top of page

Oliver Haupt

Coherent

* All members of the platform can watch the entire presentation.

 

Please register to become a member.

Oliver Haupt | Coherent: If up to 80% of microLED display cost is material, what can a laser toolmaker do to cut cost per display?

04:07 - 05:27

Other snippets from this talk

Summary of the clip:

If up to 80% of microLED display cost is material, what can a laser toolmaker do to cut cost per display?

Display makers put material at up to 80% of microLED display cost, and the compound-semiconductor side is not where an equipment maker can work. The lever discussed is different: an EP wafer costs more or less the same, so cost per display depends on how many displays can be made from one wafer. That pushes die size down and forces process work to be judged by wafer utilisation.

The clip notes many small attempts to work on processes, but argues that cost will never really come down without mass production. The two requirements have to come together: smaller die to raise displays per wafer, and a manufacturing route scaled enough to make the process economics work. That combination is the stated route to cost competitiveness.

The scope described is equipment for back planes, not back-plane production itself. The open question is whether that is enough, or whether something else must be done now to support microLED volume manufacturing. The answer implied is that tool and process development must connect to mass production rather than remain isolated attempts, because cost competitiveness depends on the whole chain coming together.

In this short video, you can learn:
* Display makers say material accounts for up to 80% of microLED display cost, which a compound-semiconductor supplier cannot address directly.
* Because EP wafer cost stays roughly the same, cost per display hinges on how many displays can be made from one wafer.
* The argument is that smaller die and mass production must arrive together, not as isolated process attempts, for microLED cost to become competitive.

📋 **Clip Abstract** The clip frames microLED cost as a material-dominated problem—up to 80% according to display makers—that cannot be solved on the compound-semiconductor side. It argues the equipment route is to increase displays per wafer through smaller die and to combine process work with mass production so the cost can become competitive.

About the speaker:
* Speaker: Oliver Haupt
* Company: Coherent
* Event: MicroLED Connect 2024
* Location: High Tech Campus, Eindhoven, Netherlands

Join the most event MicroLED, AR/VR display and Optical I/O Connect event of the year
Learn more https://www.microledconnect.com/

#MicroLEDCost, #CompoundSemiconductor, #EPWafer, #DisplaysPerWafer, #MicroLEDDisplays, #DisplayManufacturing

This is a highlight of the presentation:

MicroLED Display Volume Manufacturing Enabled By Laser Technology.

MicroLED Connect 2024

​

25-26 September 2024

High Tech Campus, Eindhoven, Netherlands

Organised By:

Metal Grass and TechBlick

​

More Highlights from the same talk.

07:27 - 08:38

Can laser lift-off shrink from 20–40 minutes per wafer to a few seconds with a 248 nm line beam?

Can laser lift-off shrink from 20–40 minutes per wafer to a few seconds with a 248 nm line beam?

Customers already know how to handle shorter wavelengths in display fabs, so the discussion moves to mass production and productivity. Several attempts exist for laser lift-off. One approach uses a single, very small laser spot and wobbles the full wafer; that can take 20, 30 or 40 minutes to lift off one wafer. This is the productivity baseline the next laser approaches are measured against.

A small-field-sized laser can do the job in minutes, cutting the lift-off time versus the single-spot, full-wafer wobble. For the next productivity step, line-beam technology—already used for annealing—has been converted to 248 nanometres, a shorter wavelength, and can process a wafer in a few seconds. That minutes-to-seconds change is how laser lift-off is framed for industrialisation.

Getting the laser beam to the workpiece is a separate problem for a laser manufacturer. Different ways to do this are identified, without detail in the clip. That beam-delivery question sits alongside the lift-off productivity comparison: 20–40 minutes for single-spot full-wafer wobble, minutes for a small-field-sized laser, and a few seconds for the 248 nm line beam.

In this short video, you can learn:
* A single small spot wobbled over the full wafer can take 20, 30 or 40 minutes for one wafer's lift-off.
* A small-field-sized laser cuts lift-off to minutes, and line-beam technology converted to 248 nm can process a wafer in a few seconds.
* Beam delivery to the workpiece is a separate laser-manufacturer task with different ways to get the beam there.

📋 **Clip Abstract** In display fabs, laser lift-off productivity ranges from a single small spot wobbled over the full wafer at 20–40 minutes per wafer to a 248 nm line beam that can process a wafer in a few seconds. A small-field-sized laser sits in between at minutes, while getting the laser beam to the workpiece remains a separate beam-delivery problem.

About the speaker:
* Speaker: Oliver Haupt
* Company: Coherent
* Event: MicroLED Connect 2024
* Location: High Tech Campus, Eindhoven, Netherlands

Join the most event MicroLED, AR/VR display and Optical I/O Connect event of the year
Learn more https://www.microledconnect.com/

#LaserLiftOff, #LineBeamLaser, #SmallFieldLaser, #BeamDelivery, #MicroLEDDisplays, #DisplayFabs

16:16 - 18:41

How small and how tightly spaced can microLEDs get before a laser can't pick one without touching its neighbours?

How small and how tightly spaced can microLEDs get before a laser can't pick one without touching its neighbours?

Cost pressure pushes microLEDs smaller, and material utilisation is the stated driver. A five-micron emitter held at a five-micron gap leaves roughly 60% or more of the wafer as unused area — only about 30% of the material is doing work. Keeping a five-micron pitch but shrinking the emitter to one micron lifts active area to about 75% or more.

Shrinking the emitter is only half the problem: the talk explicitly invites compound semiconductor companies to raise EQE so devices can keep getting smaller. Meanwhile, the display makers surveyed are converging on laser mass transfer as the most promising route for small microLEDs at high throughput — the volume-manufacturing argument the whole cost case rests on.

The experiment mimics tighter gaps — five, four, three, two, one and finally 0.5 micron — and then, in a second round, targets only the middle microLED. Selection held all the way down to 0.5 micron without disturbing neighbours, an explicit placement-accuracy demonstration. The conclusion: the industry must pursue small emitters and small gaps together, and that combination is a good job for laser.

In this short video, you can learn:
* A five-micron emitter on a five-micron gap leaves 60% or more of the wafer unused.
* Shrinking emitters to one micron at the same pitch raises active area to about 75% or more.
* Laser mass transfer isolated a single middle microLED at 0.5-micron gaps without touching neighbours.

đź“‹ **Clip Abstract** MicroLED cost is governed by how much wafer area actually emits light, and the talk quantifies the penalty of generous spacing: a five-micron emitter on a five-micron gap uses only about 30% of the material. Pushing emitters to one micron at the same pitch lifts that to roughly 75%, and laser mass transfer was shown selecting a single middle device at gaps down to 0.5 micron without disturbing its neighbours.

About the speaker:
* Speaker: Oliver Haupt
* Company: Coherent
* Event: MicroLED Connect 2024
* Location: High Tech Campus, Eindhoven, Netherlands

Join the most event MicroLED, AR/VR display and Optical I/O Connect event of the year
Learn more https://www.microledconnect.com/

#LaserMassTransfer, #MicroLEDSpacing, #HalfMicronGap, #WaferUtilisation, #MicroLEDDisplay, #CompoundSemiconductors

13:36 - 15:29

Serial transfer needs 2,400 seconds for a TV — how fast can parallel laser transfer realistically go?

Serial transfer needs 2,400 seconds for a TV — how fast can parallel laser transfer realistically go?

Serial processing uses a single laser beam or needle, transferring microLEDs one by one. At typical single-spot rates of 10 kHz or below, the arithmetic is brutal: roughly 600,000 microLEDs, about 9 million at 30–60 µm smartphone pitch needing 900 seconds, and 24 million for a TV needing 2,400 seconds.

Parallel processing changes the field interaction: a stamp or laser exposes a field and transfers many microLEDs in one shot. With the optical setup and appropriate laser sources, the stated path reaches up to 50 or 5,000 microLEDs per second; at TV-scale 400 µm pitch, reducing pitch keeps field size constant while covering more devices.

The net-time comparison is the argument for parallel laser transfer. The same 600,000 devices can be done in half a second, against the serial route’s hundreds or thousands of seconds. That is why the focus is on parallel processing: it has more potential, with movement speed of donor and receiver still a practical variable to keep in mind.

In this short video, you can learn:
* Single-spot laser processing at about 10 kHz implies 2,400 seconds net transfer time for a 24-million-microLED TV.
* Parallel stamp or laser processing exposes a field and transfers many microLEDs in one shot instead of one by one.
* The stated parallel path reaches up to 50 or 5,000 microLEDs per second, making 600,000 devices a half-second net-time calculation.

đź“‹ **Clip Abstract** The clip compares serial and parallel microLED transfer using net-time arithmetic: single-spot 10 kHz processing implies 900 seconds for a smartphone and 2,400 seconds for a TV. Parallel stamp or laser fields, with up to 50 or 5,000 microLEDs per second cited, reduce 600,000 devices to a half-second net-time calculation and explain the focus on parallel processing.

About the speaker:
* Speaker: Oliver Haupt
* Company: Coherent
* Event: MicroLED Connect 2024
* Location: High Tech Campus, Eindhoven, Netherlands

Join the most event MicroLED, AR/VR display and Optical I/O Connect event of the year
Learn more https://www.microledconnect.com/

#SerialLaserTransfer, #ParallelLaserTransfer, #MicroLEDPitch, #NetTransferTime, #MicroLEDDisplays, #LaserProcessing

More Snippets
CONTACT US

KGH Concepts GmbH

Mergenthalerallee 73-75, 65760, Eschborn

+49 17661704139

admin@techblick.com

TechBlick is owned and operated by KGH Concepts GmbH

Registration number HRB 121362

VAT number: DE 337022439

  • LinkedIn
  • YouTube

Sign up for our newsletter to receive updates on our latest speakers and events AND to receive analyst-written summaries of the key talks and happenings in our events.

Thanks for submitting!

© 2026 by KGH Concepts GmbH

bottom of page