Thin-film materials are playing an ever-increasing role in basic and applied research and are the basis for a large number of devices. In many cases, thin films are used in multilayered configurations so that the interfaces between different materials play a key role in their properties. The thicknesses of thin films being investigated approach interatomic spacing, so interfacial structure and chemistry become dominant. The field is now at a stage where it is no longer sufficient to only obtain qualitative information; it is also necessary to understand the structure and chemistry quantitatively and to determine limits on the accuracy of the various characterization tools. The present and next issues of the MRS Bulletin bring together a number of experts in the most commonly used characterization tools. To highlight and encourage international collaboration, the articles presented here are jointly written by scientific groups in the United States and Europe.

The first issue of this series is dedicated to the most commonly used techniques: electron microscopy, x-ray diffraction, and surface analysis. The article on electron diffraction by Ourmazd, Scheffler, Heinemann, and Rouviere presents studies of atomic resolution microscopy and their limitations. Aspects highlighted are the importance of comparing simulations to the images obtained in the electron microscope, and the importance of theoretical studies to obtain quantitative information. The article by Fullerton, Bruynseraede, and Schuller describes recent developments in the field of x-ray diffraction. This article focuses on a refinement technique capable of quantitative determination of interfacial roughness, interdiffusion, and crystal structure changes in a variety of novel multilayer configurations.

Microscopic properties of thin films are often strongly influenced by departures from “perfection.” These can take the form of extended defects such as dislocations, interfacial roughness, or point defects. Direct imaging of extended defects was one of the early contributions of electron microscopy to solid-state science. Since then, the role of extended defects in controlling the fabrication and properties of thin films has been extensively studied and reviewed. Recently, in-situ observation of strain relaxation in thin-film structures has increased our understanding of dislocation kinetics and its effect on properties of thin films.

In this article, we focus on electron microscopic studies of interfacial roughness, the effect of processing on thin films, and the role and properties of intrinsic point defects in solids. Concurrent development of sophisticated theoretical and experimental approaches has substantially facilitated the investigation of point-defect properties. Here, we illustrate how results from theory and experiment can be combined to form a detailed picture of point-defect diffusion in solids, and highlight areas needing increased attention. Microscopic properties of thin films cannot be covered in a single review. For this reason, and because fabrication of semiconducting thin films has reached unprecedented levels of sophistication, we illustrate this article with references to semiconducting materials.

Our main conclusions can be summarized as follows, (a) Thin films of the highest quality are bounded by interfaces that are microscopically rough. Moreover, thin-film interfaces contain roughness on many length scales, each affecting a subset of the physical properties of interest.

The physical properties of superlattices have been the subject of considerable interest because a wide range of phenomena associated with very thin films, interfaces, and coupling effects can be studied. Recent areas of activity in metallic superlattices include antiferromagnetic coupling of ferromagnetic layers across nonmagnetic spacer layers, giant magnetoresistance, magnetic surface anisotropy, low-dimensional superconductivity, and anomalous mechanical properties. All of these phenomena are strongly affected by the chemical and physical properties of the individual layers and by the superlattice structure. Therefore, a detailed understanding of the properties of superlattices requires a nondestructive, quantitative determination of the superlattice structure.

Because superlattices are not in thermodynamic equilibrium, their structure is sensitive to preparation methods and growth conditions. A dramatic example of superlattice structural dependence on growth conditions is shown in Figure 1, for sputtered Nb/Si superlattices. Increasing the Ar pressure during sputtering decreases the kinetic energy of the deposited atoms, thereby changing their surface mobility, and thus altering growth dynamics. Figure 1 shows the low-angle x-ray diffraction and cross-sectional transmission electron microscopy (TEM) images of [Nb(35 Å)/Si(25 Å)]40, superlattices sputtered in, respectively, 3 and 15 mTorr of Ar. The TEM image of the 3 mTorr superlattice clearly shows the smooth and continuous layering across the entire cross section of the image (≈5 μm). This is characteristic of sputtered metal/semiconductor superlattices used for x-ray optics.

In 1925, P. Auger first observed the so-called Auger electrons in a Wilson cloud chamber. He explained this occurrence as being due to a radiationless transition in atoms excited by a primary x-ray photon source. In 1953, Lander first pointed out that Auger electrons arising from solid samples can be detected in the energy distribution curve of secondary electrons from surfaces subjected to electron bombardment. Moreover, low-energy Auger electrons (∼1 keV kinetic energy) can escape from only the first several atomic layers of a surface since they are strongly absorbed by even a monolayer of atoms. Thus Auger electron spectroscopy (AES) possesses high surface sensitivity. This is one characteristic that makes AES very useful for the study of thin films. For such applications, an important development in AES occurred when Harris showed that the sensitivity of the detection of Auger electrons can be improved by differentiating the electron energy distribution curve with respect to the energy. Furthermore, Weber and Johnson demonstrated that, provided the Auger line profile does not change, the peak-to-peak height in the differentiated energy distribution curves is proportional to the Auger current in the peak. Therefore, in addition to its surface sensitivity, AES also can be used for quantitative studies of thin films.

Like AES, x-ray photoelectron spectroscopy (XPS) is a surface-sensitive technique that uses the energy distribution of electrons ejected from a thin film for quantitative analysis. However, in many ways the information provided by AES and XPS is complementary.