Due to technical problems on the CERN SSO system, logging in is currently failing intermittently.
September 28, 2015 to October 2, 2015
Europe/Zurich timezone

Low-Cost Bump-Bonding Process for High Energy Physics Pixel Detectors

Sep 29, 2015, 6:13 PM
Hall of Civil Engineering (Lisbon)

Hall of Civil Engineering


IST (Instituto Superior Técnico ) Alameda Campus Av. Rovisco Pais, 1 1049-001 Lisboa Portugal
Poster Packaging Poster


Dr Michele Caselle (KIT - Karlsruhe Institute of Technology (DE))


Karlsruhe Institute of Technology (KIT) is one of the module production centres for the upgrade of the CMS pixel detector (phase 1). We will present recent developments of high-density low-cost bump-bonding technologies for both single prototype ICs and mass-production. In addition we will discuss our latest results on high-density copper pillar, gold-stud and Precoat-by-Powder Sheet (PPS) bumping technologies. Furthermore a new low-temperature bonding process for irradiated silicon sensors and Cd(Zn)Te sensors will be presented. Process optimization and production quality control methods will be discussed as well.


In the next generation of collider experiments detectors will be challenged by unprecedented particle fluxes. We will need to build large area arrays of highly pixelated detectors with minimal dead area at reasonable costs. Bump-bonding of pixel detectors has been shown to be a major cost-driver. Karlsruhe Institute of Technology (KIT), Germany, is one of production centres of the upgrade of the barrel pixel detector of the CMS experiment. In this contribution the SnPb bump-bonding procedure will be presented and the production quality control methods and the production yield will be discussed.
In parallel to the production of the new CMS pixel detector, KIT is developing several alternatives to the expensive lithography electroplating/electroless metal deposition technologies.
We present recent progress and challenges faced in the development of bump-bonding technology based on gold-stud bonding by thin (12.5 um) gold wire at KIT. This technique allows producing metal bumps with diameters down of 20 um without using photolithography processes, which are typically required in order to provide suitable under-bump metallization (UBM). The short setup time for the bumping process makes gold-studs bump-bonding highly attractive (and affordable) for the flip-chipping of single prototype ICs which is the main limitation of the current lithography processes.
A second mass-production wafer-level bumping technology based on Precoat-by-Powder Sheet (PPS) processes is currently under development at KIT. This bumping technology is very attractive and could be employed to produce large-area, low-cost pixelated tracking detectors. The process consists of two steps. First micro solder balls, with diameters about 10 um, are transferred from the PPS sheet to the UBM by a thermo-compression process. In a second step the solder balls are reflowed in a spherical shape. A bump diameter size of 44 um is possible on a UBM opening of 30 um with a bump diameter uniformity of the 4%. A process overview is given, the optimization of procedures and results are discussed.
Irradiated silicon sensors and Cd(Zn)Te sensors require a low temperature bonding technology. A novel “quasi-room temperature” bonding process will be presented as an alternative to the low mechanical strength indium technology. The process is based on the flip-chip of dice with two different metal bumps, one “soft” metal bump, like SnPb, deposed in one die and a relatively “hard” metal bump, like gold, on the second die. The bonding connection is established by applying the bond force at room temperature. The resulting structure, with the gold-bump enclosed by the SnPb-bump, results in a strong pull-test force of up to 1.5 grams/bump. The process optimization and the yield are discussed as well.

Primary author

Dr Michele Caselle (KIT - Karlsruhe Institute of Technology (DE))


Alexander Dierlamm (KIT - Karlsruhe Institute of Technology (DE)) Fabio Colombo (KIT - Karlsruhe Institute of Technology (DE)) Marc Weber (KIT - Karlsruhe Institute of Technology (DE)) Simon Kudella (KIT - Karlsruhe Institute of Technology (DE)) Thomas Blank (KIT) Ulrich Husemann (KIT - Karlsruhe Institute of Technology (DE))

Presentation materials