当前位置: 主页 > 中文 > 新闻中心 >
Recent progress in stability of perovskite solar cells
时间:2020-06-23 09:05来源:未知 作者:admin 点击:
1.   Introduction
  • Perovskite solar cells have shown extremely high performance and now their power conversion efficiency (PCE) is close to the traditional inorganic semiconductors thin film solar cells, such as CIGS and CdTe[1-8]. It was predicted that the efficiency of perovskite solar cells could be beyond 25% in the near future by controlling perovskite layer growth or passivation of surface defects. At present, the most challenging part in perovskite solar cells is the long-term stability, which must be solved before putting it into practical applications. In this mini-review, we will focus on the stability of perovskite solar cells, and we will firstly discuss the degradation mechanism of perovskite solar cells, and then show some recent progress on improving the stability of perovskite solar cells.

2.   Degradation mechanism

2. 1.   Perovskite material degradation

2. 1. 1.   Moisture stability
2. 1. 2.   Thermal and photo stability
2. 1. 3.   Ion movement

2. 2.   Interface degradation

2. 2. 1.   Electrode degradation
2. 2. 2.   Charge transport layers degradation
  • The degradation of perovskite solar cells mainly could be due to two reasons, one is degradation of perovskite material itself and another one could be from interfacial degradation.

  • The typical halide perovskite material ABX3 (A = CH3NH3, HC(NH2)2; B = Pb, Sn; X = Cl, Br, I) shows severe moisture instability. The highly hydrophilic properties of perovskite can cause the materials to easily absorb moisture from the surrounding environment and induce the formation of hydrate products similar to (CH3NH3)4PbI6▪2H2O[9]. Another possibility is that the perovskite could be decomposed into a lead salt (e. g. PbI2) and an organic salt (e. g. CH3NH3I) under a moisture environment (Fig. 1)[10].

  • Recent studies showed that perovskite begins to transform into PbI2 at lower temperatures up to 140 ℃[11]. Conings et al. even found that the perovskite could be decomposed into PbI2 while heated in nitrogen at 85 ℃ for 24 h (Fig. 2)[12]. It must be noted that slightly over heating of perovskite could lead to the formation of a small amount of PbI2, which could enhance the device performance due to the passivation[13]. While heating too much could lead to serious decomposition of perovskite, and lead to poor device performance. In addition to thermal instability, perovskite solar cells showed inferior photo-stability. Although the device performance may be recovered after storing in the dark for a short period of time, such poor photo-stability could be due to a local phase change under a higher temperature after exposure to light.

  • Ion movement could be a special issue for the perovskite solar cells. For the halide perovskite materials, the ion, either anion/cation, could be moved under voltage bias or thermal drift, which could lead to instability of the devices. This migration of defects/ion such as iodine vacancies across the interface can induce interfacial degradation, and affect device operational mechanisms and finally cause device failure during operation[1415]; more proof is needed to confirm this argument.

  • In addition to the instability of perovskite material itself, the interfacial stability is also very important for the overall stability of the devices. The first interface degradation part is mainly from the chemical reaction between perovskite and the electrode. Recently, it was found that some traditional metal electrode materials such as Al and Ag could react with perovskite materials[16]. The reaction mechanism could be that the decomposition product of HI from perovskite materials react with Al or Ag to form AlI3 or AgI[16]. You et al. . found that the reaction between perovskite and Ag or Al accelerates the decomposition of perovskite while exposing the perovskite material in an ambient environment (Fig. 3)[16]. Therefore, totally insulating perovskite from the electrode (Ag/Al) could enhance the stability, however, diffusion of halide ions through the transport layer could affect the long stability of the device.

  • Organic semiconductors were usually adopted as charge transport layers in perovskite solar cells, while these layers could easily be either oxidized or water absorbing, leading to the instability of the devices. For example, in an inverted structure, n-type of fullerene e. g. PCBM was used as an efficient charge transport layer in perovskite solar cells, however, it was found that the PCBM was not stable in ambient air due to chemical states or band structure variation, which has been explained as the reason for the device degradation[16]. P-type of PEDOT:PSS was usually used as the hole transport layer in the inverted structure; this layer could easily absorb water from the environment, and also etch the transparent conductive electrode such as ITO due to the acidic properties of PEDOT:PSS, both of these lead to the degradation of the devices[1617]. For the Spiro-OMeTAD layer, dopant of Li salt could absorb water and lead to water penetration into the perovskite layer, as a result of failure of the devices[18]. Another issue is the degradation of the TiO2 layer, several results indicated that the de-absorption of oxygen from the TiO2 surface while light soaking, which leads to failure of the devices[19].

    As mentioned regarding the perovskite stability, the main issues for less stability of perovskite solar cells was due to the perovskite materials and interface; several methods have been adopted to improve the device stability.

3.   Stability improvement

3. 1.   Perovskite materials modification

3. 1. 1.   Organic "A" cation
3. 1. 2.   Halide "X" anion

3. 2.   Interface or electrode engineering for improving device stability

3. 2. 1.   Hole transport layers modification
3. 2. 2.   Electron transport layers improvement
3. 2. 3.   Electrode
  • The intrinsic structural properties of perovskite materials may be directly responsible for instability of perovskite solar cells, as the formation enthalpy of MAPbX3 is relatively high and is inversely related to its tolerance factor (MAPbI3: 0. 91; MAPbBr3: 0. 92 and MAPbCl3: 0. 94)[1520]. As inorganic oxide perovskites can have an ideal cubic structure with a tolerance factor close to 1, they can serve as a reference for organic halide perovskites as a deviation from the ideal cubic structure[1520]. The poor structural stabilities of perovskites could thus be improved by modifying the Goldschmidt tolerance factor tf=(rA+rB)/2(rB+rX),whererA,rB, and rX are the radii of cation A, cation B, and halogen, respectively, in the ABX3 structure[1520]. Several strategies have been adopted to modify the perovskite structures to obtain a stable structure of perovskite, that include increasing the radius of A or reducing the X site; changing the B site is a great challenge, which could be because Pb is greatly important in delivering the high performance of the devices. As follows, we will show some progress in modifying the perovskite layer according doping or replacing the A site cations or X site of anions.

  • The most investigated lead halide perovskite was CH3NH3PbI3 (MAPbI3), where it was found that the tolerance factor is 0. 91; theoretical calculation and experimental results showed that MAPbI3 is intrinsically instable (Fig. 4). Replacing the MA organic cation with a smaller size (MA:2. 17Å)by HC (NH2)2+ (FA:2. 53 Å) can push the tolerance factor tf of 0. 99 with an xxx structure, which could potentially enhance the thermal stability[15]. It was found that the FAPbI3 does not discolor even at 150 ℃ under ambient conditions, while MAPbI3 discolors in 30 min[21]. It must be noted that the FAPbI3 could easily show the yellow phase (δ-phase), and the tolerance factor is larger than 1, which also showed instability. To reduce the tolerance to less than 1, smaller cations should be incorporated (Fig. 5), for example, Cs or Rb could be incorporated into the alloy with FA[22-25]. Cs doped perovskite was initially invented by Kim et al. , where they initialized to improve the film morphology[26]. Later, Park et al. found that the CsI could improve the device stability of FAPbI3 in the humidity[23], the reason could be that the pure FAPbI3could be changed into δ-phase (yellow phase), and the tolerance factor is beyond 1, which leads to less stability, and the yellow phase will be formed; after doping Cs into the FAPbI3, the tolerance could be reduced to 0. 9-1 and the α-phase could be stabilized. The doping could lead to the enhancement of the thermal stability. It was also found that the Cs could enhance the device performance.

    Further enlarging the "A" cation site leads to the formation of a two dimensional structure of perovskite, while the inefficient out-plan electron will limit the efficiency. Han et al. doped a small amount (5%) of A by 5-aminovaleric acid (5-AVA) cation in the A site of MAPbI3, forming the new mixed cation perovskite, (5-AVA)x(MA)1-xPbI3, combined with a stable carbon electrode, the devices showed much enhanced stability[27]. The improved stability has been considered as the enhanced crystallinity of the perovskite layer, the role of tuning the tolerance has not been realized[27]. A more stable new perovskite, (PEA)2(MA)2Pb3I10 (PEA = C6H5(CH2)2NH3+, MA = CH3NH3+) has also been reported (Fig. 6)[28]. Films of these materials are more moisture resistant than films of MAPbI3 and devices can be fabricated under ambient humidity levels. This moisture resistance ability may be due to the more hydrophobic tail 'R' group, which may mask the hydrophilic nature of the materials. However, the fundamental reason for alloy stabilization of the structures requires further study. For quasi-two dimensional perovskite, the out-of-plane charge transport conducted by the organic cations, which act like insulating spacing layers between the conducting inorganic slabs, limiting the vertical charge transportation. The initial power conversion efficiency achieved is only about 4. 7%[28]. Later, Sargent et al. . adopted (PEA)2(MA)n-1PbnI3n+1 structure, while more MAI has been used which keeps its almost three dimensional structure and also shows high efficiency (15%)[29]. More importantly, the shelf-device stability has been significantly improved[29]. Recently, a breakthrough in quasi-two dimensional perovskite by controlling the orientation, specifically, they produced thin films of near-single-crystalline quality, in which the crystallographic planes of the inorganic perovskite component have a strongly preferential out-of-plane alignment with respect to the contacts in planar solar cells to facilitate efficient charge transport[30]. A photovoltaic efficiency of 12. 52% with no hysteresis was obtained, and the devices exhibit greatly improved stability in comparison to their three-dimensional counterparts when subjected to light, humidity and heat stress tests[30]. Unencapsulated two-dimensional perovskite devices retain over 60% of their efficiency for over 2250 h under constant, standard (AM1. 5G) illumination, and exhibit greater tolerance to 65% relative humidity than do three-dimensional equivalents (Fig. 7). When the devices are encapsulated, the layered devices do not show any degradation under constant AM1. 5G illumination or humidity for 2250 h[30].

  • Reducing the size of the X site could also enlarge the tolerance factor and enhance the device stability. The smaller size of Br (1. 96 Å) and Cl (1. 81 Å) compared with I (2. 2 Å) could be a good candidate for replacing or doping the I site. Unfortunately, it was found that the Cl doping is not efficient due to the too small size of Cl- and it cannot occupy the I- site. Experimental results showed that Cl could easily escape from the I site even when there is a large ratio of Cl in the precursor[3132]. Partial replacement of I by Br to form an alloy such as MAPb(I1-xBrx)3 or FAPb(I1-xBrx)3 are very popular compositions to tune the tolerance factor[2133]. Seok et al. has demonstrated the doping Br into I site can enhance the humidity stability of MAPbI_3. Interestingly, the MAPb(I1-xBrx)3 (x = 0, 0. 06) hybrid solar cells exhibited serious PCE degradation after exposure to 55% humidity, whereas the other MAPb(I1-xBrx)3 (x = 0. 2, 0. 29) cells maintained their PCE (Fig. 8(a))[33]. They also found that while doping the A site with FA, and the X site with Br in FAPbI3, (FAPbI3)1-x(MAPbBr3)x also enhanced the phase stability (Fig. 8(b))[34]. These results show that the role of bromine inclusion in thermal, light and ambient stabilities should be identified. Such a comparative study could help towards improvements of better stability.

  • For the general n-i-p structure, Spiro-OMeTAD has been used as the hole transport layer, as mentioned before, the Li-salt doped for Spiro-OMeTAD is very sensitive to water, which is detrimental for device stability. Developing some small molecular or polymer with a high charge mobility without the need for doping is needed. Liu et al. showed several small molecular with high mobility as the hole transport layer, and the MoO3 has been used as the dopant in the small molecular or polymers for further enhancing the charge transport ability[253536]. The devices with Li-salt free dopant was anticipated to show good efficiency, although the results have not been shown in the reference. In addition to the dopant free organic charge transport layer, several inorganic compounds have been used to replace the Li-salt doped Spiro-OMeTAD, such as CuSCN[37], and CuI[38]. Grätzel et al. have tried to use CuSCN as the top electrode to improve the stability, the devices showed 12. 4% power conversion efficiency, while the device stability has not been shown[37].

    For the p-i-n structure, generally, the hole transport layer, such as PEDOT:PSS was used, as this layer can easily absorb water and also has acidic properties, and it will lead to the instability of the devices. The most successful results are based on using NiOx as the hole transport layer to improve the device stability. Guo et al. first reported the use of solution processed NiOas the hole transport layer, demonstrating an efficiency of about 8%[39]. Jen et al used Cu-doped NiOx as a hole transport layer and achieved open circuit voltages as high as 1. 1 V with an efficiency of 15. 4% , and showed 250 h stability while storing in ambient air[40]. Later, You et al. adopted a sol-gel process for high quality NiOx films, demonstrating an efficiency of 16. 1% from a device combined with the metal oxide ZnO as the electron transport layer, and keeping its 90% original efficiency while storing in ambient air for 60 days, further confirming the role of NiOx in improving the device stability[16]. Recently, Seok et al. reported an inverted structure using a NiOx film deposited via pulsed laser deposition (PLD), and pushed the device efficiency as high as 17. 13% (Fig. 7(d))[41]. In addition to NiOx, another promising hole transport layer is CuSCN. Recently, adopting electro-deposited CuSCN as a hole transport layer in an inverted structure, Huang et al. . showed that as high as 16. 6% efficiency has been obtained[42].

  • As mentioned before, for a traditional n-i-p structure, a TiO2 compact and mesoporous layer were usually adopted, while the TiO2 layer had easy adsorption of oxygen and lead to the instability of the devices. Liu et al. . have demonstrated that the ZnO could be used as an electron transport layer for perovskite solar cells, and showed 15. 3% efficiency[43]. While these metal oxides showed good performance, the long term stability was raised due to the chemical reaction between perovskite and ZnO. More recently, it was found that the SnO2 acted as a good electron transport layer due to better conduction band alignment between perovskite and SnO2 and also the high carriers mobility of SnO2[44-48]. Solar cells using SnO2 as the electron transport layer showed high efficiency and less hysteresis[49]. More importantly, it was found that the SnO2 based devices showed improved stability, which could be due to the resistance to ultraviolet light because of the larger band gap of SnO2[49]. For the p-i-n structure, the fullerene, such as PCBM, and C60 were used as the top electron transport layers; these materials showed good device performance while inferior stability in ambient air. Developing inorganic electron transport such as TiO2, ZnO, and SnO2 has been successful. You et al. used the ZnO on the top of perovskite directly, and found that the device showed much improved stability; while the devices are stored in ambient air for 60 days, the devices can keep their original efficiency as high as 90% (Figs. 9(a) and 9(b))[16]. Chen et al used Nb doped TiO2 as an electron transport layer on PCBM and significantly enhanced the stability under 1000 h of continuous one sun soaking (Figs. 9(c) and9(d))[50]. More recently, Jen et al. has deposited SnO2 colloid on the PCBM surface to avoid the exposure of PCBM directly into the ambient air. They have showed that over 90% of its initial PCE can be retained after 30 days storage in ambient with >70% relative humidity[51]. Such crystalline SnO2 ETL not only provides a simple way to improve the performance and stability of PVSCs but also shows the great merits in future development of efficient interconnecting layers for perovskite-based tandem cells. The unique SnO2 could be more stable than that of ZnO due to the acidity resistance.

  • The electrode is the uppermost layer which is closest to the environment. Therefore, the electrode must be robust enough to slow down the moisture penetrating into the perovskite layer, and also enhance the device stability. The most successful example could be the utilization of a thick carbon electrode (>10 μm). Han et al. adopted the thick carbon as the electrode and the device's own hole transport layer, and delivered the efficiency as high as 12. 8%; further, the devices showed good stability under 1000 h light soaking (Fig. 10)[27].

4.   Conclusions
  • Above all, the ABX3 structure of halide perovskite materials is intrinsically instable, which is the main reason for the instability of perovskite solar cells. In the past several years, the stability of perovskite solar cells has been improved from several minutes to 3000 thousands of hours. This accomplishment was completed by modifying the material itself, such as doping in A or X to tune the tolerance factor to form a much more stable phase; or modifying the interface to form a stable interfacial layer which is resistive to moisture, oxygen or ultraviolet; or using thick electrode materials such as carbon electrode to slow down the moisture penetration into the perovskite layer. Although there is significant improvement, the stability is not good enough for practical applications; we should further extend the life time of perovskite solar cells to several years and even to more than ten years. To achieve this great goal, one cannot guarantee the long life time by just modifying the present perovskite materials or interface, we should call on the scientists working in this area to design and develop some new materials with high stability in severe conditions, such as high humidity, high temperature and ultraviolet illumination. We believe this is the future direction in perovskite solar cells.

联系我们
电话:+86-512-62870366
传真:+86-512-62872996
Email: kevin.xia@sicreat.com
地址:苏州工业园区金鸡湖大道99号苏州纳米城
2015-2016 苏州新美光纳米科技有限公司 版权所有 Suzhou Sicreat Nanotech Co.,Ltd. All rights reserved.