Abstract:
In recent years, nanowire (NW) based ultrasensitive sensors have been widely investigated for the potential
of real time, high sensitivity and label-free detection. Among different nanowire materials, silicon has the
potential advantage of compatibility with very large scale integration (VLSI) and complementary metal
oxide semiconductor (CMOS) technologies. These benefits are some of the main reasons for the significant
interest in silicon nanowire based sensors with quite a large number of studies on the detection of analytes in
aqueous environment, mainly within the context of biosensing (e.g., to detect biological species like DNA,
proteins and viruses etc.). However, commercial silicon NW based sensors are still unavailable for analyte
detection even in aqueous environment due to the difficulty in device manufacturing processes, reproducible
sensing/integration issues and most importantly, due to the unavailability of economically viable route for
mass fabrication. Silicon nanowire fabrication platform comprises bottom up, top down and spacer etch
processes where conventional top down/bottom up processes usually realize sensors using single crystal
silicon material whereas spacer etch process realizes sensors on polysilicon material. The choice of material
is application/facility dependent and each of these material platforms have own advantages/disadvantages.
The performance of nanowire as biosensor is inherently dependent on the choice of materials and also on
nanotechnology variables like nanowire thickness, doping etc. A rigorous study of the effects of nanowire
thickness/doping on the performance of siliconnanowire based sensors are rare in the literature and there is
no study available on the critical comparison of the single crystal and polycrystalsilicon nanowire biosenors.
We study for the first time the effect of nanowire thickness and doping concentration on the electrical
characteristics ofsingle crystal and polycrystalline silicon nanowire biosensors and compare the performance
of single crystal/polycrystal silicon nanowire biosensors to achieve a performance benchmark of sensors
realized in these two material platforms. The intention is to appraise in depth the choice of nanotechnology
variables for chosen material platforms for appreciable sensing. For nanowire thicknesses of 100 nm and 75
nm, a plausible sub-threshold slope around 100 mV/decade for a viable biosensor operation is achievable
only if doping concentration is 2×1016/cm3 or below both for single crystal and poly Si nanowires. For
a 50nm nanowire thickness a relatively wide doping concentration range with a maximum doping up to
4×1017/cm3 is viable for biosensor design while maintaining decentsub-threshold characteristics. The widest
range of doping concentrations can be chosen for 25nm and 10nm nanowire thicknesses with a maximum
doping up to 1018/cm3 for feasible biosensor design using single crystal and polycrystalline silicon nanowires.
In general poly Si NW shows inferior characteristics than single crystal Si NW. However, for 10nm, Si NW
single crystal & poly Si NW show same sub-threshold slopes at all doping densities. Considering the fact
that spacer etch process provides the cheapest & mass manufacturable platform for biosensor fabrication
using poly Si material in comparison to the available single crystal platforms, it can be decided that poly Si
NW biosensor with Si thickness ≤ 10nm is the possible commercial route of sensor fabrication with similar
performance like single crystal silicon nanowires.