Search

International audience; The maximum explosion overpressure and the maximum rate of pressure rise, which characterize the dust explosion severity, are commonly measured in apparatuses and under specific conditions defined by international standards. However, those standards conditions, designed for micropowders, may not be fully adapted to nanoparticles. Investigations were conducted on different nanopowders (nanocellulose, carbon black, aluminum) to illustrate their specific behaviors and highlight the potential inadequacy of the standards. The influence of the sample preparation was explored. Various testing procedures were compared, focusing on the dust cloud turbulence and homogeneity. Dust dispersion experiments evidenced the importance of the characterization of the dust cloud after dispersion, due to the fragmentation of agglomerates, using metrics relevant with nanoparticles reactivity (e.g. surface diameter instead of volume diameter). Moreover, the overdriving phenomenon (when the experimental results become dependent of the ignition energy), already identified for micropowders, can be exacerbated for nanoparticles due to their low minimum ignition energy and to the high energy used under standard conditions. It was evidenced that for highly sensitive nanopowders, pre-ignition phenomenon can occur. Finally, during severe explosions and due to a too long opening delay of the ‘fast acting valve’, the flame can go back to the dust container.

This paper addresses a need for information on nanoparticle emissions and related issues such as worker exposure, filtration efficiency, and dustiness. A survey has been conducted on the working conditions and safety measures used in thermal spray companies and the results compared to scientific literature and previous surveys. Responses to questions on matters of health and safety reveal a lack of information and awareness of the risks posed by the emissions of ultrafine particles generated by thermal spraying processes.

Due to their low sedimentation rate, nano-objects offer the opportunity to study flame propagation at low turbulence. The burning velocity was then estimated by flame visualization in two apparatuses: a vertical 1 m long tube with a square cross-section and a 20 L sphere equipped with visualization windows and a vent. This works aims to study the laminar burning velocity of nanocellulose by a direct visualization of the flame propagation within these devices. A high-speed video camera was used to record the flame propagation, and an estimation of the unstretched burning velocity was obtained through linear and nonlinear relationships relating the flame stretching and the flame velocities. Although these methods were initially established for gases, the organic nature of nanocellulose implies a fast devolatilization, which makes the application of the methods possible in this work. Similar results were obtained in both apparatuses in different turbulence conditions, proving the laminar burning velocity was approached. The laminar burning velocity for the nanocellulose was determined to be 21 cm s−1. This value, estimated through flame propagation visualization, was then compared to the value calculated by applying a semi-empiric correlation to the pressure-time evolution recorded during standard explosion tests in the 20 L vessel.

International audience; To avoid the influence of external parameters, such as the vessel volume or the initial turbulence, the explosion severity should be determined from intrinsic properties of the fuel-air mixture. Therefore, the flame propagation of gaseous mixtures is often studied in order to estimate their laminar burning velocity, which is both independent of external factors and a useful input for CFD simulation. Experimentally, this parameter is difficult to evaluate when it comes to dust explosion, due to the inherent turbulence during the dispersion of the cloud. However, the low inertia of nanoparticles allows performing tests at very low turbulence without sedimentation. Knowledge on flame propagation concerning nanoparticles may then be modelled and, under certain conditions, extrapolated to microparticles, for which an experimental measurement is a delicate task. This work focuses on a nanocellulose with primary fiber dimensions of 3 nm width and 70 nm length. A one-dimensional model was developed to estimate the flame velocity of a nanocellulose explosion, based on an existing model already validated for hybrid mixtures of gas and carbonaceous nanopowders similar to soot. Assuming the fast devolatilization of organic nanopowders, the chemical reactions considered are limited to the combustion of the pyrolysis gases. The finite volume method was used to solve the mass and energy balances equations and mass reactions rates constituting the numerical system. Finally, the radiative heat transfer was also considered, highlighting the influence of the total surface area of the particles on the thermal radiation. Flame velocities of nanocellulose from 17.5 to 20.8 cm/s were obtained numerically depending on the radiative heat transfer, which proves a good agreement with the values around 21 cm/s measured experimentally by flame visualization and allows the validation of the model for nanoparticles.

This work presents an overview about the explosion behaviour of metallic powders from micron to nanosize. Aluminium, magnesium, titanium, iron and zinc were considered and their explosion safety parameters were analysed as a function of their mean primary particle size either determined by BET measurements, particle size distribution. To depict the course of explosion behaviour for these metals, extensive literature review has been performed and additional experimental tests were also performed. Generally, decreasing the particle size in a metallic powder leads to a higher explosion severity. It appears that this statement is true till a critical diameter below which the explosion severity (pmax, dp/dtmax) decreases for all the considered powders. This critical size can be explained by theoretical considerations on the nature of thermal transfer in the flame, namely by analysing the Cassel model. Finally, semi-empirical models were also developed for aluminium to highlight the specific micrometre and nanometre behaviour and the influence of turbulence, particle burning time, diameter and concentration. The influence of these key parameters needs to be further assessed in a future work in order to better understand the mechanisms involved and to extend the scope to other powdered materials.

The evaluation of aerosol processes which alter the size distribution of accidentally released nanomaterial particles in an indoor environment can provide size resolved exposure estimations, and subsequently contribute to a more comprehensive risk analysis on engineered nanomaterials (ENMs). In this work processed suspensions of TiO2 and SiO2 nanopowders were introduced into the reservoir of a nebuliser, and injected continuously as nano-aerosols into a room-size dispersion facility for a fixed release time. Following injection, the concentration of the dispersed aerosols was allowed to naturally decay for a prolonged period. Airborne particle number concentration (PNC) and particle size distributions (PSD) were continuously measured at a point within the breathing zone near to the source, while deposited particles were collected for transmission electron microscopy (TEM) analysis. A log10-normal fitting program was used to determine the evolution of the modal groups present within the measured PSDs. A modelling approach that considered the experimentally determined particle decay rate as a sum of the pair to pair coagulation and deposition rates was employed to estimate the relative importance of size-resolved deposition compared to coagulation. Results indicated that the variation of PNC with time was accurately modelled, providing size-dependent insights into the contribution of the two particle removal mechanisms to the change of PNC over time. The deposition patterns obtained from the TEM images qualitatively supported the model results. Since a limited set of input parameters were used, we concluded that the proposed model could be an effective tool for a reasonable quantification of worker's exposure to aerosols originating from the accidental release of diffuse ENM in real workplaces.

Among the factors influencing dust explosion, the particle size distribution (PSD) is both one of the most important and complex to consider. For instance, it is commonly accepted that the explosion sensitivity increases when the particle size decreases. Such an assertion may be questionable for nano-objects which easily agglomerate. However, agglomerates can be broken during the dispersion process. Correlating the explosion parameters to the actual PSD of a dust cloud at the moment of the ignition becomes then essential. The effects of the moisture content and sieving were investigated on a nanocellulose powder and the impact of a mechanical agglomeration was evaluated using a silicon coated by carbon powder. Each sample was characterized before and after dispersion using in situ laser particle size measurement and a fast mobility particle sizer, and explosion and minimum ignition energy tests were conducted respectively in a 20 L sphere and in a modified Hartmann tube. It was observed that drying and/or sieving the nanocellulose mainly led to variations in terms of ignition sensitivity but only slightly modified the explosion severity. In contrast, the mechanical agglomeration of the silicon coated by carbon led to a great decrease in terms of ignition sensitivity, with a minimum ignition energy varying from 5 mJ for the raw powder to more than 1J for the agglomerated samples. The maximum rate of pressure rise also decreased due to modifications in the reaction kinetics, inducing a transition from St2 class to St1 class when agglomerating the dust.

Industrial requirements present some unique challenges that can be met only by the application of engineered nanomaterials (ENMs). The completion of risk assessments based on the knowledge of the change in the explosion severity of turbulent gas/air mixtures caused by the accidental dispersion of nanopowders is vital before integrating such materials into existing systems. In this work, known amounts of selected carbon black nanopowders were mixed with methane near the lower-flammability-limit (LFL) to form hybrid mixtures of variable dust concentration. Mixtures were ignited in a 23 L cylindrical combustion vessel that allowed the control of isotropic turbulence through specially designed fans. The particle size distribution, the explosion pressure history and the flame speed derived from high speed Schlieren cine photographs, were measured. The influence of dispersed amounts of nanopowders on explosion severity was investigated by comparing the results with those obtained for pure methane-air explosions. Results indicated that the release of a relatively low mass of nanopowder into methane-air mixtures resulted in a more severe explosion than that of a higher amount. Also, despite the very low content of volatiles in the selected nanopowders, a hybrid mixture can be ignitable at an equivalence ratio well below the LFL of the gas. However this ignitability was shown to be related to the agglomeration state of powder in the dispersion phase, indicating that as the mean particle size decreases, a hybrid mixture with an extremely low content of flammable gas could be ignitable. From a risk assessment point of view, these results may have great significance for specific industrial processes aiming to utilise ENMs.

This work aims to study the influence of the dispersion conditions in a standard 20L sphere on the explosibility of a nanopowder. Even more than for micropowders, the dispersion conditions have a strong impact on the dust cloud homogeneity and its particle size distribution (PSD). Due to their high surface energy, nanoparticles are prone to agglomeration, but such structures can be broken during the dispersion process. Varying the dispersion procedure, the ignition delay time and even the nozzle type (rebound or symmetric) leads to modifications of the dust specific surface area, and thus of its reactivity. Tests were performed on aluminum and carbon black nanopowders. They were characterized before and during their dispersion in the sphere, notably using in situ laser PSD measurement. The initial turbulence level of the dust cloud was determined by particle image velocimetry. A lower shear stress is exerted on the agglomerates by using the symmetric nozzle, which causes less fragmentation and decreases the explosion severity. However, the results obtained with this nozzle are more reproducible than with the rebound nozzle. Because pre-ignition phenomenon was observed for aluminum nanopowders during their injection through the electrovalve, the dust dispersion by dust lifting was experimented. With regard to the standard procedure, the maximum rate of pressure rise decreased by approximately 30% whereas the maximum overpressure remains nearly unchanged. It means that the same amount of dust reacts in both kinds of experiments but that less fragmentation occurs, which is confirmed by PSD measurements. This work brings some questions about the interpretation of the explosivity results related to nanopowders and highlights the potential need to introduce recommendations in the current standards.

International audience; This work presents the results of the experimental characterization of the ignition sensitivity of solid inertant/combustible powders mixtures. Three inert solids (alumina, Kieselguhr, aerosil) and eleven organic powders have been considered and the following parameters have been determined: (1) the minimum ignition energy, (2) the minimum ignition temperature in cloud and (3) the minimum ignition temperature in 5 mm layer. The effects of the addition of inert solids are described and a simple model is proposed to represent the experimental results. Generally, increasing inert solid content in a powder leads to a higher minimum ignition energy as well as a higher minimum ignition temperatures in cloud and in layer. In some cases, the flammability is influenced above a threshold concentration value, which can be quite high (up to 85 wt.%). Indeed, the proposed model shows a zone below the minimum ignition concentration (MIC), which does not enable an efficient or safe inerting: either the admixed inert solid does not provide a sufficient effect, or it can even facilitate the ignition of the dust by notably improving its dispersability. The influence of key parameters such as the thermal conductivity or optical properties on the efficiency of the inerting by admixed solid need to be further assessed in a future work in view to better understand the mechanisms involved and to extend the scope to other types of oxidizable materials.

International audience; The increasing use of nanoparticles in the production of nanomaterials has led to the identification of nanoparticle agglomeration in air or in a given medium as a key problem, both on the standpoint of process performance (i.e. product homogeneity) and above all on inhalation risks. This paper suggests a new method to evaluate the structure and the dynamics of agglomeration of a nanopowder, based upon monitoring of the shear stress of a powder which is submitted to the mechanical solicitation of a rheometer. At low and increasing shear rates, the powder flow will evolve from a Newtonian regime (dense powder) to a Coulombic regime (slightly rheofluidized dense phase). At larger shear rates, the powder will be set in suspension, which characterizes a kinetic regime governed by particle collisions. The specific shear energy, related to the specific agglomeration energy, has been calculated for different nanopowders (carbon, aluminum, silica) and compared to non-cohesive micrometric powders (glass beads). From the measurements of shear rates and stresses at the frictional/kinetic transition, agglomerate diameters have been evaluated for carbonaceous material and for silica. Values of these agglomerates can range from 200 to 500 μm and are related to their propensity to break apart. The carbonaceous materials seem to be the more difficult to deagglomerate, whereas silica nanopowder agglomerates are more easily breakable. Measures of cohesiveness (or specific agglomeration energy) can be useful for assessing the dispersibility of nanopowders and their relationships to inhalation or explosion risks.

International audience; Previously, an extensive study has been carried out in order to assess the ignition sensitivity and explosivity of aluminum nanopowders. It showed notably that, as the particle size decreases, minimum ignition temperature and minimum ignition energy decrease, indicating higher potential inflammation. However, the explosion severity decreases for diameters lower than 1 μm. As a consequence, this study leads to the conclusions that the ignition sensitivity and explosion severity of aluminum nanopowders may be affected by various phenomena, as pre-ignition, agglomeration/aggregation degree and the intrinsic alumina content. The presence of wall-quenching effects and the predominance of radiation compared to conduction in the flame propagation process have to be discussed to ensure the validity of the 20 L sphere and of the results extrapolation. Based on the peculiar behaviours that had been previously highlighted, a specific risk analysis has been developed in order to assess the fire and explosion risks of such materials. It has been applied to an industrial plant of aluminum nanopowders production. The hazard identification and the consequence modelling steps, especially the quantification of the likelihood and consequences, have been designed specifically. The application of this method has led to the definition of the most adequate safety barriers.

International audience; In this paper, an inhalation health and explosion safety risk assessment methodology for nanopowders is described. Since toxicological threshold limit values are still unknown for nanosized substances, detailed risk assessment on specific plants may not be carried out. A simple approach based on occupational hazard/exposure band expressed in mass concentrations is proposed for nanopowders. This approach is consolidated with an iso surface toxicological scaling method, which has the merit, although incomplete, to provide concentration threshold levels for which new metrological instruments should be developed for proper air monitoring in order to ensure safety. Whenever the processing or use of nanomaterials is introducing a risk to the worker, a specific nano pictogram is proposed to inform the worker. Examples of risk assessment of process equipment (i.e., containment valves) processing various nanomaterials are provided. Explosion risks related to very reactive nanomaterials such as aluminum nanopowders can be assessed using this new analysis methodology adapted to nanopowders. It is nevertheless found that to formalize and extend this approach, it is absolutely necessary to develop new relevant standard apparatuses and to qualify individual and collective safety barriers with respect to health and explosion risks. In spite of these uncertainties, it appears, as shown in the second paper (Part II) that health and explosion risks, evaluated for given MWCNTs and aluminum nanoparticles, remain manageable in their continuous fabrication mode, considering current individual and collective safety barriers that can be put in place. The authors would, however, underline that peculiar attention must be paid to non-continuous modes of operations, such as process equipment cleaning steps, that are often under-analyzed and are too often forgotten critical steps needing vigilance in order to minimize potential toxic and explosion risks.

Risks associated with the end-of-life of nanomaterials are an issue that needs to be addressed so that the public perception and opinion, with regard to these emerging technological products, can effectively be supported by experimental evidences. In order to find new ecological ways to treat nanoproducts at their end-of-life, a new home-made demonstrator system was setup at INERIS, specifically designed to perform burning tests, coupled to a differential thermal analyzer to monitor the combustion kinetics. To assess nanoobject release during combustion, a high-performance nanocomposite polymer commonly used in the automotive industry, namely the polymeric compound acrylonitrile butadiene styrene matrix mixed with 3 wt% of multiwalled carbon nanotubes (MWCNTs) was tested. To assess the potential release of carbon nanotubes (CNTs) during the combustion with this tool, the particle size distribution in the fumes was measured using an electrical low pressure impactor, and CNTs were collected using an aspiration-based transmission electron microscopy grid sampler. One of primary objective of these preliminary tests described in this study consisted in validating whether CNT fibers can be released in the gas phase during the combustion of a polymeric matrix filled with CNTs. It was found indeed that MWCNT of about 12-nm diameter and 600-nm length can be released in the ambient environment during combustion of 3 % MWCNT ABS. Such information is critical to assess whether a nanoproduct can be deemed to be considered as 'nanosafe by design' in its risk assessment.

Characterization methods with regard to nanopowder flammability and explosivity are presented and illustrated for few nanopowders. Analytical models are developed in order to explain the dependency of the combustion times on the particle diameter. Experimental evidence shows that there exists, for carbonaceous and metallic materials, mainly two combustion regimes that are either kinetically controlled, for small size particles, or diffusion controlled, for large size particles. From the experimentally measured combustion data of those materials, the dependencies of the ignition temperature and the minimal explosive concentration (MEC) with regard to the particle size have been analyzed. We found that the two combustion regimes yield two different tendencies with respect to the particle size. Overall, it is found that as the particle size decreases, minimum ignition temperature (MIT) and minimum ignition energy (MIE) decrease, indicating higher potential inflammation and explosion risks for the use of nanopowders. By contrast, the minimal explosion concentration (MEC) did not show strong variations as the particle size decreases. Rather, a theoretical plateau is observed, which was experimentally confirmed. We also observed that carbon nanopowders exhibit a low propensity to explode while metallic nanopowders can be very reactive, thus delineating high potentials for explosion risks in manufacturing facilities.

International audience; Emerging nanomanufactured products are being incorporated in a variety of consumer products ranging from closer body contact products (i.e. cosmetics, sunscreens, toothpastes, pharmaceuticals, clothing) to more remote body-contact products (electronics, plastics, tires, automotive and aeronautical), hence posing potential health and environmental risks. The new field of nanosafety has emerged and needs to be explored now rather than after problems becomes so ubiquitous and difficult to treat that their trend become irreversible. Such endeavour necessitates a transdisciplinary approach. A commonly forgotten and/or misunderstood risk is that of explosion/detonation of nanopowders, due to their high specific active surface areas. Such risk is emphasized and illustrated with the present development of an appropriate risk analysis. For this particular risk, a review of characterization methods and their limitations with regard to nanopowders is presented and illustrated for a few organic and metallic nanopowders.

International audience; The origin of the kinetic theory of granular flow was originally credited to Bagnold [1]. By using a very primitive expression of the particle collision frequency, he derived an expression for the repulsive pressure of the particles in uniform shear flows. His repulsive pressure was proportional to the square of the velocity gradient and the particle diameter and directly proportional to the particle density. This theory was later extended by Savage [2] and Gidaspow [3]. Such theories provide insight on the dependence of the viscosity, and various moduli (elastic, non elastic, viscous...) in terms of the granular temperature and the associated shear-rates. Until recently, such parameters were difficult to measure because of the lack of specifically designed equipment. This challenge was successfully taken up and resolved by P. Marchal of ENSIC who designed a new rheometer for powders (figure 1). This equipment can put in evidence the importance of the granular temperature on the elastic and viscous behaviors of the granular flows. Such rheological behavior is important in risk analysis for nanopowders, because as the nanopowder may be subjected to process shear rates and stresses, its structural and topological changes, in terms of the transformation of agglomerates into primary nanoparticles, have strong impacts on emission factors of nanosized particles that can be released in the environment or into a workplace from such dense-phase nanopowder processes. Such transformation can be analyzed by studying the nano-granular rheological signature of the system. Such risk assessment approach using these new fundamental rheological safety parameters is described in this paper.

International audience; This work deals with the study of ignition and explosion characteristics of nanoparticles. It has been carried out on various powders: zinc, aluminum, carbon blacks... Specific behaviours have been highlighted during the first phase of this project (Nanosafe 2). For instance, it has been demonstrated that there mainly exists two combustion regimes that are either kinetically controlled, for small size particles, or diffusion controlled, for large size particles (generally with diameters greater than 1 or 2 µm). It has been found that as the particle size decreases, minimum ignition temperature and minimum ignition energy decrease (even lower than 1 mJ), indicating higher potential inflammation and explosion risks for metallic nanopowders. Moreover, the presence of agglomerates in the nanopowders could modify their reactivity. Thus, the explosion severity of Al powders tends to increase as the specific surface area decreases, before reaching a peak for 1 µm particle size. These results are essential for industries producing or handling nanopowders in order to propose/design new and proper prevention and protection means. Nevertheless, the validity of the classical characterization tools with regard to nanopowders should be discussed. For example, the experimental laminar flame velocity of Al dusts has been compared to a theoretical one, determined by Huang's model, which assumes that the propagation of the flame is run mainly by conduction. It has shown a good agreement. However, under certain conditions, the Al flame propagation is expected to be mainly conducted by radiation. Two hypotheses can then be made. On the one hand, it can be assumed that the 20 L sphere probably disturbs the flame propagation and thermal mechanisms by absorbing radiation (wall quenching effect). On the other hand, it has been observed, thanks to the use of a high speed camera that the preheating zone is smaller for some nanopowders than for micro-particles (figure below). It could notably be explained by the fact that the flame radiation is absorbed by the cloud of unburnt Al nanopowders. Several other factors may have an impact on the explosion severity. If these points are correctly addressed, it will be possible to get more reliable ignition and explosion characteristics.