Driving with OPO improves CARS system
aser spectroscopy en compasses many techniques, each with notable advantages and disadvantages. Over the years, the decision on which specific approach to use for a particular problem has been driven by a careful balance between research requirements and experimental difficulty. One well-known method that historically has been plagued by the complexities and limitations of dye-based tunable lasers is coherent anti-Stokes Raman spectroscopy (CARS). The most common of the nonlinear Raman processes
Driving with OPO improves CARS system
Peter C. Chen
aser spectroscopy en compasses many techniques, each with notable advantages and disadvantages. Over the years, the decision on which specific approach to use for a particular problem has been driven by a careful balance between research requirements and experimental difficulty. One well-known method that historically has been plagued by the complexities and limitations of dye-based tunable lasers is coherent anti-Stokes Raman spectroscopy (CARS). The most common of the nonlinear Raman processes, conventional CARS produces strong signals with high spatial resolution, making it a powerful tool for such applications as combustion and plasma analysis. In the past, however, CARS has been hindered by phase-matching problems and the short scan range of dye lasers.
By replacing the dye-laser source with a broadly tunable optical parametric oscillator (OPO; see photo), our re search team at Spelman College (Atlanta, GA) has developed a new approach to CARS that promises to overcome these problems.1 It uses a new, single-wavelength detection method that virtually eliminates interference caused by ambient light, potentially expanding the usefulness of the technique beyond the laboratory to brightly illuminated field environments.2 No longer forced to conduct Raman research in total darkness, researchers may be able to pursue new real-world applications, ranging from combustion and plasma analysis to environmental or industrial monitoring, tasks where high levels of background light have previously been problematic.
Optical parametric oscillator
In the modified CARS system, the frequency-tripled output from a Nd:YAG laser pumps an optical parametric oscillator. Part of the Nd:YAG fundamental output is split off and routed through a beam-delay loop and coherently interferes with the signal and idler beams of the OPO prior to traversing the sample cell (see Fig. 1). The single-wavelength (532 nm) optical signal produced by interaction with the sample is filtered prior to passing through a monochromator and continuing on to the detector.
The CARS technique has been beleaguered by problems associated with maintaining phase-matching, a necessary condition in coherent spectroscopy that results in intense output signals. By mixing the signal and idler beams of the master oscillator-power oscillator (MOPO) while at the same time synchronously scanning them, we can reduce the phase-matching problems that have traditionally challenged CARS and other nonlinear spectroscopies. Col linear alignment of the signal and idler beams is used to cancel changes in their respective k vectors. These two beams are combined with the Nd:YAG fundamental, giving us a total of three input beams to focus into the sample.
The OPO used in this new interference-free CARS ap proach increases the wavelength scan range, while eliminating the traditional complexities of dye-laser operation and maintenance. Pumped by the 400-mJ, 355-nm third-harmonic output of a pulsed Nd:YAG laser, the master oscillator-power oscillator (MOPO-730, Spectra Physics Lasers; Mountain View, CA) consists of two coupled OPOs and is capable of producing wavelengths ranging from 440 to 1800 nm, with high energy (10-80 mJ, 5-6 ns) and narrow linewidth (0.1-0.2 cm-1). Though not a requirement in this application, a frequency-doubling option extends the system tuning range to 220 nm.
The broad tunability of this system allows us to plot the entire vibrational spectrum of a given sample, an achievement that should help dispel the conventional notion that CARS is an impractical analytical tool. With the MOPO, we were able to develop a detailed and complete fingerprint of molecules, a crucial step for accurate identification and extraction of molecular information (see Fig. 2). By comparison, dye lasers have a very narrow scan range, a characteristic that results in an incomplete spectral picture. As a result, users of dye-laser-based CARS systems are compelled to paste together pieces of spectra to cover a full vibrational spectrum, a process that is both inconvenient and time-consuming.
Conventional CARS requires that the operator continually stop the experiment to adjust the angles of the lasers as they are tuned to new wavelengths. Halting the process like this every few nanometers is inconvenient. Likewise, each dye must be replaced as the system is tuned to a new spectral region. Broadband coatings and compensating optics in the new system ensure that no mirror change or realignment is necessary while scanning the MOPO across the spectrum.
A key advantage of nonlinear spectroscopy is high spatial resolution. The three beams used in our system permit us to define a single analysis point in space where all three beams overlap. The combined signal, idler, and pump beams produce one coherent signal beam at a fixed wavelength of 532 nm. This strong signal beam can be easily separated from ambient light, thus permitting easy detection. This attribute gives us very high spatial resolution, while enabling the system to reject high levels of background light. The Nd:YAG-pumped MOPO-730 automatically generates all three beams, adding to the simplicity and accuracy of this new technique.
Our new CARS method is relatively easy to use, a benefit facilitated by the computer-controlled OPO. The MOPO helps us avoid the inherent problems associated with using dye lasers so that we can take better advantage of the payoffs associated with CARS. Furthermore, single-wavelength detection helps remove the spectral crosstalk that is often a major concern in using lasers and light for monitoring and taking measurements.
1. P. C. Chen, Analytical Chem. 68, 3068 (1996).
2. P. C. Chen, Appl. Spectros. 51(3), 376 (1997).
Optical parametric oscillator source improves performance of coherent anti-Stokes Raman spectroscopy system.
FIGURE 1. In the modified CARS system, phase-matching is maintained by mixing the signal and idler beams of the master oscillator power oscillator (MOPO) while at the same time synchronously scanning them. Collinear alignment of the signal and idler beams cancels changes in the respective k vectors. These two beams are combined with output from the Nd:YAG pump source and focused on the sample to generate a coherent signal with a fixed wavelength of 532 nm.
FIGURE 2. Coherent vibrational spectrum generated using synchronously scanned OPO CARS covers a Raman shift range from 600 cm-1 (OPO signal beam wavelength of 550 nm) to 3400 cm-1 (OPO signal beam wavelength of 650 nm). The y axis corresponds to the intensity of the generated 532-nm beam. The various peaks correspond to vibrations from benzene, cyclohexane, nitrogen, and oxygen.
PETER C. CHEN is an assistant professor in the department of chemistry at Spelman College, Atlanta, GA 30314.