Upconversion effects on the performance of near-infrared laser-driven polymer photovoltaic devices

Upconversion effects on the performance of near-infrared

laser-driven polymer photovoltaic devices

Jyh-Lih Wu

, Fang-Chung Chen

, Shu-Hao Chang

, Kim-Shih Tan

, Hsing-Yu Tuan

Department of Photonic and Display Institute, National Chiao Tung University, Hsinchu 30010, Taiwan

b

Department of Chemical Engineering, National Tsing Hua University, Hsinchu 30013, Taiwan

a r t i c l e

i n f o

Article history: Received 8 May 2012

Received in revised form 28 May 2012 Accepted 29 May 2012

Available online 28 June 2012 Keywords: Polymer Upconversion Photovoltaic Laser

a b s t r a c t

We have explored the upconversion (UC) effects resulting from the presence of NaYF4:Yb/ Er nanocrystals in near-infrared laser-driven organic photovoltaic devices (OPVs). Illumi-nation of these UC nanomaterials with monochromatic light (980 nm) led to emission of visible light, which was subsequently absorbed by the polymer photoactive layer. We infer that both the long-wavelength charge-transfer absorption of the polymer blend and the UC effect contributed to the obvious near-infrared photovoltaic response. The resulting UC effects led to noticeable enhancements in the photocurrent and the efficiencies of the OPV devices under illumination at 980 nm.

Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction

Organic photovoltaic devices (OPVs) feature several attractive properties, including light weight, semi-trans-parency, mechanical flexibility, and fabrication at low cost

and low temperature[1–5]. Because the energy payback

time (i.e., the time required for any energy-producing sys-tem or device to produce as much energy as was required in its manufacture) of flexible OPVs is potentially as short as several weeks [5], they are promising candidates for use in next-generation solar energy sources. Recently, we demonstrated a potential application of OPVs as wireless electrical sources that power biologically functionalized nanodevices when driven by near-infrared (NIR) lasers

[6]. Because of the high transparency of biological tissues toward NIR photons, OPVs featuring a high photovoltaic re-sponse in the NIR spectral region could provide electrical power for biomedical sensing or diagnosis[6,7]. The NIR photovoltaic responses of such near-infrared laser-driven (NIRLD) polymer photovoltaic devices prepared from

blends of poly(3-hexylthiophene) (P3HT) and

[6,6]-phe-nyl-C61-butyric acid methyl ester (PCBM) was resulted

from their long-wavelength absorption of charge-transfer (CT) states[6]. Further efforts will be needed to improve the device efficiency toward NIR photons to meet the power requirements of future biological applications. In addition, the quantum efficiencies of most current OPVs in the NIR region remain very low. Because of the excitonic nature of the photogenerated species in organic semicon-ductors, the OPV active layer must feature both electron donor and acceptor components to overcome the strong exciton binding energy (e.g., typically 0.3–0.5 eV) at elec-tron donor–acceptor (D–A) interfaces [3,8]. The need for energetic offset at the D–A interfaces limits the applicabil-ity of low-bandgap organic materials[8]. Therefore, taking advantage of the long-wavelength region of the solar spec-trum is one potential means of further increasing the effi-ciencies of OPVs[8,9].

One feasible approach toward harvesting NIR photons is to exploit upconversion (UC) effects—that is, converting NIR radiation into visible photons[10–13]. Among

estab-lished UC media, hexagonal NaYF4 codoped with Yb3+

and Er3+ _{ions is one of the most promising materials}

[11,14]. In such a system, efficient UC can be achieved by

1566-1199/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.orgel.2012.05.057

⇑ Corresponding author. Tel.: +886 3 5131484; fax: +886 3 5735601. E-mail address:[email protected](F.-C. Chen).

Contents lists available atSciVerse ScienceDirect

Organic Electronics

harvesting NIR photons with Yb3+ _{and emitting visible}

luminescence with Er3+_{—i.e., through energy transfer from}

Yb3+ _{to Er}3+ _{ions—followed by cross relaxation between}

two nearby Er3+ions. The optical properties of these NaY-F4:Yb/Er nanocrystals (NCs) can be manipulated through

tuning their sizes, shapes, and phases [10,14]. UC effects have been used previously to enhance the device efficien-cies of silicon-based [15] and dye-sensitized [16] solar cells. Although preliminary results of the successful use of UC materials in organic solar cells have also been briefly reported[17], complete device characterization has not. In this study, we aimed to enhance the performance of NIRLD polymer photovoltaic devices, based on a blend of P3HT and PCBM, through the introduction of NaYF4:Yb/Er NCs.

Illumination of these UC phosphors with monochromatic light (980 nm) led to emission of visible light, which was subsequently absorbed by the polymer photoactive layer. In addition, to investigate the UC effects without influenc-ing the electrical properties of the devices, we sprayed the

NaYF4:Yb/Er phosphors onto the back side of the

com-pleted device, rather than incorporating the material into

the devices. This simple method boosted the conversion efficiency of the NIRLD polymer photovoltaic devices. 2. Experimental

Monodisperse hexagonal-phase NaYF4:Yb/Er

nanocrys-tals were prepared following the method as described by Chen et al.[18]. The NCs were synthesized through decom-position of NaF and lanthanide oleate complex in the pres-ence of oleic acid and octadecene (as bifunctional linkers to control the shape and size of the NCs) at 320 °C. The result-ing particle size, determined usresult-ing transmission electron microscopy (TEM), was 40–50 nm (Fig. 1(a)). The steady-state photoluminescent (PL) spectrum was obtained using a continuous-wave laser at 980 nm as the excitation source. The devices were fabricated on ITO-coated glass substrates[6]. After a routine cleaning process, the glass substrates were dried in an oven for at least 12 h and then treated with UV ozone prior to use. The anodic buffer, poly(3,4-ethylenedioxythiophene):polystyrenesulfonate (PEDOT:PSS), was spin-coated on top of the indium tin

Fig. 1. (a) The left figure shows the schematic representation of the structure of the OPV incorporating NaYF4:Yb/Er phosphors. The right figure displays the

TEM image of the NaYF4:Yb/Er NCs used in this study. (b) Absorption spectrum of a P3HT:PCBM (1:1, w/w) film and PL spectrum of the NaYF4:Yb/Er NCs,

oxide (ITO)-coated substrates and then the samples were baked at 120 °C for 1 h. The photoactive layer, a solution of P3HT (Rieke Metals) and PCBM (Solenne) in 1,2-dichlo-robenzene, was deposited on top of PEDOT:PSS. The thick-ness of the active layer was approximately 600 nm— thicker than those of most conventional devices to allow efficient NIR absorption. After undergoing

solvent-anneal-ing [19], the film was thermally annealed at 110 °C for

15 min in a N2-filled glove box. To complete the device, a

bilayer cathode comprising Ca (30 nm) and Al (100 nm) was deposited sequentially through thermal evaporation. The completed devices were encapsulated with a cover glass. The I–V characteristics of the OPVs were measured using a Keithley 2400 source measure unit. The photocur-rent response was measured under illumination either from a 150-W Thermal Oriel solar simulator (AM 1.5G) or a 980-nm laser. The irradiation area of the 980-nm laser (ca. 3 mm2_{) was smaller than the device area (10 mm}2_).

The absorption spectrum was recorded using a UV–vis– NIR spectrometer (PerkinElmer Lambda 950).

3. Results and discussion

To exploit UC effects, we sprayed a thick layer of NaY-F4:Yb/Er NCs from a hexane solution onto the back side

of the glass substrate (Fig. 1(a)). We observed UC visible

luminescence when we excited the NCs at 980 nm (Fig. 1(b)). The emission peaks (green and red) corre-sponded to different energy transfer processes [10,11].

Fig. 1(b) also displays the absorption spectrum of the P3HT:PCBM blend; the UC emission, especially the green one near 550 nm, overlapped well with the absorption spectrum of the active layer, suggesting efficient absorp-tion in this UC OPV system.

Fig. 2(a) displays a cross-sectional SEM image of the UC film obtained after spray-coating; the average film

thick-ness was approximately 5

l

film (Fig. 2(b)) suggests that the film surface was quite rough. The cotton-like morphology of the UC material probably resulted from the high volatility of the solvent (hexane) used for spray-coating. Together with the high pressure of the carrier gas (N2), the UC NCs solidified

quickly, resulting in a rough surface. Nevertheless, although the surface of the UC film was rough, we could still observe an apparent UC process. Considering that this approach is compatible with the fabrication of large-area devices, our results suggest that a simple spray-coating process might be a promising means of preparing practi-cally useful UC thin films.

Fig. 3(a) illustrates the effect of the UC on the devices’ current–voltage (I–V) characteristics, recorded under illu-mination from a 980-nm laser at a power of 146 mW.

The reference device possessing the structure ITO/PED-OT:PSS/P3HT:PCBM/Ca/Al exhibited an open-circuit volt-age (Voc) of 0.41 V, a short-circuit current (Isc) of 31.2

l

and a fill factor (FF) of 0.55, yielding a power conversion efficiency (PCE) of 0.0048%. These values are comparable with those of previously reported devices[6]. After apply-ing the UC NCs, the values of Vocand FF increased slightly,

to 0.43 V and 0.58, respectively; more importantly, the

va-lue of Isc increased to 36.2

l

0.0062%. Meanwhile, the maximum output power, defined as the product of the maximum photocurrent multiplied by the maximum voltage, was enhanced by approximately

30% (from 7.00 to 9.05

l

device under illumination with simulated solar irradiation (AM 1.5G), however, the UC effects did not improve the de-vice efficiency.Fig. 3(b) reveals that the photocurrent of the reference device decreased significantly when using the UC NCs. We suspect that a huge portion of the incom-ing photons was scattered or blocked by the NaYF4:Yb/Er

thin film, leading to decreased device performance. Fur-thermore, because efficient UC occurs only at high levels of light intensity, we expected very minor UC effects to ex-ist in our device system under normal solar illumination conditions.

The NIR photovoltaic response of the device resulted from two major effects. First, the 980-nm photons could directly excite the CT states present in the polymer/fuller-ene blends, leading to a pronounced photocurrent

[6,20,21]. Second, as revealed inFig. 1(b), the UC visible

Fig. 3. I–V characteristics of OPV devices, recorded under illumination with (a) a 980-nm laser at a power of 146 mW and (b) simulated solar light (AM 1.5G) at 100 mW cm 2_.

luminescence could be absorbed by the P3HT:PCBM thin film, resulting in another possible route for charge carrier generation. Therefore, we infer that the additional photo-current contributed by the UC process improved the device performance.

Next, we studied the effect of the irradiation power on the photovoltaic characteristics of the NIRLD devices to ob-tain further insight into the mechanism responsible for the enhanced device performance. Fig. 4(a) reveals that the photocurrent of the reference device deviated from a linear dependence at higher laser powers. In a previous report, we attributed such deviation to possible photodegradation of the photoactive layer under high-intensity illumination

[6]. The photocurrent of the UC device, however, scaled lin-early with the illumination intensity—even at high inten-sity. This behavior might be due to the NaYF4:Yb/Er NCs

on the front of the device absorbing a large number of the NIR photons, thereby minimizing any possible photo-degradation under high-intensity illumination. Further-more, the incorporation of the UC materials also modified the effect of the irradiation power on the PCEs (Fig. 4(b)). The decreased FFs and the deviation of the photocurrent from linear dependence resulted in the PCEs decreasing upon increasing the laser power. In addition, at lower lev-els of laser power (ca. <30 mW), the presence of these UC NCs in the NIRLD devices did not improve the PCE (Fig. 4(b)). This observation is not surprising because mul-tiple photons are involved in the UC process[22]. Because a high laser power was required to obtain efficient UC lumi-nescence, the PCEs for the UC device increased upon increasing the power of the 980-nm laser. The distinct trends in the dependence of the irradiation power further account for the presence of the UC effects.

4. Conclusion

The presence of UC NaYF4:Yb/Er NCs improves the

de-vice performance of NIRLD OPVs, with both the long-wave-length CT absorption and the UC contributing to the obvious NIR photon response. We observed entirely differ-ent effects of the laser power on the photovoltaic charac-teristics of the devices prepared with and without the UC NCs. The UC effects led to noticeable enhancements in the photocurrents and efficiencies of the OPV devices

un-der illumination with monochromatic light at 980 nm, sug-gesting that UC nanomaterials have great potential for application in OPVs.

Acknowledgements

We thank the National Science Council of Taiwan (NSC 100-2221-E-009-082, NSC 100-2628-E-007-029-MY2 and NSC 101-3113-E-009-005) and the Ministry of Education of Taiwan (through the ATU program) for financial support. References

[1] C.J. Brabec, N.S. Sariciftci, J.C. Hummelen, Adv. Funct. Mater. 11 (2001) 15.

[2] G. Li, V. Shrotriya, J.S. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Nat. Mater. 4 (2005) 864.

[3] J.D. Servaites, M.A. Ratner, T.J. Marks, Energy Environ. Sci. 4 (2011) 4410.

[4] S.H. Park, A. Roy, S. Beaupre, S. Cho, N. Coates, J.S. Moon, D. Moses, M. Leclerc, K. Lee, A.J. Heeger, Nat. Photonics 3 (2009) 297.

[5] A.L. Roes, E.A. Alsema, K. Blok, M.K. Patel, Prog. Photovoltaics Res. Appl. 17 (2009) 372.

[6] J.L. Wu, F.C. Chen, M.K. Chuang, K.S. Tan, Energy Environ. Sci. 4 (2011) 3374.

[7] Z.G. Chen, L.S. Zhang, Y.G. Sun, J.Q. Hu, D.Y. Wang, Adv. Funct. Mater. 19 (2009) 3815.

[8] G. Dennler, M.C. Scharber, C.J. Brabec, Adv. Mater. 13 (2009) 1338. [9] M.C. Scharber, D. Mühlbacher, M. Koppe, P. Denk, C. Waldauf, A.J.

Heeger, C.J. Brabec, Adv. Mater. 18 (2006) 789.

[10] H.X. Mai, Y.W. Zhang, L.D. Sun, C.H. Yan, J. Phys. Chem. C 111 (2007) 13721.

[11] S. Schietinger, L. de S. Menezes, B. Lauritzen, O. Benson, Nano Lett. 9 (2009) 2477.

[12] S. Heer, K. Kömpe, H.U. Güdel, M. Haase, Adv. Mater. 16 (2004) 2102. [13] H.Q. Wang, T. Nann, ACS Nano 3 (2009) 3804.

[14] L. Wang, Y. Li, Chem. Mater. 19 (2007) 727.

[15] A. Shalav, B.S. Richards, T. Trupke, K.W. Krämer, H.U. Güdel, Appl. Phys. Lett. 86 (2005) 013505.

[16] G.B. Shan, G.P. Demopoulos, Adv. Mater. 22 (2010) 4373. [17] H.Q. Wang, M. Batentschuk, A. Osvet, L. Pinna, C.J. Brabec, Adv.

Mater. 2 (2011) 2675.

[18] C. Liu, H. Wang, X. Li, D. Chen, J. Mater. Chem. 19 (2009) 3546. [19] G. Li, Y. Yao, H. Yang, V. Shrotriya, G. Yang, Y. Yang, Adv. Funct.

Mater. 17 (2007) 1636.

[20] K. Vandewal, K. Tvingstedt, A. Gadisa, O. Inganäs, J.V. Manca, Nat. Mater. 8 (2009) 904.

[21] K. Vandewal, A. Gadisa, W.D. Oosterbaan, S. Bertho, F. Banishoeib, I.V. Severen, L. Lutsen, T.J. Cleij, D. Vanderzande, J.V. Manca, Adv. Funct. Mater. 18 (2008) 2064.

[22] B. Ahrens, P. Löper, J.C. Goldschmidt, S. Glunz, B. Henke, P.T. Miclea, S. Schweizer, Phys. Status Solidi A 205 (2008) 2822.