#726 1959 · Pilkington Brothers (Alastair Pilkington) · Glass manufacturing
Pilkington stopped trying to grind glass flat and let a bath of molten tin do it for free
the problem
Flat glass needed expensive grinding and polishing to remove surface distortion
background
By the 1950s, sheet glass could be mass-produced continuously by drawing or rolling it, but the resulting surface was never quite optically flat. Any application that needed true flatness -- mirrors, car windshields, shop windows -- still had to go through a separate grinding-and-polishing stage afterward, an expensive, slow, capital-heavy process that consumed a large share of the glass itself as waste. Decades of industry effort had gone into making that finishing stage faster and cheaper, not into eliminating it.
Alastair Pilkington, a young technical director at Pilkington Brothers, began experimenting in December 1952 with a different premise entirely: instead of shaping glass and then correcting its surface by force, let the glass shape itself. He proposed pouring molten glass onto a bath of molten tin and letting it float there, on the theory that a liquid surface is already perfectly flat and would pass that flatness on to the glass resting on it.
what everyone would do
The industry's decades of effort went into making the grinding-and-polishing finishing stage faster, cheaper, and less wasteful, because a rough or distorted glass surface straight out of forming was treated as an unavoidable cost of production to be corrected afterward, not a defect that could be prevented from happening in the first place.
what they saw
A liquid surface is already perfectly flat on its own, for free, without any tool touching it. If molten glass could be made to rest on a liquid rather than be dragged or rolled across a solid surface, the flatness wouldn't need to be manufactured afterward -- it would already be there, inherited from the liquid it floated on.
the move
Molten glass poured continuously onto a bath of molten tin spreads out and settles at its own natural equilibrium thickness, held flat on both faces by gravity and surface tension rather than by any tool touching it; as it moves along the bath it cools enough to be lifted off as a solid ribbon, already flat and fire-polished, with no grinding step at all. Pilkington and colleague Kenneth Bickerstaff spent seven years turning the idea into a continuous 24-hour industrial process before it worked reliably.
why it works
Molten tin is denser than glass, stays liquid across the exact temperature range glass needs to be formed and then cooled, and does not dissolve into it, so glass poured onto it floats rather than sinking or mixing. Gravity pulls the glass into a flat layer and surface tension holds it at a stable equilibrium thickness on both faces at once, without any external tool ever contacting the surface. As the ribbon travels down the bath and cools enough to hold its shape, it can be lifted off already flat and fire-polished, eliminating the entire grinding-and-polishing stage the rest of the industry still depended on.
the payoff
The float process was announced to the industry on 20 January 1959, after roughly seven years of development and a reported £28 million in R&D spending -- every sheet of glass made during that development period had still come out imperfect until a fully viable production run was achieved in July 1958. Rather than keep the process exclusive, Pilkington licensed it to competing glassmakers worldwide starting in the early 1960s.
where it breaks
The trick only exists if a companion liquid with the right properties is available -- dense enough to float the target material, chemically inert toward it, and liquid across the needed temperature window without visibly contaminating the product; not every material has an equivalent of molten tin waiting for it. It also demands extreme continuous-process discipline: a float line has to run as an unbroken 24-hour flow, since any interruption ruins the ribbon in progress, which makes it a poor fit for intermittent, small-batch, or highly variable production.
what came after
The float process became the global standard for flat glass manufacturing: roughly 500 float lines operate worldwide today, and it is now how the overwhelming majority of the world's flat glass is made. Asia's first float line, built under license from Pilkington, began operating at NSG's Maizuru plant in Japan in November 1965.
references
- [1]Science: Float GlassTIME, 1959time.com
- [2]Brief history of the flat glass patent – Sixty years of the float process (doi:10.1016/j.wpi.2014.04.006)World Patent Information (Elsevier), 2014sciencedirect.com
- [3]Invention of Float GlassPilkington (NSG Group), 2023pilkington.com
- [4]Invention of the Float ProcessNSG Group, 2018100th.nsg.com