Ternary solar cells with a mixed face-on and edge-on orientation enable an unprecedented efficiency of 12.1%

Abstract

Ternary organic solar cells (OSCs), with a simple structure, can be easily adopted as sub-cells in a tandem design, thereby further enhancing the power conversion efficiency (PCE). Considering the potential to surpass the theoretical PCE limit in OSCs, we incorporated a benzo[1,2-b;4,5-b′]dithiophene-based small molecule into a poly(4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b;4,5-b′]dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl):[6,6]-phenyl-C₇₁-butyric acid methyl ester host system. A hitherto unrealized PCE of 12.1% was achieved at the optimized composition of the ternary blend. The ternary blend surprisingly had a face-on and edge-on co-existent texture, which is far better than that of the face-on orientated host film. To the best of our knowledge, this intriguing result refutes for the first time a general paradigm that high-performance OSCs are unambiguously linked to face-on structures. Therefore, our study provides a new platform for refining the theoretical underpinning of multiple blending OSCs.

---

Broader context

With enormous research interest in the past few years, ternary blend organic solar cells (OSCs) comprising multiple donor or acceptor materials in an active layer have emerged as promising alternatives to enhance photovoltaic parameters for spectrally broad light harvesting while retaining the simplicity of a single step for processing the active layer as compared with traditional binary or tandem OSCs. Recent ternary structures of conjugated polymers have achieved around 10% power conversion efficiencies (PCEs), showing the great potential of ternary systems. Herein, we report ternary OSCs fabricated by incorporating DR3TSBDT as an additional donor into a PTB7-Th:PC₇₁BM host matrix at different weight ratios, exhibiting the record-PCE of 12.1% under highly optimized conditions. Through a series of characterization techniques, we identified that our ternary OSCs have mixed face-on and edge-on orientations, reaching much improved PCEs unlike common high-performing OSCs that preferentially favor face-on orientated host systems. Additionally, our study reveals that the exceptionally high PCE in this ternary structure results from not only improving the photon absorption range but also facilitating charge transport while reducing recombination. This is achieved through a combination of cascade energy levels and optimized morphology.

Introduction

An in-depth understanding of the operating mechanism of binary blend organic solar cells (OSCs), consisting of a polymeric donor and a fullerene acceptor, has enabled several research efforts aimed at increasing the power conversion efficiencies (PCEs) of OSCs.^1–4 An optimization process involving a synergistic combination of active material design,^2,5,6 nanoscale morphology control,^7–9 and device engineering^10–12 resulted in a significant breakthrough of over 10% PCEs for single-junction binary OSCs.^3,4,13–16 In a promising approach for further increasing the PCEs, tandem OSCs – in which two (or more) sub-cells absorbing light in different regions of the solar spectrum are connected in either series or parallel – have been studied.^17,18 The impressive progress in the field of tandem OSCs has laid a solid foundation for improving the PCEs by ∼40–50%,^19,20 demonstrating their immense potential for a practically realizable OSC technology. Nonetheless, there are several drawbacks in using tandem architectures, such as the complex fabrication process and high production cost, which limits their practical application.^17,21

Ternary blend OSCs comprising two donors and one acceptor (or one donor and two acceptors) are emerging as a fascinating alternative to overcome the challenges encountered during spectrally broad light harvesting using multi-junction OSC processing while retaining the simplicity of single-step processing of the active layer.^14,22–25 Despite some successful examples of ternary OSCs achieved by carefully selecting multiple components,^7,26–32 the current PCEs generally continue to be far less than the state-of-the-art binary and tandem systems. This is because the third component within the host binary systems can act as a recombination center or a morphological trap rather than a control agent to extend the absorption of the solar spectrum, since unfavourable interactions between the third component and host blend are inevitable.^33,34 Therefore, the molecular compatibility of the active materials used in ternary OSCs is believed to be critical in achieving high PCEs.^3,14,35–41 We speculate that in a vast pool of available active materials, the use of an archetype of the high-performance poly(4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b;4,5-b′]dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl) (PTB7-Th):[6,6]-phenyl-C₇₁-butyric acid methyl ester (PC₇₁BM) binary host can pave a shortcut for discovering ideal ternary systems. In addition to its appropriate energy level alignment and high-crystalline characteristics, a benzo[1,2-b;4,5-b′]dithiophene (BDT)-based small molecule, namely DR3TSBDT, is intuitively expected to have good compatibility with PTB7-Th because of the molecular similarity in their backbones, which is based on identical BDT units.

Given this background, a ternary OSC was designed and fabricated by incorporating DR3TSBDT as the additional donor into the PTB7-Th:PC₇₁BM host matrix. The ternary system demonstrated an unprecedented PCE of 12.1% under optimal conditions. Unlike common high-performance OSCs that favour a face-on orientation, our ternary OSCs have a mixed orientation, which is a combination of face-on and edge-on orientations, thus enabling much higher PCEs compared to the preferentially face-on orientated host system. Our in-depth study reveals that the exceptionally high PCE results from not only improving the photon absorption range but also facilitating charge transport while reducing recombination. This is achieved through a combination of cascade energy levels and optimized morphology.

Results and discussion

Characterization and optical properties

The chemical structures of PTB7-Th, DR3TSBDT, and PC₇₁BM, as well as the device architecture and corresponding energy levels, are given in Fig. 1a and b. We first carried out a detailed spectroscopy study to understand the changes in the chlorobenzene (CB) solution and in the chlorobenzene based thin film following the incorporation of the third component. The two donor components exhibit complementary absorption spectra (Fig. 2a and ESI,† Fig. S1a). The maximum absorption for the PTB7-Th film is at 706 nm and that of the DR3TSBDT film is centered at 586 nm. As the DR3TSBDT content in PTB7-Th increased, the absorption intensity gradually increased in the region of 400–600 nm, while the absorption at around 700 nm decreased. Note that the maximum absorption of PTB7-Th is red-shifted following the incorporation of DR3TSBDT. Fig. 2b shows the effect of incorporating various concentrations of DR3TSBDT into the PTB7-Th : PC₇₁BM host at a certain blending ratio (1 : 1.5 wt/wt). As for the absorption spectra of the two donor mixtures (Fig. 2a), it is seen that incorporating DR3TSBDT to form ternary blends leads to a small red-shift in the low-energy absorption band, implying an increased molecular ordering in PTB7-Th, which is induced by the favourable interaction with DR3TSBDT. Therefore, we expect that an optimal amount of DR3TSBDT will facilitate the crystallization of PTB7-Th in ternary OSCs. In addition, the ternary films formed using 10–30 wt% DR3TSBDT provide relatively high absorption intensity over a broad wavelength range of 300 to 800 nm (Fig. 2b).

Fig. 1 (a) Chemical structures of PTB7-Th, DR3TSBDT, and PC₇₁BM. (b) Device architecture with illustration of the active layer morphology in ternary OSCs and an energy level diagram.

Fig. 2 (a) Absorption spectra of two donor blend films. (b) Absorption spectra of ternary blend films. (c) Photoluminescence spectra (PL) of two donor binary blend films. (d) Photoluminescence spectra (PL) of two donor ternary blend films with PC₇₁BM in chlorobenzene solution (excitation at ∼500 nm).

Fig. 2c and d show the photoluminescence (PL) spectra of each neat donor, two-donor binary blends, and ternary blend solutions under excitation at ∼500 nm. PTB7-Th exhibits a rather structured emission with a maximum at 760 nm and DR3TSBDT shows a broad emission peak from 580 to 800 nm, which overlaps with the absorption spectrum of PTB7-Th. A gradual increase in the emission intensity of DR3TSBDT was observed, while the emission intensity of PTB7-Th decreased with increasing DR3TSBDT loading ratios, reflecting the existence of Förster energy transfer between DR3TSBDT and PTB7-Th.^22,25,27,30 Interestingly, the quenching of the emission was more efficient in ternary films with 25 wt% DR3TSBDT than in other ternary blends, implying the feasibility of a relatively optimized charge transfer pathway at the two donor/acceptor interface.

Photovoltaic performance

The photovoltaic performance of OSCs fabricated using different DR3TSBDT concentrations was evaluated using a conventional architecture of indium-tin oxide (ITO)/poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS)/active layer/Al under simulated AM 1.5G irradiation (100 mW cm⁻²). The overall donor to PC₇₁BM ratio in the active layer was fixed at 1 : 1.5 wt/wt. 1,8-Diiodooctane (DIO, 3.0 vol%) was used as a processing additive. This paper presents a discussion on the representative ternary blend systems (0, 10, 25, 30, 40, and 100 wt% DR3TSBDT loading ratios). Additional details of the ternary blends examined in this study are included in the ESI,† Fig. S2 and S3.

The current density–voltage (J–V) characteristics of the OSCs are shown in Fig. 3a and the corresponding device parameters are summarized in Table 1. Using highly optimized conditions for PTB7-Th:PC₇₁BM binary OSCs, a maximum PCE of 10.10% was obtained, with a short-circuit current density (J_SC) of 19.43 ± 0.30 mA cm⁻², an open-circuit voltage (V_OC) of 0.785 ± 0.004 V, and a fill factor (FF) of 64.90 ± 0.40%. The PCE value reported in this study is comparable with that of the previous best OSCs based on PTB7-Th:PC₇₁BM.^25,42 Upon the addition of 25 wt% DR3TSBDT into the host system, the J_SC value increases continuously together with a moderate enhancement in the FF (up to 70.44%), whereas a marginal decrease in the V_OC value was observed in the range of 0.794–0.772 V. Meanwhile, at DR3TSBDT loadings higher than 30 wt%, a large drop in the J_SC value was observed. Therefore, ternary OSCs with a 25 wt% DR3TSBDT content exhibited the best photovoltaic performance with a J_SC value of 22.63 ± 0.67 mA cm⁻², a V_OC value of 0.765 ± 0.007 V, and a FF value of 68.5 ± 1.9%, in an unprecedented PCE of 12.1% (average PCE = 11.78 ± 0.34%). One of our best cells was sent to an independent solar cell calibration laboratory (Nano Convergence Practical Application Center, South Korea) for certification, confirming a PCE of 11.76%, with V_OC = 0.756 V, J_SC = 23.76 mA cm⁻², FF = 65.5% (see the ESI,† Fig. S4), and is detailed in the Experimental section. To the best of our knowledge, this value is among the highest certified PCE reported to date for any type of OSC. Evidence of a hysteresis less J–V characteristic curve for one of the best devices is also provided (ESI,† Fig. S5). Additionally, we observed that V_OC is composition-dependent in the blend systems studied, rather than being determined by the difference between the PC₇₁BM LUMO (lowest unoccupied molecular orbital) and the lowest-available donor HOMO (highest occupied molecular orbital) level. The observed behavior in V_OC suggests the formation of an alloy-like model induced by the compatible donors.^1,3,43

Fig. 3 (a) J−V characteristics of OSCs under AM 1.5G irradiation at 100 mW cm⁻². (b) The corresponding EQE curves.

Table 1 Photovoltaic parameters and charge transport properties of OSCs with different DR3TSBDT loading ratios

DR3TSBDT ratios	J SC (mA cm^−2)	V OC (V)	FF^a (%)	PCE^a (%)	μ h (cm^2 V^−1 s^−1)	μ e (cm^2 V^−1 s^−1)	μ h/μe^b
a The average values obtained from at least 16 devices with standard deviation. The data in parentheses are the highest values. b The average values obtained from at least 4 devices with standard deviation.
0%	19.43 (19.70)	0.785 (0.789)	64.90 (65.30)	9.87 (10.10)	1.578 × 10^−4	1.066 × 10^−4	1.478
10%	20.72 (21.69)	0.772 (0.782)	67.83 (69.90)	10.79 (11.30)	1.683 × 10^−4	1.134 × 10^−4	1.483
25%	22.63 (23.31)	0.765 (0.772)	68.51 (70.44)	11.78 (12.10)	2.200 × 10^−4	1.268 × 10^−4	1.735
30%	19.90 (21.38)	0.762 (0.770)	67.78 (69.30)	10.97 (11.30)	2.119 × 10^−4	1.176 × 10^−4	1.801
40%	17.37 (18.00)	0.756 (0.766)	67.60 (69.10)	8.88 (9.26)	2.020 × 10^−4	1.048 × 10^−4	1.927
100%	12.92 (13.41)	0.897 (0.902)	54.80 (56.72)	4.99 (5.52)	0.208 × 10^−4	0.159 × 10^−4	1.309

---

From the differential scanning calorimetry (DSC) data, as shown in the ESI† Fig. S6, the heating curve of the pristine DR3TSBDT shows a clearly endothermic transition (i.e., melting peak), confirming its relatively high crystallinity. In contrast, no such thermally induced transition occurs during the heating cycle of both the pristine PTB7-Th and all the blend systems of PTB7-Th:DR3TSBDT, suggesting a lack of pure DR3TSBDT domains in the blends.^3,44,45 Similar surface energies of 28.4 and 30.6 mJ m⁻² for DR3TSBDT and the PTB7-Th:PC₇₁BM host blend, respectively, calculated using the Young's equation,⁴⁶ provide additional proof of the good miscibility in the ternary systems (ESI,† Fig. S7). This was further evidenced using energy dispersive X-ray analysis (EDAX) and optical microscopy. The EDAX elemental mapping indicates that the nitrogen signals from DR3TSBDT are well-spread and are adjacent to the fluoride peaks of PTB7-Th in the ternary blends (Fig. 4 and ESI,† Fig. S8). Each ternary film appears to be uniformly blended, at least at the scale shown in the optical microscopy, lacking any gross aggregation (ESI,† Fig. S9).

Fig. 4 Energy dispersive X-ray elemental mapping; fluorine (purple dots) and nitrogen (red dots) indicate PTB7-Th and DR3TSBDT, respectively, and the inset is the corresponding zoomed images.

External quantum efficiency (EQE) spectra revealed an excellent photocurrent response over the absorption range of 300–800 nm (Fig. 3b). For all devices, the shapes of the EQE plots are similar to the corresponding absorption spectra, indicating that the absorption over the entire wavelength range contributes to photocurrent generation.

Carrier transport and recombination dynamics

To understand the working mechanism and different photovoltaic performances in the tested OSCs, we characterized the charge transport properties and recombination dynamics. The charge carrier mobilities of the hole- and electron-only devices were estimated using a space-charge limited curve (SCLC),^47,48 as detailed in the Methods section (Table 1 and ESI,† Fig. S10). With the electron mobilities being relatively constant for all ternary cases, the hole mobilities are sensitive to the DR3TSBDT loading ratios. All blends investigated in this study exhibited highly balanced charge-transport properties (hole/electron mobility ratios of 1.30–1.92). In addition, it was found that both the hole and electron mobilities gradually increased at low DR3TSBDT loadings (0–25 wt%), while a further increase in the DR3TSBDT content caused a decrease in the mobilities. Thus, the ternary OSCs with 25 wt% DR3TSBDT showed a superior mobility level compared to the other samples, implying better percolation pathways for the charge carriers. This is due to the optimized morphology, which is discussed in detail in the following section.

We then analyzed the dependence of the J–V characteristics of the devices on light intensity, in order to elucidate the recombination dynamics that can affect the performance. It is well known that a correlation between J_SC and illumination intensity (I) exists and follows the relation J_SC ∝ I^α (Power law), where α is the unity when all the carriers are swept out prior to the bimolecular recombination process.^49,50 As shown in the ESI,† Fig. S11a, the logarithmic plots of J_SCversus I show slopes (α) in the range of 0.95–0.98 for the devices with 0–40 wt% DR3TSBDT, along with a slope of 0.92 for the DR3TSBDT:PC₇₁BM binary system. This suggests that the bimolecular recombination loss is minor for all devices except for DR3TSBDT:PC₇₁BM, which correlates with the unsatisfactory performance of DR3TSBDT:PC₇₁BM, especially the low FF.

The device with 25 wt% DR3TSBDT exhibited the weakest bimolecular recombination with an α value of 0.98, which partially explains the highest J_SC value and mobility levels observed. Fig. S11b (ESI†) shows a plot of V_OC as a function of the logarithm of light intensity. For all cases, a strong dependence of V_OC on light intensity with a slope larger than 2kT/q is observed, where k is Boltzmann's constant, T is temperature, and q is elementary charge, suggesting the presence of a trap-assisted recombination.

Film morphology and microstructure

The morphology and microstructure of the blend films were characterized using atomic force microscopy (AFM), high-resolution transmission electron microscopy (HR-TEM), and grazing incidence wide-angle X-ray scattering (GIWAXS). The PTB7-Th:PC₇₁BM host blend film possessed a small-length scale morphology with a root-mean-square (RMS, R_q) roughness of 1.24 nm (Fig. 5a and ESI,† Fig. S12). Incorporating up to 30 wt% DR3TSBDT into the host matrix led to a gradual increase in the uniform features with a nanoscale phase separation. However, a further increase in DR3TSBDT caused a higher roughness with large aggregated regions. In the HR-TEM images of ternary films containing 25ߝ40 wt% DR3TSBDT showed finely dispersed fibrils with domain sizes ranging from 10 to 15 nm, plausibly reflecting an interpenetrating network (Fig. 5b). However, such features were not observed in other films.

Fig. 5 (a) AFM topography images (scan size 1 × 1 μm). (b) HR-TEM images (scale bar 50 nm) of blend films with different DR3TSBDT loading ratios. Different color bars are used for height (AFM) variance.

As seen from the GIWAXS patterns in Fig. 6 and in the ESI,† Fig. S13, PTB7-Th:PC₇₁BM produced a (010) π–π stacking peak centered at about 1.67 Å⁻¹ in the out-of-plane direction (q_z) with an arc-like lamellar (100) peak along the in-plane direction (q_xy). This indicates a favoured face-on orientation relative to the substrate (see the ESI,† Table S1 for detailed crystallographic parameters). In contrast, in the case of DR3TSBDT:PC₇₁BM, another binary system, the (010) π–π stacking peak was located along the q_xy direction in conjunction with the highly ordered lamellar (100), (200), and (300) peaks in the q_z axis, suggesting a preferential edge-on alignment. Interestingly, adding DR3TSBDT to form the ternary films resulted in the formation of (010) π–π stacking peaks in both the q_xy and q_z profiles, providing conclusive evidence for the formation of mixed edge-on and face-on orientations, the so-called 3-D textured structures. In particular, we found that the intensities of both the out-of-plane and in-plane (010) peaks were strengthened for the ternary film in the case of the 25% DR3TSBDT blend (Fig. 7a), which is largely responsible for the enhanced charge transport. Additionally, for all samples the (010) coherence lengths (L_Cs), which are a measure of cumulative lattice distortions in both the q_xy and q_z axes, were also calculated using the Scherrer equation⁵¹ (Fig. 7b). With the L_C values along the q_z being relatively consistent for all the cases (4–7 nm), the L_C values along the q_xy increased from ∼6 to 20 nm with an increase in the DR3TSBDT loading ratios, illustrating that the ternary systems could form larger nanocrystallites than the PTB7-Th:PC₇₁BM host. These changes in the molecular stacking and crystallinity, following the addition of a guest molecule to the host matrix, provide additional evidence for the alloy model proposed in this study.^1,3,41,43,52 Another observation that needs to be emphasized is that, compared to the PTB7-Th:PC₇₁BM host system, which has a predominantly face-on orientation, our data for the optimized ternary systems with 3-D textures indicate far better PCEs. This observation reported for the first time, challenges the traditional assumption that a face-on orientation is more favourable for photovoltaic devices because of its vertical charge transportation channel.

Fig. 6 Grazing incidence wide angle X-ray scattering (GIWAXS) patterns of blend films with different DR3TSBDT loading ratios. Different color bars are used for intensity variance.

Fig. 7 (a) Exemplary fittings of the GIWAXS line cuts at the higher q region in the horizontal direction for PTB7-Th:PC₇₁BM, PTB7-Th–DR3TSBDT (25%):PC₇₁BM and DR3TSBDT:PC₇₁BM blend films. (b) In-plane crystallite coherence length (CCL) of π–π stacking. Gaussian fits were used to determine the peak position and FWHM.

Conclusions

A ternary system comprising DR3TSBDT:PTB7-Th as a donor and PC₇₁BM as an acceptor was investigated in detail, considering both cascaded energy levels and molecular compatibility. At low DR3TSBDT loadings (≤25 wt%) into the host system, synergistic effects such as increased J_SC and FF values were observed, which is attributed to enhanced light absorption and charge transport, reduced recombination, and optimized morphology. A notable PCE as high as 12.1% was obtained from ternary OSCs with 25 wt% DR3TSBDT, which is among the highest value reported so far. The intimate mixing properties of two donors enable the formation of an alloy-like model in the present ternary systems, as verified by not only elemental mapping and optical microscopy but also the V_OC values and molecular packing/crystallinity. Moreover, we discovered that a mixed face-on and edge-on orientation in ternary OSCs, rather than an almost entirely face-on orientation, yields better photovoltaic performances, which challenges the fundamental idea successfully used in the past to describe high-performance OSCs. This finding provides further impetus for continued research in the field of ternary OSCs in order to understand the interesting and sometimes surprising behaviour of the multiple blending systems.

Experimental section

Materials

PTB7-Th was purchased from 1-Material and PC₇₁BM from Ossila. All other commercial reagents, including the solvents used in this study, were purchased from Aldrich Co., Alfa Aesar, and TCI Co. and used without further purification. DR3TSBDT was synthesized as described previously.¹³ All active materials used in this study were sourced from the same batch, in order to ensure a fair comparison between the experimental and control devices.

Device fabrication and characterization

A conventional device was used to assess the performance of the PTB7-Th:PC₇₁BM host system and PTB7-Th:PC₇₁BM-based ternary OSCs by varying the amounts of DR3TSBDT and a total of 108 samples were measured. For device fabrication, ITO-coated glass substrates (12 Ω □⁻¹) were sequentially cleaned via ultrasonication in liquid detergent, de-ionized water, acetone, and isopropanol for 15 min each, and dried in an oven at 70 °C overnight. A thin layer of PEDOT:PSS (Baytron® PVP AI 4083 – H.C. Starck, filtered using a 0.45 μm PVDF syringe filter) was first spin coated on the pre-cleaned and UV-Ozone-treated ITO-coated glass substrates at 4000 rpm. After subsequent annealing at 140 °C for 20 min in an ambient atmosphere, the substrates were immediately transferred to a nitrogen-filled glove box. The active layer was then spin coated to yield a ∼140–150 nm thick film (see the ESI,† Fig. S14). The films were dried in a vacuum for 30 min and a 100 nm-thick Al cathode was thereafter thermally evaporated on top of the organic layer using a mask (device area: 0.13 cm²) under a pressure of <9 × 10⁻⁷ Torr. For the preparation of the active layer solution, 14 mg of donor (PTB7-Th or DR3TSBDT) and 21 mg of PC₇₁BM (1 : 1.5 ratio) were dissolved in 1 ml of chlorobenzene. The solutions were heated overnight at 60 °C under minimal stirring. Later, 3.0 vol% of DIO was added to the solutions and stirred well for homogeneity. For ternary blending, the two solutions were mixed according to the different weight ratios of the two donors and heated at 60 °C for 1 h under minimal stirring in the glove box.

After fabrication, the J–V measurements for the photovoltaic devices were performed under simulated sunlight (AM 1.5G) at an intensity of 100 mW cm⁻² using a Keithley 2365A source measurement unit. The measurements were performed in a nitrogen-filled glove box. Prior to device measurement, the irradiance was calibrated using a standard silicon photodiode. For the certified PCE data, it should be noted that the current, power and device area is the sum of 4 devices (each of 0.075 cm²). The measurement of all four devices in a single time was done according to the rule of the calibration laboratory in order to minimize any overestimation of the PCE. The measurement of 4 devices at a single time resulted in a dramatic loss in the fill factor as the total device area was increased which subsequently increased the number of recombination centers. EQE measurements were performed under an ambient atmosphere by using a QEX7 quantum efficiency measurement system (PV Measurements, Inc.). In addition, the deviations between the integral current densities and the J_SC values from the J–V measurements were found to be below 10%. For mobility measurements, the hole- and electron-only devices were fabricated using a device architecture of ITO/PEDOT:PSS/active layer/Au and ITO/Al/active layer/Al, respectively. The hole (μ_h) and electron (μ_e) mobilities were measured using the SCLC method, which is described by the Mott–Gurney square law as

J_SCLC = (9/8)ε₀ε_Tμ((V²)/(L³))

where J_SCLC is the current density, ε₀ is the dielectric constant of free space, ε_r is the relative permittivity of the transport medium (assumed to be 3), μ is the hole (μ_h) or electron (μ_e) mobility, V is the applied voltage across the device, and L is the thickness of the active layer.

Measurements

Thin films for all characterization methods were prepared using the same processing conditions as used for device fabrication and were characterized under an ambient atmosphere. For the absorption, PL emission, optical microscopy, and thickness measurements of the ternary films, the samples were fabricated on a glass substrate. For all other characterization methods, the samples were fabricated on a silicon substrate. UV-vis spectra were obtained using a UV-1800 (SHIMADZU) spectrophotometer and the absorption coefficient was obtained using the following equation:

α = 2.303 × (A/l)

where α is the absorption coefficient of the material, A is the absorbance, and l is the thickness of the thin film (in cm). PL spectra were obtained using a Varian Cary Eclipse fluorescence spectrometer. The DSC data were obtained using a DSC Q200 TA Instrument, USA. The heating rate was 5 °C min⁻¹. Static contact angles of water and glycerol were measured using Phoenix 300 (SEO) for the surface energy measurement using Young's equation. For EDAX, a Hitachi SU 8220 Cold FE-SEM was used. To achieve sufficient X-ray excitation, the FE-SEM was operated at an acceleration voltage of 10 kV. Note that only a relative concentration distribution of elements can be obtained from the resulting elemental mapping over the mapped area. Optical microscopy images were acquired using an inverted metallurgical microscope (Alpha300S, WlTec). AFM investigations were performed using a multimode V microscope (Veeco, USA) in the tapping mode with a nanoscope controller using Si tips (Bruker) at a resonance frequency of 300 kHz. HR-TEM analysis was performed using a JEOL USA JEM-2100F (Cs corrector) transmission electron microscope equipped with an EDAX at an acceleration voltage of 200 kV. The specimen for HR-TEM measurement was prepared by spin coating the blend solution on a PEDOT:PSS-coated silicon substrate and floating the film on a de-ionized water surface. The film was then transferred to lacey carbon grids (HC200-Cu). GIWAXS characterization was carried out at the PLS-II 9A U-SAXS beam line of the Pohang Accelerator Laboratory in Korea. The scattering signal was recorded using a 2-D CCD detector (Rayonix SX165). The X-ray light had an energy of 11.24 keV (λ = 1.103 Å). The incidence angle of X-rays was adjusted to 0.13–0.145° to maximize the signal to background ratio. The detector was located at a distance of approximately 232 mm from the center of the sample. The raw data were processed and analyzed using an Igor-Pro software package. The thickness of the binary and ternary films was measured using a stylus profilometer (P6, KLA Tencor).

Acknowledgements

C. Y. conceptualized the project. T. K. carried out the experiments and performed the device fabrication, as well as analyzed the data. S. M. L. synthesized DR3TSBDT. T. K. and C. Y. wrote the manuscript. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (2015R1A2A1A10053397). GIWAXD measurements at PLS-II 6D UNIST-PAL beamline and 9A beamline were supported in part by MEST, POSTECH, and UNIST UCRF.

Notes and references

1. Q. An, F. Zhang, J. Zhang, W. Tang, Z. Deng and B. Hu, Energy Environ. Sci., 2016, 9, 281–322.
2. J. Subbiah, B. Purushothaman, M. Chen, T. Qin, M. Gao, D. Vak, F. H. Scholes, X. Chen, S. E. Watkins, G. J. Wilson, A. B. Holmes, W. W. H. Wong and D. J. Jones, Adv. Mater., 2015, 27, 702–705.
3. J. Zhang, Y. Zhang, J. Fang, K. Lu, Z. Wang, W. Ma and Z. Wei, J. Am. Chem. Soc., 2015, 137, 8176–8183.
4. S. Zhang, L. Ye and J. Hou, Adv. Energy Mater., 2016, 60, 1502529.
5. M. C. Scharber, D. Mühlbacher, M. Koppe, P. Denk, C. Waldauf, A. J. Heeger and C. J. Brabec, Adv. Mater., 2006, 18, 789–794.
6. J. D. Servaites, B. M. Savoie, J. B. Brink, T. J. Marks and M. A. Ratner, Energy Environ. Sci., 2012, 5, 8343–8350.
7. W. Chen, Z. Du, M. Xiao, J. Zhang, C. Yang, L. Han, X. Bao and R. Yang, ACS Appl. Mater. Interfaces, 2015, 7, 23190–23196.
8. S. Guo, J. Ning, V. Körstgens, Y. Yao, E. M. Herzig, S. V. Roth and P. Müller-Buschbaum, Adv. Energy Mater., 2015, 5, 1401315.
9. Y. Liu, J. Zhao, Z. Li, C. Mu, W. Ma, H. Hu, K. Jiang, H. Lin, H. Ade and H. Yan, Nat. Commun., 2014, 5, 5293.
10. F. C. Krebs, Sol. Energy Mater. Sol. Cells, 2009, 93, 394–412.
11. J. M. Lee, J. Lim, N. Lee, H. I. Park, K. E. Lee, T. Jeon, S. A. Nam, J. Kim, J. Shin and S. O. Kim, Adv. Mater., 2015, 27, 1519–1525.
12. W. Shockley and H. J. Queisser, J. Appl. Phys., 1961, 32, 510–519.
13. B. Kan, Q. Zhang, M. Li, X. Wan, W. Ni, G. Long, Y. Wang, X. Yang, H. Feng and Y. Chen, J. Am. Chem. Soc., 2014, 136, 15529–15532.
14. S. Liu, P. You, J. Li, J. Li, C.-S. Lee, B. S. Ong, C. Surya and F. Yan, Energy Environ. Sci., 2015, 8, 1463–1470.
15. Y. Liu, J. Zhao, Z. Li, C. Mu, W. Ma, H. Hu, K. Jiang, H. Lin, H. Ade and H. Yan, Nat. Commun., 2014, 5, 5293.
16. J. Zhao, Y. Li, G. Yang, K. Jiang, H. Lin, H. Ade, W. Ma and H. Yan, Nat. Energy, 2016, 1, 15027.
17. T. Ameri, G. Dennler, C. Lungenschmied and C. J. Brabec, Energy Environ. Sci., 2009, 2, 347–363.
18. J. Gilot, M. M. Wienk and R. A. J. Janssen, Appl. Phys. Lett., 2007, 90, 143512.
19. J. D. Kotlarski and P. W. M. Blom, Appl. Phys. Lett., 2011, 98, 053301.
20. J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C.-C. Chen, J. Gao, G. Li and Y. Yang, Nat. Commun., 2013, 4, 1446.
21. J. Yang, R. Zhu, Z. Hong, Y. He, A. Kumar, Y. Li and Y. Yang, Adv. Mater., 2011, 23, 3465–3470.
22. H. Cha, D. S. Chung, S. Y. Bae, M.-J. Lee, T. K. An, J. Hwang, K. H. Kim, Y.-H. Kim, D. H. Choi and C. E. Park, Adv. Funct. Mater., 2013, 23, 1556–1565.
23. S. Honda, H. Ohkita, H. Benten and S. Ito, Chem. Commun., 2010, 46, 6596–6598.
24. M. Koppe, H.-J. Egelhaaf, G. Dennler, M. C. Scharber, C. J. Brabec, P. Schilinsky and C. N. Hoth, Adv. Funct. Mater., 2010, 20, 338–346.
25. L. Lu, W. Chen, T. Xu and L. Yu, Nat. Commun., 2015, 6, 7327.
26. Q. An, F. Zhang, L. Li, J. Wang, J. Zhang, L. Zhou and W. Tang, ACS Appl. Mater. Interfaces, 2014, 6, 6537–6544.
27. Q. An, F. Zhang, Q. Sun, J. Wang, L. Li, J. Zhang, W. Tang and Z. Deng, J. Mater. Chem. A, 2015, 3, 16653–16662.
28. M. E. Farahat, D. Patra, C.-H. Lee and C.-W. Chu, ACS Appl. Mater. Interfaces, 2015, 7, 22542–22550.
29. J.-H. Huang, M. Velusamy, K.-C. Ho, J.-T. Lin and C.-W. Chu, J. Mater. Chem., 2010, 20, 2820–2825.
30. T.-Y. Huang, D. Patra, Y.-S. Hsiao, S. H. Chang, C.-G. Wu, K.-C. Ho and C.-W. Chu, J. Mater. Chem. A, 2015, 3, 10512–10518.
31. P. P. Khlyabich, B. Burkhart and B. C. Thompson, J. Am. Chem. Soc., 2011, 133, 14534–14537.
32. L. Yang, H. Zhou, S. C. Price and W. You, J. Am. Chem. Soc., 2012, 134, 5432–5435.
33. L. Lu, M. A. Kelly, W. You and L. Yu, Nat. Photonics, 2015, 9, 491–500.
34. B. M. Savoie, S. Dunaisky, T. J. Marks and M. A. Ratner, Adv. Energy Mater., 2015, 5, 1400891.
35. H. Lu, X. Zhang, C. Li, H. Wei, Q. Liu, W. Li and Z. Bo, Macromol. Rapid Commun., 2015, 36, 1348–1353.
36. L. Lu, T. Xu, W. Chen, E. S. Landry and L. Yu, Nat. Photonics, 2014, 8, 716–722.
37. B. C. Thompson, Y.-G. Kim and J. R. Reynolds, Macromolecules, 2005, 38, 5359–5362.
38. L. Xiao, K. Gao, Y. Zhang, X. Chen, L. Hou, Y. Cao and X. Peng, J. Mater. Chem. A, 2016, 4, 5288–5293.
39. S. Zhang, L. Zuo, J. Chen, Z. Zhang, J. Mai, T.-K. Lau, X. Lu, M. Shi and H. Chen, J. Mater. Chem. A, 2016, 4, 1702–1707.
40. N. Gasparini, X. Jiao, T. Heumueller, D. Baran, G. J. Matt, S. Fladischer, E. Spiecker, H. Ade, C. J. Brabec and T. Ameri, Nat. Energy, 2016, 1, 16118.
41. L. Yang, L. Yan and W. You, J. Phys. Chem. Lett., 2013, 4, 1802–1810.
42. Z. He, B. Xiao, F. Liu, H. Wu, Y. Yang, S. Xiao, C. Wang, T. P. Russell and Y. Cao, Nat. Photonics, 2015, 9, 174–179.
43. D. Angmo, M. Bjerring, N. C. Nielsen, B. C. Thompson and F. C. Krebs, J. Mater. Chem. C, 2015, 3, 5541–5548.
44. S. Holliday, R. S. Ashraf, A. Wadsworth, D. Baran, S. A. Yousaf, C. B. Nielsen, C.-H. Tan, S. D. Dimitrov, Z. Shang, N. Gasparini, M. Alamoudi, F. Laquai, C. J. Brabec, A. Salleo, J. R. Durrant and I. McCulloch, Nat. Commun., 2016, 7, 11585.
45. N. Li, F. Machui, D. Waller, M. Koppe and C. J. Brabec, Sol. Energy Mater. Sol. Cells, 2011, 95, 3465–3471.
46. C. M. S. Vicente, P. S. André and R. A. S. Ferreira, Rev. Bras. Ensino Fis., 2012, 34, 1–5.
47. C. Goh, R. J. Kline, M. D. McGehee, E. N. Kadnikova and J. M. J. Fréchet, Appl. Phys. Lett., 2005, 86, 122110.
48. G. G. Malliaras, J. R. Salem, P. J. Brock and C. Scott, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 58, R13411.
49. I. Riedel, J. Parisi, V. Dyakonov, L. Lutsen, D. Vanderzande and J. C. Hummelen, Adv. Funct. Mater., 2004, 14, 38–44.
50. P. Schilinsky, C. Waldauf and C. J. Brabec, Appl. Phys. Lett., 2002, 81, 3885–3887.
51. T. Kumari, M. Moon, S.-H. Kang and C. Yang, Nano Energy, 2016, 24, 56–62.
52. T. Ameri, P. Khoram, J. Min and C. J. Brabec, Adv. Mater., 2013, 25, 4245–4266.

---

Footnotes

† Electronic supplementary information (ESI) available: Additional figures (UV vis, J–V characteristics, surface energy measurement, EDAX, optical microscopy, AFM phase images, charge transport characteristics (SCLC), GIWAXS analyses and thickness measurement of the ternary OPVs) and table (crystallographic parameters of the ternary OPV thin films). See DOI: 10.1039/c6ee02851a

‡ T. K. and S. M. L. contributed equally to this work.

---