Home / Chroniques / Photovoltaics: new materials for better efficiency
π Energy π Science and technology

Photovoltaics: new materials for better efficiency

Pere Roca i Cabarrocas
CNRS Research Director at LPICM* at École Polytechnique (IP Paris) and scientific director of Institut Photovoltaïque d'Île-de-France (IPVF)
Key takeaways
  • The global solar energy market today is 95% silicon-based – although, silicon is not actually the most ideal material for photovoltaic panels because it does not absorb light very well.
  • Researchers are looking at alternatives such as thin-film solar cell technology and perovskites.
  • Perovskites have now reached the same level of performance as silicon (with an energy conversion of more than 25%), but they are unstable, a problem that still needs to be resolved.
  • Photovoltaics in tandem (silicon and perovskites together) are promising since their efficiency can exceed 30%.

Solar pho­to­volta­ic (PV) tech­nol­o­gy has grown almost expo­nen­tial­ly over the past 15 years and is now cost-com­pet­i­tive with fos­sil fuels. Remark­ably, the under­ly­ing PV struc­ture has remained vir­tu­al­ly unchanged since its devel­op­ment at Bell Lab­o­ra­to­ries in 1954. Indeed, mod­ern solar cells are still based on a sim­ple junc­tion between ‘n’ type (elec­tron-rich) sil­i­con and ‘p’ type (hole-rich) sil­i­con. The first solar cells con­vert­ed sun­light into elec­tric­i­ty with an effi­cien­cy of about 5%, a fig­ure that has soared to over 25% in recent years thanks to more sophis­ti­cat­ed cell design, name­ly the addi­tion of high­ly doped sil­i­con and anti-reflec­tion layers.

Although sil­i­con still accounts for about 95% of the glob­al solar ener­gy mar­ket, it has one major draw­back: it does not absorb light very well. Large thick­ness­es of mate­r­i­al – to the order of hun­dreds of microns – are there­fore required. Since this is a long dis­tance for elec­trons to trav­el, PV-grade sil­i­con must be high­ly crys­talline and very pure for the charge car­ri­ers to effi­cient­ly pass through. Fab­ri­cat­ing such mate­r­i­al is com­plex, how­ev­er, and there­fore expensive.

Changing our approach to silicon

To reduce pro­duc­tion costs and the amount of mate­r­i­al need­ed, researchers have long been look­ing into alter­na­tive mate­ri­als. My team is focus­ing on thin-film solar cell tech­nol­o­gy, so-called because the films only need to be a few microns thick for suf­fi­cient opti­cal absorp­tion. Low­er qual­i­ty, low­er puri­ty mate­ri­als are also accept­able and they can be fab­ri­cat­ed using rapid depo­si­tion meth­ods: evap­o­ra­tion, direct-to-glass sput­ter­ing or plas­ma enhanced chem­i­cal vapour depo­si­tion (PECVD). These mate­ri­als, which include hydro­genat­ed amor­phous sil­i­con, cad­mi­um tel­luride (CdTe) and cop­per indi­um gal­li­um selenide (CuIn1-xGaxSe2, or CIGS for short), make for very effi­cient cells and can be grown on any type of substrate.

The effi­cien­cy of solar cells that con­vert sun­light into elec­tric­i­ty has increased from 5% to 25% in recent years.

Today, crys­talline sil­i­con wafers are con­ven­tion­al­ly fab­ri­cat­ed by draw­ing ingots and then cut­ting them into wafers about 180 µm thick. We are try­ing to break new ground in the way crys­talline sil­i­con is made by using new growth tech­niques that rely on low-tem­per­a­ture PECVD process­es – that is, between 150 and 300 degrees Cel­sius. We are also using this tech­nique to make ‘III‑V’ mate­ri­als which, although wide­ly used in opto­elec­tron­ics, are about 100 times more expen­sive than crys­talline sil­i­con. In the world of pho­to­voltaics, costs must be reduced to com­pete with crys­talline silicon.

The stan­dard meth­ods for cre­at­ing III‑V mate­ri­als are mol­e­c­u­lar beam epi­taxy (MBE) and met­al organ­ic chem­i­cal vapour decom­po­si­tion (MOCVD). These epi­tax­i­al growth meth­ods require ultra-high vac­u­um envi­ron­ments for MBE and high tem­per­a­tures (700‑1000°C) for MOCVD, mak­ing them expen­sive. The plas­ma depo­si­tion process­es that we are devel­op­ing at LPICMt in col­lab­o­ra­tion with the Insti­tut Pho­to­voltaïque d’Ile de France (IPVF) aim to reduce this cost. This is one of the last vari­ables we have any con­trol over as III‑V com­pounds are already at the max­i­mum of their effi­cien­cy when it comes to con­vert­ing solar radi­a­tion into electricity.

Perovskites: a new material

Per­ovskites are anoth­er class of mate­ri­als we are work­ing on. These are crys­talline mate­ri­als with the struc­ture ABX3, where A is cae­sium, methy­lam­mo­ni­um (MA) or for­mami­dini­um (FA), B is lead or tin and X is chlo­rine, bromine or iodine. They are promis­ing can­di­dates for thin-film solar cells because they can absorb light over a wide range of wave­lengths of the solar spec­trum thanks to their tune­able elec­tron­ic band gaps 1. Charge car­ri­ers (elec­trons and holes) can also dif­fuse quick­ly and over long dis­tances. These prop­er­ties mean that per­ovskite solar cells now boast an ener­gy con­ver­sion effi­cien­cy of over 25%, putting their per­for­mance on a par with estab­lished solar cell mate­ri­als such as sil­i­con, gal­li­um arsenide and cad­mi­um telluride.

While we know how to make them cheap­ly and effi­cient­ly, the prob­lem is that per­ovskites con­tain nat­ur­al sur­face defects and suf­fer from struc­tur­al changes known as ion migra­tion. Both of these fac­tors tend to make per­ovskite films unsta­ble and these insta­bil­i­ties become even more pro­nounced in the pres­ence of mois­ture and high­er tem­per­a­tures. To improve their sta­bil­i­ty, we need to under­stand these mate­ri­als and the inter­faces between the dif­fer­ent com­po­nents that make up the solar cell.

This will be a chal­lenge, but it is worth it, because per­ovskites are very ver­sa­tile: their opto­elec­tron­ic prop­er­ties can be manip­u­lat­ed quite eas­i­ly by sim­ple chem­i­cal mod­i­fi­ca­tions. Thanks to their incred­i­ble light absorp­tion capac­i­ty, they can be used not only in solar cells, but also in light-emit­ting diodes and oth­er elec­tron­ic appli­ca­tions. Research on per­ovskites is boom­ing and thou­sands of stud­ies are pub­lished every year.

Tandem cells 

The next ques­tion is: how do we go beyond cur­rent effi­cien­cies? While opti­mis­ing mate­ri­als and inter­faces is cru­cial, per­ovskites can also be added to estab­lished solar cell tech­nolo­gies (such as sil­i­con) to build so-called tan­dem solar cells. This is the sub­ject of research at the IPVF and it is an extreme­ly inter­est­ing way to increase the over­all effi­cien­cy of devices. Sil­i­con-only- and per­ovskite-only cells can both achieve effi­cien­cies of 26%, but if you put them togeth­er you can push the effi­cien­cy to a high­er val­ue (to beyond 30%). High­er effi­cien­cies mean, for exam­ple, that you can cov­er a small­er area with your PV pan­el to get the same ener­gy out­put – in oth­er words, it costs less.

Cells made sole­ly of sil­i­con or per­ovskite can achieve effi­cien­cies of 26%, but togeth­er this val­ue can exceed 30%.

So, what are the best mate­ri­als? If we are able to solve the sta­bil­i­ty of per­ovskites, they seem the most promis­ing. III-Vs are also inter­est­ing, but we need to reduce their cost. To address cli­mate change, our chal­lenge is to devel­op ter­awatts of pho­to­volta­ic pan­els, which means man­u­fac­tur­ing large quan­ti­ties of pho­to­volta­ic pan­els that require large instal­la­tion areas. Increas­ing their effi­cien­cy while decreas­ing the thick­ness of the cells is the best way to reduce the amount of mate­r­i­al used.

There are also oth­er prob­lems to be solved, such as recy­cling the pho­to­volta­ic mate­ri­als and keep­ing them free of dust so that they can con­tin­ue to absorb solar radi­a­tion effi­cient­ly. We are work­ing on the eco-design of solar cells that can be recy­cled to recov­er the con­stituent mate­ri­als. Pho­to­volta­ic plants are in fact pre­cious met­al ‘mines’. Per­ovskites also con­tain lead, which is tox­ic and could leach out of a cell in the event of flood­ing or fire. This aspect of PV tech­nol­o­gy is a research top­ic in itself and could be the sub­ject of a future article.

Interview by Isabelle Dumé 
1The ener­gy ranges between the valence band and the con­duc­tion band where elec­tron­ic states are for­bid­den.

Our world explained with science. Every week, in your inbox.

Get the newsletter