal the overall device performance enhancement to be due entirely to the inclusion of the surfactant, rather than a beneficial change in photoactive layer morphology.
The following examples are provided for the purpose of illustrating, not limiting, the invention.
In this example, the preparation, characterization, and use of representative fullerene surfactants, ETL-1 and ETL-2, is described. The fabrication and characterization of devices that include the surfactants is also described.
All reactions dealing with air- or moisture-sensitive compounds were carried out using standard Schlenk technique. All 1H (500 MHz) and 13C (125 MHz) spectra were recorded on Bruker AV500 spectrometers. Spectra were reported in parts per million from internal tetramethylsilane (δ 0.00 ppm) or residual protons of the deuterated solvent for 1H NMR and from solvent carbon (e.g., δ 77.00 ppm for chloroform) for 13C NMR. The matrix for MALDI-TOF-MS used 2:1 mixture of alpha-cyano-4-hydroxycinnamic acid (CHCA)/2,5-dihydroxybenzoic acid (DHB) in acetonitrile. Elemental analyses were performed by QTI, Whitehouse, N.J. (www.qtionline.com). AFM images under tapping mode were taken on a Veeco multimode AFM with a Nanoscope III controller. 2,3,4-Tris(2-(2-methoxyethoxy)ethoxy)benzaldehyde and fulleropyrrolidines were synthesized according to literature methods (Benzaldehyde: Nielsen, C B.; Johnsen, M.; Arnbjerg, J.; Pittelkow M.; Mclroy, S P.; Ogilby, P R.; Jrgensen, M. J Org. Chem. 2005, 70:7065. Fulleropyrrolidines and fulleropyrrolidiums: Bosi, S.; Feruglio, L.; Milic, D.; Prato, M. Eur. J. Org. Chem. 2003, 4741). C60 was purchased from American Dye Source. Unless otherwise noted, materials were purchased from Aldrich Inc., and used after appropriate purification.
Synthesis of Fulleropyrrolidiums
A solution of C60 (300 mg, 0.35 mmol), 2,3,4-tris(2-(2-methoxyethoxy)ethoxy)benzaldehyde (478 mg, 1.04 mmol) and sarcosinic acid (111 mg, 1.25 mmol) in chlorobenzene (100 mL) was refluxed under N2 for 4 h. After evaporation of the solvent, the residue was subjected to chromatograph purification on a silica gel column. Elution with toluene gave little unchanged C60. Fraction containing mono adduct was collected with PhMe/EtOAc (1:2) eluent. One fraction of bisadducts consisting mixture of regioisomers was then collected with EtOAc eluent. Each sample was precipitated from toluene solution with methanol or hexane, and gave monofulleropyrrolidine (115 mg, 27%), bisfulleropyrrolidine (90 mg, 15%).
Quaternization of neutral fulleropyrrolidines was achieved by heating a solution of mono or bis fulleropyrrolidine (0.05 mmol) in chloroform (2 mL) and MeI (1.5 mL) in a screw-topped Schlenk tube under N2. Reaction mixture was kept at 80 OC for 40 h. After evaporation of the solvent, the product was dissolve in chloroform and precipitated with hexane. After thoroughly washed with n-hexane, black fulleropyrrolidiums, ETL-1 or ETL-2, were obtained in quantitative yield.
Monofulleropyrrolidine.
1H NMR (500 MHz, CDCl3): δ 2.78 (s, 3H, NCH3), 3.34 (s, 3H, OCH3), 3.36 (s, 3H, OCH3), 3.40 (s, 3H, OCH3), 3.48-3.50 (m, 2H, OCH2), 3.53-3.58 (m, 4H, OCH2), 3.63-3.80 (m, 10H, OCH2), 3.86 (t, J=5.5 Hz, 2H, OCH2), 3.05-4.18 (m, 4H, OCH2), 4.27-4.32 (m, 2H, OCH2), 4.37-4.40 (m, 2H, OCH2), 4.94 (d, J=9.5 Hz, 1H, NCH2), 5.56 (s, 1H, NCH2), 6.77 (d, J=8.5 Hz, 1H, Ar—H), 7.63 (d, J=8.5 Hz, 1H, Ar—H). 13C NMR (125 MHz, CDCl3): δ 39.89, 58.99, 59.00, 59.03, 59.05, 59.18, 59.19, 68.36, 69.25, 69.76, 69.83, 70.21, 70.35, 70.62, 70.72, 70.74, 71.95, 71.98, 72.05, 72.23, 73.18, 75.74, 77.20, 109.34, 123.31, 124.57, 134.82, 136.06, 136.49, 136.59, 139.47, 139.53, 140.12, 140.14, 141.19, 141.58, 141.67, 141.89, 141.99, 142.08, 142.11, 142.16, 142.28, 142.29, 142.54, 142.57, 142.63, 142.65, 143.00, 143.08, 144.36, 144.45, 144.61, 145.11, 145.12, 145.18, 145.23, 145.26, 145.31, 145.55, 145.76, 145.94, 146.07, 146.09, 146.12, 146.20, 146.27, 146.77, 146.95, 147.30, 152.20, 152.47, 154.15, 154.33, 155.16, 156.87. MALDI-TOF-MS (+): calcd. for [C84H41NO9]−, 1208.225. found. [M]−, 1207.893.
Bisfulleropyrrolidine.
1H NMR (500 MHz, CDCl3): δ 2.55-2.88 (m, NCH3), 3.29-3.40 (m, OCH3), 3.42-4.00 (m, OCH2&OCH3), 4.06-4.68, 4.92-5.57, 5.74-5.75, 6.52-6.98, 7.35-7.49, 7.59-7.69, 7.73-7.88, 8.00-8.03; 13C NMR (125 MHz, CDCl3): δ 39.66-39.83 (m, NCH3), 53.21-53.43 (m), 58.97-59.22 (m), 68.17-68.43, 69.42, 69.70-69.94, 70.14-70.91, 71.90-72.35, 72.92-73.50, 75.34-76.00, 77.40-77.66, 109.12-109.48, 123.58-123.98, 124.42-124.61, 134.87, 136.53, 139.39, 140.76-141.93, 142.14, 142.00, 142.23, 142.30, 142.37, 142.51, 142.58, 142.95, 142.97, 143.38, 143.39, 143.58, 144.12, 144.36, 144.85, 144.96, 145.08, 145.21, 145.26, 145.44-145.74, 146.05, 146.07, 147.25, 147.47, 147.72, 147.84, 148.64, 148.77, 149.03, 150.75-151.39, 151.97-152.83, 153.66, 154.28-154.98, 155.54; MALDI-TOF-MS (+): calcd. for [C108H82N2O18], 1695.809. found. [M-I]+, 1695.929.
Fulleropyrrolidium ETL-1.
1H NMR (500 MHz, CDCl3): δ 3.36 (s, 3H, OCH3), 3.38 (s, 3H, OCH3), 3.52-3.56 (m, 7H, OCH2&OCH3), 3.67-3.77 (m, 8H, OCH2), 3.80-3.84 (m, 2H, OCH2), 3.89 (m, 2H, OCH2), 3.97 (s, 3H, NCH3), 4.02 (d, J=8.5 Hz, 2H, OCH2), 4.20-4.39 (m, 4H, OCH2), 4.48 (s, 3H, NCH3), 4.66-4.68 (m, 2H, OCH2), 5.80 (d, J=12.5 Hz, 1H, NCH2), 6.84 (d, J=13.0 Hz, 1H, NCH2), 6.88 (d, J=9.0 Hz, 1H, Ar—H), 7.28 (d, J=13.0 Hz, 1H, NCH2), 7.71 (d, J=8.5 Hz, 1H, Ar—H); 13C NMR (125 MHz, CDCl3): δ 45.69, 53.44, 59.04, 59.07, 59.08, 59.28, 67.89, 68.44, 69.38, 69.96, 70.43, 70.54, 70.69, 70.72, 71.42, 71.65, 71.93, 71.97, 72.53, 72.56, 73.13, 73.64, 78.60, 108.66, 111.57, 127.48, 134.10, 134.75, 135.52, 136.11, 139.03, 139.87, 139.98, 140.26, 140.93, 141.24, 141.38, 141.43, 141.45, 141.62, 141.82, 142.03, 142.09, 142.11, 142.13, 142.35, 142.39, 142.51, 142.52, 142.76, 142.84, 142.96, 143.01, 143.12, 143.33, 144.19, 144.23, 144.36, 144.42, 144.82, 144.89, 145.14, 145.26, 145.30, 145.45, 145.54, 145.61, 145.66, 145.77, 145.82, 145.96, 146.02, 146.13, 146.18, 146.35, 146.40, 147.42, 147.56, 149.32, 150.51, 151.18, 152.66, 153.66, 153.83, 155.77; MALDI-TOF-MS (+): calcd. for [C85H44NO9]+.I−, 1350.16. found. [M-I]+, 1222.144; Anal. Calcd for C85H44NO9: C, 75.61; H, 3.28; N, 1.04. Found: C, 73.29; H, 2.76; N, 0.76.
Fulleropyrrolidium ETL-2 (Mixture of Regioisomers).
1H NMR (500 MHz, CDCl3/CD3OD): δ 3.32-3.40 (m, OCH3), 3.42-4.03 (m, OCH2&OCH3), 4.12-4.50 (m, OCH2), 4.56-4.61 (m, OCH2), 4.70-4.72 (m, OCH2), 5.36-5.69 (m, NCH2), 6.02-6.07, 6.68-6.97, 7.04-7.13, 7.20-7.21, 7.32-7.34, 7.37-7.61, 7.77-7.89, 7.98-8.04, 8.10-8.14, 8.27-8.31; 13C NMR (125 MHz, CDCl3): δ 45.28-46.47 (m), 53.21-53.43 (m), 58.99-59.49 (m), 66.06, 66.83, 68.45, 68.49-69.41, 69.51-70.89, 71.27-71.52, 71.96-72.06, 72.46-72.67, 73.55-73.84, 78.65, 78.75, 109.24, 109.35, 109.48, 111.29, 111.44, 136.24, 136.67, 140.04, 140.44, 140.84, 140.96, 141.13, 141.35, 141.58, 141.60, 141.67, 141.73, 141.77, 141.81, 141.83, 141.94, 142.14, 142.17, 142.21, 142.32, 142.38, 142.40, 142.51, 142.61, 145.38, 145.48, 145.59, 146.14, 146.20, 147.21, 147.40, 147.53, 147.79, 147.96-148.09, 148.40, 148.70-148.82, 149.08-149.32, 150.06, 151.57, 153.67-153.76, 155.68-155.89; MALDI-TOF-MS: calcd. for [C110H88N2O18]2+.2I−, 1979.69. found [M-2I-NMe3]+ 1666.278; Anal. Calcd for C110H88I2N2O18: C, 66.74; H, 4.48; N, 1.42. Found: C, 66.07; H, 4.23; N, 1.35.
CV Measurements
Cyclic voltammetry (CV) measurements were carried out in a one-compartment cell under N2, equipped with a glassy-carbon working electrode, a platinum wire counter electrode, and an Ag/Ag+ reference electrode. Measurements were performed in THF solution containing tetrabutylammonium hexafluorophosphate (0.1 M) as a supporting electrolyte with a scan rate of 100 mV/s. All potentials were corrected against Fc/Fc+. Due to close vicinity of the electron-deficient cationic nitrogen to the fullerene core, the LUMO level of ETL-1 to that of ETL-2 has a small difference in 0.03 eV.
Fabrication and Characterization of PSCs
[6,6]-Phenyl-C61 (or C71)-butyric methyl ester was purchased from American Dye Source. PEDOT:PSS (Baytron P VP AI 4083) was purchase from H. C. Stark. Materials were used as received. The fullerene surfactant solutions in methanol were sonicated for 2 hrs prior to spin-coating in ambient at 5000 RPM. The surfactant layer thickness was about 8-10 nm as measured by AFM. The ITO substrates were cleaned by ultrasonication in acetone for 15 min, followed by manual scrubbing with detergent and deionized water, then sonication in deionized water and isopropanol for 15 min each. The substrates were blown dry under a nitrogen stream and immediately exposed to air plasma for 20 seconds. A 40 nm thick layer of PEDOT:PSS was spin coated onto each substrate and subsequently annealed in air at 140° C. for 30 min. The mixture of PIDT-PhanQ:PC71BM in o-dichlorobenzene (20 mg/ml, 1:3, w:w) was then spin-coated on the PEDOT:PSS layers at 800 RPM, and subsequently annealed at 110° C. for 10 min under nitrogen atmosphere to obtain a film thickness approximately 80 nm. After fullerene surfactant solutions was spin coated on the BHJ layer. The substrates were then transferred back into the glovebox and annealed at 110° C. for 5 min. Finally, aluminum (100 nm) or calcium (30 nm) topped with aluminum (100 nm), or silver (100 nm) was thermally evaporated onto the active layer through shadow masks.
Photocurrent-voltage (J-V) measurements were performed using a Keithley 4200 in a nitrogen-filled glove box under AM1.5 illumination conditions at intensity of 100 mW/cm2. A NREL certified silicon photodiode with a KG5 filter was used to calibrate. Device EQE spectra were obtained in air by comparison to a known AM1.5 reference spectrum for a calibrated silicon photodiode.
Organic Field-Effect Transistors
Top contact OFETs were fabricated as typical top contact, bottom gate devices on silicon substrates. Heavily doped p-type silicon <100> substrates from Montco Silicon Technologies INC. with a 300 nm (±5 nm) thermal oxide layer acted as a common gate with a dielectric layer. After cleaning the substrate by sequential ultrasonication in acetone, methanol, and isopropyl alcohol for 15 min flowed by air plasma treatment, the different fullerene surfactant films were spin-coated from a 0.5 wt % chloroform solution in ambient. Interdigitated source and drain electrodes (W=1000 μm, L=12 μm) were defined by evaporating a 50 nm Au film through a shadow mask from the resistively heated Mo boat at 10−6 Torr. OFET characterization was carried out in a N2-filled glovebox using an Agilent 4155B semiconductor parameter S6 analyzer. The field-effect mobility was calculated in the saturation regime from the linear fit of (Ids)1/2 VS Vgs. The threshold voltage (Vt) was estimated as the x intercept of the linear section of the plot of (Ids)1/2 VS Vgs. The sub threshold swing was calculated by taking the inverse of the slope of Ids VS Vgs in the region of exponential current increase.
Work Function Measurements by XPS
Samples for work function analysis were prepared on glass substrates coated with ITO to ensure good electrical contact. Work functions were measured with a PHI Versa Probe X-ray photoelectron spectrometer (ULVAC-PHI, Kanagawa, Japan) employing a monochromatic focused Al—Kα X-ray source and hemispherical analyzer. The Au 4f7/2 (84.00 eV) and Cu 2p3/2 (932.66 eV) photoemission peaks were used to calibrate the binding energy scale. A bias voltage (−5 V) was applied to the sample, and the location of the secondary electron cut-off was determined at normal emission by a linear extrapolation to the background level. To account for the instrument width, 0.14 eV were added to the work function values thus obtained. This procedure gives a work function for argon ion sputtered gold foil of 5.17 eV.
In this example, the preparation and characterization of representative photovoltaic devices with a fullerene surfactant-containing layer intermediate the active layer and cathode is described.
Fabrication of Photovoltaic Devices
ITO-coated glass substrates (15 Ωsq−1) were cleaned sequentially by sonication in detergent and deionized water, acetone and isopropanol. After drying under a N2 stream, substrates were air-plasma treated for 30 s. A about 35 nm layer of PEDOT:PSS (Baytron® P VP Al 4083, filtered through a 0.45 μm nylon filter) was spin-coated onto the clean substrates at 5 kRPM and annealed at 140° C. for 10 min. The substrates were transferred to a N2-filled glovebox where a homogeneously blended solution of PIDTPhanQ:PC71BM (40 mg/ml in o-dichlorobenzene stirred overnight in glovebox, 1:3 polymer:fullerene by weight) was spin-coated at 2 k RPM, producing an active layer about 100 nm thick, and annealed at 110° C. for 10 min. Substrates requiring a layer of fullerene surfactant were briefly transferred out of the glovebox (total ambient exposure<10 min) and about 2-5 nm thick film of C60-bis surfactant (1 mg/ml in methanol) was spin-coated at 5 k RPM. The substrates were then transferred back into the glovebox and annealed at 110° C. for 5 min to drive off any remaining solvent prior to metal deposition. Metal electrodes were deposited at a base pressure<1×10−6 Torr through a shadow mask, defining an active device area of 4.64 mm2. Ag and Cu were deposited at a rate of 1 Å s1 and Al was deposited at a rate of 4 Å s−1.
Preparation of XPS Samples
ITO-coated glass substrates were prepared as above without air-plasma treatment. Al, Ag, and Cu were deposited over the entire substrate surface at a rate of 1 Å s−1. Substrates requiring a thin layer of fullerene surfactant were transferred out of the glovebox and a solution of C60-bis surfactant was spin-coated from methanol using the same conditions as above. After transfer back into the glovebox, all substrates were heated at 70° C. for 5 min to evaporate any remaining methanol prior to being sealed with parafilm in 20 ml glass vials under N2 for transport to the XPS.
Measurement and Characterization
J-V characteristics of the unencapsulated devices were measured in ambient conditions using a Keithley 2400 source meter under AM 1.5 G (100 mW cm−2) irradiation simulated by an Oriel xenon lamp (450 W). AM 1.5 G illumination was confirmed by means of calibration to a standard silicon photodiode (Hammamatsu) which can be traced to the National Renewable Energy Laboratory. External quantum efficiency spectra were obtained by measuring the photocurrent response of the device using chopped, monochromated light from the same xenon lamp in conjunction with a Stanford Research Systems SR830 lock-in amplifier under ambient conditions. Mott-Schottky analysis was performed in a N2-filled glovebox in the dark using a Signatone probe station interfaced with a Hewlett-Packard HP4284A LCR meter. The 1 kHz AC field applied during measurement was kept at an amplitude of 25 mV to maintain response linearity. Capacitance-voltage characteristics measured thusly were obtained using devices prepared as above with an active area of 10.08 mm2. Work function determination via XPS is described below. Briefly, the secondary electron cutoff (SEC) spectrum of each sample was measured under ultra-high vacuum (<5×10−9 Torr) using a PHI 5000 VersaProbe (Ulvac-Phi, Inc.) employing a focused, monochromated Al K-α x-ray source and a hemispherical analyzer. Proper referencing of the SEC edge to that of Ar+ ion sputter-cleaned, polycrystalline gold allowed for accurate determination of the sample work functions with a reproducibility of about 0.05 eV.
Cyclic Voltammetry Measurements
Cyclic voltammetry measurements were carried out under N2 in a one-compartment cell equipped with a glassy carbon working electrode, a platinum wire counter electrode, and an Ag/Ag+ reference electrode. Measurements were performed in THF solution containing tetrabutylammonium hexafluorophosphate (0.1 M) as a supporting electrolyte with a scan rate of 100 mV/s. All potentials were corrected against the Fc/Fc+ couple and LUMO levels were estimated using the following equation: LUMO=−(4.8+E1/2red1) eV.
Work Function Determination
Work function values were obtained following a modified method previously described (M. M. Beerbom et al., Journal of Electron Spectroscopy and Related Phenomena 152, 2006, 12-17). The spectrometer's analyzer was calibrated according to the manufacturer's guidelines to yield photoemission lines of Ar+ ion sputter-cleaned Cu and Au foils for Cu 2p 3/2 and Au 4f 7/2 at 932.62 eV and 83.96 eV, respectively, following ISO 15472 (M. P. Seah, Surf. Interface Anal., 31, 2001, 721-723). This procedure ensures the linearity of the binding energy scale for the instrument, extrapolated out to the secondary electron cutoff (SEC) near the photon energy of the system (1486.6 eV for monochromated Al K-α x-rays). SEC spectra were measured at an x-ray power of 25 W and 15 kV acceleration at normal emission. For all SEC spectra a bias of −15V was applied during measurement to ensure sufficient separation of the sample SEC and that of the analyzer. Under these conditions a SEC value of 1466.24 eV for clean, polycrystalline gold was obtained, corresponding to a work function of 5.36 eV. Because the Cu and Au core level spectra mentioned above are referenced to the Fermi level, set at zero binding energy, the work function of Au was obtained by ΦAu=(hυ−qVapp−ESEC) where hυ is the x-ray photon energy, Vapp is the applied bias and ESEC is the position of the secondary electron cutoff on the binding energy scale. Ideally, the SEC edge should be a step function at 0 K, however experimental conditions include thermal and instrumental broadening. Hence, the position of the SEC is taken as the local maximum of the first derivative of the SEC feature.
While illustrative embodiments have been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention.
This application is a continuation of U.S. application Ser. No. 13/706,230, filed Dec. 5, 2012, which claims the benefit of U.S. Application No. 61/566,943, filed Dec. 5, 2011, each application is expressly incorporated herein by reference in its entirety.
This invention was made with Government support under Contract Nos. FA2386-11-1-4072 awarded by the Air Force Office of Scientific Research, N00014-11-1-0300 awarded by the Office of Naval Research, and DE-FC3608GO18024/A000 awarded by the Department of Energy. The Government has certain rights in the invention.
Number | Date | Country | |
---|---|---|---|
61566943 | Dec 2011 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 13706230 | Dec 2012 | US |
Child | 14948027 | US |