From a news story in seekingalpha.com:
“Some time ago, we started a series on solar energy innovation that proved quite popular (part 1, part 2, part 3). While there is steady and gradual progress in conversion efficiencies (the part of sunlight converted into electricity) and cost, there are companies and labs all around the world working on more radical concepts.

Radical concepts that could offer existing solar players (almost all of them already deeply plagued by market circumstances) a run for their money, but at the same time offer a critical break through to a future without subsidies and mass adoption.

We feel that investors in existing solar companies (or any company) tend to be insufficiently aware of industry characteristics. For instance, many investors were focusing on the growth in demand and steady cost reduction, while having insufficient appreciation for the lack of pricing power, due to the fact that solar cells are a near commodity. This often turned out to be a costly mistake.

Also, insufficient weight is often given to the lack of a dominant design and the fact that literally hundreds of labs and relatively unknown, often non-public companies are trying new approaches, any one of which might lead to a more radical break-through in price and/or efficiency.

For existing players (and investors in these companies), all this innovation from so many different corners presents a clear risk, the risk that some other company will eat their cheese, but for society as a whole, this is a giant trial and error race towards affordable clean energy, enlisting some of the best and the brightest.

So here is the fourth installment of our survey of promising new technologies and concepts on the horizon.

MIT
One approach to get solar cells to convert more sunlight into electricity is to build cells that react to a wider gamut of frequencies. This is what researchers at MIT are doing, creating an all carbon solar cell that captures much of the near infrared spectrum (40% of the sunlight’s energy hitting earth) which normal solar cells don’t convert.

The idea is to uses these cells in combination with normal cells capturing the visible light:
The carbon-based cell is most effective at capturing sunlight in the near-infrared region. Because the material is transparent to visible light, such cells could be overlaid on conventional solar cells, creating a tandem device that could harness most of the energy of sunlight. [MIT]

Before you’ll get overly enthusiastic, the very much experimental cells at present offer only a 0.1% conversion efficiency, which is terribly low. But there are good hopes this can be significantly improved in the near future.

New Jersey institute of Technology (NJIT)
The NJIT is working on a concept that will ultimately enable consumers to print sheets of these solar cells with inexpensive home-based inkjet printers… Imagine someday driving in your hybrid car with a solar panel painted on the roof, which is producing electricity to drive the engine. The opportunities are endless [Dailytech]

The concept is based on solar cells made of polymers, a material that is way cheaper than the purified silicon (the same material used for making microprocessors), and is much easier to handle.

North Carolina State University
Researchers from North Carolina State University have developed a method to sandwich the active layers of cells between a ‘nano sandwich.’ This allows the active material to be way thinner (amorphous silicon can be 70nm, in stead of the usual 300-500nm thin), without compromising their conversion efficiency.

This method can be applied to a wide range of active materials, not only amorphous silicon, but also cadmium telluride, copper indium gallium selenide and organic materials.

Fraunhover and Dow Chemical
The US research institute and chemical giant have together developed an ‘elixer’ to shield solar panels from all kinds of environmental influences that could lead to a degradation in performance. This is rather important, as the economics of panels are such (almost all up-front cost) that it depends critically on durability. Panels have to be able to keep working with little performance degradation for periods up to 25 years.

Instead of laminating panels with ethylene-vinyl acetate (EVA), the scientists used liquid silicone. After this hardened, they subjected it to rigorous tests, which showed that these were more durable.

Berkeley National Laboratory, the University of California, and the DOE
Researchers from these institutes have arrived at a method to use virtually any semiconductor material for solar cells, opening the door to low-cost, high efficiency cells. At present, expensive semiconductor materials are used, such as large crystals of silicon, or thin films of cadmium telluride (CaTd) or copper indium gallium selenide (CGIS).

Previously, cheaper semiconductor material like metal oxides, sulfides and phosphides have been difficult because it was expensive to taylor their properties by chemical means (called ‘doping’). What the researchers did was to tailor these materials simply by applying an electric field:

Our technology reduces the cost and complexity of fabricating solar cells and thereby provides what could be an important cost-effective and environmentally friendly alternative that would accelerate the usage of solar energy [greenbuildingelements]

Thin Film
Despite very tough market circumstances, particularly for start-up thin film producers, there is still plenty of activity and innovation in this sector. This sector faces two sets of problems. First is that more traditional silicon based technologies, which are generally more efficient, have come way down in price, negating much of the cost advantage of most thin film based manufacturers.

Second, since there is so much overcapacity in the market, clients are very hesitant to go with new companies, these might not be around in 5 or 10 years time, and it’s also more difficult to get project financing.

Stion
The American producer of thin film panels is already commercially active, with plants in California and Mississippi. It is backed by famous venture capitalist Vinod Khosla, but also a host of others (AVACO, a private Korean equity fund, Taiwan Semiconductor, Lightspeed Venture Partners, Braemar Energy Ventures and General Catalyst Partners).

The company didn’t start production before it was ready and was very frugal with capital. The company has achieved a 13.4% conversion efficiency for it’s CIGA based modules, which is a record for commercial panels. It is presently offering panels for $0.75 per watt, which is very competitive if you realize that market leader First Solar’s production cost (first quarter figures) are 73 cents per watt.

Flexible cells from Stanford University
One of the problems with solar panels is that because of the plunging price of cells and panels, the cost of installation is often higher than the panels themselves, and these costs are less susceptible to efficiency improvements or cost cuts.

One way to deal with that is to make solar panels light and flexible. The active material in thin film technologies is thin enough to be just that, but the substrate isn’t, as that’s usually rigid material like glass. Now, there are thin-film panels that are flexible, but these have drawbacks, like having complex manufacturing processes or expensive flexible substrates with extremely uniform surfaces.

Here comes Xiaolin Zheng, from Stanford University who can transfer the active materials of thin film cells to other surfaces, such as a sheet of paper(!) or plastic, the roof of a car, or the back of a smartphone. His trick is to use a layer of nickel between the fixed substrate and the active layer.

Using water which reacts with the nickel, the substrate becomes loose and can be peeled away and deposited onto another material, not affecting the efficiency of the cells. Very much a work in progress, but any commercial success could make cells ubiquitous.

UCLA
Researchers at UCLA are using polymer solar cells to arrive at much the same end result: flexible (70%) transparent cells that can be integrated in windows, buildings, and electrical devices. The transparency is achieved by absorbing infrared light, not visible light and using a transparent conductor made of a mixture of silver nanowire and titanium dioxide nanoparticles, which was able to replace the opaque metal electrode used in the past. At just 4%, it’s electricity conversion efficiency leaves something to be desired though.”

—Exerpt from http://seekingalpha.com/article/1128221-solar-innovation-marches-on-part-iv