Tuesday, June 19, 2012

Rutherford Lab, Tanner and TowerJazz Announce 200mm Wafer-Sized X-Ray Sensor

Business Wire: Science and Technology Facilities Council’s (STFC) Rutherford Appleton Laboratory (RAL) and TowerJazz have developed 120mm x 145mm X-Ray sensor wich is said to effectively use an entire 200mm wafer. The sensor's focal plane has a size of 139.2 x 120 mm, and has 6.7MP resolution (2800 x 2400 format) with a 50um pixel pitch, 32 analog outputs and also features low noise, a high DR and a programmable region-of-interest readout. Each 3T pixel has low-noise partially pinned photodiode, offering ‘charge-binning’ capability. The sensor offers a very high frame rate of 40fps at full resolution and ‘binned’ images can be read at an increasingly faster rate.

The device has sensing pixels right up to the edges on three sides of the imager. This allows multiple sensors, manufactured on cost-effective 200 mm silicon wafers, to be ‘butted’ or ‘tiled’ together in a 2 x 2 arrangement to form a significantly larger imaging area and to meet the requirements for mammography applications. Additionally, any 2 x N sensor arrangements are possible, thus making the device ideal for applications that demand even larger area coverage, such as chest imaging or security scans.

Tanner is also announcing that the project relied exclusively on its tools.


  1. I wonder.. is the detector the SI diode itself? or do they use the scintillator in front?
    Because if you need the scintillator what is new?

  2. If you use a scintillator, spatial resolution becomes worse.
    So, I guess a pure Silicon detector.

  3. I am tired of people asking what is new...what is new. This may not be intended for a student conference you know?

  4. It is strange, for mammography applications the X-ray energies (correct me if I am wrong) are not lower than 30keV.
    So the detector would have to be thick (in case of Si > 300um) and depleted to have good efficiency.

    I hope that we will see some details of this..

    1. Actually, it is just the reverse. Mammography energies are rarely above 30 keV. However direct detection with silicon is not generally useful in mammography. The 1/e absorption for 300 um is about 13 keV. Thus is is possible to make nice direct detectors for crystallography applications but not for anything much higher in energy. Some end-on silicon detectors have been made for higher energies for physics work but, for example, the 1/e thickness for 25 keV is about 2 mm. In comparison, the 1/e thickness for CsI at 25 keV is only 200 microns and for gadolinium oxysulfide, only 60 microns (at typical 30% density). Selenium has also been used as a direct converter but for that, no silicon photodiodes are needed.

  5. Unless you use very high resistivity Si, it is just too transparent at mammography energies (20 - 30 keV). Also you generate so many e-h pairs per detected photon - about 10,000 for one 25 keV photon. So unless you read very quickly then detector is saturated. Also, no mention of this device being radiation-hard so would quickly die in a mammography beam - hence scintillator plus glass fibre - optic plate.
    BTW - There are quite a few wafer-scale imagers announced now and some in production. With one device per 200 mm wafer, yield is everything.


All comments are moderated to avoid spam.