Encapsulation against Extrinsic Degradation Factors and Stability Testing of Perovskite Solar Cells

*Edwin Ramírez, Rafael Betancur, Juan F. Montoya, Esteban Velilla, Daniel Ramírez and Franklin Jaramillo*

## **Abstract**

Commercialization of perovskite solar technology depends on reaching a stable functioning of the devices. In this regard, both intrinsic (chemistry phenomena of the different device layers) and extrinsic factors (environmental) need to be considered. In this chapter, we report the state of the art of encapsulation techniques against extrinsic degradation mechanisms. Our analysis includes the most common encapsulation structures, materials employed and their by-products, standard methods to test the stability of the devices (accelerated testing, outdoor and degradation monitoring), and security requirements to prevent the health/environmental hazard of lead leakage.

**Keywords:** encapsulant materials, extrinsic degradation, stability measurements, perovskite, solar cells

### **1. Introduction**

Commercialization of perovskite photovoltaic technology (PPT) relies on the golden triangle of solar cell performance whose vertices are lifetime, cost, and efficiency. In the last years, PPT has achieved photovoltaic conversion efficiency (PCE) up to 25.7% at lab scale, almost matching their silicon counterparts [1]. Moreover, the impressive advances in the fabrication of solar modules by scalable solution deposition techniques [2] such as doctor blade, slot-die, or ink-jet printing have enabled a rapid performance growth of large-area devices. According to NREL, the PCE of perovskite modules has increased from 11.8% to 17.9% in the last 4 years. Also the scale of devices has evolved from "Submodule" with an active area of 200–800 cm2 , to "Small module" with area ranging from 800 to 6500 cm2 [3]. A recent technoeconomic model established a cost range for solution processed perovskite modules of \$3.30/W–0.53 USD/W [4], which is competitive with silicon solar cells. The low cost projected for PPT is linked to the processability by scalable solution-based deposition techniques. Therefore, the figures of merit for cost and efficiency of perovskite solar cells (PSCs) are almost pairing, in few years—those achieved for silicon after decades of technological development. However, the recent record for stability of PSC reached

#### **Figure 1.**

*(a) State of the art of the stability of perovskite photovoltaic (PV) devices based on the data available in the open-access Perovskite Database [6], and (b) stability issues in perovskite solar cells. Reproduced with permission from references [6, 7].*

9000 hours under operational tracking [5], which is far behind the proven lifetime of 25 years (>200,000 hours) reported for silicon solar cells.

In recent years, several research articles have reported stability measurements of perovskite solar cells. **Figure 1a** shows the historical evolution of *T*80 for 1833 PSC devices recorded in the Perovskite Database [6]. *T*80 is a figure of merit defined as the time taken to observe a drop of 20% in the initial power output of the solar cell. Before 2016, only few articles reported *T*80 because the main focus was related to improving efficiency and processability of PSC. Since then, an increasing number of scientific articles report the *T*80 of PSC devices achieving in most cases values up to 1000 hours. Noteworthy, less articles report 1000 hours < *T*80 < 2000 hours and only a small proportion surpasses 2000 hours. This analysis of the scientific literature reveals the urgent necessity of increasing the PSC lifetime targeting to hundreds of thousands of hours.

A complex interaction of factors determines the stability of perovskite solar cells as shown in **Figure 1b**. Devices are degraded by multiple variables such as heat, light, electric load, moisture, and oxygen, which act simultaneously in real operation conditions. Such complexity explains why the progress in stability remains behind the rapid advancements in PCE and processability of PSC. Thus, understanding the degradation mechanisms is of crucial importance to overcome stability issues. Degradation mechanisms can be classified into *intrinsic***,** which are related to the compositional and crystallographic structure of the perovskite material, and *extrinsic***,** associated with the interaction of the PSC with external factors during their operational life.

### **2. Intrinsic and extrinsic degradation mechanisms**

#### **2.1 Intrinsic degradation mechanisms**

Hybrid halide perovskites (HHPs) have been considered as "soft crystalline materials" due to their low formation energy and stability dictated by a delicate

#### *Encapsulation against Extrinsic Degradation Factors and Stability Testing of Perovskite Solar Cells DOI: http://dx.doi.org/10.5772/intechopen.106055*

thermodynamic balance [8]. Active layers of solar cells have been obtained from HHPs with the general chemical formula ABX3, where A is an organic or inorganic cation, B is a metallic cation, and X is a halogen anion. This structure has enabled the use of a wide range of atoms as A cations or X anions without significantly losing their high photovoltaic (PV) performance [9]. However, the structure can only accommodate a certain combination of ions due to restriction of sizes in order to have a stable cubic or tetragonal 3D structure. This restriction is often expressed by the Goldschmidt tolerance factor (*t*), which is calculated in terms of the ionic radii of the constituent ions. The 3D perovskite structure is favored by 0.8 < *t* < 1 [8]. In halide perovskites used for PV, the A-site is a monovalent cation, commonly methylammonium (CH3NH3 + /MA+ ), formamidinium (CH(NH2)2 + /FA+ ), cesium (Cs<sup>+</sup> ), or a combination thereof. The B-site is a divalent metallic cation, mostly lead (Pb2+) but in some cases can be tin (Sn2+), and the X-site is a halide anion that is commonly iodide (I− ) or bromide (Br<sup>−</sup> ). A stable structure must preserve charge neutrality. Therefore, the valences of A and B should sum the charge of X multiplied by three. The mentioned restrictions in sizes and charges of the constituent ions determine that only certain perovskite formulations are stable at operational temperature ranges. For instance, the archetypical methylammonium lead iodide (MAPbI3) has a tolerance factor of 0.911 forming a tetragonal structure at room temperature. This phase is stable up to 327.4 K turning into a cubic structure at higher temperatures [10]. During the fabrication of perovskite films, as well as during the solar cell operation, the material is subjected to temperature cycles that can promote the formation of unwanted PV inactive phases. This phenomenon has been observed for formamidinium lead iodide (FAPbI3), which crystallizes into a non-perovskite hexagonal phase (yellowish phase) at room temperature due to its high tolerance factor (*t* = 1.04). This perovskite turns into a cubic structure after annealing over 150 °C [7]. FAPbI3-based PSC devices have been developed motivated by the higher thermal stability of FA+ compared with MA+ cation, but a careful material processing is required in order to avoid the formation of the non-active hexagonal phase of FAPbI3. Although pure FAPbI3 or CsPbI3 perovskites yield higher thermal stability of the A-site cation, they are not the preferred choice due to thermodynamic phase stability issues. Alloys of A-site cations have been used to produce materials such as FAxCs1 − *x*PbI3 or FA0.75MA0.15Cs0.1Pb(I0. 83Br0.17)3 used as active layers in high performance devices [11]. Thus, compositional tuning is a suitable strategy to overcome thermodynamic stability issues. Note that the proportion between the A-site cations must be strictly controlled to achieve a suitable tolerance factor. In addition, since the perovskite bandgap depends in part on the energy level of the anion occupied p orbital, such compositional tuning also determines the resulting bandgap.

The benchmark MAPbI3 perovskite has shown high defect tolerance preserving its opto-electronic properties even at high trap density (1014–1016 cm−3). In contrast, PV grade gallium arsenide (GaAs) must have a defect concentration as low as 107 cm−3 [12]. This high tolerance to defects enables the synthesis of HHPs by solution processing techniques. Density functional calculations show that the growth conditions of the perovskite correlate with the concentration and type of defects. Specifically, the I/Pb ratio determines the formation energy of defects. It was found that formation energy of deep trap states is very high, therefore the probability of having nonradiative recombination centers is low. As a result, shallow trap states found in the MAPbI3 perovskite are not detrimental of the photovoltaic performance [13]. This outstanding defect tolerance of the MAPbI3 perovskite derives from its exceptional band structure. Notably, related perovskites with compositional mixtures at the A-site cations or

halides have also shown defect tolerance if they have an adequate Goldschmidt factor. Accordingly, a HHP with high structural stability is also defect tolerant.

In summary, compositional tuning is a suitable strategy to increase intrinsic stability, which means a thermodynamically stable structure with high defect tolerance. Both properties are of utmost importance for a photovoltaic material.

#### **2.2 Extrinsic degradation mechanisms**

Beyond compositional tuning, additional strategies must be developed to increase extrinsic stability. During solar cell operation, the perovskite active layer may interact with external factors such as heat, light, moisture, oxygen, electric bias, and other interface or external agents. Here, we briefly review some of the most important degradation mechanisms. Further information can be found elsewhere.

*Water:* When water interacts with perovskite, it can form hydrate or dehydrate phases according to reactions (1) and (2).

$$\text{CH}\_3\text{NH}\_3\text{PbI}\_3(\text{s}) + \text{H}\_2\text{O}(\text{g}) \longleftrightarrow \text{CH}\_3\text{NH}\_3\text{PbI}\_3 \cdot \text{H}\_2\text{O} \tag{1}$$

$$4\left[\text{CH}\_3\text{NH}\_3\text{PbI}\_3\cdot\text{H}\_2\text{O}\right] \longleftarrow \longrightarrow \left(\text{CH}\_3\text{NH}\_3\right)\_4\text{PbI}\_6\cdot\text{2H}\_2\text{O} + \text{3PbI}\_2 + 2\text{H}\_2\text{O} \tag{2}$$

These reactions are reversible. Thus, perovskite can be regenerated exposing it to an inert environment. However, some irreversibility can appear due to phase segregation. Water forms hydrogen bonds with the A-site cation weakening its interaction with the lead halide octahedra [14]. As a result, the perovskite becomes prone to degradation by external other stressors such as heat or electric bias. Once the perovskite is saturated by moisture, it fully decomposes to PbI2 and MAI.

*Heat:* During solar cell operation, the device is subjected to temperature cycles. As mentioned before, PV active phases of perovskites are stable in a temperature range. Moreover, some A-site cations are volatile organic molecules, which can be converted into gas-phase products when the perovskite reaches some critical temperatures. For instance, the MAPbI3 can be decomposed to PbI2, ammonia and methyl iodide when it is heated to 85°C in an inert atmosphere [15]. Additionally, the materials commonly used in hole or electron transporting layers are organic molecules, which can also be degraded at some temperatures. The main strategies to avoid degradation by heat are compositional engineering of the A-site cation and encapsulation with materials with good heat dissipation properties.

*Light:* light-induced degradation has been attributed to the migration of vacancies in the perovskite layer [16]. UV degradation takes place in the absence of moisture and particularly in the device stack denoted as n-i-p, especially those containing TiO2, which can induce photo-degradation. However, it has been demonstrated that UV-degraded devices can be subsequently recovered by 1-sun light soaking. The UV degradation/recovery phenomenon has been attributed to the free carriers generated by light soaking after neutralization of accumulated trap states and generated free charges [17].

Moreover, light can induce redistribution of halide and metal ions in the MAPbI3 perovskite film. This phenomenon causes the increment of the photoluminescence under illumination due to the diffusion of I− species [18]. Ionic migration leads to phase segregation in the perovskite layer with halide-rich or halide-deficient areas being the most commonly found. These defects are carrier trapping states with

*Encapsulation against Extrinsic Degradation Factors and Stability Testing of Perovskite Solar Cells DOI: http://dx.doi.org/10.5772/intechopen.106055*

smaller bandgaps, which generate the increment in the photoluminescence. In the presence of moisture or oxygen, this phenomenon is more pronounced due to the passivation effect of superoxide molecules generated by the reaction of O2 and H2O with light [19]. In alloyed perovskites, light-induced A-site cation segregation has also been observed [20].

The main strategy against extrinsic degradation mechanisms is based on the encapsulation process, which is reviewed in the following.
