Enhanced on-Chip modification and intracellular hydrogen peroxide detection via gigahertz acoustic streaming microfluidic platform

Abstract

Developing effective strategies for the flexible control of fluid is vital for microfluidic electrochemical biosensing. In this study, a gigahertz (GHz) acoustic streaming (AS) based sonoelectrochemical system was developed to realize an on-chip surface modification and sensitive hydrogen peroxide (H2O2) detection from living cells. The flexible and controlled fluid surrounding the electrochemical chip was optimized theoretically and applied in the sonoelectrochemical deposition of Au nanoparticles (AuNPs) first. Under the steady and fast flow stimulus of AS, AuNPs could be synthesized with a smaller and evener size distribution than the normal condition, allowing AuNPs to show an excellent peroxidase-like activity. Moreover, the AS also accelerated the mass transport of target molecules and improved the catalytic rate, leading to the enhancement of H2O2 detection, with an extremely low detection limit of 32 nM and a high sensitivity of 4.34 μA/ (mM·mm²). Finally, this system was successfully applied in tracking H2O2 release from different cell lines to distinguish the cancer cells from normal cells. This study innovatively integrated the surface modification and molecules detection process on a chip, and also proposed a simple but sensitive platform for microfluidic biosensing application.

Full text

Ultrasonics Sonochemistry 100 (2023) 106618
Available online 23 September 2023
1350-4177/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Enhanced on-Chip modication and intracellular hydrogen peroxide
detection via gigahertz acoustic streaming microuidic platform
Feng Zhu
a
,
1
, Zeyu Liu
a
,
1
, Xiaoyu Wu
a
, Die Xu
a
, Quanning Li
a
, Xuejiao Chen
a
, Wei Pang
a
,
Xuexin Duan
a
, Yanyan Wang
a
,
*
a
State Key Laboratory of Precision Measuring Technology and Instruments, College of Precision Instruments and Opto-electronics Engineering, Tianjin University, Tianjin
300072, China
ARTICLE INFO
Keywords:
Sonoelectrochemistry
Electrodeposition
Microuidics
Hydrogen peroxide
ABSTRACT
Developing effective strategies for the exible control of uid is vital for microuidic electrochemical biosensing.
In this study, a gigahertz (GHz) acoustic streaming (AS) based sonoelectrochemical system was developed to
realize an on-chip surface modication and sensitive hydrogen peroxide (H
2
O
2
) detection from living cells. The
exible and controlled uid surrounding the electrochemical chip was optimized theoretically and applied in the
sonoelectrochemical deposition of Au nanoparticles (AuNPs) rst. Under the steady and fast ow stimulus of AS,
AuNPs could be synthesized with a smaller and evener size distribution than the normal condition, allowing
AuNPs to show an excellent peroxidase-like activity. Moreover, the AS also accelerated the mass transport of
target molecules and improved the catalytic rate, leading to the enhancement of H
2
O
2
detection, with an
extremely low detection limit of 32 nM and a high sensitivity of 4.34
μ
A/ (mM⋅mm
2
). Finally, this system was
successfully applied in tracking H
2
O
2
release from different cell lines to distinguish the cancer cells from normal
cells. This study innovatively integrated the surface modication and molecules detection process on a chip, and
also proposed a simple but sensitive platform for microuidic biosensing application.
1. Introduction
Hydrogen peroxide (H
2
O
2
) plays a crucial role in regulating cellular
metabolism, proliferation, differentiation and apoptosis [1]. Excessive
production and accumulation of intracellular H
2
O
2
can also lead to
oxidative stress, destroy cell components, and then result in a variety of
diseases such as neurodegenerative [2,3], cardiovascular disease [4],
leukemia [5] and cancer [6]. Therefore, the level of H
2
O
2
can be
regarded as a biomarker for early diagnosis of various diseases. How-
ever, the H
2
O
2
in cells is extremely unstable and difcult to be measured
dynamically, developing methods to detect intracellular H
2
O
2
accu-
rately in high demand.
To date, a variety of analytical techniques, including chem-
iluminescence [7], uorescence [8], colorimetry [9] and electrochem-
istry [10] have been established for the H
2
O
2
measurement. Among
these, the electrochemical method exhibits the advantages of fast
detection, high sensitivity and easy operation, providing a powerful way
for intracellular H
2
O
2
detection [11]. Conventionally, natural enzymes,
including horseradish peroxidase (HRP) and microperoxidase-8 (MP8),
have always been used to modify the electrochemical working electrode
for selectively detecting H
2
O
2
. These enzymes can catalyze H
2
O
2
selectively but suffer drawbacks of high cost and poor stability.
Recently, nanomaterials based enzymes (nanozymes) have emerged as a
substitute [12,13], which mainly include types of noble metal nano-
materials, transition metal nanomaterials, carbon-based nanomaterials
and others [14]. These nanomaterials show distinct enzyme-like cata-
lytic properties and sensing performance with low cost and high sta-
bility. Among all the nanozymes reported, the Au nanoparticles (AuNPs)
are the most attractive sensing element for H
2
O
2
determination, attrib-
uting to their advantages in high catalytic activity, fast electron transfer
and stable property [15]. However, the common preparation of AuNPs-
based substrates is usually chemical bonding or physisorption, which
has difculties in modifying AuNPs controllably and evenly on the
sensing surface, hindering their practical applications. Electrodeposition
synthesis of nanomaterials is a potentially superior method for its high
controllability and one-step process [16,17], but most synthesized
* Corresponding author.
E-mail address: yanyanwang@tju.edu.cn (Y. Wang).
1
These authors contributed equally to this paper
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultson
https://doi.org/10.1016/j.ultsonch.2023.106618
Received 8 June 2023; Received in revised form 20 September 2023; Accepted 22 September 2023

Ultrasonics Sonochemistry 100 (2023) 106618
2
nanoparticles are in large and irregular size, which is not appropriate for
the electrochemical analysis of living cells. Recently, the sonoelec-
trochemical deposition of nanoparticles has shown advantages over
conventional methods such as more uniform size, higher surface activity
and better catalytic performance [17,18].
Microuidics is an emerging technology capable of operating and
analyzing micro-liters sample. Combining electrochemical methods
with microuidics, specic biomolecule detection with higher detecting
sensitivity and shorter response time can be achieved. A variety of
microuidic based sensing systems have been reported based on the
implementation of electrodes in a microuidic platform for H
2
O
2
detection [19–21]. However, the existing microuidic devices have been
largely limited by a single function, lack of exible control and intensive
mixing, making the electrochemical detection suffer from limited the
mass transport of target molecules and consequently low detection ef-
ciency or long incubation time [22]. Recently, the coupling of acoustics
with microuidics has been demonstrated to be a promising tool for
precisely manipulating micro/nano-scale uids [23–25]. By integrating
the acoustic device with the micro-channel, the sound waves generated
by devices suffer energy dissipation when crossing the solid–liquid
interface, which drives local and highly controllable streaming, named
acoustic streaming (AS). This kind of AS is superior in solution mixing
and pumping, which has been successfully implemented in microuidic
and applied for sample separation, concentration and manipulation
[26–28]. However, besides these biological and chemical analysis ap-
plications, few studies have been devoted to on chip surface modica-
tion and analysis applications [29,30].
In this study, a microuidic sonoelectrochemical platform was pro-
posed by integrating the electrochemistry and gigahertz (GHz) resonator
on a chip. The AS generated by the resonator can be exibly and accu-
rately controlled by simply adjusting the power and distance of the
resonator, which has been proved by our present work [30–33].
Beneting from the AS, sonoelectrochemical deposition of AuNPs with
smaller size and even distributing on the working electrode was ach-
ieved, which showed excellent enzyme -mimic activity. The fabricated
system was further applied to sense the H
2
O
2
, and the sensing perfor-
mance of H
2
O
2
in micro-channel was signicantly improved under the
GHz acoustic streaming. Finally, as a proof of concept, the optimized
platform was applied for detecting hydrogen peroxide in different cell
lines. We believe that the strategy proposed in this work can be extended
to the sensitive detection of other biological or chemical samples, and
the combination of GHz acoustic resonator and electrochemistry will be
a potential mean to boost the analytical eld.
2. Experiment section
2.1. Reagents and materials
Sulfuric acid (H
2
SO
4
, >98 %), hydrogen peroxide (H
2
O
2
, AR),
acetone (C
3
H
6
O, >98 %), ethanol (CH
3
CH
2
OH, >99 %), dopamine (DA,
>98 %), uric acid (UA, >98 %), glucose (Glu, >98 %) and L-ascorbic
acid (AA, >98 %) were purchased from Sigma-Aldrich. Gold (III) chlo-
ride trihydrate (HAuCl
4
, >99 %), catalase (CAT, >2x10
5
unit/g),
phorbol 12-myristate 13-acetate (PMA, 99.50 %) were obtained from
Aladdin. Phosphate-buffered saline (PBS, pH 7.40) was purchased from
Solarbio Life Sciences. The polydimethylsiloxane (PDMS) precursor and
curing agent was obtained from Dow Corning. All chemicals were used
without further purication. All solutions were prepared using ultrapure
water (18.20 MΩ) unless otherwise specied.
2.2. Device fabrication
The 1.50 GHz solidly mounted resonator (SMR) was fabricated using
a standard microelectromechanical process, which has been described in
the previous work [34]. The single SMR device was cut into 18 mm ×25
mm from a 4-inch wafer to facilitate the arrangement of a ow channel
above. The device is then mounted and wire bonded on an evaluation
board (5 cm ×5 cm) for further use.
The electrochemical chip (EC) was fabricated by UV-
photolithography and metal-evaporation. Briey, the SU-8 photoresist
was spin coated onto the SiO
2
wafer and baked according to manufac-
turer instructions. The SiO
2
wafer was prepatterned via UV exposure
using a chromium mask with the designed pattern of electrodes. Then, a
layer of chromium (thickness: 50 nm) and gold (thickness: 150 nm) were
deposited orderly on the wafer by evaporation. After the removal of the
resist, the wafer with three patterned gold electrodes was prepared, two
of which were used as working electrode (WE) and counter electrode
(CE). To maintain the stability of reference electrode (RE), Ag/AgCl ink
(ALS Co., Ltd) was coated on one of the three electrodes and left to dry at
120 ◦C for 5 min.
A PDMS lm with the thickness of 1 mm was fabricated to make the
micro-channel, and the micro-channel was then obtained by slicing a
groove of 12 mm in length and 3 mm in width.
2.3. System setup
The system was constructed in the process shown in Fig. 1. The
evaluation board with SMR, micro-channel and EC was placed from
bottom to top orderly, making sure that their working area was in micro-
channel and working electrode was facing the center of the resonator.
Then, a PDMS gland obtained by molding the EC was placed above them
to x the EC and seal the micro-channel. Two silicone capillary tubes
were inserted into the gland for liquid injecting and outow. Finally, the
resonator was connected with a sinusoidal signal source (Keysight,
N5171B) via a power amplier (Mini-Circuits, ZHL-5W-422+) to pre-
amplify the signal. The electrochemical chip was connected to the
electrochemical workstation (CHI660E) via the collet at the end.
2D nite element simulation (FEM) with the uid–structure inter-
action model was used to simulate the acoustic streaming eld gener-
ated from the resonator. The area of streaming eld can be characterized
by distance in plane. The liquid velocity eld was described by the
incompressible Navier-Stocks equation. Different dimensions of the
model and input power of the resonator were simulated respectively to
explore their inuence on acoustic streaming eld.
2.4. Electrode modication
One-step sonoelectrochemical deposition of AuNPs on the working
electrode was performed by cyclic voltammetry (CV) scanning at a po-
tential range of −0.90 to 0.90 V (vs Ag/AgCl) for 20 segments in the
electrolyte, which contained 0.50 mM H
2
SO
4
and 0.10 mM HAuCl
4
, and
were purged with pure nitrogen for 20 min to remove the oxygen. In the
experimental group, the SMR was excited to vibrate in the micro-
channel, and AuNPs were electrodeposited on the EC under the effect
of AS produced by the vibration of SMR. In the control group, the
electrodeposition was executed in a normal situation without AS. Then,
the AuNPs modied electrode was rinsed with ultrapure water to
remove physically adsorbed substances. Scanning electron microscopy
(SEM, FEI F50) images were used to observe the synthesis of nano-
particles. Image analysis was carried out by using image processing tool
(ImageJ) for nanoparticles counting and average diameter measure-
ment. The electrochemical performance of AuNPs modied EC was
characterized by CV and electrochemical impedance spectroscopy (EIS)
in the solution containing 10 mM K
3
[Fe(CN)
6
] as mediators. The CV was
conducted from −0.30 to 0.80 V with the scan rate of 0.10 V/s. The
frequency range of EIS was from 0.01 Hz to 100 k Hz, the initial voltage
was 0.30 V, and the amplitude was 5 mV.
2.5. H
2
O
2
detection
CV measurements for H
2
O
2
were conducted from −0.60 to 0.20 V at
different scan rates (0.05 – 0.50 V/s). Amperometric measurements of
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
3
H
2
O
2
were performed at the reduction peak potential of −0.40 V. The
injection of various kinds of solutions and H
2
O
2
solution with different
concentrations was done via manual injector.
2.6. Cell experiment
Hela and MCF-7 were cultured in DMEM (Gbico) with 10 % fetal
bovine serum (FBS, Invitrogen). MCF-10A cells were cultured in RPMI-
1640 (Gbico) with 10 % fetal bovine serum. NE-4C stem cells were
cultured with MEM (Gbico) containing 10 % FBS and 1 % penicillin −
streptomycin solution. The cell lines were incubated in a humidied
incubator with 5 % CO
2
at 37 ◦C. After growing to 80 % conuence, the
cells were collected by centrifugation and washed three times with PBS.
For the adherent cell detection, the collected cells (1x10
5
/ml) were
directly cultured on the surface of the prepared working electrode. After
keeping in humidied incubator with a condition of 5 % CO
2
at 37 ◦C for
48 h, the electrode with adhered cells was mounted on the microuidic
platform. During the detection process, PBS was rst injected into the
micro-channel as electrolyte. After the baseline was steady, 100 ng/mL
phorbol-12-myristate-13-acetate (PMA) was injected to stimulate the
release of H
2
O
2
from cells. For the cell suspension detection, the cells
were resuspended in PBS with 1x10
6
/ml and injected into the micro-
channel rstly, then PMA was injected into the micro-channel to stim-
ulate the cells after the baseline was stable.
3. Results and discussion
3.1. Working principle of the acoustic streaming microuidic system
The microuidic system set up is demonstrated in Fig. 1. GHz SMR
resonator is placed on the opposite side of the electrochemical chip with
a micro-channel between them. H
2
O
2
and other solutions are injected
into the micro-channel through the inlet tube, ensuring that both the EC
and SMR are fully exposed to the liquid environment. The SMR converts
electric energy into periodic mechanical vibrations due to the inverse
piezoelectric effect, when excited by a gigahertz alternating voltage
signal. Under the action of resonator, the steady and fast owing vortex
of AS is generated in the micro-channel. The range and ow rate of AS
are determined by the input power of the resonator and the height of the
micro-channel. FEM was used to analyze the ow of liquid on the
electrochemical chip surface under the act of AS. According to the
simulation results (Fig. S1), the range of liquid vortex is limited by the
height of the channel, resulting in an inuence area with a diameter
approximately twice the height of the channel. With the channel height
increasing, the vortex will inuence a larger area of the opposite elec-
trochemical chip while the ow rate will decrease at the same time. To
ensure the entire working electrode area was fully covered by liquid
vortex in a relatively high ow rate, the channel height was set as 1 mm
in the following work. The inuence of resonator input power was also
explored, and found that both ow rate and vortex area were positively
correlated with the input power (Fig. S2). Thus, in the following work,
the power supply was set at 1000 mW, this is the largest input power
that the resonator can work properly for a long time.
3.2. Improved electrodeposition of AuNPs with AS
AuNP is one of the attractive nanozyme that can catalyze the H
2
O
2
and the size and uniformity of AuNPs directly determine the catalytic
efciency, thus on chip electrodeposition of AuNPs under the effect of
AS was studied rst. The H
2
SO
4
and HAuCl
4
mixture was injected into
the micro-channel through the inlet tube, ensuring that both the EC and
SMR were fully exposed to the liquid environment. When excited by a
gigahertz alternating voltage signal, vortexes were generated in the
micro-channel (Fig. 2a). The ow rate on the electrochemical chip
surface was about 1 m/s according to the simulation (Fig. 2b). The Au
3+
surrounding the EC would be reduced to Au and adsorbed on the elec-
trode due to the low overpotential eld on the working electrode sur-
face, which called nucleation of AuNPs, when applying CVs with
potential range of −0.90 to 0.90 V on the EC. Under the action of low
overpotential eld, the nucleated AuNPs on the surface of EC attracted
Fig. 1. (a) schematic illustration and (b) fabrication process of microuidic platform integrated with the acoustic resonator and electrochemical chip.
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
4
each other and showed a growth trend. After 20 segments of CVs, AuNPs
were deposited on the working electrode surface. SEM images were used
to characterize the morphology of AuNPs electrodeposited without and
with the inuence of acoustic streaming. The distribution and average
diameter of synthesized AuNPs were analyzed and shown in Fig. 2c. It
was found that the particle size of AuNPs synthesized with AS was much
smaller at about 52.36 nm, and the distribution was relatively more
uniform. While without AS, the AuNPs showed a broad size distribution
and their size was around 137.69 nm, which was more than two and a
half times bigger than that of AuNP synthesized with AS. Meanwhile, the
number of AuNPs synthesized with AS was also more than that without
AS, suggesting that the AS effectively improved the synthesis of parti-
cles, making them smaller, more uniform, and higher density.
During the electrodeposition of AuNPs, the typical process included
an initial rapid nucleation stage (kinetical control) and a gradual growth
(diffusion control) stage [35,36]. At the early stage of the electrodepo-
sition, AuCl
4
−
was reduced to Au and then distributed as Au nuclei at the
electrode surface [37]. By applying AS in the micro-channel, the ow of
liquid in the channel driven by the AS accelerated the molecules in so-
lution contact with the working electrode, giving rise to more nucleation
of Au on the working electrode surface, which increased the surface
nucleation density. Meanwhile, the growth of AuNPs would lead to a
decreased concentration of AuCl
4
−
in the diffusion layer [38], this
phenomenon was even worse in static micro-uidic circumstances, thus
hindering the homogeneous growth of nanoparticles. The above SEM
results demonstrated that this problem could be solved effectively by
using AS to accelerate the diffusion of AuCl
4
−
in the micro-channel.
Therefore, the synthesized AuNPs showed smaller and more uniform
sizes, with the advantages of increased nucleation density and diffusion
rate by AS. It could be expected that this procedure could also be
exploited for the sono-electrodeposition of other noble metal nano-
particles or materials.
In general, nanoparticles with smaller sizes and more uniform dis-
tribution exhibit better conductivity and catalytic properties [38,39].
CVs and EIS were implemented in the solution of 10 mM Fe(CN)
6
3-/4-
containing 100 mM KCl to characterize the conductivity of EC, EC
modied with AuNPs (EC-AuNPs) and EC modied with AuNPs under
AS (EC-AuNPs-AS) in Fig. 3a, b. It is found that all kinds of ECs show a
pair of redox peaks, but the peak current of the EC-AuNPs-AS is higher
than the EC-AuNPs and EC (Fig. 3a). The results from the EIS are dis-
played in Fig. 3b within an equivalent circuit model, of which the
semicircle diameter corresponds to the charge transfer resistance (R
CT
)
of the electrode in high frequency zones. It is obvious that EC-AuNPs-AS
has the smallest charge transfer resistance, which is in agreement with
CVs, demonstrating that the AS assisted synthesized AuNPs reduced
electrode impedance and improved the electronic conductivity of
Fig. 2. Sono-electrodeposition of AuNPs with AS. (a) Schematic illustration of AS acting with electrochemical chip, and photos of SMR resonator, the schematic of
electrochemical chip. (b) 2D nite element simulation at optimized working condition with the height of 1000
μ
m and input power of 1000 mW. (c) SEM images of
AuNPs modied EC without and with AS, and statistic results of AuNPs size distribution. (d) Schematic diagram of nanoparticle synthesis process with and without
acoustic streaming.
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
5
electrode.
To further discuss the catalytic activity of EC, EC-AuNPs, and EC-
AuNPs-AS towards H
2
O
2
detection, the CV experiments were per-
formed on various electrodes in the presence of 1 mM H
2
O
2
(Fig. 3c). It
can be seen that in the absence of AuNPs
,
there is no oxidation/reduction
peak in the electrode. However, a reduction peak around −0.36 V ap-
pears in the presence of AuNPs, and a higher peak is observed at −0.40 V
for EC-AuNPs-AS (Orange color, Fig. 3c). It indicates that enhanced
electrocatalytic activity of AuNPs toward the reduction of H
2
O
2
, and
AuNPs synthesized under AS action show higher catalytic activity.
Fig. 3d shows the i-t amperometric curve of different modied electro-
chemical chips at −0.40 V upon the addition of 1 mM H
2
O
2
. This result is
also consistent with the CVs that EC-AuNPs-AS exhibits a higher current
signal to H
2
O
2
than the other two kinds of ECs. It is further conrmed
that AS facilitates the synthesis of AuNPs on the electrode surface and
enables the interface to be an excellent sensing platform for H
2
O
2
detection. Considering the optimal performance of EC-AuNPs-AS, this
kind of electrode was utilized in all the following experiments.
3.3. Improved sensing performance by acoustic streaming
The mechanism for the reduction of H
2
O
2
on nanozyme modied
electrode is the charge transfer reaction occurring at the interface be-
tween the catalyst and the electrolyte solution [40]. When the H
2
O
2
molecules diffuse from the solution to interface, nanozyme mediates the
reduction and electron transfer of H
2
O
2
. We hypothesized that the
intrinsic properties of accelerating the solution diffusion with AS could
also be utilized for improving the diffusion of H
2
O
2
. The electrochemical
catalytic kinetics of H
2
O
2
under AS was further explored. The CV curves
of EC-AuNPs-AS in 1 mM H
2
O
2
at varying scan rates from 0.05 V/s to
0.50 V/s with AS and without AS are shown in Fig. 4c, d, they can be
seen that the value of reduction peak current (I
pc
) is continuously
increasing with the scan rate.
The linear relationship of I
pc
with the scan rate is plotted in Fig. 4e, f,
and the calibration equation is I(
μ
A) =-21.02v(V/s)-2.82 (R
2
=0.99)
and I(
μ
A) =-5.64(V/s)-3.87 (R
2
=0.99), respectively, illustrating that
the H
2
O
2
reduction on the electrode is in a typical adsorption electrode
process and AS promotes this process [41]. Meanwhile, the linear rela-
tionship of oxidation peak potential (E
pa
) and reduction peak potential
(E
pc
) with the log of scan rate is plotted in Fig. S3, the electron transfer
rate constant (k
s
) can be estimated from the Laviron equation:
logks=
α
log(1−
α
) + (1−
α
)log
α
−log(RT/nFv) −
α
(1−
α
)nFΔEp/2.30RT
(1)
where R is the gas constant, T is the temperature, F is the faraday con-
stant,
α
is the transfer coefcient, v is scan rate, n is the number of
electron transfers, and ΔE
p
is the peak to peak separation.
It is found that k
s
of the H
2
O
2
sensing with AS (EC-H
2
O
2
-AS, 45.78
s
−1
) is about 1.63 times as much as it without AS (EC-H
2
O
2
, 27.94 s
−1
),
indicating the kinetic parameter for the reduction of H
2
O
2
is enhanced
by AS [42]. All the above results highlight the superiority of AS, which is
favoring the fast electrode reaction and electron transfer in the elec-
trochemistry detection system.
After nding that the AS enhanced kinetic parameter for the reduc-
tion of H
2
O
2
, the H
2
O
2
sensing performance of EC-AuNPs-AS was further
investigated in detail. The concentration of H
2
O
2
was 0.10 mM and 0.20
mM H
2
O
2
in the micro-channel successively, and the current response of
EC-H
2
O
2
-AS and EC-H
2
O
2
were recorded (Fig. 5a). It can be seen that the
current response increases with the increase of H
2
O
2
concentration, a
higher and faster current response is obtained with the effect of AS
(Fig. S4). The response time of H
2
O
2
with AS was shorter (<4.35 s) than
that without AS (<20.83 s). This further validated that AS could
signicantly improve the sensing performance of EC to H
2
O
2
.
Fig. 5b depicts the typical current response of EC-AuNPs-AS to H
2
O
2
at a potential of −0.40 V. During the detection process, the concentra-
tion of H
2
O
2
was from low to high (from 100 nM to 1 mM) in the micro-
channel continuously. The data clearly shows that with the gradual in-
crease of the concentration of hydrogen peroxide standard solution, the
response of current is more stable under the effect of AS. The linear
Fig. 3. Electrochemical performance of the synthesis of AuNPs. (a) CVs and (b) EIS on EC (blue line), EC-AuNPs (green line) and EC-AuNPs-AS (orange line) in the
solution of 10 mM Fe (CN)
6
3-/4-
and 100 mM KCl. (c) CVs and (d) amperometric responses of EC, EC-AuNPs and EC-AuNPs-AS in 1 mM H
2
O
2
.
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
6
relationships between current responses and H
2
O
2
concentrations from
5 to 700
μ
M are shown in Fig. 5c, the calibration equation of EC-H
2
O
2
-
AS and EC-H
2
O
2
is I(
μ
A) =-8.21c(mM)-0.55 (R
2
=0.96) and I(
μ
A) =
-4.70(mM)-0.79 (R
2
=0.95), respectively. Besides, the detection sensi-
tivity of EC-H
2
O
2
-AS and EC-H
2
O
2
is calculated as 4.34 and 2.50
μ
A/
(mM⋅mm
2
), respectively, and the detection limit of EC- H
2
O
2
-AS and EC-
H
2
O
2
is 32 nM and 62 nM (LOD =3S/b). From these above results, it can
be concluded that the integrated AS in the H
2
O
2
detection possesses
exhibited the faster response time, higher detection sensitivity and lower
detection limit, conrming AS can improve the sensing performance of
EC both in detection sensitivity and detection limit, by actively trans-
porting analyte to the electrode surface in a steady and fast owing
vortex.
Then, the selectivity of EC-AuNPs-AS toward H
2
O
2
was extensively
examined. Fig. 5d presents the amperometric curve of the electro-
chemical chip upon the addition of H
2
O
2
and other common coexisting
compounds like DA, Glu, UA, and, AA successively. There is a sharp rise
in current response after the addition of H
2
O
2
, but no obvious current
changes are observed followed by the addition of identical concentra-
tion of interferents, implying a superior selectivity of this sensor to H
2
O
2
over other biological compounds. H
2
O
2
can be reduced at an optimized
negative potential (−0.40 V), at which other interfering species could
not react at this potential, which can be attributed to the catalyze
activity of AuNPs-AS.
3.4. Monitoring of H
2
O
2
release from cells
Monitoring H
2
O
2
release from cells is of great importance owing to
the diverse biological functions of H
2
O
2
. Taking advantage of the small
volume of solution in micro-channel and high sensitivity of EC-AuNPs-
AS, the above optimized system was further utilized to detect the
intracellular H
2
O
2
release from cells. PMA, a free radical initiator
inducing oxidative stress in cells, was always utilized as a stimulant for
H
2
O
2
generation from living cells [43,44]. To monitor the H
2
O
2
release
from cells, two kinds of detection methods were proposed, one of which
was culturing cells adherent on electrochemical chip (Fig. 6a) and the
other was injecting suspension cell solution into micro-channel directly
(Fig. 6b). Acoustic streaming was also used to facilitate the monitoring
of intracellular H
2
O
2
releases. Firstly, two control experiments were
carried out with no cells added into the detection system, or PMA and
hydrogen peroxide catalase were injected together to stimulate MCF-7,
since catalase is known as a selective scavenger of H
2
O
2
. There were no
any obvious current changes in these two control groups, indicating that
current response was caused by H
2
O
2
, and H
2
O
2
was produced by the
stimulation of cells (Fig. 6c, d). To deeply gure out the AS effect on
cellular detection, experiments of intracellular H
2
O
2
release monitoring
Fig. 4. Inuence of acoustic streaming on electrochemical catalytic kinetics. Schematic illustration of H
2
O
2
sensing (a) with AS and (b) without AS. CVs of EC-AuNPs-
AS in 1 mM H
2
O
2
at different scan rates (0.05 – 0.50 V/s) (c)with AS and (d) without AS. Plots of cathodic peak current of H
2
O
2
versus corresponding scan rate (e)
with AS and (f) without AS.
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
7
are performed with or without AS (Fig. S5). As the data shows, after
simulation with PMA, a higher current response can be seen under
acoustic streaming, indicating that the sensor is more sensitive with the
assistance of AS.
As the abnormal level of H
2
O
2
has been reported to link with cancer,
Hela and MCF-7 cell lines were selected as tumor cell models,
conversely, MCF-10 and NE-4C were used as normal cell lines in this
study. Fig. 6c, d depict the i-t amperometric curves of different cell lines
upon the addition of PMA. The current response from all these four cell
lines is shown in Fig. 6e, f, and the corresponding H
2
O
2
concentrations
are calculated and given in Fig. S6. The two methodologies have illus-
trated complementary experimental results, indicating that the current
response of tumor cell lines (MCF-7/ Hela) is much higher than normal
cell lines (MCF-10/ NE-4C), and the highest current response was
Fig. 5. Amperometric records of EC-AuNPs-AS to H
2
O
2
with AS (orange line) and without AS (blue line) (a) with the 0.10 mM and 0.20 mM H
2
O
2
in sequence. (b)
With the successive increase of H
2
O
2
concentration (0.10–1000
μ
M) at −0.40 V, the inset shows a magnication of the black rectangle region of H
2
O
2
(0.10–1
μ
M).
(c) Calibration curve of EC-AuNPs-AS for the increasing concentrations of H
2
O
2
and current respond with AS (orange line) and without AS (blue line). (d)
Amperometric response to 0.50 mM H
2
O
2
, DA, Glu, UA, AA and 1 mM H
2
O
2
, respectively.
Fig. 6. Schematic illustration of cellular H
2
O
2
monitoring with (a) adherent cell method and (b) suspension cell method under AS. The current response of H
2
O
2
with
different cell lines (MCF-7, MCF-10, NE-4C and Hela) by the (c) adherent cell and (d) suspension cell method. Corresponding current responses in (e) adherent cell
method and (f) suspension cell method.
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
8
observed with the MCF-7 cell lines. These results may imply that the
production of H
2
O
2
is possibly connected to the types of cell lines, and
the system of EC-AuNPs-AS is highly sensitive to identify this. The
proposed system is expected to be applied to efciently monitor intra-
cellular reactive oxygen in practice.
4. Conclusion
In conclusion, a microuidic detection system integrated with an
acoustic resonator and an electrochemical chip for detecting intracel-
lular H
2
O
2
was developed in this work. Beneting from the acoustic
streaming generated by the resonator, exible control of the microuid
ow in micro-channel were achieved. We nd that the AS increases the
gold nucleation density and molecule diffusion rate, which provides a
novel sonoelectrochemical deposition method for the synthesis of uni-
form AuNPs. Additionally, the AS improves the diffusion and electron
transfer of analyte, making the detection sensitivity and response time of
microuidic electrochemical biosensors be highly improved. Finally, the
real-time monitoring of H
2
O
2
release from different cell lines has been
realized, owing to the excellent effect of this sonoelectrochemical sys-
tem. We believe the GHz AS based platform present in this work paves a
new way toward the development of microuidic sonoelectrochemical
system for bioanalysis and sensing.
CRediT authorship contribution statement
Feng Zhu: Validation, Formal analysis, Investigation, Writing –
original draft, Writing – review & editing. Zeyu Liu: Validation, Formal
analysis, Investigation, Writing – original draft. Xiaoyu Wu: Writing –
review & editing. Die Xu: Validation. Quanning Li: Visualization.
Xuejiao Chen: Visualization. Wei Pang: Resources. Xuexin Duan:
Resources, Methodology. Yanyan Wang: Conceptualization, Method-
ology, Writing – original draft, Writing – review & editing, Supervision,
Project administration, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
This work was supported by the National Natural Science Foundation
of China [grant numbers 61971302]. The authors acknowledge the
technical support from Wenlan Guo, Chen Sun, Bohua Liu and Chongling
Sun.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ultsonch.2023.106618.
References
[1] S.G. Rhee, H
2
O
2
, a necessary evil for cell signaling, Science 312 (5782) (2006)
1882–1883.
[2] K.J. Barnham, C.L. Masters, A.I. Bush, Neurodegenerative diseases and oxidative
stress, Nat. Rev. Drug Discov. 3 (3) (2004) 205–214.
[3] M. Giorgio, M. Trinei, E. Migliaccio, P.G. Pelicci, Hydrogen peroxide: a metabolic
by-product or a common mediator of ageing signals? Nat. Rev. Mol. Cell Biol. 8 (9)
(2007) 722–728.
[4] B. Steinhorn, A. Sorrentino, S. Badole, Y. Bogdanova, V. Belousov, T. Michel,
Chemogenetic generation of hydrogen peroxide in the heart induces severe cardiac
dysfunction, Nat. Commun. 9 (1) (2018) 4044.
[5] D. Huang, C. Zhang, M. Xiao, X. Li, W. Chen, Y. Jiang, Y. Yuan, Y. Zhang, Y. Zou,
L. Deng, Y. Wang, Y. Sun, W. Dong, Z. Zhang, L. Xie, Z. Yu, C. Chen, L. Liu, J. Wang,
Y. Yang, J. Yang, Y. Zhao, J. Zheng, Redox metabolism maintains the leukemogenic
capacity and drug resistance of AML cells, PNAS 120 (13) (2023), e2210796120.
[6] A. Bachi, I. Dalle-Donne, A. Scaloni, Redox Proteomics: Chemical Principles,
Methodological Approaches and Biological/Biomedical Promises, Chem. Rev. 113
(1) (2013) 596–698.
[7] X. Li, X. Qin, Z. Wang, Y. Wu, K. Wang, X. Xia, S. Liu, In Situ Imaging of
endogenous hydrogen peroxide efux from living cells via bipolar gold
nanoelectrode array and electrochemiluminescence technology, ACS Sensors 7 (8)
(2022) 2446–2453.
[8] S.E. Salamifar, R.Y. Lai, Use of combined scanning electrochemical and
uorescence microscopy for detection of reactive oxygen species in prostate cancer
cells, Anal. Chem. 85 (20) (2013) 9417–9421.
[9] Y. Peng, X. Yu, W. Yin, W. Dong, J. Peng, T. Wang, Colorimetric assay using
mesoporous fe-doped graphitic carbon nitride as a peroxidase mimetic for the
determination of hydrogen peroxide and glucose, Acs Applied Bio Materials 3 (1)
(2020) 59–67.
[10] R. Karthik, J. Vinoth Kumar, S.M. Chen, P. Sundaresan, B. Mutharani, Y. Chi Chen,
V. Muthuraj, Simple sonochemical synthesis of novel grass-like vanadium disulde:
a viable non-enzymatic electrochemical sensor for the detection of hydrogen
peroxide, Ultrason. Sonochem. 48 (2018) 473–481.
[11] X. Zhao, R. Zhu, M. Anikovskiy, Q. Wu, Z. Ding, Proling H2O2 from single COS-7
cells by means of scanning electrochemical microscopy, Biosens. Bioelectron. 227
(2023), 115123.
[12] G.H. Yang, A. Abulizi, J.J. Zhu, Sonochemical fabrication of gold nanoparticles-
boron nitride sheets nanocomposites for enzymeless hydrogen peroxide detection,
Ultrason. Sonochem. 21 (6) (2014) 1958–1963.
[13] J. Wu, X. Wang, Q. Wang, Z. Lou, S. Li, Y. Zhu, L. Qin, H. Wei, Nanomaterials with
enzyme-like characteristics (nanozymes): next-generation articial enzymes (II),
Chem. Soc. Rev. 48 (4) (2019) 1004–1076.
[14] P. Hu, H. Qin, K. Hu, R. Dai, Z. Wang, K. Huang, Constructing a defect-rich
hydroxide nanoenzyme sensor based on dielectric barrier discharge microplasma
etching for sensitive detection of thiamine hydrochloride and hydrogen peroxide,
J. Colloid Interface Sci. 628 (2022) 597–606.
[15] Y. Lin, J. Ren, X. Qu, Nano-Gold as Articial Enzymes: Hidden Talents, Adv. Mater.
26 (25) (2014) 4200–4217.
[16] U.S. Mohanty, Electrodeposition: a versatile and inexpensive tool for the synthesis
of nanoparticles, nanorods, nanowires, and nanoclusters of metals, J. Appl.
Electrochem. 41 (3) (2011) 257–270.
[17] D.S. Karousos, K.I. Desdenakis, P.M. Sakkas, G. Sourkouni, B.G. Pollet, C. Argirusis,
Sonoelectrochemical one-pot synthesis of Pt – Carbon black nanocomposite PEMFC
electrocatalyst, Ultrason. Sonochem. 35 (2017) 591–597.
[18] H.E. Hansen, F. Seland, S. Sunde, O.S. Burheim, B.G. Pollet, Frequency controlled
agglomeration of Pt-nanoparticles in sonochemical synthesis, Ultrason. Sonochem.
85 (2022).
[19] P. Farinazzo Bergamo Dias Martins, I. Plazl, D. Strmcnik, B. Genorio, Prospect of
microuidic devices for on-site electrochemical production of hydrogen peroxide,
Current Opinion Electrochemistry 38 (2023), 101223.
[20] B. Liu, B. Ran, C. Chen, L. Shi, J. Jin, Y. Zhu, High-throughput microuidic
production of bimetallic nanoparticles on MXene nanosheets and application in
hydrogen peroxide detection, ACS Appl. Mater. Interfaces 14 (50) (2022)
56298–56309.
[21] C.D.R. Mata, R.S. Moakhar, I.I. Hosseini, M. Jalali, S. Mahshid, A nanostructured
microuidic device for plasmon-assisted electrochemical detection of hydrogen
peroxide released from cancer cells, Nanoscale 13 (34) (2021) 14316–14329.
[22] F. Paratore, V. Bacheva, M. Bercovici, G.V. Kaigala, Recongurable microuidics,
Nat. Rev. Chem. 6 (1) (2022) 70–80.
[23] C. Chen, B. Ran, B. Liu, X. Liu, Y. Liu, M. Lan, R. Manasseh, Y. Zhu, Development of
a novel microuidic biosensing platform integrating micropillar array electrode
and acoustic microstreaming techniques, Biosens. Bioelectron. 223 (2023),
114703.
[24] Y. Zhang, X. Chen, Particle separation in microuidics using different modal
ultrasonic standing waves, Ultrason. Sonochem. 75 (2021), 105603.
[25] Z. Chen, P. Liu, X. Zhao, L. Huang, Y. Xiao, Y. Zhang, J. Zhang, N. Hao, Sharp-edge
acoustic microuidics: principles, structures, and applications, Appl. Mater. Today
25 (2021), 101239.
[26] W. Wei, Y. Wang, Z. Wang, X. Duan, Microscale acoustic streaming for biomedical
and bioanalytical applications, Trac-Trends in Analytical Chemistry 160 (2023),
116958.
[27] M. Wiklund, R. Green, M. Ohlin, Acoustouidics 14: Applications of acoustic
streaming in microuidic devices, Lab Chip 12 (14) (2012) 2438–2451.
[28] M. Zhang, S. He, W. Pang, W. Wei, F. Zhou, X. Wu, H. Qi, X. Duan, Y. Wang, On
chip manipulation of carbon dots via gigahertz acoustic streaming for enhanced
bioimaging and biosensing, Talanta 245 (2022), 123462.
[29] B.G. Pollet, J.-Y. Hihn, M.-L. Doche, J.P. Lorimer, A. Mandroyan, T.J. Mason,
Transport limited currents close to an ultrasonic horn: equivalent ow velocity
determination, J. Electrochem. Soc. 154 (10) (2007).
[30] S. He, Z. Wang, W. Pang, C. Liu, M. Zhang, Y. Yang, X. Duan, Y. Wang, Ultra-rapid
modulation of neurite outgrowth in a gigahertz acoustic streaming system, Lab
Chip 21 (10) (2021) 1948–1955.
[31] S. He, W. Pang, X. Wu, Y. Yang, W. Li, H. Qi, K. Yang, X. Duan, Y. Wang,
Bidirectional regulation of cell mechanical motion via a gold nanorods-acoustic
streaming system, ACS Nano 16 (5) (2022) 8427–8439.
[32] Y. Yang, L. Zhang, K. Jin, M. He, W. Wei, X. Chen, Q. Yang, Y. Wang, W. Pang,
X. Ren, X. Duan, Self-adaptive virtual microchannel for continuous enrichment and
separation of nanoparticles, Sci. Adv. 8 (30) (2022).
[33] H. Qi, S. Zhang, J. Liang, S. He, Y. Wang, Controllable blood–brain barrier (BBB)
regulation based on gigahertz acoustic streaming, Nanotechnology and Precision
Engineering 5 (4) (2022).
F. Zhu et al.

Ultrasonics Sonochemistry 100 (2023) 106618
9
[34] H. Zhao, X. Guo, Y. Wang, X. Duan, H. Qu, H. Zhang, D. Zhang, W. Pang, Microchip
based electrochemical-piezoelectric integrated multi-mode sensing system for
continuous glucose monitoring, Sens. Actuators B 223 (2016) 83–88.
[35] C. Bosch-Navarro, J.P. Rourke, N.R. Wilson, Controlled electrochemical and
electroless deposition of noble metal nanoparticles on graphene, RSC Adv. 6 (77)
(2016) 73790–73796.
[36] R.D. Vais, N. Sattarahmady, H. Heli, Green electrodeposition of gold
nanostructures by diverse size, shape, and electrochemical activity, Gold Bull. 49
(3) (2016) 95–102.
[37] J. Fang, H. You, B. Ding, X. Song, Large-area and high-density gold nanoparticle
arrays with sub-10nm gaps, Electrochem. Commun. 9 (9) (2007) 2423–2427.
[38] M. Keddam, Electrochimie, principes, methodes et applications : A.J. Bard et L. R.
Faulkner Masson, Paris, 1983, 781 pp., Electrochimica Acta 30 (1985) 835.
[39] T. Hezard, K. Fajerwerg, D. Evrard, V. Colli`
ere, P. Behra, P. Gros, Gold
nanoparticles electrodeposited on glassy carbon using cyclic voltammetry:
application to Hg(II) trace analysis, J. Electroanal. Chem. 664 (2012) 46–52.
[40] V.S. Haritha, S.R. Sarath Kumar, R.B. Rakhi, Non-enzymatic electrocatalytic
detection of hydrogen peroxide using tungsten disulphide nanosheets modied
electrodes, Mater. Sci. Eng. B 285 (2022), 115932.
[41] L. Chen, E.E.L. Tanner, C. Lin, R.G. Compton, Impact electrochemistry reveals that
graphene nanoplatelets catalyse the oxidation of dopamine via adsorption, Chem.
Sci. 9 (1) (2018) 152–159.
[42] E. Laviron, General expression of the linear potential sweep voltammogram in the
case of diffusionless electrochemical systems, J. Electroanal. Chem. Interfacial
Electrochem. 101 (1979) 19–28.
[43] P. Wu, Y. Qian, P. Du, H. Zhang, C. Cai, Facile synthesis of nitrogen-doped
graphene for measuring the releasing process of hydrogen peroxide from living
cells, J. Mater. Chem. 22 (13) (2012).
[44] A.W.G.A.J.M. Cooper, PMA was utilized as a stimulant to induce H2O2 generation
from living cells immediately, Anal. Chem. 70 (1998) 2607–2612.
F. Zhu et al.