Interpixel capacitance in nondestructive focal plane arrays

Abstract

Inter-pixel capacitive coupling can exist in a non-destructive detector array if the detector nodes change voltage as they integrate charge and the design of the device allows for an electric field to exist between adjacent collection nodes. Small amounts of inter-pixel capacitance can cause large errors in the measurement of poissonian noise versus signal, and all subsequently derived measurements such as nodal capacitance and quantum efficiency. Crosstalk and MTF can also be significantly influenced by interpixel capacitance. Two 1k × 1k Raytheon SB226-based hybridized silicon PIN arrays were tested for nodal capacitance and MTF. Initial results indicated unexpected and unexplainably large nodal capacitance, poor MTF, and odd edge spread. It was hypothesized that inter-pixel capacitive coupling was responsible for these discrepancies. A stochastic method of measuring the coupling using 2D autocorrelation and Fourier Transform techniques was devised and implemented. Autocorrelation of the shot noise in the images revealed a correlation consistent with 3.2% interpixel capacitive coupling. When the effects of the measured interpixel capacitance were taken into account, the initially measured nodal capacitance of 56 fF was found to be 30% higher than the corrected nodal capacitance measurement of 43 fF. Large discrepancies between the theoretical and observed edge spread response were also greatly reduced. A simulation of the electric field in the PIN detector intrinsic region predicted an interpixel coupling very close to the observed coupling. Interpixel capacitance was also observed in 2k × 2k Raytheon SB304-based InSb detector arrays, but was not strongly evident in a bare Raytheon SB226 multiplexer.

Development of sensitive long-wave infrared detector arrays for passively cooled space missions

“…Thus the determined nodal capacitance for H1RG-16885, which is not epoxy back-filled, must be reduced by the factor (1 þ 8α), or ∼10%. 26 The epoxy will increase this factor to ∼13% with backfill. NEOCam arrays will be epoxy backfilled and substrate removed.…”

“…26 It is important to note that this device is not epoxy backfilled: for use in space, the devices must be epoxy backfilled so that substrate removal can take place. The epoxy backfill impacts the value for the IPC.…”

NIR Detector Nonlinearity and Quantum Efficiency

“…The variance in the recorded signal, in units of electron counts squared, then equals the mean signal in electron counts plus an offset due to other noise sources. 3 The matter is complicated in the presence of capacitive coupling between adjacent pixels (Moore, Ninkov, & Forrest 2004;Finger et al 2005;Moore 2006;. Adjacent pixels share a small fraction of the difference in their charges.…”

Space qualification and performance results of the SIDECAR ASIC

“…The typical conversion gain of a HgCdTe detector hybridized to a HAWAII-2RG and measured with the variance method (noise^2) is somewhere between 3 and 4µV/e-, sometimes even above 4µV/e-. Taking the effect of the interpixel capacitance into account, which causes a systematic underestimation of the shot noise, the true conversion gain is about 20% higher (Finger et al [4] and Moore et al [5]). The actual conversion gain therefore spreads from about 3.6µV to over 4.8µV/e-, and the mid-level value is at 4.2µV/e-.…”