Ignition, Combustion, and Passivation of Nanopowders

Abstract

The nanopowders of metals are pyrophoric, i.e., they are capable of self-ignite contacting with the air because of high chemical activity and large specific surface. There are two methods of providing explosion safety of the process: a passive one, based on the systems of warning of undesirable changes in the process, and an active one, assuming the change in a technological chain, which leads to providing maximal safety of the process. In the chapter, the primary attention is paid to the second method in relation to processes of production, processing, transportation, and storage of metallic nanopowders. To make safe a further processing of nanopowders to products, the powders are passivated. The passivation means the creation of a protective thin oxide film on a surface of nanoparticles, which prevents self-ignition of metallic nanopowders. There have been no reliable, scientifically grounded methods of passivation of metallic nanopowders until now. The passivation takes tens of hours, which is a restrictive factor to get an increase in output of nanopowders. However, the theoretical models explaining ignition of ultra-disperse powders at room temperature are not still developed. These models should allow both revealing the main parameters influencing critical conditions of ignition and optimizing process of passivation of nanopowders. Application of the models developed for the description of ignition of metal particles, which take into account the interaction of particles with an oxidizer as well as diffusion of reagents through a product layer on the surface of the particles, is generally unacceptable. According to these models, ignition at temperatures close to the room ones is impossible because of low diffusivities. Therefore, development of the theory explaining pyrophoric behavior of metallic nanopowders and considering the coalescence of nanoparticles at temperature increase and the role of diffusion of an oxidizer into a sample is much urgent.

The experimental study of influence of conditions of passivation on the kinetics of formation of a protective oxide film on the surface of nanoparticles is carried out below. The influence of the factors limiting the diffusion of active components of the gas environment into a powder sample is investigated and recommendations for carrying out the process of passivation under optimal conditions are developed. Both physical and chemical properties and the morphology of the passivated powders are investigated by methods of electron microscopy, the chemical and X-ray phase analysis that will further allow establishing interrelation both between parameters and duration of passivation and the fine structure of passivated nanopowders.

The overwhelming number of publications on ignition and combustion of pyrophoric nanopowders is associated with combustion of aluminum nanopowders. From the known literature, it is possible to conclude that modeling and, respectively, understanding of the mechanism of combustion of Al nanopowders are still at an initial stage.

In case of oxidation in nanoscale, the process is kinetically controlled because of a small diffusive path. Heterogeneous oxidation through an oxide layer is under diffusion control due to slow diffusion of Al cations and oxygen anions. However, oxidation through oxygen from a gas phase can be either under kinetic control or under diffusion control depending on Damkohler number. For nanoparticles, the ignition stage always begins with heterogeneous reactions and phase transitions and is observed at rather low temperatures in comparison with the microscale. Because of fast heterogeneous reactions, nanoparticles are completely consumed in them before achievement of Al boiling temperature, thus the flame front (the wave mode) does not occur [1].

The models presented in the literature are based on two transfer mechanisms. According to the first one, a driving force of the process is either diffusion or a heterogeneous reaction velocity at the possible occurrence of a pressure difference in an Al core covered with an oxide layer, and oxygen out of the core. In the second case, a driving force of the process is the double electric layer at the phase boundary between the kernel and the oxide layer. Notice that at numerical calculation macroscopic parameters, namely diffusivities, thermal capacities, etc., always appear in spite of the fact that the research object is a nanoparticle.

Authors [2] used both a mass spectrometry method and transmission electron microscopy. The investigations showed emergence of hollow particles at Al nanopowder oxidation that points to the importance of the accounting of diffusion of Al atoms in the overall process. The experiments specify that oxidation proceeds in two regimes. Before melting of aluminum there is a slow oxidation caused by diffusion of oxygen through an aluminum oxide layer. Above melting point, fast oxidation takes place, at which both Al and oxygen diffuse through the oxide layer. It leads to an increase in the process velocity. The dependence between oxidation time and a particle radius is obtained. Various authors showed that for macro-particles the dependence is typical of the model of mobile oxide boundary [3, 4]. In other words, the difference between combustion of micro- and nanoparticles is determined by diffusion control for macro- and kinetic control for nanoparticles. With the reduction of Al and Zn particle size, their minimum energy of ignition sharply decreases [5].

For aluminum particles, the abnormal dependence of the flammability on diameter is found. Aluminum particles of 200 nm in diameter ignite more effective, than 100 nm ones. The authors suggested that it is connected with both the extent of Al particles passivation of aluminum, which differs for these two types of particles, and the extent of agglomeration, which is more pronounced for 100 nm particles [5]. It means that in the case of Al nanopowder, the passivation has to have a noticeable impact on flammability. We remind that passivation of metal nanoparticles is a research objective of the present chapter.

In [6], thermomechanical behavior of Al nanoparticles covered with a crystal or amorphous oxide layer during melting is studied by the method of molecular dynamics. In [7] Al nanopowder combustion with MoO₃ is analytically studied. The results showed that macroscopic flame propagation in the ensemble of small pyrophoric particles could be simulated as the laminar flame in mixed gas. A rather simple mechanism of combustion of Al particle is suggested in [8]. The model assumes two-stage combustion of a particle. At the first stage, there occur both a phase transition and heterogeneous reactions until the moment when the melting temperature of oxide is reached. At the second stage, the layer of oxide is missing and the steady diffusion flame occurs. It allows applying the standard assumptions of a flame front propagation. It is noteworthy that metal additives to jet fuels significantly intensify combustion and it makes the investigation of combustion of metallic nanopowders as Al, B, Mg, and Zr even more urgent [9, 10]. However, taking into account that the sizes of particles are in the range of micron and sometimes even less, there are numerous side effects, including ignition delays, slow speed of combustion, and incomplete combustion of comparatively large metal particles of the micron size [11].

The role of a double electric layer is considered in [12]. The mechanochemical behavior of the aluminum nanoparticles about 10 nm in diameter covered with oxide is examined by the molecular dynamics method. It is shown that the ignition mechanism for the aluminum particles covered with oxide could be determined by the built-in electric field in an oxide cell. It differs from the contemporary ideas of diffusion as a driving force, caused by the pressure difference at the cell border. In [13], Al nanopowder combustion modeling in the spherically symmetric case is carried out. In calculations, it is considered that Al particle of 10–50 nm is covered with thin oxide (1–4 nm) layer and surrounded with oxidizer. The nonlinear model of Cabrera–Motta [14] with a self-consistent potential is used to calculate oxidation reaction velocity as a function of temperature and size of the oxidized metal powder.

Thus, from the known literature, it is possible to conclude that modeling and, respectively, understanding of combustion of metallic nanopowders is still at an initial stage. This has not allowed creating up to now reliable, scientifically grounded methods of decrease in risk of emergence of technogenic accidents by production, processing, transportation, and storage of nanopowders at maximal preservation of their unique properties.

In this chapter, a model of the porous sample ignition is proposed, based on an assumption of a limiting role of the oxidizer diffusion in the ignition mechanism. It is shown that the ignition process can have a two-stage character. The duration of the stages is estimated by the methods of combustion theory. The applicability limits of the semi-infinite body model are determined. The role of a finite size of a sample in the ignition process is analyzed. The nonuniform quasi-two-dimensional mode of combustion of iron nanopowders and fingering patterns in combustion of nickel nanopowders in the absence of external flows is revealed for the first time. The method of estimation of the extent of passivation of Fe nanopowders with the use of a method of color high-speed filming is offered. It is experimentally established that both the dependencies of the period of a delay of ignition and quantity of the primary centers of combustion on the time of passivation can be used for estimation of the extent of passivation. On the basis of the experimental data for the certain sample, the approximate equation for estimation of the minimum time of complete passivation for the sample of arbitrary thickness is offered. By the method of X-ray phase analysis, it is established that 1 mm thick samples of iron nanopowder treated in a stream of 3% of air + Ar for the time interval more than 6 min contain only metallic iron. Therefore, the method of passivation suggested is rather effective. The effective means of stabilization of iron nanoparticles synthesized by the method of chemical metallurgy by means of passivation in argon stream + 0.6% O₂ within 6–60 min is offered.

It is established that at storage of iron nanopowder in a vessel equipped with the ground-in cover within 5 months in the ambient air any noticeable change of chemical composition of the powder was not observed. It was shown that nanoparticles form crystallites with a size ~20 –100 nm. The results of Auger spectroscopy method are consistent with the fact that nanoparticles of iron contain an iron kernel and an oxide layer 2–4 nm thick.

The average specific surface area of the passivated nanoparticles of iron determined by the BET method makes up ~9.2 m²/g and does not practically depend on the time of passivation.

The effective method of stabilization of iron nanoparticles synthesized by the method of chemical metallurgy by means of passivation in the dry air at subzero temperatures is offered for the first time. It is experimentally shown that at certain subzero temperature Fe nanoparticles do not ignite in dry air; however, passivation occurs and makes the particles stable at room temperature. It was shown that combustion modes at room temperature and subzero temperatures differ qualitatively. It is detected that both the content of oxides in the iron nanopowder sample after combustion and the maximum warming up decrease with a decrease in initial temperature. It was shown for the first time that the concepts of the classical macroscopic theory of a thermal explosion are quite applicable to nanoobjects.