News

RECENT DEVELOPMENTS IN DYE SOLAR CELLS AT 3GSOLAR
by Dr. Jonathan Goldstein, Dr. Ilya Yakupov and Barry Breen

ABSTRACT:

Dye solar cells (DSCs) offer a cheaper alternative to conventional silicon cells on the basis of 1. bill of materials (bulk titania powder is the photoactive material in place of silicon) and 2. process costs (screen printing and oven treatment in air in place of vacuum deposition). Furthermore, small research dye cells have been reported by various developers having promising energy conversion efficiencies of more than 10% under one sun illumination. Many developmental dye cells, however, have scale up, efficiency and stability limitations that have hindered commercialization. According to the approach of 3GSolar Ltd. (formerly Orionsolar Ltd.) here described, a novel, intrinsically corrosion-resistant, robust current collecting grid has been developed that allows scale-up to large area dye cells having increased stability and with reduced wastage of cell active area. Our prototype dye cells, which are glass based and have been scaled up to a full commercial single cell size of 225 sq cm, presently give 5.2% efficiency under one sun. We are steadily ramping up cell efficiency and durability by optimization of titania and other cell components, and are on schedule towards pilot production of cells and modules by the end of 2008. In this paper preliminary results are presented on the scale-up, fabrication and performance of full size prototype glass-based single dye cells of size 225 sq cm that incorporate the corrosion-resistant robust current collectors.

1. INTRODUCTION

Dye cells since their invention (1) by Michael Graetzel in 1987 offer a route to low cost PV because not only the raw materials are cheap (titanium dioxide powder is the main photoactive material, and this is a commodity material used in paints and toothpaste) but also cell manufacturing is potentially low cost (print and bake in air). There is no need for expensive materials (such as pure silicon) or costly production methods (no vacuum or doping) as in other types of PV. Many companies and research groups are active in PV dye cells. However commercialization has been delayed because of scale up, materials and stability problems. In this paper we indicate a new route to low cost cells.

2. DYE CELL SCHEMATIC AND OPERATING MECHANISM

Figure 1 is a schematic showing the components of a classical PV dye cell (DSC) and a proposed operating mechanism. The photoanode (facing the light source) is a glass plate whose inner surface

has been coated with a thin layer of transparent conducting tin oxide (TCO). Onto this layer is sintered a several micron thick porous layer of nanocrystalline titanium dioxide (particle size about 20 nanometers) on which is absorbed a monolayer of sensitizer dye. The cell contains also an electrolyte containing a redox species based on the iodide/tri-iodide couple, and a counter electrode (cathode) consisting of a glass plate with a conducting tin oxide layer coated with catalyst. The two glass plates are sealed off at the edges and current take off to an external load is via the TCO layer extremities. Lightweight plastic dye cells are also feasible but have poor efficiency and life.
A photon impacting a sensitizer dye molecule (coating a titanium dioxide particle) causes activation of the titanium dioxide to visible light and enables an electron to enter the conduction band of the titanium dioxide, leaving behind an activated dye molecule. The electron passes out of the cell via the TCO layer (that supports the titanium dioxide) and does useful work via the load, re-entering the cell via the counter electrode. In the presence of the catalyst on the counter, tri-iodide in the electrolyte is reduced to iodide, and iodide is correspondingly oxidized at the activated dye molecule back to tri-iodide. The dye molecule thus returns to its non activated state ready for the next photon.

Fig.1: Dye Cell Schematic and Operating Mechanism

Small research glass-based dye cells (areas usually below 1sq cm) of the above type have been reported by various developers having promising energy conversion efficiencies of more than 10% under one sun illumination. Cell open circuit voltages using ruthenium based sensitizer dyes in such cells are above 700mV and short circuit currents are in excess of 20mA per sq cm.

3. 3GSOLAR APPROACH TO LARGE AREA DYE CELLS

Despite encouraging results with small cells by many workers, scale up to larger cells of practical dimensions (~200 sq cm) has proven difficult. Basically, use of conductive glass alone in a dye cell allows effective current take-off only from long, narrow strip cells, which design wastes often over 30% of the cell footprint due to seal areas and edge effects. Attempts to use silver conductors in the cell to withdraw current and enable construction of broad, large area single cells have not been successful in the long term due to corrosion of silver by the iodine in the cell electrolyte and the poisoning effect of silver on the dye. Even when the silver grids were protected by a layer of polymer or glaze it was impossible to completely prevent pinholes in practical manufacture, and the silver corroded. 3GSolar in Israel (formerly Orionsolar) have developed (2,3) a system of silver-free proprietary robust conductors as an enabling platform for use in current take-off from large area cells, which materials are intrinsically inert to the cell electrolyte. Individual large area cells are sealed, mounted in a module and electrically connected.
Presently we are at the cell prototype building & testing stage, but our near term goal for full sized (15cm x 15cm or 11cm x 20cm) cells for low cost applications is to meet three main parameters:
1. 7.5% cell efficiency at cell level (allowing an efficiency at the module level of 7%)
2. 7 year lifetime warranty
3. $70 per sq meter for bill of materials and labor.
This is consistent with our initial manufacturing cost goal of $1.40 per peak watt for an 8MW per year sized dye cell plant. This is less than half the cost of rival silicon and thin film PV systems, not only because of the lower cost of raw materials for dye cells (titania vs. silicon) but note the investment needed for building such a plant is much smaller than for rival systems because of freedom from vacuum processing and simplicity of technology (screen-print and air baking of active layers). This manufacturing cost is expected to drop to below $1 per peak watt based on economies of scale in multiple plant production and through incremental increase in cell efficiency through incorporation of new materials (such as dyes or titania pastes) and processes.

4. EXPERIMENTAL RESULTS

Fig. 2 shows a full size 3GSolar anode plate (size 11cm x 20cm) comprising tin oxide glass fitted with robust current collectors. The plate shown has been coated with titania, followed by sintering and staining with ruthenium dye. The robust current collectors can be sintered without loss of

Fig. 2: Anode plate (20cm x 11cm) with
robust conductors, titania and dye

conductivity or corrosion resistance in the cell electrolyte, and give low shading (presently ~7%) of available active area. The cell is completed by juxtaposing onto the dye coated anode a counter electrode, covering with a plain glass cover, rim sealing with polymer, and filling with electrolyte. Commercially available electrolytes are used. Due to the large cell area, the rim seal takes up a relatively small fraction of the cell face, and about 90% of the cell geometric area can be made active, a significant improvement over many previous DSC designs. Fig. 3 shows a full size prototype cell with robust conductors (225 sq cm).

Fig. 3: Prototype for pilot production
– full size 225 sq cm cell with robust conductors

Fig. 4 shows the I-V curve for the full size 225 sq cm prototype single cell fitted with robust conductors under one sun illumination.. The open circuit voltage is 721mV, short circuit current 11.7 mA/sq cm, fill factor 62% and efficiency 5.2%. The maximum current drawn from the cell is 2.6A and this is believed to be a world record current for a single dye cell.

Fig. 4: 5.2% efficiency under one sun achieved
for 225 sq cm cell with robust conductors

The robust conductors used in cells are showing exceptional stability in accelerated testing, consistent

Fig: 5 Ongoing weight loss test, full size 225 sq cm cell held at 85 C

with cell lifetimes exceeding 10 years. Rim sealing effectiveness in cells using selected polymers is being confirmed by weight loss tests at 85 C, as shown in Fig. 5 for an ongoing test. We are also assembling single cells into modules and life testing in an outdoor Roof Station (Fig. 6).

Fig. 6: Testing of module in Roof Station

5. FURTHER WORK

We are continuing to refine our prototypes, steadily ramping up efficiency and durability, with a goal of proceeding to pilot production of cells and modules by the end of 2008.

6. REFERENCES

[1] B. O’Regan and M. Graetzel, Nature 335, 737 (1991)
[2] J. Goldstein, I. Markovich, A. Zaban and L. Grinnis, Proceedings of 13th Symposium on Solar Energy Production, Sde Boqer, Israel, Oct. 2005.
[3] J. Goldstein, I. Yakupov and B. Breen, Proceedings of 14th Symposium on Solar Energy Production, Sde Boqer, Israel, Feb 2007.