A Twisted Dimeric Perylene Diimide Electron Acceptor for Efficient Organic Solar Cells

Authors


Abstract

image

A novel solution-processable twisted dimeric perylene diimide (IDT-2PDI) with a bulky fused-ring indacenodithiophene as a bridge is developed as an electron acceptor for organic solar cells (OSCs). Poly(3-hexylthiophene) (P3HT):IDT-2PDI based OSCs have power conversion efficiencies (PCEs) up to 2.61% with a high fill factor of 0.668, and benzodithiophene-diketopyrrolopyrrole (BDT-2DPP):IDT-2PDI based OSCs exhibit a promising PCE of 3.12%.

Solution processed bulk heterojunction (BHJ) organic solar cells (OSCs) have attracted considerable attention because they have advantages including being light weight, low cost, and flexible, in addition to the possibility for large area fabrication.[1-6] Nowadays, OSCs based on blends of polymer donors and fullerene acceptors have shown the best performance with power conversion efficiencies (PCEs) that exceed 10%.[7] Compared to the rapid development of new electron donor materials, the development of novel electron acceptors has lagged behind. Fullerenes and their derivatives such as [6,6]-phenyl C61 butyric acid methyl ester (PC61BM) and its C70 homologue (PC71BM) have been the dominant electron acceptor materials in BHJ OSCs because of their high electron mobility, large electron affinity, isotropy of charge transport, and good ability to form favorable nanoscale network with donor materials.[8] However, there remain incentives to develop non-fullerene electron acceptors that will not only retain the favorable properties of fullerenes, but that will also overcome their insufficiencies, which include a weak, narrow absorption in the visible region and limited energy level variability.

Recently, considerable effort has been dedicated to the synthesis of non-fullerene acceptors for OSCs,[9-11] and a popular strategy for tailoring properties of electron acceptors is to introduce electron-withdrawing building blocks exemplified by cyano,[12] imide,[13-25] amide,[26-28] thiadiazole,[29-31] and others.[32, 33] Most non-fullerene acceptors have been investigated by blending with the classical donor poly(3-hexylthiophene) (P3HT), and the highest PCE of P3HT:non-fullerene acceptors based OSCs is 2.9%.[17]

Rylene diimides such as perylene diimide (PDI) and naphthalene diimide (NDI) have good thermal, chemical, and light stability, strong electron-accepting abilities, and high electron mobilities. Rylene diimides and their derivatives are the most promising non-fullerene acceptors. OSCs based on P3HT:rylene diimides blends have exhibited PCEs up to 2.35%.[20] OSCs based on blends of rylene diimide small molecules or polymers and narrow bandgap donors afforded even higher PCEs up to 3–4%.[16, 18, 19, 24, 34-38] However, the parent PDIs possess high planarity and strong intermolecular interaction, and form large crystalline domains and large scale phase separation in BHJ films, leading to reduced exciton diffusion/separation efficiencies and finally low PCEs of the OSCs.[39] Therefore, restricting the crystallinity without adversely weakening charge transport properties of PDIs is a design principle for PDI-based acceptors.

Here, we design, theoretically calculate, and synthesize a twisted PDI dimer (IDT-2PDI, Scheme 1) with a bulky fused-ring indaceno[1,2-b:5,6-b′]dithiophene(IDT) as a bridge. The steric hindrance twists the IDT-2PDI molecular structure with ca. 49.6° dihedral angle between the IDT and PDI planes (Figure S1, Supporting Information). OSCs based on P3HT:IDT-2PDI exhibit PCEs up to 2.61% with high fill factor (FF) of 0.668; and the PCE of 2.61% is higher than the best PCE (2.35%) reported for P3HT:rylene diimides blends and among the highest reported for P3HT:non-fullerene acceptors. Furthermore, IDT-2PDI is blended with the previously reported small molecular donor, benzodithiophene-diketopyrrolopyrrole (BDT-2DPP)[40] (molecular structure see Figure S2, Supporting Information) with complementary absorption spectra and well matched energy levels, and PCEs as high as 3.12% were achieved, which are among the highest reported for solution-processed fullerene-free small molecular OSCs.[41]

Scheme 1.

Route for synthesis of IDT-2PDI.

Compound IDT-2PDI was synthesized in 78% yield via Stille coupling reaction between PDI bromide (1) and IDT ditin (2) using Pd(PPh3)4 as catalyst (Scheme 1) and was fully characterized by matrix-assisted laser desorption ionization time of flight (MALDI-TOF) mass spectroscopy (MS), 1H NMR, 13C NMR, and elemental analysis. The IDT-2PDI product is a mixture of 1,7,1′,7′-, 1,7,1′,6′-, and 1,6,1′,6′-regioisomers, as indicated from the 1H NMR spectrum, and the molar ratio of the 1,7- to 1,6-isomer was 3:1. Due to solubilizing alkyl substituents, this compound is readily soluble in common organic solvents such as dichloromethane and dichlorobenzene at room temperature. The thermal properties of IDT-2PDI were investigated by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). This compound exhibits excellent thermal stability with decomposition temperature (5% weight loss) at 360 °C in nitrogen atmosphere (Figure S3, Supporting Information). The DSC trace for this compound shows no melting peak, and meanwhile the X-ray diffraction (XRD) pattern (Figure S4, Supporting Information) does not show crystalline reflection, suggesting that IDT-2PDI is amorphous.

IDT-2PDI in dichloromethane solution (10−6 M) exhibits strong absorption in 300–800 nm region with a maximum extinction coefficient of 5.1 × 104 M−1 cm−1 at 532 nm (Figure S5, Supporting Information). Relative to its solution, the thin film of IDT-2PDI shows broader absorption with a similar profile, suggesting that there is weak intermolecular interaction and molecular aggregation in the film, due to twisted molecular structure. The optical band gap of IDT-2PDI film estimated from the absorption edge (804 nm) is 1.54 eV (Figure 1a and Figure S5, Supporting Information).

Figure 1.

a) UV-vis absorption spectra and b) energy levels of IDT-2PDI, P3HT, and BDT-2DPP in thin film.

The electrochemical properties of IDT-2PDI were investigated by cyclic voltammetry (CV) method in film on a glassy carbon working electrode in 0.1 M [nBu4N]+[PF6] CH3CN solution at a potential scan rate of 100 mV s−1. IDT-2PDI exhibits quasi-reversible reduction and oxidation waves (Figure S6, Supporting Information). The onset oxidation and reduction potentials versus FeCp2+/0 (0.46 V) are 0.73 and −0.97 V, respectively, thus the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energies (Figure 1b) are estimated to be −5.53 and −3.83 eV from the onset oxidation and reduction potentials, respectively, assuming the absolute energy level of FeCp2+/0 to be 4.8 eV below vacuum.[42] The HOMO and LUMO energies are obviously lower than those (−4.76 and −2.74 eV) of P3HT,[43] respectively. The LUMO gap and HOMO gap between P3HT and IDT-2PDI are large enough to promote efficient exciton dissociation.[44] The difference between the LUMO of IDT-2PDI and the HOMO of P3HT is 0.93 eV, thus the open-circuit voltage (VOC) of P3HT:IDT-2PDI based OSCs can be estimated to be ≈0.63 V.[45]

To demonstrate the potential application of IDT-2PDI in OSCs, we used IDT-2PDI as an electron acceptor and the classical polymer P3HT as an electron donor, and fabricated BHJ OSCs with a structure of indium tin oxide (ITO)/poly(3,4-eth­ylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS)/P3HT:IDT-2PDI/Ca/Al. Table 1 summarizes VOC, short circuit current density (JSC), FF, and PCE of the devices at different donor/acceptor weight ratios. These devices yield VOC of 0.64–0.66 V, which are similar to that (0.63 V) we estimated above and are also almost insensitive to the donor:acceptor weight ratio. As the P3HT content is decreased from 60% to 33% in the BHJ film, the JSC of devices decreases from 1.94 to 0.21 mA cm−2, indicating the photocurrent mainly results from absorption of P3HT. When P3HT weight content is 40% to 50%, the FF of devices is higher. The blend with donor:acceptor weight ratio of 1:1 yields a PCE of 0.43% with a VOC of 0.66 V, JSC of 1.59 mA cm−2 and FF of 0.413.

Table 1. The device data of OSCs based on P3HT:IDT-2PDI under the illumination of AM 1.5G, 100 mW cm−2
P3HT:IDT-2PDI [w/w]VOC [V]JSC [mA cm−2]FFPCE [%]
  1. a)Thermal annealing at 150 °C for 10 min;

  2. b)Solvent annealing.

1.5:10.641.970.2910.37
1:10.661.590.4130.43
1:1.50.650.940.4460.27
1:20.650.210.3420.05
1:1a)0.693.070.5091.08
1:1b)0.705.580.6682.61

Thermal annealing at 150 °C for 10 min improves JSC to 3.07 mA cm−2, FF to 0.509 and PCE to 1.08%, 150% enhancement compared to that for the as-cast device (Table 1, Figure 2a). Furthermore, after solvent annealing (the active layer is dried in o-dichlorobenzene vapor atmosphere and then annealed at 150 °C for 10 min), OSC devices show even better performance: VOC of 0.70 V, JSC of 5.58 mA cm−2, FF of 0.668 and PCE of 2.61% (over six times of that for the as-cast device). The FF of 0.668 and PCE of 2.61% are among the highest values reported for solution-processed OSCs based on P3HT:non-fullerene acceptors.

Figure 2.

a) JV curves and b) IPCE spectra of devices with the structure ITO/PEDOT:PSS/P3HT:IDT-2PDI (1:1, w/w)/Ca/Al.

The incident-photon-to-converted current efficiency (IPCE) of the blend film with donor:acceptor weight ratio of 1:1 without and with post-treatment are shown in Figure 2b. These blend films show broad IPCE spectra from 300 to 800 nm. The maximum peaks at 555 nm and shoulders at ≈600 nm are attributed to P3HT absorption, while the maximum peaks at 404 nm and shoulders at 650–800 nm are attributed to IDT-2PDI absorption. Obviously, P3HT makes more contributions to JSC than the acceptor IDT-2PDI. Thermal annealing and solvent annealing improve the IPCE maximum at 555 nm from 11% to 18% and 36%, respectively.

To understand the effect of post-treatment on device performance, the BHJ active layer was examined using atomic force microscopy (AFM), transmission electron microscopy (TEM), and XRD techniques. The actual surface morphology of blend films of P3HT:IDT-2PDI (1:1, w/w) without and with post-treatment is shown in Figure 3. The film as cast exhibits crystalline grains with a root-mean-square (RMS) roughness of 4.9 nm. After thermal annealing at 150 °C for 10 min, the roughness has a little change (5.0 nm) along with increased size of crystalline grains. The TEM images (Figure S7, Supporting Information) resemble the AFM images; the internal structure of the P3HT:IDT-2PDI blend film after thermal annealing exhibits larger crystalline grains than the as-cast BHJ film. Compared to the as-cast film, the P3HT:IDT-2PDI blend film after thermal annealing shows a stronger reflection peak in the XRD pattern (Figure S8, Supporting Information), which is consistent with the surface and internal morphologies observed using AFM and TEM, respectively.

Figure 3.

AFM height images (5 μm × 5 μm) of P3HT:IDT-2PDI (1:1, w/w) blend film without any post-treatment, by thermal annealing at 150 °C for 10 min, and by solvent annealing (left to right).

The BHJ film after solvent annealing shows even stronger crystalline reflection peak (Figure S8, Supporting Information) than the thermally annealed film. Slow growth of BHJ film leads to several large-sized crystalline domains along with much rougher surface (20 nm RMS roughness, Figure 3). The crystalline grains or domains are a result of P3HT self-organization, which is beneficial to charge transport in the thin film.[46]

To confirm this point and understand the influence of charge carrier transport on photovoltaic performance, the hole and electron mobilities in the BHJ blend film were measured using teh space charge limited current (SCLC) method (Figure S9,S10, Supporting Information). The average hole and electron mobilities for the blend film as cast were found to be 2.0 × 10−4 and 8.5 × 10−5 cm2 V−1 s−1, respectively. After thermal annealing at 150 °C for 10 min or solvent annealing, the BHJ film exhibits enhanced charge transport with the hole mobility of 5.0 × 10−4 or 7.5 × 10−4 cm2 V−1 s−1 and the electron mobility of 4.6 × 10−4 or 3.9 × 10−4 cm2 V−1 s−1, respectively. The enhanced charge transport is beneficial to improvement in JSC of OSCs so that the OSC devices after thermal annealing or solvent annealing yield higher JSC compared to the as-cast devices. The balanced charge transport with electron/hole mobility ratios of 0.4–0.8 leads to the high FF of up to 0.668. Meanwhile, due to better charge transport, OSCs with thermal annealing or solvent annealing have much lower series resistances than the as-cast devices, which can be easily calculated from JV curves (Figure 2a), thus the VOC of annealed devices improves by 0.03–0.04 V relative to that of the as-cast devices.[47]

As shown in Figure 1a,b, absorption spectra of P3HT and IDT-2PDI considerably overlap, and mismatching of the energy levels between P3HT and IDT-2PDI leads to energy loss and relatively low VOC. Our previously reported small molecule BDT-2DPP has complementary absorption and suitable energy levels that are matched with IDT-2PDI (Figure 1), and thus we used IDT-2PDI as an electron acceptor and BDT-2DPP as an electron donor to fabricate BHJ OSCs. Small molecule OSCs based on a BDT-2DPP:IDT-2PDI (1:1, w/w) blend film exhibit a promising PCE of 3.12% with a high VOC of 0.95 V, JSC of 7.75 mA cm−2, and FF of 0.424 (Figure 4a). This blend film shows a broad IPCE spectrum from 300 to 800 nm with peaks at 410 and 620 nm and a shoulder at 680 nm, and the maximum is up to 40% at 620 nm (Figure 4b). The hole and electron mobilities in BDT-2DPP:IDT-2PDI BHJ blend film were measured by SCLC method (Figures S11, Supporting Information). The BHJ film exhibits a hole mobility of 2.0 × 10−5 cm2 V−1 s−1 and an electron mobility of 2.3 × 10−6 cm2 V−1 s−1. Relative to P3HT:IDT-2PDI system, BDT-2DPP:IDT-2PDI blend films exhibit lower hole and electron mobilities as well as unbalanced charge transport, which are responsible for the lower FF value in BDT-2DPP:IDT-2PDI devices.

Figure 4.

a) JV curve and b) IPCE spectrum of devices with the structure ITO/PEDOT:PSS/BDT-2DPP:IDT-2PDI (1:1, w/w)/Ca/Al.

In summary, a novel solution-processable twisted dimeric PDI (IDT-2PDI) with a bulky fused-ring IDT as a bridge was developed as a non-fullerene acceptor for OSCs. IDT-2PDI exhibits broad and strong absorption in visible region and suitable energy levels. P3HT:IDT-2PDI blend films exhibit relatively high hole and electron mobility (10−4 cm2 V−1 s−1) and balanced charge transport. P3HT:IDT-2PDI based OSCs yield PCEs up to 2.61% and FF up to 0.668, both of which are among the highest values reported for P3HT:non-fullerene acceptors. Furthermore, small molecule OSCs based on BDT-2DPP:IDT-2PDI exhibit a promising PCE of 3.12%, which is among the highest reported for solution-processed fullerene-free small molecular OSCs. These preliminary results indicate that IDT-2PDI is a promising small molecular non-fullerene acceptor for OSCs and combining IDT-2PDI with other narrow bandgap donors may produce even better performance.

Experimental Section

Synthesis of IDT-2PDI: Compounds 1 (305 mg, 0.4 mmol), 2 (245 mg, 0.2 mmol), Pd(PPh3)4 (50 mg, 0.043 mmol) and toluene (20 mL) were added to a three-necked round bottom flask. The mixture was deoxygenated with argon for 30 min. The mixture was refluxed for 72 h and then cooled down to room temperature. A saturated KF aqueous solution (5 mL) was added and stirred overnight. Water (25 mL) was added and the mixture was extracted with chloroform (2 × 50 mL). The organic phase was dried over anhydrous MgSO4 and filtered. After removing the solvent from filtrate, the residue was purified by column chromatography on silica gel using petroleum ether/chloroform (1:4) as eluent yielding a black solid (350 mg, 78%). 1H NMR (400 MHz, CDCl3): δ 9.40 (d, J = 8.0 Hz, 2H), 8.64 (m, 3.5H), 8.57 (m, 2.5H), 8.42 (d, J = 8.8 Hz, 2H), 8.35 (d, J = 8.0 Hz, 0.5H), 8.23 (d, J = 7.6 Hz, 1.5H), 7.52 (s, 2H), 7.14 (d, J = 7.2 Hz, 8H), 7.08 (d, J = 7.2 Hz, 8H), 7.02 (s, 2H), 4.50 (m, 4H), 4.15 (m, 8H), 2.53 (m, 8H), 2.04 (m, 4H), 1.97 (m, 4H), 1.64 (m, 16H), 1.29 (m, 52H), 1.07 (t, J = 7.2 Hz, 6H), 0.96 (m, 12H), 0.86 (m, 24H). 13C NMR (100 MHz, CDCl3): δ 163.77, 163.51, 163.36, 157.77, 157.22, 153.44, 146.72, 143.36, 141.81, 141.24, 135.37, 134.99, 134.17, 133.41, 133.26, 132.74, 131.08, 129.92, 129.31, 128.63, 128.43, 128.10, 127.75, 127.59, 127.44, 123.70, 123.53, 122.98, 121.90, 121.55, 121.47, 121.32, 121.04, 70.25, 63.18, 44.28, 38.10, 37.88, 35.58, 31.72, 31.29, 30.90, 30.76, 29.25, 28.82, 28.72, 24.07, 23.16, 23.08, 22.62, 19.53, 14.16, 14.09, 13.86, 10.68. MS (MALDI): m/z 2276 (M+). Anal. calcd. for C152H170N4O10S2: C, 80.17; H, 7.52; N, 2.46. Found: C, 79.53; H, 7.45; N, 2.49%.

Fabrication and Characterization of Photovoltaic Cells: Photovoltaic cells were fabricated with a structure of ITO/PEDOT:PSS/P3HT or BDT-2DPP:IDT-2PDI/Ca/Al. The patterned ITO glass (sheet resistance = 30 Ω −1) was pre-cleaned in an ultrasonic bath of acetone and isopropanol, and treated in an ultraviolet-ozone chamber (Jelight Company, USA) for 30 min. A thin layer (30 nm) of PEDOT:PSS (Baytron P VP AI 4083, Germany) was spin-coated onto the ITO glass and baked at 150 °C for 30 min. An o-dichlorobenzene solution (30 mg mL−1 for P3HT:IDT-2PDI, 35 mg mL−1 for BDT-2DPP:IDT-2PDI) of blend of P3HT or BDT-2DPP:IDT-2PDI was subsequently spin-coated (1000 rpm for P3HT:IDT-2PDI, 2000 rpm for BDT-2DPP:IDT-2PDI) on PEDOT:PSS layer to form a photosensitive layer. Calcium (≈20 nm) and aluminium (≈50 nm) layers were subsequently evaporated onto the surface of the photosensitive layer under vacuum (≈10−5 Pa) to form the negative electrode. The active area of the device was 4 mm2. The JV curves were measured using a computer-controlled Keithley 236 Source Measure Unit. A xenon lamp coupled with AM1.5 solar spectrum filters was used as the light source, and the optical power at the sample was 100 mW cm−2. The IPCE spectrum was measured using a Stanford Research Systems model SR830 DSP lock-in amplifier coupled with a WDG3 monochromator and a 500 W xenon lamp.

Acknowledgements

The authors thank the 973 Program (2013CB834702, 2011CB808401), the NSFC (21025418, 51261130582), and the Chinese Academy of Sciences for financial support. The Supercomputing Center of Chinese Academy of Sciences is acknowledged for molecular modeling.

Ancillary