Sub-micron-sized dust grains originate in the interstellar medium, and get incorporated to the disk during the proto-stellar collapse. The smallest of the dust particles easily grow to larger sizes, since random Brownian motions facilitate non-destructive collisions in a short timescale. Large dust particles decouple from the gas and quickly settle toward the disk mid-plane, sweeping up other particles and growing. But once macroscopic sizes are reached (centimeter to decimeter sizes), collisional coagulation efficiencies drop and the outcome of collisions is either fragmentation or bouncing. At this point, the interaction of these grains with the gaseous disk is critical: gas drag reduces their angular momentum and large particles rapidly spiral inwards into the star, failing to stay in the disk reservoir. Yet, it is these macroscopic grains that will eventually become the building blocks of planets, and somehow they must survive. There is currently an active effort to develop better theoretical models and design more realistic laboratory experiments, but direct observational constraints on particle growth in circumstellar disks are critical to improve our understanding of this astrophysical process.
From an observational point of view, grain growth can be inferred by measuring the spectral energy distribution of circumstellar disks at long wavelengths, which traces the continuum dust emission spectrum and hence the dust opacity. Several observational studies indicate an evolution of the dust component in protoplanetary disks, suggesting at least 4 orders of magnitude growth in particle size. Furthermore, thanks to now-available spatially resolved observations of circumstellar disks at multiple millimeter and centimeter wavelengths, evidence on a radial dependance of the dust properties inside the disk has been obtained. In this talk, I will discuss these recent observational constraints and its comparison with physical models of grain evolution that include collisional coagulation, fragmentation, and the interaction of these grains with the gaseous disk (the radial-drift problem). Current results suggest that the first step toward planet formation is modulated by the interaction of millimeter and centimeter-sized dust grains with the gaseous disk.