Device Optimization Through Simulations
Optimization of Active Layer:- The thickness of the photoactive layer affects the incident light absorption and hence exciton generation in OSC, therefore its optimization is very crucial. To improve the PCE of the inverted OSC, the active layer was varied between 100-300 nm thickness[32]. Fig. 2 depicts the current density-voltage curves and the extracted electrical parameters for different active layer thicknesses. The J–V characteristics exhibited sensitive relation with the active layer thickness since Jsc largely depends upon absorption inside the photoactive layer. It is observed that Jsc, Pmax, η and FF show considerable dependence on the active layer thickness whereas Voc appears to be insensitive. The charge carrier generation and movement expressed by Jsc depends on the absorption and the diffusion length. For active layer thicknesses less than 200 nm, the absorption is incomplete whereas thickness above 200 nm enhances the exciton recombination and hence reduces the charge carrier collection. In view of this, the active layer was optimized to be 200 nm. For the 200 nm thick active layer, Pmax and PCE of 62.82 W/m2 and 6.28 % respectively have been obtained through simulations. Also, FF falls rapidly with an increase in active layer thickness while Voc falls slightly from 0.62V to 0.60V. The decrease in FF is because of the enhanced series resistance. The simulated results show the maximum efficiency for the 200 nm thick active layer, hence it was selected to fabricate the actual device.
Optimization of ETL:- The electron transport layer commonly known as the buffer layer is used to increase carrier collection in cells. This study shows the dependence of the performance of simulated inverted OSC on the variation in ETL thickness. The photoactive layer PTB7:PC71BM is fixed at the optimized thickness of 200 nm. The ETL material between the front cathode (AZO) and the photoactive layer (PTB7:PC71BM) should favour the electron conduction and restrict the transport of holes toward the cathode, thereby, preventing the recombination of charge carriers near the cathode. The alignment of energy levels for the active layer, ETL and cathode clearly shows that ZnO having a large bandgap (Eg) and electron affinity similar to AZO is the most suitable material for ETL. Zinc oxide (ZnO) is chosen as ETL and its thickness is varied between 10-50 nm. Fig. 3 depicts the observed J-V characteristics and the electrical parameters extracted. The variations of Jsc, Pmax and PCE show a similar trend with the increase in ETL thickness. Also, the increase in ETL thickness has only minimal effect on all the parameters in our simulated device. The Voc and FF remain almost constant with variations in ETL thickness[33]. Overall, it can be concluded that ETL thickness slightly affects the performance parameters of the OSC and the optimized thickness for ETL observed to be 30 nm.
Optimization of HTL:- From the simulations results of the preceding sections, the thickness of ETL and active layer is optimized to 30 nm and 200 nm respectively. In this section, the variation of HTL thickness is investigated through simulations. MoO3 is chosen as HTL due to its concomitant properties like better transparency, high conductivity and better hole transportation[34]. The MoO3 thickness is varied between 5-15 nm. The J-V characteristic curves and other parameters obtained through the simulations are drawn in Fig. 4 (a) and Fig. 4 (b) respectively. As seen in Fig. 4(b), the variation in HTL thickness has no prominent effect on FF and VOC, which is due to the low hole mobility. The optimized inverted organic solar cell is obtained at 10 nm thickness of HTL. Jsc, Pmax and PCE slightly vary with variation in the thickness of HTL and exhibit the optimum values at 10 nm. FF and Voc remain almost insensitive to variation in thickness of HTL. Consequently, the thickness of HTL has a minor effect on electrical parameters of simulated inverted OSC due to the low hole mobility[35].
Cathode Investigation
As the aim of this work is to deposit AZO as TCO having comparable characteristics to ITO, we started with deposition and characterization of AZO first. The AZO film over glass substrate was deposited using rf sputtering and the AZO film thickness was determined to be 800 nm with the help of Dektak profilometer. The XRD pattern for the AZO cathode film is shown in Fig. 5 (a). The AZO crystallites exhibit a hexagonal wurtzite structure (ICDD-00-051-0037). The spectra show a major peak at 62.85˚ reflecting the lateral growth along the crystallographic axis (103). Other peaks at 34.28˚, 36.18˚, 47.39˚ and 56.41˚ with small intensity are detected at axes (002), (101), (102) and (110), respectively. The (002) plane is related to vertical growth, however, the presence of a relatively intense peak along (103) is indicative of non-vertical growth. Another critical parameter of TCO is the low sheet resistance, which corresponds to better conductivity. The sheet resistance of AZO film was calculated using a four-point probe. Fig. 5 (b) shows the room temperature I-V curve, with sheet resistance around 20 Ω/sq which approaches the sheet resistance of ITO electrode ⁓ 15 Ω/sq.
The UV-Vis spectra in transmission mode for AZO and ITO films were measured and the wavelength dependent transmittance for both the films are plotted Fig. 5(c). It is evident from the plot that AZO film has an average transmittance of around 80% in the visible region (400 nm to 800 nm), comparable to ITO/glass. Also, the transmittance curve of AZO depicts a cut off at the wavelength ~ 380 nm which is due to the UV absorption properties of AZO (bandgap 3.26 eV). Although AZO blocks the UV part of the solar spectrum, most of the sunlight passes through it as the major portion of the solar radiation lies in the spectrum having λ > 380 nm, hence enabling efficient absorption in the device. This UV filtering capacity of the AZO film enhances the stability of the device by preventing device degradation due to UV exposure. Uniformity is a prerequisite for the cathode of a solar cell. To ensure the uniformity in AZO cathode, raman mapping was done at various locations over the film. Fig. 5(d) depicts the optical microscopic image of the film in the inset, indicating six different points where raman mapping was performed. The AZO characteristic anomalous peaks at 274, 475 and 511 cm−1 were present and for all the locations similar spectra were recorded, indicative of uniformity of the AZO cathode[36]. Since Al-doped zinc oxide has demonstrated electrical and optical properties that are comparable to ITO, it can effectively replace ITO to achieve cost-effective and stable photovoltaic devices.
Gold Nanoflowers Characterization
Efficient light absorption in inverted OSCs is important for their better performance and the inclusion of gold nanoflowers in the active layer serves the purpose without compromising their architecture. The incorporated plasmonic nanoparticles (5-50 nm) act like antennas and the incoming light energy is stored in LSPR modes resulting in multiple times enhancement of the local electromagnetic field leading to enhanced light absorption. The increase in exciton dissociation is also possible through interactions between the plasmons and the excitons thereby lowering the recombination loss of excitons. The introduction of the gold nanoflowers inside the active layer improves the PCE of the OSCs using the optical property of LSPR without affecting their electrical characteristics. Gold is preferably used as it is highly resistive to oxidation and hence ensures a longer lifetime of the cell. The size and shape of nanoflowers were investigated using TEM apparatus. The TEM picture shown in Fig. 6(a) suggests the formation of gold nanoflowers. The average size for gold nanoflowers analyzed by ImageJ software was 33 nm. The core of the nanoflowers was around 25 nm with the length of each petal around 4 nm. The absorption spectrum for the gold nanoflowers is the convolution of two absorption peaks as seen from fig. 6(b). There exist two absorption peaks within the visible regime, attributed to the surface plasmons in the gold nanoflowers. These nanoparticles show surface plasmon peaks at two different frequencies, the first peak at 588 nm arises from the core of nanoflowers and the other peak at 748 nm is due to free electrons present in petals.
Device Characterization
The simulation results of section 4.1 provided the optimized thickness of ETL (ZnO), active layer (PTB7:PC71BM) and HTL (MoO3) to be 30 nm, 200 nm and 10 nm respectively. Utilizing these optimized thicknesses, we fabricated the three devices:- (1) AZO based device. (2) ITO based control device. (3) plasmonic assisted AZO based device. The first AZO based device is fabricated based on the device geometry AZO/ZnO/PTB7:PC71BM/MoO3/Ag. The second device contains ITO as a front cathode keeping the remaining structure identical and is used for comparison studies (control device). The third device consists of plasmonic gold nanoflowers (average size of 33 nm) embedded into the active layer and using AZO as the front transparent conducting cathode. The J-V measurements for all three fabricated inverted OSCs were recorded using 100 mW/cm2 AM 1.5 G and it was observed that AZO based device showed a PCE of 6.19% which was slightly less than PCE 6.83% for the ITO based control device. Also, the PCE of the AZO based OSC was further enhanced to 7.01% when gold nanoflowers were embedded within the active layer of the OSC. The J-V characteristic curves for the three fabricated cells are shown in Fig. 7(a) and the extracted important parameters like PCE, FF, Pmax, VOC and JSC are included in Table 1. Also, the parameters extracted from the J-V curves of the experimentally fabricated AZO based inverted OSC matches well with the theoretically optimized device using GPVDM simulation which ensures the credibility of the simulation results. The EQE measurements for the experimentally fabricated cells are depicted in Fig. 7(b) and it is evident that the cell with AZO front transparent cathode shows a red-shift in the short wavelength region associated with cut off in transmittance (λ < 380 nm) for AZO cathode. The EQE curve for the AZO based device follows the absorbance in the visible portion of the solar spectrum indicating the non-contribution of the UV in the measured JSC. This can be attributed to the UV blocking property of AZO and is further augmented by the use of ZnO as ETL. It is interesting that AZO is acting as a transparent cathode and blocks the UV, which can increase the lifetime of the fabricated OSC.
Upon introduction of gold nanoflowers inside the active region, there is a considerable enhancement in EQE in the wavelength regime 400-780 nm for the AZO cathode based device. This enhancement in EQE is due to near field enhancement shown by plasmonic gold nanoflowers which increases absorption inside the active layer. Fig. 7(c) reports the absorbance spectra of the PTB7, PC71BM and PTB7:PC71BM for the wavelength range 330-800 nm. PTB7 exhibits stronger light absorption ranging from 550 to 750 nm while PC71BM presents an overwhelming absorption of light in the wavelength range 330 to 550 nm. Therefore, PC71BM provides complimentary light absorption when mixed with PTB7 and the blend becomes a champion active layer material.
The OSC degradation due to UV exposure can be eliminated by restricting the UV portion from the solar spectrum to reach the photoactive layer and this can be effectively done by the AZO front cathode. It is clear that AZO film has very low transmittance in the UV region showing its potential candidature as a UV filtering layer. The UV assisted acceleration ageing test was done to investigate the stability of the AZO based and ITO based inverted OSCs[37]. Both the devices with different front cathodes were illuminated under UV irradiation for an exposure time of 0 to 30 min from the cathode side. This test was performed inside the nitrogen filled glove box which
Table 1. Comparison of electrical parameters for the simulated and the fabricated devices.
Electrical Parameters
|
GPVDM Simulated AZO based device
|
Fabricated ITO based control device
|
Fabricated AZO based device
|
Fabricated plasmonics assisted AZO based device
|
VOC(V)
|
0.61
|
0.62
|
0.60
|
0.61
|
JSC(mA/cm2)
|
15.21
|
16.10
|
15.20
|
16.15
|
PMAX(W/m2)
|
62.82
|
68.30
|
61.92
|
70.12
|
Fill Factor(%)
|
67.71
|
68.42
|
67.89
|
71.18
|
PCE(%)
|
6.28
|
6.83
|
6.19
|
7.01
|
neglects the possible degradation of OSC due to moisture and oxygen. Fig. 7 shows the change in device parameters (PCE, FF, VOC, JSC) recorded for the AZO and ITO (control) based inverted OSC under UV exposure of 0 to 30 min. Though ITO based control device possesses slightly higher efficiency initially, a fast decrease in all important device parameters is observed from the graphs in Fig. 8. For the ITO based control device, parameters like PCE, FF, VOC and JSC dropped from 6.83 %, 68.42 %, 0.62 V, 16.10 mA/cm2 to 3.08 %, 47.23 %, 0.46 V, 14.20 mA/cm2 after 30 min of UV exposure. However, AZO based device exhibits only a slight decrease in the device performance parameters under the same test and hence ensures better stability. For the AZO based inverted OSC, the important parameters including PCE, FF, VOC and JSC changed from 6.19 %, 67.89 %, 0.60 V, 15.20 mA/cm2 to 5.48 %, 64 %, 0.57 V, 15.02 mA/cm2 after 30 min of UV exposure. This study reveals the better stability of AZO based OSC in comparison to ITO based devices and this is because of the UV blocking properties of the AZO layer. This work clearly indicates that AZO can effectively replace ITO cathode in OSCs to enhance the lifetime of the device.