As many of you know, RD39 has planned to use silicon microchannel cooling for extracting the heat of readout electronic chips on their cryogenic detector modules. Silicon microchannels had been already used much earlier for cooling various electronic devices at cryogenic temperatures, but never for the cooling of pixel electronics, for example.

We have been recently in contact with a group which has developed substantial expertise in micromachined channels buried in silicon. They are interested in setting up a collaboration and submitting a proposal to EU for a programme to develop the microchannel cooling techniques for various electronic applications in a wide range of temperatures. This would combine their experience in silicon micromachining, with the experience gained by the RD39 institutes in two-phase microchannel flow and heat transfer. Institutes involved in pixel detector design and construction, and interested in the potential of the technique, would also be welcome to join.

to discuss the formation of a collaboration and the proposal to be submitted. We would like to invite you and your colleagues who might be interested in such a low-mass cooling technique, to attend to this one-day event. The programme will consist of a few review talks on the relevant topics, followed by round-table discussions and a planning session.

There are several outstanding problems which need study and R&D, before reliable silicon microchannel cooling systems can be built. Among these are the pipe conections to silicon, and the fluid delivery systems, both of which are presently on the level of ideas only . The hermeticity of the microchannel is also a topic of interest.

The main application is the acceleration of very fast DSM CMOS electronics by cooling to a low temperature. Not only do the gates flip faster, but also the signals propagate in the metal lines with smaller rounding and losses. Both of these entail lower power dissipation. A secondary benefit comes from the noise reduction due to the lower temperature of the analog parts of the circuits.

In RD39 the interest is in the cooling of the pixel detectors for LHC after the first-generation vertex detectors are phased out, and in particular after its luminosity upgrade. We should stress that the principle can be applied equally well at room temperature, at -20 °C, and all the way down to temperatures below 100 K.