FTIR Instrumentation

Transcription

1 FTIR Instrumentation Adopted from the FTIR lab instruction by H.-N. Hsieh, New Jersey Institute of Technology: 1. IR Instrumentation Two types of instrumentation are used to obtain infrared spectra: Dispersive spectrophotometers, which use a monochromator to produce an infrared spectrum one resolution element at a time. Michelson interferometers, which use a moving mirror to create an interference pattern, or interferogram, from which all resolution elements are determined simultaneously. 2. FTIR Spectroscopy History Chemical infrared spectroscopy emerged as a science in the 1880s. In the 1890s, A. A. Michelson, to further his studies of the speed of light, invented the interferometer. In the early 1940s, chemical infrared spectroscopy was still an immature scientific field. But with commercial development of the optical null dispersive spectrophotometer, later that decade, chemical infrared spectroscopy came into widespread use. Dispersive instruments proved the tremendous value of infrared analysis, and soon became the mainstay of organic characterization laboratories. In 1949 astrophysicist Peter Fellgett used an interferometer to measure light from celestial bodies and produced the first Fourier transform infrared spectrum. But for many years, only a few advanced research groups with access to large, expensive computers and with personnel able to wait up to 12 hours to transform an interferogram into a spectrum used Fourier transform infrared (FTIR) spectroscopy. For two decades more, FTIR spectrometers were limited to studying problems not solvable with dispersive techniques. In the late 1960s when microcomputers able to do the Fourier transform became available, commercial FTIR spectrometers appeared. The 1966 development of the Cooley-Tukey algorithm, which quickly does a Fourier transform (the Fast Fourier Transform or FFT), was also instrumental in the commercialization of FTIR spectrometers. However, the first FTIR spectrometers were large and expensive, and were found primarily in a few well-to-do research labs. Gradually, technology reduced the cost, increased the availability, and enhanced the capabilities of FTIR spectroscopy systems. The performance-to-price ratio provided by FTIR spectrometers today was unthinkable only a decade ago. 1

2 3. Theory of Interferometer Operation The optical system in an FTIR spectrometer is very simple: the interferometer requires two mirrors, an infrared light source, an infrared detector, and a beamsplitter. fixed mirror beam splitter collimator moving mirror detector source focussing mirror Figure 1. The Michelson interferometer The beamsplitter is the heart of the interferometer. Essentially a half-silvered mirror, the beamsplitter reflects about half of an incident light beam while simultaneously transmitting the remaining half. One half of this split light beam travels to the interferometer's moving mirror while the other half travels to the interferometer's stationary mirror. The two mirrors reflect both beams back to the beamsplitter where each of the two beams is again half reflected and half transmitted. Two output beams result: one travels to the detector as the other travels to the source. When the two beams return to the beamsplitter, an interference pattern, or interferogram, is generated. This interference pattern varies with the displacement of the moving mirror, that is, with the difference in pathlength in the two arms of the interferometer. The interference pattern, detected by the infrared detector as variations in the infrared energy level, is what ultimately yields spectral information. 4. More Interferometer Theory An example of how light is affected in an interferometer further illustrates interferometer operation. Consider the interference pattern generated when 2

3 monochromatic light, such as that produced by a helium-neon (HeNe) laser, is the light source. An idealized beamsplitter: - neither generates nor absorbs radiation. - transmits half of all incident light. - reflects half of all incident light. All light energy that enters the interferometer from the source either returns back to the source or travels to the infrared detector. As the moving mirror scans back and forth, more or less than half of the total light energy that originally entered the interferometer reaches the detector. All light that does not reach the detector reaches the source instead. For example, if both interferometer mirrors are equidistant from the beamsplitter, the two light beams travel equal distances. As each beam returns to the beamsplitter, each is again split and then recombined with half of the light from the other interferometer arm. The two component beams that make up the output beam passed to the detector are completely in phase with each other and, therefore, interfere constructively. The two component beams that make up the output beam passed to the source are completely out of phase with each other and, therefore, interfere destructively. In this case, all of the original light energy goes to the detector and none goes to the source. By contrast, if the moving mirror is displaced one-quarter wavelength of the HeNe laser radiation (to produce optical retardation of one-half wavelength), light returning from the interferometer's moving mirror reaches the beamsplitter one-half wavelength out of phase with respect to the light returning from the fixed mirror. This results in complete destructive interference in the detected output beam so no light energy reaches the detector, and all light returns to the source. Figure 2. Constructive (left) and destructive (right) interference. As the moving mirror scans back and forth, the detector observes light and dark bands corresponding to partial and total constructive and destructive interference in the detected output beam. In this monochromatic light example, the interference signal the detector receives produces a cosine interferogram. This wave goes from full intensity at the point where both mirrors are equidistant from the beamsplitter to zero intensity when the moving mirror has traveled one-quarter wavelength. Full intensity is again achieved when the moving mirror has traveled one-half or multiples of one-half wavelength. Similarly, zero intensity occurs at each multiple of one-quarter wavelength of mirror travel. 3

4 An infrared source is polychromatic, emitting light over a broad range of frequencies. Each frequency still yields a unique cosine signal: the resulting interferogram represents the sum of all cosine waves generated by each individual infrared frequency. However, only at the point where both mirrors are equal distances from the beamsplitter does complete constructive interference occur--simultaneous constructive interference of all wavelengths. At any other point, only some wavelengths interfere constructively while others do not. The Fourier transform resolves the frequency and intensity of each cosine wave in the interferogram. In other words, this algorithm converts the measured.intensity-versusmirror-displacement signal (the interferogram) into a plot of intensity versus frequency (a spectrum). Δx Figure 3. Spectrum with three frequencies with different amplitudes (top), interferograms for the three separate frequencies (middle), and Interferograms for the three mixed frequencies. Δx Δx = 0 5. Advantage 5.1 Multiplex Advantage (Fellgett's Advantage) An interferometer does not separate light into individual frequencies before measurement. This means each point in the interferogram contains information from each wavelength in the input light. In other words, if 8,000 data points along the interferogram are collected, each wavelength in the input light is sampled 8,000 times. By contrast, a dispersive spectrophotometer that measures 8,000 individual points across a spectrum samples each wavelength only once. 5.2 Throughput Advantage (Jacquinot's Advantage) The simple optical path of the interferometer (no slits and fewer optical elements) means more energy gets to the sample than is possible with dispersive 4

5 spectrophotometers. This means more energy reaches the detector, increasing the spectrum's potential signal-to-noise ratio. Together the multiplex and throughput advantages allow an FTIR spectrometer to obtain a high-quality infrared spectrum in a fraction of the time needed to get the same spectrum on a dispersive instrument. Also, to increase the resolution of dispersive instruments, the slits through which light must pass are narrowed, thereby decreasing energy throughput. In an FTIR spectrometer, resolution is increased by lengthening the moving mirror stroke length with no decrease in energy throughput. As wavelength resolution increases, the advantages of interferometric versus dispersive measurements increase. 5.3 Frequency Precision (Conne's Advantage) With dispersive instruments, frequency precision and accuracy depend on: 1) calibration with external standards and 2) the ability of electromechanical mechanisms to uniformly move gratings and slits during and between scans. By contrast, the interferometer has an internal frequency standard, generally a heliumneon laser. Also, both mirror movement and detector sampling are clocked by the interferometer fringes from the laser's monochromatic light. All frequencies in the output spectrum are calculated from the known frequency of the laser light. An FTIR spectrometer easily achieves frequency precision and accuracy of better than 0.01 wavenumbers. This means that spectra collected with an FTIR spectrometer can be quantitatively compared whether they were collected five minutes or five years apart. 5.4 Constant Spectral Resolution In an FTIR spectrometer, the resolution of the measured spectrum is the same for all frequencies, not varied throughout the spectrum as is often true with dispersive instruments. 6. Resolution In a dispersive spectrophotometer, resolution depends on slit width and on the reciprocal of the grating's linear displacement. As the slits are narrowed, the range of frequencies passed to the detector decreases, resulting in higher resolution. In an FTIR spectrometer, resolution is increased by increasing optical retardation, that is, by increasing how far the moving mirror moves. A relationship between resolution and optical retardation can be empirically derived by considering a spectrum made up of two sharp bands. Each band is at a distinct frequency and generates a sinusoidal signal. The total signal at the detector is the sum of the two signals. If the two spectral frequencies are relatively far apart, the resulting signals are quite different and only a short movement of the mirror is needed to measure a complete 5

6 cycle of the pattern made by the addition of the two waveforms. On the other hand, if the two frequencies are closer together, the interferometer needs to scan a larger distance to measure a full cycle of the addition pattern. The relationship between spectral line spacing and the interferogram is R = 1/d where R is the spectral resolution in wavenumbers and d is the optical retardation in centimeters. Thus, the length of mirror travel (retardation) must be doubled to reduce instrument resolution by half, or, in other words, to improve instrument resolution by a factor of two. Several different criteria are used to measure FTIR spectrometer resolution. These include the Rayleigh criterion and the full-width-at-half-height (FWHH) method. Measuring instrument resolution with the FWHH method involves collecting an absorption spectrum with very sharp peaks, such as that obtained from low-pressure CO at about 5 torr pressure. fixed mirror electronics beam splitter moving mirror source aperture detector sample Figure 4: Left: Outer view of the Bruker IFS 66v/S FTIR spectrometer. (A) Sample chamber, (B) Interferometer chamber, (C) Laser compartment, (D) Electronic compartment and vacuum control, (E) Detector chamber. Right: Schematic internal view. Bold lines mark the optical pathway. 6