Facile preparation of porous Co3O4 nanocubes for directly screen-printing an ultrasensitive glutamate biosensor microchip

Journal Pre-proof Facile preparation of porous Co3 O4 nanocubes for directly screen-printing an ultrasensitive glutamate biosensor microchip Fuhao hu, Tao Liu, Jun Pang, Zhenyu Chu, Wanqin Jin

PII:

S0925-4005(19)31786-1

DOI:

https://doi.org/10.1016/j.snb.2019.127587

Reference:

SNB 127587

To appear in:

Sensors and Actuators: B. Chemical

Received Date:

16 October 2019

Revised Date:

5 December 2019

Accepted Date:

15 December 2019

Please cite this article as: hu F, Liu T, Pang J, Chu Z, Jin W, Facile preparation of porous Co3 O4 nanocubes for directly screen-printing an ultrasensitive glutamate biosensor microchip, Sensors and Actuators: B. Chemical (2019), doi: https://doi.org/10.1016/j.snb.2019.127587

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Facile preparation of porous Co3O4 nanocubes for directly screen-printing an ultrasensitive glutamate biosensor microchip Fuhao hu, Tao Liu, Jun Pang, Zhenyu Chu*, Wanqin Jin.

State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemical Engineering, Nanjing Tech University, Nanjing 210009, P. R. China

ro of

Corresponding Author

-p

*E-mail: [email protected]; Fax: +86 25–8317–2292; Tel: +86 25–8317–2266.

Jo

ur

na

lP

re

Graphical abstract

Highlights



A novel Co3O4 architecture was created with a well-defined porous cubic feature.



The Co3O4 crystals were directly served as the ink to screen-print a biosensor

microchip. 

The chip shows ultrahigh glutamate sensitivity with excellent usage stability for

Jo

ur

na

lP

re

-p

ro of

more than 300-times test in 30 days.

Abstract: Although L-glutamate is an important biological analyte for the clinic diagnosis and food industry, rare capable glutamate biosensor has been developed in the practical detection. The electrochemical biosensor is an advanced technique due to its on-site and real-time concentration recognition, however, the unsatisfactory performance in the glutamate detection blocks its production transfer. In this work, we proposed a novel screen-printed glutamate biosensor chip fabricated in large-scale through

ro of

designing a porous nanocubic Co3O4. This material is prepared via a facile thermal

oxidation strategy by using the cobalt hexacyanoferrate (CoHCF) as a precursor to construct its regular geometric morphology. The obtained Co3O4 crystal is of both

-p

regular nanocubic outline and abundant interconnected pores to generate high electrocatalysis on the glutamate enzymatic oxidation. It was directly served as the

re

screen-printing ink to fabricate a microchip which integrated all three-electrode

lP

components. In detection, with loading a small amount of glutamate oxidase, this biosensor chip can still exhibit a high sensitivity of 20.12 μA mM-1 cm-2, as well as a

na

wide linear range from 10 to 600 μM and a low LOD=10 μM. Besides, under a very low working potential of -0.2V. Moreover, the biosensor chip possesses an excellent

days.

ur

anti-interference ability and usage stability to withstand more than 300 tests in 30

Jo

Keywords: Porous Co3O4 nanocube; glutamate biosensor; screen-printed microchip; ultrahigh sensitivity

1. Introduction L-glutamate is one of the main excitatory neurotransmitters in the central nervous system [1-3], which plays an essential role in various neurological functions, therefore it can be served as an important analyte in medicine and food analysis [4]. As reported, the glutamate concentration in blood is closely related to the brain function [5-7]. It has been demonstrated that chronically high level of extracellular glutamate is excitotoxic and easily causes the neuronal cell death. In addition, fluctuating

ro of

glutamate levels for high frequency is associated with many neurodegenerative diseases, such as Parkinson's and Alzheimer's, amyotrophic lateral sclerosis and

ischemic stroke [6]. Therefore, several analytical methods for the glutamate detection

-p

have been continuously reported, including chromatography [8] and fluorescent

spectrometry [9] etc. However, most of these methods always require some special

re

and high-cost instruments with complex and time-consuming sample pretreatments, causing the difficulty in the on-site and immediate analysis.

lP

In order to simplify the detection process, the electrochemical biosensor has attracted more and more research attentions on the glutamate recognition due to its

na

outstanding sensitivity and accuracy, as well as low cost and portability. Since the first electrochemical biosensor was delivered in 1962 [10], numerous novel electrode

ur

materials have been developed to pursue better electrochemical biosensing

Jo

performance [11-13]. However, rare achievement can be harvested in the technique transfer to the commercial biosensor device due to the unsatisfactory reproducibility and stability of their applied materials and preparation methods. Recent progress of the screen-printing technique [14-16] provides a low-cost and stable platform through a large-scale fabrication of the biosensor micro-chip to integrate the traditional three electrodes into one. Usually, most researches just modify their electrode materials on

a bare screen-printed chip for studying, however, weak interaction between the material and the chip substrate strongly restricted the usage stability. Therefore, some work directly prepared the nanomaterials based ink [11,17] for printing to address above issue. In this case, the electrocatalysis of the printing material is the key to determine the biosensing performance of the screen-printed chip. Among various reported screen-printing materials, the nanostructured metal oxide [18-20] shows the superior detection sensitivity than others due to its high catalysis on

ro of

the enzymatic oxidation reaction. As reported, the Co3O4 nanoparticle [21-24] was advanced in biosensing due to its superior electrocatalytic activity, specific surface

area and good biocompatibility. Its stable co-existence of both Co2+ and Co3+ in the

-p

unit cell can construct a natural electron transport channel, showing the excellent

catalytic capacity to H2O2 produced from the enzymatic oxidation of the glutamate

re

oxidase (GMOx) [25-27]. Hence, the glutamate concentration can be determined by

lP

the current response according to the following reactions: 𝐺𝑀𝑂𝑥

L-Glutamate + O2 →

α-Ketoglutarate + H2O2

na

2Co2++ H2O2 →2Co3++ 2OHCo3+ + e- → Co2+

(1) (2) (3)

ur

Different with other detection targets, due to the high expense of the GMOx, higher performance with a lower enzyme concentration is always desired for the wide and

Jo

practical applications of the glutamate biosensor [28]. In this case, the enhancement of the Co3O4 electrocatalysis is highly expected to strengthen the biosensing signal [12, 26]. Although various Co3O4 nanostructures have been developed, most obtained crystals are of the low roughness and surface area to hardly satisfy the printing requirement [27-28]. Therefore, rare literatures directly adopt the Co3O4 as the screen-printing ink to fabricate the biosensor chip.

In this work, we designed a porous Co3O4 nanocube material to directly screen-print a novel ultrasensitive glutamate biosensor microchip. As shown in Scheme. 1, this crystal architecture was achieved through a facile and fast thermal oxidation method by using the cobalt hexacyanoferrate (CoHCF) as a precursor. In addition to the large specific surface area, the porous structure can benefit the immobilization of more glutamate oxidase as well as the permeation of the target into the inner pores, promoting the electron generation and transfer. Therefore, the high

ro of

electrochemical catalysis and conductivity are reflected. The Co3O4 crystals were further treated as a functional ink to print a microchip involving the working, counter

and reference electrodes into one. Only using the small amount of GMOx, the

-p

as-prepared biosensor chip can perform an ultrahigh sensitivity to the glutamate

detection, as well as the wide detection range and low detection limit under a very low

2. Experimental

na

2.1 Reagents and apparatus

lP

usage stability for 300-times test.

re

potential -0.2 V. Besides, this chip possesses the outstanding selectivity and ultrahigh

All chemicals were of analytical purity and used without any further purification. Potassium ferricyanide (Ⅲ) (K3[Fe(CN)6]·3H2O) and L-Glutamate Oxidase (GMOx)

ur

from Streptomyces sp. (EC1.4.3.11, 5U mg-1) were purchased from Sigma-Aldrich.

Jo

Cobalt chloride hexahydrate (Ⅱ) (CoCl2·6H2O) and albumin from bovine serum (BSA) were acquired from Aladdin. Sodium glutamate acid salt monohydrate was obtained from Alfa-Aesar. Dipotassium hydrogen phosphate (K2HPO4), potassium dihydrogen phosphate (KH2PO4), potassium chloride (KCl), hydrochloric acid (HCl) and glutaraldehyde 25% (v/v) were purchased from Shanghai Lingfeng Chemical Reagent Co. Ltd. (China). Ascorbic acid (AA), uric acid (UA) and hydrogen peroxide (H2O2,

30%, w/v, solution) were obtained from Sinopharm Chemical Reagent Co. Ltd. (China). Conductive carbon ink and silver chloride ink were bought from Henkel-Loctite Co. Ltd. All solutions were prepared with deionized water. The screen-printed electrode (SPE) chips were fabricated via a 245 DEK screen-printing machine (Weymouth, UK). The morphology of CoHCF and Co3O4 powders and the energy-dispersive X-ray (EDX) spectrum of the working electrode were examined by using a field emission scanning electron microscope (FESEM)

ro of

(Hitachi, ModelS-4800II, Japan). Transmission electron microscopy (TEM) experiments were performed by a JEOL JEM-2010 UHR. The X-ray diffraction (XRD) maps of the CoHCF and Co3O4 powders were measured on an X-ray diffractometer

-p

(D/MAX 2500 V/PC) with a Cu-Ka line (0.15419 nm). The X-ray photoelectron

spectroscopic (XPS) measurements were tested by an X-ray photoelectron

re

spectrometer (Thermo, ESCALAB 250, USA). The enzyme solution was immobilized

lP

on the surface of the working electrode with the NORDSON EFD three-dimensional dotting-enzyme machine (JR-V2303MI, Taiwan). The concentration of real fermentation samples was determined by High Performance Liquid Chromatography

na

(Shimadzu, LC20A, Japan). All the electrochemical measurements were carried out in a 0.05 M phosphate buffered saline (PBS, pH 6.5) containing 0.1 M KCl by using

ur

electrochemical workstation (CHI 660E, Shanghai Chenhua Instrument Co. Ltd.

Jo

China). All experiments were operated at room temperature except for some special requirements.

2.2 Synthesis of the CoHCF and Co3O4 powders Firstly, the synthesis of the CoHCF nanocubes was based on the chemical reaction between K3[Fe(CN)6] and CoCl2·6H2O. Their solutions were prepared as follows: solution A, 0.01 M K3[Fe(CN)6]+0.1 M KCl+0.1 M HCl, and solution B, 0.01 M

CoCl2·6H2O+0.1 M KCl+0.1 M HCl. Then, above two 100 mL solutions were loaded into two syringes, respectively. The syringes were fixed on an injection pump with the injection speed of 500 μL/min. At the same time, a beaker with stirring was served as the reaction vessel. After the reaction finished, the CoHCF powder was obtained by the centrifugation with the speed of 7000 rpm, and then dried in the air. In order to obtain the porous Co3O4, the dried CoHCF powder was calcinated at 400 ℃ for 2h in air with the heating and cooling rates of 2 ℃·min−1. The C and N elements in the -CN-

ro of

were oxidized and released from the crystals in the form of gas and the metal atom Co were oxidized to form metal oxides Co3O4. Because of the gas release, the pores can be gradually generated in the cubic bulk from the outer to inner.

-p

2.3 Preparation of the Co3O4 based printing ink

The mixture ink was prepared by mixing the Co3O4 powder and the carbon ink.

re

Due to the viscosity of the mixed ink is significant for the quality of screen printing.

lP

Therefore, the proportion of the Co3O4 powder and the carbon ink was constantly changed to obtain the optimum viscosity of the mixture ink before printing. Then a small quantity of ethanol was added into mixture ink. Furthermore, the Co3O4 powder

na

and carbon ink were uniformly blended by constantly. Until the viscosity of the mixture ink can be printed, the mixture ink stopped stirring.

ur

2.4 Fabrication of the biosensor chips

Jo

A commercial Polyethylene terephthalate (PET) plate was used as the substrate for screen-printing. The as-prepared mixture ink was served to construct the working electrode (WE). Meanwhile, carbon ink and silver chloride ink were used to print the counter electrode (CE) and reference electrode (RE), respectively. After the printing operation, the obtained chips were further dried at 45 ℃ in air before usage. In order to immobilize the GMOx, the enzyme solution contained 0.25% glutaraldehyde (v/v),

1% BSA (v/v) for cross-linking, in which 0.1 U/μL GMOx were included. Then, 5 μL of enzyme solutions were dropped evenly onto the surface of the working electrode of biosensor chip, followed by drying at 4oC. Finally, Co3O4 based biosensor chip was stored at 4 oC when not in use.

3. Results and discussion 3.1 Structure evaluation of the Co3O4 nanocrystals In order to obtain a regular framework of the desired Co3O4 crystals, the CoHCF

ro of

precursor was first synthesized to be of a certain regular shape via a low-speed chemical synthesis method. As shown in Fig. 1A, it can be observed that the CoHCF

particles showed a well-defined nanocubic shape with a smooth surface. According to

-p

our previous work, we have demonstrated that the geometric edge of the PB (Prussian blue) nanocube and its analogues can provide higher catalysis than other sites [19].

re

Hence, the cubic skeleton of these CoHCF crystals will be expected to maintain after

lP

the thermal oxidation process for the Co3O4 preparation. The TGA results in Fig. 1B present that the weight loss occurs in two different temperature ranges. The first slightly weight loss was below 254 ℃ attributed to the evaporation of the physically

na

adsorbed water. The second weight loss stage in the range of 254 to 345 ℃ indicated

ur

the decomposition of the CoHCF framework to release a large amount of gas, including H2O, CO2 and NOx. Rare significant weight loss was observed at the

Jo

temperature above 400 ℃, which confirmed the complete decomposition of CoHCF crystals. During the whole experiment, the thermal oxidation process from precursor CoHCF to Co3O4 was very mind due to a low rate of the temperature rising (2 o

C·min−1). Furthermore, the thermal conductivity of CoHCF is weak as well as any

other PBA materials [29-31], avoiding the structure collapse from the inner of the cubic crystal. During the transition process, the release of CO2 and NOx was slow and

smooth with the temperature rising, which enables the final Co3O4 to maintain a cubic morphology with a uniform porous surface. For more clarity, the XRD patterns of varying the temperature to the CoHCF were compared to monitor this heating process. As shown in Fig. 1C, till 100 ℃, this crystal was able to keep its typical diffraction peaks at 17.2°, 24.4°, 34.8° and 39.0° which represent the crystal faces of (200), (220), (400) and (420), respectively. It is worth mentioning that related peaks shifted a little bit to the right when the temperature rises to 200 ℃. The loss of the water in the unit

ro of

cell resulted in the decrease of crystal plane spacing, which leads to the right shift of the peaks. Actually, the structure of CoHCF has not changed substantially. However, when the temperature was increased to 300 ℃, all above peaks were disappeared to

-p

indicate the change of crystal structure. Fig. 1D shows the magnified XRD pattern of the crystal treated at 400 ℃. It can be observed that there are several obvious peaks

re

located at 19.0o, 31.3o, 36.8o, 44.8o, 59.4o and 65.2o which are corresponding to the

lP

Co3O4 crystal planes of (111), (220), (311), (222), (511) and (440), respectively [25]. XPS results of O1s and Co2p orbits were further applied to study the bond information of the prepared CoHCF powders before and after the thermal oxidation

na

process. As depicted in Fig. 2A, the peaks of the uncalcined CoHCF located at 532.0 eV were obtained, which can be assigned to the adsorbed oxygen from water species

ur

on the surface. However, there is an obvious peak shift after the process. This peak

Jo

can be separated as two independent peaks to indicate two kinds of oxygen species. The peak at a lower binding energy (~529.6 eV) was associated with the surface lattice oxygen, and another peak at ~531 eV may be usually assigned to the surface oxygen species adsorbed on the oxygen vacancy. This confirms that the CoHCF has already changed to the oxide state after the thermal oxidation at 400 oC for 2 h. In Fig. 2B, two main peaks located at 782.2 eV and 797.6 eV (uncalcined) or 779.6 eV and

794.8 eV (calcined) belongs to Co 2p3/2 and Co 2p1/2 spin orbitals, respectively [22,27]. These two orbit peaks can be further divided to four characteristic peaks at ca. 779.4, 780.7, 794.6 eV and 795.7 eV which are attributed to the CoIII3/2, CoII3/2, CoIII1/2 and CoII3/2, respectively [31,32]. Therefore, it can be concluded that the calcined material possesses both Co3+ and Co2+ elements to form the Co3O4 nanomaterial. As above mentioned, we expect to keep the original cubic morphology of the CoHCF after the Co3O4 formation in order to increase the catalytic activity. Therefore,

ro of

the nanostructure evolution of the CoHCF crystals thermal-treated from 200 to 500 ℃ was investigated to present in Fig. 3A to D. During heating to 200 ℃, the CoHCF

crystal can survive from the thermal treatment to still keep its original smooth surface

-p

(Fig. 3A), however, it was changed to present several folds to indicate the initial

collapse of the CoHCF structure at 300 ℃ (Fig. 3B). As shown in Fig. 3C, although

re

the crystal outline can still maintain the cubic shape under 400 ℃, numerous surface

lP

deficiencies were produced to show lots of small pores which obviously enhanced the surface area and roughness. Further increasing the temperature would totally destroy all geometric faces of the crystal to only leave the irregular and small particles after

na

melting (Fig. 3D). The feature of a single Co3O4 nanocube was magnified to focus its deficiencies. TEM image in Fig. 3E showed that despite of the high thickness from

ur

the crystal top to button surface, the electron beam can still realize the transmission to

Jo

indicate the through-hole structure. This may be ascribed to the gas release during the thermal oxidation process of the precursors. Fig. 3F displayed an interlayer spacing of 0.473 nm which is corresponding to the (111) lattice plane of Co3O4 unit cell. Associating above XRD results, 400 ℃ was preferred as the optimum temperature parameter to realize the Co3O4 formation with the regular porous morphology. To evaluate the photostability of porous Co3O4 nanocube, we experimented with

sunshine treatment of the material. Firstly, the freshly synthesized Co3O4 powder was characterized by XRD, and the corresponding results are shown as the red curve in Fig. S1. Then we exposed the material under sunshine for 8 hours per day which continuously lasts to 10 days for next XRD characterization. As shown in Fig. S1 (black curve), rare changes of the peak locations and strengths were produced after 10 consecutive days of illumination, confirming the good photostability of the prepared Co3O4 nanocube.

ro of

3.2 Electrochemical behaviors of the biosensor chip In order to satisfy the suitable viscosity for screen-printing, the prepared Co3O4

powder was mixed with the carbon ink to serve as the screen-printing paste for

-p

fabricating the working electrode of the biosensor chip. Meanwhile, the counter and

the reference electrodes were printed by the carbon ink and the AgCl ink, respectively.

re

It is worth noting that our prepared biosensor chip has good flexibility for storage and

lP

carrying, (Fig. 4A) and the bending behavior can not cause cracks or peeling of the printed electrode. The surface of working electrode was investigated to reveal the Co3O4 distribution in the mixture paste through scanning the location of the Co

na

element (Fig. 4C) and O element (Fig. 4D) in the selected region (Fig. 4B). The results illustrated that the Co3O4 crystals were uniformly distributed on the electrode

ur

surface, which is beneficial to enhance the electrocatalysis behavior and decrease the

Jo

entire electron resistance of the working electrode. To confirm the advances of the porous structure to the specific surface area, two

types of biosensor chips printed by the CoHCF and Co3O4 pastes were prepared to study their electrochemical effective surface area, respectively. Cyclic voltammetry curves (CVs) were carried out to follow the Randles–Sevcik equation. 𝐼𝑝 = (2.69 × 105 )𝑛3 𝐷𝑜 1/2 𝐶𝑜 ∗ 𝐴 𝑣 1/2

Where A is the electrode area, n is the number of electrons transferred in the redox event, v is the scan rate, Do is the diffusion coefficient of the mole cules, Co* is the concentration of the probe molecules, and Ip is the peak current of the redox couple. From this equation, the surface area is proportional to Ip/v1/2. As shown in Fig. 5A and B, both CoHCF or Co3O4 chips can produce a reversible electrochemical process according to the increase of the peak potential difference (△Ep) with the scan rate raise. Besides, the correlation between the peak current and

ro of

the square root of the scan rate can well satisfy the linear fitting (Fig. 5C) which implies that both redox processes of CoHCF or Co3O4 chips are controlled by the linear diffusion of the electroactive species. It is worth mentioning that the 5 mM

-p

[Fe(CN)6]4/3- indicator solution was required to apply into the PBS solution during

such CV scanning. Therefore, both CV results of CoHCF and Co3O4 based chips

re

showed their catalysis to the redox reaction of the Fe elements in [Fe(CN)6]4/3-.

lP

Although the shapes of curves in Fig. 5A and B may be similar, the current intensity and peak potential can be observed to show quite different feature. Therefore, the effective specific surface area of the CoHCF and the Co3O4 were calculated from the

na

Randles’ slope as 22.6 mm2 and 32.9 mm2, respectively. This result demonstrates that the formation of a porous structure on the cubic crystal surfaces can remarkably

ur

increase the electrochemical effective surface to provide more active catalytic sites for

Jo

the enzyme immobilization and reaction. As well known, both CoHCF and Co3O4 materials are semi-conductors, showing

weak conductivities to affect the electron transfer. Normally, Co3O4 possesses a higher resistance than CoHCF due to the larger band gap between the highest and lowest unoccupied molecular orbital. Because the prepared Co3O4 was derived from CoHCF, it is worthy to study the influence of this transformation on the chip conductivity.

Therefore, these two materials-based chips were further tested via the EIS technique. The bare, CoHCF and Co3O4 chips were respectively studied in a 0.1 M KCl solution containing 5 mM K3Fe(CN)6/K4Fe(CN)6 (molar ratio is 1:1) with a frequency ranging from 0.1 Hz to 1000 kHz. Observed in Fig. 5D, the bare chip printed by only carbon ink was of the highest transfer resistance (Rct) of 108.0Ω. After the introduction of CoHCF or Co3O4 to the carbon ink as two pastes, the printed CoHCF and Co3O4 chips exhibited much lower Rct of 79.6 and 73.3 Ω, respectively. This result illustrates that

ro of

the transformation from CoHCF to Co3O4 through the thermal oxidation method has not obviously reduced the conductivity of the fabricated biosensor chip.

In order to achieve the optimal ratio of Co3O4 and conductive carbon ink, we

-p

fabricated a series of biosensor chips based on several different mass ratios (Co3O4 powder to carbon ink), including 1:5, 1:10, 1:15 and 1:20. The electrochemical

re

response to H2O2 was measured via a Chronoamperometry method, which is used for

lP

evaluate the properties of the biosensor chips. As shown in Fig. S2A, all the biosensor chips can exhibit a uniform current step regardless of the ratio, which also verified the electrocatalytic capacity of porous Co3O4 nanocubes. Although the same

na

concentration of H2O2 was added, the chips prepared by different Co3O4 amount produced quite different strengths of the response current. We performed a linear

ur

fitting between the current density obtained in Fig. S2A and the concentration value

Jo

of H2O2. As shown in Fig. S2B, we found that while the mass ratio of Co3O4 to carbon ink was 1:10, the biosensor chip exhibited the highest sensitivity of 41.70 µA·mM−1·cm−2. Therefore, we obtain the conclusion that the optimal mass ratio of the material to the carbon ink is 1:10. Repetitive CV scanning was used for investigating the stabilities of different biosensor chips based on CoHCF and Co3O4. It can be found in Fig. S3A, after 50

times of repetitive experiments, both peak currents of CoHCF chip showed the obvious decreases, indicating the weak stability of this material during the continuous redox reactions. Differently, there were rare changes in the electrochemical redox curves of Co3O4 based chip after the same repetitive scanning. This confirms that the thermal oxidation for the Co3O4 preparation from CoHCF can also improve the electrochemical stability. 3.3 Biosensing performance of the as-prepared Co3O4 chip

ro of

The electrocatalytic activity of the as-prepared SPE-chip determines the biosensing performance on the glutamate detection. This property was evaluated through the

electrochemical reaction on H2O2 which is served as a main product in the GMOx

-p

reaction. As shown in Fig. 6A, it can be found that the oxidation current decreased

and the reduction current increased between -0.4 and 0.1 V after the addition of 1 mM

re

H2O2, indicating the electrochemical reduction function of the Co3O4 chip to H2O2.

lP

The most change of the current was particularly obvious at the potential of -0.2 V, therefore, this potential was selected as an optimum parameter for the further chronoamperometry test. It's worth noting that there weren’t significant characteristic

na

peaks of Co3O4 in the CV diagram. As mentioned above, the optimal mass ratio of Co3O4 to carbon ink was only 1:10, therefore, the Co element redox peaks were not

ur

obvious. We also repeated the CV experiment at a wider range of potentials, and the

Jo

results were shown in the black curve in Fig. S4, which had a similar trend to Fig. 6A. In addition, the extra little peaks around 0.1V and -0.1V were from conductive carbon ink. To verify it, we also performed the CV scanning on the micro-chip based on pure carbon ink under the same condition (0.05 M PBS, pH 6.5, 25 ˚C), which was shown as the red inset in Fig. S4. We found that the peaks around 0.1V and -0.1V were more distinct. As shown in Fig. 6B, the biosensor chip can successively and rapidly

generate the stable current steps for each H2O2 addition. It can be found that the response current can show equal decrease with the increase of the same H2O2 concentration in the buffer solution, showing the enhancing reduction reaction to H2O2. As calculated, its current density shows a well-defined linear relation to the H2O2 concentration with the fitting equation of j=-41.70C+94.34 (R2=0.999), where j and C represent the current density (µA·cm−2) and target concentration (mM), respectively. Hence, the as-prepared chip possesses the sensitivity of 41.70

ro of

µA·mM−1·cm−2 on H2O2 detection. Actually, the activity of the porous Co3O4 nanocube can be evaluated via such sensitivity value. Compared with some other similar modified electrodes based on Co3O4 material, the porous Co3O4 nanocube can

-p

exhibit high sensitivity at a low working potential (-0.2V). The comparison results are listed in Table S1.

re

For achieving the glutamate detection, 0.5 U GMOx was immobilized on the

lP

working electrode by the glutaraldehyde cross-link. After drying in the 4 ℃, the biosensing performance of the as-prepared enzymatic chip was also evaluated by the chronoamperometry method. As shown in Fig. 7A, after each addition of 0.05 mM

na

glutamate, a current step can rapidly appear to show the increase of the negative current. Compared with the curve in the H2O2 detection, also under the -0.2 V, the

ur

current signal shows some weak noises which may be attributed to the weak

Jo

conductivity of the loaded oxidase. However, the response current value is much higher than the noise value to neglect this interference. According to the linear calibration (Fig. 7B), the current density can present a linear relation to the glutamate concentration with the equation of j=-20.12 C+80.77 (R2=0.998). As calculated, this SPE-chip shows a high sensitivity of 20.12 µA mM−1 cm−2 for the glutamate detection and a wide linear response from 10 to 600 μM (R2=0.998) with a short response time

within 5 s. The limit of detection was tested through the decrease of the added glutamate concentration to distinguish the response current. As shown in Fig. 7A, till decreasing to 10 μM, a clear current increase was also produced to indicate the detection limit. The performance of the as-prepared SPE-chip was compared with those of the recently reported glutamate biosensors. As shown in Table 1, carbon nanotubes, polymers and metal oxides are mainly applied to fabricate the glutamate biosensors.

ro of

Among these materials, the porous Co3O4 nanocubes show a lowest operation potential due to the reduction mechanism on H2O2 produced by the glutamate oxidation. Low potential is beneficial to avoid the oxidation reaction of other

-p

co-existed substances to confirm its detection accuracy. Besides, different with the composite materials based biosensor, it can be found that our prepared biosensor can

re

be of the almost highest sensitivity which is only attributed to the Co3O4. More

lP

important, the enzyme loading amount is only 0.5 u on each chip, which is lowest but harvests the superior performance. These evidences indicate that the prepared Co3O4 with the special architecture can greatly promote the electrocatalysis on the enzymatic

na

reaction of glutamate.

3.4 Anti-interference ability, reproducibility and stability of the biosensor chips

ur

The accuracy of the biosensor is an essential parameter in the practical applications

Jo

in the real samples, such as the fermentation solution and blood. Therefore, this parameter of the as-prepared SEP-chip was evaluated by using ascorbic acid (AA) and uric acid (UA) as two interfering species during the glutamate detection process because of their very low oxidation potentials. In this experiment, these two substances were successively added after the addition of glutamate with the same concentration of 0.05 mM during the chronoamperometry test. As shown in Fig. 8A,

the addition of either AA or UA (100 folds to standard concentration of glutamate) has not aroused any obvious current fluctuation, however, the further addition of glutamate can still produce the current step. This is because that the operation potential (−0.2 V) of our SPE-chip is much lower than the oxidation potentials of AA and UA (0.2 V and 0 V), which effectively prevented their oxidations for the current generation in the detection process. Besides, we also find that the addition of glucose and lactate for 100 folds concentration to the standard glutamate solution could not

ro of

produce any obvious current fluctuation. Due to the specificity nature of the applied GMOx, the as-prepared chip can show excellent selectivity to avoid the interferences, such as glucose and lactate. Furthermore, the reproducibility of the fabricated chips

-p

was tested by of the random selection of ten chips printed in different batches. As

calculated, the relative standard deviation (RSD) of the sensitivity was 4.8% to

re

indicate the good reproducibility which is mainly attributed to the superior material

lP

stability. Besides, after the storage of a fresh chip at 4 ℃ for 30 days, the sensitivity can retain about 89.59% of the initial one, which indicated the good storage stability. In order to investigate the usage stability, one chip was repetitively tested by days to

na

evaluate the fluctuation of its sensitivity. In each day, more than 20 times of glutamate additions were operated to calculate this parameter. As shown in Fig. 8B, after 30

ur

days with 300 times of glutamate test, the sensitivity can still remain 79.32% level of

Jo

the fresh chip. This remarkable performance is attributed to the direct printing of nanomaterials instead of the subsequent deposition, enhancing the stability during application.

3.5 Application of the biosensor chip in real fermentation sample In order to evaluate the accuracy of the biosensor chip in practical analysis, the prepared biosensor chips were employed to measure concentration of glutamate in the

real fermentation sample. As shown in Fig. S5, 1μL glutamate (2 M) was first added into 10 mL PBS as the standard sample. Then 1 mL glutamate fermentation solution was added to evaluate the practical performance of the biosensor chip. The tested glutamate concentrations in the fermentation sample were tested 3.92 and 3.95 mM by our prepared biosensor chip and HPLC, respectively. Moreover, three repetitive experiments were operated to show the average deviation of 3.13% in Table S2, demonstrating the good accuracy of the as-prepared biosensor chip in the practical

ro of

fermentation application. 4. Conclusions

In summary, a novel Co3O4 crystal with porous nanocubic structure was

-p

successfully constructed through a facile thermal oxidation of the CoHCF precursor. This material was served as the printing ink to directly fabricate a glutamate biosensor

re

microchip. Using a very low working potential of -0.2 V, this chip exhibited a high

lP

sensitivity with an excellent anti-interference ability, as well as the good storage and usage stability. Besides of the glutamate detection, this chip is promising to extent the recognitions of more physiological substances by changing different oxidases.

na

Besides, this large-scale production route from the nanomaterial synthesis to the biosensor chip printing can provide the inspiration to promote the product transfer in

Jo

ur

the practical applications.

Declaration of Interest Statement

This work is funded by National Natural Science Foundation of China (Nos. 21706116 and 21727818), the Innovative Research Team Program

by the Ministry of Education of China (No. IRT_17R54), the Top-notch Academic Programs Project of Jiangsu Higher Education Institutions (TAPP) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). This submission has been agreed by all authors.

ro of

Zhenyu Chu, Ph.D

Acknowledgement

-p

This work was financially supported by National Natural Science Foundation of

China (Nos. 21706116 and 21727818), the Innovative Research Team Program by the

re

Ministry of Education of China (No. IRT_17R54), the Top-notch Academic Programs

lP

Project of Jiangsu Higher Education Institutions (TAPP) and the Priority Academic

Reference

na

Program Development of Jiangsu Higher Education Institutions (PAPD).

[1] T. Salt, S. Eaton, Functions of ionotropic and metabotropic glutamate receptors in

55-72.

ur

sensory transmission in the mammalian thalamus, Progress In Neurobiology 48 (1996)

Jo

[2] T. Borisova, D. Kucherenko, O. Soldatkin, I. Kucherenko, A. Pastukhov, A. Nazarova, M. Galkin, A. Borysov, N. Krisanova, A. Soldatkin and A. El Skaya, An amperometric glutamate biosensor for monitoring glutamate release from brain nerve terminals and in blood plasma, Anal. Chem. Acta. 1022 (2018) 113-123. [3] S. Chakraborty, C. Retna Raj, Amperometric biosensing of glutamate using carbon nanotube based electrode, Electrochem. Commun. 6 (2007) 1323-1330. [4] M.A. Haxhiu, J.C. Chavez, P. Pichiule, B. Erokwu, I.A. Dreshaj, The excitatory

amino acid glutamate mediates reflexly increased tracheal blood flow and airway submucosal gland secretion, Brain Res. 1 (2000) 77-86. [5] G. Alfredsson, F.A. Wiesel, A. Tylec, Relationships between glutamate and monoamine metabolites in cerebrospinal fluid and serum in healthy volunteers, Biol. Psychiatry 7 (1988) 689-697. [6] M. Ganesana, E. Trikantzopoulos, Y. Maniar, S.T. Lee, B.J. Venton, Development of a novel micro biosensor for in vivo monitoring of glutamate release in the brain, Biosens. Bioelectron. 130 (2019) 103-109.

immunosensor

based

on

multi-Walled

carbon

ro of

[7] X. Sun, Y. Cao, Z. Gong, X. Wang, Y. Zhang, J. Gao, An amperometric nanotubes-thionine-chitosan

nanocomposite film for chlorpyrifos Detection, Sensors 12 (2012) 17247-17261.

[8] O. Soldatkin, A. Nazarova, N. Krisanova, A. Borysov, D. Kucherenko, I.

-p

Kucherenko, N. Pozdnyakova, A. Soldatkin, T. Borisova, Monitoring of the velocity of high-affinity glutamate uptake by isolated brain nerve terminals using

re

amperometric glutamate biosensor, Talanta 135 (2015) 67-74.

[9] C. Zuo, Y. Zhang, J. Wang, L. Han, C. Peng, D. Peng, Deciphering the

lP

intervention mechanism of Taohong Siwu Decoction following the abnormal uterine bleeding rats based on serum metabolic profiles, J. Pharmaceut. Biomed. 170 (2019)

na

204-214.

[10] L.C. Clark, C. Lyous, Electrode system for continuous monitoring in cardiovascular surgery, Annals New York Academy of Science (1962) 29-45.

ur

[11] P. Yang, J. Peng, Z. Chu, D. Jiang, W. Jin, Facile synthesis of Prussian blue nanocubes/silver nanowires network as a water-based ink for the direct screen-printed

Jo

flexible biosensor chips, Biosens. Bioelectron. 92 (2017) 709-717. [12] X. Dong, H. Xu, X. Wang, Y. Huang, M.B. Chan-Park, H. Zhang, L. Wang, W. Huang, P. Chen, 3D graphene-cobalt oxide electrode for high-performance supercapacitor and enzymeless glucose detection, Acs Nano 4 (2012) 3206-3213. [13] M. Haddaoui, N. Raouafi, Chlortoluron-induced enzymatic activity inhibition in tyrosinase/ZnO NPs/SPCE biosensor for the detection of ppb levels of herbicide, Sens. Actuators B 219 (2015) 171-178.

[14] M. Bilgi, E.M. Sahin, E. Ayranci, Sensor and biosensor application of a new redox mediator: Rosmarinic acid modified screen-printed carbon electrode for electrochemical determination of NADH and ethanol, J. Electronal. Chem. 813 (2018) 67-74. [15] D. Jiang, Z. Chu, J. Peng, W. Jin, Screen-printed biosensor chips with Prussian blue nanocubes for the detection of physiological analytes, Sens. Actuators B 228 (2016) 679-687. [16] G. Hughes, R.M. Pemberton, P.R. Fielden, J.P. Hart, Development of a novel screen-printed

amperometric

biosensor

based

on

glutamate

ro of

reagentless,

dehydrogenase and NAD+, integrated with multi-walled carbon nanotubes for the

determination of glutamate in food and clinical applications, Sens. Actuators B 216 (2015) 614-621.

sensors, Sens. Actuators B 243(2017) 919-926.

-p

[17] Z. Chu, J. Peng, W. Jin, Advanced nanomaterial inks for screen-printed chemical

re

[18] D. T., K.V. Rao, K. Bikshalu, V. Malapati, K.K. Sadasivuni, Non-enzymatic sensing of glucose using screen-printed electrode modified with novel synthesized

lP

CeO2@CuO core shell nanostructure, Biosens. Bioelectron. 111 (2018) 166-173. [19] D. Jiang, C. Li, T. Liu, L. Li, Z. Chu, W. Jin, X. Ren, A regular nanostructured

na

dithiolene metal complex film for ultrasensitive biosensing of liver enzyme, Sens. Actuators B 241 (2017) 860-867.

[20] A. Dudina, U. Frey, A. Hierlemann, Carbon-nanotube-based monolithic CMOS

ur

platform for electrochemical detection of neurotransmitter glutamate, Sensors 19 (2019) 3080.

Jo

[21] Y. Zheng, Z. Liu, H. Zhan, J. Li, C. Zhang, Studies on electrochemical organophosphate pesticide (OP) biosensor design based on ionic liquid functionalized graphene and a Co3O4 nanoparticle modified electrode, Anal. Methods 8 (2016) 5288-5295. [22] H. Dai, Y. Chen, X. Niu, C. Pan, H. Chen, X. Chen, High-performance electrochemical biosensor for nonenzymatic H2O2 sensing based on Au@C-Co3O4 heterostructures, Biosens. Bioelectron. 118 (2018) 36-43.

[23] C. Kaçar, B. Dalkiran, P.E. Erden, E. Kiliç, An amperometric hydrogen peroxide biosensor based on Co3O4 nanoparticles and multiwalled carbon nanotube modified glassy carbon electrode, Appl. Surf. Sci. 311 (2014) 139-146. [24] Y. Ye, Y. Zheng, J. Yu, Y. Zhang, Y. Song, Q. Guo, L. Wang, F. Xu, Flexible cobalt hexacyanoferrate/carbon cloth nanocomposites for H2O2 detection, Anal. Methods 12 (2016) 2666-2672. [25] L. Hu, Q. Peng, Y. Li, Selective Synthesis of Co3O4 Nanocrystal with Different Shape and Crystal Plane Effect on Catalytic Property for Methane Combustion, J. Am.

ro of

Chem. Soc. 48 (2008) 16136-16137. [26] J. Liu, J. Jiang, C. Cheng, H. Li, J. Zhang, H. Gong, H.J. Fan, Co3O4 nanowire@MnO2

ultrathin

nanosheet

core/shell

arrays:

A

new

class

of

high-performance pseudocapacitive materials, Adv. Mater. 18 (2011) 2076-2081.

-p

[27] Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier, H. Dai, Co₃ O₄

nanocrystals on graphene as a synergistic catalyst for oxygen reduction reaction, Nat.

re

Mater. 10 (2011) 780-786.

[28] A.I. Rekertaitė, A. Valiūnienė, P. Virbickas, A. Ramanavicius, Physicochemical

lP

characteristics of polypyrrole/(Glucose oxidase)/(Prussian blue)-based biosensor modified with Ni- and Co-hexacyanoferrates, Electroanal 1 (2019) 50-57.

na

[29] B. Kong, C. Selomulya, G.F. Zheng, D.Y. Zhao, New faces of porous Prussian blue: interfacial assembly of integrated hetero-structures for sensing applications, Chem. Soc. Rev. 44 (2015) 7997-8108.

ur

[30] A. Zanotto, A. Spinella, G. Nasillo, E. Caponetti, A.S. Luyt, Macro-micro relationship in nanostructured functional composites, Express Polym. Lett. 5 (2012)

Jo

410-416.

[31] N. Yan, L. Hu, Y. Li, Y. Wang, H. Zhong, X. Hu, X. Kong, Q. Chen, Co3O4 nanocages for high-performance anode material in Lithium-Ion batteries, J. Phys. Chem. C, 12 (2012) 7227-7235. [32] Y. Chen, L. Hu, Novel Co3O4 porous polyhedrons derived from metal–organic framework toward high performance for electrochemical energy devices, J. Solid. State. Chem, 239 (2016)23-29.

[33] S.A.B. Asif, S.B. Khan, A.M. Asiri, Electrochemical sensor for H2O2 using a glassy carbon electrode modified with a nanocomposite consisting of graphene oxide, cobalt(III) oxide, horseradish peroxidase and nafion, Microchim. Acta. 11 (2016) 3043-3052. [34] S. Liao, S. Lu, S. Bao, Y. Yu, L. Yu, Electrospinning Synthesis of Porous CoWO4 Nanofibers as an Ultrasensitive, Nonenzymatic, Hydrogen-Peroxide-Sensing Interface with Enhanced Electrocatalysis, ChemElectroChem 12 (2015) 2061-2070. [35] B. Dalkiran, P.E. Erden, E. Kilic, Graphene and tricobalt tetraoxide nanoparticles

ro of

based biosensor for electrochemical glutamate sensing, Artif. Cells. Nanomed. Biotechnol. 2 (2017) 340-348.

[36] Z.Y. Chu, L. Shi, Y.N. Zhang, W.Q. Jin, S. Warren, D. Ward, E. Dempsey,

J. Mater. Chem. 22 (2012) 14874–14879.

-p

Single layer prussian blue grid as a versatile enzyme trap for low-potential biosensors,

[37] E. Kaivosoja, N. Tujunen, V. Jokinen, V. Protopopova, S. Heinilehto, J.

re

Koskinen, T. Laurila, Glutamate detection by amino functionalized tetrahedral amorphous carbon surfaces, Talanta 141 (2015) 175-181.

lP

[38] R. Monošík, M. Streďanský, E. Šturdík, A biosensor utilizing L-Glutamate dehydrogenase and diaphorase immobilized on nanocomposite electrode for

na

determination of L-glutamate in food samples, Food Anal. Methods 6 (2013) 521-527. [39] V.M. Tolosa, K.M. Wassum, N.T. Maidment, H.G. Monbouquette, Electrochemically deposited iridium oxide reference electrode integrated with an

ur

electroenzymatic glutamate sensor on a multi-electrode arraymicroprobe, Biosens. Bioelectron. 42 (2013) 256-260.

Jo

[40] L. Kergoat, B. Piro, D.T. Simon, M. Pham, V. Noël and M. Berggren, Detection of Glutamate and Acetylcholine with Organic Electrochemical Transistors Based on Conducting Polymer/Platinum Nanoparticle Composites, Adv. Mater. 32 (2014) 5658-5664.

Biographies

Zhenyu Chu is Associate Professor of chemical engineering at Nanjing Tech University and his research mainly focuses on the nanomaterial design for biosensors and screen-printed microchip. He received his B.Sc. and Ph.D. degree in chemical engineering from the Department of Chemistry and Chemical Engineering, Nanjing

ro of

Tech University, China in 2008 and 2013. He studied in Institute of Technology

Tallaght, Dublin, Ireland as a collaborative visitor in 2011. He worked as a visiting professor in Curtin University, Australia from 2017 to 2018. He has published over 50

-p

internationally refereed journal papers on nanomaterials based biosensors. He serves

re

as an associate editor of Frontiers in Nanotechnology and member of Medical, Biological Membrane Committee in Chinese Industry Association.

lP

Wanqin Jin is Professor of chemical engineering at Nanjing Tech University and His

na

current research focuses on the development of membrane and biosensing materials, electrochemical analysis and the production transfer of biosensors by screen-printing

ur

technique. He was an Alexander von Humboldt Research Fellow (2001), and a visiting Professor at Arizona State University (2007) and Hiroshima University (2011,

Jo

JSPS invitation fellowship). He has published over 250 internationally refereed journal papers and edited a book on materials-oriented chemical engineering. He serves as associate editor and an editorial board member for several journals and is a council member of the Aseanian Membrane Society. Fuhao Hu received his B.Sc. degree in chemical engineering from the Department of Chemical and Environmental Engineering, Jiangsu University of Technology, China

in 2016. Currently he is undertaking the master research at Nanjing Tech University. Tao Liu received his B.Sc. degree in chemical engineering from the Department of Chemical Engineering, Nanjing Tech University, China in 2017. Currently he is undertaking the doctor research at Nanjing Tech University. Jun Pang received his B.Sc. degree in Resource Science and Engineering from the College of environment, Nanjing Tech University, China in 2016. Currently he is

Jo

ur

na

lP

re

-p

ro of

undertaking the master research at Nanjing Tech University.

Figure Captions Fig. 1. (A) and (B) present the FESEM image and TGA curve of the synthesized CoHCF nanocubes, respectively; (C) XRD patterns of varying the temperature to the CoHCF from 25 to 500 ℃, (D)XRD pattern of the calcined material from the CoHCF

re

-p

ro of

precursor under 400 ℃ for 2h.

Fig. 2. XPS spectra of (A) O 1s, (B) Co 2p patterns of the prepared CoHCF before

Jo

ur

na

lP

and after the thermal oxidation at 400 ℃ for 2 h.

Fig. 3. FESEM images of CoHCF nonocubes in different calcining temperatures for 2h, respectively: (A) 200℃, (B) 300℃, (C) 400℃ and (D) 500℃. (E) and (F) present the TEM images of the thermal oxidized material under different magnifications. The inset in (F) shows its diffraction image.

ro of

Fig. 4. (A) Digital image of the self-made flexible biosensor chip (B) FESEM image of working electrode. EDX im-ages of the selected region of working electrode, (C)

na

lP

re

-p

cobalt element and (D) oxygen element.

Fig. 5. CV diagrams of the CoHCF SPE-chip (A) and Co3O4 SPE-chip (B) at different

ur

scan rates of 50, 100, 150, 200 and 250mV s−1 in a 5 mM [Fe(CN)6]4/3- indicator solution. (C) provides the linear fitting of the peak current value to the root of the scan

Jo

rate from 50 to 250 mV s−1. (D) shows the EIS diagrams of the bare substrate, CoHCF and Co3O4 based SPE-chips in a 5 mM [Fe(CN)6]4/3- indicator solution.

ro of

Fig. 6. (A) CV diagrams of the prepared SPE-chip with and without the addition of 1

mM H2O2 under the scan rate of 50 mV/s in a 0.05M PBS solution at 25 ℃. (B)

-p

Chronoamperometry result of the SPE-chip to the H2O2 detection under -0.2 V working potential in a 0.05M PBS solution at 25 ℃. The inset shows a linear

ur

na

lP

re

calibration of the response current to the added H2O2 concentration.

Jo

Fig. 7. (A) Amperometric responses with different glutamate concentrations in a 0.05 M PBS solution at 25℃. The inset shows the response time. (B) Linear calibration result of the response current to the added glutamate concentration. The inset shows the partial enlargement.

ro of

Fig. 8. (A) The anti-interference test of the SPE-chip to AA, UA, glucose and lactate.

Jo

ur

na

lP

re

-p

(B) Usage stability test of the as-prepared SPE-chip for 30 days.

Scheme 1. Fabrication of the screen-printed biosensor coating of Co3O4 porous

Jo

ur

na

lP

re

-p

ro of

nanocubes and the electrocatalytic mechanism for glutamate detection.

Table 1. Performance comparisons of the reported glutamate biosensors.

Table 1. Performance comparisons of the reported glutamate biosensors. Enzyme loading amount (U)

Sensitivity (mA M−1 cm−2)

Detection limit (µM)

Linear range (μM)

-0.2

0.5

20.12

10

10-600

CNT/GCE

-0.1

27

0.71

2

NA

IR/PPD/Pt

0.6

0.6

Co3O4/GR/CS/GCE

0.7

0.24

7.37

GS-GMOD/PB/Pt

-0.1

0.5

12.36

Ta-C/APTES/Pt

0.6

0.6

MWCNT/Chit/Au

0.3

3.3

PPy/IrOx/Pt

0.6

na 0.4

Jo

ur

PSS/PtNPs/Glass

0.097(mA

This work [3]

0.044

5-150

[6]

2

4-600

[35]

NA

NA

[36]

2.9

10

10-500

[37]

28

5.4

NA

7.7

0.32

5-300

[39]

NA

4.3

5

1-14

[40]

-p

M-1)

re

nanocubes/SPE

lP

Co3O4

Ref.

ro of

Potential (V vs. Ag/AgCl)

Modifiers

10-349 5

[38]