**1. Introduction**

Glucose is an essential carbohydrate involved in major catabolic pathways, including oxidative phosphorylation and glycolysis for the creation of proteins, glycogens, and lipids [1,2]. Glucose is absorbed through the intestines, and, converted by the liver into a more stable form of glycogen, regulated by the hormone insulin [3,4]. Diabetes mellitus (DM) has been termed the "invisible killer" as a consequence of both hyperglycemia and hypoglycemia [5]. A fasting blood glucose concentration less

than 100 mg/dl (5.6 mmol/L) is normal, a level from 100 to 125 mg/dL (5.6 to 6.9 mmol/L) is considered prediabetes and greater than 126 mg/dL (7 mmol/L) on two separate tests allows the diagnosis of diabetes. Hypoglycemia is defined by a blood glucose concentration <70 mg/dl (3.9 mmol/L) and concentrations of both <54 mg/dL (3.0 mmol/L) and <50mg/dL (2.8 mmol/L) cause defective glucose counterregulation and impaired awareness of hypoglycemia. Hyperglycemia can result in multiple metabolic abnormalities associated with long term microvascular and macrovascular complications [6–10]. The global prevalence of diabetes in 2019 was estimated at 463 million people, and has been predicted to rise 10.2% by 2030 and 10.9% by 2045. The prevalence is higher in developed countries (10.4%) than in developing countries (4.0%). Furthermore, one in two people living with diabetes do not know that they have diabetes. The rising burden of diabetes in low- and middle-income countries may cause financial strain on individuals and health systems. Among all countries worldwide, the United States and China have the highest diabetes related medical expenditure. Between 2019 and 2045, the global expenditure for diabetes treatment is expected to grow from USD 760 billion to USD 845 billion. Diagnosis and management of diabetes require accurate, sensitive, reliable, rapid, and attentive monitoring of glucose in day to day life [11,12]. Generally, H2O<sup>2</sup> is generated during enzyme/glucose reactions and so the monitoring of H2O<sup>2</sup> is also of great importance. H2O<sup>2</sup> is an unstable compound found in nature that plays a vital role as an intermediate in several biological reactions such as the metabolism of proteins, carbohydrates, cell signaling, and immune responses [13,14]. However, excess H2O<sup>2</sup> can damage DNA or proteins via the generation of reactive oxygen species [15]. Hence, the monitoring of both H2O<sup>2</sup> and glucose with a novel sensing approach in humans and the environment is of great significance. Such non-enzymatic glucose and H2O<sup>2</sup> (NEGH) sensors have applications in biomedical devices, catalysis, and the environment.

Several analytical approaches have been reported to quantify glucose and H2O<sup>2</sup> levels, namely calorimetric, titrimetric analysis, spectrometry, fluorescence, chemiluminescence, and high-pressure liquid chromatography [16–20]. However, these methods have certain limitations, such as cumbersome fabrication processes, low reproducibility, matrix interference, high cost, and short shelf time. Hence, there is a need for the development of more efficient techniques for glucose and H2O<sup>2</sup> quantification, and, in this context, electrochemical methods have much influence. Electrochemical techniques for glucose and H2O<sup>2</sup> sensing have good accuracy, specificity, response time, simplicity, lower detection limits, high physical and chemical stability, enhanced electron transfer rate, practical detectability, easy to scale up, and biocompatibility [21]. The first enzyme-based glucose sensors were explored in 1960, and have served to drive work in this area for many researchers. Thereafter, first, second, and third generation enzyme-based glucose biosensors have been established. Third-generation sensors are still in their infancy, but those based on nano-mesoporous electrode surfaces show promise but with some drawbacks [22,23]. The mechanism of these sensors is based on the detection of oxygen or H2O2, the electron mediator, or the enzyme. Immobilized glucose oxidase (GOx) sensing results in the detection of gluconolactone and H2O<sup>2</sup> [24]. Hence, the sensing of both glucose and H2O<sup>2</sup> exists in correlation and has significance in food, pharmaceutical, clinical, and environmental studies [25,26]. However, enzymatic glucose and H2O<sup>2</sup> sensors (EGHS) have certain limitations, including enzyme denaturation due to environmental changes (pH, humidity, and temperature), digestion by proteases, expensive preparation, time-consuming purification, high cost, thermo-chemical deformation, poor reproducibility, lack of stability, and tedious enzyme immobilization techniques [27,28]. These disadvantages of EGHS, as mentioned, can be adequately defined by nanomaterial assisted electrochemical processes through NEGH sensing.

The most significant challenges faced while designing NEGH sensing are the high working potential, unpredicted redox reactions, slow electro kinetics, intermediate poisoning and weak sensing parameters [29]. Therefore, recent efforts have been devoted primarily on discovering novel nanomaterials with high conductivity, efficient catalytic activity, and excellent physical and chemical strength for the construction of non-enzymatic sensors [30,31]. Nanomaterials have a large surface area, applied potential window, low charge transfer resistance, and flexibility, which makes them ideal

electrode materials [32,33]. These novel nanomaterials include metal/metal oxide, carbon, and polymer nanocomposites in different nano morphologies such as crystals, rods, wires, fibers, twisters, core shell, and quantum dots (Figure 1) [34]. ideal electrode materials [32,33]. These novel nanomaterials include metal/metal oxide, carbon, and polymer nanocomposites in different nano morphologies such as crystals, rods, wires, fibers, twisters, core shell, and quantum dots (Figure 1) [34].

*Biosensors* **2020**, *10*, x FOR PEER REVIEW 3 of 37

**Figure 1.** Schematic illustration of advanced nanomaterials for non-enzymatic electrochemical glucose and H2O2 sensing: (**a**) AuNBP/MWCNT/GCE nanocomposites [35]; (**b**) Ni3N/GA samples ([36]); (**c**) 3D N-Co-CNT@NG ([37]); (**d**) Cu2O PLNWs/Cu foam ([38]); (**e**) core shell NixCo3-xN/NG **Figure 1.** Schematic illustration of advanced nanomaterials for non-enzymatic electrochemical glucose and H2O<sup>2</sup> sensing: (**a**) AuNBP/MWCNT/GCE nanocomposites [35]; (**b**) Ni3N/GA samples [36]; (**c**) 3D N-Co-CNT@NG [37]; (**d**) Cu2O PLNWs/Cu foam [38]; (**e**) core shell NixCo3-xN/NG [39]; (**f**) Ni (OH)<sup>2</sup> /RGO/Cu2O@Cu electrode [40].

([39]); (**f**) Ni (OH)2/RGO/Cu2O@Cu electrode ([40]).

**2. Metal Nanocomposites for Dual-in-Line NEGH Sensing** 

in designing novel materials.

A wide variety of nanomaterials are fabricated; however, only a limited number of nanomaterials have been utilized for NEGH sensing due to their enhanced conductivity, surface area, electro kinetics, and the electro catalytic activity in acid, and base media. The nanoparticle concentration, synergistic effect, charge carrier type, surface charge, bandgap, mobility and density of electrons on the surface of a nanomaterial can be tuned by considering a combination of materials, and efficient preparation method, which has enabled their applications in a wide range of electrochemical devices [41–43]. Significant research effort was dedicated to the development of NEGH sensing with advanced nanomaterials to obtain high conductivity, suitably applied potential, and portable sensing of glucose and H2O2. Hence, this article focuses on recent advancements in the development of various nanocomposites for NEGH sensing with same electrode materials and comparatively addresses their sensing parameters in terms of wide linear range, limits of detections, response time, stability, reproducibility, sensitivity, and selectivity with critical aspects in real-time clinical, health, and environmental applications. The specific applications of different nanocomposites in real and analytical situations have been discussed and their limitations have been comprehensively addressed. Additionally, we believe that this article help to provide research directions by specifying the existing hindrances faced by advanced nanomaterial-equipped NEGH sensing and can also aid A wide variety of nanomaterials are fabricated; however, only a limited number of nanomaterials have been utilized for NEGH sensing due to their enhanced conductivity, surface area, electro kinetics, and the electro catalytic activity in acid, and base media. The nanoparticle concentration, synergistic effect, charge carrier type, surface charge, bandgap, mobility and density of electrons on the surface of a nanomaterial can be tuned by considering a combination of materials, and efficient preparation method, which has enabled their applications in a wide range of electrochemical devices [41–43]. Significant research effort was dedicated to the development of NEGH sensing with advanced nanomaterials to obtain high conductivity, suitably applied potential, and portable sensing of glucose and H2O2. Hence, this article focuses on recent advancements in the development of various nanocomposites for NEGH sensing with same electrode materials and comparatively addresses their sensing parameters in terms of wide linear range, limits of detections, response time, stability, reproducibility, sensitivity, and selectivity with critical aspects in real-time clinical, health, and environmental applications. The specific applications of different nanocomposites in real and analytical situations have been discussed and their limitations have been comprehensively addressed. Additionally, we believe that this article help to provide research directions by specifying the existing hindrances faced by advanced nanomaterial-equipped NEGH sensing and can also aid in designing novel materials.

#### **2. Metal Nanocomposites for Dual-in-Line NEGH Sensing**

Most of the metal nanocomposites or hybrids benefit from their integrated properties without any alteration in structure and morphology, which can overcome limitations of the traditional noble and non-noble metals [44,45]. Technological advances in metal nanocomposite-based electrodes in several fields have stimulated their exploration in the field of NEGH sensing [46]. The ability of multiple oxidation states, stress-free oxidation of redox reactions, fast formation of intermediate compounds, and easy activation of reaction centers of metal nanocomposites is further utilized in NEGH sensing [47]. Several limitations, such as poor electrochemical activity in alkaline solutions, low diffusion of analytes towards the electrode, the solubility of the electrode, and the aggregation of metal nanoparticles during the electro catalytic process, have been efficiently addressed by the formation of nanocomposites with graphene/carbon nanotubes (both single walled and multi walled)/quantum dots/polymers. This section covers the most widely used metals and their nanocomposites for efficient glucose and H2O<sup>2</sup> sensing.

#### *2.1. Gold and Silver Metal Nanocomposites*

Gold (Au) and silver (Ag) have shown excellent glucose non-enzymatic sensing performance because of their excellent conductivity and electro catalytic activity. These nanoparticles are notable for their antimicrobial activity and in enhancing the durability of sensors and thus are specifically significant in fabricating sensors with a longer lifetime, whereas, for H2O2, the gold-based electrode is inactive except at a very negative potential to form an O-O bond on the surface of the electrode as platinum during electrochemical sensing. Recent studies have shown O-O bond formation on the Au (100) plane surface, reflecting the different crystal facets of Au having different peroxide-like activities. In order to avoid the agglomeration issues of gold nanoparticles, carbon/polymers were used as supporting materials for electrochemical analysis. For example, Mei et al. (2019) synthesized gold nanohybrids by seed-mediated growth on Multiwalled carbon nanotubes (MWCNTs) to develop Gold Nanobipyramids (AuNBP) on MWCNTs as shown in Figure 1a. The AuNBP/MWCNT electrode showed better electrocatalytic activity than the bulk Au, Au Nanoparticle (NP), AuNBP, and MWCNT electrodes because of the more incipient gold oxide provided by AuNBPs. Electrochemical reactions in neutral pH conditions lead to glucose electro-oxidation, which is a diffusion-controlled process, whereas H2O<sup>2</sup> reduction is a surface-controlled process. The major limitation of the AuNBP/MWCNTs hybrid is that the sensor needs to work in a strong alkaline solution to allow glucose detection. They tested the ability of the sensor to detect glucose in human serum while its ability to detect H2O<sup>2</sup> was evaluated in antibacterial lotion (3%). Acceptable recovery with reasonable relative standard deviation (RSD) values for practical applications were reported [35]. Kundu et al. (2015) fabricated ordered assemblies of noble Ag NPs over Graphitic carbon nitride quantum dot (g-CNQD) sheets using the microwave assisted method. The Ag-CNx composites were assembled through an evaporation and condensation process by thermal-ultrasonic treatment. They observed superior electro catalytic activity towards H2O<sup>2</sup> reduction/oxidation compared to 0.01 M PBS buffer than 0.05 M NaOH solution. In this work, reported H2O<sup>2</sup> sensing at +0.7 and −0.7 V applied potential and achieved a lower detection limit of 0.6 nM (+0.7 V). Nucleation and growth of AgNPs on the voids of CNx sheets were strengthened by the Ag-N affinity and the ordered assembly of Ag particles triggered electrochemical sensing. However, the authors did not explore a wide range of molecules for selectivity and other limitations, such as low water solubility, demand further analysis of Ag NPs with the g-CNQD system [48].

## *2.2. Copper Metal Nanocomposites*

The electro catalytic activity of copper metal nanocomposites is mediated by the exchange of oxidation states from Cu (II) and Cu (III) or vice versa. Economically Cu is low cost and easily available and avoids the interference compounds during sensing than the Au/Pt/Ag due to its high isoelectric point (net surface charge). Moreover, the catalytic activity of Cu-based particles are promising, making them applicable in manufacturing sensors for catalysis. Thus, major attention has been given to Cu-based electrodes for NEGH sensing in the last few years. Cu metal-based nanocomposites with different shapes and active support materials such as graphene, reduced graphene oxide, carbon nanotubes, and polymers have improved the NEGH sensing performance. The synthesis strategies of Cu-based materials improved the active surface of electrodes to form intimate contact between highly electroactive nanomaterials. During the sensing mechanism, it acts as an efficient current collector for enhancing electronic conductivity. In this regard, Babu et al. (2014) carried out work on the electropolymerization with electrodeposition technique to develop copper nanoparticles using ionic liquid on a paraffin wax-impregnated graphite electrode (PIGE). The modified electrode exhibited positive working potentials (0 V and +0.35 V) for oxidation of glucose and H2O2. A good response was achieved for glucose concentration ranging from 6.6 × 10−<sup>6</sup> to 1.3 × 10−<sup>3</sup> M with a detection limit of 2.2 × 10−<sup>6</sup> M. The modified electrode catalyzes the electro oxidation of glucose to gluconolactone through the formation of Cu2<sup>+</sup> ions. For H2O2, the electrode exhibited a rapid response in <4 s with a change in concentration. A linear response was achieved for 8.3 × 10−<sup>6</sup> to 1.5 × 10−<sup>3</sup> M with a detection limit of 2.7 × 10−<sup>6</sup> M. This modified copper hybrid electrode showed the advantages of ease of preparation, excellent analytical sensing performance and carries a reduction in over potential to avoid interference. For both glucose and H2O<sup>2</sup> detection, respective applied potentials of +0.35 V and 0 V were reported by this study, which was the least compared with concurrent studies. The practical applications for H2O<sup>2</sup> and glucose concentrations were evaluated in solutions of stain remover and human urine samples, respectively, achieving 99.6% and 103.7% recovery rates [49]. Another research group (Mani et al., 2015) avoided the easy oxidation of Cu NPs by considering the biopolymers (pectin) as scaffold through stabilizing methods and fabricated highly stable, uniform, electroactive Cu NPs using graphene as support. The sensor displayed appreciable repeatability (five measurements), reproducibility (five different electrodes with standard deviation 2.92%) and operational stability (with 6.2% reduction in initial current when rotated in 0.1 M NaOH/2 µM H2O<sup>2</sup> for 3000 s). The real-time applications were performed in contact lens cleaning solution and human serum for H2O<sup>2</sup> and glucose, respectively [50]; however, reasonable data and explanations were not demonstrated. In another report, Lu et al. (2016) discussed Cu chalcogenides, i.e., sulfur-doped Cu in enhancing the sensitivity and low detection limits of glucose and H2O2. This group synthesized Cu2S nano rods on 3D copper foam (Cu2S NRs@Cu) via in situ facile electrodeposition method. The enhanced electrocatalytic activity of Cu2S NRs@Cu was due to its high surface-to-volume ratio and the presence of more active sites, which improved mass and electron transfer between the Cu2S NRs and Cu foam. In addition, it displayed ultra-high sensitivity (glucose: 11,750.8, and H2O2: <sup>745</sup> <sup>µ</sup>A mM−<sup>1</sup> cm−<sup>2</sup> ), excellent reproducibility, selectivity, low detection limits, and also investigated real-time measurements, indicating the promising prospect for NEGH sensors in designing other biomedical applications. Stability of the sensors was explored only for two weeks, and retained the glucose and H2O<sup>2</sup> response by 94.8% and 95.6%. These values may decrease further over time, as there is a possible degradation in the fouling resistance. This opens up the chance of more detailed analysis of the materials reproducibility and durability [51].

#### *2.3. Nickel Metal Nanocomposites*

Ni-based nanocomposite seems to be an excellent material for the fabrication of an NEGH sensor due to their attractive catalytic activity resulting from the redox/oxidation states of Ni3+/Ni2<sup>+</sup> in alkaline media. However, the reaction mechanism is found to be different from Au-, Pt- and Ag-based electrodes. Ni-based hydroxides and oxides showed poor electrical conductivity in electro sensing. As a result, substrates with promising electron transfer ability need to be developed to sustain these active materials. Babu et al. (2013) used an ionic liquid as an electrolyte for the electropolymerization of nanomaterials on PIGE. The modified electrode was used to determine the concentrations of glucose and H2O<sup>2</sup> along with clinically important compounds such as vitamin B6, vanillin, etc. Both cyclic voltammetry and amperometric studies were performed to study the sensing characteristics and the

latter demonstrated a <3 s response time. Good linear range, low working potential and detection limit were achieved with effective applications in flow systems [52]. Furthermore, Wu et al. (2016) doped Ni with Sulphur for morphological change to enhance stability and reproducibility. They synthesized different phases of nickel sulfides (NiS, Ni3S4, Ni7S6, Ni9S8) by a facile hydrothermal method using thiourea and ethanolamine. Among them, they obtained 3D flower-like Ni7S<sup>6</sup> for NEGH sensing. The cyclic voltammetry (CV) graph Ni7S6/Glassy carbon electrode (GCE) contained two redox peaks, which explained the electrocatalytic mechanism 0.1 M NaOH. The first reduction peak was attributed to the conversion of Ni7S<sup>6</sup> to Ni7S6OH. After 20 cycles, a second reduction peak appeared corresponding to the conversion of Ni(OH)<sup>2</sup> to NiOOH. The first reduction peak gradually became weaker, but the second reduction peak gradually became stronger, indicating that Ni7S<sup>6</sup> was consumed gradually and converted to Ni(OH)2. The applicability of the sensor for H2O<sup>2</sup> sensing in antibacterial lotion (3%), and for glucose in human serum samples was evaluated [53]. The values calculated for the glucose in serum was 5.55–5.64 mM, in close proximity to the glucometer data (5.60 mM), and with 4.51% to 3.28% relative standard deviation. To further enhance the electrocatalytic activity of Ni7S6, the same research group doped different concentrations of cobalt (x = 0.5, 1.0, 1.5, 2.0, 2.5, 3.0 and 3.5). Among the synthesized compounds, Ni5.5Co1.5S<sup>6</sup> had shown high sensing performance having an aloe-like morphology. This sensor evaluated glucose and H2O<sup>2</sup> in antibacterial lotions and water samples (lake/tap/pickle water) and showed a reasonable recovery rate of 99–103% for glucose, H2O<sup>2</sup> and nitrite [54]. To avoid the drawback of sulphur poisoning during real-time application, the material was doped with nitrate, and, to enhance conductivity and stability, Yin et al. (2018) prepared novel Ni3N NPs on conductive Graphene aerogels (GA) via hydrothermal method cum freeze-drying and calcination under NH<sup>3</sup> atmosphere. Figure 1b represents the Nickel nitride (Ni3N)/GA-modified electrode for NEGH sensing. The authors were successful in demonstrating the influence of structural characteristics on the sensing performance of the Ni3N/GA sample. The three-dimensional aerogel provides multifunctional electronic or ionic pathways through their interconnected macroporous framework and allows for the easy transportation of electrons and ions. In addition, the aerogels prevent Ni3N aggregation and thus increases the active sites for electrocatalysis. In short, the results demonstrated that the prepared sensor with an efficient conductive nature is applicable for perfect charge transfer and avoided the agglomeration problems, which is a huge benefit for non-enzymatic electrocatalytic application [36].

## *2.4. Cobalt Metal Nanocomposites*

The Co-based catalysts for NEGH sensing was explored and the effect of different morphological structures on electroanalytical properties, and ideal support material for active catalyst loading was studied. As of now, various Co-based oxides, phosphides, sulfides, and complex structures with carbon/polymers have been explored for non-enzymatic monitoring of glucose and H2O2. Electrochemical sensors based on transition metal sulfides offers a more active and cheaper catalyst for sensing both glucose and H2O2. Among various metal sulfides, Cobalt sulphide (CoS) has attracted intense research interest due to their outstanding physical and chemical property with excellent catalytic properties and have been excellent in glucose and H2O<sup>2</sup> detection. In this direction, Wu et al. (2017) studied the ability of different phases of cobalt sulfides to sense glucose and H2O2. One-pot hydrothermally synthesized CoS had a tremella-like nanostructure, and the sensor based on this material exhibited simple operation, good selectivity, stability, and reproducibility [55]. The good electrochemical response towards glucose and H2O<sup>2</sup> was due to the absorption of intermediate species with higher glucose concentration on the electrode surface. The CoS sensor was compared with different H2O<sup>2</sup> and glucose sensors. The CoS sensor showed 17 times wider range of detectability than NP-PtCo and 24 times higher value than nano CoPc-Gr. The detection limit (1.5 µM) was also comparable with the other sensors. Recovery of the sensors from the detection of human serum sample was also appreciable with a relative standard deviation of 2.82%. Furthermore, a novel, scalable, and one-pot method was illustrated by Balamurgan et al. (2017) to prepare a 3D N<sup>2</sup> doped Co-CNT over the

graphene sheet (3D N-Co-CNT@NG). This novel biosensor contains a porous architecture with a high conductive nature for efficient charge transfer at low oxidation and reduction potentials. The pictorial representation of 3D N-Co-CNT@NG for NEGH sensing is shown in Figure 1c. The synergetic interaction between metallic cobalt and nitrogen-doped CNT provided an outstanding electrocatalytic activity, and the fabricated sensor showed promising application in human serum samples and had great potential in health and environment applications [37]. To improve the performance of Co-based NEGH sensing, Fengyu et al. (2018) developed a high electroactive electrode by utilizing a cobalt nitrate nanowire array on Ti mesh. They produced Co3N nanowires (NW) array by hydrothermal with NH<sup>3</sup> gas heat treatment method using Co (NO3)2·6H2O, NH4F, and urea as source materials and found that the array was embedded with Co, N, and Ti ions. The X-ray diffraction (XRD) profile, Scanning electron microscope (SEM) images, corresponding CV curves, ampherometric i-t response, and interferences study results show the structural and morphological features for stable sensing performance of Co3N NW/Titanium mesh (TM) for glucose and H2O<sup>2</sup> detection as shown in Figure 2a–i. The non-enzymatic Co3N NW/TM sensor reported a 0.1 µM to 2.5 mM detection range, a 50 nM low detection limit, a 3325.6 µA mM−<sup>1</sup> cm−<sup>2</sup> response sensitivity and a <5 s response time for glucose. For H2O2, the detection range was from 2 µM to 28 mM with a 1 µM limit. This study suggests a low cost, simple preparation method to prepare 3D doped nano array for NEGH sensing applications, practically useful in electronics and catalytic devices [56].

**Figure 2.** Co3N NW/TM: (**a**) XRD pattern; (**b** and **c**) SEM images of Co (Co(OH)<sup>2</sup> /TM; (**d**) ampherometric i-t response of Co3N NW/TM at 0.55 V (vs Hg/HgO with successive addition of glucose with varying concentration from 20 µM to 5.5 mM); (**e**) corresponding calibration curve of Co3N NW/TM for the detection of glucose; (**f**) interference studies in the presence of glucose; (**g**) ampherometric i-t response of Co3N NW/TM at 0.55 V (vs Hg/HgO with successive addition of H2O<sup>2</sup> with varying concentration from 20 µM to 5.5 mM); (**h**) corresponding calibration curve of Co3N NW/TM for detection H2O<sup>2</sup> ; (**i**) interference studies in the presence of H2O<sup>2</sup> [56].

1

### *2.5. Other Metal Nanocomposites*

Co, Cu, and Ni are the most studied transition metals for generating electrochemical sensors for glucose and H2O<sup>2</sup> [56–58]. However, a few other metals are also reported for their role in developing sensing elements [59]. Barman et al. (2016) obtained improved electron transfer coefficient and catalytic rate constant for the prepared vanadium-based samples. In this work, they modified a gold electrode using a bis(acetylacetonate) oxo vanadium (IV) transition metal complex with 4-(pyridine-40-amido) thiol phenol (PATP) for NEGH sensing in neutral medium. This sensor showed good selectivity in the presence of AA, UA, L-dopa, L-Cysteine and Na+, K<sup>+</sup> and Cl<sup>−</sup> ions. It obtained an excellent recovery rate in human serum for glucose and for H2O<sup>2</sup> in processed milk samples. This work provided a simple preparation process, stability, and low-cost sensor for the clinical and food industry [60]. The study also addressed the influence of scan rate, accumulation potential, time, and pH on the electrochemical sensing properties of the electrode. High scan rates shift the oxidation peak potential of glucose to more positive and reduction potential of H2O<sup>2</sup> to more negative which confirms the kinetic limitation of the electrochemical reaction. When the accumulation time was changed from 0 to 300 s, the oxidation peak currents of both glucose and H2O<sup>2</sup> remained the same; however, the accumulation potential variation (to more positive) decreased the glucose peak current due to oxidation. For H2O2, the change in accumulation potential to a negative value caused reduction. Both the oxidation peak potential of glucose and the reduction peak potential of H2O<sup>2</sup> were pH dependent and they respectively shifted to more negative and more positive with increased pH values (5–10). Good reproducibility (relative standard deviation of 0.2% for glucose and 0.3% for <sup>H</sup>2O2, in 10 repeated cycles), sensitivity (120.24 <sup>µ</sup>A cm−<sup>2</sup> mM−<sup>1</sup> for glucose and 326.66 µA cm−<sup>2</sup> mM−<sup>1</sup> for H2O2), stability (retained 100% response after 20 days) and selectivity were achieved, making it applicable in clinical diagnosis and the food industry.

Sarkar et al. (2018), developed sensors using transition-metal dichalcogenide-based vanadium sulfide (VS2) via template free-solvothermal decomposition process and utilized vanadium for the first time to study sensing parameters. The developed sensor electrode reported a selective and sensitive non-enzymatic detection of H2O<sup>2</sup> with a sensitivity of 41.96 <sup>µ</sup>A mM−<sup>1</sup> , linear range of 0.5 µM to 2.5 mM with a lower detection limit of 0.224 µM. The high conductivity, abundant source, and low cost of VS<sup>2</sup> NPs motivated to study NEGH sensing [61]. Tian et al. (2013) converted bulk C3N<sup>4</sup> into ultrathin graphitic carbon nitrate (g-C3N4) using ultra sonication-assisted liquid exfoliation, which offered a low-cost synthesis method with an efficient electro catalyst for NEGH sensing. The modified g-C3N<sup>4</sup> Nano sheet/GCE showed enhanced electro catalytic activity at a very low negative operational potential of −0.60 V towards H2O2. In the same way, amperometric responses towards glucose were obtained at 0.81 V. It is important to note that g-C3N<sup>4</sup> nano sheets have advantages over noble metal nanomaterials in the form of low-cost fabrication and bulk preparations of samples. It also showed a detection limit of 11 and 45 µM, respectively, for the buffer and human serum media [62].

This section reviews the significant roles of metals such as Au, Ag, Cu, Ni, Co and V in combination with MWCNTs/graphene/reduced graphene/graphitic carbon nitrate/biopolymers in NEGH sensing. The development of simple preparation techniques and attention towards transition metal chalcogenides (TMDs) such as NiN2, CoN2, Cu2S, CoS and V2S has overcome the limitations of poor conductivity, less exposed active sites, poor measurement stability, low contacted target analytes, low electron transfer, chemical instability, wide band gap and also reduced over potential issues. In a few reports, the modification of carbon materials into reduced graphene oxide, 3D graphene aerogels, and graphitic carbon nitride with unique geometry has improved porous structure for electron/ion transfer, high conductivity, strong adhesion property to catalyst particles, high mechanism strength, thermal stability, agglomeration of nanoparticles, etc., and favored reproducibility. The development of an advanced electro deposition process, and in situ fabrication techniques for metal nanocomposites has improved long-term stability for the NEGH sensors. The detection of H2O<sup>2</sup> requires high over potentials, which in turn causes interference issues and decreases the sensitivity. These can be overcome by considering efficient synthesis strategies, such as the aforementioned synthesis methods. The working potential of modified electrodes is a key parameter for dual sensing application, which was effectively changed by modifying the morphology of metal nanocomposites into sheets, nano wires, nano rods, and flower-like structures that enhanced the surface to volume ratio to increase mass and electron transfer issues. Overall, the metal nanocomposites performed excellent catalytic activity, and exhibited notable NEGH sensing performance. By doping the different concentration of metals in sulfides and nitrides, especially transition metals can further improve the sensitivity, detection limits and linear ranges of glucose and H2O<sup>2</sup> analytes. The sensing performance of metal nanocomposites in alkaline and acid conditions are still not clear and need to be improved further by considering core-shell-like nanostructure morphologies. Electrochemical sensing parameters such as sensitivity, detection limits, linear ranges, working potentials, storage stability, repeatability, reproducibility and real-time applications are compared for different modified metal nanocomposite electrodes for both glucose and H2O<sup>2</sup> sensing, in Table 1.

#### **3. Metal Oxide Nanocomposite for Dual-In-Line NEGHS**

Metal nanocomposites had limitation for NEGH sensing such as inferior performance under neutral or low pH conditions and easy oxidation in harsh environments because of the dependency of MOOH species on the electro oxidation/reduction of glucose and H2O2. These limitations have increased focus on the development of metal oxide nanocomposites in NEGH sensing. Metal oxide sensors have the advantages of rapid electro catalytic response with specific morphology of nanoparticles, nanotubes, nanowires, nanofibers, graphene/CNTs, among others. In this section, we discuss advanced developments of NEGH sensors based on various metal oxides. In Table 2, the information on metal oxides for NEGH sensors is reported, and a brief comparison of H2O<sup>2</sup> and glucose with the same electrode materials is given in terms of the sensing parameters.



#### *Biosensors* **2020**, *10*, 151


*Biosensors* **2020**, *10*, 151

**Table 1.** *Cont.*



#### *3.1. Copper Oxide (CuO) Nanocomposite*

The natural abundance, low cost, and unique optical and electro catalytic properties of CuO mark them as one of the suitable nanomaterials for heterogeneous catalysis, magnetic storage devices, lithium-ion electrodes, gas sensors, and photovoltaic devices. Compared with the unstable Cu and Cu2O, CuO nanostructures are relatively stable for electro sensing analysis. NEGH detection technology can be used to design CuO nanomaterials with enhanced non-enzymatic intrinsic characteristics [63]. They are synthesized in the form of nano spheres, rods, wires, and flowers. Prathap et al. (2012) conducted a study to control the morphology of copper oxide (CuO) using different acids such as ammonia/citric/tartaric acids via the hydrothermal method and proposed a CuO formation mechanism based on the experimental results. According to the mechanism, crystal formation fully depends on nucleation and crystal growth. The addition of NH<sup>3</sup> and NaOH to the reaction medium forms a Cu(NH3)<sup>4</sup> <sup>2</sup><sup>+</sup> complex followed by the precipitation of orthorhombic Cu(OH)2. This is in the form of a sheet-like structure connected through H-bonding, the length of which enhances with amino acid interaction. During the hydrothermal reaction, the amino acid functional group forms a co-ordinate bond with Cu2<sup>+</sup> resulting in its adsorption on the crystal particle surface, preventing the re-dissolution/re-precipitation. This causes formation of flower-like morphologies compared to the dumbbell morphology during normal reaction conditions. In fact, the chemical nature of acids and hydrothermal time modified the morphology of CuO. The tyrosine amino acid synthesized CuO showed the best electro catalytic activity in this study and the results were compared with conventional CuO nanoparticles. This work provided new insight for the fabrication of CuO with different morphologies using different chemical additives and demonstrated the influence of large specific surface area and porosity in enhancing electron transfer and thus sensitivity [64]. Recently, Liu et al. (2019) developed a novel electrochemical sensor with hollow CuO/Polyalanine (PANI) nano-hybrid co-axial fibers via. electrospinning using poly(acrylic acid)(PAA) as a sacrificial template. The utilization of PANI in this work achieved excellent stability, high specific capacitance, strong adsorption, large surface area and many reactive sites. The three-dimensional porous structure of the developed sensor elements and the hollow structure of the hybrid nanofiber enhanced the surface area and the reactive sites and enabled the electrochemical sensing at ultra-low concentration levels. The developed electrode also retained its initial current response after 10 days and showed a promising application in clinical and food testing [65]. In addition, Chakraborty et al. (2019) synthesized 1D nanomaterials (CuO nanorods) over Fluorine doped Tin Oxide (FTO) substrate via the novel hydrothermal method and suggested that the 1D nanostructure electrodes are favorable to NEGH sensing due to their low fabrication cost, high electro active surface area, and excellent charge transfer property compared to other nanostructures. The pictorial representation of glucose oxidation, H2O<sup>2</sup> reduction, and interference studies of this work are shown in Figure 3a–d. In glucose sensing, the high valence Cu3<sup>+</sup> mediates the electro oxidation of glucose on the CuO surface. This happens when the glucose oxidation converts Cu2<sup>+</sup> to Cu3<sup>+</sup> and the formed ion acts as an electron delivery system for the glucose-gluconolactone-gluconic acid conversion. Similarly, the electro catalytic reduction of H2O<sup>2</sup> reduces Cu2<sup>+</sup> to Cu1+, which intermediates the H2O<sup>2</sup> to water conversion. This group accurately performed simultaneous sensing of glucose in the presence of H2O2. The data demonstrated negligible current density changes with the addition of interfering agents compared to the current density variation with glucose/H2O<sup>2</sup> addition. Thus, the dual sensor developed with a stability of 30 days was observed to be useful in practical applications from the point of manufacturing biodevices [66].

*Biosensors* **2020**, *10*, x FOR PEER REVIEW 15 of 37

**Figure 3.** Schematic representation of CuO NRs: (**a**) glucose oxidation; (**b**) H2O2 reduction; (**c**) interference study during glucose sensing after the addition of 0.1 mM of DA, UA, AA, UR and SU and 0.5 mM of H2O2 along with 0.5 mM glucose; (**d**) interference during H2O2 sensing during the addition of 0.1 mM DA, UA, AA, UR, SU and 0.5 mM of glucose along with 0.5 mM H2O2 [66]. **Figure 3.** Schematic representation of CuO NRs: (**a**) glucose oxidation; (**b**) H2O<sup>2</sup> reduction; (**c**) interference study during glucose sensing after the addition of 0.1 mM of DA, UA, AA, UR and SU and 0.5 mM of H2O<sup>2</sup> along with 0.5 mM glucose; (**d**) interference during H2O<sup>2</sup> sensing during the addition of 0.1 mM DA, UA, AA, UR, SU and 0.5 mM of glucose along with 0.5 mM H2O<sup>2</sup> [66].

#### *3.2. Cuprous Oxide (Cu2O) Nanocomposite 3.2. Cuprous Oxide (Cu2O) Nanocomposite*

Cu2O is a well-known p-type semiconducting material with a 2.17 eV band gap and is applied in many potential applications, such as lithium ion batteries, solar cells and gas sensors. Zhang et al. (2009) provided a promising Cu2O microstructure for NEGH sensing application. They fabricated porous cuprous oxide (Cu2O) microcubes by a simple sonochemical route and its sensing results were compared with smooth surface Cu2O microcubes under similar experimental conditions. The porous cubes had much higher performance compared to that of the smooth Cu2O attributing to the porous microstructure, which provided abundant active sites for glucose and H2O2 sensing [67]. Li et al. (2011) implemented a low-temperature chemical method for the preparation of hierarchical Cu2O nanocrystals with the help of sodium borohydride (NaBH4), polyvinyl pyrrolidone (PVP) and N, Ndimethylformamide (DMF). The high charge-transport channels in hierarchical Cu2O nanocrystals was due to the self-assembly of nanocrystals and the presence of many grain boundaries with a compact attachment of nanocrystals. The increased electro active surface area showed a fast amperometric response and sensitivity for H2O2, which was much higher than glucose detection. The response time of less than 0.5 s was required to achieve steady current during H2O2 detection with 0.39 × 10−7 mol L−1 detection limit. However, the developed sensor showed 1.2 × 104 times higher detection limit for the glucose compared with the H2O2, the reason for which was not fully addressed [68]. In another work, Gao et al. (2012) successfully prepared mesocrystalline Cu2O hollow nanocubes (MCHNs) via a facile reduction reaction and studied the effects of reaction parameters. To identify factors contributing to unique characteristics for the formation of MCHNs, experiments were performed by changing CuCl2 to CuSO4. Hierarchical mesoporous spheres were formed with CuSO4. At the same time, when LiOH was changed to NaOH, a cubic shaped product with a solid or hollow appearance was obtained, suggesting the leading role of Cl ions in the formation of distinctive MCHN structure. By varying the temperature, the final product was analyzed at low and high temperatures, and the formation of nanocubes was observed with some wide size distributions. Cu2O is a well-known p-type semiconducting material with a 2.17 eV band gap and is applied in many potential applications, such as lithium ion batteries, solar cells and gas sensors. Zhang et al. (2009) provided a promising Cu2O microstructure for NEGH sensing application. They fabricated porous cuprous oxide (Cu2O) microcubes by a simple sonochemical route and its sensing results were compared with smooth surface Cu2O microcubes under similar experimental conditions. The porous cubes had much higher performance compared to that of the smooth Cu2O attributing to the porous microstructure, which provided abundant active sites for glucose and H2O<sup>2</sup> sensing [67]. Li et al. (2011) implemented a low-temperature chemical method for the preparation of hierarchical Cu2O nanocrystals with the help of sodium borohydride (NaBH4), polyvinyl pyrrolidone (PVP) and N,N-dimethylformamide (DMF). The high charge-transport channels in hierarchical Cu2O nanocrystals was due to the self-assembly of nanocrystals and the presence of many grain boundaries with a compact attachment of nanocrystals. The increased electro active surface area showed a fast amperometric response and sensitivity for H2O2, which was much higher than glucose detection. The response time of less than 0.5 s was required to achieve steady current during H2O<sup>2</sup> detection with 0.39 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mol L−<sup>1</sup> detection limit. However, the developed sensor showed 1.2 × 10<sup>4</sup> times higher detection limit for the glucose compared with the H2O2, the reason for which was not fully addressed [68]. In another work, Gao et al. (2012) successfully prepared mesocrystalline Cu2O hollow nanocubes (MCHNs) via a facile reduction reaction and studied the effects of reaction parameters. To identify factors contributing to unique characteristics for the formation of MCHNs, experiments were performed by changing CuCl<sup>2</sup> to CuSO4. Hierarchical mesoporous spheres were formed with CuSO4. At the same time, when LiOH was changed to NaOH, a cubic shaped product with a solid or hollow appearance was obtained, suggesting the leading role of Cl<sup>−</sup> ions in the formation of distinctive MCHN structure. By varying the temperature, the final product was analyzed at low and high temperatures, and the formation of nanocubes was observed with some wide size distributions. These results confirm that the kinetics of reactions are essential for

These results confirm that the kinetics of reactions are essential for the formation of different morphologies of Cu2O products. Finally, they showed high resistance to interference species with the formation of different morphologies of Cu2O products. Finally, they showed high resistance to interference species with excellent reproducibility and high stability [69]. In another study, Liu et al. (2013) improved the electrochemical cycling stability of Cu2O nanocubes by wrapping with graphene. The resulting nanocomposite showed a glucose-sensing response with a low detection limit of 3.3 µM and a linear response of 0.3 to 3.3 mM. The non-enzymatic H2O<sup>2</sup> sensor exhibited an electrocatalytic response with a linear range of 0.3 to 7.8 mM and a low detection limit of 20.8 µM. While other studies use interferences with 1/20 to 1/10 glucose concentration to study the selectivity of the sensor, this study tested interferents with 1/2 glucose concentration in 0.1 M KOH. Lower potentials generated negligible current responses, and at 0.7 V, the responses become <3.5%, which is a comparatively good sign of selectivity. Moreover, the high chloride tolerance was also confirmed for Cu2O/GNs as it did not change the current of glucose oxidation. The very good linear response, selectivity and detection range are associated with the higher electron transfer ability and increased electro catalytic surface area [70]. In another reearch study, Cu2O was combined with carbon quantum dots (CQDs) to enhance the stability and sensitivity for NEGH sensing. The Cu2O/CQDs were synthesized by a hydrothermal with ultrasonic treatment method, and the presence of low-index (111)-octahedral planes showed good electrochemical performance and stability in the sensing of all low-indexed planes. The scan rate also affected the glucose oxidation, as increasing scan rates increased both oxidation and reduction currents. The water solubility and biocompatibility of CQD with octahedral Cu2O further enhanced linear response ranges and selectivity issues. In short, the CQDs/octahedral Cu2O/Nafion/GCE provided wider detection range, shorter detection limit and response time than the octahedral Cu2O/Nafion/GCE, attributed mainly to the synergistic interaction between CQD and (111) planes of Cu2O [71]. Ding et al. (2015) reported a superior NEGH sensing electrode with excellent conductivity using Cu2O microspheres (MSs) decorated on reduced graphene oxide (RGO). Cu2O MSs of different sizes and uniform shapes were obtained on the surface of RGO by varying the mass ratio (1:20 to 1:80) using sodium ascorbate in the presence of sodium hydroxide. The RGO sheets cover the Cu2O and act as additional surfactant. This reduces the microsphere size, prevents particle aggregation, protects Cu2O MSs and improves the electrochemical stability. The typical reaction method controls the Cu2O nanocrystal morphology with the addition of a capping agent and the Cu2O MS grows on RGO sheets by the Ostwald ripening mechanism. When the mass ratio was 1:80, the sensor produced the best performance, i.e., a 0.005 to 2.775 mM linear detection range and a 0.0108 mM detection limit for <sup>H</sup>2O<sup>2</sup> and a 0.001 to 0.419 mM linear detection range and a 7.288 <sup>×</sup> <sup>10</sup>−<sup>4</sup> M detection limit for glucose. In addition, this sensor showed improved stability with excellent selectivity and good reproducibility because of the extraordinary high surface property of RGO, which reduced the size of Cu2O MSs to improve the catalytic activity and the synergetic interaction between RGO and Cu2O MSs [72]. In another report, Lu et al. (2016) developed a self-supporting NEGH sensing electrode by modifying 3D copper foam into a pod-like Cu2O nanowire array as shown in Figure 1d. The Cu foam acted as a current collector and facilitated charge and mass transfer, while the open framework of the foam provided large amounts of anchoring sites for the deposition of Cu2O NWs. The Cu2O PLNWs/Cu foam, respectively, showed the sensitivity of 6.6807 mA mM−<sup>1</sup> cm−<sup>2</sup> and 1.4773 mA mM−<sup>1</sup> cm−<sup>2</sup> to glucose and H2O<sup>2</sup> and detection limits of 0.67 and 1.05 µM. It further exhibited high stability (retained 98.9% of initial response after a week) and resistance to interference studies. The relative standard deviation was 4.61% for six tests for 0.1 mM glucose, substantiating good reproducibility [38] and thus promising that enzymeless glucose and H2O<sup>2</sup> sensors can be developed by manipulating the structural integrity of the Cu-based nanocomposites.
