The next generation of Si devices requires thermal treatments of 1200°C – 1300°C but can only withstand temperatures above 800°C for a few milliseconds. Current rapid thermal processing techniques cannot meet these requirements. We have designed, constructed, and tested a microwave reactor that heats Si to 1300°C in only a few milliseconds and cools the wafer at a rate that exceeds a million degrees per second. Applying millisecond microwave annealing to ultra-shallow junction formation in advanced Si devices shows that this technique meets or exceeds the thermal processing requirements for the next several generations of Si devices.

Electromagnetic heating provides a novel alternative to “illumination-based” rapid thermal processing techniques. Exposure to radiation, in the RF and microwave frequency regimes, rapidly heats silicon (∼125°C/sec) to temperatures in excess of 1000°C without the use of a susceptor. These ramp rates make this technology suitable for the activation of shallow implanted dopants, and satisfaction of the 100 nm technology node has been achieved. Furthermore, the presence of high frequency electric fields creates ponderomotive forces that may alter the kinetics of dopant activation and diffusion. These additional driving forces could, once fully understood, lead to an enhanced activation mechanism that activates sufficient dopants to satisfy the 70 nm technology node at temperatures less than 1000°C.

A theoretical model of microwave absorption in linear dielectric (non-ferroelectric) ionic crystals that takes into account the presence of point defects was synthesized and verified using NaCl single crystals. In the next stage of this research, we will introduce a controlled density of dislocations into the single crystal NaCl samples and study the effect on the microwave absorption mechanisms (ionic conduction, dielectric relaxation and multi-phonon processes) both theoretically and experimentally. Qualitative outlines of this modified theory are presented. The loss factor ε’ has been measured in the dislocation-free case by a cavity resonator insertion technique and the experimental results are in good agreement with the theoretical model. We describe the sample preparation technique that will be used to produce a controlled dislocation density in single crystal samples that will also be studied in our cavity resonator insertion system.

Models for nonthermal effects on ionic motion during microwave heating of crystalline solids are considered to explain the anomolous reductions of activation energy for diffusion and the overall faster kinetics noted in microwave sintering experiments and other microwave processing studies. We propose that radiation energy couples into low (microwave) frequency elastic lattice oscillations, generating a nonthermal phonon distribution that enhances ion mobility and thus diffusion rates. Viewed in this manner, it is argued that the effect of the microwaves would not be to reduce the activation energy, but rather to make the use of a Boltzmann thermal model inappropriate for the inference of activation energy from sintering-rate or tracer-diffusion data. A highly simplified linear oscillator lattice model is used to qualitatively explore coupling from microwave photons to lattice oscillations. The linear mechanism possibilities include resonant coupling to weak-bond surface and point defect modes, and nonresonant coupling to zero-frequency displacement modes. Nonlinear mechanisms such as inverse Brillouin scattering are suggested for resonant coupling of electromagnetic and elastic traveling waves in crystalline solids. The models suggest that nonthermal effects should be more pronounced in polycrystalline (rather than single crystal) forms, and at elevated bulk temperatures.

Reports of enhanced sintering rates associated with microwave heating may be due to nonthermal lattice fluctuation statistics. Recent theoretical analyses reviewed in this paper confirm the feasibility of this phenomenon for a wide variety of situations involving very different microwave absorption mechanisms. For materials with weak microwave absorption coefficients, the effect is expected to be uniformly distributed throughout the volume. For strongly absorbing materials, however, the effect is expected to be concentrated near the material surface, with a characteristic exponential penetration depth of Lnt ∼ 10 - 100 μm. An “observable” nonthermal effect depends on the relative magnitude of the microwave electric field strength |E| and the lattice ion energy relaxation rate γ with the most pronounced effects occurring for larger values of |E| and smaller values of γ.