Hidrokarbonların detonasyonu

Abstract

Yanıcı ve patlayıcı maddelerin üretilmesinde, taşınmasında, depolanmasında ve kullanılmasında, kullanılan araç ve emniyet elemanlarının tasarımında reaktantın tek ve çok fazlı sistemlerindeki kritik değerlerinin bilinmesi büyük önem taşır. Oluşan kazaların önlenmesinde yine bu kritik noktalar göz önünde bulundurularak yapılacak işlem basamakları tesbit edilebilir. Birinci bölümde teknik emniyet açısından detonasyonun tehlikesinin önemi, özellikleri ve eksplozyondan farkı vurgulanmıştır. İkinci bölümde tek fazın, yaygın olarak kullanılan basit C-J yaklaşımı ile stasyoner ve tek boyutlu kabul edilen şok dalgasının korunum denklemleri ve ses hızı kullanılarak, matematiksel modeli verilmiştir. Çok fazlı yapılarda ise tek boyutlu stasyoner dalga yaklaşımının tatmin edici sonuçlar vermediği tespit edildiğinden kütle ve ısı transferini iki boyutlu ya da zamana bağlı olarak hesaplayan yöntemler kullanılmaktadır. Bu tip bir problemin hem tanecik kopmasımn hem de difüzyonun karışım oluşumuna etkilerini göz önüne alarak oluşturulan bir çözümü ve bu çözümün dayandığı prensipler ayrıntılı olarak ikinci bölümde verilmiştir. İkinci bölümde ayrıca hücresel yapı ve dönen detonasyon konusunda geliştirilen teoriden de bahsedilmiştir. Üçüncü bölümde LPG/LNG bileşenlerinin detonasyonunu değişik açılarda inceleyen deneysel çalışmalar örneklenmiştir. Bu tip çalışmalarda basit ZND yaklaşımıyla kurulan matematiksel modeller yerine çok boyutlu hücresel yapı yaklaşımı tercih edilmiş ve bu konuda daha önce yapılan çalışmalar sonucu önerilen hücre boyutu kriterlerinden faydalanılmıştır. Ayrıca detonasyonun oluştuğu kap yüzeyinin, hidrokarbon yakıtın moleküler özelliklerinin ve derişikliğinin detonasyona etkilerini inceleyen farklı çalışmalardan da bahsedilmiştir. Tüm bu çalışmaların incelenmesinin sonucunda görülmüştür ki çoğu durum için tek boyutlu matematiksel model tatmin edici ve deneysel çalışmalara göre daha güvenilir sonuçlar vermektedir. Ancak yine de bu yaklaşımın yeterli olmadığı durumlar söz konusu olmaktadır ve bu durumlarda yan ampirik ifadeler kullanmak daha faydalıdır.

Gas detonation has attracted increased interest in the last decade because of technological problems, explosion hazards in mining and other problems in technology and power engineering, including nuclear. Thus at this study it is tried to prepare a monograph about gas detonation, in particular hydrocarbon detonation. Due to this aim an extent number of paper and book is reviewed and some examples of these studies are given and explained. Firstly, it is emphasized that differences between two phenomena; detonation and explosion which are generally used as synonyms by the members of the scientific and engineering communities, because of their common features such as external manifestations. Then initiation and propagation of detonation in hydrocarbon-air mixtures are explained simply and possible hazardous effects caused by detonation are shown. Then basic principles of theoretical treatment of detonation is given. The one dimensional ZND model is analyzed and derivation of basic expression used at this model are explained. Recently developed three dimensional spin detonation behavior theory and cellular structure is also mentioned. With a detailed analyze, a mathematical model study about heterogeneous detonation and some experimental studies about hydrocarbon detonation, researching the behavior of it under different conditions are given. And the results of both one dimensional theory and experimental studies are compared. Conclusions of this survey can be stated by following expressions: Although on the whole the one dimensional theory describes detonation wave behavior quite satisfactorily, detailed studies have shown that the front structure is always cellular. Most current studies envisage these structure and transverse waves. The cell size is a characteristic dimension used to scale the phenomena: reaction-zone sizes, detonation tube and free-charge diameters, channel dimensions, roughness and obstacles sizes, initiation energy-distribution zones, turbulent pulsating scale,... etc. For example, the transverse cell size, s, is an easily measurable parameter characterizing the non-stationary and three-dimensional detonation wave properties, and therefore it has been applied for characterizing of different limiting cases of detonation propagation. Knystautas et al. have recently shown that a combustion to detonation transition can occur only if 5 < d,where d is the tube diameter The above criterion was proposed by Shchelkin-not, however, for transition but for detonation limits. The argument behind the criterion was that the tube diameter must be at least large enough to accommodate one detonation cell. Shchelkin's criterion was supported by Vasiliev's experiments, where it was found that self sustaining propagation can only propagate in a flat channel of width h > s. Another suggestion has been made by Kogarko and Zeldovich, who proposed as early as 1948 that s=nd be used as a criterion for detonation limits in circular tubes. The limiting case of detonation propagation in circular tubes is spinning detonation. This usually appears in a tube with a diameter slightly larger than the cell size for the same mixture in a large tube. Although an actual detonation front structure in a gas is always in more than one dimensional (cellular or spin), the one dimensional theory gives good predictions not only for the speed but also for certain internal characteristics, in particular the peak in the mean pressure acting at the wall in the forward and reflected waves. There are some wave parameters where the predictions from the theory deviate from experiment, although some of them tend to approach the theoretical values as d/a increases, while other differ substantially from the predictions but their scales are correlated with the cell structures size, such as parameter profiles, effective front thickness and distance to the C-J surface. There are macrokinetic effects from the inhomogeneous structure; these are two types. On the one hand, the effective induction time is reduced, with the reaction starting directly at the front, while on the other, inhomogeinities lengthen the energy deposition zone, which is found to be larger by an order of magnitude that than envisaged in the one dimensional theory. Consequently, the parameter profiles deviate from the almost rectangular form corresponding to one dimensional theory and becomes triangular, even for high activation energies. Extremely detailed experiments have been performed on major inhomogeneous structure elements in stationary detonation wave, as well as on certain transient phenomena such as emergence from a tube into a large volume or direct initiation. The theoretical and numeric descriptions sometimes fit the observations well, but sometimes they lead us to desire better. Semi empirical models appear to be the most effective in some cases, such as for the initiation energy. One of these detailed studies about inhomogeneous structure elements in stationary detonation is given at this study for sampling. This is a theoretical and experimental study of the problems of deflagration and detonation structure in heterogeneous media, which contains an oxidant in the gaseous phase and fuel in the form of either dispersed droplets in the oxidant flow or a thin film on the chamber walls. Detonation in such systems is shown to have complex unsteady - state structure: the detonation front can exhibit mobile discontinuities and can pulsate periodically. A physical model of pulsating and spin detonation in heterogeneous media is XVU developed. A system of governing equation with boundary conditions is composed that makes it possible to simulate mathematically the transition of deflagration to detonation. The transition process and the initiation of the detonation are calculated numerically and studied experimentally. The comparison shows good agreement of theoretical and experimental results. According to this study; the internal structure of detonation in heterogeneous systems of gaseous oxidant condensed fuel type differs essentially from the structure of the familiar homogeneous detonation: A much more extended reaction zone sustaining the bow shock wave and in the presence in this zone of strong discontinuities (secondary shock waves ) generated by local explosions. This difference in the structure of the detonation zone is explained by the fact that at the initial instant the oxidant and fuel occur in different phases and are non mixed and therefore not only combustion of the fuel air mixture takes place behind the bow shock wave, but also the process of mixture formation. Inter phase mass transfer occurs as result of thermal mechanical influence of a gas flow behind a shock wave on the droplets and liquid layer on the inner tube surface. The shedding and atomization of the droplets of fuel followed by the evaporation and diffusion of vapors into the oxidant are responsible for the origin of a combustible gas mixture behind a leading wave. One of the main factors that determine the mode of detonation propagation in two phase systems is the process of fuel air mixture formation behind a bow shock wave. This type of detonation is studied with the aid of different models of mixture formation mechanisms. Among these are models of instantaneous shedding mixing and evaporation of fuel droplets over the entire flow cross section; the shedding of droplets from the liquid surface due to the formation of surface waves and phase interface instability; for relatively thick fuel films at small detonation rates the shedding of droplets is considered to be the main mechanisms of mass supply into the boundary layer, while at high (>1000 m /s) rates an initially combustible mixture is considered to be formed as a result of evaporation and diffusion. Among theoretical publications there are two types of considerations. One of them is purely shedding mechanisms of mixture formation, and the other one is the purely diffusion type. In this study both of them are taken in to account. The shedding and atomization of droplets occur some distance dawn the shock wave and lead to a sharp increase in the interfacial area. At the same time, the formation of the volumes of combustible mixture capable of exploding results from fuel evaporation and mixing of vapor with an oxidant. Also the processes of evaporation and combustion that takes place behind the bow shock wave before the origination of perturbations on the surface and start of shedding are essential. XVHl Thus on detonation in a heterogeneous medium containing dispersed droplets and a liquid film on the tube walls, phase transitions are possible: 1) Evaporation of fuel from the film surface and liquid droplets. 2) Shedding of fine droplets from the film surface and atomization of large droplets (it is convenient to consider the film and dispersed droplets from the film and dispersed droplets as different phases). 3) Deposition of large droplets on the tube surface, film formation and coalescence of small droplets into large ones. Heat liberation during combustion substantially increases the heat flux into the liquid layer and enhances evaporation. In many systems the rate of fuel mass supply into the boundary layer by means of evaporation and shedding of droplets exceeds the rate of combustion, and the fuel accumulates in the boundary layer farther and farther away from the bow shock wave. As result of this, at a distance of about 10 cm from the bow shock wave, the zone of possible detonation, 3 is formed where the conditions developed, as to the concentration limits and critical diameter, which are a prerequisite for detonation propagation of a homogeneous, proper detonation wave. Thus, detonation in two phase systems is represented by a complex which contains a bow shock wave, an extended zone of reaction with fuel evaporation and combustion and a proper detonation wave degenerating in to a secondary shock wave. The mode of two phase detonation propagation is determined by the net effect of flow on the bow shock wave for the period. The flow net effect is determined by the influence of the proper detonation wave (by means of the secondary shock wave and subsequent rarefaction ) and also by the influence of week perturbations, because of the presence of a boundary layer on the tube walls (due to friction, heat fluxes to the walls and fuel layer, evaporation and combustion of fuel ) and by the dispersed phase influence (shedding, atomization, acceleration droplets, warming up, evaporation and combustion. The self sustaining mode of detonation propagation in non mixed systems is determined by the equality of loses and supplies of energy to the bow shock wave for the period, such as by the net flow effect being equal to zero. The character of propagation of the proper detonation wave and of the associated secondary shock wave determines two different limiting modes of detonation propagation in non mixed two phased systems: pulsating and spin like. Current topics are stationary and nonstationary flame propagation in the speed range 10-103 m/s, which is a range intermediate for classical combustion and detonation, particularly in channels, encumbered spaces, and in free volumes; spin and cellular detonation wave stability in long tubes; detonation limits, in particular for spin and galloping modes; the effects of scale factors on detonation wave initiation and propagation in large explosions; and the physical and chemical processes in detonation wave interactions with suspended particles, obstacles, and walls, including explosion safety and engineering gas detonation applications. XDC For the marginal detonation of aliphatic natural gas constituents it is concluded that a shock front temperature of around 1200 K is a critical condition for detonation in mixtures of these aliphatic natural gas constituents with oxygen and nitrogen, while contained at ambient conditions in tubes or pipes, but that the justification for this remains incompletely understood and will be particularly complicated in fuel-rich soot producing mixtures The diminishing detonability sequence obtained from this study is ethane, n-propane, n-butane, methane. For predictive use, for instance in safety field, it may of course be better to use always the lowest Ts value of 1200 K. Allowing for experimental error in preparation, this will effectively bring virtually all measured fuel-lean limits within the Ts predicted detonation regime. For the fuel-lean limit of ethane, oxygen and 40% nitrogen it is still just within the mixture accuracy. This will have to be reviewed in the light of future results. Strong deflagrations similar to those recorded in more reactive mixtures and in the absence of obstacles can be induced by using short obstacles. The resulting deflagrations appear to be quasi-stable, displaying a continuos, initially weak, acceleration. These deflagrations have been found to serve as an excellent basis for transition to detonation and can easily be induced in mixtures with fuels of low reactivity, such as methane. By using longer obstacles it has been demonstrated that, compared with results obtained in smooth tube experiments, there are differences in ordering, as well as in magnitudes, of fuel sensitivity to DDT and flame acceleration in general. The differences in behavior indicate that considerable caution is required when applying hazard criteria based direct initiation work to an anticipated turbulent environment, as the relative hazard assessments will be in grave error or even irrlevant. Within the uncertainty of the data, there appears to be no effect of temperature between 25 °C and 100 °C on detonation cell with for sitoichiometric alkanes and alkenes. Similarly within the uncertainty of the data, all sitoichiometric n-alkane mixtures from ethane to decane have a similar likelihood of detonating. Molecular structure effects detonability, as shown for C hydrocarbons where the saturated ring structure is more sensitive than the straight-chain alkane which is more sensitive than the branched chain alkane. Unsaturated alkenes and alkynes are more sensitive to detonation than saturated alkanes. However, the degree of sensitization decreases with increase with increasing molecular weight and appears to be unchanged by increasing the number of unsaturated bonds. Addition of functional groups such as nitro, nitrate, epoxy, and ethers can significantly reduce the detonation cell width from the parent n-alkane, to the degree that nitrated n-alkanes can be as sensitive as hydrogen-air mixtures. For conclusion; 1- Most gaseous mixtures that will combust, will detonate under the right conditions. 2- The detonation wave travels at a very high speed. XX 3- Pressure in the detonating zone is very high and multiplying effects of impact and reflection can make a detonation more destructive than a normal explosion. 4- That detonations are not rare occurrences is demonstrated by the fact that as many as many as five in a single year period have been recognized. 5- Recognizing the above facts, the only means of operating a unit with complete safety is to avoid mixtures of oxygen and fuel gases. Processes which require mixtures of air and oil vapor to be present must be supplied with adequate controls and instrumentation to keep the mixtures outside the flammable range.