Solid State Nanopores for Selective Sensing of DNA - Nanopores: Sensing and Fundamental Biological Interactions - page 112

and deposition processes to reduce the diameter of SSNs. Danelon et al. had drilled a
nanopore of 50-80 nm in a thin Si 3 N 4 membrane, and they then reduced it to less than
10 nm by applying electron-beam-assisted silicon dioxide deposition to the selected
pore area [ 13 ]. The gas precursor, tetraethyl orthosilicate (TEOS) decomposed by
electron beam and deposited all over and inside the pore surface, resulted in reduction
of the pore diameter. The typical values used for the electron beam voltage and
current were 10 KV and 130 pA respectively. The TEOS reservoir was maintained at
a constant temperature in order to have stable gas pressure inside the system. This had
ensured a constant flow rate of gas and a controlled deposition.
Nilsson et al. had used the same procedure as used by Danelon, with H 2 O vapor
introduction to increase the deposition rate [ 14 ]. However, Nilsson et al. used ion-
beam-assisted deposition of SiO 2 instead of using electron-beam. They successfully
reduced the pore diameter to 25 nm. Harrell et al. used a single, conically-shaped
gold tube embedded inside a polymer membrane [ 15 ]. They reduced the diameter
of the pore via thiol-modified linker. Similar attempts had been made by Chen
et al. [ 16 ]. They used atomic layer deposition to fine tune the surface properties
and reduce the diameter of the pore. Alumina layers were deposited one by one and
each cycle required only 12 s; it was proven to be a very controlled process.
Alumina coating also had a few additional advantages: it covered the surface
defects made by FIB, it eliminated the surface charge, and it caused a reduction
of 1 / f noise during IV measurements.
5.2.1.1 TEM Shrinking
As reported by Heng et al. [ 17 ] TEM was used to fabricate a single nanopore in
thin membranes of tens-of-nanometers thickness. The same TEM was used to
shrink the SSN diameter. When the electron beam is focused on a small membrane
area, it melts the surface and drills a hole in the membrane. When the same beam is
focused on a larger area, it shrinks the pore due to controlled surface melting.
Dekker and co-workers at Delft university also used TEM to reduce the pore
diameter and then in other reports to drill pores in Si 3 N 4 to single-nanometer
precision [ 3 ]. The advantages of their process are in situ visualization of the pore
diameter and the fact that the composition of the material remains the same after
the process. After TEM melting, when the desired morphology was obtained, the
control to lower the electron beam intensity was established; the soft material
solidified immediately upon doing this.
They reported that the pore would shrink only when the pore radius satisfied
the condition of r < h / 2 , where “ r ” is the radius of the pore and “ h ” is the thickness
of the membrane. The soft membrane material tried to solidify to a morphology
where it satisfied the minimum surface free energy, and the final pore diameter
depended on the ratio of “ r ” and “ h ” of the membrane. Kim et al. reported similar
results, but they updated the relation as r < h / 3 for pore shrinking to occur
under the electron beam [ 18 , 19 ].