It is known that unconfined dust explosions consist of a relatively weak primary (turbulent) deflagrations followed by a devastating secondary explosion. The secondary explosion may propagate with a speed of up to 1000 m/s producing overpressures of over 8-10 atm. Since detonation is the only established theory that allows a rapid burning producing a high pressure that can be sustained in open areas, the generally accepted view was that the mechanism explaining the high rate of combustion in dust explosions is deflagration to detonation transition. In the present work we propose a theoretical substantiation of the alternative propagation mechanism explaining origin of the secondary explosion producing the high speeds of combustion and high overpressures in unconfined dust explosions. We show that clustering of dust particles in a turbulent flow gives rise to a significant increase of the thermal radiation absorption length ahead of the advancing flame front. This effect ensures that clusters of dust particles are exposed to and heated by the radiation from hot combustion products of large gaseous explosions sufficiently long time to become multi-point ignition kernels in a large volume ahead of the advancing flame front. The ignition times of fuel-air mixture by the radiatively heated clusters of particles is considerably reduced compared to the ignition time by the isolated particle. The radiation-induced multi-point ignitions of a large volume of fuel-air ahead of the primary flame efficiently increase the total flame area, giving rise to the secondary explosion, which results in high rates of combustion and overpressures required to account for the observed level of overpressures and damages in unconfined dust explosions, such as e.g. the 2005 Buncefield explosion and several vapor cloud explosions of severity similar to that of the Buncefield incident.

It is known that unconfined dust explosions typically starts off with a relatively weak primary flame followed by a severe secondary explosion. We show that clustering of dust particles in a temperature stratified turbulent flow ahead of the primary flame may give rise to a significant increase in the radiation penetration length. These particle clusters, even far ahead of the flame, are sufficiently exposed and heated by the radiation from the flame to become ignition kernels capable to ignite a large volume of fuel-air mixtures. This efficiently increases the total flame surface area and the effective combustion speed, defined as the rate of reactant consumption of a given volume. We show that this mechanism explains the high rate of combustion and overpressures required to account for the observed level of damage in unconfined dust explosions, e.g., at the 2005 Buncefield vapor-cloud explosion. The effect of the strong increase of radiation transparency due to turbulent clustering of particles goes beyond the state-of-the-art of the application to dust explosions and has many implications in atmospheric physics and astrophysics.

Condensation of water vapor on active cloud condensation nuclei produces micron-size water droplets. To form rain, they must grow rapidly into at least 50-100 $\mu$m droplets. Observations show that this process takes only 15-20 minutes. The unexplained physical mechanism of such fast growth, is crucial for understanding and modeling of rain, and known as "condensation-coalescence bottleneck in rain formation". We show that the recently discovered phenomenon of the tangling clustering instability of small droplets in temperature-stratified turbulence (Phys. Fluids 25, 085104, 2013) results in the formation of droplet clusters with drastically increased droplet number densities. The mechanism of the tangling clustering instability is much more effective than the previously considered by us the inertial clustering instability caused by the centrifugal effect of turbulent vortices. This is the reason of strong enhancement of the collision-coalescence rate inside the clusters. The mean-field theory of the droplet growth developed in this study can be useful for explanation of the observed fast growth of cloud droplets in warm clouds from the initial 1 micron size droplets to 40-50 micron size droplets within 15-20 minutes.

We study turbulent diffusion of chemically reacting gaseous admixtures in a developed turbulence. In our previous study [Phys. Rev. Lett. {\bf 80}, 69 (1998)] using a path-integral approach for a delta-correlated in time random velocity field, we demonstrated a strong modification of turbulent transport in fluid flows with chemical reactions or phase transitions. In the present study we use the spectral tau approximation, that is valid for large Reynolds and Peclet numbers, and show that turbulent diffusion of the reacting species can be strongly depleted by a large factor that is the ratio of turbulent and chemical times (turbulent Damk\"{o}hler number). We have demonstrated that the derived theoretical dependence of turbulent diffusion coefficient versus the turbulent Damk\"{o}hler number is in a good agreement with that obtained previously in the numerical modelling of a reactive front propagating in a turbulent flow and described by the Kolmogorov-Petrovskii-Piskunov-Fisher equation. We have found that turbulent cross-effects, e.g., turbulent mutual diffusion of gaseous admixtures and turbulent Dufour-effect of the chemically reacting gaseous admixtures, are less sensitive to the values of stoichiometric coefficients. The mechanisms of the turbulent cross-effects are different from the molecular cross effects known in irreversible thermodynamics. In a fully developed turbulence and at large Peclet numbers the turbulent cross-effects are much larger than the molecular ones. The obtained results are applicable also to heterogeneous phase transitions.

We study particle clustering in a temperature stratified turbulence with small finite correlation time. It is shown that the temperature stratified turbulence strongly increases the degree of compressibility of particle velocity field. This results in the strong decrease of the threshold for the excitation of the tangling clustering instability even for small particles. The tangling clustering instability in the temperature stratified turbulence is essentially different from the inertial clustering instability that occurs in non-stratified isotropic and homogeneous turbulence. While the inertial clustering instability is caused by the centrifugal effect of the turbulent eddies, the mechanism of the tangling clustering instability is related to the temperature fluctuations generated by the tangling of the mean temperature gradient by the velocity fluctuations. Temperature fluctuations produce pressure fluctuations and cause particle clustering in regions with increased pressure fluctuations. It is shown that the growth rate of the tangling clustering instability is much larger than that of the inertial clustering instability. It is found that depending on the parameters of the turbulence and the mean temperature gradient there is a preferential particle size at which the particle clustering due to the tangling clustering instability is more effective. The particle number density inside the cluster after the saturation of this instability can be in several orders of magnitude larger than the mean particle number density. It is also demonstrated that the evaporation of droplets drastically change the tangling clustering instability, e.g., it increases the instability threshold in the droplet radius. The tangling clustering instability is of a great importance, e.g., in atmospheric turbulence with temperature inversions.