Plasmonic organic solar cell employing Au NP:PEDOT:PSS doped rGO

Abstract

We demonstrate the first facile synthetic process to prepare a highly promising composite material by combining our frequently used p-type conducting polymer poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS) and reduced graphene oxide (rGO), and further employed the composite as a hole transport layer (HTL) in plasmonic organic solar cells. The conductivity of the PEDOT:PSS:rGO mixture can be tuned by varying the concentration of rGO in the PEDOT:PSS solution. Ultimately, the integration of gold nanoparticles (Au NPs) in the PEDOT:PSS:rGO plasmonic organic solar cell demonstrated a 9.34% improvement in the power conversion efficiency (PCE), measured using an AM1.5 G solar simulator at 100 mW cm⁻² light illumination intensity to generate localized surface plasmon resonance (LSPR). The enhanced performance was caused by local enrichment of the electromagnetic field surrounding the Au NPs.

---

Introduction

Recently, we have seen rapid progress in the development of high performance organic solar cells (OSCs). OSCs attracted attention from industries and academics because they offer a variety of advantages such as creating cost-effective, flexible, lightweight, and large area devices.^1–4 Today, the best reported power conversion efficiency (PCE) for single junction OSCs is 12% by Heliatek,⁵ while a PCE of 11.83% has been reported to be the best efficiency for multi-junction OSCs.⁶ These values are relatively low if one compares them with inorganic solar cells. The quantum efficiency of OSCs is still limited by low carrier mobility.⁷ A thinner active layer can lower the probability of charge recombination and increase the carrier drift velocity with a higher electric field, thus enhancing the internal quantum efficiency (IQE). However, a minimum film thickness is required to ensure sufficient photon absorption.^7,8 Nowadays, various approaches have been put forward to enhance light absorption in order to optimize cell absorption without increasing the thickness of the active layer, and to avoid increasing the charge recombination.^9,10 Recently, plasmonic resonant metallic nanostructures^11–29 have been introduced into OSCs with the introduction of metallic nanoparticles (NPs). Two primary schemes are commonly employed for the excitation of the SPP, in order to create coherent collective oscillation of the conduction electrons surrounding the metallic surfaces. In one, surface plasmon polaritons (SPPs) propagated along the metal-dielectric interface are triggered by incorporating metallic nanostructures, such as periodic arrays or gratings.^7,30,31 In the other, surface plasmons are localized by noble metallic nanoparticles (NPs), such as Cu, Ag, Pt, and Au, resulting in localized surface plasmon resonance (LSPR).^32–37 However, there are limited reports of conversion efficiencies higher than 8% from integrating metallic NPs into OSCs.^38,39 Further in-depth investigation needs to be done on the influence of NPs on the charge separation and transport inside high efficiency OSCs.

In this paper, we report a comprehensive study of the influence of NPs on OSCs by introducing Au NPs into OSCs fabricated using the polymers polythieno[3,4 b]-thiophene/benzodithiophene (PTB7) and [6,6]-phenyl C₇₁-butyric acid methyl ester (PC₇₀BM). Excitation of the LSPR, triggered by adding Au NPs into the anodic buffer layer, significantly enhanced the overall PCE of the OPV devices. Electrical characterization results revealed that the presence of metallic NPs had a negligible effect on the charge transport process. Moreover, steady state and dynamic photoluminescence (PL) measurements provided strong evidence that the LSPR, induced by the Au NPs, not only increased the degree of light absorption but also enhanced the degree of exciton dissociation. As a result, the photocurrent and overall device efficiency were both improved considerably after exploiting the optical effects of the LSPR. Thus, our OSCs demonstrated the following advantages: (i) broad light absorption enhancement; (ii) improved exciton generation rate and dissociation efficiency; and (iii) increased charge carrier density and lifetime. These improvements led to an improved PCE of 8.08%, which was 9.34% higher than that of our control device.

Fig. 1 shows the thin film characteristics of pristine rGO compared with our newly synthesized PEDOT:PSS:rGO layer. The FTIR spectra contain absorption peaks at 3435 and 1168 cm⁻¹ which correspond to the remaining –OH group and C–O–C bonds on the surface of the rGO, respectively.⁴⁰ The peaks located at 1521, 1312, and 1196 cm⁻¹ are assigned to the C=C and C–C bonds of the thiophene ring and the sulfonic acid group of the PSS, respectively.^40–42 From the FTIR spectra, one can conclude that the rGO has polar groups for the dispersion and that the polymerization of PEDOT:PSS occurred with the presence of rGO. This assumption was supported by the Raman spectra as shown in Fig. 1b.

Fig. 1 (a) FTIR, Raman, XRD, and transmittance spectra of the rGO and PEDOT:PSS:rGO layers.

Both spectra exhibit five typical bands corresponding to a C=C anti-symmetric stretching (1570 cm⁻¹), C=C asymmetrical stretching (1501 cm⁻¹), C=C symmetrical stretching (1440 cm⁻¹), single C–C stretching (1365 cm⁻¹), and C–C inter-ring stretching (1262 cm⁻¹).⁴³ In comparison with the spectrum of PEDOT:PSS, the peaks of PEDOT:PSS:rGO are slightly shifted (for example, from 1437.1 cm⁻¹ to 1441.5 cm⁻¹ in the case of the C=C symmetrical stretching) according to the strong π–π interactions of the aromatic structures of the PEDOT:PSS and the electron-rich rGO.⁴⁴ To confirm the formation of the composite, X-ray diffraction (XRD) spectra of the rGO and PEDOT:PSS:rGO were obtained as shown in Fig. 1c. Pristine rGO showed a broad peak at 21.5° and, after dispersion with PSS and in situ polymerization with EDOT, the peaks of PEDOT:PSS at 17.5° and 25.8° appeared as the main peaks. This indicates that the diffraction peak of rGO disappeared and the peaks of the nanocomposite were almost identical to PEDOT:PSS. This result clearly shows that the rGO was well-dispersed by the polyelectrolyte PSS. The transparency of the PEDOT:PSS:rGO was slightly decreased compared with that of the rGO in the entire visible region. However, it is worth noting that although the transparency of the PEDOT:PSS:rGO is slightly less, it is still above 90% transparency in the visible region.

Table S1† reveals the conductivity values for the pristine PEDOT:PSS, pristine rGO and PEDOT:PSS:rGO mixture. The conductivities of the pristine PEDOT:PSS and rGO are 456 and 4.42. S cm⁻¹, respectively. However, the conductivity of the mixture increases significantly depending on the rGO concentration.

In addition, an atomic force microscopy (AFM) study revealed that the pure rGO film had a smooth surface with an average roughness (R_a) of 2.41 nm, while the PEDOT:PSS:rGO film (R_a = 5.36 nm) had a rougher surface as shown in Fig. 2a and b.

Fig. 2 Atomic force microscopy images of the rGO and PEDOT:PSS:rGO layers.

Fig. 3a and b show the OSC schematic device structure, and its energy diagram. TEM images demonstrated that the size of the Au NPs was around 20–30 nm (Fig. 3c), and it can be assumed that these NPs were embedded in the PEDOT:PSS:rGO and PTB7:PC₇₀BM active material. Fig. 3d shows an analysis of the Au NP distribution. This hypothesis is further confirmed by the atomic force microscopy (AFM) images.

Fig. 3 (a) Schematic of the plasmonic OSC, ITO/Au NP:PEDOT:PSS:rGO/PTB7:PC₇₀BM/PFN/Al, (b) energy level diagram of the OSC materials used in this study, (c) TEM image of the Au NP:PEDOT:PSS:rGO, and (d) size distribution of the Au NPs.

Fig. 4a and b display the AFM images of the PEDOT:PSS with and without the presence of Au NPs, while Fig. 4c and d show the AFM images of the PTB7:PC₇₀BM with and without Au NPs. The root-mean-square (RMS) roughness of the PEDOT:PSS layer on ITO glass was around 0.77 nm, while the PEDOT:PSS blended with the Au NPs exhibited increased RMS roughness of 2.28 nm. The opposite trend was seen for the Au NPs blended into the PTB7:PC₇₀BM. The RMS roughness decreased with the presence of Au NPs from 3.88 to 0.8 nm. To further clarify the role of the Au NPs, PL decay measurements were obtained to probe the singlet quenching in PTB7 under the influence of oxygen. It is worth noting that the Au NP resonance has an excitation lifetime of a few picoseconds. Fig. 4e shows the PL decay plots for the pristine PEDOT:PSS and the Au NP:PEDOT:PSS:rGO samples. As shown in Fig. 4e, the presence of Au NPs in the PEDOT:PSS:rGO delayed the PL intensity decay rate of the device. Donor–acceptor interactions between the triplet excitons of PTB7 and the Au NPs cause the quenching of the triplet state and, thus, the photo-oxidation rate.

Fig. 4 AFM images of the (a) PEDOT:PSS:rGO, (b) Au NP:PEDOT:PSS:rGO, (c) Au NP:PTB7:PC₇₀BM, (d) Au NP:PEDOT:PSS:rGO:PTB7:PC₇₀BM layers, and (e) normalized PL decay measurements of the pristine PEDOT:PSS (black) and Au NP:PEDOT:PSS:rGO (red).

Fig. 5a demonstrates the J–V characteristics of our control and plasmonic devices under 100 mW cm⁻² illumination (AM1.5 G). The control device prepared with pristine PEDOT:PSS:rGO demonstrated an open-circuit voltage (V_oc) of 0.76 V, a short-circuit current (J_sc) of 14.86 mA cm⁻², and a fill factor (FF) of 65.49%. The obtained power conversion efficiency (PCE) was 7.39%. After the incorporation of Au NPs into the PEDOT:PSS:rGO hole transport layer, the value of V_oc remained unchanged (0.76 V) but the FF improved significantly (Table 1). We also observed a noticeably increasing trend in the J_sc with the presence of Au NPs. For the plasmonic devices fabricated with the PEDOT:PSS:rGO blended with 1%, 2%, and 4% Au NP solutions, the J_sc values were 15.24, 15.42, and 15.54 mA cm⁻². The PCE also improved from 7.39% to 7.98%, 8.01%, and 8.08%, respectively. However, a further increase in the concentration of the Au NPs to greater than 4% led to a significant decrease in the J_sc. We attributed the abrupt plunge of the J_sc to enhanced backward scattering and/or the increased resistivity of the hole transport layer. To confirm our speculation, we carried out four-point probe measurements and the resistivity of the hole transport layer improved after being blended with the Au NPs (Table 1). We attributed the increase in resistivity to the morphology changes upon the utilization of Au NPs in the hole transport layer.

Fig. 5 Comparison of OSC results with and without Au NPs in the PEDOT:PSS layer: (a) J–V characteristics (dark) of the OSCs, (b) J–V characteristics of the OSCs under AM1.5 G (100 mW cm⁻²) illumination, and (c) EQE of the OSCs.

Table 1 Photovoltaic parameters of the solar cells with different Au NP concentrations in the PEDOT:PSS:rGO layer under AM1.5 G illumination at 100 mW cm⁻²

Au NPs	Jsc (mA cm^−2)	Voc (V)	FF (%)	PCE (%)
Control	14.86 ± 0.03	0.76 ± 0.01	65.49 ± 0.02	7.39 ± 0.02
1%	15.24 ± 0.02	0.76 ± 0.01	68.83 ± 0.11	7.98 ± 0.03
2%	15.42 ± 0.02	0.76 ± 0.01	68.31 ± 0.03	8.01 ± 0.02
4%	15.54 ± 0.03	0.76 ± 0.01	68.24 ± 0.01	8.08 ± 0.02

---

Accordingly, the values of the device series resistance (R_s) extracted from the inverse slopes of the dark J–V curves at the V_oc, also increased after the incorporation of the Au NPs (Table 1), indicating that the improved performance of the plasmonic devices did not result from a reduction in the device resistance. Table 1 tabulates the photovoltaic parameters extracted from Fig. 5.

On the other hand, IPCE is a very useful parameter for determining the PCE of plasmonic devices, and reaches 100% when all incident photons generate electron–hole pairs and are collected. Nevertheless, IPCE is usually less than 100% because of losses from: i) the reflection of incident photons, ii) imperfect absorption of photons by the semiconductor, and iii) recombination of charge carriers within the semiconductor. The IPCE spectra for the control and the plasmonic devices incorporating Au NP:PEDOT:PSS:rGO (1, 2, and 4%) are shown in Fig. 5c. Even though all the IPCE spectra are almost identical, the IPCE values for the photovoltaic devices containing Au NP:PEDOT:PSS:rGO are higher compared to those of the PEDOT:PSS from 300 to 700 nm. For instance, the IPCE of the control device is around 68.87% at 480 nm, and the IPCEs of the plasmonic devices with 1, 2, and 4% Au NP:PEDOT:PSS:rGO are 72.03, 72.77, and 72.87%, respectively, at the same wavelength. The difference may result from the improved charge carrier mobility, and consequently the absorption of the Au NP:PEDOT:PSS:rGO hole transport layer. The integrated J_sc values from the IPCE spectra of the photovoltaic devices incorporating 1, 2, and 4% of Au NPs are 14.32, 14.95, 15.07, and 15.13 mA cm⁻², respectively. These values are in agreement with the experimental J_sc values of 14.86, 15.24, 15.42, and 15.54 mA cm⁻². Please note that, throughout this work, the error between the experimental J_sc and the integrated J_sc was less than 4%.⁴⁵ These values are tabulated in Table 1 for better comparison and the slight differences are also a consequence of the spectral mismatch of the xenon lamp, which exhibits a higher photon flux than the sun in the UV region.⁴⁶

Although the enhanced performance is shown through the photovoltaic behaviour, the origin of the enhanced PCE is clearly different when the Au NPs are blended with the active layer compared to when the Au NPs are blended with the PEDOT:PSS:rGO. To confirm this hypothesis, we also carried out another set of experiments, where we blended Au NPs with an PTB7:PC₇₀BM active layer. Fig. S1a† represents the J–V characteristics of the control and plasmonic photovoltaic devices with different Au NP concentrations, under the illumination of AM1.5 G with 100 mW cm⁻². The photovoltaic parameters are illustrated in Table S2.† From Table S2,† it is clear that the addition of 1, 2, and 4% of Au NPs in the PTB7:PC₇₀BM active layer triggers a notable enhancement of both the J_sc and FF. Note that the V_oc remains unchanged. It is worth mentioning that further increments of Au NPs (>4%) negatively impact the photovoltaic performance. We believe that too many Au NPs may perturb the formation of the PTB7 and PCBM domains resulting in an inhomogeneous morphology, and might lower the carrier mobility.

Fig. S1b† demonstrates the IPCE spectra of the control and plasmonic devices showing the best PCE. The IPCE enhances notably upon the utilization of Au NPs, which complies with the improved J_sc and FF observed. In particular, compared to the control device, the IPCE of the plasmonic device with 2% Au NPs, becomes enhanced in a broad spectral range (400 to 700 nm), and maximizes at 570 nm. This wavelength range coincides with the spectral range in which the optical absorption of the Au NPs is embedded in the PTB7:PCBM medium by the LSPR effect. The IPCE improvements can be attributed to the local enhancement of the incident electromagnetic irradiation field in the vicinity of the small-sized Au NPs as well as to multiple scattering by the Au NPs. Previously, it was reported that the integration of NPs in the active layer enhances both the device performance and structural stability.⁴⁷ This behaviour can explain the observed discrepancy between the absorption and the IPCE enhancement, whereas the NPs enhance the structural stability, leading to superior performance. Thus, the improved performance due to the presence of NPs also indicates a better active layer morphology.

To assess in detail the influence of Au NPs on the exciton generation and dissociation behaviours, we evaluated the maximum exciton generation rate (G_max) and exciton dissociation probability P(E, T) of the control and plasmonic devices. The devices were biased, sweeping from +1V to −1V.

Fig. 6a illustrates the photocurrent density (J_ph) versus the internal voltage (V_int) of the control and plasmonic devices, under illumination at 100 mW cm⁻². Here J_ph is defined as:

J_ph = J_L − J_D (1)

where J_L and J_D are the current densities under illumination and in the dark, respectively, and

V_int = V_BI − V_app (2)

where V_BI and V_a are the voltage at which J_ph equals zero, and the applied voltage, respectively. The two distinctive regions can be seen from Fig. 6; low and high V_int. In the low V_int region, it appears that J_ph is enhanced in proportion to the voltage at low voltage, while in the high V_int region, the J_ph is saturated with the increased voltage, where the internal field is high enough to sweep out all carriers to the electrodes. Thus, the J_ph saturation is limited by the number of absorbed photons. Assuming J_ph saturation is independent of the bias and temperature, we obtained the G_max using the following equation:

J_ph = qLG_max (3)

where q is the electronic charge, and L is the thickness of the active layer,^48,49 assuming that all of the photogenerated excitons dissociated into holes and electrons and contributed to the current in the saturated regime. The G_max values for the control and plasmonic devices (1, 2, and 4% Au NP:PEDOT:PSS:rGO) were 5.17 × 10²⁷ m⁻³ s⁻¹ (J_ph,sat = 114 A m⁻²), 5.41 × 10²⁷ m⁻³ s⁻¹ (J_ph,sat = 120 A m⁻²), 5.46 × 10²⁷ m⁻³ s⁻¹ (J_ph,sat = 123 A m⁻²), and 5.55 × 10²⁷ m⁻³ s⁻¹ (J_ph,sat = 127 A m⁻²), respectively. These values demonstrate significant improvement in G_max because the Au NPs were mixed into the PEDOT:PSS:rGO layer. It is worth noting that G_max is the maximum photoinduced carrier generation rate per unit volume.^48,49 Such enhancement indicates improved light absorption in the plasmonic device. To gain deeper insight into the exciton dissociation probability, Fig. 6b compares the control and plasmonic devices, which are related to the electric field (E) and the temperature (T). In brief, only a fraction of the photogenerated excitons can be dissociated into free carriers. Therefore, the J_ph of the solar cell can be written as^48,49

J_ph = qG_maxP(E, T)L. (4)

Fig. 6 (a) Photocurrent density (J_ph) versus internal voltage (V_int) for the control and plasmonic solar cells, (b) exciton dissociation probability [P(E, T)] versus internal voltage (V_int) for control and plasmonic solar cells.

The P(E, T) can be extracted from the normalized photocurrent density with the saturated photocurrent density (J_ph/J_sat).⁵⁰ As shown in Fig. 6b, we can see the increment of the P(E, T) under the short-circuit conditions (V_a = 0 V) from 75% for the control device to 83, 89, and 90%, respectively for the plasmonic devices (1, 2, and 4% Au NP:PEDOT:PSS:rGO). This implies that the excitation of the LSPR induced the photogenerated excitons to dissociate into free carriers. Hence, it is concluded that the excitation of the LSPR not only improved the exciton generation rate but also the dissociation probability and thereby improved the J_sc of the OPVs.

To get a better understanding of the charge recombination kinetics, we studied a variation of J_sc on our OPV devices as a function of the light intensity (10, 30, 50, 70 and 100 mW cm⁻²). A few authors have proposed a power law dependence of J_sc on the light intensity, using the following equation^48,51

J_sc α I^α. (5)

Fig. 7 illustrates the variation of J_sc versus the incident light intensity in a log–log scale and fitted to a power law using eqn (5). From Fig. 7, the linear dependence of J_sc is consistent with sweep-out at short circuit and also indicates that bimolecular recombination is negligible.⁵² The α values for the control and plasmonic devices (1, 2, and 4% Au NP:PEDOT:PSS:rGO) are 0.983, 0.999, 0.999 and 0.999, respectively. Hence, we expect that the introduction of Au NPs into the PEDOT:PSS:rGO hole transport layer has negligible impact on the nature of the charge transport process in the device.

Fig. 7 Current density (J_sc) versus light intensity. Straight lines were fitted using the equation J_sc α I^α.

We carried out time resolved PL spectroscopy, using a 470 nm excitation source, to understand the decay profiles of the PTB7:PC₇₀BM films that had been deposited on the PEDOT:PSS:rGO layers with and without Au NPs. Fitting the data with two exponential decays curves yields the lifetimes of the carriers and/or excitons. For the control sample, a PL lifetime of 0.37 ns was observed coupling the process between the plasmonic field and the photogenerated excitons within the photoactive blend.^53–56 Fig. 8 shows the PL intensity. When the PTB7:PC₇₀BM forms a contact with the Au NP:PEDOT:PSS:rGO, it was observed that the PL lifetimes are decreased remarkably. For the plasmonic sample, the PL lifetime decreases to 0.27 ns, 0.23 ns, and 0.20 ns were observed for the Au NP:PEDOT:PSS:rGO:PTB7:PC₇₀BM with 1%, 2% and 4% Au NPs, respectively. These observations indicate that charge transfer in the plasmonic samples is faster than in the control sample. The vivid change in the value of the PL lifetime after the incorporation of Au NPs also accounts for the presence of the strong coupling between the plasmonic field and the excitons.^48,57,58

Fig. 8 PL decay profile for the PTB7:PC₇₀BM blends in the control and plasmonic solar cells. Excitation source: 470 nm pulsed laser.

Although we have seen such a rapid improvement in device efficiency, the lifetime of organic devices is also significant, especially in OSCs. Previous works have reported different approaches to overcome stability issues such as (i) employing an inverted structure, (ii) use of metal-oxide transitions, and (iii) avoidance of acidic material such as PEDOT:PSS on ITO. Hence, we carried out lifetime measurements for our unencapsulated, control and plasmonic devices over a four week period. The normalized photovoltaic parameters are shown in Fig. 9.

Fig. 9 Degradation comparison of (a) control, and plasmonic devices (b) 1% Au NP:PEDOT:PSS:rGO, (c) 2% Au NP:PEDOT:PSS:rGO, and (d) 4% Au NP:PEDOT:PSS:rGO.

J–V characteristics were taken every week under simulated AM1.5 solar irradiation (100 mW cm⁻²). Particularly, the V_oc of the control and plasmonic devices remained stable over the four weeks of stability measurements. In the control device, the J_sc remained constant after the first week although the FF dropped significantly. The J_sc started to decrease during the second week, when the J_sc plunged from 14.86 to 14.10 mA cm⁻². The efficiency of the control device dropped remarkably at the fourth weeks measurement to 5.92%, which is a 19.89% drop. Compared to the control device, the plasmonic devices also suffered some degradation over the 4 weeks. Among the plasmonic devices, the worst degradation occurred for the device integrated with a 1% Au NP:PEDOT:PSS:rGO hole transport layer, where the PCE decreased to 6.21% after 4 weeks of measurements. This is due to the significant decrease of the J_sc and FF during the first week of measurements. On the other hand, the best performance came from the 2% Au NP:PEDOT:PSS:rGO. The efficiency dropped almost 15% from its initial value. The mechanism behind the improved stability is now under investigation and will be published later.

Conclusions

In conclusion, we demonstrated the facile synthesis of a highly conducting composite material of PEDOT:PSS:rGO and the positive effects of Au NPs in PEDOT:PSS:rGO based on PTB7:PC₇₀BM photovoltaic devices. The PEDOT:PSS:rGO embedded with Au NPs had better J_sc and PCE values than the control devices. These results are attributed to the localized surface plasmon resonance-induced local field enhancement. Finally, the plasmonic devices demonstrated 9.34% improvement compared to the control devices.

Experimental section

Materials

Au NPs, PTB7, and PCBM were purchased from Sigma Aldrich and used without any further purification.

Synthesis of PEDOT:PSS doped reduced graphene oxide

The PSS was used as a dispersant and dopant for the rGO and PEDOT simultaneously. The required amount of rGO powder was directly blended with an aqueous PSS solution before the polymerization of EDOT to prepare the PEDOT:PSS:rGO with the rGO content ranging from 2 wt% to 10 wt%. For the synthesis of the PEDOT:PSS:rGO with 6 wt% of rGO to total solid content of PSS, the rGO (0.17 g) was added to the blended solution of distilled water (1 kg) and PSS (5.85 g), and then the mixed solution was stirred for 30 min and ultrasonicated with a tip sonicator in a 10 °C bath for 30 min to prepare a well-dispersed solution. After the sonication process, the solution was bubbled using nitrogen gas (99.999%) for 1 h at a rate of 3 L min⁻¹ to avoid oxidation from the dissolved oxygen in the water. To this solution, the EDOT monomer (7.31 g) was slowly added and stirred for 30 min. The direct synthesis of the PEDOT:PSS:rGO was carried out in the presence of an Fe³⁺-catalyzed oxidative. The oxidizing reagents of iron(III) sulfate (0.21 g, 5.2 × 10⁻⁴ mol) and sodium persulfate (8.81 g, 3.7 × 10⁻² mol) were each dissolved in 30 mL of distilled water using a sonication bath and added to the reaction solution. The polymerization was performed for 24 h at 10 °C with bubbling nitrogen gas. After the polymerization of the PEDOT:PSS:rGO, the product was mixed with 400 mL of a mixture of cation and anion ion exchange resin for 1 h and filtered with a 30 μm mesh filter. For comparison, the pristine PEDOT:PSS solution was synthesized as described above without the addition of rGO.

Preparation of the Au NP:PEDOT:PSS doped reduced graphene oxide solution

The Au NP:PEDOT:PSS:rGO solution was prepared by mixing Au NPs, PEDOT:PSS and rGO according to the blend ratio of 4 : 1 : 3 for Au NP : PEDOT : PSS : rGO, respectively. The mixed solution was stirred overnight.

Device fabrication

The OSCs were fabricated on pre-cleaned indium tin oxide (ITO) coated glass substrates, with a sheet resistance of 10–15 Ω/sq. The thickness of the Au NP:PEDOT:PSS:rGO layer at about 40 nm was obtained at 4000 rpm spin speed, and then annealed at 130 °C for 20 min. The PTB7:PC₇₀BM (at a ratio of 1 : 1.5) was dissolved in chlorobenzene with 3% diiodooctane and was then spin-coated on the Au NP:PEDOT:PSS:rGO layer at 500 rpm for 25 s. A thin layer of PFN (5 nm) was spin-coated at 4000 rpm for 25 s. Next, the thermal evaporation of 100 nm of Al, as the top electrode, was carried out in a vacuum (base pressure of ∼1 × 10⁻⁸ Torr). The active areas of our organic solar cells were 0.2 cm in length and width, respectively. After being transferred into a glove-box, the completed organic solar cells were encapsulated with a transparent encapsulation glass, glued with UV-curable resin, and exposed to ultraviolet light for 1 min. The performance of the organic solar cells was obtained from the J–V characteristics measured using a Keithley 2400 LV source meter. The solar cell performance was measured using a solar simulator, with Air Mass 1.5 Global (AM1.5 G) and with an irradiation intensity of 100 mW cm⁻². All measurements were carried out at room temperature, under a relative humidity of 60%. The EQE measurements were performed using the EQE system (Model 74 000) obtained from Newport Oriel Instruments USA and Hamamatsu calibrated silicon cell photodiodes were used as a reference diode. The wavelength was controlled with a monochromator of 200–1600 nm.

Acknowledgements

This work was supported by the Human Resources Development program (no. 20134010200490) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry and Energy.

References

1. C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct. Mater., 2001, 11, 15–26.
2. W. J. da Silva, H. P. Kim, A. R. B. M. Yusoff and J. Jang, Nanoscale, 2013, 5, 9324–9329.
3. K. Kim, S.-H. Bae, C. T. Toh, H. Kim, J. H. Cho, D. Whang, T.-W. Lee, B. Özyilmaz and J.-H. Ahn, ACS Appl. Mater. Interfaces, 2014, 6, 3299–3304.
4. M. Hösel, R. R. Søndergaard, M. Jørgensen and F. C. Krebs, Adv. Energy Mater. DOI:10.1002/aenm.201301625.
5. http://www.heliatek.com/newscenter/latest_news/neuer-weltrekord-fur-organische-solarzellen-heliatek-behauptet-sich-mit-12-zelleffizienz-als-technologiefuhrer/?lang=en.
6. A. R. B. M. Yusoff, D. Kim, H. P. Kim, F. K. Shneider, W. J. da Silva and J. Jang, Energy Environ. Sci., 2015, 8, 303–316.
7. K. Tvingstedt, N. K. Persson, O. Inganas, A. Rahachou and I. V. Zozoulenko, Appl. Phys. Lett., 2007, 91, 113514.
8. F. J. Beck, A. Polman and K. R. Catchpo, J. Appl. Phys., 2009, 105, 114310.
9. P. W. M. Blom, V. D. Mihailetchi, L. J. A. Koster and D. E. Markov, Adv. Mater., 2007, 19, 1551–1566.
10. H. A. Atwater and A. Polman, Nat. Mater., 2010, 9, 205–213.
11. (a) V. E. Ferry, L. A. Sweatlock, D. Pacifici and H. A. Atwater, Nano Lett., 2008, 8, 4391–4397; (b) M. A. M. Teridi, M. Sookhakian, W. J. Basirun, R. Zakaria, F. K. Schneider, W. J. da Silva, J.-Y. Kim, S. J. Lee, H. P. Kim, A. R. B. M. Yusoff and J. Jang, Nanoscale 10.1039/c4nr05874g; (c) A. R. B. M. Yusoff, S. J. Lee, J. Jang, G. Long, X. Wan and Y. Chen, J. Mater. Chem. A, 2014, 2, 19988–19993.
12. S. P. Sundararajan, N. K. Grady, N. Mirin and N. J. Halas, Nano Lett., 2008, 8, 624–630.
13. R. A. Pala, J. White, E. Barnard, J. Liu and M. L. Brongersma, Adv. Mater., 2009, 21, 3504–3509.
14. C. Hagglund, S. P. Apell and B. Kasemo, Nano Lett., 2010, 10, 3135–3141.
15. D. M. Schaadt, B. Feng and E. T. Yu, Appl. Phys. Lett., 2005, 86, 063106.
16. S. Pillai, K. R. Catchpole, T. Trupke and M. A. Green, J. Appl. Phys., 2007, 101, 093105.
17. V. Resta, A. P. Caricato, A. Loiudice, A. Rizzo, G. Gigli, A. Taurin, M. Catalano and M. Martino, J. Nanopart. Res., 2013, 15, 2017.
18. D. H. Wang, D. Y. Kim, K. W. Choi, J. H. Seo, S. H. Im, J. H. Park, O. O. Park and A. J. Heeger, Angew. Chem., Int. Ed., 2011, 50, 5519.
19. G. D. Spyropoulos, M. Stylianakis, E. Stratakis and E. Kymakis, Photonics and Nanostructures, 2011, 9, 184.
20. W.-J. Yoon, K.-Y. Jung, J. Liu, T. Duraisamy, R. Revur, F. L. Teixeira, S. Sengupta and P. R. Berger, Sol. Energy Mater. Sol. Cells, 2010, 94, 128.
21. K. Topp, H. Borchert, F. Johnen, A. V. Tunc, M. Knipper, E. Von Hauff, J. Parisi and K. Al-Shamery, J. Phys. Chem. A, 2010, 114, 3981.
22. K. Kim and D. L. Carroll, Appl. Phys. Lett., 2005, 87, 203113.
23. R. S. Kim, J. Zhu, J. H. Park, L. Li, Z. Yu, H. Shen, M. Xue, K. L. Wang, G. Park, T. J. Anderson and Q. Pei, Opt. Express, 2012, 20, 163911.
24. D. D. S. Fung, L. Qiao, W. C. H. Choy, C. Wang, W. E. I. Sha, F. Xie and S. He, J. Mater. Chem., 2011, 21, 16349.
25. C. C. D. Wang, W. C. H. Choy, C. Duan, D. D. S. Fung, W. E. I. Sha, F.-X. Xie, F. Huang and Y. Cao, J. Mater. Chem., 2012, 22, 1206.
26. J.-L. Wu, F.-C. Chen, Y.-S. Hsiao, F.-C. Chien, P. Chen, C.-H. Kuo, M. H. Huang and C.-S. Hsu, ACS Nano, 2011, 5, 959.
27. J. Yang, J. You, C.-C. Chen, W.-C. Hsu, H.-R. Tan, X. W. Zhang, Z. Hong and Y. Yang, ACS Nano, 2010, 5, 6210.
28. J. H. Lee, J. H. Park, J. S. Kim, D. Y. Lee and K. Cho, Org. Electron., 2009, 10, 416.
29. A. J. Morfa, K. L. Rowlen, T. H. Reilly III, M. J. Romero and J. van de Lagemaat, Appl. Phys. Lett., 2008, 92, 013504.
30. M. G. Kang, T. Xu, H. J. Park, X. Luo and L. J. Guo, Adv. Mater., 2010, 22, 4378–4383.
31. N. C. Lindquist, W. A. Luhman, S. H. Oh and R. J. Holmes, Appl. Phys. Lett., 2008, 93, 123308.
32. S. S. Kim, S. I. Na, J. Jo, D. Y. Kim and Y. C. Nah, Appl. Phys. Lett., 2008, 93, 073307.
33. Y. C. Chang, F. Y. Chou, P. H. Yeh, H. W. Chen, S. H. Chang, Y. C. Lan, T. F. Guo, T. Tsai and C. T. Lee, J. Vac. Sci. Technol., B, 2007, 25, 1899.
34. F. C. Chen, J. L. Wu, C. L. Lee, Y. Hong, C. H. Kuo and M. H. Huang, Appl. Phys. Lett., 2009, 95, 013305.
35. A. J. Morfa, K. L. Rowlen, T. H. Reilly III, M. J. Romero and J. van de Lagemaat, Appl. Phys. Lett., 2008, 92, 013504.
36. J. H. Lee, J. H. Park, J. S. Kim, D. Y. Lee and K. Cho, Org. Electron., 2009, 10, 416–420.
37. A. P. Kulkarni, K. M. Noone, K. Munechika, S. R. Guyer and D. S. Ginger, Nano Lett., 2010, 10, 1501–1505.
38. X. H. Li, W. C. H. Choy, L. J. Huo, F. X. Xie, W. E. I. Sha, B. F. Ding, X. Guo, Y. F. Li, J. H. Hou, J. B. You and Y. Yang, Adv. Mater., 2012, 24, 3046–3052.
39. L. Lu, Z. Luo, T. Xu and T. Yu, Nano Lett., 2013, 13, 59–64.
40. Y. Si and E. T. Samulski, Nano Lett., 2008, 8, 1679–1682.
41. Y. Wu, B. Wang, Y. Ma, Y. Huang, N. Li, F. Zhang and Y. Chen, Nano Res., 2010, 3, 661–669.
42. J. Kim, J. You and E. Kim, Macromolecules, 2010, 43, 2322–2327.
43. X. J. Wang and K. Y. Wong, Thin Solid Films, 2006, 515, 1573–1578.
44. J. Zhang and X. S. Zhao, J. Phys. Chem. C, 2012, 116, 5420–5426.
45. E. Hutter and J. H. Fendler, Adv. Mater., 2004, 16, 1685–1706.
46. K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys. Chem. B, 2003, 107, 668–677.
47. M. Pelton, J. Aizpurua and G. Bryant, Laser Photonics Rev., 2008, 2, 136–159.
48. O. K. Varghese and C. A. Crimes, Sol. Energy Mater. Sol. Cells, 2008, 92, 374–378.
49. A. Duret and M. Gratzel, J. Phys. Chem. B, 2005, 109, 17184–17191.
50. B. Paci, G. D. Spyropoulos, A. Generosi, D. Bailo, V. R. Albertini, E. Stratakis and E. Kymakis, Adv. Funct. Mater., 2011, 21, 3573–3582.
51. V. D. Mihailetchi, H. X. Xie, B. de Boer, L. J. A. Koster and P. W. M. Blom, Adv. Funct. Mater., 2006, 16, 699–708.
52. V. D. Mihailetchi, L. J. A. Koster, J. C. Hummelen and P. W. M. Blom, Phys. Rev. Lett., 2004, 93, 216601.
53. V. Shrotriya, Y. Yao, G. Li and Y. Yang, Appl. Phys. Lett., 2006, 89, 063505.
54. P. Schilinsky, C. Waldauf and C. J. Brabec, Appl. Phys. Lett., 2002, 81, 3885.
55. V. Shrotriya, E. H. E. Wu, G. Li, Y. Yao and Y. Yang, Appl. Phys. Lett., 2006, 88, 064104.
56. J. H. Huang, F. C. Chien, P. L. Chen, K. C. Ho and C. W. Chu, Anal. Chem., 2010, 82, 1669–1673.
57. J. H. Huang, C. Y. Yang, Z. Y. Ho, D. Kekuda, M. C. Wu, F. C. Chien, P. L. Chen, C. W. Chu and K. C. Ho, Org. Electron., 2009, 9, 27.
58. J. H. Huang, K. C. Li, F. C. Chen, Y. S. Hsiao, D. Kekuda, P. Chen, H. C. Lin, K. C. Ho and W. Chu, J. Phys. Chem. C, 2010, 114, 9062–9069.

---

Footnote

† Electronic supplementary information (ESI) available: J–V curves, EQE spectra. See DOI: 10.1039/c5ra02878g

---