Au nanorods-incorporated plasmonic-enhanced inverted organic solar cells*
Peng Linga),b)§, Mei Yanga)§, Chen Shu-Fena), Zhang Yu-Peia), Hao Jing-Yua), Deng Ling-Linga), Huang Weia),c)
Key Laboratory for Organic Electronics and Information Displays & Institute of Advanced Materials, Nanjing University of Posts & Telecommunications Synergistic Innovation Center for Advanced Materials, Nanjing 210023, China
School of Opto-Electronic Engineering, Nanjing University of Posts & Telecommunications, Nanjing 210023, China
Key Laboratory of Flexible Electronics & Institute of Advanced Materials, National Synergistic Innovation Center for Advanced Materials, Nanjing Tech University, Nanjing 211816, China

These authors contributed equally to this work.

Corresponding author. E-mail: iamsfchen@njupt.edu.cn

Corresponding author. E-mail: iamdirector@fudan.edu.cn

*Project supported by the Ministry of Science and Technology, China (Grant No. 2012CB933301), the National Natural Science Foundation of China (Grant Nos. 61274065, 51173081, 61136003, BZ2010043, 51372119, and 51172110), and the Priority Academic Program Development of Jiangsu Provincial Higher Education Institutions and Synergetic Innovation Center for Organic Electronics and Information Displays, China.

Abstract

The effect of Au nanorods (NRs) on optical-to-electric conversion efficiency is investigated in inverted polymer solar cells, in which Au NRs are sandwiched between two layers of ZnO. Accompanied by the optimization of thickness of ZnO covered on Au NRs, a high-power conversion efficiency of 3.60% and an enhanced short-circuit current density ( JSC) of 10.87 mA/cm2 are achieved in the poly(3-hexylthiophene): [6,6]-phenyl-C61-butyric acid methyl ester (P3HT:PC60BM)-based inverted cell and the power conversion efficiency (PCE) is enhanced by 19.6% compared with the control device. The detailed analyses of the light absorption characteristics, the simulated scattering induced by Au NRs, and the electromagnetic field around Au NRs show that the absorption improvement in the photoactive layer due to the light scattering from the longitudinal axis and the near-field increase around Au NRs induced by localized surface plasmon resonance plays a key role in enhancing the performances.

Keyword: 52.35.Hr; 52.25.Tx; 68.37.Lp; 68.37.Ps; organic solar cells; nanorods; LSPR effect; scattering
1. Introduction

Polymer-fullerene bulk heterojunction solar cells (SCs) composed of interpenetrating networks of electron donors and acceptors, are now attracting a great deal of attention and show increasing importance due to their flexibility, low-cost, and simple manufacture process. However, a low charge mobility and a high bandgap for most of polymers lead to the carrier recombination and the limitation of the light absorption in an infrared region, thus restricting the overall performances of these polymer SCs (PSCs).[1, 2] The recent developments of narrow-bandgap photoactive materials and employing tandem structures are efficient to broaden the light absorption band of the active layer, thus enhancing the power conversion efficiency (PCE) of PSCs to over 9%.[35] In addition, incorporating metallic nanoparticles (NPs) into devices to enhance light harvesting is also an attractive approach to improving the performances of SCs. Metallic NPs are well known for their strong interactions with electromagnetic waves due to the collective oscillations of the conduction electrons within the metallic particles.[68] This near-field strength generated by the localized surface plasmon resonance (LSPR) is beneficial to improving light harvesting in the photoactive layer, [9] and hence increasing the short-circuit current density (JSC) and PCE in organic solar cells (OSCs).

The devices made via incorporating Au or Ag NPs into a carrier extraction layer, [1012] a photoactive layer, [1315] or depositing them onto the electrode surface, [1618] have been widely reported with significant performance improvement. One of the most significant PCE enhancements was reported, [18] in which an enhancement factor of 70% was obtained through inserting Ag NPs below the poly(3, 4-ethylenedioxythiophene): poly(4-styrenesulfonate) (PEDOT:PSS) layer. Unfortunately, its overall efficiency was rather low. Qu et al.[19] numerically simulated the absorption enhancement values using the finite element method with a three-dimensional model and demonstrated that plasmonic enhancement of above 100% could be obtained by optimizing the shapes, the sizes, and the positions of Ag NPs. However, despite these reports on performance improvement, a significant PCE degradation was also observed when Ag NPs without being capped by any insulator shell were embedded into a photoactive layer[20] and the continuous decline of PCE with the increase in Ag NPs concentration revealed that the intimate contact between the metallic NPs and the photoactive layer would promote trap-assisted recombination and degrade the performance of the plasmonic SCs. On the other hand, simulation results in the literature indicated that the near-field intensity induced by LSPR is commonly confined in a range of ∼ 20  nm around metallic NPs surface, and at the position far away from NPs surface, the local field weakens very quickly, [2123] resulting in no improvement in absorption of the photoactive layer. Hence, the distance between NPs and photoactive layer must be considered to obtain an obvious performance enhancement.

The LSPR wavelength is another important factor influencing the optimization of performances in plasmonic PSCs. Several reports have demonstrated that the greater enhancement in PCE was observed when the extinction peak wavelength of metallic NPs was located in a poor light absorption region instead of near the absorption peak of the photoactive material.[24, 25] At this point, the improvement on photocurrent using spherical Ag or Au NPs is restricted because of their LSPR peaks around 430  nm and 530  nm, respectively. Compared with nanospheres (NSs), rod-shaped metallic NPs exhibit a very good performance due to their widely tunable resonance peaks from the visible to near-infrared region by a simple alteration of longitudinal/transverse ratio. Janković et al.[24] reported a 14.4% PCE increase by inserting Au/SiO2 core/shell nanorods (NRs) with an extinction peak of 830  nm into poly [2, 6-4, 8-di (5-ethylhexylthienyl) benzo [1, 2-b; 3, 4-b] dithiophene-alt-5-dibutyloctyl-3, 6-bis (5-bromothiophen-2-yl) pyrrolo [3, 4-c] pyrrole-1, 4-dione]: [6, 6]-phenyl-C61-butyric acid methyl ester (PBDTT-DPP:PC60BM), which is far higher than the PCE value of the same structure Au/SiO2 core/shell NSs. In Lu et al.’ s research, [25] employing Au NRs with transverse and longitudinal axes of 15  nm and 40– 50  nm into PEDOT:PSS also shows good behaviors including photocurrent and PCE compared with those with Au NSs.

In this paper, we report a plasmonic-enhanced PSC based on Au NRs and instead of the commonly used PEDOT:PSS, ZnO is used as the ambient medium around Au NRs due to its significantly high refractive index, which will generate an obviously enhanced near-field intensity around Au NRs. Considering a quick electric field attenuation with distance, the ZnO thickness between the photoactive layer and Au NRs is vital and explored in this paper in order to obtain an improved device performance. Results demonstrate that 28-nm ZnO as the electron extraction layer generates the highest cell performance and has an optimized Au NRs’ distribution density and distance to the active layer, a 19.6% PCE enhancement is observed compared with those without Au NRs. The possible influence factors including changes in the light field inside the device and scattering induced by Au NRs are investigated to explore the origin of performance improvement.

2. Experiment

The solution of the photoactive layer was prepared as follows. The mixed solution consisting of poly(3-hexylthiophene) (P3HT) and PC60BM were dissolved into 1, 2-dichlorobenzene with a concentration of 15:15  mg/mL and stirred at 50  ° C overnight in a N2-filled glove box. The ZnO precursor was prepared by dissolving zinc acetate dihydrate and monoethanolamine in ethanol solution, respectively, with concentrations of 21.9  mg/ml and 6.7  μ L/ml, followed by at least 12 hours of stirring at 60  ° C.

Figure  1(a) shows the absorption spectrum of Au NRs in aqueous solution and the inset shows the transmission electron microscopy (TEM) image of Au NRs. Their average lengths of longitudinal and transverse axes are 42  nm and 12  nm, corresponding to the LSPR peaks of 515  nm and 713  nm, respectively. The statistics on the aspect ratio (AR) of Au NRs in Fig.  1(b) shows that the ARs for most Au NRs used in our experiment fall in a range of 2.9– 3.5. The Au NRs were obtained by centrifuging twice at 14000  rpm for 10  min to remove dispersant, and then dispersed in deionized water again by vigorously stirring to prevent their aggregation. Note that prior to spin coating, the solution containing Au NRs should be sonicated for at least 20  min to ensure that the Au NRs are well dispersed.

Fig.  1. (a) UV– vis absorption spectrum of Au NRs in aqueous solution. The inset shows the TEM image of Au NRs with a bar of 100  nm in length. (b) Statistics on the Au NRs' ratios from 400 particles in the TEM image.

Inverted devices were fabricated on patterned ITO-coated glass substrates. The substrates were washed by sonication in acetone, ethanol, and deionized water for 10  min, in sequence, and then dried at 100  ° C for 10  min in an ambient atmosphere. Before spin-coating Au NRs, a 20-nm ZnO film was formed onto the cleaned ITO-coated substrate by spin-coating ZnO precursor at 3000  rpm for 60  s and subsequently pyrolysis on a hot plate in air at 150  ° C for 30  min. The solution containing Au NRs was then spin-coated onto the ZnO layer, followed by 5  min drying at 100  ° C to evaporate the solvent. After that another ZnO overlayer ranging from 4 to 20  nm was deposited, followed by pyrolyzation at 150  ° C for 20  min. The thickness of the ZnO overlayer is controlled by adjusting the spin-coating speed and measured with ellipsometry. The ZnO-covered substrates were transferred into the glove box, followed by spin-coating the P3HT:PC60BM blend layer at 800  rpm for 60 s with an approximate film thickness of 70  nm. The film was then solvent annealed for 90  min in the glove box at room temperature. Finally, the devices were completed by sequential thermal evaporation of an 8-nm MoO3 hole extraction layer and an 80-nm Ag anode under a high vacuum of 6.0 × 10− 4  Pa. The device structure is shown in Fig.  2.

Fig.  2. Schematic structure for our cells.

Photovoltaic characteristics were measured with a solar simulator (Oriel 94023A, AAA, 300  W, USA) with an AM 1.5  G filter (Oriel, USA). The power of the simulated light was calibrated to 100  mW/cm2 by using an Oriel Solar standard Si solar cell. The curves of current density versus voltage (JV) were measured with a Keithley 2400 source meter and recorded with a computer. The absorption spectra were studied using ultraviolet (UV)– visible spectroscopy (Shimadzu UV-3600). Scanning electron microscope (SEM) imaging was carried out with JEOL JSM-6700 SEM system. The curves of incident photon versus conversion efficiency (IPCE) were measured with normal incident light from the monochromator (Oriel Instrument, model 74100).

3. Results and discussion

Three groups of devices are fabricated with different ZnO cathode buffer layer thickness values, where 20  nm ZnO/Au NRs/20  nm ZnO overlayer, 20  nm ZnO/Au NRs/8  nm ZnO overlayer, and 20  nm ZnO/Au NRs/4  nm ZnO overlayer are denoted as groups 1, 2, and 3. Note that in each group the control device that does not contain Au NRs but has the same ZnO thickness as that in the device with Au NRs, is also fabricated for comparison. Keeping the same ZnO thickness in each group would eliminate the difference in electron transport ability induced by the ZnO thickness variance. Figure  3(a) shows the JV characteristic curves for all SCs under AM1.5 illumination. The results show that both the plasmonic device and the control device with an 8-nm ZnO overlayer exhibit the best performances in the three groups of devices. The Au NRs-incorporated device with an 8-nm ZnO overlayer exhibits a PCE enhancement factor of about 19.6% compared with that of its control device. The detailed parameters are listed in Table  1, and the increase in JSC is mainly responsible for the performance improvement, which increases from 8.90  mA/cm2 to 10.51  mA/cm2 accompanied by the employment of Au NRs.

Fig.  3. (a)  JV characteristic curves for the plasmonic devices with various thickness values of ZnO overlayer and their control devices without Au NRs. (b)  IPCE curves for the plasmonic device with an 8-nm ZnO overlayer and its control device.

Table 1. Photovoltaic characteristics for the plasmonic device with various thickness values of ZnO overlayer and its control device.

To analyze the origin of the increase in JSC, we first measure the IPCE curves for the device with 20  nm ZnO/Au NRs/8  nm ZnO overlayer and its control device without Au NRs. As shown in Fig.  3(b), the device with incorporating Au NRs exhibits obviously increased IPCE values in a range of 450– 620  nm, with the peak value of efficiency enhanced from 0.43 to 0.53 at 525  nm. The theoretical JSC value converted from the IPCE curve is 8.35  mA/cm2 for the 8-nm ZnO overlayer device, showing a 20.3% enhancement factor compared with 6.94  mA/cm2 in the control one, which is approximately consistent with the enhancement ratio of 18.1% among the measured JSC values. It should be pointed out that the calculated JSC value slightly lower than the measured one in Table  1 is mainly ascribed to a smaller device area of 5  mm2 than the light spot in the IPCE measurement, thereby leading to loss of a fraction of light in these devices. Because the IPCE depends on the light absorption in the active layer, the approximate enhancement factors from the measurement and calculation results indicate that the increased JSC in the device with 20  nm ZnO/Au NRs/8  nm overlayer is due to the absorption improvement in the photoactive layer.

To further confirm our analysis above, we prepare two structures ITO/ZnO (28  nm)/P3HT: PC60BM and ITO/ZnO (20  nm)/Au NRs/ZnO (8  nm)/P3HT: PC60BM to measure their light absorption spectra as shown in Fig.  4. The inset shows an SEM image of Au NRs deposited on the ZnO layer surface, from which one observes that Au NRs are well dispersed on the bottom ZnO film with their longitudinal axes parallel to the film, but their orientations are arbitrary in the film. Besides, the Au NRs' density on the ZnO surface is calculated to be 120  particles/μ m2. The light absorption spectra in Fig.  4 show that Au NRs inserted between the two ZnO layers indeed improve the light absorption of P3HT:PC60BM, consistent with our IPCE results. From the absorption spectrum of Au NRs in Fig.  1, we clearly find that the resonance peak excited by Au NR transverse axis is near maximum absorption wavelength of P3HT:PC60BM, indicating that the enhanced light absorption in the active layer is attributed to the increased light field excited by Au NRs. As we know, when sunlight is incident on metallic NPs, the surface charges of metallic NPs interact with the electromagnetic field, then the electric field in the vicinity of metallic NPs is strongly increased due to the collective oscillations of surface electrons induced by incident light. The overlap between the LSPR wavelength and absorption band of active material will help to enhance the absorption efficiency of the solar cell. This is further proved to be true by calculating the incident electric field distribution within the devices (the effect of magnetic field is negligible, thus only electric field intensity is considered). From Figs.  5(a)– 5(c), we can see that the electric field intensities are enhanced by inserting the Au NRs at 550, 600, and 635  nm, respectively. We also observe that among all the wavelengths, the electric field increase is the most obvious at 635  nm, which means that the LSPR wavelength moves towards a longer wavelength compared with the plasmonic peak of the transverse axis in aqueous solution in Fig.  1. We attribute this red shift to the fact that the refractive index of ZnO is higher than that of the aqueous solution.

Fig.  4. Absorption spectra for the structures of ZnO (28  nm)/P3HT:PC60BM and ZnO (20  nm)/Au NRs/ZnO overlayer (8  nm) /P3HT:PC60BM. The inset shows an SEM image of Au NRs spin-coated onto the ZnO film with a bar of 400  nm.

Fig.  5. Simulated electric field distributions around Au NR in the plasmonic device with 8-nm ZnO overlayer at (a)  550  nm, (b)  600  nm, and (c)  635  nm along the XZ plane. (d)  Variances of the electric field intensity at 635  nm along the Z axis at x = 0 in the devices (1: ZnO (28  nm)/P3HT:PC60BM; 2: ZnO (20  nm)/Au NRs/ ZnO (8  nm)/P3HT:PC60BM; 3: ZnO (40  nm)/P3HT:PC60BM; 4: ZnO (20  nm)/Au NRs/ZnO (20  nm)/P3HT:PC60BM). The X and Z axes are along the longitudinal axis of Au NR and the incident direction of sun light, respectively. The inset shows the difference in electric field of P3HT:PC60BM layer between the plasmonic and its control devices.

Note that the LSPR effect from the longitudinal axis is not effectively utilized because its plasmon resonance band fails to fall in the P3HT:PC60BM absorption range of 400– 650  nm. It was reported that for the metallic NPs with small sizes of 5– 20  nm, the near-field intensity generated by the LSPR effect is mainly responsible for the light harvesting enhancement. While for the larger size NPs, light scattering makes the incident light travel a longer path, thus giving rise to an obvious absorption increase within the active layer.[2628] Figure  6 shows the simulation results of the light scattering cross sections of the glass/ITO/ZnO (28  nm) with/without Au NRs/P3HT:PC60BM/MoO3/Ag multilayer films. The scattering cross section (SCS) without Au NRs is about 10− 28, while it increases up to 10− 18 when incorporating Au NRs into the ZnO extraction layer, about 10 orders of magnitude. From the above results, we speculate that light scattering from the longitudinal axis is beneficial to increasing light harvesting and further raising the OPV’ s PCE because of the relatively large size Au NPs used in our experiments. The role of Au NRs’ light scattering in absorption enhancement is also proved by comparing the plasmonic and its control devices in group  1. As shown in Table  1, the JSC in the plasmonic device with a 20-nm ZnO overlayer increases from 7.91 to 8.47  mA/cm2, corresponding to an enhancement factor of 7.1%. Figure  5(d) reveals that the variations of electric field intensity with the incident direction of sun light in the plasmonic devices with 8-nm and 20-nm ZnO overlayer and their control devices. It is clear that although the electric field increase induced by LSPR can be observed around Au NRs, the intensive local field can hardly reach the photoactive layer because of the thick ZnO overlayer between P3HT:PC60BM and Au NRs, which means that Au NR-induced LSPR has no contribution to light harvesting enhancement in a thick ZnO overlayer, so we conclude that light scattering by Au NRs is the main factor for the JSC improvement in the device with a 20-nm ZnO overlayer. While for an 8-nm ZnO overlayer, the electric field in the photoactive layer shows a significant enhancement. As shown in the inset in Fig.  5(d), the intensive local field induced by LSPR can extend to the depth of about 15  nm in P3HT:PC60BM. Through our discussion above, we attribute the JSC improvement in the plasmonic device with an 8-nm ZnO overlayer to the absorption enhancement in the photoactive layer induced by both Au NRs’ LSPR effect and their scattering to the incident light.

Fig.  6. Calculated scattering cross sections for glass/ITO/ZnO (28  nm) without (a) and with (b) Au NRs/P3HT:PC60BM/MoO3/Ag multilayer films.

To investigate the influence of the insertion of Au NRs on the active layer morphology, we measure the morphologies  of  P3HT/PC60BM with and without Au NRs by using atom force microscope (AFM). Figure  7 shows the typical height and phase images of P3HT/PC60BM layer spin-coated on ZnO with and without Au NRs. We can see that the active layer with Au NRs is rougher than that without Au NRs, which has root-mean-squared surface roughness of 4.3  nm in Fig.  7(b) compared with 3.3  nm in Fig.  7(a). Besides, islands and valleys are apparent, while ordered P3HT chain alignment is clearly visible in Figs.  7(b) and7(d), but they are absent in Figs.  7(a) and7(c). These two features may be beneficial to the enhancement of charge transport in active layer and charge collection by electrode.

Fig.  7. AFM images of P3HT:PC60BM films spin-coated on ZnO without ((a), (c)) and with ((b), (d)) Au NRs. Panels (a) and (b) are height images, and panels (c) and (d) are phase images.

Unlike the improved performances in the device with a 20-nm or 8-nm ZnO overlayer, the device with an only 4-nm ZnO overlayer exhibits a seriously declined performance compared with its control device 3, which is mainly caused by low VOC and FF values of 0.50  V and 0.45. Such a large VOC and an FF loss indicate a direct contact of the Au NRs with P3HT:PC60BM due to a very thin ZnO overlayer, leading to a nonohmic contact between the electrode and the active layer.[2931] To confirm our deduction, x-ray photoelectron spectroscopy (XPS) spectra of ZnO/Au NRs/ZnO (4  nm) and ZnO/Au NRs/ZnO (20  nm) are measured. Compared with that in the ZnO/Au NRs/ZnO (20  nm) multilayer in Fig.  8(a), the significant Au 4f signal intensity in the ZnO/Au NRs/ZnO (4  nm) multilayer in Fig.  8(b) confirms our speculation that the 4-nm ZnO cannot completely cover the Au NRs and the exposure of most of the Au NRs under the P3HT:PC60BM active layer results in a non-ohmic contact between the cathode and the photoactive layer. Therefore, an increased interface energetic barrier due to a high work function of Au (5.1  eV) accounts for the decrease in VOC.[31] Moreover, these exposed Au NRs serving as the recombination centers lead to the quenching of excitons in the vicinity of Au NRs, and thus a further decrease in FF. This can be proved with the measured photoluminescence (PL) spectra of three multilayers of ZnO/P3HT:PC60BM, ZnO/Au NRs/ZnO (8  nm)/P3HT:PC60BM, and ZnO/Au NRs/ZnO (4  nm)/P3HT:PC60BM as shown in Fig.  9. By spin-coating an 8-nm ZnO layer on the top of Au NRs, the PL peak intensity of the photoactive layer is remarkably enhanced by 51% at 734  nm, meaning more photogenerated excitons due to the coupling between the plasmonic field of the Au NRs and the excited state of P3HT molecules. This is consistent with the improvement on the absorption and JSC. In contrast, the obviously decreased PL intensity for the film with a 4-nm ZnO overlayer indicates that the excitons quenching in the vicinity of Au NRs is dominant when the inserted ZnO film between the Au NRs and the photoactive layer is too thin.

Fig.  8. XPS spectra of (a) ZnO/Au NRs/ZnO (20  nm) and (b) ZnO/Au NRs/ZnO (4  nm) multilayers.

Fig.  9. PL spectra of three multilayers of ZnO/P3HT:PC60BM, ZnO/Au NRs/ZnO (8  nm)/P3HT:PC60BM, and ZnO/Au NRs/ZnO (4  nm)/P3HT:PC60BM. A 480-nm laser is used as the excitation source.

4. Conclusions

In this paper, we report the enhanced performance by sandwiching Au NRs between two layers of ZnO in the polymer solar cell with a structure of ITO/ZnO/Au NRs/ZnO overlayer/P3HT:PC60BM/MoO3/Ag. The JV curves show that the device with an 8-nm ZnO overlayer exhibits the best performance with a PCE of 3.60% and a JSC of 10.51  mA/cm2, corresponding to increase of 19.6% and 18.1% compared with the control device. Our analysis reveals that the near-field strength induced by LSPR around Au NRs plays a major role in improving the absorption of P3HT:PC60BM. In addition, light scattering from the longitudinal axis of Au NRs, resulting in the increase of OPL in active layer, may also be responsible for the absorption enhancement of active layer.

Reference
1 Coakley K M and McGehee M D 2004 Chem. Mater. 16 4533 DOI:10.1021/cm049654n [Cited within:1]
2 Moon J S, Takacs C, Cho S, Coffin R C, Kim H, Bazan G C and Heeger A J 2010 Nano Lett. 10 4005 DOI:10.1021/nl101923m [Cited within:1]
3 Peet J M L, Senatore M L, Heeger A J and Bazan G C 2009 Adv. Mater. 21 1521 DOI:10.1002/adma.v21:14/15 [Cited within:1]
4 You J, Dou L, Yoshimura K, Kato T, Ohya K, Moriarty T, Emery K, Chen C, Gao J, Li G and Yang Y 2013 Nat. Commun. 4 1446 DOI:10.1038/ncomms2411 [Cited within:1]
5 Zhou H, Zhang Y, Seier J, Collins S D, Luo C, Bazan G C, Nguyen T Q and Heeger A J 2013 Adv. Mater. 25 1646 DOI:10.1002/adma.201204306 [Cited within:1]
6 Huang Q, Wang J, Cao L R, Sun J, Zhang X D, Geng W D, Xiong S Z and Zhao Y 2009 Acta Phys. Sin. 58 1980(in Chinese) [Cited within:1]
7 Zhou R L, Chen X S, Zeng Y, Zhang J B, Chen H B, Wang S W, Lu W, Li H J, Xia H and Wang L L 2008 Acta Phys. Sin. 57 3506(in Chinese) [Cited within:1]
8 Barnes W L, Dereux A and Ebbesen T W 2003 Nature 424 824 DOI:10.1038/nature01937 [Cited within:1]
9 Xiang C P, Jin Y, Liu J T, Xu B Z, Wang W M, Wei X, Song G F and Xu Y 2014 Chin. Phys. B 23 038803 DOI:10.1088/1674-1056/23/3/038803 [Cited within:1]
10 Mahmoud A Y, Zhang J, Ma D, Izquierdo R and Truong V V 2012 Org. Electron. 13 3102 DOI:10.1016/j.orgel.2012.09.015 [Cited within:1]
11 Tan K S, Chuang M K, Chen F C and Hsu C S 2013 ACS Appl. Mater. Interfaces 5 12419 DOI:10.1021/am403374p [Cited within:1]
12 Hao J Y, Xu Y, Zhang Y P, Chen S F, Li X A, Wang L H and Huang W 2015 Chin. Phys. B 24 045201 DOI:10.1088/1674-1056/24/4/045201 [Cited within:1]
13 Heo S W, Lee E J, Song K W, Lee J Y and Moon D K 2013 Org. Electron. 14 1931 DOI:10.1016/j.orgel.2013.04.042 [Cited within:1]
14 Li X, Choy W C H, Lu H, Sha W E I and Ho A H P 2013 Adv. Funct. Mater. 23 2728 DOI:10.1002/adfm.201202476 [Cited within:1]
15 Wang D H, Kim D Y, Choi K W, Seo J H, Im S H, Park J H, Park O O and Heeger A J 2011 Angew. Chem. Int. Ed. 50 5519 DOI:10.1002/anie.201101021 [Cited within:1]
16 Chen X, Zhao C, Rothberg L and Ng M K 2008 Appl. Phys. Lett. 93 123302 DOI:10.1063/1.2988190 [Cited within:1]
17 Heo M, Cho H, Jung J W, Jeong J R and Park S 2011 Adv. Mater. 23 5689 DOI:10.1002/adma.v23.47 [Cited within:1]
18 Morfa A J, Rowlena K L, Reilly T H, Romero M J and Lagemaatb J V D 2008 Appl. Phys. Lett. 92 013504 DOI:10.1063/1.2823578 [Cited within:2]
19 Qu D, Liu F, Huang Y, Xie W and Xu Q 2011 Opt. Express 19 24795 DOI:10.1364/OE.19.024795 [Cited within:1]
20 Wu B, Wu X, Guan C, Tai K F, Yeow E K L, Fan H J, Mathews N and Sum T C 2013 Nat. Commun. 4 2004 DOI:10.1038/ncomms3004 [Cited within:1]
21 Anger P, Bharadwaj P and Novotny L 2006 Phys. Rev. Lett. 96 113002 DOI:10.1103/PhysRevLett.96.113002 [Cited within:1]
22 Chen S F, Cheng F, Mei Y, Peng B, Kong M, Hao J Y, Zhang R, Xiong Q H, Wang L H and Huang W 2014 Appl. Phys. Lett. 104 213903 DOI:10.1063/1.4880575 [Cited within:1]
23 Zhou J, Xue M, Shen H, Wu Z, Kim S and Ho J J 2011 Appl. Phys. Lett. 98 151110 DOI:10.1063/1.3577611 [Cited within:1]
24 Janković V, Yang Y, You J, Dou L, Liu Y, Cheung P, Chang J P and Yang Y 2013 ACS Nano 7 3815 DOI:10.1021/nn400246q [Cited within:2]
25 Lu L, Luo Z, Xu T and Yu L 2013 Nano Lett. 13 59 DOI:10.1021/nl3034398 [Cited within:2]
26 Atwater H A and Polman A 2010 Nat. Mater. 9 205 DOI:10.1038/nmat2629 [Cited within:1]
27 Dijk M A V, Tchebotareva A L, Orrit M, Lippitz M, Berciaud S, Lasne D, Cognetc L and Lounisc B 2006 Phys. Chem. Chem. Phys. 8 3486 DOI:10.1039/b606090k [Cited within:1]
28 Hsiao Y S, Charan S, Wu F Y, Chien F C, Chu C W, Chen P and Chen F C 2012 J. Phys. Chem. C 116 20731 DOI:10.1021/jp306124n [Cited within:1]
29 Kim K and Carroll D L 2005 Appl. Phys. Lett. 87 203113 DOI:10.1063/1.2128062 [Cited within:1]
30 Mihailetchi V D, Blom P W M, Hummelen J C and Rispens M T 2003 J. Appl. Phys. 94 6849 DOI:10.1063/1.1620683 [Cited within:1]
31 Sun J, Zhu Y, Xu X, Lan L, Zhang L, Cai P, Chen J, Peng J and Cao Y 2012 J. Phys. Chem. C 116 14188 DOI:10.1021/jp3009546 [Cited within:2]