For CCD imagers, there are three main sources of dark current: (1) thermal generation and diffusion in the neutral bulk, (2) thermal generation in the depletion region, and (3) thermal generation due to surface states at the silicon-silicon dioxide interface. Of these sources, the contribution from surface states is the dominant contributor for multiphase CCDs. Dark current generation at this interface depends on two factors: the density of interface states and the density of free carriers (holes and electrons) that populate the interface. Electrons that thermally “hop” from the valence band to an interface state (sometimes referred to as “a mid-band state”) and to the conduction band produce a dark e-h pair. The presence of free carriers will fill interface states and, if the states are completely populated, will suppress hopping and conduction and substantially reduce dark current to the bulk dark level. Normal CCD operation depletes the signal channel and the interface of free carriers, maximizing dark current generation. Under depleted conditions, dark current is determined by the quality of the silicon-silicon dioxide interface or the density of mid-band states.
In MPP technology, dark current is significantly curtailed by inverting the signal-carrying channel by populating the silicon-silicon dioxide interface with holes that, as mentioned above, suppress the hopping conduction process. In fact, MPP CCD technology has achieved dark floors of 25 pA/square-cm, a factor 400 times lower than non-MPP CCDs.
MPP mode is applied to the CCD by significantly biasing the array clocks negatively to invert the n-buried channel and “pin” the surface potential beneath each phase-to-substrate potential (hence the name “multi-pinned phase”). Biasing the array clocks in this manner causes holes from the p+ channel stops to migrate and populate the silicon-silicon dioxide interface, eliminating surface dark current generation. Unfortunately, when inverting conventional CCDs, the sensor’s full well capacity is annihilated since the potential wells within a pixel all assume the same level. This condition results in severe blooming up and down the signal-carrying channel, since there is no preferential location for charge to collect. To circumvent this difficulty in MPP CCD technology, a weak implant is employed beneath the phases during the fabrication of the sensor. The extra implant creates a potential difference between phases, allowing charge to accumulate in collecting sites when biased into inversion.
There are additional advantages of MPP CCD technology besides eliminating surface dark current. For example, the charge-transfer efficiency of a CCD generally degrades with decreasing operating temperature. Therefore, MPP technology can also assist in the charge-transfer process since it permits the use of higher operating temperatures. The MPP CCD also eliminates residual image, a serious problem that has plagued low-signal-level CCD users for many years. Residual image results when the sensor is either overexposed or when a CCD camera is first powered up. Under these circumstances, electrons are found trapped at the silicon-silicon dioxide interface that slowly de-trap into the pixel’s potential well during the course of an exposure. For very cold operating temperatures (-120°C), residual charge may take hours or even days (depending on how long the sensor integrates) before the level of residual charge seen is lost into the sensor’s read-noise floor. Inverting the CCD causes holes to immediately recombine with the trapped residual electrons, eliminating remnant image effects during integration as well as readout. The implications of the successful implementation of MPP technology are far-reaching and, for many applications that formerly required the use of cryogens (liquid nitrogen) or other bulky and complicated mechanical refrigeration schemes, CCDs can now use thermoelectric cooling.