Durable nonenzymatic electrochemical sensing using silver decorated multi-walled carbon nanotubes for uric acid detection

In this study, we demonstrate a facile, durable and inexpensive technique of producing silver nanoparticles-decorated multi-walled carbon nanotubes (MWCNT/AgNP) on the easy-to-use screen-printed carbon electrodes (SPCE) for non-enzymatic detection of uric acid (UA) in an electrochemical sensor. The developed sensors show great durability for three months in storage, and high specificity performance for preclinical study using spiked UA in a synthetic urine sample. A simple route for this hybrid nanocomposite was proposed through an oxidation–reduction with reflux (ORR) process. A significant increase in the electroactive surface area of SPCE was achieved by modifying it with MWCNT/AgNP. Scanning electron microscopy (SEM), transmission electron microscopy (TEM), energy-dispersive x-ray spectroscopy (EDX), Fourier-transform infrared (FT-IR) spectroscopy, Raman spectroscopy, and x-ray diffraction (XRD) analysis confirmed this synthesis. The nanocomposite nanostructure electrodes achieved an outstanding UA detection with sensitivity of 0.1021 μA μM−1 and a wide dynamic range of 10–1000 μM. In phosphate-buffered saline (PBS), the measurements achieved a detection limit of 84.04 nM while in pure synthetic urine; it was 6.074 μM. The constructed sensor exhibits excellent stability and durability for several months, and great specificity against interfering compounds, including dopamine (DA), urea, and glucose. Overall, the present outcomes denote the potential of MWCNT/AgNP-decorated SPCE for early uric acid diagnostics tools in health monitoring.


Introduction
Uric acid is a byproduct of purine metabolism, formed in the human body during the breakdown of nucleic acids [1].Ordinarily, the kidney expels uric acid from the body, but its accumulation in the bloodstream may occur because of excessive production or inadequate excretion [2].The typical range for uric acid concentration in serum is 155-416 μmol l −1 [3,4], while in urine, it is around 300-400 mg d −1 , which corresponds to 1.8-2.4μmol l −1 [5].Elevated levels of uric acid, known as hyperuricemia, have been linked to various diseases such as gout, metabolic syndrome, cardiovascular disease, and chronic kidney disease [6][7][8][9][10].Consequently, measuring uric acid levels has become crucial in diagnosing and predicting these conditions [11].To facilitate early diagnostics and prevent worsening health conditions, the development of a point-of-care (PoC) testing method for convenient serum uric acid monitoring is imperative.However, creating biosensor components for PoC applications presents certain challenges, notably in terms of durability and lifespan.The ideal components should be practical, storable in a simple system, and able to endure extended periods of use.Such advancements would significantly benefit PoC devices in remote areas with limited resources [12].
Nonenzymatic detection is an excellent approach in biosensing that observes the electrocatalytic activity of the biomarkers.Using this technique, the analyte detection can be measured without the use of enzymes, as a promising alternative because of its simplicity, low cost, and high stability [13,14].The fundamental mechanism of the electrocatalytic analysis in electrochemistry process is by using specific electrodes with metal oxide or carbon-based materials.These materials in nanostructures effectively catalyze the oxidationreduction process of the analyte with no enzymes [15][16][17].One of the popular carbon-based materials for the nonenzymatic electrodes is carbon nanotubes [18][19][20][21].
The increasing popularity of multi-walled carbon nanotubes (MWCNTs) in non-enzymatic biosensor development stems from their advantageous properties, including low cost, good biocompatibility, and a high surface-area-to-volume ratio (SA:V) [22,23].The elevated SA:V in these materials offers numerous active sites that facilitate efficient electron transfer on electrode surfaces.Compared to single-walled carbon nanotubes (SWCNTs), MWCNTs have simpler synthesis procedures, allowing for higher aspect ratios, greater purity with no catalyst, and improved mechanical strength [24].An illustrative case involves the use of silver nanoparticles (AgNP) to decorate MWCNT/AgNP nanocomposites, which has been the subject of several studies for biochemical sensing [22,[25][26][27][28]. Incorporating AgNP onto MWCNTs gives excellent electrocatalytic properties, enhancing the electron transfer rate between the sensor surface and the analyte.As a result, the obtained electrical signal is significantly amplified [29].
Prior investigations have explored various electrodes and materials for non-enzymatic uric acid biosensors [6,[30][31][32][33][34][35][36].However, to the best of our knowledge, the utilization of MWCNT/AgNP nanocomposites for this purpose has not been addressed specifically.This article demonstrates nonenzymatic detection of uric acid for preclinical study using electrochemical sensors, incorporating durable, long-lasting, and easily synthesized MWCNT/AgNP electrodes.Screenprinted carbon electrodes (SPCE) offer several advantages, such as convenience, design flexibility, and the requirement of only a small sample volume because of their compact size [17,[37][38][39].This research is focused on developing a nonenzymatic uric acid biosensor that is cost-effective and easy to use, achieved through the modification of the SPCE surface with MWCNT/AgNP nanocomposites.A facile route was used to synthesize the nanocomposites by anchoring AgNPs through −COOH functionalization of MWCNT [24,40].The presence of AgNPs on MWCNTs with their higher surface area and curvatures significantly enhances electrochemical signal amplification.
In this article, we proposed a novel method for uric acid (UA) detection in synthetic urine samples using differential pulse voltammetry (DPV) method in an electrochemical sensor platform with the electrode of MWCNT/AgNP.The proposed structure performed a durable and long span lifetime of the electrodes up to 3 months.The proposed sensor undergoes a series of electrochemical detections for uric acid, exhibiting rapid screening, exceptional detection limit around 84.04 nM, broad dynamic range up to 1 mM and excellent specificity against common uric acid interference in the urine sample, such as dopamine (DA), urea, and glucose.Moreover, our results show the sensor's durability for preclinical study, measuring the spiked UA in synthetic urine samples with LOD 0.167 to 6.074 μM.These promising results pave the way for point-of-care (PoC) applications in early diagnostics.

Apparatus and instrumentations
The successful synthesis of MWCNT/AgNP nanocomposites was confirmed using various instruments.Scanning electron microscopy with energy dispersive x-ray (JIB-4610F, JEOL, Tokyo, Japan) and Transmission Electron Microscopy (TECNAI G220 S-TWIN, Thermo Fisher Scientific, Waltham, Massachusetts, USA) were used to examine the morphology of the material and were used to determine the percentage of constituent elements.Raman spectroscopy (Labspec 6-HORIBA Scientific Type iHR320, Kyoto, Japan) investigated material vibrations and determined graphitization and defects.Fourier-transform Infrared (Shimadzu IR Prestige-21, Kyoto, Japan) was used to identify functional groups, and x-ray diffraction (SmartLab Rigaku, Tokyo, Japan) was used for crystallinity analysis of the electrode.
Commercial screen-printed carbon electrodes (SPCEs) were obtained from Zimmer Peacock Ltd (Oslo, Norway), with a carbon working electrode and Ag/AgCl counter and reference electrodes.Electroanalytical measurements were conducted using Zimmer Peacock AnaPot EIS (Oslo, Norway) and PalmSens Sensit Smart potentiostats (Houten, the Netherlands).For statistical analysis, we used OriginPro software (Northampton, Massachusetts, USA).Analysis of particle distributions was using ImageJ (image J 1.53t, Wayne Rasband and contributors, National Institute of Health, USA).

Synthesis and characterization of MWCNT/AgNP nanocomposites
Functionalized MWCNT was prepared by refluxing it in a mixture of H 2 SO 4 and HNO 3 (3:1 ratio).First, 4 g of MWCNT was mixed with 75 ml of 95%-97% H 2 SO 4 and 25 ml of 65% HNO 3 and refluxed for 7 h at 32 °C until black precipitate was yielded.This precipitate was washed with dH 2 O and centrifuged until the filtrate reached a neutral pH.The resulting MWCNT-COOH powder was obtained by drying the precipitate in an oven at 80 °C [25,41].
MWCNT/AgNP nanocomposites were synthesized using functionalized MWCNT-COOH.NaOH was used as the reducing agent.Initially, 0.5 g of MWCNT-COOH was mixed with 200 ml of dH 2 O and sonicated for an hour.After adding 200 ml of 0.1 M AgNO 3 to the suspension, sonication continued for another half hour.Next, the black suspension was heated and stirred at around 85 °C for ten minutes.During this time, 25 ml of 8 M NaOH solution was added drop-by-drop.Following a wash with dH2O, the suspension was centrifuged until the filtrate reached a neutral pH.The resulting MWCNT/AgNP powder was obtained by drying the precipitation in a furnace at 80 °C. Figure 1 illustrates schematics of MWCNT/AgNP nanocomposites.
MWCNT-COOH and MWCNT-AgNP nanocomposites were studied with SEM-EDX to determine their shape and elements.The utilization of transmission electron microscopy (TEM) enables the examination of the structure of the MWCNTs/AgNPs nanocomposites.The crystal structure of MWCNT/AgNPs nanocomposites can be identified using x-ray diffraction (XRD).Raman spectroscopy was used to measure the vibrational spectra of MWCNT-COOH and MWCNT/AgNPs.MWCNT/AgNPs functional groups are detected using Fourier-transform infrared (FT-IR).We also use ImageJ to calculate the particle size distribution of MWCNT and AgNP.The particle size distribution shows the percentage of particles with specific sizes (or within specific size intervals).After analyzing the image, we get a table with length data representing particle diameters.
Particle size analysis using ImageJ (National Institute of Health, USA) was done manually.The feature of normalization scale was used to analyze the size of the particles.Furthermore, because the particles to be analyzed have varying sizes, it is necessary to calibrate the scale bar.The procedure is to select a picture, then draw a straight line on it according to the scale bar and set the known distance with a value shown on the scale bar and adjust the unit of length.The next step is the measurement of the size of each particle.Similar to the scale adjustment procedure, a straight line was drawn, but this time according to the diameter of the particle to be measured.A table containing the diameter length data of the particles will be obtained.The same procedure for all particles was repeated.After the measurement process is complete, the last step is to export the data to retrieve the measurement results into a worksheet editor (Microsoft Excel, Microsoft, USA).

Fabrication of MWCNT/AgNP nanocomposite electrode
Fabrication of the silver-lined nanocomposite electrode is performed by drop-casting the nanocomposite suspension onto the surface of the working electrode of a screen printed carbon electrode (SPCE) with Nafion as a binder.A nanocomposite suspension was prepared by mixing MWCNT/ AgNP powder with dH 2 O at the desired concentration and sonicating for 40 min.Nafion solution was prepared by diluting 5% Nafion in ethanol at a 1:20 (v/v) ratio.SPCE activation was also performed before modifying the working electrode by conducting cyclic voltammetry (CV) using 50 μl of 0.5 M H 2 SO 4 at a potential range of 0.9 to 1.5 V versus Ag/AgCl.
Fabrication of the nanocomposite electrode involved dropping a certain volume of MWCNT/AgNP suspension onto the SPCE working electrode and allowing it to dry for 30 min at room temperature.Nafion layer was added by dropping 0.5 μl of Nafion solution onto the modified electrode surface and drying for 5 min at room temperature.For multiple layering of Nafion, each layer was added by repeating this step.The modified electrode fabrication was optimized for the following parameters: During each optimization process, as one parameter was varied, the other two were kept constant.An optimal value of the parameter was obtained for each optimization process.This optimal value was then used as a constant value for the next optimization process.For example, during droplet volume optimization, droplet volumes were varied while the number of Nafion layers and nanocomposite suspension concentration was kept constant.Once an optimal value for the droplet volume was obtained, this value was then used during the next optimization process for the number of layers of Nafion.
First, the droplet volume was varied from 0.5 to 2 μl with an increment of 0.5 μl, while the nanocomposite suspension concentration was kept at 1 mg.ml −1 and 4 layers of Nafion were used.Afterwards, the number of layers for the Nafion was varied from 1 to 4 layers, the droplet volume was kept constant (at the optimal value obtained from the previous optimization process) and 1 mg.ml −1 of nanocomposite suspension concentration was used.Finally, the nanocomposite suspension concentration was varied to 1, 3, and 5 mg.ml −1 , while results from previous optimization were used for the other two parameters.Besides MWCNT/AgNP-SPCE, MWCNT-COOH-SPCE was also prepared using the obtained optimal drop-casting steps.The electrochemical surface area (ECSA) of the MWCNT/AgNP-SPCE (precisely, the modified working electrode) was also calculated.The area was calculated using the data from the cyclic voltammetry (CV) measurement of the modified electrode in solution containing 10 mM K 3 [Fe(CN) 6 ] + 10 mM K 4 [Fe(CN) 6 ].3H 2 O + 0.1 M KCl.CV parameters used were scan rate of 0.1 V s −1 and potential range of −0.3 to 0.6 V versus Ag/AgCl.Bare SPCE was also tested for comparison.

Uric acid electrochemical sensing
Uric acid (UA) sensing performance of MWCNT/AgNP-SPCE was evaluated through differential pulse voltammetry (DPV) method.Sample solution (50 μl) containing various concentrations of UA was measured at a potential range of 0 to 0.4 V versus Ag/AgCl with 50 mV.s−1 scan rate.UA sample solutions with different concentrations (10, 25, 50, 100, 250, 500, and 1000 μM) were prepared by dissolving UA powder in 10 mM PBS pH 7.4 solution.UA sensing performance of the modified electrode was also evaluated in synthetic urine solution to mimic the real sample from a human for preclinical implementation purpose.The UA samples were spiked to the urine samples and the same DPV measurements were performed.There were three variations of spiked samples used: (i) 100× dilution of synthetic urine in 10 mM PBS pH 7.4, (ii) 10× dilution of synthetic urine in 10 mM PBS pH 7.4, and (iii) pure synthetic urine.Three repetitions of measurement were conducted for all experiments in this section.The acid treatment targets the introduction of the carboxylic group on the MWCNT and aims to increase the interaction of carbon nanotubes (CNT) and organic solvents and polymer matrices, which is typically weak [40].Covalent functionalization is also preferable since the dispersibility of CNT could be enhanced, the toxicity is reduced and the hydrophilic nature of CNT is ensured [42] which is substantial in facilitating aqueous target capture in the detection stage.This functionalization step also added defects, which simultaneously provides strong interaction forces for metal nucleation.Subsequently, through electrostatic interaction, Ag + ions are attached to the COOH-modified MWCNT.AgNPs-speckled MWCNT can be seen in previous reports [25,43].

Synthesis and characterization of MWCNT/AgNP nanocomposites
To validate the uniformity of MWCNT and AgNPs sizes, particle size distribution analysis was performed using Ima-geJ.Figure 3(a) shows that the as-prepared MWCNT dominantly generated fine fibrous structures with an average diameter of 19.576 nm ± 0.245.Meanwhile, the average diameter of the AgNPs was calculated to be 3.063 nm ± 0.052 (figure 3(b)).Undergoing the energy dispersive x-ray analysis (EDX) displayed in figure 3(c), we observed the nanocomposites were dominated by C (85.5 Wt%), followed by Ag (9.9 Wt%) and O (4.6 Wt%).This implies the purity of the synthesized nanocomposites, as no other elements were detected in the spectrum.Raman spectroscopy was also used to characterize the synthesized nanocomposites.The obtained Raman spectrum presented in figure 3 These results are in good agreement with the findings in the previous study [44].The appearance of D′+D′ bands provide further insight into the structural and electronic characteristics of the MWCNT/AgNP material.It provides additional information about the local structure and electronic properties of carbon materials.
Figure 3(e) presents the FT-IR spectrum of the MWCNT/AgNP.The peak observed at ∼1444 cm −1 showed C=C stretching band of the MWCNT backbone, while peaks at ∼1640 cm −1 and ∼1059 cm −1 can be assigned to C=O and C-O stretching bands of the carboxylic acid functional group (-COOH).Peaks at ∼3206 cm −1 and ∼3387 cm −1 are absorption peaks of the hydroxyl group (-OH) [45].XRD spectrum in figure 3   remaining AgNO 3 crystalline from the MWCNT/AgNP synthesis.

Optimization of various parameters in the fabrication of MWCNT/AgNP nanocomposite electrode
Some parameters included in the drop-casting process were varied and evaluated to fabricate a MWCNT/AgNP-modified electrode, which would produce the optimal electrochemical response.The first drop-casting parameter to be optimized was the droplet volume of MWCNT/AgNP suspension dropped on the surface of the electrode.Figures 4(a) and (b) indicate that the highest current density value was produced from the electrode modified with 2 μl of nanocomposite suspension.This outcome is obtained for both anodic and cathodic peaks, where j pa and j pc are 1.68 mA.cm −2 and −0.6 mA.cm −2 , respectively.Subsequent optimisation was performed for Nafion layers, coating the modified electrodes.Nafion was used as a binder to help stabilise MWCNT/AgNP deposited on the electrode surface, thus preventing disintegration when performing electrochemical measurements [50][51][52].However, Nafion film is proton conductive and might repel anionic molecules like uric acid in a neutral pH environment.Still, the ionic strength of UA is not high enough to be thoroughly repelled by Nafion film with a certain thickness [36].Therefore, optimization was conducted to find the optimal amount of Nafion layers that can properly bind the nanocomposites, yet still produce a high enough response.Modified electrodes with different number of Nafion layers were tested, and the results are presented in figures 4(c) and (d).Using two layers of Nafion yielded the highest density current, where j pa and j pc obtained are 2.41 mA.cm −2 and −0.82 mA.cm −2 respectively.Last, various concentrations of MWCNT/AgNP suspension were tested.Results in figures 4(e) and (f) indicate that the best outcome was obtained from electrodes modified with 5 mg.ml −1 MWCNT/AgNP suspension.j pa and j pc obtained were 9.98 mA.cm −2 and −10.96 mA.cm −2 , respectively.From this optimization, it can be concluded that the dropcasting process gives the most optimal yield when using 2 μl of MWCNT/AgNP suspension with a concentration of 5 mg.ml −1 , followed by 2 layers of Nafion coating.Effect of potential scan rate values used during CV measurements was also investigated in a 10 mM PBS solution.The scan rate values examined ranged from 10 to 200 mV.s−1 .Figures 5(c) and (d) display the cyclic voltammograms and corresponding calibration curves obtained.The calibration demonstrates that current density values of anodic and cathodic peaks have a linear correlation with the square roots of the scan rate (ν 1/2 ) variation, with correlation coefficient values (R 2 ) of 0.9996 and 0.9871, respectively.This correlation infers the diffusion-controlled mechanism of the electrochemical reaction on the surface of MWCNT/AgNP-SPCE [54].
The stability towards consecutive measurements (repeatability) and the stability based on storage test (durability) of MWCNT/AgNP-SPCE were also tested.The test results were evaluated based on the values of average current change and coefficient of variation of the anodic and cathodic peak current.Average current change is the average of the relative error percentage of the peak current value for all 50 cycles of the cyclic voltammetry measurement for MWCNT/ AgNP-SPCE in 10 mM PBS pH 7.4 (see figure 6(a)).The relative error is calculated using equation (1).Meanwhile, coefficient of variation is the percentage value of the standard deviation of the peak current value data divided by the mean  of the related data (see equation ( 2)).
where Error rel is the relative error; I x is the cyclic voltammetry's anodic or cathodic peak current value of cycle x; I ref is the reference current value of the cyclic voltammetry's anodic or cathodic peak taken from the first cycle of each test; CoV is the coefficient of variation; and SD is the standard deviation of the anodic or cathodic peak current value data, while Mean is the average of the related data.
The modified electrode shows good stability towards consecutive measurements until 50 cycles of CV, as shown in figures 6(a) and (b), with average current changes of 7.95% and 7.50%, as well as coefficient of variation (CoV) values of 2.68% and 2.06% for anodic and cathodic peaks, respectively.In addition, excellent stability was also achieved for a storage test of the electrochemical response.Three different MWCNT/AgNP-modified electrodes were stored for 3 months in a closed container at room temperature.Each electrode received different treatments regarding CV measurements: weekly, monthly, and no in-between measurements during the 3-month storage period, as presented in figures 6(c)-(h).Anodic and cathodic peaks of the weekly measured MWCNT/AgNP-SPCE showed 49.38% and 33.16% remaining current responses.After three months, the monthly measured electrode indicated 88.13% anodic and 79.94% cathodic current values.Weekly measurements showed CoV values of 22.38% and 32.88%, while monthly measurements had CoV values of 5.11% and 8.88% for anodic and cathodic peaks, respectively.The best storage stability was displayed when the electrode did not undergo any in-between measurements, with the final responses of 92.69% and 87.69% with CoV as low as 3.79% and 6.56% for anodic and cathodic peaks, respectively.
Figure 7 shows the cyclic voltammograms of the MWCNT/AgNP-SPCE versus bare SPCE in the solution containing 10 mM K 3 [Fe(CN) 6 ] + 10 mM K 4 [Fe(CN) 6 ]. 3H 2 O + 0.1 M KCl.As it can be seen, there is an increase in anodic and cathodic peak currents between bare SPCE and MWCNT/AgNP-SPCE.These results indicate the nanocomposite provides higher conductivity to the electrode.A smaller peak-to-peak separation value of MWCNT/AgNP-SPCE (V versus Ag/AgCl) compared to bare SPCE (V versus Ag/AgCl) is also observed, indicating that the modified electrode produces a faster electron transfer [55].The data from figure 7 was also used to calculate the electrochemical surface area (ECSA) of the MWCNT/AgNP-SPCE's working electrode by using Randles-Ševčík equation shown in equation (3) [56].Using I P value of 116 μA from the CV's anodic/oxidation peak current of the experiment (figure 7), n (electron transfer number) value of 1, C (concentration of K 3 [Fe(CN) 6 ]), the substance that undergoes oxidation reaction) value of 10 mM, ν (scan rate) value of 0.1 V s −1 , and D (diffusion coefficient) value of 7.60 × 10 −6 cm 2 •s −1 (0.1 mol•l −1 KCl) [6].The calculated A (ECSA value) of MWCNT/AgNP-SPCE's working electrode was 0.0495 cm 2 .The roughness factor (ρ) was also calculated by dividing ECSA value with the geometric surface area of SPCE's working electrode, which is 0.0415 cm 2 , resulting in a value of 1.1928.

. Uric acid sensing performance of MWCNT/AgNP-SPCE
MWCNT/AgNP-SPCE responses to various UA concentrations were recorded to evaluate the sensor performance toward UA detection.The measurement was done using differential pulse voltammetry (DPV) in 10 mM of PBS at pH 7.4 solution containing UA with concentrations of 0, 10, 25, 50, 100, 250, 500, and 1000 μM.This dynamic range comprehensively covers the clinical cut-off in uric acid standard screening (155-416 μmol.l−1 ) [3,4].The response for each UA concentration was measured using different MWCNT/AgNP-modified electrodes.Figure 8(a) shows the recorded differential pulse voltammograms after being refined using the baseline subtraction analysis tool on OriginPro software.The shifts were observed in the anodic peak potential values along with the increase in UA concentration, from +0.08 V versus Ag/AgCl for 10 μM to +0.21 V versus Ag/AgCl for 1000 μM.While this suggests that more energy was needed for the oxidation reaction when a higher amount of UA was present in the working solution [25], the observed shift may be a manifestation of the confounded effects between the complex interaction between this surface modification (MWCNT/AgNP-SPCE), the analyte and the matrix of the working solution.Figure 8(b) depicts the corresponding calibration curve for the peak current values from the DPV measurement.The curve presents a linear correlation between anodic current and UA concentration from 10 to 1000 μM with an R 2 value of 0.9857.In addition, the limit of detection (LOD) and limit of quantification (LOQ) values can also be determined using the following equations [57,58]: where σ is the standard deviation obtained from 3 DPV measurements in a blank solution of 10 mM PBS pH 7.4 (0.0026 μA), and s is the slope of the calibration curve represents the sensitivity of 0.1021 μA μM −1 .The calculated LOD is 84.04 nM, while the LOQ is 254.65 nM.These results showed that MWCNT/AgNP-SPCE can sense UA concentration values below and above the normal range found in human serum (155-416 μM) [7], which is suitable for the application to clinical assessment [59].
The modified electrode showed good UA sensing performance in synthetic urine solution.We used the same parameters as in 10 mM of PBS at pH Table 1 summarizes the results of the UA detection variations.The LOQ and LOQ values of the detection increased when the solution containing UA was more complex.The complexity of the solution was because of the addition of synthetic urine, which contained several substances.These substances might affect the background signal, causing the baseline to fluctuate and the value of σ to increase, which corresponded to the increase of LOD and LOQ values.However, the LOD values of each detection variation were still within the linear detection range (10 to 1000 μM).

Sensor selectivity and stability
To evaluate the selectivity of MWCNT/AgNP-SPCE as a UA biosensor, commonly identified bioanalytes in human serum alongside UA were tested as interfering substances.Measurements were conducted using the DPV method with the same parameters previously used to evaluate UA sensing performance.The substances used were dopamine (DA), urea, and glucose.Before they were mixed with UA in the same solution, MWCNT/AgNP-SPCE response to each of the substances was observed at the same potential range used during UA detection.Solutions of 250 μM UA, 2 μM DA, 3 mM urea, and 3 mM glucose were prepared in 10 mM PBS pH 7.4.The tested concentrations were chosen with consideration of each substance's concentration range in human serum [7,25,26,50].MWCNT/AgNP-SPCE responses towards UA with the interference of each substance and the mixture of UA and each substance were then measured.
Figure 9(a) depicts the selectivity of the proposed sensor towards UA as displayed by the peak current of UA as compared to the negligible current resulting from the measurement of other substances measured separately.The specificity and selectivity of the proposed sensor is also attested by the measurement of the mixture containing UA with one or all three interfering substances where the peak current of UA was still maintained regardless of the mixture components (figure 9(b)).When measurement is performed in the mixed solution, all mixture solutions containing UA and any single interfering agents generated peak currents.This is plausible since the particular potential window is highly selective for UA, yet with some signal attenuation because of the presence of the interfering compounds.
The stability of the sensor is substantial considering that, in real application, the electrode could be stored or transported within a certain period or setup.Figure 9(c) displayed the stability of the signal obtained for UA detection on five different MWCNT/AgNP-SPCEs.These electrodes were fabricated and then stored at the same starting point (t storage = 0 for each electrode is the same, where t storage arbitrarily refers to the storage period).One electrode was used immediately (at t storage = 0 week).Each of the remaining four electrodes was then used for UA measurement in the following weeks.One electrode is used for each week.The modified electrodes were stored in a closed container at room temperature when they were not used.It is shown that the peak current is relatively stable, with the reduction peak current ranging from around 50-54 μA (figure 9(d)).The sensor indicates good stability of sensing performance, with a final response of 93.25% from the initial response.Thus, the stability of the MWCNT/AgNP-SPCE has been tested for up to 4 weeks of storage period.
As compared with the previously published reports on UA detection on a modified electrochemical electrode shown in table 2, our constructed MWCNT/AgNP-modified SPCE outstandingly performs by detection limit and comparable durability.The detection limit achieved in this study emphasizes the capability of the proposed structure toward the advancement as early diagnostic PoC in uric acid detection.

Conclusions
The fabricated MWCNT/AgNP-SPCE exhibited excellent electroanalytical performance as a uric acid electrochemical biosensor.The synthesized MWCNT/AgNP provided exceptional nanostructuring pathway of an SPCE with high coverage, outstanding morphological parameters and effective increase of surface area, beneficial in signal augmentation for biosensing.Applied as uric acid electrochemical sensor, the proposed structure achieved an impressive detection limit down to 84.04 nM and sensitivity of 0.1021 μA μM −1 for uric acid with a sufficiently wide linear range of 10-1000 μM.Performed in the detection of UA with several interfering agents, including dopamine, urea and glucose, our system resulted in a high selectivity with robust reduction peak current signal relatively stable in response to uric acid regardless of the presence of other interfering substances.The stability of the proposed biosensor is validated by the electrochemical measurements of the MWCNT/AgNP-SPCE.The proposed biosensor is stable up to three months when tested in PBS pH 7.4.It is stable up to four weeks when tested with uric acid samples.Overall, the system holds an alternative route for the high-cost and complex nanostructuring technique, which could be applied as an early diagnostic point-of-care system for uric acid detection.In the future work, we envisage the integration of the proposed sensor with microfluidic system toward automatization and the detection of complex clinical matrices.

Figure 2 (
Figure2(a) shows that the functionalization of pristine MWCNT with mixed strong acids resulted in a clearer threadlike structure, which confirms the effects of impurity removal in the as-prepared MWCNT.The acid treatment targets the introduction of the carboxylic group on the MWCNT and aims to increase the interaction of carbon nanotubes (CNT) and organic solvents and polymer matrices, which is typically weak[40].Covalent functionalization is also preferable since the dispersibility of
The addition of NaOH solution reduces Ag + to Ag 0 , forming silver nanoparticles attached to the MWCNT surface.As shown in figure 2(b), the silver nanoparticles (AgNPs) are distributed on the surface of MWCNTs which are depicted as spherical shapes and bright images of AgNPs resulting in a considerably regular size of interconnecting thread-like structure that provides effective

Figure 4 .
Figure 4. (a) Cyclic voltammograms of SPCE modified with various droplet volumes of MWCNT/AgNP suspension measured in 10 mM PBS pH 7.4, and (b) the corresponding peak current.(c) Cyclic voltammograms of modified SPCE coated with various amounts of Nafion layers measured in 10 mM PBS pH 7.4, and (d) the corresponding peak current.(e) Cyclic voltammograms of SPCE modified with various concentrations of MWCNT/AgNP suspension measured at 10 mM PBS pH 7.4, and (f) the corresponding peak current.Each data point has standard deviation from 3 repetitions.
(d) shows the formation of D band, G band, 2D band, as well as D+D′ band around the Raman shift values typical for MWCNT/AgNP.The Raman spectrum of MWCNT-COOH and MWCNT/AgNP shows that the addition of AgNP to the MWCNT making the ratio of 2D/G band drop significantly.It indicates that the defects of the MWCNT increase as the AgNPs attached to it.

3. 3 .
Electrochemical characterization of MWCNT/AgNP-SPCE The electrochemical response of MWCNT/AgNP-SPCE measured with cyclic voltammetry (CV) is compared with the response of the bare SPCE and MWCNT-COOH-SPCE.The CV measurement was conducted in 10 mM PBS pH 7.4 solution with a scan rate of 50 mV.s−1 .From the recorded cyclic voltammograms presented in figure 5(a), both MWCNT/AgNP-SPCE and MWCNT-COOH-SPCE showed an increase in both anodic and cathodic peak current value compared to the bare SPCE.The current density peaks produced by MWCNT/AgNP-SPCE were significantly higher compared to MWCNT-COOH-SPCE.This shows the electrocatalytic property of AgNP, which increased the electron transfer rate on the electrode surface, thus improving the electrochemical properties of MWCNT/AgNP-SPCE as an electrochemical sensor [53].MWCNT/AgNP-SPCE has three peaks corresponding to the oxidation and reduction reaction of AgNP with PBS as the electrolyte solution [25].The anodic peak at +0.06 V versus Ag/AgCl indicated oxidation of Ag 0 to Ag + , while cathodic peaks at −0.10 V versus Ag/AgCl and −0.45 V Ag/AgCl demonstrated a reduction of Ag + to Ag 0 and Ag 2+ to Ag + , respectively.Figure 5(b) shows the photograph of (i) bare SPCE, (ii) MWCNT-COOH-SPCE, and (iii) MWCNT/AgNP-SPCE.
7.4 solution to measure the DPV response of MWCNT/AgNP-SPCE towards UA in different solutions: 100× dilution of synthetic urine in 10 mM PBS pH 7.4 (figure 8(c)), 10× dilution of synthetic urine in 10 mM PBS pH 7.4 (figure 8(e)), and pure synthetic urine (figure 8(g)).The calibration curves for the peak current values are shown in figures 8(d), (f), and (h), respectively.The LOD and LOQ values for each solution were: 0.167 μM and 0.505 μM for 100× dilution of synthetic urine in 10 mM PBS pH 7.4; 1.930 μM and 5.850 μM for 10× dilution of synthetic urine in 10 mM PBS pH 7.4; and 6.074 μM and 18.405 μM for pure synthetic urine.

Figure 9 .
Figure 9. Baseline-subtracted differential pulse voltammograms of MWCNT/AgNP-SPCE response to (a) 250 μM UA, 2 μM DA, 3 mM urea, and 3 mM glucose in 10 mM PBS, and (b) UA with the addition of interfering substances mixed separately and all together with the stated concentration.(c) Differential pulse voltammograms and (d) The peak current characteristics of 4-week storage stability of MWCNT/ AgNP-SPCE sensing performance in UA 500 μM.

Table 1 .
Comparison of UA detection's LOD and LOQ using MWCNT/AgNP-SPCE sensor in solutions of 10 mM of PBS at pH 7.4, 100× dilution of synthetic urine in 10 mM PBS pH 7.4, 10× dilution of synthetic urine in 10 mM PBS pH 7.4, and pure synthetic urine.

Table 2 .
Comparison of analytical performance from various reported modified electrodes implemented for uric acid detection.