Low frequency noise in heavily doped polysilicon thin film resistors

Abstract

Low frequency noise and current-voltage measurements in several heavily doped polysilicon resistors of varying geometry and both p and n type, and over a limited range of temperatures from Ϫ60 to 50°C were conducted for the first time. We found that the noise in p-type polysilicon was independent of temperature, but not the n-type polysilicon. For the p-type resistors, linear currentvoltage characteristics were observed, and the relative noise spectral density was independent of bias and inversely proportional to frequency. For the n-type resistors, linear current-voltage characteristics were observed, and the relative noise spectral density was independent of bias. Finally, the normalized noise level in the linear n-type resistors was almost an order of magnitude lower than for the p-type resistors. We believe that this difference is because n-type dopants segregate to the grain boundaries, thus passivating some of the traps there. Boron (p-type dopant͒, on the other hand, does not segregate to the grain boundaries, leaving more unpassivated grain-boundary traps which capture and emit more carriers, resulting in more low frequency noise.

Exploitation of spatiotemporal information and geometric optimization of signal/noise performance using arrays of carbon black-polymer composite vapor detectors

“…A standard method was used to determine the noise of the detector films [21,24]. Briefly, the films were placed into a metal box and were biased with a stack of batteries (18 V total) that was connected in series to a 1 MO resistance.…”

3-D Integration and Through-Silicon Vias in MEMS and Microsensors

“…With high frequency signals, high resistances can cause significant RC delay and signal decay; at high currents, high resistance will cause large power consumption and heat generation. In addition to resistivity, low frequency noise may be another consideration affecting polysilicon TSVs [121], [122]. It has been found that p-type polysilicon shows a higher noise level than n-type polysilicon, possibly due to a larger number of unpassivated grain-boundary traps [121].…”

A Thin-Film Cochlear Electrode Array With Integrated Position Sensing

“…An LT1057 JFET input high-precision amplifier was used for the instrumentation amplifier, which had a measured input circuit noise of 10 −9 -10 −8 V 2 /Hz [21]. Polysilicon thin-film resistors (220 kΩ) have been reported to have a 1/f noise of 10 −10 -10 −7 V 2 /Hz [22], [23], which far exceeds their thermal noise. The peak-to-peak noise level of the amplified sensor signal is on the order of a few millivolts when the gain is 30.…”

“…These performance limitations are imposed by the difficulties in achieving deep insertion in the scala tympani of the cochlea (Fig. 1), by the small number (16)(17)(18)(19)(20)(21)(22) of wires that can be accommodated in the scala tympani, and by the current spreading that occurs due to the physical separation between the sites and the receptors in the modiolus. Ultimately, frequency resolution will likely be determined by this separation and by the effectiveness of multiple sites in shaping the stimulus currents.…”

“…(a) Use of a closed-loop position-sensing and positioning-control system consists of four steps-sensor calibration, position sensing, shape recovery, and feedback to the implant tool, and (b) a pneumatically actuated device made of PET tubes[22] is attached to the back of a thin-film electrode array and functions as its implant tool.…”