Contact potential measurement: Spacing-dependence errors

Abstract

We examine the causes of spacing dependence of the nulling bias voltage in the vibrating capacitor contact potential measurement technique. In addition to effects already recognized in the literature, namely, nonuniform work functions, nonparallel surfaces, fringe fields, and capacitive coupling to distant surfaces, we investigate the effects of finite gain and spurious signals in feedback loop systems. We argue that much of the spacing dependence reported in the literature may be due to microphonic signals, which are very difficult to eliminate. We also discuss the means by which existing spacing dependence can be minimized.

Surface photovoltage phenomena: theory, experiment, and applications

“…However, many practical designs are subject to geometrical, ease of handling, or other limitations which preclude such arrangements. A particularly problematic type of driver frequency noise is contributed by microphonics, which is a general name for ac electrical signals produced as a result of the vibration of conductors and insulators [176]. Since a maximal ac current is obtained when the ac capacitance is maximal, the modulation index should be large.…”

“…Thus, a signi®cant effort has been devoted to eliminating any sources of mechanical vibrations [130,133,141,175,178± 180]. Noise contributions at the driver frequency may be modeled as parasitic ac current and voltage sources, which are connected in parallel and in series, respectively, to the vibrating capacitor [176]. These prevent the ac current from being perfectly nulli®ed and produce a systematic error in the CPD reading.…”

“…These prevent the ac current from being perfectly nulli®ed and produce a systematic error in the CPD reading. Worse, the relative magnitude of the error term with respect to the real one, and therefore the CPD reading, may become spacing dependent [176].…”

“…The stray capacitance of the shield itself is eliminated if the shield and probe are made of the same material [160] or if an auxilliary shield bias is used [139,181]. Many different shield designs have been proposed (see, e.g., [131,150,176,179,184,188,189]). …”

“…Addition of a deliberate dc offset inside the lock-in ampli®er has been shown to compensate for microphonicsinduced spacing-dependence [176]. Stray capacitance-induced spacing-dependence has been compensated for analytically by extracting the stray coef®cients and adding a correction term [161].…”