• Print

The view from Moses Lake, Part I: REC Silicon makes a big solar-poly bet on a fluidized bed


Tom Cheyney
Tom Cheyney
Tom Cheyney, former senior editor of PV-Tech and Photovoltaics International, is now chief curator of SolarCurator.com and director of Impress Labs’ solar practice.

The tallest structure in rural Grant County, WA, is not a grain silo or airport control tower, but a nine-story, 56-meter-tall factory unit on REC Silicon’s Moses Lake campus that will play the key role in the company’s solar-grade polysilicon production business going forward. There’s quite a view from the observation deck, both of the sprawling facility and the surrounding east-central Washington flatlands.

Part of a nearly $1.7 billion capital investment outlay ongoing at the site (which also includes two new silane plants and other support facilities), the building contains a line of 24 fluidized bed reactors, or FBRs. Not to be confused with those other FBRs--the fast breeder reactors in the nuclear energy field-- these solar material processing tools represent a significant, somewhat risky technological jump from the traditional Siemens process used to produce most of the world’s poly.

Instead of the usual chunks and rods that come out of the standard process, the FBR reactors synthesize millimeter-ish-diameter polycrystalline silicon granules. The low-cost process runs continuously and, perhaps most importantly, is much more energy efficient than the plain-wrap Siemens method, according to REC Si. (That energy efficiency was cited as a major reason for the Feds’ decision to award nearly $155 million in advanced energy manufacturing investment tax credits to the company as part of the U.S. stimulus package.)

Another plus, the end-products are apparently easy to handle:  those millions of little poly BBs are pourable and can be stored in flexible bulk containers, attractive both from the manufacturer’s shipping perspective and at the customer’s loading dock.

Although REC Si rightly claims that it is all about the SiH4 molecule, the big bet it has made on FBR was not something taken in the heat of the market moment. Work on fluidized-bed technology has been going on at Moses Lake since the pre-REC days of the 1990s when the concept was proven, although R&D was shelved in the latter part of that decade and not restarted until REC got involved in 2002.

Three successive generations of pilot reactors led to the development of the FBR-A demo reactor in the mid-2000s, the 150MT workhorse unit on which the company spent five years optimizing its commercial process, tightening up its control modeling, and getting the first samples of the granular poly out for customer qualification.

Leveraging the learning achieved with the demo reactor, the Silicon 3.0 factory with its larger reactors began ramping in earnest in early 2009, but has encountered some turbulence (aside from the normal agitation that occurs in the process tools) along the way.

Glitches and delays are to be expected when a new technology moves from pilot development to the volume production stage, and REC Si’s experience with FBR has had its share. The project has run about 67% over the original budget because of delays and slowdowns in the construction and production ramp.

The most significant hiccup came when the company had to shut the new plant down for safety reasons from late March to June 2009, as a result of materials fatigue problems it had identified with the original discharge piping on the reactors. Silane source-gas shortages in July and October also brought production to a halt. A few small fires and reactor plugging issues in the silane areas didn’t help matters.  

But the company feels that it has effectively dealt with those issues—thanks to significant modifications in the affected FBR piping, steadily increasing silane output, and across-the-board process optimization—and the fluidized-bed line has become ever-more stable and productive.

REC Si says that 14 of the 24 reactors are processing granular poly, their production cycles lengthening, pushing the plant toward its slated 10,500MT nameplate annual capacity (an average per-reactor capacity of 437.5MT per year, or normalized per-week output of just over 201MT).

FBR is the cornerstone of REC Si’s efforts to whack at least 40% out of poly production costs by 2012 and to push the cash cost per kilo of poly granules below $20 by the end of 2010--and even lower in subsequent years. The company forecasts that it will have 12,000MT of poly production capacity coming online by the end of this year (both from its Siemens and FBR ops), with the amount rising to 15,000MT and 17,000MT in 2011 and 2012, respectively.

If these estimates hold, it would mean that the vertically integrated Norwegian PV firm’s silicon unit will soon hold more than 10% of the projected available global supply among the established incumbent poly-producing companies (a bit less when the new entrants’ capacities are taken into account). 

On the silane side, the first of the pair of new 9000MT sister plants at Moses Lake—part of Silicon 3.0--is fully operational and ramping capacity, while construction on the second—known as Silicon 4.0--will be done in April, with commissioning slated for completion in June, according to plant management.

With the sharpest growing pains seemingly behind it and feeling more confident about the company’s ambitious buildout, the RECing crew recently flew a group of European and North American investors, analysts, and even a couple of journalists like myself out to Moses Lake for a first-hand look at the operations.

The fieldtrip was split between a morning full of presentations from company execs and an afternoon that mixed one-on-one meets between tripsters and the V- and C-level folks with a series of plant tours. At the end of the day, I felt like I had received a comprehensive tutorial on the current status of REC Si, the history of the Moses Lake site, and silane and polysilicon production/technology à la REC. 

In the next part of the Moses Lake blog series, REC Silicon’s VP of technology compares and contrasts the Siemens and FBR processes, and provides further technical insights into the fluidized bed that the company has made for itself. Click here to go to the next installament.



  • Photovoltaics International 29th Edition

    Forecasting the evolution of a young, dynamic industry is by definition an uncertain business, and solar is no exception. Rarely, if ever, do the numbers broadcast by any of the various bodies involved in the PV prediction game tally, and even historical deployment rates remain the subject of hot debate. The paradox is that getting forecasts broadly right is going to become increasingly important over the next few years, particularly for those involved in producing the equipment that will support whatever levels of demand come to pass.



Solar Media


We won't share your details - promise!