In order to obtain more detailed characteristics and information in highly turbulent flames, for a better understanding of the transient behavior of eddies in such flames, a measurement technique with sufficient temporal resolution is requested. However, the probing of species distributions relevant in combustion (e.g. OH, CH2O) with ultra-high-speed laser diagnostics still remains a challenge. Nd:YAG clusters commercially available can generate only 4-8 pulses, although with high laser energy. Systems based on a diode-pumped solid-state Nd:YAG laser combined with a dye laser produce only about 100 μJ pulse energy at ultra-high repetition rates (≥50 kHz). Even more comprehensive information on the flame structure can be gained if simultaneous recording of multi-species is performed. In the present work, the development of the first ultra-high-speed diagnostic technique capable of simultaneous probing of hydroxyl radicals and formaldehyde distributions at a repetition rate of 50 kHz is outlined. This has been achieved by employing a burst laser pumped optical parametric oscillator system for the simultaneous detection of CH2O excited at 355 nm and OH-radicals excited at 283 nm, where the interference of scattering laser light can be avoided. The applicability of the proposed technique was demonstrated in a highly turbulent jet flame. Moreover, the presented improvement in terms of the number of consecutive images recorded with ultra-high-speed planar laser induced fluorescence imaging is significant. Due to the high temporal resolution, the movement of CH2O pocket enclosed by OH at the flame tip can be clearly captured. The transport velocity of the CH2O pocket was calculated and found to be in good agreement with previous LDV results.
© 2017 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Turbulent combustion commonly occurs in most practical advanced combustors nowadays, such as internal combustion engines (ICE) and gas turbines. Comprehensive understanding of turbulent combustion calls for more advanced modelling and diagnostic techniques. Eddies in the flow field of a turbulent flame can form, interact and propagate rapidly. The time-scales of both turbulence and chemical reactions span widely from nanoseconds to seconds. For a better understanding of the transient behavior (small integral time-scales) of wrinkled structures and eddies in highly turbulent flames, sufficient temporal resolution is needed. However, that still remains challenging experimentally.
During the last two decades, the field of high speed diagnostics has seen significant development. Initially, burst-mode lasers were introduced. These are based on sequential amplification of a high repetition rate seed laser [1–3]. Soon after, Nd:YAG clusters were developed that can produce bursts with short pulse separation (theoretically with no minimum time limitation between pulses) making them capable of resolving very small time scales, e.g. 125 µs  and 10 µs , while maintaining pulse energies similar to what is normally found in 10 Hz systems. The architecture of complete parallel systems in the YAG clusters implies that they can be faster than the burst-mode systems but their main drawback is the limited number of pulses. However, the burst mode systems allow for high temporal resolution under prolonged time scales . In several studies as mentioned below burst-mode systems have been used for the investigation of turbulent flames.
Using a burst-mode YAG-laser, probing of species distributions relevant in combustion, e.g. CH2O excited at 355 nm can be achieved through planar laser induced fluorescence (PLIF). Gabet et al. demonstrated the visualization of CH2O in a turbulent CH4-air flame with excellent signal-to-noise ratio . Taking it one step further, Slipchenko et al. developed a burst-mode laser and one of the applications exhibited the detection of CH2O in a CH4-air diffusion flame at 20 kHz repetition rate . This approach has later been improved by Michael et al.  to facilitate imaging of CH2O at 100 kHz.
Besides CH2O species, OH is also an important radical in combustion, and its formation is commonly employed as a marker of the reaction zone in a flame. However, excitation of OH radicals cannot be accessed directly by the harmonics from the commonly employed Nd:YAG laser systems. This could present a challenge for ultra-high-speed diagnostics since although the pump lasers might be available, rapid high energy pumping of dye-lasers has proven to be problematic . Albeit there has been a significant development of dye-lasers since then, there are no reports demonstrating that dye-lasers operate efficiently in combination with high-energy and high-repetition-rate pump lasers. The first ultra-high-speed (50 kHz and above) OH measurement was conducted by Miller et al.  in a H2-air diffusion flame. They used the output of a seeded optical parametric oscillator (OPO), mixed with residuals at 532 nm from the pump source, to achieve laser radiation at around 313 nm. This was used to probe the P2(10) line in the (0,0) vibrational band of OH. Although this excitation scheme is favored by a high fluorescence yield, scattering light from the laser prohibits such a diagnostic approach from being employed in practical combustion systems with confined combustion chambers, such as ICE.
An alternative excitation scheme commonly employed for the detection of OH radicals is to probe the (1,0) vibrational transition, at approximately 284 nm. The red-shifted fluorescence, at 308 nm, facilitates the spectral filtering of the spurious laser radiation. High speed OH PLIF can be achieved by using a diode-pumped solid-state (DPSS) Nd:YAG laser combined with a dye laser, which has been reported in, for example, premixed CH4-air flames [11, 12]. Nevertheless, the detection of OH radicals at repetition rates higher than 20 kHz by means of this laser system significantly has disadvantage of the low pulse energy (about 100 µJ) available, leading to low signal-to-noise ratio (SNR). Hammack et al. demonstrated OH-PLIF imaging in a Henchen burner and a DC transient-arc plasmatron at 50 kHz repetition rate with a DPSS and a dye laser . However, in a CH4-air turbulent flame, the OH concentration is very low, .e.g. several thousand ppm calculated by numerical simulation with modeling provided in .
Experience from earlier work, where dye-lasers were pumped using the high energy output from the Multi-YAG system , indicates that such technology would not be favorable for the applications of burst-mode laser system. For ultra-high-speed OH detection, the combination of a burst-mode YAG-laser and an OPO could be a viable solution for providing a decent compromise between pulse energy and repetition rate. Sjöholm et al. have successfully demonstrated the utilization of a YAG-cluster/OPO combination to excite the A-X (1,0) vibration band of the OH molecule at 282.97 nm with 8-pulses per burst, and to detect the fluorescence of OH through the A-X (0,0) transition around 308 nm . Another example is that Jiang et al. employed a burst-mode laser system in combination with an OPO at 10 kHz repetition rate to investigate the presence of CH in a DLR Flame A .
It has been demonstrated extensively that multi-scalar detection plays an important role in the understanding of turbulent flame structures [17–20]. Further comprehension of the flame structure can be gained due to detailed information that simultaneous recording of multi-species provide. Naturally, different species represent different characteristics of the flame. The highest gradient of OH is a well-established marker of the flame front as mentioned before. CH2O is mostly present in the preheat zone and is, hence, often used as a marker of that zone. Furthermore, simultaneous measurement of OH and CH2O has been proposed to provide a good qualitative measure of the heat release rate  which is an important quantity for studying turbulent combustion. For these reasons, simultaneous ultra-high-speed OH and CH2O measurements can provide both scalar and temporal information on the evolution of the pre-heat zone and flame topology. This can potentially aid the understanding of turbulent flames and thus be useful for model validation. At present, this combination of multiple scalars, probed at such ultra-high repetition rates, has not yet been accomplished according to the authors’ best knowledge.
This work aims at developing the first ultra-high-speed diagnostic technique capable of simultaneous probing of hydroxyl radicals and formaldehyde distributions in a highly turbulent flame at a repetition rate of 50 kHz. This has been achieved by employing a burst laser pumped OPO system. The output from such high speed multi-scalar measurements can bring out new insights to the interaction between turbulence and chemical reactions.
Moreover, 100 consecutive images were recorded in the present work. Therefore, this brings the possibility of following events that occur over longer sequences than in the previous work by Sjöholm and Li et al. with 8 images [18, 22], and Miller et al. who presented 28 pulses .
The novelty of the present study and the difference compared to previous studies are as follows. Firstly, the OH-PLIF is performed utilizing the 283.93 nm Q1(9) transition, which has less temperature dependence. Secondly, a 310 nm band-pass filter is used in order to suppress most of the scattered light at 283.93 nm, which facilitates measurements also in confined combustion chambers. The latter cannot be achieved by exciting at 309 nm due to lower SNR. Thirdly, the sequence of 100 images (simultaneous CH2O/OH) is significantly longer than reported in previous work [10, 22]. This would be highly beneficial, for example, in ICE applications where the entire combustion phase in a single cycle can be captured.
2. Experimental setup
Figure 1 illustrates the experimental setup for simultaneous multi-species visualization in a premixed turbulent jet flame. The burner is a hybrid porous-plug/jet type burner (LUPJ burner), described in more detail in [17, 18, 23–25]. Its main parts are a porous sintered stainless steel plug with a diameter of 61 mm and a 1.5 mm-diameter nozzle in the center for the creation of a jet flame. Premixed CH4 and air mixture were fed through the center nozzle to create the jet flame which was stabilized by a reacting co-flow in the form of a premixed CH4-air flat-flame above the porous plug. The gas flows were regulated by six mass flow controllers (Bronkhorst), all calibrated at 300 Kelvin with higher than 98.5% accuracy. Various conditions were studied by altering the flow speed in the jet and co-flow as well as the air-fuel ratio as shown in Table 1. The turbulent Reynolds numbers, Kolmogorov length scales and Karlovitz numbers of the jet flame are depict in Table 2. Detailed calculation is presented in the previous study .
The laser system used in the experiment to pump an OPO (GWU, PremiScan/ MB) is a state-of-the-art ultra-high-speed laser (QuasiModo by Spectral Energies, LLC), which is similar but not identical to the one developed by Slipchenko et al. . In short, the laser system is a hybrid-pumped burst-mode laser with two diode-pumped amplifiers followed by five flashlamp-pumped amplifiers. Two 4-mm flashlamp-pumped amplifiers serve as a double-pass for the laser beam in order to increase the gain. The beam is being further amplified after passing through three flashlamp-pumped amplifiers. In the future, the system will be equipped with one extra 12-mm amplifier which will result in an energy output twice as high as for the present configuration.
The measured output energy of this laser is approximately 210 mJ/pulse at 1064 nm and 50 kHz repetition rate. The maximum burst duration is 10 ms which refers to 500 consecutive laser pulses per burst, at 50 kHz. However, a burst duration of 2 ms was used in the presented work with a 10 seconds separation between the bursts, in order to avoid thermal lensing effects in the amplifiers. The pulse duration was set to 15 ns at full width at half maximum (FWHM) and the spectral bandwidth was below 0.03 cm−1 at 1064 nm.
A third harmonic generator (THG) crystal (KDP type) was utilized in order to produce radiation at 355 nm. The peak output energy after the THG was measured at 80 mJ/pulse. The beam size was reduced to approximately 4 mm in diameter, using a telescope before the β-barium borate (BBO) crystal, in order to increase the power density and consequently the conversion efficiency of the OPO. The output wavelength of the OPO signal beam was measured by a wavemeter (GWU, LambdaScan) and the desired wavelength was selected before frequency doubling in a KDP type crystal.
In the present work an OH excitation scan from 282.3 nm to 284.5 nm was performed initially. The emission from the A-X (0, 0) transition was collected at around 308 nm, through an interference filter (λT = 310 ± 10 nm) mounted in front of a CMOS camera (Photron SA-Z) equipped with a lens coupled high speed intensifier (Lavision HS-IRO).
In Fig. 2 the excitation scan is presented together with a fitted curve generated by LIFBASE . In the LIFBASE simulation the settings were selected based on a 0.06 nm resolution Gaussian profile and 1600 K flame temperature. After comparing the OH excitation scan with the LIFBASE simulation, an experimental resolution of approximately 0.06 nm could be estimated. The specific transition at 283.93 nm (A2Σ+-X2Π, 1-0 transition) was selected because of its high population and limited dependence on the flame temperature in the 1500-2500 K range. Furthermore, it is spectrally well separated from other transitions lines. Further information on this specific transition line can be found in .
The OH excited wavelength will be expressed as 284 nm instead of 283.93 nm below.
In Fig. 1, after the doubling crystal, a Pellin Broca prism was used to separate 284 nm from 568 nm. The 284 nm beam was then formed into a laser sheet by a cylindrical lens (f = −40 mm) in combination with a spherical lens (f = 200 mm). The resulting pulse energy of the 284 nm at 50 kHz was about 350 µJ/pulse, which is sufficient for getting good SNR in this application. The shot-to-shot variation in pulse energy is less than 10%.
CH2O was excited in several rotational transitions at 355 nm . For this purpose the residuals of the OPO pump beam were collected by a prism (#1) as shown in Fig. 1. A telescope in the 355 nm beam path was used to locate the focal points of both 284 nm and 355 nm beams at the center of the burner with only one set of sheet forming optics. The 355 nm beam was redirected, along the same direction as the 284 nm beam, by another prism (#2) which was positioned just below the 284 nm beam path. The height of the 284 nm and the 355 nm laser sheets were around 23 mm and 14 mm respectively and the thickness of both was coarsely measured to be less than 200 µm.
Two high speed cameras in combination with two high speed intensifiers were mounted opposite to each other in order to capture both OH and CH2O fluorescence simultaneously. For detecting the former species, a CMOS Photron Fastcam SA-Z (see Fig. 1: High speed camera #1) was used in combination with a high speed intensified (LaVision HS-IRO). For the CH2O imaging, a CMOS Photron Fastcam SA-X2 (see Fig. 1: High speed camera #2) was employed which was also combined with a high speed intensifier (Lambert HS IRO 2 stage). The resolution had to be reduced due to the limited read-out speed of both cameras at 50 kHz. In the present work, the resolution of the CMOS cameras was reduced to 1024 x 400 and 640 x 384, respectively. The exposure gate-width of both high speed intensifiers was set to 100 ns to make sure that any temporal jittering in the laser/detector system would not affect the results. At 100 ns exposure time, background signals caused by the chemiluminescence of the flame were proved to be negligible. Both image intensifiers were operated in the middle of their respective gain range. The parameters of camera lens and filters utilized for the cameras are described in Table 3.
The residual 355 nm and the OPO output at 568 nm were monitored using fast photo diodes (PDs). By observing the signal of PDs, the gate of intensifiers and the exposure timing of cameras on an oscilloscope, accurate timing could be achieved for the image acquisition. By carefully aligning the two cameras the corresponding field of view (FOV) which is 27 mm x 27 mm measuring at full chip could be made virtually identical. A transparent grid object was used to generate reference images for both cameras. Hence, pixel to pixel correspondence between images recorded by the two cameras could be achieved through image registration.
3. Results and discussion
As a 2ms burst contains 100 pulses in the experiment, thermal load on the BBO crystal of the OPO is quite high leading to potential problems. Due to the increased thermal load of the BBO in longer bursts, optimal phase matching conditions cannot be maintained throughout the full duration of the burst. This causes a decrease in the conversion efficiency for subsequent pulses and, hence, a steady drop in available laser energy at 284 nm. One might expect a similar behavior from the pump source at 355 nm, but this was verified to show only a marginal drift in pulse energy during a burst. It was also noted that the output wavelength from the OPO was drifting shot-by-shot after a ‘cold start’. After being ‘warming-up’ by the burst laser, the wavelength shift stabilized and it could be compensated by adjusting the angle of the BBO.
To demonstrate the applicability of the proposed diagnostics, a practical application of this ultra-high-speed imaging technique in a turbulent jet flame is described below. 100 consecutive images of both species in the turbulent flame were captured with a 20-μs interval between successive images, corresponding to a 50 kHz data acquisition rate.
The SNR was calculated based on dividing the average value of the signal by the standard deviation of the background noise from the region where laser radiation is present but there is no fluorescence signal. Figure 3 shows the raw images of OH and CH2O signals prior to post-processing. One solid line and one dashed line were selected for both OH and CH2O at different heights above the burner for SNR calculation. Figure 4 contains plotted signal profiles of OH and CH2O along the red lines, respectively. The averaged value of the signal is marked as Is and I’s while the standard deviation of the background noise is indicated as In and I’n. The region for signal or noise calculation is illustrated as the width of the arrows in Fig. 4. The resulting SNR for OH and CH2O were calculated to be around 40 and 18, respectively.
It is worth pointing out that both CH2O and Polycyclic aromatic hydrocarbon (PAH) mark the fuel-rich regions of the flame. However, from numerical simulations, the CH2O concentration in CH4-air flames is about one thousand ppm at stoichiometric condition while PAH concentration is much lower (<10 ppm) . Furthermore, a PLIF measurement was conducted in a CH4-air flame at stoichiometric condition with 266 nm and 355 nm wavelengths respectively to check the contribution of PAH fluorescence on the CH2O signal. The results show that the PAH fluorescence is barely detected, which is as low as the background noise, while the CH2O signal is well distinguished. Therefore, the influence of PAH on CH2O signal is considered to be negligible in this experiment. PAH can also be excited by 284 nm laser beam. Any possible cross talk was evaluated by blocking one laser at the time, which revealed no detectable cross talk as well. For high PAH concentration flames, preferred solution can be delaying the OH excitation beam significantly (on the order to 50 ns) with respect to the CH2O excitation beam to avoid the resulting cross talk.
The first case in Table 1 is presented in Fig. 5, where the red region indicates the OH distribution. The green region, which is enclosed by the OH distribution, represents the distribution of CH2O, a species commonly used as a marker of low temperature reactions in the preheat zone. In the post-processing, both OH and CH2O images have been normalized by their maximum value. In Fig. 5 one can notice the strong spatial correlation between the distributions of these two species. The number in the upper right corner of each sub-image indicates the number of the frame in a consecutive sequence of images. For the conditions prevailing, the time scale, τ (τ = (ν/ε)0.5), at 45 mm height above the burner (HAB) is about 0.05 ms for a jet speed of 66 m/s and it is around 0.02 ms for 110 m/s. In addition, the integral time scale, τ0 (τ0 = l0/u), is approximately 0.44 ms and 0.22 ms, respectively. Thus, the temporal resolution achieved is capable of following the turbulent development of this jet flame.
For brevity, only a couple of consecutive OH and CH2O images are presented in the “results and discussion” section. But additional data are available in the appendix and supplementary materials (see Visualization 1, Visualization 2, Visualization 3).
It is worth mentioning that in some situations, 100 images of OH and CH2O are critically useful for the combustion characterization within a specific time domain. For example, in the homogeneous charge compression ignition (HCCI) and partially premixed combustion (PPC) engines, the duration of the main combustion phase is less than 15 crank angle degrees which corresponds to 2 ms at 1200 rpm (100 images at 50 kHz). Furthermore, in turbulent combustion studies, the chance of capturing extinction and re-ignition events can be significantly increased when more consecutive images are available.
In Fig. 5, the CH2O signal is absent from the upper and lower part of the imaged area, because of the limitation in laser energy and the non-uniform energy distribution in the residual OPO pump beam. However, this does not mean that there is no CH2O present in the flame.
The CH2O was transported downstream the jet flame by convection, e.g. several flame structures (spatial structure a, b and c) in the sub-images of Fig. 5 (start from No. 35, No. 40 and No. 70) can be observed and followed as the combustion process progress. The temporal resolution is sufficient to visualize the process in which the CH2O region was disconnected into small pockets at the ‘neck’ point of the flame. Later, the CH2O pocket was deformed into thin and long ligaments due to the turbulence straining and shearing, while being consumed.
These results are in good agreement with the results obtained by Sjöholm et al.  and Zhou et al. . In those publications single-shot imaging captured at 10 Hz shows that the CH2O is usually an extended thin ligament at the tip of the flame. However, with the utilization of this ultra-high-speed diagnostics, the development of highly turbulent flame structures can be visualized and investigated for a better understanding of turbulent combustion processes.
In case 2, the equivalence ratio of the jet flow was reduced to 0.3 and the co-flow speed was decreased to one-third of the previous case (case 1), leading to local extinction as shown in Fig. 6. As a consequence, the jet flame became unstable because less heat was supplied by the surrounding co-flow flame as well as leaner combustion for the jet flow and local quenching of OH could be identified. For instance, from sub-images No. 33 to No. 42,OH was hardly observed at the height above 44 mm HAB while CH2O distributed in this domain. It is evident that starting from No. 44 (Fig. 6), re-ignition occurs at the flame-tip as indicated by the OH layer propagating to a higher position.
Figure 7(a) illustrates the comparison between the velocity of CH2O pockets and the velocity data measured by laser Doppler velocimetry (LDV) in a previous study . The grey box represents the data from LDV and red triangle and black circle symbols indicate the data calculated from 66 m/s and 110 m/s PLIF images, respectively. The pocket was evaluated by the OH PLIF image due to the large FOV as shown in Fig. 7(b). The velocity of the CH2O pocket was calculated by the movement of the centroid of a CH2O pocket between two frames (Δt = 20 µs). The result of CH2O pocket imaging velocimetry and LDV data are in very good agreement for both case 1 and case 3, as shown in Fig. 7(a). Therefore, by using such ultra-high-speed diagnostics, proper data evaluation and analysis can provide complementary velocity information in a turbulent jet flame.
4. Summary and conclusion
According to the literature survey performed by the authors, this is the first time that simultaneous PLIF imaging of two species relevant to combustion studies has been conducted at such high repetition rate, 50 kHz, including 100 consecutive images. Furthermore, the ability to extract more than 30 pulses from an OPO being pumped at 50 kHz is also being demonstrated for the first time. The OH radicals were excited in the Q1(9) transition at 283.93 nm generated by 355 nm pumping of an OPO followed by frequency doubling of the signal beam. Formaldehyde distributions were accessed through excitation at 355 nm utilizing the residuals of the OPO pumping beam. By applying a 2 ms long burst at 50 kHz, a pulse train of 100 individual pulses could be generated from the burst laser. The peak output energy was 80 mJ/pulse at 355 nm and after frequency conversion in the OPO this resulted in approximately 350 µJ/pulse at 284 nm and 6 mJ/pulse remaining at 355 nm. Using appropriate filters in front of two intensified CMOS detectors the feasibility of this ultra-high-speed diagnostics technique could be successfully demonstrated in a premixed turbulent jet flame featuring flow velocities in excess of 60 m/s. Even at these high velocities the demonstrated diagnostics approach was capable of following the temporal evolution of the reacting flow.
The financial support from CECOST and KCFP through the STEM (Swedish Energy Agency) are gratefully acknowledged. The authors would also like to thank Prof. Xuesong Bai, Prof. Alexander Konnov, Dr. Mikhail Slipchenko and Dr. Jason Mance for their valuable help and contribution.
References and links
1. W. Lempert, P. F. Wu, R. Miles, B. Zhang, J. Lowrance, V. Mastrocola, and W. Kosonocky, “Pulse-burst laser system for high-speed flow diagnostics,” in Proceedings of 34th Aerospace Sciences Meeting and Exhibit, AIAA Paper 96–0179 (1996).
2. W. Lempert, P. F. Wu, and R. Miles, “Filtered rayleigh scattering measurements using a MHz rate pulse-burst laser system,” in Proceedings of 35th Aerospace Sciences Meeting and Exhibit (1997), paper 97–0500 (1997).
3. P. F. Wu, W. Lempert, and R. Miles, “Megahertz pulse-burst laser and visualization of shock-wave/boundary-layer interaction,” AIAA J. 38(4), 672–679 (2000).
4. C. F. Kaminski, J. Hult, and M. Aldén, “High repetition rate planar laser induced fluorescence of OH in a turbulent non-premixed flame,” Appl. Phys. B 68(4), 757–760 (1999).
5. J. Nygren, J. Hult, M. Richter, M. Aldén, M. Christensen, A. Hultqvist, and B. Johansson, “Three-dimensional laser induced fluorescence of fuel distributions in an HCCI Engine,” Proc. Combust. Inst. 29, 679–685 (2002).
6. B. Thurow, N. Jiang, and W. Lempert, “Review of ultra-high repetition rate laser diagnostics for fluid dynamic measurements,” Meas. Sci. Technol. 24, 012002 (2013).
7. K. N. Gabet, R. A. Patton, N. Jiang, W. Lempert, and J. A. Sutton, “High-speed CH2O PLIF imaging in turbulent flames using a pulse-burst laser system,” Appl. Phys. B 106, 569–575 (2012).
8. M. N. Slipchenko, J. D. Miller, S. Roy, J. R. Gord, S. A. Danczyk, and T. R. Meyer, “Quasi-continuous burst-mode laser for high-speed planar imaging,” Opt. Lett. 37(8), 1346–1348 (2012). [PubMed]
9. J. B. Michael, P. Venkateswaran, J. D. Miller, M. N. Slipchenko, J. R. Gord, S. Roy, and T. R. Meyer, “100 kHz thousand-frame burst-mode planar imaging in turbulent flames,” Opt. Lett. 39(4), 739–742 (2014). [PubMed]
10. J. D. Miller, M. Slipchenko, T. R. Meyer, N. Jiang, W. R. Lempert, and J. R. Gord, “Ultrahigh-frame-rate OH fluorescence imaging in turbulent flames using a burst-mode optical parametric oscillator,” Opt. Lett. 34(9), 1309–1311 (2009). [PubMed]
11. I. Boxx, C. M. Arndt, C. D. Carter, and W. Meier, “High-speed laser diagnostics for the study of flame dynamics in a lean premixed gas turbine model combustor,” Exp. Fluids 52, 555–567 (2012).
12. R. Wellander, M. Richter, and M. Aldén, “Time-resolved (kHz) 3D imaging of OH PLIF in a flame,” Exp. Fluids 55, 1746 (2014).
13. S. Hammack, C. Carter, C. Wuensche, and T. Lee, “Continuous hydroxyl radical planar laser imaging at 50 kHz repetition rate,” Appl. Opt. 53(23), 5246–5251 (2014). [PubMed]
14. A. A. Konnov, “Implementation of the NCN pathway of prompt-NO formation in the detailed reaction mechanism,” Combust. Flame 156, 2093–2105 (2009).
15. J. Sjöholm, E. Kristensson, M. Richter, M. Aldén, G. Göritz, and K. Knebel, “Ultra-high-speed pumping of an optical parametric oscillator (OPO) for high-speed laser-induced fluorescence measurements,” Meas. Sci. Technol. 20, 025306 (2009).
16. N. Jiang, R. A. Patton, W. Lempert, and J. A. Sutton, “Development of high-repetition rate CH PLIF imaging in turbulent nonpremixed flames,” Proc. Combust. Inst. 33, 767–774 (2011).
17. B. Zhou, C. Brackmann, Z. Li, M. Aldén, and X.-S. Bai, “Simultaneous multi-species and temperature visualization of premixed flames in the distributed reaction zone regime,” Proc. Combust. Inst. 35, 1409–1416 (2015).
18. J. Sjöholm, J. Rosell, B. Li, M. Richter, Z. Li, X.-S. Bai, and M. Aldén, “Simultaneous visualization of OH, CH, CH2O and toluene PLIF in a methane jet flame with varying degrees of turbulence,” Proc. Combust. Inst. 34, 1475–1482 (2013).
19. B. Zhou, Q. Li, Y. He, P. Petersson, Z. Li, M. Aldén, and X.-S. Bai, “Visualization of multi-regime turbulent combustion in swirl-stabilized lean premixed flames,” Combust. Flame 162, 2954–2958 (2015).
20. B. Zhou, Advanced laser-based multi-scalar imaging for flame structure visualization towards a deepened understanding of premixed turbulent combustion (Doctoral Thesis, Lund University 2015).
21. B. Böhm, C. Heeger, R. L. Gordon, and A. Dreizler, “New perspectives on turbulent combustion: multi-parameter high-speed planar laser diagnostics,” Flow Turbul. Combus. 86, 313–341 (2011).
22. Z. Li, J. Rosell, M. Aldén, and M. Richter, “Simultaneous burst imaging of dual species using planar laser-induced fluorescence at 50 kHz in turbulent premixed flames,” Appl. Spectrosc. 71(6), 1363–1367 (2017). [PubMed]
23. B. Zhou, C. Brackmann, Q. Li, Z. Wang, P. Petersson, Z. Li, M. Aldén, and X.-s. Bai, “Distributed reactions in highly turbulent premixed methane/air flames,” Combust. Flame 162, 2937–2953 (2015).
24. B. Zhou, C. Brackmann, Z. Wang, Z. Li, M. Richter, M. Aldén, and X.-S. Bai, “Thin reaction zone and distributed reaction zone regimes in turbulent premixed methane/air flames: scalar distributions and correlations,” Combust. Flame 175, 220–236 (2017).
25. J. Rosell, X.-S. Bai, J. Sjoholm, B. Zhou, Z. Li, Z. Wang, P. Pettersson, Z. Li, M. Richter, and M. Alden, “Multi-species PLIF study of the structures of turbulent premixed methane/air jet flames in the flamelet and thin-reaction zones regimes,” Combust. Flame 182, 324–338 (2017).
26. J. Luque and D. R. Crosley, “LIFbase: database and spectral simulation (Version 2.0.60, 2008),” Sri. Int. Report MP. 99–009 (2008).
27. B. Halls, P. Hsu, N. Jiang, E. Legge, J. Felver, M. N. Slipchenko, S. Roy, T. R. Meyer, and J. R. Gord, “kHz-rate four-dimensional fluorescence tomography using an ultraviolet-tunable narrowband burst-mode optical parametric oscillator,” Optica 4, 897–902 (2017).
28. C. Brackmann, J. Nygren, X. Bai, Z. Li, H. Bladh, B. Axelsson, I. Denbratt, L. Koopmans, P.-E. Bengtsson, and M. Aldén, “Laser-induced fluorescence of formaldehyde in combustion using third harmonic Nd:YAG laser excitation,” Spectrochim. Acta A Mol. Biomol. Spectrosc. 59(14), 3347–3356 (2003). [PubMed]
29. A. D’Anna and A. Violi, “A kinetic model for the formation of aromatic hydrocarbons in premixed laminar flames,” in Proceedings of 27th Symposium on Combustion, 425–433 (1998).