Characterizing turbulence within a premixed turbulent flame brush: separating intrinsic turbulence from reactant-product intermittency and mean-velocity-jump effects

Can the turbulence inside a premixed turbulent flame brush be characterized by a well-defined quantity? The classical root-mean-square turbulent velocity $u'$ that drives virtually all burning-rate models is contaminated inside the brush by the mean velocity jump that thermal expansion imposes across flame fronts, while statistics conditioned on reactants or products are contaminated by reactant-product intermittency; model problems and dedicated constant-density direct numerical simulations (DNS) show that neither yields the intrinsic turbulence the flame actually interacts with. The 2009 Combustion and Flame paper by A. N. Lipatnikov that posed this question called for (1) quantitative regime maps of the intermittency and jump contributions to conditioned Reynolds stresses, (2) constant-density DNS of flame-front propagation in homogeneous isotropic turbulence, (3) closures of the conditioned balance equations in variable-density flames, and (4) a practical prescription for the turbulence input to premixed-combustion models. Later literature has substantially executed (2), with a negative verdict on all second-moment characterizations and a pointer toward velocity-gradient-based quantities, and has developed new diagnostics (conditioned structure functions; Helmholtz-Hodge decomposition), but as of 2026 no validated characterization, no consensus regime map or importance criterion, and no closure or model prescription exists: the problem remains open.

Background

Premixed turbulent combustion - the burning of a homogeneous fuel-air mixture in a turbulent flow - is the working principle of spark-ignition engines, stationary gas engines, and gas-turbine combustors, and its predictive modeling is a central goal of combustion science. In the flamelet regime the reaction is confined to thin sheets (flamelets) that wrinkle under turbulence; the ensemble of these sheets occupies a finite region, the mean flame brush, statistically described by the progress variable $c$ ($c=0$ in reactants, $c=1$ in products). A burning-rate model - a Reynolds-averaged (RANS) or large-eddy-simulation (LES) closure of the turbulent burning velocity $S_T$ - needs a turbulence input, and the vast majority of such models, for example flame-speed-closure-type approaches (Lipatnikov and Chomiak, 2002), take the root-mean-square (rms) turbulent velocity $u'$ as the primary characteristic controlling how turbulence accelerates the flame (Sabelnikov and Lipatnikov, 2017).

Defining $u'$ inside a flame brush is nevertheless not straightforward. Because heat release expands the gas (the density ratio between reactants and products is typically 5-7), the mean velocity conditioned on products exceeds that conditioned on reactants by a slip velocity of order $(\sigma-1)S_L$, where $S_L$ is the laminar flame speed. The standard (Reynolds- or Favre-averaged) second moment of the velocity then contains a jump term proportional to $\tilde c(1-\tilde c)(\bar u_{k,b}-\bar u_{k,u})^2$ that scales as $(\sigma S_L)^2$ and has nothing to do with turbulent eddies. This follows from the classical statistical picture of a flame brush, developed in the 1970s, as an intermittent mixture of reactant and product lumps within one statistical ensemble. Experiments therefore often report turbulence characteristics conditioned on the reactant stream (Cheng, 1984), and balance equations for conditioned Reynolds stresses were derived and assessed, e.g., in the zone-conditional DNS study of flame-generated turbulence by Im et al. (2004) and in the conditionally averaged balance equations of Lipatnikov (2008).

In the 2009 source paper, Lipatnikov asked whether any of these quantities actually characterizes turbulence in a flame brush. The difficulty is twofold. First, unconditioned moments are contaminated by the thermal-expansion velocity jump. Second, a conditional average is taken over a subensemble (reactants at a given point and time) whose composition itself fluctuates randomly with the flame-sheet motion - the classical intermittency problem of turbulent flows with randomly moving interfaces. By analyzing simple model problems, the paper showed that even for a passively self-propagating interface in a constant-density flow, where the true turbulence is known and unaffected by the interface, the conditioned second moments differ from the true ones; the magnitudes were estimated from the conditioned balance equations. The paper thus posed a research program: quantify the intermittency and jump contributions across flow and flame regimes (parameters $u'/S_L$, $L/\delta_L$ with $L$ the integral turbulence length scale and $\delta_L$ the laminar flame thickness, the density ratio $\sigma$, and flame development time); perform dedicated constant-density DNS of flame propagation in homogeneous isotropic turbulence, where the true turbulence is known, to measure the effects exactly; extend the analysis to variable-density reacting flow by closing the conditioned balance equations; and on that basis formulate a practical prescription that supplies combustion models with the turbulence input at controlled accuracy, including guidance on when upstream or nonreacting $u'$ values are acceptable proxies.

Subsequent work sharpened the question without settling it. Constant-density DNS of self-propagating fronts in homogeneous isotropic turbulence (Yu, Lipatnikov and Bai, 2014; Yu, Bai and Lipatnikov, 2015) confirmed that conditioned velocity moments deviate from the true turbulence, while conditioned enstrophy (the squared vorticity magnitude) and strain statistics remain close to their means - pointing to velocity-gradient-based quantities as better turbulence characteristics. The authors' 2017 review (Sabelnikov and Lipatnikov) concluded that neither mean nor conditioned second moments of the velocity field properly characterize turbulence within a premixed turbulent flame brush, that no model can yet predict the conditioned second moments in the flamelet regime, and that burning-rate models using the classical $u'$ as the primary input should be revisited. New diagnostics were developed - conditioned structure functions (two-point velocity-difference statistics conditioned on the mixture states at both points) and Helmholtz-Hodge decomposition of the velocity into solenoidal (rotational) and potential (dilatational) parts - and revealed that combustion-induced pressure perturbations significantly modify the incoming reactant turbulence even at high Karlovitz number $Ka$ (the ratio of the flame chemical time to the smallest-eddy turnover time). The adjacent, long-debated issues of countergradient turbulent scalar transport and flame-generated turbulence are reviewed in Lipatnikov and Chomiak (2010).

Problem Statement

Determine whether - and by what quantity - turbulence within a premixed turbulent flame brush can be characterized, and determine how the resulting characterization should supply the turbulence input to premixed-combustion models in place of the classical rms velocity $u'$. The core open question posed by the source (Lipatnikov, 2009) and still unresolved is:

(i) Identify turbulence characteristics that are well defined inside a flame brush and are not artifacts of reactant-product intermittency or of the thermal-expansion mean-velocity jump across flamelets. The standard unconditioned rms velocity $u'$ is contaminated by the jump term, and the reactant- or product-conditioned second moments are contaminated by intermittency (model problems and constant-density DNS show both deviate from the true turbulence), so the open question is whether any such characteristic exists - velocity-gradient statistics such as enstrophy (the squared vorticity magnitude) or total strain, solenoidal/potential-decomposed statistics, or conditioned structure functions (two-point velocity-difference statistics conditioned on the mixture state at each point) - or whether no well-defined one does.

Related open sub-directions recorded by the source's authors and the citing literature - not conjunctive acceptance conditions - include:

(ii) Quantify, across flow and flame regimes - from the wrinkled/corrugated-flamelet regimes (the classical premixed-combustion regimes in which the reaction zone remains a thin sheet, wrinkled or corrugated by eddies too large to enter it) to the thin-reaction-zones regime (penetrated by small eddies that do not disrupt the inner reaction zone), parameterized by $u'/S_L$, $L/\delta_L$, density ratio $\sigma$, Karlovitz number $Ka$ and Damköhler number $Da$ (dimensionless numbers comparing the flame chemical time scale with turbulent time scales) and flame development time; here $S_L$ is the laminar flame speed, $L$ the integral turbulence length scale, and $\delta_L$ the laminar flame thickness - the magnitude of the intermittency and mean-velocity-jump contributions to the conditioned Reynolds stresses, carried by the unclosed terms (conditional correlations of velocity, pressure-gradient, and flamelet-surface quantities) of the conditioned balance equations, and thereby map the parameter ranges in which the classical $u'$ measured in the reactants or upstream of the flame is an acceptable proxy for the turbulence interacting with the flame - and where it fails.

(iii) Close the conditioned balance equations for the Reynolds stresses (whose unclosed terms are conditional correlations of velocity, pressure-gradient, and flamelet-surface quantities) in variable-density flame brushes with heat release, so that the conditioned or replacement turbulence characteristics can be predicted rather than only measured.

(iv) On this basis, formulate and validate a practical closure prescription that supplies premixed-combustion models (Reynolds-averaged or large-eddy-simulation closures of the turbulent burning rate, which currently take $u'$ as their primary turbulence input) with the appropriate turbulence characteristic at controlled accuracy, including explicit guidance on when upstream or nonreacting $u'$ values are acceptable proxies.

The constant-density DNS measurement sub-direction of this program (source item 2) has already been substantially executed (see Background); the characterization question itself remains open.

References

1. A. N. Lipatnikov, 'Can we characterize turbulence in premixed flames?', Combustion and Flame 156(6):1242-1247, 2009. DOI: 10.1016/j.combustflame.2009.01.002, https://doi.org/10.1016/j.combustflame.2009.01.002 (source paper)
2. A. N. Lipatnikov, 'Conditionally averaged balance equations for modeling premixed turbulent combustion in flamelet regime', Combustion and Flame 152(4):529-547, 2008. DOI: 10.1016/j.combustflame.2007.11.004, https://doi.org/10.1016/j.combustflame.2007.11.004
3. R. K. Cheng, 'Conditional Sampling of Turbulence Intensities and Reynolds Stress in Premixed Turbulent Flames', Combustion Science and Technology 41(3-4):109-142, 1984. DOI: 10.1080/00102208408923826, https://doi.org/10.1080/00102208408923826
4. Y.-H. Im, K. Y. Huh, S. Nishiki, T. Hasegawa, 'Zone conditional assessment of flame-generated turbulence with DNS database of a turbulent premixed flame', Combustion and Flame 137(4):478-488, 2004. DOI: 10.1016/j.combustflame.2004.03.006, https://doi.org/10.1016/j.combustflame.2004.03.006
5. A. N. Lipatnikov, J. Chomiak, 'Turbulent flame speed and thickness: phenomenology, evaluation, and application in multi-dimensional simulations', Progress in Energy and Combustion Science 28(1):1-74, 2002. DOI: 10.1016/S0360-1285(01)00007-7, https://doi.org/10.1016/S0360-1285(01)00007-7
6. A. N. Lipatnikov, J. Chomiak, 'Effects of premixed flames on turbulence and turbulent scalar transport', Progress in Energy and Combustion Science 36(1):1-102, 2010. DOI: 10.1016/j.pecs.2009.07.001, https://doi.org/10.1016/j.pecs.2009.07.001
7. A. Lipatnikov, 'Conditioned moments in premixed turbulent reacting flows', Proceedings of the Combustion Institute 33(1):1489-1496, 2011. DOI: 10.1016/j.proci.2010.06.138, https://doi.org/10.1016/j.proci.2010.06.138
8. A. Lipatnikov, 'A test of conditioned balance equation approach', Proceedings of the Combustion Institute 33(1):1497-1504, 2011. DOI: 10.1016/j.proci.2010.07.012, https://doi.org/10.1016/j.proci.2010.07.012
9. R. Yu, A. N. Lipatnikov, X. S. Bai, 'Three-dimensional direct numerical simulation study of conditioned moments associated with front propagation in turbulent flows', Physics of Fluids 26(8):085104, 2014. DOI: 10.1063/1.4891735, https://doi.org/10.1063/1.4891735
10. R. Yu, X.-S. Bai, A. N. Lipatnikov, 'A direct numerical simulation study of interface propagation in homogeneous turbulence', Journal of Fluid Mechanics 772:127-164, 2015. DOI: 10.1017/jfm.2015.211, https://doi.org/10.1017/jfm.2015.211
11. V. A. Sabelnikov, A. N. Lipatnikov, 'Recent Advances in Understanding of Thermal Expansion Effects in Premixed Turbulent Flames', Annual Review of Fluid Mechanics 49:91-117, 2017. DOI: 10.1146/annurev-fluid-010816-060104, https://doi.org/10.1146/annurev-fluid-010816-060104
12. V. A. Sabelnikov, A. N. Lipatnikov, S. Nishiki, T. Hasegawa, 'Investigation of the influence of combustion-induced thermal expansion on two-point turbulence statistics using conditioned structure functions', Journal of Fluid Mechanics 867:45-76, 2019. DOI: 10.1017/jfm.2019.128, https://doi.org/10.1017/jfm.2019.128
13. V. A. Sabelnikov, A. N. Lipatnikov, S. Nishiki, T. Hasegawa, 'Application of conditioned structure functions to exploring influence of premixed combustion on two-point turbulence statistics', Proceedings of the Combustion Institute 37(2):2433-2441, 2019. DOI: 10.1016/j.proci.2018.08.029, https://doi.org/10.1016/j.proci.2018.08.029
14. P. Brearley, U. Ahmed, N. Chakraborty, A. Lipatnikov, 'Statistical behaviors of conditioned two-point second-order structure functions in turbulent premixed flames in different combustion regimes', Physics of Fluids 31(11):115109, 2019. DOI: 10.1063/1.5124143, https://doi.org/10.1063/1.5124143
15. V. A. Sabelnikov, A. N. Lipatnikov, N. V. Nikitin, S. Nishiki, T. Hasegawa, 'Solenoidal and potential velocity fields in weakly turbulent premixed flames', Proceedings of the Combustion Institute 38(2):3087-3095, 2021. DOI: 10.1016/j.proci.2020.09.016, https://doi.org/10.1016/j.proci.2020.09.016 (postprint: arXiv:2007.10127)
16. A. N. Lipatnikov, V. A. Sabelnikov, N. V. Nikitin, S. Nishiki, T. Hasegawa, 'Influence of Thermal Expansion on Potential and Rotational Components of Turbulent Velocity Field Within and Upstream of Premixed Flame Brush', Flow, Turbulence and Combustion 106(4):1111-1124, 2021. DOI: 10.1007/s10494-020-00131-3, https://doi.org/10.1007/s10494-020-00131-3
17. V. A. Sabelnikov, A. N. Lipatnikov, N. V. Nikitin, F. E. Hernandez-Perez, H. G. Im, 'Conditioned structure functions in turbulent hydrogen/air flames', Physics of Fluids 34(8):085103, 2022. DOI: 10.1063/5.0096509, https://doi.org/10.1063/5.0096509
18. V. A. Sabelnikov, A. N. Lipatnikov, N. V. Nikitin, F. E. Hernandez-Perez, H. G. Im, 'Effects of thermal expansion on moderately intense turbulence in premixed flames', Physics of Fluids 34(11):115127, 2022. DOI: 10.1063/5.0123211, https://doi.org/10.1063/5.0123211
19. A. Lipatnikov, V. A. Sabelnikov, N. V. Nikitin, 'Opposite effects of flame-generated potential and solenoidal velocity fluctuations on flame surface area in moderately intense turbulence', Proceedings of the Combustion Institute 40(1-4):105238, 2024. DOI: 10.1016/j.proci.2024.105238, https://doi.org/10.1016/j.proci.2024.105238
20. V. A. Sabelnikov, A. N. Lipatnikov, 'A Priori Study of Inter-Scale Kinetic Energy Transfer and Energy Exchange in a Turbulent Premixed Flame', Energies 19(3):822, 2026. DOI: 10.3390/en19030822, https://doi.org/10.3390/en19030822
21. J. Lee, J. F. MacArt, M. E. Mueller, 'Heat release effects on the Reynolds stress budgets in turbulent premixed jet flames at low and high Karlovitz numbers', Combustion and Flame 216:1-8, 2020. DOI: 10.1016/j.combustflame.2020.02.014, https://doi.org/10.1016/j.combustflame.2020.02.014
22. A. Kazbekov, A. M. Steinberg, 'Physical space analysis of cross-scale turbulent kinetic energy transfer in premixed swirl flames', Combustion and Flame 229:111403, 2021. DOI: 10.1016/j.combustflame.2021.111403, https://doi.org/10.1016/j.combustflame.2021.111403