Temperature dependence and defect related structure, photoluminescence, (ferro)magnetism and ammonia sensitivity of un-doped nanocrystalline ZnO

https://doi.org/10.1016/j.mseb.2020.114748Get rights and content

Highlights

  • ZnO prepared by decomposition of Zn-propionate and annealing at 400–970 °C show RTFM.

  • g = 2.0065 and g = 1.9632 EPR signals. A prominent violet-blue PL emission for 470–800 °C.

  • Very intense green-yellow PL emission and large number of GBs for 800–970 °C.

  • Strained lattice for 400 °C; Ms enhanced to 0.022 emu/g; TC > 700 °C; high NH3 adsorption.

  • Intrinsic defects in a distorted lattice or pinned by GBs generate RTFM.

Abstract

ZnO nanostructures with intrinsic and extended defects were prepared by rapid decomposition of zinc-propionate and annealing at 400 °C–970 °C. The correlation between the structure/morphology, type of native defects (photoluminescence (PL), EPR) and ferromagnetism was investigated, together with ammonia adsorption capacity. All the samples show room temperature ferromagnetism (RTFM). Crystallite size increases while the unit cell volume, c-axis constant, microstrain and saturation magnetization relax with increasing temperature; morphology varies from aggregated nanoparticles to frameworks of well-welded crystals. 400 °C–800 °C annealed samples show a broad visible and/or a prominent violet-blue PL emission and, two narrow g = 2.0065 and g = 1.9632 EPR signals. 800 °C–970 °C annealed samples exhibit very intense green-yellow photoluminescence. The intrinsic defects in conjunction with a deformed lattice and/or pinned by grain-boundaries appear responsible for RTFM and Curie temperature exceeding 700 °C. Tuning the morphology, PL intensity and ferromagnetic signal by choice of annealing temperature can find applications in (gas) sensing, photonic/optoelectronic and spintronic devices.

Introduction

Over the past two decades, ZnO has gained more attention primarily due to promising ferromagnetic properties [1] with potential applications in spintronic devices in addition to applications in optoelectronics of p-type ZnO semiconductors or n-type ZnO on the (currently available) high quality p-type substrates [2]. Zinc oxide found applications in a large number of other technological (electrical, piezoelectric, photonic, sensing) devices. Some advantages of ZnO over other materials with similar applications are its large exciton binding energy (60 meV), direct wide band gap (3.37 eV) at room temperature and easy fabrication in a wide variety of shapes and morphologies, resulting in inexpensive ZnO-based devices. ZnO is naturally a n-type semiconductor [3] and its properties (structural, optical, electrical, (gas) sensing, magnetic, etc.) are largely influenced by native point defects.

Special attention has been paid to zinc oxide since Dietl et al [1] predicted for transition metal (TM)-doped ZnO a Curie transition, TC, at room temperature. There are some reports for RTFM on un-doped ZnO, although more intense results were reported for transition metal-doped ZnO. Nevertheless, controversial opinions remain regarding the origin and mechanism of ferromagnetism in both un-doped and transition metal-doped ZnO. It is increasingly questionable whether, or to what extent, magnetic ions participate in the ferromagnetism of TM-doped ZnO. More evidence is provided that intrinsic (native) point defects are crucial to generate room temperature ferromagnetism in un-doped ZnO [4], [5], [6], [7], [8], [9], [10] as well as in TM-doped ZnO. There are experimental and theoretical reports [11], [12], [13] that evidence that in the absence of defects TM-doped ZnO is not ferromagnetic. It has been predicted that the paramagnetic, antiferromagnetic (AFM) or spin-glass states are more favourable to the ferromagnetic state and that extra carriers are needed to induce the ferromagnetism in TM-doped ZnO [11], [14]. An experimental confirmation was provided by X-ray magnetic circular dichroism on Co-doped ZnO, which showed that the Co atoms were paramagnetic in the ferromagnetic Co-ZnO films [15]. It was reported that in some TM-ZnO systems, extra carriers could be supplied by n-type native defects [11], [14], [16], [17], such as Zn interstitials, Zni, and oxygen vacancies, VO.

For un-doped ZnO, VO-induced ferromagnetism was predicted by Lany et al. [7], whereas Wang et al. [18] proposed Zn vacancy-induced ferromagnetism. VO and Zni are the most reported, experimentally and theoretically, defects that can provide shallow donor levels in the band gap of n-type ZnO and contribute to ferromagnetism. At the same time, it is believed that VO and Zni are deep levels in the band gap and are unlikely to exist in high concentrations in n-type ZnO. Density-functional theory (DFT) calculations have demonstrated that a low concentration of VO and Zni is expected in n-type ZnO under equilibrium conditions (with a Fermi level near the conduction band minimum) due to their high formation energy [3], [19], [20]. The type and concentration of a native defect depends on its formation energy, which in turn depends on the growth and annealing conditions [21], [22]. The growth environment (e.g., oxygen-rich, zinc-rich or oxygen deficiency) is one of most important parameters. Experimentally, it was shown that when processed in non-equilibrium conditions [30] (high energy electron irradiation [23], high formation rate [24], [25], [26], under a zinc-enriched environment [24], [27], [28], [29], etc.) a high concentration of VO and Zni (providing shallow donors) could be achieved in ZnO. Another challenge is the assignment of an emission luminescence band to a certain type of defect. This problem arises because the different transitions in ZnO obtained by the different methods of preparation and growth can emit very close wavelengths (e.g., [19] and references therein).

There are few reports, especially for un-doped ZnO, about the connection of FM and defect-related PL band (native defects) with lattice parameters [30]. For doped ZnO, it is thought that the substitution of Zn2+ with a foreign ion of a dissimilar radius is mostly responsible for the change in the lattice parameters. A connection between the ferromagnetism and the concentration of Zni was reported in Co-doped ZnO [31], [32]. The variation of saturation moment, Ms, was associated with the increase of Zni as the Co concentration increased; at the same time, an increase in the c-axis lattice constant was observed [31]. For un-doped ZnO, the theoretical calculations show that the formation volume of a defect depends on its type (and charge state) [18]. Experimental results of un-doped ZnO show the impact of Zni on the lattice constants consistent with the processing in the non-equilibrium conditions (e.g., [33], [34]). The observed modification of the c-axis constant and the lattice disorder along the c-axis in zinc oxide was associated with Zni [34].

The aim of this work is to investigate the properties and performance (e.g. Ms, TC, sensitivity to NH3 vapors) of un-doped ZnO nanostructures with different intrinsic and extended defect states as well as to provide more insights into the correlation between structure/morphology, type of (native) defects (PL emission, EPR signal) and RTFM. For this, the Zn-propionate powder was heated to 400 °C at a high rate and used as a precursor for the production of a batch of nanocrystalline ZnO by annealing between 400 °C and 970 °C. We argue, based on experimental facts, that the presence of the intrinsic defects is not a sufficient condition for the occurrence of RTFM in un-doped ZnO. Instead, the coexistence of a non-equilibrium concentration of intrinsic defects with a deformed and strained lattice (here in the samples annealed at low temperatures) and/or pinned by GBs (here in the samples annealed at high temperatures) can generate RTFM in un-doped ZnO. In adition, the work comprises the conditions for achieving high Ms (22.0 × 10−3 emu/g) and Curie temperature exceeding 700 °C (one of the highest TC values reported for un-doped ZnO). The adsorption capacity of ammonia vapors was also evaluated since it in conjunction with the presence of donor intrinsic defects controls the performance of ammonia gas sensors.

Section snippets

Powder preparation

Zinc acetate dehydrate [Zn(OOCCHH3)2⋅2H2O] (Alfa Aesar) and propionic acid (CH3CH2COOH) (Merck), used as a complexing agent, were mixed at room temperature under permanent stirring. Ammonia solution was added drop wise to this mixture with continuous stirring. The mixture was than heated to 80 °C and kept at this temperature for several hours until a gel was produced. The gel was heated, with an average heating rate of 2 °C/min, to 200 °C in air and calcined for a few minutes at this

Decomposition of the Zn-propionate precursor and the choice of processing parameters to ensure the formation of well-crystallized ZnO with large concentration of native defects

The DSC/TG curve in Fig. 1 illustrates the decomposition rout of the precursor. The FTIR spectrum of the precursor powder is shown in Fig. 2 together with the spectra of the p400, p600 and Com970 samples. Fig. 3 displays the XRD patterns for the precursor (the inset) together with the patterns of the 400 °C–970 °C annealed samples (main panel). The FTIR spectra (Fig. 2) and the XRD pattern (inset of Fig. 3) of the precursor powder were identified as Zn-propionate C6H10O4Zn [35]. An analysis of

Discussions

The p470 sample shows the highest RT ferromagnetic signal. Although native defects are thought to be an important factor for the occurrence of ferromagnetism in ZnO, the Com970 sample does not show an important ferromagnetic signal even though it shows a very high intensity of defect-related PL signals (Fig. 9). On the other hand, p400 shows a high Ms despite a relatively low intensity of the PL visible band. In this subsection, by correlating the results of the measurements described in the

Conclusions

Un-doped ZnO nanostructures with different intrinsic and extended defect states were produced by rapid heating (20 °C/min) the Zn-propionate precursor to 400 °C and annealing at 400 °C − 970 °C.

  • -

    -All samples show RTFM with Ms between 12 × 10−3 and 0.092 × 10−3 emu/g.

  • -

    -The as-prepared sample (400 °C) is characterized by (i) a highest c/a ratio, c-axis constant and microstrain and (ii) a broad defect-related PL emission and two narrow g = 2.0065 and g = 1.9632 EPR signals consistent with the

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors gratefully acknowledge the financial support from the CCDI-UEFISCDI PN-III-P1-1.2-PCCDI-2017-0871: project 47PCCDI/2018 and Core Program PN19-03 (contract no. 21N/08.02.2019) of Romanian Ministry of Research and Innovation. The authors thank Dr. I. Pasuk, Dr. G. Stan and Dr. G. Aldica (NIMP, Romania) for their experimental and technical support.

Data Availability Statement

The data will be provided if requested in future.

References (65)

  • A.A. Tsyganenko et al.

    Infrared study of surface species arising from ammonia adsorption on oxide surfaces

    J. Mol. Struct.

    (1975)
  • W. Yu et al.

    New insight into the enhanced photocatalytic activity of N-, C- and S-doped ZnO photocatalysts

    Appl. Catal. B: Environ.

    (2016)
  • T. Dielt et al.

    Zener model description of ferromagnetism in zinc-blende magnetic semiconductors

    Science

    (2000)
  • Ü. Özgür et al.

    J. Appl. Phys.

    (2005)
  • A. Janotti et al.

    Fundamentals of zinc oxide as a semiconductor

    Rep. Prog. Phys.

    (2009)
  • D.E. Motaung et al.

    Shape-selective dependence of room temperature ferromagnetism induced by hierarchical ZnO nanostructures

    ACS Appl. Mater. Interfaces

    (2014)
  • M.A. Garcia et al.

    Magnetic properties of ZnO nanoparticles

    Nano Lett.

    (2007)
  • X. Xu et al.

    Size dependence of defect-induced room temperature ferromagnetism in undoped ZnO nanoparticles

    J. Phys. Chem., C

    (2012)
  • S. Lany et al.

    Anion vacancies as a source of persistent photoconductivity in II-VI and chalcopyrite semiconductors

    Phys. Rev. B

    (2005)
  • M. Venkatesan et al.

    Unexpected magnetism in a dielectric oxide

    Nature (London)

    (2004)
  • N.H. Hong et al.

    Observation of ferromagnetism at room temperature in ZnO thin films

    J. Phys.: Condens. Matter

    (2007)
  • J.B. Yi et al.

    Ferromagnetism in ZnO nanowires derived from electro-deposition on AAO template and subsequent oxidation

    Adv. Mater.

    (2008)
  • S.J. Pearton et al.

    Dilute magnetic semiconducting oxides

    Semicond. Sci. Technol.

    (2004)
  • T. Shi et al.

    Structures and magnetic properties of wurtzite Zn1−xCoxO dilute magnetic semiconductor nanocomposites

    Appl. Phys. Lett.

    (2007)
  • T. Shi et al.

    The role of Zn interstitials in cobalt-doped ZnO diluted magnetic semiconductors

    Appl. Phys. Lett.

    (2010)
  • Q. Wang et al.

    Magnetic properties of transition-metaldoped Zn1−xTxO (T=Cr, Mn, Fe Co, and Ni) thin films with and without intrinsic defects: a density functional study

    Phys. Rev. B

    (2009)
  • T. Tietze et al.

    XMCD studies on Co and Li doped ZnO magnetic semiconductors

    New J. Phys.

    (2008)
  • E. Biegger et al.

    Defect induced low temperature ferromagnetism in Zn1-xCoxO fi De

    J. Appl. Phys.

    (2007)
  • K. Sato et al.

    Abinitio study on the magnetism in ZnO-, ZnS-, ZnSe and ZnTe-based diluted magnetic semiconductors

    Phys Stat. Sol. (b)

    (2002)
  • Q. Wang et al.

    Vacancy-induced magnetism in ZnO thin films and nanowires

    Phys. Rev. B: Condens. Matter. Mater. Phys.

    (2008)
  • A. Janotti et al.

    Native point defects in ZnO

    Phys. Rev. B

    (2007)
  • P. Erhart et al.

    First-principles study of intrinsic point defects in ZnO: role of band structure, volume relaxation, and finite-size effects

    Phys. Rev. B

    (2006)
  • Cited by (0)

    View full text