RESEARCH ◥ losses. In contrast, thin-film single-crystalline REVIEW SUMMARY GaAs cells (28.8%) show only minimal recom- bination losses but can be improved by better light management. Polycrystalline CdTe thin- PHOTOVOLTAICS film cells (21.5%) offer excellent light absorption but have relatively high recombination losses; perovskite cells (21.0%) and Cu(In,Ga)(Se,S)2 Photovoltaic materials: Present (CIGS) cells (21.7%) have poorer light manage- ment, although CIGS displays higher electrical efficiencies and future challenges quality. Aside from these five materials (Si, GaAs, CdTe, Albert Polman,* Mark Knight, Erik C. Garnett, Bruno Ehrler, Wim C. Sinke CIGS, perovskite) with efficiencies of >20%, a broad range of other thin-film materials have been developed with efficiencies of 10 to 12%: BACKGROUND: Photovoltaics, which directly tries and illuminated under the standard AM1.5 micro/nanocrystalline and amorphous Si, Cu convert solar energy into electricity, offer a solar spectrum, and compare these to the fun- ◥ (Zn,Sn)(Se,S)2 (CZTS), dye- practical and sustainable solution to the chal- damental limits based on the S-Q model. Cells ON OUR WEBSITE sensitized TiO , organic J 2 lenge of meeting the increasing global energy that show a short-circuit current ( sc)lower polymer materials, and demand. According to the Shockley-Queisser than the S-Q limit suffer from incomplete light Read the full article at http://dx.doi. quantum dot solids. So far, (S-Q) detailed-balance model, the limiting absorption or incomplete collection of gener- org/10.1126/ cell designs based on these photovoltaic energy conversion efficiency for ated carriers, whereas a reduced open-circuit science.aad4424 materials all suffer from V FF .................................................. a single-junction solar cell is 33.7%, for an voltage ( oc)orfillfactor( )reflectsunwanted both light management optimum semiconductor band gap of 1.34 eV. bulk or interfacial carrier recombination, para- and carrier management problems. Organic Parallel to the development of wafer-based Si sitic resistance, or other electrical nonideal- and quantum dot solar cells have shown sub- solar cells, for which the record efficiency has ities. The figure shows the experimental values stantial efficiency improvements in recent years. on April 23, 2017 continually increased during recent decades, for Jsc and the Voc × FF product relative to the a large range of thin-film materials have been S-Q limiting values for the different materials. OUTLOOK: The record-efficiency single- developed with the aim to approach the S-Q This graph enables a direct identification of crystalline materials (Si, GaAs) have room limit. These materials can potentially be de- each material in terms of unoptimized light man- for efficiency improvements by a few abso- posited at low cost, in flexible geometries, and agement and carrier collection (Jsc/JSQ <1)or lute percent. The future will tell whether the using relatively small material quantities. carrier management (Voc × FF/VSQ × FFSQ <1). high-efficiency polycrystalline thin films (CdTe, Monocrystalline Si cells (record efficien- CIGS, perovskite) can rival the efficiencies of ADVANCES: We review the electrical charac- cy 25.6%) have reached near-complete light Si and GaAs. Because the cost of photovoltaic teristics of record-efficiency cells made from 16 trapping and carrier collection and are mostly systems is only partly determined by the cost widely studied photovoltaic material geome- limited by remaining carrier recombination of the solar cells, efficiency is a key driver to reduce the cost of solar energy, and therefore large-area photovoltaic systems require high- http://science.sciencemag.org/ efficiency (>20%), low-cost solar cells. The lower-efficiency (flexible) materials can find applications in building-integrated PV systems, flexible electronics, flexible power generation systems, and many other (sometimes niche) markets. High-efficiency (>20%) materials find applications in large-area photovoltaic power generation for the utility grid as well as in small Downloaded from andmedium-sizedsystemsforthebuiltenvi- ronment. They will enable very large-scale pen- etration into our energy system, starting now and growing as the cost per kilowatt-hour is reduced further by a factor of 2 to 3. This can be achieved by nanophotonic cell designs, in which optically resonant and nonresonant struc- tures are integrated with the solar cell archi- tecture to enhance light coupling and trapping, in combination with continued materials engi- neering to further optimize cell voltage. Making Limiting processes in photovoltaic materials. An efficient solar cell captures and traps all big steps forward in these areas will require a incident light (“light management”) and converts it to electrical carriers that are efficiently collected coordinated international materials science and (“carrier management”). The plot shows the short-circuit current and product of open-circuit engineering effort.▪ voltage and fill factor relative to the maximum achievable values, based on the Shockley-Queisser detailed-balance limit, for the most efficient solar cell made with each photovoltaic material. The data indicate whether a particular material requires better light management, carrier management, The list of author affiliations is available in the full article online. h *Corresponding author. E-mail: [email protected] or both. Colors correspond to cells achieving <50% of their S-Q efficiency limit SQ (red), 50 to 75% Cite this article as A. Polman et al., Science 352,aad4424 (green), or >75% (blue). (2016). DOI: 10.1126/science.aad4424 SCIENCE sciencemag.org 15 APRIL 2016 • VOL 352 ISSUE 6283 307 RESEARCH ◥ substantially lower than the S-Q limit for a given REVIEW band gap. Ideal and record-efficiency solar PHOTOVOLTAICS cells compared We distinguish three classes of PV materials: (i) ultrahigh-efficiency monocrystalline materials with Photovoltaic materials: Present efficiencies of >75% of the S-Q limit for the corre- sponding band gap: Si (homojunction and hetero- junction), GaAs, and GaInP; (ii) high-efficiency efficiencies and future challenges multi- and polycrystalline materials (50 to 75% “ ” 1 1 1 1 1,2 of the S-Q limit): Si, Cu(In,Ga)(Se,S)2 ( CIGS ), Albert Polman, * Mark Knight, Erik C. Garnett, Bruno Ehrler, Wim C. Sinke CdTe, methyl ammonium lead halide perovskite [CH3NH3Pd(I,Cl,Br)3], and InP; and (iii) low- Recent developments in photovoltaic materials have led to continual improvements efficiency materials (<50% of the S-Q limit): micro- in their efficiency. We review the electrical characteristics of 16 widely studied geometries or nanocrystalline and amorphous Si, Cu(Zn,Sn) of photovoltaic materials with efficiencies of 10 to 29%. Comparison of these characteristics (Se,S)2 (“CZTS”), dye-sensitized TiO2, organic and to the fundamental limits based on the Shockley-Queisser detailed-balance model provides polymer materials, and quantum dot materials. a basis for identifying the key limiting factors, related to efficient light management and The record efficiency for each of these mate- charge carrier collection, for these materials. Prospects for practical application and large-area rials is plotted in Fig. 1B (see also table S1). The fabrication are discussed for each material. experimental values for Jsc, Voc, and FF for the record-efficiency cell reported for each individual material are shown in Fig. 2, A to C, together hotovoltaics (PV), which directly convert to electrical energy because of thermalization of with the limiting values calculated using the S-Q solar energy into electricity, offer a practical charge carriers (Fig. 1A, inset). Taking these two model (2). The experimental values for Jsc gener- and sustainable solution to the challenge factors into account, ∼45% of the incident spectrum- ally follow the trend given by the S-Q limit, with of meeting the increasing global energy integrated solar power remains for semiconductors some materials closely approaching this limit. on April 23, 2017 P FF demand. In recent years, the decreasing with a band gap of 1.1 to 1.4 eV. This is the max- Values for Voc and are much more scattered, price of PV systems has levelized the cost of PV- imum power that would be generated if the cell with only a few materials approaching the S-Q produced electricity to the point that it can now were operated at a voltage corresponding to the limit. To analyze these trends, we evaluated two compete with the variable portion of consumer band gap energy and a current corresponding to characteristic parameters for each material: (i) the electricity prices in many countries worldwide: full capture of all photons with energy above the current ratio j = Jsc/JSQ, which indicates the de- The point of “socket parity” has been reached (1). band gap, followed by full collection of all gen- gree of light coupling, absorption, and trapping Substantial further cost reduction is needed, how- erated carriers. in the active layer(s) of the cell, and also depends ever, to allow PV to compete in more electricity Even in an ideal case, however, the open-circuit on the carrier collection efficiency; and (ii) the volt- markets and to enter the multi-terawatt regime. voltage Voc is always lower than the band gap age ratio v = Voc/VSQ, which is primarily related Aside from the solar cell and module fabrication energy because thermodynamic detailed balance to the degree of recombination of carriers in the costs, a major and increasing fraction of the cost requires the cell to be in equilibrium with its en- bulk, surfaces, and interfaces. Together, the voltage http://science.sciencemag.org/ of PV generation (typically 50%) is related to com- vironment, which implies that there is spontane- ratio v and fill factor ratio f = FF/FFSQ indicate the ponent and installation requirements such as in- ous light emission from the cell. The corresponding total electrical limitations of a cell (6). A plot of verters, cabling, mounting structures, and labor radiative carrier recombination represents a dark j versus v×ffor all evaluated materials (Fig.