Introduction

During the past decades, electrode buffer layers have played a vital role in promoting the performance of organic optoelectronic devices, in particular for organic photovoltaic devices (as shown in Figure 1a), which consist of electrodes (anode and cathode), photoactive layers and electrode buffer layers (EBLs)1,2,3,4. Both inorganic and organic materials can be used for making the EBLs. Recently, important progresses have been achieved in developing new n-type organic EBLs and the power conversion efficiencies (PCEs) of polymer solar cells (PSCs) over 8% have been achieved in various groups via polymer interlayers such as amino-substituted polyfluorene derivatives (PFN)5,6,7,8,9, ethoxylated polyethyleneimine (PEIE)10, fullerene derivatives11,12. Compared to the rapid progresses in n-type organic EBL materials, the development of solution processable p-type organic materials is comparatively lagged behind. Although, the applications of a few p-type organic materials in PSCs were reported in recent works13,14,15,16, the classical p-type EBL material, poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT: PSS), is predominantly used as the p-type EBL material in PSCs and also the highest PCE values in PSCs were obtained by utilizing PEDOT: PSS as EBLs17. To explore a new solution processable organic p-type EBL material with superior properties is still an interesting and important topic for the field of PSCs.

Figure 1
figure 1

(a) Device diagram of PSCs and the chemical structure of the materials involved in the current work: hydrochloric acid doped polyaniline (HAPAN) as well as the model polymer (PBDTTT-EFT); (b) Energy level diagram of the state-of-the-art photovoltaic materials.

Full size image

A superior p-type organic EBL material should meet several requirements: (i) Processing solvent of the buffer layer should be orthogonal with that of the active layer. (ii) As known, the state-of-the-art donor polymers often exhibited HOMO levels of 5.0 ~ 5.4 eV (see Figure 1b)18,19,20,21,22,23,24,25,26,27 and thereby the work function (WF) of the p-type EBL material should exceed 5.0 eV to form ohmic contact with the active layers. (iii) In PSCs, the EBL should have very high transparency and thus the active layer can utilize more sunlight. Usually, more than 5% of the incident light was loss in the 35 nm PEDOT:PSS layer. To minimize the light loss, two strategies can be considered: the first choice is to explore new materials with no or extremely low absorption in visible region; an alternative strategy is to utilize ultrathin buffer layers. For example, Heeger et al.16 reported a novel p-type conjugated polyelectrolyte, which can fulfill all these above requirements simultaneously and thus exhibited comparable performance with PEDOT:PSS in PSCs.

As a type of classical conducting polymers, polyanilines have been widely studied in the past decades due to their high conductivities, environmental stabilities and ease of synthesis for multiple energy-related applications such as sensors, supercapacitors and organic light-emitting diodes28,29,30,31. Polyanilines with good solution processibilities in commonly used solvents can be prepared through internal or external doping. For instance, when doped by sulfonic acid, polyanilines can be dissolved into water32 and the water soluble polyanilines can be used as the p-type EBL materials in organic solar cells33. A few of acid doped polyanilines like acid-doped polyaniline nanotubes (a-PANIN), self-doped sulfonic acid polyaniline (SPAN) and graft copolymers (PAN-PEG) have been explored in P3HT:PCBM based PSCs and achieved relatively poor or comparable performance with PEDOT: PSS34,35,36. Overall, although polyaniline based p-type EBLs have been utilized in classical PSCs, their applications and underlying potentials in highly efficient PSCs based on novel photovoltaic polymers have not been intensively investigated yet.

Herein, a water-soluble hydrochloric acid doped polyanilines (named as HAPAN in Figure 1a) were employed as the p-type EBL material in highly efficient PSCs based on Poly(2-ethylhexyl 6-(4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b:4,5-b′]dithiophen-2-yl)-3-fluorothieno[3,4-b]thiophene-2-carboxylate) (designated as PBDTTT-EFT in Figure 1a)11. Notably, when the blend of PBDTTT-EFT:PC71BM was used as the active layer, the PSC with only 5.0 nm HAPAN film as the p-type EBL exhibited excellent photovoltaic performance with PCE exceeding 9%. Interestingly, the HAPAN-based PSC still achieved a high PCE up to 8.94% when the thickness of HAPAN film is reduced to only 1.3 nm under the illumination of AM 1.5G 100 mW/cm2.

Results

HAPAN is prepared by a cost-effective method, namely a direct sulfonation of emeraldine salts with chlorosulfonic acid in dichloroethane at 80°C and subsequently hydrated in water at 100°C and the synthesis details have been reported elsewhere37. The FT-IR spectrum and XPS spectra of HAPAN are provided (see Figure S1 and Table S1 in supporting information). HAPAN exhibits good solubility (up to 30 mg/ml) in water while insoluble in o-DCB. Under the concentration of 1 mg/ml, the corresponding pH of HAPAN solution is 4.2, which is slightly higher than that of the well-known PEDOT:PSS (VP AI 4083 from H. C. Stark, PEDOT4083).

Initially, the work functions (WF) of HAPAN films were characterized by ultraviolet photoelectron spectroscopy (UPS). Figure 2a presented the UPS results of the pristine ITO and HAPAN films with different thickness (1.3 nm, 5.0 nm and 30 nm) covered indium-tin-oxide (ITO) samples. The thicknesses of HAPAN were measured and controlled by spectroscopic ellipsometer and surface profilometer (see Figure S2a), meanwhile, the photographs of HAPAN films with different thickness were depicted in Figure S2b. The WF value of ITO measured by UPS was consistent with the reported results35. The WF of 30 nm HAPAN film was determined to be 5.12 eV, which is similar as the WF (5.16 eV) of PEDOT:PSS35. It is interesting to note that the ultrathin HAPAN film (5.0 nm) exhibited the same WF value of 5.12 eV as that of the thicker HAPAN film (30 nm). The results demonstrated that the WF of ITO can be well modified by coating 5.0 nm HAPAN buffer layer. As the thickness of HAPAN film reduced to 1.3 nm, the WF was 4.75 eV and close to that of the bare ITO film (4.72 eV).

Figure 2
figure 2

(a) UPS spectra of indium-tin oxide (ITO), PEDOT: PSS (35 nm), different thickness of HAPAN (1.3 nm, 5.0 nm and 30 nm) film on ITO substrate, respectively. (b) UV-Vis absorption spectra of the different thickness of HAPAN layers (1.3 nm, 5.0 nm, 10 nm, 30 nm) on the quartz glass. (c) AFM height images of bare ITO film, 5.0 nm HAPAN-coated ITO film and 1.3 nm HAPAN-coated ITO film.

Full size image

The transmittance of the EBL is an important parameter that affects the photovoltaic performance especially short-circuit current density (Jsc) of PSCs. As known, the optimized thickness of the PEDOT:PSS layers are ca. 35 nm, which blocks some of solar light38. Figure 2b depicted the optical transmittance spectra of the HAPAN films with varied thicknesses on quartz glass in the UV-Vis-NIR region. We found that the transmittance of the 30 nm HAPAN film is lower than that of the PEDOT:PSS (35 nm)35. When the thickness of HAPAN reduced to 5.0 nm or even to 1.3 nm, the corresponding transmittance of the whole absorption range was 95% ~ 98%, which was much higher than that of the PEDOT:PSS film (35 nm).

Surface morphologies of the HAPAN film on the ITO coated glass were probed by tapping-mode atomic force microscopy (AFM) measurements. Height topography images and the corresponding root-mean-square roughness (Rq) were taken for each film and shown in Figure 2c. After modified with 5.0 nm HAPAN film, the Rq value of ITO is significantly reduced from 3.50 nm to 1.32 nm. Interestingly, when 1.3 nm thick HAPAN was coated, the surface roughness of ITO still can be reduced effectively, i.e. the Rq value is 1.67 nm. Clearly, the ultrathin HAPAN (1.3 nm–5.0 nm) films are suitable p-type buffer material to meet the three key characteristics: high transmittance, suitable WF as well as relatively smooth surface.

Recently, Benzo[1,2-b:4,5-b′]dithiophene-based photovoltaic materials like PBDTTT-EFT, PBDTTT-C-T, PTB7 made a series of outstanding progress in polymer photovoltaic field5,6,7,8,11,19,20,21,22,23,24,25,38,39,40. Herein, a newly developed polymer PBDTTT-EFT11,39 was selected as the model polymer to verify the suitability of ultrathin HAPAN in highly efficient PSCs. Initially, to investigate the thickness dependence of HAPAN on photovoltaic performance, the PSCs based on PBDTTT-EFT: PC71BM with the conventional device configuration (ITO/p-type EBL/active layer/Ca/Al) were fabricated and characterized under AM 1.5G illumination (100 mW/cm2), as depicted in Figure 3a. The optimal D/A ratio (1:1.5) and processing solvent (DCB/3% DIO) were consistent with the reported work11. As shown in Figure S3 and Table 1, when the thickness of the HAPAN buffer layer increased from 1.3 nm to 30 nm, the Voc values of the devices were ca. 0.79 V, indicating that the thickness of the HAPAN film does not affect the Voc of the PSC at all. Noting that the ITO anode modified with 1.3 nm HAPAN showed a WF of 4.75 eV, it is interesting to get a Voc of 0.79 V for the device with 1.3 nm HAPAN. We speculate that the UPS result of the 1.3 nm HAPAN sample might be affected by the bared ITO part on the substrates, because the Rq of ITO (Rq = 3.5 nm, see Figure 2c) is much higher than the thickness of the ultrathin HAPAN (1.3 nm).

Table 1 The photovoltaic performance of PBDTTT-EFT/PC71BM-based PSC employing HAPAN with different thickness
Full size table
Figure 3
figure 3

(a) Device architecture of the conventional PSC utilized in this study; J-V characteristics of PBDTTT-EFT based PSC devices employing ultrathin HAPAN with different thickness under (b) 100 mW/cm2 AM 1.5G illumination and (c) in the dark; (d) The corresponding EQE curves of the PBDTTT-EFT based PSC devices employing ultrathin HAPAN with different thickness.

Full size image

When the thickness of the HAPAN buffer layer was changed from 1.3 nm to 30 nm, the Jsc values kept ~15 mA/cm2 with a peak value of 16.54 mA/cm2 (see Figure 3b and Table 1). For the PSCs without p-type EBL, poor PCE of 4.28 ± 0.28% were recorded. When modified with 5.0 nm HAPAN, the PSCs delivered the best PCE (9.00 ± 0.12%) along with the highest Jsc of 16.51 mA/cm2, a Voc of 0.789 V and a FF of 0.695, which were among the highest values from conventional PSC devices. The active layers also exhibited favorable morphology with appropriate nanoscale feature size of ca. 30 nm on the ultrathin HAPAN film. (See AFM and TEM images in Figure S4). It is worth noting that when the thickness of HAPAN is 1.3 nm only, a considerable PCE of 8.84 ± 0.10% still can be obtained and the results reveal that the ITO anode can be effectively modified by coating an ultrathin layer of HAPAN with even 1.3 nm. The series resistance (Rs) and shunt resistance (Rsh) were derived from the J-V curves. The series resistance (Rs) reduced to 7.96 Ω·cm2 and the shunt resistance (Rsh) increased to 0.89 kΩ·cm2 for the devices with 5.0 nm HAPAN buffer layer. As clearly seen in the dark J-V curves (Figure 3c), a relatively low leakage current and a high rectification ratio was recorded in ultrathin HAPAN-based PSC. In the parallel experiment, the best-performing PSC with PEDOT: PSS (35 nm) as p-type EBL showed a slightly lower PCE of 8.78% (see Figure S5).

The corresponding incident photon conversion efficiency (IPCE) spectra of the PSCs were characterized and collected in Figure 3d. The Jsc values integrated from the IPCE curves were 13.45 mA/cm2, 16.45 mA/cm2, 16.30 mA/cm2 correspond to the no buffer layer, 1.3 nm and 5.0 nm HAPAN device, respectively. Clearly, the Jsc values integrated from the IPCE curves agreed well with the Jsc of J-V tests. Compared with the PSC without p-type EBL, the best-performing PSC with 5.0 nm HAPAN film as p-type EBL exhibited a higher IPCE ranging from 300 nm to 800 nm, particular in 550 nm–750 nm. The 1.3 nm HAPAN also exhibited better IPCE in the region from 350 nm to 550 nm with a peak value of 72% at 450 nm, indicating efficient photon harvest and charge collection.

In addition, the long-term stability of PSCs based on ultrathin p-type EBL was also investigated (see Figure 4 and Figure S6). The PBDTTT-EFT: PC71BM-based unencapsulated PSCs in nitrogen glove-box employing 1.3 nm-thick HAPAN was collected and the PSCs employing PEDOT:PSS (35 nm) was also as controls. The PCE measured from PEDOT:PSS-based PSC was reduced to 70% of the pristine value within 30 days. In contrast, as observed in Figure 4, the PSC employing the ultrathin HAPAN film as the p-type EBL still attained more than 85% of the pristine PCE in a long period, which implied the device stability based on ultrathin HAPAN (1.3 nm) is much better than that of PEDOT:PSS (35 nm). The possible reasons might be the morphological changes of the active layer41, the high pH value as well as minimal usage of ultrathin HAPAN film.

Figure 4
figure 4

The long-term stability of the ultrathin HAPAN (1.3 nm) and PEDOT: PSS (35 nm) based-PSC stored in N2 filled glove box.

Full size image

To further investigate the applicability of this ultrathin p-type EBL, ultrathin HAPAN was also employed in the other efficient BHJ systems consisting PC71BM and photovoltaic polymers, such as PBDTTT-C-T9,21, PTB720, PBDTBDD40,42, PBTTDPP-T43 and PDPP3T26,44 and their molecular structures were provided in Figure 5a. The identical device fabrication processes were used according to the previous works besides of the p-type EBLs. All of the photovoltaic parameters were summarized in Table 2. The similar trends can be observed in PBDTTT-C-T/PC71BM, PTB7/PC71BM, PBDTBDD/PC71BM, PBTTDPP-T/PC71BM and PDPP3T/PC71BM-based PSC devices (see Figure 5b–5f). Besides, the ultrathin HAPAN was also utilized in the PSC based on the classical P3HT (see Figure 5g). It can be concluded that the ultrathin HAPAN film is a universal and effective p-type EBL for various polymer: PCBM blends.

Table 2 The photovoltaic properties of PSCs based on several efficient polymer: PCBM blends under AM 1.5G, 100 mW/cm2 solar illumination
Full size table
Figure 5
figure 5

(a) Molecular structures of other efficient polymers and P3HT; The J-V curves of PSC based on (b) PBDTTT-C-T: PC71BM, (c) PTB7: PC71BM, (d) PBDTBDD: PC71BM, (e) PBTTDPP-T: PC71BM, (f) PDPP3T: PC71BM and (g) P3HT: PC61BM.

Full size image

In summary, PBDTTT-EFT based PSCs with the conventional architecture were fabricated with ultrathin HAPAN film as anode buffer layer and high PCEs of approximately 9% were obtained on only 5.0 nm or 1.3 nm-thick HAPAN film. This study suggests that the application of ultrathin HAPAN film as the anode interfacial layer is also an effective method to improve the device stability. To the best of our knowledge, this is the first report about utilizing ultrathin p-type EBLs in highly efficient PSCs. More importantly, this ultrathin layer works well in several efficient photovoltaic polymers including PBDTTT-C-T, PTB7, PBDTBDD, PBTTDPP-T, PDPP3T and P3HT, so it gives a new option for fabricating higher efficiency and long-term stability solar cells as well as ultrathin PSC devices45. Further advances in photovoltaic performance would be achieved by optimizing the intrinsic conductivity and pH values of the polyaniline-based p-type anode buffer layers.

Methods

Materials

PBDTTT-EFT, PTB7, PDPP3T and PC71BM were commercial available from Solarmer Material Inc. PBDTTT-C-T21, PBDTBDD42, PBTTDPP-T43 were synthesized in our laboratory according to our previous works. The other materials were commercially available and used as received.

Device fabrications

The different concentration HAPAN solution was prepared in distilled water at 60°C for 3 hours in air. The anode buffer layer was spin-coated on pre-cleaned ITO substrates from the HAPAN solution prepared at 4000 rpm for 40 s and annealed at 150°C for 15 min. The thickness of HAPAN is 1.3 nm, 2.5 nm, 5.0 nm, 10 nm, 20 nm, 30 nm respectively corresponding to a concentration of 1 mg/ml, 2 mg/ml, 4 mg/ml, 10 mg/ml, 20 mg/ml 30 mg/ml. The concentration of PBDTTT-EFT and PC71BM in o-dichlorobenzene (DCB) solution is 10 mg/ml and 15 mg/ml, respectively. These HAPAN films were spin-coated at 4000 rpm for 40 s. The film thickness of HAPAN was verified by several methods such as spectroscopic ellipsometer and the film roughness was characterized using AFM in tapping mode. For morphology modulation, 3% DIO (vol) was added into the blend solution. Blend solutions based on other polymers are prepared according to the previous works and also described in details in Table S2. Prior to evaporating metal cathodes, the blend films were treated with ~60 μL methanol according to our recent report46. The device fabrication was completed by thermal evaporation of a 20 nm thick Ca and a 80 nm thick Al layer as cathode under vacuum at a base pressure of 1 × 10−4 Pa.