As a geologist, George Rossman often examined crystals to determine their purity. Commercially available spectrometers were able to characterize the absorption spectra of large samples, but for samples measuring between 5 and 12 µm, no device provided the needed spatial resolution. Rossman and colleagues at the California Institute of Technology's division of geological and planetary science developed their own spectrometer that features two critical design differences from many Fourier transform infrared spectrometers. The first was the light source: While tungsten halogen lamps were adequate for many spectroscopic applications, they did not produce the coherence or spectral brilliance of an Nd:YAG laser. The second modification was employing an optical parametric oscillator (OPO), a laser-pumped crystal with nonlinear optical properties that essentially split a beam of incoming photons into two lower-power beams of different wavelengths. Using an OPO allowed the scientists to generate coherent light that was tunable over a large number of infrared and visible wavelengths. Design test To test their designs, they set up several configurations using two Nd:YAG lasers. For measurements in the 640- to 3200-nm range, they pumped the OPO cavity with a Continuum Q-switched Nd:YAG laser emitting in the second harmonic (532 nm). For the 410- to 710-nm range, they used a Spectra-Physics laser emitting in the third harmonic (355 nm). To generate a particular wavelength within the respective ranges, they adjusted the angle of the two counter-rotating BBO (ß-BaB2O4) crystals inside the oscillator cavity. To test the spectrometer's sensitivity Rossman covered a sample of topaz with a mask that had a 5-µm pinhole. In another test, he placed a mask with a 12.5-µm hole over a sample of beryl. In each case, even with 99.99 percent of the light reflecting off the mask, enough incident light passed through the pinhole to get an accurate reading. Though still in its prototype form, the researchers have eyed the spectrometer as a tabletop alternative to a synchrotron source for biological sampling. They anticipate significant gains in sensitivity, quality and spatial resolution through such modifications as different beam focusing techniques and low-noise amplifiers.