April 21, 2008 – Researchers at the Energy research Center of the Netherlands (ECN) say they have developed a method to improve solar cell efficiency in crystalline silicon industrial solar cells, by making them from n-type silicon wafers instead of p-type ones.

The key, they say, was a novel (patented) method for passivating the emitter of n-type solar cells that “significantly” increases their efficiency vs. p-type ones, by forming an ultrathin (1.5 nm) SiO₂ layer (formed by nitric acid oxidation of Si NAOS) between the silicon emitter and the silicon nitride antireflection coating film. The method uses the same PECVD-SiN_x technology used in industry to passivate n+ (i.e., phosphorous) emitters, which suggests rapid and low-cost implementation could be possible in modified existing process lines. (Wafers used in the work were standard 125mm, 1.4-1.6 Ωcm resistivity, 210μm thickness.) Other process steps, such as screen printing, were “of the same low-cost character as are presently used in mainstream industrial production,” they noted in a statement.

Fig. 1. The patented achievement is in the anti-reflection coating and passivating layer (green layer). N (wafer) and P+ (emitter) types constitute the fundamental difference with ‘normal’ cells, where p-type is wafer and n+ is the emitter. (Source: ECN)

The researchers say they observed a 2% point enhancement in conversion efficiency for n-type solar cells passivated using nitric acid SiO₂/SiN_x stack layers, vs. the solar cells passivated without the ultrathin SiO₂ layer. (For multicrystalline silicon, they said, results depend on the quality of the wafer). Record-high best efficiency of 18.3% was demonstrated on single-crystalline Czochralski wafers (17.9% average), and 16.4% on multicrystalline silicon.

Paul Wyers, manager of ECN Solar Energy, noted in a statement that the group’s work focuses on n-type cell technology because of several advantages it may offer — e.g., tolerances for some common impurities in lower-grade silicon feedstock, as well as advantages in the cell process itself. “We are well on our way to develop several cell processes that should be an interesting proposition to industry,” he stated.

Results of the work were published in the Feb. 11 issue of Applied Physics Letters.

Fig. 2. Accelerated ageing test of solar cells passivated using either a SiN_x layer or a stack of nitric acid SiO₂/SiN_x layers. (Source: ECN)