Enhancement of open circuit voltage in organic solar cells by doping a fluorescent red dye

Abstract

The open circuit voltage (V_OC) of small-molecule organic solar cells (OSCs) could be improved by doping suitable fluorescent dyes into the donor layers. In this paper, 4-(dicyanomethylene)-2-t-butyl-6-(1,1,7,7-tetramethyljulolidyl-9-enyl)-4H-pyran (DCJTB) was used as a dopant, and the performance of the OSCs with different DCJTB concentration in copper phthalocyanine (CuPc) was studied. The results showed that the V_OC of the OSC with 50% of DCJTB in CuPc increased by 15%, compared with that of the standard CuPc/fullerene (C₆₀) device. The enhancement of the V_OC was attributed to the lower highest occupied molecular orbital (HOMO) level in the DCJTB than that in the CuPc. Also, the light absorption intensity is enhanced between 400 and 550 nm, where CuPc and C₆₀ have low absorbance, leading to a broad absorption spectrum.

Keywords

organic solar cells (OSCs) / open circuit voltage / fluorescent dye doping / 4-(dicyanomethylene)-2-t-butyl-6-(1,1,7,7-tetramethyljulolidyl-9-enyl)-4H-pyran (DCJTB)

Introduction

In recent years, researches in organic solar cells (OSCs) have spurred much interest owing to their highly tunable optical and physics properties, mechanical flexibility and low cost [1-3]. The initial research into organic donor/acceptor heterojunction solar cell was reported by Tang [4], which triggered further research into small-molecular OSCs with various materials and device structures [5-7]. To date, the introduction of planar-mixed heterojunction (PM-HJ) provides a useful means to improve OSCs performance. Currently, state-of-the-art planar-mixed heterojunction OSCs achieved a power conversion efficiency of up to 5.7% for small molecular devices [8], and 9.2% for polymer bulk devices [9].

Although great improvement in device performance has been obtained, the power conversion efficiencies of OSCs are still low for widespread commercial viability [10]. The power conversion efficiency (η_p) of OSCs can be described as

(1){\eta }_{\mathrm{p}}={\displaystyle \frac{V_{\max }{J}_{\max }}{P_{\mathrm{i}\mathrm{n}}}}={\displaystyle \frac{V_{\mathrm{O}\mathrm{C}}{J}_{\mathrm{S}\mathrm{C}}\mathrm{F}\mathrm{F}}{P_{\mathrm{i}\mathrm{n}}}},

where P_in is the incident optical power density, V_max and J_max are the current density and voltage, respectively, of which the maximum power is generated in the device, and FF is the fill factor. From Eq. (1), it can be observed that one of the key parameters that influence the performance of organic solar cells is the V_OC, the increase of which is a useful means to improve the power conversion efficiency.

In the past few years, lots of effort has been made for the enhancement of V_OC. First, by stacking series-connected subcells with complementary absorption spectra, V_OC can be doubly enhanced [11]. However, the realization of tandem structure needs the scrupulous optimization of the thickness of every layer and interlayer between the two subcells. Second, V_OC can be improved by inserting fluorescent materials with the low highest occupied molecular orbital (HOMO) level between the donor and the acceptor layer to form multicharge separation (MCS) interfaces [12]. Finally, the optimization of the donor/acceptor energy levels by doping proper materials is considered as a simple and effective way for increasing V_OC. For small-molecular OSCs, it has been reported that by introducing 5,6,11,12-tetraphenylnaphthacene (rubrene) as a dopant, the V_OC can increase from 0.42 to 0.55 V [13], which is attributed to the low HOMO level in the rubrene.

Fluorescent dyes have been widely used as dopant in organic light-emitting devices, but this approach is seldom used in OSCs. In this paper, a fluorescent red dye of 4-(dicyanomethylene)-2-t-butyl-6-(1,1,7,7-tetramethyljulolidyl-9-enyl)-4H-pyran (DCJTB) with a low HOMO level as a dopant in the OSCs is systematically investigated, which proves favorable for the enhancement of V_OC. In addition, the DCJTB has a complementary light absorption for CuPc and C₆₀. The current density-voltage characteristics, photocurrent, and absorption spectra of OSCs are compared and discussed to reveal the underlying mechanism responsible for the variation of the OSC performance.

Experiments

The molecular structure of the DCJTB fluorescent dye, the structure of the OSC device, and the corresponding energy level diagram are illustrated in Figs. 1-3. The structure of the device studied in this paper is indium-tin oxide (ITO)/DCJTB: CuPc (20 nm)/C₆₀ (40 nm)/Bphen (2.5 nm)/Ag (100 nm). Bathophenanthroline (Bphen) was used as a cathode buffer layer [14]. The CuPc from Aldrich was twice sublimed, and the C₆₀ (99.5%, Aldrich) was used as received without further purification. The devices were fabricated onto the ITO coated glass substrates by an organic light emitting device (OLED)-V organic multifunctional vacuum deposition equipment. The ITO coated glass substrates with a sheet resistance of 10 Ω/sq were cleaned consecutively in ultrasonic baths containing detergent, acetone, ethanol, deionized water for 10 min each, and finally dried by high purity nitrogen blow. The substrates were treated by O₂ plasma for 5 min prior to the deposition of organic layers. And the organic layers were deposited onto the ITO substrates successively at a pressure of 0.3 mPa with a deposition rate of 0.1-0.2 nm/s. The mixed layer was grown by co-evaporation from two separately controlled sources. Then Ag top electrode was deposited through a mask with 3×4 mm² at a pressure of 3×10^-3 Pa with a deposition rate of ~1 nm/s. The deposition rate and film thickness were in situ monitored using a quartz crystal oscillator mounted to the substrate holder.

The J-V curves in the dark and under light circumstance were measured with Keithley 4200 semiconductor measuring system. A light source integrated with a Xe lamp (CHF-XM35, Beijing Trusttech Co., Ltd.) with an illumination power of 100 mW/cm² was used as solar simulator. The absorption spectra were measured with a SHIMATZU UV1700 system. All the measurements were performed in air at room temperature.

Results and discussion

Figure 4 demonstrates the J-V characteristics of the ITO/DCJTB:CuPc (20 nm)/C₆₀ (40 nm)/Bphen (2.5 nm)/Ag devices with different concentration of DCJTB in the CuPc under illumination at an intensity of 100 mW/cm² (AM 1.5G). The concentration of the DCJTB in the CuPc is 5%, 20%, and 50%. The reference device with pristine CuPc exhibited a V_OC of 0.40 V, a short circuit current (J_SC) of 8.78 mA/cm², and a fill factor (FF) of 0.47. The calculated power conversion efficiency (η_p) is 1.62 %. It can be seen that after the addition of DCJTB to the donor material, the values of V_OC gradually increases. This is caused by the lower HOMO level of 5.4 eV in the DCJTB (as shown in Fig. 3). Chan et al. [13] and Taima et al. [15] have reported that by doping rubrene with a lower HOMO level into the donor layer, a high V_OC and power conversion efficiency can be obtained.

Table 1 summarizes the main parameters extracted from the fitting results of the J-V characteristics of OSCs. It can be noticed that as the concentration of DCJTB increases, the V_OC is enhanced. In particular, when the 50% DCJTB is dopted in the CuPc, the V_OC increases to 0.46 V by a factor of 1.15 due to the lower HOMO level of the DCJTB than CuPc. Cheyns et al. [16] have derived an analytical model for V_OC based on exciton dissociation at the interface between two materials. According to their results, the V_OC can be described as

(2)V_{\mathrm{O}\mathrm{C}}=\left|\mathrm{H}\mathrm{O}\mathrm{M}{\mathrm{O}}_{\mathrm{D}}\right|-\left|\mathrm{L}\mathrm{U}\mathrm{M}{\mathrm{O}}_{\mathrm{A}}\right|+{\displaystyle \frac{kT}{q}}\left({\displaystyle \frac{p_i}{N_{\mathrm{D}}}}\right)+{\displaystyle \frac{kT}{q}}\left({\displaystyle \frac{n_{\mathrm{i}}}{N_{\mathrm{A}}}}\right)+{\mathrm{\Delta }}_{\mathrm{l}\mathrm{o}\mathrm{w}}(F_{\mathrm{c},\mathrm{D}})+{\mathrm{\Delta }}_{\mathrm{l}\mathrm{o}\mathrm{w}}(F_{\mathrm{c},\mathrm{A}}),

where k and T are Boltzmann constant and Kelvin temperature, respectively, q is electronic charge, p_i and n_i are the hole and electron concentration at the donor/acceptor interface, N_D and N_A are the effective density of the states of the donor and acceptor, and Δ_low(F_c,D) and Δ_low(F_c,A) represent the effective reduction of energy barrier at the electrode contact, LUMO denotes lowest unoccupied molecular orbital. It can be seen that the terms which determine the V_OC are mainly HOMO_D, LUMO_A and the carrier concentration achievable at the interface. The concentration of photogenerated charge carriers at the interface can never exceed the density of states. As a consequence, the V_OC will be limited by the effective energy offset between the HOMO of the donor and the LUMO of the acceptor, and by choosing a donor material with a lower HOMO level such as DCJTB, the V_OC can be significantly improved. On the other hand, the J_SC and FF do not increase after DCJTB was adopted in the devices. Therefore, the performances of the OSCs with a mixed donor layer are greatly affected by DCJTB/C₆₀ interfaces.

To analyze the influence of the DCJTB/C₆₀ heterojunction on the performance of doped CuPc devices, the OSC with a structure of ITO/DCJTB (5 nm)/C₆₀ (40 nm)/Bphen (2.5 nm)/Ag (100 nm) was fabricated. Figure 5 displays the current density-voltage (J-V) characteristics of the devices with different donor layers in the dark and under illumination. As the V_OC is determined by the HOMO_D-LUMO_A, and the HOMO of DCJTB is lower than CuPc, the device with heterojunction formed by the DCJTB and C₆₀ has a higher V_OC than that of the reference structure. A large V_OC of 0.67 V is achieved for the OSC using DCJTB as a donor material. On the other hand, the current density of the device based on DCJTB in the dark and under illumination are both smaller than that of the device based on CuPc, revealing the poor hole transport properties of DCJTB. It has been known that charge carrier mobility in dye materials is on the order of 10^-5-10^-7 cm²/(V·s), which is comparatively lower than those in the CuPc. In contrast, the hole mobility in CuPc is on the order of 10^-4 cm²/(V·s) [17]. The lower charge carrier transport properties of the DCJTB leads to a higher series resistance of 58.17 Ω·cm² in the DCJTB/C₆₀ device, while the series resistance of the CuPc/C₆₀ device is only 4.09 Ω·cm². As a result, the V_OC is improved by doping DCJTB in the donor layer, while the J_SC and FF are not.

To further investigate the mechanism responsible for the change of J_SC and FF by doping fluorescent red dye DCJTB, the maximum exciton generation rates (G_max) and exciton dissociation probabilities for devices with pristine CuPc and CuPc doped with 50% of DCJTB was compared. The photocurrent density (J_ph) against the effective applied bias voltage (V_eff=V₀-V) is depicted in Fig. 6. The photocurrent density is given by J_ph=J_L-J_D, where J_L and J_D are the current density under illumination and in the dark, respectively. The compensation voltage (V₀) is determined by the voltage at which J_ph= 0. It can be seen that the value of J_ph was proportional to bias at a low value of V_eff, and then reached a saturated level at a sufficiently high value of V_eff. The saturation photocurrent density (J_sat), which is independent of bias and temperature, correlates with the value of G_max, given by J_sat=qG_maxL, where q is the electronic charge and L is the thickness of the active layer (L=60 nm). The values of J_sat for the devices prepared with pristine CuPc and the CuPc doped with 50 % DCJTB are 94.5 and 82.0 A/m², respectively, corresponding to the calculated G_max values of 9.84×10²⁷ m^-3·s^-1 and 8.54×10²⁷ m^-3·s^-1, respectively. This reveals that the number of excitons generated in the DCJTB is less than that in the CuPc due to the weak light absorption of the DCJTB in the long wavelength region, which also causes the reduction of J_SC.

The exciton dissociation probability can be obtained from the normalized photocurrent density (J_ph/J_sat). After incorporating 50% DCJTB in the CuPc, the exciton dissociation probability under short circuit condition (V = 0) decreases from 92.7% to 88.2%, indicating that the recombination rate increases. This is attributed to the introduction of the DCJTB, which act as the recombination cites. In general, the increase of recombination rate reduces the FF of OSCs [18]. Consequently, the decreased FF of doped devices is primarily attributed to the decrease of exciton dissociation probability, which may result from the low charge carrier mobility of the DCJTB.

The strength and width of the absorption spectrum of an active layer determines to a large extent its potential for harvesting incident light, and the improvement of J_SC accordingly. The absorption spectra of devices without and with 50% DCJTB in CuPc are presented in Fig. 7. For comparison, the absorption spectra of the CuPc, C₆₀ and DCJTB films are also given in the Inset. It can be seen that the DCJTB has two absorption peaks at 359 and 538 nm, which can broaden the absorption spectra, where CuPc and C₆₀ do not absorb. In addition, it is evident that the light absorption intensity is enhanced between 400 and 550 nm, indicating that besides the contribution from both CuPc and C₆₀, the additional DCJTB doped in the CuPc can introduce its parasitic absorption. However, the J_SC of the devices with CuPc doped with DCJTB does not increase due to the weak absorption of the DCJTB in the long wavelength region. Therefore, by introducing proper material with low HOMO level and strong absorption in the near infrared region (NIR) to ensure effective harvesting of the solar photos the device performance can be further improved.

Conclusions

In summary, a simple means to improve the performance of small-molecular organic solar cells is demonstrated. By incorporating 50% DCJTB in CuPc donor layer, V_OC increased to 0.46 V by a factor of 1.15, compared with the device with pristine CuPc. This can be attributed to the lower HOMO level of DCJTB. In addition, the absorption spectra revealed that the total absorption of cell was enhanced. However, the low charge carrier mobility of fluorescent dye material which increases the recombination in DCJTB layer between electron and hole leads a decrease of J_SC and FF. Also, the small maximum excition generation rate of DCJTB owing to the weak absorption in the long wavelength is detrimental to the improvement of J_SC. Therefore, by doping novel fluorescent dyes with low HOMO energy level, strong absorption in the near infrared region (NIR) and high charge carrier mobility, higher performance of OSC devices can be obtained.

References

[1] Kim J Y, Lee K, Coates N E, Moses D, Nguyen T Q, Dante M, Heeger A J. Efficient tandem polymer solar cells fabricated by all-solution processing. Science, 2007, 317(5835): 222-225

[2] Yu J S, Huang J, Zhang L, Jiang Y D. Energy losing rate and open-circuit voltage analysis of organic solar cells based on detailed photocurrent simulation. Applied Physics Letters, 2009, 106(6): 063103

[3] Rider D A, Worfolk B J, Harris K D, Lalany A, Shahbazi K, Fleischauer M D, Brett M J, Buriak J M. Stable inverted polymer/fullerene solar cells using a cationic polythiophene modified pedot: pss cathodic interface. Advanced Functional Materials, 2010, 20(15): 2404-2415

[4] Tang C W. Two-layer organic photovoltaic cell. Applied Physics Letters, 1986, 48(2): 183-185

[5] Brabec C J, Sariciftci N S, Hummelen J C. Plastic solar cells. Advanced Functional Materials, 2001, 11(1): 15-26

[6] Xue J, Uchida S, Rand B P, Forrest S R. 4.2% efficient organic photovoltaic cells with low series resistances. Applied Physics Letters, 2004, 84(16): 3013-3015

[7] Wang N N, Yu J S, Zang Y, Huang J, Jiang Y D. Effect of buffer layers on the performance of organic photovoltaic cells based on copper phthalocyanine and C₆₀. Solar Energy Materials and Solar Cells, 2010, 94(2): 263-266

[8] Xue J, Uchida S, Rand B P, Forrest S R. Asymmetric tandem organic photovoltaic cells with hybrid planar-mixed molecular heterojunctions. Applied Physics Letters, 2004, 85(23): 5757-5759

[9] Service R F. Solar energy. Outlook brightens for plastic solar cells. Science, 2011, 332(6027): 293

[10] Coakley K M, Mcgehee M D. Conjugated polymer photovoltaic cells. Chemistry of Materials, 2004, 16(23): 4533-4542

[11] Cheyns D, Rand B P, Heremans P. Organic tandem solar cells with complementary absorbing layers and a high open-circuit voltage. Applied Physics Letters, 2010, 97(3): 033301

[12] Kinoshita Y, Hasobe T, Murata H. Control of open-circuit voltage in organic photovoltaic cells by inserting an ultrathin metal-phthalocyanine layer. Applied Physics Letters, 2007, 91(8): 083518

[13] Chan M Y, Lai S L, Fung M K, Lee C S, Lee S T. S. L. Lai S L, Fung M K, Lee C S, Lee S T. Doping-induced efficiency enhancement in organic photovoltaic devices. Applied Physics Letters, 2007, 90(2): 023504

[14] Wang N N, Yu J S, Lin H, Jiang Y D. Organic photovoltaic cells with improved performance using bathophenanthroline as a buffer layer. Chinese Journal of Chemical Physics, 2010, 23(1): 84-88

[15] Taima T, Sakai J, Yamanari T, Saito K.Doping effects for organic photovoltaic cells based on small-molecular-weight semiconductors. Solar Energy Materials & Solar Cells, 2009, 93(6,7): 742-745

[16] Cheyns D, Poortmans J, Heremans P, Deibel C, Verlaak S, Rand B P, Genoe J. Analytical model for the open-circuit voltage and its associated resistance in organic planar heterojunction solar cells. Physical Review B: Condensed Matter and Materials Physics, 2008, 77(16): 165332

[17] Lai S L, Lo M F, Chan M Y, Lee C S, Lee S T. LoM F, Chan M Y, Lee C S, Lee S T. Impact of dye interlayer on the performance of organic photovoltaic devices. Applied Physics Letters, 2009, 95(15): 153303

[18] Mandoc M M, Veurman W, Koster L J A, de Boer B, Blom P W M. origin of the reduced fill factor and photocurrent in mdmo-ppv: pcnepv all-polymer solar cells. Advanced Functional Materials, 2007, 17(13): 2167-2173

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg