Investigation of supramolecular interactions between liquid crystals and PCBM for improved morphological stability in solar cells

Abstract

Supramolecular interactions between liquid crystals (LCs) of different chemical structures and PC₆₁BM molecules have been studied, showing that 4-cyano-4′-pentylterphenyl (5CT) containing electron-withdrawing cyano substituents on the phenyl ring exhibited the strongest interactions with PC₆₁BM, as revealed via density functional theory (DFT) calculations and experimental analysis based on Fourier transform infrared spectrometry (FTIR), differential scanning calorimetry (DSC) and polarized optical microscopy (POM). In contrast, 4-octyloxy-4′-cyanobiphenyl (8OCB) comprising an electro-donating octyloxyl group and dioctylterthiophene (8TTP8) with thiophene rings and long alkyl groups showed weaker interactions with PC₆₁BM. After electric field treatment at 600 V mm⁻¹ in an air environment, the P3HT:PC₆₁BM:8OCB specimen showed a higher power conversion efficiency (PCE) of 2.9% and a more stable morphology than P3HT:PC₆₁BM:8TTP8 with a PCE of 2.7%. Upon annealing at 150 °C for 1 h, 5CT is most effective in restricting the aggregation and crystallization of PC₆₁BM molecules, thus stabilizing the morphology of P3HT:PC₆₁BM. Moreover, the supramolecular interaction between LCs and PCBM could also influence the thermal stability in the narrow bandgap system of PTB7-Th:PC₇₁BM, with 8TTP8 showing the highest ability to restrict the depression of the PCE value.

---

Introduction

Organic photovoltaics (OPVs) have made substantial progress recently and have shown their potential in low-cost, flexible, and lightweight solar energy conversion devices.^1,2 Power conversion efficiencies (PCEs) over 10% have been achieved for polymer-based OPVs with bulk heterojunction (BHJ) architectures.^3,4 Nevertheless, a further increase of their PCE is needed for future applications, and the long-term stability still remains to be a great challenge.

In order to solve the above problems, various approaches including chemical methods such as synthesis of novel donor and acceptor materials via molecular design and development of novel interfacial materials,^5–7 and physical methods such as thermal and solvent annealing,^8,9 and the addition of third components have been demonstrated to improve the PCE and thermal stability of OPVs. It is noted that the third components such as light-absorbing small molecules¹⁰ or polymers,¹¹ fullerene¹² or non-fullerene acceptors,¹³ metal- or carbon-based nanomaterials,^14,15 quantum dots¹⁶ and polymer or small molecule nonvolatile additives^17,18 can improve the absorption behavior or optimize the phase separation of the active layers. Among the different types of third additives, liquid crystalline molecules such as 2,3,6,7,10,11-hexaacetoxytriphenylene discotic liquid crystals (DLCs) can improve the PCE from 3.03% to 3.97% at a weight ratio of 3 wt% in the poly(3-hexyl thiophene) (P3HT) and [6,6]-phenyl-C₆₁-butyric acid methyl ester (PC₆₁BM) system.¹⁹ Meanwhile, the nematic liquid crystals (NLCs) were also demonstrated to improve the performance of P3HT:PC₆₁BM solar cells.²⁰ It is well accepted that liquid crystals (LCs) can facilitate the crystallization of P3HT, leading to an improvement in absorption and charge transportation. However, the interactions between LCs and PC₆₁BM molecules are not well understood. Jeong et al.²¹ speculated that the incorporation of nematic liquid crystals into the P3HT:PC₆₁BM system may lead to the formation of slightly larger PC₆₁BM domains according to the X-ray diffraction spectrum. However, they did not have more evidence to confirm their speculations. In our previous study,²² we found that there existed specific interactions between the 4-cyano-4′-pentylterphenyl (5CT) and PC₆₁BM molecules, leading to the disappearance of PC₆₁BM melting peaks at the 5CT weight ratio above 6 wt% in PC₆₁BM:5CT binary blends as revealed by differential scanning calorimetry (DSC). Additionally, 5CT and PC₆₁BM could form rod-like complexes as observed using transmission electron microscopy (TEM). The reason why noncovalent interactions exist between 5CT and PC₆₁BM molecules is not known. We believe that that the complexes might originate from the supramolecular interactions between liquid crystals and PC₆₁BM molecules, depending on the chemical structure of the liquid crystals.

The supramolecular concept was first proposed by Lehn in 1978,^23,24 and has been extensively studied by scientists, as it plays a vital role in the aggregation and self-assembly of molecules. As one of the supramolecular interactions, the π–π interaction widely exists in systems containing aromatic rings.²⁵ Nakamura²⁶ found that there was a favorable face-to-face interaction between a perfluoroaromatic ring and the π surface of fullerene. Moreover, Cheng²⁷ found that the supramolecular interaction between [6,6]-phenyl-C₆₁ butyric acid pentafluorophenyl ester (PC₆₁BP^F) and the C₆₀ cores of PC₆₁BM can effectively suppress the PC₆₁BM materials from extensive thermal-driven aggregation to overcome morphological stability.

In this article, we aimed to demonstrate the influence of the chemical structure of LCs on the strength of the supramolecular interactions between LCs and PC₆₁BM. Additionally, the effect of LCs on the morphology, photovoltaic performance and morphological stability of solar cells was extensively investigated. It is revealed that the 5CT containing electron-withdrawing substituents of the cyano group (CN) on the benzene ring is the most effective in restricting the aggregation and crystallization of the PC₆₁BM molecules, thus stabilizing the morphology of the active layer and improving the stability of the solar cells. The P3HT:PC₆₁BM specimen containing 4-octyloxy-4′-cyanobiphenyls (8OCB) exhibited a higher short circuit current (J_sc) and PCE than those of dioctylterthiophene (8TTP8) after electric field treatment, showing a stronger interaction of 8OCB with PC₆₁BM in contrast to 8TTP8. However, in the narrow bandgap system of PTB7-Th:PC₇₁BM, 8TTP8 was found to be the most effective in enhancing the thermal stability. Besides the supramolecular interaction between the LCs and PCBM, the interaction between the liquid crystals and the donor polymer should also be taken into consideration to improve the thermal stability of the solar cells.

Results and discussion

In our previous study, the 5CT liquid crystal was found to induce the crystallization of P3HT and restrict the crystallization of the PC₆₁BM molecules.²² It is well known that PC₆₁BM contains phenyl rings in the molecular structure. Based on Cheng's findings,²⁷ there exist supramolecular perfluorophenyl–C₆₀ interactions in the active layer. Moreover, Sherrill²⁸ found that the dimmers containing electron-withdrawing and electron-donating substituents bind more strongly than the benzene dimmer, based on the second-order perturbation theory (MP2). The electron-withdrawing substituents stabilize the transition state by decreasing the repulsion between the π electrons on each aryl ring, while electron-donating substituents destabilize the transition state by increasing the repulsion between the two π systems. Thus, the benzene containing electron-withdrawing substituents should have stronger interactions with benzene, due to the electrostatic interactions. The benzene containing electron-donating substituents should bind more weakly with benzene due to the electrostatic repulsions. However, the interactions may be slightly stronger than the benzene–benzene interactions when considering the dispersion interactions. In this article, as shown in Fig. 1, the 5CT molecule contains one CN group, one pentyl group and three benzene rings, in contrast to the 8OCB molecule containing one CN group, one octyloxyl group and two benzene rings. It is noted that the 5CT molecule contains more phenyl rings than the 8OCB molecule, in addition to the shorter alkyl group and the absence of electron-donating substituents. The electron-donating octyloxyl group in 8OCB may depress the electron-withdrawing effect of the CN group. Therefore, 5CT is believed to have stronger interactions with the PC₆₁BM molecules than 8OCB. Where 8TTP8 is concerned, the thiophene ring exhibiting a π₅⁶ structure with a higher electron density shows a stronger electrostatic repulsion than the phenyl ring. The relatively longer alkyl group may expand the distance between the thiophene ring and the phenyl ring. It is speculated that the interaction between 8TTP8 and PCBM should be weaker than that of 5CT and 8OCB.

Fig. 1 The chemical structure of different liquid crystals.

In order to confirm the above speculation, we performed theoretical calculations based on the density functional theory (DFT).²⁹ The initial configurations of the three complexes including PC₆₁BM:5CT, PC₆₁BM:8OCB and PC₆₁BM:8TTP8 are shown in Fig. 2, where a pentagon of PC₆₁BM is parallel to 5CT, 8OCB and 8TTP8, while the intermolecular distances were set to be 3.5 Å. The configurations of the complexes were then optimized using DFT at the B3LYP/6-31G** level. In order to calculate the binding energies between the LCs and PC₆₁BM, the single point energies of these optimized complexes were computed considering the Basis Set Superposition Error (BSSE), and the corresponding result is shown in Table 1. The binding energy of PC₆₁BM:5CT, PC₆₁BM:8OCB and PC₆₁BM:8TTP8 is calculated to be −6.19 eV, −3.98 eV and −4.20 eV, respectively. It is concluded that the interaction between 5CT and PCBM should be the highest. The binding energy between PC₆₁BM:8OCB and PC₆₁BM:8TTP8 is close to each other. However, by taking into account the alkyl groups in 8TTP8, the relatively longer alkyl groups in 8TTP8 may depress the interaction with PC₆₁BM, thus showing the weakest ability to restrict the crystallization of PC₆₁BM. The theoretical calculation result is consistent with the above analysis.

Fig. 2 (A) Initial configurations and (B) optimized configurations of the PC₆₁BM:5CT, PC₆₁BM:8OCB and PC₆₁BM:8TTP8 complexes.

Table 1 The calculated binding energies (E_b, in kcal mol⁻¹) of the three optimized complexes

	PC61BM:5CT	PC61BM:8OCB	PC61BM:8TTP8
E b (kcal mol^−1)	−6.19	−3.98	−4.20

---

Furthermore, we performed experimental investigations on the supramolecular interactions between the liquid crystals and the PC₆₁BM molecules. The Fourier transform infrared spectroscopy (FTIR) spectra (Fig. S1, ESI†) of the binary blends based on PC₆₁BM:8TTP8, PC₆₁BM:8OCB and PC₆₁BM:5CT present a novel absorption peak at 752 cm⁻¹, ascribed to the supramolecular interaction between the liquid crystals and the PC₆₁BM molecules. However, the influence of the chemical structure of the liquid crystals on the strength of the supramolecular interactions is unobvious based on the FTIR analysis.

The interactions between the LCs and PC₆₁BM were then investigated via differential scanning calorimetry (DSC) analysis. According to the previous result, the melting peak of PC₆₁BM almost disappeared at the 5CT weight ratio above 6 wt%, illustrating the strong supramolecular interaction.²² In the PC₆₁BM:8TTP8 blend as shown in Fig. 3a, the liquid crystal phase transition temperature for 8TTP8 remained almost unchanged after blending with PC₆₁BM, indicating that 8TTP8 could crystallize in its own region. The melting peak of PC₆₁BM remains obvious, shifting gradually to lower temperatures as the 8TTP8 weight ratio is increased to 20 wt%. In addition, the melting peaks of 8OCB change from three sharp peaks to one major round peak after blending with PC₆₁BM in the PC₆₁BM:8OCB blend (Fig. 3b). The melting peak of PC₆₁BM becomes less obvious after the incorporation of 8OCB, in contrast to the disappearance of the melting peak of PC₆₁BM in the PC₆₁BM:5CT blends. In order to further confirm the interactions between the liquid crystals and the PC₆₁BM molecules, the DSC cooling curves are shown in Fig. 3c and d. In the PC₆₁BM:5CT blend, no discernible crystallization peak contributing to PC₆₁BM could be observed at the 5CT weight ratio of 10 wt%.²² In the PC₆₁BM:8TTP8 blend, the crystallization peak temperature of PC₆₁BM shifts from 240.6 °C to a lower temperature of 176.8 °C as the 8TTP8 weight ratio is increased to 20 wt%. Additionally, the crystallization peaks of the PC₆₁BM molecules are obvious even at high contents of 8TTP8. In contrast, the crystallization peak of PC₆₁BM in the PC₆₁BM:8OCB blends almost disappears at the 8OCB weight ratio of 10 wt%, indicating that the crystallization of the PC₆₁BM molecules becomes very difficult in the presence of 8OCB liquid crystals.

Fig. 3 DSC heating curves for (a) PC₆₁BM:8TTP8 and (b) PC₆₁BM:8OCB, and cooling curves for the (c) PC₆₁BM:8TTP8 and (d) PC₆₁BM:8OCB blends at different weight ratios of liquid crystals.

To further explore the effect of liquid crystals on the crystallization of the PC₆₁BM molecules, we investigated the morphology of the P3HT:PC₆₁BM:LC (0.05 : 1 : 0.06) films after annealing at 200 °C for 10 min via polarized optical microscopy (POM) (Fig. S2, ESI†). The P3HT:PC₆₁BM (0.05 : 1) specimen forms of large spherulites contributing to the PC₆₁BM crystals with clear interfaces. By the incorporation of 6 wt% 8TTP8, PC₆₁BM develops into fibrillar bundles of crystals. 8OCB could induce the formation of relatively small PC₆₁BM spherulites as compared with the P3HT:PC₆₁BM (0.05 : 1) specimen. Moreover, 5CT could seriously restrict the cold crystallization of PC₆₁BM showing only a few undeveloped PCBM spherulites.

Based on the above experimental data, it is concluded that there exists supramolecular interactions between the liquid crystals and PC₆₁BM molecules, depending on the chemical structure of the liquid crystals. On the DSC heating curves, the melting peak of PC₆₁BM almost disappeared at the 5CT weight ratio above 6 wt%.²² In contrast, the melting peak of PC₆₁BM decreased after the incorporation of 8OCB, while the melting peak of PC₆₁BM was still obvious even at the 8TTP8 weight ratio of 20 wt%. Similarly, on the DSC cooling curves, the crystallization peak of PC₆₁BM could not be discerned at a 5CT and 8OCB weight ratio above 6 wt%, while the crystallization peak of PC₆₁BM was still obvious in the PC₆₁BM:8TTP8 blend. It is believed that 5CT exhibits the strongest supramolecular interactions with PC₆₁BM, and 8TTP8 shows the weakest supramolecular interaction with PC₆₁BM, depending on the type of aromatic rings, electron-donating or electron-withdrawing substituents. According to the POM images, the diameter and shape of the PC₆₁BM crystallites are dependent on the nucleating density and the migration of the PC₆₁BM molecules. The existence of liquid crystals facilitates the nucleation of PC₆₁BM. Due to the relatively weak supramolecular interactions between 8TTP8 and PC₆₁BM, the PC₆₁BM molecules could migrate to the crystallization growth front of the PC₆₁BM crystals more easily, leading to the appearance of fibrillar bundles in a higher density. In the PC₆₁BM:8OCB film, the supramolecular interaction between 8OCB and PC₆₁BM should be stronger, the migration of PC₆₁BM molecules onto the crystallization growth front becomes difficult, leading to the appearance of spherulites with a smaller size. Due to the strongest interaction between 5CT and PC₆₁BM, the growth of the PC₆₁BM spherulites is seriously restricted by 5CT.

Taking the supramolecular interaction into consideration, the incorporation of liquid crystals into P3HT:PC₆₁BM is supposed to enhance the photovoltaic properties. In our previous study, the P3HT:PC₆₁BM:5CT (1 : 1 : 0.06) system showed the highest photovoltaic performance with a PCE of 3.5%, a short-circuit current (J_sc) of 9.87 mA cm⁻², an open circuit voltage (V_oc) of 0.621 V, and a fill factor (FF) of 56.6% under an external electric field strength of 600 V mm⁻¹.²²Fig. 4 presents the current–voltage (J–V) characteristics of the solar cells based on the P3HT:PC₆₁BM:8TTP8 and P3HT:PC₆₁BM:8OCB blends after electric field treatment of different strengths in an air environment, and the corresponding parameters are shown in Table 2. The device based on pristine P3HT:PC₆₁BM exhibits a PCE of 2.4%, with a J_sc of 5.94 mA cm⁻², V_oc of 0.618 V, and FF of 65.5%, respectively. When 6 wt% 8TTP8 and 8OCB were added into the P3HT:PC₆₁BM system, the PCE value slightly increases to 2.5%. Where the P3HT:PC₆₁BM:8TTP8 specimen is concerned, the PCE value reaches the maximal value of 2.7% as the electric field strength is increased to 600 V mm⁻¹. Then, the PCE significantly drops to 1.7% at the electric field strength of 1000 V mm⁻¹. Moreover, the influence of the electric field strength on the photovoltaic performance of the P3HT:PC₆₁BM:8OCB specimen is similar to that of the P3HT:PC₆₁BM:8TTP8 specimen, showing the highest PCE of 2.9% at an electric field strength of 600 V mm⁻¹. In general, the electric field treatment seems to play a vital role in improving the photovoltaic performance of the ternary blend system. The appropriate electric field strength is found to be 600 V mm⁻¹, which is the same as in our previous study.²² However, the photovoltaic performance of the devices seems to be dependent on the chemical structure of the liquid crystals. The specimen containing 8OCB exhibits a higher PCE value than the specimen containing 8TTP8 at an electric field strength of 600 V mm⁻¹. According to the relationship of short-circuit current versus electric field strength, it is noted that the incorporation of liquid crystals into P3HT:PC₆₁BM could enhance the J_sc value. Moreover, the J_sc value reached the highest at an electric field strength of 600 V mm⁻¹. The specimen containing 8OCB showed a higher J_sc value as compared to the specimen containing 8TTP8.

Fig. 4 Current–voltage (J–V) characteristics of solar cells based on (a) P3HT:PC₆₁BM:8TTP8 (1 : 1 : 0.06) and (b) P3HT:PC₆₁BM:8OCB (1 : 1 : 0.06) specimens after electric field treatment of different strengths. (c) Relationship of short-circuit current versus electric field strength of the P3HT:PC₆₁BM:8TTP8 (1 : 1 : 0.06) and P3HT:PC₆₁BM:8OCB (1 : 1 : 0.06) specimens after electric field treatment.

Table 2 Summary of the photovoltaic parameters of the P3HT:PC₆₁BM:8TTP8 and P3HT:PC₆₁BM:8OCB solar cells after electric field treatment of different strengths under AM 1.5G solar illumination

Specimens	Electric field strength (V mm^−1)	J sc (mA mm^−2)	V oc (V)	FF (%)	PCE (%)
P3HT:PCBM (1 : 1)	0	5.94	0.618	65.5	2.4
P3HT:PCBM:8TTP8 (1 : 1 : 0.06)	0	6.09	0.619	67.5	2.5
	400	6.03	0.612	69.7	2.6
	600	6.22	0.617	70.0	2.7
	1000	5.86	0.612	46.2	1.7
P3HT:PCBM:8OCB (1 : 1 : 0.06)	0	6.40	0.608	65.3	2.5
	400	6.36	0.612	68.0	2.6
	600	6.56	0.623	70.6	2.9
	1000	6.11	0.602	55.8	2.1

---

The morphology of the active layer before and after the electric field treatment was analyzed via transmission electron microscopy (TEM) and atomic force microscopy (AFM). As shown in Fig. 5, the morphology of the films displays a strong dependence on the electric field strength. Where the P3HT:PCBM:8TTP8 specimen is concerned, the black rod-like aggregates should contribute to the 8TTP8 crystals, due to the relatively weak interaction between 8TTP8 and PC₆₁BM. The width distribution of the nanofibrils is centered at around 50–70 nm. After the electric field assisted treatment at a strength of 400 V mm⁻¹, the morphology of the film shows no significant change as compared with the specimen without any electric field assisted treatment, showing that the width of the nanofibrils reduced to 30–50 nm. At an electric field strength of 600 V mm⁻¹, the black dots with a diameter of about 20 nm distribute homogeneously in the matrix. Upon increasing the electric field strength to 1000 V mm⁻¹, some black aggregates in the diameter range of about 80–140 nm could be observed, implying that the nanorods could orient to the direction of the electric field. In contrast, the P3HT:PC₆₁BM:8OCB specimen exhibits an obviously different morphology showing no bundles in the matrix. The black dots ascribed to 8OCB and the PC₆₁BM complexes distribute homogeneously in the matrix with sizes ranging from 40 to 60 nm. Upon treatment with an electric field of 400 V mm⁻¹, the diameter of the black dots slightly reduces and they are distributed more homogeneously in the matrix. At an electric field strength of 600 V mm⁻¹, the diameter of the black dots significantly reduces to about 20 nm. As the electric field strength continuously increases to 1000 V mm⁻¹, the diameter of the black dots shows no serious change. Based on the above analysis, we can conclude that 8OCB exhibits stronger supramolecular interactions with PC₆₁BM. Thus, the 8TTP8 molecules tend to self-aggregate into crystalline phases, while 8OCB tends to form complexes with PC₆₁BM due to supramolecular interactions, facilitating the morphological stability in the presence of an external force. Even at an electric field strength of 1000 V mm⁻¹, the morphology of the specimen containing 8OCB remains the same without significant change. The specimen containing 8TTP8 tends to form larger black dots, resulting in much lower values of J_sc and PCE.

Fig. 5 TEM images of the (a–d) P3HT:PC₆₁BM:8TTP8 (1 : 0.06) and (e–h) P3HT:PC₆₁BM:8OCB (1 : 0.06) specimens after electric field treatment of different strengths.

In order to further explore the meso-scale film morphology in the lateral direction of films before and after electric field (600 V mm⁻¹) assisted treatment, the AFM images are presented in Fig. S3 (ESI†). The pristine P3HT:PC₆₁BM film shows a relatively large phase-separated morphology with a root mean square (RMS) surface roughness of 7.18 nm. Where the specimens containing 8TTP8 and 8OCB are concerned, the phase domain size significantly reduces in contrast to the pristine P3HT:PC₆₁BM specimen. After electric field assisted treatment, the phase domain size of the films further decreases, in addition to the reduction of the RMS surface roughness of the films. For example, the RMS value of the specimen containing 8TTP8 decreases from 2.01 nm to 0.74 nm. Moreover, the RMS value of the specimen containing 8OCB decreases from 2.49 nm to 2.03 nm. It is believed that 8TTP8 tends to distribute more homogeneously in the matrix upon treatment with an electric field, leading to a reduction in the RMS value. However, due to stronger supramolecular interactions between 8OCB and PC₆₁BM, the morphology seems to be more stable, accompanied by less reduction in the RMS value. It may lead to the formation of PC₆₁BM clusters in appropriate sizes, facilitating the electron transport.

According to the above data, it is recognized that the supramolecular interactions between the liquid crystals and PC₆₁BM may stabilize the morphology of active layers. Fig. 6 shows the POM images of the P3HT:PC₆₁BM:LC (1 : 1 : 0.06) films after thermal annealing at 150 °C for 1 h. For comparison, the POM images of specimens prior to thermal annealing are presented in Fig. S4 (ESI†). The pristine P3HT:PC₆₁BM film shows no significant structure in the whole image, indicating mixing of P3HT and PC₆₁BM at the molecular level. After the incorporation of 8TTP8, bright dots can be noticed, attributed to the 8TTP8 crystals. In contrast, the specimen containing 8OCB shows smaller bright dots. However, the morphology of the specimen containing 5CT is similar to that of the pristine P3HT:PC₆₁BM film, showing a few bright dots. The above result indicates that 8TTP8 tends to self-assemble into the liquid crystalline phase while 5CT tends to form complexes with PC₆₁BM, demonstrating that 5CT exhibits the strongest supramolecular interaction with the PC₆₁BM molecules. In Fig. 6, it is observed that some needle-like crystals ascribed to PC₆₁BM distributes in the whole image of the P3HT:PC₆₁BM specimen. By the incorporation of 6 wt% 8TTP8, the density of the PC₆₁BM crystals slightly reduces, showing that the presence of 8TTP8 liquid crystals could restrict the crystallization of PC₆₁BM. As the P3HT:PC₆₁BM:8OCB specimen is concerned, the density of the PC₆₁BM crystals is much lower than that of pristine P3HT:PC₆₁BM and the P3HT:PC₆₁BM:8TTP8 specimens. In the P3HT:PC₆₁BM:5CT specimen, only a few PC₆₁BM crystals could be discerned, confirming that 5CT has the strongest supramolecular interaction with the PC₆₁BM molecules. It is speculated that 5CT could enhance the morphological stability of the active layer under thermal treatment.

Fig. 6 POM images of the specimens of (a) P3HT:PC₆₁BM, (b) P3HT:PC₆₁BM:8TTP8 (1 : 1 : 0.06), (c) P3HT:PC₆₁BM:8OCB (1 : 1 : 0.06) and (d) P3HT:PC₆₁BM:5CT (1 : 1 : 0.06) after annealing at 150 °C for 1 h. The images were taken after adding a quarter-wave optical filter.

To further explore the effect of liquid crystals on the thermal stability of devices based on the narrow bandgap poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b:4,5-b′]dithiophene-co-3-fluorothieno[3,4-b]thiophene-2-carboxylate] (PTB7-Th):[6,6]-phenyl-C₇₁-butyric acid methyl ester (PTB7-Th:PC₇₁BM) system, 8TTP8, 8OCB and 5CT were incorporated into PTB7-Th:PC₇₁BM and the corresponding parameters are illustrated in Table 3 and Fig. S5 (ESI†). It is noted that the PCE value of all the devices reduced significantly upon thermal treatment. The pristine PTB7-Th:PC₇₁BM exhibits a PCE value of 7.3%, which sharply decreases to 4.0% after heating at 150 °C for 20 min. In PTB7-Th:PC₇₁BM:8TTP8, the PCE value reduces from 7.0% to 4.4% upon thermal treatment, showing that the incorporation of 8TTP8 could slightly improve the thermal stability of PTB7-Th:PC₇₁BM. In PTB7-Th:PC₇₁BM:8OCB, the PCE value decreases from 7.5% to 3.3%. In contrast, the PCE value decreases from 6.8% to 3.6% in PTB7-Th:PC₇₁BM:5CT. It is believed that 8OCB and 5CT are not beneficial to the thermal stability of PTB7-Th:PC₇₁BM, especially for 8OCB. It should be pointed out that the change in V_oc upon heating is dependent on the chemical structure of the liquid crystals. It is observed that the V_oc continuously decreases due to heating time in the pristine PTB7-Th:PC₇₁BM and PTB7-Th:PC₇₁BM:8OCB systems. However, the reduction in V_oc upon heating time in PTB7-Th:PC₇₁BM:5CT seems to be slower. Surprisingly, the V_oc value in PTB7-Th:PC₇₁BM:8TTP8 remains constant irrespective of heating time. It is well accepted that the change in V_oc should be related to the morphology evolution of the active layer. The corresponding TEM images of the active layer before and after thermal treatment are shown in Fig. S6 and S7 (ESI†). It is noted that the specimens before thermal treatment exhibit an optimized phase-separated morphology, especially for pristine PTB7-Th:PC₇₁BM and specimens containing 8OCB and 8TTP8. PTB7-Th:PC₇₁BM:5CT shows obvious black regions contributing to the 5CT phase, which exhibit a relatively lower PCE. After thermal treatment, the phase domain size of PTB7-Th:PC₇₁BM:8OCB and PTB7-Th:PC₇₁BM:8TTP8 reduces, showing less obvious phase separation. However, the phase domain size of PTB7-Th:PC₇₁BM:5CT increases.

Table 3 Summary of the photovoltaic parameters of the PTB7-Th:PC₇₁BM:LCs solar cells after heating for different times under AM 1.5G solar illumination

Systems	Time (min)	V oc (V)	J sc (mA cm^−2)	FF (%)	PCE (%)
PTB7-Th:PC71BM (1 : 1.5)	0	0.781	16.34	57.0	7.3
	5	0.758	11.04	49.3	4.1
	10	0.751	12.00	47.0	4.2
	20	0.726	12.04	46.1	4.0
PTB7-Th:PC71BM:8TTP8 (1 : 1.5 : 0.06)	0	0.775	15.53	58.3	7.0
	5	0.788	10.12	49.5	4.0
	10	0.791	11.14	49.5	4.4
	20	0.787	10.99	50.4	4.4
PTB7-Th:PC71BM:8OCB (1 : 1.5 : 0.06)	0	0.773	15.60	61.8	7.5
	5	0.732	10.48	48.6	3.7
	10	0.714	10.09	50.0	3.6
	20	0.699	9.29	51.9	3.3
PTB7-Th:PC71BM:5CT (1 : 1.5 : 0.06)	0	0.780	14.00	62.0	6.8
	5	0.764	10.52	47.0	3.8
	10	0.744	10.20	47.6	3.6
	20	0.745	10.14	47.7	3.6

---

In the PTB7-Th:PC₇₁BM system, the phase separation between PTB7-Th and PC₇₁BM should be the main factor in determining the morphological stability upon thermal treatment. Due to the similarity in chemical structure between 8OCB and 5CT, they tend to reside in the PC₇₁BM phase. During the heating process, the stronger supramolecular interaction between 5CT and PC₇₁BM could restrict the serious aggregation of PC₇₁BM, showing better thermal stability than the specimen containing 8OCB. In contrast, 8TTP8 may distribute in the interface between PTB7-Th and PC₇₁BM, which could reduce the interfacial force and thus stabilize the morphology upon thermal treatment. It is concluded that the use of molecules to restrict the aggregation and crystallization of PCBM should be the major route to stabilizing the morphology of the active layer, enhancing the thermal stability in the systems containing semicrystalline polymers such as P3HT. In the systems containing narrow band gap polymers, how to reduce the interfacial force in order to improve the morphology stability as well as the thermal stability should be taken into consideration. The selection of appropriate molecules which could interact well with both the donor and acceptor molecules through π–π or hydrogen bond interactions should be important. Researchers should pay more attention to the supramolecular interaction between the third additive and the donor and acceptor to explore how to stabilize the morphology of the active layer upon thermal treatment, enhancing the thermal stability of organic solar cells.

Conclusions

The supramolecular interactions between liquid crystals and PCBM molecules have been investigated via theoretical calculations and experimental analyses, with the strength depending on the chemical structure of the liquid crystals. The DFT calculation results demonstrate that the binding energy between 5CT and PCBM was the highest. Based on the FTIR, DSC and POM analyses, 5CT exhibited the strongest interaction to restrict the crystallization of PC₆₁BM, while 8TTP8 seemed to show the weakest interaction to restrict the crystallization of PC₆₁BM. In the P3HT:PC₆₁BM:LC system, the specimen containing 8OCB showed a higher J_sc and PCE than that of 8TTP8 after annealing at different electric field strengths. The highest PCE of 2.9% was achieved for the P3HT:PC₆₁BM:8OCB specimen at an electric field strength of 600 V mm⁻¹, showing higher morphological stability under an external force due to the stronger interaction between 8OCB and PC₆₁BM. 5CT was found to seriously restrict the crystallization of PCBM after thermal annealing at 150 °C in the P3HT:PC₆₁BM specimen, attributing to the strongest supramolecular interactions. In the narrow bandgap PTB7-Th:PC₇₁BM system, 8TTP8 could slightly improve the thermal stability while 5CT and 8OCB are less effective. Besides the supramolecular interaction between the liquid crystals and PCBM, the interaction between the liquid crystals and donor polymers should also be taken into consideration. Corresponding research should be further carried out to explore the relationship between the chemical structure of the third additive and the thermal stability of other systems showing high PCE values.

Experimental

Materials

Regioregular P3HT (M_w = 48 300 g mol⁻¹, head-to-tail regioregularity >90%) and poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b:4,5-b′]dithiophene-co-3-fluorothieno[3,4-b]thiophene-2-carboxylate] (PTB7-Th) used in this study were purchased from Rieke Metals, Inc. and 1-Materail, respectively. PC₆₁BM (99.5% purity) and PC₇₁BM were supplied by Nano-C, whereas the indium tin oxide (ITO) glass was purchased from Delta Technologies Limited. Zinc acetate dihydrate (Zn(CH₃COO)₂·2H₂O) and ethanolamine (NH₂CH₂CH₂OH) were purchased from Sigma Aldrich. 4-Octyloxy-4′-cyanobiphenyls (8OCB), dioctylterthiophene (8-TTP-8) and 4-cyano-4′-pentylterphenyl (5CT) were supplied by Energy Chemical Co. Chlorobenzene (CB) and dichlorobenzene (DCB) were purchased from Aldrich. All the reagents were used directly as received without further purification.

Device fabrication

Solution-processed solar cells with the conventional device architecture of ITO/ZnO/P3HT:PC₆₁BM:LCs/MoO₃/Ag or ITO/ZnO/PTB7-Th:PC₇₁BM:LCs/MoO₃/Ag were fabricated according to the following procedure. The ITO glass was cleaned through sequential ultrasonic treatment in acetone, detergent, deionized water, and isopropanol, and then treated in an ultraviolet-ozone chamber for 10 min. After treating with UV/ozone for 20 min, the ZnO precursor solution was spin coated at 4000 rpm and the ZnO layer was generated at 220 °C in an ambient atmosphere. The active layers were spun at 800 rpm from solutions of P3HT:PC₆₁BM:LCs (1 : 1 : 0.06) and at 1000 rpm from solutions of PTB7-Th:PC₇₁BM:LCs (1 : 1.5 : 0.06) containing DIO (0.3 v%). The electric field assisted treatment for the P3HT:PC₆₁BM:LCs (1 : 1 : 0.06) specimens was then performed by applying various magnitudes of voltage during the solvent drying process for 30 min in air. After evaporation of the solvents, MoO₃ was deposited via sequential thermal evaporation of 7 nm followed by 90 nm of Ag.

Measurements

Differential scanning calorimetry (DSC). The miscibility between P3HT and the LCs, and interactions between PCBM and the LCs were characterized through differential scanning calorimetry (DSC) analysis. The P3HT/LCs and PCBM/LCs blends at different weight fractions of LCs were prepared via a solution blending method. The specimens of about 3 mg were heated to 300 °C at a rate of 10 °C min⁻¹. After 3 min, the specimens were cooled to room temperature at a rate of 10 °C min⁻¹. The liquid crystalline transition temperature of the LCs, as well as the melting temperature (T_m) of P3HT or PCBM, was analyzed based on the DSC heating curves (Fig. S8, ESI†). The crystallization peak temperature (T_c) of PCBM was analyzed based on the DSC cooling curves.

Polarized optical microscopy (POM). POM images were obtained using a NIKON E600 POL instrument equipped with a camera to investigate the morphology of the films.

UV-visible absorption spectra. The UV-visible spectra of the films were recorded using a Perkin-Elmer Lambda 750 UV/VIS spectrometer. The films were prepared by spin-coating the requisite solution onto pre-cleaned glass slides (Fig. S9, ESI†).

Photoluminescence spectra. The PL spectra of the films were recorded on a Hitachi F-7000 PC spectrofluorophotometer equipped with a xenon lamp as the light source (Fig. S10 and S11, ESI†).

Transmission electron microscopy (TEM). The morphology of the films was observed using JEM-2010 HR TEM with an accelerating voltage of 200 kV. The films floating on the surface were transferred onto the copper screen after dipping the slides into deionized water.

Atomic force microscopy (AFM). The topography observation of the films was performed using the AFM (Bruker, MultiMode 8) PeakForce Tapping module.

Current–voltage (J–V) characteristics. J–V curves were characterized using a Keithley 2400 instrument. The currents were measured in the dark and under 100 mW cm⁻² simulated AM 1.5G irradiation (Abet Solar Simulator Sun2000). All the measurements were performed under an ambient atmosphere at room temperature.

Theoretical calculations

The configurations of the PC₆₁BM:8TTP8, PC₆₁BM:8OCB and PC₆₁BM:5CT complexes were optimized using density functional theory (DFT) at the B3LYP/6-31G** level. The single point energies of these optimized complexes were computed considering Basis Set Superposition Error (BSSE) to compute the binding energies between the three small molecules and PC₆₁BM. All the calculations were carried out using the Gaussian-09 program.³⁰

Author contributions

W. Zhou and K. Hu contributed equally to this work.

Conflicts of interest

The authous declare no competing financial interest.

Acknowledgements

This work was financially supported by the National Science Fund for Distinguished Young Scholars (51425304) and the National Natural Science Foundation of China (51303077 and 51563016).

References

1. N. Espinosa, M. Hösel, D. Angmo and F. C. Krebs, Solar cells with one-day energy payback for the factories of the future, Energy Environ. Sci., 2012, 5, 5117–5132.
2. A. C. Arias, J. D. MacKenzie, I. McCulloch, J. Rivnay and A. Salleo, Materials and applications for large area electronics: Solution-based approaches, Chem. Rev., 2010, 110, 3–24.
3. Z. He, B. Xiao, F. Liu, H. Wu, Y. Yang, S. Xiao, C. Wang, T. P. Russell and Y. Cao, Single-Junction Polymer Solar Cells with High Efficiency and Photovoltage, Nat. Photonics, 2015, 9, 174–179.
4. B. Kan, M. Li, Q. Zhang, F. Liu, X. Wan, Y. Wang, W. Ni, G. Long, X. Yang and H. Feng, A series of simple oligomer-like small molecules based on oligothiophenes for solution-processed solar cells with high efficiency, J. Am. Chem. Soc., 2015, 137, 3886–3893.
5. R. Tao, T. Umeyama, T. Higashino, T. Koganezawa and H. Imahori, Synthesis and isolation of cis-2 regiospecific ethylene-tethered indene dimer–[70] fullerene adduct for polymer solar cell applications, ACS Appl. Mater. Interfaces, 2015, 7, 16676–16685.
6. H. Hu, K. Jiang, G. Yang, J. Liu, Z. Li, H. Lin, Y. Liu, J. Zhao, J. Zhang and F. Huang, Terthiophene-based D–A polymer with an asymmetric arrangement of alkyl chains that enables efficient polymer solar cells, J. Am. Chem. Soc., 2015, 137, 14149–14157.
7. H. L. Yip and A. K. Y. Jen, Recent advances in solution-processed interfacial materials for efficient and stable polymer solar cells, Energy Environ. Sci., 2012, 5, 5994–6011.
8. O. Oklobia and T. Shafai, A quantitative study of the formation of PCBM clusters upon thermal annealing of P3HT/PCBM bulk heterojunction solar cell, Sol. Energy Mater. Sol. Cells, 2013, 117, 1–8.
9. S. Nam, S. Woo, J. Seo, W. H. Kim, H. Kim, C. R. McNeill, T. J. Shin, D. D. Bradley and Y. Kim, Pronounced cosolvent effects in polymer: Polymer bulk heterojunction solar cells with sulfur-rich electron-donating and imide-containing electron-accepting polymers, ACS Appl. Mater. Interfaces, 2015, 7, 15995–16002.
10. 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, Complementary absorbing star-shaped small molecules for the preparation of ternary cascade energy structures in organic photovoltaic cells, Adv. Funct. Mater., 2013, 23, 1556–1565.
11. M. Koppe, H. J. Egelhaaf, G. Dennler, M. C. Scharber, C. J. Brabec, P. Schilinsky and C. N. Hoth, Near IR sensitization of organic bulk heterojunction solar cells: Towards optimization of the spectral response of organic solar cells, Adv. Funct. Mater., 2010, 20, 338–346.
12. K. Sivula, Z. T. Ball, N. Watanabe and J. M. Fréchet, Amphiphilic diblock copolymer compatibilizers and their effect on the morphology and performance of polythiophene: Fullerene solar cells, Adv. Mater., 2006, 18, 206–210.
13. Y. Lin, P. Cheng, Y. Li and X. Zhan, A 3D star-shaped non-fullerene acceptor for solution-processed organic solar cells with a high open-circuit voltage of 1.18 V, Chem. Commun., 2012, 48, 4773–4775.
14. C. H. Kim, S. H. Cha, S. C. Kim, M. Song, J. Lee, W. S. Shin, S. J. Moon, J. H. Bahng, N. A. Kotov and S. H. Jin, Silver nanowire embedded in P3HT:PCBM for high-efficiency hybrid photovoltaic device applications, ACS Nano, 2011, 5, 3319–3325.
15. C. Y. Nam, Q. Wu, D. Su, C. Chiu, N. J. Tremblay, C. Nuckolls and C. T. Black, Nanostructured electrodes for organic bulk heterojunction solar cells: Model study using carbon nanotube dispersed polythiophene-fullerene blend devices, J. Appl. Phys., 2011, 110, 064307.
16. C. Liu, J. Li, X. Zhang, Y. He, Z. Li, H. Li, W. Guo, L. Shen and S. Ruan, Improving the efficiency of inverted polymer solar cells by introducing inorganic dopants, Phys. Chem. Chem. Phys., 2015, 17, 7960–7965.
17. M. Han, H. Kim, H. Seo, B. Ma and J. W. Park, Photovoltaic efficiency enhancement by the generation of an embedded silica-like passivation layer along the P3HT/PCBM interface using an symmetric block copolymer additive, Adv. Mater., 2012, 24, 6311–6317.
18. A. Sharenko, N. D. Treat, J. A. Love, M. F. Toney, N. Stingelin and T. Q. Nguyen, Use of a commercially available nucleating agent to control the morphological development of solution-processed small molecule bulk heterojunction organic solar cells, J. Mater. Chem. A, 2014, 2, 15717–15721.
19. S. Jeong, Y. Kwon, B. D. Choi, H. Ade and Y. S. Han, Improved efficiency of bulk heterojunction poly(3-hexylthiophene):[6,6]-phenyl-C61-butyric acid methyl ester photovoltaic devices using discotic liquid crystal additives, Appl. Phys. Lett., 2010, 96, 183305.
20. S. W. Heo, K. H. Baek, H. J. Song, T. H. Lee and D. K. Moon, Improved performance of P3HT:PCBM based solar cells using nematic liquid crystals as a processing additive under low processing temperature conditions, Macromol. Mater. Eng., 2014, 299, 353–360.
21. S. Jeong, Y. Kwon, B. D. Choi, G. Kwak and Y. S. Han, Effects of nematic liquid crystal additives on the performance of polymer solar cells, Macromol. Chem. Phys., 2010, 211, 2474–2479.
22. W. Zhou, J. Shi, L. Lv, L. Chen and Y. Chen, A mechanistic investigation of morphology evolution in P3HT-PCBM films induced by liquid crystalline molecules under external electric field, Phys. Chem. Chem. Phys., 2015, 17, 387–397.
23. J. M. Lehn, Cryptates: The chemistry of macropolycyclic inclusion complexes, Acc. Chem. Res., 1978, 11, 49–57.
24. J. M. Lehn, In supramolecular chemistry, Proc. - Indian Acad. Sci., Chem. Sci., 1994, 915–922.
25. L. M. Salonen, M. Ellermann and F. Diederich, Aromatic rings in chemical and biological recognition: Energetics and structures, Angew. Chem., Int. Ed., 2011, 50, 4808–4842.
26. C. Z. Li, Y. Matsuo, T. Niinomi, Y. Sato and E. Nakamura, Face-to-face C₆F₅-[60] fullerene interaction for ordering fullerene molecules and application to thin-film organic photovoltaics, Chem. Commun., 2010, 46, 8582–8584.
27. M. H. Liao, C. E. Tsai, Y. Y. Lai, F. Y. Cao, J. S. Wu, C. L. Wang, C. S. Hsu, I. Liau and Y. J. Cheng, Morphological stabilization by supramolecular perfluorophenyl-C₆₀ interactions leading to efficient and thermally stable organic photovoltaics, Adv. Funct. Mater., 2014, 24, 1418–1429.
28. M. O. Sinnokrot and C. D. Sherrill, Unexpected substituent effects in face-to-face π-stacking interactions, J. Phys. Chem. A, 2003, 107, 8377–8379.
29. X. Ye, Z.-H. Li, W. Wang, K. Fan, W. Xu and Z. Hua, The parallel π–π stacking: A model study with MP2 and DFT methods, Chem. Phys. Lett., 2004, 397, 56–61.
30. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Gaussian, Inc., Wallingford, CT, USA, 2009.

---

Footnote

† Electronic supplementary information (ESI) available: FTIR spectra, POM images, AFM images, DSC curves, UV-vis absorption spectra, and photoluminescence spectra. See DOI: 10.1039/c6qm00199h

---