LONGi bet on mono silicon and ate the loss on diamond wire until the industry followed

Through the 2000s nearly the whole solar industry built on cheaper multicrystalline silicon; monos higher conversion efficiency did not pay for its pricier production. Li Zhenguo, who had pulled monocrystalline rods since the 1990s, kept questioning the consensus. In 2006 he and classmate Zhong Baoshen worked from the metric that matters, cost per kilowatt-hour: cost falls through engineering, but the efficiency gap between routes is fixed by crystal physics. Their conclusion, recounted in Chinese business writing, was that even if polys ingot cost fell to zero, mono would still win.

The second decision attacked monos cost. In 2009 LONGi entered wafer slicing, where mature slurry cutting wasted material: one ton of ingot consumed some 300 kg each of steel wire, silicon carbide and polyethylene glycol, against roughly 40 kg of diamond wire cutting several times faster and saving about $0.10 of silicon per wafer. Diamond wire was immature and dear, so LONGi sold diamond-cut wafers for five straight months at slurry prices. By 2018 the technology had gone from first adoption to nearly 100% of the industry, per pv magazine, and LONGi led the world in mono ingots and wafers.

The route followed the cost curve. China's 2015 Top Runner program rewarded higher efficiency, mono's share climbed as wafers cheapened, and in 2019 mono capacity overtook poly for the first time, completing the reversal the 2006 analysis had predicted. The company that had been told its route was a dead end had set both the technology standard and, for a stretch, the price floor of the industry's most important input.

Bet on the physics, then eat the cost curve

Conversion efficiency is set by crystal structure and does not yield to cost-cutting, so a route bet on mono could not be undercut by poly makers improving castings.

Diamond wire replaced three consumables with one and ran three to five times faster, so its cost fell steeply with scale while slurry's economics were already mature.

Selling diamond-cut wafers at slurry prices for five months bought adoption: customers got the better product at the old price while LONGi learned the curve first.

By 2018 the whole industry sliced with diamond wire, which removed poly's cost shield precisely when mono's efficiency premium mattered most.

When one options advantage is permanent (physics) and the others temporary (cost), wait on the permanent side and spend to shrink the temporary one. First up the curve sets the price.

Mono's share passed multicrystalline in 2019, and diamond-wire slicing spread beyond solar into semiconductor wafering. LONGi rode the transition to become the world's largest wafer and module supplier and made Li Zhenguo and his co-founders, by 2020, the richest people in Shaanxi province. The company's later years brought the sector's usual overcapacity pains, but the mono-plus-diamond-wire bet remains the standard Chinese case of a technology route decided by physics and won on the cost curve.