Selective Iron(III) ion uptake using CuO-TiO2 nanostructure by inductively coupled plasma-optical emission spectrometry

Background CuO-TiO2 nanosheets (NSs), a kind of nanomaterials is one of the most attracting class of transition doped semiconductor materials due to its interesting and important optical, electrical, and structural properties and has many technical applications, such as in metal ions detection, photocatalysis, Chemi-sensors, bio-sensors, solar cells and so on. In this paper the synthesis of CuO-TiO2 nanosheets by the wet-chemically technique is reported. Methods CuO-TiO2 NSs were prepared by a wet-chemical process using reducing agents in alkaline medium and characterized by UV/vis., FT-IR spectroscopy, X-ray photoelectron spectroscopy (XPS), powder X-ray diffraction (XRD), and field-emission scanning electron microscopy (FE-SEM) etc. Results The structural and optical evaluation of synthesized NSs were measured by XRD pattern, Fourier transform infrared (FT-IR) and UV–vis spectroscopy, respectively which confirmed that the obtained NSs are well-crystalline CuO-TiO2 and possessing good optical properties. The morphological analysis of CuO-TiO2 NSs was executed by FE-SEM, which confirmed that the doped products were sheet-shaped and growth in large quantity. Here, the analytical efficiency of the NSs was applied for a selective adsorption of iron(III) ion prior to detection by inductively coupled plasma-optical emission spectrometry (ICP-OES). The selectivity of NSs towards various metal ions, including Au(III), Cd(II), Co(II), Cr(III), Fe(III), Pd(II), and Zn(II) was analyzed. Conclusions Based on the selectivity study, it was confirmed that the selectivity of doped NSs phase was the most towards Fe(III) ion. The static adsorption capacity for Fe(III) was calculated to be 110.06 mgg−1. Results from adsorption isotherm also verified that the adsorption process was mainly monolayer-adsorption onto a surface containing a finite number of CuO-TiO2 NSs adsorption sites.


Introduction
Transition-metal doped semiconductor nanomaterial have attracted significant attention due to their potential applications in fabrication of nano-scale electronics, electroanalytical, selective metal-ions detection, opto-electronics, biological devices, electron field-emission sources for emission displays, bio-chemical detections, surface-enhanced Raman properties, and immense-potential applications etc. [1,2]. It is exhibited a regular morphological nanostruc ture, which composed a number of regular phases with geometrically-coordinated metals and oxide atoms along the axes. Doped materials have also concerned significant research effort for their exceptional and outstanding properties as well as versatile applications [3,4]. In last decade, nano-sized transition-metal oxides have been widely studied as promising anodes for Laser-Induced Break-down Spectroscopy's since they were first reported by Tarascon et al. [5,6]. Among them, undoped copper oxide (CuO) has attracted much interest owing to their high theoretical capacity, highly-stable, non-toxic, economical approach, and facile synthesis. It is a p-type semiconductor material with a band-gap energy [7][8][9], which is studied for various applications in bio-materials [10], photo-conductive [11] electro-magnetic [12], and super-conductors mico-devices [13] etc. Various efforts have been focused toward the fabrication of nano-structured CuO to improve their performance in currently existing applications, which is considered as one of the promising artificial mediators owing to their properties and functionalities [14][15][16][17][18][19][20][21]. However, it is still a big challenge to attain high-rate capability and crystallinity of copper oxide doped semiconductor nanomaterials. It is well accepted that a smaller size of CuO can lead to higher capacity and higher surface capability. This reduces the over potential and allows faster reaction kinetics to detect the metal ions [22][23][24]. Therefore, the synthesis of low-dimensional CuO nanostructures is a promising approach to make a significant improvement of largesurface area and high-aspect-ratio in doped nanostructures. Due to the significant properties of semiconductors, the doped nanomaterials were attained a considerable attention in terms of controlled growth of crystalline materials in huge quantity. Various growth mechanism have been employed, including the vapor-liquid-solid growth [25], thin-film growth [26], vapor-solid growth [27], wetchemical methods [28], and electro-spinning [29] etc. Titanium dioxide is the promising host-material as semiconductor having high photo-chemical stability and largesurface area with low economical-cost. Well-dispersed undoped titania dioxide nanostructure materials with very fine sizes are promising in many significant applications such as pigments, adsorbents and catalytic supports [30][31][32]. In almost all of these cases, when the dimension of nanomaterials is reduced significantly, especially to several nano-meter scales, some novel optical, morphological, and structural properties are expected, owing to the large surface-to-volume ratio [33]. In addition, the development of simple, rapid and efficient techniques has achieved a huge interest for monitoring metal ions in the environment. Several analytical methods have been applied to analyze metal ions in aqueous solutions, such as atomic absorption spectrometry [34], inductively coupled plasmaoptical emission spectrometry (ICP-OES) [35], anodic stripping voltammetry [36], and ion chromatography [37]. However, analytical methods can not directly measures the metal ions, in particular at ultra-trace concentration, in aqueous systems due to the lack of sensitivity and selectivity of these methods. Therefore, an efficient separation procedure is urgently required prior to the determination of noble metals for sensitive, accurate and interferencefree determination of noble metals [38].
Several analytical methods are introduced for separation of analytes including liquid-liquid extraction [39], ion-exchange [40], co-precipitation [41], cloud-point extraction [42] and solid-phase extraction (SPE) [43]. SPE is considered to be one of the most powerful methods because it minimized the solvent usage and exposure, disposal costs, and extraction time for sample preparation. Several adsorbents have appeared due to the popularity of SPE for selective extraction of analytes, such as alumina [44], C18 [45], molecular imprinted polymers, cellulose [46], silica-gel [47,48], activated carbon [49,50] and carbon nanotubes [51,52]. This study was also planned to perform the analytical efficiency of CuO-TiO 2 NSs phase as adsorbent on the selectivity and adsorption capacity of Fe(III) prior to its determination by ICP-OES. The selectivity of CuO-TiO 2 NSs towards different metal ions, including Au(III), Cd(II), Co(II), Cr(III), Fe(III), Pd(II), and Zn(II), was performed in order to study the effectiveness of CuO-TiO 2 NSs on the adsorption of selected metal ions. Here the calcined CuO-TiO 2 NSs have significant properties such as largesurface area (surface-to-volume ratio) and static adsorption capacity. These offered high adsorbent features that enhanced the direct solid-phase adsorption towards the target metal-ions for the selective detection of Fe(III) ions. Based on the selectivity study, it was found that the selectivity of nanosheet-phase was the most towards Fe (III). The static adsorption capacity for Fe(III) was also executed in this study. Results of adsorption isotherm are confirmed that the adsorption process is mainly monolayer adsorption onto surface-phase containing a finite number of adsorption sites. Adsorption data of Fe (III) are well fit with the Langmuir-classical adsorption isotherm.

Materials and methods
The λ max (364.7 nm) of calcined CuO-TiO 2 NSs was evaluated with UV/visible spectroscopy (UVO-2960, LABOMED Inc.). FT-IR spectra were measured with a spectro-photometer (Spectrum-100 FT-IR) in the mid-IR range, which was purchased from Bruker, USA. The powder X-ray diffraction (XRD) prototypes were measured with X-ray diffractometer (Rigaku X-ray difractometer, Mini-Flex 2) equipped with Cu-K α 1 radiation (λ = 1.5406 nm) using a generator voltage (40.0 kV) and a generator current (35.0 mA). Morphology of CuO-TiO 2 NSs was investigated on FE-SEM instrument (FESEM; JSM-7600F, Japan). The XPS measurements were executed on a Thermo Scientific K-Alpha KA1066 spectrometer (Germany). Monochromatic AlKα x-ray radiation sources were used as excitation sources, where beam-spot size was kept in 300.0 μm. The spectra was recorded in the fixed analyzer transmission mode, where pass energy was kept at 200 eV. The scanning of the spectra was performed at lower pressures (<10 −8 Torr). 1000.0 mgL −1 stock standard solution of each Au(III), Cd(II), Co(II), Cr(III), Fe(III), Pd(II), and Zn(II) were purchased from Sigma-Aldrich (Milwaukee, WI, USA). All reagents were used of analytical and spectral purity grade. Doubly distilled de-ionized water was also used throughout the experimental studies. Analytical grade of copper chloride, titanium dioxide, and sodium hydroxide was used and purchased from Sigma-Aldrich Company. ICP-OES measurements were acquired by use of a Perkin Elmer ICP-OES model Optima 4100 DV, USA. The ICP-OES instrument was optimized daily before measurement and operated as recommended by the manufacturers. The ICP-OES spectrometer was used with following parameters: FR power (1300 kW), frequency (27.

Samples preparation and detection procedure
Stock solutions of Au(III), Cd(II), Co(II), Cr(III), Fe(III), Pd(II) and Zn(II) were prepared in 18.2 MΩ · cm distilled deionized water and stored in the dark at 4°C. For selectivity study, standard solutions (1.0 mgL -1 ) of each metal ion were prepared and adjusted to pH value of 5.0 with acetate buffer. Then, each standard solution was individually mixed with 25.0 mg CuO-TiO 2 NSs. In this study, a fixed pH value of 5.0 was chosen for all metal ions in order to avoid any precipitation of other species, in particular for Fe(III). For example, Fe(III) usually forms a precipitation of Fe(OH) 3

Synthesis of CuO-TiO 2 nanosheets by a wet-chemical process
The liquid-phase precipitation was applied to prepare CuO-TiO 2 NSs by a wet-chemical method from CuCl 2 (0.1 M, 1.7048 g, 100.0 ml) and titanium dioxide (0.1 M, 0.799 g, 100.0 ml) as the precipitating agent in basic medium (pH~10.5). The starting materials are put in de-ionized water to make 0.1M solution separately in round conical flask at room conditions. After addition of reducing agent (NaOH, adjusting pH at 11.0) into the reactant mixtures (CuCl 2 & TiO 2 ), it was stirred gradually for 12 hours and placed on a hot-plate (at 150.0°C, active solution temperature~92°C). The starting materials (CuCl 2 , TiO 2 , and NaOH) were used without further purification for precipitation technique. Then the solution was washed with acetone and water consecutively and kept for drying at room conditions. The as-grown powder powders were calcined at 450.0°C in muffle furnace for 5 hours. Finally, the calcined products were characterized in features of their structural, morphological, and optical properties as well as applied for the detection of metal ions uptake.

Results and discussion
Optical characterization of CuO-TiO 2 nanosheets The optical property of the calcined CuO-TiO 2 NSs structure is one of the most important features for the evaluation of their photo-catalytic activity. The optical absorption spectra of NSs are measured by UV-visible spectrophotometer in the visible range (250.0 to 800.0 nm). In UV/visible absorption method, the outer-electrons of atoms or molecules are absorbed by incident radiation sources, which undergo electron transition from lower to higher energy levels. According to the phenomenon, the spectrum is obtained due to the optical absorption, which can be used to analyze the band-gap energy of CuO-TiO 2 NSs. The optical absorption measurement was carried out at ambient conditions. From the absorption spectrum, an absorbance maximum is measured using NSs at~364.7 nm, which is presented in Figure 1a. Band-gap energy (E bg ) is calculated on the basis of the maximum absorption band of NSs and found to be 3.404 eV, according to following equation (i).
Where E bg is the band-gap energy and λ max is the wavelength (~364.7 nm) of the nanosheets. No extra peaks associated with impurities and structural defects were observed in the spectrums, which proved that the synthesized NSs control crystallinity of CuO-TiO 2 NSs [53].
The CuO-TiO 2 nanosheets were studied in term of atomic and molecular vibrations. FT-IR spectra basically in the region of 400-4000 cm -1 are executed at room conditions. Figure 1b displays the FT-IR spectrum of the calcined NSs structures. It represents band at 459 cm -1 . This observed wide vibrational band (at 459 cm -1 ) could be assigned as a metal-oxygen (Cu-O & Ti-O) stretching vibration [54], which is demonstrated the configuration of doped nanomaterials. At low-frequency region, the vibrational bands are indicated the formation of CuO-TiO 2 nanosheets. To confirm the crystal phases and crystallinity of the calcined CuO-TiO 2 nanosheets, XRD pattern was employed and presented in Figure 1c. The obtained diffraction pattern is well matched with the base centered monoclinic CuO form. According to the JCPDS cards (45-0937), the lattice constants for calcined CuO NSs are found to be a = 4.6853 Å, b = 3.4257 Å, and c = 5.1303 Å. The phases are found the major characteristic peaks (indicated with black-color) with indices for calcined crystalline CuO at 2θ values of (110), (002), (020), (105), (113), (113), (311), and (222), which is presented in Figure 1c. The high-intensity of diffraction peaks in the obtained pattern clearly confirmed that the doped products are well-crystalline [55]. The CuO-TiO 2 NSs are showed the crystalline nature of TiO 2 peaks lying at 2θ = 25.3°( 101), 2θ = 38.1°(004), 2θ = 48.6°(200), and 2θ = 66.1(204). The preferred orientation related to the plane (004) is observed the highest magnitude. All the peaks in XRD patterns can be indexed as crystalline phases of TiO 2 and the diffraction data were in good agreement with JCPDS files # 021-1272 [56,57]. Crystallite size was calculated by Debye-Scherrer's formula given by equation (ii) Where D is the crystal size; λ is the wavelength of the X-ray radiation (λ = 0.15406 nm) for CuKα; K is usually taken as 0.9; and β is the line width at half-maximum height (FWHM) [58]. The average diameter and thickness of CuO-TiO 2 nanosheets is close to~1.07 μm and~21 ± 5.0 nm respectively.
High resolution FE-SEM images of calcined CuO-TiO 2 NSs are presented in Figure 2a to Figure 2b. Low to highmagnified FE-SEM images [ Figure 2(a1) to Figure 2(b1)] of aggregated nanosheets composed microstructure materials are displayed in sheet-shape. Microstructures are composed with nano-dimensional thickness of CuO-TiO 2 sheets. The average length of NSs is calculated in the range of 1.0 μm to 1.3 μnm, which is close to~1.07 ±10.0 μm. It is clearly exhibited from FE-SEM that the synthesized NS products are microstructure in regular sheet-shape with high-density of CuO-TiO 2 . The average cross-sectional thickness of sheets is measured from the FESEM image, which is close tõ 21.0 ± 5.0 nm. It is interesting to note that most of the calcined CuO-TiO 2 NSs are uniformly grown and possessing uniform sheet-shapes. In addition, the nano-dimensional sheets are perpendicularly arranged and aligned on the upper-portion of the calcined nanosheets. It is also proposed that the wet-chemically prepared nanostructures are composed with aggregated nanosheet composed microstructure of CuO-TiO 2 NSs [59]. X-ray photoelectron spectroscopy (XPS) is a quantitative spectroscopic method that determined the elementalcomposition, kinetic energy, empirical-formula, chemicalstate, binding energy, and electronic-state of the elements that present within a material. XPS spectra are acquired by irradiating a material with a beam of X-rays, while simultaneously determining the kinetic energy and number of electrons that get-away from the top one to ten nm of the material being analyzed. Here, XPS measurements were executed for CuO-TiO 2 nanosheets to investigate the chemical states of CuO and TiO 2 . The XPS spectra of Cu2p and O1s are presented in Figure 4a. In Figure 4b, the spin orbit peaks of the Cu2p (3/2) and Cu2p (1/2) binding energy for all the samples appeared at around 935.2 eV and 956.1 eV respectively, which is in good agreement with the reference data for CuO [60]. In Figure 4c, the spin orbit peaks of the Ti2p (3/2) binding energy for codoped sample appeared around at 559.5 eV, which is in good agreement with the reference data for TiO 2 [61]. The O1s spectrum shows a peak at 531.9 eV in Figure 4d. The peak at 531.9 eV is assigned to lattice oxygen, may be indicated to oxygen (ie, O 2 -) presence in the CuO-TiO 2 NSs [62]. XPS compositional analyses investigated the co-existence of the two single-phase of CuO and TiO 2 materials. Therefore, it is concluded that the wet-chemically prepared CuO-TiO 2 materials have NSs phase contained two different mate rials. Also, this conclusion is reliable with the XRD data significantly.

Selectivity study of CuO-TiO 2 nanosheets
Selectivity of newly introduced CuO-TiO 2 NSs phase towards various metal ions was measured based on determination of the distribution coefficient of NSs phases. The distribution coefficient (K d ) can be obtained from the following equation (iii) [63]: where C o and C e refer to the initial and final concentrations before and after filtration with CuO-TiO 2 NSs, respectively, V is the volume (mL) and m is the weight of CuO-TiO 2 NSs phase (g). Distribution coefficient values of all metal ions investigated in this study are reported in Table 1. It can be clearly observed from the Table 1 that the greatest distribution coefficient value was obtained for Fe(III) with CuO-TiO 2 (50546.392 mLg −1 ) as compared to pure TiO 2 adsorbent (3557.885 mLg −1 ). As shown in Table 1, the amount of Fe(III) is almost extracted by CuO-TiO 2 phase. In addition, minimal to no change in the selectivity is observed for both CuO-TiO 2 and TiO 2 adsorbents toward other metal ions included in this study. Results of selectivity study provided that the newly synthesized CuO-TiO 2 phase is the most selective towards Fe(III) among all metal ions. The highest selectivity of CuO-TiO 2 adsorbent toward Fe(III) can be attributed to additional incorporated oxygen atoms ( donor) presented in CuO-TiO 2 phase, as a result of modification of TiO 2 with CuO. Thus, both incorporated oxygen donor atoms presented in TiO 2 and CuO-TiO 2 phases are able to selectively bind with Fe (III) through an electrostatic attraction or a chelating mechanism.

Static adsorption capacity
For determination of the static uptake capacity of Fe(III) on CuO-TiO 2 NSs phase, 25.0 mL of Fe(III) sample solutions with different concentrations (0-150.0 mgL −1 ) were adjusted to pH 5.0 and individually mixed with 25.0 mg of CuO-TiO 2 NSs. These mixtures were mechanically shaken for a hr at room temperature. Static adsorption capacity was obtained using equation (iv) as follows: Where q e represents the adsorbed Fe(III) by the CuO-TiO 2 NSs phase (mgg −1 ), C ο and C e are the initial and equilibrium concentrations of Fe(III) ion in solution (mgL −1 ), respectively, V is the volume (L) and m is the weight of NSs phase (g). Figure 5 shows the static adsorption capacity of NSs for Fe(III) obtained from the   experiment of adsorption isotherm. In this study, the adsorption capacity of NSs was determined for Fe(III), which is close to be 110.06 mgg −1 . This is the higher adsorption capacity value obtained, which is compared with previously reports for Fe(III) in various studies such as 7.00 [64], 18.30 [65], 28.69 [66], 28.90 [67], and 173.14 [68] mgg −1 . Moreover, under the same batch conditions, the adsorption capacity of Fe(III) on pure TiO 2 and doped CuO-TiO 2 NSs adsorbent is estimated to be 56.06 mgg −1 and 110.06 mgg −1 respectively, which is presented in Figure 5. From this comparative study, it is concluded that the adsorption capacity for Fe(III) was improved by 96.33% with the newly synthesized CuO-TiO 2 NSs adsorbent as compared to pure TiO 2 phase. The sensitivity (slope) and linearity (R 2 ) of Fe(III) ion using CuO-TiO 2 NSs phase is calculated from the calibration plot, which is close to 0.8881 Lg -1 and 0.9971 respectively.

Adsorption isotherm models
Experimental equilibrium adsorption data were analyzed using different models in order to develop an equation that accurately represents the obtained results. Langmuir equation is based on an assumption of a monolayer adsorption onto a completely homogeneous surface with a finite number of identical sites and a negligible interaction between the adsorbed molecules. The Langmuir adsorption isotherm model is governed by the following relation (v) [69]: Where C e corresponds to the equilibrium concentrations of Fe(III) ion in solution (mgmL −1 ) and q e is the adsorbed metal ion by the adsorbate (mgg −1 ). The symbols Q o and b refer to Langmuir constants related to adsorption capacity (mgg −1 ) and energy of adsorption (Lmg −1 ), respectively. These constants can be determined from a linear plot of C e /q e against C e with a slope and intercept equal to 1/Q o and 1/Q o b, respectively. Moreover, the essential characteristics of Langmuiradsorption isotherm can be represented in terms of a dimension-less constant separation factor or equilibrium parameter, R L , which is defined as R L = 1/(1 + bC o ), where b is the Langmuir constant (indicates the nature of adsorption and the shape of the isotherm); C o the initial concentration of the analyte. The R L value indicates the type of the isotherm, and R L values between 0 and 1 represent a favorable adsorption [70].
The experimental isotherm data were fit-well with the Langmuir equation based on the least square fit, as shown in Figure 6, supporting the validity of Langmuir adsorption isotherm model for the adsorption process. Consequently, adsorption isotherm data was indicated that the adsorption process was mainly monolayer on a homogeneous adsorbent surface of NSs. Langmuir constants Q o and b are found to be 109.46 mgg −1 and 1.30 Lmg -1 , respectively. The correlation coefficient obtained from the Langmuir model is found to be R 2 = 0.998 for adsorption of Fe(III) on CuO-TiO 2 NSs. Furthermore, the static adsorption capacity (109.46 mgg −1 ) calculated from Langmuir equation was in agreement with that (110.06 mgg −1 ) of the experimental isotherm study. The R L value of Fe(III) adsorption on the NSs is 0.01, indicating a highly favorable adsorption process based on the Langmuir classical adsorption isotherm model.

Conclusions
The calcined CuO-doped TiO 2 NSs are successfully prepared by a wet-chemical technique combined with a heat-treatment and characterized in detail in terms of their morphological, structural, and optical properties. It is displayed that the synthesized microstructures are possessed monoclinic structure having good optical properties. The static-uptake capacity of the NSs phase for selective adsorption and determination of Fe(III) in aqueous solution was investigated. Reasonable static-uptake capacity of 110.06 mgg −1 with NSs adsorbent for Fe(III) in aqueous solution was achieved. Adsorption data of Fe(III) was well-fit with the Langmuir adsorption isotherm model. Thus, the method may show considerable promise for using it as an effective approach for a selective separation and determination of Fe(III) in complex matrices. The obtained nanosheets composed CuO-TiO 2 microstructures is a promising candidate for potential application in metal ions uptake.