ArticlePDF Available

On-chip surface modified nanostructured ZnO as functional pH sensors


Abstract and Figures

Zinc oxide (ZnO) nanostructures are promising candidates as electronic components for biological and chemical applications. In this study, ZnO ultra-fine nanowire (NW) and nanoflake (NF) hybrid structures have been prepared by Au-assisted chemical vapor deposition (CVD) under ambient pressure. Their surface morphology, lattice structures, and crystal orientation were investigated by scanning electron microscopy (SEM), x-ray diffraction (XRD), and transmission electron microscopy (TEM). Two types of ZnO nanostructures were successfully integrated as gate electrodes in extended-gate field-effect transistors (EGFETs). Due to the amphoteric properties of ZnO, such devices function as pH sensors. We found that the ultra-fine NWs, which were more than 50 μm in length and less than 100 nm in diameter, performed better in the pH sensing process than NW–NF hybrid structures because of their higher surface-to-volume ratio, considering the Nernst equation and the Gouy–Chapman–Stern model. Furthermore, the surface coating of (3-Aminopropyl)triethoxysilane (APTES) protects ZnO nanostructures in both acidic and alkaline environments, thus enhancing the device stability and extending its pH sensing dynamic range.
Content may be subject to copyright.
This content has been downloaded from IOPscience. Please scroll down to see the full text.
Download details:
IP Address:
This content was downloaded on 13/08/2015 at 00:35
Please note that terms and conditions apply.
On-chip surface modified nanostructured ZnO as functional pH sensors
View the table of contents for this issue, or go to the journal homepage for more
2015 Nanotechnology 26 355202
Home Search Collections Journals About Contact us My IOPscience
On-chip surface modied nanostructured
ZnO as functional pH sensors
Qing Zhang, Wenpeng Liu, Chongling Sun, Hao Zhang, Wei Pang,
Daihua Zhang and Xuexin Duan
State Key Laboratory of Precision Measuring Technology and Instruments, College of Precision Instrument
and Opto-electronics Engineering, Tianjin University, Tianjin 300072, Peoples Republic of China
E-mail: and
Received 4 March 2015, revised 18 May 2015
Accepted for publication 21 May 2015
Published 12 August 2015
Zinc oxide (ZnO) nanostructures are promising candidates as electronic components for
biological and chemical applications. In this study, ZnO ultra-ne nanowire (NW) and nanoake
(NF) hybrid structures have been prepared by Au-assisted chemical vapor deposition (CVD)
under ambient pressure. Their surface morphology, lattice structures, and crystal orientation were
investigated by scanning electron microscopy (SEM), x-ray diffraction (XRD), and transmission
electron microscopy (TEM). Two types of ZnO nanostructures were successfully integrated as
gate electrodes in extended-gate eld-effect transistors (EGFETs). Due to the amphoteric
properties of ZnO, such devices function as pH sensors. We found that the ultra-ne NWs, which
were more than 50 μm in length and less than 100 nm in diameter, performed better in the
pH sensing process than NWNF hybrid structures because of their higher surface-to-volume
ratio, considering the Nernst equation and the GouyChapmanStern model. Furthermore, the
surface coating of (3-Aminopropyl)triethoxysilane (APTES) protects ZnO nanostructures in both
acidic and alkaline environments, thus enhancing the device stability and extending its
pH sensing dynamic range.
Keywords: ZnO nanostructures, pH sensor, eld-effect-transistors
(Some gures may appear in colour only in the online journal)
1. Introduction
In recent years, construction of functional bio-sensing devices
using nanomaterials has triggered tremendous interest in the
eld of nanotechnology [17]. The high surface-to-volume
ratio (S/V ratio) of the nanostructures has the unique advan-
tages of absorbing a lot more biomolecules (e.g., biomarkers
[8], enzymes [9], and bio-receptors [10]) at the sensor surface,
thus increases the sensor sensitivity [810]. Nanostructure-
based biosensors promise to revolutionize the eld of bio-
analytical research by offering ultrasensitive, rapid, and label-
free detection of the target biomolecules [1113]. ZnO is one
of the most promising materials to conduct as a nano-bio-
sensor. It has a wide, direct bandgap and large exciton
binding energy [14,15]. Besides, the diverse and abundant
ZnO nanostructures, such as nanowires (NWs) [1618],
nanoakes (NFs) [19], nanobelts [20,21], nanobows [22],
and nanohelices [23], open novel designs and applications of
Recently, ZnO nanostructures have been applied for
pH sensing applications due to their unique amphoteric
properties (reacting with both acidic and alkaline solutions)
[24]. Many biological processes release protons during bio-
chemical reactions [25], and various bioelectronics have been
developed based on pH sensors [26], such as glucose [27],
cholesterol [28], and intracellular sensing devices [25]. Ion-
sensitive eld-effect transistors (ISFETs) are the most popular
miniature pH sensors [26]. The mechanism of pH sensing
with an ISFET is based on the change of the surface prop-
erties of the gate dielectric, which responds to different
pH values. The amphoteric nature of the hydroxyl groups at
the solution-dielectric interface allows the surface charge state
to change as it protonates and deprotonates. The most
important component of an ISFET sensor is the pH-sensitive
Nanotechnology 26 (2015) 355202 (9pp) doi:10.1088/0957-4484/26/35/355202
0957-4484/15/355202+09$33.00 © 2015 IOP Publishing Ltd Printed in the UK1
membrane, which can respond to the H
ions in solutions
[29]. Previous studies have demonstrated the possibilities of
integrating ZnO as pH-sensitive membranes in ISFET.
However, most studies focused on the ZnO nanostructures
prepared by the liquid-phase deposition, which suffers non-
uniformity and relatively large feature size with uncontrol-
lable defects [30]. Therefore, when integrated into functional
devices, they have poor measurement repeatability. Besides,
due to their amphoteric properties, ZnO nanostructures tend to
dissolve in strong acidic or alkaline conditions, which limited
their pH sensing stability. Great efforts have been made to
modify the ZnO nanostructures with different doping mate-
rials (e.g., Si [31], Al [32], and Ta [33]), which stabilized the
ZnO nanostructures and extended their sensing dynamic
range. However, these methods require expensive and com-
plicated procedures, and potentially they would introduce
In this work, we fabricated ZnO nanostructures in gas
phase with Au-assisted chemical vapor deposition (CVD),
which has the advantage of fabricating defect-free nanos-
tructures with ultra-high S/V ratio. Two types of nanos-
tructures with different S/V ratios were prepared. Their
pH sensing properties have been compared before and after
the integration of nanostructures with functional eld effect
transistor (FET) devices. In addition, in order to stabilize the
ZnO nanostructures in ionic solutions, a simple amino-sila-
nization step was introduced to modify the ZnO surface, and
therefore it extended their pH sensing range and further
increased their pH sensitivity.
2. Materials and methods
2.1. Materials
ZnO powder and graphite powder were purchased from
Beijing Gold Crown for the New Material Technology Co.,
Ltd (3-Aminopropyl) triethoxysilane (APTES) was purchased
from the Aladdin Industrial Corporation. NaH
were purchased from Tianjin Heowns Biochem LLC.
All materials were used as received without additional
2.2. Growth of ZnO Nanostructures by CVD
The ZnO nanostructures were prepared with the following
four steps. Firstly, a 300 nm SiO
lm was coated on a Si
substrate by chemical deposition (Applied Materials, Preci-
sion 5000) process to isolate Si from the catalyst metal. Then,
a thin Cr/Au layer was sputtered on the substrate by electron
beam evaporation (CHA-600), where the 5 nm Cr and
25 nm Au lm acted as conducting layer and catalyst,
respectively. Afterwards, NWs and NWNF hybrid nanos-
tructures were grown on the substrate by a vapor liquid solid
(VLS) process. Finally, the prepared substrates were synthe-
sized on a Si/SiO
substrate in a tube furnace.
Before the VLS process, the four- inch Si/SiO
wafer was cut into 0.8 cm ×0.8cm squares. For the CVD
process, source materials consisted of ZnO powder (0.8 g,
purity 99.99%) and graphite powder (0.8 g, purity 99.99%)
(w:w = 1:1). The source material was put in a tube furnace
under O
/Ar atmosphere. The furnace was heated up from
room temperature (20 °C) to 820 °C at the rate of
10 °C min
, and then from 820950 °C at the rate of
5 °C min
. The furnace was kept at 950 °C for growing
nanostructures. In order to grow NWs, the substrates were
mounted vertically in the chamber at the downstream of the
source materials, and to grow NWNF hybrid structures, the
substrates were put horizontally. The distance between the
substrates and the source materials was 1 2 mm. The
growing time was around 30 min. After these steps, the whole
system was naturally cooled down to room temperature.
2.3. Surface functionalization of ZnO
Before functionalization, ZnO NWs and NWNF hybrid
structures were treated with oxygen plasma for 30 s to remove
all the impurities and organic contaminants. This step was
followed by an overnight APTES vapor deposition under
reduced pressure in a desiccator to generate an amino layer on
the ZnO surface.
2.4. Preparation of testing pH solution
The pH test solution was prepared by dissolving NaH
and C
in mill-Q water and adjusting to the desired
pH with a strong acid or alkali.
2.5. Characterization and measurement
The top-view and tilted-view microstructures of ZnO NWs
and NWNFs were characterized by scanning electron
microscopy (SEM, FEI Inspect F50). X-ray diffraction (XRD)
was carried out by D/MAX-2500 (Rigaku, Japan). Trans-
mission electron microscopy (TEM, FEI Tecnai G2 F20) was
used to characterize the lattice structure of ZnO NWs.
The pH sensing experiments were measured using an
extended-gate FET (EGFET) setup. A front-end sensing chip
and back-end transistors were integrated on a single printed
circuit board, accompanying signal amplication, and a data
acquisition interface for personal computers. The front-end
sensing chip was a micro-fabricated Au electrode with a
typical width of 20 μm. On top of the Au electrode, different
ZnO nanostructures were incorporated. A mixing cell (solu-
tion chamber) was then created on top of the sensing electrode
by epoxying thin-walled, 8 mm diameter PTFE tubing to the
chip surface and by inserting thinner tubing (0.5 mm) to serve
as the uid supply and return. The solution input tube was
placed directly over the central region of the sensor. This
system enabled continual mixing (equivalent to pipetting up-
and-down) throughout the course of the sensing measure-
ments. A miniature Ag/AgCl reference electrode (Harvard
Apparatus) was inserted into the solution chamber as a
solution gate during all tests. The I
characteristic of off-
chip EGFET at a constant V
(100 mV) could be tested
thoroughly off-line before performing any measurements,
serving as a look-up table for converting the measured drain
Nanotechnology 26 (2015) 355202 Q Zhang et al
30 32 34 36 38 40
2 Theta (Degree) 2 Theta (Degree)
(arb. unit)
(arb. unit)
Zn0 (002)
Zn0 (100)
Zn0 (101)
Zn0 (002)
Si (100)
Figure 1. Characterizations of ZnO nanostructures. SEM images of (a) top-view of NWNFs, (b) tilted-view of NWNFs, (c) tilted-view of
NWs, and (d) zoom-in of (c). TEM images of (e) a typical view of a single NW with 10 μm in length and a selected area electron diffraction
(SAED) image (inset) give the [0001] growth direction, respectively. (f) Typical TEM image with HRTEM picture (top inset) and other
SAED picture (lower inset). XRD patterns of ZnO NWs with (g) wide range from 2080°; (h) narrow range from 3040°.
Nanotechnology 26 (2015) 355202 Q Zhang et al
current I
back into the interface potential. IVand DC time
measurements were carried out by a customized system using
a National Instruments data acquisition card and two Keithley
2400 source measure units. All measurements were per-
formed at room temperature.
3. Results and discussions
3.1. Characterization of ZnO nanostructures
In this work, ZnO NWs and NWNFs were successfully
prepared through carefully controlling the CVD process
conditions. As shown in SEM images (gures 1(a)(d)), the
NWNF hybrid structures are around 500 nm in width, while
the NWs are less than 100 nm; both structures extended more
than 50 μm in length. Thus, the nanostructures feature an
ultra-high S/V ratio. As shown in gures 1(a) and (b), the
NWNF hybrid structures are uniformly distributed
throughout the whole substrate, and the NFs are grown on the
top of NWs. As for the NW structures (gures 1(c) and (d)),
the NWs are arranged vertically on the substrate with a great
uniformity in diameter and length. Compared to the NWs, the
NWNF hybrid structures are more diverse in position,
dimension, and shape.
TEM is employed to investigate the crystal structures of
ZnO nanostructures. Figures 1(e) and (f) show the TEM
images of a typical single ZnO NW with a diameter of 80 nm.
The inset of gure 1(e) is the electron diffraction pattern of a
selected area, which demonstrates that the NW grows along
[0001], and the volume is free from dislocations. Inset of
gure 1(f) presents a high-resolution TEM (HRTEM) image,
which shows a lattice distance of c= 5.2 Å. This conrms the
high crystallization quality of ZnO NWs with a clean and
structurally perfect surface. The surface of the NW is atom-
ically sharp and has no sheathed amorphous phase.
Figures 1(g) and (h) show the XRD patterns of ZnO NWs
and NWNF hybrid structures. The XRD images clearly
show the diffraction peaks of ZnO (002) and Si (100) in
gure 1(g). Here, the ZnO (002) diffraction peak is from ZnO
nanostructures, and the Si (100) diffraction peak is from the
Si substrate. Figure 1(h) shows the XRD result from 30° to
40°, which conrms the hexagonal wurtzite phase of the ZnO
nanostructures. The ZnO nanostructures are highly crystal-
line, grown by the CVD process at high temperature (950 °C).
The ZnO diffraction peaks are clearly detected, and the peaks
intensity agrees with SEM and TEM results.
3.2. Device integration
An EGFET setup is used to evaluate the pH sensing proper-
ties of the ZnO nanostructures (gure 2). An EGFET is
developed from an ISFET and has the advantage of main-
taining higher device stability by connecting an auxiliary
sensing electrode to the gate terminal of a conventional
metallic oxide semiconductor eld effect transistor (MOSFET),
thus separating the FET device from the chemical or biological
environment and preventing potential damage of the FET [6
8]. Such a system also facilitates the integrating of ZnO
nanostructures into the FET measurement. Au is used as the
catalyst for ZnO growth in the CVD process, and, in principle,
such an Au layer can be used as the sensing electrode of
EGFET. Thus ZnO nanostructures and EGFETs can share the
same Au layer. However, the Au layer will melt below 900°C
and is not able to survive the high temperature during ZnO
growth (typically during 900950 °C). To integrate the
nanostructures into the EGFET, a metal layer (Cr) is deposited
between the Au and the substrate before growing ZnO. This Cr
layer can survive the high temperature process and strengthen
the adhesion between ZnO nanostructures and the substrate.
The high conductivity of Cr effectively allows for the electrical
changes of ZnO during pH sensing.
3.3. pH sensing results
Figure 3shows the pH sensing results of bare ZnO nanos-
tructures. Figures 3(a) and (b) are typical static measurements
of the I
curve from the ZnO nanostructures, which are
connected to the gate terminal of the EGFET. A clear dif-
ference can be observed after exposure to different
pH solutions, which indicates successful sensing. We calcu-
lated the transconductance (g
)bylineartting of the IV
curve and found that the g
change is very small (<4.74%),
and the change of the I
induced by different pH solutions
can be described as a parallel (threshold voltage) shift of the
. It is also noticed that by changing the same pH, the
threshold voltage shift of the ZnO NW modied electrode
Figure 2. Schematic diagram of (a) ZnO nanostructure-based
EGFET pH sensor, and (b) APTES-modied ZnO nanostructures.
(c) The test circuit picture.
Nanotechnology 26 (2015) 355202 Q Zhang et al
(gure 3(b)) is larger than that of the NWNF-modied elec-
trode (gure 3(a)) (average 34.74 mV pH
for NWNFs,
while 36.65 mV pH
for NWs).
More reliable pH determination is achieved by real-time
measurements from the same device. Figure 3(c) shows real-
time FET current (I
) responses of ZnO NWs by exposure to
different pH solutions, with V
xed at 0.7 V and V
xed at
0.1 V. This sensor immersed for 300s in each buffer solution.
The I
decreased each time when the pH value increased
from 49 and increased as the pH value decreased from 94.
This measurement exhibits excellent real-time response with
multiple scans, which proves the reliability of ZnO NWs as
pH sensors. The potentials of the sensor in gure 3(c) exhibit
almost complete overlapping with less than 0.05 mV differ-
ences, indicating excellent recovery.
Since the solution pH change will induce the surface
potential change of ZnO, we converted the device current
responses to surface potential changes by dividing the
FET transconductance. The surface potential changes of
ZnO NWs and NWNFs in different pH solutions are com-
pared in gure 3(d). The ZnO NWNF sensor is found to
have weak pH sensitivity at 34.74 mV pH
, compared with
36.65 mV pH
for the ZnO NW sensor.
The ZnO NWs perform better than ZnO NWNF hybrid
structures for pH sensing (125% times higher in signal
intensity). This is due to the fact that NW structures have a
higher S/V ratio than NWNF hybrids, which indicates higher
surface-site densities. The pH response of ZnO nanostructures
is due to the ZnO surface, which can protonate or deproto-
nate, based on the electrolyte solution pH. The surface-site
densities of the ZnO are critical in their pH sensing ability,
according to the Nernst equation and the GouyChapman
Stern model [26]. As shown in formulas [1] and [2] below,
the H
absorption percentage of ZnO NWs is larger than that
of NWsNFs when they function as pH sensors.
In acid buffer: ZnO H ZnOH (1)
() ()
In basic buffer: ZnO 2H O Zn(OH) H (2)
() 2 3( )
+= +
Besides the morphology effects, the chemical surface
functionalization of ZnO nanostructures has also been
investigated to improve their pH sensing performance. In this
work, APTES is used to modify the surface of the ZnO
nanostructures. The results are shown in gure 4. Figures 4(a)
and (b) show the comparison of the static sensor responses for
the APTES-modied ZnO nanostructure sensors, in which the
is xed at 0.1 V. The threshold voltage is observed to shift
from lower to higher potential as the pH increases, and the
voltage shift of the NWs is higher than the NWNF hybrids,
which is similar to non-functionalized ZnO. The results in
Figure 3. Sensing response of V
versus I
under different buffer solutions of (a) ZnO NWNF hybrid structure and (b) ZnO NW-based
EGFET sensor. (c) Real-time I
responses to buffer solutions of a ZnO NW-based sensor in a pH 4-9-4 loop cycle at V
of 0.7 V. (d) The
variation of the reference electrode potential (V
) as a function of pH value for ZnO NWs and ZnO NWNF hybrid-based sensors.
Nanotechnology 26 (2015) 355202 Q Zhang et al
gure 4(c) show that this sensor is stable during the whole
test. The overlapping difference is less than 0.01 mV, indi-
cating good reproducibility. The APTES- modied sensor
maintains good repeatability under this pH sensing procedure;
this is because APTES can protect ZnO nanostructures in both
acid and alkaline solutions. The real-time response of the
APTES-modied ZnO NW sensor has also been investigated
(gure 4(d)). This sensor has been immersed for 300 s in each
pH buffer solution. The I
differs with the pH value change,
Figure 4. After APTES modication, the sensing responses of V
versus I
under different buffer solutions of (a) ZnO NWNF hybrid
structures and (b) a ZnO NW-based EGFET sensor. (c) Repeatability and stability measurement of an APTES-modied ZnO NW-based
sensor under a pH value 4-6-9 and repeated four times. (d) Real-time I
response to buffer solutions of an APTES-modied ZnO NW-based
sensor in a pH 2-9-2 loop cycle at V
of 0.7 V.
Figure 5. The variation of gate potential (V
) of an EGFET as a
function of pH value change for both APTES-modied and non-
modied ZnO NWs and ZnO NWNF hybrid-based sensors.
Figure 6. Statistical analysis of pH sensing results of ZnO NWs and
NWNF hybrid structures with and without APTES coatings.
Nanotechnology 26 (2015) 355202 Q Zhang et al
which also indicates reliable performance in the pH sensing
procedure. The potentials of the sensors in gure 4(d) (and
also gure 3(c)) exhibit a pH response hysteresis [38], which
shows average differences of less than 0.1 mV. Considering
that the sensitivities of ZnO NWs and APTES-coated ZnO
NWs are 36.65 mV pH
and 43.22 mV pH
, respectively,
the sensitivity and hysteresis are acceptable for the
pH electrode.
Figure 5compares the pH sensitivities of the APTES-
modied and non-modied ZnO nanostructures. In both NWs
and NWNF sensors, the APTES-modied sensors have
higher pH sensitivity than the non-modied sensors
(approximately 185% times higher). This is due to the fact
that APTES can induce more surface sites on ZnO nanos-
tructures, which is critical in pH sensing, as discussed above.
Besides the improvement of pH sensitivity, APTES can
protect ZnO nanostructures in an extremely acidic environ-
ment. We are able to get a pH response at 2, which is not
possible for bare ZnO.
Another consideration is the stability of the APTES
coatings. Silane can be coated either in solution or in gas
phase. In the solution phase, silane molecules tend to react
with each other via cross-linking of the alkoxy units, resulting
in rough, defective, and unordered multilayers. Such multi-
layers of silanization are less stable in aqueous solutions due
to the aqueous hydrolysis of the alkoxysilanes, thus suffering
degradation over time. However, in gas-phase silanization of
small silane molecules (e.g., APTES), usually a monolayer is
formed, and such a coating has better stability in aqueous
solutions. In a recent published paper [39], it has been
demonstrated that APTES-coated ZnO nanocrystals show
rather high aqueous stability. Regarding our results, as is
shown in gure 4(d), the APTES-modied ZnO NWs have
been exposed in pH value 2-4-6-9 and repeated four times.
For each step, the APTES-modied ZnO NW-based sensor
would immerse into the pH solution for 300 s during the test
then be washed in a buffer solution for another 300 s. So the
total test for stability is around 2 h. During this period, we did
not observe degradation of the APTES coatings.
It is also noticed that the device integrated with NWs and
modied with APTES performs better as a pH sensor, which
is in agreement with the results from the non-modied
sensors. The NWs have a higher S/V ratio compared with
NWNF hybrid structures; thus the APTES coating will result
in higher NH
density as well. To get more statistical results,
we plotted the pH sensing results with ve different samples.
As shown in gure 6, the differences of ZnO NWs and NW-
NF hybrid structures with and without APTES are statistically
3.4. Discussion of the role of APTES
The amino silane functionalization can protect ZnO in both
acid and alkaline solutions and enhance their chemical sta-
bility. More importantly, according to the Nernst equation
[3037], when all the surfaces of ZnO are protonated (in acid
buffer) or de-protonated (in alkaline buffer) under ideal
conditions, the surface sites can achieve the highest level; thus
we can get the highest surface potential changes. The sensi-
tivity of such a device is supposed to reach 59.1 mV pH
(Nernst potential). However, the protonation/deprotonation
process actually cannot reach such extreme conditions and
can only keep dynamic equilibrium according to H
concentration. As a result, enhancing the S/V ratio and
increasing surface protonation/ deprotonation sites can
effectively push the pH response closer to the limit value of
the Nernst equation. The introduction of APTES functiona-
lization can ideally increase more sites for protonation/
deprotonation compared with the bare ZnO, as indicated in
gure 7(a). Therefore, it is likely that the pH response of such
a device is contributed not only by ZnO (equations (1), (2))
but also by APTES (equation (3)). Covalently linking APTES
to the ZnO surface results in a surface terminating in both
and ZnO groups, which have different dissociation
constants, pKa. At low pH, the NH
group is protonated to
+(equation (3)), which is a dominant reaction compared
with ZnO protonation (equation (1)). This has been proven by
the extension of the pH sensing range (gure 5) and is less
damaging of ZnO after APTES coating. Thus the APTES
protonation functions as a positive gate and decreases the
conductance. At high pH, ZnOH is deprotonated to
(equation (2)), which correspondingly causes an
increase in conductance. The observed linear response can be
attributed to an approximately linear change in the total sur-
face charge density (versus pH) because of the combined acid
and alkaline behavior of both surface groups.
APTES in solution : NH H NH (3)
Besides the binding sites, the conductance of the ZnO is
another critical factor for reaching the limit value of the
Nernst equation. For a typical n-type ZnO semiconducting
material, APTES modication can provide electronics so as to
reduce the bandgap (the distance of conduction band E
valence band E
) of ZnO, as gure 7(b) shows. Obtaining
additional electronics, the E
of ZnO is getting close to the
Fermi level (E
), making it much easier to translate the
pH signal from the solution to the sensing board. The
pH response can be delivered with less loss in the transfer
process through more conductive ZnO. As a result, such
combinations of these effects indicate that the APTES
Figure 7. (a) The schematic diagram shows ZnO surface sites to
protonate/deprotonate before (right) and after (left) APTES mod-
ication. (b) Energy-band diagrams of the interface of ZnO and
APTES, where APTES provides electronics to ZnO so as to decrease
the E
and make ZnO more conductive.
Nanotechnology 26 (2015) 355202 Q Zhang et al
functionalization leads to higher pH sensitivity, as shown in
gure 5.
4. Conclusion
ZnO nanostructures are known for their wide, direct bandgap
and large exciton binding energy. They are promising nano-
materials to conduct as nanoelectronics. ZnO nanostructures
grown from the gas phase (CVD process) usually have better
quality compared with the solution process; however, the high
temperature process hinders their device integration. In this
study, defect-free ZnO nanostructures grown from the CVD
process are successfully prepared and integrated into EGFET.
Due to the unique amphoteric properties of ZnO, such devices
function as pH sensors. Two types of ZnO nanostructures
(ultra-ne NWs and NWNF hybrid structures) are prepared,
and their surface morphology, lattice structures, crystal
orientations, and pH sensing properties are carefully com-
pared. Furthermore, ZnO nanostructures are coated with
APTES monolayers to enhance their stability and pH sensing
dynamic range. Among the four types of ZnO nanostructures,
pH sensors made from the APTES-modied ZnO NWs
exhibit the highest pH sensitivity and largest dynamic
pH sensing range, which are due to their high surface to
volume ratio and surface coating effects (increasing more
surface sites for protonation/deprotonation and higher con-
ductivity). Such surface modication methods are much
simpler compared with the previously reported doping
method and can be applied to other types of nanomaterial-
based devices.
The authors gratefully acknowledge nancial support from
Tianjin Applied Basic Research and Advanced Technology
(14JCYBJC41500) and the 111 Project (B07014).
[1] Zhong W 2009 Nanomaterials in uorescence-based
biosensing Anal. Bioanal. Chem. 394 4759
[2] Lei J and Ju H 2012 Signal amplication using functional
nanomaterials for biosensing Chem. Soc. Rev. 41 212234
[3] Yang W et al 2010 Carbon nanomaterials in biosensors: should
you use nanotubes or grapheme Angew. Chem., Int. Ed. 49
[4] Luo J et al 2013 A new type of glucose biosensor based on
surface acoustic wave resonator using Mn-doped ZnO
multilayer structure Biosens. Bioelectron. 49 5128
[5] Lee CT, Chiu YS, Ho SC and Lee YJ 2011 Investigation of
a photoelectrochemical passivated ZnO-based glucose
biosensor Sensors 11 464855
[6] Kwon M J, Lee J, Wark A W and Lee H J 2012 Nanoparticle-
enhanced surface plasmon resonance detection of proteins at
attomolar concentrations: comparing different nanoparticle
shapes and sizes Anal. Chem. 84 17027
[7] Jang H R, Wark A W, Baek S H, Chung B H and Lee H J 2013
Ultrasensitive and ultrawide range detection of a cardiac
biomarker on a SPR platform Anal. Chem. 86 8149
[8] Yue H Y et al 2014 ZnO nanowire arrays on 3D heirarchical
graphene foam: biomarker detection of Parkinsons disease
ACS Nano 8163946
[9] Ma Q and Nakazato K 2014 Low-temperature fabrication of
ZnO nanorods/ferrocenylalkanethiol bilayer electrode and
its application for enzymatic glucose detection Biosens.
Bioelectron. 51 3625
[10] Lu X et al 2011 Piezoelectric biosensor platform based on ZnO
micro membrane Curr. Appl. Phys. 11 S2857
[11] Nath N and Chilkoti A 2004 Label-free biosensing by surface
plasmon resonance of nanoparticles on glass: optimization
of nanoparticle size Anal. Chem. 76 53708
[12] Gasparac R et al 2004 Ultrasensitive electrocatalytic DNA
detection at two- and three-dimensional nanoelectrodes
J. Am. Chem. Soc. 126 122701
[13] Chen X et al 2012 Enzyme-free and label-free ultrasensitive
electrochemical detection of human immunodeciency virus
DNA in biological samples based on long-range self-
assembled DNA nanostructures Anal. Chem. 84 827783
[14] Tan S T et al 2005 Blueshift of optical bandgap in ZnO thin
lms grown by metalorganic chemicalvapor deposition
J. Appl. Phys. 98 013505
[15] Sun H et al 2002 Enhancement of exciton binding energies in
ZnO/ZnMgO multiquantum wells J. Appl. Phys. 91 19937
[16] Pachauri V, Vlandas A, Kern K and Balasubramanian K 2010
Site-specic self-assembled liquid-gated AnO nanowire
transistors for sensing applications Small 658994
[17] Park J, Lee S, Lee J and Yong K 2013 A light incident angle
switchable ZnO nanorod memristor: reversible switching
behavior between two non-volatile memory devices Adv.
Mater. 25 64239
[18] Khaderbad M A et al 2012 Electrical actuation and readout in a
nanoelectromechanical resonator based on a laterally
suspended zinc oxide nanowire Nanotechnology 23 025501
[19] Wen W, Wu JM and Wang YD 2012 Large-size porous ZnO
akes with superior gas-sensing performance Appl. Phys.
Lett. 100 262111
[20] Gao P, Ding Y and Wang Z 2003 Crystallographic orientation-
aligned ZnO nanorods grown by a tin catalyst Nano Lett. 3
[21] Pan Z W, Dai Z R and Wang Z L 2001 Nanobelts of
semiconducting oxides Science 291 19479
[22] Hughes W L and Wang Z L 2004 Formation of piezoelectric
single-crystal nanorings and nanobows J. Am. Chem. Soc.
126 67039
[23] Gao P X et al 2005 Conversion of zinc oxide nanobelts into
superlattice-structured nanohelices Science 309 17004
[24] Batista P D and Mulato M 2005 ZnO extended-gate eld-effect
transistors as pH sensors Appl. Phys. Lett. 87 143508
[25] AlHilli S, AlMofarji R and Willander M 2006 Zinc oxide
nanorod for intracellular pH sensing Appl. Phys. Lett. 89
[26] Rothberg J M et al 2011 An integrated semiconductor device
enabling non-optical genome sequencing Nature 475
[27] Usman Ali S M, Nur O, Willander M and Danielsson B 2009
Glucose detection with a commercial MOSFET using a ZnO
nanowires extended gate IEEE Trans. Nanotechnol. 867883
[28] Ahmad R, Tripathy N and Hahn YB 2013 High-performance
cholesterol sensor based on the solution-gated eld effect
transistor fabricated with ZnO nanorods Biosens.
Bioelectron. 45 2816
[29] Menzel A et al 2011 Multifunctional ZnO-nanowire-based
sensor Funct. Mater. 21 43428
[30] Fulati A et al 2009 Miniaturized pH sensors based on zinc
oxide nanotubes/nanorods Sensors 9891123
Nanotechnology 26 (2015) 355202 Q Zhang et al
[31] Huang BR, Lin JC and Yang YK 2013 ZnO/silicon
nanowire hybrids extended-gate eld-effect transistors as
pH sensors J. Electrochem. Soc. 160 B7882
[32] Yang PYet al 2011 pH sensing characteristics of extended-
gate eld-effect transistor based on Al-doped ZnO
nanostructures hydrothermally synthesized at low
temperatures IEEE Electron Device Lett. 32 16035
[33] Chiu YS, Lee CT, Lou LR, Ho SC and Chuang CT 2013
Wide linear sensing sensors using ZnO: Ta extended-gate
eld-effect transistors Sensors Actuators B188 9448
[34] Chen CPet al 2011 Ultrasensitive in situ label-free DNA
detection using a GaN nanowire-based extended-gate eld-
effect transistor sensor Anal. Chem. 83 193843
[35] Chiu YS, Tseng CY and Lee CT 2012 Nanostructured
EGFET pH sensors with surface-passivated ZnO thin-lm
and nanorod array IEEE Sensors J. 12 9304
[36] Li HHet al 2013 Coaxial-structured ZnO/silicon nanowires
extended-gate eld-effect transistor as pH sensor Thin Solid
Films 529 1736
[37] AlHilli S M, Willander M, Ost A and Stralfors P 2007 ZnO
nanorods as an intracellular sensor for pH measurements
J. Appl. Phys. 102 084304
[38] Tsaia C-N, Choub J-C, Sunc T-P and Hsiung S-K 2005
Study on the sensing characteristics and hysteresis effect
of the tin oxide pH electrode Sensors Actuators B108
[39] Li S, Sun Z, Li R, Dong M, Zhang L, Qi W,
Zhang X and Wang H 2015 ZnO nanocomposites
modied by hydrophobic and hydrophilic silanes with
dramatically enhanced tunable uorescence and aqueous
ultrastability toward biological imaging applications Sci.
Rep. 58475
Nanotechnology 26 (2015) 355202 Q Zhang et al
... The pH sensors using ZnO nanorods, [162,163,164,165] nanotubes, [ 162] nanowalls, [166] nanowires, [167] and nanoflakes [167] have been produced. Quantitative measurements [162,168] contingent on the site-binding model [169] of the detection findings of ZnO nanorod based pH sensor have expected the highest Nernstian sensitivity of 59.1 mV/pH at 25 • C in the large pH region (1 -14). In another study, the pH monitoring based on an individual ZnO nanorod was reported [170]. ...
... It found that a variation in surface potential at the ZnO/solution interface was identified as a linear change in conductance of the individual nanorod in the pH scale (2 -12) of 8.5 nS/ pH with a resolution of ∼0.1 pH. It reported that the pH sensing performance was significantly affected by the structural properties and surface morphology (i.e., surface site density) of ZnO nanostructures [165,168]. Since the broader surface-to-volume ratio ZnO nanowires exhibited superlative sensitivity of 36.65 mV/pH compared to that of 34.74 mV/pH for the nanowires-nanoflakes hybrids [167]. Furthermore, because of the broader surface-to-volume area with the existence of surface and subsurface oxygen vacancies, ZnO nanotubes produce sensitivity (45.9 mV/pH) as high as twice that of ZnO nanorods (28.4 mV/pH) [162]. ...
Full-text available
In the last few decades’ electrochemical biosensors have witnessed vast developments due to the broad range of different applications, including health care and medical diagnosis, environmental monitoring and assessment, food industry, and drug delivery. Integration of nanostructured material with different disciplines and expertise of electrochemistry, solid-state physics, material science, and biology has offered the opportunity of a future generation of highly rapid, sensitive, stable, selective, and novel electrochemical biosensor devices. Among metal oxide nanomaterials, ZnO nanostructures are one of the most important nanomaterials in today’s nanotechnology research. Such nanostructures have been studied intensely not only for their extraordinary structural, optical, and electronic properties but also for their prominent performance in diverse novel applications such as photonics, optics, electronics, drug delivery, cancer treatment, bio-imaging, etc. However, functionality of these nanomaterials is eventually dictated by the capability to govern their properties including shape, size, position, and crystalline structure on the nanosized scale. This review aims to update the outstanding advancement in the developments of the enzymatic and non-enzymatic biosensors using a different structure of ZnO nanomaterials. After a coverage of the basic principles of electrochemical biosensors, we highlight the basic features of ZnO as a potential anticancer agent. focused attention gives to functionalized biosensors based on ZnO nanostructures for detecting biological analytes, such as glucose, cholesterol, L-lactic acid, uric acid, metal ions, and pH.
... When the sensor is immersed in an acidic solution, −NH 2 will be protonated to −NH 3 + . The reaction formula of APTES during measurement is as follows [17]: APTES in solution: ...
Full-text available
In this study, we deposited zinc oxide (ZnO) and aluminum-doped zinc oxide (AZO) on the electroless nickel immersion gold (ENIG) of a flexible printed circuit board (FPCB) as a potentiometric pH sensor. The sensing films of the pH sensor were fabricated by a radio frequency (RF) sputtering system and analyzed by field emission scanning electron microscope (FE-SEM) and X-ray photoelectron spectroscopy (XPS). In the pH 2 to 10 buffer solutions, it was observed that the characteristics of the pH sensor through the voltage–time (V-T) measurement system include average sensitivity and linearity, drift effect, and repeatability. According to the experimental results, the pH sensors in this study could exhibit good characteristics.
Full-text available
An electrochemical zinc oxide (ZnO) pH sensor is proposed and prototyped using a flexible printed circuit board (FPCB). The device is resistant to acidic and alkaline solutions and exhibits excellent electrical conductivity. The sensing film was deposited by radio-frequency sputtering, providing a dense ZnO layer, and (3-aminopropyl)triethoxysilane (APTES) was used to enhance the chemical stability of the ZnO in acidic solutions. The proposed pH sensor has an average sensitivity of 37.52 mV/pH, linearity of 0.995, and a repeatable RSD value of 2.4%. Testing was conducted with Carmody’s buffer pH 2–10 and a voltage-time measurement system. The stability of the prototype was determined by the measurement of the drift and hysteresis effects.
With the rapid spread and multigeneration variation of coronavirus, rapid drug development has become imperative. A major obstacle to addressing this issue is adequately constructing the cell membrane at the molecular level, which enables in vitro observation of the cell response to virus and drug molecules quantitatively, shortening the drug experiment cycle. Herein, we propose a rapid and label-free supported lipid bilayer-based lab-on-a-chip biosensor for the screening of effective inhibition drugs. An extended gate electrode was prepared and functionalized by an angiotensin-converting enzyme II (ACE2) receptor-incorporated supported lipid bilayer (SLB). Such an integrated system can convert the interactions of targets and membrane receptors into real-time charge signals. The platform can simulate the cell membrane microenvironment in vitro and accurately capture the interaction signal between the target and the cell membrane with minimized interference, thus observing the drug action pathway quantitatively and realizing drug screening effectively. Due to these label-free, low-cost, convenient, and integrated advantages, it is a suitable candidate method for the rapid drug screening for the early treatment and prevention of worldwide spread of coronavirus.
A multifunctional ion-sensitive floating gate Fin field-effect transistor (ISFGFinFET) for hydrogen and sodium detection is demonstrated. The ISFGFinFET comprises a FGFET and a sensing film, both of which are used to detect and improve sensitivity. The sensitivity of the ISFGFinFET can be adjusted by modulating the coupling effect of the FG. A nanoseaweed structure is fabricated via glancing angle deposition (GLAD) technology to obtain a large sensing area to enhance the sensitivity for hydrogen ion detection. A sensitivity of 266 mV per pH can be obtained using a surface area of 3.28 mm². In terms of sodium ion detection, a calix[4]arene sensing film to monitor sodium ions, obtaining a Na⁺ sensitivity of 432.7 mV per pNa, is used. In addition, the ISFGFinFET demonstrates the functionality of multiple ions detection simultaneously. The sensor arrays composed of 3 × 3 pixels are demonstrated, each of which comprise of an FGFET sensor and a transistor. Furthermore, 16 × 16 arrays with a decoder and other peripheral circuits are constructed and simulated. The performance of the proposed ISFGFinFET is competitive with that of other state-of-the-art ion sensors.
Full-text available
In this article, the znic oxide (ZnO) nanorods (NRs) and Pt-nanoparticles (NPs) decorated on ZnO (Pt@ZnO) NRs based on extended-gate field-effect-transistor (EG-FET) sensor were prepared and expolred through a simple hydrothermal method (HTM) and a direct current (DC) magnetron sputtering system (0 and 30 s). The results showed that all crystals preferentially grew in the c-axis direction. The ZnO and Pt@ZnO pH sensors are also called ps-0 and ps-30. The Pt sheet as reference electrode was used to test the sensitivity and linearity of buffer solutions with various pH values. It was found that both of them revealed good linearity and sensitivity, and ps-30 sample showed notably enhanced sensing characteristic. As a result, the average current and voltage sensitivities of the ps-0 samples were 15.50 μA/pH and 28.95 mV/pH, and linearity curves were 0.990 and 0.978, whereas that of the ps-30 samples were 47.82 μA/pH and 49.83 mV/pH with a linearity of 0.985 and 0.994, respectively. Furthermore, the ps-30 samples have superior output response voltage, which demonstrated that the devices will be extremely useful in pH-sensing applications.
Full-text available
For highly sensitive pH sensing, an electrolyte insulator semiconductor (EIS) device, based on ZnO nanorod-sensing membrane layers doped with magnesium, was proposed. ZnO nanorod samples prepared via a hydrothermal process with different Mg molar ratios (0–5%) were characterized to explore the impact of magnesium content on the structural and optical characteristics and sensing performance by X-ray diffraction analysis (XRD), atomic force microscopy (AFM), and photoluminescence (PL). The results indicated that the ZnO nanorods doped with 3% Mg had a high hydrogen ion sensitivity (83.77 mV/pH), linearity (96.06%), hysteresis (3 mV), and drift (0.218 mV/h) due to the improved crystalline quality and the surface hydroxyl group role of ZnO. In addition, the detection characteristics varied with the doping concentration and were suitable for developing biomedical detection applications with different detection elements.
In this study, pH sensors with ZnO nanorods (NRs) and ZnO film were fabricated by low-temperature hydrothermal method. The 100nm ZnO film was deposited by radio-frequency magnetron sputtering as a seed layer on a glass substrate. By using a Pt electrode as reference electrode, the potentiometric method was used to measure the potential difference between the two ends, and an operation amplifier (OPA) was employed as the readout circuit. The sensor and reference electrode were placed in a buffer solution of different pH values (4, 6, 7, 8, and 10) for potentiometric analysis. Results showed that the average sensitivity of the ZnO NRs sensor was 44.56 mV/pH, and linearity was 0.983, whereas that of the ZnO film pH sensor was 34.82 mV/pH with a linearity of 0.992. Both structures of the pH sensors exhibited high linearity for pH measurement.
Full-text available
A new type of zinc oxide/silicon nanowire (ZnO/SiNW) hybrid is developed for use in an extended-gate field-effect transistor (EGFET) for pH sensors. SiNWs are first formed using the Ag-assisted electroless etching technique and are then covered with ZnO nanostructures through a combination of sol-gel and hydrothermal processes. The ZnO nanostructures were synthesized at 90 degrees C for 3 h using precursor solutions with molar concentrations of 10 mM and 25 mM. The ZnO nanostructures provide a larger surface area than the pristine SiNWs for adsorbing additional H+ and OH- ions, along with increased oxygen-related binding to effectively sense H+ ions in the acid solution region. The 25 mM ZnO/SiNW sensors exhibited higher sensitivity (66 mV/pH) than pristine SiNW sensors (52 mV/pH). This simple and low-cost sensing device can be applied in disposable biosensors.
Full-text available
Multicolor ZnO quantum dots (QDs) were synthesized and further modified with hydrophobic hexadecyltrimethoxysilane (HDS) and then hydrophilic aminopropyltriethoxysilane (APS) bilayers, resulting in amine-functionalized ZnO@HDS@APS nanocomposites with tunable fluorescence from blue to green yellow. Systematic investigations verify that the resultingZnO@HDS@APS could display extremely high stability in aqueous media and unexpectedly, dramatically-enhanced fluorescence intensities, which are about 10-fold higher than those of bare ZnO QDs. The feasibility of the as-prepared ZnO nanocomposites for blood, cell, and tissue imaging was preliminarily demonstrated, promising the wide bio-applications for cell or tissue imaging, proteome analysis, drug delivery, and molecular labeling.
Full-text available
We report vertically aligned ZnO nanowire arrays (ZnO NWAs) were fabricated on 3D graphene foam (GF) and used to selectively detect uric acid (UA), dopamine (DA), and ascorbic acid (AA) by a differential pulse voltammetry method. The optimised ZnO NWA/GF electrode provided a high surface area and high selectivity with a detection limit of 1 nM for UA and DA. The high selectivity in the oxidation potential was explained by the gap difference between the lowest unoccupied and highest occupied molecular orbitals of a biomolecule for a set of given electrodes. This method was further used to detect UA levels in the serum of patients with Parkinson's disease (PD). The UA level was 25% lower in PD patients than in healthy individuals. This finding strongly implies that UA can be used as a biomarker for PD.
Full-text available
A simple top-down route is developed to fabricate large size porous ZnO flakes via solution combustion synthesis followed by a subsequent calcination in air, which is template-free and can be easily enlarged to an industrial scale. The achieved porous ZnO flakes, which are tens to hundreds of micrometers in flat and tens of nanometers in thickness, exhibit high response for detecting acetone and ethanol, because the unique two-dimensional architecture shortens effectively the gas diffusion distance and provides highly accessible open channels and active surfaces for the target gas.
To obtain wide linear pH sensing range, the tantalum doped zinc oxide (ZnO:Ta) thin film was deposited as the sensing membrane of extended-gate field-effect-transistor (EGFET) pH sensors using the vapor cooling condensation system. Compared with the ZnO EGFET pH sensors, the experimental results exhibited that the linear sensing pH range of the ZnO:Ta EGFET pH sensors was extended from the pH range of 4-12 to the pH range of 1.3-12. Furthermore, the ZnO:Ta pH sensors was stable in the whole extended pH range and showed favorable sensing sensitivity of 41.56 mV/pH. The AFM images of the ZnO:Ta sensing membrane after the measurement in strong acidic solution showed no observable surface damage, which further verified the high corrosion resistance of the ZnO:Ta sensing membrane.
One of the main challenges in the development of new analytical platforms for ultrasensitive bioaffinity detection is also achieving a wide dynamic range in target analyte concentration, especially for approaches that rely on multi-step processes as a part of the signal amplification mechanism. In this paper, a new surface-based sandwich assay is introduced for the direct detection of B-type natriuretic peptide (BNP), an important biomarker for cardiac failure, at concentrations ranging from 1 aM to 500 nM. This was achieved using nanoparticle-enhanced surface plasmon resonance (SPR) where a DNA aptamer is immobilized on a chemically modified gold surface in conjunction with the specific adsorption of antiBNP coated gold nanocubes in the presence of the biomarker target. A concentration detection range greater than eleven orders of magnitude was achieved through dynamic control of only the secondary nanoparticle probe concentration. Furthermore, detection at low attomolar concentrations was also achieved in undiluted human serum.
A simple fabrication of ZnO-nanowire-based device and their implementation as a pH sensor, temperature sensor, and photo detector is reported. The presented multifunctional ZnO multiple-nanowire sensor platform contains a Au finger structure, which is realized by conventional photolithography on a SiO2 substrate. The nanowires are grown using thermal chemical vapor deposition. In order to detect the physical signals, changes in electrical signals were measured (conductance and current). For temperature sensing, the current behavior from 90 to 380 K under vacuum conditions exhibit a tunneling behavior between spaced nanowires. For photo sensing, the current response between the “on” and “off” states of light was measured when exposed to different wavelengths ranging from UV to visible light. Finally, for pH sensing the conductance was measured between a pH of 5 and 8.5. The ZnO nanowires were protected from chemical attacks by a thin layer of C4F8-plasma-based coating.
An extended-gate field-effect transistor (EGFET) of coaxial-structured ZnO/silicon nanowires as pH sensor was demonstrated in this paper. The oriented 1-μm-long silicon nanowires with the diameter of about 50 nm were vertically synthesized by the electroless metal deposition method at room temperature and were sequentially capped with the ZnO films using atomic layer deposition at 50 °C. The transfer characteristics (IDS–VREF) of such ZnO/silicon nanowire EGFET sensor exhibited the sensitivity and linearity of 46.25 mV/pH and 0.9902, respectively for the different pH solutions (pH 1–pH 13). In contrast to the ZnO thin-film ones, the ZnO/silicon nanowire EGFET sensor achieved much better sensitivity and superior linearity. It was attributed to a high surface-to-volume ratio of the nanowire structures, reflecting a larger effective sensing area. The output voltage and time characteristics were also measured to indicate good reliability and durability for the ZnO/silicon nanowires sensor. Furthermore, the hysteresis was 9.74 mV after the solution was changed as pH 7 → pH 3 → pH 7 → pH 11 → pH 7.
A novel ZnO nanorods/ferrocenyl-alkanethiol (FcC11SH) bilayer structure was prepared and applied for the fabrication of glucose enzymatic biosensor. ZnO nanorod matrix was synthesized by low temperature aqueous method and provided a favorable environment for the immobilization of glucose oxidase (GOx). A monolayer of FcC11SH molecular was self-assembled on the surface of gold electrode and introduced a shuttling way for electronic communication between GOx and electrode. The morphology and structure of prepared ZnO nanorods were characterized by employing scanning electron microscopy (SEM), and X-ray powder diffraction (XRD). Electrochemcial measurements of the sensor revealed a high and reproducible sensitivity of 27.8μAcm(-2)mM(-1), and a linear range from 0.05 to 1.0mM with a detection limit of 20μM. A relatively low value of Michaelis-Menten constant about 2.95mM indicates the enhanced affinity of GOx to glucose. To the best of our knowledge, this is the first time to fabricate the glucose biosensor by using ZnO and FcC11SH bilayer structure.
A light incident angle selectivity of a memory device is demonstrated. As a model system, the ZnO resistive switching device has been selected. Electrical signal is reversibly switched between memristor and resistor behaviors by modulating the light incident angle on the device. Moreover, a liquid passivation layer is introduced to achieve stable and reversible exchange between the memristor and WORM behaviors.