Investigation of Ti doping on the structural, optical, and magnetic properties of ZnO nanoparticles

Ti-doped ZnO (TixZn1-xO x = 0.00, 0.05, 0.10, 0.15) nanoparticles have been synthesized through co-precipitation approach. X-ray diffraction (XRD), scanning electron microscopy (SEM), photoluminescence (PL), UV–Visible spectroscopy, and Vibrating Sample Magnetometer (VSM) have been used to characterize the samples. X-Ray Diffraction (XRD) analysis manifested the hexagonal wurtzite structure. The crystallite size decreased from 37 to 29 nm as dopant concentration is increased. Fourier transform infrared analysis showed the absorption bands of ZnO, with few within the intensities. SEM investigation showed the irregular shape and agglomeration of the particles. Ti, Zn, and O composition were determined from EDX analysis and confirmed the purity of the samples. PL spectra showed a near-band edge emission and visible emission. Vibrating sample magnetometer (VSM) demonstrated pure and doped samples exhibited ferromagnetism behavior at room temperature.


Introduction
In recent years, semimagnetic semiconductors (SMSs) have attracted significant interest because of their versatile capability packages in spintronics devices [1][2][3][4]. They had been generally acquired by doping a small amount of Transition metallic (TM: Fe, Co, Ni, Mn, Cr, etc.) into semiconductors.
In Diluted Magnetic oxides, ferromagnetism at room temperature (RTFM) observed may be due to the intrinsic defects (oxygen vacancies) or presence of secondary phases (extrinsic) and it depends on the methods of preparation. Among II-VI semiconductors, Zinc oxide has a large excitation binding energy (60 meV) with wide direct bandgap [5,6]. Still, there are a lot of requirements to enlighten the origin of the RTFM in Transition metal-doped ZnO.
Among Transition metals [7][8][9][10], Titanium can be easily substituted for host ions due to its d subshells filled in partial. It is a non-ferromagnetic element with a lesser bond length compared with Zn [11]. As the ionic radii of Ti 4? (0.069 nm) is small related with Zn 2? (0.083 nm), Ti 4? can substitute easily in Zn 2? in the ZnO lattice. Moreover, Ti ([Ar] 3d 2 4s 2 ) can donate two electrons due to the valence difference between Ti 4? and Zn 2? . Hence, Ti is a suitable transition metal compared with other materials. The properties of ZnO nanoparticles could be easily changed and controlled by doping with Ti. Although doped ZnO-based DMS has been widely studied, investigations on magnetic properties of Ti-doped ZnO are sporadic. Moreover, Ti provides more carriers in the ZnO lattice, it is interesting to study the magnetic moment shown by Ti-doped ZnO.
Venkatesan et al. [12] and Antony et al. [13] synthesized Ti-doped ZnO films exhibited room temperature Ferromagnetism with a saturation moment of about 0.15lB/Ti. Akilan et al. [14] reported defects induced room temperature ferromagnetism in Tidoped ZnO prepared by solid-state reaction method.
Several methods such as sol-gel [21], solid-state reaction [22], hydrothermal [23], and polymerization solution method [24] have been successfully used to synthesize ZnO nanoparticles. To the best of our knowledge, there are no reports on magnetic properties Ti-doped ZnO nanopowders by co-precipitation carried out so far. It has the advantages of simple, low cost, easy modification of dopant concentration, controlled particle size, and bulk scale synthesis of nanoparticles [25].
In this article, we focus on the synthesis of Tidoped ZnO nanoparticles by co-precipitation method. Influence of Ti doping on structural, optical, and magnetic properties of ZnO nanoparticles have been studied by X-ray diffraction (XRD) method, Fourier transform infrared (FTIR)spectroscopy, Ultraviolet-visible (UV-vis) spectroscopy, Photoluminescence (PL) spectroscopy, and Scanning electron microscopy (SEM) with compositional analysis and vibrational sample magnetometer (VSM). Initially, Zinc chloride and Titanium tetrachloride were dissolved separately in 100 mL of pure distilled water to make 0.2 M of solution. These two solutions are mixed with each other in a stoichiometric ratio under magnetic stirring for 30 min. Then, sodium hydroxide solution was added to the above mixture drop by drop and stirred continuously. The addition of sodium hydroxide was stopped until pH = 13, and stirring was continued for 2 h. It was observed that white precipitation continued to be formed. The white precipitate was washed repeatedly with deionized water and absolute ethanol to remove impurities. Then the precipitate was dried at 150°C for 2 h and the dried powder was calcined in air at 600°C for 5 h.

Characterization
The phase structure and crystalline size of the samples were determined by X-Ray diffraction (XRD) using a PANalytical X'pertPro diffractometer with Cu-Ka radiation (wavelength of 1.5406 Å ). The surface morphology of the samples were studied by SEM (Carl Zeiss SUPRA-55). Chemical composition of the synthesized nanoparticles was examined by energy-dispersive X-ray spectrometer (EDS). UV-vis absorption spectra of all the samples were recorded by Shimadzu-UV 2450 spectrophotometer. The photoluminescence (PL) measurements were carried out by Perkin Elmer-LS 45 spectrofluorometer with an excitation wavelength 325 nm. Fourier transformed infrared (FTIR) spectra of the samples were recorded using a Shimadzu-FTIR spectrometer and room temperature magnetic measurements were obtained by LAKESHORE-7410 vibrating sample magnetometer (VSM).  (201) planes. The intensities of peaks continuously decrease with Ti doping indicating that Ti 4? ions (ionic radius: 0.068 nm) are substituted in Zn 2? ions (ionic radius: 0.074 nm) [25].

Structural characterization
From the d spacing values, the lattice constant 'a' and 'c' can be calculated [26] and their values are given in Table 1.
where 'h', 'k', and 'l' are miller indices, 'a' and 'c' are lattice parameters, and 'd' is the inter planner spacing. The lattice constant values are found slightly lower than the standard values of a = 3.2498 Å and c = 5.2066 Å (JCPDS card no: 36-1451). The lattice parameters (a and c) and microstrain of the samples considerably increased with increasing Ti concentration [27].With increasing Ti concentration, the lattice parameters ( a and c) increase gradually from 3.248 Å to 3.2488 Å and from 5.2051 Å to 5.2059 Å , respectively. This increase can be explained basically by the volume increase due to the larger ionic radius of Ti as compared with Zn. The average crystallite size of the sample was calculated using Debye-Scherrer's formula [26].  where k is the wavelength of X-ray (1.54Ȧ ), b is the full width at half maximum, and h is the angle of diffraction. The average crystallite size was found to be 37.46 nm for ZnO and it decreased to 29.36 nm for 15% Ti-doped ZnO. The decrease in the crystallite size is mainly due to the doped Ti 4? ions that reduce nucleation and rate of growth of ZnO NPs. Consequently, the broadening of XRD peaks as the Ti concentration increases, indicating a reduction of crystallite size due to additional grain boundaries originated from the interstitial Ti. The dislocation density is calculated using the following relation [26],

DS
The dislocation density indicates the crystallinity of a crystal, and it will increase with increasing Ti concentration.
Using the following equation, the volume of the unit cell is calculated [26].
Atomic packing fraction (APF) is estimated using the formula [26], The value of APF does not change from that for bulk ZnO sample and this means that the doping of Ti to ZnO has not created new voids. The APF of ideal hexagonal ZnO is 74% but the calculated value is about 75.4%. It means that APF of nanocrystals is slightly larger than that of bulk materials. It may be due to the size effect in nanocrystalline powders.
The atomic displacements 'u' is given as [26] u ¼ a 2 3c 2 þ 1 4 The average bond length between the cations and the anions can be calculated by the following expression [26].
Bond length has been found to 1.971 Å for Zn-O and it increases to 1.987 Å for x = 15 (wt.%) which also confirms the substitution of Ti in ZnO matrix.
The value of Young's modulus, energy density, and stress of the NPs are calculated and tabulated (Table 1). It is observed that the energy density increases when the percentage of Ti increases. Figure 2 shows SEM images of pure and Ti-doped ZnO nanoparticles. Random agglomeration with cluster shape were observed in all SEM images and agglomeration of particles on the surface occurs due to the high surface energy of the as synthesized nanoparticles. From the SEM images, it is also observed that the grain sizes fall in the nanoscale regime. Figure 3 represents the FTIR spectra for undoped and Ti-doped ZnO nanoparticles. The strong absorption band observed at 3500 cm -1 is owing to the stretching vibration of O -H in water molecules [29].The presence of CO 2 molecules is confirmed by the absorption band at 2300 cm -1 [30]. A sharp absorption band at 1600 cm -1 is ascribed to bending vibration of H-O-H [31]. The band observed at 1500, 1000, and 700 cm -1 is owing to the precursor material [32].

Fourier transform infrared spectroscopy (FTIR)
The band at 465 cm -1 is due to stretching frequency of Zn-O [33] and it is shifted to a low-frequency mode with increasing doping concentration. The

UV-Vis spectroscopy
The optical absorption properties of undoped ZnO and Ti-doped samples were analyzed using UV-VIS spectrometer. The bandgap of undoped and Ti- doped ZnO nanoparticles was calculated by Tauc's plot using the relation [34] ahm ¼ B ðhm À EgÞ n where a-absorption coefficient, B-constant, h-Planck's constant, m-photon frequency, and E goptical bandgap. The energy bandgap of ZnO and Ti-doped ZnO nanoparticles were found to be 3.25, 3.28, and 3.4 and 3.67 eV, respectively (Fig. 4). The redshift in the bandgap is due to the Burstein-Moss effect [35]. The conduction band lower levels are filled with free electrons generated When Ti 4? ions replace the Zn 2? ions. Subsequently, it increases the Fermi Level and also widening the bandgap [36]. According to quantum confinement effect, the smaller crystallites have a larger bandgap in metal oxide systems [37]. This is correlated with the findings in the XRD section in which the crystallite size of ZnO nanoparticles is reduced with the addition of Ti. Figure 5 shows Room temperature PL emission spectra of undoped ZnO, 5%, 10%,15% Ti-doped ZnO samples. Technically, all the samples were excited at 325 nm. It is noticed that undoped and doped ZnO samples exhibits two peaks (i) 390 nm (UV range) originates from the exciton recombination corresponding to the near-band edge (NBE) and (ii) 412 nm (violet range), to be the recombination from the defect centers such as O and Zn interstitials [38].

Photoluminescence (PL)
Defects such as structural defects or vacancies are the main reason for the emission of different deeptrap or colors [39]. Interstitial zinc and oxygen are ascribed to structural defects, while vacancies are produced due to the presence of Zinc and oxygen vacancy [40]. High content of zinc vacancy is the main reason for the violet emission peak observed at 418 nm [41].

Magnetic property
Pure and Ti-doped ZnO samples magnetic properties were studied at room temperature shown in Fig. 6. All samples are ferromagnetic as observed from the M-H hysteresis loops and the values are given in Table 2. The plots of coercivity and retentivity as a function of dopant concentration are shown for all the samples. The trends of variations are nearly similar to those of dislocations density as a function of dopant concentration.
The room temperature ferromagnetism of the nanoparticles could arise due to extrinsic magnetism or intrinsic magnetism. But no traces of impurity or secondary phases in XRD ruled out the RTFM due to extrinsic source. Hence, the attained ferromagnetism is an intrinsic magnetic property of the nanoparticles. The bound magnetic polarons (BMPs) theoretical model [42] is a defect-induce ferromagnetism model. According to this, the ferromagnetism arises due to the oxygen vacancies and dopant ions indirect interaction. The stable ferromagnetic interaction arises due to extend over of polarons. Exchange interaction between oxygen vacancies or interstitial Zn atom with overlap of 3d orbital electrons in ZnO nanoparticles is the origin of Ferromagnetism [43]. The PL results also suggested the oxygen vacancies in ZnO. As the presence of such defects is not huge, the extent of ferromagnetic ordering and hence coercivity is relatively week. Incorporation of Ti ion into the ZnO lattice could also introduce more oxygen vacancy. A singly occupied oxygen vacancy with a spin of 1/2 has been ascribed to the origin of RTFM in ZnO without and with Ti doping.

Conclusions
Ti-doped ZnO nanoparticles were successfully synthesized by co-precipitation technique. From XRD, it is found that our samples are having single-phase hexagonal wurtzite structure and the average crystallite size 37-29 nm. Scanning Electron Microscopic investigation revealed the morphology of pure ZnO and 5%, 10%, and 15% nanoparticle seemed to be approximately spherical in shape. The EDX spectrum confirmed the presence of Ti in ZnO nanoparticles. Fourier Transform Infrared spectra were recorded to analyze the effect of Ti doping on the nanoparticles. The UV-visible spectroscopic study the intercepts yield the direct energy band for the Ti-doped ZnO 5%, 10%, and 15% samples are 3.25, 3.28, and 3.4 and 3.67 eV, respectively. Room temperature VSM study demonstrates that all the samples are suitable for spintronic applications.