Improving quantum efficiency in organic solar cells with a small energetic driving force

Abstract

Organic solar cells based on a polymer donor (PM7) and a non-fullerene acceptor (Y5) with a very small energetic difference between the local excited state and the charge transfer (CT) state are investigated. We find that the small energetic difference (ΔE_CT) leads to a low voltage loss (0.44 eV). However, the short-circuit current density (J_sc) of the solar cell based on PM7:Y5 is very low, due to monomolecular recombination of the CT state excitons, limiting the internal quantum efficiency of the device. To solve the problem with the inefficient exciton dissociation, a polymer donor (PBDB-T) with a similar chemical structure to PM7 is employed as a second donor component for constructing ternary solar cells. We find that the frontier energy levels of the two donors are hybridized, allowing us to realize fine-tuning of the effective energy of the CT state and ΔE_CT of the ternary blend, by varying the PBDB-T content. As a result, a significantly improved CT state dissociation efficiency is achieved by adding a small amount of PBDB-T in the active layer. Meanwhile, the low voltage loss property of PM7:Y5 is very well maintained in the ternary solar cell, due to the energy level hybridization of the donor materials.

---

1. Introduction

The development of non-fullerene acceptor materials in the last few years has increased the power conversion efficiency (PCE) of the bulk-heterojunction (BHJ) organic solar cells up to 17–18%.^1–4 Many of the highly efficient solar cells are realized based on the ternary concept consisting of adding an additional donor or acceptor into the binary BHJ active materials system:^5–7 The use of the third component can often lead to improved spectral coverage for photon harvesting and a more desired phase separation in the active layer for a more efficient charge carrier collection. Today, high quantum efficiencies (QE) are observed in organic solar cells based on different donor–acceptor combinations, and the short-circuit current density (J_sc) of the most efficient organic solar cells has already exceeded 25 mA cm⁻².^8–10 Thus, it has become increasingly more important to increase the open-circuit voltage (V_oc) to further improve the performance of organic solar cells.

Minimizing the voltage lost in the indispensable charge transfer process in organic solar cells is crucial. Generally, voltage losses (V_loss) in organic solar cells can be divided into three parts.^11–18 ΔE_CT, ΔV_r, and ΔV_nr. ΔE_CT (also known as the energetic driving force) is defined as the energetic difference between the charge transfer (CT) state and the local excited state of the donor or the acceptor material (S₁). The typical value for ΔE_CT is in the range of 0.05 to 0.2 eV for the most efficient organic solar cells. ΔV_r and ΔV_nr are the voltage loss terms related to the radiative and non-radiative decay rate of CT states, respectively.^15,19,20 Because the CT state can be regarded as the lowest energy of state in the blend photoactive layer,^21,22 increasing the energy of CT state (E_CT) for a reduced ΔE_CT can directly lead to a reduced V_loss in the solar cell. Besides, the higher E_CT can result in a reduced ΔV_nr, due to the weakened electronic coupling between the vibrational excited and ground states.¹⁷ Furthermore, the coupling between S1 and CT state is stronger in the blend system with a higher E_CT and a smaller ΔE_CT. This can lead to a further weakened CT to ground state coupling.^16,19 As a result, the non-radiative decay rate of CT states in the blend system with a smaller ΔE_CT is slower, and ΔV_nr of the solar cell with a smaller ΔE_CT is lower.

However, the strong coupling between S1 and CT states in the blend active layer with a too-small ΔE_CT can lead to inefficient exciton dissociation, which limits QE and thus J_sc of the solar cell. This has been observed in solar cells based on several different BHJ systems.^23–26 The inefficient exciton dissociation can be ascribed to a rapid back transfer of CT states to S1 states, enabled by the small ΔE_CT at the heterointerfaces.²⁷ This reduces the probability for the CT state to dissociate into free charge carriers, since the energy of the free charge carrier state (E_FC) is generally higher than that of the CT state.²⁸ Therefore, there is always a tradeoff between QE and V_loss which must be balanced to construct efficient organic solar cells using the blends with small ΔE_CT. Currently, there is no solution for the poorly performing solar cells with QE limited by a too-small ΔE_CT, apart from designing new materials with different frontier energy levels.

In this work, a ternary blend strategy is employed to solve the high QE loss problem in the low V_loss organic solar cells based on low ΔE_CT systems. First, we perform a detailed investigation on the organic blend system based on poly[4,8-bis[5-(2-ethylhexyl)-4-chloro-2-thienyl]benzo[1,2-b:4,5-b′]dithiophene-2,6-diyl]-2,5-thiophenediyl[5,7-bis(2-ethylhexyl)-4,8-dioxo-4H,8H-benzo[1,2-c:4,5-c′]dithiophene-1,3-diyl]-2,5-thiophenediyl] (PM7) and (2,2′-((2Z,2′Z)-((12,13-bis(2-ethylhexyl)-3,9-diundecyl-12,13-dihydro[1,2,5]thiadiazolo[3,4e]thieno[2′′,3′′:4′,5′]thieno[2′,3′:4,5] pyrrolo[3,2-g]thieno[2′,3′:4,5] thieno[3,2-b]indole-2,10-diyl)bis(methanylylidene))bis(3-oxo-2,3-dihydro-1H-indene-2,1-diylidene))dimalononitrile) (Y5) with a very small ΔE_CT (close to 0 eV), and we identify a very low V_loss (0.44 V) in the solar cell associated with the small ΔE_CT. However, the ΔE_CT is too-small, providing an insufficient driving force for CT states to dissociate. This limits QE and J_sc of the solar cell. To overcome this problem, poly[(2,6-(4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)-benzo[1,2-b:4,5-b’]dithiophene))-alt-(5,5-(1′,3′-di-2-thienyl-5′,7′- bis(2-ethylhexyl)benzo [1′,2′-c:4′,5′-c’]dithiophene-4,8-dione) (PBDB-T) with a chemical structure similar to PM7 is introduced as a third component for constructing ternary solar cells. We find that the addition of PBDB-T can lead to a hybridization of the frontier orbitals of the donor materials. As a result, fine-tuning of the effective E_CT and ΔE_CT of the ternary solar cell is achieved by varying the PBDB-T content: a significantly improved CT state dissociation efficiency, and thus an increased QE is realized by adding a small amount of PBDB-T ((PBDB-T : PM7 donor ratio = 10 wt%)) in the photoactive layer. More importantly, we find that the low V_loss property of the PM7:Y5 blend is well kept in the ternary solar cell with a low PBDB-T content, due to the hybridization of the energy levels of the donor materials.

2. Results and discussion

Fig. 1a shows the chemical structures of the active materials used in this work. First, we focus on investigating the V_oc (0.98 V, Table 1) of the solar cell based on PM7 : Y5 (ref. 29 and 30) (1 : 1.2, weight ratio), which is higher than any of the other solar cells based on PM7 and the Y-series acceptors reported in literature.^31,32 To understand the reason for the high V_oc, optical transitions in the BHJ system is investigated. The absorption properties of the blend and the pure acceptor material are studied using sensitive external quantum efficiency (EQE) and electroluminescence (EL) measurements, as shown in Fig. 1b. From the onset of the EQE spectrum, we determine E_CT, using a Gaussian fit derived based on the Marcus theory:^12,33,34where E is photon energy, k is the Boltzmann constant, T is temperature, λ is the reorganization energy, and f is a pre-factor, proportional to the density of the ground states formed at the donor–acceptor interfaces and the absorption oscillator strength of CT state (f_osc).

Fig. 1 (a) Chemical structures of the active materials used in this work. (b) EQE and EL of the solar cell based on PM7 : Y5 (1 : 1.2 wt ratio). The tail of the EQE spectrum is calculated from the measured EL using the reciprocal relation, and it is used to calculated E_CT based on the Marcus theory. V_oc,rad is also calculated by integrating the product of EQE and ϕ_BB, and ΔV_nr is equal to the difference between V_oc,rad and V_oc. To avoid an arbitrary fitting on the EQE spectrum, two boundary conditions are applied. Details are provided in ESI-2.† (c) Dependence of J_sc of the solar cell on illumination intensity. A slope of one is obtained indicating monomolecular recombination limiting J_sc of the solar cell.

Table 1 Performance parameters of the solar cells based on PM7:Y5, PM7:PBDB-T(10%):Y5, and PBDB-T:Y5. Details regarding the determination of these parameters are provided in ESI-2 (see ESI)

	J sc (mA cm^−2)	V oc (V)	PCE (%)	E S1,A (eV)	E CT (eV)	ΔECT (eV)	V loss (V)	J 0,rad (mA cm^−2)	V oc,rad (V)	ΔVnr^a (V)	ΔVnr^b (V)	EQEEL (%)
a ΔVnr calculated from J0,rad. b ΔVnr calculated from measured EQEEL.
PM7:Y5	7.8	0.98	3.28	1.42	1.39	0.03	0.44	1.5 × 10^−18	1.10	0.12	0.13	0.751
PM7:PBDB-T(10%):Y5	12.0	0.95	5.27	1.42	1.39	0.03	0.47	1.3 × 10^−18	1.10	0.15	0.16	0.142
PBDB-T:Y5	17.2	0.88	9.04	1.42	1.37	0.05	0.54	2.7 × 10^−18	1.08	0.21	0.22	0.015

---

More specifically, to avoid an arbitrary fitting on the EQE spectrum, a series of fittings are performed: radiative recombination limit of the voltage loss derived using the detailed balance theory³⁵ is employed for determining the minimum value for E_CT, while the upper limit for E_CT is restricted by the assumption that the contribution of the weak CT state absorption to the EQE of the solar cell is at most 10%.^12,36 These boundary conditions allow us to identify a range for the values of E_CT with real physical meaning, and the upper and the lower boundary values are found to differ by only about 0.02–0.04 eV. For the analysis of the energetics and voltage losses in the solar cells studied in this work, the mean value of the range is used. More details regarding the fitting of the EQE spectrum are provided in ESI (SI-2).†

The value of E_CT from the EQE spectrum for the solar cell based on PM7:Y5 is found to be very close to the energy of the optical bandgap (E_g) of the active layer, which is equivalent to the energy of the singlet excited state of the acceptor material (E_S1,A) (ESI-3†).³⁷ Therefore, ΔE_CT is very small, approaching zero.

Furthermore, the radiative recombination limit for the saturation current (J_0,rad) is also calculated from the EQE spectrum using the detailed balance theory:^11,38

where q is the elementary charge and ϕ_BB is the black body spectrum at 300 K. The calculated values are listed in Table 1. From J_0,rad, we further calculate V_oc,rad, the maximum V_oc of the solar cell in which only radiative recombination exists.

We find that V_oc,rad of the solar cell based on PM7:Y5 is 1.10 V, thus ΔV_nr is very low, only 0.12 V (ΔV_nr = V_oc,rad– V_oc). Therefore, recombination of charge carriers in the solar cell is expected to be highly emissive. The low ΔV_nr of the solar cell and the high emission efficiency of CT state recombination are also confirmed by directly measuring the external quantum efficiency of electroluminescence (EQE_EL) of the solar cell, which is close to 1%, corresponding to a ΔV_nr value of 0.13 V ^12,38 The value of ΔV_nr from EQE_EL agrees very well with that calculated from V_oc,rad. Consequently, due to the small ΔE_CT and low ΔV_nr, the total voltage loss defined as is also very low for the solar cell based on PM7:Y5 (0.44 V), considerably lower than the generally reported V_loss for organic solar cells (0.5–0.6 V).^39,40

J _sc of the solar cell based on PM7:Y5 is rather low, because IQE of the solar cell calculated using a transfer matrix model is only about 30% (ESI-4†).^41,42 For organic solar cells, a low IQE is normally due to the bimolecular recombination loss of charge carriers.^43–46 However, fast bimolecular recombination in the solar cell often leads to a low EQE_EL and a high ΔV_nr, contradictory to the experimental results discussed above. Therefore, light intensity dependent current density–voltage (J–V) (Fig. 1c) measurements are performed on the solar cell based on PM7:Y5, and we find a linear dependence of J_sc on the light intensity, which suggests that the IQE and J_sc of the solar cell are limited by inefficient charge carrier generation, a result of monomolecular recombination losses.^47,48

The monomolecular recombination loss in BHJ solar cells can be associated with inefficient exciton dissociation in the active layer. This is indeed observed for the PM7:Y5 blend system in the PL quenching measurements. As shown in Fig. 2a, the characteristic PL emission spectra of the neat PM7 and the neat Y5 film are peaked at 680 and 920 nm, respectively. In the blend active layer of PM7:Y5, the PM7 emission peak is found completely quenched, but the Y5 emission peak is only slightly reduced. This suggests that the singlet excitons generated in the Y5 phase of the blend active layer do not dissociate into free charge carriers. However, the atomic force microscopic image reveals that there is no obvious phase separation in the active layer (ESI-5†), and the efficient quenching of the PM7 PL emission implies that the degree of phase separation in the active layer is low. Therefore, the reason for the inefficient dissociation of the singlet Y5 excited state (S_1,A) is ascribed to the too-small ΔE_CT providing an insufficient driving force for the already formed CT state excitons to dissociate, which is very common for the BHJ systems with small ΔE_CT.^23–26 In this case, an electric field-dependent PL quenching of the excited state is expected, because dissociation of the weakly bound CT excitons is often facilitated by an electric field, resulting in less populated S_1,A states, and thus reduced S_1,A emission intensity.^23,24

Fig. 2 (a) PL spectra of pure PM7, pure Y5, and the PM7:Y5 blend, measured with laser excitation at 460 nm. The emission peaks of pure PM7 and pure Y5 are located at 680 and 920 nm, respectively, and the PL of the blend is dominated by the Y5 emission. (b) Emission intensity values of the PL peaks of the solar cells based on pure PM7, pure Y5, and the PM7:Y5 blend plotted as a function of applied voltage. The intensity values are normalized by the intensity value of the peak measured under short circuit conditions. Complete spectra are provided in ESI-6.†

The field-dependent PL spectra are measured for the solar cell based on PM7:Y5, as well as the photovoltaic device based on pure Y5 (ESI-6†). In Fig. 2b, the intensities of the emission peak are plotted as a function of applied voltage. We find that for the solar cell based on the blend active layer, the Y5 emission peak at 920 nm is effectively quenched by the applied voltage, whereas the degree of the electric field induced quenching of the emission from the device based on pure Y5 is significantly smaller. These field-dependent PL results suggest that CT states are indeed formed in the solar cell based on PM7:Y5, but they do not easily dissociate into free charge carriers under a short-circuit condition.

The chemical structure of PBDB-T⁴⁹ is similar to PM7, but the highest-occupied-molecular-orbital (HOMO) level of PBDB-T (−5.3 eV) is slightly upper laying compared to that of PM7 (−5.4 eV).²⁹ Therefore, a higher ΔE_CT exists in the solar cell based on the blend of PBDB-T:Y5, allowing for more efficient dissociation of CT excitons and higher J_sc (Table 1). However, V_loss in the solar cell is significantly higher, not only because of the higher ΔE_CT but also a higher ΔV_nr.

Therefore, to overcome the problem of inefficient CT dissociation in the low V_loss solar cell based on PM7:Y5, PBDB-T is employed as a third component, acting as an additional donor for constructing ternary solar cells. Note that the ternary solar cells based on PBDB-T:PM7:Y5 are with a fixed donor : acceptor weight ratio (1 : 1.2). The representative J–V characteristic curves of the solar cells are provided in ESI-7,† and J_sc are plotted as a function of the PBDB-T content in Fig. 3a. Surprisingly, we find that with the addition of a very small amount of PBDB-T (PBDB-T : PM7 donor ratio = 10 wt%), J_sc of the solar cell is increased by 30%, from 7 to 10 mA cm⁻², which is further doubled, when the donor ratio is increased to 20%. The J_sc of the solar cell does not noticeably increase when the donor ratio is increased to over 20%, and when the donor ratio exceeds 40%, J_sc is reduced.

Fig. 3 (a) J_sc and V_oc of the solar cells based on PBDB-T:PM7:Y5, plotted as a function of the PBDB-T : PM7 donor ratio. (b) PL emission spectra of the solar cell based on different PBDB-T : PM7 donor ratios. More efficient quenching of the Y5 emission is found in the solar cell with higher PBDB-T content. (c) Emission intensities of the PL peaks of the solar cells based on different PBDB-T : PM7 donor ratios plotted as a function of applied voltage. The intensity values are normalized by the intensity value of the peak measured under short circuit conditions. The dependence of the PL quenching efficiency on the applied voltage reduces with the increasing PBDB-T content. Complete spectra are provided in ESI-6.†

These results indicate that the IQE of the ternary solar cell is increased significantly with the PBDB-T : PM7 donor ratio exceeding 20%, which is remarkable as it suggests that adding PBDB-T into the PM7:Y5 blend can lead to substantially more efficient CT state dissociation. Indeed, from Fig. 3b, we find that the PL peak of the blend at 920 nm, corresponding to Y5 emission, is much more efficiently quenched, and the dependence of the PL intensity on the applied electric field is significantly reduced in the ternary solar cell with PBDB-T (Fig. 3c).

There are two possible mechanisms for the increased dissociation efficiency of the CT excitons in the ternary blend:⁵⁰ (1) There is no interaction of the frontier orbitals of the different donor materials, and most of the holes generated in Y5 are transferred to PBDB-T to contribute to free charge carrier generation. (2) There is a hybridization of the energy levels of PBDB-T and PM7, leading to effectively reduced E_CT and increased driving force for the whole system. This can result in a decreased electronic coupling strength between S_1,A and CT states, and thus reducing the probability for CT states to back transfer to S_1,A state, and increasing the dissociation probability of the CT states. Both mechanisms require a homogeneous distribution of PBDB-T in the blend active layer.

When there is no interaction between the orbitals of the donor materials in the ternary solar cells, the recombination dynamics of the photogenerated charge carriers should be primarily determined by the electronic property of the CT state formed at the PBDB-T:Y5 interfaces, even when the blend is with a low PBDB-T content since the PBDB-T:Y5 CT state is the lowest energy of state in the blend system. Therefore, with a small amount of PBDB-T in the ternary blend, V_oc of the solar cell should be similar to that of the binary solar cell based on PBDB-T:Y5. In addition, the characteristics of the recombination current of the ternary solar cell should be the same as that of the binary solar cell based on PBDB-T:Y5. Because in the ternary solar cell, the electrically injected holes are expected to rapidly relax to the HOMO of PBDB-T, and then find electrons at the PBDB-T:Y5 interfaces to recombine. However, we observe that V_oc of the ternary solar cell almost linearly decreases with the increasing PBDB-T content (Fig. 3a), and the recombination current in the solar cell gradually increases with the PBDB-T content (ESI-7†). These results indicate that the energy levels of PBDB-T and PM7 are hybridized to form an effective CT state in the ternary blend, which are also observed for other ternary blends reported in literature.⁵⁰

To elucidate that the energy levels of the PBDB-T and PM7 in the blend are hybridized, we further investigate the charge generation process in the devices based on PBDB-T : PM7. We find that the J_sc and EQE of the device based on PBDB-T : PM7 (1 : 1 wt%) are considerably higher than that of the device based on pure PBDB-T or pure PM7 (Fig. S11a and b, ESI†), but the absorption spectra of the two materials are very similar (Fig. S3b†). PL measurements, shown in Fig. S11c (ESI),† suggest that the emissions from PBDB-T and PM7 are significantly quenched, in the blend of PBDB-T:PM7. This suggests that the singlet excited states of PBDB-T and PM7 are hybridized, resulting in the formation of CT states at the interface between the two polymers, facilitating the dissociation of the excitons generated in the blend active layer. Sensitive electroluminescence and EQE measurements (Fig. S11d and e, ESI†) reveal that the absorption and emission spectra of the blend thin film are slightly red-shifted compared to those of the films based on pure PBDB-T or PM7, also confirming that the energy of the excited states, i.e., the energy levels of PBDB-T and PM7 are hybridized in the blend.

Therefore, we can also use the Marcus theory to evaluate the CT state property in the ternary solar cells (SI-2). We find that E_CT is indeed gradually reduced with the increasing PBDB-T content (Fig. 4a), leading to a continuously tuned ΔE_CT (Fig. 4b). The electronic coupling between the S_1,A and CT state is thus reduced.

Fig. 4 (a) E_CT and (b) ΔE_CT of the solar cells based on PBDB-T:PM7:Y5, plotted as a function of the PBDB-T : PM7 donor ratio. (c) V_oc and V_loss of the solar cells based on PBDB-T:PM7:Y5, plotted as a function of the PBDB-T : PM7 donor ratio. Note that the donor : acceptor ratio for the ternary blends is also kept at 1 : 1.2 (weight ratio).

Inevitably, the V_oc of the ternary solar cell reduces with increasing PBDB-T content (Fig. 4c). However, due to the hybridization of the energy levels of PBDB-T and PM7, the V_oc does not directly reduce to that of the binary solar cell based on PBDB-T:Y5. Therefore, the low V_loss property of the solar cell based on PM7:Y5 can be well kept in the ternary solar cell, while QE of the solar cell is significantly increased, with the addition of a small amount of PBDB-T. As shown in Fig. 4c, V_loss is about 0.47 V in the ternary solar cell with a PBDB-T : PM7 donor ratio of 10%, which is slightly higher than that of the binary solar cell based on PM7:Y5 (0.44 eV), but the J_sc of the ternary solar cell is higher by a factor of 1.30; and the V_loss is increased to 0.50 V for the ternary solar cell with a donor ratio of 20%, while the J_sc is doubled. The gain of using the ternary blend strategy is remarkably high.

The increased V_loss in the ternary solar cell is partially due to the reduced effective E_CT, since E_CT is the electronic bandgap of the blend active layer.¹² Meanwhile, the reduced E_CT can lead to increased ΔV_nr due to a stronger vibrational coupling between the excited and ground state.¹⁷ In addition, the electronic coupling between the S_1,A and CT state is reduced due to the reduced E_CT: although the reduction in the electronic coupling is needed to prevent the back transfer of the CT state to the S_1,A state, the reduced coupling can also lead to a reduced the effective dipole moment for the CT to ground state transition,¹⁹ resulting in the increased non-radiative decay rate of CT state, reduced EQE_EL, and thus increased ΔV_nr in the solar cell. EQE_EL measurements (Fig. S8, ESI†) confirm that ΔV_nr in the ternary solar cell is indeed higher than that in the binary solar cell based on PM7:Y5, and it gradually increases with the PBDB-T content, as shown in Fig. 5a, and transient photovoltage decay time plotted as a function of bias illumination intensity confirms that the increased ΔV_nr is a result of the increased non-radiative decay rate of CT state (Fig. 5b).

Fig. 5 (a) ΔV_nr of the solar cells based on PBDB-T:PM7:Y5 with different PBDB-T : PM7 donor ratios. The ΔV_nr values are calculated from the measured V_oc and V_oc,rad determined from the sensitive EQE spectra, and they are verified by EQE_EL measurements. (b) Transient photovoltage decay lifetime measured with a pulsed LED and a bias illumination at different intensities for the solar cells with different PBDB-T : PM7 donor ratios. The x-axis represents the photovoltage of the solar cell generated by the bias illumination. To compare the decay lifetime at the same charge carrier concentration, the x-axis is corrected for the difference in E_CT of the solar cells based on different PBDB-T contents. Photovoltage vs. time plots for these solar cells are provided in ESI-9.†

3. Conclusion

To conclude, we investigated organic solar cells based on PM7:Y5, in which an extremely low V_loss (0.44 V) was found, and we showed that the low V_loss was a result of a small ΔE_CT and a low ΔV_nr. However, the QE of the solar cell was low, due to the small ΔE_CT, leading to inefficient dissociation of CT state excitons. To solve this problem, a ternary blend strategy was employed. We demonstrated that the use of PBDB-T with slightly upper laying HOMO, compared to PM7, as a second donor in the blend active layer could effectively lead to a hybridization of the energy levels of PBDB-T and PM7. This resulted in reduced E_CT and increased ΔE_CT in the ternary system, giving rise to a significantly improved CT state dissociation efficiency in the active layer, and thus an increased J_sc of the solar cell. More importantly, the low V_loss property of the PM7:Y5 based blend was very well maintained: J_sc of the ternary solar cell with a PBDB-T : PM7 donor ratio of 20% was increased by two folds, compared to that of the binary solar cell based on PM7, but V_loss was only increased by 0.06 eV. Thus, we propose that the ternary blend strategy is highly effective for realizing a balance between QE and V_loss in organic solar cells with a small ΔE_CT. To achieve a high degree of hybridization of the energy levels of the active materials, and the tuning of ΔE_CT of the ternary blend, the second donor should have different energy levels compared to the primary donor, and the chemical structures of the donor materials should be similar to allow a high degree of intermixing in the ternary blend. This is a newly discovered advantage of the ternary blend strategy, which has not yet been fully explored.

4. Experimental

Materials and devices

PEDOT:PSS (CLEVIOS P VP AL 4083) was purchased from Heraeus, PBDB-T, PM7, Y5 and PFN-Br were purchased from Solarmer. The solar cells are based on a standard architecture of ITO/PEDOT:PSS active layer/PFN-Br/Ag. The indium tin oxide (ITO) substrates were sonicated with acetone, isopropanol, and ethanol, and then rinsed with deionized water. PEDOT:PSS interlayers were spin-coated on the clean ITO substrates at a spin-coating speed of 5000 rpm for 60 s, and annealed on a hotplate at 150 °C for 20 min. The thickness of the PEDOT:PSS interlayers was about 30 nm. The organic photoactive layers (100 nm) were spin-coated onto the PEDOT:PSS coated substrate in a glovebox filled with nitrogen, at a speed of 2000 rpm for 30 s, from chloroform solutions with a total concentration of the active materials of 8 mg mL⁻¹ (the donor : acceptor weight ratio is kept to 1 : 1.2 for both the binary and the ternary systems). The solutions were stirred at 40 °C for 4 hours, prior to use, and the active layers were annealed at 110 °C for 10 min in the glove box. Then, the PFN–Br interlayers (1 nm) were spin-coated on the active layers at a spin-coating speed of 3000 rpm for 30 s. Construction of the solar cells was finalized by a thermal evaporation of 100 nm Ag electrode in a vacuum chamber at a pressure of 1 × 10⁻⁶ Pa. The photoactive area of the solar cells is about 4 mm².

Characterizations

Details regarding characterization of the materials and devices are provided in ESI (ESI-1).†

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This work is financial supported by the National Natural Science Foundation of China (Grant No. 51973031, 51933001, 52073056), Shanghai Pujiang Program (Grant No. 19PJ1400500), the Natural Science Foundation of Shanghai (Grant No. 19ZR1401400) and the Fundamental Research Funds for the Central Universities (Grant No. 2232021A09).

References

1. Q. Liu, Y. Jiang, K. Jin, J. Qin, J. Xu, W. Li, J. Xiong, J. Liu, Z. Xiao, K. Sun, S. Yang, X. Zhang and L. Ding, Sci. Bull., 2020, 65, 272–275.
2. H. N. Tran, S. Park, F. T. A. Wibowo, N. V. Krishna, J. H. Kang, J. H. Seo, H. Nguyen-Phu, S. Jang and S. Cho, Adv. Sci., 2020, 7, 2002395.
3. L. Meng, Y. Zhang, X. Wan, C. Li, X. Zhang, Y. Wang, X. Ke, Z. Xiao, L. Ding, R. Xia, H.-L. Yip, Y. Cao and Y. Chen, Science, 2018, 361, 1094–1098.
4. Y. Cui, H. Yao, J. Zhang, K. Xian, T. Zhang, L. Hong, Y. Wang, Y. Xu, K. Ma, C. An, C. He, Z. Wei, F. Gao and J. Hou, Adv. Mater., 2020, 32, 1908205.
5. Q. An, F. Zhang, J. Zhang, W. Tang, Z. Deng and B. Hu, Energy Environ. Sci., 2016, 9, 281–322.
6. N. Gasparini, A. Salleo, I. McCulloch and D. Baran, Nat. Rev. Mater., 2019, 4, 229–242.
7. P. Bi and X. Hao, Sol. RRL, 2019, 3, 1800263.
8. P. Chao, H. Chen, Y. Zhu, H. Lai, D. Mo, N. Zheng, X. Chang, H. Meng and F. He, Adv. Mater., 2020, 32, 1907059.
9. W. Gao, H. Fu, Y. Li, F. Lin, R. Sun, Z. Wu, X. Wu, C. Zhong, J. Min, J. Luo, H. Y. Woo, Z. Zhu and A. K. -Y. Jen, Adv. Energy Mater., 2020, 2003177.
10. J. Yuan, Y. Zhang, L. Zhou, G. Zhang, H.-L. Yip, T.-K. Lau, X. Lu, C. Zhu, H. Peng, P. A. Johnson, M. Leclerc, Y. Cao, J. Ulanski, Y. Li and Y. Zou, Joule, 2019, 3, 1140–1151.
11. K. Vandewal, K. Tvingstedt, A. Gadisa, O. Inganäs and J. V. Manca, Nat. Mater., 2009, 8, 904–909.
12. K. Vandewal, K. Tvingstedt, A. Gadisa, O. Inganäs and J. V. Manca, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 125204.
13. Z. Tang, J. Wang, A. Melianas, Y. Wu, R. Kroon, W. Li, W. Ma, M. R. Andersson, Z. Ma, W. Cai, W. Tress and O. Inganäs, J. Mater. Chem. A, 2018, 6, 12574–12581.
14. K. Vandewal, K. Tvingstedt and O. Inganäs, in Semiconductors and Semimetals, Elsevier, 2011, vol. 85, pp. 261–295.
15. M. Azzouzi, J. Yan, T. Kirchartz, K. Liu, J. Wang, H. Wu and J. Nelson, Phys. Rev. X, 2018, 8, 031055.
16. D. Qian, Z. Zheng, H. Yao, W. Tress, T. R. Hopper, S. Chen, S. Li, J. Liu, S. Chen, J. Zhang, X.-K. Liu, B. Gao, L. Ouyang, Y. Jin, G. Pozina, I. A. Buyanova, W. M. Chen, O. Inganäs, V. Coropceanu, J.-L. Bredas, H. Yan, J. Hou, F. Zhang, A. A. Bakulin and F. Gao, Nat. Mater., 2018, 17, 703–709.
17. J. Benduhn, K. Tvingstedt, F. Piersimoni, S. Ullbrich, Y. Fan, M. Tropiano, K. A. McGarry, O. Zeika, M. K. Riede, C. J. Douglas, S. Barlow, S. R. Marder, D. Neher, D. Spoltore and K. Vandewal, Nat. Energy, 2017, 2, 17053.
18. K. D. Rosenthal, M. P. Hughes, B. R. Luginbuhl, N. A. Ran, A. Karki, S.-J. Ko, H. Hu, M. Wang, H. Ade and T.-Q. Nguyen, Adv. Energy Mater., 2019, 9, 1901077.
19. F. D. Eisner, M. Azzouzi, Z. Fei, X. Hou, T. D. Anthopoulos, T. J. S. Dennis, M. Heeney and J. Nelson, J. Am. Chem. Soc., 2019, 141, 6362–6374.
20. K. Vandewal, Annu. Rev. Phys. Chem., 2016, 67, 113–133.
21. V. Coropceanu, X.-K. Chen, T. Wang, Z. Zheng and J.-L. Brédas, Nat. Rev. Mater., 2019, 4, 689–707.
22. K. Vandewal, K. Tvingstedt, J. V. Manca and O. Inganäs, IEEE J. Sel. Top. Quantum Electron., 2010, 16, 1676–1684.
23. K. Vandewal, Z. Ma, J. Bergqvist, Z. Tang, E. Wang, P. Henriksson, K. Tvingstedt, M. R. Andersson, F. Zhang and O. Inganäs, Adv. Funct. Mater., 2012, 22, 3480–3490.
24. Z. Ma, W. Sun, S. Himmelberger, K. Vandewal, Z. Tang, J. Bergqvist, A. Salleo, J. W. Andreasen, O. Inganäs, M. R. Andersson, C. Müller, F. Zhang and E. Wang, Energy Environ. Sci., 2013, 7, 361–369.
25. C. Yang, J. Zhang, N. Liang, H. Yao, Z. Wei, C. He, X. Yuan and J. Hou, J. Mater. Chem. A, 2019, 7, 18889–18897.
26. Y. Xie and H. Wu, Materials Today Advances, 2020, 5, 100048.
27. L. Benatto, M. de J. Bassi, L. C. W. de Menezes, L. S. Roman and M. Koehler, J. Mater. Chem. C, 2020, 8, 8755–8769.
28. S. Ullbrich, J. Benduhn, X. Jia, V. C. Nikolis, K. Tvingstedt, F. Piersimoni, S. Roland, Y. Liu, J. Wu, A. Fischer, D. Neher, S. Reineke, D. Spoltore and K. Vandewal, Nat. Mater., 2019, 18, 459–464.
29. H. Zhang, H. Yao, J. Hou, J. Zhu, J. Zhang, W. Li, R. Yu, B. Gao, S. Zhang and J. Hou, Adv. Mater., 2018, 30, 1800613.
30. J. Yuan, Y. Zhang, L. Zhou, C. Zhang, T.-K. Lau, G. Zhang, X. Lu, H.-L. Yip, S. K. So, S. Beaupré, M. Mainville, P. A. Johnson, M. Leclerc, H. Chen, H. Peng, Y. Li and Y. Zou, Adv. Mater., 2019, 31, 1807577.
31. M.-A. Pan, T.-K. Lau, Y. Tang, Y.-C. Wu, T. Liu, K. Li, M.-C. Chen, X. Lu, W. Ma and C. Zhan, J. Mater. Chem. A, 2019, 7, 20713–20722.
32. R. Ma, T. Liu, Z. Luo, Q. Guo, Y. Xiao, Y. Chen, X. Li, S. Luo, X. Lu, M. Zhang, Y. Li and H. Yan, Sci. China: Chem., 2020, 63, 325–330.
33. R. A. Marcus, J. Phys. Chem., 1989, 93, 3078–3086.
34. I. R. Gould, D. Noukakis, L. Gomez-Jahn, R. H. Young, J. L. Goodman and S. Farid, Chem. Phys., 1993, 176, 439–456.
35. W. Shockley and H. J. Queisser, J. Appl. Phys., 1961, 32, 510–519.
36. Z. Tang, Z. Ma, A. Sánchez-Díaz, S. Ullbrich, Y. Liu, B. Siegmund, A. Mischok, K. Leo, M. Campoy-Quiles, W. Li and K. Vandewal, Adv. Mater., 2017, 29, 1702184.
37. K. Vandewal, J. Benduhn and V. C. Nikolis, Sustainable Energy Fuels, 2018, 2, 538–544.
38. U. Rau, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 085303.
39. Y. Xu, H. Yao, L. Ma, J. Wang and J. Hou, Rep. Prog. Phys., 2020, 83, 082601.
40. J. Wang, H. Yao, Y. Xu, L. Ma and J. Hou, Mater. Chem. Front., 2021, 5, 709–722.
41. L. A. A. Pettersson, L. S. Roman and O. Inganäs, J. Appl. Phys., 1999, 86, 487–496.
42. G. F. Burkhard, E. T. Hoke, S. R. Scully and M. D. McGehee, Nano Lett., 2009, 9, 4037–4041.
43. J. Vollbrecht, V. V. Brus, S.-J. Ko, J. Lee, A. Karki, D. X. Cao, K. Cho, G. C. Bazan and T.-Q. Nguyen, Adv. Energy Mater., 2019, 9, 1901438.
44. G.-J. A. H. Wetzelaer, N. J. V. der Kaap, L. J. A. Koster and P. W. M. Blom, Adv. Energy Mater., 2013, 3, 1130–1134.
45. L. J. A. Koster, V. D. Mihailetchi and P. W. M. Blom, Appl. Phys. Lett., 2006, 88, 052104.
46. G. Lakhwani, A. Rao and R. H. Friend, Annu. Rev. Phys. Chem., 2014, 65, 557–581.
47. P. Hartnagel and T. Kirchartz, Adv. Theory Simul., 2020, 3, 2000116.
48. A. K. K. Kyaw, D. H. Wang, V. Gupta, W. L. Leong, L. Ke, G. C. Bazan and A. J. Heeger, ACS Nano, 2013, 7, 4569–4577.
49. D. Qian, L. Ye, M. Zhang, Y. Liang, L. Li, Y. Huang, X. Guo, S. Zhang, Z. Tan and J. Hou, Macromolecules, 2012, 45, 9611–9617.
50. W. Tress, B. Beyer, N. Ashari Astani, F. Gao, S. Meloni and U. Rothlisberger, J. Phys. Chem. Lett., 2016, 7, 3936–3944.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ta00576f

---