Subpicosecond Pulse Techniques [and Discussion]
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
Subpicosecond optical techniques based on the passively model-locked continuous wave (c.w.) dye laser are reviewed. Recent advances in high power amplification, frequency conversion and tunable probe gating greatly increase the range of applica tions for these systems. Although the first picosecond pulses were generated over a decade ago (De Maria et al. 1966) , we are now witnessing a renewed interest in the development of ultra-short pulse techniques and their application to dynamic studies.
... o dynamic studies. This is largely due to the emergence of new laser systems that offer broader wavelength coverage, subpicosecond resolution, more sensitivity, high repetition rates and greater reliability. Some of these systems are based on the passively modelocked c.w. dye laser, which was the first laser to produce subpicosecond pulses (Shank & Ippen 1974 ) and which has recently made possible advances in high-power amplification of ultrashort pulses and broadband subpicosecond spectroscopy. In this paper we discuss the various experimental capabilities that have been achieved with this type of system. Several applications to molecular studies are described later in this Symposium in the paper by C. V . Shank. Figure 1 compares, in two dimensions at least, some of the different ultra-short pulse laser systems currently available. The three middle systems are directly tunable sources. The broad band coverages shown for Nd: glass and the passively mode-locked c.w. dye laser have been demonstrated by nonlinear optical, frequency conversion techniques. It should be noted that specific experimental considerations influence the choice of system for a particular application, pulse duration may not be the only factor determining temporal resolution in each case. Never theless, the passively mode-locked c.w. dye laser has a clear advantage in this regard. It is also a high repetition rate system that can take advantage of powerful signal averaging techniques. The various capabilities that are now based upon the subpicosecond c.w. dye laser are out lined in figure 2. We divide up our discussion accordingly. Pulse generation The passively mode-locked c.w. dye laser used in all of our studies has been described pre viously . It utilizes two free-flowing dye streams. The Rhodamine 6G gain stream is located near the centre of the resonator and is optically pumped by 5 W of c.w. argon laser power at 514.5 nm. The second stream, near one end of the resonator, contains a mixture of two saturable absorbers, DODCI and malachite green, in relative concentration (around 2 x 10-4 m for each) such that their absorptivities at 615 nm are about equal. With this mixture the laser produces its shortest pulses in a stable regime well above threshold. Since this is a completely passive system, there is no need to accurately stabilize any drive frequency or cavity length. These lasers can operate for long time periods without adjustment and with the pulse width changing by less than 0.1 ps. 27 r i Vol. 298. A