• No results found

Al$_2$O$_3$ nanoparticle polymorphs: effects of Zn$^{2+}$ doping on the structural, optical and cytotoxic properties

N/A
N/A
Protected

Academic year: 2022

Share "Al$_2$O$_3$ nanoparticle polymorphs: effects of Zn$^{2+}$ doping on the structural, optical and cytotoxic properties"

Copied!
13
0
0

Loading.... (view fulltext now)

Full text

(1)

Al

2

O

3

nanoparticle polymorphs: effects of Zn

2+

doping on the structural, optical and cytotoxic properties

JE´ SSICA DE LARA ANDRADE1,* , ANGE´ LICA GONC¸ALVES DE OLIVEIRA1,

LEONARDO SOBREIRA RODRIGUES2, MYCHELLE VIANNA PEREIRA COMPANHONI3, CELSO VATARU NAKAMURA3, SANDRO MARCIO LIMA4, LUIS HUMBERTO DA CUNHA ANDRADE4, LUIZ FERNANDO CO´ TICA5, ANA ADELINA WINKLER HECHENLEITNER1, EDGARDO ALFONSO GO´ MEZ PINEDA1and DANIELA MARTINS FERNANDES DE OLIVEIRA1

1Programa de Po´s-Graduac¸a˜o em Quı´mica, State University of Maringa´, Maringa´ 87020-900, Brazil

2Programa de Po´s-Graduac¸a˜o em Materiais, Federal University of Alagoas, Maceio´ 57072-970, Brazil

3Departamento de Cieˆncias Ba´sicas da Sau´de, State University of Maringa´, Maringa´ 87020-900, Brazil

4Programa de Po´s-Graduac¸a˜o em Recursos Naturais, State University of Mato Grosso do Sul, Dourados 79804-970, Brazil

5Programa de Po´s-Graduac¸a˜o em Fı´sica, State University of Maringa´, Maringa´ 87020-900, Brazil

*Author for correspondence (jessika_delara@hotmail.com) MS received 16 May 2019; accepted 28 August 2020

Abstract. Al2O3-Znx% NPs, withx= 0, 1, 3, 5 or 10 mol% Zn2?were synthesized by a modified sol–gel method. The influence of the insertion of Zn2?dopant on the crystal lattice, morphology, optical and cytotoxic properties of Al2O3was investigated. Rietveld refinement applied to DRX data revealed that the oxides are constituted by four crystalline phases:

a-Al2O3, h-Al2O3, d-Al2O3 and a-Al2O3(*), and that the doping promoted changes in unit cell volume for all the crystalline phases. Raman signals indicated that the insertion of Zn2?caused changes in the vibrations of bonds Al–O, mainly in tetrahedral sites of transition phases of Al2O3, which are preferentially occupied by Zn ions. The oxides exhibited photoluminescence emission in the visible and near-infrared region, but Al2O3-Zn 10% showed increased emission intensity in the visible region. The nanoparticles with spherical and elongated morphologies did not exhibit cytotoxic effects on L929 fibroblast cells.

Keywords. Nanostructured materials; crystal structure; optical properties; biomedical applications.

1. Introduction

Aluminium is the third most abundant element in the Earth’s crust, while aluminium oxide (Al2O3) has very interesting properties such as high chemical and mechanical stability, high hardness, low ion release, biocompatibility, low cost and easy control of their textural properties [1–4]. These prop- erties are dependent on a variety of factors, such as its crys- talline structure and the presence of impurities [4,5].

Aluminium oxide is structurally complex due to its polymorphism, which is characterized by different transi- tion crystallographic phases as gamma, delta, eta, theta, kappa and chi (c,d,g,h,j andv) [3]. The polymorphism presence depends upon the process techniques, precursors (usually hydroxides or oxyhydroxides) and of the annealing temperature used in the synthesis of Al2O3, until reaching the most thermodynamically stable phase, the corundum structure (a-Al2O3) [6]. It has been known that fora-Al2O3

synthetized from the boehmite (c-AlO(OH)), the phase transitions in temperatures between c- and d-Al2O3, between d- and h-Al2O3, and between h- anda-Al2O3are

*700–800, 900–1000 and 1000–1100°C, respectively [7].

Outstanding to its advantages, a-Al2O3 and their metastable crystalline phases have triggered great interest and applied in several areas, including abrasives, substrates for electronic circuits, insulators, catalysts or catalyst support in petroleum and petrochemical industries, bioremediation and also in biomedical applications as biological implants in human bone tissue, dentistry, biomolecule immobilization and separation, biosensors and in wastewater cleaning processes [8–12]. The excellent mechanical properties and chemical stability of Al2O3make it suitable for applications in nano- medicine, such as drug delivery systems [13].

Nemade and Waghuley [14] prepareda-Al2O3quantum dots (QDs) using low temperature to sinter their precipitate.

They observed a-Al2O3 nanoparticles (NPs) were with

Electronic supplementary material: The online version of this article (https://doi.org/10.1007/s12034-020-02308-y) contains supplementary material, which is available to authorized users.

https://doi.org/10.1007/s12034-020-02308-y

(2)

trigonal structure and with average crystallite size of 4.2 nm. Al2O3QDs showed intense absorption in the ultraviolet region with semiconducting behaviour, being the bandgap energy value determined as 3.6 eV. This behaviour was attributed to the presence of different sub-energy levels in the bandgap, which are related to surface defects and due to the strong quantum confinement effect. Besides that, Al2O3

QDs exhibited emission of intense blue luminescence under excitation at 254 nm and room temperature.

Nowadays, there are several discussions regarding the safety of nanostructured biomaterials [15,16], since nanostructured biomaterials have the potential to be used in a wide range of applications, such as drug delivery, diagnosis, medical imaging and tissue engineering [17].

The interaction between a biomaterial and the host tissue depends on various factors, such as the physical proper- ties of the biomaterial, structural defects, surface area, size, shape and mechanical properties [18]. Consequently, when a nanostructured material is considered for biomedical applications, their properties must be evaluated.

Fibroblast cells are the most abundant cell type of the connective proper tissue of animals, and its main purpose is to produce and maintain the extracellular matrix [19], playing an important role in growth control and cell dif- ferentiation and therefore is suitable for cell viability assays in the presence of biomaterials [20]. L929 fibroblast cell line is derived from normal subcutaneous areolar and adi- pose tissues of 100-day-old male C3H/An mouse and it is cited as a reference for cytotoxicity tests on biomaterials [21]. In this study, fibroblast cells (L929) were selected as the model system for cytotoxicity tests and is in ISO 10993-5 compliant cell line [22], commonly used in bio- compatibility testing given that its sensitivity to the cyto- and genotoxic effects of materials under test is higher than human fibroblasts [23,24].

The cytotoxicity of Al2O3NPs has been investigated by several researchers. Radziun et al [2] also studied the cytotoxic effect of c-Al2O3 on L929 cells. Their results revealed that Al2O3NPs can penetrate the cell membrane, but do not cause increased apoptosis or decreased cell viability. This study demonstrated that aluminium oxide NPs do not have a cytotoxic effect on the tested cells with a concentration of 10–400 lg ml-1, thereby indicating their biocompatibility.

Recently, Song et al[25] reported one study comparing the cytotoxic effects of Al2O3and ZnO. They evaluated the effect of size and different crystalline phases of these oxides on the Caco-2 cell line, which is widely used as a model of intestinal epithelial cells and forin vitromodes to stimulate the gastrointestinal tract. They found that ZnO induced serious toxicity to Caco-2 cells compared to Al2O3 NPs.

Significant toxicity was detected by ZnO NPs at 18 lg ml-1, whereas slightly cytotoxicity was found for Al2O3at concentrations higher than 50 lg ml-1. The authors also showed that the size of Al2O3 (nano or micro) did not

demonstrate an effect on its cytotoxicity. From this study, it was believed that the dose and dissolution were the most important influencing factors [25]. ZnO showed higher cytotoxicity due to its dissolution, while Al2O3 was less toxic as it did not dissolve in the culture medium. These reports motivated us to investigate the influence of Zn2?

doping in the Al2O3lattice and the in vitrocytotoxicity of Zn-doped Al2O3materials against L929 fibroblasts cells.

Doping is a way of modulating the structural, morpho- logical and optical properties of Al2O3and its cytotoxicity due to the synergistic effect of two materials in one and the different properties that can emerge from this new material [26]. There is one study that reports the effects of Zn ion doping on the microstructure, phase composition and microwave dielectric properties of Al2O3 ceramics [27].

The authors showed that doping is an easy way to improve the dielectric and microwave absorption properties of alu- minium oxide, which could make this material appropriate for use in military applications, such as microwave-ab- sorbing materials. Prior to the present study, previous work about Zn-doped Al2O3have reported mainly on the purpose of technological applications, such as for insulators or memory devices [27–29] and only one study that reports in vitroexperiments with Zn-doped Al2O3, which revealed this material outstanding osteogenic activity and antibac- terial effects was found [30].

In a previous study, we investigated the effect of Al3?

concentration (from 0 to 10 mol% Al3?) on the structural, optical, photocatalytic and cytotoxic properties of Al-doped ZnO synthesized by a modified sol–gel method [31]. We verified that the incorporation of Al3? in the ZnO host promoted changes in its electronic, optical and photocat- alytic properties. In addition, all the investigated samples (undoped ZnO and Al-doped ZnO) were highly cytotoxic on fibroblast cells, exhibiting potential action anticancer, but in contrast, this high cytotoxicity could limit its use in biomedical applications.

In this study, we synthesized Zn-doped Al2O3 NPs (Al2O3-Zn x%) and investigated the effects of different levels of Zn2?doping on the structural, morphological and optical properties. Thein vitrocytotoxicity of the NPs was also investigated on fibroblast cells (ATCC L929), with the aim of evaluating its potential use in biomedical devices.

2. Experimental

2.1 Preparation of Al2O3-Zn x% NPs

Al2O3-Znx% NPs withx =0, 1, 3, 5 or 10 mol% Zn2?were synthesized by a modified sol–gel method [32] and all the reagents used in this synthesis were of analytical grade and were used without further purification. Firstly, a 10% (w/v) aqueous solution of poly(vinyl alcohol) (98–99% hydrol- ysed, Aldrich) and saturated metal nitrate solutions (Al(NO3)39H2O and Zn(NO3)26H2O, both synthesized

(3)

C98.0% purity) were separately prepared and then mixed at specific metallic ion:monomer unit ratios. The solutions were maintained at room temperature under stirring for 2 h and then heated until 250–350°C for total water evaporation and initial thermal degradation of organic material. The nanostructured materials were obtained after calcination of the precursor powder under air atmosphere at 1100°C.

Then, Zn-doped Al2O3samples were successfully prepared and labelled as undoped Al2O3, Al2O3-Zn 1%, Al2O3-Zn 3%, Al2O3-Zn 5% and Al2O3-Zn 10%.

2.2 Characterization of Al2O3-Zn x% NPs

Crystal structures were analysed by X-ray diffraction (XRD) using a Shimadzu XRD 7000 diffractometer, with Ka of 1.5406 A˚ radiation at 40 kV and 30 mA in the 2h range from 20 to 80°with a scan rate of 2°min-1and a step of 0.02°. Rietveld refinement was performed using TOPAS software in order to identify what the phases of Al2O3 present in XRD of the samples. Fourier-transform infrared (FTIR) spectra were obtained using a Fourier transform infrared spectrometer Bruker Vertex 70Z, at resolution of 4 cm-1in the range of 4000–400 cm-1. The nanostructured oxides were well mixed with dried KBr powder, pressed to form disk tablets and used for measurements.

Micro-Raman measurements were carried out using a Bruker Senterra Raman microscope under a 532 nm laser radiation with a power of 20 mW, 50 scans and the spectra were acquired by averaging a hundred acquisitions of 3 s at room temperature (*300 K). Photoacoustic absorption (PA) spectra were obtained using monochromatic light from a 1000 W xenon arc lamp (Model 68820, Oriel Corpora- tion), a monochromator (Model 77250, Oriel Instruments) and a mechanical chopper (Model SR540, Stanford Research Systems). The spectral range was between 200 and 800 nm.

Photoluminescence (PL) spectra were carried out with kexc= 330 nm. The optical emission was collected with an optical fibre and analysed with a Jobin Yvon iHR550 monochromator (1200 grooves mm-1holographic grating) coupled to an iCCD. All the measurements were performed at room temperature (*300 K) with at least three repeti- tions. The morphology and shape of the NPs were evaluated by electron microscopy. Al2O3-Znx% NPs were observed by transmission electron microscopy in a JEM-1400 JEOL operated at 120 kV. Scanning electron microscopy (SEM) micrographs were assessed using a FEI Quanta 250 microscope coupled with energy dispersive X-ray spectroscopy.

2.3 In-vitro cytotoxicity of Al2O3-Zn x% NPs

Cell viability assays were performed using anin vitrostudy employing fibroblast cells (ATCC L929). The fibroblasts

cells were cultured in Dulbecco’s modified Eagle medium (DMEM, Gibco Invitrogen) supplemented with 10% fetal bovine serum (FBS), 1% penicillin and 1% streptomycin, in a humidified atmosphere 5% CO2 incubator at 37°C. The culture and co-confluent were washed with phosphate-buf- fered saline, trypsinized, resuspended and seeded at an initial density of 2.5 9105cells per well in DMEM sup- plemented with 10% FBS into a 96-well microplate. After 24 h of incubation, the culture media with Al2O3, Al2O3-Zn 1% and Al2O3-Zn 5% NPs were prepared by serial dilution in concentrations of 800, 400, 200, 100, 50 and 10lg ml-1, and 200 ll of these dispersions were added to each before being well incubated for 24 h. Two independent experi- ments were performed in duplicate.

The cell viability was evaluated by the metabolic activity of the cells through a colorimetric method based on the ability of the mitochondria of viable cells to reduce MTT (tetrazolium salt) into a purple insoluble compound called formazan. After 24 h, the 96-well cell culture plates were observed under an optical microscope. Afterwards, a 40-ll MTT solution at a concentration of 2.5 mg ml-1was added to each well, and the cells were incubated for 4 h at 37°C and 5% CO2. The supernatant was removed and 100ll of DMSO was added to each well to solubilize the cells. The reading was performed in an ELISA spectrophotometer (Bio-Tek FL-600 Microplate Fluorescence Reader) at 540 nm with a reference at 630 nm. Cells grown in the absence of Al2O3-Znx% NPs were used as a negative control and taken as 100% growth. The number of viable cells was calculated based on the absorbance value obtained.

3. Results and discussion

Figure1 shows the X-ray diffractograms of Al2O3-Znx%

(in different compositions) and the Rietveld refinement applied for each sample using TOPAS software. After adjustments were obtained, a structural model indicated the coexistence of four crystalline phases of aluminium oxide in the samples, being thema-Al2O3,h-Al2O3,d-Al2O3anda- Al2O3(*), and the crystallographic planes in XRD pattern are indicated in figure1 and described in detail in supple- mentary table S1. Thea-Al2O3(corundum) phase belongs to the space group R3c, crystallized in trigonal crystalline system [33], in which Al3?cations occupy octahedral sites.

h-Al2O3 and d-Al2O3 are transition phases to corundum structure, the first is crystallized in the monoclinic system and space group C2/m [34,35], and d-Al2O3is crystallized in tetragonal structure and space group P4m2 with cubic spinel lattice and for both Al3? ions are distributed in an equal number on the octahedral and tetrahedral sites [36,37]. a-Al2O3(*) is a compact crystalline phase with oxygen deficiency [38], obtained by transformation of corundum phase in high temperature, it has hexagonal structure and can be assigned to the space group P63/mmc.

This structure is a spinel with aluminium ions in octahedral

(4)

Figure 1. XRD patterns and Rietveld refinements of (a) undoped Al2O3, (b) Al2O3-Zn 1%, (c) Al2O3-Zn 3%, (d) Al2O3-Zn 5% and (e) Al2O3-Zn 10%.

(5)

and tetrahedral sites and it is established by defects as vacancies with two excess electrons [38].

Table 1 shows the abundance of each crystalline phase present in Al2O3-Zn oxides with different Zn2? content.

Undoped Al2O3is constituted approximately by 5%a-(*), 20% h-, 30% a- and 45% d-Al2O3, which is the most abundant among the crystalline phases. The percentage of h-Al2O3 anda-Al2O3(*) remained practically constant for all the different Zn2?content. However, the abundance of a-Al2O3exhibited a tendency of decreasing, whiled-Al2O3 considerably increased with Zn2?content. In Al2O3-Zn 1%

the most thermodynamically stable a-Al2O3was identified as the most abundant crystalline phase (&60%) and the percentage of the metastabled-Al2O3phase decreased from 43 to 21%. For samples with higher Zn2?contents (3, 5 and 10 mol%), d-Al2O3is a most abundant crystalline phase.

Probably, for Al2O3 doped with small Zn2? content (as Al2O3-Zn 1%), Zn2?can be occupying interstitial sites in the a-Al2O3 lattice, while for Zn-doped Al2O3containing Zn2?higher than 1 mol% (3, 5 and 10 mol%), the doping process possibly produced substitutional defects, in which some Al3?were replaced by Zn2?ions, favouring the for- mation and structural organization of d-Al2O3 phase.

Although the doping could occur in all the crystalline phases, it is known that Zn2?ions prefer to occupy tetra- hedral sites due to its high field and electronegativity [39,40]. Sinced-Al2O3is a cubic spinel with octahedral and tetrahedral sites, it is probable that for samples with high content of dopant zinc ions, they preferentially substituted

some Al3? ions in tetrahedral sites of d-Al2O3 lattice, increasing the chemical stability of this metastable phase of aluminium oxide.

The values of lattice parameters and unit cell volume are given in table1. In order to ease of understanding these parameters were also plotted in function of Zn2?content as shown in supplementary figure S1. In a-Al2O3 crystalline phase a slight increase in unit cell volume for Al2O3-Zn 1%

and Al2O3-Zn 3%, and a small decrease for Al2O3-Zn 5% and Al2O3-Zn 10% compared to undoped Al2O3were observed.

As previously discussed, this behaviour indicates that for samples doped with small Zn2?content (1 and 3 wt%), the dopant ions could be occupying interstitial sites in the a- Al2O3lattice, promoting thus a slight increase in the unit cell volume ofa-Al2O3, and reaching a ‘saturation’ of Zn2?in the a-Al2O3lattice with 3 mol%. Nevertheless, for the doped oxides with Zn2?content higher than 3 mol% (5 and 10 mol%

Zn2?), the dopant ions may be acting as structure-directing agents, inducing to the formation of metastable phases of alumina. This elucidates the formation of a greater amount of d-Al2O3(table 1), which has tetrahedral sites that are pref- erentially occupied by Zn2?[39,40]. The predominance of metastable phases of aluminium oxide in Zn-doped Al2O3

and the preferential occupation of Zn2? dopant in these phases, might be reducing the amount of structural defects in a-Al2O3, and causing a contraction in this unit cell volume compared to undoped Al2O3.

Inh-Al2O3a tendency of decreasing in unit cell volume with Zn2?content was observed, except for Al2O3-Zn 10%, Table 1. Structural parameters determined from XRD/Rietveld refinement for different crystalline phases of Al2O3present in undoped Al2O3and Al2O3-Znx% samples.

Crystalline phases

Composition Abundance (%) a(A˚ ) b(A˚ ) c(A˚ ) Cell volume (A˚3)

Undoped Al2O3 a-Al2O3 30.48 4.7559 4.7559 12.9840 254.3394

h-Al2O3 20.21 11.6239 3.1286 5.5893 195.8744

d-Al2O3 43.53 5.6562 5.6562 23.4868 751.4191

a-Al2O3(*) 5.78 3.1040 3.1040 5.0085 41.7920

Al2O3-Zn 1% a-Al2O3 59.89 4.7571 4.7571 12.9855 254.6726

h-Al2O3 13.95 11.8609 2.9466 5.5705 188.0865

d-Al2O3 21.91 5.6345 5.6345 24.0576 765.8032

a-Al2O3(*) 4.25 3.1297 3.1297 5.1282 43.4368

Al2O3-Zn 3% a-Al2O3 24.29 4.7571 4.7571 12.9921 254.6175

h-Al2O3 15.93 11.6686 3.0888 5.5150 191.5493

d-Al2O3 53.80 5.6784 5.6784 23.8171 767.9686

a-Al2O3(*) 5.98 3.2325 3.2325 5.0221 45.4465

Al2O3-Zn 5% a-Al2O3 20.96 4.7524 4.7524 12.9733 253.7496

h-Al2O3 9.27 11.1367 2.9540 5.7742 185.0584

d-Al2O3 67.69 5.6663 5.6663 23.7122 761.3803

a-Al2O3(*) 2.08 3.2081 3.2081 5.0126 44.6770

Al2O3-Zn 10% a-Al2O3 10.42 4.7536 4.7536 12.9749 253.5130

h-Al2O3 10.33 11.5974 3.1075 5.7169 202.7370

d-Al2O3 69.73 5.6851 5.6851 23.9998 777.5427

a-Al2O3(*) 9.53 3.2103 3.2103 5.0070 44.4913

(6)

for which this parameter increased from 195.8744 A˚3(for undoped Al2O3) to 202.7370 A˚3. For all the doped oxides, it was noted an increase in unit cell volume of thea-Al2O3(*) phase in relation to undoped Al2O3. Additionally,d-Al2O3 that is the predominant phase in the doped oxides with high contents of Zn2?, showed an expressive increase in unit cell volume from 751.4191 A˚3(for undoped Al2O3) to 777.5427 A˚3(for Al2O3-Zn 10%). A tendency of increase in unit cell volume of the crystalline phases of Al2O3with Zn2?con- tent is expected, since Zn2? (0.74 A˚ ) is bigger than Al3?

(0.53 A˚ ) ions, then it could change the lattice parameters and consequently the unit cell volume, as was observed mainly in a-(*) and d-Al2O3 crystalline phases. It was detected that the doping of Al2O3 NPs polymorphs with Zn2?influenced more considerably the unit cell volume of d-Al2O3, probably because this is the most abundant crys- talline phase and consequently, the doping preferentially occurs on it.

It has been reported that Zn-doped Al2O3 with Zn2?

contents higher than 3 mol% results in the emergence of ZnAl2O4 segregated phase [27]. However, no diffraction peak characteristic of secondary phases attributed to ZnO and/or ZnAl2O4 was detected in X-ray diffractograms of Al2O3-Zn NPs containing up to 10 mol% Zn2? dopant, suggesting the efficient incorporation of Zn2? into Al2O3 host. This is possibly related to existence of h-,d- and a- Al2O3(*) crystalline phases in aluminium oxide that possess tetrahedral sites, which are preferentially occupied by Zn2?

ions [39,40].

Besides that, to investigate in detail the limit Zn2?con- centration that can dope Al2O3 Zn-doped Al2O3 samples containing 12, 15 and 20 mol% Zn2?were synthesized and XRD patterns of these samples are shown in supplementary figure S2. In XRD of Zn-doped Al2O3 containing C12 mol% is possible to detect the formation of ZnAl2O4over the polymorphic alumina, in accordance to JCPDS card no.

74-1138, from evident peaks of this crystalline phase at 2h

= 31.4 and 37.0°. However, no peak of ZnO phase was identified from XRD patterns of these samples. Therefore, it was established that for Al2O3 doped with Zn2? content between 10 and 20 mol% occurs preferentially the forma- tion of a spinel-like phase of ZnAl2O4.

FTIR spectra obtained for Al2O3-Znx% NPs are exhib- ited in figure 2. A large absorption band between 1000 and 400 cm-1 was observed for all nanostructured oxides.

Stretching frequencies in the range of 700–400 cm-1can be attributed to metal oxides, such as Al–O and/or Zn–O [41].

Precisely, bands at 450, 490, 584, 638 and 760 cm-1cor- respond to aluminium ions in octahedral coordination (AlO6), characteristic of the a-Al2O3 phase [36,42,43].

These bands are more intense in the spectra of undoped Al2O3and Al2O3-Zn 1%. It was noted that the intensity of these bands decreased significantly with the increase of Zn2?content, and a broad band was observed at 400 to 650 cm-1region, which can be related to the Al3?in tetrahedral sites (AlO4) present inh-Al2O3,a-Al2O3(*) and especially,

d-Al2O3 phases, because there are reports that Al3? in octahedral sites (AlO6) exhibit narrow bands [36]. All the samples showed weak bands at 1630 and the 3600–3200 cm-1region, which are characteristics of in-plane bending and symmetric stretching of O–H groups, respectively.

Hydroxyl group are due to molecular water adsorbed on the surface of the nanostructured oxides [44].

The influence of Zn2?on the structural organization of aluminium oxide could also be analysed from Raman spectra obtained for Al2O3-Znx% NPs shown in figure3.

The a-Al2O3 phase belongs to ditrigonal-scalenohedral class of the trigonal symmetry D3d6 (R3c) [45], and in this crystalline phase the Al3?ions are on sites C3and are sit- uated at octahedral sites that are coordinated with two layers of oxygen. The oxygen atoms are on sites with C2 sym- metry, and then the octahedron is severely distorted [46].

Based on factor group analysis, the optical modes in the crystal ofa-Al2O3are 2A1g?5Eg?2A2u?4Eu?3A2g? 2A1u [46–48], which are highlighted in the Raman spectra of Zn-doped Al2O3. A1gmode at*417 cm-1is commonly associated with Al–O stretching vibrations [49] on octahe- dral sites and its intensity slightly increased with Zn2?

content, indicating that the doping promoted changes on the structural organization of Al2O3. It is noted in table2 that the centre peak position attributed to this vibration mode (AlO6) shifted to higher wavenumber for Al2O3-Zn 10%.

Furthermore, the Raman spectra of the samples showed a signal centred at 259 cm-1(for undoped Al2O3), which has been assigned to Al–O stretching vibrations in tetrahedral sites (AlO4) present in metastable phases: h-, d- and a- Al2O3(*) [50]. This signal had its intensity increased for the oxides containing high concentrations of Zn2?probably due to the more abundance of these metastable phases in doped oxides. Besides that, as can be seen in table2, this signal Figure 2. FTIR spectra of undoped Al2O3 and Al2O3-Zn x%, with x corresponding to different Zn2? ion contents in mol%

(indicated).

(7)

appears shifted to lower wavenumber. This shift is more an indicative that Al3?was replaced by Zn2?, causing changes in the crystal lattice, and therefore in the vibrations of bonds Al–O, preferentially in tetrahedral sites of transition phases of Al2O3.

The Raman signal exclusively observed in Zn-doped Al2O3 spectra around 327 cm-1 is associated with an acoustic vibration mode of Zn2?, confirming that Zn ions were incorporated into the network of aluminium oxide [51–54]. This vibration mode also appears shifted for higher wavenumber with increased Zn2?content (table2).

Photoacoustic absorption spectra obtained for undoped Al2O3and Al2O3-Znx% NPs are shown in figure4. All the oxides exhibited a maximum photoacoustic signal in the ultraviolet region, between 230 and 325 nm, which is assigned to the energy for promoting electrons from the valence band (VB) to the conduction band (CB) [53], in accordance to reported data for absorption of the a-Al2O3 NPs [54]. Further, only the spectra of Zn-doped Al2O3 exhibited a shoulder between 355 and 400 nm, which is related to emergence of additional electronic levels due to the incorporation of Zn2?ions into crystal lattice of Al2O3.

The PA spectra of these samples also show an optical absorption slightly shifted to the visible region compared to undoped Al2O3, indicating that the insertion of Zn2?also affects the optical properties of Al2O3NPs polymorphs. All the oxides exhibited bandgap energy values (Eg) charac- teristic of semiconductor materials [54], around 4.0 eV, which were calculated using the Tauc relation [31]. In general, the Egvalues estimated for Al2O3-Znx% NPs are in accordance with those reported for Al2O3NPs [14,50].

PL spectra obtained for Al2O3-Znx% NPs in the visible (400–600 nm) and near-infrared (600–900 nm) regions are shown in figure 5a and b, respectively. In figure5a it is observed that undoped Al2O3 and Zn-doped Al2O3 con- taining 1, 3 and 5 mol% Zn2? exhibit less intense PL emission, between 425 and 600 nm. However, Al2O3-Zn 10% exhibits increased emission intensity in this region, demonstrating that the doping with 10 mol% Zn2?

improved the emission of Al2O3 in the visible region. A broad and nonlinear PL emission between 700 and 860 nm is shown in figure5b for all the oxides, where Al2O3-Zn 3%

showed a more intense emission. The broad emission in the near-infrared region (NIR) might be attributed to the phase transitions of aluminium oxide, such asc- andh-Al2O3[55], although some studies relate this emission to uncontrolled impurities of alumina, like transition metals Cr3?, Fe3?and Ti3?[56], which were not detected in the samples by other characterization techniques.

In order to analyse the PL spectra in the visible region, their areas were normalized and deconvoluted between 375 and 575 nm. Five Gaussian components allowed us to get the best fit of experimental and calculated curves. The deconvolutions obtained for undoped Al2O3 (figure5c) showed a multicomponent luminescence, being the most intense emission with kmax = 479 nm and low-intensity peaks atkmax= 421, 439, 518 and 541 nm. Figure5d, e, f Table 2. Centre peak position Raman (cm-1) with specific

attributions.

Peak position (cm-1)

AlO6 AlO4 Zn2?

Undoped Al2O3 416 259 —

Al2O3-Zn 1% 417 256 317

Al2O3-Zn 3% 417 255 326

Al2O3-Zn 5% 417 257 327

Al2O3-Zn 10% 428 254 328

Figure 4. PA spectra obtained at 300 K for undoped Al2O3and Al2O3-Zn x% NPs, with x corresponding to different Zn2? ion contents in mol% (indicated).

Figure 3. Raman spectra of undoped Al2O3and Al2O3-Znx%, with x corresponding to different Zn2? ion contents in mol%

(indicated).

(8)

Figure 5. PL spectra obtained at 300 K upon excitation at akexc= 330 nm for undoped Al2O3and Al2O3-Zn x%: (a) in the visible and (b) in the near-infrared regions. Deconvolution into Gaussian components of these spectra in the visible region for (c) undoped Al2O3, (d) Al2O3-Zn 1%, (e) Al2O3-Zn 3%, (f) Al2O3-Zn 5% and (g) Al2O3-Zn 10%.

(9)

Figure 6. TEM images obtained for (a) undoped Al2O3and (b) Al2O3-Zn 10%. SEM images obtained for (c) undoped Al2O3, (d) Al2O3-Zn 1%, (e) Al2O3-Zn 3%, (f) Al2O3-Zn 5% and (g) Al2O3-Zn 10%.

(10)

and g shows the deconvolution obtained for Al2O3-Zn 1%, Al2O3-Zn 3%, Al2O3-Zn 5% and Al2O3-Zn 10% NPs, respectively. They showed the most intense emission with kmax = 477, 486, 465, 466 nm for each oxide, and other peaks with low intensity. It can be noted that the doped oxides, in general, exhibit maximum absorption shifted to lower wavelengths in relation to undoped Al2O3, possibly due to the emergence of additional electronic levels and/or defects created by the insertion of Zn2?dopant between VB and CB.

PL emission in the visible region for Al2O3 have been assigned to structural defects as oxygen vacancies in dif- ferent charge states and their aggregate centres [57].

Emission bands between 330–410 nm are mainly attributed to F?and F centres (oxygen vacancies that trapped one and two electrons, respectively). Besides that, emission bands between 410 and 550 nm are ascribed to F2centres, which are defects due to an aggregation of centres F (two neutral oxygen vacancies that trapped four electrons) [9,58,59]. As shown in figure 5a, Al2O3-Zn 10% predominantly consti- tuted by d-Al2O3 crystalline phase exhibits more intense visible emission than other oxides. This behaviour probably is related to greater amount of oxygen vacancies in the lattice of the metastable phases. The presence of these structural defects owed the doping process, may be improving the PL emission of the Al2O3-Zn 10% in this region.

Moreover, in Zn-doped Al2O3NPs, some trivalent Al3?

ions could be replaced by divalent Zn2? ions into tetrahe- dral sites, as indicated by XRD, Raman and FTIR results (figures1–3), causing possible distortions in the Al2O3host structure [60]. It has been reported that Zn2? ions substi- tuting Al3? can promote disorders and defects related to oxygen vacancies [61]. In addition, Al3?or Zn2?ions may also be occupying interstitial sites, causing internal stress in the crystal lattice in all crystalline phases (a-Al2O3, h- Al2O3, d-Al2O3anda-Al2O3(*)). All these possible struc- tural changes can contribute for this intense PL emission.

The more intense PL emission observed for Al2O3-Zn 3% in the NIR might be associated with highest level of internal stress, in accordance with the analysis of the variation in the lattice parameters and unit cell volume with Zn2? content (see supplementary figure S1). For the doped samples with Zn2?content higher than 3 mol% probably occurs substi- tutional doping, resulting in a lower lattice stress and con- sequently in a less intense PL emission in the NIR. An interesting aspect of our perception is that the intense emission in the NIR showed by our materials can make them important for biomedical applications, for example, in photoacoustic and photothermal imaging or in drug delivery systems [62,63]. We intend to investigate these possible applications in future studies.

Transmission electron microscopic images of undoped Al2O3and Al2O3-Zn 10% NPs (figure6) show the presence of NPs with diameter varying 10–20 nm. In addition, undoped Al2O3 (figure6a) and Al2O3-Zn 10% (figure6b)

are constituted by NPs with defined edges and a predomi- nantly cubic morphology and in both are present some elongated structures.

SEM images obtained for Al2O3, Al2O3-Zn 1%, Al2O3- Zn 3%, Al2O3-Zn 5% and Al2O3-Zn 10% are shown in figure6c, d, e, f and g, respectively. All the samples exhibit some structures with morphology similar to nanoneedles, with an average diameter of *10 nm and length in micrometre scale. The elongated structures might be owing to d-Al2O3, the predominant crystalline phase in doped oxides. In addition, the nanoneedles appear supported in structures with a compact morphology, which could be characteristic of the transition process or particle rear- rangement from d-Al2O3 and h-Al2O3 to a-Al2O3 and a- Al2O3(*), the thermodynamically stable crystalline phases of alumina [64]. Energy dispersive X-ray spectra were obtained for Al2O3-Zn% NPs (supplementary figure S3) and showed that aluminium atoms are predominant and the content of zinc ions are close to those expected for each sample based on its preparation.

In this study, we synthesized Al2O3-Zn% NPs with the aim of their use in biomedical applications, thus its safety should be investigated. Figure7 shows the results of the in vitrocytotoxicity of undoped Al2O3, Al2O3-Zn 1% and Al2O3-Zn 5% NPs after 24 h of incubation on L929 fibroblast cells. The concentration of the respective nano- structured oxides varied from 0 to 800 lg ml-1 and the cytotoxicity was measured by enzymatic assays of MTT.

For undoped Al2O3NPs in all the tested concentrations, the cell viability of the L929 fibroblast was higher than 90%

compared to the viability of cells in the negative control (100% cell viability). For Zn-doped Al2O3 with 1 and 5 mol%, the viability of the L929 fibroblast cell was higher than 96%. Therefore, these results demonstrate that Zn- doped Al2O3 NPs synthesized in this work did not exhibit

Figure 7. L929 cell viability expressed as a percentage of the viability of cells in the control after incubation for 24 h with a culture in different concentrations of undoped Al2O3, Al2O3-Zn 1% and Al2O3-Zn 5%.

(11)

cytotoxic effects on L929 cells, and they can thereby be considered as non-toxic materials. In addition, the high L929 fibroblast cell viability observed indicates that this material can help to produce and maintain the extracellular matrix [19] and then it might be applied as scaffold in tissue engineering [18], for example, for cell growth in wounds.

Until now, no study on the cytotoxicity of Zn-doped Al2O3 NPs has been reported. Krishnamurithy et al [65]

investigated the cytotoxicity of Al2O3 doped with carbon (Al2O3/C) for biomedical applications and they verified that the doping did not affect the attachment and proliferation of mesenchymal stromal cells (hBMSCs). In addition, Al2O3/C facilitated the osteogenic differentiation of hBMSCs, demonstrating the potential to be used in tissue engineering.

In order to confirm the non-toxic characteristics of undoped Al2O3and Al2O3-Znx% NPs, the morphology of L929 fibroblast cells was analysed using an optical micro- scope (figure 8). In the presence of Al2O3NPs (at 10, 100 and 200 ll ml-1), the typical stellar morphology of the fibroblast cells [66] was maintained. When L929 cells were cultivated in contact with Al2O3-Zn 1% and Al2O3-Zn 5%

NPs, we observed that its morphology was again not affected.

There are many discussions regarding the safety of alu- minium oxide by toxicologists. Some researchers have

suggested that Al2O3is less toxic than other nanostructured metal oxides. Alternatively, others researchers have con- sidered this oxide as a potential neurotoxin for a long time [8,15]. Therefore, we believe that the lack of cytotoxicity of Al2O3NPs could be related to its chemical stability, which can be obtained through the synthesis method used.

Figure8 also confirmed that Zn-doped Al2O3NPs did not show cytotoxicity on L929 fibroblast cells. This demon- strates that Zn-doped Al2O3NPs exhibit performance to be used in biomedical applications as photoacoustic and pho- tothermal imaging, drug delivery systems or in biocom- patible polymer composites employed in tissue engineering for cell growth and proliferation, further we have not found a similar report for the cytotoxicity of Zn-doped Al2O3yet.

4. Conclusions

Al2O3-Zn NPs with different Zn2?contents were success- fully prepared by a simple modified sol–gel method. Riet- veld refinement showed that aluminium oxide NPs are constituted by a mix of four crystalline phases:a-,h-,d- and a-Al2O3(*). The phase mixing facilitated the doping pro- cess, due to the tetrahedral sites present in the metastable phases, which are preferentially occupied by Figure 8. Optical images of L929 cells cultured. (a) Control cells; (b), (c) and (d) exposed to Al2O3; (e), (f) and

(g) exposed to Al2O3-Zn 1%; (h), (i) and (j) exposed to Al2O3-Zn 5% at concentrations of 200, 100 and 10lg ml-1, respectively.

(12)

Zn2? ions. In general, d-Al2O3 is a most abundant crys- talline phase present in Al2O3-Zn NPs. FTIR and Raman spectra showed typical absorption bands of aluminium ions in octahedral and tetrahedral coordinations, characteristic from the a-Al2O3 and metastable phases of aluminium oxide with their positions slight shifts due to the insertion of Zn2?dopant. Al2O3-Zn NPs also exhibited PL emission in the visible and NIRs. Al2O3-Zn 10% had its PL emission intensity increased in the visible region, demonstrating that the doping with higher contents of Zn2? enhanced the emission of Al2O3NPs. All these results indicate the effi- cient incorporation of Zn ions into the network of alu- minium oxide. The NPs in this study exhibited diameter varying between 10 and 20 nm and morphology similar to nanoneedles supported in structures compacted. Studies in vitrorevealed that Al2O3-Zn NPs were not cytotoxic for L929 fibroblast cells, indicating that these materials exhibit potential to be employed in biomedical applications.

Acknowledgments

We are grateful to the Department of Physics/UEM, to COMCAP/UEM, for providing the equipment used in this study and to CAPES and CNPq (Process no. 405381/2016- 6) for financial support.

References

[1] Piconi C and Maccauro G 2016 in Reference module in materials science and materials engineeringSaleem Hashmi (ed) (Oxford: Elsevier) p 1

[2] Radziun E, Wilczynska J D, Ksiazek I, Nowak K, Anus- zewska E L, Kunicki Aet al2011Toxicol. in Vitro251694 [3] Chagas L H, Carvalho G S G, San Gil R A S, Chiaro S S X,

Leita˜o A A and Diniz R 2014Mater. Res. Bull.49216 [4] Pardo P and Alarcon J 2018Ceram. Int.4411486 [5] Zhang X, Pfeiffer S, Rutkowski P, Mokowska M, Kata D,

Yang Jet al2020Appl. Surf. Sci.520146304

[6] Terzic A, Pezo L, Andric L and Mitic V V 2015Ceram. Int.

4111908

[7] Kim H N and Lee S K 2013Am. Mineral.981198 [8] Park E, Lee G, Yoon C, Jeong U, Kim Y, Cho Met al2015

J. Appl. Toxicol.36424

[9] Itou M, Fujiwara A and Uchino T 2009J. Phys. Chem. C113 20949

[10] Zhang T, Zhou Y, Bu X, Wang Y, Zhang M and Hu J 2014 Ceram. Int.4013703

[11] Mozalev A, Sakairi M, Takahashi H, Habazaki H and Hubalek J 2014Thin Solid Films550486

[12] Treccani L, Yvonne T, Lein K, Meder F, Pardun K and Rezwan K 2013Acta Biomater.97115

[13] Lakade S H, Harde M T and Deshmukh P K 2019Particul.

Sci. Technol.02726351

[14] Nemade K R and Waghuley S A 2014Ceram. Int.406109

[15] Mestres G, Espanol M, Xia W, Tenje M and Ott M 2016ACS Omega1706

[16] Periasamy V S, Athinarayanan J and Alshatwi A A 2016 Biotechnol. Appl. Biochem.63320

[17] Choi J and Wang N S 2011 inNanoparticles in biomedical applications and their safety concerns. In biomedical engi- neering from theory to applications R Fazel-Rezai (ed) (Rijeka: InTech Publisher) p 299

[18] Mitragotri S and Lahann J 2009Nat. Mater.815

[19] Dick M K, Miao J H and Limaiem F 2020 Histology, fibroblast(Treasure Island, FL: Stat Pearls Publishing) [20] Melo A, Bet A C, Assreuy J, Debacher N A and Soldi V J

2009Braz. Chem. Soc.201753

[21] Serrano M C, Pagani R, Vallet-Regi M, Pena J, Ramila A, Izquierdo Iet al2004Biomaterials255603

[22] ISO 1099355 2009 International standard ISO specification 10993-5:biological evaluation of medical devices– Part 5.

Tests for in vitro cytotoxicity, Geneva, Switzerland [23] Lopes V R, Schmidtke M, Helena Fernandes M, Martins R

and Vasconcelos V 2011Toxicol. In Vitro25944

[24] Wadajkar A S, Ahn C, Nguyen K T, Zhu Q and Koma- bayashi T 2014ISRN Dentist.2014191068

[25] Song Z, Tang H, Xiaoyong D, Xiang K, Cao A, Liu Yet al 2017J. Nanosci. Nanotechnol.172881

[26] Kannan K, Radhika D, Sadasivuni K K, Reddy K R and Raghu A V 2020Adv. Colloid Interface Sci.281102178 [27] Wang Y, Luo F, Wei P, Zhou W and Zhu D 2015J. Elec.

Materi.442353

[28] Chen S, Cui X, Ding S, Sun Q, Nyberg T, Zhang Set al2013 IEEE Electron. Device Lett.341008

[29] Wang H, Wei Y, Yang Y and Lee J 2005 J. Electron.

Spectrosc. Relat. Phenom.144–147817

[30] Weng Y, Liu H, Ji S, Huang Q, Wu H, Li Zet al2018Appl.

Surf. Sci.4571025

[31] Andrade J L, Oliveira A G, Mariucci V V G, Bento A C, Companhoni M V, Nakamura C V et al 2017 J. Alloys Compd.729978

[32] Fernandes D M, Silva R, Hechenleitner A A W, Radova- novic E, Melo M A C and Pineda E A G 2009Mater. Chem.

Phys.115110

[33] Bouhamed H and Baklouti S 2014Powder Technol.264278 [34] Wang H, Wang F, Liao Q and Li X 2015 Ceram. Int.41

4959

[35] Li J, Wu Y, Pan Y, Liu W and Guo J 2007Ceram. Int.33919 [36] Boumaza A, Favaro L, Ledion J, Sattonnay G, Brubach J B,

Berthet Pet al2009J. Solid State Chem.1821171 [37] Repelin Y and Husson E 1990Mat. Res. Bull.25611 [38] Dan’ko A, Rom M A, Sidelnikova N S, Nizhankovskiy S V,

Budnikov A T, Grin’ L A et al2008 Crystallogr. Rep.53 1112

[39] Goldman, Alex 1999 Handbook of modern ferromagnetic materials (The Kluwer international series in engineering and computer science; SECS 505) p 216,https://doi.org/10.

1007/978-1-4615-4917-8

[40] Kim N, Bassiri R, Fejer M M and Stebbins J F 2014J. Non- Cryst. Solids4051

[41] Ajamein H and Haghighi M 2016Ceram. Inter.4217978 [42] Cunha G C, Roma˜o L P C and Macedo Z S 2014 Powder

Technol.254344

[43] Arier U O A and Tepehan F Z 2014Compos. Part B: Eng.58 147

(13)

[44] Costa T M H, Gallas M R, Benvenutri E V and Jornada J A H 1999J. Phys. Chem. B1034278

[45] Aminzadeh A and Sarikhani-Fard H 1999Spectrochim. Acta A551421

[46] Porto S P S and Khishnan R S 1967J. Chem. Phys.471009 [47] Misra A, Bista H D, Navati M S, Thareja R K and Narayan J

2001Mater. Sci. Eng. B7949

[48] Liu Y, Cheng B, Wang K, Ling G, Cai J, Song Cet al2014 Solid State Commun.17816

[49] Laha T, Balani K, Agarwal A, Patil S and Seal S 2005 Metall. Mat. Trans. A36A3001

[50] Jbara A S, Othaman Z and Saeed M A 2017J. Alloys Compd.

7181

[51] Stojadinovic´ S, Tadic´ N, Radic´ N, Stojadinovic´ B, Grbic´ B and Vasilic´ R 2015Surf. Coat. Technol.276573

[52] Zhang R, Yin P, Wang N and Guo L 2009Solid State Sci.11865 [53] Chang H and Chang Y 2008J. Mater. Process. Technol.207193 [54] Bajaj N S and Omanwar S K 2015J. Sol-Gel Sci. Technol.

751

[55] Trinkler L, Berzina B, Jakimovica D, Grabis J and Steins I 2011Opt. Mater.33817

[56] Trinkler L, Berzina B, Jevsjutina Z, Grabis J, Steins I and Baily C J 2012Opt. Mater.341553

[57] Kostyukov A I, Zhuzhgov A V, Kaichev V V, Rastorguev A A, Snytnikov V N and Snytnikov V N 2018Opt. Mater.75 757

[58] Pustovarov V A, Perevalov T V, Gritsenko V A, Smirnova T P and Yelisseyev A P 2011Thin Solid Films5196319 [59] Amirsalari A, Shayesteh S F and Ghahrizjani R T 2017Chin.

Phys. B26036101

[60] Li P G, Lei M and Tang W H 2010Mater. Lett.64161 [61] Shen Y, He W, Zhang D, Zhang X, Xue Y and Liu C 2011J.

Lumin.1312725

[62] Zhang Y, Yu J, Kahkoska A R and Gu Z 2017Sensors172 [63] Lee J, Kim J and Kim W J 2016Chem. Mater.286417 [64] Lamouri S, Hamidouche M, Bouaouadja N, Belhouchet H,

Garnier V, Fantozzi G et al 2017Boletı´n de la Sociedad Espan˜ola de Cera´mica y Vidrio5647

[65] Krishnamurithy G, Yahya N A, Mehrali M, Mehrali M, Mohan S, Murali M Ret al2016Ceram. Inter.4218247 [66] Ravikanth M, Soujanya P, Manjunath K, Saraswathi T R and

Ramachandran C R 2011J. Oral Maxillofac. Pathol.15247

References

Related documents

The observed trap depth parameters (TL), transition probabili- ties, radiative lifetime and emission cross-section parameters (UV and PL) suggest that the Y 2 O 3 -doped glass sample

These research results revealed that obtaining a bimodal microstructure, i.e., abnormally grown Si 3 N 4 grains sur- rounded by fine glass–ceramic matrix, is an effective method

As shown in figure 3 that an increase in Young’s modulus, shear modulus and bulk modulus of the bioglass samples is due to increase in bridging oxygens (–O–) by the addition of Al 2

Various glass compositions melted in mol percentage of oxides, density, optical band gap, type of vibrations and peak position in the FT-IR spectra of G1, G2, G3, G4 and G5

3.3b Effect of normal load on sliding wear regimes: Figure 11a shows the change in volume loss with an increasing applied load for A10Z composite.. It is noted that

12 In the present study, FTIR spectra of the SFMO samples show three absorption bands corresponding to Fe–O and Mo–O vibrations, namely, strong bands in the high-wavenumber range ∼

Effect of copper doping on structural, optical and electrical properties of Cd 0⋅8 Zn 0⋅2 S films prepared by chemical bath. deposition 53

Optical absorption and fluorescence spectra of Eu 3+ ions in Al(NO 3 ) 3 –SiO 2 sol–gel glass have been investigated using the Judd–Ofelt theory.. The lifetime ( ττ r ) of 5 D 0