On the origin of enhanced sensitivity in nanoscale FET-based biosensors

Abstract

Electrostatic counter ion screening is a phenomenon that is detrimental to the sensitivity of charge detection in electrolytic environments, such as in field-effect transistor-based biosensors. Using simple analytical arguments, we show that electrostatic screening is weaker in the vicinity of concave curved surfaces, and stronger in the vicinity of convex surfaces. We use this insight to show, using numerical simulations, that the enhanced sensitivity observed in nanoscale biosensors is due to binding of biomolecules in concave corners where screening is reduced. We show that the traditional argument, that increased surface area-to-volume ratio for nanoscale sensors is responsible for their increased sensitivity, is incorrect.I n recent years, there has been a major drive to use field-effect transistor (FET)-based devices to detect biological molecules in electrolytic environments (1). These biosensors use the charge of biomolecules to gate the current through a transistor (2). Frequently, the transistor is based on a quasi-1D nanostructure, such as a nanowire (NW) or nanotube, and the biomolecules bind directly to the surface of the nanoscale structure (1, 3). The use of such nanostructures is justified by the belief that nanoscale biosensors are more sensitive, with sensitivity defined as the relative change in drain current or a shift in threshold voltage in response to a change in bound biomolecule density. A few experiments specifically studied the effect of shrinking nanowire radii on sensitivity, albeit with varying structures, analytes, and sensing circumstances, and found that shrinking a sensor's dimensions indeed improves its sensitivity (4-6). The enhanced sensitivity has been loosely attributed to the increase in the sensor's surface area-to-volume ratio, which is a direct result of shrinking its dimensions. This argument has been analytically justified in the context of gas sensors (7). However, there is a fundamental difference between gas and biomolecule sensing: biomolecule sensing is performed in an electrolyte, and the ions therein will screen the charge of bound biomolecules in a phenomenon known as Debye screening (8, 9). The direct application of the gas sensing result to the biosensing environment implicitly assumes that the screening effect does not change with shrinking dimensions, an assumption we believe to be false. There have been studies that included a rigorous treatment of screening in biosensors, but they studied neither the specific cause of increased sensitivity at the nanoscale, nor the effect of varying size on screening behavior (10). We believe the phenomenon responsible for the increased sensitivity of nanowires in particular, and nanostructured biosensors in general, have not yet been uncovered by the research community.We have previously dissected the operation of biosensors into two independent parts to better understand the underlying physics (11): first, biomolecule charges cause a change in the local electrostatic potential at the outer surface of the gate dielectri...

Field-effect sensors – from pH sensing to biosensing: sensitivity enhancement using streptavidin–biotin as a model system

“…The commonly stated argument that increased surface area-to-volume ratio increases the normalised change in current 28,36,143,144 is the rationalisation for much of the focus in publications on creating devices with nanoscale dimensions. This argument is still subject to debate, with some authors suggesting that nanoscale dimensions offer increased response via a different mechanism 125,145 or that the argument is not generally applicable to all structures. 145 Operating in the linear region, the SNR was calculated and measured to increase with the square root of the device area for constant device thickness.…”

“…The change in current can be instead normalised by I low instead of I 0 , resulting in the 'unbounded' normalised change in current (I norm and Chui hypothesised that no such limit for biomolecule binding exists because the high affinity of biomolecularligand binding would result in the extent of binding being independent of the surface potential, in contrast to protonbinding where the binding is exponentially related to the potential of the surface. 124,125 The largest shift in threshold voltage for a pH sensor can be calculated based on the Nernstian response of 59 mV per pH. The steepest subthreshold slope for a classical transistor is approximately 59 mV dec −1 .…”

On the capacity of diffusion-based molecular communications with SiNW FET-based receiver

“…where C dl is the diffusion layer capacitance resulting from the double layer created by the medium counterions accumulated at the interface between oxide layer, i.e., SiO 2 layer, and the electrolyte medium; C ox is the capacitance of the oxide layer; and C nw is the SiNW capacitance, which is again a double layer capacitance caused by the accumulation of carriers to the SiO 2 /SiNW interface [12], [13]. For a nanowire-on-insulator (NWoI) configuration of RX, we can consider the SiNW as a hemicylinder with an oxide layer of thickness t ox covering the surface, and a diffusion layer of thickness λ D covering the oxide layer [13].…”

“…For a nanowire-on-insulator (NWoI) configuration of RX, we can consider the SiNW as a hemicylinder with an oxide layer of thickness t ox covering the surface, and a diffusion layer of thickness λ D covering the oxide layer [13]. Accordingly, the diffusion layer capacitance can be given by…”

Two different device physics principles for operating MoS2 transistor biosensors with femtomolar-level detection limits

“…This result indicates that the charges introduced by TNF-a molecules electrostatically affect the free carrier concentration in the MoS 2 channel and induce a shift of the threshold voltage (i.e., DV T measured at the back gate), but these charges do not cause a noticeable change of the ON-state transconductance (g m ¼ @I DS @V G ) (or the field-effect mobility) of the FET. For such insulating-layercoated sensors, the response behavior can be well described by the conventional capacitor model, 14 and a calibrated sensor response quantity (S) is expressed in Eq. (1), 15,16 where I DS(n¼0) is the I DS measured from an as-functionalized sensor (i.e., n ¼ 0) biased under a set of fixed V DS and V G , and FIG.…”