Copper Zinc Tin Sulphide (CZTS) Solar Cells

Adam Jorna
December 11, 2012

Submitted as coursework for PH240, Stanford University, Fall 2012


Fig. 1: Relative abundance of the elements. (Source: Wikimedia Commons.)

Of all the possible sources of renewable energy, solar has perhaps the most potential. This is because of the sheer size of the solar energy resource when compared to wind, wave, or tidal power. (Table 1) The estimated recoverable solar resource of 1,000 TW is easily capable of meeting the current world energy consumption of 18 TW.

Solar energy is, however, nothing like as concentrated as oil or coal, so harvesting it requires very large scale installations. [2] As an example the Charanka solar park in Gujarat, India, is expected to generate some 500 MW when complete and covers 2,000 hectares

As Thin as Possible?

Because of solar's diffuse nature, a key consideration in the production of solar cells is the cost of raw materials and ensuring that these are used as efficiently as possible.

There are currently two main types of solar cells in mass production, bulk crystalline (primarily silicon based) and thin films. The thin film technologies tend to have lower efficiencies but are easier to manufacture and, as the name suggests, use much less active material. This means that overall the two types of cell have a comparable cost per watt.

At first sight crystalline silicon is an odd choice of material for a photovoltaic device. It has has a bandgap of only 1.12eV (1.4eV gives the highest possible efficiencies for a single junction cell) and is a poor absorber of photons. However owing to the integrated circuit industry, hundreds of billions of dollars have been spent researching silicon and incredibly pure single crystals of silicon are produced in vast quantities. The solar cell industry has piggybacked on this work and as a result the highest efficiency commercial solar cells are generally made from crystalline silicon.

Current thin film technologies are not as efficient as crystalline silicon but they are cheaper to manufacture and have several other important advantages as well.

Resource Total (TW) Estimated Recoverable (TW)
Solar 90,000 1,000
Wind 1,200 10
Wave 3 0.5
Tides 30 0.1
Table 1: Estimated Global Energy resources. [1]

The two main types of thin film solar cells are CdTe, cadmium telluride, and CIGS, copper indium gallium, sulphide. Unfortunately there are drawbacks to both of these technologies. Cadmium is toxic and there are issues around the potential scarcity, and consequent volatility in cost, of both Indium, In, and Tellurium, Te. [3] This can about seen in Fig. 1 which shows that Tellurium is about as scarce as gold. Indium, whilst somewhat more common, is in great demand for many alternative uses and has seen enormous price volatility over the past 10 years - around a 10-fold increase in price between 2003 and 2009.

One possible alternative is thin film solar based on a CZTS semiconductor. The constituents of this semiconductor, copper, zinc, tin and sulfur, have the advantages of being both abundant in the earth's crust and non-toxic. CZTS also has near ideal properties for solar photovoltaics, as it is a very strong absorber and has a band gap of around 1.4eV.

IBM has an active research department in this area and recently announced a CZTS based solar cell that had a conversion efficiency of 11.1% and could potentially be printed. [4] This efficiency is lower than the equivalent laboratory records of 19.6% for CIGS and 17.3% for CdTe, however CZTS is at a much earlier stage of development. [5]

It is worth noting that the IBM solar cell used a combination of selenium (Se) and sulfur, but selenium, unlike indium does not have significant industrial uses and hence is much cheaper - around $80 per kg.

Is CZTS the Future?

So is CZTS destined to supplant the alternative technologies and become the material of choice for photovoltaics? Clearly it is impossible to say anything definitive at such an early stage, especially with so many different types of cell in the market. It is however worth making a few observations. The first of these it that China in particular has invested $bn's in crystalline silicon facilities and it will be very difficult for any new start-up to reach the economies of scale necessary to entirely supplant silicon from the market. The second is that module price is not the be all and end all when comparing the cost per watt. This is because the module cost has fallen dramatically over the last 10 years and as a consequence the balance of system and installation costs have become much more important. Module efficiency is therefore a powerful lever in reducing the cost per watt of these factors.

The maximum theoretical efficiency of single junction solar cells is around 30% - 33%, some way in excess of the current production efficiencies for either CIGS, CdTe or crystalline silicon. [5,6] Unfortunatley significant improvements to curent production efficiency levels will be hard to achieve through optimization of the current manufacturing process, or by replacing one semiconductor for another. Instead tandem or multi-junction cells have perhaps the greatest potential for significantly increasing the efficiency and hence reducing the total cost per watt.

Multi-junction cells work by first capturing the high energy photons, and extracting as much energy as possible from them, and then the second layer captures the lower energy level photons. The first layer could be an organic semiconductor with a large band gap and the second layer an inorganic semiconductor such as CZTS. Such a structure would have a theoretical maximum efficiency of 42%. [7,8]

In summary no one type of solar cell is likely to dominate the market, instead different designs and materials will be optimized for various operating conditions. If the cells need to operate at high temperatures, or need to be flexible then thin film technologies (including CZTS) may be favored, if efficiency is of more importance then this may swing the pendulum towards a silicon / organic tandem cell. Either way the future looks bright for solar power.

© Adam Jorna. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.


[1] B. Sorensen, "Renewable Energy: A Technical Overview," Energy Policy 19, 386 (1991).

[2] S Herron, "Solar Irradiation and Energy from Deserts," Physics 240, Stanford University, Fall 2010.

[3] C. Candelise, J. F. Speirs and R. J. K. Gross, "Materials Availability For Thin Film (TF) PV Technologies Development: A Real Concern?" Renew. Sustain. Energy Rev. 15, 4972 (2011).

[4] M. LaMonica, "IBM Breaks Efficiency Mark with Novel Solar Material," Technology Review, 20 Aug 12.

[5] M. A. Green et al., "Solar Cell Efficiency Tables (Version 40)," Prog. Photovolt. Res. Appl. 20, 606 (2012).

[6] W. Shockley and H. J. Queisser, "Detailed Balance Limit of Efficiency of pn Junction Solar Cells," J. Appl. Phys. 32, 510 (1961).

[7] A. De Vos, "Detailed Balance Limit of the Efficiency of Tandem Solar Cells," J. Phys. D 13, 839 (1980).

[8] C. Peters, "Thin Film Tandem Solar Cells," Physics 240, Stanford University, Fall 2010.