From stereolithography to today: a short history
📍 Where we are: Foundations — Chapter 2 of 11. Chapter 1 defined additive manufacturing (AM). Now we look back at how it was invented, who owned which ideas, and how it grew from a lab curiosity into a factory tool.
Think of the history of the electric guitar or the personal computer. First a few tinkerers in labs show that the idea can work. Then someone patents a clever version and sells expensive machines to companies. Years later the patents run out, hobbyists copy and improve the designs, prices fall, and suddenly everyone has one. AM followed a similar path, over about four decades.
What this chapter covers
We walk through the main milestones in order: the first layer-by-layer experiments in the early 1980s, the patents that created the first companies, the arrival of metal machines, the open-source RepRap movement, and the move into industry. To keep to the rule that every fact must be checkable, we only state dates we could confirm in patent records, journal articles or other sources, and we say plainly where sources disagree.
A short vocabulary note: a patent is a legal right, granted for a limited number of years, that stops others from making or selling an invention without permission. When a patent expires, anyone may use the idea freely.
The 1980s: the first layers
Kodama, 1981
Hideo Kodama of the Nagoya Municipal Industrial Research Institute published "Automatic method for fabricating a three-dimensional plastic model with photo-hardening polymer" in the journal Review of Scientific Instruments in November 1981 (volume 52, pages 1770-1773). He described building a solid plastic model by exposing a liquid photopolymer (a liquid plastic that hardens when ultraviolet light shines on it) and stacking the hardened cross-section layers [1]. It is an early, published demonstration of the layer-by-layer idea.
Hull and stereolithography, 1983-1986
Charles (Chuck) Hull filed US patent 4,575,330, "Apparatus for production of three-dimensional objects by stereolithography", on 8 August 1984; it was granted on 11 March 1986 [2]. The patent describes forming a cross-sectional pattern at the surface of a fluid that changes state when stimulated (for example by radiation) and building the object from successive layers. The word stereolithography (SLA) refers to this approach of hardening a liquid layer by layer with light. The American Society of Mechanical Engineers, which designated the process a Historic Mechanical Engineering Landmark in 2016, dates Hull's prototype system to 1983, says he co-founded 3D Systems in 1986, and says the company's first printer, the SLA-1, followed in 1987 [3].
The first machines were sold as rapid prototyping tools: they made quick plastic models of a design so that engineers could look at and hold a shape before paying for a mold. In 1987 the STL file format was created for 3D Systems' first commercial machines by the Albert Consulting Group; STL describes a surface as a mesh of triangles, and it remains common today (see Chapter 3) [4], [5].
Patents create the process families
The late 1980s and early 1990s produced patents for three more processes. The table lists what the patent records show.
| Process idea | Inventor and institution | Patent | Filed (or priority) | Granted |
|---|---|---|---|---|
| Selective laser sintering (SLS): a laser fuses a bed of powder | Carl R. Deckard, University of Texas | US 4,863,538 | 17 Oct 1986 | 5 Sep 1989 [6] |
| Fused deposition modeling (FDM): melted plastic is squeezed out of a nozzle | S. Scott Crump, Stratasys | US 5,121,329 | 30 Oct 1989 | 9 Jun 1992 [7] |
| Three-dimensional printing: an inkjet head sprays liquid binder onto powder | Sachs, Haggerty, Cima, Williams, MIT | US 5,204,055 | 8 Dec 1989 | 20 Apr 1993 [8] |
The MIT process is what the standard now calls binder jetting. According to the sources we found, MIT licensed it to Z Corporation in the mid-1990s; the sources give 1994 and 1995 for the founding or licensing, so we do not pin it down [8].
"FDM" is a trademark of Stratasys. The neutral name used by the standard for the same idea is material extrusion, which is why you will often see "FFF" (fused filament fabrication) used for open-source printers.
The 1990s and 2000s: metal arrives
Plastic models were useful, but industry wanted metal parts. Two routes appeared in Europe:
- Laser melting of metal powder. At the Fraunhofer Institute for Laser Technology (ILT) in Aachen, Germany, a project in 1995 led to the "basic" selective laser melting patent. The German application DE 19649865 was filed on 2 December 1996 and published on 12 February 1998 [9]. The idea was to melt metal powder completely rather than only partly bonding it, as SLS does. Today this family is called laser powder bed fusion (Book 2, Chapter 5).
- Electron-beam melting. The Swedish company Arcam AB was formed in 1997, building on a patent application of 1993 with Chalmers University of Technology. A 2006 company presentation says beta machines reached the market in 2002 and the first EBM S12 machine was delivered in March 2003 [10].
Patents expire and RepRap appears
An open-source project is one whose designs are published so that anyone can study, copy and improve them. Adrian Bowyer and colleagues at the University of Bath started RepRap ("replicating rapid prototyper"), an open project to build a 3D printer that could print many of its own parts; the project's own 2011 paper gives 2004 as its start. The same paper reports an estimated reproductive success of about 4,500 machines in two and a half years [11].
The timing mattered because key patents were running out. Google Patents lists an anticipated expiration of 30 October 2009 for Crump's FDM patent, and 17 October 2006 for Deckard's SLS patent; Google itself notes these are assumptions, not legal conclusions [6], [7]. We did not find an authoritative source that ties a specific expiry to the arrival of low-cost desktop machines, and a company can hold several patents on one process, so treat any single "expiry year" as approximate.
One commercial milestone of the desktop era is documented by the buyer: Stratasys announced on 19 June 2013 that it would acquire MakerBot, a transaction it valued at 403 million US dollars at announcement [12]. That is a company press release about its own deal, so we use it only for that fact.
2010s: attention and the industrial turn
In his State of the Union address on 12 February 2013, US President Barack Obama said that a once-shuttered warehouse in Youngstown, Ohio, was "now a state-of-the art lab where new workers are mastering the 3D printing that has the potential to revolutionize the way we make almost everything" [13]. That speech is a convenient marker of how much public attention the technology had gained. Whether the excitement of those years was justified is a matter of opinion; later chapters look at what AM does and does not do well.
Since then the story has been less about hobby printers and more about industry. The annual Wohlers Report estimated worldwide AM industry revenue at about 20.0 billion US dollars for 2023 (growth of 11.1 percent), with about 3,793 metal AM machines sold that year, up from 3,049 in 2022 [14]. The 2025 edition reports growth of 9.1 percent to about 21.9 billion US dollars, driven mostly by Asia [15]. In 2015 an international vocabulary standard first appeared, and the second edition, ISO/ASTM 52900:2021, replaced it, so that the field's words are now shared worldwide [16].
Key dated milestones, drawn from patent records and papers: from the first photopolymer layers (1981) to RepRap (2004) and industry-scale use.
Original diagram by the authors, created with AI assistance.
Why it matters
Knowing the history explains three things you will meet later. First, the strange names: SLA, SLS and FDM are old trade or inventor terms, which is why the standard introduced neutral ones. Second, why open-source machines took off when they did. Third, why industrial metal AM is relatively young: the laser and electron-beam metal machines date from the 1990s and early 2000s, and qualification (proving that parts are safe) is still maturing.
In the real world
Almost every process in the next chapters traces back to one of the inventions above. A dental lab using resin printers is using Hull's family of ideas. A hobbyist's desktop printer descends from FDM and RepRap. A factory metal printer is a descendant of the Fraunhofer ILT or Arcam work. When you read a machine brochure, you are usually reading the latest chapter of one of these stories.
Key terms
- Patent — a legal right, for a limited number of years, to stop others from making or selling an invention without permission.
- Rapid prototyping — an older name for using AM-like methods to make quick models of a design before final production.
- Photopolymer — a liquid plastic that hardens when exposed to light of a suitable color, usually ultraviolet.
- Stereolithography (SLA) — the process, coined by Hull, that builds objects by hardening liquid photopolymer layer by layer with light.
- Selective laser sintering (SLS) — a process where a laser selectively fuses regions of a powder bed.
- Fused deposition modeling (FDM) — Stratasys' trademarked name for melting plastic and extruding it through a nozzle; the neutral term is material extrusion.
- Binder jetting — a process where an inkjet-style head deposits liquid binder to glue powder particles together.
- RepRap — an open-source project begun in 2004 (the date given in its own 2011 paper) to create a 3D printer that can print many of its own parts.
- STL file — a file format that describes a 3D surface as a mesh of triangles, created in 1987 for 3D Systems machines.
- Open source — a way of sharing designs or software so that anyone may use, study, modify and share them.
Where this leads
With the story in hand, the next chapter follows one object through the modern digital workflow: From idea to object: the digital workflow.