- Posted by Riccardo Gnudi
- On September 3, 2020
- 0 Comments
ALPES LASER – Defect categorization, management and yield increase in high variant production
Interview to Tobias Gresch, Process Design Manager at Alpes Laser SA.
What is the core activity of your company and which technology(ies)/sector(s) is(are) mainly developed/involved in the iQonic project?
Alpes Lasers SA is a Swiss company specialized in the manufacturing of long-wavelength semiconductor lasers, in particular the quantum cascade laser (QCL). These are semiconductor lasers emitting in the wavelength range between 4 and 13um with a broad spectrum of applications in spectroscopy in the case of single-mode DFB devices. Another important market, for high-power multi-mode emitters is in defense. Our involvement in the iQonic project is as an end user of iQonic technology with test use cases in the back-end part of the manufacturing including wafer separation, die-bonding and testing.
Tobias, can you please describe your company’s use case and the main objectives you’re pursuing through iQonic?
We are involved in three use case scenarios concerning the back-end part of the laser manufacturing. The first one concerns the optimization of the chip-separation strategy based on observations and inspections during the manufacturing process. In the second use case scenario we will evaluate whether an automated die-bonding machine can be used to mount our chips efficiently. In this process the laser chips are soldered on a ceramic heat-spreader that also serves as landing zone for the wire-bonds. The difficulty is that the chips we mount vary in geometry and properties for almost every client. Therefore a highly flexible tool is required where the benefits (reliability, precision, …) overcome the additional effort for setting-up the machine for very small number of devices. Lastly, the third use case scenario involves accelerated lifetime testing.
What do you expect from the iQonic project?
Our goal is to improve our overall production yield. The expectations we have from the project are manifold. On one hand the project allows us to gain some experience with an automated assembly machine and to evaluate whether such a tool is suitable in our production chain that is optimized for heavy customization at moderate volumes. From the iQonic technologies we are involved in, we hope to be able to locally improve our efficiency and yields as well as simplifying inspection data collection and analysis and information flows.
Which activities have you carried out within the project?
As iQonic end-user partners we were involved in the definition of the different use case scenarios. We were also the main author for deliverable D3.1 that describes optoelectronic functionalities (public deliverable soon available in the “Deliverables and Publications” section). Furthermore, we provide information and devices to different iQonic partners. For example, we sent a HHL-packaged QCL device to the Politecnico of Milan (POLIMI) who will study its disassembly and create models based on that. These models then will be used to define strategies for re-working this kind of packages in case of failure. Internally, we have developed a setup for accelerated lifetime testing that will allow to spot lifetime affecting defects before devices are being mounted.
Which partners are you collaborating with the most and how?
Our collaborations change over time as the project advances. Currently, we are collaborating with Polimi in the scope of WP5 – Defect Life-Cycle Management, Dissassembly and Remanufacturing, on the cyber physical system and reverse assembly and we are establishing a data exchange channel for iQonic middleware with Atlantis Engineering. We planned to perform some assembly steps — in particular visual facet inspection and die-bonding — for a series of our devices in ficonTEC‘s facilities to evaluate how one of their assembly machines performs in our highly variable and client-specific production line.