Capture and Translocation of Nucleic Acids into Sub-5 nm Solid-State Nanopores - Nanopores: Sensing and Fundamental Biological Interactions - page 245

Fig. 10.13 Dwell-time distribution, shown in a linear-linear scale for translocations of a 6,000 bp
DNA fragment (for a 4-nm pore at 21 C, 300 mV applied voltage). The distributions obey first-
passage time distributions in the limit of strong pore-polymer interactions. The mean transport
time can be well-approximated by fitting an exponentially decaying function to the tail of the
distribution (i.e. t>t P ). The emergence of a second dominant timescale ( t2 2 ) is visible for longer
DNA molecules. Reproduced with permission from Wanunu et al. [ 49 ], Copyright Elsevier Inc
10.3.2 Effect of Nanopore Size on DNA Transport
One obvious way to control the polymer-pore interactions is to fine-tune the pore size.
Control over the nanopore size can be achieved with the sub-nmprecision fabrications
capabilities of a TEM [ 23 , 25 , 44 ]. This facilitated an investigation into the effect of
nanopore size on its transport dynamics. This investigation involved preparing a set of
nanopores in the range of ~2.5-5 nm in diameter, and measuring the typical dwell
times of double-stranded DNA in each pore, as explained above [ 49 ]. As shown in
Fig. 10.14 , a decrease of over an order of magnitude in dwell time was observed when
the pore diameter was increased by a mere ~2.3 nm. This increase is much larger than
the expected increase based on viscous drag alone, illustrating the dominance of
DNA-pore interactions in determining the biopolymer's dynamics. Additionally,
the collision timescale appears to be only mildly affected by the pore size, as shown
by the open circles in Fig. 10.15 . Since colliding molecules do not make extended
contact with the pore, logically pore size should not affect their dynamics as much as it
affects actual translocations.
10.3.3 DNA Length Dependence on the Transport Dynamics
As explained above, the dwell-time distributions show two distinct timescales
( t 1 and t 2 ), obtained by fitting bi-exponential functions to the distributions. Notably,