A non-enzymatic sensor based on three-dimensional graphene foam decorated with Cu-xCu2O nanoparticles for electrochemical detection of glucose and its application in human serum

Journal Pre-proof A non-enzymatic sensor based on three-dimensional graphene foam decorated with Cu-xCu2O nanoparticles for electrochemical detection of glucose and its application in human serum Zahra Khosroshahi, Fathallah Karimzadeh, Mahshid Kharaziha, Alireza Allafchian PII:

S0928-4931(19)31239-1

DOI:

https://doi.org/10.1016/j.msec.2019.110216

Reference:

MSC 110216

To appear in:

Materials Science & Engineering C

Received Date: 2 April 2019 Revised Date:

6 September 2019

Accepted Date: 16 September 2019

Please cite this article as: Z. Khosroshahi, F. Karimzadeh, M. Kharaziha, A. Allafchian, A nonenzymatic sensor based on three-dimensional graphene foam decorated with Cu-xCu2O nanoparticles for electrochemical detection of glucose and its application in human serum, Materials Science & Engineering C (2019), doi: https://doi.org/10.1016/j.msec.2019.110216. 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 B.V.

Article title: A non-enzymatic sensor based on three-dimensional graphene foam decorated with Cu-xCu2O nanoparticles for electrochemical detection of glucose and its application in human serum

Graphical abstract:

1 2 3 4 5 6 7 8 9

A non-enzymatic sensor based on three-dimensional graphene foam decorated with Cu-xCu2O nanoparticles for electrochemical detection of glucose and its application in human serum Zahra Khosroshahi1, Fathallah Karimzadeh1,*, Mahshid Kharaziha1, Alireza Allafchian2 1

2

Department Materials Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran. Researcher Institute for Nanotechnology and Advanced Materials, Isfahan University of Technology, Isfahan 84156-83111, Iran.

*Corresponding author Email: [email protected]

10

Abstract

11 12 13 14 15 16 17 18 19 20 21 22 23 24

In the present study, a porous structure consisting of three-dimensional graphene (3DG) decorated with Cu-based nanoparticles (NPs) (Cu or Cu-Cu2O) was synthesized in order to develop an enzyme-free electrochemical glucose sensor. Moreover, the effects of Cu-based nanoparticle concentrations on electrochemical properties and glucose detection were evaluated by cyclic voltammetry, electrochemical impedance spectroscopy, and differential pulse voltammetry. Cu-based NPs@3DG showed markedly better electrochemical performance in glucose oxidation in alkaline solution compared to 3DG foam. Moreover, CuCu2O NPs@3DG foam with the lowest concentration of Cu precursor showed excellent performance in glucose detection. A high sensitivity of 230.86 µA.mM-1.cm-2 was obtained for this electrode in a linear range of 0.8-10 mM (R2=0.9951) and a detection limit of 16 µM. This sensor also showed good repeatability (relative standard deviation of 1.02%) and high selectivity (no observation of significant interference by interfering species such as ascorbic acid, dopamine, urea, and acetaminophen). The results confirmed that this electrode could be applied as a feasible and inexpensive non-enzymatic electrochemical glucose sensor.

25 26

Keyword: Electrochemical sensor; Glucose; Three dimensional graphene; Cu-based nanoparticles; Non-enzymatic; Differential pulse voltammetry.

27

1. Introduction

28

Improvement of glucose sensors is very important in a number of fields, such as

29

environmental monitoring, medical science, and the food processing industry. It is

30

particularly important in blood glucose testing, pharmaceutical analysis and energy

31

conversion [1, 2]. Because of increases in the number of patients with diabetes, many

32

researchers are currently focusing on the development of glucose sensors. In 2015, the

33

International Diabetes Federation reported that approximately 387 million people were

34

suffering from diabetes [3]. As a result of deficiencies in insulin production or the poor

35

response of the body cells to insulin, metabolic disorders are widely distributed, leading to

36

disabilities like blindness, nerve degeneration, kidney failure and sometimes death.

37

Therefore, specific monitoring is required for the detection and control of glucose blood

38

levels that are higher than normal (4.4–6.6 mM) [4, 5]. A variety of analytical approaches 1

1

have been applied to glucose monitoring, including electrochemical, optical, piezoelectric,

2

thermistors

3

electrochemical methods are widely used for glucose testing because of their sensitivity,

4

practicability, simplicity, time efficiency, ease of operation, and low production costs [3, 7-

5

13]. Electrochemical-based sensors are divided into two types: enzymatic and non-enzymatic.

6

Enzymatic glucose sensors that are based on immobilization of glucose oxidase (GOx) to

7

oxidize

8

immobilization procedures, chemical and thermal instability, degradation of GOx, low

9

stability, high cost, and dependence of the response on operational conditions. In contrast, in

10

non-enzymatic approaches, glucose is directly oxidized to gluconolactone at nanostructured

11

electrodes, providing a large reaction region for effective electrocatalytic activity [3, 7, 14].

12

Non-enzymatic approaches have drawn a lot of interest because of their high sensitivity and

13

reliability, fast response, good selectivity, and low detection limit [15, 16].

and

glucose

semiconductor-based

to

gluconolactone

electrodes

have

some

[6].

Among

these

disadvantages:

approaches,

complex

enzyme

14

In the last decade, graphene, consisting of two-dimensional (2D) monolayers of sp2

15

carbon atoms, has attracted interest for bioensors because of its superior electron mobility,

16

high thermal conductivity, excellent mechanical flexibility and strength, large specific surface

17

area, and good biocompatibility [17-20]. However, graphene nanosheets have a tendency to

18

agglomerate, presenting significant challenges for the use of two-dimensional sheets of

19

graphene in sensor applications. High contact resistance between graphene sheets and

20

restacking of graphene sheets confine the active surface area and electron pathways, leading

21

to poor electrochemical performance [21, 22]. Consequently, development of three-

22

dimensional graphene (3DG) foam with interconnected porosity could resolve these issues

23

and result in the formation of excellent conductive foam with high specific surface areas,

24

robust mechanical strength, and fast electron transport kinetics to replace conventional

25

electrodes [23-25]. Up to the present, various approaches have been introduced for the

26

synthesis of porous 3DG foam: self-assembly (i.e., gelation [26] and hydrothermal reduction

27

[27]), template-guided methods [22, 28, 29] and a leavening strategy [30]. Template

28

approaches provide well-defined 3D porous networks with controllable pore sizes. However,

29

most template techniques require a post-treatment approach that is time-consuming and often

30

leads to the collapse of the original structure [22]. Polystyrene (PS) colloid templates are

31

typical sacrificial templates that are often applied to form porous 3DG foam with

32

micrometer-sized porosity based on electrostatically induced assembly [24, 31]. In this

33

procedure, at a specific pH value, graphene layers and PS particles have negative and positive 2

1

charges, respectively. Uniform 3D foam can be formed by electrostatic interaction between

2

negative and positive charges and, subsequently, removing the PS particles [28, 29]. The

3

physiological characteristics of graphene-based sensors generally depend on the number and

4

dispersion of the graphene layers. Incorporation of nanoparticles between graphene

5

nanosheets decreases agglomeration, leading to nanocomposites with improved super-

6

capacitive, conductivity and other properties depending on the nanoparticle types [32, 33].

7

Therefore, a variety of nanocomposites consisting of metal and metal oxide nanoparticles

8

decorated graphene (or reduced graphene oxide [rGO]) have been reported for non-enzymatic

9

glucose detection, including Ni-rGO [34], Pt-rGO [35], Au-graphene [36], Cu-graphene [37],

10

NiO-rGO [38, 39], SnO2-rGO [40], Mn3O4-rGO [41], CuO-graphene [42], and Cu2O-

11

graphene [43]. Copper (Cu) and its oxides are cost-effective components that are easy to

12

synthesize and show high chemical stability, good electrochemical activity, and excellent

13

electrical conductivity properties [44-47]. In addition to pure Cu nanoparticles, cuprous oxide

14

(Cu2O), a non-toxic semiconductor with a band gap of 2.17 eV and specific optical and

15

magnetic characteristics, has received a lot of attention as a biosensing material [48-51].

16

Recent studies have demonstrated that Cu and Cu2O nanoparticles have the prominent

17

electrochemical characteristic of catalyzing biomolecules, making them suitable for direct

18

electrochemical detection of glucose without further mediators [52-58]. Ju et al. [59]

19

synthesized a Cu nanowire-rGO nanocomposite for non-enzymatic glucose detection that

20

showed improved electrochemical performance compared to rGO for glucose detection.

21

Wang et al. [60] developed an inexpensive flexible electrochemical non-enzymatic glucose

22

sensor by in situ growth of Cu nanoflowers on rGO sheets. They showed that this sensor

23

performed better than rGO in glucose detection.

24

Because 3DG foam has electrochemical properties superior to those of graphene

25

nanosheets, and because of the effective role of metal and metal oxide nanoparticles in the

26

reduction of rGO sheet agglomeration, recent research has focused on porous 3DG-metal or

27

metal oxide nanoparticle foam for non-enzymatic sensors [61-66]. Wang et al. [67]

28

synthesized a glucose sensor with ultra-high sensitivity by in situ growth of Cu2O

29

nanoparticles on a 3DG network. In another study, Ma et al. [68] synthesized 3DG foam

30

decorated with CuO nanoflowers for ascorbic acid detection. They confirmed that 3DG foam

31

was favorable for mass transport and encapsulation of bioactive molecules. Moreover, CuO

32

nanoflowers promote the active surface area and catalyze the redox of ascorbic acid. It is

3

1

thought that 3DG foam decorated with Cu nanopoartiacles might be an excellent substrate for

2

non-enzymatic glucose detection.

3

The aim of the present study was to synthesise 3DG foam decorated with Cu-based

4

nanoparticles (Cu-xCu2O NPs@3DG nanocomposite) using a PS template guide procedure to

5

develop a sensor with large surface area, a well-ordered 3D structure, and ultra-high

6

sensitivity for glucose detection. Moreover, the effect of Cu-based nanoparticle

7

concentrations on the efficiency of CuO NPs@3DG foam for in vitro glucose detection was

8

evaluated.

9

2. Experimental

10

2.1. Reagents and materials

11

Graphene oxide powder (purity>99%, thickness: 3.4-7 nm, 6-10 layers) was obtained

12

from Nanosay Co., Iran. Hydrazine hydrate (N2H4, purity≥80%), hydrochloric acid (HCl,

13

37%), L(+)-ascorbic acid (C6H5O6, purity≥99.7%), 2-propanol (C3H8O), sodium hydroxide

14

(NaOH, purity≥99%) and an ammonia solution (NH4OH, purity≥28%) were obtained from

15

Merck Co., Germany. Copper (II) chloride dihydrate (CuCl2.2H2O, purity≥99%), Styrene,

16

methyl methacrylate, PVP K-30, ethyl alcohol and ammonium persulfate was obtained from

17

Sigma-Aldrich Co., USA. Glassy carbon electrode (GCE) was prepared by Azar Electodr

18

Co., Iran..

19

2.2. Synthesis procedures

20

2.2.1. Preparation of PS nanoparticles

21

Styrene (30 wt. % based on total, as monomer), methyl methacrylate (1.0 wt. % based on

22

monomer, as co-monomers copolymerized with styrene to form co-stabilizers in situ), PVP

23

K-30 (1.8 wt. % based on monomer, as stabilizer) and ethyl alcohol were weighed and poured

24

into a three-neck round-bottom flask equipped with a condenser and immersed in an oil bath

25

with an actual temperature. Nitrogen gas was bubbled through the solution to remove oxygen

26

from the system. Then agitation was started at a given speed. As soon as the temperature

27

reached the actual desired value, a solution of ammonium persulfate (1.0 wt. % based on

28

monomer) in ethyl alcohol was added to initiate polymerization. After 30 minutes,

29

hydrochloric acid (0.2 vol. %, based on monomer) was added to the reaction system.

30

Polymerization was carried out at 75 °C for 24 hours in a nitrogen atmosphere.

31

2.2.2. Synthesis of 3D graphene (3DG) foam

4

1

The fabrication process for 3DG foam is schematically represented in Fig. 1. Briefly, GO

2

nanopowder was dispersed in deionized (DI) water (0.5 mg/ml) using ultrasonication for 1

3

hour. Subsequently, 5 µl ammonia, followed by a hydrazine hydrate solution (weight ratio of

4

GO to hydrazine was 10:7,) were added to 5 ml of the GO solution. After the formation of a

5

homogeneous mixture, the solution was placed in a water bath at 95 ⁰C for 1 hour to reduce

6

the GO. The PS solution was prepared by dispersing PS particles in DI water (10 wt%) at

7

pH=2 (the pH was adjusted by adding HCl solution). The previously prepared reduced

8

graphene oxide (rGO) solution and the PS solution were mixed (95:5 in weight ratio) at pH=2

9

and stirred vigorously for 30 minutes. Subsequently, the mixed solution was placed under

10

ultrasonication at pH 6-8 for 30 minutes (the pH was adjusted by adding NaOH solution).

11

The PS/rGO foam was obtained by centrifuging (1600 rpm, 5 min) the suspension and

12

washing three times with DI water and ethanol; it was then dried in a vacuum oven at 50-60

13

⁰C for 15 hours. Finally, the 3DG foam was prepared by immersing the PS/rGO construct in

14

toluene for 1 hour, washing it with DI water and ethanol, and drying it in the vacuum oven at

15

50-60 ⁰C.

16

2.2.3. Synthesis of Cu-xCu2O NPs@3DG foam

17

The synthesis process for Cu-xCu2O NPs@3DG foam is schematically presented in Fig.

18

1. After preparation of CuCl2.2H2O solution in DI water and 2-propanol with volume ratio of

19

1:1, the previously synthesized 3DG foam, NaOH and ascorbic acid were added to it

20

sequentially under vigorous stirring. The estimated amounts of the materials are presented in

21

Table 1. In order to show the role of Cu-xCu2O in the electrochemical properties of the final

22

Cu-xCuO NPs@3DG foam, 3DG foam were dipped in various concentrations of CuCl2.2H2O

23

solutions. The samples were labeled according to the concentration of CuCl2.2H2O solution

24

(Table 1). The mixtures were heated to 95 ⁰C and maintained for 15 minutes. Subsequently,

25

the samples of Cu-xCu2O NPs@3DG foam were extracted using centrifugation (5 min, 1600

26

rpm) of the solution, washing the precipitated mixture three times with DI water and ethanol,

27

and, finally, drying them in an oven at 50-60 °C for 15 hours.

28

Table 1. The estimated amount of applied material for preparation of Cu-xCu2O NPs@3DG foam

Sample

3DG/CuCl2.2H2O solution (mg/ml)

3DGC1

0.3

CuCl2. 2H2O solution concentration (M) 0.05

6.32

Ascorbic acid/ CuCl2.2H2O (weight ratio) 9.6

3DGC2

0.3

0.1

6.32

9.6

3DGC3

0.3

0.15

6.32

9.6

5

NaOH/CuCl2.2H2O (weight ratio)

3DGC4

0.3

0.2

6.32

9.6

3DGC5

0.3

0.5

6.32

9.6

3DGC6

0.3

1

6.32

9.6

1 2 3

Figure 1. Schematic of the procedure for fabrication of 3DG and Cu based NPs@3DG foam.

2.3. Characterization of 3DG-based foam

4

The crystal structures of GO, 3DG, and Cu-xCu2O NPs@3DG nanocomposite foam were

5

evaluated using X-ray diffraction (XRD, Philips X’Pert-MPD, Holland) with Cu Kα radiation

6

(λ=0.542); phase analysis was performed by X’Pert High Score software. The surface

7

morphology of GO, 3DG, and Cu-xCu2O NPs @rGO foam were characterized using

8

scanning and transmission electron microscopy (SEM, Philips XL30 SERIES and TEM, EM

9

208S) and analyzed using ImageJ software. Before SEM imaging, the samples were coated

10

with a thin layer of gold. The presence of copper in Cu-xCu2O NPs@rGO foam was deduced

11

by energy dispersive X-ray (EDS SERON AIS2300C). Functional groups of 3DG and Cu-

12

xCu2O NPs@3DG foam were identified using a SENTERRA II Raman spectroscope

13

(Bruker, Germany) equipped with a 785 nm laser. Fourier-transform infrared spectroscopy

14

(FTIR) was performed with a Bruker Tensor 27 over a range of 400-4000 cm-1 with a

15

resolution of 2 cm-1. The specific surface area and porosity size were obtained using

16

Brunauer–Emmett–Teller (BET) analysis performed with a BELSORP-mini II (Japan).

17

Finally, the thermal stability of 3DG and Cu-xCu2O NPs@3DG nanocomposite foam

18

was evaluated using Thermogravimetric analysis (TGA, STA BAHR 503). The

19

heating program was performed in a temperature range from 25 to 600 ⁰C at a

20

heating rate of 20 ⁰C.min-1 in the Ar atmosphere. 6

1

2.4. Electrochemical evaluation of 3DG-based foam

2

Before electerochemical evluation, the glassy carbon electrode (GCE) was first polished

3

with 0.05 µm alumina slurries and washed with DI water. After dropping 10 µL of 3DG and

4

Cu-xCu2O NPs@3DG suspensions (5 mg/ml in DI water) on the GCE, they were air-dried.

5

Cyclic voltammetry (CV) (potential window=-0.2-0.8 V and scan rate=100 mV/s),

6

differential pulse voltammetry (DPV) (potential window= 0-1.2 V, step potential=0.003 V,

7

module amplitude=0.2 V, module time=0.05 s, and interval time=0.5 s) and electrochemical

8

impedance spectroscopy (EIS) (Frequency window= 0.01-100000 Hz) were run on an

9

Autolab (204 SERIES) electrochemical work station. Electrochemical measurements were

10

performed in NaOH solution with pH= 13 using a conventional three-electrode system at

11

room temperature. While the GCE modified with 3DG-based foam was used as the working

12

electrode, a saturated calomel electrode and platinum sheet were applied as reference and

13

counter electrodes, respectivly. For evaluation of glucose sensing behaviour of modified

14

electrode, 10 µl glucose stock solution with concentration of 0.5 M was added to

15

electrochemical cell each time and the electrochemical response of modified electrode was

16

determined.

17

3. Results and discussion

18

3.1. Characterization of 3DG-based foam

19

Before development of Cu-xCu2O NPs@3DG foam, 3DG foam was fabricated by a

20

simple template-guided method using PS microparticles (Fig. 1). According to the SEM

21

image and particle size distribution (Supplementary Fig. S2(a and b)), PS particles with

22

spherical-like morphology and an average diameter of 2.3±0.4 µm were synthesized. During

23

the 3DG foam fabrication process under the pH values of 6-8, PS and GO nanosheets were

24

positively and negatively charged, respectively, leading to electrostatic interaction between

25

them [28]. According to the SEM image (Supplementary Fig. S2(c)), in this situation, PS

26

particles were wrapped using GO nanosheets. Moreover, PS particles were distributed

27

uniformly between GO layers. Consequently, 3DG foam with uniform pore size distribution

28

(1.8±0.2 µm) was formed by removing PS particles by exposure to toluene (Supplementary

29

Fig. S2(d)). These pores were detected in the whole structure, providing 3DG foam with

30

interconnected porosity. This morphology may be helpful in promoting electrochemical

31

responses of 3DG foam. Similarly, Choi et al. [29] used a similar procedure to synthesize 3D

32

graphene frameworks with interconnected pores with an average size of 2 µm and showed

33

that an open porous structure with no collapse was developed after removal of PS particles. 7

1

X-ray diffraction (XRD) patterns of primary GO nanosheets and 3DG are presented in

2

Fig. 2(a). The results confirmed that the intensive peak of GO nanosheets related to (001)

3

lattice plane at 2θ = 11.7 ° vanished in the XRD pattern of the 3DG sample, confirming the

4

reduction of GO during foam preparation. The XRD pattern of 3DG consisted of a broad

5

peak at 2θ=21⁰,which was similar to reports in previous research on 3DG foam [22].

6

Moreover, the FTIR spectra of GO and 3DG foam (Fig. 2(b)) also confirmed reduction of GO

7

nanosheets during the foaming process using hydrazine hydrate solution. The FTIR spectrum

8

of GO consisted of peaks centered at about 3400 cm−1 and 1430 cm−1 (O-H groups), 1730 cm-

9

1

(C=O stretching vibration), 1630 cm-1 (the stretching mode of C=C), 1168 (C-OH

10

stretching) and 1050 cm-1 (C-O stretching) [69, 70]. After reduction of GO using hydrazine

11

hydrate solution and formation of 3DG foam, the intensity of the O-H band was considerably

12

reduced, and other peaks relating to the oxygen functional group (C=O, C-OH and C-O-C)

13

were removed, confirming the reduction of GO in the presence of hydrazine hydrate solution.

14

Reduction of GO using hydrazine hydrate solution during 3DG foam formation was also

15

demonstrated by Raman spectroscopy (Fig. 2(c)). Generally, the Raman spectrum of

16

graphene consists of two main peaks: G mode, attributed to the first-order scattering of the

17

E2g phonon of sp2 carbon atoms; and D mode, related to the breathing mode of k-point

18

photons of A1g symmetry [71]. According to Fig. 2(c), the Raman spectrum of GO consisted

19

of two characteristic peaks at 1590 cm-1 (G peak), representing the symmetry and

20

crystallizability of carbon materials, and 1280 cm-1 (D peak), which was a defect peak and

21

characterized the surface defects and disorder of the graphitic layers. The intensity ratio of

22

ID/IG was used to identify the degree of disorder of carbon materials. The ID/IG ratio of

23

graphene in the 3DG structure was 0.95, which was slightly higher than ID/IG ratio of GO

24

(0.92). This could be due to the presence of more defects and a decrease in the average size of

25

the sp2 domains upon the reduction of GO to rGO and the restoration of C=C bonds after

26

reduction of GO and formation of the 3DG structure. In addition, this may be related to the

27

reduction process, in which oxygen groups are wiped off and graphene is restored to its

28

original layer structure [30, 72, 73]. Tung et al. [73] synthesized 3DG by a self-assembly

29

procedure and reported a slight increase in the ID/IG ratio compared to that of GO. Tu et al.

30

[74] synthesized 3DG by a chemical vapor deposition method and reported a strong D peak

31

for 3DG related to the disorder that was due to edges and the 3D structure of 3DG.

8

1 2

Figure 2. (a) XRD pattern, (b) FTIR, and (c) Raman spectra of GO and 3DG.

3

After optimizaton of 3DG foam, a number of Cu-xCu2O NPs@3DG foams were

4

synthesized based on a simple process (Fig. 1). The XRD patterns of Cu-xCu2O NPs@3DG

5

foams (Fig. 3(a)) revealed that a broad peak related to (001) lattice plane of GO at 2θ = 11.7 °

6

had disappeared, and new peaks relating to Cu-based components were detected. Depending

7

on the concentration of Cu precursor solution, these characteristic peaks were different. For

8

instance, for 3DGC1, 3DGC2, and 3DGC3 foams, in which the concentration of CuCl2.2H2O

9

solution was less than 0.2 M, Cu2O and Cu were synthesized. However, while Cu2O was

10

identified as the main phase for 3DGC1 foam, the small characteristic peak of Cu located at

11

2θ=43.31⁰ corresponding to (111) lattice plane was detected. In other words, although the

12

main peak of 3DGC3 was Cu, the small characteristic peaks of Cu2O were detected at

13

2θ=36.49⁰ and 42.40⁰ relating to (111) and (200) lattice planes, respectively. Moreover, the

14

XRD pattern of 3DGC2 foam consisted of Cu as the main phase and Cu2O as the secondary

15

phase, owing to the presence of the characteristic peaks at 2θ=36.49⁰, 42.40⁰, 61.52⁰, and

16

73.69⁰. Furthermore, increasing the concentration of Cu precursor solution more than 0.2 M

17

(in 3DGC4, 3DGC5, and 3DGC6 samples) resulted in the synthesis of pure Cu components.

18

The crystallite size of the Cu-based components was also estimated from XRD patterns. The

19

results showed that there was no significant difference between the crystallite size of the Cu9

1

based components, confirming that the crystallite size was independent from the

2

concentration of Cu precursor solution. The crystallite size of 3DGC1, 3DGC2, 3DGC3,

3

3DGC4, 3DGC5, and 3DGC6 was estimated at around 39.8±10 nm, 26.8±7 nm, 34±8 nm,

4

33.4±8, 33.4±8, and 30.5±6 nm, respectively. The FTIR spectra of the Cu-Cu2O NPs@3DG

5

(3DGC1) and Cu NPs@3DG (3DGC5) foams (Fig. 3(b)) clearly confirmed the formation of

6

Cu2O and Cu, depending on the sample condition. Both FTIR spectra consisted of three

7

characteristic peaks centered at 786, 1328, and 1635 cm-1 related to Cu-O, C-O-Cu, and C-O

8

functional groups, confirming the existence of Cu [75]. However, the FTIR spectrum of Cu-

9

Cu2O NPs@3DG foam (3DGC1) showed the existence of a peak at 636 cm-1 that was the

10

characteristic peak of Cu2O, confirming decoration of 3DG with Cu2O in addition to Cu [76,

11

77]. Kumar et al. [78] synthesized a Cu2O-rGO nanocomposite and reported the presence of

12

an absorption band at 620 cm-1 of the FTIR spectrum of a nanocomposite attributed to Cu2O.

13 14

Figure 3. (a) XRD pattern of Cu based NPs@3DG foams. (b) FTIR spectra of 3DGC1 and 3DGC5.

15

Cu NPs@3DG and Cu-Cu2O NPs@3DG foams were formed by a redox reaction between

16

3DG (consisting of rGO-COO- surface layers) and Cu2+ ions. During soaking of the 3DG

17

foams in a Cu precursor solution, Cu2+ ions completely covered the rGO foam surface 10

1

through electrostatic interaction with –COOH groups. First, a ligand exchange absorbed Cu2+

2

ions onto the rGO foam surface to form Cu2+-rGO. The Cu-based NPs-rGO foam was

3

obtained by a transformation process of calcination at 95 °C. Choi et al. synthesized

4

Co2O3/rGO film using the same precedure and reported the same trends, including

5

incorporation of ions on the 3DG film, formation of Co(OH)2/rGO film with NH3OH

6

solution, and a calcination process at 450 °C. At high concertrations of Cu precursor

7

(3DGC4, 3DGC5, and 3DGC6 foams), accumulation of high concentrations of Cu2+ ions on

8

the surface of the 3DG foams led to incomplete calcination of Cu2+-rGO due to low amounts

9

of available oxygen gas. Therfore, Cu was detected as the main component on the 3DG

10

foams. In other words, at low concentrations of Cu precursor (3DGC1, 3DGC2, and 3DGC3),

11

a complete calcination process occurred, resulting in the formation of Cu2O-Cu nanoparticles

12

instead of Cu nanoparticles.

13

SEM images of hybrid Cu-xCu2O NPs@3DG foams consisting of various amounts of Cu-

14

based components (Fig. 4) clearly showed the distribution of nanoparticles within the 3DG

15

foams. However, the particle size, concentration and distribution of nanoparticles on the

16

surface of the 3DG foams were strictly dependent on Cu precursor concentrations. In 3DGC1

17

foam, agglomerated Cu-Cu2O nanoparticles with an approximate size distribution of

18

260±0.04 nm were detected on or between porous 3DG foam with a pore size of 2.2±0.03

19

µm. Therefore, it can be concluded that no collapse occurred during the decoration of 3DG

20

foam with Cu-Cu2O nanoparticles. According to higher magnification of SEM image and

21

EDS-mapping (Fig. S3(a)), spherical-like Cu-based components were distributed throughout

22

the 3DG foam. Moreover, the EDS spectrum (Fig. S3(a)) indicated that 3DGC1 foam was

23

composed of Cu, and C and O elements without any impurity of precursors. Increasing the

24

concentration of the Cu precursor resulted in the agglomeration and formation of high-density

25

Cu-based components on the 3DG foam with no collapse of the porous three-dimensional

26

structure. Notably, the SEM image of the 3DGC5 sample clearly revealed that Cu-based

27

components with an agglomeration size of 587±100 nm were distributed on a 3DG surface

28

with a pore size of 2.0±0.2 µm. Due to high concentration of Cu precursors, the entire surface

29

of the 3DG foam was covered with Cu-based components. This behavior could be clearly

30

observed in the EDS-mapping image of 3DGC5 foam (Fig. S3(b)). Moreover, the EDS

31

mapping of 3DGC5 foam showed a uniform distribution of Cu nanoparticles with less O

32

element in this foam. Furthermore, whenthe concentration of Cu precursors was increased,

33

the average pore size of the 3DG foam was estimated to be about 2.1±0.1, 2.2±0.1, 2.3±0.2, 11

1

and 2.0±0.3 µm for 3DGC2, 3DGC3, 3DGC4, and 3DGC6 foam, respectively. These results

2

clearly confirmed that the pore size of the 3DG foam did not significantly change after

3

deposition of Cu-based components. In addition, different concentrations of Cu precursor in

4

solution led to the formation of agglomerated nanoparticles of various sizes. The SEM

5

images revealed that the average sizes of the agglomerations for the 3DGC1,3DGC2,

6

3DGC3, 3DGC4, 3DGC5, and 3DGC6 samples were 260±38 nm, 303±29 nm, 446±54 nm,

7

587±100 nm, 582±94 nm, and 587±0.103 nm, respectively.

8 9 10

Figure 4. SEM images of Cu-based NPs@3DG foams: (a) 3DGC1, (b) 3DGC2, (c) 3DGC3, (d) 3DGC4, (e) 3DGC5, (f) 3DGC6 samples.

11

The TEM images of the 3DG and 3DGC1 samples (Fig. 5) clearly revealed the 3D

12

structure of the graphene and the distribution of nanoparticles within the 3DG foams. The

13

TEM images of the 3DG foam showed that in addition to a microporous framework with a 12

1

pore size of about 1.8-2 µm (supplementary Fig. S2(d)), mesopores with a size of 200-250

2

nm were developed in the entire structure. These mesopores could be attributed to the PS

3

nanoparticles removed during immersion in tolouen. However, a few PS nanoparticles with a

4

size of less than 60 nm could be identified in the TEM images (Fig. 5(b and c), which could

5

not be removed from the 3DG foams due to being trapped between 3DG layers of high

6

thickness. After incorporation of Cu-based components (for instance, the 3DGC1 sample),

7

the 3DG layers were decorated with nanoparticles with a size of 30-40 nm without any

8

deterioration of micro- and mesopores of 3DG (Fig. 5(d)). Cu-based components grew

9

around the pores without filling them (Fig. 5(c)), confirming the SEM images.

10 11

Figure 5. TEM images of (a-c) 3DG and (d-f) 3DGC1 sample, at various magnifications.

12

In order to confirm the SEM and TEM images, the specific surface area and pore size of

13

the foams was estimated using the BET method based on nitrogen absorption (Fig. S4). The

14

specific surface area of 3DG and 3DGC1 foams was 11.7 and 6.8 m2/g, respectively. The

15

specific surface area of a single layer of graphene is around 2600 m2/g. Therefore, the

16

specific surface area of graphene stacks should be 2600/n m2/g, where n is the number of

17

layers in the stacks [79]. Since the thickness of the 3DG foam was determined to be around

18

300 nm by the TEM image, the number of layers in the 3DG foam was 600. Therefore, the

19

surface area of graphene stacks with the same number of layers should be about 4.3 m2/g.

20

Consequently, development of the 3DG foam resulted in an enhanced surface area of

21

graphene, with an increase from 4.3 to 11.7 m2/g. After incorporation of Cu-Cu2O

22

nanoparticles into the 3DG foam (the 3DGC1 sample), the specific surface area was reduced

23

to 6.8 m2/g, although that was still higher than the graphene stacks. Zhang et al. [80] similarly 13

1

synthesized 3D porous graphene decorated with MoO2 nanoparticles and reported a reduction

2

in the surface area with decorated nanoparticles on the 3DG. The reduced surface area after

3

decoration with Cu-Cu2O nanoparticles might be due to the high number of nanoparticles that

4

accumulated around the 3DG mesopores. Furthermore, the mean diameter of the mesopores

5

calculated by BET analysis for 3DG and 3DGC1 was 30 and 29 nm, respectively, confirming

6

that the decoration of 3DG with Cu-based nanoparticles did not have a negative effect on

7

pore size.

8

The thermal stability of 3DG and 3DGC1 foam was investigated by TGA analysis (Fig.

9

S5). The TGA curve of GO shows three weight loss step when exposed to high temperatures.

10

The first weight loss (0.2%) at around 100 ⁰C is ascribed to the removal of water content in

11

3DG (three dimensional structure of rGO) after the reduction of GO. Whereas the second

12

(14.9%) from 150 ºC to 400 ºC is related to decomposition of non-reduced hydroxyl and

13

carboxyl function groups. And finally, weigh loss up to 600 °C (77%) can be attributed to the

14

loss of the remained carbonyl groups. The remaining weight percent (7.9%) is ascribed to

15

carbon skeleton [81-83]. In comparison, 3DGC1 (Cu/Cu2O NPs@3DG) shows a total mass

16

loss of 33.2% when the temperature reaches 600 °C indicating higher thermal stability than

17

3DG. According to the mass loss in 3DGC1, the mass percentages of Cu/Cu2O NPs in

18

3DGC1 is calculated to be 66.8%. Therefore, the incorporation of the Cu/Cu2O NPs on the

19

3DG leads to more excellent thermal stabilization than 3DG.

20

3.2. Electrochemical evaluation of Cu-xCu2O NPs@3DG foams

21

The electrochemical properties of 3DG, Cu-Cu2O NPs@3DG (3DGC1), and Cu

22

NPs@3DG (3DGC5) were evaluated by CV assay. In the absence of glucose (Fig. 6(a)), no

23

obvious redox peaks were detected in the working potential ranges for all samples. Therefore,

24

there was no electrochemical activity toward glucose detection. After addition of 8 mM

25

glucose (Fig. 6(b)), there were no obvious redox peaks in the CV curves for 3DG and GCE,

26

confirming that these electrodes could not electrochemically detect glucose. Moreover, there

27

was an increase in anodic current in the CV curve of the 3DG foam compared to that of GCE

28

because the highly porous interconnected network of 3DG provided more active sites for

29

glucose detection [62]. After incorporation of Cu-based components into 3DG foams

30

(3DGC1 and 3DGC5), a pair of redox peaks were observed, corresponding to oxidation of

31

glucose. Therefore, Cu-Cu2O NPs@3DG and Cu NPs@3DG foams revealed electrochemical

32

activity toward sensing glucose.

14

1 2 3 4

Figure 6. CVs of GCE, 3DG, Cu/Cu2O NPs@3DG (3DGC1), and Cu NPs@3DG (3DGC5) in NaOH solution with a scan rate of 100 mV/s: (a) absence, (b) presence of 8 mM glucose. Nyquist plot of (c) GCE, 3DG, 3DGC1, and 3DGC5 and (d) Cu-based NPs@3DG foams in NaOH solution (insert: magnified Nyquist plot at high frequencies).

5

The electrochemical behavior of Cu-xCu2O NPs@3DG foams on a GCE electrode was

6

assessed using CV, DPV and EIS assays. The impact of Cu-xCu2O nanoparticles on the 15

1

ability of the electrode for electron transfer between electrolyte and the electrode surface was

2

primarily investigated using EIS assay. Generally, Nyquist curves of EIS assay are divided

3

into two regions: a semicircular region at high frequencies corresponding to charge-transfer-

4

limited process which its diameter is equal to charge transfer resistance; and linear region at

5

low frequencies related to a diffusion-limited process. Nyquist plots for samples in NaOH

6

solution (Fig. 6(c, d)) consists of a nearly straight line without an obvious semicircle at high

7

frequencies. However, high magnification of a Nyquist plot at high frequencies (inset in Fig.

8

6(c)) showed that the diameter of the semicircular region of curves after deposition of

9

nanoparticles was significantly lower than those for 3DG and GCE, implying higher electron

10

transfer ability of the 3DG foam in the presence of nanoparticles than 3DG and GCE. This

11

observation was due to the presence of Cu-based nanoparticles, which promote electron

12

transfer between the electrode and the solution. The results showed the promising

13

electrochemical properties of these components, making them suitable for sensing

14

applications. This phenomenon may be due to the very large interface between nanoparticles

15

and 3DG in the 3DG foams, which has also been reported in other research [67]. The effect

16

of Cu-xCu2O nanoparticle concentrations on the electrochemical properties of Cu-xCu2O

17

NPs@3DG foams was estimated using EIS assay (Fig. 6(d)). The Nyquist curves of 3DGC1

18

and 3DGC5 showed lower diameters in the semicircular regions than the other samples,

19

confirming that 3DGC5 and 3DGC1 had better ability of electron transfer and might be more

20

suitable for glucose detection.

21

The electrocatalytic oxidation mechanism of glucose on the Cu or Cu-Cu2O nanoparticles

22

in NaOH solution is a multi-step process. First, Cu and Cu2O is oxidized to Cu(II) species,

23

including CuO and Cu(OH)2 (equations 1-3). Subsequently, Cu(II) species are oxidized to

24

strong oxidizing agents of Cu(III) species, including CuOOH and Cu OH

25

Then, glucose is catalytically oxidized by Cu(III) species to produce gluconic acid (equation

26

7,8), while Cu(III) compounds return to Cu(II) compounds:

4

(equations 4-6).

Cu+2OH- → CuO+H2 O+2e-

(1)

Cu+2OH- → Cu(OH)2 +H2 O+2e-

(2)

Cu2 O+2OH- +H2 O → 2Cu(OH)2 +2e-

(3)

Cu(OH)2 +OH- → CuOOH+H2 O+e-

(4)

CuO+OH- → CuOOH+e-

(5) 16

CuO+2OH- +H2 O → Cu(OH)-4 +2e-

(6)

Cu III +glucose → Cu II +gluconolactone

(7)

hydrolyzation

(8)

Gluconolactone

Gluconic acid

1

Therefore, Cu(III) acts as and electro-transformative material, and non-enzymatic

2

electrochemical detection of glucose is performed by the Cu(III)/Cu(II) redox couple [60, 67,

3

84, 85]. Moreover, detection of glucose is dependent on the number of Cu and Cu-Cu2O

4

active sites in the foam.

5

Cu-xCu2O nanoparticle concentration also played a significant role in glucose detection.

6

The CV curves of the Cu-Cu2O NPs@3DGC foams (3DGC1, 3DGC2, and 3DGC3) and the

7

Cu NPs@3DG foams (3DGC4, 3DGC5, and 3DGC6) are presented in Fig. 7. Among the CV

8

curves of the Cu-Cu2O NPs@3DGC foams (Fig. 7(a)), 3DGC1 was more promising for

9

detecting glucose, due to presence of redox peaks at lower potentials and sharper peaks than

10

3DGC2 and 3DGC3, indicating that 3DGC1 could reduce the over-potential in glucose

11

oxidation. In addition, 3DGC2 revealed lower currents and less ability for electron transfer.

12

In other words, among the Cu NPs@3DG foams, 3DGC5 revealed superior glucose detection

13

ability, having a redox peak at higher currents and lower potentials, indicating high

14

electrocatalytic activity. Therefore, the 3DGC1 and 3DGC5 foams were better

15

electrocatalysts for glucose oxidation and, hence, glucose detection. Ma et al. [68]

16

synthesized 3DG/CuO foam for ascorbic acid detection and observed the same results. They

17

reported that the 3DG/CuO foam showed higher electrocatalytic activity than the 3DG foam.

18

Moreover, the ascorbic acid oxidation peak of the 3DG/CuO foam revealed a negative shift

19

compared to the 3DG foam, representing the ability of the 3DG/CuO foam to reduce over-

20

potential of ascorbic acid oxidation [68].

17

1 2 3 4

Figure 7. CVs of Cu-based NPs@3DG foams in NaOH solution with a scan rate of 100 mV/s: (a) Cu-Cu2O NPs@3DG (3DGC1, 3DGC2, and 3DGC3), (b) Cu NPs@3DG (3DGC4, 3DGC5, 3DGC6).

3.3. Electrochemical activity of Cu-xCu2O NPs@3DG foams toward glucose detection

5

In order to estimate the best electrodes for glucose detection, the DPV technique was

6

applied in order to investigation and selection the optimum modified electrode for glucose

7

detection owning to the small contribution of charging current to the background. Since the

8

normal range of glucose in the human blood is 4-7 mM, the modified electrode response was

9

monitored by using DPV technique in the range of 3-8 mM glucose in NaOH solution (Fig.

10

8). It was observed that the oxidation peak current of glucose for all samples had a linear

11

relationship with glucose concentration (Inset in Fig. 8). Also the results showed the

12

oxidation peak of glucose appeared at higher potential in DPV voltammogram in compare

13

with the oxidation peak of glucose in CV voltammogram. DPV is highly sensitive technique

14

that is related to non-faradic process and the charging current is minimized whereas in CV

15

technique, both faradic and non-faradic process takes place. This result was reported by other

16

researchers too [86, 87]. The calculated sensitivity of these samples (3DGC1, 3DGC5, 3DG,

17

and GCE) is presented in Table 2. The results confirmed that all the 3DG-based electrodes

18

revealed better sensitivity than GCE and were more suitable for glucose detection. This may

19

be due to the presence of carboxyl acid functional groups of GO, which could be covalently

20

attached to glucose, leading to improvement of glucose detection [88]. Similar results have

21

been reported in previous research [72]. Moreover, among the various Cu-xCu2O NPs@3DG

22

foams, 3DGC1 revealed the highest sensitivity (6.1254 µA.mM-1), owning to the presence of

23

Cu2O nanoparticles within the structure. Similarly, Wang et al.[57] compared the

24

electrocatalytic ability of Cu2O and Cu nanoparticles toward glucose sensing and reported

25

that Cu2O nanoparticles had higher ability to detect glucose [57]. Moreover, according to the 18

1

previous results, 3DGC1 had fewer nanoparticles than 3DGC5. Therefore, 3DG could play a

2

significant role in electron and ion transfer, leading to more sensitivity toward glucose

3

sensing due to the porous interconnected structure of 3DG. In addition, there was an extra

4

peak at 0.2-0.4 V for 3DGC1 and 3DGC5 electrode, compared to the GCE and 3DG samples,

5

which corresponds to Cu nanoparticles [89, 90].

19

1 2 3

Figure 8. DPV curves of (a) GCE, (b) 3DG, (c) 3DGC1, and (d) 3DGC5 in NaOH solution containing 3 mM to 8 mM glucose (insert: calibration curve).

20

1 2

Table 2. The calculated sensitivity (µA.mM-1) of GCE, 3DG, 3DGC1, and 3DGC3 from the calibration curve of DPV results in NaOH solution containing 3 mM to 8 mM glucose

Sample GCE 3DG 3DGC1 3DGC5

Sensitivity (µA.mM-1) 2.2187 6.2017 7.1292 6.1254

3 4

Fig. 9 presents the DPV curve of 3DGC1 in NaOH solutions with different concentrations

5

of glucose and the calibration curves of the oxidation peak current. The peak current of

6

3DGC1 was linear for glucose concentration in the range of 0.8-10 mM. The linear equation

7

was Ip(µA)=230.86C(mM)-149.86, where Ip is the peak current and C is the glucose

8

concentration with good correlation (R2=0.9951). According to this equation, sensitivity was

9

calculated to be about 230.86 µA.mM-1.cm-2. The detection limit (DL) of this electrode was

10

16 µM, determined using DL=3Sb/S, where Sb is the standard deviation of three-time blank

11

measurements and S is the sensitivity calculated from calibration curve [72]. Moreover, the

12

relative standard deviation (RSD) was calculated to be 1.02%, indicating good repeatability

13

for the sensor produced in this study. The stability of 3DGC1 modified electrode was tested

14

by monitoring 8 mM glucose current response through DPV technique in NaOH solution.

15

The current response remained 92.2% and 95.1% of the initial value after 6000 seconds and

16

20 days respectively, which indicated excellent stability. These results indicated that the

17

3DGC1 modified electrode had good electron transfer for oxidation of glucose due to the

18

porous interconnected structure of 3DG, providing active sites for glucose detection and

19

increased electron transfer. Furthermore, 3DG was decorated with electrocatalyst Cu-Cu2O

20

NPs, which featured oxidization of glucose at the active sites by the Cu(III)/Cu(II ) redox

21

couple. Table 3 shows a comparison between some previous reports for glucose detection and

22

the results of the current study [43, 56, 67, 91-94]. Among the various samples, the sensitivity

23

and linear range of the electrode in the current study was very high; it could detect glucose up

24

10 mM. Moreover, the detection limit of this electrode was acceptable, compared to the

25

results of other research. Therefore, this electrode showed satisfactory performance in

26

glucose detection.

21

1 2 3

Figure 9. (a and b) DPV and calibration curves of 3DGC1 in NaOH solution containing different concentrations of glucose (a-w: 0, 0.4, 0.6, 0.8, 1, 1.2, 1.4, 1.6, 1.8, 2, 2.4, 2.8, 3.2, 3.6, 4, 5, 6, 7, 8, 9, 10 mM) respectively.

4

Table 3. Performance comparison of various glucose sensors.

Cu-Cu2O NPs@3DG

Sensitivity (µA.mM-1.cm-2) 230.89

Linear range (mM) 0.8-10

Detection limit (µM) 16.0

Cu NPs@graphene

48.13

4.50

1.3

[91]

Cu2O nanocubes@graphene

285

0.3-3.3

3.3

[43]

Cu/Cu2O@rGO

145.2

0.005-7

0.5

[92]

Cu/Cu2O nanoporous NPs

123.8

0.01-5.5

0.05

[56]

Cu2O/rGO

185

0.01-6

0.05

[93]

3DG thin film

1.63

0.3-6

0.2

[94]

Cu2O NPs@3DG

2310

1.8

0.14

[67]

Electrode

Reference This work

5 6

3.4. Selectivity study of Cu-xCu2O NPs@3DG foam

7

Anti-reference capability was also investigated for selectivity evaluation of the 3DGC1

8

electrode toward glucose detection. The normal level of glucose in the blood is at least ten 22

1

times higher than interfering species in human blood, such as ascorbic acid, acetaminophen,

2

urea, and dopamine [7, 95]. Fig. 10(a) presents the DPV results for the modified electrode

3

upon injection of 2 mM glucose and 0.2 mM of interfering species (ascorbic acid, urea,

4

acetaminophen, and dopamine). There was no oxidation peak in the potential range of the

5

glucose oxidation peak. Moreover, the response of the modified electrode after the addition

6

of glucose was significantly higher than its response after the addition of interfering species.

7

In addition, the oxidation glucose peak shifted to a higher potential, owing to saturation of the

8

active sites of the electrode surface with interfering species, which reduced the rate of the

9

glucose oxidation reaction. Therefore, other interfering species in the human blood did not

10

interfere with the detection of glucose, and the sensor made from the 3DGC1 electrode could

11

be applied to analysis of real samples. Also the effect of other interfering species including

12

fructose, betalactose, maltose, xylose, sucrose and GSH was investigated (Fig. 10(b)). The

13

current response of 6 mM Glucose in NaOH solution at presence and absence of these species

14

was similar (23.4 and 24.7 µA respectively). So, these species have no oxidation peak and

15

interference in potential range of glucose oxidation.

16 17 18 19

Figure 10. DPV curve of 3DGC1 in NaOH solution toward (a) 2 mM glucose and 0.2mM interference species (including ascorbic acid, urea, dopamine, and acetaminophen, (b) 6 mM glucose with and without interfering species including fructose, betalactose, maltose, xylose, sucrose and GSH

23

1

3.5. Glucose detection in real human serum sample

2

The performance of 3DGC1 electrode for glucose sensing in human serum was tested

3

with standard addition method [96-100]. 0.1 ml of human serum was injected into 4.9 ml of

4

NaOH solution with pH= 13 and the standard addition of glucose was done for evaluation of

5

glucose sensing with this electrode. The recovery is defined as following equation [101]:

6

recovery%=(Cfound -Creal /Cadded )×100

7

(9)

8

Where Cfound is the concentration of analyte was detected with electrode, Creal is the

9

concentration of glucose in real sample and Cadded is the concentration of glucose was spike

10

in the real sample. The results are presentation in Table 4. This table shows good recovery

11

with average of about 94% for glucose detection in real sample. Also there is good agreement

12

between results of glucose detection in serum sample with the real values of glucose in

13

sample. Therefore, this electrode has potential of using for glucose detection in real serum

14

sample.

15

Table 4. Results of analysis of glucose spiked in human serum sample.

Spiked (mM)

Found (mM)

Recovery %

Uncertainly %

3

2.7

90

-10

4

3.6

89

-11

5

5.2

104

4

16 17

Also, In order to investigate the feasibility of 3DGC1 electrode in practical analysis, the

18

glucose concentration of four human blood serum samples obtained from a hospital was

19

calculated with the proposed detection strategy in this work. For this purpose, the serum

20

samples were diluted with NaOH solution before measurements, and the obtained results are

21

listed in Table 5. The results show that the glucose concentration measured by the 3DGC1

22

electrode is consistent with that measured with a hospital, indicating excellent reliability and

23

accuracy of the suggested sensor in real sample analysis.

24

Table 5. Results of glucose determination in human serum samples.

Serum sample 1

Concentration measured by glucose sensor of this work (mM) 4.52

RSD% (n=3) 1.49

Concentration measured by hospital (mM) 4.61

2

6.80

1.20

6.94

97.96

3

7.79

0.66

7.99

97.45

24

Recovery% 98.13

4

11.47

0.92

11.93

96.13

1 2

4. Conclusion

3

In the current study, a novel non-enzymatic glucose sensor was developed by decorating

4

porous three-dimensional graphene (3DG) foam with Cu-Cu2O nanoparticles (Cu-xCu2O

5

NPs@3DG foam) without any linker. 3DG foam was fabricated using a simple, low-cost

6

polystyrene colloid template. The 3DG foam was subsequently decorated with Cu-xCu2O

7

nanoparticles using an efficient immersion technique. The results demonstrated uniform

8

dispersion of Cu-Cu2O NPs with a diameter of 30-40 nm and an agglomeration diameter of

9

about 250 nm on the 3DG surface with micro- and meso-porosity. The GCE-modified Cu-

10

xCu2O NPs@3DG foam displayed high electrocatalytic activity in glucose oxidation and

11

good conductivity. While Cu-Cu2O NPs played an effective role as active sites for glucose

12

oxidation, the interconnected porous structure of the 3DG foam promoted electron and ion

13

transfer. Furthermore, this electrode exhibited excellent performance in glucose detection

14

(wide linear range, low detection limit, high sensitivity, good repeatability and appropriate

15

selectivity and recovery). Therefore, this foam is an inexpensive, feasible, and appropriate

16

candidate for glucose sensors.

17

Acknowledgements

18 19

The authors would like to thanks Dr. Yalda Shoja for her many beneficial comments and suggestions.

25

References [1] E. Zhang, Y. Xie, S. Ci, J. Jia, Z. Wen, Biosensors and Bioelectronics, 81 (2016) 4653. [2] C.-L. Hsu, J.-H. Lin, D.-X. Hsu, S.-H. Wang, S.-Y. Lin, T.-J. Hsueh, Sensors and Actuators B: Chemical, 238 (2017) 150-159. [3] C.-W. Hsu, F.-C. Su, P.-Y. Peng, H.-T. Young, S. Liao, G.-J. Wang, Sensors and Actuators B: Chemical, 230 (2016) 559-565. [4] S. Alexander, P. Baraneedharan, S. Balasubrahmanyan, S. Ramaprabhu, Materials Science and Engineering: C, 78 (2017) 124-129. [5] H.-B. Wang, Y. Chen, N. Li, Y.-M. Liu, Microchimica Acta, 184 (2017) 515-523. [6] B.D. Malhotra, A. Turner, Advances in biosensors: Perspectives in biosensors, Elsevier, 2003. [7] A.A. Ensafi, N. Ahmadi, B. Rezaei, Sensors and Actuators B: Chemical, 239 (2017) 807-815. [8] S. Liu, B. Yu, T. Zhang, Electrochimica Acta, 102 (2013) 104-107. [9] X. Zhuang, D. Chen, S. Zhang, F. Luan, L. Chen, Microchimica Acta, 185 (2018) 166. [10] N. Zhou, H. Chen, J. Li, L. Chen, Microchimica Acta, 180 (2013) 493-499. [11] D. Chen, X. Zhuang, J. Zhai, Y. Zheng, H. Lu, L. Chen, Sensors and Actuators B: Chemical, 255 (2018) 1500-1506. [12] X. Zhuang, D. Chen, S. Wang, H. Liu, L. Chen, Sensors and Actuators B: Chemical, 251 (2017) 185-191. [13] N. Zhou, J. Li, H. Chen, C. Liao, L. Chen, Analyst, 138 (2013) 1091-1097. [14] S. Liu, J. Tian, L. Wang, X. Qin, Y. Zhang, Y. Luo, A.M. Asiri, A.O. Al-Youbi, X. Sun, Catalysis Science & Technology, 2 (2012) 813-817. [15] M. Baghayeri, A. Amiri, S. Farhadi, Sensors and Actuators B: Chemical, 225 (2016) 354-362. [16] S. Liu, Z. Wang, F. Wang, B. Yu, T. Zhang, RSC Advances, 4 (2014) 33327-33331. [17] L. Yang, Z. Wang, Y. Ji, J. Wang, G. Xue, Macromolecules, 47 (2014) 1749-1756. [18] X. Dong, X. Wang, L. Wang, H. Song, H. Zhang, W. Huang, P. Chen, ACS applied materials & interfaces, 4 (2012) 3129-3133. [19] S. Zhang, X. Zhuang, D. Chen, F. Luan, T. He, C. Tian, L. Chen, Microchimica Acta, 186 (2019) 450. [20] X. Zhuang, H. Wang, T. He, L. Chen, Microchimica Acta, 183 (2016) 3177-3182. [21] C. Wu, X. Huang, G. Wang, L. Lv, G. Chen, G. Li, P. Jiang, Advanced Functional Materials, 23 (2013) 506-513. [22] K.G. Lee, J.-M. Jeong, S.J. Lee, B. Yeom, M.-K. Lee, B.G. Choi, Ultrasonics sonochemistry, 22 (2015) 422-428. [23] D.A. Brownson, L.C. Figueiredo-Filho, X. Ji, M. Gómez-Mingot, J. Iniesta, O. Fatibello-Filho, D.K. Kampouris, C.E. Banks, Journal of Materials Chemistry A, 1 (2013) 5962-5972. [24] C. Li, G. Shi, Nanoscale, 4 (2012) 5549-5563. 26

[25] L. Lu, Biosensors and Bioelectronics, 110 (2018) 180-192. [26] H. Bai, C. Li, X. Wang, G. Shi, Chemical Communications, 46 (2010) 2376-2378. [27] Y. Xu, K. Sheng, C. Li, G. Shi, ACS nano, 4 (2010) 4324-4330. [28] B.G. Choi, S.-J. Chang, Y.B. Lee, J.S. Bae, H.J. Kim, Y.S. Huh, Nanoscale, 4 (2012) 5924-5930. [29] B.G. Choi, M. Yang, W.H. Hong, J.W. Choi, Y.S. Huh, ACS nano, 6 (2012) 40204028. [30] Z. Niu, J. Chen, H.H. Hng, J. Ma, X. Chen, Advanced Materials, 24 (2012) 41444150. [31] B. Yu, D. Kuang, S. Liu, C. Liu, T. Zhang, Sensors and Actuators B: Chemical, 205 (2014) 120-126. [32] Y. Sudhakar, H. Hemant, S. Nitinkumar, P. Poornesh, M. Selvakumar, Ionics, 1-10. [33] C. Sarkar, S.K. Dolui, RSC Advances, 5 (2015) 60763-60769. [34] B. Zhang, Y. He, B. Liu, D. Tang, Microchimica Acta, 182 (2015) 625-631. [35] M. Li, X. Bo, Z. Mu, Y. Zhang, L. Guo, Sensors and Actuators B: Chemical, 192 (2014) 261-268. [36] H. Shu, G. Chang, J. Su, L. Cao, Q. Huang, Y. Zhang, T. Xia, Y. He, Sensors and Actuators B: Chemical, 220 (2015) 331-339. [37] J. Luo, S. Jiang, H. Zhang, J. Jiang, X. Liu, Analytica chimica acta, 709 (2012) 4753. [38] L. Wang, X. Lu, C. Wen, Y. Xie, L. Miao, S. Chen, H. Li, P. Li, Y. Song, Journal of Materials Chemistry A, 3 (2015) 608-616. [39] X. Zhuang, C. Tian, F. Luan, X. Wu, L. Chen, RSC Advances, 6 (2016) 9254192546. [40] Y. Ye, P. Wang, E. Dai, J. Liu, Z. Tian, C. Liang, G. Shao, Physical Chemistry Chemical Physics, 16 (2014) 8801-8807. [41] S. Yang, L. Liu, G. Wang, G. Li, D. Deng, L. Qu, Journal of Electroanalytical Chemistry, 755 (2015) 15-21. [42] Y. Tian, Y. Liu, W.-p. Wang, X. Zhang, W. Peng, Electrochimica acta, 156 (2015) 244-251. [43] M. Liu, R. Liu, W. Chen, Biosensors and Bioelectronics, 45 (2013) 206-212. [44] M. Devaraj, R. Saravanan, R. Deivasigamani, V.K. Gupta, F. Gracia, S. Jayadevan, Journal of Molecular Liquids, 221 (2016) 930-941. [45] P.M. Nia, P.M. Woi, Y. Alias, Applied Surface Science, 413 (2017) 56-65. [46] H.-B. Wang, H.-D. Zhang, Y. Chen, Y. Li, T. Gan, RSC Advances, 5 (2015) 7790677912. [47] K. Justice Babu, S. Sheet, Y.S. Lee, G. Gnana Kumar, ACS Sustainable Chemistry & Engineering, 6 (2018) 1909-1918. [48] B. Wang, J. He, F. Liu, L. Ding, Journal of Alloys and Compounds, 693 (2017) 902908. [49] J. Tian, H. Li, Z. Xing, L. Wang, Y. Luo, A.M. Asiri, A.O. Al-Youbi, X. Sun, Catalysis Science & Technology, 2 (2012) 2227-2230.

27

[50] T. Gan, Z. Wang, J. Gao, J. Sun, K. Wu, H. Wang, Y. Liu, Sensors and Actuators B: Chemical, 282 (2019) 549-558. [51] T. Gan, Z. Wang, Z. Shi, D. Zheng, J. Sun, Y. Liu, Biosensors and Bioelectronics, 112 (2018) 23-30. [52] Y. Zhang, L. Su, D. Manuzzi, H.V.E. de los Monteros, W. Jia, D. Huo, C. Hou, Y. Lei, Biosensors and Bioelectronics, 31 (2012) 426-432. [53] T.-K. Huang, K.-W. Lin, S.-P. Tung, T.-M. Cheng, I.-C. Chang, Y.-Z. Hsieh, C.-Y. Lee, H.-T. Chiu, Journal of Electroanalytical Chemistry, 636 (2009) 123-127. [54] Y. Zhao, J. Zhao, D. Ma, Y. Li, X. Hao, L. Li, C. Yu, L. Zhang, Y. Lu, Z. Wang, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 409 (2012) 105111. [55] S. Felix, P. Kollu, B.P. Raghupathy, S.K. Jeong, A.N. Grace, Journal of Chemical Sciences, 126 (2014) 25-32. [56] Y. Zhao, Y. Li, Z. He, Z. Yan, RSC Advances, 3 (2013) 2178-2181. [57] L. Wang, J. Fu, H. Hou, Y. Song, Int J Electrochem Sci, 7 (2012) 12587-12600. [58] S.N.A.M. Yazid, I.M. Isa, N. Hashim, Materials Science and Engineering: C, 68 (2016) 465-473. [59] L. Ju, G. Wu, B. Lu, X. Li, H. Wu, A. Liu, Electroanalysis, (2016). [60] B. Wang, Y. Wu, Y. Chen, B. Weng, C. Li, Sensors and Actuators B: Chemical, 238 (2017) 802-808. [61] J.S. Chung, S.H. Hur, Sensors and Actuators B: Chemical, 223 (2016) 76-82. [62] N.S. Nguyen, G. Das, H.H. Yoon, Biosensors and Bioelectronics, 77 (2016) 372377. [63] S. Kogularasu, M. Govindasamy, S.-M. Chen, M. Akilarasan, V. Mani, Sensors and Actuators B: Chemical, 253 (2017) 773-783. [64] A. Hatamie, R. Rahmati, E. Rezvani, S. Angizi, A. Simchi, ACS Applied Nano Materials, 2 (2019) 2212-2221. [65] D. Yin, X. Bo, J. Liu, L. Guo, Analytica chimica acta, 1038 (2018) 11-20. [66] Y. Du, Y. He, Z. Zheng, X. Shen, Y. Zhou, T. Wang, Z. Zhu, C. Wang, Journal of The Electrochemical Society, 166 (2019) B42-B48. [67] M. Wang, X. Song, B. Song, J. Liu, C. Hu, D. Wei, C.-P. Wong, Sensors and Actuators B: Chemical, 250 (2017) 333-341. [68] Y. Ma, M. Zhao, B. Cai, W. Wang, Z. Ye, J. Huang, Biosensors and Bioelectronics, 59 (2014) 384-388. [69] R. Ortega-Amaya, Y. Matsumoto, M. Pérez-Guzmán, M. Ortega-López, Journal of Nanoparticle Research, 17 (2015) 397. [70] W. Zhang, X. Li, Z. Yang, X. Tang, Y. Ma, M. Li, N. Hu, H. Wei, Y. Zhang, Nanotechnology, 27 (2016) 265703. [71] L. Malard, M. Pimenta, G. Dresselhaus, M. Dresselhaus, Physics Reports, 473 (2009) 51-87. [72] S. Darvishi, M. Souissi, F. Karimzadeh, M. Kharaziha, R. Sahara, S. Ahadian, Electrochimica Acta, 240 (2017) 388-398.

28

[73] N.B. Trung, T. Van Tam, D.K. Dang, K.F. Babu, E.J. Kim, J. Kim, W.M. Choi, Chemical Engineering Journal, 264 (2015) 603-609. [74] Z. Tu, S. Wu, F. Yang, Y. Li, L. Zhang, H. Liu, H. Ding, P. Richard, Materials Letters, 152 (2015) 264-267. [75] M. Nasrollahzadeh, M. Atarod, S.M. Sajadi, Journal of colloid and interface science, 486 (2017) 153-162. [76] H.-H. Chen, R. Anbarasan, L.-S. Kuo, M.-Y. Tsai, P.-H. Chen, K.-F. Chiang, NanoMicro Letters, 2 (2010) 101-105. [77] S. Bouson, A. Krittayavathananon, N. Phattharasupakun, P. Siwayaprahm, M. Sawangphruk, Royal Society Open Science, 4 (2017) 170654. [78] J.S. Kumar, M. Jana, P. Khanra, P. Samanta, H. Koo, N.C. Murmu, T. Kuila, Electrochimica Acta, 193 (2016) 104-115. [79] K. Sohn, Y.J. Na, H. Chang, K.-M. Roh, H.D. Jang, J. Huang, Chemical Communications, 48 (2012) 5968-5970. [80] L. Zhang, H. Lin, L. Zhai, M. Nie, J. Zhou, S. Zhuo, Journal of Materials Research, 32 (2017) 292-300. [81] M. Mokhtar, S.A. El Enein, M. Hassaan, M. Morsy, M. Khalil, International Journal of Nanoparticles and Nanotechnology, 3 (2017). [82] G. Chen, H. Guan, C. Dong, X. Xiao, Y. Wang, Journal of Physics and Chemistry of Solids, 98 (2016) 209-219. [83] R.G. Bai, K. Muthoosamy, F.N. Shipton, A. Pandikumar, P. Rameshkumar, N.M. Huang, S. Manickam, RSC Advances, 6 (2016) 36576-36587. [84] Q. Zhang, Z. Wu, C. Xu, L. Liu, W. Hu, Nanotechnology, 27 (2016) 495603. [85] C.-T. Hsieh, W.-H. Lin, Y.-F. Chen, D.-Y. Tzou, P.-Q. Chen, R.-S. Juang, Journal of the Taiwan Institute of Chemical Engineers, 71 (2017) 77-83. [86] S.N. Topkaya, Biosensors and Bioelectronics, 64 (2015) 456-461. [87] S. Xie, Y. Chai, Y. Yuan, L. Bai, R. Yuan, Biosensors and Bioelectronics, 55 (2014) 324-329. [88] Y. Yang, A.M. Asiri, Z. Tang, D. Du, Y. Lin, Materials Today, 16 (2013) 365-373. [89] P.-G. Su, L.-G. Lin, P.-H. Lin, Sensors and Actuators B: Chemical, 191 (2014) 364370. [90] A.e.N. Tufan, S. Baki, K. Güçlü, M. Özyürek, R.a. Apak, Journal of agricultural and food chemistry, 62 (2014) 7111-7117. [91] D. Jiang, Q. Liu, K. Wang, J. Qian, X. Dong, Z. Yang, X. Du, B. Qiu, Biosensors and Bioelectronics, 54 (2014) 273-278. [92] H. Huo, C. Guo, G. Li, X. Han, C. Xu, RSC Advances, 4 (2014) 20459-20465. [93] D.-L. Zhou, J.-J. Feng, L.-Y. Cai, Q.-X. Fang, J.-R. Chen, A.-J. Wang, Electrochimica Acta, 115 (2014) 103-108. [94] N. Mansouri, A.A. Babadi, S. Bagheri, S.B.A. Hamid, International Journal of Hydrogen Energy, 42 (2017) 1337-1343. [95] M. Chowdhury, F. Cummings, M. Kebede, V. Fester, Electroanalysis, 29 (2017) 578-586.

29

[96] M. Behbahani, F. Najafi, S. Bagheri, M.K. Bojdi, P.G. Hassanlou, A. Bagheri, Environmental monitoring and assessment, 186 (2014) 2609-2618. [97] C. Hou, D. Zhao, Y. Wang, S. Zhang, S. Li, Journal of Electroanalytical Chemistry, 822 (2018) 50-56. [98] M. Khan, X. Liu, J. Zhu, F. Ma, W. Hu, X. Liu, Biosensors and Bioelectronics, 108 (2018) 76-81. [99] X. Zhou, X. Dai, J. Li, Y. Long, W. Li, Y. Tu, Materials Science and Engineering: C, 49 (2015) 579-587. [100] P.M. Nia, W.P. Meng, F. Lorestani, M. Mahmoudian, Y. Alias, Sensors and Actuators B: Chemical, 209 (2015) 100-108. [101] Y. Shoja, A.A. Rafati, J. Ghodsi, Materials Science and Engineering: C, 76 (2017) 637-645.

30

Supporting figures:

Figure S1: (a) SEM and (b) TEM images and XPS survey spectrum (c) of initial GO nanosheets.

Figure S2: (a) SEM image and (b) particle size distribution of PS particles. (c) SEM images of PS-encapsulated graphene sheets and (d) 3DG foam after removal of PS spheres.

31

Figure S3: EDS results for (a) 3DGC1 and (b) 3DGC5 foams.

Figure S4: N2 adsorption–desorption isotherm and corresponding pore size distribution (inset) of (a) 3DG and (b) 3DGC1.

32

Figure S5: TGA curves of 3DG and 3DGC1 samples

33

Conflict of Interest and Authorship Conformation Form

Please check the following as appropriate: •

All authors have participated in (a) conception and design, or analysis and interpretation of the data; (b) drafting the article or revising it critically for important intellectual content; and (c) approval of the final version.

•

This manuscript has not been submitted to, nor is under review at, another journal or other publishing venue.

•

The authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript

o

The following authors have affiliations with organizations with direct or indirect financial interest in the subject matter discussed in the manuscript:

Author’s name

Affiliation

,

34

Article title: A non-enzymatic sensor based on three-dimensional graphene foam decorated with Cu-xCu2O nanoparticles for electrochemical detection of glucose and its application in human serum

Research highlights:

•

A sensitive non-enzymatic glucose electrochemical sensor was designed based on CuxCu2O NPs/3DG film

•

Characterization sensor was done by SEM, XRD and EDX

•

The performance of proposed sensor was investigated by DPV and CV techniques

•

The impact of Cu-xCu2O nanoparticles on the response of sensor was investigated

•

The sensor showed good sensitivity, high electrocatalytic efficiency, repeatability, and selectivity