October 15, 2014 / Vol. 39, No. 20 / OPTICS LETTERS
6005
Active FTIR-based stand-off spectroscopy using a femtosecond optical parametric oscillator Zhaowei Zhang,1,* Rhea J. Clewes,2 Christopher R. Howle,2 and Derryck T. Reid1 1
Scottish Universities Physics Alliance (SUPA), Institute of Photonics and Quantum Sciences, School of Engineering and Physical Sciences, Heriot–Watt University, Edinburgh EH14 4AS, UK 2
Defence Science and Technology Laboratory, Porton Down, Salisbury SP4 0JQ, UK *Corresponding author: [email protected]
Received August 25, 2014; revised September 17, 2014; accepted September 17, 2014; posted September 18, 2014 (Doc. ID 221689); published October 14, 2014 We presented the first demonstration of stand-off Fourier transform infrared (FTIR) spectroscopy using a broadband mid-infrared optical parametric oscillator, with spectral coverage over 2700–3200 cm−1 . For vapor-phase water and nitromethane (NM), stand-off spectra was recorded using a concrete target at from 1-m to 2-m range and showed good agreement with reference spectra, and in NM a normalized detection sensitivity of 15 ppm · m · Hz−1∕2 was obtained. Spectra from 50-μL droplets of liquid thiodiglycol were detected at a stand-off distance of 2 m from aluminum, concrete and painted metal surfaces. Our results imply that OPO-based active FTIR stand-off spectroscopy is a promising new technique for the detection of industrial pollutants and the identification of chemical agents, explosives or other hazardous materials. © 2014 Optical Society of America OCIS codes: (300.6300) Spectroscopy, Fourier transforms; (120.6200) Spectrometers and spectroscopic instrumentation; (120.0280) Remote sensing and sensors; (190.4970) Parametric oscillators and amplifiers. http://dx.doi.org/10.1364/OL.39.006005
Fourier transform infrared (FTIR) spectroscopy, widely used for the identification and quantitative analysis of molecular species, is normally implemented by use of an incandescent light source, e.g., a globar. Broadband mid-infrared (mid-IR) optical parametric oscillators (OPOs), which provide excellent spatial coherence and high spectral brightness compared to thermal sources, have shown themselves to be particularly suitable for FTIR spectroscopy, enabling excellent detection sensitivity, resolution and acquisition rates [1–5]. In the context of open-path spectroscopy, the exceptional spatial coherence of mid-IR light from a femtosecond OPO enables it to be collimated over long distances, with the capability for even low-power OPOs to out-perform the best thermal sources over several tens of meters. Indeed, for a diffraction-limited mid-IR OPO source with an average power of 1 mW and a spectral bandwidth of 1000 cm−1 , its spectral power density is 1 μW∕cm−1 , and at 3100 cm−1 , its spectral brightness of 0.1 W∕mm2 · cm−1 · sr is approximately 104 and 109 times higher than for 1200 and 300 K blackbody sources, respectively [6]. Stand-off detection is an important and promising approach for the remote identification of molecular species and is commonly employed in circumstances where direct access to the samples is difficult or hazardous [7]. Typical applications include: environmental pollution diagnosis; detection of gas leaks or toxic industrial contamination; identification of chemical warfare agents; analysis of contaminated soils and the remote identification of explosives or hazardous materials. Passive stand-off FTIR spectroscopy [8], relying on the reflected infrared radiation from the cold sky, allows detection over a long distance, but identification of molecular species can be complicated and sometimes impossible due to the influence of the background spectrum from the sky. Active stand-off FTIR spectroscopy using a high-temperature blackbody source [9,10] facilitates the identification of molecular species, but it is inefficient and only offers short-range remote sensing as a result of 0146-9592/14/206005-04$15.00/0
the poor spatial coherence and low spectral brightness of the source. Stand-off spectroscopy using wavelengthtunable mid-IR lasers (e.g., quantum cascade lasers or narrow-linewidth OPOs) has been shown to provide sensitive detection of solid or gas samples at medium or long distances [11–13], however combining broad spectral coverage with rapid tuning remains challenging. Recently, stand-off detection using spatially coherent, broadband mid-IR OPO or supercontinuum sources was demonstrated with grating-based monochromators and lock-in detection [14,15]. In the grating-based system detailed in [14], the acquisition time for one spectrum was 150 s, whereas rapid-scanning FTIR spectrometers of the kind we used here achieved a measurement rate of several Hz. Indeed, the advantages of FTIR spectrometers over grating spectrometers are well documented, and lead to their ability to provide higher signal-to-noise ratio (SNR) measurements in a shorter acquisition time [16], suggesting them as a promising modality for broadband OPO-based stand-off detection. Access to the mid-IR from a 1-μm pump laser is more efficient using an OPO than a super-continuum source. We have comparable spectral brightness in the 3–4-μm region to the super-continuum source detailed in [14], but with a 976-nm pump power of 7 W rather than 104 W. Furthermore, the superior coherence of an OPO source over a super-continuum generated by modulation-instability/ four-wave-mixing opens the possibility of using heterodyne techniques to improve the detection sensitivity available in stand-off systems. In this Letter, we present the first implementation, to our knowledge, of stand-off FTIR spectroscopy using a coherent mid-IR source. The use of such a spatially coherent OPO light source overcomes limits in the stand-off distance and detectivity encountered when using a thermal light source. We have demonstrated stand-off spectroscopy from noncompliant surfaces at distances of up to 2 m with a SNR that implies a greater stand-off range should be readily achievable. © 2014 Optical Society of America
6006
OPTICS LETTERS / Vol. 39, No. 20 / October 15, 2014
A schematic of our configuration is shown in Fig. 1(a). Details of the broadband mid-IR OPO appear in [17]. This signal-resonant OPO employed a 20 mm-long MgO-doped periodically poled lithium niobate (PPLN) crystal that was synchronously pumped by a fiber amplifier seeded by a mode-locked Yb:KYW laser. The average power, pulse duration and repetition rate of the idler output were 100 mW, 3 ps and 100 MHz, respectively. A typical spectrum, covering 2700–3200 cm−1 , is shown in Fig. 1(b). The right axis of Fig. 1(b) indicates its SNR, defined as the ratio of the measured spectral intensity and standard deviation of the noise. Mid-IR idler light with a collimated beam diameter of 8 mm was directed to a FTIR spectrometer consisting of a Michelson interferometer with a scanning mirror. A second beam from a 632.816-nm He–Ne laser was coupled into the interferometer for absolute delay calibration. After the interferometer, the beam was directed to the surface of interest. Light diffusely reflected from the surface was collected by a CaF2 lens with a diameter of 50 mm and a focal length of 100 mm onto a thermoelectrically cooled MCZT (HgCdZnTe) photovoltaic point detector with an active area of 1 mm2 . The moving mirror in the Michelson interferometer was scanned at approximately 5 Hz, allowing a single interferogram to be
acquired in 100 ms by a 16-bit digital acquisition card with a sampling rate of 5 M samples/s. A typical series of detected interferograms is shown in Fig. 1(c). We assessed the stand-off performance of the system by recording the amplitude of the measured interferogram at different distances for two different surfaces: a chemical agent resistant coating (CARC) painted metal plate [Fig. 2(a)] and a smooth concrete surface [Fig. 2(b)]. The stand-off distance is defined as the distance between the target surface and the full spectrometer system including light source, interferometer and the light collection/detection system. As shown in Fig. 2(c), the amplitude of the measured interferogram diminished with increasing stand-off distance for both surfaces, but at different rates. For the concrete, the amplitude at 2 m was 80% of that at 0.2 m, whereas for the CARC-coated surface the amplitude at 2 m was 5% of that at 0.2 m. These results are expected since the concrete surface used here was flatter and less scattering than the CARC-coated surface. In both cases, stand-off spectra at a distance of 2 m were obtained, only limited by the dimensions of the optical bench used in the experiment. An angular study was also conducted for both surfaces at a stand-off distance of 0.5 m [Fig. 2(d)]. The amplitude of the interferogram diminished with increasing angle, and in both cases an angular tolerance of 10° was obtained. The CARC surface exhibited a larger detection angle compared with concrete, because its surface was more irregular on the scale of the beam spot size, causing the light to be scattered into a larger solid angle. We first present an example of a stand-off measurement of atmospheric absorption by recording the midIR spectrum diffusely reflected by a concrete surface at a distance of 2 m. A Fourier transform was performed on a single interferogram, and after normalization the absorption spectrum shown in Fig. 3 was obtained. The total path travelled by the mid-IR light from the OPO to the detector was 5 m. In the spectral region from 2700–3200 cm−1 the main atmospheric absorption is due to water vapor, and a HITRAN [18] simulation (Fig. 3) of
Fig. 1. (a) Layout of the stand-off detection system, showing the definition of the stand-off distance, L; (b) measured OPO idler spectral intensity (left axis) and its SNR (right axis); (c) a typical interferogram sequence.
Fig. 2. Images of (a) the CARC-coated metal plate, and (b) the concrete plate. The maximum interferogram (IFM) signal as a function of (c) stand-off distance and (d) scattering angle for CARC and concrete plates at a stand-off distance of 0.5 m.
October 15, 2014 / Vol. 39, No. 20 / OPTICS LETTERS
6007
Fig. 3. Atmospheric absorption spectrum measured at a distance of 2 m from a concrete substrate (blue) and a simulated water vapor absorption spectrum, assuming 3% water vapor at one atmosphere and a 5-m path length (red).
3% concentration water vapor at room temperature, is in good agreement with the experimental data. With a longer path length, other ambient gases such as methane could also be measured, suggesting applications in environmental sensing of greenhouse gases and stand-off detection of industrial emissions [12]. In another vapor-phase experiment we measured the absorption spectrum of nitromethane (NM), which is widely used in the manufacture of pharmaceuticals, pesticides and explosives. We introduced 2 μL of liquid NM into a sealed cell of air at atmospheric pressure, with a diameter of 40 mm and a length of 100 mm. After 30 min, the NM was vaporized at the room temperature of 25°C. The gas cell was placed along the beam path and the transmitted light was diffusely reflected by a concrete sample at a distance of 1 m from the system. Figure 4 shows the measured and simulated transmission spectra [19]. For the single measurement shown in Fig. 4, the acquisition time was 100 ms and the SNR was 20, indicating a potential detection sensitivity of 100 nL. This corresponds to a normalized detection sensitivity of
Fig. 4. Measured (solid line) and simulated (dashed line) transmission spectrum of vaporized NM obtained from a concrete substrate at a stand-off distance of 1 m. The simulated data take no account of any potential substrate contribution.
Fig. 5. Solid lines, measured transmission spectrum of (a) TDG on a concrete substrate, (b) TDG on an Al plate, (c) TDG on a CARC plate, and (d) CARC plate, all at a stand-off distance of 2 m. Dashed lines, reference transmission spectrum of TDG.
15 ppm · m · Hz−1∕2 , similar to detection sensitivities obtained using a quantum cascade laser [20]. We investigated the detection of liquid thiodiglycol (TDG), a chemical warfare agent (CWA) simulant, on different surfaces. A 50-μL drop of TDG was applied on a concrete surface, a black anodized aluminum plate and a CARC plate. The mid-IR light was diffusely reflected by the TDG at a stand-off distance of 2 m. The spectra measured are shown in Figs. 5(a)–5(c), respectively, together with a reference TDG spectrum [19] (dashed lines). The main absorption peaks of TDG were clearly identified on the concrete surface [Fig. 5(a)] but with small deviations from the reference spectrum. Independent measurements made in reflectance mode with an analytic FTIR system (Tensor 37 spectrometer with a Specac Golden Gate Attenuated Total Reflectance accessory) differed by a similar amount from the reference TDG spectrum, implying that the exact spectral shape is sensitive to the specific sample preparation and illumination conditions. On the aluminum plate [Fig. 5(b)], the measured TDG spectrum showed some greater differences from the reference. In an independent measurement, we confirmed that there was no absorption by the aluminum plate within the measured spectral region. As observed by Neely et al. [15], factors such as beam collimation, illumination/collection angle, power and polarization may play a role in distorting the measured spectra in a way that prevents exact registration with reference data. On the CARC plate [Fig. 5(c)] the measured spectrum was significantly different from the reference spectrum, explained by interference from the absorption features of the CARC paint. To illustrate this, a stand-off measurement of the CARC paint over a 2-m distance was implemented and is shown in Fig. 5(d). The absorption band from 2820–3080 cm−1 by the CARC paint can be
6008
OPTICS LETTERS / Vol. 39, No. 20 / October 15, 2014
attributed to CH3 and CH2 asymmetric stretches [21], with features at 2860, 2930 and 2980 cm−1 . Unique discrimination of TDG from CARC would be possible in the long-wave mid-IR molecular fingerprint region, e.g., at ∼1050 cm−1 , where TDG has a strong absorption band. In summary, we have demonstrated the first stand-off FTIR spectroscopy based on a mid-IR femtosecond OPO, across a measurement range of 2700–3200 cm−1 . By using a broadband mid-IR supercontinuum source [14] or an MgO:PPLN crystal comprising an array of gratings with suitable poling periods, the full atmospheric window from 2000–3300 cm−1 could be addressed. Spectra of ambient water vapor, NM vapor, and liquid TDG on different surfaces have been obtained. Our results imply that to unambiguously identify materials by use of broadband stand-off spectroscopy, prior knowledge of the chemical and optical properties of the substrate background and target chemicals is essential. The demonstrated standoff distance of 2 m is currently limited by the dimensions of the optical bench, not by SNR considerations. Based on our data, we estimate a stand-off distance of 5–10 m is feasible with no changes to our experimental arrangement. Compared with a blackbody source, a spatially coherent broadband mid-IR light source such as a femtosecond OPO is inherently more suitable for stand-off spectroscopy because of its excellent spatial coherence, enabling its output light to propagate over a long distance without substantial diffraction, and high spectral brightness, which means that the signal is still detectable at a stand-off distance of a few meters even when it has experienced significant loss from highly-scattering real-world surfaces. In this initial demonstration, no averaging was applied, and only detection at a single spatial location was attempted. The acquisition rate of 5 Hz could be improved significantly by using a faster scanning mirror for the Michelson interferometer. Much higher acquisition rates can also be achieved by use of a dual-comb configuration, for which two asynchronous OPO frequency combs are required [5]. The scheme could be extended to obtain feature-rich hyper-spectral 1-D or 2-D images by integrating a galvanometer scanner into the system [10]. A further extension could be to implement the system at wavelengths within the 500–1500 cm−1 spectral region, which forms the molecular fingerprint band that is commonly used for the identification and analysis of molecular species. The authors gratefully acknowledge financial support from the Centre for Defence Enterprise, part of the UK Ministry of Defence. References and Note 1. K. A. Tillman, R. R. J. Maier, D. T. Reid, and E. D. McNaghten, Appl. Phys. Lett. 85, 3366 (2004). 2. A. Schliesser, M. Brehm, F. Keilmann, and D. van der Weide, Opt. Express 13, 9029 (2005).
3. F. Adler, P. Masłowski, A. Foltynowicz, K. C. Cossel, T. C. Briles, I. Hartl, and J. Ye, Opt. Express 18, 21861 (2010). 4. A. Schliesser, N. Picqué, and T. W. Hänsch, Nat. Photonics 6, 440 (2012). 5. Z. Zhang, T. Gardiner, and D. T. Reid, Opt. Lett. 38, 3148 (2013). 6. The spectral brightness of a diffraction–limited broadband mid-IR OPO source is calculated as P∕λ2 · Δλ, where P is the average power, λ wavelength, and Δλ the bandwidth. The spectral brightness of a high temperature blackbody at different temperatures is calculated using Planck’s law. 7. F. J. Fortes and J. J. Laserna, Spectrochimica Acta Part B 65, 975 (2010). 8. J.-M. Theriault, E. Puckrin, J. Hancock, P. Lecavalier, C. Jackson Lepage, and J. O. Jensen, Appl. Opt. 43, 5870 (2004). 9. R. Harig, R. Braun, C. Dyer, C. Howle, and B. Truscott, Opt. Express 16, 5708 (2008). 10. F. Rosi, C. Miliani, R. Braun, R. Harig, D. Sali, B. G. Brunetti, and A. Sgamellotti, Angew. Chem. 125, 5366 (2013). 11. F. Fuchs, S. Hugger, M. Kinzer, R. Aidam, W. Bronner, R. Losch, and Q. Yang, Opt. Eng. 49, 111127 (2010). 12. D. J. Stothard, M. H. Dunn, and C. F. Rae, Opt. Express 12, 947 (2004). 13. J. D. Suter, B. Bernacki, and M. C. Phillips, Appl. Phys. B 108, 965 (2012). 14. M. Kumar, M. N. Islam, F. L. Terry, M. J. Freeman, A. Chan, M. Neelakandan, and T. Manzur, Appl. Opt. 51, 2794 (2012). 15. T. W. Neely, L. Nugent-Glandorf, and S. Diddams, Lasers, Sources, and Related Photonic Devices, OSA Technical Digest (CD) (Optical Society of America, 2012), paper LT2B.3. 16. P. R. Griffiths and J. A. De Haseth, Fourier Transform Infrared Spectrometry (Wiley, 2007). 17. Z. Zhang, J. Sun, T. Gardiner, and D. T. Reid, Opt. Express 19, 17127 (2011). 18. L. S. Rothman, I. E. Gordon, Y. Babikov, A. Barbe, D. Chris Benner, P. F. Bernath, M. Birk, L. Bizzocchi, V. Boudon, L. R. Brown, A. Campargue, K. Chance, E. A. Cohen, L. H. Coudert, V. M. Devi, B. J. Drouin, A. Fayt, J.-M. Flaud, R. R. Gamache, J. J. Harrison, J.-M. Hartmann, C. Hill, J. T. Hodges, D. Jacquemart, A. Jolly, J. Lamouroux, R. J. Le Roy, G. Li, D. A. Long, O. M. Lyulin, C. J. Mackie, S. T. Massie, S. Mikhailenko, H. S. P. Müller, O. V. Naumenko, A. V. Nikitin, J. Orphal, V. Perevalov, A. Perrin, E. R. Polovtseva, C. Richard, M. A. H. Smith, E. Starikova, K. Sung, S. Tashkun, J. Tennyson, G. C. Toon, Vl., G. Tyuterev, and G. Wagner, J. Quant. Spectrosc. Radiat. Transfer 130, 4 (2013). 19. NIST Mass Spec Data Center, S. E. Stein, director, “Infrared Spectra” in NIST Chemistry WebBook, NIST Standard Reference Database Number 69, P. J. Linstrom and W. G. Mallard, eds., National Institute of Standards and Technology, Gaithersburg Md., 20899, http://webbook.nist.gov/, (retrieved March 20, 2013). 20. N. A. Macleod, R. Rose, and D. Weidmann, Opt. Lett. 38, 3708 (2013). 21. J. A. Escarsega, D. M. Crawford, and P. J. Kaste, Physical Properties of Polymers Handbook (Springer, 2007).
6005
Active FTIR-based stand-off spectroscopy using a femtosecond optical parametric oscillator Zhaowei Zhang,1,* Rhea J. Clewes,2 Christopher R. Howle,2 and Derryck T. Reid1 1
Scottish Universities Physics Alliance (SUPA), Institute of Photonics and Quantum Sciences, School of Engineering and Physical Sciences, Heriot–Watt University, Edinburgh EH14 4AS, UK 2
Defence Science and Technology Laboratory, Porton Down, Salisbury SP4 0JQ, UK *Corresponding author: [email protected]
Received August 25, 2014; revised September 17, 2014; accepted September 17, 2014; posted September 18, 2014 (Doc. ID 221689); published October 14, 2014 We presented the first demonstration of stand-off Fourier transform infrared (FTIR) spectroscopy using a broadband mid-infrared optical parametric oscillator, with spectral coverage over 2700–3200 cm−1 . For vapor-phase water and nitromethane (NM), stand-off spectra was recorded using a concrete target at from 1-m to 2-m range and showed good agreement with reference spectra, and in NM a normalized detection sensitivity of 15 ppm · m · Hz−1∕2 was obtained. Spectra from 50-μL droplets of liquid thiodiglycol were detected at a stand-off distance of 2 m from aluminum, concrete and painted metal surfaces. Our results imply that OPO-based active FTIR stand-off spectroscopy is a promising new technique for the detection of industrial pollutants and the identification of chemical agents, explosives or other hazardous materials. © 2014 Optical Society of America OCIS codes: (300.6300) Spectroscopy, Fourier transforms; (120.6200) Spectrometers and spectroscopic instrumentation; (120.0280) Remote sensing and sensors; (190.4970) Parametric oscillators and amplifiers. http://dx.doi.org/10.1364/OL.39.006005
Fourier transform infrared (FTIR) spectroscopy, widely used for the identification and quantitative analysis of molecular species, is normally implemented by use of an incandescent light source, e.g., a globar. Broadband mid-infrared (mid-IR) optical parametric oscillators (OPOs), which provide excellent spatial coherence and high spectral brightness compared to thermal sources, have shown themselves to be particularly suitable for FTIR spectroscopy, enabling excellent detection sensitivity, resolution and acquisition rates [1–5]. In the context of open-path spectroscopy, the exceptional spatial coherence of mid-IR light from a femtosecond OPO enables it to be collimated over long distances, with the capability for even low-power OPOs to out-perform the best thermal sources over several tens of meters. Indeed, for a diffraction-limited mid-IR OPO source with an average power of 1 mW and a spectral bandwidth of 1000 cm−1 , its spectral power density is 1 μW∕cm−1 , and at 3100 cm−1 , its spectral brightness of 0.1 W∕mm2 · cm−1 · sr is approximately 104 and 109 times higher than for 1200 and 300 K blackbody sources, respectively [6]. Stand-off detection is an important and promising approach for the remote identification of molecular species and is commonly employed in circumstances where direct access to the samples is difficult or hazardous [7]. Typical applications include: environmental pollution diagnosis; detection of gas leaks or toxic industrial contamination; identification of chemical warfare agents; analysis of contaminated soils and the remote identification of explosives or hazardous materials. Passive stand-off FTIR spectroscopy [8], relying on the reflected infrared radiation from the cold sky, allows detection over a long distance, but identification of molecular species can be complicated and sometimes impossible due to the influence of the background spectrum from the sky. Active stand-off FTIR spectroscopy using a high-temperature blackbody source [9,10] facilitates the identification of molecular species, but it is inefficient and only offers short-range remote sensing as a result of 0146-9592/14/206005-04$15.00/0
the poor spatial coherence and low spectral brightness of the source. Stand-off spectroscopy using wavelengthtunable mid-IR lasers (e.g., quantum cascade lasers or narrow-linewidth OPOs) has been shown to provide sensitive detection of solid or gas samples at medium or long distances [11–13], however combining broad spectral coverage with rapid tuning remains challenging. Recently, stand-off detection using spatially coherent, broadband mid-IR OPO or supercontinuum sources was demonstrated with grating-based monochromators and lock-in detection [14,15]. In the grating-based system detailed in [14], the acquisition time for one spectrum was 150 s, whereas rapid-scanning FTIR spectrometers of the kind we used here achieved a measurement rate of several Hz. Indeed, the advantages of FTIR spectrometers over grating spectrometers are well documented, and lead to their ability to provide higher signal-to-noise ratio (SNR) measurements in a shorter acquisition time [16], suggesting them as a promising modality for broadband OPO-based stand-off detection. Access to the mid-IR from a 1-μm pump laser is more efficient using an OPO than a super-continuum source. We have comparable spectral brightness in the 3–4-μm region to the super-continuum source detailed in [14], but with a 976-nm pump power of 7 W rather than 104 W. Furthermore, the superior coherence of an OPO source over a super-continuum generated by modulation-instability/ four-wave-mixing opens the possibility of using heterodyne techniques to improve the detection sensitivity available in stand-off systems. In this Letter, we present the first implementation, to our knowledge, of stand-off FTIR spectroscopy using a coherent mid-IR source. The use of such a spatially coherent OPO light source overcomes limits in the stand-off distance and detectivity encountered when using a thermal light source. We have demonstrated stand-off spectroscopy from noncompliant surfaces at distances of up to 2 m with a SNR that implies a greater stand-off range should be readily achievable. © 2014 Optical Society of America
6006
OPTICS LETTERS / Vol. 39, No. 20 / October 15, 2014
A schematic of our configuration is shown in Fig. 1(a). Details of the broadband mid-IR OPO appear in [17]. This signal-resonant OPO employed a 20 mm-long MgO-doped periodically poled lithium niobate (PPLN) crystal that was synchronously pumped by a fiber amplifier seeded by a mode-locked Yb:KYW laser. The average power, pulse duration and repetition rate of the idler output were 100 mW, 3 ps and 100 MHz, respectively. A typical spectrum, covering 2700–3200 cm−1 , is shown in Fig. 1(b). The right axis of Fig. 1(b) indicates its SNR, defined as the ratio of the measured spectral intensity and standard deviation of the noise. Mid-IR idler light with a collimated beam diameter of 8 mm was directed to a FTIR spectrometer consisting of a Michelson interferometer with a scanning mirror. A second beam from a 632.816-nm He–Ne laser was coupled into the interferometer for absolute delay calibration. After the interferometer, the beam was directed to the surface of interest. Light diffusely reflected from the surface was collected by a CaF2 lens with a diameter of 50 mm and a focal length of 100 mm onto a thermoelectrically cooled MCZT (HgCdZnTe) photovoltaic point detector with an active area of 1 mm2 . The moving mirror in the Michelson interferometer was scanned at approximately 5 Hz, allowing a single interferogram to be
acquired in 100 ms by a 16-bit digital acquisition card with a sampling rate of 5 M samples/s. A typical series of detected interferograms is shown in Fig. 1(c). We assessed the stand-off performance of the system by recording the amplitude of the measured interferogram at different distances for two different surfaces: a chemical agent resistant coating (CARC) painted metal plate [Fig. 2(a)] and a smooth concrete surface [Fig. 2(b)]. The stand-off distance is defined as the distance between the target surface and the full spectrometer system including light source, interferometer and the light collection/detection system. As shown in Fig. 2(c), the amplitude of the measured interferogram diminished with increasing stand-off distance for both surfaces, but at different rates. For the concrete, the amplitude at 2 m was 80% of that at 0.2 m, whereas for the CARC-coated surface the amplitude at 2 m was 5% of that at 0.2 m. These results are expected since the concrete surface used here was flatter and less scattering than the CARC-coated surface. In both cases, stand-off spectra at a distance of 2 m were obtained, only limited by the dimensions of the optical bench used in the experiment. An angular study was also conducted for both surfaces at a stand-off distance of 0.5 m [Fig. 2(d)]. The amplitude of the interferogram diminished with increasing angle, and in both cases an angular tolerance of 10° was obtained. The CARC surface exhibited a larger detection angle compared with concrete, because its surface was more irregular on the scale of the beam spot size, causing the light to be scattered into a larger solid angle. We first present an example of a stand-off measurement of atmospheric absorption by recording the midIR spectrum diffusely reflected by a concrete surface at a distance of 2 m. A Fourier transform was performed on a single interferogram, and after normalization the absorption spectrum shown in Fig. 3 was obtained. The total path travelled by the mid-IR light from the OPO to the detector was 5 m. In the spectral region from 2700–3200 cm−1 the main atmospheric absorption is due to water vapor, and a HITRAN [18] simulation (Fig. 3) of
Fig. 1. (a) Layout of the stand-off detection system, showing the definition of the stand-off distance, L; (b) measured OPO idler spectral intensity (left axis) and its SNR (right axis); (c) a typical interferogram sequence.
Fig. 2. Images of (a) the CARC-coated metal plate, and (b) the concrete plate. The maximum interferogram (IFM) signal as a function of (c) stand-off distance and (d) scattering angle for CARC and concrete plates at a stand-off distance of 0.5 m.
October 15, 2014 / Vol. 39, No. 20 / OPTICS LETTERS
6007
Fig. 3. Atmospheric absorption spectrum measured at a distance of 2 m from a concrete substrate (blue) and a simulated water vapor absorption spectrum, assuming 3% water vapor at one atmosphere and a 5-m path length (red).
3% concentration water vapor at room temperature, is in good agreement with the experimental data. With a longer path length, other ambient gases such as methane could also be measured, suggesting applications in environmental sensing of greenhouse gases and stand-off detection of industrial emissions [12]. In another vapor-phase experiment we measured the absorption spectrum of nitromethane (NM), which is widely used in the manufacture of pharmaceuticals, pesticides and explosives. We introduced 2 μL of liquid NM into a sealed cell of air at atmospheric pressure, with a diameter of 40 mm and a length of 100 mm. After 30 min, the NM was vaporized at the room temperature of 25°C. The gas cell was placed along the beam path and the transmitted light was diffusely reflected by a concrete sample at a distance of 1 m from the system. Figure 4 shows the measured and simulated transmission spectra [19]. For the single measurement shown in Fig. 4, the acquisition time was 100 ms and the SNR was 20, indicating a potential detection sensitivity of 100 nL. This corresponds to a normalized detection sensitivity of
Fig. 4. Measured (solid line) and simulated (dashed line) transmission spectrum of vaporized NM obtained from a concrete substrate at a stand-off distance of 1 m. The simulated data take no account of any potential substrate contribution.
Fig. 5. Solid lines, measured transmission spectrum of (a) TDG on a concrete substrate, (b) TDG on an Al plate, (c) TDG on a CARC plate, and (d) CARC plate, all at a stand-off distance of 2 m. Dashed lines, reference transmission spectrum of TDG.
15 ppm · m · Hz−1∕2 , similar to detection sensitivities obtained using a quantum cascade laser [20]. We investigated the detection of liquid thiodiglycol (TDG), a chemical warfare agent (CWA) simulant, on different surfaces. A 50-μL drop of TDG was applied on a concrete surface, a black anodized aluminum plate and a CARC plate. The mid-IR light was diffusely reflected by the TDG at a stand-off distance of 2 m. The spectra measured are shown in Figs. 5(a)–5(c), respectively, together with a reference TDG spectrum [19] (dashed lines). The main absorption peaks of TDG were clearly identified on the concrete surface [Fig. 5(a)] but with small deviations from the reference spectrum. Independent measurements made in reflectance mode with an analytic FTIR system (Tensor 37 spectrometer with a Specac Golden Gate Attenuated Total Reflectance accessory) differed by a similar amount from the reference TDG spectrum, implying that the exact spectral shape is sensitive to the specific sample preparation and illumination conditions. On the aluminum plate [Fig. 5(b)], the measured TDG spectrum showed some greater differences from the reference. In an independent measurement, we confirmed that there was no absorption by the aluminum plate within the measured spectral region. As observed by Neely et al. [15], factors such as beam collimation, illumination/collection angle, power and polarization may play a role in distorting the measured spectra in a way that prevents exact registration with reference data. On the CARC plate [Fig. 5(c)] the measured spectrum was significantly different from the reference spectrum, explained by interference from the absorption features of the CARC paint. To illustrate this, a stand-off measurement of the CARC paint over a 2-m distance was implemented and is shown in Fig. 5(d). The absorption band from 2820–3080 cm−1 by the CARC paint can be
6008
OPTICS LETTERS / Vol. 39, No. 20 / October 15, 2014
attributed to CH3 and CH2 asymmetric stretches [21], with features at 2860, 2930 and 2980 cm−1 . Unique discrimination of TDG from CARC would be possible in the long-wave mid-IR molecular fingerprint region, e.g., at ∼1050 cm−1 , where TDG has a strong absorption band. In summary, we have demonstrated the first stand-off FTIR spectroscopy based on a mid-IR femtosecond OPO, across a measurement range of 2700–3200 cm−1 . By using a broadband mid-IR supercontinuum source [14] or an MgO:PPLN crystal comprising an array of gratings with suitable poling periods, the full atmospheric window from 2000–3300 cm−1 could be addressed. Spectra of ambient water vapor, NM vapor, and liquid TDG on different surfaces have been obtained. Our results imply that to unambiguously identify materials by use of broadband stand-off spectroscopy, prior knowledge of the chemical and optical properties of the substrate background and target chemicals is essential. The demonstrated standoff distance of 2 m is currently limited by the dimensions of the optical bench, not by SNR considerations. Based on our data, we estimate a stand-off distance of 5–10 m is feasible with no changes to our experimental arrangement. Compared with a blackbody source, a spatially coherent broadband mid-IR light source such as a femtosecond OPO is inherently more suitable for stand-off spectroscopy because of its excellent spatial coherence, enabling its output light to propagate over a long distance without substantial diffraction, and high spectral brightness, which means that the signal is still detectable at a stand-off distance of a few meters even when it has experienced significant loss from highly-scattering real-world surfaces. In this initial demonstration, no averaging was applied, and only detection at a single spatial location was attempted. The acquisition rate of 5 Hz could be improved significantly by using a faster scanning mirror for the Michelson interferometer. Much higher acquisition rates can also be achieved by use of a dual-comb configuration, for which two asynchronous OPO frequency combs are required [5]. The scheme could be extended to obtain feature-rich hyper-spectral 1-D or 2-D images by integrating a galvanometer scanner into the system [10]. A further extension could be to implement the system at wavelengths within the 500–1500 cm−1 spectral region, which forms the molecular fingerprint band that is commonly used for the identification and analysis of molecular species. The authors gratefully acknowledge financial support from the Centre for Defence Enterprise, part of the UK Ministry of Defence. References and Note 1. K. A. Tillman, R. R. J. Maier, D. T. Reid, and E. D. McNaghten, Appl. Phys. Lett. 85, 3366 (2004). 2. A. Schliesser, M. Brehm, F. Keilmann, and D. van der Weide, Opt. Express 13, 9029 (2005).
3. F. Adler, P. Masłowski, A. Foltynowicz, K. C. Cossel, T. C. Briles, I. Hartl, and J. Ye, Opt. Express 18, 21861 (2010). 4. A. Schliesser, N. Picqué, and T. W. Hänsch, Nat. Photonics 6, 440 (2012). 5. Z. Zhang, T. Gardiner, and D. T. Reid, Opt. Lett. 38, 3148 (2013). 6. The spectral brightness of a diffraction–limited broadband mid-IR OPO source is calculated as P∕λ2 · Δλ, where P is the average power, λ wavelength, and Δλ the bandwidth. The spectral brightness of a high temperature blackbody at different temperatures is calculated using Planck’s law. 7. F. J. Fortes and J. J. Laserna, Spectrochimica Acta Part B 65, 975 (2010). 8. J.-M. Theriault, E. Puckrin, J. Hancock, P. Lecavalier, C. Jackson Lepage, and J. O. Jensen, Appl. Opt. 43, 5870 (2004). 9. R. Harig, R. Braun, C. Dyer, C. Howle, and B. Truscott, Opt. Express 16, 5708 (2008). 10. F. Rosi, C. Miliani, R. Braun, R. Harig, D. Sali, B. G. Brunetti, and A. Sgamellotti, Angew. Chem. 125, 5366 (2013). 11. F. Fuchs, S. Hugger, M. Kinzer, R. Aidam, W. Bronner, R. Losch, and Q. Yang, Opt. Eng. 49, 111127 (2010). 12. D. J. Stothard, M. H. Dunn, and C. F. Rae, Opt. Express 12, 947 (2004). 13. J. D. Suter, B. Bernacki, and M. C. Phillips, Appl. Phys. B 108, 965 (2012). 14. M. Kumar, M. N. Islam, F. L. Terry, M. J. Freeman, A. Chan, M. Neelakandan, and T. Manzur, Appl. Opt. 51, 2794 (2012). 15. T. W. Neely, L. Nugent-Glandorf, and S. Diddams, Lasers, Sources, and Related Photonic Devices, OSA Technical Digest (CD) (Optical Society of America, 2012), paper LT2B.3. 16. P. R. Griffiths and J. A. De Haseth, Fourier Transform Infrared Spectrometry (Wiley, 2007). 17. Z. Zhang, J. Sun, T. Gardiner, and D. T. Reid, Opt. Express 19, 17127 (2011). 18. L. S. Rothman, I. E. Gordon, Y. Babikov, A. Barbe, D. Chris Benner, P. F. Bernath, M. Birk, L. Bizzocchi, V. Boudon, L. R. Brown, A. Campargue, K. Chance, E. A. Cohen, L. H. Coudert, V. M. Devi, B. J. Drouin, A. Fayt, J.-M. Flaud, R. R. Gamache, J. J. Harrison, J.-M. Hartmann, C. Hill, J. T. Hodges, D. Jacquemart, A. Jolly, J. Lamouroux, R. J. Le Roy, G. Li, D. A. Long, O. M. Lyulin, C. J. Mackie, S. T. Massie, S. Mikhailenko, H. S. P. Müller, O. V. Naumenko, A. V. Nikitin, J. Orphal, V. Perevalov, A. Perrin, E. R. Polovtseva, C. Richard, M. A. H. Smith, E. Starikova, K. Sung, S. Tashkun, J. Tennyson, G. C. Toon, Vl., G. Tyuterev, and G. Wagner, J. Quant. Spectrosc. Radiat. Transfer 130, 4 (2013). 19. NIST Mass Spec Data Center, S. E. Stein, director, “Infrared Spectra” in NIST Chemistry WebBook, NIST Standard Reference Database Number 69, P. J. Linstrom and W. G. Mallard, eds., National Institute of Standards and Technology, Gaithersburg Md., 20899, http://webbook.nist.gov/, (retrieved March 20, 2013). 20. N. A. Macleod, R. Rose, and D. Weidmann, Opt. Lett. 38, 3708 (2013). 21. J. A. Escarsega, D. M. Crawford, and P. J. Kaste, Physical Properties of Polymers Handbook (Springer, 2007).
