A PV manufacturing state of the union


Click here to enlarge image

Paula Mints,
Navigant Consulting,
Palo Alto, CA USA

Historically, photovoltaic (PV) manufacturers have developed in-house capability one machine at a time, and since the industry was small, there was little outside interest in changing this paradigm. Equipment manufacturers serving the semiconductor industry had no interest in developing machinery for an industry with a relatively small revenue stream. In 1998, total cell and module revenues were $510.8 million. Five years later in 2003, cell and module revenues were $2 billion — tiny when compared to semiconductor revenues.

In 2004, boom times came to the PV industry and it became worthwhile to pay attention. In 2004, cell and module revenues grew 72% to $3.4 billion. By 2006, cell and module revenues were over $6 billion; in 2007, cell and module revenues had increased to over $11 billion — significant growth over 2003 revenues of $2 billion.

The world (and the media) began to pay attention, and the spotlight turned to technology development. The silicon shortage, which constrained supply of crystalline (wafer-based) technologies during the boom, drew attention to thin films, and investment dollars flowed into amorphous silicon (a-Si), cadmium telluride (CdTe), and copper indium gallium selenide (CIGS). Equipment manufacturers such as U.S.-based Applied Materials and Switzerland-based Oerlikon, with solar manufacturing capability acquired from Unaxis, began serious development of next-generation amorphous silicon, referred to as micromorph (tandem junction amorphous).

Uni-Solar, Kaneka, and other a-Si technology manufacturers were already developing micromorph technology with approach as the primary difference. Semiconductor equipment manufacturers, seeing a high-growth business, believed that a semiconductor equipment model could be made to fit the not-invented-here (NIH) PV industry, particularly where flat panel synergies existed. Sharp Solar, a Japan-based manufacturer of crystalline and a-Si technology, plans to make use of these synergies in a 1GW facility located in Sakai, Osaka Prefecture, Japan, and in 2008 formed its own a-Si equipment joint venture with chip equipment maker Tokyo Electron.

Customers for turnkey facilities have (for the most part) come from outside the traditional PV industry attracted by high growth, seemingly endless demand, and the belief that technologies with lower manufacturing costs had a significant advantage over crystalline technologies. The new entrants overlooked certain painful truths, among them the long timeline from pilot-scale to commercialization of product (this has everything to do with repeatability), the necessity to work towards higher efficiencies, the fact that thin film manufacturing costs (for the most part) are currently not significantly lower than crystalline, and that price is a market function. Most new entrants assumed that they would be at multi-MW production in a matter of months and that 100% of production would be sold.

2008 over 2007, sales of PV technologies grew by 79% to 5.5GW, with ~1.4GW of this languishing in inventory at the beginning of economically challenged 2009. Changes in Spain’s program and a global recession have severely affected demand for PV products, while casting a pall on investments in large-scale systems.

Though the PV industry is likely to experience one, if not two years of slow growth, strong demand will return, and the U.S. has voiced a desire to return to the manufacturing leadership role it held up until the late 1990s. To do so, the U.S. will need to develop a stronger market in conjunction with a domestic manufacturing sector. Manufacturing incentives and grants will attract new entrants (the semiconductor model, turnkey buyers) to locate in the U.S.

Europe, Japan, China, the U.S., and other regions all have significant interest and good reason to develop robust PV manufacturing sectors. In the case of the U.S., a strong PV manufacturing sector could help turn the U.S. back into a country that makes things instead of just buying them. For all new manufacturers hoping to enter, particularly those starting with turnkey facilities, one to two years of slower growth is a good time to invest in a technology that needs time to develop and commercialize.

Paula Mints, principal analyst PV services program and associate director, energy practice, Navigant Consulting, 3000 El Camino Real, 5 Palo Alto Square, Palo Alto, CA 94306 USA;