DarkSide-20K: The Search for Dark Matter
21 Jul 2023



Scientists work hard to increase our understanding of the universe we occupy. We look out at the cosmos and amongst the beauty, we see physics in action at a truly impressive scale.



A wafer of the silicon photo-multiplier devices made 
for Darkside 20k 

However, there is much about these observations which cannot be explained, unless a lot more matter exists than we can currently see.

An international collaboration is building a huge detector 1.4km underground in Gran Sasso, Italy, to broaden the search for this unseen matter. It is widely believed that the reason we can't yet see it is because it is 'dark'. That is, it does not emit, absorb or reflect any form of electromagnetic radiation. The very darkness of this matter makes it very difficult to observe – you certainly can't do it with any conventional telescope. ​​

Any interactions that would indicate the presence of dark matter are very scarce, creating a problem that is worse than looking for a needle in a haystack. It is more like looking for a needle in a haystack that is buried by every other haystack that has ever existed. The underground facility at the Gran Sasso National Laboratory (LNGS) has been home to many physics experiments since it opened in 1985. They take advantage of the low background radiation environment to improve their sensitivity. Removing the background radiation is like removing all the additional haystacks, meaning that anything left is far more likely to be the needle, although still hidden in its own haystack.

An array of silicon photo-multipliers assembled 
together into a veto sensor (vTile+) 
However, even when operating in a low-radiation environment, there are a lot of other sources that can still mask our ability to find the needle. The Darkside 20k detector employs additional radiation sensors which will allow reliable identification of these unwanted strands of hay. They can then be removed from (or vetoed out of) the data stream which needs to be analysed to identify candidates which may be indicative of dark matter. At this point, it becomes much more possible to find the needle!

STFC​ Interconnect plays a key role in this international effort. We are making half of the sensors that are used to make these veto decisions. The veto sensors consist of arrays of incredibly sensitive silicon photomultipliers. The devices are so sensitive that they can identify when a single photo-electron is generated by light impinging on the sensor.

The measurements and environment where these devices operate dictate that they too have a part to play in not adding to the radiation background. It would be pointless going to all these measures to shield out external radiation if the sensor electronics inside the detector resembled a radioactive source! So radiopurity of every component and all the build and storage environments also have a key role to play. Every component and subassembly has undergone extensive radio assay screening, and all the work and storage areas have been radon screened to ensure that no avoidable radiation sources are built-in or plated onto the assemblies we make.

The experiment will be left running in an inaccessible state for many years, so during the pre-production phase, we've placed a strong emphasis on root cause analysis, to identify and eradicate any yield and reliability problems encountered. The emphasis on quality has also required us to build dedicated test benches. These make use of standard test equipment adjacent to the build environment. The test benches have been automated using Python to ensure consistency and scalability to the planned production volumes.

Test system used to establish that the sensors work correctly before
they leave our lab ​

The sensor production utilises many of our mainstream assembly processes, such as solder paste deposition, solder-based die-attach, micro-alignment during die attach, aluminium wedge wire bonding and metrology. The production is running through 2023 and should complete by Autumn 2024. The sensors will then undergo further testing at cryogenic temperatures before getting installed in the detector, which is scheduled to run for 10 years from 2026. We look forward to seeing the scientific discoveries that we are helping enable!

If you are interested in learning more about DarkSide, watch "DarkSide-20k: an experiment for direct Dark Matter detection" on YouTube.​

​Written by John Lipp, Interconnect ​group leader.​