On the nature of the room temperature ferromagnetism in nanoparticulate co-doped ZnO thin films prepared by EB-PVD
- a Universidade Federal de Alfenas, 37130-000, Alfenas, MG, Brazil
- b Centro de Desenvolvimento da Tecnologia Nuclear, CDTN, 31270-901, Belo Horizonte, MG, Brazil
- c Departamento de Física, Universidade Federal de Lavras, 37200-000, Lavras, MG, Brazil
- d Instituto de Física de São Carlos, USP – Universidade de São Paulo, 13560-970, São Carlos, SP, Brazil
- Received 16 September 2016, Revised 10 November 2016, Accepted 13 November 2016, Available online 14 November 2016
Highlights
- •
Co-doped ZnO (Zn1−xCoxO, x = 0.02 and 0.08) films were prepared by EB-PVD method.
- •
Structural analyses detects a lower density of VO for sample with x = 0.08.
- •
Optical results show the existence of a defect sub-band close to conduction band.
- •
Magnetic data reveal a huge MS for the film with x = 0.08, not correlated to VO.
- •
Magnetic properties are analyzed in the scope of BMP theory related to Zni defects.
Abstract
The complete comprehension of the magnetic properties of the dilute magnetic oxides has emerged as one of the major challenges in the fields of materials science and condensed matter physics. Up to now, there is no consensus about the mechanism behind the so often observed room temperature ferromagnetism. However it is well known that defects plays an important role in the context. Here we present the study of the correlation between the structural and the magnetic properties of nanoparticulate Co-doped ZnO (Zn1−xCoxO, x = 0.02 and 0.08) thin films prepared via Electron Beam-Physical Vapor Deposition. The structural results confirms the incorporation of the Co2+ ions into the ZnO host matrix with no secondary phases. Magnetic measurements show a robust ferromagnetic order for the thin film with x = 0.08, whereas for the sample with x = 0.02 only a tiny ferromagnetism were observed. We could explain the differences in the magnetic behavior entirely under the scope of the spin-split impurity band model. In this work we also present convincing evidences of the no correlation between the oxygen vacancies and the desired room temperature ferromagnetism.
Graphical abstract
Keywords
- Dilute magnetic oxides;
- Defect ferromagnetism;
- Co-doped ZnO thin films;
- Structural and magnetic characterization
1. Introduction
The development of spintronic devices based on metallic systems is undoubtedly a successful research area [1]; [2] ; [3]. Besides, interesting theoretical and experimental properties have been reported on metal/semiconductor hybrid systems [4] ; [5]. However, the initial and direct applicability of such systems was hindered by the low spin injection efficiency due to the well know impedance mismatch across the interface between both materials [6]. This problem was further overcome by using either ballistic electrons [7] or very thin tunnelling barriers between the ferromagnet and the semiconductor [8] ; [9]. Besides, another approach to solve the spin-injection problem would be the use of ferromagnetic semiconductors instead of magnetic metals. Particularly, a new class of materials called dilute magnetic semiconductors (DMS) are attracting intense interest. Mn-doped III-V semiconductors were found to be ferromagnetic, introducing very special new properties, like the electrical control of the magnetization and the Curie temperature (TC) [10]. Nevertheless, the TC for these materials is still far below room temperature, not exceeding 200 K [11] ; [12].
In this context, the study of the magnetic properties of transition metal (TM) doped large band gap semiconductors was triggered by the theoretical work of Dietl et al. reporting a room temperature ferromagnetism (RTFM) for the Mn-doped GaN and ZnO [13]. After all, a huge effort has been concentrated on TM-doped oxides, the dilute magnetic oxides (DMOs), as ZnO, TiO2 and SnO2[14]; [15] ; [16]. However, in spite of the extensive investigations, the origin of the observed RTFM for the DMOs remains inconclusive and controversial. Up to now, the consensus is that defects play an important role to drive the ferromagnetic behavior [17] ; [18], which could explain the disparity among experimental reports and the attested high sensibility to the preparation conditions observed for the DMOs [19].
From the theoretical point of view, in the literature one can find several different models to explain the usual observed RTFM. Early works on such systems attributed the ferromagnetic order to a carrier-mediated mechanism based on the Zener model [13]. However, it could not account for the RTFM in insulating systems. For those cases Coey et al. proposed in 2005 the spin-split impurity band model, also called bound magnetic polaron (BMP) theory [20]. The main common feature of these models is the importance of an exchange interaction between the charge carriers introduced in the oxide host through specific point defects and the magnetic dopants. It is also important to mention a sub-category of BMP theory named F-center exchange (FCE) [21] and the charge-transfer ferromagnetism (CTF) [22]. By your turn, the CTF model were introduced to account for more specific unexplained experimental observations, like moments exceeding the maximum possible spin moment per cation [23]. In contrast to the other models, the magnetic moments in the CTF are entirely localized on itinerant carriers confined in small regions, such as grain boundaries and interfaces [22]. In the same direction, Straumal et al. have also proposed another model relating the RTFM in nanoparticulate doped and undoped oxides [24] ; [25] to magnetic moments of unpaired electrons residing in specific electronic states at the grain boundaries of the nanoparticles. There, they have established an empirical grain size threshold value of around 33 nm below what the ferromagnetism would occur [24].