- Email: [email protected]
TFSA doped interlayer for efficient organic solar cells
Organic Electronics 15 (2014) 3702–3709
Contents lists available at ScienceDirect
Organic Electronics journal homepage: www.elsevier.com/locate/orgel
TFSA doped interlayer for efficient organic solar cells Yubin Xiao a, Shuang Zhou a, Yaorong Su a, Lei Ye a, Sai-wing Tsang b,1, Fangyan Xie c, Jianbin Xu a,⇑ a Department of Electronic Engineering and Materials Science and Technology Research Centre, The Chinese University of Hong Kong, Shatin, Hong Kong Special Administrative Region b Department of Physics and Materials Science, The City University of Hong Kong, Kowloon Tong, Hong Kong Special Administrative Region c Instrumental Analysis and Research Center, Sun Yat–sen University, Guangzhou 510275, PR China
a r t i c l e
i n f o
Article history: Received 19 August 2014 Received in revised form 15 October 2014 Accepted 16 October 2014 Available online 30 October 2014 Keywords: Molecular doping Interlayer Charge injection OPVs
a b s t r a c t Organic solar cells based on bis(trifluoromethanesulfonyl)amide (TFSA, [CF3SO2]2NH) doped poly[N-900 -hepta-decanyl-2,7-carbazole-alt-5,5-(40 ,70 -di-2-thienyl-20 ,10 ,30 -benzothiadiazole) (PCDTBT) were fabricated to investigate the effect of molecular doping. By replacing poly(3,4-ethylenedioxythiophene) (PEDOT): poly(styrenesulfonate) (PSS) with a thin TFSA layer, we have found more efficient charge injection at anode/active interface enhanced photovoltaic performance. The doping effect is confirmed by photoemission spectroscopy that the Fermi level of doped PCDTBT shifts downward to its HOMO level and results in higher carrier concentration. The reduced injection barrier also evidented by impedance spectroscopy that the real impedance of the TFSA doped PCDTBT solar cell decreases more than 50%. Using the molecular doping approach, the overall power conversion efficiency (PCE) was largely increased from 4.70% to 5.98%. Our results suggest that TFSA functions not only as a surface doping molecule, but also an anode interfacial layer to replace the conventional PEDOT:PSS. Ó 2014 Published by Elsevier B.V.
1. Introduction Polymer solar cell is an emerging technology for renewable photovoltaic with the merits of easy processing, low cost, light weight, as well as the transparency and flexibility [1–3]. To date, the state-of-art organic solar cell has reached the power conversion efficiency (PCE) threshold of 10%, which has great competitiveness against other emerging solar cell technologies [4,5]. The major challenges hindering the advance of organic solar cells are the short exciton migration distance, narrow light absorption, and low stability [2,6–8]. Hence, plenty of efforts have ⇑ Corresponding author. E-mail addresses: [email protected] (S.-w. Tsang), [email protected]. edu.hk (J. Xu). 1 Co-corresponding author. http://dx.doi.org/10.1016/j.orgel.2014.10.024 1566-1199/Ó 2014 Published by Elsevier B.V.
been devoted to improving the device performance, such as incorporating second or third solvents (1,8-diiodooctane (DIO) or chloroform (CF)) to optimize donor–acceptor morphology and thus minimizing the recombination rate [9– 12], adding quantum dots (QDs) or nanoparticles to enhance the light absorption with a broader spectrum [13–16], or employing tandem structure to increase the open circuit voltage [17,18]. Besides the above issues, the carrier injection barrier at electrode/active layer interface, one of the most important figures of merit for organic materials also plays a significant role in determining the device performance. Molecular doping, one of the facile method to manipulate carrier density has been investigated for a long time and be considered as an effective approach to control the electrical property [19–22]. Recently, researchers found that the device performance and photoconductivity could be efficiently
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
improved by doping polymers (Poly(3-hexylthiophene-2,5diyl)(P3HT), Poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-cyclopenta[2,1-b,3,4-b0 ]dithiophene)-alt-4,7(2,1,3-benzothiadiazole)](PCPDTBT)) with widely used organic p-type dopant tetrafluoro–tetracyanoquinodimethan (F4-TCNQ). The resulting polymers had downward shifted the donor Fermi level and the increased carrier density [23,24]. Moreover, a hole-collecting favourable contact on P3HT diodes can be created by means of p-type doping strategy [25]. On the other hand, bis(trifluoromethanesulfonyl)amide (TFSA) has been reported as a valid p-type chemical dopant for carbon based materials (carbon nanotube, graphene) due to its strong electron withdrawing group [26–28]. Schottky junction solar cells fabricated with TFSA doped graphene have exhibited a PCE of 8.6% [29]. In addition, in polymer light emitting diodes, the work function of graphene electrode was tuned by TFSA doping for efficient hole injection [30]. In the present work, the effects of TFSA on the performance of solution processed bulk heterojunction organic photovoltaics (BHJ-OPV) are investigated. It is found that by utilizing a thin layer of TFSA between ITO and the active layer, the hole injection barrier at anode/polymer interface is reduced and the carrier injection becomes more efficient. Moreover, the TFSA incorporated device (without PEDOT:PSS) shows an improved PCE of 5.98% when compared with the traditional ITO/PEDOT:PSS/active layer/LiF/Al structure (PCE of 4.70%). Therefore, it is suggested that besides acting as a p-type doping molecule at the anode interface, TFSA also has the potential to serve as an alternative buffer layer to replace the commonly used PEDOT:PSS, which is always regarded as the cause for device degradation [31–34]. 2. Experiments In order to demonstrate the effect of TFSA, we fabricated a control device with configuration of ITO (180 nm)/PEDOT:PSS(40 nm)/PCDTBT:PC71BM (80 nm)/LiF (1 nm)/Al (100 nm). PCDTBT was purchased from 1-material Chemscitech, Inc., and used as received. Firstly, a pre-patterned ITO (conductivity: 10–15 X/square) glass was cleaned by detergent, deionized water, acetone and isopropanol in sequence, followed by oxygen plasma treatment for 90 s. Secondly, for control sample, a thin layer of PEDOT:PSS was spun-cast onto ITO glass with a thickness of ca. 40 nm, and then annealed at 145 °C for 10 min in air. For comparison, device without PEDOT:PSS, TFSA (dissolved in nitromethane and stirred overnight) was spun coated onto ITO and formed a thin film of 20 nm. Thereafter, the substrates were transferred to a N2 filled glove box. PCDTBT and PC71BM with a mass ratio of 1:3 were dissolved in a mixed solution composed of 1,2 o-dichlorobenzene (o-DCB) and chloroform (CF) with volume ratio of 1:1, and then spun-cast atop the PEDOT:PSS (or TFSA) layer to form the active layer, giving rise to a thickness of 80 nm. Following that, a 20-min soft annealing and a further annealing at 120 °C for 10 min were conducted. Finally, a thin LiF interfacial layer (1 nm) and a 100 nm thick Al electrode were deposited sequentially by thermal evaporation. For each condition, there are at least 60 solar cells were measured and the figures were drawing from the average value.
3703
The evaporator was BOC Edwards Auto 306 and the active layer area of the device was defined by a shadow mask of 2 mm 6 mm. Electrical measurements were performed by a semiconductor characterization system (Keithley 236) at room temperature in air under the spectral output from solar simulator (Newport) using an AM 1.5G filter with a light power of 100 mW/cm2. The light intensity was precisely calibrated by a calibrated silicon solar cell. The morphologies of the PCDTBT and PC71BM blend thin films were characterized by atomic force microscopy (AFM) in tapping mode. The thickness of the active layer and evaporated layers were recorded with a thickness monitor (Sigma SQM-160), and also verified by AFM.
3. Results and discussion Fig. 1(a) shows the chemical structure of materials that used in this study. Fig. 1(b) shows the solar cell configuration and Fig. 1(c) presents the energy band diagram in which PEDOT:PSS is replaced by TFSA. Once TFSA is in contact with PCDTBT, electrons will transfer from PCDTBT to TFSA due to the strong electron-accepting nature of TFSA. As electrons move to TFSA, the Fermi level of PCDTBT will shift downward to its HOMO level as indicated by the green arrow in Fig. 1(c). Such doping effect is confirmed by photoemission spectroscopy measurement that will be discussed later. As an initial step, the thickness and concentration of TFSA were optimized, as shown in Figs. 1s and 2s. Firstly, the thicknesses of TFSA were varied from 5 to 25 nm and the concentration was fixed at 20 mM. With only 5 nm TFSA, the Voc was increased to 0.87 V, when further increased to 10 nm, the Voc reached 0.9 V and kept constant in the following thicker films. Overall, with 20 nm TFSA atop, the highest PCE was obtained. There are two possible reasons for the increased Voc with thicker TFSA. Firstly, the different penetration depth of TFSA to the bulk active layer. At thin TFSA layer, like 5 nm, it is thick enough to cover the beneath ITO electrode, on the top side, after spin cast the active layer, there will be interpenetration of active layer into TFSA. At thin TFSA layer, the active layer could penetrate through the TFSA and may contact with ITO electrode to some extent, so there will be current leakage pathway there. For thicker TFSA, active layer could no longer penetrate through it, therefore the chance of leaking current was reduced and lead to stable Voc. Secondly, more TFSA, 10 nm or thicker film compared to 5 nm, facilitate more complete doping of PCDTBT at the interface. Generally the energy difference between the HOMO of the donor and the LUMO of the acceptor (effective bandgap) set the upper limit of the Voc in organic solar cell. However, the work-function of electrode has strong influence on the Voc below that limit. As the effective bandgap in PCDTBT:PC71BM is around 1.4 eV, improving the WF in this material system still can enhance the Voc. So in the rest of the study, the film thickness was fixed at 20 nm. For the concentration variation (5–25 mM), the cell with 20 mM TFSA doping showed the optimum PCE. Therefore, the film thickness and concentration were determined to be 20 nm
3704
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Fig. 1. (a) Chemical structure of TFSA, PCDTBT, PC71BM used in the study. (b) Device architecture of the PCDTBT:PC71BM solar cells. (c) Energy diagram of the BHJ solar cell. With the p-type doping, the green arrow indicates the Fermi level of PCDTBT shifted toward its HOMO level. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and 20 mM in this study. The detailed parameters were summarized in Tables 1s and 2s, respectively. Next in order to examine the carrier transport at the anode/PCDTBT and anode/active layer interface, an series of hole-only devices were fabricated with both undoped and doped PCDTBT (or PCDTBT:PC71BM) sandwiched between ITO/PEDOT:PSS and MoO3/Al or ITO/TFSA and MoO3/Al respectively. MoO3 in here is an electron blocking layer (EBL), so only hole can transport in the devices. Room temperature J1/2–V characteristics are given in Fig. 2(a), with solid and empty curves corresponding to PCDTBT and PCDTBT:PC71BM compounds respectively. For PCDTBT only devices, the doped PCDTBT shows nearly 2.5 times
higher root square current density at the range around open circuit voltage (0.9 V), while for PCDTBT:PC71BM hole only devices, the doped device shows nearly an order higher root square current density as well. Especially, noted that the doped PCDTBT:PC71BM hole-only diode exhibits even higher current density compared with undoped single PCDTBT layer devices, despite the scattering effect of PC71BM. Yet, doping is well known to lower the effective injection barrier at the interface by changing the molecular level alignment or/and creating a narrow depletion region for free carriers tunnelling through. Hence, it is believed that the TFSA p-type doping will decrease the injection barrier at the anode/active layer
Fig. 2. (a) J1/2–V characteristics of doped and undoped PCDTBT (solid symbol) and PCDTBT:PC71BM (open symbol) hole-only devices. (b) Nyquist plots of impedance spectra for PCDTBT:PC71BM solar cells without and with TFSA treatment and measured at the Voc condition under dark.
3705
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Table 1 Photovoltaic performance of PCDTBT:PC71BM solar cells with different interfacial structure between ITO and BHJ layer (Rsh, Rs extracted from illuminated J–V curve).
PEDOT:PSS TFSA PEDOT/TFSA TFSA/PEDOT
Voc (V)
Jsc (mA/cm2)
FF (%)
PCE (%)
Rsh (X cm2)
Rs (X cm2)
J (EQE) (mA/cm2)
0.84 0.90 0.88 0.85
10.2 11.1 11.2 8.9
55.0 59.9 58.9 59.2
4.70 5.98 5.80 4.50
423 497 477 448
15 10 8 13
9.9 10.9 11.0 8.8
interface which is responsible for the increased current density in the diode. As commonly recognized, the hole density of a p-doped polymer will increase and the increasing carrier concentration will result in a high hole mobility. Therefore, the hole mobility was calculated from the space-charge-limited current (SCLC) region to verify its variation. As shown in Fig. 2(a), for the undoped single PCDTBT layer device, the hole mobility is 3.66 104 cm2/V s while for the doped device, the hole mobility increases to 9.35 104 cm2/ V s. For the hole mobility in BHJ devices (with PCDTBT:PC71BM), it decreases compared with PCDTBTonly device. The possible reason is the scattering effect of PC71BM when it is mixed with PCDTBT. However, still the hole mobility of doped PCDTBT:PC71BM device (4.67 104 cm2/V s) presents nearly 4 times higher value compared with undoped one (1.37 104 cm2/V s). Since the mobility of electron in PCDTBT:PC71BM is on the order of 104 cm2/V s [35], a more balanced charge transport is beneficial for higher charge collection efficiency [36,37]. In order to understand the complete charge transport process, we also fabricated and characterized electron-only devices for pure PC71BM and TFSA treated PC71BM. As shown in Fig. 3s, the results show that the electron mobility is slightly decreased from 4.59 103 cm2/V s for pristine to 3.23 103 cm2/V s for doped one which indicate that TFSA has very less effect for the electron transport. To further understand the doping effect of TFSA, the impedance of PCDTBT:PC71BM solar cells was measured in a frequency range from 1 MHz to 0.1 Hz by utilizing impedance spectroscopy under dark. Fig. 2(b) presents the Nyquist plots of the impedance spectra for PCDTBT:PC71BM solar cells without and with TFSA treatment and measured under an external bias voltage at the Voc condition. For PCDTBT solar cells without TFSA treatment (black line), the real impedance is 1101 X, while for the one doped with TFSA (red 2line), the real impedance decreases to 545 X. The decreasing impedance indicates that the overall device resistance (bulk and contact resistance) is decreased due to the TFSA doping effect. According to the generally accepted concept that by doping conducting polymers, the charge injection barrier at the electrode and active layer contact can be optimized. Hence, the lowered resistance will undoubtedly conduce to reduce the current lost across the metal–organic junction which is attributed to the increment of the short circuit current [38].
2 For interpretation of color in Fig. 2, the reader is referred to the web version of this article.
For the next step, we studied the current–voltage characteristic and PCE, based on the device structures of (1) ITO/PEDOT:PSS/PCDTBT:PC71BM/LiF/Al and (2) ITO/TFSA/ PCDTBT:PC71BM/LiF/Al. The device performances are summarized in Table 1. As shown in Fig. 3(a), the control sample by using PEDOT:PSS has an open circuit voltage (Voc) of 0.84 V, a short circuit current (Jsc) of 10.2 mA/cm2, a fill factor (FF) of 55.0% and a PCE of 4.70%. Whereas for the TFSA treated PCDTBT solar cell, enhanced Voc to 0.90 V, Jsc to 11.1 mA/cm2, FF of 59.9% and PCE of 5.98% were obtained. The increased Jsc was further confirmed by external quantum efficiency (EQE) (Fig. 3(d)). From the above discussion, we could infer that the improved solar cell performance is attributed to the decreased carrier injection barrier and cell resistance of PCDTBT which leads to an efficient hole transportation and results in enhanced current density and fill factor. Moreover, Voc is increased from 0.84 to 0.90 V which is owing to the reduced charge recombination rate at the interface after TFSA doping. However, it has also been demonstrated that the work function of the metal and/or metal oxide electrodes can be modified by organic molecular, e.g., Zhou et al. have found that polymers such as polyethylenimine ethoxylated (PEIE) and branched polyethylenimine (PEI) have an universal effects on different electrodes such as ITO, graphene and Al [39]. Therefore, it is interesting to investigate whether the doping effect of TFSA is on ITO or PCDTBT. To clarify this problem, two more devices with slightly different structures were fabricated. For one device, it was configured as (3) ITO/PEDOT:PSS/TFSA/active layer/LiF/Al, where TFSA could only contact with active layer. For comparison, the other device had a structure of (4) ITO/TFSA/ PEDOT:PSS/active layer/LiF/Al, in other words, TFSA was spun-cast between ITO electrode and PEDOT:PSS, hence TFSA no longer contacted with active layer and the doping effects could be eliminated. From Fig. 3(a and b), we can see that device (3) has a PCE of 5.80%, while device (4) shows a lower PCE of 4.50%. With TFSA next to PCDTBT, the solar cell displays relatively better performance with Voc of 0.88 V, Jsc of 11.2 mA/cm2 and FF of 58.9%. However, for device (4) with TFSA aparting from PCDTBT:PC71BM, it shows lower Voc of 0.85 V, Jsc of 8.9 mA/cm2 and FF of 59.2%. Hence, except FF, device (4) has a similar Voc with device (1) and an even lower Jsc. Therefore, from the above results, we can confirm that the doping of TFSA is effective on PCDTBT instead of ITO electrode. Fig. 3(c) shows the dark current versus bias voltage characteristics. After doping with TFSA, reduced dark currents under reverse bias and lowered leakage currents at zero bias were observed which led to a relatively high
3706
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Fig. 3. (a) Current density versus voltage characteristics of PCDTBT:PC71BM solar cells under 100 mW/cm2 irradiation. (b) Jsc, PCE, Voc and FF of control sample, with TFSA doping, with TFSA next to PCDTBT and TFSA next to ITO electrode. (c) Dark current density versus voltage. (d) External quantum efficiency of PCDTBT:PC71BM solar cells.
diode rectification ratio, indicating a promisingly high level of the shunt resistance and thus contributing to a relatively high Voc as discussed previously [40]. Whereas for the dark current in the space charge region (1–2 V) that dominated by the faster carriers, a similar dark current level can be observed for all cases with or without doping. Apparently, it suggests that the p-doping with TFSA does not affect the electron transport in PCDTBT:PC71BM solar cells. In addition, active layer morphology is another important issue for organic solar cells since the charge transport is largely depended on it. Therefore, we investigated the influence of TFSA doping on the morphology characteristics of the active layer. Fig. 4(a)–(d) shows 1 lm 1 lm topographic and phase atomic force microscopy (AFM) images of the photoactive layer without and with TFSA doping under air, respectively. Without TFSA, the surface morphology of PCDTBT: PC71BM shows relatively smooth and featureless structure with a root mean square (RMS) roughness of 0.33 nm. While after doping with TFSA, an apparent surface reconstruction was observed, the active layer surface turns coarse with relatively larger domain size and the RMS increases to 0.60 nm. From the corresponding phase images, it can be seen that the domain without TFSA doping shows very blurry edges, while for the one treated with TFSA, it displays sharp and clear
domain boundaries. Generally, unclear morphology indicates impure domains and would lead to enhanced bimolecular recombination and reduced charge transport, therefore leads to a lower Jsc which is in good agreement with the previous electrical characterizing results [12]. The reasons of the changed morphology after TFSA doping are not fully understood yet. One cause of the modified morphology may due to the variation of polymer stacking, therefore we did grazing incidence X-ray diffraction (GIXRD) analysis as shows in Fig. 4s. However, it shows very similar peaks indicates no sign of changed polymer orientation. Also from UV–vis spectroscopy as shows in Fig. 5s, no peak shifting or new absorption peak were observed too. Since the feature of the polymer blends could be affected by the substrates [41], one possibility is that the difference in specific surface energy between TFSA and PEDOT:PSS which leading to the variation of the donor–acceptor domains and the increasing RMS. To further verify the above speculations, more investigations are still needed to be done and the related results will be reported elsewhere. In order to confirm the doping effect of TFSA on PCDTBT, the UPS measurement was conducted and the results are shown in Fig. 5. The value of work function (WF) of the polymer is defined by the energy difference between vac-
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
3707
Fig. 4. AFM images of PCDTBT:PC71BM, (a) morphology and (c) phase without TFSA doping; (b) morphology and (d) phase with TFSA doping.
Fig. 5. The UPS spectra of undoped and doped PCDTBT film at various TFSA concentrations.
uum level and Fermi level of the polymer. As determined by UPS, work function is equal to the difference between the low cutoff energy and the bias voltage. In our case, a 5 V bias voltage was applied in order to obtain the low cutoff energy levels of the secondary photo-electrons. For pristine and TFSA doped PCDTBT at each concentrations (5–25 mM), the lower cut-off energy levels of the secondary photo-electrons were 9.66, 9.71, 9.77, 9.86, 9.87, and 9.85 eV, respectively, as shown in Fig. 5. Concomitantly, the work function of the film increases from 4.66 to 4.71, 4.77, 4.86, 4.87 and 4.85 eV for the same doping concentration values respectively. Notice that the Fermi level at
25 mM did not showing further downside shifting, this could explain the slight decreasing PCE at 25 mM compare with 20 mM. Overall, it is apparent that with TFSA doping, the Fermi level of PCDTBT is effectively downward shifted to its HOMO level. With the lower Fermi level, the hole concentration at the anode/PCDTBT interface will increase and therefore reduce the charge injection barrier in between. Finally, to explore the doping applicability of TFSA to other donor polymers, solar cells with active layer composed of P3HT:PC61BM was fabricated. As shown in Fig. 6(a), the doped P3HT hole-only device presents a higher current density compared with undoped one, indicating a more efficient hole injection from ITO electrode to P3HT. This result is consistent with the previous study of PCDTBT hole-only diode. In comparison of the undoped P3HT devices, TFSA doped P3HT solar cells show a slightly increasing Voc from 0.60 V to 0.61 V, an increasing Jsc from 8.80 to 9.59 mA/cm2 and a similar FF. Because of the increasing Voc and Jsc, the overall PCE was increased from 3.31% to 3.63% after doping and the detail data was summarized in Table 2. Fig. 6(c) exhibits the UPS spectra of undoped and doped P3HT, the lower cut-off energy levels of the secondary photo-electrons were 9.08 and 9.58 eV, respectively. Hence, the corresponding work function of P3HT also goes deeper from 4.08 to 4.58 eV. Therefore, according to the above results, we can conclude that the TFSA molecular doping can be a generally applicable way to improve the performance of polymer solar cells.
3708
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Fig. 6. (a) J–V characteristics of doped and undoped P3HT hole-only devices. (b) Current density versus voltage characteristics of P3HT:PC61BM solar cells under 100 mW/cm2 irradiation. (c) The UPS spectra of undoped and doped P3HT film at 20 mM TFSA.
Voc (V)
Jsc (mA/cm2)
FF (%)
PCE (%)
WF (eV)
donors such as P3HT. Overall, our findings suggest a facile method to effectively enhance the hole mobility of PCDTBT and simultaneously reduce the solar cell resistance for more efficient organic photovoltaic cells.
0.60 0.61
8.80 9.59
62.7 62.1
3.31 3.63
4.08 4.58
Acknowledgements
Table 2 Photovoltaic performance of pristine and TFSA doped P3HT:PC61BM solar cells.
Undoped Doped
4. Conclusions In summary, we have shown that by inserting a thin layer of TFSA between the organic active layer and the ITO anode, the injection barrier at ITO/active layer interface can be greatly reduced and it leads to an improvement in Jsc, Voc, FF and the overall PCE of the PCDTBT:PC71BM based BHJ solar cells. The hole mobility (lh) of PCDTBT increases by nearly 4 times and results in a largely enhanced PCE from 4.70% to 5.98% for the undoped and doped devices, respectively. Impedance spectroscopy reveals that the real impedance of PCDTBT solar cell decreases more than 50% after TFSA doping. Although the roughness of the active layer increases a little, from the phase images, it can be inferred that the charge transportation is more efficient in well defined phase separation domains. Finally, the downward shifting of the Fermi level of PCDTBT is confirmed by UPS. Furthermore, the doping effects of TFSA can also be applicable to other polymer
This work is in part supported by Research Grants Council of Hong Kong, particularly, via Grant Nos CUHK2/ CRF/08, CUHK4182/09E, CUHK4179/10E, N-CUHK405/12 and AoE/P-03/08.J Xu would like to thank the National Science Foundation of China for the support, particularly, via Grant Nos. 60990314, 60928009 and 61229401. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.orgel.2014.10.024. References [1] G. Li, R. Zhu, Y. Yang, Nat. Photon. 6 (3) (2012) 153. [2] J.L. Brédas, J.E. Norton, J. Cornil, V. Coropceanu, Acc. Chem. Res. 42 (11) (2009) 1691. [3] C. Deibel, V. Dyakonov, Rep. Prog. Phys. 73 (9) (2010). [4] J.B. You, C.C. Chen, Z.R. Hong, K. Yoshimura, K. Ohya, R. Xu, S.L. Ye, J. Gao, G. Li, Y. Yang, Adv. Mater. 25 (29) (2013) 3973.
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709 [5] J.B. You, L.T. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C.C. Chen, J. Gao, G. Li, Y. Yang, Nat. Commun. 4 (2013). [6] A. Moliton, J.M. Nunzi, Polym. Int. 55 (6) (2006) 583. [7] H.J. Son, B. Carsten, I.H. Jung, L.P. Yu, Energy Environ. Sci. 5 (8 815) (2012). [8] H. Spanggaard, F.C. Krebs, Sol. Energy Mater. Sol. C 83 (2–3) (2004) 125. [9] B.R. Aich, J.P. Lu, S. Beaupre, M. Leclerc, Y. Tao, Org. Electron. 13 (9) (2012) 1736. [10] R.B. Aich, Y.P. Zou, M. Leclerc, Y. Tao, Org. Electron. 11 (6) (2010) 1053. [11] S. Alem, T.Y. Chu, S.C. Tse, S. Wakim, J.P. Lu, R. Movileanu, Y. Tao, F. Belanger, D. Desilets, S. Beaupre, M. Leclerc, S. Rodman, D. Waller, R. Gaudiana, Org. Electron. 12 (11) (2011) 1788. [12] L. Ye, S.Q. Zhang, W. Ma, B.H. Fan, X. Guo, Y. Huang, H. Ade, J.H. Hou, Adv. Mater. 24 (47) (2012) 6335. [13] E.A. Parlak, T.A. Tumay, N. Tore, S. Sarioglan, P. Kavak, F. Turksoy, Sol. Energy Mater. Sol. C 110 (2013) 58. [14] P.A. Staniec, A.J. Parnell, A.D.F. Dunbar, H.N. Yi, A.J. Pearson, T. Wang, P.E. Hopkinson, C. Kinane, R.M. Dalgliesh, A.M. Donald, A.J. Ryan, A. Iraqi, R.A.L. Jones, D.G. Lidzey, Adv. Energy Mater. 1 (4) (2011) 499. [15] D.M. DeLongchamp, R.J. Kline, D.A. Fischer, L.J. Richter, M.F. Toney, Adv. Mater. 23 (3) (2011) 319. [16] M. He, F. Qiu, Z.Q. Lin, J. Phys. Chem. Lett. 4 (11) (2013) 1788. [17] L.T. Dou, J.B. You, J. Yang, C.C. Chen, Y.J. He, S. Murase, T. Moriarty, K. Emery, G. Li, Y. Yang, Nat. Photon. 6 (3) (2012) 180. [18] M. Wright, A. Uddin, Sol. Energy Mater. Sol. C 107 (2012) 87. [19] L. Ley, Y. Smets, C.I. Pakes, J. Ristein, Adv. Funct. Mater. 23 (7) (2013) 794. [20] M.T. Greiner, M.G. Helander, W.M. Tang, Z.B. Wang, J. Qiu, Z.H. Lu, Nat. Mater. 11 (1) (2012) 76. [21] O.V. Kozlov, S.A. Zapunidi, Synth. Met. 169 (2013) 48. [22] Y.L. Gao, Mater. Sci. Eng. R 68 (3) (2010) 39. [23] X.Y. Han, Z.W. Wu, B.Q. Sun, Org. Electron. 14 (4 111) (2013). [24] X.F. Lei, F.T. Zhang, T. Song, B.Q. Sun, Appl. Phys. Lett. 99 (23) (2011) 233305.
3709
[25] A. Dai, Y.H. Zhou, A.L. Shu, S.K. Mohapatra, H. Wang, C. FuentesHernandez, Y.D. Zhang, S. Barlow, Y.L. Loo, S.R. Marder, B. Kippelen, A. Kahn, Adv. Funct. Mater. 24 (15) (2014) 2197. [26] M. Herstedt, M. Smirnov, P. Johansson, M. Chami, J. Grondin, L. Servant, J.C. Lassègues, J. Raman Spectrosc. 36 (8) (2005) 762. [27] S. Tongay, K. Berke, M. Lemaitre, Z. Nasrollahi, D.B. Tanner, A.F. Hebard, B.R. Appleton, Nanotechnology 22 (42) (2011). [28] S.M. Kim, Y.W. Jo, K.K. Kim, D.L. Duong, H.J. Shin, J.H. Han, J.Y. Choi, J. Kong, Y.H. Lee, ACS Nano 4 (11) (2010) 6998. [29] X.C. Miao, S. Tongay, M.K. Petterson, K. Berke, A.G. Rinzler, B.R. Appleton, A.F. Hebard, Nano Lett. 12 (6) (2012) 2745. [30] B. Ray, M. Ashraful, IEEE J. Photovolt. 3 (1) (2013) 310. [31] W. Tress, A. Merten, M. Furno, M. Hein, K. Leo, M. Riede, Adv. Energy Mater. 3 (5) (2013) 631. [32] S.W. Tsang, J.R. Manders, M.J. Hartel, T. H Lai, S. Chen, J.R. Reynolds, C.M. Amb, F. So, Adv. Funct. Mater. 23 (299) (2013). [33] H.Q. Zhou, Y. Zhang, J. Seifter, S.D. Collins, C. Luo, G.C. Bazan, T.Q. Nguyen, A.J. Heeger, Adv. Mater. 25 (11) (2013) 1646. [34] B.B. Chen, X.F. Qiao, C.M. Liu, C. Zhao, H.C. Chen, K.H. Wei, B. Hu, Appl. Phys. Lett. 102 (19) (2013). [35] A.V. Tunc, A. De Sio, D. Riedel, F. Deschler, E. Da Como, J. Parisi, E. von Hauff, Org. Electron. 13 (2) (2012) 290. [36] B.Y. Qi, J.Z. Wang, Phys. Chem. Chem. Phys. 15 (23) (2013) 8972. [37] H.S. Shim, H.J. Kim, J.W. Kim, S.Y. Kim, W.I. Jeong, T.M. Kim, J.J. Kim, Proc. SPIE 8477 (2012). [38] M.A. Muth, W. Mitchell, S. Tierney, T.A. Lada, X. Xue, H. Richter, M. Carrasco-Orozco, M. Thelakkat, Nanotechnology 24 (48) (2013). [39] Y.H. Zhou, C. Fuentes-Hernandez, J. Shim, J. Meyer, A.J. Giordano, H. Li, P. Winget, T. Papadopoulos, H. Cheun, J. Kim, M. Fenoll, A. Dindar, W. Haske, E. Najafabadi, T.M. Khan, H. Sojoudi, S. Barlow, S. Graham, J.L. Bredas, S.R. Marder, A. Kahn, B. Kippelen, Science 336 (6079) (2012) 327. [40] V. Shrotriya, Y. Yao, G. Li, Y. Yang, Appl. Phys. Lett. 89 (6) (2006). [41] T. Wang, A.J. Pearson, A.D.F. Dunbar, P.A. Staniec, D.C. Watters, H.N. Yi, A.J. Ryan, R.A.L. Jones, A. Iraqi, D.G. Lidzey, Adv. Funct. Mater. 22 (7) (2012) 1399.
Contents lists available at ScienceDirect
Organic Electronics journal homepage: www.elsevier.com/locate/orgel
TFSA doped interlayer for efficient organic solar cells Yubin Xiao a, Shuang Zhou a, Yaorong Su a, Lei Ye a, Sai-wing Tsang b,1, Fangyan Xie c, Jianbin Xu a,⇑ a Department of Electronic Engineering and Materials Science and Technology Research Centre, The Chinese University of Hong Kong, Shatin, Hong Kong Special Administrative Region b Department of Physics and Materials Science, The City University of Hong Kong, Kowloon Tong, Hong Kong Special Administrative Region c Instrumental Analysis and Research Center, Sun Yat–sen University, Guangzhou 510275, PR China
a r t i c l e
i n f o
Article history: Received 19 August 2014 Received in revised form 15 October 2014 Accepted 16 October 2014 Available online 30 October 2014 Keywords: Molecular doping Interlayer Charge injection OPVs
a b s t r a c t Organic solar cells based on bis(trifluoromethanesulfonyl)amide (TFSA, [CF3SO2]2NH) doped poly[N-900 -hepta-decanyl-2,7-carbazole-alt-5,5-(40 ,70 -di-2-thienyl-20 ,10 ,30 -benzothiadiazole) (PCDTBT) were fabricated to investigate the effect of molecular doping. By replacing poly(3,4-ethylenedioxythiophene) (PEDOT): poly(styrenesulfonate) (PSS) with a thin TFSA layer, we have found more efficient charge injection at anode/active interface enhanced photovoltaic performance. The doping effect is confirmed by photoemission spectroscopy that the Fermi level of doped PCDTBT shifts downward to its HOMO level and results in higher carrier concentration. The reduced injection barrier also evidented by impedance spectroscopy that the real impedance of the TFSA doped PCDTBT solar cell decreases more than 50%. Using the molecular doping approach, the overall power conversion efficiency (PCE) was largely increased from 4.70% to 5.98%. Our results suggest that TFSA functions not only as a surface doping molecule, but also an anode interfacial layer to replace the conventional PEDOT:PSS. Ó 2014 Published by Elsevier B.V.
1. Introduction Polymer solar cell is an emerging technology for renewable photovoltaic with the merits of easy processing, low cost, light weight, as well as the transparency and flexibility [1–3]. To date, the state-of-art organic solar cell has reached the power conversion efficiency (PCE) threshold of 10%, which has great competitiveness against other emerging solar cell technologies [4,5]. The major challenges hindering the advance of organic solar cells are the short exciton migration distance, narrow light absorption, and low stability [2,6–8]. Hence, plenty of efforts have ⇑ Corresponding author. E-mail addresses: [email protected] (S.-w. Tsang), [email protected]. edu.hk (J. Xu). 1 Co-corresponding author. http://dx.doi.org/10.1016/j.orgel.2014.10.024 1566-1199/Ó 2014 Published by Elsevier B.V.
been devoted to improving the device performance, such as incorporating second or third solvents (1,8-diiodooctane (DIO) or chloroform (CF)) to optimize donor–acceptor morphology and thus minimizing the recombination rate [9– 12], adding quantum dots (QDs) or nanoparticles to enhance the light absorption with a broader spectrum [13–16], or employing tandem structure to increase the open circuit voltage [17,18]. Besides the above issues, the carrier injection barrier at electrode/active layer interface, one of the most important figures of merit for organic materials also plays a significant role in determining the device performance. Molecular doping, one of the facile method to manipulate carrier density has been investigated for a long time and be considered as an effective approach to control the electrical property [19–22]. Recently, researchers found that the device performance and photoconductivity could be efficiently
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
improved by doping polymers (Poly(3-hexylthiophene-2,5diyl)(P3HT), Poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-cyclopenta[2,1-b,3,4-b0 ]dithiophene)-alt-4,7(2,1,3-benzothiadiazole)](PCPDTBT)) with widely used organic p-type dopant tetrafluoro–tetracyanoquinodimethan (F4-TCNQ). The resulting polymers had downward shifted the donor Fermi level and the increased carrier density [23,24]. Moreover, a hole-collecting favourable contact on P3HT diodes can be created by means of p-type doping strategy [25]. On the other hand, bis(trifluoromethanesulfonyl)amide (TFSA) has been reported as a valid p-type chemical dopant for carbon based materials (carbon nanotube, graphene) due to its strong electron withdrawing group [26–28]. Schottky junction solar cells fabricated with TFSA doped graphene have exhibited a PCE of 8.6% [29]. In addition, in polymer light emitting diodes, the work function of graphene electrode was tuned by TFSA doping for efficient hole injection [30]. In the present work, the effects of TFSA on the performance of solution processed bulk heterojunction organic photovoltaics (BHJ-OPV) are investigated. It is found that by utilizing a thin layer of TFSA between ITO and the active layer, the hole injection barrier at anode/polymer interface is reduced and the carrier injection becomes more efficient. Moreover, the TFSA incorporated device (without PEDOT:PSS) shows an improved PCE of 5.98% when compared with the traditional ITO/PEDOT:PSS/active layer/LiF/Al structure (PCE of 4.70%). Therefore, it is suggested that besides acting as a p-type doping molecule at the anode interface, TFSA also has the potential to serve as an alternative buffer layer to replace the commonly used PEDOT:PSS, which is always regarded as the cause for device degradation [31–34]. 2. Experiments In order to demonstrate the effect of TFSA, we fabricated a control device with configuration of ITO (180 nm)/PEDOT:PSS(40 nm)/PCDTBT:PC71BM (80 nm)/LiF (1 nm)/Al (100 nm). PCDTBT was purchased from 1-material Chemscitech, Inc., and used as received. Firstly, a pre-patterned ITO (conductivity: 10–15 X/square) glass was cleaned by detergent, deionized water, acetone and isopropanol in sequence, followed by oxygen plasma treatment for 90 s. Secondly, for control sample, a thin layer of PEDOT:PSS was spun-cast onto ITO glass with a thickness of ca. 40 nm, and then annealed at 145 °C for 10 min in air. For comparison, device without PEDOT:PSS, TFSA (dissolved in nitromethane and stirred overnight) was spun coated onto ITO and formed a thin film of 20 nm. Thereafter, the substrates were transferred to a N2 filled glove box. PCDTBT and PC71BM with a mass ratio of 1:3 were dissolved in a mixed solution composed of 1,2 o-dichlorobenzene (o-DCB) and chloroform (CF) with volume ratio of 1:1, and then spun-cast atop the PEDOT:PSS (or TFSA) layer to form the active layer, giving rise to a thickness of 80 nm. Following that, a 20-min soft annealing and a further annealing at 120 °C for 10 min were conducted. Finally, a thin LiF interfacial layer (1 nm) and a 100 nm thick Al electrode were deposited sequentially by thermal evaporation. For each condition, there are at least 60 solar cells were measured and the figures were drawing from the average value.
3703
The evaporator was BOC Edwards Auto 306 and the active layer area of the device was defined by a shadow mask of 2 mm 6 mm. Electrical measurements were performed by a semiconductor characterization system (Keithley 236) at room temperature in air under the spectral output from solar simulator (Newport) using an AM 1.5G filter with a light power of 100 mW/cm2. The light intensity was precisely calibrated by a calibrated silicon solar cell. The morphologies of the PCDTBT and PC71BM blend thin films were characterized by atomic force microscopy (AFM) in tapping mode. The thickness of the active layer and evaporated layers were recorded with a thickness monitor (Sigma SQM-160), and also verified by AFM.
3. Results and discussion Fig. 1(a) shows the chemical structure of materials that used in this study. Fig. 1(b) shows the solar cell configuration and Fig. 1(c) presents the energy band diagram in which PEDOT:PSS is replaced by TFSA. Once TFSA is in contact with PCDTBT, electrons will transfer from PCDTBT to TFSA due to the strong electron-accepting nature of TFSA. As electrons move to TFSA, the Fermi level of PCDTBT will shift downward to its HOMO level as indicated by the green arrow in Fig. 1(c). Such doping effect is confirmed by photoemission spectroscopy measurement that will be discussed later. As an initial step, the thickness and concentration of TFSA were optimized, as shown in Figs. 1s and 2s. Firstly, the thicknesses of TFSA were varied from 5 to 25 nm and the concentration was fixed at 20 mM. With only 5 nm TFSA, the Voc was increased to 0.87 V, when further increased to 10 nm, the Voc reached 0.9 V and kept constant in the following thicker films. Overall, with 20 nm TFSA atop, the highest PCE was obtained. There are two possible reasons for the increased Voc with thicker TFSA. Firstly, the different penetration depth of TFSA to the bulk active layer. At thin TFSA layer, like 5 nm, it is thick enough to cover the beneath ITO electrode, on the top side, after spin cast the active layer, there will be interpenetration of active layer into TFSA. At thin TFSA layer, the active layer could penetrate through the TFSA and may contact with ITO electrode to some extent, so there will be current leakage pathway there. For thicker TFSA, active layer could no longer penetrate through it, therefore the chance of leaking current was reduced and lead to stable Voc. Secondly, more TFSA, 10 nm or thicker film compared to 5 nm, facilitate more complete doping of PCDTBT at the interface. Generally the energy difference between the HOMO of the donor and the LUMO of the acceptor (effective bandgap) set the upper limit of the Voc in organic solar cell. However, the work-function of electrode has strong influence on the Voc below that limit. As the effective bandgap in PCDTBT:PC71BM is around 1.4 eV, improving the WF in this material system still can enhance the Voc. So in the rest of the study, the film thickness was fixed at 20 nm. For the concentration variation (5–25 mM), the cell with 20 mM TFSA doping showed the optimum PCE. Therefore, the film thickness and concentration were determined to be 20 nm
3704
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Fig. 1. (a) Chemical structure of TFSA, PCDTBT, PC71BM used in the study. (b) Device architecture of the PCDTBT:PC71BM solar cells. (c) Energy diagram of the BHJ solar cell. With the p-type doping, the green arrow indicates the Fermi level of PCDTBT shifted toward its HOMO level. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and 20 mM in this study. The detailed parameters were summarized in Tables 1s and 2s, respectively. Next in order to examine the carrier transport at the anode/PCDTBT and anode/active layer interface, an series of hole-only devices were fabricated with both undoped and doped PCDTBT (or PCDTBT:PC71BM) sandwiched between ITO/PEDOT:PSS and MoO3/Al or ITO/TFSA and MoO3/Al respectively. MoO3 in here is an electron blocking layer (EBL), so only hole can transport in the devices. Room temperature J1/2–V characteristics are given in Fig. 2(a), with solid and empty curves corresponding to PCDTBT and PCDTBT:PC71BM compounds respectively. For PCDTBT only devices, the doped PCDTBT shows nearly 2.5 times
higher root square current density at the range around open circuit voltage (0.9 V), while for PCDTBT:PC71BM hole only devices, the doped device shows nearly an order higher root square current density as well. Especially, noted that the doped PCDTBT:PC71BM hole-only diode exhibits even higher current density compared with undoped single PCDTBT layer devices, despite the scattering effect of PC71BM. Yet, doping is well known to lower the effective injection barrier at the interface by changing the molecular level alignment or/and creating a narrow depletion region for free carriers tunnelling through. Hence, it is believed that the TFSA p-type doping will decrease the injection barrier at the anode/active layer
Fig. 2. (a) J1/2–V characteristics of doped and undoped PCDTBT (solid symbol) and PCDTBT:PC71BM (open symbol) hole-only devices. (b) Nyquist plots of impedance spectra for PCDTBT:PC71BM solar cells without and with TFSA treatment and measured at the Voc condition under dark.
3705
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Table 1 Photovoltaic performance of PCDTBT:PC71BM solar cells with different interfacial structure between ITO and BHJ layer (Rsh, Rs extracted from illuminated J–V curve).
PEDOT:PSS TFSA PEDOT/TFSA TFSA/PEDOT
Voc (V)
Jsc (mA/cm2)
FF (%)
PCE (%)
Rsh (X cm2)
Rs (X cm2)
J (EQE) (mA/cm2)
0.84 0.90 0.88 0.85
10.2 11.1 11.2 8.9
55.0 59.9 58.9 59.2
4.70 5.98 5.80 4.50
423 497 477 448
15 10 8 13
9.9 10.9 11.0 8.8
interface which is responsible for the increased current density in the diode. As commonly recognized, the hole density of a p-doped polymer will increase and the increasing carrier concentration will result in a high hole mobility. Therefore, the hole mobility was calculated from the space-charge-limited current (SCLC) region to verify its variation. As shown in Fig. 2(a), for the undoped single PCDTBT layer device, the hole mobility is 3.66 104 cm2/V s while for the doped device, the hole mobility increases to 9.35 104 cm2/ V s. For the hole mobility in BHJ devices (with PCDTBT:PC71BM), it decreases compared with PCDTBTonly device. The possible reason is the scattering effect of PC71BM when it is mixed with PCDTBT. However, still the hole mobility of doped PCDTBT:PC71BM device (4.67 104 cm2/V s) presents nearly 4 times higher value compared with undoped one (1.37 104 cm2/V s). Since the mobility of electron in PCDTBT:PC71BM is on the order of 104 cm2/V s [35], a more balanced charge transport is beneficial for higher charge collection efficiency [36,37]. In order to understand the complete charge transport process, we also fabricated and characterized electron-only devices for pure PC71BM and TFSA treated PC71BM. As shown in Fig. 3s, the results show that the electron mobility is slightly decreased from 4.59 103 cm2/V s for pristine to 3.23 103 cm2/V s for doped one which indicate that TFSA has very less effect for the electron transport. To further understand the doping effect of TFSA, the impedance of PCDTBT:PC71BM solar cells was measured in a frequency range from 1 MHz to 0.1 Hz by utilizing impedance spectroscopy under dark. Fig. 2(b) presents the Nyquist plots of the impedance spectra for PCDTBT:PC71BM solar cells without and with TFSA treatment and measured under an external bias voltage at the Voc condition. For PCDTBT solar cells without TFSA treatment (black line), the real impedance is 1101 X, while for the one doped with TFSA (red 2line), the real impedance decreases to 545 X. The decreasing impedance indicates that the overall device resistance (bulk and contact resistance) is decreased due to the TFSA doping effect. According to the generally accepted concept that by doping conducting polymers, the charge injection barrier at the electrode and active layer contact can be optimized. Hence, the lowered resistance will undoubtedly conduce to reduce the current lost across the metal–organic junction which is attributed to the increment of the short circuit current [38].
2 For interpretation of color in Fig. 2, the reader is referred to the web version of this article.
For the next step, we studied the current–voltage characteristic and PCE, based on the device structures of (1) ITO/PEDOT:PSS/PCDTBT:PC71BM/LiF/Al and (2) ITO/TFSA/ PCDTBT:PC71BM/LiF/Al. The device performances are summarized in Table 1. As shown in Fig. 3(a), the control sample by using PEDOT:PSS has an open circuit voltage (Voc) of 0.84 V, a short circuit current (Jsc) of 10.2 mA/cm2, a fill factor (FF) of 55.0% and a PCE of 4.70%. Whereas for the TFSA treated PCDTBT solar cell, enhanced Voc to 0.90 V, Jsc to 11.1 mA/cm2, FF of 59.9% and PCE of 5.98% were obtained. The increased Jsc was further confirmed by external quantum efficiency (EQE) (Fig. 3(d)). From the above discussion, we could infer that the improved solar cell performance is attributed to the decreased carrier injection barrier and cell resistance of PCDTBT which leads to an efficient hole transportation and results in enhanced current density and fill factor. Moreover, Voc is increased from 0.84 to 0.90 V which is owing to the reduced charge recombination rate at the interface after TFSA doping. However, it has also been demonstrated that the work function of the metal and/or metal oxide electrodes can be modified by organic molecular, e.g., Zhou et al. have found that polymers such as polyethylenimine ethoxylated (PEIE) and branched polyethylenimine (PEI) have an universal effects on different electrodes such as ITO, graphene and Al [39]. Therefore, it is interesting to investigate whether the doping effect of TFSA is on ITO or PCDTBT. To clarify this problem, two more devices with slightly different structures were fabricated. For one device, it was configured as (3) ITO/PEDOT:PSS/TFSA/active layer/LiF/Al, where TFSA could only contact with active layer. For comparison, the other device had a structure of (4) ITO/TFSA/ PEDOT:PSS/active layer/LiF/Al, in other words, TFSA was spun-cast between ITO electrode and PEDOT:PSS, hence TFSA no longer contacted with active layer and the doping effects could be eliminated. From Fig. 3(a and b), we can see that device (3) has a PCE of 5.80%, while device (4) shows a lower PCE of 4.50%. With TFSA next to PCDTBT, the solar cell displays relatively better performance with Voc of 0.88 V, Jsc of 11.2 mA/cm2 and FF of 58.9%. However, for device (4) with TFSA aparting from PCDTBT:PC71BM, it shows lower Voc of 0.85 V, Jsc of 8.9 mA/cm2 and FF of 59.2%. Hence, except FF, device (4) has a similar Voc with device (1) and an even lower Jsc. Therefore, from the above results, we can confirm that the doping of TFSA is effective on PCDTBT instead of ITO electrode. Fig. 3(c) shows the dark current versus bias voltage characteristics. After doping with TFSA, reduced dark currents under reverse bias and lowered leakage currents at zero bias were observed which led to a relatively high
3706
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Fig. 3. (a) Current density versus voltage characteristics of PCDTBT:PC71BM solar cells under 100 mW/cm2 irradiation. (b) Jsc, PCE, Voc and FF of control sample, with TFSA doping, with TFSA next to PCDTBT and TFSA next to ITO electrode. (c) Dark current density versus voltage. (d) External quantum efficiency of PCDTBT:PC71BM solar cells.
diode rectification ratio, indicating a promisingly high level of the shunt resistance and thus contributing to a relatively high Voc as discussed previously [40]. Whereas for the dark current in the space charge region (1–2 V) that dominated by the faster carriers, a similar dark current level can be observed for all cases with or without doping. Apparently, it suggests that the p-doping with TFSA does not affect the electron transport in PCDTBT:PC71BM solar cells. In addition, active layer morphology is another important issue for organic solar cells since the charge transport is largely depended on it. Therefore, we investigated the influence of TFSA doping on the morphology characteristics of the active layer. Fig. 4(a)–(d) shows 1 lm 1 lm topographic and phase atomic force microscopy (AFM) images of the photoactive layer without and with TFSA doping under air, respectively. Without TFSA, the surface morphology of PCDTBT: PC71BM shows relatively smooth and featureless structure with a root mean square (RMS) roughness of 0.33 nm. While after doping with TFSA, an apparent surface reconstruction was observed, the active layer surface turns coarse with relatively larger domain size and the RMS increases to 0.60 nm. From the corresponding phase images, it can be seen that the domain without TFSA doping shows very blurry edges, while for the one treated with TFSA, it displays sharp and clear
domain boundaries. Generally, unclear morphology indicates impure domains and would lead to enhanced bimolecular recombination and reduced charge transport, therefore leads to a lower Jsc which is in good agreement with the previous electrical characterizing results [12]. The reasons of the changed morphology after TFSA doping are not fully understood yet. One cause of the modified morphology may due to the variation of polymer stacking, therefore we did grazing incidence X-ray diffraction (GIXRD) analysis as shows in Fig. 4s. However, it shows very similar peaks indicates no sign of changed polymer orientation. Also from UV–vis spectroscopy as shows in Fig. 5s, no peak shifting or new absorption peak were observed too. Since the feature of the polymer blends could be affected by the substrates [41], one possibility is that the difference in specific surface energy between TFSA and PEDOT:PSS which leading to the variation of the donor–acceptor domains and the increasing RMS. To further verify the above speculations, more investigations are still needed to be done and the related results will be reported elsewhere. In order to confirm the doping effect of TFSA on PCDTBT, the UPS measurement was conducted and the results are shown in Fig. 5. The value of work function (WF) of the polymer is defined by the energy difference between vac-
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
3707
Fig. 4. AFM images of PCDTBT:PC71BM, (a) morphology and (c) phase without TFSA doping; (b) morphology and (d) phase with TFSA doping.
Fig. 5. The UPS spectra of undoped and doped PCDTBT film at various TFSA concentrations.
uum level and Fermi level of the polymer. As determined by UPS, work function is equal to the difference between the low cutoff energy and the bias voltage. In our case, a 5 V bias voltage was applied in order to obtain the low cutoff energy levels of the secondary photo-electrons. For pristine and TFSA doped PCDTBT at each concentrations (5–25 mM), the lower cut-off energy levels of the secondary photo-electrons were 9.66, 9.71, 9.77, 9.86, 9.87, and 9.85 eV, respectively, as shown in Fig. 5. Concomitantly, the work function of the film increases from 4.66 to 4.71, 4.77, 4.86, 4.87 and 4.85 eV for the same doping concentration values respectively. Notice that the Fermi level at
25 mM did not showing further downside shifting, this could explain the slight decreasing PCE at 25 mM compare with 20 mM. Overall, it is apparent that with TFSA doping, the Fermi level of PCDTBT is effectively downward shifted to its HOMO level. With the lower Fermi level, the hole concentration at the anode/PCDTBT interface will increase and therefore reduce the charge injection barrier in between. Finally, to explore the doping applicability of TFSA to other donor polymers, solar cells with active layer composed of P3HT:PC61BM was fabricated. As shown in Fig. 6(a), the doped P3HT hole-only device presents a higher current density compared with undoped one, indicating a more efficient hole injection from ITO electrode to P3HT. This result is consistent with the previous study of PCDTBT hole-only diode. In comparison of the undoped P3HT devices, TFSA doped P3HT solar cells show a slightly increasing Voc from 0.60 V to 0.61 V, an increasing Jsc from 8.80 to 9.59 mA/cm2 and a similar FF. Because of the increasing Voc and Jsc, the overall PCE was increased from 3.31% to 3.63% after doping and the detail data was summarized in Table 2. Fig. 6(c) exhibits the UPS spectra of undoped and doped P3HT, the lower cut-off energy levels of the secondary photo-electrons were 9.08 and 9.58 eV, respectively. Hence, the corresponding work function of P3HT also goes deeper from 4.08 to 4.58 eV. Therefore, according to the above results, we can conclude that the TFSA molecular doping can be a generally applicable way to improve the performance of polymer solar cells.
3708
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709
Fig. 6. (a) J–V characteristics of doped and undoped P3HT hole-only devices. (b) Current density versus voltage characteristics of P3HT:PC61BM solar cells under 100 mW/cm2 irradiation. (c) The UPS spectra of undoped and doped P3HT film at 20 mM TFSA.
Voc (V)
Jsc (mA/cm2)
FF (%)
PCE (%)
WF (eV)
donors such as P3HT. Overall, our findings suggest a facile method to effectively enhance the hole mobility of PCDTBT and simultaneously reduce the solar cell resistance for more efficient organic photovoltaic cells.
0.60 0.61
8.80 9.59
62.7 62.1
3.31 3.63
4.08 4.58
Acknowledgements
Table 2 Photovoltaic performance of pristine and TFSA doped P3HT:PC61BM solar cells.
Undoped Doped
4. Conclusions In summary, we have shown that by inserting a thin layer of TFSA between the organic active layer and the ITO anode, the injection barrier at ITO/active layer interface can be greatly reduced and it leads to an improvement in Jsc, Voc, FF and the overall PCE of the PCDTBT:PC71BM based BHJ solar cells. The hole mobility (lh) of PCDTBT increases by nearly 4 times and results in a largely enhanced PCE from 4.70% to 5.98% for the undoped and doped devices, respectively. Impedance spectroscopy reveals that the real impedance of PCDTBT solar cell decreases more than 50% after TFSA doping. Although the roughness of the active layer increases a little, from the phase images, it can be inferred that the charge transportation is more efficient in well defined phase separation domains. Finally, the downward shifting of the Fermi level of PCDTBT is confirmed by UPS. Furthermore, the doping effects of TFSA can also be applicable to other polymer
This work is in part supported by Research Grants Council of Hong Kong, particularly, via Grant Nos CUHK2/ CRF/08, CUHK4182/09E, CUHK4179/10E, N-CUHK405/12 and AoE/P-03/08.J Xu would like to thank the National Science Foundation of China for the support, particularly, via Grant Nos. 60990314, 60928009 and 61229401. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.orgel.2014.10.024. References [1] G. Li, R. Zhu, Y. Yang, Nat. Photon. 6 (3) (2012) 153. [2] J.L. Brédas, J.E. Norton, J. Cornil, V. Coropceanu, Acc. Chem. Res. 42 (11) (2009) 1691. [3] C. Deibel, V. Dyakonov, Rep. Prog. Phys. 73 (9) (2010). [4] J.B. You, C.C. Chen, Z.R. Hong, K. Yoshimura, K. Ohya, R. Xu, S.L. Ye, J. Gao, G. Li, Y. Yang, Adv. Mater. 25 (29) (2013) 3973.
Y. Xiao et al. / Organic Electronics 15 (2014) 3702–3709 [5] J.B. You, L.T. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C.C. Chen, J. Gao, G. Li, Y. Yang, Nat. Commun. 4 (2013). [6] A. Moliton, J.M. Nunzi, Polym. Int. 55 (6) (2006) 583. [7] H.J. Son, B. Carsten, I.H. Jung, L.P. Yu, Energy Environ. Sci. 5 (8 815) (2012). [8] H. Spanggaard, F.C. Krebs, Sol. Energy Mater. Sol. C 83 (2–3) (2004) 125. [9] B.R. Aich, J.P. Lu, S. Beaupre, M. Leclerc, Y. Tao, Org. Electron. 13 (9) (2012) 1736. [10] R.B. Aich, Y.P. Zou, M. Leclerc, Y. Tao, Org. Electron. 11 (6) (2010) 1053. [11] S. Alem, T.Y. Chu, S.C. Tse, S. Wakim, J.P. Lu, R. Movileanu, Y. Tao, F. Belanger, D. Desilets, S. Beaupre, M. Leclerc, S. Rodman, D. Waller, R. Gaudiana, Org. Electron. 12 (11) (2011) 1788. [12] L. Ye, S.Q. Zhang, W. Ma, B.H. Fan, X. Guo, Y. Huang, H. Ade, J.H. Hou, Adv. Mater. 24 (47) (2012) 6335. [13] E.A. Parlak, T.A. Tumay, N. Tore, S. Sarioglan, P. Kavak, F. Turksoy, Sol. Energy Mater. Sol. C 110 (2013) 58. [14] P.A. Staniec, A.J. Parnell, A.D.F. Dunbar, H.N. Yi, A.J. Pearson, T. Wang, P.E. Hopkinson, C. Kinane, R.M. Dalgliesh, A.M. Donald, A.J. Ryan, A. Iraqi, R.A.L. Jones, D.G. Lidzey, Adv. Energy Mater. 1 (4) (2011) 499. [15] D.M. DeLongchamp, R.J. Kline, D.A. Fischer, L.J. Richter, M.F. Toney, Adv. Mater. 23 (3) (2011) 319. [16] M. He, F. Qiu, Z.Q. Lin, J. Phys. Chem. Lett. 4 (11) (2013) 1788. [17] L.T. Dou, J.B. You, J. Yang, C.C. Chen, Y.J. He, S. Murase, T. Moriarty, K. Emery, G. Li, Y. Yang, Nat. Photon. 6 (3) (2012) 180. [18] M. Wright, A. Uddin, Sol. Energy Mater. Sol. C 107 (2012) 87. [19] L. Ley, Y. Smets, C.I. Pakes, J. Ristein, Adv. Funct. Mater. 23 (7) (2013) 794. [20] M.T. Greiner, M.G. Helander, W.M. Tang, Z.B. Wang, J. Qiu, Z.H. Lu, Nat. Mater. 11 (1) (2012) 76. [21] O.V. Kozlov, S.A. Zapunidi, Synth. Met. 169 (2013) 48. [22] Y.L. Gao, Mater. Sci. Eng. R 68 (3) (2010) 39. [23] X.Y. Han, Z.W. Wu, B.Q. Sun, Org. Electron. 14 (4 111) (2013). [24] X.F. Lei, F.T. Zhang, T. Song, B.Q. Sun, Appl. Phys. Lett. 99 (23) (2011) 233305.
3709
[25] A. Dai, Y.H. Zhou, A.L. Shu, S.K. Mohapatra, H. Wang, C. FuentesHernandez, Y.D. Zhang, S. Barlow, Y.L. Loo, S.R. Marder, B. Kippelen, A. Kahn, Adv. Funct. Mater. 24 (15) (2014) 2197. [26] M. Herstedt, M. Smirnov, P. Johansson, M. Chami, J. Grondin, L. Servant, J.C. Lassègues, J. Raman Spectrosc. 36 (8) (2005) 762. [27] S. Tongay, K. Berke, M. Lemaitre, Z. Nasrollahi, D.B. Tanner, A.F. Hebard, B.R. Appleton, Nanotechnology 22 (42) (2011). [28] S.M. Kim, Y.W. Jo, K.K. Kim, D.L. Duong, H.J. Shin, J.H. Han, J.Y. Choi, J. Kong, Y.H. Lee, ACS Nano 4 (11) (2010) 6998. [29] X.C. Miao, S. Tongay, M.K. Petterson, K. Berke, A.G. Rinzler, B.R. Appleton, A.F. Hebard, Nano Lett. 12 (6) (2012) 2745. [30] B. Ray, M. Ashraful, IEEE J. Photovolt. 3 (1) (2013) 310. [31] W. Tress, A. Merten, M. Furno, M. Hein, K. Leo, M. Riede, Adv. Energy Mater. 3 (5) (2013) 631. [32] S.W. Tsang, J.R. Manders, M.J. Hartel, T. H Lai, S. Chen, J.R. Reynolds, C.M. Amb, F. So, Adv. Funct. Mater. 23 (299) (2013). [33] H.Q. Zhou, Y. Zhang, J. Seifter, S.D. Collins, C. Luo, G.C. Bazan, T.Q. Nguyen, A.J. Heeger, Adv. Mater. 25 (11) (2013) 1646. [34] B.B. Chen, X.F. Qiao, C.M. Liu, C. Zhao, H.C. Chen, K.H. Wei, B. Hu, Appl. Phys. Lett. 102 (19) (2013). [35] A.V. Tunc, A. De Sio, D. Riedel, F. Deschler, E. Da Como, J. Parisi, E. von Hauff, Org. Electron. 13 (2) (2012) 290. [36] B.Y. Qi, J.Z. Wang, Phys. Chem. Chem. Phys. 15 (23) (2013) 8972. [37] H.S. Shim, H.J. Kim, J.W. Kim, S.Y. Kim, W.I. Jeong, T.M. Kim, J.J. Kim, Proc. SPIE 8477 (2012). [38] M.A. Muth, W. Mitchell, S. Tierney, T.A. Lada, X. Xue, H. Richter, M. Carrasco-Orozco, M. Thelakkat, Nanotechnology 24 (48) (2013). [39] Y.H. Zhou, C. Fuentes-Hernandez, J. Shim, J. Meyer, A.J. Giordano, H. Li, P. Winget, T. Papadopoulos, H. Cheun, J. Kim, M. Fenoll, A. Dindar, W. Haske, E. Najafabadi, T.M. Khan, H. Sojoudi, S. Barlow, S. Graham, J.L. Bredas, S.R. Marder, A. Kahn, B. Kippelen, Science 336 (6079) (2012) 327. [40] V. Shrotriya, Y. Yao, G. Li, Y. Yang, Appl. Phys. Lett. 89 (6) (2006). [41] T. Wang, A.J. Pearson, A.D.F. Dunbar, P.A. Staniec, D.C. Watters, H.N. Yi, A.J. Ryan, R.A.L. Jones, A. Iraqi, D.G. Lidzey, Adv. Funct. Mater. 22 (7) (2012) 1399.