3D Vertical Scale Integration vs 2D CMOS technology

Detectors for particle tracking can be revolutionized by 3D-IC technology [1]. The advantage of this approach is that different 3D‑IC vertical levels (called tiers) can be individually manufactured and optimized, by independently fabricating 2D process, and then bonded together after precise alignment and thinning and interconnecting them through deep metal vias known as through silicon vias (TSV).

Theoretically, heterogeneous wafers, i.e., from different foundries or even different process families may be combined, optimally distributing tasks like photon sensing, analog amplification, and digital processing. The sensor layer may be tailored specifically to the needs dictated by the radiation to be detected.

Factors affecting sensor optimization include material, pixel granularity, and the choice of front or back side illumination. Each pixel can then be equipped with read-out electronics comprising tens or hundreds of transistors distributed on separate tiers.

A fully processed sensor layer (made with high resistivity silicon for operation in depletion) can be attached to a 3D multi-tier read-out circuit in a separate step. Thus, a side benefit of the 3D-IC development is emergence of die-to-wafer or die-to-die fusion bonding techniques that may be used as a replacement of In or PbSn bump bonding.

Vertical integration technologies have already become quite popular among IC designers, as they can alleviate some important performance limitations correlated with CMOS feature size scaling. They are already widely used in the design of high density storage devices and promise to provide a means to overcome the bandwidth bottleneck in modern microprocessors by vertically integrating processor and memory subsystems in a single chip.

Another possible application of 3D-IC technology, proposed by the INFN Perugia group, is for realizing particle tracking detectors based on CMOS Active Pixel Sensors (APS) layers, monolithically integrated in a all-in-one chip featuring multiple, stacked, fully functional detector layers capable to provide momentum measurement (particle impact point and direction) within a single detector [2], [3]. This will results in a very low material detector, thus dramatically reducing multiple scattering issues.

In particular, instead of using different tiers of the stacked 3D structure for heterogeneous integration (namely, by devoting different tiers to the sensing layer, and to the analog and digital circuitry), identical fully-functional CMOS APS matrix detectors, including both sensing area and control/signal elaboration circuitry, could be stacked in a monolithic device by means of TSV connections.

The information coming from thinned multiple stacked layers could be usefully exploited to extend the detection capability of the monolithic sensor. In principle, such a detector would be capable of giving accurate estimation not only of the impact point of a ionizing particle, as well as of its incidence angle. A single detector allowing particle momentum measurement therefore could be built, at the same time being a low material detector: multiple scattering effects are expected to be negligible with respect to conventional sensors, since incoming particles have to cross only few micrometers of bulk silicon.

References

1. P. Garrou, C. Bower and P. Ramm, Handbook of 3D Integration, Wiley-VCH, (2008).
2. D. Passeri, L. Servoli, S. Meroli, Analysis of 3D stacked fully functional CMOS Active Pixel Sensor detectors, JINST 4 (2009) P04009.
3. 1.22 D. Passeri et al., 3D monolithically stacked CMOS active pixel sensor detectors for particle tracking applications, JINST 7 (2012) C08008 doi:10.1088/1748-0221/7/08/C08008.