A compact noise-immune cavity-enhanced optical heterodyne molecular spectroscopy (NICE-OHMS) system based on a narrow linewidth distributed-feedback laser and fiber-coupled acousto-optic and electro-optic modulators has been developed. Measurements of absorption and dispersion signals have been performed at pressures up to 1/3 atmosphere on weak acetylene transitions at 1551 nm. Multiline fitting routines were implemented to obtain transition parameters, i.e., center frequencies, linestrengths, and pressure broadening coefficients. The signal strength was shown to be linear with pressure and concentration, and independent of detection phase. The minimum detectable on-resonance absorption with a cavity with a finesse of 460 was 2 × 10−10 cm−1 for 1 minute of integration time.
© 2010 OSA
Noise-immune cavity-enhanced optical heterodyne molecular spectroscopy (NICE-OHMS) is a highly sensitive absorption technique that relies on a unique combination of cavity enhancement, for increased interaction length with the sample, with frequency modulation spectroscopy (FMS), for reduction of 1/f noise [1,2]. Due to the use of an external cavity, the technique has the capability to provide both Doppler-broadened [3–8] and sub-Doppler [1,9–11] signals. The main advantage of sub-Doppler detection is that it provides high spectral selectivity and resolution in combination with reasonably large signals (the peak-to-peak value of the sub-Doppler dispersion signal can be up to 1/4 of the on-resonance Doppler-broadened absorption signal ). However, it has the drawback that it requires optically saturating conditions, which for most molecular transitions occur at sub-torr pressures, limiting the detectability for atmospheric pressure samples in terms of analyte concentration. Detection of Doppler-broadened transitions is therefore still the preferred mode of operation for many types of applications, primarily for trace gas detection or detection of weak molecular transitions, since the signal is present also at tens or hundreds of torr pressures.
NICE-OHMS has been developed with the use of highly stable solid state lasers (mainly Nd:YAG [1,13]) for frequency standard applications, in which the absolute frequency stability of the laser is more important than its tunability. A record sensitivity of 1 × 10−14 cm−1 in 1 s integration time was achieved for sub-Doppler detection using a cavity with an ultra-high finesse (105) . More recent systems, incorporating tunable lasers and cavities with finesse in the 102 – 104 range, routinely obtain sensitivities in the 10−11 – 10−10 cm−1 range for Doppler-broadened detection, and in the 10−13 – 10−11 cm−1 range for sub-Doppler detection . External cavity diode lasers (ECDL) have often been employed for NICE-OHMS [3,5,8] due to their wide tunability and relatively low price. These lasers are, however, susceptible to mechanical disturbances due to the use of an external grating, which limits their use for practical field applications. A promising alternative are quantum cascade lasers , which operate in the mid-IR wavelength range that encompasses the strong fundamental vibrational transitions of molecules, since these can provide a high detectability in terms of concentration. However, the availability of optical components and sensitive high-speed detectors in the mid-IR is still restricted, which presently limits the sensitivity of NICE-OHMS in this spectral range. Recently, much work has been devoted to development of a NICE-OHMS system based on an erbium-doped fiber laser, whose inherent narrow linewidth simplifies considerably the locking of the laser frequency to a cavity mode [6,7,11]. Its fiber-coupled output also allows the use of integrated optics devices, making the system compact. The disadvantage of such a system is the limited tunability of the fiber laser (presently up to 3 GHz), which allows for detection of only a single Doppler-broadened transition per scan.
Because of their wide tunability (up to several tens of GHz by current tuning) and rugged design, distributed-feedback (DFB) lasers are today often used for spectrometric applications, e.g., in commercially available tunable diode laser absorption spectrometry systems. Although the total (temperature) wavelength tunability of a DFB laser is smaller than the grating-controlled tunability of ECDLs, the mode-hop-free current tuning range is similar. The laser is, however, inherently more stable, and its fiber-coupled output allows the use of integrated optics components, such as fiber-coupled electro-optic and acousto-optic modulators. It is therefore of interest to investigate to which extent such a laser can be incorporated into a NICE-OHMS system.
In this work we demonstrate the first NICE-OHMS system based on a DFB laser with a free-running linewidth of 70 kHz, operating in the 1549-1555 nm wavelength range. Although sub-Doppler detection was possible with the current setup, in this first assessment of DFB-laser-based NICE-OHMS we stress the advantage of the wide tunability of the DFB laser by focusing on multipeak detection of pressure-broadened acetylene transitions. Since the operating wavelength range of the laser overlaps with a number of fairly poorly characterized weak transition bands of acetylene, multiline fitting was employed to demonstrate the ability of DFB-NICE-OHMS to retrieve relevant line parameters of such lines (i.e., their center positions, linestrengths, and widths). Significant deviations from the data given in the HITRAN database  were found. The linearity of the technique was demonstrated at pressures up to 1/3 of an atmosphere. Lineshapes were modeled using the theory by Ma et al. , which is shortly recalled here.
The Doppler-broadened NICE-OHMS signal can be expressed as a product of a signal strength, , and a general NICE-OHMS lineshape function, , as 
The general NICE-OHMS lineshape function is a combination of absorption lineshape functions, , for the sidebands, and dispersion lineshape functions, , for the sidebands and the carrier15]15], which are independent of pressure.
3. Experimental setup and procedures
A schematic illustration of the experimental setup is shown in Fig. 1 . The system was based on a narrow linewidth DFB laser (NEL, NLK1C6DAAA) and a cavity with a finesse of 460 and a free spectral range (FSR) of 381 MHz. The free-running linewidth of the laser was specified to be 70 kHz, the wavelength was temperature tunable between 1549 and 1555 nm, and the mode-hop-free injection current tuning range was 65 GHz. The laser was driven by a custom-made high-bandwidth low-noise current driver (Metrinova AB), which provided means of external modulation with a bandwidth of a few MHz and maximum amplitude of 8.5 GHz. The flat input mirror and concave (radius of curvature 1 m) output mirror (Layertec) of the optical cavity were glued to low voltage ring PZT actuators (Piezomechanik, HPSt 150/20), which were used to tune and scan the cavity length. The gas chamber, which also served as a mirror spacer to which the PZTs were glued, was made of Zerodur (Schott AG).
The light emitted from the DFB laser was first led through some fiber-coupled components: two optical isolators (OI, Opneti, IS-D-A-1530-900-5) to prevent optical feedback to the laser diode, an acousto-optic modulator (AOM, AA Opto-Electronic, MT110-IIR25-3FIO), and an electro-optic modulator (EOM, General Photonics, LPM-001-15). All fibers in the setup were polarization maintaining single-mode fibers with FC/APC connectors. The light left the fiber through a collimator (Thorlabs, F240APC-1550) and was mode matched to the cavity TEM00 mode by the use of a single lens. The light reflected from the cavity was picked up with the help of a polarizing beam splitter (PBS) cube and a quarter-wave plate, and directed onto a 1 GHz bandwidth photodetector (PD1, New Focus, 1611). A free-space polarizer was used to clean the polarization state of the light coming from the fiber and to adjust the optical power in front of the cavity. A fraction of the power was redirected to a reference photodetector (PD3, New Focus, 1611) with the help of a half-wave plate and the PBS. The light transmitted through the cavity was detected with a 1 GHz bandwidth photodetector (PD2, New Focus, 1611).
Since FMS is insensitive to multiple reflections originating from optical surfaces that are separated by a multiple of the cavity length (i.e., it is not affected by background signals from etalons whose FSR is equal to the modulation frequency or a multiple thereof at dispersion phase or to a multiple of half the modulation frequency at absorption phase ), free space optical components were placed at distances close to the cavity length (40 cm) whenever possible. In order to minimize the background signal originating from multiple reflections between the detectors PD1 and PD2 and the cavity mirrors, these detectors were placed 40 cm away from the input and output cavity mirrors, respectively. Detector PD3 was placed 40 cm away from the beamsplitter as in this way no additional background signals from reflections involving the PBS were created. The distance between the collimator and the input cavity mirror, which is fixed for a given choice of mode-matching lens, was 70 cm, away from the optimum of a multiple of 40 cm, wherefore a free-space optical isolator (Isowave, I-15-UHP-4) was placed directly after the collimator to reduce multiple reflections between the collimator and the cavity mirrors, as well as other free space optical components.
The laser frequency was locked to a cavity mode with the Pound-Drever-Hall (PDH) frequency stabilization technique . An RF signal at 20 MHz was applied to the EOM and the PDH error signal was obtained by demodulating the signal from PD1 at this frequency. The correction signal was applied to the laser diode solely through the current driver. The closed loop bandwidth of the servo was ~200 kHz, limited by the phase response of the laser diode, which developed to −180° already at 1 MHz. The optical power incident on the cavity was stabilized to 0.96 mW by the use of an AOM. The error signal was derived from the reference detector PD3, integrated and fed back to the AOM RF driver (AA Opto-Electronic, MODA-110-4W), operating at 110 MHz. The laser wavelength was monitored with a wavemeter (Burleigh, WA-1500), to which the 1st order output of the AOM (shifted by 110 MHz with respect to the 0th order output) was connected.
For detection of NICE-OHMS signals an RF signal at 381 MHz was applied to the EOM from a tunable voltage controlled oscillator (VCO, Metrinova AB), producing phase modulation with a modulation index of 0.4. The frequency of the VCO was tunable within ± 100 kHz range and was locked to the cavity FSR with the deVoe-Brewer method . An error signal was obtained by demodulating the cavity reflected signal at 361 MHz, i.e. at the difference between the two RF-frequencies already existing in the NICE-OHMS setup, integrated and fed as a correction signal to the VCO. The RF signal from PD2 was amplified (Mini-Circuits, ZRL-700), demodulated in a double-balanced mixer (Mini-Circuits, ZLW-1), and low-pass filtered (Mini-Circuits, PLP-5). A phase shifter (Mini-Circuits, JSPHS-446) was placed in the reference arm to provide adjustment of the detection phase. The NICE-OHMS signal was further amplified and low-pass filtered with a relatively high corner frequency of 500 Hz to allow for fast signal acquisition. The signals were collected by a 16-bits data acquisition card (National Instruments, PCI-6251) and stored on a computer for post-detection processing.
Scanning of the cavity length, and thus the laser frequency, was achieved by feeding a triangular voltage ramp to one of the cavity PZTs. The maximum stroke of the longer and shorter PZT corresponded to a shift of the cavity mode frequency by 16.5 GHz and 7.6 GHz, respectively. The optical frequency scale was calibrated with the laser unlocked by scanning the PZT with the same amplitude as during the signal acquisition and using the cavity modes as frequency markers. Signals from a single acetylene line were acquired at a rate of 1 Hz and averaged 10 times, while for measurements of spectra consisting of four consecutive lines the PZT was scanned at a rate of 100 mHz with no averaging. A background signal was measured after each (series of) measurement(s); either with an empty cavity and the cavity mode frequency adjusted to compensate for the pressure shift of the cavity mode (for phase calibration and measurements on a single line) or with the cavity filled with the buffer gas at the same total pressure as during the measurement (for measurements of spectra). No background was taken in connection to the determination of the Allan variance.
The maximum pressure range that could be used without unlocking of the laser was 250 Torr, limited by the tuning range of the cavity PZTs. The pressure in the gas chamber was measured with a capacitive sensor (Leybold, Ceravac CTR-90) covering a pressure range from 1 to 1000 Torr. A gas mixture of 1000 ppm of C2H2 in N2 and pure acetylene were used.
The detection phase was calibrated on the Pf(30) acetylene transition at 6448.343 cm−1 (1550.786 nm), with a molecular linestrength of 1.51 × 10−23 cm−1/(molecule cm−2)  corresponding to S = 3.74 × 10−4 cm−2/atm at 23°C, and a Doppler width of 233 MHz. NICE-OHMS signals from pure acetylene in the Doppler limit, i.e., at pressures in the mTorr range, were measured and Eq. (1) (with the Gaussian approximation of the absorption and dispersion lineshapes, i.e. with the homogenous linewidth set to zero) was fitted to the signals with the phase, center frequency, signal strength, and linear background as fitting parameters. Optical saturation, which was observed at sub-torr pressures, did not affect the calibration, since the shape of pure Doppler-broadened absorption and dispersion signals is not affected by optical saturation in the Doppler limit [15,19]. Using this method, the experimental phase setting corresponding to absorption and dispersion signals could be determined with an accuracy of 0.1 rad.
4. Results and discussion
Figure 2 shows background corrected absorption and dispersion NICE-OHMS signals from four acetylene transitions around 1550.8 nm [from left to right Pe(21) in the band, Pf(30) and Pe(30) in the band, and Pe(21) in the band], taken with 1000 ppm of acetylene at total pressures of 81, 137, and 183 Torr, respectively. The sum of four expressions for the NICE-OHMS signal, Eq. (1), was fitted to each spectrum with the center frequencies, homogenous linewidths, signal strengths and a linear background of the four transitions as fitting parameters. The fits are shown by the gray curves and the corresponding residuals are displayed below each panel, proving satisfactory agreement between fits and signals. The various transitions overlap partly at pressures above 100 Torr, and the overlap is larger at absorption than at dispersion phase, due to the larger width of the absorption signal.
Table 1 lists the parameters of the four lines determined experimentally in this work, together with the corresponding values found in the literature. The line center positions were measured with the wavemeter (with an accuracy of 30 MHz) at sub-torr pressures with the laser frequency tuned to the center of the sub-Doppler dispersion signals. The linestrengths and pressure broadening coefficients are obtained from fits shown in Fig. 2 (with uncertainties given by one standard deviation of the mean). The spacings of the lines obtained from the fits agreed well with those measured with the wavemeter (to within 20-80 MHz). We assumed that the linestrength of the strongest line, Pf(30), listed by El Hachtouki and Vander Auwera , is the most accurate, and we therefore rescaled the linestrengths of the other three lines with respect to this. The linestrength and the position of the Pe(21) line at 6448.59 cm−1 agreed then reasonably well with that given by Tran et al. . The linestrengths of the other two transitions differed significantly from those listed in the HITRAN database ; the Pe(21) line at 6448.176 cm−1 was found to be 70% stronger and the Pe(30) 17% weaker. Moreover, the position of the Pe(30) line was found to be shifted by −0.007 cm−1 (−210 MHz). This discrepancy originates most probably from the fact that the positions and linestrengths of these two lines listed in HITRAN are only calculated and have not been confirmed experimentally . No data for the pressure broadening of the measured lines by N2 was found in the literature. The obtained values are therefore compared to the broadening coefficients for the corresponding lines in the similar acetylene band at 1550 nm measured by Arteaga et al. .
In order to demonstrate the linearity of the signal strength with partial pressure of the analyte, absorption and dispersion NICE-OHMS signals from 1000 ppm acetylene in N2 were measured at total pressures up to 250 Torr at the Pf(30) line. Equation (1) was fitted to the measured signals with the center frequency, homogenous linewidth, signal strength and a linear background as fitting parameters. The fits included the presence of the two neighboring partly overlapping lines (whose parameters were held fixed). Figure 3a shows the signal strength at absorption and dispersion phase (solid square and open circular markers, respectively) as a function of pressure, together with the corresponding linear fits (forced through zero). The figure shows that the signal strength, as defined by the Eqs. (1) and (2), is linear with partial pressure of the analyte for both detection phases. The slopes of the response curves are 2.67(6) × 10−3 V/Torr at absorption phase and 2.63(2) × 10−3 V/Torr at dispersion phase; the small difference originates from an error in the phase setting.
Figure 3b shows the homogenous width of the Pf(30) transition as a function of pressure with the corresponding linear fits (forced through zero), whose slopes, i.e., the pressure broadening coefficient, were found to be 2.23(1) and 2.15(1) MHz/Torr at absorption and dispersion phase, respectively. The difference between the coefficients obtained at the two phases (3.5%) can originate from an error in the calibration of the detection phase or the use of a simplified model of the lineshapes, i.e., the Voigt function, which neglects velocity-changing collisions (i.e. Dicke-narrowing) and speed-dependent effects, which play a role under the prevalent pressure conditions . Theoretical simulations show that a 0.1 rad error in the phase setting results in a 20% error in the signal strength and 30% error in the homogeneous width at the dispersion phase. These errors are smaller at absorption phase, equal to 15% and 3%, respectively. Since the determination of the pressure broadening coefficient is less sensitive to errors in the detection phase around the absorption phase, the pressure broadening coefficients listed in Table 1 were taken from absorption measurements only. The influence of velocity-changing collisions on the NICE-OHMS lineshape function (and the transition parameters) is under investigation and will be reported elsewhere.
Finally, Fig. 3c shows the peak-to-peak values of the NICE-OHMS signals as a function of pressure together with curves calculated from Eq. (1) using the pressure broadening coefficient from Table 1. The maximum peak-to-peak values of the absorption and dispersion signals occur at 150 Torr and 88 Torr, respectively, as predicted by the theory.
The sensitivity of the instrumentation was determined using the Allan variance. Absorption and dispersion NICE-OHMS signals from 1000 ppm of acetylene were measured continuously for 3 hours at a 1 Hz acquisition rate, at pressures of 120 Torr for absorption and 80 Torr for dispersion phase (thus close to those that maximize the peak-to-peak value of the signals for a given relative concentration). Figure 4 shows examples of these signals, together with fits of Eq. (1), where the fitting parameters were the center frequency and the signal strength of the targeted line, together with a linear background. All other parameters (detection phase and linewidth), as well as those of the neighboring lines, were held fixed.
The signal strengths obtained from each fit were recalculated to the corresponding on-resonance absorption, , defined as , giving rise to the two curves shown in Figs. 5a and b (for absorption and dispersion phase, respectively). Figure 5c shows the square root of the Allan variance for the two phases. The solid line that is fitted to the data at short integration times has the characteristic dependence of white noise  and a slope of 1.5 × 10−9 cm−1 Hz-1/2. The noise level at the two detection phases is similar, but the dispersion phase deviates earlier from the white noise behavior, which indicates that the two detection phases are affected by different types of background signals at longer integration times. The dispersion phase is presumably more influenced by relative drifts of the phases of the three modes of the FM triplet induced by the PM fibers, whereas the absorption phase is dominated by drifts of etalons from thin optical elements (such as the sloping background visible in Fig. 4a). The square root of the Allan variance in the white noise dominated regime is equivalent to the standard deviation, wherefore it can be used to determine the sensitivity. The minimum detectable on-resonance absorption is 2 × 10−10 cm−1 at 70 s for absorption and 5 × 10−10 cm−1 at 10 s for dispersion phase. Due to the low linestrength of the addressed transitions, these sensitivities correspond to concentrations of acetylene in the hundreds of ppb range, namely 125 ppb at 120 Torr and 375 ppb at 80 Torr for absorption and dispersion phase, respectively.
The shot-noise limited single pass absorption in NICE-OHMS is given by 26].
The incorporation of a DFB laser into a NICE-OHMS setup is an important step in the development of NICE-OHMS towards a user-friendly technique for trace gas detection, due to the rugged design and tens of GHz injection current tunability of the laser. The fiber-coupled output of the laser allows straightforward use of other fiber-coupled optical components, which makes the system compact and versatile. The wide tunability, in turn, allows acquisition of scans encompassing several pressure-broadened transitions (with a scanning range that is limited by the maximum stroke of the cavity piezo), which facilitates the analysis of dense spectra (including determination of line parameters) and allows for assessment of gases at optimum pressures (in the hundreds of torr range).
It was shown that the NICE-OHMS dispersion and absorption signals from a given concentration are maximized for pressures of 90 and 150 Torr, respectively, for the particular transition and modulation frequency used. Thus, in order to obtain an optimum signal to noise ratio, the pressure of an atmospheric sample should be reduced by a factor of 5 - 9. Moreover, it has been shown that the NICE-OHMS signal strength is linear with partial pressure of the analyte and that it does not depend on the detection phase, which allows rapid determination of gas concentration by line fitting. The system has demonstrated detection of pressure-broadened signals with a sensitivity of 1.5 × 10−9 cm−1 Hz-1/2, which corresponds to a minimum detectable on-resonance absorption of 2 × 10−10 cm−1 at 70 s integration time at absorption phase.
A multipeak fitting procedure was implemented to analyze data and retrieve line positions, relative linestrengths, and linewidths of several weak, partly overlapping acetylene transitions. Some of the parameters were found to be significantly different than that given in the HITRAN database . Although the Voigt lineshape fits well to the measured data, some residuals are still observed, and the use of other lineshape profiles, in particular those that take velocity-changing collisions into account, e.g., Rautian and Galatry, could improve the quality of the fits in the tens and hundreds of torr pressures range.
In conclusion, the high sensitivity and wide tunability of the laser give the system a high potential for trace gas detection, as well as for accurate assessment of line parameters, e.g. pressure broadening coefficients and linestrengths. The latter is of particular interest in this wavelength range in which there is a lot of unassigned lines as well as lines that are only theoretically predicted.
This work was supported by the Swedish Research Council under the project 621-2008-3674. The authors would also like to acknowledge the Kempe foundations and Carl Trygger’s foundation for support. †The present address of A. F. is JILA, National Institute of Standards and Technology and University of Colorado, and the Department of Physics, University of Colorado, Boulder, Colorado 80309-0440, USA.
References and links
1. J. Ye, L. S. Ma, and J. L. Hall, “Ultrasensitive detections in atomic and molecular physics: demonstration in molecular overtone spectroscopy,” J. Opt. Soc. Am. B 15(1), 6–15 (1998). [CrossRef]
2. A. Foltynowicz, F. M. Schmidt, W. Ma, and O. Axner, “Noise-immune cavity-enhanced optical heterodyne molecular spectroscopy: Current status and future potential,” Appl. Phys. B 92(3), 313–326 (2008). [CrossRef]
3. L. Gianfrani, R. W. Fox, and L. Hollberg, “Cavity-enhanced absorption spectroscopy of molecular oxygen,” J. Opt. Soc. Am. B 16(12), 2247–2254 (1999). [CrossRef]
4. J. Bood, A. McIlroy, and D. L. Osborn, “Measurement of the sixth overtone band of nitric oxide, and its dipole moment function, using cavity-enhanced frequency modulation spectroscopy,” J. Chem. Phys. 124(8), 084311 (2006). [CrossRef] [PubMed]
5. N. J. van Leeuwen and A. C. Wilson, “Measurement of pressure-broadened, ultraweak transitions with noise-immune cavity-enhanced optical heterodyne molecular spectroscopy,” J. Opt. Soc. Am. B 21(10), 1713–1721 (2004). [CrossRef]
6. F. M. Schmidt, A. Foltynowicz, W. Ma, and O. Axner, “Fiber-laser-based noise-immune cavity-enhanced optical heterodyne molecular spectrometry for Doppler-broadened detection of C2H2 in the parts per trillion range,” J. Opt. Soc. Am. B 24(6), 1392–1405 (2007). [CrossRef]
8. C. L. Bell, G. Hancock, R. Peverall, G. A. D. Ritchie, J. H. van Helden, and N. J. van Leeuwen, “Characterization of an external cavity diode laser based ring cavity NICE-OHMS system,” Opt. Express 17(12), 9834–9839 (2009). [CrossRef] [PubMed]
9. C. Ishibashi and H. Sasada, “Near-infrared laser spectrometer with sub-Doppler resolution, high sensitivity, and wide tunability: A case study in the 1.65-μm region of CH3I spectrum,” J. Mol. Spectrosc. 200(1), 147–149 (2000). [CrossRef] [PubMed]
10. M. S. Taubman, T. L. Myers, B. D. Cannon, and R. M. Williams, “Stabilization, injection and control of quantum cascade lasers, and their application to chemical sensing in the infrared,” Spectrochim. Acta A Mol. Biomol. Spectrosc. 60(14), 3457–3468 (2004). [CrossRef] [PubMed]
11. A. Foltynowicz, W. Ma, and O. Axner, “Characterization of fiber-laser-based sub-Doppler NICE-OHMS for quantitative trace gas detection,” Opt. Express 16(19), 14689–14702 (2008). [CrossRef] [PubMed]
12. O. Axner, W. Ma, and A. Foltynowicz, “Sub-Doppler dispersion and noise-immune cavity-enhanced optical heterodyne molecular spectroscopy revised,” J. Opt. Soc. Am. B 25(7), 1166–1177 (2008). [CrossRef]
13. L. S. Ma, J. Ye, P. Dube, and J. L. Hall, “Ultrasensitive frequency-modulation spectroscopy enhanced by a high-finesse optical cavity: theory and application to overtone transitions of C2H2 and C2HD,” J. Opt. Soc. Am. B 16(12), 2255–2268 (1999). [CrossRef]
14. L. S. Rothman, I. E. Gordon, A. Barbe, D. C. Benner, P. E. Bernath, M. Birk, V. Boudon, L. R. Brown, A. Campargue, J. P. Champion, K. Chance, L. H. Coudert, V. Dana, V. M. Devi, S. Fally, J. M. Flaud, R. R. Gamache, A. Goldman, D. Jacquemart, I. Kleiner, N. Lacome, W. J. Lafferty, J. Y. Mandin, S. T. Massie, S. N. Mikhailenko, C. E. Miller, N. Moazzen-Ahmadi, O. V. Naumenko, A. V. Nikitin, J. Orphal, V. I. Perevalov, A. Perrin, A. Predoi-Cross, C. P. Rinsland, M. Rotger, M. Simeckova, M. A. H. Smith, K. Sung, S. A. Tashkun, J. Tennyson, R. A. Toth, A. C. Vandaele, and J. Vander Auwera, “The HITRAN 2008 molecular spectroscopic database,” J. Quant. Spectrosc. Radiat. Transf. 110(9-10), 533–572 (2009). [CrossRef]
15. W. Ma, A. Foltynowicz, and O. Axner, “Theoretical description of Doppler-broadened noise-immune cavity-enhanced optical heterodyne molecular spectroscopy under optically saturated conditions,” J. Opt. Soc. Am. B 25(7), 1144–1155 (2008). [CrossRef]
16. E. A. Whittaker, M. Gehrtz, and G. C. Bjorklund, “Residual amplitude modulation in laser electro-optic phase modulation,” J. Opt. Soc. Am. B 2(8), 1320–1326 (1985). [CrossRef]
17. R. W. P. Drever, J. L. Hall, F. V. Kowalski, J. Hough, G. M. Ford, A. J. Munley, and H. Ward, “Laser phase and frequency stabilization using an optical resonator,” Appl. Phys. B 31(2), 97–105 (1983). [CrossRef]
18. R. G. DeVoe and R. G. Brewer, “Laser frequency division and stabilization,” Phys. Rev. A: At. Mol. Opt. Phys. 30, 2827–2829 (1984).
19. A. Foltynowicz, W. Ma, F. M. Schmidt, and O. Axner, “Doppler-broadened noise-immune cavity-enhanced optical heterodyne molecular spectroscopy signals from optically saturated transitions under low pressure conditions,” J. Opt. Soc. Am. B 25(7), 1156–1165 (2008). [CrossRef]
20. R. El Hachtouki and J. Vander Auwera, “Absolute line intensities in acetylene: the 1.5-μm region,” J. Mol. Spectrosc. 216(2), 355–362 (2002). [CrossRef]
21. H. Tran, J. Y. Mandin, V. Dana, L. Regalia-Jarlot, X. Thomas, and P. Von der Heyden, “Line intensities in the 1.5-μm spectral region of acetylene,” J. Quant. Spectrosc. Radiat. Transf. 108(3), 342–362 (2007). [CrossRef]
22. D. Jacquemart, N. Lacome, J. Y. Mandin, V. Dana, H. Tran, F. K. Gueye, O. M. Lyulin, V. I. Perevalov, and L. Regalia-Jarlot, “The IR spectrum of (C2H2)-C-12: Line intensity measurements in the 1.4 μm region and update of the databases,” J. Quant. Spectrosc. Radiat. Transf. 110(9-10), 717–732 (2009). [CrossRef]
23. S. W. Arteaga, C. M. Bejger, J. L. Gerecke, J. L. Hardwick, Z. T. Martin, J. Mayo, E. A. McIlhattan, J. M. F. Moreau, M. J. Pilkenton, M. J. Polston, B. T. Robertson, and E. N. Wolf, “Line broadening and shift coefficients of acetylene at 1550 nm,” J. Mol. Spectrosc. 243(2), 253–266 (2007). [CrossRef]
24. L. Fissiaux, M. Dhyne, and M. Lepere, “Diode-laser spectroscopy: Pressure dependence of N-2-broadening coefficients of lines in the ν(4) + ν(5) band of C2H2,” J. Mol. Spectrosc. 254(1), 10–15 (2009). [CrossRef]
25. P. Werle, R. Mucke, and F. Slemr, “The limits of signal averaging in atmospheric trace-gas monitoring by Tunable Diode-Laser Absorption Spectroscopy (TDLAS),” Appl. Phys. B 57, 131–139 (1993). [CrossRef]
26. A. Foltynowicz, I. Silander, and O. Axner, are preparing a manuscript to be called 'Reduction of background signals from fiber-coupled electro-optic modulators in NICE-OHMS'.