Open Access
Add to BibSonomy
Abstract
We report the high-speed imaging of multi-species and multi-parameter combustion diagnostics for turbulent non-premixed jet flames using a three-legged burst-mode laser system. Simultaneous OH/CH2O planar laser-induced fluorescence and Rayleigh-scattering imaging measurements at a 10-kHz rate are obtained. OH and CH2O concentrations, flame temperatures, and heat-release rates are simultaneously acquired in two-dimensions at 10 kHz.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
High-speed laser-based measurement techniques have been developed and applied to flow and combustion diagnostics in recent decades [1–3]. Species concentrations [4–6], flow temperature fields [7], velocity fields [8–10], and other important parameters [11] can be spatially and temporally resolved using high-speed laser diagnostics, which enables understanding of turbulent combustion dynamics and provides useful data for validating combustion models. Unlike traditional 10-Hz laser diagnostic techniques which can only provide flow statistics, high-speed laser diagnostic techniques can track combustion dynamics and instability at rates governed by laser repetition rates of 10 kHz or greater [2,6,7]. To advance the understanding of turbulent combustion, simultaneous measurements of multiple flow parameters (e.g., temperature, velocity, heat-release rate, etc.) and multiple key combustion species (e.g., OH, CH2O, CH, etc.) at high repetition rates are desired. However, such multi-species and multi-parameter combustion imaging measurements remain challenging because of the requirement for multiple high-speed, high-energy lasers. State-of-the-art 10–50 kHz diode-pumped solid-state (DPSS) lasers are commonly used for single-parameter detection because of limited laser energy. Burst-mode lasers can provide ~20 times more energy, though not continuously, at the same repetition rate compared to DPSS systems, and hence they are a more promising approach for multi-species and multi-parameter measurements. Nevertheless, standard single-legged burst-mode lasers still have limited capability for the desired multi-parameter measurements such as simultaneous OH/CH2O planar laser-induced fluorescence (PLIF). Although a single-legged laser could excite OH and CH2O simultaneously with 283 nm and 355 nm respectively [12], the captured CH2O PLIF images will have interference from Polycyclic Aromatic Hydrocarbons (PAH) fluorescence which is efficiently excited by 283 nm. The PAH fluorescence is usually difficult to be removed by bandpass filters. Although an optical delay with a long path length for laser beam at 355 nm could somewhat resolve this problem, optically delayed beams usually have distorted beam profiles, and available delay times are usually less than 50 ns [13], which is not sufficient to eliminate the 283-nm-excited PAH contamination effects. In addition to the signal-interference problem, the output energy of a single-legged burst-mode laser may not be sufficient to pump two or more optical parametric oscillators (OPOs) for simultaneous detection of multiple species. As a result, multiple high-speed lasers are typically required for multi-species and multi-parameter combustion imaging [14]. Recently, a three-legged, high-speed burst-mode laser system was developed to provide sufficient energy for simultaneous high-speed measurements of multi-flow parameters and multiple species in turbulent combustion flows [15]. This laser system can simultaneously generate three beams of high-repetition-rate and high-energy laser pulses with the capability of controlling the time delay between the three beams, making it possible to avoid optical-interference problems in combustion diagnostics.
In this paper, we present simultaneous high-speed measurements of temperature, main heat-release rate, and multiple species in a turbulent non-premixed DLR-A jet flame (a standard flame used in turbulent combustion studies [16]) using a three-legged burst-mode laser system. OH PLIF, CH2O PLIF, and Rayleigh scattering at a 10-kHz rate were implemented in this study. OH and CH2O are often used as indicators for the flame front and preheat zone, respectively. The direct product of the reaction of OH and CH2O, HCO, represents the main heat-release source in hydrocarbon-related combustion [17]. Because of the short lifetime and low concentrations of HCO, its direct measurement remains difficult, and hence the product of the OH and CH2O concentrations is usually used as an indicator of the heat release rate (HRR) [18]. Therefore, the presented high-speed simultaneous OH/CH2O PLIF imaging measurements provide information on flame dynamics from the OH and CH2O species concentration distributions as well as HRR information. Temperature is another important parameter in combustion studies. There are a few techniques for two-dimensional (2D) temperature measurement in reacting and non-reacting flows such as coherent anti-Stokes Raman scattering (CARS) [19] and two-color OH PLIF [20,21]. However, the measurement speeds for such approaches are usually below 1 kHz. Rayleigh scattering can be used for 2D temperature measurements in a medium with uniform Rayleigh scattering cross sections. For the experiments on the DLR-A flame reported herein, the local mixture-averaged Rayleigh-scattering cross section of this flame has previously been experimentally verified to be nearly constant (within 3% uncertainty) [16]. Therefore, the Rayleigh-scattering signal from the DLR-A flame can be assumed to be inversely proportional to the flame temperature [7].
2. Experimental details
A schematic diagram of the experimental setup is shown in Fig. 1. The three-legged burst-mode laser has three different laser beams as outputs, each of which consists of a 10-kHz train of pulses with 10-ns pulse widths. The timing diagram for the outputs of the three-legged burst-mode laser is also shown in Fig. 1. The details of the three-legged burst-mode laser are available in [15]. The 532-nm laser beam (~700 mJ/pulse) from Leg 1 is used for the Rayleigh scattering measurements, and the 355-nm laser beam (~110 mJ/pulse) from Leg 2 is used for CH2O-PLIF imaging. The 355-nm laser beam from Leg 3 is first converted to a signal beam at approximately 609 nm, using an optical parametric oscillator (OPO), which mixes with the residual 532 nm beam to generate a 284-nm laser beam (∼5 mJ/pulse and ∼300 MHz linewidth), via sum-frequency mixing, for the excitation of the Q1(9) transition of OH. These three laser beams, i.e., at 532, 355, and 284 nm, having inter-pulse intervals of ~100 ns, are combined and then transmitted through a common set of sheet-forming optics before entering the test section. A cylindrical lens (f = −25 mm) and a spherical lens (f = + 400 mm) are used to generate laser sheets of ~150-mm in height and and ~0.5-mm in thickness. Only the uniform central section (~70 mm) of the laser sheets are used for the PLIF and Rayleigh imaging. Three high-speed CMOS cameras (Photron SA-Z) equipped with high-speed intensifiers (LaVision IRO) are utilized for the OH/CH2O PLIF and Rayleigh imaging; all of the intensifier gains are set to ~50% to ensure that the maximum intensities in the images are close to 85% of the saturation count of the cameras. Spectral filters for the OH/CH2O/Rayleigh imaging are placed in front of the respective cameras to eliminate background noise. Both the laser and cameras are operated at 10 kHz, and a master timing box is used to control the laser pulse delays and camera triggers.
Fig. 1 Schematic diagram (top) of the experimental setup for simultaneous Rayleigh/OH-PLIF/CH2O-PLIF measurements, and a timing diagram (bottom) for the three pulses respectively used to generate the Rayleigh, OH-PLIF, and CH2O-PLIF excitations.
The turbulent jet burner used in these experiments consists of an 8-mm-diameter central fuel jet sheathed by an annular 300-mm-diameter co-flow jet. To generate a DLR-A flame, the central fuel jet is a mixture of CH4, H2, and N2 with concentrations of 22.1%, 33.2%, and 44.7% respectively. The total flow rate is 127.2 SLPM. The bulk flow speed at the jet exit is ~42 m/s with a cold-flow Reynolds number of 15200. The co-flow gas is purified air with a flow speed of ~0.3 m/s. Since the flow speed is relatively slow (~50 m/s), we can assume that the flow is frozen during the total three-pulse time period of 200 ns.
3. Results and discussion
Simultaneously acquired sequences of 10-kHz images of OH, CH2O, and flow density are shown in Figs. 2–4; in each case, the raw data was processed, by removing background signal and making laser beam profile corrections, to prepare the image. The pulse-to-pulse laser energy fluctuations were recorded by photodiodes. The laser beam profile averaged over a burst was captured by a fluorescent plate for correction purposes. The field-of-view of these images is ~50 mm (X, width) × 60 mm (Y, height) and the spatial resolution is ~65 µm/pixel, in both the X and Y directions, for all three cameras. All the measurements are performed at ~Y/D = 10 (Y is the height above the jet exit, D is the jet diameter which is ~8 mm, i.e., Y = 80 mm here). The intensities in the images shown in Fig. 2 directly reflect the OH radical concentrations. The evolution and propagation of large-scale turbulent structures in the flame front are clearly shown in these OH PLIF images. The average signal-to-noise ratio (SNR) of the OH images is ~50:1 defined as the peak OH PLIF signal divided by the standard deviation of the OH signal fluctuations [6]. The observed OH layer structures vary slowly, because the OH layer is within the outbound flow of the reaction zone or mixing layer, which has relatively low flow-speed as compared to that of the core jet flow.
The image sequence displayed in Fig. 3 illustrates the evolution of the CH2O concentration field. CH2O is an important combustion intermediate occurring within the “cool flame” region, prior to the primary reaction zone, for hydrocarbon-fueled flames; thus, CH2O is a good indicator of the preheat zone in combustion studies. The laser wavelength is carefully tuned to a strong absorption line of CH2O at 28183.65 cm−1 [6]. The minimum wavelength-tuning step for the seeding laser of the three-legged burst-mode laser system is ~40 MHz, and the laser wavelength is monitored by a wavemeter during the wavelength scanning. It was observed that with resonant excitation, the CH2O PLIF signal intensity increases by a factor of 10 as compared to that with off-resonant excitation (the shift in the laser wavelength for excitation of the off-resonant signal was ~0.5 cm−1 with respect to the peak of absorption line). The relatively high intensity (and the corresponding SNR) of the CH2O fluorescence is most likely due to the combination of high input laser energy (~110 mJ/pulse) and resonant excitation. As expected, the CH2O distribution is located within the preheat zone (between the fuel core and high-temperature reaction zones). The CH2O distribution is much broader than that of OH, because OH only occurs within the high-temperature region. The preheating area indicated by the CH2O distribution is in good agreement with the high-speed velocity area from previous particle image velocimetry (PIV) measurements [15], and the OH-PLIF signal is principally located at its the outer edges. The observed CH2O-PLIF signal is weak near the center of the jet exit because of the low temperature of the fuel/N2 mixture.
Figure 4 shows a sequence of Rayleigh-scattering images. Rayleigh scattering intensities directly reflect the density of the scattering medium, thus in reacting flow, higher Rayleigh-scattering signals occur in lower temperature regions. As expected, the highest Rayleigh-scattering signal is generated in the room-temperature air area (outside the jet flame), which has the highest density. Low Rayleigh-scattering signals (appearing blue in Fig. 4) occur in the mixing layer region where the gas density is low due to the high flame temperature. Near the central jet exit, the Rayleigh signal is also high because this area is the flame preheat zone where the temperature is relatively low and contains a mixture of unburned fuels. Because the density of the ambient air is uniform, we used the Rayleigh scattering signal from the ambient air to correct the image data for the effects of fluctuations in pulse energy and the non-uniform beam profile.
The product of the OH and CH2O fluorescence is a good indicator of the instantaneous heat-release rate (HRR) for hydrocarbon-fueled flames. Based on the simultaneous OH/CH2O PLIF measurements [Figs. 2 and 3], the product of their PLIF profiles allows for a qualitative mapping of the HRR in DLR-A flame, as presented in Fig. 5. The spatial distribution of the HRR can also provide the locations of combustion reaction zones. Typically, the locations with higher HRR values have greater heat release and hence higher temperatures (i.e., they are hot zones). A few hot zones (i.e., high HRR locations) and flame holes (i.e., low HRR locations) were observed in the turbulent DLR-A flame. We note that for a DLR-A flame, the HCO concentration only indicates the heat release pathway for CH4 fuel, not for H2. Based on fuel concentrations and released heat energies (891 kJ/mol for CH4 and 242 kJ/mol for H2), we estimated that CH4 supports ~70% of the total heat release for the studied flame. Therefore, the HRR values mapped in Fig. 5 indicate the main HRRs for the entire combustion system.
As mentioned earlier, for a DLR-A flame, the Rayleigh-scattering signal is inversely proportional to the flame temperature because of the constant Rayleigh-scattering cross section. Figure 6 shows the corresponding temperature image sequence, obtained from the Rayleigh-scattering images shown in Fig. 4. The surrounding room-air temperature is ~300 K, and the turbulent-flame temperature varies in the range of 1000–2000 K. Here, the temperature of the jet core flow in the preheat zone is ~400 K. Due to the modest asymmetry of the jet, Figs. 1–5 show that both OH/CH2O concentration and the HRRs are higher on the left hand side of the image which is also confirmed by the high temperature from Rayleigh measurement shown in Fig. 6.
Both the combustion HRR and flame temperature are important parameters in combustion studies. The reacting-flow structures move with a certain flow speed while the chemical reactions release heat energy, i.e., the maximum temperature might be reached at a time delay (Δt) after the occurrence of the maximum heat-release rate. Here, the delay time comes from heat transfer through diffusion processes involving turbulent mixing and flow convection. Considering the thermal-lag effect in the reacting flows, the local maximum flame temperature and local maximum HRR might occur at different places. To our knowledge, this hypothesis has not yet been experimentally confirmed. However, it could be estimated by a simultaneous high-speed multi-parameter imaging technique such as the one demonstrated in this work. With the aim of clarifying this point, cross-sectional profiles of OH/CH2O concentrations, HRRs, and temperatures from an instantaneous image set are plotted together in Fig. 7. The cross-section at Y = 52 mm (i.e., Y/D = 6.5) was selected so that a clear zone of the cold core flow is included in the profiles. Another cross-section at Y = 96 mm (Y/D = 12), where the mixing layers already overlap with each other, is also displayed. For the case of Y/D = 6.5, it is observed that in the left layer the positions of the peaks for the flame temperature and HRR are offset by ~1 mm. This difference is mainly because of the local turbulent-flow convection. However, for the right layer, the peak of the temperature coincides with the HRR peak, which means that the local turbulence effect is relatively small. For the Y/D = 12 case, a broad temperature profile is observed because of turbulent mixing. Thus, a comparison of exact peak locations for the local flame temperatures and HRRs is not possible.
In addition, dynamic combustion events can be tracked in time and better understood by using multi-parameter measurements. Complex events, such as extinction, lead to both thermal and chemical effects which are difficult to understand simultaneously. Figure 8 shows a sequence of images that includes the progression of an extinction event. The upper and lower rows of images show simultaneous measurements of the thermal dissipation (∇T)2 [22] obtained from Rayleigh scattering images and CH2O-PLIF signal, respectively. A level-set threshold was calculated from the OH PLIF measurements and is overlaid as a white line on both sets of images. The OH-PLIF contour begins as a single connected element which starts to shrink and subsequently breaks at 200 μs. The top row of images represents a scaled heat-loss rate which begins to interact with the OH layer at 0 μs. At 100 μs, the thermal dissipation structure interacts with the inner portion of the OH contour, and thus with the reaction layer. By 200 μs, the thermal dissipation structure creates excessive heat loss and breaks the OH contour, leading to an extinction event. Further, the lower set of images in Fig. 8 shows the CH2O-PLIF signal. At 0 μs, a few pockets of CH2O are formed inside the OH layer which are not connected to the intense area of CH2O signal inside the jet. Thus, we observe that the thermal-dissipation structure is already beginning to quench reactions, as a larger quantity of the intermediate species CH2O is being formed. This localized CH2O region can be tracked in time and its intensity appears to decrease throughout the sequence of images.
Fig. 8 A sequence of three 10-kHz-rate thermal dissipation (∇T)2 (top row) and CH2O-PLIF (bottom row) images of a flame segment undergoing extinction. The solid white lines overlaid on the images represents a level-set threshold contour of the OH-PLIF signal.
4. Conclusions
In conclusion, simultaneous Rayleigh/CH2O-PLIF/OH-PLIF measurements at a rate of 10-kHz have been demonstrated for turbulent reacting-flow diagnostics using a three-legged burst-mode laser system. Such a diagnostic technique provides the capacity for simultaneous measurements of flow temperature, heat-release rate, and OH/CH2O concentration fields at high data acquisition rates. Our initial demonstration of the technique shows that such measurements could be very useful for the validation of combustion-flow modeling and to understand flame and flow interactions in turbulent flames. Our observations on the chemical and physical processes occurring during an extinction event would not have been possible without simultaneous multi-species and multi-parameter high-speed measurements. Although additional information can be extracted from the data set we have presented, that work is beyond the scope of this paper and will be presented in a future report.
Funding
Air Force Research Laboratory (AFRL) (FA8650-16-C-2725, FA8650-15-D-2518); Air Force Office of Scientific Research (AFOSR) LRIR (14RQ06COR).
Acknowledgments
The authors thank Drs. Josef J. Felver and Mikhail N. Slipchenko for their help during the experiment. Dr. Stephen Grib was funded under the National Research Council AFRL Science and Technology Postdoctoral Fellowship Program. High-speed Rayleigh scattering was originally developed under NASA SBIR (80NSSC17C0008). Approved for public release (# 88ABW-2018-1017).
References
1. B. Thurow, N. Jiang, and W. Lempert, “Review of ultra-high repetition rate laser diagnostics for fluid dynamic measurements,” Meas. Sci. Technol. 24(1), 012002 (2013). [CrossRef]
2. 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). [CrossRef]
3. N. Jiang and W. R. Lempert, “Ultrahigh-frame-rate nitric oxide planar laser-induced fluorescence imaging,” Opt. Lett. 33(19), 2236–2238 (2008). [CrossRef] [PubMed]
4. N. Jiang, R. A. Patton, W. R. Lempert, and J. A. Sutton, “Development of high-repetition rate CH PLIF imaging in turbulent non-premixed flames,” Proc. Combust. Inst. 33(1), 767–774 (2011). [CrossRef]
5. N. Jiang, M. Webster, W. R. Lempert, J. D. Miller, T. R. Meyer, C. B. Ivey, and P. M. Danehy, “MHz-rate nitric oxide planar laser-induced fluorescence imaging in a Mach 10 hypersonic wind tunnel,” Appl. Opt. 50(4), A20–A28 (2011). [CrossRef] [PubMed]
6. K. N. Gabet, R. A. Patton, N. Jiang, W. R. Lempert, and J. A. Sutton, “High-speed CH2O PLIF imaging in turbulent non-premixed flames using a pulse burst laser system,” Appl. Phys. B 106(3), 569–575 (2012). [CrossRef]
7. R. A. Patton, K. N. Gabet, N. Jiang, W. R. Lempert, and J. A. Sutton, “Multi-kHz temperature imaging in turbulent non-premixed flames using planar Rayleigh scattering,” Appl. Phys. B 108(2), 377–392 (2012). [CrossRef]
8. J. D. Miller, N. Jiang, M. N. Slipchenko, J. G. Mance, T. R. Meyer, S. Roy, and J. R. Gord, “Spatiotemporal analysis of turbulent jets enabled by 100-kHz, 100-ms burst-mode particle image velocimetry,” Exp. Fluids 57(12), 192 (2016). [CrossRef]
9. S. Beresh, S. Kearney, J. Wagner, D. Guildenbecher, J. Henfling, R. Spillers, B. Pruett, N. Jiang, M. Slipchenko, J. Mance, and S. Roy, “Pulse-burst PIV in a high-speed wind tunnel,” Meas. Sci. Technol. 26(9), 095305 (2015). [CrossRef]
10. N. Jiang, M. Nishihara, and W. R. Lempert, “Quantitative NO2 molecular tagging velocimetry at 500 kHz frame rate,” Appl. Phys. Lett. 97(22), 221103 (2010). [CrossRef]
11. R. A. Patton, K. N. Gabet, N. Jiang, W. R. Lempert, and J. A. Sutton, “Multi-kHz mixture fraction imaging in turbulent jets using planar Rayleigh scattering,” Appl. Phys. B 106(2), 457–471 (2012). [CrossRef]
12. Z. Wang, P. Stamatoglou, Z. Li, M. Aldén, and M. Richter, “Ultra-high-speed PLIF imaging for simultaneous visualization of multiple species in turbulent flames,” Opt. Express 25(24), 30214–30228 (2017). [CrossRef] [PubMed]
13. J. Kojima and Q.-V. Nguyen, “Laser pulse-stretching with multiple optical ring cavities,” Appl. Opt. 41(30), 6360–6370 (2002). [CrossRef] [PubMed]
14. M. Tanahashi, S. Murakami, G.-M. Choi, Y. Fukuchi, and T. Miyauchi, “Simultaneous CH-OH PLIF and stereoscopic PIV measurements of turbulent premixed flames,” Proc. Combust. Inst. 30(1), 1665–1672 (2005). [CrossRef]
15. S. Roy, N. Jiang, P. S. Hsu, T. Yi, M. N. Slipchenko, J. J. Felver, J. Estevadeordal, and J. R. Gord, “Development of a three-legged, high-speed, burst-mode laser system for simultaneous measurements of velocity and scalars in reacting flows,” Opt. Lett. 43(11), 2704–2707 (2018). [CrossRef] [PubMed]
16. V. Bergmann, W. Meier, D. Wolff, and W. Stricker, “Application of spontaneous Raman and Rayleigh scattering and 2D LIF for the characterization of a turbulent CH4/H2/N2 jet diffusion flame,” Appl. Phys. B 66(4), 489–502 (1998). [CrossRef]
17. H. N. Najm, P. H. Paul, C. J. Mueller, and P. S. Wyckoff, “On the adequacy of certain experimental observables as measurements of flame burning rate,” Combust. Flame 113(3), 312–332 (1998). [CrossRef]
18. B. O. Ayoola, R. Balachandran, J. H. Frank, E. Mastorakos, and C. F. Kaminski, “Spatially resolved heat release rate measurements in turbulent premixed flames,” Combust. Flame 144(1–2), 1–16 (2006). [CrossRef]
19. S. Roy, P. S. Hsu, N. Jiang, M. N. Slipchenko, and J. R. Gord, “100-kHz-rate gas-phase thermometry using 100-ps pulses from a burst-mode laser,” Opt. Lett. 40(21), 5125–5128 (2015). [CrossRef] [PubMed]
20. S. Kostka, S. Roy, P. J. Lakusta, T. R. Meyer, M. W. Renfro, J. R. Gord, and R. Branam, “Comparison of line-peak and line-scanning excitation in two-color laser-induced-fluorescence thermometry of OH,” Appl. Opt. 48(32), 6332–6343 (2009). [CrossRef] [PubMed]
21. P. S. Hsu, N. Jiang, A. K. Patnaik, V. Katta, S. Roy, and J. R. Gord, “All fiber-coupled OH planar laser-induced-fluorescence (OH-PLIF)-based two-dimensional thermometry,” Appl. Spectrosc. 72(4), 604–610 (2018). [CrossRef] [PubMed]
22. J. H. Frank and S. A. Kaiser, “High-resolution imaging of dissipative structures in a turbulent jet flame with laser Rayleigh scattering,” Exp. Fluids 44(2), 221–233 (2008). [CrossRef]
