Your daily dose of photovoltaic technology developments and solar news

• Most Popular Articles

  • Name change: Showa Shell Solar soon to be Solar Frontier, targets 1GW global sales of CIS modules
  • Suntech sales reach US$1.7 billion in 2009: capex, capacity and shipments to rise in 2010
  • SolarEdge partners with Flextronics for manufacturing of distributed PV power harvesting systems
  • LDK Solar acquires Best Solar’s crystalline module manufacturing plant - updated
  • Solyndra taps another Uni-Solar distributor

• Popular Technical Papers

  • Photovoltaic Fab Managers’ Forum Review
  • Photovoltaic industry 2009: a journey into uncertainty
  • Overview of automation in the photovoltaic industry
  • Thin-film CIGS starts to come of age
  • CIGS, CdTe thin-film PV equipment sector emerges, standardization remains elusive

• Subscribe to our weekly newsletter

  Name
  Job title
  Company
  Email

If you want to get into the thin-film PV game as a serious player, you have to benchmark your costs against First Solar and beat them. That’s a recurring theme among some TFPV start-ups I’ve talked to recently, companies that put as much emphasis on their cost and performance models as they do on touting their enhanced manufacturing of existing approaches (like CIGS) or new-fangled, next-gen technologies (like nanocomposite cells). One such firm gearing up for the long haul is Solexant, which hopes to leverage its inorganic semiconductor nanocrystalline secret sauce(s) with very-low-cost, monolithically integrated roll-to-roll processing and improved balance-of-system schemes.

Solexant’s Damoder Reddy (pictured) wouldn’t divulge the elemental truths behind the company’s UC Berkeley-licensed crystalline entities when I spoke with him at the recent Solar Power International show. But the president/CEO did explain that the approach is to synthesize and functionalize the single-digit-nanometer-scale binary semiconductors in solutions or inks and then print them as films on a flexible, inexpensive metal-foil substrate in a nonvacuum deposition process. Micron-sized crystals grown from those printed films become high-quality photoactive materials capable of absorbing light from across the spectrum.

If this sounds somewhat similar to Nanosolar’s approach, keep in mind that three of Solexant’s top management team members spent time working for the much-publicized printed-CIGS-on-flex company. A diagram of the Solexant substrate architecture (part of Reddy’s presentation at the recent NREL Clean Energy Industry Growth Forum) reveals a film stack reminiscent of CIGS—largely standard-looking back metal, absorber, window, and TCO layer scheme—although the company’s glass-on-top, foil-on-bottom configuration deviates from the norm.

But Solexant is not Nanosolar 2.0 (or ISET 3.0, for that matter). No CIGS here. One key difference, Reddy told me, is that the technology is material agnostic—the nanocrystal solution process remains the same, regardless of which semiconducting material is used. The manufacturing plant can handle any material set, which could then be ramped quickly at high yields in production.

A perusal of Reddy’s/Solexant’s patent applications reveals a wide range of “embodiments” for the proprietary technologies, in material choice and processing configuration. (Photo at left shows nanoparticle film, photo below depicts micron-scale crystals.)

Solexant’s 2MW pilot line in San Jose, which has just been completed and should be in full operation in a few months, is already running webs of the cheap (and still-proprietary) foil in widths up to 750mm. By the middle of 2010, Reddy expects the line (made up of tools of his team’s own design) to produce “modules good enough for field testing” with the company’s three customers (two European solar farm installers, one U.S. utility).

The next step will be to select a location and then build a 100MW production line in the U.S., which would come online in 2011, as long as financing is secured, according to Reddy. The first volume facility would incorporate cell and module production on the same site, but the eventual game plan is to process the cells in the States and assemble the panels in various locations proximate to regions with strong demand for solar.

Here’s where Reddy starts to pound home Solexant’s “we’re gonna beat First Solar at its own cost game” strategy. The price tag of that envisioned 100MW fab, according to the exec? Less than $40 million, and an 800MW plant would come in under $250 million.

These scenarios factor in capital efficiency and manufacturing costs of less than 50 cents per watt, engendered in part by claims of better than 90% material utilization, as well as 200%-500% faster throughputs and a thermal budget less than 50% compared to glass module production.

(First Solar says it expects manufacturing costs per watt to be in the 65-70 cents range in the next few years; also, it can process a CdTe module, from glass in to panel out, in about 2.5 hours, so Solexant’s throughput/cycle time number breaks down to about an hour or less.)

Single-layer conversion efficiencies would fall in the 10-12% range (second-generation materials in R&D could hit the high teens, he said). The Solexant modules would be significantly larger (though relatively lightweight) and more powerful than First Solar’s. More interestingly, the nano-PV upstart claims its technology and design eliminates the so-called “balance of system penalty” suffered by the
thin-film market leader in relation to the crystalline-silicon crowd.

Solexant says it has the potential to reach total BOS costs of about 97 cents versus First’s $1.21, total installed costs of close to two bucks per watt, and a grid-parity-like LCOE (levelized cost of energy/electricity) of less than a dime per kilowatt-hour—all while supposedly maintaining gross margins of better than 40% at 100MW production scale.

(Note that First’s updated BOS roadmap says it should drive costs down from the current ~1.40/W to the 91-98 cents range by 2014.)

Solexant’s proposed 1m long × 2m wide modules will pack nearly three times as much wattage as First’s (220W vs. 80W), with 6 amps of current compared to 1 amp. It will take less than a third as many of the start-up’s modules to power a 10MW PV power plant (33,600 vs. 106,000), and about half the number of mounting rails (22,440 vs. 42,400), saving some $600,000 in cost, according to company estimates.

On the DC cable-to-combiner box side of the cost analysis envelope, Solexant says it will deploy less than a sixth as many units (35,900 vs. 243,000) at about half the cost as First ($2 million vs. $3.9 million). The new guys say they will fit 18 modules to an array table (to First’s 50) and sport 30 modules to a string (to First’s 10).

Despite the admirable cost modeling and aggressively optimistic outlook by Solexant, the young firm’s grandiose claims require a serious reality check. It is just now ramping up its pilot line. It needs to get some working module prototypes out into the field soon, and send panels to be shaked, baked, chilled, beat up, and otherwise go through the battery of accelerated-life, safety, and reliability tests. It will then have to join the queue for IEC and UL certification. The nanocrystal glass-flex modules need to be more than just “cool”—they will have to work, and work well, to be competitive and instill confidence in project developers and system integrators. Realistically, it will take at least two years, possibly more, to deliver on all these action items.

Then there’s that small matter of financing. The company must raise a trunkload of cash--it's already garnered some $22.5 million in two VC rounds, and Reddy told me he's in discussions on a Series C--and secure adequate credit lines to stay on track with its bold roadmap. 

If things continue apace, Solexant may not have to do this all on its own—the firm could/would make an intriguing acquisition prospect for a big fish that's in the market for disruptive yet commercially imminent technology. Perhaps even a piscine purchaser with low-cost bona fides sporting the initials “F” and “S.”