Solar glass: Through thick and thin
Australians are used to solar panels that are about 300 microns thick and they’ve stood the test of time. Phil Kreveld finds out more about a new technology that shrinks the panels to about 80 microns, the thickness of human hair.
There’s a new technology just around the corner that should revolutionise the way Australians consume solar energy and change the way electricians install solar panels.
Thin solar, and we’re talking as thin as human hair, is exactly as it sounds and will have huge ramifications on the way we gather energy from the sun. Additionally, there are huge benefits for building integrated photovoltaic (PV), or BIPV.
Rather than arranging solar panels in rectangular formations on roofs, thin solar cells can be incorporated into a vast range of spaces that we haven’t thought of before. Given their size, this solar technology could be incorporated into curtain walls, spandrels, atriums and even just installed horizontally.
Thin solar lets through light while also generating electricity, opening a world of opportunities that weren’t there with current solar panel technology.
Vertical surfaces require different calculations to determine insolation factors compared to roof-mounted panels and part of this article delves into the features of a new software package, BIPV Enabler, that aims to do just that.
The BIPV Enabler is an RMIT University project that’s partly funded by the Australian Renewable Energy Agency (ARENA) The primary function of the software is to establish the economic viability of buildings with proposed BIPV, taking account of the building coordinates and its latitude and longitude orientation.
Some electricians who install solar panels may be tempted to read no further but please, go on! BIPV, although in its infancy in Australia, will be the future of new office constructions for the new energy-consuming data centres, hospitals, sports/entertainment venues and much more.
On top of that, it will demand new electrical layouts, junction boxes and inverters, so a fresh market is opening up for solar installation companies with the foresight to stay abreast of developments in BIPV.
An outline of thin-film PV
Australia, thus far, is a laggard in the market despite thin panel technology being actively researched by our academic institutions.
Having established the desirability of thin-film solar because of its lower weight compared to traditional roof-mounted solar, the next question is how similar this technology is compared to the PV we’re familiar with.
Solid-state physics is an abstruse topic. We don’t have to bother too much with it, but a few comments on principles of thin versus thick will hopefully enforce the utility of the relative ‘new chum on the block’.
An easy way to explain how it works is to compare it to silicon. While thin-film PV uses semiconductors like copper indium gallium (di) selenide (referred to as ‘CIGS’), it’s difficult to get our minds away from silicon because, in its poly-crystalline (p-Si) or mono-crystalline (c-Si) form, some 300 microns (0.3mm) are necessary to ‘absorb’ a photon in the sun’s visible spectrum. It might undergo 10,000 interactions with the silicon, diamond-shaped crystalline structure (note: p-Si contains mini-crystals scattered in all directions with respect to each other) traversing 0.3mm before loosening an electron needed for electric current contribution.
For thin-film PV, the distance needed is as small as one-tenth of that encountered in silicon. In a lighter-weight solar cell, particularly one that offers transparency, photons only interact to a small extent, with the bulk passing through. In other words, with a low electron conversion efficiency.
In practice, individual photocells, with a much-reduced transparency are part of glass curtains that have a covering of a light-absorbing film to assist the conversion efficiency of the photocells. The gaps between the photocells have sufficient transparency to aid the illumination of the building interior.
As is the case in rooftop solar panels, the photocells are connected in series to provide a power output but wiring between the individual cells is avoided by laser ruling, as shown in Figure 1.
A spandrel panel containing cells is manufactured as one unit and laser ruling separates the cells from one another providing an architecturally pleasing appearance. There is much investigative work going on to improve the efficiency of BIPV because of the tradeoff between transparency and efficiency for light capture and conversion to useful electrons.
The interior world of thin and thick PV cells
Let’s dive into the electron conversion process a bit. It’s essentially no different from conventional solar cells but varying arrangements are used for the doping of the semiconductor material. The so-called intrinsic material of a photocell will create electrons and holes when exposed to light. These holes are vacancies left in the host atoms when valence electrons are moved to the conduction band.
Silicon has four valence electrons which it shares with neighbouring silicon atoms, including the valence electrons that hold the show together, structurally. For CIGS PV cells, selenium can also exhibit tetra-valency as is the vase for silicon.
Externally, the wires connecting it to an electrical load, say a resistor, only pass electrons whereas, within the cell, electrical current is comprised of electrons and holes.
But hang on, a hole has no electrical charge, unlike a negatively charged electron. In fact, a vacancy created in the valence electrons (by lifting one of them to the conduction band), leaves the host atom nucleus positively charged (another term is ‘ionised’).
The vacancy created is soon filled by another nearby valence electron because, at any temperature other than absolute zero (-273°C), the atoms and surrounding electrons vibrate, making these exchanges possible. The electric current consists of negatively charged electrons and positively charged holes within the intrinsic material. Or, if that sounds weird, positively charged nuclei.
But in the outside world, holes meet electrons, cancelling each other out.
In Figure 2, you can see there are six electrons moving to the right terminal of the intrinsic cell and three electrons to the left, leaving a net contribution of three electrons. Finally, an electric field to help move electrons in one direction, rather than wandering left and right in the conduction band, will allow them to produce electrical power.
The electric field is provided by doping with foreign elements. For example, one with five valence electrons would have a loosely bound fifth electron as only four are needed to bind with neighbouring intrinsic atoms.
The spare fifth electron is so close to the conduction band that it can readily jump there. With plenty of doping atoms, as in the above example, donor atoms because they are ‘donating’ electrons, a negative charge would be built up. In the same way, but using doping atoms with three valence electrons, a vacancy is created for neighbouring intrinsic material valence electrons to jump into, thus building up a negative charge. These doping atoms are referred to as acceptors. The way to picture this is to imagine the donor as a ‘spare’ valence electron to have an energy level just below the conduction band, and for the acceptor atom’s valence electrons to have an energy just above that of the intrinsic material atom valence electrons.
In short, by making a PN sandwich of the donor (N-contributing, N-egative electrons) and acceptor (P-contributing, P-ositve ‘hole’) doped regions, an electric field is created, electrons having been ‘sucked up’ into the conduction band, i.e., creating the negative ‘pole’ and likewise a positive ‘pole’ in the acceptor-doped region.
So far, all of this is no different to the way all solar cells operate – with the exception of thin-film solar which has different ways of creating the sandwich. With thin-film solar, rather than a PN junction, a PIN arrangement might be used, the I standing intrinsic material, i.e., no doping atoms present.
To make matters more interesting, thin-film PV cells often comprise several PN junctions in series, with the structure referred to as a ‘hetero-junction device’.
Figure 3 shows the operating schematic of an idealised PV cell. The circle signifies the current generator, i.e., within the PN region – note that the doping concentrations of acceptors and donors are not necessarily equal, and the lighter doped region is referred to as the ‘bulk’ region. The parallel diode signifies the polarity of the electric field created by the PN sandwich, or PIN sandwich.
As shown, the resulting current-voltage profile is the sum of the photon-created electron flow and the ‘dark’ electron flow of the PN diode, provided of course that there is an external load for the current generator. However, unlike the thick film PV cell, the thin cell has in addition to a light flux-dependent ‘current generator’ also a voltage dependent one (shown as a diamond shape). Actually, the diamond shape signifies that there always is a possibility of electrons flowing the ‘wrong way’ and that is influenced by the probability of an electron heading to the negative terminal.
Building integrated PV
With the very significant recent progress in the field of Cu(In, Ga)Se (CIGS), several characteristics have been attributed to PV cells: high photovoltaic efficiency, stability of performance and a low-cost industrial manufacturing method.
Various methods make it possible to obtain the active absorbing layer in CIGS. Co-evaporation is the technique giving the best current photovoltaic yields (22.6% has been recorded) but electro-deposition and magnetron cathode sputtering techniques also see use. This latter technique is generally compatible with the industrial mass production of thin films on a large surface. Figure 4 shows the essential construction features of a thin-film solar cell.
In rooftop solar panels, the size of individual cells is limited by the silicon ingots. Monocrystalline cells are sliced from circular ingots, therefore having rounded edges, polycrystalline cells are not limited by ingot dimensions, but their size is also limited.
Either way, cells have dimensions of the order of 140mm, width and length. Thin film cells, by virtue of the deposition of layers, can be much larger, for example, deposited as long strips, making them less affected by partial shadowing. Nevertheless, shadowing on vertical surfaces is far more the case than on rooves so the layout of bypass diodes is important.
Large buildings require larger lighting and climate conditioning loads. Given the relative scarcity of land, these large buildings go for limited footprints, thus creating occupation volume by extension of height. This creates a suitable application for BIPV because there is much more vertical than rooftop area available for solar.
Additionally, the roof is usually densely populated with HVAC plants. By a rather elementary comparison, the difference between vertical and horizontal insolation is not easy. At high latitudes, roof areas are more favourable than vertical areas and the reverse applies to low latitudes.
Either way, overshadowing from neighbouring buildings is likely to have a dominating influence on the energy-generating capacity of the building surfaces. The evaluation of electrical energy creation for new projects is critical as that can then interact with architectural and consultant engineering design at an early stage.
Retrofitting of BIPV might not be practical in many, if not most, cases. The BIPV Enabler program is an excellent tool for the conceptual design stage.
Location and geographical coordinates should already be given, and building energy load can be computed or based on energy profiles derived from like buildings in preferably the same location as the new project (or guesstimated). The input of building load and its variation in diurnal power as well as seasonal variation due to seasonal diurnal pattern consumption are all part of the RMIT-created program.
The economics of BIPV are more complicated to work out than the average domestic solar installation, hence the value of the RMIT University BIPV-Enabler program.
Projections that buildings might become solar farms are somewhat hyped up. That said, considering BIPV for new buildings is consistent with the general focus on reducing humanity’s CO2 footprint on the planet. Combining architectural elements with BIPV can provide sound economics to buildings that are hoping to achieve their green targets.