Crystal growth

This report introduces briefly some concepts and materials on crystal growth presented by Dr. Zhen-yu Zhang from the Oak Ridge (TN) National Laboratory, and Dr. En-ge Wang from the Institute of Physics, Chinese Academy of Sciences, in a session on crystal growth at the first Chinese-American Frontiers of Science Symposium.

Crystal growth involves a variety of research fields ranging from surface physics, crystallography, and material sciences to condenser mater physics. Though it has been studied extensively more than 100 years, crystal growth still plays an important role in both theoretical and experimental research fields, as well as in applications. For example, how to growth ideal high Tc superconductor crystal has become an dominant subject both for testing of superconductor theories and physical properties. Furthermore, carbon 60 and carbon nano-tubes have opened a new field to both condensed mater physics and chemistry. From the recent discoveries in high Tc superconductors and C60, which brought the Nobel prize to the pioneer researchers in this field, one can understand the importance of crystal growth today.

As the development of scientific instruments and analytical methods, such as x-rays, electron microscopy, NMR, and scanning tunneling microscopy continues, research on crystal growth and structure characterization has entered an atomic level, which makes it possible for further understanding of the physical, chemical, and other properties of the structure nature of various crystals. Especially for the crystals with low dimension and nano-structures, such as carbon nanotubes, blue-light emitting GaN thin films, and magnetic multilayers with giant magneto-resistance, their abnormal properties that have great potential in application can be understood only with the knowledge of structure at the atomic level. Moreover, a further improvement of crystal quality also depends on the structure characterizations.

Based on its importance described above, crystal growth had been chosen as one of the topics in the first Chinese-American Frontiers of Science Symposium. Two speakers, as mentioned above, were invited to discuss electron growth of metal overlayers on semiconductor substrates and the attempt of synthesis of hard materials.

Electron Growth of Metal Overlayers on Semiconductor Substrates

With the development of the information industry, people have paid more attention to thin films used for making various types of sensor and laser devices. From extensive studies, we have seen that the physical properties of thin films closely depend on crystal quality. For important scientific and technological reasons, it often is desirable to prepare metallic thin films on semiconductor substrates with atomically flat interface and growth front. Nevertheless, atomically flat overlayers do not grow in many heteroepitaxial systems. Recent research in thin-film growth has been focused primarily on gaining atomic-scale understanding of various kinetic processes and the stress effects. In the symposium, Dr. Zhen-yu Zhang reported on a concept of electronic growth of metallic overlayers on semiconductor substrates (1, 2).

In a metal thin film on a semiconductor, the conduction electrons are confined in a quantum-size region bounded by the vacuum on one side and the metal/semiconductor interface on the other, which produces quantum levels or sub-bands known as quantum well states. Such quantum-size effect can influence the stability of metal thin films on a supporting substrate. Their model contains three central ingredients: (i) quantum confinement, (ii) charge spilling, and (iii) interface-induced Fried oscillations. Electronic confinement within the metal overlayer can mediate an effective repulsive force between the interface and the metal surface, acting to stabilize the overlayer. Electron transfer from the overlayer to the substrate leads to an attractive force between the two interfaces, acting to destabilize the flat overlayer. Interface-induced Friedel oscillatory modulation in electron density can further impose an oscillatory modulation onto the two previous interfaces. These three competing factors, all of electronic nature, can make a flat metal overlayer critically or marginally stable or totally unstable against roughening.

The electronic growth concept also can be schematically described as the following: As a metal is added onto a semiconductor substrate layer by layer, the motion of the conduction electrons in the metal film is confined by the two vacuum-metal and metal-semiconductor interfaces, forming electronic standing waves. These waves resist being squeezed any further, helping to stabilize the film. If some electrons leak into the semiconductor, the stabilization force would be weakened. These two competing effects determine the critical thickness for smooth film growth.

This theoretical work provides additional understanding of the crystal growth of smooth film and the role of metal-semiconductor interface during the film growth. It also provide a possible way to do quantum engineering of metallic overlayers down to the atomic scale, which may enable fabrication of special films needed for developing new-generation electronic devices.

Carbon Nitride and Related Materials

The investigation and development of carbon nitride and related materials have been a subject of intense research for more than 10 years (3). Much of this research is motivated by the extraordinary combination of physical properties possessed by the covalently bonded materials made from light atomic weight elements from the first row of the periodic table. For example, Cohen (4–6) proposed that carbon nitride should have diamond-like properties with a relatively isotropic arrangement of short (1.47 Å in length) and covalent bonds. Such materials are important in high-performance engineering applications for high-hardness, high-temperature, high-power, or high-frequency devices ranging from microelectronic to space flight applications.

In addition to the potential applications, the goal of the effort is to see whether one can design a high-performance material by beginning with theories to select candidates for laboratory synthesis. As one of the computer-designed structures, this study provides a test of the effectiveness of first-principle calculations in materials science. The recent research on the fabrication of carbon nitrides can be traced back to early 1970s. Since a theoretical prediction by Cohen in 1985 (4), a large variety of the more readily available techniques, such as plasma, sputtering, laser ablation, chemical vapor deposition, ion beam deposition, and high-pressure pyrolysis, have been used for depositing thin films of carbon-based materials.

Most of the early C-N films presented amorphous nature with layer-like structure. To test for the presence of crystalline carbon nitride phases the x-ray diffraction (XRD) data were compared with the patterns calculated for hypothetical crystal structures. A detailed study of the experimental and theoretical XRD results has been summarized (4–6). The structures of these C-N materials also were investigated by electron diffraction. Some groups reported the diffraction rings matched with theoretical β-phase (7–9). Recently, Wang et al. (10) have made an extensive study of the samples grown on nickel substrate by selected area electron diffraction (SAED) analysis performed in transmission electron microscopy. Based on the SAED result, they can obtain the average information of the lattice spacing and symmetry of the crystalline planes. This technique usually supplies a complement to the information obtained in the lattice image. It provides more accurate information than that obtained by measuring the lattice fringes in high-resolution images.

It should be noticed that carbon nitride makes a new material system. Some new phases and related materials have been observed. Very recently, two new C-N structures with tetragonal and monoclinic phases have been identified by Guo et al. (11, 12). The lattice parameters for the tetragonal C-N structure are a = 5.65 Å and c = 2.75 Å with a N/C ratio of about 0.8:1.0. The lattice parameters of the monoclinic C-N structure are a = 5.065 Å, b = 11.5 Å, c = 2.80 Å, and β = 96^o with a N/C ratio of about 0.5:1.0.

A better approach to integrate the superior property of C-N film with the mature technology of diamond on silicon substrate has been practiced by Wang’s group with a two-step growth mode in an electron cyclotron resonance-enhanced chemical vapor deposition. Detailed x-ray photoelectron spectroscopy analyses of the chemical bonding state are given before and after C-N deposition. The nitrogen concentration in the films remains unchanged when the substrate temperature varies from 100°C to 700°C, which suggests a steady nonpolar C-N phase formed. Johansson et al. (13) reported that CN/BNC multilayers had been deposited on Si, high-speed steel, cemented carbide, martensitic steel, and TiN substrates by unbalanced dual cathode reactive magnetron sputtering from C (graphite) and BC targets in an Ar/N² discharge.

A related material, silicon carbonitride (Si-C-N) with β-phase, was identified experimentally and theoretically. Elemental profiles revealed a strong phase separation between B-N layers and carbon layers along the radial direction. Boron carbonitride films have been deposited in our laboratory by using bias-assisted hot filament chemical vapor deposition from gaseous mixtures of CH₄, B₂H₆, N₂, and H₂ on polycrystalline nickel, quartz, molybdenum, silicon, and graphite substrates. The largest BCN rods are of about 20 μM in diameter and 100 μM in length. Those rods with full hollow or part solid insides are composed of many small crystalline particles shown by scanning electron microscopy. X-ray photoelectron spectroscopy and energy dispersive x-ray analysis were used further to confirm their chemical composition and atomic-level hybrid. A hydroazafullerene was synthesized by Keshavazok-K et al. (14) that should open the door to a new chemistry of heterofullerenes.

In addition to the known structures, Wang’s group (10–12) has successfully grown large-scale aligned carbon nitrogen C_1-xN_x (x = 0.15–0.17) nanotubes by using microwave plasma-enhanced chemical vapor deposition with a mixture of methane (CH₄) and nitrogen gases only. The substrate temperature was kept constant (500°C) throughout the deposition process, which is well below the operating temperature for the arc-discharge method or the laser evaporation method. Therefore, the present technique can be conveniently scaled up in more practical nanodevice applications. The films are composed of large quantities of well-aligned nanotubes with a uniform diameter of about 100–200 nm. The individual nanotubes are mostly perpendicular to the substrate surface, and the formed array is about 15–20 μm in height after 3-hr deposition. Furthermore, Wang et al. have studied their field emission properties. Rather uniform emission was observed from large-area film. The density of emitting nanotubes increases with increasing gap field.

The covalently bonded carbon nitride and related films are an interesting, challenging, and technologically important material system that is not only important for basic research but also has the potential for industrial use. In addition to their earlier application in the tribological area, for example, carbon nitride films may be useful in microelectronics. Although the recent spate of success is impressive, much work remains to be accomplished for this system to even distinguish itself as one of the leading materials. A continued effort toward the growth and processing of high-quality film is most important. There is also a need for further study of the fundamental physical and chemical properties of these materials to evaluate the true potential of this new material system.