Growth and magnetism of metallic thin films and multilayers by pulsed-laser deposition

Abstract

Pulsed-laser deposition (PLD) is a powerful method to grow thin films and multilayers of complex materials such as transition metal oxides. In this case, the most cited advantage of PLD is the simplicity of preserving the stoichiometry of the target material. Recently, there are many reports showing that PLD can significantly improve the growth of even simple metallic thin films/multilayers. Here it is the ultrahigh instantaneous deposition rate and the high kinetic energy of PLD that play the most crucial roles. The improved growth, in particular for the first several monolayers, provides great opportunities to design artificial thin film structures that have promising physical properties.

Introduction

Tailoring the properties of functional materials at atomic scale represents the ultimate goal in materials science research. The most desirable approach is to artificially fabricate structures with atomic precision, i.e. in layer-by-layer, row-by-row, or even atom-by-atom manner. For this purpose, in the past two decades, a great amount of effort has been devoted to improving growth techniques as well as growth simulations. The most noticeable reward came out of the research on two-dimensional magnetic ultrathin films and multilayers [1], [2], [3]. The ability to control thin film growth layer-by-layer has not only advanced our current understandings of two-dimensional physics, but also directly led to the observation of important and technologically relevant physical phenomena such as giant magnetoresistance (GMR) [4], [5], tunneling magnetoresistance (TMR) [6], [7], and ferromagnetism in dilute doped semiconductors [8], [9], [10]. GMR and TMR multilayers have already had a tremendous impact in magnetic data storage devices, and the magnetic semiconductors are defining a radically new avenue in spintronics [11], [12], [13].

Depending on the characteristics of materials and the cost concerns, several techniques have been employed for the growth of thin films and multilayers. These include thermal deposition (TD), sputter deposition (SD), chemical vapor deposition (CVD), and pulsed-laser deposition (PLD). Thermal deposition is the oldest and the most accessible technique to grow thin films. In its refined form, known as molecular beam epitaxy (MBE), thermal deposition is particularly useful for growing high-quality metallic and semiconductor films in a layer-by-layer manner. Typically MBE is conducted in an ultrahigh vacuum (UHV) environment, a necessary condition to keep the films free of contamination for basic science research. While MBE can also be used to grow multi-element thin films by co-deposition from multiple sources, it is much less straightforward than the congruent transfer of multi-elemental materials by sputter deposition. The CVD method, which grows films on a substrate by decomposing a chosen precursor vapor, is preferred in industrial applications for its high and spatially homogeneous growth rates. CVD is also very useful to grow certain metastable thin films such as single crystal CrO₂ [14].

Compared to these growth methods, PLD has a unique capability to produce high-quality thin films of various kinds of materials (for a review, see [15]). The process of PLD involves focusing an intense laser pulse on the surface of a target, and removing target materials in the form of some volatile phases, e.g. a gas or a plasma. The distinct advantages of PLD include its simplicity of use, since the laser is totally decoupled from the growth chamber, and the ability to preserve the stoichiometry of compound materials under optimum conditions. The latter has made PLD the primary option for growing complex materials such as transition metal oxides, particularly after its first successful application in high-T_c superconductor thin films [16].

As for the growth of metallic thin films and multilayers, PLD has played a much smaller role so far. One may argue that this is partly due to its success in growing complex materials, and partly due to the maturity of other techniques such as MBE and sputter deposition in growing metallic systems. However, early experiments have shown that there are not any principle obstacles or challenges to grow metals with PLD [17], [18]. More recent work has given strong evidence showing the superiority of PLD for layer-by-layer growth [19] as well as for growing high-quality thin films of refractory metals [20], [21] and unstable alloys [22]. In particular, the successful application of PLD in magnetic ultrathin films and multilayers, which show clearly improved magnetic behavior as compared with similar MBE-grown structures [23], is making a compelling case that PLD should become a major force in the study of artificial metallic structures.

In this article, we review some recent developments in PLD growth of magnetic films and multilayers, which, in our opinion, represent the most significant impact of PLD in the fabrication of metallic systems. The readers are referred to previous review articles for some earlier work in this area [24], [25]. Whenever possible, the growth and properties of PLD grown films are compared with those of MBE-grown films, since they both can be prepared in a UHV environment to greatly limit contamination effects. The paper is organized as follows: after some general discussions on fundamental issues related to PLD growth of metallic films in Section 2, we give extensive discussions on the ability of PLD to grow metastable structured magnetic thin films and alloys, as well as their corresponding magnetic properties in Section 3. The growth and magnetism of multilayers is described in Section 4, followed by some examples of the congruent deposition of multi-element thin films in Section 5. Section 6 is devoted to summarizing the most recent and promising applications of PLD in metallic systems. In the end (Section 7), we give conclusions and outlook.