3.2.4. SEM

gCuHCF was characterized by SEM coupled with X-ray microanalysis (SEM-XRM). It was found in our previous work [40] that SEM can supply information on the size, distribution, and shape of the tested sample. Figure 3a–d presents the overall morphology of the flower-like particles formed in the process. The XRM images of the gCuHCF film show the characteristic peaks for Cu and Fe (Figure 3e). According to the SEM results, the synthesized gCuHCFs are not nano-sized but rather microparticles.

**Figure 3.** The results of gCuHCF study using SEM with RSM: (**<sup>a</sup>**–**d**)—SEM images at different magnifications; (**e**)—X-ray spectral characteristics.

Likewise, the different analytical approaches demonstrated that the synthesized gCuHCF is a suspension of micro-sized particles. These observations were confirmed by different methods: means of particle counting, dynamic light scattering, zeta-potential analysis, and SEM.

Based on the gCuHCF images presented in Figure 3, the studied catalytically active composite material may be described as "organic-inorganic micro/nanoflowers" (hNFs).

hNFs belong to a class of flower-like hybrid materials that self-assemble from metal ions and organic components, such as enzymes, DNA, and amino acids, into flowerlike micro/nano superstructures [59,60]. hNFs are widely used for the development of stable, robust, reusable, efficient and cost-effective systems for the immobilization of biomolecules. Some hNFs were shown to exhibit an intrinsic PO-like activity [61,62]. Due to their remarkable performance—the simplicity of their synthesis; their high surface area; excellent thermal, storage, and pH stability; and catalytic activity—hNFs have various

potential applications in bioremediation, bioassays, biomedicine, industrial biocatalysis and wastewater treatment [60]. Promising results were reported for hNFs in biosensing, including electrochemical biosensors, colorimetric biosensors and point-of-care diagnostic devices [60–62].

#### *3.3. Application of the gCuHCF as a PO Mimetic in Amperometric (Bio)sensors*

The applicability of the gCuHCF as a chemo-sensor for H2O2 detection was demonstrated in our previous work [40]. Quantitative analysis of a real sample of commercial disinfectant was carried out. The average H2O2 concentration determined by the gCuHCFbased chemo-sensor was shown to be well correlated with the manufacturer's data, with an error of less than 10%.

#### 3.3.1. Properties of gCuHCF

Selectivity of the ABS towards the target analyte is of grea<sup>t</sup> importance, especially for the analysis of real samples. In this paper, to study the selectivity of the gCuHCF, a modified GE was tested for its ability to respond to a number of analytes: glucose, alcohols, organic acids, and ammonium ions, etc. The selectivity of the constructed chemo-sensor was estimated for the individual natural substrates (Figure 4a) as well as for their mixture with hydrogen peroxide (Figure 4b). The results presented in Figure 4b demonstrate that the presence of various compounds in the analyzed mixture does not interfere with H2O2 determination.

**Figure 4.** The selectivity tests for gCuHCF/GE: (**a**)—current responses in relative units (%), on the added analytes up to 2 mM concentration, as a ratio of the detected signals to the value of the highest current response; (**b**)—chronoamperograms as outputs on the added analytes (1–7) up to 0.5 mM concentration: (1)—H2O2, (2)—glucose, (3)—glycerol, (4)—methanol, (5)—sodium citrate, (6)—sodium lactate, (7)—ammonium chloride. Conditions: working potential −50 mV vs. Ag/AgCl (reference electrode), 50 mM NaOAc buffer, pH 4.5 at 23 ◦C.

The amperometric analysis was performed using CV and chronoamperometry at different potentials (−50 and +150–200 mV) in different buffer solutions, with a pH from 4.0 to 8.0 (data not shown). It was demonstrated that neither methanol, glycerol, organic acids, nor glucose elicited any signals, while hydrogen peroxide (at −50 mV), ammonium ions and L-lactate (both at +200 mV) were found to elicit significant current responses on the gCuHCF/GE under the tested conditions. Current responses to L-lactate and ammonium under the potential −50 mV were insignificant (Figure 4).

Moreover, we demonstrated that in gCuHCF formation, Fc*b*2 was concentrated from the diluted solutions due to co-precipitation with the gCuHCF-based hNFs. When immobilized on a GE, the gCuHCF may become an ABS for L-lactate. CV analysis showed that the current output due to the L-lactate addition correlated with Fc*b*2 activity in the sensing layer (data not shown). Thus, the proposed method of hNF formation, using oxido-reductase

in the presence of its substrate, may be a promising platform for the concentration and stabilization of any enzyme.

Additionally, using a laccase as a model oxidase, we demonstrated that the gCuHCF not only displayed enzymatic (PO) activity but also an electro-mediator ability (data not shown).

Preliminary experiments for the development of biosensors for primary alcohols and L-amino acids (based on alcohol oxidase and L-amino acid oxidase, respectively) were carried out (data not shown). The obtained results indicated that the gCuHCF and other gHCFs have a potential for use as PO-like composites for the construction of amperometric biosensors with any oxidase.

We conclude that the gCuHCF that was obtained with Fc*b*2 assistance, forming a flower-like micro-superstructure, is a prospective organic-inorganic composite material for biosensor construction. It is a stable, catalytically and electrochemically active carrier for enzyme concentration, immobilization and stabilization.

#### 3.3.2. Optimization of H2O2 Sensing

To improve the conditions for exploiting the biosensor, the optimal buffer, pH and working potential were estimated. For optimization of the chemo-sensor and further biosensor construction, the quantity of gCuHCF material on the surface of the GE, as well as the enzyme/gCuHCF ratio, were determined experimentally.

We analyzed the correlation of PO-mimetic activity with the effectiveness of H2O2 sensing, using the gCuHCF/GE under different conditions of pH and working potential.

The dependence of the chemo-sensor's analytical characteristics on the quantities of gCuHCF placed on the GE surface was studied under the working potential −50 mV in 50 mM NaOAc, pH 4.5. The results are presented in Figure S4 and are summarized in Table 3. Based on the data, the optimal PO-like activity of the gCuHCF for achieving the highest sensitivity under the described conditions is 2–5 mU.

**Table 3.** Effect of gCuHCF PO-mimetic activity on the analytical characteristics of the modified GEs at pH 4.5.


The optimal working potential for H2O2 sensing was determined using a CV study (Figure 5), followed by chronoamperometry experiments at pH 6.0 (data not shown). The decision to change the conditions of the experiments, and work under a pH range of 6–8, was necessitated by our plans to develop biosensors using different oxido-reductases. Many microbial enzymes have shown optimal activity near these pH values.

As seen in Figure 5, the optimal working potentials for H2O2 sensing under pH 6.0 were lower than −100 mV. To select the best conditions for achieving the highest gCuHCF/GE sensitivity, we determined its analytical parameters under different potentials, namely, −50 and −200 mV (Figure S5). According to the data, the chemo-sensor sensitivity under −200 mV was 2.7-fold higher than under −50 mV.

**Figure 5.** Cyclic voltammograms (CV) of the gCuHCF/GE. CV profiles (1–3) as outputs upon addition of H2O2 up to concentrations: (1)—0 mM (black); (2)—0.17 mM (red); (3)—0.5 mM (blue) mM. Conditions: scan rate 50 mV·s<sup>−</sup>1; Ag/AgCl (reference electrode) in 50 mM PB, pH 6.0. The sensing layer contains 0.35 mU of PO-like activity.

3.3.3. Development of an Amperometric Biosensor for Glucose Determination

In our previous work [40], we reported on the construction of a mono-enzyme amperometric biosensor (ABS) for glucose, using gCuHCF as the PO mimetic and commercial glucose oxidase (GO). It is worth mentioning that the control gCuHCF/GE did not show any amperometric output in response to glucose. The sensitivity of the developed GO/gCuHCF/GE was rather low (76 <sup>A</sup>·M−1·m<sup>−</sup>2). In the current study, we set a goal to develop an improved GO/gCuHCF/GE with elevated/optimized analytical characteristics. We carried out the investigation of the gCuHCF as an artificial PO in more detail by studying the influence of various experimental stages on the effectiveness of H2O2 sensing; we describe these results in Section 3.3.2. The next task was the optimization of glucose biosensing.

According to Figure 6, the optimal working potential for glucose sensing determined via CV measurement was −450 mV. However, to avoid a possible interference of various substances on the electrode response in the presence of oxygen at high voltage, we chose a lower working potential, namely, −250 mV. This requirement is relevant for the application of the biosensor for the analysis of real samples, e.g., food products.

**Figure 6.** Cyclic voltammograms (CV) of the GO/gCuHCF/GE. CV profiles (1–4) as outputs upon addition of glucose up to concentrations: (1) 0, (2) 0.17, (3) 0.5, (4) 1.3 mM. Conditions: scan rate 50 mV·s<sup>−</sup>1; Ag/AgCl (reference electrode) in 50 mM PB, pH 6.5. The sensing layer of the biosensor contains 0.5 mU of PO-like gCuHCF and 40 mU of GO.

For optimization of the biosensor composition, the enzyme/gCuHCF ratio on the GE surface was determined experimentally (data not shown). It was found that the optimal ratio, calculated from total activities (GO and PO-like gCuHCF), was 80. Activities of the GO and gCuHCF were estimated with o-dianisidine, as described in Sections 2.2 and 2.5, respectively.

Figure 7 demonstrates the best results obtained from the constructed GO-ABSs. To select the optimal working potential for GO-ABS exploitation, we estimated its analytical parameters under two potentials, at −250 and at −300 mV (Figure 7). Taking into account the parameters (b) from the linear regression graphs (Figure 7b,d) and the square of the electrode surface (7.3 mm2), we calculated the sensitivities of the GO-ABS to glucose. These and other analytical characteristics of the developed GO/gCuHCF/GEs are summarized in Table 4. According to Table 4, the sensitivity (A <sup>M</sup>−1m−2) at the potential −250 mV was 2.2-fold higher than at −300 mV, and 9.4-fold higher than at −50 mV. Thus, −250 mV was chosen as the optimal working potential for the exploitation of a GO/gCuHCF-based ABS.

**Figure 7.** Characteristics of the GO/gCuHCF/GE under different working potentials: (**<sup>a</sup>**,**b**)—dependences of the current response on increasing concentrations for glucose determination; (**<sup>c</sup>**,**d**)—calibration graphs. Conditions: working potentials −250 (**<sup>a</sup>**,**<sup>c</sup>**) and −300 mV (**b**,**d**) vs. Ag/AgCl (reference electrode), 50 mM phosphate buffer, pH 6.0 at 23 ◦C. The GE contains 0.5 mU of PO-like activity and 40 mU GO.


**Table 4.** Analytical characteristics of the developed GO/gCuHCF/GEs.

Thus, we determined the optimal conditions for construction and exploitation of the most effective and highly sensitive GO-based ABS: the ratio of GO activity to PO-like activity of gCuHCF was shown to be 80 under conditions of −250 mV working potential, 50 mM phosphate buffer, and pH 6.0.

#### 3.3.4. Testing of GO/gCuHCF/GE Biosensor for Glucose Analysis in Juice Samples

In order to demonstrate the practical feasibility of the constructed ABS, the developed biosensor was used for glucose analysis in three fruit juice samples using the graphical method known as the standard addition test (SAT). Graphical SAT is a type of quantitative analysis often used in analytical chemistry when a standard is added directly to the aliquots of the analyzed sample. SAT is used in situations where sample components also may contribute to the analytical signal, which makes it impossible to use routine calibration methods. Estimation of glucose concentration in the initial sample was performed using the equation C = AN/B, where A and B are parameters of a linear regression and N is the dilution factor.

Figure 8 demonstrates in detail the algorithm of glucose estimation using two juices as the examples. The results of glucose determination in the juices sampled by the proposed biosensor and by a commercial enzymatic kit are presented in Table 5. The average glucose concentrations determined from the data in Figure 8 differ by less than 10% from the data obtained using the reference method (Table 5).

**Figure 8.** The example of glucose analysis using the biosensor in samples of juices: Multivitamin "Sadochok" (**<sup>a</sup>**–**<sup>c</sup>**), and apple–pear "Galicia" (**d**), in two dilutions; chronoamperograms (**<sup>a</sup>**,**b**), and corresponding linear graphs (**<sup>c</sup>**,**d**). Conditions: working potential −250 mV vs. Ag/AgCl (reference electrode), 50 mM phosphate buffer, pH 6.0 at 23 ◦C.


**Table 5.** Results of glucose estimation in the samples of fruit juices.
