Achieving Highly Efficient Nonfullerene Organic Solar Cells with Improved Intermolecular Interaction and Open-Circuit Voltage.

: A new acceptor–donor–acceptor structured nonfullerene acceptor ITCC is Q1 designed and synthesized via simple end-group modiﬁcation. ITCC shows improved electron transport properties and a high-lying lowest unoccupied molecular orbital level. A power conversion efﬁciency of 11.4% with an impressive V OC of over 1 V is recorded in photovoltaic devices, suggesting great potential for applications in tandem organic solar cells.

Please note: It is not permitted to present the PDF file on the internet or on company homepages Special Offer If you order 200 or more reprints you will get a PDF file for half price.

Information regarding VAT
Please note that from German sales tax point of view, the charge for Reprints, Issues or Posters is considered as "supply of goods" and therefore, in general, such delivery is a subject to German sales tax. However, this regulation has no impact on customers located outside of the European Union. Deliveries to customers outside the Community are automatically tax-exempt. Deliveries within the Community to institutional customers outside of Germany are exempted from the German tax (VAT) only if the customer provides the supplier with his/her VAT number. The VAT number (value added tax identification number) is a tax registration number used in the countries of the European Union to identify corporate entities doing business there. It starts with a country code (e.g. FR for France, GB for Great Britain) and follows by numbers.  4  345  395  425  445  548  752  5-8  490  573  608  636  784  1077  9-12  640  739  786  824  1016  1396  13-16  780  900  958  1004  1237  1701  17-20  930  1070  1138  1196  1489  2022   for every additional 4 pages  147  169  175  188  231 315

Special Offer
If you order 200 or more reprints you will get a PDF file for half price.
A new acceptor-donor-acceptor structured nonfullerene acceptor ITCC is designed and synthesized via simple endgroup modification. ITCC shows improved electron transport properties and a high-lying lowest unoccupied molecular orbital level. A power conversion efficiency of 11.4% with an impressive V OC of over 1 V is recorded in photovoltaic devices, suggesting great potential for applications in tandem organic solar cells.

ABSTRACT:
A new acceptor-donor-acceptor structured nonfullerene acceptor ITCC is Q1 designed and synthesized via simple end-group modification. ITCC shows improved electron transport properties and a high-lying lowest unoccupied molecular orbital level. A power conversion efficiency of 11.4% with an impressive V OC of over 1 V is recorded in photovoltaic devices, suggesting great potential for applications in tandem organic solar cells.
Organic solar cells (OSCs) with bulk heterojunction (BHJ) structures consisting of electron-donor and electron-acceptor materials have achieved impressive progress over the past decade, [1,2] demonstrating their great potential in practical applications. An efficient active layer must possess several features, including strong absorption over a broad range of the solar spectrum, proper alignment of molecular energy levels to allow for charge separation, high charge carrier mobility, and favorable blend morphology in the thin film. [3] All of these  aspects play a role in dictating device performance, which is generally characterized by the short-circuit current density (J SC ), open-circuit voltage (V OC ), and fill factor (FF). In the past decades, research has focused on using fullerene derivatives as the electron acceptor in OSC devices. Furthermore, considerable efforts have been devoted to developing efficient molecular design strategies to optimize polymer or small-molecule (SM) electron donors, [4][5][6][7] with power conversion efficiencies (PCEs) of OSC devices based on fullerenes having surpassed 11%. [8] However, continued progress in fullerene-based OSCs is challenging, partly owing to their intrinsically large voltage losses (usually over 0.7 V), [9] defined as the difference between the optical bandgap of the active layer and the output voltage of the device.
The past few years have witnessed the rapid progress of nonfullerene acceptors, and the PCEs of nonfullerene-based OSC devices have been catching up and even surpassed that of their fullerene counterparts. [10][11][12] Compared to fullerenes, nonfullerene acceptors have some advantages such as greater tunability of the absorption spectra and molecular energy levels, potential for lower cost preparation and purification processes. Furthermore, it is worth noting that several studies have shown that the inevitable voltage losses are reduced in nonfullerene based OSCs, [13][14][15] which is beneficial for achieving high V OC devices with outstanding photovoltaic performance. For example, when the electronic structure is tuned such that a V OC reaches 1 V, devices based on fullerene acceptors can only yield relatively low PCEs of around 6%-7% [16,17] whereas nonfullerenebased OSCs can deliver PCEs of around 10%. [12,14] Over the past decade, many nonfullerene electron acceptors based on naphthalene diimide, [18] perylene diimide, [14,[19][20][21][22][23][24][25] or indacenodithiophene (IDT) [13,15,[26][27][28][29] units as core structures have been designed and applied to OSCs. Among these, SMs with an acceptor-donor-acceptor (A-D-A) structure such as ITIC, [27] IEIC, [26] and IDTBR, [30] have shown very good photovoltaic per- formance. For instance, as the first SM acceptor that breaks the 10% efficiency, ITIC has been widely studied and achieved many great successes. Although ITIC can work well with multiple polymer donors, the OSC devices only deliver relatively low V OC (≈0.82-0.94 V) [9,27,31] because ITIC has a relatively low-lying lowest unoccupied molecular orbital (LUMO) level of −3.83 eV. [29] What's more, the electron mobility of ITIC (2.6 × 10 −4 cm 2 V −1 s −1 ) is still lower than of the fullerene-based acceptors, hence further improvement is required for efficient charge transfer in the devices. From the aspect of molecular structure, the highly efficient A-D-A SMs all have large, branched side chains attached to the core units to provide solubility, which impedes the compact stacking of the core moieties of the molecules (Figure 1a). Therefore, the stacking of electron-withdrawing end-groups in this type of acceptor molecules may provide the main electron transport channels, determining overall electron transport properties. What's more, calculations based on density functional theory (DFT) reveal that the electron density distribution of the LUMO extends over the end-groups, but the electron density distribution of the highest occupied molecular orbital (HOMO) resides mostly on the central donor unit. This suggests that the properties of the LUMO could be easily tuned by end-groups modification, making end-group modification of this type of SM acceptor potentially a very important factor that could influence molecular packing and frontier orbital properties.
Here, we report the design and synthesis of a novel nonfullerene acceptor, ITCC (Figure 1b), which is based on introducing thienyl-fused indanone as end-groups on a previously reported SM acceptor ITIC. Due to enhanced intermolecular interactions, ITCC exhibits a closer π -π stacking distance compared to ITIC. Theoretical predictions and experimental results demonstrate that ITCC has improved charge transport prop-erties relative to ITIC. In addition, the LUMO level of ITCC elevated by about 110 meV relative to ITIC. In OSC devices consisting of PBDB-T [32] as the polymer donor, ITCC-based OSC device obtains an impressive V OC of greater than 1 V while ITIC-based device only yields a V OC of 0.93 V. Morphology studies of the blend films imply that purer domains are formed in the ITCC-containing film than that of ITIC-containing film, and this could suppress bimolecular recombination in the devices. This contribution highlights the influence of end-group modification on the properties of an A-D-A SM acceptor. The high PCE of 11.4% suggests that ITCC is a promising acceptor candidate for efficient OSCs.
Since a sulfur atom is more easily polarized compared to a carbon atom due to its more loosely held electrons relative to a benzene ring, the thiophene-containing conjugated molecules usually have stronger intermolecular π -π interactions. [29] Hence, the thienyl-fused indanone end-groups of ITCC are expected to enhance intermolecular π -π interactions and facilitate charge transport compared to ITIC which contains phenyl-fused indanone end-groups. In addition, the more electron-donating nature of thiophene compared to benzene should result in an up-shifted LUMO for ITCC, which contains the thienyl-fused end-group, relative to ITIC. This increased LUMO can also help obtain a higher V OC . Quantum chemical calculations were performed to help understand the rational design of new nonfullerene acceptors. As shown in Figure 1b, the optimized molecular conformations and wavefunction distributions of the frontier orbitals (determined using DFT at a B3LYP level of theory and a 6-31G(d,p) basis set) do not show much difference between ITIC and ITCC. Compared with ITIC, however, the calculated HOMO and LUMO levels of ITCC are simultaneously up-shifted by 34 and 110 meV, respectively, leading to an increased calculated optical bandgap for ITCC. Geometry optimization of isolated ITIC and ITCC molecules was performed by DFT-D (DFT including the D3 version of Grimme's dispersion [33] to accurately describe the weak interactions) and the long alkyl side chains were replaced by methyl groups to reduce the computation time. The optimized geometry of dimer molecules was used to evaluate the intermolecular interactions during intermolecular self-assembly ( Figure S1, Supporting Information), and the intermolecular binding energies of ITIC and ITCC based on optimized dimers were calculated to be −24.06 and −26.65 Kcal mol −1 , respectively, under the correction of basis set superposition error (BSSE). [34] The decreased binding energy of ITCC by 2.59 Kcal mol −1 suggests that ITCC molecules have a stronger tendency for intermolecular stacking, which will enhance its crystalline properties and improve charge carrier mobility.
As presented in Scheme 1, ITCC was prepared from thiophene-2-carboxylic acid via a four-step process. First, acylchlorination of carboxylic acid was conducted using thionyl chloride as a chlorinating agent. After removing the solvent and excess reactant, the resulting compound 2 was used in the next step without further purification. Then, Friedel-Crafts reaction between compound 2 and malonyl dichloride catalyzed by aluminium trichloride yielded compound 3. The key intermediate, compound 4, was obtained by treating compound 3 with malononitrile using dimethylsulfoxide as a solvent at ambient temperature. Finally, the target product of ITCC was prepared by the Knoevenagel Condensation reaction between compound 4 and compound 5 in a high yield of 80% with the assistance of a microwave reactor. ITCC can be easily dissolved in common solvents like dichloromethane, chloroform, and chlorobenzene. The detailed synthetic procedures and characterizations of the intermediates are provided in the Experimental Section of the Supporting Information. The normalized ultraviolet-visible (UV-Vis) absorption spectra of neat ITCC and ITIC thin films are presented in Figure 2a. In chloroform, the main absorption band of ITCC extends from 450 to 700 nm with a maximum absorption peak at 640 nm ( Figure S2, Supporting Information). From solution to thin film, a red-shift by 30 nm was observed with an absorption onset at 742 nm, corresponding to an optical bandgap of 1.67 eV. Compared to ITIC (E opt g = 1.59 eV) thin films, the blueshifted absorption spectrum of ITCC thin films (Figure 2a) can be attributed to the decreased intermolecular charge transfer (ICT) due to the more electron-donating thiophene units relative to benzene units, and less solar photons will be absorbed for the corresponding OSC devices. The frontier orbital levels of ITCC and ITIC were evaluated by cyclic voltammetry (CV) measurements to study the influence of introducing thiophene units, and ferrocene was used as an external standard in parallel. As shown in Figure 2b The crystalline properties of ITIC and ITCC thin films were investigated by grazing-incidence wide-angle X-ray scattering (GIWAXS). As shown in Figure S3 in the Supporting Information, ITCC and ITIC films show (010) diffraction peaks in the out-of-plane (OOP) direction and a relatively uniform (100) signal, implying preferential face-on orientation of the crystallites. It is worth noting that the π -π stacking distance extracted from the peak location decreases slightly to 3.7Å for ITCC from 3.8Å for ITIC. This is in agreement with the enhanced intermolecular interactions predicted by DFT. The tight π -π stacking of ITCC molecules should help obtain a high electron mobility, and this is supported by space-charge-limited current (SCLC) [35] measurements ( Figure S4, Supporting Information) which show the electron mobility of ITCC (9.26 × 10 −4 cm 2 V −1 s −1 ) is much higher than that of ITIC (1.76 × 10 −4 cm 2 V −1 s −1 ), which is favorable for the effective charge transport and extraction in the devices.
To investigate the photovoltaic performance of the newly designed SM acceptor, OSC devices with a conventional configuration of indium tin oxide/poly(3,4ethylenedioxythiophene):poly(styrenesulfonate) (PE-DOT:PSS)/active blend/poly[(9,9-bis(3 ′ -((N,N-dimethyl)-N-ethylammonium)-propyl)-2,7-fluorene)-alt-2,7-fluorene)-alt-2,7-(9,9-dioctylfluorene)]dibromide (PFN-Br) [36] /Al ( Figure  2c) were fabricated, in which the PEDOT:PSS and PFN-Br performed as the anode and cathode electrode buffer layers, respectively. Our previous publications have demonstrated that the polymer donor, PBDB-T [32] (Figure 2c), works well with A-D-A structured nonfullerene acceptors such as ITIC [10] and IT-M. [11] Therefore, we fabricated the OSC devices using PBDB-T as the polymer donor. After device optimization of donor:acceptor (D/A) weight ratio and solvent additive (Table S1, Supporting Information), we found the processing conditions of ITCC-based OSC devices are the same as for ITIC-based devices. As shown in the current density-voltage (J-V) curves (Figure 3a), the PBDB-T:ITCC device shows a maximum PCE of 11.4% with an impressive V OC of 1.01 V, a J SC of 15.9 mA cm −2 and a FF of 0.71. Meanwhile, the OSC device based on PBDB-T:ITIC shows a lower PCE of 10.6% with a V OC of 0.93 V, a J SC of 17.0 mA cm −2 and a FF of 0.67 under the same processing conditions and the same PBDB-T batch, which has some small differences (PCE over 11% could also be achieved with a higher FF but a relatively low V OC ) when inverted device structure was adopted in our previous work. [10] The enhancement of V OC for the ITCC-based devices can be attributed to the up-shifted LUMO level of ITCC, and a high efficiency over 11% with V OC over 1 V has been rarely reported previously. Compared with the ITIC-based device, the slightly decreased J SC of ITCC-based device at least in part due to its narrowed absorption band as discussed below. Then we fabricated ITIC-and ITCC-based OSC devices with different film thicknesses, and the results ( Figure S5, Supporting Information) suggest PBDB-T:ITCC-based devices can keep ≈90% efficiency at a film thickness around 200 nm while the efficiency of PBDB-T:ITIC-based device drops to its ≈80%. [37] The external quantum efficiency (EQE) curves of devices are demonstrated in Figure 3b. Compared to ITIC, ITCC-based device has a narrower photoresponse range. However, as the polymer donor PBDB-T and the small molecule acceptor ITCC all have absorption bands covering 450-750 nm, the device has very high EQE response in this region with a maximum value of 78% at 550 nm. The integrated current densities are 15.2 and 16.3 mA cm −2 for ITCC-and ITIC-based devices, respectively, which are consistent with the J-V measurements.
In order to investigate the origin for the improved FF in the ITCC-based devices, we conducted photocurrent analysis and characterized blend film morphology. Figure 3c shows the photocurrent dependence on effective voltage (V eff ) of the devices, [38] from which the J ph /J sat (J ph is defined as the difference between the light current and dark current, J sat represents the saturation current) ratios under short-circuit conditions and is calculated to be 0.90 and 0.87 for ITCC-and ITIC-based OSC devices, implying the overall exciton dissociation and charge collection processes all are quite efficient. Fitting the relationship of photocurrent and light intensity, [39] as shown in Figure  3d, the power-law exponents of the equation J ph ∝ P s light for the ITCC-and ITIC-based devices are 1.00 and 0.96, respectively, suggesting that the bimolecular recombination is more efficiently suppressed in the ITCC-based device. The hole and electron mobilities of the blends were also measured by using SCLC method ( Figure S6, Supporting Information). PBDB-T:ITCC blend has a higher electron mobility (6.74 × 10 −4 cm 2  V −1 s −1 ) than PBDB-T:ITIC blend (1.28 × 10 −4 cm 2 V −1 s −1 ) while they have similar hole mobilities.
The surface and bulk morphology of the blend films were first studied by atomic force microscopy (AFM) and transmission electron microscopy (TEM). As displayed in Figure 4, the AFM height images reveal that the root-mean-square surface roughness (R q ) of the PBDB-T:ITCC blend film is greater (2.78 nm) compared to PBDB-T:ITIC blends (1.85 nm), and larger aggregation domains were observed in the phase image. The TEM images also confirm the larger domains in the bulk of the PBDB-T:ITCC blend film, which may be attributed to the stronger intermolecular interactions of ITCC and thus stronger aggregation during casting.
In order to gain a deeper understanding about the influence of morphology on the device performance, we conducted GIWAXS and resonant soft X-ray scattering (R-SoXS) measurements on the blend films, from which the detailed morphology information such as molecular packing/orientation, domain length scales, and average composition variation (ACV) in blend films can be extracted. From the 2D-GIWAXS patterns ( Figure S7, Supporting Information), we can find that both of the blend films display strong π -π stacking (010) scattering in