On current blockade upon analyte

translocation in nanopores
Cite as: J. Appl. Phys. 129, 064702 (2021); https://doi.org/10.1063/5.0035113
Submitted: 26 October 2020 . Accepted: 23 January 2021 . Published Online: 09 February 2021

Chenyu Wen, and Shi-Li Zhang

ARTICLES YOU MAY BE INTERESTED IN

Deep learning for nanopore ionic current blockades

The Journal of Chemical Physics 154, 044111 (2021); https://doi.org/10.1063/5.0037938

Solid-state nanopore hydrodynamics and transport

Biomicrofluidics 13, 011301 (2019); https://doi.org/10.1063/1.5083913

Dynamic and weak electric double layers in ultrathin nanopores

The Journal of Chemical Physics 154, 134703 (2021); https://doi.org/10.1063/5.0048011

J. Appl. Phys. 129, 064702 (2021); https://doi.org/10.1063/5.0035113 129, 064702

© 2021 Author(s).
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

On current blockade upon analyte translocation

in nanopores
Cite as: J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113
Submitted: 26 October 2020 · Accepted: 23 January 2021 · View Online Export Citation CrossMark
Published Online: 9 February 2021

Chenyu Wen and Shi-Li Zhanga)

AFFILIATIONS
Division of Solid-State Electronics, Department of Electrical Engineering, Uppsala University, SE-751 03 Uppsala, Sweden

a)
Author to whom correspondence should be addressed: [email protected]

ABSTRACT
Nanopore sensing primarily concerns quantifying the amplitude and shape of blockage current as well as the frequency of translocation
events by analyzing the variation of the ionic current upon analyte translocation in a nanopore that represents an extremely simple device
structure. To facilitate such an analysis, most reported physical-phenomenological models focus on geometrical factors. Here, we systemati-
cally analyze several other factors that may influence the amplitude and waveform of the blockage current. Our theoretical analysis starts
with an analytical model based on geometry. It is then extended to include effects of surface conductance, electroosmotic flow, ionic concen-
tration polarization, and induced charge on nanopore membranes. This approach allows for the examination of related electrokinetic and
electrohydrodynamic aspects of analyte translocation in nanopores. The model results are confirmed using numerical simulation. The prin-
cipal outcome of our theoretical scrutiny includes the identification of the respective determinatives of various factors as well as criteria for
safely neglecting some of them when correlating the amplitude and waveform of blockage current to the properties of the translocating
analyte. Our attempt to categorize these factors can be of practical implications in understanding the translocation process and for develop-
ing advanced data analysis algorithms as an effort to promote nanopore sensor applications.

Published under license by AIP Publishing. https://doi.org/10.1063/5.0035113

I. INTRODUCTION ionic current during the translocation. The blockage concerns both
A nanopore sensor features a strikingly simple device structure the blockade in the nanopore region and analyte–nanopore interac-
with a nanoscale pore chemically and electrically bridging two elec- tions. This has been the practice used to distinguish different pro-
trolyte reservoirs that are otherwise separated by a membrane. A teins from their fingerprint features in the blockage ionic current
similar idea was first developed for pores of micrometer dimen- traces reflecting their individual morphological and dynamic prop-
sions known as the Coulter counter.1 By shrinking the pore size to erties,7 and further to extract the shape, volume, and dipole
the nanoscale, the resultant nanopore is capable of single molecule moment of proteins,8–10 infer the folding state11 and flexibility of
analysis. The prevailing mechanism for nanopore sensing is based proteins,12 and measure nanoparticle size.13–15 Therefore, under-
on monitoring the ionic current between the two electrolyte reser- standing analyte translocation mechanisms and establishing reliable
voirs.2 It can also be combined with other sensing principles such models to decode the ionic current signal are the keystones of
as transverse tunneling current3 and plasmonics4 to boost the nanopore-based sensing.
sensing fidelity. A celebrated application of nanopore technology is In the models developed for the Coulter counter, the geometry
the DNA sequencer based on biological nanopores and commer- of the translocating analyte is adequately considered for its decisive
cialized by Oxford Nanopore Technologies.5,6 The success of this role in the current change. Maxwell devised a model that shows an
nanopore sensor lies in its ability in deciphering the subtle fluctua- increase in the resistance of a homogeneous conductive media
tions in blockage current caused by the four nucleotides to the proportional to the fractional volume of dispersed insulating
sequence of long DNA strands. In addition, for the translocation of particles.16 Applying this theoretical assessment in a micropore/
analytes such as proteins and nanoparticles, it is anticipated that nanopore structure finds that the resistance increase during the
the analyte properties, e.g., size, shape, charge, and concentration, translocation of an analyte is 1.5 times the volume ratio of the
can be extracted by analyzing the characteristic blockage of the analyte to the pore.17,18 For various spheroidal analytes, e.g.,

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-1

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

sphere,19 prolate ellipsoid, oblate ellipsoid,20–22 and disk,23 correc- desirable that the action conditions and ranges of the various
tion factors are introduced to account for the geometrical effects of factors be systematically explored. In this work, all related factors,
the analytes.24,25 Furthermore, the orientation of a spherically i.e., geometry, surface conductance, EOF, ICP, and ICE, are thor-
asymmetric analyte relative to the pore geometry is considered oughly discussed in the framework of a set of analytical models
in analyzing small fluctuations of the blockage current during supported by numerical simulations. Specifically, their effects on
translocation caused by possible rotation7,26,27 and off-central axis the amplitude and waveform of translocation current are studied.
movements of an analyte.8 However, electrokinetic and electrohy- In addition, the respective determinatives of these factors as well as
drodynamic effects originating from the surface phenomena criteria for safely neglecting some of them are discussed. Providing
become increasingly dominant when the pore size is shrunk from such a complete picture of how various factors determine the nano-
the micrometer- to nanometer-scale. It has been reported that the pore translocation process can help develop advanced data analysis
models based on geometry considerations usually bear 10%–40% algorithms and promote the technology development for nanopore
errors in predicting the analyte volume, especially when performing sensors. In the remainder of this work, “blockade” is used to
experiments with dilute electrolytes.8,10,11 emphasize the process of obstruction, while “blockage” is used to
Surface conductance caused by the drift of the surface charge express the state in obstruction. Similarly, “translocating” refers
in the diffuse layer of the electrical double layer (EDL) has been to the process, while “translocation” is for the state.
incorporated in some of the models.28,29 Including the surface con-
ductance in the model can help predict the amplitude of blockage
current due to translocation of heavily charged analytes with differ- II. METHODS
ent electrolyte concentrations. One of the successful cases is the
A. Analytical models
unique current increase upon DNA translocation.30,31 The underly-
ing mechanism is the high density of negative charge DNA mole- Analytical models are usually constructed based on the frame-
cules carry. The resultant surface conductance on the DNA work of a geometrical blockade model. It is often built on the
molecules may become so high to compensate for and even surpass simple picture that an insulating analyte of a given size and shape
the conductance loss due to the volumetric occupation of translo- occupies a certain volume of the pore during its translocation.
cating DNA.30,32,33 In addition, ionic concentration polarization Therefore, this occupation causes an increase in the resistance of
(ICP), a prominent effect for relatively small pores (tens of nm in the system and a decrease in the nanopore current. However, as
diameter) at large bias voltages (hundreds of mV), can totally shown in Fig. 1(a) for a cylindrical pore for simplicity, several other
change the concentration distribution around the pore.34 Because factors than geometrical occupation are also at work, and they can
of a larger population of one ion species than the other in the pore modulate and even totally dominate the blockage current. These
region, the current is carried more by the major ions, thereby factors include the drift of the surface charge, EOF, ICP, and ICE,
breaking the balance of the ions and counterions in the bulk which is discussed as follows. Two representative shapes, a sphere
region. To compensate for this unbalance, ion accumulation and and a long cylinder, are selected for the model demonstration of
depletion regions leading to a concentration gradient build up in translocating analytes. Analytes neither spherical nor cylindrical
the vicinity of the pore. Under certain conditions, ICP may induce can, in principle, be approximated with a shape between these two
a biphasic translocation waveform, meaning that both an increase extremes.
and a decrease of current can appear during translocation.35–37
Factors that influence ICP will, thus, affect the blockage current,
e.g., bias voltage,38 salt concentration,39,40 and surface charge 1. Spherical analytes
density of nanopore sidewalls.38 In the presence of ICP, the charge
By focusing on the geometry of both the nanopore and trans-
density of translocating analytes41 and the translocation direction
locating analyte, a simple phenomenological model mathematically
in asymmetric pores39,42 will contribute to the shape, duration, and
describing the blockage current can be established based on resist-
amplitude of translocation waveforms. As for electrohydrodynamic
ance arguments. During the translocation of a spherical analyte of
factors, electroosmotic flow (EOF), i.e., a water flow driven by the
radius r0, the cylindrical pore can be sectioned in three parts for
mobile surface charge, should be considered. The EOF can not
model development, as shown in Fig. 1(b). Let Rb represent the
only modulate the translocation velocity through co-actions from
resistance of the spherical blockage segment, while Rs the sum of
electrophoresis, hydro-pressure, and viscous retardation,43,44 but
the resistance components of the upper and lower undisturbed
also further shape the translocation waveform.45,46 Moreover, other
parts including their respective access region. Two critical assump-
mechanisms such as induced charge effects (ICEs, the effects
tions should be noticed here: (1) the conductivity of the electrolyte
caused by the charging–discharging process of membrane surface
is homogeneous, irrespective of the location in the system and; (2)
under the external bias voltage),47 transport-induced-charge elec-
the equipotential surface at the nanopore opening is flat and per-
troosmosis,48 and pressure gradient flow49 may largely affect the
pendicular to the central axis. Thus, the total blockage resistance is
translocation current under certain conditions.
the sum of Rb and Rs, respectively, expressed by2
It has been found that numerous simulations and experiments
have been implemented to study various mechanisms, besides the ðr0
geometrical factor, involved in the nanopore translocation 8ρ dz 4ρ 2r0
Rb ¼ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi arctan qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (1)
process.33,50–54 However, a comprehensive study to structure all π dp2  4r02 þ 4z 2 π d2  4r 2 dp  4r02
2
0 p 0
related factors in a unified picture is generally lacking. It is also

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-2

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

FIG. 1. Structure of a nanopore sensor. (a) Different physical mechanisms involved in the translocation process. Assuming a negative surface charge of the membrane,
the reddish layer on the membrane surface represents accumulated positive charge inside EDL. Under bias, the mobile surface charge contributes to surface conductance
and generates EOF, with a velocity of uw and an uneven distribution. The higher concentration of cations than anions in the pore region induces a redistribution of the elec-
trolyte concentration. An accumulation region with a higher concentration, cH, and a depletion region with a lower concentration, cL, appear at the lower and upper vicinities
of the pore, respectively, shown by the darker and lighter bluish color compared to that in the reservoir with a concentration of c0. Considering the ICE, the EDL on
the upper surface of the membrane is slightly thicker than that on the lower surface. The weakly charged analytes are translocating nanopores driven by EOF instead of
the electrophoretic force. (b) and (c) Resistance model for the translocation of a spherical analyte and a long cylindrical analyte, respectively. (d) Domain and boundary
definition for numerical simulation of the nanopore system.

and This expression is precisely the exact solution for a rod analyte
of radius r0 and height 2r0. Thus, the relative current change refer-
4ρ(Leff  2r0 ) ring to the open-pore current I0 can be
Rs ¼ : (2)
πdp2
ΔI Rb  R0
¼ : (5)
I0 Rb þ Rs
Detailed mathematical derivations can be found in
supplementary material, Note 1. For the open-pore state, Rb is sub-
The effects of surface charge and EOF are fused into the
stituted by R0 that represents the resistance of the middle segment
model by their contribution to the surface conductance. In the per-
free from the spherical analyte,2
spective of the equivalent circuit, the extra conductance contributed
by the surface charge and EOF is connected in parallel with the
8ρr0
R0 ¼ : (3) above-considered geometrical conductance (inverse of resistance).
πdp2 Therefore, the expressions of Rb, Rs, and R0 with surface charge
contribution become, respectively,
Here, ρ is the resistivity of electrolyte and dp is the diameter of
the pore. An important parameter is Leff, the effective transport " #1
πdp2 πr0
length of the pore that is defined as equal to the sum of the dis- Rb ¼ (1 þ Du) þ μπσ s  , (6)
tances from the nanopore center where the electric field is at its 8ρr0 2ρ
maximum magnitude to the points along the central axis where
their electric field falls to e−1 of this maximum.55 The Leff region
4ρ(Leff  2r0 )
has the highest electric field in the system and bears most of the Rs ¼ , (7)
external bias voltage drops. Leff is equal to h + 0.92dp for a cylindri- πdp2 (1 þ Du)
cal pore with the membrane thickness of h. Utilizing the Taylor
expansion of the arctan function, Eq. (1) can be approximated by and

8ρr0 8ρr0
Rb ¼ : (4) R0 ¼ , (8)
π(dp2  4r02 ) πdp2 (1 þ Du)

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-3

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

with and

2σ 0 4ρLeff
Du ¼ : (9) Rb ¼ : (16)
dp c0 qNA π(dp2  4r02 )

Here, Du is the Dukhin number defined as the ratio of the surface The resulting relative current blockade is
conductance to the bulk conductance.56 σ0 and σs are the surface
charge density of the nanopore sidewall and translocating analyte, ΔI R0
respectively, c0 is the electrolyte concentration, q is the element ¼1 : (17)
I0 Rb
charge, and NA is the Avogadro constant.
Furthermore, the water flow driven by the mobile surface Including the contributions of the surface charge turns R0 and
charge, i.e., EOF, can induce an extra surface current component Rb into
due to the directional transport of the net charge in EDL by the
water flow. This current can be calculated by integrating the pro- 4ρLeff
R0 ¼ (18)
duction of water flow velocity and charge density on the cross πdp2 (1
þ Du)
section of the nanopore.57 The final expressions for the resistance
in the presence of EOF current are57,58 and
"  2 #1 " #1
πdp2 (1 þ Du) 2πdp ε20 ε2w kT πdp2 2r0 μπσ s πr2
Rs ¼ þ γ(σ 0 ) , (10) Rb ¼ (1 þ Du) þ  0 : (19)
4ρ(Leff  2r0 ) (Leff  2r0 )λD η q 4ρLeff Leff ρLeff

(   )1 Further incorporating the contributions of EOF makes R0 and

πdp2 πr0 πε20 ε2w kT 2
Rb ¼ (1 þ Du) þ μπσ s  þ [dp γ(σ 0 ) þ 2r0 γ(σ s )] , Rb look as follows:
8ρr0 2ρ r0 λD η q
(11)
"   #1
πdp2 (1 þ Du) 2πdp ε20 ε2w kT 2
R0 ¼ þ γ(σ 0 ) (20)
"  2 #1 4ρLeff Leff λD η q
πdp2 (1 þ Du) πdp ε20 ε2w kT
R0 ¼ þ γ(σ 0 ) , (12)
8ρr0 r0 λD η q and
(  
with πdp2 2r0 μπσ s πr 2 2πε20 ε2w kT 2
Rb ¼ (1 þ Du) þ  0 þ
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 4ρLeff Leff ρLeff Leff λD η q
ε0 εw kT )1
λD ¼ (13)
2q2 NA c0  [dp γ(σ 0 ) þ 2r0 γ(σ s )] : (21)

and
The analytical models developed above are implemented on
 
qλD σ MATLAB R2015b, and the modeling results will be compared with
γ(σ) ¼ arsinh2 , (14) the simulation ones.
2kTε0 εw

where λD is the Debye length, ε0 is the permittivity of vacuum, εw 3. Ionic concentration polarization, ICP
is the relative permittivity of electrolyte, η is the viscosity of electro-
lyte, k is the Boltzmann constant, and T is temperature in Kelvin. The ionic current of a pore comprises four ionic flux compo-
nents: drift and diffusion contributed by both cations and anions.
A nanopore with charged sidewalls can show ionic selectivity
during the transport of ions. Specifically, the cations and anions in
2. Cylindrical analytes the electrolyte generate different drift fluxes under the same exter-
nal electric field because the population of cations and anions in
For a long cylindrical analyte of radius r0, it occupies
the pore region is no longer the same as induced by the surface
the entire Leff region at the same time during the translocation
charge on the nanopore sidewalls. A concentration gradient is built
[Fig. 1(c)]. Thus, one resistor is adopted to represent the entire Leff
up across the pore region to generate diffusion fluxes toward the
region. Considering the geometrical effects only, the open-pore and
same direction for both cations and anions to balance the drift
blockage resistances are59
fluxes. At steady state, the cation and anion fluxes are equal all
4ρLeff along their paths. The phenomenon and process in establishing
R0 ¼ (15) this concentration gradient is termed ICP, and it is pronounced in
πdp2 the relatively small pores (tens of nm in diameter) with low

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-4

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

concentrations of electrolyte (∼100 mM). It generates severe inho- simulation implementation followed the convention with the
mogeneity in conductivity, leading to a greatly modified waveform Nernst–Planck equation (NPE) to govern the dynamics of ions in
of the ionic blockage current, not to mention its amplitude. An the electrolyte, achieved by the transport of diluted species module;
enhanced ionic current followed by a current decrease can be the Poisson equation (PE) to describe the electric field distribution
observed. Such a severe inhomogeneity in conductivity makes it through the electrostatics module; and the Navier–Stokes equation
challenging in establishing an analytical model for the translocation (NSE) to dictate the movement of water, realized by the laminar
dynamics. Nonetheless, the concentration gradient of ICP in the flow module. The simulation domain was composed of two reser-
open-pore state can be estimated to assess criteria for situations in voirs and a connecting pore in between, as shown in Fig. 1(d). For
which ICP is necessary to consider. a clear illustration, only the topological structure is shown here
In the perspective of fluxes through a nanopore of cross- without accounting for the length and scale of each boundary.
sectional area S0, the current can be expressed as60 The simulation starts with that considering all aforementioned
physical processes, i.e., geometrical blockage, surface conductance,
2qNA S0 Dc Δc EOF, ICP, and ICE. Domain 1 [light blue in Fig. 1(d)] represents
I ¼ qNA S0 ( fel c þ fel a ) ¼ , (22)
tþ þζtþ ζ h the electrolyte, where NPE, PE, and NSE are fully coupled and
applied in a self-consistent way. Domain 2 (gray) and 3 (light
where fel_c/a represents the drift fluxes of cations/anions, Dc/a is orange) denote the nanopore membrane and the translocating
the diffusivity of cations/anions, Δc is the maximum concentra- analyte, respectively, where the PE is applied to describe the electric
tion difference across the nanopore, h is the thickness of the field distribution. The simulation is implemented on a two-
nanopore membrane, ζ is the diffusivity ratio of cations to anions, dimensional axial-asymmetrical domain, with the rotational asym-
i.e., ζ = Dc/Da, and t+ is the transfer number of cations that metrical axis AH (green color). Boundaries DL, DE, EM, and IJK
calculates the fraction of the drift current carried by cations in (red) are assigned a uniform charge density. Boundaries AB and
the pore region.61 t+ is related to the surface charge density of HG (blue) are the inlet and the outlet for the fluid, respectively,
nanopore, σ0, via62 with a constant electrolyte concentration and zero-pressure. In
addition, HG is electrically grounded, while AB is assigned a bias
χ voltage. The rest of the boundaries (black) are non-slip walls for
tþ ¼ , (23)
χþ1 fluid, insulators for the electric field, and non-permeable walls for
ions. Details regarding the definition of each boundary condition
with and parameter used can be found in supplementary material,
s
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi!2 Note 2.
2 For other situations where only one or some target physical
σ0 σ0
χ¼ þ þ1 : (24) processes are involved, conditions for the corresponding domains
qNA dp c0 qNA dp c0
and boundaries need be adjusted accordingly. When only consid-
ering the geometrical effects, Domain 2 and 3 are excluded. The
In the perspective of a circuit loop, the total current can be
surface charge on both the nanopore membrane and translocating
expressed according to Ohm’s law,
analyte [red boundaries in Fig. 1(d)] was set to zero. Cations and
anions would now distribute evenly in the entire solution
V VS0 (1  δ 2 )
I¼ ¼ , (25) domain, leading to a homogeneous conductivity. The contribu-
R ρLeff tion of the surface conductance could be included by introducing
the surface charge density in the system. In order to exclude the
with δ = Δc/2c0, where V is the voltage drop across the nanopore. consequent ICP caused by the ionic selectivity of the pore at this
Combining Eqs. (22) and (25), Δc can be solved as point, the ion distribution was assumed “frozen” under the zero-
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi bias condition (i.e., thermal equilibrium). Then, the ionic current
Δc ¼ α2 þ 4c20  α, (26) was calculated at other bias voltages by keeping the same conduc-
tivity distribution inherited from the equilibrium condition with
with corresponding ion mobility. When ICP was considered, a full
coupling between NPE and PE was performed in Domain 1.
4qNA Dc ρLeff c20 Finally, EOF was introduced by fully coupling NSE to the system
α¼ : (27) and ICE was incorporated by including the membrane and trans-
hV(tþ þζtþ ζ)
locating object zones into the solution domain of PE with
assigned permittivity.
In simulation, the bias voltage was in the range of −500 mV to
B. Numerical simulation 500 mV, consistent with the commonly adopted experimental con-
In order to quantify the effects of aforementioned factors, ditions. To investigate the dynamics of current changes during the
these factors were separately considered with appropriate boundary translocation, i.e., translocation waveform, the analytes were placed
conditions and coupling relationships in numerical simulation. at different positions along the central symmetrical axis, and the
Following a similar procedure as our previous work,63,64 the simu- current was calculated by integrating various ionic fluxes on a
lation was performed using COMSOL Multiphysics 5.4. The cross-sectional area of the nanopore.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-5

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

III. RESULTS AND DISCUSSION making the signal extraction a severe challenge. Thus, boosting the
In what follows, effects of the various factors are discussed signal amplitude, such as using small pores, and suppressing the
stepwise by starting from the most basic ones related to the geome- noise are two important aspects to achieve a high signal-to-noise
try of a nanopore. ratio.
There have been numerous publications on modeling the
translocation of differently shaped analytes, all motivated by the
A. Geometry Maxwell model.16 As discussed above, the relative resistance
If the resistivity of the electrolyte is assumed to be uniform increase, ΔR/R0, for a dilute dispersion with insulating spheres in a
and the current decrease is only caused by the geometrical block- uniform conductive material can be approximated to be 1.5 times
ade, the current changes during the translocation of a sphere along the fractional volume of the spheres Vs/V0. This assessment is well
the central axis, i.e., the translocation waveform, are shown as the supported by the model proposed here [Eqs. (1)–(5)], with a devia-
inset of Fig. 2(a). The waveform is symmetric for cylindrical pores, tion <20% for the volume ratio, Vs/V0 < 0.05 (see supplementary
and the shape does not change with the bias voltage, while the material, Note 3 for details).
amplitude of translocation signal is affected by the concentration of
the electrolyte, diameter, and thickness of nanopore and the radius
B. Surface conductance
of the translocating analyte (sphere or cylinder in this study).
As expected, the relative amplitude of blockage current is seen in The surface charge is the origin for the formation of EDL on
Figs. 2(a) and 2(b) to decrease with increasing nanopore diameter. the surface of the nanopore sidewall as well as on that of translocat-
The dots represent the simulation data, while the lines are the mod- ing analytes. The presence of EDL not only contributes to surface
eling results by Eqs. (1)–(5) and (15)–(17) for spherical and cylin- conductance but also causes ICP. The latter will be discussed in
drical analytes, respectively, without invoking any fitting Sec. III D. Increasing pore diameter can decrease the relative block-
parameter. The comparison indicates that the model predicts well age amplitude, as shown in Figs. 3(a) and 3(d). Furthermore, the
about the blockage current, which provides a reliable baseline for conductance enhancement due to the presence of the surface
including other factors in Secs. III B–III F. In practice, noise is an charge can reduce the relative blockage amplitude; the higher the
inevitable issue, and the blockage signal from relatively large nano- surface charge density, the smaller the amplitude is, cf. Figs. 3(b)
pores can be lost in the background noise. For example, a and 3(e).
6 nm-diameter sphere can generate a blockage current of 1.3% I0 in If the nanopore is large in diameter compared to the thickness
a 20 nm-diameter pore, which usually shows a noise level of of EDL (e.g., dp > 8 nm in a 100 mM electrolyte with a ∼1 nm-thick
∼0.2 nA peak-to-peak amplitude at I0 = 20 nA with 10 kHz band- EDL), the EDLs on the pore sidewall and on the analyte are far
width. The noise is, thus, comparable with the blockage signal, away from each other. Thus, the contributions of surface

FIG. 2. Current change caused by geometrical blockade. (a) and (b) Relative amplitude of blockage current caused by the translocation of a 6 nm-diameter sphere and a
6 nm-diameter, 150 nm-long cylinder in cylindrical nanopores of different diameters, respectively. Dots: simulation; solid lines: modeling. The inset in (a) translocation wave-
form of a 6 nm-diameter sphere in a 15 nm-diameter, 25 nm-thick cylindrical pore. The x-axis is the position of the sphere along the central translocation axis marked in
the schematic beside in which the orange area marks the region of the nanopore.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-6

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

FIG. 3. Relative blockage amplitude by considering surface conductance and EOF. (a) and (d) Relative blockage amplitude caused by the translocation of a sphere and
a long cylinder in nanopores of different dp, respectively. (b) and (e) Relative blockage amplitude of a sphere and a long cylinder in nanopores of different negative
σ0 without considering EOF, respectively. (c) and (f ) Relative blockage amplitude of a sphere and a long cylinder with σs = ± 0.01 C/m2 (blue for − and red for +) in the
electrolyte of different concentrations, respectively. Dots: simulation; solid lines: modeling.

conductance from the two EDLs can be considered independent. other until they overlap as seen in Fig. 4(a). From the simulation
Our simple model represented by Eqs. (6)–(9), (18) and (19) is results, the entire volume of the nanopore has a much larger ionic
valid. By changing the electrolyte concentration, the Debye length concentration, as well as conductivity, than that in the reservoirs
can be tuned. The EDL on the pore sidewall and that on the when (dp/2 − r0) < 4λD. For example, the Debye length for a
analyte do not overlap for a 3 nm-diameter sphere/cylinder translo- 100 mM electrolyte is ∼1 nm. The EDL on the nanopore sidewall
cating a 15 nm-diameter pore in electrolytes of 100 mM concentra- and that on the analyte then overlap for nanopores of diameter
tion and above, see Figs. 3(c) and 3(f ). The polarity of the surface smaller than 12 nm, yielding a relatively high ionic concentration
charge does not influence the blockage amplitude. It indeed shows in the entire region of the nanopore as shown in Fig. 4(a). The
that the relative blockage amplitudes are identical for positively and overlapping EDLs are also seen in Fig. 4(b) and supplementary
negatively charged analytes. For lower electrolyte concentrations material, Fig. S7. Furthermore, the polarity of the surface charge
and thereby thicker and overlapping EDL, the blockage amplitudes plays a role. If the charge polarity of the pore sidewall is the same
for translocation of positively and negatively charged analytes as that of the analyte, the concentration enhancement is more
deviate from each other. It is worth noting that if the analyte has a obvious resulting in a lower blockage amplitude [Fig. 4(c)].
high surface charge density and if the salt concentration of the elec- To summarize, the influence of surface conductance on the
trolyte is low, the mobile charge carried by the analyte can com- blockage current is more pronounced for smaller pores (with its
pensate for the blockage effect, thereby resulting in a current geometrical dimensions comparable with the Debye length) with
increase instead of decrease. The effect of this phenomenon is a higher surface charge density and lower electrolyte concentrations.
negative ΔI/I0 and has been observed in DNA translocation in SiO2 These conditions give rise to a larger fraction of the surface con-
nanopores65 and predicted by several theoretical models and ductance in the total conductance. In general, the Dukhin number
simulations.30,32,39 [Eq. (9)] is introduced to evaluate the significance of the surface
This phenomenon can also be observed by directly decreasing conductance.56 In practical nanopore experiments, the Dukhin
the nanopore diameter. The accumulated ion layers on the surface number can be calculated. If it is much smaller than one, the
of the pore sidewall and analyte asymptotically approach each surface conductance effects can be safely ignored. However, if the

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-7

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

FIG. 4. Surface conductance and EDL overlap. (a) Distribution of total ion concentration along the radial cutline at the narrowest restriction defined by a translocating
sphere and the nanopore sidewall for different dp, shown in the inset as an example. (b) Distribution of net charge [qNA(c+ − c−)] for an 8 nm- and a 30 nm-diameter nano-
pores during the translocation of a 6 nm-diameter sphere. The white lines in the figures are the equal-charge density contours at 10% of the maximum charge density,
which mark a general region of EDL. It can be clearly seen that the EDL on the nanopore sidewall and that on the analyte overlap in the 8 nm case but not in the 30 nm
case. (c) Translocation waveform of a 6 nm-diameter sphere with σs = ± 0.01 C/m2 in 10 nm- and 15 nm-diameter nanopores.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-8

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

Dukhin number is close to or larger than one, the contribution and modulates the blockage current. By including the electroos-
of the surface conductance should be considered in the model, as motic current, the translocation waveform becomes slightly asym-
Eqs. (6)–(9). For example, a correction factor can be introduced in metry as seen in Fig. 5(a). The polarity of the surface charge on the
the model so as to compensate for the effects and to more precisely pore sidewall and on the analyte shows a more pronounced influ-
extract the analyte size. ence on the blockage amplitude. Since the surface charge on the
pore sidewall and on the analyte drags water towards opposite
directions if they do not have the same polarity, a relatively small
C. Electroosmosis EOF velocity appears compared to that with the charge on the pore
The surface charge can also cause EOF under an external elec- sidewall and the analyte is of the same polarity. It can be seen
tric field. The directional movement of water flow carries the ions clearly by the direct comparison of the extracted electroosmotic
in it. In EDL, the convective transport of the charge generates an current component from the total current between the cases with posi-
extra component of the surface current, i.e., electroosmotic current, tively and negatively charged nanopore sidewall, as shown in Fig. 5(b).

FIG. 5. EOF effect on current. (a) Translocation waveform and (b) electroosmotic current component of a 6 nm-diameter sphere with σs = ± 0.01 C/m2 in a 15 nm-diameter
nanopore involving EOF. Inset: distribution of electroosmotic current density during the translocation of the sphere with σs = −0.01 C/m2 in a nanopore with σ0 = −0.02 C/m2.
(c) and (d) Relative blockage amplitude of a sphere and a long cylinder in nanopores of different negative σ0 by considering EOF, respectively.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-9

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

FIG. 6. ICP effect on current. (a) Translocation waveform of a 6 nm-diameter sphere in a 15 nm-diameter cylindrical nanopore at 500 mV bias voltage with ICP involved.
Relative blockage amplitude generated by the translocation of (b) a 6 nm-diameter sphere and (c) a 6 nm-diameter, 150 nm-long cylinder, at different bias voltage. Legends
in the figure: Geo, pure geometrical blockade; Surf C −/+, involving surface conductance with surface charge of −/ + 0.01 C/m2 on the translocating analyte; EOF −/+,
involving EOF with surface charge of −/ + 0.01 C/m2 on the translocating analyte. The surface charge on the nanopore sidewall is always kept to −0.02 C/m2.

In addition, the electroosmotic current mainly concentrates near EOF weakens the geometrical blockage amplitude. Moreover, the
the surface of both the nanopore sidewall and the analyte as a surface conductance, including the EOF contribution, is mainly
result of the convective charge moving in the EDLs. determined by the surface charge density and polarity that are
The electroosmotic current enhances the conductance synerg- independent of the bias voltage [see supplementary material,
istically with the drift of the surface charge discussed in Sec. III B. Figs. S8(b) and S8(c)].
The EOF velocity as well as the electroosmotic current increases
with increasing the surface charge density on the pore sidewall and
on the translocating analyte. This dependence results in a further D. Ionic concentration polarization, ICP
decrease of the relative amplitude of the blockage current (Figs. 5(c) Both the symmetric waveform and voltage independence are
and 5(d)) compared to the results of the case only considering the broken by ICP. This effect is evidenced by a current increase fol-
drift of the surface charge contribution (Figs. 3(b) and 3(e)). lowed by a decrease along the symmetric axis of the nanopore
However, the enhancement of surface conductance by EOF is observed in Fig. 6(a) exemplified for a 6 nm-diameter sphere trans-
limited, amounting only to 10% – 25% of the total surface conduc- locating a 15 nm-diameter pore from the higher potential side
tance in general situations. Details related to the theoretical deriva- toward the ground side under positive bias. This phenomenon has
tion can be found in supplementary material, Note 4. Therefore, been confirmed both experimentally and by numerical
the influence of electroosmotic current on the blockage current can simulation.35–37 Influence by ICP, the ionic concentrations inside
be measured by a feature number of 0.1Du – 0.25Du. the pore as well as in its vicinity severely deviate from the unaf-
Merely tuning the surface conductance will not break the sym- fected concentration c0 in the reservoir far away from the nanopore.
metry of the translocation waveform. This observation is evident by At positive bias, an ionic depletion region (c < c0) and enhancement
comparing the waveforms in supplementary material, Fig. S8(a). region (c > c0) are built on the upper and lower sides of the nano-
The contribution of the surface conductance from both the surface pore opening. It can be seen from the distribution of the total ion
charge drift by the electric field and the convectional transport by concentration, which also represents conductivity, that the ions

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-10

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

FIG. 7. Distribution of the ion concentration compared to the bulk concentration (c+ + c− − 2c0) caused by ICP. Panels (i) to (ix) are the chronological sequence of snap-
shots during the translocation. Black lines and red arrows mark the border between the ion accumulation and depletion regions. The position of the sphere is marked by
the grey dots.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-11

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

redistribute during the translocation of a sphere, as shown in Fig. 7. waveform and the blockage amplitude. The coupling effects of EOF
The border between the ionic depletion and enhancement regions with ICP will be discussed in Sec. III E.
locates precisely at the position of the sphere, i.e., the narrowest The translocation waveform also relies on the bias conditions
restriction in the system, and follows the movement of the sphere (polarity and amplitude) and the translocation direction. In addi-
(marked by red arrows in Fig. 7). Therefore, when the translocating tion to the observation of the increasing–decreasing translocation
sphere approaches the upper opening of the pore [Fig. 7(ii)], the current in Fig. 6(a), the surface charge polarity of the pore sidewall
entire pore region is occupied by the high concentration electrolyte. and the analyte can also significantly modify the ionic concentra-
A current increase is obtained, i.e., the rising edge of the green line tion distribution and affect the waveform (comparing the green
from 260 to 275 nm in Fig. 6(a) for the sphere to enter the and blue lines). There is no obvious current increase in the wave-
pore. When the sphere moves near the lower opening of the pore form if the analyte carries a positive surface charge [blue line in
[Fig. 7(viii)], the entire pore region is filled by the low concentra- Fig. 6(a)]. Furthermore, since ICP is stimulated by the electric field,
tion electrolyte. A current decrease is evident by the falling edge of the concentration gradient built by ICP positively depends on the
the green line from 275 to 300 nm in Fig. 6(a). These observations bias voltage [Eq. (26)]. Thus, the blockage amplitude also closely
confirm that the ICP effect can be so strong to make the blockage relies on the bias voltage, which can be seen in Figs. 6(b) and 6(c).
current totally be determined by the concentration distribution, Based on our model for the concentration difference built by
instead by the geometrical blockade. It is worth noting that all sim- ICP [Eqs. (22)–(27)], the significance of ICP can be assessed. It can
ulation results shown in this section are valid for the cases without be seen in Fig. 8 and supplementary material Fig. S9 that the frac-
considering EOF. The EOF effect can be coupled with the ICP tional concentration change Δc/c0 caused by ICP increases with
effect to yield an even more obvious effect on the translocation decreasing electrolyte concentration and/or increasing bias voltage.

FIG. 8. Fractional concentration change Δc/c0 caused by ICP in a 15 nm-diameter, 25 nm-thick cylindrical nanopore with σ0 equal to (a) −0.01 C/m2, (b) −0.02 C/m2, (c)
−0.05 C/m2, and (d) −0.1 C/m2, in electrolyte of different concentrations at various bias voltages. The color code is shown in the logarithm scale, and the numbers
marked on the contours are in exponent of base 10.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-12

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

ICP is more obvious in a smaller pore with higher surface charge

density. For example, ICP can be safely ignored (Δc/c0 < 1%) if the
electrolyte concentration is higher than 300 mM and at a bias voltage
lower than 100 mV for a 15 nm-diameter pore with −0.02 C/m2
surface charge. The other way around, in a 100 mM electrolyte at
100 mV bais voltage, the diameter of the nanopore should be larger
than 50 nm in a 20 nm-thick membrane with 0.02 C/m2 surface
charge density. With ICP, the analyte information is hidden since the
translocation waveform does not directly correlate to the size of the
analyte. Consequently, it is difficult to convert the blockage amplitude
to the size information. However, other approaches including wave-
form analysis, calibration schemes, and deep learning can be adopted
in practice to extract such information.

E. ICP coupling with EOF

EOF not only contributes to surface conductance in general but
also couples with ICP to further modulate the concentration distribu-
tion through the moving water. The coupling of ICP with EOF can
cause more severe current changes than the 10%–25% contribution
via the surface conductance at all practical bias voltages. An intuitive
impression of the EOF effect coupled to ICP can be gained from a
comparison of the translocation waveform of a sphere through a
nanopore with and without taking into account the EOF effect [green
vs purple lines and blue vs black lines in Fig. 6(a)]. By involving EOF,
the waveform can totally change for a translocating analyte having an
either positively or negatively charged surface.
In order to scrutinize the effect of EOF-ICP coupling, four sim-
ulation scenarios are compared: #1, without EOF and without ICP;
#2, with EOF but without ICP; #3, without EOF but with ICP; and
#4, with EOF and with ICP. For the cases with EOF (#2 and #4), the
current contribution by EOF can be obtained by integrating the
fluxes over a cross-sectional area of the pore, S0, by the following:57
ð
Ieof ¼ qNA (cþ c )vz ds, (28)
S0

where c+ and c− are the concentration of cations and anions, respec- FIG. 9. Electroosmotic current extracted from different simulation configurations
tively, and vz is the water flow velocity along the axial direction. By in (a) open-pore state and (b) blockage state of a spherical translocating analyte
comparing the difference in the total current between #1 and #2, as with negative surface charge σs = −0.01 C/m2.
well as that between #3 and #4, with the electroosmotic current com-
ponents calculated using Eq. (28) in #2 and #4, which include EOF,
the effects of EOF can be thoroughly assessed. The obtained electro- bulk region, in addition to the contribution of the current in the
osmotic related currents are depicted in Fig. 9 for both open-pore surface region. The higher the bias voltage, the larger are ICP and
and blockage states when a negatively charged translocating sphere is the consequent coupling effect. In detail, EOF tends to retard the
at the center of the pore. Similar results for a positively charged establishment of concentration gradient by ICP. Furthermore, it
translocating sphere can be found in supplementary material, tends to weaken the effects of the concentration gradient (see supple-
Fig. S10. In Figs. 9(a) and 9(b), it is apparent that the red plot (#2) mentary material, Fig. S11). In summary, the effects of EOF can be
almost overlaps with the dark yellow one (#2–#1). It indicates that amplified by the assistance of ICP in small pores with low electrolyte
the EOF component in #2 is equal to the difference in the total concentrations.
current between #1 and #2, indicating that the current difference
between #1 and #2 is precisely equal to the EOF component.
However, the difference in the total current between #3 and #4 F. Discussion
remarkably deviates from the EOF component in #4 if ICP is
involved. This deviation is especially prominent when the bias 1. Induced charge effect, ICE
voltage is increased above 200 mV. Hence, EOF coupling with ICP In a nanopore sensor, the thin membrane hosting the pore is
can lead to the modulation of the concentration distribution in the sandwiched by two conductive electrolyte reservoirs, a natural

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-13

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

capacitor. Therefore, the upper and lower surfaces of the mem- achieved charge density by the deprotonation of oxide.33,64,67
brane can accumulate an extra amount of charge induced by an Therefore, ICE can be safely ignored under common experimen-
external field. If the absolute density of the induced charge is tal conditions, which is also confirmed by the identical translo-
comparable to that of the initial fixed charge, ICE on the ionic cation waveform for a 6 nm-diameter 150 nm-long cylinder
current and EOF will dominate.66 The thinner the membrane, the translocating a 15 nm-diameter pore involving or ignoring ICE
higher the permittivity of the membrane, and/or the higher the (see supplementary material, Fig. S12).
applied voltage, the higher is the density of induced charge and
the more obvious effects it can cause. Despite the potential
effects, the density of induced charge is usually much smaller 2. Influence of various factors on blockage amplitude
than that of the initial fixed charge under ordinary experimental Sections III A–III E have separately (mostly) discussed the
conditions for translocation studies. For example, the density of factors that can affect the translocation signal, waveform and ampli-
induced charge is about Vε0εr/h = 0.005 C/m2 for an extreme tude. These factors include geometry of nanopore and translocating
case of a 20 nm-think silicon membrane biased at 1 V. This analyte, surface conductance due to the drift of ions in EDL, and
density is four times smaller than 0.02 C/m2, a commonly surface current generated by EOF, ICP, and ICE. With reference to

FIG. 10. Relative blockage amplitude by the sphere normalized to that caused by the geometrical blockade only (the flat reference plane) with (a) different c0 and σ0 and
with (b) different σ0 and dp. Relative blockage amplitude at different conditions caused by the translocation of a 6 nm-diameter sphere through a 15 nm-diameter cylindrical
pore. The translocating sphere carries σs equal to (c) −0.01 C/m2 and (d) +0.01 C/m2, while the nanopore sidewall is characterized by σ0 = −0.02 C/m2. For a better com-
parison, the dash lines mark the relative amplitude of the geometrical-blockade-only case.

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-14

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

TABLE I. Determinatives and neglect criteria for various factors influencing the typical experimental scenarios. Three categories of nanopores are
translocation waveform. σ0/s, surface charge density of nanopore sidewall/translocat- included in Table II, i.e., two-dimensional (2D) nanopores exempli-
ing analyte; λD, Debye length; c0, electrolyte concentration; Δc, concentration differ- fying atomically thin membranes, SiNx nanopore being most
ence caused by ICP; geo, geometry; dp, nanopore diameter; h, membrane
thickness; r0, radius of translocating sphere/cylinder; V, bias voltage; ε0, vacuum widely studied, and nanopipette representing excessively thick
permittivity; εr, relative permittivity of the membrane; Du, the Dukhin number. conical pores. Typical values of related parameters are selected,
such as σ0 = − 0.02 C/m2 and V = 200 mV. The criterion for a par-
Factors Determinative Neglect criteria ticular factor to be significant is its influence to be larger than 10%
of the corresponding affected physical quantities (conductance for
Surface σ0, σs, λD (c0), geo Du ≪ 1 [Eq. (9)]
the surface conductance effect and EOF, electrolyte concentration
conductance (dp, r0)
for ICP, and surface charge density for ICE).
Electroosmotic σ0, σs, λD (c0), geo (0.1 ∼ 0.3) Du ≪ 1
In general, surface conductance and electroosmotic current
current (dp, r0) [Eq. (9)]
cannot be neglected for a sub-5 nm pore in an electrolyte with less
ICP σ0, σs, λD (c0), geo Δc/c0  1 [Eq. (26)]
than 100 mM. However, these two factors will not severely change
(dp, r0), V
the translocation waveform as discussed earlier. Hence, a compen-
ICE V, εr, geo(h, r0) Vε0 εr /h  σ 0 , σ s
sation factor can be introduced in the model for analyte size extrac-
tion. The ICP effects can be obvious in most of the pores smaller
than 20 nm, especially for those in a membrane thicker than
the pure geometrical blockade, Figs. 10(a) and 10(b) show the rela- 20 nm. As the ICP effects can totally change the shape of transloca-
tive blockage amplitude of a 6 nm-diameter sphere translocating a tion waveform, it is difficult to build a model for analyte size
cylindrical nanopore of different dp, at different c0 and σ0, by consid- extraction. In this case, calibration curves are a feasible and practi-
ering both surface conductance and EOF. The relative blockage cal approach to correlating the analyte information to the blockage
amplitude is normalized to the current level of the pure geometri- current. Such curves can be constructed by simulation and standar-
dized experiments. Furthermore, caution should be exercised to the
cal effect. High c0, low σ0, and large dp all favor the dominance of
ICE effects for the nanopore in 2D materials, such as graphene and
the geometrical effect with the normalized relative amplitude
single-layer MoS2. The extremely thin membrane comes with a
close to 1. Same observations can be made for the translocation of
high parasitic capacitance and the changing-discharging process
a cylindrical analyte, as shown in supplementary material, can significantly modulate the surface charge density of a
Figs. S13(a) and S13(b). nanopore.
Furthermore, the relative blockage amplitude for the same
sphere translocation is extracted from the simulation and compared
in Figs. 10(c) and 10(d) for sphere analyte and supplementary 3. Further considerations
material, Figs. S13(c) and S13(d) for a cylinder analyte at different In this work, a cylindrical nanopore is selected as a typical
bias voltages by involving various factors. For both positively and and the simplest example. For other nanopore shapes, the influen-
negatively charged sphere, the ICP coupling with EOF can modu- tial factors should give the same effects discussed for the cylindrical
late the relative amplitude determinatively, especially at large bias pore. By invoking the concept of Leff, other nanopore shapes can
voltages. The influence of these factors on the blockage amplitude be approximated by a cylinder and our models reviewed earlier can
is less significant for the translocation of the cylinder than that of be applicable for the study of factors potentially affecting the trans-
the sphere. As summarized in Table I, these factors have their location signal.
respective determinatives and criteria for safe neglect. In the simulation, the translocating analyte was positioned at
In order to elucidate the influence of these factors in practical different locations along the central axis for snapshots of the ionic
situations, the significance of these effects is evaluated for a few current and other physical quantities during a designed

TABLE II. Effects of different factors in typical experimental cases.

Diameter Surface conductance EOF ICP

dp (nm) 0.1M 1M 0.1M 1M 0.1 M 1M ICE
2D nanopore h = 1 nm 1 ✓ ✓ ✓ – ✓ ✓ ✓
5 ✓ – ✓ – ✓ ✓ ✓
20 – – – – – – ✓
SiNx nanopore h = 20 nm 5 ✓ – ✓ – ✓ ✓ –
20 – – – – ✓ ✓ –
50 – – – – – – –
Nanopipette h = 1 μm 5 ✓ – ✓ – ✓ ✓ –
20 – – – – ✓ ✓ –
50 – – – – ✓ – –

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-15

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

translocation. The velocity of the analyte that may affect the trans- DATA AVAILABILITY
location current45,68–70 was not considered. Interaction between the The data that support the findings of this study are available
translocating analyte and the current is a significant subject and from the corresponding author upon reasonable request.
will motivate our further work.

IV. CONCLUSIONS
We have scrutinized major possible mechanisms in nanopore REFERENCES
systems, which may influence the translocation blockage current by 1
J. Hurley, “Sizing particles with a Coulter counter,” Biophys. J. 10(1), 74–79
using a sphere and a long cylinder object through a cylindrical (1970).
nanopore as typical examples. The related factors include geometry, 2
C. Wen and S.-L. Zhang, “Fundamentals and potentials of solid-state nano-
surface conductance, EOF, ICP, and ICE. For a nanopore with a pores: A review,” J. Phys. D: Appl. Phys. 54, 023001 (2020).
larger diameter and a weaker surface charge density in a higher 3
M. Di Ventra and M. Taniguchi, “Decoding DNA, RNA and peptides with
concentration electrolyte at a lower bias voltage, the blockage quantum tunnelling,” Nat. Nanotechnol. 11(2), 117–126 (2016).
amplitude is more determined by pure geometry. On the other
4
J. D. Spitzberg et al., “Plasmonic nanopore biosensors for superior single mole-
hand, the effects originated from the surface charge are more pro- cule detection,” Adv. Mater. 31(23), 1900422 (2019).
5
D. Deamer, M. Akeson, and D. Branton, “Three decades of nanopore sequenc-
nounced in smaller pores with lower electrolyte concentration.
ing,” Nat. Biotechnol. 34(5), 518–524 (2016).
The surface charge causes two direct phenomena influencing 6
M. Rhee and M. A. Burns, “Nanopore sequencing technology: Research trends
the translocation waveform: (a) surface conductance, weakening the and applications,” Trends Biotechnol. 24(12), 580–586 (2006).
relative amplitude; (b) ICP, making the waveform asymmetrical. 7
E. C. Yusko et al., “Real-time shape approximation and fingerprinting of
The surface conductance of the translocating object and pore side- single proteins using a nanopore,” Nat. Nanotechnol. 12(4), 360–367
wall independently contribute to the current when (dp/ (2017).
2 − r0) > 4λD. Otherwise, the surface charge in EDL of the translo- 8
J. Houghtaling et al., “Estimation of shape, volume, and dipole moment of indi-
cation object and pore sidewall overlaps and determines the radial vidual proteins freely transiting a synthetic nanopore,” ACS Nano 13, 5231
conductivity distribution together according to their polariton and (2019).
density. For a nanopore with the diameter much larger than the
9
J. Sha et al., “Identification of spherical and nonspherical proteins by a solid-
state nanopore,” Anal. Chem. 90(23), 13826–13831 (2018).
Debye length, the ICP can be ignored and the translocation wave- 10
J. Larkin et al., “High-bandwidth protein analysis using solid-state nanopores,”
form is symmetrical. On the contrary, the waveform becomes more Biophys. J. 106(3), 696–704 (2014).
and more asymmetrical when the ICP is pronounced. 11
D. S. Talaga and J. Li, “Single-molecule protein unfolding in solid state nano-
The current contribution by EOF is only 10%–25% of surface pores,” J. Am. Chem. Soc. 131(26), 9287–9297 (2009).
conductance, which will weaken the relative amplitude of the spike 12
R. Hu et al., “Differential enzyme flexibility probed using solid-state nano-
signal. However, EOF can couple with ICP resulting in a more pores,” ACS Nano 12(5), 4494–4502 (2018).
obvious influence on the distribution of ions, as well as the translo- 13
W.-J. Lan et al., “Nanoparticle transport in conical-shaped nanopores,” Anal.
cation waveform, than just contributing the surface conductance, Chem. 83(10), 3840–3847 (2011).
especially at higher bias voltages. Polarity of the surface charge on
14
I. W. Leong, et al., “Back-side polymer-coated solid-state nanopore sensors,”
ACS Omega 4(7), 12561–12566 (2019).
the translocating object and pore sidewall influences the EOF and 15
M. Tsutsui et al., “Temporal response of ionic current blockade in
ICP obviously. Furthermore, the ICE can be safely ignored in
solid-state nanopores,” ACS Appl. Mater. Interfaces 10(40), 34751–34757
general measurement situations with bias voltage less than 1 V. (2018).
16
J. C. Maxwell, A Treatise on Electricity and Magnetism, 3rd ed (Dover
SUPPLEMENTARY MATERIAL Publications, Inc., New York, 1954).
17
R. W. DeBlois and C. P. Bean, “Counting and sizing of submicron parti-
See the supplementary material for additional discussions on cles by the resistive pulse technique,” Rev. Sci. Instrum. 41(7), 909–916
formula derivations, numerical simulation configuration, error (1970).
analysis, and other supporting results such as concentration distri- 18
N. B. Grover et al., “Electrical sizing of particles in suspensions,” Biophys. J.
bution in nanopores with different diameters, blockage current fea- 9(11), 1398–1414 (1969).
tures without ICP, fractional concentration change caused by ICP, 19
H. Fricke, “A mathematical treatment of the electrical conductivity of colloids
electroosmotic current of a positively charged sphere translocating, and cell suspensions,” J. General Physiol. 6(4), 375–384 (1924).
distribution of ion concentration during the translocation of a
20
H. Fricke, “The electric permittivity of a dilute suspension of membrane
covered ellipsoids,” J. Appl. Phys. 24(5), 644–646 (1953).
sphere, translocation waveform of a long cylinder, and relative 21
J. A. Osborn, “Demagnetizing factors of the general ellipsoid,” Phys. Rev.
blockage amplitude of the cylinder translocation.
67(11–12), 351–357 (1945).
22
S. Velick and M. Gorin, “The electrical conductance of suspensions of ellip-
ACKNOWLEDGMENTS soids and its relation to the study of avian erythrocytes,” J. Gen. Physiol. 23(6),
753–771 (1940).
The authors would like to thank Dr. Ralph Scheicher for 23
W. R. Smythe, “Flow around a spheroid in a circular tube,” Phys. Fluids 7(5),
assistance in the COMSOL simulation implemented on the 633 (1964).
Swedish National Infrastructure for Computing (SNIC). This work 24
A. Han et al., “Label-free detection of single protein molecules and protein−protein
partially benefited from the grants from Stiftelsen Olle Engkvist interactions using synthetic nanopores,” Anal. Chem. 80(12), 4651–4658
Byggmästare (Nos. 2016/39 and 194-0646). (2008).

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-16

Published under license by AIP Publishing.
 Journal of ARTICLE scitation.org/journal/jap
Applied Physics

25
T. Ito, L. Sun, and R. M. Crooks, “Simultaneous determination of the size and 48
W.-L. Hsu, J. Hwang, and H. Daiguji, “Theory of transport-induced-charge
surface charge of individual nanoparticles using a carbon nanotube-based electroosmotic pumping toward alternating current resistive pulse sensing,” ACS
Coulter counter,” Anal. Chem. 75(10), 2399–2406 (2003). Sens. 3(11), 2320–2326 (2018).
26
D. C. Golibersuch, “Observation of aspherical particle rotation in Poiseuille 49
L. Jubin, et al., “Dramatic pressure-sensitive ion conduction in conical nano-
flow via the resistance pulse technique. II. Application to fused sphere ‘dumb- pores,” Proc. Natl. Acad. Sci. U.S.A. 115(16), 4063–4068 (2018).
bells,’” J. Appl. Phys. 44(6), 2580–2584 (1973). 50
H. Liu, S. Qian, and H. H. Bau, “The effect of translocating cylindrical parti-
27
D. C. Golibersuch, “Observation of aspherical particle rotation in Poiseuille cles on the ionic current through a nanopore,” Biophys. J. 92(4), 1164–1177
flow via the resistance pulse technique,” Biophys. J. 13(3), 265–280 (1973). (2007).
28
D. Coglitore et al., “Metal alloy solid-state nanopores for single nanoparticle 51
C.-H. Huang et al., “Electrophoresis of a soft particle within a cylindrical pore:
detection,” Phys. Chem. Chem. Phys. 20(18), 12799–12807 (2018). Polarization effect with the nonlinear Poisson−Boltzmann equation,” J. Phys.
29
D. Coglitore et al., “Revealing the nanoparticle-protein corona with a solid- Chem. B 114(31), 10114–10125 (2010).
state nanopore,” Materials 12(21), 3524 (2019). 52
N. Giamblanco et al., “Detection of protein aggregate morphology through
30
R. M. M. Smeets et al., “Salt dependence of ion transport and DNA transloca- single antifouling nanopore,” Sens. Actuators B: Chem. 260, 736–745 (2018).
tion through solid-state nanopores,” Nano Lett. 6(1), 89–95 (2006). 53
C. Gu et al., “Experimental study of protein translocation through MoS2 nano-
31
K. E. Venta et al., “Gold nanorod translocations and charge measurement pores,” Appl. Phys. Lett. 115(22), 223702 (2019).
through solid-state nanopores,” Nano Lett. 14(9), 5358–5364 (2014). 54
J. E. Reiner et al., “Theory for polymer analysis using nanopore-based single-
32
B. Luan and G. Stolovitzky, “An electro-hydrodynamics-based model for the molecule mass spectrometry,” Proc. Natl. Acad. Sci. U.S.A. 107(27),
ionic conductivity of solid-state nanopores during DNA translocation,” 12080–12085 (2010).
Nanotechnology 24(19), 195702 (2013). 55
C. Wen, Z. Zhang, and S.-L. Zhang, “Physical model for rapid and accurate
33
Y. He et al., “Controlling DNA translocation through gate modulation of determination of nanopore size via conductance measurement,” ACS Sens.
nanopore wall surface charges,” ACS Nano 5(7), 5509–5518 (2011). 2(10), 1523–1530 (2017).
34
L.-H. Yeh et al., “Ion concentration polarization in polyelectrolyte-modified 56
L. Bocquet and E. Charlaix, “Nanofluidics, from bulk to interfaces,” Chem.
nanopores,” J. Phys. Chem. C 116(15), 8672–8677 (2012). Soc. Rev. 39(3), 1073–1095 (2010).
35
S. Pal et al., “Effect of single nanoparticle-nanopore interaction strength on 57
B. Henrik, “Electroosmosis,” in Theoretical Microfluidics (Oxford University
ionic current modulation,” Sens. Actuators B Chem. 325, 128785 (2020). Press, New York, 2008), pp. 157–170.
36
W.-J. Lan et al., “Effect of surface charge on the resistive pulse waveshape 58
N. Di Fiori et al., “Optoelectronic control of surface charge and translocation
during particle translocation through glass nanopores,” J. Phys. Chem. C 118(5), dynamics in solid-state nanopores,” Nat. Nanotechnol. 8(12), 946–951 (2013).
2726–2734 (2014). 59
S. W. Kowalczyk et al., “Modeling the conductance and DNA blockade of
37
K. Chen et al., “Biphasic resistive pulses and ion concentration modulation solid-state nanopores,” Nanotechnology 22(31), 315101 (2011).
during particle translocation through cylindrical nanopores,” J. Phys. Chem. C 60
C. Wen et al., “On rectification of ionic current in nanopores,” Anal. Chem.
119(15), 8329–8335 (2015). 91(22), 14597–14604 (2019).
38
Y. Qiu, “Optimal voltage for nanoparticle detection with thin nanopores,” 61
J. Cervera, et al., “Asymmetric selectivity of synthetic conical nanopores
Analyst 143(19), 4638–4645 (2018). probed by reversal potential measurements,” J. Phys. Chem. C 111(33),
39
K. Chen et al., “Direction- and salt-dependent ionic current signatures for 12265–12273 (2007).
DNA sensing with asymmetric nanopores,” Biophys. J. 112(4), 674–682 (2017). 62
L.-J. Cheng and L. J. Guo, “Rectified ion transport through concentration gra-
40
S. Zhai and H. Zhao, “Influence of concentration polarization on DNA trans- dient in homogeneous silica nanochannels,” Nano Lett. 7(10), 3165–3171
location through a nanopore,” Phys. Rev. E 93(5), 052409 (2016). (2007).
41
Y. Qiu et al., “Highly charged particles cause a larger current blockage in 63
C. Wen et al., “Autogenic analyte translocation in nanopores,” Nano Energy
micropores compared to neutral particles,” ACS Nano 10(9), 8413–8422 (2016). 60, 503–509 (2019).
42
Y. Qiu et al., “Direction dependence of resistive-pulse amplitude in conically 64
S. Zeng et al., “Rectification of protein translocation in truncated pyramidal
shaped mesopores,” Anal. Chem. 88(9), 4917–4925 (2016). nanopores,” Nat. Nanotechnol. 14(11), 1056–1062 (2019).
43
J.-P. Hsu et al., “Importance of boundary on the electrophoresis of a soft cylin- 65
H. Chang et al., “DNA-mediated fluctuations in ionic current through silicon
drical particle,” J. Phys. Chem. B 116(41), 12626–12632 (2012). oxide nanopore channels,” Nano Lett. 4(8), 1551–1556 (2004).
44
S. Y. Lee et al., “Diffusiophoretic motion of a charged spherical particle in a 66
Y. Yao et al., “On induced surface charge in solid-state nanopores,” Langmuir
nanopore,” J. Phys. Chem. B 114(19), 6437–6446 (2010). 36(30), 8874–8882 (2020).
45
S. R. German et al., “Controlling nanoparticle dynamics in conical nanopores,” 67
C. Lee et al., “Large apparent electric size of solid-state nanopores due to spa-
J. Phys. Chem. C 117(1), 703–711 (2013). tially extended surface conduction,” Nano Lett. 12(8), 4037–4044 (2012).
46
M. Firnkes et al., “Electrically facilitated translocations of proteins through 68
M. Tsutsui et al., “Solid-state nanopore time-of-flight mass spectrometer,”
silicon nitride nanopores: Conjoint and competitive action of diffusion, electro- ACS Sens. 4(11), 2974–2979 (2019).
phoresis, and electroosmosis,” Nano Lett. 10(6), 2162–2167 (2010). 69
M. Tsutsui et al., “Particle trajectory-dependent ionic current blockade in
47
H. Wu et al., “The estimation of field-dependent conductance change of nano- low-aspect-ratio pores,” ACS Nano 10(1), 803–809 (2016).
pore by field-induced charge in the translocations of AuNPs-DNA conjugates,” 70
S. Movahed and D. Li, “Electrokinetic motion of a rectangular nanoparticle in
J. Phys. Chem. C 118(46), 26825–26835 (2014). a nanochannel,” J. Nanopart. Res. 14(8), 1032 (2012).

J. Appl. Phys. 129, 064702 (2021); doi: 10.1063/5.0035113 129, 064702-17

Published under license by AIP Publishing.