Invisible conductors - coating to reduce costs of PV cells
A new high-technology coating has the potential to signiﬁcantly reduce the manufacturing costs of new-generation solar photovoltaic cells and other optoelectronic devices. Melanie Rutherford talks to researchers at the University of Oxford and their colleagues at its technology company, Isis Innovation Ltd.
It’s hard to remember life before smart phones, tablets and high-definition LCD televisions. While the development of modern display applications has experienced a relatively recent surge, the organic LEDs (OLEDs) used by many mobile phones and, more recently, large-screen displays, are actually based on advances in small molecule dyes pioneered by Kodak in the USA back in the 1960s and 1970s. While technology has come a long way since then, the theory remains the same.
Photovoltaic (PV) cells, also known as solar cells, convert light into electrical current via a semiconducting material. When a photon is absorbed by this material, an electron is knocked loose from the atom that ‘excites’ an electron in the crystal lattice that would normally be tightly bound by covalent bonds between surrounding atoms. This allows it to move freely within the semiconductor, leaving its previous covalent bond short of one electron – known as a hole. This hole allows space for neighbouring electrons to free themselves of their respective atoms, which in turn creates holes into which other electrons can move. As a result, the hole can move through the lattice. Intrinsic to this is the semiconducting material, which creates these mobile electron-hole pairs. Different semiconductor materials have different properties, and improving existing materials and finding new ones is of great interest in materials science.
Professor Peter Edwards of the University of Oxford, UK, says, ‘An intrinsic and very low mobilities of both electrons and holes. On the positive side, this means that the probability of light-emitting recombination is high, so they exhibit good properties for light emission. The principle problem appears to be the lifetime of such devices and the complexity of circuitry, as the efficiency of different colours decreases at different rates over time.’
An example is blue OLEDs that, when used for flat-panel displays, have a lifetime of around 14,000 hours until brightness is reduced by half. Furthermore, the material used to produce blue light degrades at a faster rate than that used for other colours, the resulting change in colour balance more noticeable than an overall decrease in brightness.
‘One has to recall, of course, that electron and hole production in organic-based materials in terms of solidstate physics corresponds in chemistry terms to radical anion–cation production,’ says Dr Vladmir Kuznetsov of the University of Oxford. ‘Radical anions and cations are known for their high chemical reactivity and, therefore, their high chemical instability.’
Zinc oxide (ZnO) is a semiconductor noted for its transparency, high electron mobility, wide bandgap and strong luminescence at room temperature. As such it is a common transparent conducting oxide component of thin films used for PV applications. Doping a semiconductor with impurities to modify its electronic properties is a technique widely used to enhance the electronic properties of these materials. ‘We have found that silicon dioxide (SiO2) acts as an effective n-type dopant to ZnO thin films prepared by both vacuum deposition and solution-phase deposition techniques,’ explains Edwards. ‘The doping effect arises due to a substitution of silicon for zinc in the ZnO host crystal structure, which leads to a high carrier concentration and electrical conductivity of thin films. As a result, the electrical properties of silicon-doped ZnO (ZnO:Si) significantly exceed those typical for aluminium or indium-doped ZnO films prepared by solution routes.’
A research team at the University of Oxford is developing two approaches for depositing the PV coating. The first is a vacuum-phase deposition process (also known as sputtering), involving a pressed target of ZnO:Si for use in magnetron sputtering equipment. ‘Vacuum-deposited ZnO:Si offers a direct replacement for sputtered indium tin oxide (ITO, or tin-doped indium oxide) at a lower materials cost,’ says Ferguson. ‘This should result in manufacturers being more competitive and a reduction in the price of consumer products that have a large transparent conducting oxide (TCO) component.’
The second approach is an ambient pressure approach using liquid precursors and spray pyrolysis. ‘Precursors may be deposited as an aerosol over a large area substrate, which is heated to decompose the precursors into the desired ZnO:Si thin films. The spray pyrolysis technique allows large areas to be covered – a scalable and relatively low-cost process compared to vacuum techniques. ‘The solution-phase atmospheric pressure approach would lend itself to wide-area use, as the spraying process is readily scalable,’ Kuznetsov adds. ‘Solution phase processes would also work on web production systems. We regard these developments as being important to industries such as third-generation solar cells and plastic electronics, with the drive to fabricate high-performance devices on flexible substrates and decrease manufacturing costs.’
Solution phase TCO also allows the use of non-vacuum processing equipment for liquid phase coating – not only enabling coverage of large areas at lower cost, but also offering potential for web-based processing, which is essential for low-cost plastic electronics.
‘In terms of transmittance, our silicon doped zinc oxide (ZnO:SiO2) materials compare very favourably to indium tin oxide (In2O3:SnO2, ITO),’ says Edwards (see graph above). ‘Note the optical cut off, which signifies comparable optical transparency of our SiO2 doped ZnO thin film with ITO, and indeed conventional, non-conducting glass itself.’
Optical transmittance of ZnO:Si films is around 85% over the 400nm–800nm range, a transmittance value that also includes the absorption by a glass substrate.
Other transparent coatings have recently been developed by scientists at the University of California, USA, derived from combinations of ultra-fine silver wires, carbon nanotubes and graphene within the transparent conducting composite. While electrons can move faster through graphene than they can through silicon, Edwards highlights the relative expense and availability of these materials, ‘which also demand complete control of quality and reproducibility,’ he says.
‘Silver, also expensive and a rare metal, has some similarities with indium as a component in a TCO device, hence our initial drive away from indium-based TCO materials. Silver will also degrade in the presence of atmospheric sulphur to form silver sulphide (Ag2S) as a black semiconducting material. These cost and stability issues are some of the reasons why silver is not used in conventional solar cells, which favour gold or steel alloys. It is for the same reasons that silver is never used for electronic wiring, despite having higher molar conductivity than copper.’
On the circuit
‘The use of organic devices as electronic circuit elements also has the intrinsic problem that carrier mobility is extremely low – often much less than 1cm2V-1/sec-1,’ says Edwards. ‘Compare these with ZnO:Si (see table above), which have the similar advantage of solution processing and large area deposition but with significantly higher carrier mobilities and carrier densities – typically by a factor of 10, even for the solution-based routes. This is adequate for a number of applications where the highest conductivity performance is not required.’ The Oxford researchers say these materials will allow better device and circuit characteristics including power use, frequency response, and significantly higher environmental stability.
While the team is now fabricating PV devices using ZnO:Si coatings, the next step will be the further development of low-temperature solution-phase deposition of thin film coatings from carefully designed precursor molecules. They are about to test its performance with several photovoltaic technologies, with a view to manufacturing and testing operating device demonstrators across a variety of photovoltaics, OLED lighting and LCD displays.
In as little a year’s time, a ZnO:Si-based target for vacuum deposition of TCO coatings could be introduced to market, although Edwards stresses that further careful optimisation is needed to ensure maximum performance. ‘We also have plans to scale up the solution phase work, where there is clearly a variety of parameters to optimise – particularly the role of precursors and post-deposition treatment technologies.’