Fundamental Understanding of Solar Cells

1 Fundamental Understanding of Solar Cells Kevin P. Musselman, Kianoosh Poorkazem DE PARTMENT OF MECHANICAL AND MECHATRONICS ENGINEERING, UNIVERSITY O F WATERLOO , WATERLOO, ON, CANADA

1. Operation of a Photovoltaic Solar Cell 1.1 Photovoltaic Fundamentals In photovoltaic energy conversion, an incident photon imparts energy to an electron in an absorbing material, promoting the electron to an excited state. Commonly this corresponds to an electron being excited from the valence band to the conduction band in a semiconductor. In a photovoltaic solar cell, some built-in asymmetry is present to remove the excited electrons (and the energy associated with them) before they can relax to their lower-energy states. Typically, this asymmetry is provided by an interface between an electron-conducting (n-type) material and a hole-conducting (p-type) material, as in the p-n junction illustrated in Fig. 1. The asymmetric junction separates the excited electrons and holes (vacant electron states) and collects them at opposite electrodes of the device, producing a potential that can power an external circuit. The equivalent circuit for a solar cell is shown in Fig. 2. The short-circuit current density Jsc is the current resulting from light absorption when the cell is measured in a shortcircuit configuration, such that there is no potential difference V across it. The direction of Jsc through the cell is shown in Figs. 1 and 2. The asymmetric p-n junction is represented by a diode circuit element in Fig. 2. The formation of a potential difference across the cell results in a dark current density Jdark in the diode, as shown in Fig. 2, that opposes the short-circuit current density Jsc created by light absorption. Parasitic resistances that exist in the solar cell and reduce the achievable power output are also shown in Fig. 2. The series resistance Rseries corresponds to the resistance to current in the junction materials and current-collecting electrodes. It results in a parasitic voltage drop in the solar cell, which reduces the external potential V that can be produced. Any disruption in the asymmetry of the cell results in leakage current through the cell, which corresponds to a shunt resistance Rshunt, as shown in Fig. 2. For efficient cell operation, a low Rseries and high Rshunt are desirable.

Advanced Micro- and Nanomaterials for Photovoltaics. https://doi.org/10.1016/B978-0-12-814501-2.00001-3 © 2019 Elsevier Inc. All rights reserved.

1

2

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

FIG. 1 Basic operation of a photovoltaic solar cell: a photon is absorbed, exciting an electron to a higher energy state. A built-in asymmetry separates the electron and hole, resulting in the generation of current and voltage.

+ Rseries

Rshunt

Jsc

V

Jdark

– FIG. 2 Equivalent circuit for a photovoltaic solar cell.

Following from this equivalent circuit, the net current density J and voltage V produced by the cell can be related to the given parameters:
qðV + JAR Þ  series V + JARseries mkT J ¼ Jsc  J0 e 1  ARshunt

(1)

In this equation, J0 is a constant describing the dark current density, q is the electron charge, A is the area of the solar cell, m is the ideality factor describing the dark current’s dependence on applied bias, k is the Boltzmann constant, and T is the temperature.

1.2 The Silicon Photovoltaic Solar Cell The solar cell described in Section 1 is the basic building block of a photovoltaic system. When illuminated by the sun, the voltage produced by a typical cell, such as the silicon solar cell discussed in this section, is around 1 V or less. This voltage is insufficient for most applications, so solar cells are typically connected together in series to form a photovoltaic module, a schematic of which is shown in Fig. 3A and is easily recognizable from roof-top and utility-scale photovoltaic installations.

Chapter 1 • Fundamental Understanding of Solar Cells

3

FIG. 3 Schematics of (A) a photovoltaic module and (B) a cross-section of a traditional silicon solar cell.

The majority of photovoltaic modules currently in use consist of silicon solar cells. A traditional silicon solar cell is fabricated from a p-type silicon wafer a few hundred micrometers thick and approximately 100 cm2 in area. The wafer is lightly doped (e.g., approximately 1016 cm3) and forms what is known as the “base” of the cell. It may be multicrystalline silicon or single-crystal silicon. An n-type layer is produced on the surface of the wafer using dopant diffusion to create the desired p-n junction. The n-type layer, known as the “emitter,” is more heavily doped (e.g., approximately 1019 cm3) and much thinner than the p-type base. A schematic cross-section of a silicon solar cell is shown in Fig. 3B. Prior to formation of the n-type emitter layer, the silicon wafer is typically texturized via chemical etching to reduce the reflection of light from its surface. Following the formation of the n-type layer, antireflection coatings can be added to the surface to further enhance light absorption. These coatings can consist of SiNx or TiO2, and in addition to reducing reflections, can provide some passivation to prevent charges from getting trapped at the surface. The rear surface is more heavily doped to create a back surface field that limits surface recombination. Metal contacts are then applied using a printing technique, such as screen-printing of aluminum or silver. A grid-like metal pattern is printed on the front of the cell, to limit the amount of light blocked, whereas the metal contact on the rear side of the cell can be continuous, as shown in Fig. 3B. Many variations of this silicon cell architecture have been developed to improve the performance. These include, among others, passivated emitter cells, which include a dedicated passivation layer on the front surface; passivated emitter and rear cells (PERC cells), which employ a dielectric passivation layer on their rear surface to reflect unabsorbed light back into the cell; and heterojunction with intrinsic thin layer cells (HIT cells), which sandwich the crystalline silicon between thin amorphous silicon layers. Silicon solar cells are prone to mechanical damage and corrosion, so the manufacturing of modules includes encapsulation steps to provide suitable protection. The seriesconnected cells can be encapsulated in a transparent polymer, such as ethylene vinyl acetate (EVA), and then mounted on a sheet of glass, which acts as the top, transparent

4

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

surface of the module. An additional polymer layer, such as a sheet of polyvinyl fluoride, can then be added to provide a further barrier to water and gases on the underside of the module.

1.3 Measures of Photovoltaic Performance Power is generated by a solar cell when it operates in the fourth quadrant of a current density-voltage (J-V) graph, where voltage is positive and current negative. The power conversion efficiency (PCE) is the most fundamental measure of solar cell performance. It is defined as the ratio of the electric power delivered to the external circuit to the solar power incident on the cell: PCE ¼

Pout FF  Jsc  Voc ¼ Pincident light intensity

(2)

The short-circuit current density, open-circuit voltage, and fill factor of the cell are represented by Jsc, Voc, and FF, respectively. The open-circuit voltage Voc is the cell voltage measured in an open-circuit configuration, such that there is zero net current through the cell. The fill factor FF is defined as FF ¼

ðJ  V Þmax ðJsc  Voc Þ

(3)

It is the power produced at the maximum power point on the J-V curve divided by the product of Voc and Jsc. It can be thought of as a measure of how rectangular the J-V characteristic is in the fourth quadrant. Another measure of photovoltaic performance that is often reported is the quantum efficiency. The external quantum efficiency EQE is the ratio of electrons collected by the solar cell to the number of incident photons at a particular energy. Ideally the EQE is a step-function at the band edge of the light-absorbing material. The EQE at a particular wavelength λ depends on the proportion of light that is reflected from the front surface R(λ), the percentage of non-reflected light that is absorbed a(λ), and the efficiency with which photogenerated electron-hole pairs are collected from the cell ηCE(λ), according to EQE ðλÞ ¼ ð1  RðλÞÞaðλÞηCE ðλÞ

(4)

If the EQE is known, the expected short-circuit current density can be calculated as Z Jsc ¼ q

∞

EQE ðλÞϕðλÞdλ

(5)

0

where ϕ(λ) is the solar photon flux. The internal quantum efficiency IQE accounts for reflection and parasitic absorption losses. It gives the ratio of electrons collected by the solar cell to the number of photons absorbed by the active layer, again as a function of energy (or wavelength). The distinction between EQE and IQE is important when considering whether losses in a cell are due to optical absorption properties of the device stack, or due to the photoconversion properties of the absorbing materials.

Chapter 1 • Fundamental Understanding of Solar Cells

5

The National Renewable Energy Laboratory (NREL) maintains a record of the highest certified values of power conversion efficiency for different photovoltaic technologies, as shown in Fig. 4. These PCEs are for research cells and are therefore higher than what is obtained in commercial cells and modules, but the chart nonetheless provides an excellent overview of the progress made with the various technologies. The PCE record for a crystalline silicon cell is 25.8% in Fig. 4 and 26.6% for a HIT cells. Fig. 4 also shows PCE records for a number of other photovoltaic technologies, many of which will be addressed in the following sections and chapters. There are a number of reasons the PCE of solar cells is not unity. These will be discussed in the next section.

2. Limitations in Photovoltaic Energy Conversion 2.1 Single-Junction Maximum Theoretical PCE The possible PCE of the ideal, two-band, single-junction solar cell, such as that shown in Fig. 1, depends primarily on the incident light spectrum and the band gap Eg of the absorber. For a fixed solar spectrum, the theoretical efficiency therefore depends almost exclusively on the band gap. This dependency is explained in Fig. 5. Photons with energies Ephoton less than the band gap of the absorber are transmitted through the semiconductor without being absorbed, such that their energy is lost. Photons with energies Ephoton larger than the band gap are absorbed and excite electrons to higher-lying energy states, but these electrons rapidly thermalize to the band edge by interacting with phonons, reducing their electrical potential energy. Therefore, small and large band gaps result in low PCEs. For a large Eg, few photons are absorbed to produce electrical power. For a small Eg, many photons are absorbed, but the excited electrons lose most of their electrical potential energy (the voltage that the cell can produce is always less than Eg). This trade-off between current and voltage results in a maximum PCE at an intermediate band gap. For singlejunction solar cells under standard global air mass 1.5 (AM1.5G) illumination, the maximum efficiency is around 33% at a band gap near 1.3 eV for unconcentrated solar illumination [1, 2].

2.2 Strategies to Exceed the Single-Junction Limit Several approaches have and continue to be developed to achieve PCEs beyond the singlejunction limit. Recognition that some of the losses illustrated in Fig. 5 can be avoided if photons of different energies can be absorbed in cells with different band gaps has led to the development of tandem (also known as multijunction) solar cells. Tandem cells place an absorber with a large Eg on top of an absorber with a smaller Eg. The light first passes through the absorber with the large Eg, which filters out most of the high-energy photons, then through the absorber with the smaller Eg, where photons with smaller energies are absorbed. By matching photon energies and band gaps, the thermalization losses shown in Fig. 5 can be minimized, while still absorbing many of the photons. This tandem

6 ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

FIG. 4 Highest confirmed power conversion efficiencies for photovoltaic cells, from 1976 to the present, for a range of photovoltaic technologies. This plot is courtesy of the National Renewable Energy Laboratory, Golden, Colorado.

Chapter 1 • Fundamental Understanding of Solar Cells

7

FIG. 5 Fundamental losses in a single-junction solar cell: low energy photons are not absorbed and electrons excited by high-energy photons lose energy via thermalization.

concept is shown in Fig. 6A and B in four-terminal and two-terminal designs, respectively. The four-terminal design requires independent electrical connections for each of the two cells, which can be challenging in practice, whereas the two-terminal design connects the cells in series, avoiding the need for additional connections. However, the two-terminal design requires that the currents from each cell be matched, which can also be challenging in practice, especially in conditions of variable illumination. For the four-terminal design, a maximum theoretical PCE of 55% is expected with band gaps of 1.65 and 0.75 eV under full solar concentration [2]. The maximum theoretical efficiency is slightly less for twoterminal designs, due to the current-matching requirement. Some of the materials commonly employed in tandems cells include GaAs, InGaP, Si, and now organometal halide perovskites [3]. In Fig. 4, a PCE of 32.8% is reported for a two-junction tandem cell based on GaInP and GaAs in a two-terminal design under un-concentrated illumination, and 35.5% is reported for a two-junction tandem cell based on GaInAsP and GaInAs in a two-terminal design under concentrated illumination. Adding more band gaps can further increase the PCE. In the limit of an infinite number of band gaps, the maximum theoretical efficiency reaches 69% under one sun illumination (and 86% under full concentration). It is seen in Fig. 4 that PCEs of 44.4% and 46.0% have been achieved for two-terminal tandems with three and four junctions, respectively, under concentrated illumination. Tandem cells attempt to avoid the thermalization losses illustrated in Fig. 5 by employing several absorbers with different band gaps. Another approach is to try and harness some of the excess kinetic energy of these excited electrons before they relax (thermalize). This can be done if the electron-phonon interactions are slowed down (or the carrier collection sped up, or both), so that the excited electrons can be collected while still “hot.” This process is illustrated in Fig. 6C and would result in an increase in the cell voltage. The theoretical maximum PCE for a hot carrier solar cell is approximately 85% under full concentration and 65% under one sun illumination [2]. The band gaps that would provide these maximum efficiencies are a few tenths of an eV and approximately 0.25 eV, respectively.

8

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

FIG. 6 Strategies to exceed the single-junction photovoltaic efficiency limit: (A) four-terminal and (B) two-terminal tandem cell designs, (C) hot carrier extraction and impact ionization, (D) photon upconversion, and (E) photon downconversion.

Hot carrier cells allow the band gap to be reduced to increase the number of photons absorbed, without the negative effect of increased thermalization losses. In practice, thermalization can be slowed down in quantum well and quantum dot structures, where electron-phonon interactions are inhibited, and devices with small dimensions can enable

Chapter 1 • Fundamental Understanding of Solar Cells

9

rapid carrier collection. An additional requirement for hot carrier cells is specially designed contacts that are energy selective (i.e., will only accept carriers within a narrow energy range), in order to minimize heat production of the high kinetic energy carriers in the contacts. Alternatively, the excess energy of the hot carriers could be used to create more carriers. This process, known as impact ionization, Auger generation, or multiple pair generation, is illustrated in Fig. 6C and would result in an increase in the cell current. The energetic electron collides with the lattice and gives up its kinetic energy to excite another electron across the band gap. This enables a theoretical IQE of 200% for high-energy photons. For 1 sun illumination, the theoretical maximum PCE for an impact ionization cell is 55% at a band gap of approximately 1 eV. Under full concentration, the theoretical behavior of impact ionization cells is similar to the hot carrier cells, reaching a maximum PCE of 85% for very small values of Eg [2]. Another approach to improve the cell efficiency is to control the photon energies to better match the band gap of the absorber. This can be achieved using photon converters. Two low-energy photons (that would otherwise not be absorbed by the cell) can be combined to produce a high-energy photon that can be absorbed. This process is known as upconversion and is illustrated in Fig. 6D. Alternatively, one high-energy photon (that would otherwise create a hot carrier prone to thermalization losses) can be split to produce two photons that can still be absorbed by the cell but have smaller energies and smaller associated thermalization losses. This process is known as downconversion and is illustrated in Fig. 6E. In practice, photon converters can be added as thin films to the solar cell. Inorganic Yb+3 and Er+3 have been used to dope a number of materials to obtain upconversion [4]. Since the Eg of organic molecules is tunable, organic upconversion systems, such as those based on triplet-triplet annihilation, provide flexibility (compared with inorganics) in the use of different-Eg active layers [5, 6]. Materials known to demonstrate downconversion include Gd3+, Eu3+, and Er3+ coupled in the LiGdF4 host lattice, (Y,Yb)PO4:Tb3+, CaAl2O4:Yb3+, GdAl3(BO3)4:RE3+, Yb3+(RE ¼ Pr, Tb, Tm), YbxGd1 xAl3(BO3)4:Tb3+, YVO4:Yb3+, Er3+–Yb3+:Cs3Y2Br9, ZnO: Eu3+, LaVO4:Dy3+, GdVO4:Dy3+, and YF3:Pr3+ [4, 7]. Finally, singlet fission in conjugated organic molecules has also received attention as a possible route to exceed the single-junction limit. Bound electron-hole pairs in organic semiconductors, known as “excitons,” can occupy singlet (spin ¼ 0) or triplet (spin ¼ 1) states. In some organic semiconductors, when a photon is absorbed and creates a singlet exciton, the exciton can be rapidly converted into two lower-energy triplet excitons. The result is two electrons from a single photon. Studied singlet fission materials include pentacene and tetracene, which can be combined with appropriate acceptor molecules or coated onto other absorbers such as layers of inorganic quantum dots and silicon, resulting in EQEs >100% [8–11]. In these devices, the other absorber can absorb low-energy photons, whereas high-energy photons can be absorbed by the singlet fission sensitizer to produce two electrons and avoid thermalization losses in a manner similar to the impact ionization process shown in Fig. 6C.

10

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

2.3 Other Losses in Solar Cells The absorption spectrum of a solar cell dictates, to a large degree, the loss of energy in the cell. Any light that is transmitted or reflected is incident power lost. Once energy is absorbed as excited electrons, energy can be lost by thermalization of the carriers to the band edges, as discussed in Section 2.1, and also via the recombination of excited electrons and holes. In relation to Fig. 2 and Eq. (1), recombination reduces the photocurrent and voltage by increasing the dark current J0. Recombined carriers do not generate electrical power, and hence lower the PCE of the cell. Some mechanisms of recombination are unavoidable. Radiative recombination (spontaneous or optically stimulated) is always expected; however, it is balanced somewhat by reabsorption of the emitted light. Likewise, Auger recombination is unavoidable and expected to be significant in materials with small bandgaps and high carrier concentrations. The Auger process is similar to impact ionization discussed in the last section; however, in this case, two similar carriers interact, causing one to decay across the band gap and transfer its kinetic energy to the other (rather than exciting an additional carrier across the band gap, as in impact ionization). Nonradiative recombination arises from material imperfections. In nonradiative, Shockley Read Hall recombination, relaxation of a photogenerated electron occurs via a localized trap state. When an electron is captured by a trap state in the band gap, it may subsequently be released by thermal activation (“detrapping”). However, if the trap state also captures a hole before the electron is released, recombination occurs. Trap states located deep within the band gap have a higher probability of capturing both electrons and holes, and are therefore commonly referred to as recombination centers. Recombination can occur at defects within the grains of the active materials, at impurities, and at interfaces, including heterojunctions and grain boundaries. As the density of recombination centers increases, the lifetimes of electrons in the conduction band and holes in the valence band decrease (carriers are captured more quickly by the deep trap states). The charge carrier transport length and lifetime are thus critical parameters in solar cells. The carrier transport length should be longer than the thickness of the material needed for complete light absorption, to prevent recombination en route to the collection electrodes. Charge transport lengths are generally highest in uniform, crystalline materials free from defects, impurities, and grain boundaries.

3. Emerging Photovoltaic Materials and Photovoltaic Solar Cells Considerable effort has been expended in recent years to develop new photovoltaic materials and architectures. Photovoltaic silicon has a diamond cubic crystal structure, but a variety of different silicon phases and composites are known to exist and continue to be discovered, such as clathrates and nanostructures [12]. Nano-engineered forms of silicon can be tailored to improve the efficiency of solar energy conversion. Biomimetic

Chapter 1 • Fundamental Understanding of Solar Cells 11

photovoltaic materials have also received considerable attention, such as large protein complexes similar to those involved in photosynthesis [13]. Most recent research on emerging photovoltaics, however, has centered on four classes of devices: dye-sensitized solar cells (DSSCs), organic photovoltaics (OPV), colloidal quantum dot solar cells (CQDSCs), and perovskite solar cells (PSCs).

3.1 Dye-Sensitized Solar Cells Liquid-electrolyte dye-sensitized solar cells (DSSCs) were introduced by Gratzel and O’Regan in 1991 [14]. A transparent conducting electrode (TCE), typically fluorine-doped tin(IV)-oxide (FTO) on glass, is coated with a mesoporous layer of anatase titanium(IV) oxide (TiO2), as illustrated schematically in Fig. 7A. The mesoporous layer is sintered to produce a network of connected nanoparticles (approximately 20 nm in diameter) with high porosity (typically 50%–65%). The mesoporous film is coated with a monolayer of photosensitive dye, which acts as the light absorber. For 15–20 nm particles, this

FIG. 7 Device architectures of (A) dye-sensitized solar cells, (B) bulk-heterojunction solar cells, (C) colloidal quantum dot solar cells, and (D) perovskite solar cells.

12

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

mesoporous structure delivers a 100-fold enhancement in surface area per micron thickness, as compared with a flat film [15]. A surface area approximately 1000 times greater than the planar area ensures efficient solar harvesting, such that mesoporous film thicknesses are typically around 10 μm. Traditionally the mesoporous film is filled with an iodide/triiodide-based electrolyte and capped with a platinum counter electrode. After light-absorption by the dye, electron transfer takes place from the dye’s excited state into the conduction band of the electron-transport material (ETM; i.e., TiO2), and the electrons reach the TCE electrode via diffusion through the disordered nanoparticle structure. The iodide redox couple regenerates the oxidized dye molecules, with the positive charge being transported through the electrolyte to the counter electrode. These cells have demonstrated certified power conversion efficiencies up to 11.9% [16]. An advantage of DSSCs is the variety of suitable dyes that can be employed, covering a wide range of the solar spectrum. Liquid-electrolyte DSSCs are a relatively mature technology and are now commercially available from several suppliers. Recent research directions in DSSCs have included, but are not limited to, improving light harvesting through co-sensitization, light scattering, and increasing the electrode transparency; surface engineering through hydrogenation, protonation, and plasma treatments; novel DSSC structures including p-type DSSCs and tandem devices; stability improvements through the replacement of the liquid electrolyte and Pt catalyst; and cost reductions through the use of natural dyes and the replacement of FTO and Pt with graphene. More information about current research trends can be found in recent review articles, such as the one from Gong et al. [17].

3.2 Organic Photovoltaics Organic photovoltaic devices are composed of electron-transporting and holetransporting organic semiconductors. These materials have the potential to be cheaply produced in large volumes, and cells can be made flexible and lightweight for a variety of applications. Fullerene derivatives such as PCBM ([6,6]-phenyl-C61-butyric acid methyl ester), which are small conjugated molecules, have typically been employed as the electron transporter. Copolymers of polyfluorene, and poly(phenylenevinylene) and poly(thiophene) backbones such as P3HT (poly(3-hexylthiophene)), have generally been used as the hole-transporting materials. Unlike inorganic semiconductors where photoexcitation generally results in the production of free carriers, bound states of excited electrons and holes, “excitons,” are produced in organic semiconductors (with binding energy in the order of hundreds of meV [18–21]). Bound by Coulomb attraction, the excitons are strongly localized to individual polymer chains or molecules. Following photon absorption, excitons diffuse to the interface between the electron-transporting and holetransporting semiconductors. An offset in the energy levels of the two materials at the interface facilitates dissociation of the exciton into a free electron and hole. Following dissociation, the electrons and holes are transported through their respective organic layers to either an ETM or a hole-transport material (HTM) interface, and then to the electrodes.

Chapter 1 • Fundamental Understanding of Solar Cells 13

Charge transport in these materials results from carrier hopping between the molecules or chains, such that mobilities are lower than in inorganic equivalents. The exciton diffusion length, in particular, is typically reported to be on the order of 10 nm [22]. The efficiency of organic solar cells is therefore very sensitive to the morphology of the materials on the nanoscale. Interfacial distances need to be suitably small so that photogenerated excitons can be dissociated. Continuous pathways are also necessary in the individual phases to allow transport of the carriers to the electrodes. The active layer must also be suitably thick to absorb most of the incident light. These criteria are simultaneously satisfied by producing bulk heterojunctions (BHJs), where two organic semiconductors are blended in a common solvent to form an interpenetrating network of electron and hole-transporting materials, as illustrated in Fig. 7B. Bulk heterojunctions are typically synthesized by spin-coating the blended solution onto a conducting glass substrate with a thin electron-blocking layer (hole transport layer). Many groups have studied how processing conditions and subsequent treatments influence the performance of bulk heterojunction solar cells. Optimization of these parameters has resulted in a certified PCE of 11.5% for cells with an architecture of ITO/ZnO/polymer:PC71BM/V2O5/Al, with the polymer poly[(5,6-difluoro-2,1,3-ben000 0 0 00 00 000 000 zothiadiazol-4,7-diyl)-alt-(3,3 -di(2-nonyltridecyl)-2,2 ;5 ,2 ;5 ,2 -quaterthiophen-5,5 diyl)] (PffBT4T-C9C13) [23]. Here ITO refers to a thin indium tin oxide layer. Organic photovoltaics are widely observed to degrade over time, and research efforts are ongoing to improve their lifetime to a timeframe suitable for commercial applications. Improvements in encapsulation and increased glass transition temperatures for the organic materials, among other possible improvements, have been identified and are summarized in a review by McGehee et al. [24].

3.3 Colloidal Quantum Dot Solar Cells Semiconducting quantum dots offer important advantages over conventional bulk materials such as solution-processability, low-temperature fabrication, and quantum sizeeffect tunability of their band gap, which allows their absorption properties to be optimized for photovoltaic applications. The most effective solar cell architecture employing quantum dots has been the colloidal quantum dot solar cell (CQDSC), which is shown schematically in Fig. 7C. The CQDSC is typically based on a heterojunction between an n-type material, such as TiO2 or ZnO, and a layer of p-type quantum dots, such as those based on lead chalcogenides like PbSe and PbS, and more recently perovskite materials, which are discussed further in the next section. The n-type metal oxide is deposited on conducting glass (typically FTO or ITO), and a high work function electrode (typically coated with a HTM) is placed on top. Photons are absorbed by the quantum dots, and electrons and holes are transported by drift and diffusion to the collection electrodes. A certified efficiency of 13.4% has been obtained with a cell employing CsPbI3 QDs and the following architecture: FTO/TiO2/CsPbI3 QDs/Spiro-OMeTAD/MoOx/Al [25], where Spiro-OMeTAD (2,20 ,7,70 -tetrakis-[N,N0 -di-p-methoxyphenylamino]-9,90 -spirobifluorene)

14

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

and MoOx are HTM layers. Recent progress in colloidal quantum dot solar cells, and related quantum dot sensitized solar cells, is discussed in a review article by Wei et al. [26].

3.4 Perovskite Solar Cells The solar cells that have received the most attention recently are those based on metal halide perovskite absorbers. Unlike almost all other types of solar cells, perovskite devices can be cost-effective, efficient, flexible, and lightweight at the same time. These properties have let PSCs become a promising candidate to either compete with silicon solar cell technology or to be coupled with it to produce multijunction devices. A perovskite has ABX3 composition, where A, B, and X are a monovalent cation, a divalent cation, and a monovalent anion, respectively. In the perovskite structure, A+ cations occupy the void spaces in the BX 3 inorganic framework. The Goldschmidt tolerance factor determines if a combination of cations and anions may produce a perovskite structure, rA + rX tf ¼ pffiffiffi 2 ðrB + rX Þ

(6)

where rA, rB, and rX are the ionic radius of the corresponding ions. A typical tolerance factor between 0.9 and 1 is favorable and representative of a cubic and compact phase of perovskite. The first perovskite composition used in PSCs was CH3NH3PbI3. Compositional engineering has shown that either of the three ions may be completely or partially replaced to achieve a similar or higher device efficiency. Until now, the highest-efficiency devices have usually been fabricated using the combination of CH3NH3, formamidinium (FA), and Cs+ as the A-site cation, and I and Br as the X-site anion [27]. Similar to OPVs or CQDSCs, a perovskite active layer is sandwiched between ETM and HTM layers as shown in Fig. 7D. Therefore, either of the n-i-p or p-i-n configurations are possible for PSCs, where n, i, and p are representative of n-doped ETM, intrinsic perovskite, and p-doped HTM, respectively. One of the outstanding properties of metal halide perovskites is their low exciton binding energies (i.e., 2–12 meV) [28, 29]. The thermal energy of excitons at room temperature is kBT  26 meV, where kB is the Boltzmann constant and T is the temperature. Hence, the photoexcited electrons and holes can easily be dissociated into free carriers inside the perovskite layer. As a result, PSCs are not highly dependent on the large surface area in the TiO2 layer of DSSCs or the bulk heterojunctions of OPVs. Improving efficiency and stability are two important strategies to develop the PSC technology. In terms of efficiency, PSCs have already reached their maximum theoretical values of Jsc and FF. Therefore, future research should be focused on improving the Voc. A general approach to this challenge is decreasing the density of trap states. Perovskites are known to suffer from iodide migration, a phenomenon that occurs when they are illuminated or electrically biased. The iodide migration leads to hysteresis of J-V curves, when they are scanned in reverse and forward directions, and to poor device stability in the long term. Providing extra iodide atoms while fabricating the perovskite layer was suggested as a solution to mitigate the iodide migration [30]. This has led to achieving one of the highest

Chapter 1 • Fundamental Understanding of Solar Cells 15

certified efficiencies (22.1%) reported to date. It had a device architecture of FTO/compact TiO2/mesoporous TiO2/(FA)1 xCsxPbI3 yBry/Poly(triarylamine)/Au. Although the high efficiencies of PSCs are very close to those of silicon solar cells, their stability needs to be improved before commercialization. PSCs are susceptible to exposure to humidity [31] or O2 while under illumination [32]. Research on improving the efficiency or the stability of PSCs has been based on modifications of the composition [33] or fabrication procedures for either the perovskite or other layers. A discussion about the efficiency and stability of PSCs can be found in a review article by Correa-Baena et al. [34].

3.5 Other New Materials and Structures The development of new photovoltaic materials and cells extends well beyond the DSSCs, OPV, CQDSCs, and PSCs discussed in this section. Novel materials and designs for photon converters, tandem cells, and hot carrier cells, among others, continue to be developed to address the limitations discussed in Section 2. Oxides are used in virtually every emerging PV architecture of Fig. 7, particularly as selective ETM (e.g., TiO2, SnO2, and ZnO) and HTM (e.g., MoO3, NiO, and WO3) layers. New methods, such as atmospheric pressure spatial atomic layer deposition (AP-SALD), are being developed to finely tune the properties of these oxides and manufacture them in a scalable manner. Novel nanostructures are also being developed for better light management and superior electrical contacts. Promising TCEs include metal nanowire networks and two-dimensional materials such as graphene, among others. This text will provide an overview of a variety of these advanced microscale and nanoscale photovoltaic technologies currently under development.

References [1] W. Shockley, H.J. Queisser, Detailed balance limit of efficiency of p-n junction solar cells, J. Appl. Phys. 32 (1961) 510–519. [2] J. Nelson, The Physics of Solar Cells, Imperial College Press, London, 2003. [3] M.A. Green, Y. Hishikawa, E.D. Dunlop, D.H. Levi, J. Hohl-Ebinger, A.W.Y. Ho-Baillie, Solar cell efficiency tables (version 51), Prog. Photovolt. Res. Appl. 26 (2018) 3–12. [4] D. Verma, T.O. Saetre, O.M. Midtga˚rd, Review on up/down conversion materials for solar cell application, in: 38th IEEE Photovoltaic Specialists Conference, 2012, pp. 002608–002613. [5] L. Frazer, J.K. Gallaher, T.W. Schmidt, Optimizing the efficiency of solar photon upconversion, ACS Energy Lett. 2 (2017) 1346–1354. [6] K. Poorkazem, A.V. Hesketh, T.L. Kelly, Plasmon-enhanced triplet–triplet annihilation using silver nanoplates, J. Phys. Chem. C 118 (2014) 6398–6404. [7] T. Fix, H. Rinnert, M.G. Blamire, A. Slaoui, J.L. MacManus-Driscoll, Nd:SrTiO3 thin films as photon downshifting layers for photovoltaics, Sol. Energy Mater. Sol. Cells 102 (2012) 71–74. [8] M.W.B. Wilson, A. Rao, B. Ehrler, R.H. Friend, Singlet exciton fission in polycrystalline pentacene: from photophysics toward devices, Acc. Chem. Res. 46 (2013) 1330–1338. [9] B. Ehrler, M.W.B. Wilson, A. Rao, R.H. Friend, N.C. Greenham, Singlet exciton fission-sensitized infrared quantum dot solar cells, Nano Lett. 12 (2012) 1053–1057.

16

ADVANCED MICRO- AND NANOMATERIALS FOR PHOTOVOLTAICS

[10] A. Rao, R.H. Friend, Harnessing singlet exciton fission to break the Shockley–Queisser limit, Nat. Rev. Mater. 2 (2017) 17063. [11] J. Xia, S.N. Sanders, W. Cheng, J.Z. Low, J. Liu, L.M. Campos, T. Sun, Singlet fission: progress and prospects in solar cells, Adv. Mater. 29 (2017) 1601652. €ro € s, G. Galli, Novel silicon phases and nanostructures for solar energy [12] S. Wippermann, Y. He, M. Vo conversion, Appl. Phys. Rev. 3 (2016) 040807. [13] P.I. Gordiichuk, G.-J.A.H. Wetzelaer, D. Rimmerman, A. Gruszka, J.W. de Vries, M. Saller, D.A. Gautier, S. Catarci, D. Pesce, S. Richter, P.W.M. Blom, A. Herrmann, Solid-state biophotovoltaic cells containing photosystem I, Adv. Mater. 26 (2014) 4863–4869. [14] B. O’Regan, M. Gratzel, A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2 films, Nature 353 (1991) 737–740. [15] M. Graetzel, Dye-sensitized solid-state hetereojunction solar cells, MRS Bull. 30 (2005) 23. [16] R. Komiya, A. Fukui, N. Murofushi, N. Koide, R. Yamanaka, H. Katayama, Improvement of the conversion efficiency of a monolithic type dye-sensitized solar cell module, in: Technical Digest, 21st International Photovoltaic Science and Engineering Conference, Fukuoka, 2011. 2C-5O-08. [17] J. Gong, K. Sumathy, Q. Qiao, Z. Zhou, Review on dye-sensitized solar cells (DSSCs): advanced techniques and research trends, Renew. Sust. Energ. Rev. 68 (2017) 234–246. [18] S. Kraner, R. Scholz, C. Koerner, K. Leo, Design proposals for organic materials exhibiting a low exciton binding energy, J. Phys. Chem. C 119 (2015) 22820–22825. [19] H.-W. Li, Z. Guan, Y. Cheng, T. Lui, Q. Yang, C.-S. Lee, S. Chen, S.-W. Tsang, On the study of exciton binding energy with direct charge generation in photovoltaic polymers, Adv. Electron. Mater. 2 (2016) 1600200. [20] M. Knupfer, Exciton binding energies in organic semiconductors, Appl. Phys. A Mater. Sci. Process. 77 (2003) 623–626. [21] B.-G. Kim, C.-G. Zhen, E.J. Jeong, J. Kieffer, J. Kim, Organic dye design tools for efficient photocurrent generation in dye-sensitized solar cells: exciton binding energy and electron acceptors, Adv. Funct. Mater. 22 (2012) 1606–1612. [22] A.C. Mayer, S.R. Scully, B.E. Hardin, M.W. Rowell, M.D. McGehee, Polymer-based solar cells, Mater. Today 10 (2007) 28–33. [23] J. Zhao, Y. Li, G. Yang, K. Jiang, H. Lin, H. Ade, W. Ma, H. Yan, Efficient organic solar cells processed from hydrocarbon solvents, Nat. Energy 1 (2016) 15027. [24] W.R. Mateker, M.D. McGehee, Progress in understanding degradation mechanisms and improving stability in organic photovoltaics, Adv. Mater. 29 (2017) 160394. [25] E.M. Sanehira, A.R. Marshall, J.A. Christians, S.P. Harvey, P.N. Ciesielski, L.M. Wheeler, P. Schulz, L.Y. Lin, M.C. Beard, J.M. Luther, Enhanced mobility CsPbI3 quantum dot arrays for record-efficiency, high-voltage photovoltaic cells, Sci. Adv. 3 (2017). eaao4204. [26] W. Huiyun, L. Dongmei, Z. Xinhe, M. Qingbo, Recent progress of colloidal quantum dot based solar cells, Chin. Physics B 27 (2018) 018808. [27] M. Saliba, T. Matsui, J.-Y. Seo, K. Domanski, J.-P. Correa-Baena, M.K. Nazeeruddin, S.M. Zakeeruddin, W. Tress, A. Abate, A. Hagfeldt, M. Gratzel, Cesium-containing triple cation perovskite solar cells: improved stability, reproducibility and high efficiency, Energy Environ. Sci. 9 (2016) 1989–1997. [28] Q. Lin, A. Armin, R.C.R. Nagiri, P.L. Burn, P. Meredith, Electro-optics of perovskite solar cells, Nat. Photonics 9 (2015) 106–112. [29] A.M. Soufiani, F. Huang, P. Reece, R. Sheng, A. Ho-Baillie, M.A. Green, Polaronic exciton binding energy in iodide and bromide organic-inorganic lead halide perovskites, Appl. Phys. Lett. 107 (2015) 231902.

Chapter 1 • Fundamental Understanding of Solar Cells 17

[30] W.S. Yang, B.-W. Park, E.H. Jung, N.J. Jeon, Y.C. Kim, D.U. Lee, S.S. Shin, J. Seo, E.K. Kim, J.H. Noh, S. I. Seok, Iodide management in formamidinium-lead-halide–based perovskite layers for efficient solar cells, Science 356 (2017) 1376–1379. [31] J. Yang, B.D. Siempelkamp, D. Liu, T.L. Kelly, Investigation of CH3NH3PbI3 degradation rates and mechanisms in controlled humidity environments using in situ techniques, ACS Nano 9 (2015) 1955–1963. [32] D. Bryant, N. Aristidou, S. Pont, I. Sanchez-Molina, T. Chotchunangatchaval, S. Wheeler, J. R. Durrant, S.A. Haque, Light and oxygen induced degradation limits the operational stability of methylammonium lead triiodide perovskite solar cells, Energy Environ. Sci. 9 (2016) 1655–1660. [33] K. Poorkazem, T.L. Kelly, Compositional engineering to improve the stability of lead halide perovskites: a comparative study of cationic and anionic dopants, ACS Appl. Energy Mater. 1 (2018) 181–190. €tzel, A. Abate, W. Tress, A. Hagfeldt, Promises and [34] J.-P. Correa-Baena, M. Saliba, T. Buonassisi, M. Gra challenges of perovskite solar cells, Science 358 (2017) 739–744.