Carbon-based ruthenium nanomaterial-based electroanalytical sensors for the detection of anticancer drug Idarubicin

In this work, a novel nanosensing platform was suggested based on ruthenium for the sensitive determination of Idarubicin anticancer drugs. Ruthenium/Vulcan carbon-based nanoparticles were synthesized ultrasonication method and then characterized by transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and X-ray diffraction (XRD). The mean particle size of the nanoparticles calculated by the TEM analysis was found to be 1.98 nm ± 0.29 nm, and the Ru nanoparticles were mostly dispersed on the support material. Glassy carbon electrode (GCE) surface was modified with Ruthenium/Vulcan carbon-based nanomaterials (Ru@VC), and characterization of the nanosensor was performed using electrochemical impedance spectroscopy and cyclic voltammetry. The limit of detection (LOD) and limit of quantification (LOQ) values were found as 9.25 × 10–9 M and 2.8 × 10–8 M in buffer samples. To demonstrate the applicability and validity of developed nanosensor, it was used for the determination of Idarubicin in Idamen® IV (10 mg/10 mL vial) and human serum sample. The results of recovery studies showed that the Ru@VC/GCE nanosensor was free from excipient interferences in the dosage forms of injection, and it can be successfully applied to biological samples.


experimental
Apparatus. Electrochemical impedance spectroscopy (EIS) measurements were done using a Metrohm Autolab Potentiostat/Galvanostat device. PalmSens4 Potentiostat/Galvanostat/Impedance Analyzer was used for all the other voltammetric measurements for IDA. The working conditions of cyclic voltammetry (CV) were: initial potential, − 0.2 V; scan rate, 100 mV/s; potential step, equilibration time, 5 s; 2 mV; N scans, four and current range, 1,000 µA. In the measurements of differential pulse voltammetry (DPV), initial potential, − 0.2 V; pulse amplitude, 50 mV; scan rate, 20 mV/s; pulse width, 200 ms; potential step, 20 mV; final potential, 1.2 V; current range, 10 µA, and equilibration time, 5 s were utilized. In the measurements of adsorptive stripping differential pulse voltammetry (AdSDPV), scan rate, 20 mV/s; pulse width, working potential, − 0.2 V-1.2 V; pulse amplitude, 50 mV; 200 ms; equilibration time, 5 s; potential step, 20 mV; current range, 10 µA, deposition time 120 s, and deposition potential 0 V were utilized. In addition, the three-electrode system consists of a GCE (BAS, 3 mm, diameter) as the working electrode, a Pt wire as the auxiliary electrode, and an Ag/AgCl electrode as the reference electrode. In the course of the experiment, prior to each measurement, the GCE was polished on a damp polishing cloth (BAS velvet polishing pad) by an aqueous slurry of alumina (0.01 μm, Metkon ALU-MIK) until a mirror-like finish was obtained. Before each AdSDPV measurement for the electrochemical cleaning of the modified glassy carbon electrode, 20 CV cycles were applied. A pH meter Model 538 (WTW, Germany) was utilized for the pH measurements with a combined electrode with an accuracy of ± pH. To the drying process, the Nuve EV 018 vacuum oven was employed.
Reagents and chemicals. Idamen ® IV 10 mg/10 mL vial and IDA were provided by pharmaceutical companies in Turkey. Sodium phosphate monobasic dihydrate, sulphuric acid, acetic acid, phosphoric acid, disodium phosphate, sodium acetate trihydrate, ruthenium(III) chloride (RuCl 3 ), DMAB [4-(dimethylamino) benzaldehyde], Vulcan ® XC72R, human serum samples, methanol, and ethanol were obtained from Sigma-Aldrich. Also, the water was processed to be analytical grade using the Millipore water treatment system in all experiments. Under the argon atmosphere, 99.5% tetrahydrofuran (THF) from Merck was made ready.
preparation of solutions. The  XRD was done utilizing a Panalytical Empyrean diffractometer with Ultima + theta-theta high-resolution goniometer, an X-ray generator (Cu K radiation, λ = 1.54056 Å) with 45 kV and 40 mA operating conditions. Ru@VC based nanosensor preparation. For the preparation of Ru@VC suspension, firstly, 1 mg of Ru@VC was distributed in 1 mL distilled water. Later, utilization of the ultrasonic bath, the suspension was ultrasonicated for 2 h. Before GCE was modified, the electrode surface was polished on a polishing cloth with alumina slurry, then washed by distilled water and dried with air. Lastly; the nanomaterial amount optimization study was performed by dropping various volumes of Ru@VC suspension (1 µL, 3 µL, 5 µL, 7 µL, and 10 µL) onto the surface of the electrode and drying in a vacuum oven. iDA analysis from dosage forms and human serum samples. Idamen ® IV 10 mg/10 mL vial contains 10 mg Idarubicin Hydrochloride for injection. The working solution of IDA with 20% methanol was prepared by dilution of the stock solution to 2.5 × 10 -7 M with pH 1.5 phosphate buffer solution. In order to prepare standard serum solution, 1 mL of IDA from 1 × 10 −3 M stock solution of IDA, 3.6 mL serum, and 5.4 mL acetonitrile were put together in a 10 mL centrifuge tube, and this mixture was centrifuged for 30 min at 3,500 rpm. After the centrifuge process and separation of protein residues and supernatant, the supernatant was taken and used for serum sample analysis. For further electrochemical analysis with serum samples, working solutions were prepared using supernatant as serum stock solution with methanol and phosphate buffer solution at pH 1.5 and were analyzed under optimum conditions. The recovery studies were performed using a standard addition method to prove the reliability, applicability, and accuracy of the proposed nanosensor from real samples.

Results and discussion
Synthesis of Ru@VC nanomaterial. The distribution of Ru metals on VC support materials and the particle size of the formed Ru@VC nanomaterials were investigated by TEM analysis. Figure 1a indicates well and almost homogenous distributions of the metals on the support materials. Figure 1a also shows the negligible level of particle agglomeration has been obtained on the surface of Vulcan carbon supports. The average particle size was calculated utilizing an image analyzer (ImageJ software). About 100 particles were employed for the calculation of the size distribution. As shown in Fig. 1b, the average particle size of Ru@VC nanomaterials was found as 1.98 nm ± 0.29 nm that result has a good agreement with the previous studies [43][44][45] .
XPS analysis also studied the oxidation state of Ru@VC nanomaterials according to the Gaussian-Lorentzian process. Figure 2 indicates the XPS pattern of Ru and shows the main peaks at 462.60, 484.97 eV assigned to Ru 0 and 464.60, 486.96 eV assigned to Ru 4+ , respectively. Additionally, the binding energies of 462.60 eV and 486.96 eV were observed in the Ru 3p 3/2 and 3p 1/2 orbital XPS spectra of Ru@VC nanomaterials [46][47][48] .
The crystal structures of the Vulcan carbon and synthesized Ru@VC nanomaterials were characterized by utilizing XRD, the results of which are displayed in Fig. 3. The characteristic peaks appearing at 2θ values of 26° www.nature.com/scientificreports/ and 42.3° may be ascribed to Vulcan carbon corresponding to (002) and (101) crystal plane, respectively 49 . From  Fig. 3, it detected that the XRD pattern of Ru@VC nanomaterials displays the diffraction peaks at 68.32°, and 77.2° represented by (110) and (103), the crystal planes of ruthenium, respectively. Furthermore, the 101 crystal planes of face-centered cubic (fcc) structures of ruthenium were detected at around 2θ = 42.3°5 0,51 .
Electrochemical characterization of Ru@VC/GCE. EIS is a significant electrochemical method for explaining differences between unmodified and modified electrodes in terms of conductivity or impedance for oxidation or reduction processes 52 . In this work, EIS measurements were performed using 5 mM K 3 [Fe(CN) 6 ] solution as the redox probe in order to describe and compare electrochemical characteristics of bare GCE and Ru@VC/GCE. The results are given in Fig. 4 in the form of a Nyquist plot. The parameters determined after fitting the results to the Randles equivalent circuit model are listed in Table 1. As it is displayed in Fig. 4, the EIS Nyquist plot of Ru@VC modified GCE has a smaller semicircle compared to bare GCE. These results indicate that the modified GCE has better electronic conductivity and enhances electron transfer kinetics compared to bare GCE. Besides, the surface of Ru@VC/GCE has faster electron transfer and decreased charge transfer resistance.  www.nature.com/scientificreports/ pH effect on electrochemical studies. In order to investigate the influence of the pH on the oxidation peak potentials and peak currents of IDA, H 2 SO 4 solutions, acetate, and phosphate buffer solutions pH values ranging from 0.3 to 8 were used by DPV using Ru@VC/GCE (Fig. 5). The relationship among peak potential (Ep) and pH can be described by the Eq. (1) as follows: The pH increase resulted in a shift of Ep to less positive values. While Ep demonstrated a linear response against pH, it also showed that Ep was pH-dependent. Furthermore, the slope value of the equation above is close   IDA's peak current (I p ) reached the max peak when the pH was increased from 0.3 to pH 1.5. Starting with pH 4.7, peaks of modification material interfered with IDA peaks and caused inconsistent results. Therefore, pH 1.5 phosphate buffer was chosen as the optimal pH value and was utilized in additional measurements. Scan rate effect on electrochemical studies. The electrochemical behavior of 5 × 10 -6 M IDA was studied at optimum pH, which is in phosphate buffer (pH 1.5), on bare GCE and Ru@VC/GCE by CV (Fig. 6). Same as in our previous study 4 , a well-defined, oxidation peak was obtained around 700 mV with bare GCE, whereas Ru@VC/GCE showed a more broad peak for the same concentration of IDA with very high current values.
The scan rate studies were performed to understand the oxidation behavior of IDA on Ru@VC/GCE, whether if it is diffusion or adsorption controlled. 5 × 10 -6 M IDA was investigated at Ru@VC/GCE using CV in phosphate buffer pH 1.5 in the range of 5 to 1,000 mV/s. The decrease of scan rate resulted in the shift of Ep to lower potential values and the decrease of I p : Equation (3) can represent the relationship between I p and v. The linearity of I p vs. vindicates that the oxidation mechanism is adsorption controlled.
Moreover, a log I p vs. log v graph was also obtained to understand the process deeply. If the correlation coefficient of the log I p vs. log v is close to 0.5, that indicates the diffusion-controlled electrode process, and if it is close to 1, that indicates the adsorption controlled electrode process [54][55][56] .
It can be understood from Eq. (3) that the nature of the electrode process is adsorption controlled. Therefore, deposition potential and deposition time were further optimized. In the adsorption process, the optimal deposition time and potential were studied using 5 × 10 -6 M IDA in pH 1.5 phosphate buffer using the AdSDPV methods with Ru@VC/GCE nanosensor. Amongst different potential values between − 0.1 and 0.7 V, and different time values between 10 and 300 s; using 0 V accumulation potential and 30 s accumulation time, the highest oxidation peak of IDA was acquired, and these values were chosen as optimum conditions for further experiments (Fig. 7).

Modification effect of nanomaterial on the electrochemical response.
For the modification effect studies, firstly, Ru@VC suspension was prepared by dispersing nanomaterials in distilled water (1 mg/mL, ultrasonication for 2 h) utilizing an ultrasonic bath. Prior to the modification, the bare GCE's surface was polished with alumina slurry on a polishing cloth, cleaned by distilled water and dried. After that, 5 µL of Ru@VC suspension was dropped onto the surface of bare GCE and dried in a vacuum oven. For the comparison of bare and www.nature.com/scientificreports/ modified GCEs, the voltammetric behavior of 5 × 10 -6 M IDA was studied in phosphate buffer at pH 1.5 by DPV using first bare GCE and then Ru@VC/GCE. When the acquired voltammograms examined, it was understood that with Ru@VC modification, the peak current of IDA (Fig. 8b) increased 9 times compared to the bare GCE (Fig. 8a). It indicates that the oxidation of IDA is easier on Ru@VC/GCE than bare GCE. After performing scan rate studies and determining adsorption-controlled processes for the oxidation of IDA, 5 × 10 -6 M IDA was studied in pH 1.5 phosphate buffer by AdSDPV using bare GCE (Fig. 8c) and Ru@VC/GCE (Fig. 8d). There exist µL, 5 µL, 7 µL, and 10 µL of Ru@VC suspension was dropped to the surface of the electrode, drying in a vacuum oven. The voltammetric behavior of different Ru@VC suspension amount was studied using 5 × 10 -6 M IDA by AdSDPV (Fig. 9). The results showed that the highest I p was obtained with 10 µL of Ru@VC. On the other hand, it was hard to maintain a steady drop with 10 µL of nanomaterial on the electrode surface, and it caused a longer drying time, which makes this higher amount is a non-optimal condition. Thus, the second-best option, 7 µL, was preferred as the optimal nanomaterial amount and used in the subsequent studies.
Analytical characterization and validation of the nanosensor. Quantitative analysis of IDA was performed using Ru@VC/GCE sensor by the AdSDPV method under the selected optimum conditions at 0.0 mV accumulation potential, 30 s accumulation time. The calibration graph of I p vs. concentration of IDA gave a linear response among 5 × 10 -8 M and 1 × 10 -6 M (Fig. 10). The data obtained from this graph was listed in Table 2. The AdSDPV method calibration equation was given below: The values of LOD and LOQ were determined to utilize the following equations; where s is the standard deviation's response, and m is the calibration curve's slope [57][58][59] . The values of LOD and LOQ were calculated as 9.25 × 10 -9 M and 2.8 × 10 -8 M, as summarized in Table 2 with the reproducibility of peak current and potential. IDA was also determined in the human serum sample, and the calculated results were given in Table 2. The linear range was obtained between 5 × 10 -8 M and 2.5 × 10 -7 M IDA in human serum samples with the LOD and LOQ values of 7.24 × 10 -9 M and 2.19 × 10 -8 M, respectively. When we compare obtained results with the literature, better LOD responses were received from our previous study, where we used multiwalled carbon nanotubes 4 and from Arkan et. where they used TiO 2 nanoparticles and carbon nanofibers 42 that are summarized in Table 3.
Application to pharmaceutical dosage forms and human serum. To assess the applicability and validity of developed nanosensor, it was used for the determination of IDA in Idamen ® IV (10 mg/10 mL vial) and human serum samples using the standard addition method. Idamen ® IV (10 mg/10 mL vial) contains 10 mg of Idarubicin Hydrochloride as an active substance and water for injection as excipients 60 . The recovery results for Idamen ® IV and human serum samples were listed in Table 4. The results indicated that the proposed nanosensor could be successfully applied to pharmaceutical dosage forms and real samples with acceptable precision and accuracy results. www.nature.com/scientificreports/

conclusion
The current study paves ways to investigate voltammetric behavior to determine the anticancer drug IDA in human serum samples and a pharmaceutical dosage form by AdSDPV of a novel Ru@VC/GCE nanosensor that promises practical applications. The synthesized ruthenium/Vulcan carbon-based nanomaterials were characterized by XPS, TEM, and XRD. TEM analysis displayed that Ru metals dispersed well on Vulcan carbon with an average size of 1.98 nm. XRD analysis presented that the main phases in the ruthenium and Vulcan carbon correspond to an fcc structure. For electrochemical characterization, the enhanced effect of nanosensor was studied www.nature.com/scientificreports/ using CV and EIS techniques. In addition, scan rate, pH effect, nanomaterial amount, deposition time, and potential were also investigated for the selection of optimum conditions. As a result, the proposed new nanosensor showed enhancement for the oxidation peak current of IDA due to its improved electronic conductivity and electron transfer kinetics compared to bare GCE. The values of LOD and LOQ were calculated as 9.25 × 10 -9 M and 2.8 × 10 -8 M with a linear range among 5 × 10 -8 M and 1 × 10 -6 M. To demonstrate the applicability and validity of developed nanosensor, it was used for the determination of IDA in Idamen® IV (10 mg/10 mL vial) and human serum sample. The results of recovery studies showed that the Ru@VC/GCE nanosensor was free from excipient interferences in the dosage forms of injection, and it can be successfully applied to biological samples.