9596 | New J. Chem., 2015, 39, 9596--9604 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

Cite this: NewJ.Chem., 2015,

39, 9596

Investigation of magnetic and electrochemicalsensing properties of novel Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles

Nadir S. E. Osman,†*a Neeta Thapliyal,†b Thomas Moyo†a andRajshekhar Karpoormath†*b

Novel Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles were successfully synthesized using the glycol thermal route.

The X-ray diffraction study confirmed a well defined spinel phase structure of the sample. The

microstrain was investigated based on the Williamson–Hall plot. Crystallinity, shape and size of the

nanoparticles were studied using high resolution transmission electron microscopy and high resolution

scanning electron microscopy. Brunauer–Emmet–Teller measurement revealed that the sample has

a high surface area of 116 m2 g�1. The Barrett–Joyner–Halenda test showed that the sample is

mesoporous. The magnetization was found to increase from 66.5 � 0.3 emu g�1 at 300 K to 84.4 �0.5 emu g�1 at 4 K. Furthermore, the electrochemical sensing properties of Ba1/3Mn1/3Co1/3Fe2O4 nano-

particles were investigated using cyclic voltammetry. A glassy carbon electrode was modified using

the synthesized Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles. The modified electrode demonstrated excellent

electrocatalytic activity towards didanosine, an anti-HIV drug. A linear response to the drug concen-

tration was obtained in the range from 0.001 to 5.0 mM with a detection limit of 1.0 nM. The electrode

was highly stable, reproducible and was successfully used to determine trace amounts of didanosine in

human urine samples.

1 Introduction

Nanomaterials have emerged as an active area of interest amongresearchers because of their distinctive physical and chemicalproperties, such as high surface area, catalytic effect and masstransport, due to the quantum size effect.1 Recent years havewitnessed an upsurge in the exploration of spinel nanoferrites(SNPs) because of their useful magnetic and electrical propertiesfor technological applications in diverse fields such as ferrofluids,magneto-optics, spintronics, medical diagnostics, informationstorage systems, microwave absorbers and the manufacture ofanodes for batteries.2 The magnetic properties of SNPs can betuned by substituting different elements in their sub-lattices.3

Several factors, such as purity, homogeneity and microstructure,are known to affect the magnetic properties of these ferrites.4

A number of methods for the synthesis of SNPs have beenreported. Some of these methods are hydrothermal, co-precipitation,

electrospinning, sol–gel, combustion and conventional ceramictechniques.5 However, the ease and greater control of particlesize by wet chemistry methods make them preferable, comparedto solid state reaction techniques.6 SNPs have also been found toexhibit advantages in the electrochemical detection of a widerange of analytes due to their distinctive electrical and catalyticproperties.7,8 Studies are in progress to improve their electro-chemical sensing ability to the maximum possible extent. Addingdopants to ferrites is shown to vary their electronic conductivity,thus extending their application in the field of electroanalysis.9 Inthe present study, we have synthesized novel single phase spinelstructured Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles and investigated thestructure and properties of the nanomaterial obtained. Thepresence of magnetic ions in the sample has led us to investigateits magnetic properties. Literature survey reveals that iron-basedmagnetic nanoparticles act as excellent materials for electrodemodification in electroanalysis owing to their unique propertiessuch as large surface area and high surface energy.10,11 Magneticnanoparticles have also been found to function as electron-conducting pathways, allowing the rapid transfer of electronsbetween the redox system and the electrode surface.12,13 Thisinformation motivated us to study the electrochemical propertiesof the newly synthesized magnetic Ba1/3Mn1/3Co1/3Fe2O4 nano-particles. The study further investigates the electrochemical

a School of Chemistry and Physics, Westville Campus, University of KwaZulu-Natal,

P/Bag X5401, Durban 4000, South Africa. E-mail: nadir-shams@hotmail.comb Department of Pharmaceutical Chemistry, Discipline of Pharmaceutical Sciences,

College of Health Sciences, University of KwaZulu-Natal, Durban 4000,

South Africa. E-mail: Karpoormath@ukzn.ac.za

† All authors have contributed equally in this research work.

Received (in Montpellier, France)18th June 2015,Accepted 22nd September 2015

DOI: 10.1039/c5nj01547b

www.rsc.org/njc

NJC

PAPER

This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015 New J. Chem., 2015, 39, 9596--9604 | 9597

sensing ability of the synthesized nanoparticles for didanosine(DDI), an anti-HIV drug.

Didanosine (20,30-dideoxyinosine, DDI) (Scheme 1) is a purinenucleoside analogue, which is used to treat human immuno-deficiency virus (HIV) infection as a part of highly activeantiretroviral therapy (HAART).14 Approved in 1991 by the Foodand Drug Administration (FDA), it belongs to a class of antiretroviraldrugs called nucleoside reverse transcriptase inhibitors (NRTIs).15

Adverse effects of the drug include peripheral neuropathy,pancreatitis, retinal changes, optic neuritis and lactic acidosis.16

The determination of DDI is very important for quality control andobtaining the optimal therapeutic drug concentration. A number ofanalytical methods have been reported for the estimation of thedrug, with the majority of them involving high-performance liquidchromatography and spectrophotometric techniques, which areexpensive as well as time-consuming.17–20 Due to their highsensitivity, specificity, low cost of instrumentation and simplicity,the electroanalytical methods have proved to be expedient forthe quantitative analysis of several drugs in bulk amounts,pharmaceutical formulations and biological fluids. However, todate, very few voltammetric methods have been reported for DDIdetermination. Ozoemena et al.21 fabricated electrochemicalsensors, based on metallopthalocyanine (MPc, M = Co, Fe)complexes and impregnated a carbon paste electrode to deter-mine the drug. Later, Farias et al.22 developed an adsorptivestripping method for the quantification of DDI at a mercury filmelectrode. Metal ferrite nanoparticles show promising applica-tion in the field of electroanalysis.7 Thus, in the present study, anovel electrochemical sensor based on the Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles and a modified glassy carbon electrode was developedfor the sensitive determination of DDI. The electrochemical behaviorof the drug was studied using cyclic voltammetry (CV) and themodified electrode was validated for the quantification of the drug inhuman urine samples.

2 Experimental

Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles were synthesized by theglycol thermal method in a Watlow series model PARR 4843stirred pressure reactor. The starting materials (BaCl2�6H2O:99%, MnCl2�6H2O: 99%, CoCl2�4H2O: 98% and FeCl2�6H2O:99%) were purchased from Sigma-Aldrich. Stoichiometric metalchlorides were dissolved in deionized water using a magnetic

stirrer for 30 minutes. Drops of NH4OH were added slowly tothe mixture until a pH of 9 was reached. The precipitate waswashed using deionized water over a Whatman glass microfiberfilter. A standard solution of silver nitrate was used to confirmthat all the chloride ions had been removed. The wet precipitatewas reacted in ethylene glycol at 200 1C in a pressure reactorunder continuous stirring for 6 hours at a pressure of about100 Psi. After the reaction, the mixture was washed withdeionized water and finally with ethanol. The final productwas dried using a 200 W infra-red lamp. The powder washomogenized using an agate mortar and pestle. The phaseand structural characterization of the sample was conductedusing a Phillips X-ray diffractometer (Model: PANalytical,EMPYREAN) using Co-Ka radiation (l = 1.7903 Å). The morphologyand micro-structure of the nanoparticles were investigated byhigh-resolution transmission electron microscopy (HRTEM)(Jeol_JEM-1010) and high-resolution scanning electron micro-scopy (HRSEM) (Ultra Plus ZEISS-FEG HRSEM instrument). Thetextural and porosity properties of the nanoparticles were inves-tigated using a Micromeritics TriStar II 3020 instrument, withliquid N2 as the analysis gas. A mini cryogen free magnetizationmeasurement system (Cryogenic Ltd, UK) was used to performlow temperature magnetization measurements from 4 to 300 Kwithin magnetic fields of up to 5 Tesla.

For the electrochemical experiments, DDI and potassiumferricyanide were obtained from Sigma-Aldrich, USA. All otherchemicals used were of analytical reagent grade and purchasedfrom Merck. A stock solution (2.00 � 10�3 M) of DDI wasprepared in double distilled water and stored in the dark whennot in use. Phosphate buffer solutions (PBS; m = 0.2 M) ofvarious pH values were used as supporting electrolyte. Purenitrogen gas was purged into the DDI containing solution for atleast 3 minutes before each experiment. The electrochemicalmeasurements were performed on an electrochemical analyzer(Model 800B Series, CH Instruments, Inc.) and conducted in asingle-compartment three-electrode glass cell, which consistedof a reference electrode (Ag/AgCl), a counter electrode (platinumwire) and a working electrode (a bare/modified glassy carbonelectrode). For sample preparation, the urine samples (obtainedfrom laboratory personnel) were diluted 100 times by phosphatebuffer of pH 7.2. The dilution process helps in reducing thematrix effect.23 The drug was then quantified in spiked humanurine samples using CV.

Before fabrication of the modified electrode, the glassycarbon electrode (GCE) was cleaned by polishing with 0.05 micronalumina slurry on a polishing cloth for about 30 seconds. Theelectrode was then rinsed thoroughly with double distilled waterand dried in air. A 0.5 mg mL�1 suspension of Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles in dimethyl formamide was prepared. 10.0 mL of thesuspension was then cast onto the cleaned electrode surface anddried to constitute Ba1/3Mn1/3Co1/3Fe2O4/GCE. The active surfacearea of bare and modified GCE was measured by recording CVs in1.0 mM K3[Fe(CN)6] solution at different scan rates and using theRandles–Sevcik equation.24 The active surface areas were found to be0.070 cm2 and 0.218 cm2 for bare GCE and Ba1/3Mn1/3Co1/3Fe2O4/GCE, respectively.

Scheme 1 Chemical structure of didanosine.

Paper NJC

9598 | New J. Chem., 2015, 39, 9596--9604 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

3 Results and discussion

In the present study, Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles weresuccessfully synthesized using the glycol thermal route. Thestructure, morphology, magnetic and electrochemical propertiesof the sample were investigated by XRD, HRTEM, HRSEM,magnetization and voltammetry measurements.

3.1 Characterization of Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles

Fig. 1 depicts the X-ray powder diffraction (XRD) patterns ofBa1/3Mn1/3Co1/3Fe2O4 nanoparticles. The XRD shows a well-defined spinel structure with no impurity peaks. The latticeparameter a was calculated using Bragg’s equation, where d is theinter-planar spacing and hkl are the Miller indices. The averagecrystallite size was determined using Scherrer’s equation, where b isthe full-width at half-maximum of the most intense (311) XRD peakand y is the Bragg’s angle.25 The microstrain was determined usingthe Williamson–Hall plot, as displayed in Fig. 2. The distributionof the data points reflects the homogeneity of the microstrain.

The lattice parameter, crystallite size and microstrain for the pre-pared Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles were found to be 0.839 �0.003 nm, 8.37 � 0.06 nm and 0.0015 � 0.0001, respectively.

Fig. 3 shows images of the as-prepared Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles captured via high resolution transmission electronmicroscopy (HRTEM) and high resolution scanning electronmicroscopy (HRSEM). The images in Fig. 3A and B show quasi-spherical shapes. A mono-dispersive characteristic was observed.The lattice fringes are well defined (as seen in Fig. 3A), indicatingthat the particles have a well formed crystalline structure. Themarked d spacing of 0.252 nm is anticipated to reflect 311 planes.The average particle sizes are found to be 9.3� 2 nm. Fig. 3C showsthat the particles are well dispersed. The HRTEM and HRSEMimages in Fig. 3 support the XRD result in terms of crystallinity andsize of the synthesized nanoparticles.

The sample revealed a high surface area of 116 m2 g�1, asdeduced from Brunauer–Emmet–Teller (BET) surface areameasurement. The texture and character of the pores were studiedusing isothermal N2 adsorption–desorption measurements.Fig. 4 shows N2 gas adsorption isotherms for the as-preparedBa1/3Mn1/3Co1/3Fe2O4 nanoparticles. The adsorption of N2

increased to a maximum value of 222 cm3 g�1 at a relativepressure (P/P0) of 99%, whilst desorption of the gas decreasedto a lowest value of 19 cm3 g�1, as the P/P0 decreased to 0.01%.The adsorption and desorption curves were found to merge,indicating that a hysteresis loop is associated with N2 conden-sation inside the sample pores.26 The hysteresis loop can beclassified into a type IV isotherm, which suggests the meso-porous property of the sample.27 Barrett–Joyner–Halenda (BJH)measurement was carried out to explore the scattering of pores.Most of the scatter points were placed in the range between 2and 47 nm, further confirming the mesoporous property of thenanoparticles.28 the sharp distribution peak of the pores (insetof Fig. 4) indicates a homogeneous pore distribution.

Fig. 1 X-ray diffraction pattern for the synthesized Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles.

Fig. 2 Plot of bcosy versus 4 siny for the synthesized Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles.

Fig. 3 HRTEM image with 20 nm magnification (A), HRTEM image with100 nm magnification (B), and HRSEM image (C) for the synthesizedBa1/3Mn1/3Co1/3Fe2O4 nanoparticles.

NJC Paper

This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015 New J. Chem., 2015, 39, 9596--9604 | 9599

Fig. 5 depicts M–H loops at different measuring tempera-tures for the as-prepared Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles.The measurements were achieved in external magnetic fields inthe range from�5 T to +5 T. The results show an increase in themagnetization from 66.5 � 0.3 emu g�1 to 84.4 � 0.5 emu g�1,as the measuring temperature decreased from 300 K to 4 K. Thecoercivity increased from 0.009 � 0.004 T to 1.010 � 0.004 Twith decrease in temperature from 300 K to 4 K. The significantincrease in the coercivity value with decrease in temperatureindicates that the sample becomes magnetically harder at lowtemperatures, which is related to the spin freezing effect.Moreover, the sample shows a distortion in the hysteresis loopsat low temperature (as seen in Fig. 5). This distortion isanticipated to be due to the combined effect of soft and hardmagnetic phases. The spin disorder in the surface layer is alsoexpected to lead to such behavior.29

The approach to the saturation of the magnetization (M) wasused to obtain the values of MS(T) as a function of the measuringtemperature. Fig. 6 exhibits the initial magnetization curves forthe as-prepared Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles. The bestfitting curves (in red) obtained for the initial magnetizationcurves of the sample are based on the empirical law of approach

to saturation magnetization, M ¼MSðTÞ 1� a

H� b

H2

� �þ wH,

where w is the high-field susceptibility and the parameters, aand b, are constants.30 The magnetic parameters (maximummagnetization, saturation magnetization, coercivity, remanentmagnetization and the squareness of the hysteresis loop) of theas-prepared Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles are summarizedin Table 1.

The values of the remanent magnetization (Mr) and thesquareness of the hysteresis loop Mr/MS were found to increasedramatically with decrease in temperature. As reported in theliterature, Mr values depend on magnetic anisotropy, particlemicrostructure as well as stress sensitivity.31 Fig. 7 shows thatthe decline in Mr is related to the reduction in magneticanisotropy strength. The inset in Fig. 7 displays the variation

Fig. 5 Hysteresis loop measurements at different temperatures foras-prepared Ba1/3Mn1/3Co1/3Fe2O4 ferrite nanoparticles.

Fig. 6 Initial magnetization curves at different temperatures for theas-prepared Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles.

Table 1 Maximum magnetizations at 5 T (Mm), saturation magnetizationsobtained from empirical law of approach to saturation (MS), coercive fields(HC), remanent magnetizations (Mr) and squareness of the hysteresis loops(Mr/MS) for the Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles

T (K)Mm (emu g�1)� 0.5

MS (emu g�1)� 0.1

HC (T)� 0.004

Mr (emu g�1)� 1 Mr/Ms

4 84.4 90.5 1.010 60 0.71610 84.1 90.4 0.948 60 0.65920 83.8 90.2 0.886 59 0.65040 83.5 89.6 0.742 55 0.61460 82.5 88.7 0.599 51 0.57380 82.3 87.7 0.461 46 0.522100 81.6 86.7 0.325 39 0.483120 80.5 85.1 0.211 32 0.372140 79.8 84.0 0.153 26 0.309180 77.0 80.4 0.066 14 0.169200 75.1 78.5 0.042 8 0.100240 71.8 74.9 0.042 5 0.069280 68.3 72.9 0.010 4 0.075300 66.5 67.6 0.009 4 0.071

Fig. 4 N2 gas adsorption isotherms for the as-prepared Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles. The inset shows the pore size distribution.

Paper NJC

9600 | New J. Chem., 2015, 39, 9596--9604 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

of Mr/MS with change in temperature. It is observed that as thetemperature increases, Mr/MS monotonically decreases. Basedon Stoner–Wohlfarth theory, the value of Mr/MS for cubic anduniaxial anisotropy was found to be 0.832 and 0.5, respec-tively.32 At a low temperature (T o 80 K), the values of Mr/MS

were considerably higher than 0.5, as can be seen in the insetof Fig. 7, suggesting that at a low temperature the as-preparedBa1/3Mn1/3Co1/3Fe2O4 nanoparticles tend to have cubic magneto-crystalline anisotropy.

Fig. 8 depicts the variation of the coercivity (HC) as afunction of the measuring temperature. Significant increasein HC was observed at temperatures below 200 K. This wasassociated with the thermal oscillation of blocked momentsthrough the anisotropy barrier.33 The temperature dependenceof HC for non-interacting, mono-domain particles in the tem-perature range below the blocking temperature TB, is described

by the Kneller’s law HC(T) = HC(0)[1 � (T/TB)a], where HC(0) isthe coercivity at 0 K and a is a constant.34 For modifiedKneller’s law, the value of a is 0.5. The HC was found to followTa law with a = 0.77, as reported by Pfeiffer et al.35 The collecteddata of HC as a function of measuring temperature was plottedand a correlation coefficient of 0.9955 was obtained. The valueof HC(0), TB and a were found to be 1.08 � 0.03 T, 166� 5 K and0.77 � 0.07, respectively. The inset in Fig. 8 demonstrates thevariation of HC with T0.77 for the as-prepared Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles.For ferromagnetic and ferrimagnetic systems, the tempera-

ture dependence of magnetization is associated with the spinwaves (magnons), where the magnetization decreases withincrease in the measuring temperature. The thermal behaviorof the magnetization is usually described by Bloch’s law MS(T) =MS(0)[1 � (T/T0)b], where MS(0) is the saturation magnetizationat 0 K, T0 is the temperature at zero magnetization, 1/T0 is theBloch’s constant that depends on the nanoparticle structureand b is the Bloch’s exponent.36 This model is effective for bulksystems with b of 3/2. For the nano-scale materials, the mag-nons with wavelength larger than the particle size (due of finitesize effect) need thermal energy to be excited. Thus, themodified Bloch’s law with a b value greater than 3/2 is expectedto be an appropriate model to describe such a system. As seenin Fig. 9, the temperature dependence of magnetization is fittedin accordance with the modified Bloch’s law with a correlationcoefficient of 0.9995. The values of MS(0), T0 and b, determinedfrom the fitting curves, are 90.5 � 0.1 emu g�1, 704 � 10 K and1.62 � 0.02, respectively.

3.2 Electrochemical behavior towards didanosine

3.2.1 Electrochemical activity. The electrochemical activityof the synthesized Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles towards DDIwas investigated using CV. Fig. 10 depicts cyclic voltammograms

Fig. 7 Variation of the remanent magnetization (Mr) plotted as a function ofthe measuring temperature. The inset shows the variation of the reducedremanent magnetization (Mr/MS) with temperature of Ba1/3Mn1/3Co1/3Fe2O4

nanoparticles.

Fig. 8 Coercivity temperature dependency for the as-preparedBa1/3Mn1/3Co1/3Fe2O4 nanoparticles. The red line in the inset shows theT0.77 dependence.

Fig. 9 Thermal dependency of saturation magnetization for the as-preparedBa1/3Mn1/3Co1/3Fe2O4 nanoparticles. (The red line shows the fitting curveaccording to the modified Bloch’s law.)

NJC Paper

This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015 New J. Chem., 2015, 39, 9596--9604 | 9601

recorded for 0.5 mM DDI in 0.1 M PBS (pH 7.2) at bare GCE andBa1/3Mn1/3Co1/3Fe2O4/GCE. No voltammetric peak was observedat the bare electrode. However, the voltammetric responseimproved significantly at the modified electrode, exhibitinga distinct peak at 0.75 V. This electrocatalytic activity of theBa1/3Mn1/3Co1/3Fe2O4/GCE is attributed to the unique proper-ties of the synthesized nanoparticles and the large surface areaof the electrode. Because the electrode is modified with transi-tion metal nanoparticles, it is anticipated that when an electro-chemical reaction takes place, these transition metal ionsfunction as active sites,37 acting as catalyst for the oxida-tion of DDI. Thus, Ba1/3Mn1/3Co1/3Fe2O4/GCE demonstratedan enhanced electrochemical sensing property as comparedto the bare GCE.

3.2.2 Effect of scan rate and pH. The effect of scan rate (v)on the electrochemical behavior of DDI at Ba1/3Mn1/3Co1/3Fe2O4/GCE was investigated in the range of 10–200 mV s�1. The peakcurrent (ip) was found to increase linearly with an increase in v(as seen in Fig. 11(A)), indicating that the electrooxidation of DDIon the modified electrode is a typical adsorption-controlledprocess.38 The relationship between ip and v is expressed bythe equation ip (mA) = 0.0087v + 0.1239, with a correlationcoefficient of 0.9967. Furthermore, log ip was plotted againstlog v (Fig. 11(B)) and a linear relationship was obtained,expressed by the equation log iP = 0.8089 log v � 1.6004 havinga correlation coefficient of 0.9937. A slope of 0.8 was observed,which is close to the theoretical value of 1.0 expected for an idealadsorption-controlled electrode process.39 This confirmed thatthe electrooxidation of DDI was adsorption-controlled. The peakpotential (Ep) was also observed to shift towards more positivevalues with an increase in v, suggesting the irreversible nature ofthe electrochemical process.40

On plotting Ep versus log v (Fig. 11(C)), a straight line wasobserved that followed the equation EP = 0.0681 log v + 0.6134,with a correlation coefficient of 0.9995. For an irreversibleelectron transfer, the Ep is expressed by the Laviron’s equation,

EP ¼ E00 þ 2:30RTanF

� �log

RTk0

anF

� �þ 2:30RT

anF

� �log v

where E00 is the formal redox potential, n is the number ofelectrons transferred in the electrochemical process, a is theelectron transfer coefficient, k0 is the standard heterogeneousrate constant of the reaction, and the other mentioned symbolshave their usual meanings.41 The value of an is calculated fromthe slope of the Ep versus log v plot. In the given system, the slopewas found to be 0.0681 V. Considering R = 8.314 J K�1 mol, F =96 480 C mol�1 and T = 298 K, an was found to be 0.87. The value

of n was then calculated using the equation a ¼ 47:7

Ep � Ep=2

, where

Ep/2 is the half-peak potential.42 The value of n was found to be1.64 (B2), indicating the involvement of two electrons in theoxidation of DDI at Ba1/3Mn1/3Co1/3Fe2O4/GCE.

The pH of the supporting electrolyte exhibits significantinfluence on the electrooxidation of DDI at the modifiedelectrode. Fig. 11(D) depicts the effect of pH on Ep of 0.5 mMDDI using Ba1/3Mn1/3Co1/3Fe2O4/GCE in 0.1 M PBS in the pHrange of 3.1–9.0. It was observed that as the pH value of thesolution increased, the Ep shifted linearly towards less positivepotentials, with a slope of 54.2 mV pH�1. This value was closeto the theoretical value of 59 mV pH�1, which is indicated forelectrochemical processes involving the same number of elec-trons and protons.43 Thus, it is suggested that two electronsand two protons participate in the electrooxidation of DDI, andthe anticipated reaction mechanism is depicted in Scheme 2.

3.2.3 Analytical performance. Cyclic voltammograms of DDIwith varying concentrations were recorded at Ba1/3Mn1/3Co1/3Fe2O4/GCE under optimum experimental conditions. The oxidation peakcurrent (peak height) was found to increase linearly with DDIconcentration in the range from 1.0 � 10�9 M to 5.0 � 10�6 M(as depicted in Fig. 12). The observed linear equation was ip

Fig. 10 Typical cyclic voltammograms observed (a) in the absence of DDIat Ba1/3Mn1/3Co1/3Fe2O4/GCE and in the presence of 0.5 mM DDI at (b) bareGCE and (c) Ba1/3Mn1/3Co1/3Fe2O4/GCE in 0.1 M PBS (pH 7.2) at a scan rateof 100 mV s�1.

Fig. 11 (A) Observed dependence of peak current on scan rate; (B) varia-tion of the logarithm of peak current with the logarithm of scan rate; (C) plotof Ep versus logarithm of scan rate; and (D) dependence of Ep on pH for0.5 mM DDI in 0.1 M PBS.

Paper NJC

9602 | New J. Chem., 2015, 39, 9596--9604 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

(10�6 A) = 2178.6C (R2 = 0.9989), where C is the concentration ofDDI (in mM). The detection limit (LOD) was found to be 1.0 nM.

The reproducibility and stability of Ba1/3Mn1/3Co1/3Fe2O4/GCE was also investigated. A series of five modified electrodeswere fabricated using the same procedure and then used for thedetermination of 0.5 mM DDI in 0.1 M PBS. A relative standarddeviation (RSD) of ip values recorded at the five electrodeswas found to be 4.2%, suggesting the good fabrication repro-ducibility of the proposed electrode. For six successive deter-minations of 0.5 mM DDI solution at the same modifiedelectrode, the calculated RSD was 3.6%. Moreover, the CV responseof Ba1/3Mn1/3Co1/3Fe2O4/GCE was investigated for the same DDIconcentration, after leaving the electrode unused for seven days.The peak potential did not exhibit any change and the peak currentresponse also showed insignificant variation, suggesting good

stability of the modified electrode. The selectivity of Ba1/3Mn1/

3Co1/3Fe2O4/GCE towards determination of 0.5 mM DDI was exam-ined in the presence of organic species commonly present in realsamples. The tolerance limit was fixed at the concentration ofinterfering species exhibiting a signal change of 5% or more. It wasfound that over 10-fold excess concentration of ascorbic acid,dopamine and uric acid did not interfere with the voltammetricresponse of DDI, suggesting good selectivity of the method.

To evaluate the analytical applicability of the proposed method,assays were performed for two human urine samples. The sampleswere spiked with varying quantities of DDI and the recovery studyof the drug was conducted. The observed results (summarized inTable 2) show satisfactory recoveries, thus validating that theproposed method has good accuracy. Fig. 13 displays the cyclicvoltammogram obtained for DDI spiked human urine sample(sample 1), where a peak is observed at 0.75 V for the presence ofDDI. An additional peak is observed at B0.4 V, which is attributed tothe existence of uric acid in the urine sample, suggesting that thebiological substances present in urine did not show any interferencein the determination of the drug. Thus, the method can be usedefficiently for the quantification of DDI in real matrices.

4 Conclusions

The present study reports the simultaneous substitution of Ba,Mn and Co into the ferrite spinel structure using the glycolthermal method. The spinel phase structure of the sample was

Scheme 2 Proposed electro-oxidation mechanism of DDI.

Fig. 12 Calibration plot observed for DDI at Ba1/3Mn1/3Co1/3Fe2O4/GCE atpH = 7.2.

Table 2 Recovery results obtained for DDI in human urine sample at Ba1/3Mn1/3Co1/3Fe2O4/GCE

Sr. no.

Added (M) Found (M) Recoveries (%)

Sample 1 Sample 2 Sample 1 Sample 2 Sample 1 Sample 2

1 1.00 � 10�9 1.00 � 10�9 9.91 � 10�10 1.01 � 10�9 99.1 101.02 1.00 � 10�8 1.00 � 10�8 9.84 � 10�9 9.92 � 10�9 98.4 99.23 1.00 � 10�7 1.00 � 10�7 9.87 � 10�8 1.01 � 10�7 98.7 101.04 5.00 � 10�7 5.00 � 10�7 4.96 � 10�7 5.04 � 10�7 99.2 100.85 1.00 � 10�6 1.00 � 10�6 1.01 � 10�6 1.02 � 10�6 101.4 102.3

Fig. 13 Cyclic voltammograms of (a) blank urine sample 1 and (b) urinesample 1 spiked with 0.5 mM standard solution of DDI, under optimumexperimental conditions.

NJC Paper

This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015 New J. Chem., 2015, 39, 9596--9604 | 9603

identified using X-ray powder diffraction. Microstrain wasestimated from the Williamson–Hall plot, with the scattered datashowing homogeneity in the sample microstrain. High-resolutiontransmission electron microscopy and high-resolution scanningelectron microscopy were used to monitor the morphology of theas-prepared Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles. The results showmono-dispersive crystallite particles having a high value for surfacearea, as deduced from Brunauer–Emmet–Teller testing. The samplewas found to possess mesoporous character, as observed fromBarrett–Joyner–Halenda measurement. The magnetization investi-gation revealed that the sample tends to become magneticallyharder at low temperatures. The spontaneous magnetizationincreased with decrease in temperature. The remanent magnetiza-tion and reduced remanent magnetization values were also foundto increase with decrease in the measuring temperature. Tempera-ture dependence of coercivity and magnetization were observed tofollow Kneller’s law and a modified Bloch’s law, respectively. Theelectrocatalytic activity of the synthesized nanoparticles was inves-tigated by fabricating Ba1/3Mn1/3Co1/3Fe2O4 nanoparticles on amodified glassy carbon electrode as a voltammetric sensor towardsdidanosine. The modified electrode, with advantages such assimple fabrication procedure, excellent reproducibility, wide linearconcentration range and a significantly low detection limit, wassuccessfully applied for the quantification of the drug didanosinein human urine samples, thus offering a promising substitute tothe commonly reported chromatographic and spectrophotometricmethods for quantification of didanosine.

Acknowledgements

The authors would like to thank the National Research Foundation(NRF) of South Africa for VSM and mini cryogen free measurementequipment grants, the Department of Pharmaceutical Chemistry(College of Health Sciences, University of KwaZulu-Natal, SouthAfrica) for financial support (NT) and Sudan University of Scienceand Technology for the study leave (NSEO).

References

1 G. B. H. Sohn and R. E. Cohen, Chem. Mater., 1997, 9,264–269.

2 C. Vazquez-Vazquez, M. A. Lopez-Quintela, M. C. Bujan-Nunez and J. Rivas, J. Nanopart. Res., 2011, 13, 1663–1676.

3 R. K. Gopathi, V. K. Katrapally and V. Y. Chetty, Mater. Sci.Appl., 2012, 3, 87–91.

4 Y. Purushotham, M. Singh, S. P. Sud and P. V. Reddy, Int.J. Mod. Phys. B, 1998, 12, 2247–2262.

5 M. P. Reddy, X. Zhou, A. Yann, S. Du, Q. Huang andA. M. A. Mohamed, Superlattices Microstruct., 2015, 81,233–242.

6 S. Diodati, L. Nodari, M. M. Natile, U. Russo, E. Tondello,L. Lutterotti and S. Gross, Dalton Trans., 2012, 41,5517–5525.

7 V. A. Kumary, J. Divya, T. E. M. Nancy and K. Sreevalsan, Int.J. Electrochem. Sci., 2013, 8, 6610–6619.

8 A. A. Ensafi, F. Saeid, B. Rezaei and A. R. Allafchian, Anal.Methods, 2014, 6, 6885–6892.

9 B. Ul-ain, S. Ahmed, M. A. ur Rehman, Y. Huang andC. A. Randall, New J. Chem., 2013, 37, 2768–2777.

10 M. Hasanzadeh, N. Shadjou and M. de la Guardia, TrendsAnal. Chem., 2015, 72, 1–9.

11 G. Absalan, M. Akhond, A. Bananejad and H. Ershadifar,J. Iran. Chem. Soc., 2015, 12, 1293–1301.

12 D. Grieshaber, R. MacKenzie and J. Voros, Sensors, 2008, 8,1400–1458.

13 H. Wei and E. Wang, Luminescence, 2011, 26, 77–85.14 B. Roca, C. Lapuebla and B. Vidal-Tegedor, Int. J. Infect. Dis.,

2005, 9, 195–200.15 B. A. Ahidjo, R. Veale, A. G. Duse, P. Becker and E. Marais,

Int. J. Antimicrob. Agents, 2008, 32, 186–191.16 C. E. Reust, Am. Fam. Physician, 2011, 83, 1443–1451.17 A. M. C. de Oliveira, T. C. R. Loewen, L. M. Cabral, E. M. dos

Santos, C. R. Rodrigues, H. C. Castro and T. C. dos Santos,J. Pharm. Biomed. Anal., 2005, 38, 751–756.

18 H. Sura, S. Bonthu and T. E. G. K. Murthy, J. Adv. Pharm.Educ. Res., 2013, 3, 187–195.

19 R. N. Kane, P. S. Bhokare, C. C. Nalawade, M. S. Sayyed andR. D. Paliwal, Int. J. Pharm. Chem. Sci., 2012, 1, 2277–5005.

20 J. N. Sangshetti, P. A. Kulkarni and D. B. Shinde, TrendsAppl. Sci. Res., 2007, 2, 71–75.

21 K. I. Ozoemena, R. I. Stefan-van Staden and T. Nyokong,Electroanal, 2009, 21, 1651–1654.

22 P. A. M. Farias, A. A. Castro and A. I. P. Cordoves,J. Electrochem. Sci. Eng., 2012, 2, 133–142.

23 M. M. I. Khan, A. M. J. Haque and K. Kim, J. Electroanal.Chem., 2013, 700, 54–59.

24 R. O. Kadara, N. Jenkinson and C. E. Banks, Sens. Actuators,B, 2009, 138, 556–562.

25 S. Sutradhar, K. Mukhopadhyay, S. Pati, S. Das, D. Das andP. K. Chakrabarti, J. Alloys Compd., 2013, 576, 126–133.

26 C. P. Bean and J. D. Livingston, J. Appl. Phys., 1959, 30,120S–129S.

27 M. Kruk and M. Jaroniec, Chem. Mater., 2001, 10,3169–3183.

28 P. B. Lihitkar, S. Violet, M. Shirolkar, J. Singh, O. N. Srivastava,R. H. Naik and S. K. Kulkarni, Mater. Chem. Phys., 2012, 133,850–856.

29 R. H. Kodama, C. L. Seaman, A. E. Berkowitz and M. B. Maple,J. Appl. Phys., 1994, 75, 5639–5641.

30 J. Liu, W. Ren, D. Li, N. Sun, X. Zhao, J. Li and Z. Zhang,Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75,064429/1.

31 G. F. Dionne, Proc. IEEE, 1975, 63, 777–789.32 E. C. Stoner and E. P. Wohlfarth, Trans. - R. Soc., 1948, A240,

599–644.33 K. Maaz, A. Mumtaz, S. K. Hasanain and M. F. Bertino,

J. Magn. Magn. Mater., 2010, 322, 2199–2202.34 K. Maaz, S. Karim, K. J. Lee, M. H. Jung and G. H. Kim,

Mater. Chem. Phys., 2012, 133, 1006–1010.35 H. Pfeiffer and W. Schueppel, Phys. Status Solidi A, 1990,

119, 259–269.

Paper NJC

9604 | New J. Chem., 2015, 39, 9596--9604 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

36 R. H. Kodama, J. Magn. Magn. Mater., 1999, 200, 359–372.37 D. Ye, Y. Xu, L. Luo, Y. Ding, Y. Wang and X. Liu, J. Solid

State Electrochem., 2012, 16, 1635–1642.38 E. Laviron, A. Vallat and R. Maunier-Prest, J. Electroanal.

Chem., 1994, 379, 427–435.39 D. K. Gosser, Cyclic Voltammetry: Simulation and Analysis of

Reaction Mechanisms, VCH, New York, 1993, p. 43.

40 J. I. Gowda and S. T. Nandibewoor, Asian J. Pharm. Sci., 2014,9, 42–49.

41 E. Laviron, J. Electroanal. Chem., 1979, 101, 19–28.42 A. J. Bard and L. R. Faulkner, Electrochemical methods: Funda-

mentals and applications, Wiley, New York, 2nd edn, 2004, p. 236.43 L. Fotouhi, M. Fatollahzadeh and M. M. Heravi, Int.

J. Electrochem. Sci., 2012, 7, 3919–3928.

NJC Paper