Journal of Quantitative Spectroscopy & Radiative Transfer 222–223 (2019) 122–129 Contents lists available at ScienceDirect Journal of Quantitative Spectroscopy & Radiative Transfer journal homepage: www.elsevier.com/locate/jqsrt Measurement of the mid-infrared absorption spectra of ethylene (C 2 H 4 ) and other molecules at high temperatures and pressures ∗ C.L. Strand , Y. Ding, S.E. Johnson, R.K. Hanson High Temperature Gasdynamics Laboratory, Thermosciences Division, Stanford University, 452 Escondido Mall, Bldg. 520, Stanford, CA 94305, USA a r t i c l e i n f o a b s t r a c t Article history: A methodology for the measurement of mid-infrared absorption cross sections of gaseous molecules at Received 23 July 2018 high temperatures and pressures is presented. A rapid-tuning, broad-scan external cavity quantum cas- − Revised 18 September 2018 cade laser with a tuning rate in excess of 30 0 0 0 cm −1s 1 has been employed in the measurement of full Accepted 18 October 2018 vibrational bands ( > 100 cm −1) in shock-heated test gases. The approach is demonstrated with measure- Available online 19 October 2018 ments of the absorption cross section of ethylene (C 2 H 4 ) in the 8.5 μm to 11.7 μm region for tempera- tures from 800 K to 1600 K and pressures from 1 to 5 atm. Decreasing peak strength with temperature is observed as well as pressure-insensitivity. The measurements are compared with existing experimen- tal, empirical, and ab initio databases. Additionally, illustrative absorption cross section measurements of propene (C 3 H 6 ), 1-butene (1-C 4 H 8 ), 2-butene (2-C 4 H 8 ), 1,3-butadiene (1,3-C 4 H 6 ), and methanol (CH 3 OH) are presented near 10 0 0 K and 2.3 atm. ©2018 Elsevier Ltd. All rights reserved. 1. Introduction titative Infrared Database, focus on molecules at or near room temperature and atmospheric pressures [8–11] . Additionally, there There exists a marked lack of experimental absorption spec- are a few databases, including HITEMP and ExoMol, that provide tra for high-temperature and high-pressure gaseous molecules. high-temperature spectral data but only for a limited subset of Gases in these high-enthalpy thermodynamic states are present molecules [1,3] . in a wide range of natural and man-made environments, such as The limited availability of high-temperature and high-pressure cool stars, exoplanets, plasmas, explosions, flames, volcanoes, forest absorption and emission spectra stem primarily from the consid- fires, combustion systems, hypersonic flows, industrial processes, erable challenges these conditions pose for ab initio , empirical, and exhaust stacks. Correspondingly, these gases are the subject of and experimental methods. High temperatures require the consid- scientific study and engineering employ for many fields, including eration of hot lines that emerge due to the population of higher astrophysics, environmental science, plasma physics, combustion energy levels, including high-J rovibrational transitions and vibra- science, aerospace engineering, chemical engineering, and energy tional hot bands. As a consequence, the number of lines that must systems engineering. In the endeavor to study and monitor high- be considered for accurate spectral modeling at high temperatures temperature gases, infrared absorption and emission spectroscopy often increases by several orders of magnitude [1,2,12] . This issue is offer a powerful toolset for understanding the characteristics of further compounded by increasing molecular size and correspond- these environments [1–7] . This work presents a methodology us- ingly the number of vibrational modes, leading to the number of ing shock tube facilities in conjunction with rapid-tuning, broad- infrared lines for a single gas species to count in the billions. scan, narrow-linewidth lasers to measure the high-temperature Due to the weakness of hot lines at low temperatures and prac- and high-pressure absorption spectra of molecules in the mid- tical database considerations, hot lines are not typically observed, infrared. calculated, nor included in experimental, empirical, and ab initio Several databases exist to collate the infrared spectra and spec- room-temperature databases. This makes it impossible to accu- troscopic parameters of gaseous molecules. The most comprehen- rately extrapolate room-temperature spectroscopic data to higher sive and prominent line parameter and cross section databases, temperatures [2] . such as HITRAN, GEISA, PNNL Northwest Infrared, and NIST Quan- There are presently several initiatives to calculate, measure, and collate absorption spectra at high temperatures, including the two predominant databases HITEMP and ExoMol [1,3,13] . Unfortunately, ∗ Corresponding author. the experimental validation is often quite limited, especially for E-mail address: [email protected] (C.L. Strand). larger molecules. This lack of experimental validation is primarily https://doi.org/10.1016/j.jqsrt.2018.10.030 0022-4073/© 2018 Elsevier Ltd. All rights reserved. C.L. Strand et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 222–223 (2019) 122–129 123 the result of the same challenges that make any experimentation lations, assuming frozen chemistry. The necessary inputs are pro- at high-temperature, high-pressure conditions difficult; namely the vided by the known initial driver gas conditions and the incident challenge of generating a well-known, homogeneous, chemically- shock speeds measured using five piezoelectric pressure transduc- stable, high-enthalpy gas state that remains steady for a sufficient ers distributed over the final 1.5 m of the driven section. Uncer- duration to perform the desired experiment. All laboratory ap- tainty in the experimental pressure and temperature are each less proaches achieve only a subset of these desired characteristics and than ± 1% [14–16] . The test plane is 2 cm from the end wall of must compromise on some to faithfully maintain the others. More- the shock tube with instrument access provided by eight ports. For over, the requirement of a stable chemical composition generally these experiments, one pair of ports was equipped with ZnSe win- imposes an unavoidable upper-bound on the duration of the ex- dows with a 30 wedge and a broadband anti-reflection (AR) coat- periment. At elevated temperatures, accelerated reaction rates limit ing providing less than 1% reflection per surface over the wave- the experimental test time available to any experimental approach length range of 7 to 12 μm. These window characteristics mini- (e.g., the thermal decomposition of large molecules). mize etalon fringe noise when tuning the wavelength of a laser. Standard approaches for achieving high-enthalpy conditions in- Another pair of ports was equipped with flat AR-coated ZnSe win- clude radiative/conductive heating, rapid-compression, arc-heating, dows (R avg < 5% over 8 to 12 μm) for use with fixed wavelength and shock-heating. With the exception of shock-heating, these ap- lasers where etalon fringe noise is not a concern. Additionally, one proaches generally provide extended high-temperature test du- port was instrumented with a transducer to record the pressure in rations (from 100 ms up to several minutes) at the expense the post-reflected shock region. of stable chemical composition, temperature uniformity, or ther- Helium and helium-nitrogen driver gas mixtures, as well as modynamic equilibrium, respectively. Shock-heating provides a driver-section volume inserts, were used to achieve long test times near-ideal mechanism for instantaneously achieving a well-known, with a steady pressure profile over a broad range of tempera- homogeneous, high-temperature and high-pressure environment ture and pressure conditions [17,18] . Test gas mixtures were pre- with maximum temperatures and pressures in excess of 10 0 0 0 K pared manometrically in a stainless-steel mixing tank using either and 10 0 0 atm, respectively. There is unfortunately the significant neat gases ( > 99%) or certified standard gas mixtures ( ± 2% un- caveat of a very short test time (typically 1 to 10 ms but up to certainty on reported concentration) to supply the target species 100 ms is possible) in facilities that employ this mechanism, such and high-purity argon ( > 99.999%) as a diluent. For the measure- as shock tubes. ments reported here, only dilute test mixtures in argon with a tar- In addition to the challenges of generating the desired con- get species concentration no greater than 5% were used in order to ditions in a laboratory, the experimental challenges are further limit the endothermic effects of pyrolysis at high temperatures. increased for the collection of spectroscopic reference data un- der these circumstances. This is due both to the characteristics 2.2. Lasers and optics of molecular spectra at high temperatures/pressures and the per- formance of the tools used to acquire spectra. High temperatures Two laser systems were integrated into the shock tube in or- and high pressures lead to significant line broadening. The out- der to perform absorption cross section measurements using the come of these broadening effects, in concert with the emergence fractional transmission through the test gas. The fractional trans- of hot lines, is highly congested and blended spectra; wherein, in- mission is determined by the Beer–Lambert relation: dividual lines are typically not observable and the local absorption I or emission intensity is the result of the superposition of many = exp (−α) = exp (−σ nL ), (1) neighboring lines. At these