Encyclopedia of Sustainability Science and Technology

Living Edition
| Editors: Robert A. Meyers

Airborne Nanoparticles: Control and Detection

  • Mohsen Rezaei
  • Matthew S. JohnsonEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-1-4939-2493-6_1099-1


Air pollution

“is the presence in the outdoor atmosphere of any one or more substances in quantities which are or may be harmful or injurious to human health or welfare, animal or plant life or property or unreasonably interfere with the enjoyment of life or property, outdoor recreation” [1].


is a solid or liquid particle suspended in a gas. Such particles may be produced directly, for example, from combustion or indirectly from gas-to-particle conversion. The former is primary aerosol and the latter secondary aerosol [2]. There are multiple processing mechanisms in the atmospheric including exchange with the gas phase and coagulation.


is a size category of aerosol particles with diameter between 1 and 100 nm [2].

Anthropogenic emissions

include man-made pollution from activities such as transportation, cooking and heating, industry, and energy production, involving combustion of fossil fuel and biomass. Human activities can be the main primary source of airborne nanoparticles both indoors and outdoors [3].

Filtration efficiency

characterizes the performance of a filter or any process that can trap particles passing through it. It can be determined by measuring mass or number density of particles before and after filtration, for a specific source or as a function of particle size [4].

Particle coagulation

when there is a large number of concentrations of airborne particles, they move by Brownian or turbulent diffusion; thereby their collisions increase with each other, resulting in sometimes aggregate clusters or a single new larger particle [1].

Nanoparticle characterization

is a subfield of atmospheric science and technology that uses determinations of the physical and chemical properties of nanoparticles to characterize their behavior and impacts [5].


when a particle is large and dense, its inertia will cause it to remain in the streamlines to be collided with the object. This is called impaction [1].

Light scattering

scattering of light as a result of collisions with particles called light scattering [5].

Electrical mobility

“when a charged particle is exposed to an electric field, it will migrate at a velocity that is determined by a balance between the resulting electrostatic force and aerodynamic drag that resists its motion. This characteristic migration velocity is described as the electrical mobility of the particle” [5].


This article will review the properties and behavior of nanoparticles suspended in air and measurement technologies for monitoring, characterizing, and controlling them, to reduce their adverse effects on the environment and human health. A huge variety of particles is found in the atmosphere and built environment. Many environments include unwanted gases and particles that can originate from primary emissions or from secondary transformations in the atmosphere. In addition there is significant processing within the atmosphere including condensation and evaporation, multiple cycles of activation into a droplet followed by dehydration, coagulation, photolysis, and chemical reaction. Research has shown that aerosol particles in the indoor and outdoor environments have significant detrimental impacts on health. The word aerosol was introduced in ca. 1920, in analogy to “hydrosol,” a liquid colloidal suspension of solid particles. Aerosols are particulate matter, either liquid or solid or a combination, suspended in a gaseous medium. “Aerosol particle” refers to the suspended particles themselves and can be divided into two groups, primary and secondary, depending on their origin. The former is generated by a source at the surface, for example, abrasion or combustion. The latter is often produced from gas-to-particle conversion or agglomeration of primary aerosols. Primary and secondary aerosol particles are characterized by their shape, size, and chemical composition. A spherical shape is often assumed, to simplify calculations. Aerosol particles are commonly classified by their “aerodynamic diameter,” that is, the diameter of a spherical particle with the same aerodynamic behavior. These classifications divide particulate matter (PM) into coarse (PM10), fine (PM2.5), and ultrafine (PM0.1) fractions. The subscripted number is a cutoff size, for example, PM2.5 is the total mass density of all aerosol particles with an aerodynamic diameter less than 2.5 μm. Airborne nano-sized particles, i.e., nanoparticles, belong to the ultrafine class of particles; they are often solid phase and have a dimension less than 100 nm. Nanoparticles are commonly emitted during combustion and are formed through gas-to-particle conversion.


There is a growing interest in airborne nanoparticles (NPs) because of their unique properties for health and climate and the increasing development of nanomaterials. NPs can easily enter the body and be distributed due to their small size. In addition NPs cause serious environmental problems including changing the dynamics of cloud formation, thereby contributing to drought and global warming and impacting the radiation balance of the Earth, as reported in the literature [6]. Recent studies have reported the harmful effects of NPs on human health including mortality [7, 8, 9] and their association with pulmonary disease, cancer, incident wheezing, asthma, lower spirometric values, and increased asthma-related emergency in children [10, 11]. Toxicological studies show that NPs possess unique physicochemical impacts due to their high number concentration, ultrahigh reactivity, and high surface area to mass ratio relative to other particle size classes [7], resulting in elevated bioavailability and toxicity [12]. Because of their small size, nanoparticles can enter into the bloodstream from the alveolar region of the respiratory tract or through the skin. Consequences include cardiopulmonary effects, oxidative stress, and cancer [13].

Because of their weak light scattering ability and extremely small size, NPs are often not visible. In addition they follow the streamlines of the carrier gas making them more difficult to control and measure than fine and coarse particles. Some common and effective technologies for removing coarse particles, such as centrifugation or settling, are only marginally effective for nanoparticles. The choice of controlling technology is limited to only a few methods. Filtration and electrical precipitation are widely taken to be the best options available. Fibrous filtration can filter NPs close to 99.999%, and a carbon wall membrane filter would be able to meet filtration efficiency up to 99.9999%. Newer technologies include phoretic and nanomaterial approaches, but their use in achieving clean air goals is untested.

Monitoring is one of the greatest challenges facing organizations that deal with indoor and outdoor air quality. It is clear that ultrafine particles possess quite different physicochemical properties, such as ultrahigh reactivity, high number concentration, and high surface area to mass ratio, relative to other size classes [7]. However, data concerning the composition and spatial/temporal distributions of NPs are limited, making it difficult to study their health effects [12]. There is great interest in developing new methods for monitoring NPs; key parameters include sensitivity, range, size resolution, and cost. Instruments for measuring NPs can be grouped into impaction, optical, diffusion, gravimetric, and electrical mobility techniques. Sometimes the techniques are combined. The micro-orifice uniform deposit impactor (MOUDI, TSI Nano-MOUDI 125R), laser aerosol spectrometer (LAS, TSI 3340A), condensation particle counter (CPC TSI CPC 3750), and scanning mobility particle sizer (SMPS, TSI 3938 E57 with 3082, 1 nm electrical classifier and CPC 3750) are commercialized examples of systems measuring ultrafine/fine aerosol particles.

Nanoparticle aerosol science and technology is a rapidly growing field. Many research papers describe nanoparticle control and characterization techniques in detail. It would therefore be useful to review the field’s current status and future vision. In this review, first, the source and health effect of nanoparticles are summarized. This is followed by a presentation of control technologies for nanoparticles, including filtration methods, electrostatic precipitators, thermophoretic and diffusiophoretic methods, and grouping technology, with a focus on removal mechanisms for nanoparticles, as well as the principals of each technology, and the effects of operational and structural parameters on removal efficiency. These parameters are discussed by comparing and reviewing significant studies. In addition, measurement techniques for nanoparticles along with their principal and comparative perspective are presented.

Source and Health Effect of Nanoparticles

Concern about exposure to NPs has increased dramatically due to awareness of and increase in anthropogenic sources involving transportation, industry, and energy production. Specific examples include soot from diesel engines, abrasion from tires and brake pads, smoke, and urban smog. The developing field of nanotechnology, producing new electronic, magnetic, optoelectronic, biomedical, pharmaceutical, cosmetic, energy, environmental, and catalytic products and processes, is another source of human exposure to engineered NPs. Some of the natural and anthropogenic sources of NPs are summarized in Table 1 [3]. In nature, NPs are produced by many sources such as in situ photochemical reactions, forest fires, volcanic eruptions, sea-salt spray, microorganisms, biogenic magnetite, and so on. Graphitic soot is produced by wildfires. Due to their light scattering properties, NPs impact the radiation balance of the Earth directly and indirectly and are central to the mechanisms of climate change [14]. Many kinds of NPs are known within organisms. Natural production of NPs can occur through intracellular and extracellular mechanisms including within human brain cells. Many organisms produce nanomaterials through intracellular and extracellular processes. For example, biogenic magnetite is a kind of nanomaterial that has been found in many microbial species including microorganisms (from bacteria to protozoa) and in animals (e.g., human brain) and which has been associated with neurodegenerative diseases [15].
Table 1

Natural and anthropogenic sources of NPs




Fire (forest)



Sea spray (e.g., salt)

Biogenic magnetite

Ferritin (12.5 nm)


Condensation and nucleation

Car exhaust and brakes


Power generation


Factories (e.g., cement)

Incinerators (e.g., waste)

Jet engines

Fumes, e.g., polymer and metal fume

Stove (cooking)


Laboratories engineered nanoparticle (e.g., catalysts)

Anthropogenic sources are estimated to account for about 10% of total NPs. NP sources can be divided into industrial and urban. Industrial activities release NPs into the atmosphere inside factories working with metallurgy, power and cement production, mining, incineration, and any industrial product that uses fossil fuels as an energy source for power generation. In urban areas, vehicles are a common source of NP outdoors, being produced by internal combustion engines, road dust and wear on brakes and tires. Important indoor sources of NP include cooking, cleaning and smoking. Of course, the chemical and physical nature of NPs emitted in industrial and urban areas can be quite different. Indoor emissions can be most important for determining health impacts because humans normally spend >80% of their time in the indoor environment [16]. Engineered nanoparticles are commonly used in stain-resistant clothing, sporting goods, tires, sunscreens, cosmetics, and toothpaste and as food additives, etc. and are specifically engineered in the laboratory to form nanostructures such as nanotubes, nanofibers, and nanowires [15, 16].

Epidemiological and clinical studies on humans and rodents, and also in vitro cell cultures, have shown the health effects of laboratory-generated and ambient NPs. Researchers have documented the association of ambient NPs with adverse cardiovascular and respiratory effects in humans [17, 18, 19]; however, not all studies have not seen these associations [20, 21]. Intensive deposition of laboratory-engineered NPs in the respiratory tract, even greater with asthma or chronic obstructive pulmonary disease, has been seen in controlled clinical studies. On the other hand, controlled exposure to carbonaceous NPs has been seen to cause many symptoms in the cardiovascular system including systemic inflammation and changes in pulmonary diffusion capacity and the appearance of coagulation markers in the blood. In vivo tests by Brown et al. [22] have shown an increased inflammatory response after the introduction of 64 nm polystyrene particles into the rat lung, and they have observed a pro-inflammatory effect after in vitro cell culture with the model NPs. In addition, an oxidative stress-related cellular response with ambient NPs has been reported by Li et al. [18].

The main deposition mechanism for inhaled NPs in the respiratory tract is diffusion; other deposition mechanisms including inertial impaction, gravitational settling, and electrostatic force are not significant. Figure 1 shows three exposure regions for inhaled aerosol particles in the human respiratory tract including the nasopharyngeal, tracheobronchial, and the alveolar region. The figures on the right (Fig. 1) show which sizes of PM can deposit in each region of the respiratory tract. With an increase in size from 1 to 100 nm, high percentages of inhaled NPs are deposited on the upper regions of the respiratory tract. For example, ~90% and ~10% of 1 nm particles are deposited in the nasopharyngeal and the tracheobronchial regions, respectively, while no particles are deposited in the alveolar region. In contrast, 50% of 20 nm particles are deposited only in the alveolar region [15]. Manigrasso et al. [19] estimated a respiratory deposition of 6.6 × 1010 particles of 5.6–560-nm-diameter particles per hour for persons close to traffic, in short-term peak exposure events. It has been seen that deposited NPs are transported to extrapulmonary organs to a greater extent and by a larger variety of transfer routes than for larger-sized particles. One is the conduction of NPs into blood by transcytotic transport across epithelia of the respiratory tract and subsequently into the interstitium or by lymphatics which can distribute throughout the body. Another is axonal translocation of NPs to ganglionic and CNS structures by their uptake by the sensory nerve endings of airway epithelia. Nanotoxicological studies have demonstrated that inhaled NPs can deposit in the nasal cavity, being absorbed by the olfactory epithelium and translocated into the brain through the blood-brain barrier [17].
Fig. 1

Predicted fractional deposition of inhaled particles in the nasopharyngeal, tracheobronchial, and alveolar region of the human respiratory tract during nose breathing. (Reprinted with permission [15])

Nanoparticle Control Technologies

One of the most important concerns for air pollution control is the removal of nanoparticles from airstreams. A variety of technologies have been used in recent years. It should be noted that air quality regulatory organizations are imposing increasingly strict rules as knowledge of the health effects of nanoparticles has grown. In parallel with increased knowledge, it has become increasingly urgent to implement suitable particle control systems. The most effective and important technologies for capturing nanoscale particles will be examined, based on their physical properties, the required degree of control, cost, etc. [23].

Nanoparticle Filtration

Filtration is the most commonly used method for trapping airborne particles. Different types of filters are used to target a wide range of airborne particles. The most common forms are fibrous, porous membrane, capillary pore membrane, fabric, and granular bed [24]. Fibrous filters are the simplest and most economical filters and are capable of efficiently removing nanoscale particles from gas streams [25].

It is important to quantify filtration performance in order to document and compare the abilities of particle filters. Filter efficiency and pressure drop are the most crucial parameters in this regard [26]. Equations (1) and (2) define filter efficiency in terms of mass concentration and particle number, respectively:
$$ E=\frac{{\mathrm{C}}_0-C}{C_0} $$
$$ E=\frac{{\mathrm{N}}_0-N}{N_0} $$
where C0 and C are the mass concentrations (typically as μg/m3) of particulate matter before and after filtration and N0 and N are the number concentration (usually as number per m3) before and after filtration.
The efficiency of pollution control equipment is often characterized in terms of the fraction entering versus exiting the filter. This metric is known as particle penetration (P):
$$ P=\frac{C}{C_0}=1-E $$
$$ P=\frac{N}{N_0}=1-E $$
In addition to performance against particles, HVAC system engineers require some additional information to design filtration systems and judge their efficiency: The volumetric airflow passing the filter (Q) and its velocity at the face of the filter (normal to the filter, “U”) impact P and E. They are related as shown in Eq. (5):
$$ {U}_0=\frac{Q}{A} $$
A is the area of the filter. However, the air velocity inside the filter is higher than the face velocity because of the air resistance caused by the filter material. The material comprising the filter has a cross-sectional area reducing the area through which air can flow. If Q is conserved and A decreases, U will increase. The velocity through the filter (Uf) is further defined as:
$$ {U}_f=\frac{Q}{A\ \left(1-\alpha \right)} $$
where α is the packing solidity or density of the filter, which can be obtained from Eq. (7):
$$ \alpha =\frac{V_f}{V_T}=1-\varnothing $$
where Vf is the fiber volume; VT, the total filter volume; and ∅, the porosity [24].
The goal is to create a filter that combines a high filtration efficiency (E) with a low pressure drop (Δp), thereby cleaning air while minimizing, e.g., power and noise. However, in most particulate filter systems, an increase of the filtration efficiency also increases pressure drop. The resistance to airflow across the filter is characterized by the pressure drop and arises from the combined drag force of the fibers. Assuming laminar flow, pressure drop will increase with the face velocity; to a first approximation, this relation is linear. Pressure drop is proportional to filter thickness and the packing density of fibers and is inversely proportional to the square of the fiber diameter [27]. Despite this, for many installations, it is desirable that the pressure drop of the filter should be less than ~150 Pa which is often difficult to achieve while maintaining high filtration efficiency. The quality factor is one criterion used to judge the performance of a filter, as defined by Chen [28]:
$$ Quality\ factor=\frac{-\ln \left(1-E\right)}{\varDelta p} $$

The QF shows the correlation between filtration efficiency and pressure drop. A filter with a greater filtration efficiency and lower pressure drop will have a high quality factor.

In recent years, nanoscale fiber filters have received increasing attention for air filtration because of favorable properties including small pore size and high specific surface area. These properties enable the slip effect of airflow through the fibers, reducing the drag force on the airstream as it passes nanofibers and subsequently strengthening the different filtration mechanisms on the filter [26]. Normally the surface of a fiber blocks the airflow, diverting airstreams, triggering turbulence and viscosity, and necessitating power and pressure drop for the air to flow past. However, when the fiber diameter similar to or below the size of the mean free path of air molecules (just under 70 nm), normal continuum flow behavior no longer applies. Molecules can fly past the fiber without interrupting the momentum of the flow; this is called “slip flow.” As detailed below, because of slip flow, nanofiber filters hold the promise of yielding high QF filtration for nanoparticles.

Air filtration using fibrous filters is a promising method for separating nanoparticles from an airstream while allowing air molecules to pass [29]. Filters manufactured with materials including plastic, cellulose, glass, and carbon are the most common [24]. The fibers are made into a soft flexible mat, with fiber orientations primarily perpendicular to the streamline [30]. These mats provide a high filtration efficiency for ultrafine particles along with minimal aerodynamic resistance and thus find many applications such as industrial and indoor cleaning purification, respirators, and car cabins [23, 30].

A few mechanisms are important for trapping nanoparticles from their air: interception, inertial impact, diffusion, and electrostatic deposition. Their relative importance will vary depending on gas velocity, particle and fiber size, particle charging, surface static charge, and so on. The total filtration efficiency for nanoparticles is the sum of the filtration efficiencies of particles by each mechanism (Table 2). Sound classical descriptions of fibrous filter performance can be found in many sources (see, e.g., [23, 24, 30]).
Table 2

The filtration mechanisms and their filtration efficiency

Filtration mechanism

Single-fiber filtration efficiency




Open image in new window

\( {E}_R=\frac{\left(1-\alpha \right){R}^2}{K_u\left(1+R\right)} \) or,

\( {E}_R=0.6\ \left(\frac{\left(1-\alpha \right){R}^2}{K_u\left(1+R\right)}\right)\left(1+\frac{k_n}{R}\right) \)

\( R=\frac{d_p}{d_f} \)

\( {K}_u=-\frac{Ln\ \alpha }{2}-\frac{3}{4}+\alpha -\frac{\alpha^2}{4} \)

ER: filtration efficiency by the interception

dp: particle diameter

df: fiber diameter

Ku: Kuwabara hydrodynamic factor

α: packing density (solidity)

Only a deposition mechanism does not result from a particle departing from its original streamline

Inertial impaction

Open image in new window

\( {E}_I=\frac{(STK)J}{2{K_u}^2} \)

\( STK=\frac{\rho_p{d}_p^2{C}_c{U}_0}{18\mu {d}_f} \)

J = (29.6 − 28α0.62)R2 − 27.5R2.8

EI: filtration efficiency by the impaction

STK: Stokes number

J: dimensionless

Cc: Cunningham coefficient

ρp: particle density

μ: air dynamic viscosity

The most important deposition mechanism for large particles

Diffusion (Brownian motion)

Open image in new window

\( {E}_D=1.6\ {\left(\frac{1-\alpha }{K_u}\right)}^{1/3}\ 2{Pe}^{\raisebox{1ex}{$-2$}\!\left/ \!\raisebox{-1ex}{$3$}\right.} \)

\( Pe=\frac{d_f{U}_0}{D} \)

\( D=\frac{k_bT{C}_s}{3\pi \mu {D}_p} \)

\( {C}_s=1+{K}_n\ \Big[1.027+0.44\mathit{\exp}\left(\raisebox{1ex}{$-0.78$}\!\left/ \!\raisebox{-1ex}{${K}_n$}\right.\right) \)

ED: filtration efficiency by the diffusion

Pe: Peclet number

D: particle diffusion coefficient

kb: Boltzmann constant

T: absolute temperature

Cs: Cunningham slip correction

Only the deposition mechanism that increases as the particle diameter increases

Electrostatic deposition

Open image in new window

Difficult to quantify because it requires knowing the charge on the particles and on the fibers

This mechanism is often neglected unless the particles and/or fibers have been charged in some quantifiable way

Particles larger than around 300 nm are subject to inertial separation from a curved flow streamline, leading to impact with an obstruction. Such particles are unable to adjust to an abruptly changing streamline in the vicinity of a fiber.

The diffusion constant increases as the size of particles decreases. For particles smaller than ca. 100 nm, diffusional separation of particles from the continuum flow streamline can be the dominant mechanism leading to impaction. Every filter has a minimum in efficiency for a characteristic particle size called the most penetrating particle size (MPPS), which typically falls between 100 and 300 nm. The removal efficiency of filters at the MPPS determines the filter grade. In addition there are two criteria, surface chemistry and the electrostatic potential of fibers, which play significant roles in determining filtration efficiency. For example, polar functional groups at the filter surface may lead to strong adsorption of nanoparticles.

The relative roles of the mechanisms depend on the Knudsen number (Kn) as [26]:
$$ {K}_n=\frac{2\lambda }{d_f} $$
where df is the diameter of fiber and λ is the mean free path of gas molecules. The mean free path is ca. 66 nm at the standard state of air, obtained as:
$$ \lambda =\frac{RT}{\sqrt{2}{N}_a\pi\ {d}_m^2\ p\ } $$
where R is the gas constant, T temperature, Na Avogadro’s number, dm the collision diameter of an air molecule (3.7 × 10−10 m), and p the air pressure [31].
There is a relationship between the airflow regime, as defined by the Reynolds number, across the filter and the fiber diameter in all the airborne particle filtration systems. Particles can be removed from an airstream in the continuum flow regime (Kn < 0.001, generally for milli-scale fibers), the slip flow regime (0.001 < Kn < 0.25, generally for microscale fibers), the transition flow regime (0.25 < Kn < 10, for nanoscale fibers), and the free molecular regime (Kn > 10, for ultrafine scale fibers), as shown in Fig. 2a, b [32]. In the continuum flow regime, the mean free path of the gas molecules is much smaller than the diameter of the fiber. Most filtration systems use fibers operating in the continuum flow regime. The slip airflow regime is characterized by an increase in the Knudsen number, resulting in an increased filtration efficiency via the slip effect. The free molecular flow regime is a dominant regime in which the diameter of fibers is much smaller than the mean free path of the gas molecules. In general, nanoparticle-capturing fibers, including electrospun fibrous filters and carbon nanotubes, have higher surface areas for nanoparticle deposition than microfibers that operate under the transition airflow (electrospun filters (Fig. 2c)) and free molecular airflow regimes (carbon nanotubes (Fig. 2d)). It should be noted that it is not possible for other fibers to achieve the free molecular regime by increasing λ by reducing the pressure [26]. The slip flow effect can be understood by considering the interplay between viscosity length and fiber diameter. A large fiber blocks the airflow, so the momentum of the air is zero in the vicinity of the surface. As fiber size decreases, the height of the surface layer, called the quasi-laminar layer, decreases, becoming zero. A nanofiber is not able to significantly impact the momentum of the flow and the air is said to slip past, as described in terms of the Reynold’s number above.
Fig. 2

(a) Flow pattern around fibers of different diameter. (b) The surface areas of the nanofibrous filters. (Reprinted with permission [32]). SEM images (c) of PUR electrospun nanofibers and (d) carbon nanotubes. (Reprinted with permission [31])

The filtration efficiency for a clean fibrous filter can be determined using Eq. (11):
$$ E=1-\mathit{\exp}\left[\frac{-4\alpha {E}_fZ}{\pi \left(1-\alpha \right){d}_f}\right] $$
where, α, Z, Ef, and df are fiber packing density, filter thickness, single-fiber efficiency, and mean fiber diameter, respectively. For nanofiber filters, interception and diffusion are the most important trapping mechanisms, and thus the sum of the mechanisms can be used to approximate the theoretical single-fiber efficiency, as follows [24]:
$$ {E}_f={E}_R+{E}_D $$

Electrospun Nanofibers

A series of studies have found that electrospinning is capable of producing a felt or mat of high porosity electrospun nanofibers which efficiently and effectively remove ultrafine PM from an air stream. The diameter of the electrospun nanofibers can be from <2 nm (microporous) to 50 nm (mesoporous) and are a thousand times smaller than microfibers [33]. Figure 3 shows the most common electrospinning equipment.
Fig. 3

Schematic of the electrospinning technique

Electrospinning Process

Electrospinning is a technique that uses an electrostatic force on a polymer fluid to produce nanofibers. After Formhals, Taylor developed the method mathematically, describing the effect of the electrostatic charge on the polymer droplet to form a conical structure, called a Taylor cone [33].

As can be seen in Fig. 3, polymer solutions or melts are loaded into a syringe which is held at a high potential with a voltage of kV applied between a fiber collector and the syringe needle. The combined forces of surface tension and electrostatic repulsion result in the formation of the Taylor cone at the needle tip. The electric field overcomes the surface tension and a stable jet is formed at the needle tip. The combined result is that an elongating-and-whipping process is continuously generated by electrostatic repulsion resulting in a thin string. These fibers accumulate on the collector forming a mat of nanofibers.

Finally, the nanofiber mat is removed from the collector, after evaporation of the solvent. It should be noted that the solution electrospinning method, compared to melt electrospinning, is able to generate fibers with a smaller diameter. The diameter of the fibers can be controlled by changing solution parameters, such as the concentrations of the components, and operating parameters such as the electrospinning voltage. Additionally, a variety of fiber structures, including hollow, core/shell, secondary growth, porous, and helical, can be generated by controlling the electrospinning procedure, solution components, and calcination parameters. In addition the nanofibers can be modified after spinning using thermal or chemical treatment to improve features that include pore size, conductivity, and mechanical strength. Thermal posttreatment can be used to control the diameter of nanofibers. Metal oxides or salts in the nanofibers can be used to make the fibers bactericidal, and graphitization of carbon onto the nanofibers will enhance their specific area surface. Chemical modification may add properties similar to thermal treatment. Nanofibers generated by electrospinning can make use of a wide range of inorganic, organic, and organometallic materials. Almost any soluble polymer can be incorporated into nanofibers with the electrospinning method with the constraint that their molecular weight should be sufficiently high. Examples include polyvinylpyrrolidone (PVP), polyethylene oxide (PEO), polyvinyl acetate (PVAc), and polyvinyl alcohol (PVA), which are water-soluble polymers, and non-water-soluble polymers such as polyimide (PI), polyacrylonitrile (PAN), polyvinylidene fluoride (PVDF), polylactic acid (PLA), polymethacrylate (PMMA), polystyrene (PS), polypyrrole (PPy), and polyvinylchloride (PVC) [34].

Performance of Electrospun Nanofibers

Nanofiber filters have attracted a great deal of interest since they have a high filtration efficiency due to interception and a low pressure drop because of the slip flow effect. Many parameters impact filtration performance including the physical properties of the filter (such as nanofiber diameter, surface area, pore size, packing density, basis weight, and thickness) and the conditions of the installation such as face velocity. Particulate matter builds up on the filter in dendrimers which can enhance particle filtration [25].

Nanofiber diameter is an important parameter affecting filtration performance. Mao et al. [4] evaluated the relationship between fiber diameter and filtration efficiency and pressure drop, using an yttria-stabilized zirconia (YSZ) nanofiber. They found a decrease in the filtration efficiency with an increase in nanofiber diameter. At the same time, smaller fiber diameter decreased pressure drop, regardless of particle size (Fig. 4a, b). It was recently reported that nanofiber diameters less than 100 nm enhance the slip flow effect, decreasing drag and pressure drop and subsequently increasing particle capture [37]. One study used keratin-based nanofibers to demonstrate the effect [39]. Polymeric nanofiber felts made of PU nanofibers [31], cellulose [40], and PVP [41] have been produced by electrospinning and also show improved filter performance due to their reduced fiber diameter. In sum, electrospinning provides greatly improved filtration efficiency due in part to smaller fiber diameter that is able to capture ultrafine particles; the smaller the fiber, the greater the filtration efficiency.
Fig. 4

(a) The effect of the fiber diameter of yttria-stabilized zirconia (YSZ) on filtration efficiency and pressure drop. (Reprinted with permission [4]). (b) Schematic landscape of the relationship between fiber diameter and particle capture. (c) The effect of packing density on MPPS. (Reprinted with permission [35]). (d) Relationship between fiber basis weight and filtration efficiency. (Reprinted with permission [36]). (e) Pressure drop and (f) filtration efficiency for PSU/TiO2-0 and PSU/TiO2-5 as a function of face velocity. (Reprinted with permission [37]). Relationship between the particle diameter and the MPPS. (Reprinted with permission [38]). (h) The captured particle diameter increases with air velocity. (Reprinted with permission [31])

The surface area and pore size of nanofibers are affected by the fiber structure and the composition of the blended solution. Liu et al. [42] prepared an electrospun PAN/polyacrylic acid (PAA) composite nanofiber felt to study the effects of PAN/PAA ratios on filtration performance. Blending PAA with PAN allowed creation of a composite with features that could be controlled. The tensile strength of the nanofibers greatly improved, going from 3.8 to 6.6 MPa as the PAA content increased, creating robust mechanical strength in pure PAN nanofibers. The smallest pore size in the membranes resulted in a very high filtration efficiency, over 99.99%, with minimal pressure drop, 160 Pa. A porous bead-on-string polylactic acid (PLA) membrane for capturing nanoparticles was created by Wang et al. [43]. The large surface area, coupled with microporous beads, showed enhanced filtration efficiency. A smaller pore size polyvinyl alcohol (PVA) nanofiber filter fabricated by Li et al. [44] has shown a high filtration efficiency for nanoparticles. Wang et al. [45] found that adding SiO2 nanoparticles to multilayer PAN structures created bimodal distributions of fiber sizes that increased filtration performance, with a removal efficiency of 99.99% at a pressure drop of Δp = 116 Pa, for nanoparticles relative to single-layer fibers. This was said to be due to the roughness of the surface fibers arising from the noncircular structure and also from the greater specific surface area of the multilayer fibers, as a result of adding SiO2. Two membrane fibers, PA6 and PMIA, arranged in sequence by Zhang et al. [46] showed high filtration efficiency (99.995%) for nanoparticles at a low pressure drop (100 Pa). The hybrid membrane has high porosity, small pore size, and a large specific surface. Generally, microporous fibrous filters, despite their higher filtration efficiencies, have a higher particle retention ratio and show a higher tendency for clogging [47]. Also, the smaller fiber diameter is associated with a smaller pore size due to the decrease in pore aperture with fiber diameter, improving the interception-collision of nanoparticles with the nanofiber and increasing capture efficiency [48].

The nanofiber packing density (α) depends on fiber density and filter thickness. High α values correlate with effective particle capture. Some workers have investigated the effect of α on the filtration efficiency of nanofilters. Leung et al. [35] found a good relationship between the packing density and the filtration efficiency for PEO fibers. By increasing α from 3.9 × 10−3 to 36 × 10−3, filtration efficiency increased from 15% to 90%. At the same time, the most penetrating particle size (MPPS) decreased from 140 to 90 nm (Fig. 4c). Choi et al. [38] found that a laminated nanofiber with high packing density has a lower quality factor owing to its high pressure drop when compared to a mixed fiber (nanofiber 780 nm/microfiber 11.4 μm). Furthermore, reducing the fraction of nanofiber in a microfiber filter enhanced the quality factor at a low packing density. Therefore they suggested that an optimal ratio would be to have a nanofiber mixing fraction of 5% in mass in order to decrease pressure drop and increase the quality factor for nanoparticle capture.

According to Eq. (10), a larger fiber thickness can be useful for improving the performance of nanofiber filters. By increasing the fiber thickness, the basis weight and the pore size of fibers are also increased, elevating filtration efficiency. However, Kim et al. [49] noted that if the fibers are too thick, the filtration efficiency is reduced due to the increased pressure drop. Wang et al. [50] found that a thick filter contributed to higher energy consumption and larger pressure drop. Research by Zhang at al [51] showed that a multilayer PAN fibrous membrane can be more effective for obtaining a high quality factor compared to a single thick layer nanofiber. Leung et al. [35] concluded that the effect of fibers’ thickness on the MPPS is less than the effect of fiber packing density. The work of Yun et al. [52] demonstrated that the pressure drop across an electrospun PAN filter (270 nm) increased linearly from 63 to 220 Pa as the fiber thickness increased from 0.004 to 0.02 mm. However, they reported that all fibrous filters for a given particle size had the same single-fiber efficiency values, indicating thickness-independent quality factors.

Hung et al. [53] found that larger fiber basis weights enhanced filtration efficiency nonlinearly. However, the increased basis weights decreased the quality factor due to an increased pressure drop (Fig. 4d). Also, this study found that the multi-structure nanofiber membrane produced in the electrospinning process with low basis weight showed better performance than a single-structure nanofiber membrane with a high basis weight.

Besides the filter structure, there are many filtration parameters capable of significantly changing filter performance. The air velocity across the filter and particle size distributions play a key role in this regard [35]. Many studies find that a higher face velocity results in a lower filtration efficiency. Wan et al. [37] found the filtration efficiency of a polysulfone (PSU) membrane fibrous filter without TiO2 decreased from ~100% to 94% as the face velocity increased from 15 to 90 L/min, while filtration did not decrease significantly for a filter coated with TiO2 nanoparticles at 5% by weight (Fig. 4e). In the fibrous filters, the pressure drop is proportional to the face velocity in accordance with the experimental results of this study (Fig. 4f). Thus, the results suggested low quality factors at higher face velocity. Similarly, Wang et al. [54] have tested a polyvinylidene fluoride (PVF)/polytetrafluoroethylene (PTFE) nanoparticle filter for face velocities of 2–16 cm/s. They found the filtration efficiency dropped from ~100% to 95% for pure PVF and a slight decrease to 98% for PVF with 5 weight percent PTFE.

The relationship between particle size and filtration efficiency is well established. Many studies have shown that filtration efficiency is very dependent of the particle size because of its effect on the efficacy of the capture mechanisms. For example, for particle sizes larger than the fiber diameter, the single-fiber efficiency derived using interception is not correct. The relationship between particle size and filtration efficiency can be quantified using the MPPS. Many studies found that a particle size of around 100–300 nm is usually the MPPS [38, 55] (Fig. 4g). It has been also found that the MPPS changes when the face velocity changes. Sambaer et al. have shown that with an increase of the face velocity, the MPPS decreases (Fig. 4h) [31].

Table 3 shows the filtration efficiency, pressure drop, and quality factor for different polymeric membrane nanofibers.
Table 3

The filtration performance in recent studies


Nanoparticle diameter (nm)

Fiber diameter (nm)

Filtration efficiency (%)

Pressure drop (pa)

Quality factor (pa−1)









PU (polyurethane)






















(PA: polyamide)







PSU/TiO2 (5%wt)







PAN/PU-FPU (FPU: fluorinated polyurethane)














PVDF/PTFE NPs (0.05%wt)







Carbon Nanotubes (CNTs)

CNTs were fabricated the first time by Iijima et al. [60] and have now received considerable attention regarding their use in filters for airborne particles. CNTs are classified as single-walled CNTs (SWNTs) (normally df < 10 nm) and multi-walled CNTs (MWNTs) (normally df > 10 nm) [60]. The very small diameter of fibers indicates that flow around the CNTs will be in the free pass flow regime, where the disturbance of the flow by CNTs is not significant [47]. In comparison with electrospun nanofibers, CNTs provide high filtration efficiency because of their ultrathin fibers, ultrahigh specific surface areas, and robust mechanical properties. The limitation of a trade-off between filtration efficiency and pressure drop inherent to fibrous filtration can likely be overcome with CNT-based filtration [26]. However, it is very difficult or perhaps impossible to fabricate a CNT that can balance many design constraints and trade-offs involving fiber properties such as air velocity, packing density, and basis weight. The important characteristics of the CNT nanofibers are similar to those discussed for electrospun nanofibers. Key features of the performance of the CNTs are discussed below.

Viswanathan et al. [27] coated a continuous layer of MWNT (df = 20–50 nm, thickness = 1–2 nm, porosity = 93.58%) film onto cellulose fibers (Fig. 5a, b). The diameter and thickness of the MWNT were three and two orders of magnitude smaller than the diameters and thickness of the cellulose fibers, respectively. Table 4 shows that with an increase in the MWNT/cellulose ratio, the pressure drop sharply increased. The pressure drop of cellulose filter II, consisting of five cellulose layers, was similar to that of the MWNT I, but the filtration efficiency and thus quality factor of the MWNT I were much higher. It can be seen that MWNT II–IV with a filtration efficiency higher than 99.97% could be used in HEPA class filters, which require filtration of 99.975% particles at the MPPS.
Fig. 5

SEM images of MWNT-coated filters. (a) Fibrous filter morphology. (b) The cross-sectional view shows a continuous MWNT film (∼1–2 μm thick, as seen in (c)) on top of the cellulose layer. (Reprinted with permission [27])

Table 4

Characteristics of cellulose filters and MWNT-coated filters. (Reprinted with permission [27]). Note: cellulose filter II consists of five cellulose layers in series


Coverage of MWNTs

[mg cm−2]

Pressure drop after filtration [kPa]

Filter efficiency for dp = 300 nm [%]

Filter quality for dp = 300 nm [kPa−1]

Cellulose filter I
























Cellulose filter II




Free-standing SWNT films (FSFs), an excellent example of air-filtering CNTs, were fabricated by Nasibulin et al. [61], as shown in Fig. 6a. The nanotubes have 10 μm lengths and 1.3–2 nm diameters with a few micrometer’s thickness. Figure 6b shows the filtration efficiency increased linearly as the nanotube was thickened. Figure 6c shows a lower pressure drop by FSF filters, up to ~230 Pa, as compared to commercial filters (500–4000 Pa) in various airflow rates, illustrating the high quality factor of the FSFs; more than one order of magnitude is higher than that of commercial filters (Fig. 6d). They noted that the FSF have robust mechanical properties and are durable enough to capture nanoparticles for study. It was also found from durability tests that the filter was able to operate for more than a month without a drop in the filtration efficiency at a flow rate of 300 cm3/min. The high surface area of the free-standing SWNT films can reasonably explain their high filtration efficiency because of the related improvement of the filtering mechanisms in the FMF regime. The low pressure drop results from the high porosity and low thickness of the films.
Fig. 6

(a) Set of images showing a submonolayer FSF suspended over 5 mm openings in aluminum foil. (b) Thickness dependence of transmittance and of collection efficiency for 44 nm γ-Fe2O3 particles. (c) Flow rate dependence of pressure drop for 120-nm-thick SWCNT films and various commercial filters. (d) Dependence of the filter quality factor on the particle size for commercial filters and a 120-nm-thick SWCNT film. The solid and open symbols show results obtained by two independent measurement techniques, namely, scanning over the entire particle size range studied and at fixed particle sizes. (Reprinted with permission [61])

Yildiz et al. [62] prepared aligned CNT sheets by embedding them between polypropylene melt-blown fabrics. The diameters of the CNTs were 25–40 nm with lengths of ~1 mm and thicknesses of 20–25 μm (Fig. 7a, b). The test particles were 10–300 nm. They found, as shown in Fig. 7c, that the filtration efficiency of the fabricated filters dramatically increased with the number of CNT layers, while the pressure drop also increased. They observed that while the pressure drop of the seven-layer filter was very high, the filtration performance should meet high-efficiency particulate air (HEPA) filter standards with an acceptable pressure drop. Therefore, they laid the CNTs in a cross-plied structure within the filter. The three-layer CNT cross-ply filter met the HEPA filtration standard with a filtration efficiency of 99.98% for a 300 nm particle size at 10 cm/s face velocity and also had the highest quality factor (Fig. 7d). They noted that the performance of novel CNT filters is comparable to electrospun fabrics, making them a viable option for future filtration applications.
Fig. 7

(a) SEM and (b) photographic images of the three-layer CNT filter structure with a cross-ply geometry. (c) Particle penetration fraction of the different structures as a function of pressure drop at 0.3 μm particle size and 10 cm/s face velocity. (d) Quality factor as a function of particle size, ranging from 0.01 to 0.3 μm, at 10 cm/s face velocity. (Reprinted with permission [62])

After the development of CNT filters for air filtration applications, researchers have observed that the filters can become clogged by nanoparticles due to the smaller space between CNT films than particle aggregates. To overcome the capacity-related issues, a hierarchical structure for CNT filters has been proposed. In a typical procedure, the CNT films grow on a porous material. This structure not only provides a macroporous structure and high mechanical strength, but also the specific surface area of the filter structure increased.

Li et al. [63] fabricated CNT/quartz fiber (QF) filter as a depth-type hierarchical structure through in situ growth of CNTs on quartz fiber (QF) filters using a floating catalyst chemical vapor deposition (CVD) method. They observed the CNT/QF filter met the standard of high-efficiency particulate air (HEPA) filters (Fig. 8a, b). The specific surface area of the hybrid filter was more than 12 times higher than that of the pristine QF filters. The pore size of the CNT/QF filter only has a small change. The pressure drop through the CNT/QF filter changed only a small amount relative to that of the pristine QF, while the filtration efficiency increased significantly, resulting in an overall increase of the quality factor for the CNT/QF filters. Scanning electron microscope images reveal that CNTs are very efficient at capturing submicron aerosols (Fig. 8c).
Fig. 8

(a) Penetration of particles with different sizes in the QF filter and the CNT/QF filter; (b) quality factor versus particle size of the QF filter and the CNT/QF filter. (c) SEM images of the CNT/QF filter with deposited NaCl particles after filtration testing. The inset shows NaCl particles deposited on a single CNT (the black arrow shows a NaCl particle larger than 100 nm; white arrows show NaCl particles smaller than 100 nm). (Reprinted with permission [63])

To prevent clogging and thus increase the service life of the filters, Li et al. [63] created hierarchical CNT/quartz-fiber (QF) filters with gradient structures where the content of CNTs decreases exponentially along the thickness direction of the filters. They observed that by using only 1.17 wt % CNT, penetration through the gradient filter reduced the MPPS by one order of magnitude, where the pressure drop was only 6% higher than that of the pristine QF filter, resulting in a higher quality factor for the gradient structure (Fig. 9a, b). As shown in Fig. 9c, they also tested different placements of CNT inside the quartz filter to investigate clogging and service life of the gradient filter. The service life of a composite with the CNT-rich side downstream was up to 64% longer compared to the pristine QF filter, while the service life of the composite oriented with the CNT-rich side upstream was only 41.7% longer.
Fig. 9

(a) The rate of pressure drop increase versus filtration test time for the CNT/QF filter with different placement positions and the QF filter under continuous aerosol loading; (b) the decrease in rate of efficiency at MPPS versus filtration test time for the CNT/QF filter with different placement positions and the QF filter under continuous aerosol loading. (c) An illustration of aerosols accumulating in the CNT/QF filter with different placement positions and the QF filter. Reprinted with permission [63]

Zhao et al. [64] fabricated a novel multifunctional Ag@MWCNTs/Al2O3 hybrid filter with a depth-type hierarchical structure. This filter achieved 99.9999% filtration for particles at 300 nm particle diameter. Ag@MWCNTs/Al2O3 had only 35.60% of the pressure drop of the pristine Al2O3 filter, leading to a high quality factor.

Yang et al. [65] have proposed growing CNTs on an activated carbon fiber (ACF) filter support using chemical vapor deposition in a form of hierarchical structure. The design goal is to achieve high removal efficiencies with low pressure drop and thus a superior quality factor. Indeed, compared with recent CNT-based filter media, CNTs grown on ACF achieved high particle filtration performance with a very low pressure drop, leading to a high quality factor. The results are shown in Fig. 10.
Fig. 10

(a) SEM images of the CNTs/ACF. (b) PM filtration efficiencies and (c) quality factors of the pristine ACF and the CNTs/ACF. Coordinates in parentheses represent the MPPS and associated filtration efficiency. (Reprinted with permission [65])

Nonetheless, in general, while CNT-based air filters have been used with successful results, in-depth knowledge, including their behavior in industrial-scale applications is lacking (Table 5).
Table 5

The filtration performance of CNTs

Structure type


Particle size (nm)

Air velocity (cm/s)

Pressure drop (kPa)

Filtration efficiency (%)

Quality factor (Pa−1)


CNT coating

Cellulose fiber







Free-standing SWNT film







Aligned sheet

Polypropylene fabric








Glass fiber








Metal filter








Quartz fiber








Quartz fiber







Agglomerate structure








Al2O3 filter















Electrostatic Precipitation

Electrostatic precipitators (ESPs), along with fibrous filters, are the most common particle-control devices in industry. A corona discharge was first used to remove particles from an aerosol in 1824 by Hohlfeld; in 1907 Professor F. G. Cottrell of the University of California Berkeley filed a patent (US Patent 895729) for a device to charge particles and then collect them using electrostatic attraction. ESPs function by using electrophoresis to separate charged particles from an airstream.

A Boltzmann distribution of charged particles exists in any gas, for example, charged particles and gas molecules and transient free electrons. At high electric fields, the ions and free electrons can be accelerated to collide with gas molecules, leading to the ejection of additional electrons and further ionization. The process creates a plasma containing a large number of positive ions and free electrons in the gas. In a typical ESP, a corona discharge is created at an electrode with a high positive potential which accelerates free electrons from the surrounding air. The role of the plasma is to increase charging of aerosol particles so they can be removed using electrostatic force. ESPs can also be operated in negative mode. The electrode is given a large negative charge and emits electrons that attach to particles. Two main types of collection electrodes are used in ESPs: flat plates and tubes (Fig. 11b). The typical space between the electrodes and the plates is 10–15 cm. Particles are separated from the gas stream as it passes between them (Fig. 11a) resulting in the formation of a particulate cake on the plates. Some systems include shakers or hammers to clean the plates. In the tubular configuration, which is more common for wet ESPs, the discharge electrode is located at the center of a tubular collector electrode. The diameter of tubes can be 8–25 cm with a 1–4 m length (Fig. 11b). In this design, the discharge electrode is parallel to the gas flow [5, 70].
Fig. 11

(a) Commercial plate-type ESP (General1). (b) Tube-type electrode ESP. (c) Corona formation in a wire and plate plan. (d) Particle charging [70]

An electrical potential of ~4 kV/cm is used to create the corona discharge, applied between the discharge and collection electrodes. Generally, the wires are charged at 20–100 kV below ground potential. In most cases a negative corona (Fig. 11c) is more stable than a positive corona; however, negative corona plasmas also produce much more ozone which can be an unwanted byproduct. Negative coronas are typically chosen because their stability leads to better performance (e.g., less arcing) and lower cost.

In a negative mode operation, negative ions fill the space outside the corona and then collide with particles, charging them negatively (Fig. 11d). The charged particles are driven onto the positive collector plates by an electrical field. Electrostatic charging of particles occurs via two mechanisms, bombardment charging (important for micron-sized particles) and diffusion charging (important for submicron particles). When the initial distribution has relaxed to equilibrium, the charge on the particles will be proportional to particle surface area, i.e., dp2. As submicron particles (<0.1 μm) have significant diffusional motion, they will be charged significantly by diffusional charging. Thus, ESPs are widely used to control submicron particles.

Table 6 shows many advantages and disadvantages of ESP technologies used to control particles.
Table 6

Advantages and disadvantages of ESPs [71]



Easy operation for high-temperature gases, such as boilers and steel furnaces

High initial cost

Low pressure drop, reducing energy consumption

Large footprint

High collection efficiency if operated properly

Suitable for combustible particles

Effective operation over a large range of particulate sizes

Longer chambers are required for particles with high electrical resistivity (>2 × 1011 ohm-cm), their buildup on collectors limits performance

Low operating and maintenance costs if designed and constructed properly

Some particles are slow to give up their charge when they reach the positive electrode plates. They create an insulating layer on the collectors, generating back corona and reducing current flow


Particles with low electrical resistivity cannot be held on the collector; thus re-entrainment can occur

ESP Performance

Deutsch equation is a common description of particle collection efficiency (E) for monodisperse particles:
$$ \eta =1-\exp \left(\frac{-\omega A}{Q}\right) $$
where ω is migration velocity, A is the collector electrode area, and Q is volumetric gas flow. It has been pointed out that Eq. (13) is for particles with similar drift velocities and not for mixtures of particle sizes having different drift velocities. An underlying assumption in the equation is that the particles are permanently collected.
The assumption works well when the collection efficiency is low; for high collection efficiencies (> 99%), other mechanisms besides drift velocity dominate particle collection. Re-entrainment (from rapping and scouring), low electrical resistivity, and uneven flow through the filter can limit collection efficiency leading to particle collection slower than predicted with Eq. (13). Empirical modifications to the equations have been made, to describe the behavior of ESPs with ultrahigh collection efficiency. The Hazen equation (Eq. (14)) and Matts-Ohnfeldt equation (Eq. (15)) build on the Deutsch equation:
$$ \eta =1-{\left(1+\frac{\omega A}{nQ}\right)}^{-n} $$
$$ \eta =1-\exp {\left(\frac{-\omega A}{Q}\right)}^x $$
where n is an empirical constant set at 3 to 5 and x is an empirical constant set at 0.5 based on experimental data [70, 71].

Various studies have shown that the collection efficiency for nanoparticles by ESPs is largely dependent on structural properties including stages of collection, combination with other methods, choice of wet vs. dry ESP, discharge electrode positioning, type of collection electrodes and so on, and specific collection conditions including cleaning gas velocity, type, size, and concentration of particles.

Zhuang et al. [72] have investigated the effect of various operating conditions on the collection efficiency for nanoparticles in the range of 50–500 nm in a cylindrical (tubular) ESP. The ESP was operated in either one- or two-stage mode. It was found that collection efficiencies dropped as the corona current decreased due to a too small drift effect for particles toward the collection electrodes at high air velocities (Fig. 12a). They also found that a lower electrical resistivity of particles caused a decrease in the collection efficiency owing to decreased attractive force at collector surfaces after deposition. Particles of NaCl and Al2O3 showed lower conductivity and ion emission, resulting in poor corona current and thus a decrease in the collection efficiency. The two-stage ESP had a higher collection efficiency than the one-stage ESP (Fig. 21a), and for both ESPs the collection efficiency for nanoparticles in the range of 70–100 nm was better than for other sizes (Fig. 12a, b).
Fig. 12

(a) Collection efficiency of Al2O3 and NaCl particles as a function of size for two different gas velocities in the two-stage ESP. (b) Collection efficiency of Al2O3 and NaCl particles as a function of size for a single-stage ESP. (Reprinted with permission [72])

Kim et al. [73] used a combined electrospray/ESP to enhance the collection efficiency of monodisperse nanometer-sized particles (Fig. 13). The combination enhanced the collection of particles by 21–36% depending on the particle size and in addition reduced the energy consumption of the ESP.
Fig. 13

The combined effect of electrospray and ESP on particle collection efficiency. (Reprinted with permission [73])

There are a number of technical challenges associated with collecting sticky and aggregative soft nanoparticles with a conventional ESP. Dey et al. [74] have designed a positive mode charged wet electrostatic precipitation (WESP) (Fig. 14a) to address them. The goal of their system is to collect soft nanoparticles including monodisperse polystyrene latex (PSL), polydisperse sucrose, and stearic acid particles directly into liquid media. The collection efficiency of WESP was 70–90% for particles of 80–600 nm diameter, in agreement with theoretical modeling, and 40–70% for small particles of 20–80 nm diameter. The performance for these small particles was significantly lower than the theoretical estimate, possibly due to incomplete neutralization of negative charges during air-jet atomization. Figure 14b shows transmission electron microscopy (TEM) images of aggregated nanoparticles on a collecting plate for the dry ESP and nonaggregated nanoparticles on the collecting plate of the WESP.
Fig. 14

(a) Cross-sectional view of fabricated wet electrostatic precipitator (WESP). (b) TEM images of particles collected in dry ESP (left) and WESP (right) (voltage applied on corona wire = +4.2 kV and voltage applied on collecting plate = −3.5 kV). (Reprinted with permission [74])

An efficient wire-to-plate wet electrostatic precipitator (WESP) for nanoparticles was designed by Chen et al. [75], targeting submicron- and micron-sized particles emitted from semiconductor manufacturing. A fixed voltage of −15 kV was supplied by tungsten wire discharge electrodes to produce the electric field and corona ions. Water mist at room temperature was used to quench the high-temperature exhaust gas in order to enhance particle condensation growth and improve the collection efficiency of nanoparticles. The nanoparticle collection efficiency was 67.9–92.9% without fine water mist, which increased to 99.2–99.7% when the water mist was used (Fig. 15a). As shown in Fig. 15b, the results are in good agreement with the formula η = 1 – exp(−α (NDe) β + γ, where α, β, and γ are empirical regression coefficients determined from the experiment and NDe is the Deutsch number.
Fig. 15

(a) Collection efficiency for a WESP with and without fine water mist; the applied voltage is −15 kV. (b) Comparison of particle collection efficiency for the WESP with fine water mist between the experimental data and the present predictions (the applied voltage is −15 kV). (Reprinted with permission [75])

It has been noted [76] that partial charging reduces the collection efficiency for nanoparticles smaller than 30 nm in a traditional wire-in-plate ESP. It is also known that the collection efficiency of ESPs for nanoparticles decreases sharply as operation time increases, because particles deposit on both the collection plates and high-voltage electrode, reducing the electric field [76]. Therefore, both the discharge and collection electrodes should be rapped frequently to remove deposited particle cake. However, rapping may not be practical for smaller devices such as indoor air cleaners. Furthermore, cleaning the wires by water spraying is difficult as it can cause arcing and current bridging between the wires and the collection electrodes [75]. Other problems related to traditional electrodes include particle re-entrainment [72], reverse corona [77], and ozone generation [5]. To overcome these issues, Li et al. [78] developed a new wire-on-plate ESP (WOPEP). Figure 16a compares the traditional and newer ESPs. In WOPEP, discharge wires are attached directly to the surface of a dielectric plate, leading to reduced particle deposition on the wires and reduced ozone generation while maintaining a high particle collection efficiency. The authors achieved collection efficiencies of 90.9–99.7% and 98.8–99.9% in the particle size range of 30–1870 nm, at average face velocities of 0.50 m/s (flow rate, 30 L/min; residence time, 0.36 s) and 0.25 m/s (flow rate, 15 L/min; residence time, 0.72 s). The experimental collection efficiencies were seen to be in agreement with the theoretical models. A key finding was that the WOP ESP, at the same voltage, has one to two orders of magnitude lower ozone generation, and cleaner wires and higher collection efficiency, compared to a traditional ESP (Fig. 16b).
Fig. 16

(a) Schematic of the traditional WIP-EP and proposed WOP-EP. (b) Collection efficiency of the WOP-EP and the WIP-EP at the initial clean condition and after the particle-loaded collection plates were cleaned, with an applied voltage of 18 kV and flow velocity of 0.25 m/s. (Reprinted with permission [78])

Besides the various configurations of ESP structure such as operation mode and voltages, which have a large effect on particle collection performance, there are some important parameters related to the composition of the gas stream that greatly contribute to the collection efficiency of ESPs. The velocity and particle concentration of the gas stream are the most critical operational parameters in these processes. Oliveira et al. [79] separately evaluated the influences of residence time (gas velocities) and concentration of nanoparticles on the performance of a conventional ESP. They observed a decline in collection efficiency with particle concentration and particle diameter (Fig. 17a). An increase in the gas velocity at a fixed particulate feed rate reduced the collection efficiency, which was related not only to the residence time but also to the gas-particle concentration (GAC). To prevent the dilution effect due to increased velocity, they used a particle injection method called aqueous solution concentration (AQC), to quantify the effect of the residence time. Figure 17b shows the decreasing trend of the collection efficiency in this test.
Fig. 17

Grade efficiency vs. particle diameter (a) for a gas velocity of 3.3 cm/s at different KCl AQCs of 0.2, 0.4, and 2 g/L with related GACs of 7.26, 15.55, and 45.78 μg/m3, respectively, and (b) for gas velocities of 6.6, 8.2, and 9.9 cm/s, with KCl AQCs of 4.0, 5.0, and 6.0 g/l, respectively. (Reprinted with permission [79])

Table 7 shows the collection performance of the previously described ESPs, as a function of structural properties and operational conditions.
Table 7

The collection performance of ESPs


Operation mode

Operation stage

Particles size/nm

Air velocity/cm/s

Applied voltage/kV

Collection efficiency/%




Single, two




−7.5 to −10














EHA(−4 kV/cm)/ESP(−16.2 kV)













ESP-fine water

ESP + fine water





−15 to −34




























4 kV/cm





The movement of a particle due to an external force is called phoresis. Examples include heat (thermophoresis) and an electrical field (electrophoresis).

A temperature gradient through an aerosol leads to a thermal gradient across the aerosol particles themselves. As illustrated in Fig. 18a, b, collisions with gas molecules on the high-temperature side of the particle will have more energy than collisions on the cooler side. As a result, particles are accelerated toward the cold side. Thermophoresis is not a strong mechanism compared to the aforementioned methods; however, it can be applied in many aerosol-related fields such as nanoparticle filtration. The mass of a particle depends on the volume or the third power of radius, whereas the number of collisions depends on the surface area, which depends on the second power of radius. Therefore, this mechanism is more important for small particles than for large [70, 81]. Typical temperature gradients are on the order of 20,000 K/m; for UFPs this results in thermophoretic mobilities on the order of Brownian mobilities [70].
Fig. 18

Mechanisms for thermophoretic (a) and diffusiophoretic (b) motion of particles in the gas phase

Researchers have studied this phenomenon both theoretically and experimentally [82, 83, 84]. Tsai and Lu [85] constructed a thermophoretic precipitator to collect nanoscale particles and analyzed its performance. The system built by Gonzales et al. [86] achieved near-complete deposition of NaCl and Fe nanoparticles, for distributions with mean diameters of 50 and 3.6 nm, respectively, in a particle flow below 15 mL/min. Particle filtration was homogeneous regardless of particle size.

Boddu et al. [82] reported a near-complete filtration of carbon nanoparticles with a mean size of 60 nm at a mass density of 1 × 103 μg/m3. They observed that the particle filtration performance decreased to near 70% for 190 nm particles at 1.6 × 104 μg/m3 in a thermophoresis cell. Based on these studies, it can be concluded that thermophoresis is an effective strategy for capturing nanoscale particles [87, 88].

Particle Coagulation

Particles in an aerosol collide due to their differing relative velocities, due to thermal Brownian motion, electro- or thermophoresis, momenta, and so on. Most of the time, colliding particles form an aggregate that coalesces; the process of coagulation changes the size distribution but not the mass density. From the perspective of nanoparticle control, coagulation is a mechanism that could be used to shift material into a size range that is easier to control with standard technologies. Katoshevski et al. have described using coagulation for nanoparticle control [89]. The attractive interaction mechanisms between particles such as electrostatic, Van der Waals, and reduction in surface tension drive the coagulation of nanoparticles. Soot aggregates are one example of the result of such interactions. Mendelevich et al. [90] have tested a geometric approach for clustering nanoparticles in transportation exhaust systems in which the flow velocity across a pipe is manipulated by design. They evaluated the ability of coagulation to reduce the number of nanoparticles as a function of engine speed by comparing the particle number at the outlet of the coagulation pipe relative to the PN of the original pipe. Their results demonstrate that the nanoparticle number is reduced significantly by coagulation in the coagulation pipe (Fig. 19). The authors noted that the nanoparticles probably joined larger particles and that it is also possible to absorb gaseous molecules into the particles, reducing their local concentration and thus suppressing the formation of new particles by nucleation. They concluded that using a well-designed coagulation pipe in the exhaust system of vehicles will contribute to reducing the health effects associated with exposure to nanoparticles and will help vehicles qualify within the new EURO VI regulation framework.
Fig. 19

Particle concentration at the outlet of the original pipe and the coagulation pipe during 30 s of measurement at an engine load of 200 hp. at 1900 rev/min (engine speed). (Reprinted with permission [90])

Acoustic agglomeration is a promising technologies for controlling aerosols, which has been characterized by many researchers [73, 92, 93, 94, 95].

When sound waves travel through an aerosol, particle mutual collision probability is enhanced increasing aggregation. Different particle sizes are affected differently by the pressure wave of sound, leading to different velocities and collision. The newly formed particles in turn continue to collide and grow, leading to nonlinear growth. The acoustic agglomeration can be further enhanced by orthokinetic collision [96], hydrodynamic interaction [92], and Brownian agglomeration [95]. Zu et al. [95] investigated an acoustic chamber for particle agglomeration (Fig. 20a); their model agreed well with the experimental results. The model found that there are many parameters that affect the process including particle size, acoustic frequency, and sound pressure level (SPL). The work showed that the collision efficiency between particles is increased for large particles and for higher SPLs. In addition they observed that the optimal acoustic frequency for enhancing collision, around 1000 Hz, increases with decreasing particle size. Therefore, they suggested that a higher acoustic frequency be used for smaller particles. They found the lower limit for producing an effect via that SPL is affected by the particle concentration and size distribution – lowest effect SPL increases with lower concentration and smaller particle size. Their study indicated that improved acoustic agglomeration can be achieved for a longer residence time of the particles in the agglomeration chamber. SEM images (Fig. 20b) of the particles before and after the acoustic chamber showed the formation of aggregations of particles.
Fig. 20

(a) Experimental setup for acoustic agglomeration of particles. (b) SEM photographs of the particle samples at the outlet of the agglomeration chamber: in the absence of sound (left) and in the presence of sound (right). (Reprinted with permission [95])

Despite a fair number of studies of this method, it seems there are two main challenges that remain. Both the high noise level and high energy consumption inhibit use in many cases.

Another system to enhance the agglomeration of nanoparticles was recently introduced by Zhao et al. [91]. They developed an effective pretreatment agglomeration system to reduce downstream problems related to UFPs. As shown in Fig. 21a, b, Zhao et al. used a simple method to modify the flow in ducts in which two dampers (a batch flow duct to simulate indoor stagnant air) or one damper (a continuous flow duct to simulate a continuous airflow being treated in a downstream cleaner device) was installed. The particle number in the batch chamber was decreased by 73% over 30 min. It was found that the damper movement facilitated particle collision and agglomeration by increasing the flow oscillations, in essence, by increasing turbulence. By fitting a damper with a cycle time of 1 s, the particle capture efficiency of the downstream filters increased from 36% to 48%. The particle size distribution was found to be larger after the damper demonstrating the agglomeration of the particles during the process (Fig. 21c).
Fig. 21

(a) Schematic diagrams of agglomeration chambers, (upper) for a batch process and (bottom) for a continuous flow process with downstream filters. (b) Schematic diagrams of a partial closed damper oscillation in the chamber. (c) Comparison of particle size distributions with and without a damper. (Reprinted with permission [91])

Nanoparticle Detection

Experimental aerosol research dates back more than 100 years ago when John Aitken built a device to count dust particles in air [5]. Today there are many techniques for detecting and characterizing particles that use properties such as mass, optical absorption and scattering, impaction, and electrostatic and diffusional behavior to yield information on the mass and number concentration, and size distribution, of aerosol samples. In some cases the methods are used in combination. In an urban area, as shown in Fig. 22, a majority of particles by number concentration are typically found in the nucleation and Aitken modes. The picture changes, however, if volume or surface distributions are considered. Stanier et al. [98] and Woo et al. [99] reported that nearly 25%, 75%, and 90% of total urban particle number are smaller than 10, 50, and 100 nm, respectively. Similarly, in another study of particle distributions in five European cities, it was found that more than 80% of particles were under the nucleation and Aitken mode ranges [100]. Table 8 presents a summary of the instruments used to measure nanoparticles.
Fig. 22

(a) Number, (b) surface, and (c) volume distribution for a typical urban background aerosol in North Kensington, London, during 24–29 July 2012. The APS/SMPS data set was collected during the ClearfLo Project and merged by the authors. (Reprinted with permission [97])

Table 8

Instruments used to characterize nanoparticles


Commercial model


Particle size range (nm)




APM/DMA (Aerosol Particle Mass Analyzer/Differential Mobility Analyzer)

Kanomax APM-3602

Size-resolved mass concentration



Gives only mass information

The measurement not dependent on particle size or shape


QCM (Quartz Crystal Microbalance)


Size-resolved mass concentration



Ideal for medium and lower concentration measurements down to a few μg/m3


MOUDI (Micro-Orifice Uniform Deposit Impactor (MOUDI)


Size-resolved mass concentration



13 stages-offline characterizing

Collects size-fractionated particle samples for gravimetric and/or chemical analysis


ELPI (Electric Low-Pressure Impactor)

Dekati ELPI+

Size-resolved mass concentration



High time resolution

Able to collect samples for offline analysis

Spherical particle approximation


Laser Aerosol Spectrometer (LAS)

TSI 3340A

Number size distribution



Overcomes the limitations of classic optical particle analyzers


CPC (Condensation Particle Counter)

TSI CPC 3750

Number concentration



High sensitivity for low variation of particle number

High time resolution


EAS (Electrical Aerosol Spectrometer)

Airel EAS

Number size distribution



Spherical particle approximation


SMPS (Scanning Mobility Particle Sizer)

TSI 3938 E57 with 3082 1 nm DMA classifier and CPC 3750

Mass and number size distribution



Better reliability and reproducibility of measurements

Spherical particle approximation


EEPS (Engine Exhaust Particle Sizer)

TSI 3090

Mass and number size distribution



High time resolution (10 times per second)

Low sensitivity EEPS which limits these instruments to high concentration aerosol measurements such as engine exhaust measurements

Considerably lower size resolution compared to SMPS

Spherical particle approximation


FMPS (Fast Mobility Particle Sizer)

TSI 3091 with 3082 1 nm DMA classifier

Mass and number size distribution



High time resolution (one time per second)

Spherical particle approximation


FIMS (Fast Integrated Mobility Spectrometry)

Mass and number size distribution



High time resolution

Spherical particle approximation


P-Trak UPC (Ultrafine Particle Counter)

TSI 8525

Number concentration



Measurement is based on condensation particle counting


Diffusion Battery

TSI 3041

Size distribution



Compact and simple

It does not need particle charging


NSAM (Nanoparticle Surface Aerosol Monitor)


Average (μm2/cm3) and total (μm2) surface area



An NSAM provides a total surface area concentration of particles deposited in a human lung

Underestimates the geometric surface area of particles >100 nm

The surface area underestimated in comparison with APM and TEM


ToF-AMS (Time-of-Flight Aerosol Mass spectrometer)

Aerodyne C-ToF-AMS




Refractory particles cannot detect


(T or S) EM-EDX (Transmission or Scanning electron Microscopy-Energy Dispersive X-ray)

Tecnai G2 F20

Morphology, size, and composition



It is possible to gain information about morphology and size, composition (elements and compounds), and crystallographic


Aerosol Particle Mass Analyzer (APM)

An aerosol particle mass analyzer (APM) was developed by Ehara et al.; an APM instrument is now commercially available from Kanomax (Model 3602). After charging particles in a bipolar charger, the aerosol passes between two rotating coaxial cylindrical electrodes, which rotate at the same angular velocity. Two forces, the centrifugal force and the electrostatic force, affect the particles passing between the electrodes. Only those particles for which the forces balance move the length of the sizer and exit through an annular gap. For a given radius of cylinders (r1, r2), angular velocity (w), particle charge (q), and the voltage between cylinders (V), the mass to charge ratio of particles (mc) can be obtained from [5]:
$$ {m}_c=\frac{qV}{r_c^2{\omega}^2\ln \left(\frac{r_2}{r_1}\right)} $$

For electrodes with the same angular velocity, unbalanced forces may occur in the APM. For example, particles located near the cylinder experience a larger centrifugal force and those near the inner, a smaller centrifugal force. This can lead to deposition of the particles on the outer and inner cylinders, respectively, even at the correct mass to charge ratio.

Additional complications include loss of mass from particles (volatilization) or gain (condensation), of water or other species, which is also in conventional filter-based methods. Volatilization can be encouraged by the pressure drop across the series of impactors. The method also has a relatively slow time resolution [5]. To obtain the desired measurement including size distribution, mass concentration, and chemical characteristics of the particles, a tandem setup including an APM with a differential mobility analyzer (DMA, see section “Electrical-Based Measurement”) is widely used. A coupled DMA/APM is capable of determining the effective density [117], size-resolved mass information [101], specific surface area [118], dynamic shape factor [119], and fractal dimension [119] of an aerosol. The Couette centrifugal particle mass analyzer (CPMA) is a similar technology, developed by Olfert and Collings [120] as a mass classifier similar to the APM. Unlike the APM, the CPMA applies an improved transfer function to the classifier using a stable system of forces. The system has been used recently by Liao et al. [117] and Johnson et al. [121] to determine the size-resolved effective density of rural ambient and cigarette smoke nanoparticles, respectively.


Since the development of impactors in 1860, many impactors have been used in aerosol research [103]. Cascade impactors such as the Andersen, Mercer, quartz crystal microbalance (QCM), Pilat, Berner, low-pressure impactor (LPI) (MOUDI), and the electric low-pressure impactor (ELPI) are commercialized versions [122].

Generally, a cascade impactor consists of several stages where each stage has a porous micro-orifice plate and an impaction plate under it. As a particle-laden stream passes through the stages, particles are impacted and thus collected on the plates. Because of inertia, larger particles are not able to turn with the streamlines and impact due to their forward momentum. The geometry of the impactor can be varied to control the cutoff size of particles that will pass. In the cascade, large particles are trapped first followed by successively smaller particles that are separated in the terminal stages. The size resolution of the instrument is controlled by the number of stages. Among these systems, the MOUDI and ELPI have been used to obtain a smaller cutoff allowing characterization of nanoparticles. Problems with cascade impactors include particle bounce, overloading of particles on the impaction plate, and interstage loss [103]. Some workers have used oil-coated substrates [123] and porous substrates [124] to overcome the problem of particle bounce. Also, to avoid the effect of humidity, Chen et al. [125] suggested controlling the relative humidity (RH) of the incoming aerosol of the MOUDI. Another issue is particle clogging in the nozzles because of long-term or high particle concentration sampling. This problem can increase the pressure drop across the plates; thus, the nozzles need to be cleaned periodically [122].

The MOUDI have found more widespread use than low-pressure impactors because they are equipped with a lower pressure drop-inertial filter for classifying nanoparticles, restricting loss of volatile species [122]. To facilitate much longer operations, a second class of MOUDI called MOUDI-II was developed. This model, as shown in Fig. 23a, uses internal motors to rotate the impaction plates [103], which makes more uniform deposition of particles on the impaction plates, increasing the operational lifetime and reducing the probability of particle bounce.
Fig. 23

(a) 125 nano-MOUDI II with internal motor rotation and (b) typical schematic of a quartz crystal microbalance

The ELPI is an improved low-pressure impactor in which particles are charged with a unipolar charger and the mass aerodynamic size distributions, with a time response below 5 s, are obtained by using electrometers on the stages of a cascade impactor. However, Olfert et al. [126] noted that the ELPI has a poor size resolution for submicron particles and that the sensitivity of the electrometers limits the ELPI to high aerosol concentrations.

Use of the QCM instrument was widespread in the 1980s and continues to be used for many applications. In a QCM, electrically charged particles impact a mechanically oscillating quartz crystal disk and then deposit onto an electrode attached to the center of both sides of the crystal. The resonant frequency of the disk decreases as the particles collect on it (Δf =CΔm). The changing frequency generates a signal that is proportional to the collected mass. Figure 23b shows a typical setup. One of the advantages of the QCM is that it is a direct measurement with high sensitivity and accuracy [5]. The main drawbacks include a high probability of particle bounce due to the high frequency of the electrode, which increases with particle size, and saturation at low mass levels in which case the oscillation frequency does not undergo a significant change. Researchers successfully used a QCM to monitor the deposition of nanoparticles [6, 103, 122].

Optical Measurement

Two physical phenomena, light scattering and light absorption by particles, are the basis of optical-based characterization of particles. Of the two, light scattering has a larger application than extinction. In a light scattering-based device, an aerosol passes across a light beam (usually a laser) where light is scattered by particles and received by a photodetector (Fig. 24). The photoelectric pulse’s frequency determines the number, and its height gives the size distribution of the particles. The scattering phenomena are described by Mie scattering theory for particles with diameter about equal to or larger than the wavelength of light and Rayleigh scattering theory for particles with a diameter about equal to or smaller than the wavelength of light. Also, there are two approaches for light scattering, by single particles or by an assembly of particles. For a high-concentration particle flow, the ensemble techniques are appropriate, while single particle counters are suitable for measuring low particle concentrations [5].
Fig. 24

Optical scattering system of a double-lens laser diffraction instrument. (6) The laser light, (2,4) lens, (3) the sample cell, and (1) the forward-scattering detectors, and (5) backscattering detectors. (Reprinted with permission [127])

A great deal of research and development is built on using the patterns of light scattering for particle characterization. A partial list includes forward-scattering spectrometer probe (FSSP-100) (Knollenberg 1981) and FSSP-300 (Baumgardner 1992), 90° White Light-Scattering Analyzers (Umhauer 1983), Particle Counter Sizer Velocimeter (PCSV) (Holve and Self 1979a), Laser Doppler Velocimetry (LDV) visibility based (Post 1978), Phase Doppler Particle Analyzer (PDPA) (Bachalo and Houser 1984), and Particle Dynamics Analyzer (PDA) (Saffman 1984). Single particle counters have some important limitations including collection efficiency and noise/poor counting statistics at low concentrations, the Rayleigh effect for gas molecules, and saturation at high particle concentration due to coincidence and dead time. Researchers have developed some solutions to address these limitations. It has been found that, overall, small particles scatter light at a larger angle, while larger particles scatter at a smaller angle. Therefore, forward scattering and side scattering are not sufficient to characterize light scattered from nanoparticles. Arakawa et al. [128] used a collimated light beam in a vacuum sample cell to reduce the Rayleigh effect from air molecules, enabling the detection of smaller particles.

Bauer et al. [129] noted that the nanoparticle measurement methods all remove the nanoparticles from their original environment, and the particles will rapidly change in response to the new environment. For example, charging particles, such as in the charging stage of a DMA, will change their behavior, and it will change the particles themselves. A charged particle will have different rates of growth and evaporation relative to a neutral particle. A similar problem can occur in the mass spectrometric analysis methods. In addition, Wang et al. [130] found significant diffusional deposition of nanoparticles at the bends or elbows of instruments, which may distort results. Therefore, recently, Bauer et al. [129] introduced a new technique called synchrotron small-angle X-ray scattering (SAXS) in which high-intensity X-ray beams from synchrotrons make it possible to measure nanoparticles directly in the gas phase, overcoming the drawback. By comparing the results with those of a differential mobility particle sizer (DMPS) operated in parallel, they could show that the SAXS method is able to measure the primary particles and the aggregates, whereas the DMPS measured only aggregates. They noted that in situ direct nanoparticle measurement at ultralow volume fractions of ~10−10 is feasible with SAXS under atmospheric conditions.

Nanoparticles have an extremely low polarizability because of their ultra-small size and are therefore difficult to detect by light scattering-based techniques [127]. There are physical limits, for example, when comparing the wavelength of visible light (ca. 500 nm) with the size of nanoparticles (e.g., 20 nm). Therefore, despite advances with commercial optical particle counters to maximize the collection of scattered light, their performance for nanoparticles is not satisfactory. Furthermore, optical devices determine particle size based on the scattered light intensity. Although this method is faster than other techniques, variations in the refractive index of particles with the same size, which depends on the morphology and chemical composition of the particle, will always cause variations in the scattered light intensity [131].

Condensation Particle Counter (CPC)

Since the size distribution that can be determined from optical techniques is not reliable for particles with a diameter below 100 nm, aerosol science had to find a new approach. The condensation particle counter (CPC) is a widely used optical instrument capable of characterizing aerosols. The method involves three steps: (I) generation of a supersaturation of water vapor or another working fluid, (II) growth of particles by condensation of the vapors, and (III) the optical detection of the enlarged particles, by counting individual pulses of scattered light (Fig. 25). Condensation techniques were used the first time by John Aitken (1888) on atmospheric aerosols [5]. Several models of CPC designed to detect particles within the range of 1 nm to 10 μm have been developed and commercialized by TSI (Thermo-Systems Incorporated). In a common application, a CPC is used in tandem with a differential mobility analyzer (DMA) or a diffusion battery as a detector to determine the size-resolved number concentration.
Fig. 25

Flow schematic of TSI Model of 3772 CPC. (Reprinted with permission [132])

Operational conditions such as aerosol flow rate, saturation rate, pressure drop gradients of the flow paths, the temperature difference between saturator and condenser, type of working fluid, etc. affect the detection efficiency of commercial CPCs. A number of researches have looked into some modifications of CPCs in order to enhance their use for aerosols with diameters below 2 nm. Recent studies have shown that by modifying the operating conditions of, e.g., the TSI 3025A [133], TSI 3772 [134], TSI 3025 [135], and TSI 3010 [136], detection of sub-2 nm particles is possible, with high detection efficiency. The studies commonly focus on modifying the temperature difference between the saturator and the condenser tube [133] and working fluids [137]. These operational conditions strongly affect the probability of the condensational growth of the smaller nanoparticles.

Particle Electrical Mobility

The method of determining the size distribution of an aerosol using particle’s electrical mobility was introduced about 120 years ago. Electrical mobility is perhaps the best and most widely used technique for measuring ultrafine particles. As mentioned above, there are significant obstacles toward using optical techniques for the high-resolution characterization of nanoparticles due to weak light scattering by small particles, smaller than the wavelength of light. However impaction methods, e.g., MOUDI, can use pressures below atmospheric to collect such particles. Over the last few decades, the electrostatic force has been exploited for sizing nanoparticles. As mentioned in section 4, when a particle charged by a corona or gas phase ions is exposed to an electric field, it migrates at a velocity that depends on its size and morphology. A particle’s motion is described by the balance between the applied electrostatic force and aerodynamic resistance (drag). The forces on nanoparticles are made sufficiently high to overcome diffusional effects, yielding high-resolution sizing. The determination of the size distribution of an aerosol using these devices requires knowledge of the charge distribution of the particles, i.e., the charge number on particles with a given diameter [5].

The differential mobility analyzer (DMA), introduced by Knutson and Whitby [138], is the main class of electrical mobility instruments. The fundamental principle of this technique is shown in Fig. 26a. The DMA is a cylindrical classifier in which particles are charged by colliding with a cloud of ions produced by a unipolar high voltage central rod (corona electrode). Particle trajectories then deviate radially from the airstream depending on their size, toward the outer collection electrode. The applied voltage of the corona can be varied between 0 and 5 kV. A particle-free sheath flow passes through the cylinder. Each ring collection electrode is connected to an electrometer for measuring the number of particles and their currents which is related to the particle size.
Fig. 26

(a) Schematic of a typical DMA. (Reprinted with permission [139] and (b) SMPS DMA [5])

The scanning mobility particle sizer (SMPS), now the most common type of mobility analyzer, was developed by Wang and Flagan [140] and commercialized by TSI in 1993. The SMPS system has three key components: the bipolar particle charging chamber, the DMA, and the CPC. Bipolar charging is applied in the charging chamber to create a Boltzmann distribution of charges on the particles. The charged particles then enter the DMA from near the outer electrode. Figure 26b shows a DMA used in commercial SMPS systems. The particles, depending on their charge, are attracted or repelled by the central electrode; the voltage on the electrode may be varied. Particles will thus experience a force depending on their electrical mobility. Particles with a high mobility are quickly deposited on one of the electrodes (depending on polarity). Only particles within a narrow, tunable range of mobility reach the extraction port at the end of the chamber. These selected “monodisperse” particles are then introduced to the CPC to measure their number concentration.

The DMA applies voltages to tune the mobility range, and the voltages are scanned to obtain a size-resolved number concentration. The SMPS is often selected as a reference method in aerosol research. In a classic SMPS program, a scan takes 1 min or longer, introducing a risk that the particle size distribution of the source could change during the course of a single scan [141]. Trostl et al. [142] have obtained a rapid scan time of 3 s with a precision of ±3% using a newer model of SPMS (TSI 3082). However, the scan range was limited to 2.21–60.4 nm. The newest model of these instruments (TSI 3839) provides the scan time of ~15 s.

Nano-sized particle size distributions can also be measured using more rapid instruments such as the Engine Exhaust Particle Sizer (EEPS, TSI 3090) [143], which uses a design based on the electrical aerosol spectrometer (EAS) [107], and the Fast Mobility Particle Sizer (FMPS) (TSI 3091) [110]. The time resolutions for the FMPS and the EEPS are 1 and 10 Hz, respectively. While the EEPS has been designed for engine emissions, both can be used for industrial measurements [143].

Asbach et al. [144] compared the performance of two TSI SMPS, one TSI FMPS, and one Grimm SMPS. The FMPS measured a more narrow distribution and lower concentrations for NaCl aerosols, while for diesel soot, it showed a broader distribution and higher concentrations, relative to the TSI instruments. The authors noted that this difference was probably due to the different particle morphologies or particle size-dependent effects. The SMPS recorded consistent results for both particle sources. In addition, they found that the FMPS underestimated the nanoparticle size distribution by approximately 15% compared to the SMPS. The Grimm SMPS found broader distributions and higher concentrations than the TSI instruments. In a similar study comparing the performance of an EEPS with a CPC and an SMPS, Johnson et al. [109] found that the particle number measured by the EEPS was 50% higher than that measured by the CPC. For size distributions the number concentration measured by the EEPS was less than the SMPS for particles larger than 80 nm, in agreement with the results obtained by Asbach et al. [144]. In another study the determinations of an FMPS and an SMPS did not have a significant difference in their size distributions below 200 nm, but for larger sizes, the FMPS delivered unreliable results [145].

An important disadvantage for fast-sizing instruments is the low electrometer sensitivity for high particle concentrations [141]. Also, the broad unipolar charge distribution used in these electrometer-based analyzers means that they have a lower resolution than the SMPS [126].

The development of a new class of device called fast integrated mobility spectrometry (FIMS) described by Kulkarni et al. [146] may eliminate the need for voltage scanning and increase the time resolution of the classic SMPS. As described by Kulkarni et al. [146], in a FIMS, particles are charged and electrically separated into different streamlines. The separated particles are introduced into a condenser and grow into larger droplets. The droplets are subsequently exposed to a pulse of laser light and their image captured by a CCD array. The images are processed to determine the number of particles and the particle distance from the electric field, which directly depend on particle mobility. Counting the particles as a function of their mobility means that the FIMS is able to measure the number size distribution of particles at a comparatively impressive time resolution, much faster than a classic SMPS. The ability to detect single particles with the FIMS allows it to measure distributions with a higher signal/noise ratio than for electrometer-based instruments [146].

With the exception of a few portable detectors, e.g., the NanoScan SMPS (TSI 3910), commercial mobility analyzers are heavy, large, costly, and complex. Some of the research goals such as personal exposure monitoring, deployment on small aircraft, and spatial gradient mapping, which require an array of sensors in the urban environment, cannot be approached with current tools. This need has driven research on newer approaches for the particle mobility analyzer. These instruments include the opposed migration aerosol classifier (OMAC) [147], cross-flow ion mobility spectrometer (CIMS) [148], and a miniature electrical-mobility aerosol spectrometer (MEAS) [149]. The OMAC is similar to a DMA in that the opposing forces of aerodynamic drag and electrostatic force are used to sort particles into size bins. Particles of too-high and too-low mobility are deposited on the porous electrodes; only those particles with an electrostatic migration velocity that balances the flow velocity can pass. The system uses an applied voltage of ~1 V, much lower than in a conventional DMA that would use a potential difference of ~30 V at the same flow rate. Also, the OMAC is much smaller than a DMA because the distance between electrodes is only 1 mm for the same operational condition. The CIMS consists of a number of channels between parallel plates that are 1 mm apart. The charging electrodes are located at the edge of the channel so that particles, after passing through these electrodes, migrate across the channels with the sheath flow air. While high concentrations of particles may not be suited for electrometer-based tools (FMPS, EEPS), the CIMS showed a good performance for measuring high-mobility gas ions and/or nanoparticles, at high concentrations. The MEAS is a bipartite rectangular chamber containing an electrostatic precipitator and a classifier. Particles are charged in the electrostatic section and subsequently injected into the classifier region with a narrow-range streamline at the desired location. The charged particles are then separated based on their electrical mobilities and collected on plates. The plates are located inside of classifier and are connected to electrometers. These electrometers quantify current signals that are proportional to the number of particles collected on the plates [149]. The structure allows the detection of single particles over a wide range of mobilities.

The equivalent aerodynamic diameter is only a good approximation for spherical particles; this fact limits the performance of any mobility-based analysis method.

Particle Diffusion Mobility

The diffusion battery is a simple and well-characterized diffusion-based aerosol classification method that is free from some of the limitations of the techniques discussed so far. Diffusion batteries were developed to describe aerosols according to their diffusional mobilities [150]. The diffusion battery is a versatile, compact, and simple system that can be easily cleaned after use and does not require particle charging. A diffusion battery comprises a stack of screens and air is sampled at different stages of the flow. The aerosol is passed through the screens, and small particles with a large diffusion constant are more likely to deposit; the particle diameter can be derived from the particle size-dependent deposition rate. The technique is one of a few methods that can measure nanoparticles down to 0.8 nm, i.e., a cluster of just a few molecules. To achieve the size-resolved concentration of nanoparticles, diffusion batteries are often accompanied by a CPC, but this combination limits its use as a personal monitor [151]. Diffusion battery systems have been made in several designs. Wire screen batteries are most commonly used to determine size distributions. Diffusion mobility-dependent size distributions have been used in countless studies (see, e.g., [113, 151, 152, 153, 154]).

Measurements of Nanoparticle Composition and Morphology

Measurements of the composition and morphology of airborne nanoparticles central to determinations of pollution sources, toxicity, and atmospheric behavior. The most common techniques are aerosol mass spectrometer (AMS), scanning electron microscopy (SEM), and transmission electron microscopy (TEM).

An AMS designed and developed by Aerodyne Research, Inc. (ARI), is used worldwide to determine the size-resolved mass concentrations of non-refractory aerosol. At first, the Aerodyne AMS was useful for measuring the ensemble average data of the composition of the mass size distributions of fine particles with aerodynamic diameters of ~50–1000 nm, based on a quadrupole mass spectrometer [155]. With time the sensitivity and time resolution of the AMS were developed along with its ability to analyze single particles, along with a shift to time-of-flight (ToF)-AMS which provide higher mass resolution. The instrument has three main sections: the aerosol inlet (particle beam generation), the particle size discrimination chamber, and the composition measurement chamber. Fine particles are focused into a particle beam by an aerodynamic lens in the entrance section and then pass through a sizing chamber. After that, the beam enters the particle composition detection chamber where particles reach a hot surface (~600 °C) causing them to vaporize. The chemical composition of the vapor is determined by electron impact ionization (EI) and mass spectrometry [156].

Previous studies have shown the technique can be used to measure the size-resolved mass compositions of particles [115, 155, 156, 157]. Jimenez at al. [155] used a ToF-AMS for the study of atmospheric fine particles. This study determined the size-resolved mass composition ratios and chemically resolved mass distribution for atmospheric fine particles. Also, they observed that mass concentrations of sulfate and nitrate measured with the AMS agree with those measured by ion chromatography-based instruments. TOF-AMS has been developed by Su et al. [157] to measure fine and ultrafine particles, specifically particles measuring less than 100–300 nm in diameter. The aerodynamic size of the particles is determined using light scattered as the particle passes through two continuous-wave laser beams. Onasch et al. [115] used a soot particle AMS (SP-AMS) for the physical and chemical characterization of black carbon. The SP-AMS is equipped with an intracavity laser vaporizer (1064 nm), along with a common resistively heated tungsten vaporizer. They showed that the laser vaporizer could measure both the refractory and non-refractory components.

SEM is a microscopic method that probes a sample with an electron beam and can be used to characterize the morphology and to some extent the composition of nanoparticles deposited on a conductive substrate. A typical SEM includes the electron column, scanning system, and detectors. The electron column operates in vacuum and consists of an electron gun and electromagnetic lenses. A sample is introduced into the column and is irradiated by a beam of electrons generating secondary electrons (SEs), backscattered electrons (BSEs), and X-rays [158]. Generally, the secondary electrons are produced by inelastic scattering, while backscattered electrons are caused by elastic scattering of the electrons by the sample. Normally secondary electrons have a low energy, ca. 50 eV. These electrons are analyzed, for example, using an Everhart-Thornley detector with a scintillator/photomultiplier system, yielding topographic information about the sample. For SE images, the edges of the sample’s elements are usually brighter, due to the intensity of electron emission. The generation of backscattered electrons is related to the atomic number of the elements in the sample with higher atomic number giving a brighter image, providing a rough elemental analysis especially for heavier elements. Backscattered electrons are generated at depths of 0.5–1 μm, and thus the spatial resolution of the images is less than for secondary electrons. The backscattered electrons can be detected by scintillator and solid-state detectors. When secondary electrons are produced by the interaction of the electron beams and the sample atoms, holes are left in the inner shells of the atom. X-rays are emitted when outer shell electrons relax into the vacancy. The frequencies of light are characteristic for specific elements and even contain information about their chemical binding. For technical reasons, these characteristic X-rays are analyzed for all elements except hydrogen and helium. The X-rays are detected by either EDS or wavelength dispersive spectroscopy. In an SEM, Bremsstrahlung X-rays (continuum or background X-rays) also can be produced by the deceleration of the primary beam electrons by the electric field of the nuclei of the sample atoms.

An FESEM (field emission SEM) is an SEM optimized for operation at a low energy range (<5 kV), using an electron gun that produces a low- and high-energy electron beam. This kind of operation produces high resolution and also enables scanning at low potentials [159].

Transmission electron microscopy (TEM) can be used to characterize nanoparticle size, morphology, chemical composition, crystallinity and structure, thermal volatility, and mixing state, because of its resolution down to fractions of a nanometer. A TEM is very similar to an SEM. In TEM, the objective lens is placed after the sample area in order to image electrons that have been transmitted through the sample. These electrons may have been scattered (elastically or inelastically) or transmitted (without any interactions with the sample atoms). If the sample area is thicker, fewer electrons are transmitted and the images will be darker; thus, a thinner sample will make brighter images. Elastically scattered electrons do not lose their energy when interacting with the sample atoms, but the angle at which they scatter, according to Braggs Law, can be used to characterize crystalline regions. Inelastically scattered electrons lose energy due to interactions with the atoms in the sample providing elemental analysis and bonding information [160]. Many examples of the use of SEM and TEM to characterize airborne nanoparticles can be found in the literature [116, 161, 162].

Future Prospects

In this review, the most important and feasible techniques for removing nanoparticles from an airstream have been presented. Despite their very high filtration efficiency, the existing filtration methods have a high pressure drop, resulting in high energy consumption. The electrostatic precipitator does not have this problem; however, there are limitations to its use because the high voltage generates ozone, in addition to its size, capacity, and energy use. There is an inevitable trade-off between purification performance and economic and technical feasibility. As a consequence, there is a continual desire to develop more appropriate and advanced methods. Filtration of nanofibers and nanotubes, with the lowest possible pressure drop, may be possible by designing the arrangement of nanofibers or by using the high-performance monolayers.

The most common and commercially successful nanoparticle-measuring techniques have been reviewed. Based on the available literature, the field of nanoparticle measurement has important gaps including development of new techniques better suited to different environmental conditions, improving the sensitivity of electrometer-based instruments and the response time of charging-based instruments and addressing inability to successfully use light-based and mass-based techniques. As always, it would be desirable to have instruments that are smaller and more robust.


  1. 1.
    Cooper CD, Alley FC (2010) Air pollution control: a design approach. Waveland Press, Long Grove Illinois, USAGoogle Scholar
  2. 2.
    Hinds WC (1999) Aerosol technology: properties, behavior, and measurement of airborne particles. Wiley, New YorkGoogle Scholar
  3. 3.
    Biswas P, Wu C-Y (2005) Nanoparticles and the environment. J Air Waste Manage Assoc 55(6):708–746Google Scholar
  4. 4.
    Mao X, Bai Y, Yu J, Ding B (2016) Flexible and highly temperature resistant polynanocrystalline zirconia nanofibrous membranes designed for air filtration. J Am Ceram Soc 99(8):2760–2768Google Scholar
  5. 5.
    Kulkarni P, Baron PA, Willeke K (2011) Aerosol measurement: principles, techniques, and applications. Wiley, HobokenGoogle Scholar
  6. 6.
    Liu J-Y, Hsiao T-C, Lee K-Y, Chuang H-C, Cheng T-J, Chuang K-J (2018) Association of ultrafine particles with cardiopulmonary health among adult subjects in the urban areas of northern Taiwan. Sci Total Environ 627:211–215Google Scholar
  7. 7.
    Chen R, Hu B, Liu Y, Xu J, Yang G, Xu D, Chen C (2016) Beyond PM2. 5: the role of ultrafine particles on adverse health effects of air pollution. Biochimica et Biophysica Acta (BBA)-General Subjects 1860(12):2844–2855Google Scholar
  8. 8.
    Stafoggia M, Schneider A, Cyrys J, Samoli E, Andersen ZJ, Bedada GB, Bellander T, Cattani G, Eleftheriadis K, Faustini A (2017) Association between short-term exposure to ultrafine particles and mortality in eight European urban areas. Epidemiology 28(2):172–180Google Scholar
  9. 9.
    Tobías A, Rivas I, Reche C, Alastuey A, Rodríguez S, Fernández-Camacho R, de la Campa AMS, de la Rosa J, Sunyer J, Querol X (2018) Short-term effects of ultrafine particles on daily mortality by primary vehicle exhaust versus secondary origin in three Spanish cities. Environ Int 111:144–151Google Scholar
  10. 10.
    Heinzerling A, Hsu J, Yip F (2016) Respiratory health effects of ultrafine particles in children: a literature review. Water Air Soil Pollut 227(1):32Google Scholar
  11. 11.
    Samet JM, Graff D, Berntsen J, Ghio AJ, Huang Y-CT, Devlin RB (2007) A comparison of studies on the effects of controlled exposure to fine, coarse and ultrafine ambient particulate matter from a single location. Inhal Toxicol 19(sup1):29–32Google Scholar
  12. 12.
    Liu Q, Liu D, Chen X, Zhang Q, Jiang J, Chen D-R (2019) A cost-effective, miniature electrical ultrafine particle sizer (mini-eUPS) for ultrafine particle (UFP) monitoring network. Aerosol Air Qual Res (AAQR) 23:24Google Scholar
  13. 13.
    Du Y, Xu X, Chu M, Guo Y, Wang J (2016) Air particulate matter and cardiovascular disease: the epidemiological, biomedical and clinical evidence. J Thorac Dis 8(1):E8Google Scholar
  14. 14.
    Li Y, Chen Q, Zhao H, Wang L, Tao R (2015) Variations in PM10, PM2. 5 and PM1. 0 in an urban area of the Sichuan Basin and their relation to meteorological factors. Atmosphere 6(1):150–163Google Scholar
  15. 15.
    Oberdörster G, Oberdörster E, Oberdörster J (2005) Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles. Environ Health Perspect 113(7):823–839Google Scholar
  16. 16.
    Dhananjayan V, Ravichandran B, Sen S, Panjakumar K (2019) Source, effect, and risk assessment of nanoparticles with special reference to occupational exposure. In: Nanoarchitectonics in biomedicine. William Andrew Publishing, Norwich NY, USA, pp 643–676Google Scholar
  17. 17.
    Knudsen KB, Northeved H, Ek PK, Permin A, Andresen TL, Larsen S, Wegener KM, Lam HR, Lykkesfeldt J (2014) Differential toxicological response to positively and negatively charged nanoparticles in the rat brain. Nanotoxicology 8(7):764–774Google Scholar
  18. 18.
    Li N, Sioutas C, Cho A, Schmitz D, Misra C, Sempf J, Wang M, Oberley T, Froines J, Nel A (2003) Ultrafine particulate pollutants induce oxidative stress and mitochondrial damage. Environ Health Perspect 111(4):455–460Google Scholar
  19. 19.
    Manigrasso M, Vernale C, Avino P (2017) Traffic aerosol lobar doses deposited in the human respiratory system. Environ Sci Pollut Res 24(16):13866–13873Google Scholar
  20. 20.
    Pekkanen J, Timonen KL, Ruuskanen J, Reponen A, Mirme A (1997) Effects of ultrafine and fine particles in urban air on peak expiratory flow among children with asthmatic symptoms. Environ Res 74(1):24–33Google Scholar
  21. 21.
    Tiittanen P, Timonen K, Ruuskanen J, Mirme A, Pekkanen J (1999) Fine particulate air pollution, resuspended road dust and respiratory health among symptomatic children. Eur Respir J 13(2):266–273Google Scholar
  22. 22.
    Brown DM, Wilson MR, MacNee W, Stone V, Donaldson K (2001) Size-dependent proinflammatory effects of ultrafine polystyrene particles: a role for surface area and oxidative stress in the enhanced activity of ultrafines. Toxicol Appl Pharmacol 175(3):191–199Google Scholar
  23. 23.
    Agranovski I (2011) Aerosols: science and technology. Wiley, WeinheimGoogle Scholar
  24. 24.
    Hinds WC (2012) Aerosol technology: properties, behavior, and measurement of airborne particles. Wiley, Los AngelesGoogle Scholar
  25. 25.
    Ardkapan SR, Johnson MS, Yazdi S, Afshari A, Bergsøe NC (2014) Filtration efficiency of an electrostatic fibrous filter: studying filtration dependency on ultrafine particle exposure and composition. J Aerosol Sci 72:14–20Google Scholar
  26. 26.
    Zhang R, Wei F (2019) High-efficiency particulate air filters based on carbon nanotubes. In: Nanotube superfiber materials. William Andrew Publishing, Norwich NY, USA, pp 643–666Google Scholar
  27. 27.
    Viswanathan G, Kane DB, Lipowicz PJ (2004) High efficiency fine particulate filtration using carbon nanotube coatings. Adv Mater 16(22):2045–2049Google Scholar
  28. 28.
    Chen CY (1955) Filtration of aerosols by fibrous media. Chem Rev 55(3):595–623Google Scholar
  29. 29.
    Zhang S, Rind NA, Tang N, Liu H, Yin X, Yu J, Ding B (2019) Electrospun nanofibers for air filtration. In: Electrospinning: nanofabrication and applications. William Andrew Publishing, Norwich NY, USA, pp 365–389Google Scholar
  30. 30.
    Wang C-S, Otani Y (2012) Removal of nanoparticles from gas streams by fibrous filters: a review. Ind Eng Chem Res 52(1):5–17Google Scholar
  31. 31.
    Sambaer W, Zatloukal M, Kimmer D (2011) 3D modeling of filtration process via polyurethane nanofiber based nonwoven filters prepared by electrospinning process. Chem Eng Sci 66(4):613–623Google Scholar
  32. 32.
    Gibson P, Schreuder-Gibson H, Rivin D (2001) Transport properties of porous membranes based on electrospun nanofibers. Colloids Surf Physicochem Eng Aspects 187:469–481Google Scholar
  33. 33.
    Kadam VV, Wang L, Padhye R (2018) Electrospun nanofibre materials to filter air pollutants–a review. J Ind Text 47(8):2253–2280Google Scholar
  34. 34.
    Liu Q, Zhu J, Zhang L, Qiu Y (2018) Recent advances in energy materials by electrospinning. Renew Sust Energ Rev 81:1825–1858Google Scholar
  35. 35.
    Leung WW-F, Hung C-H, Yuen P-T (2010) Effect of face velocity, nanofiber packing density and thickness on filtration performance of filters with nanofibers coated on a substrate. Sep Purif Technol 71(1):30–37Google Scholar
  36. 36.
    Zhang S, Liu H, Zuo F, Yin X, Yu J, Ding B (2017) A controlled design of ripple-like polyamide-6 nanofiber/nets membrane for high-efficiency air filter. Small 13(10):1603151Google Scholar
  37. 37.
    Wan H, Wang N, Yang J, Si Y, Chen K, Ding B, Sun G, El-Newehy M, Al-Deyab SS, Yu J (2014) Hierarchically structured polysulfone/titania fibrous membranes with enhanced air filtration performance. J Colloid Interface Sci 417:18–26Google Scholar
  38. 38.
    Choi H-J, Kumita M, Hayashi S, Yuasa H, Kamiyama M, Seto T, Tsai C-J, Otani Y (2017) Filtration properties of nanofiber/microfiber mixed filter and prediction of its performance. Aerosol Air Qual Res 17:1052–1062Google Scholar
  39. 39.
    Tonin C, Aluigi A, Varesano A, Vineis C (2010) Keratin-based nanofibres. InTechOpen, London, UK,  https://doi.org/10.5772/8151, pp 139–158
  40. 40.
    Grafe TH, Graham KM (2003) Nanofiber webs from electrospinning. In: Nonwovens in filtration-fifth international conference, Stuttgart, 2003, pp 1–5Google Scholar
  41. 41.
    Morozov VN, Mikheev AY (2012) Water-soluble polyvinylpyrrolidone nanofilters manufactured by electrospray-neutralization technique. J Membr Sci 403:110–120Google Scholar
  42. 42.
    Liu Y, Park M, Ding B, Kim J, El-Newehy M, Al-Deyab SS, Kim H-Y (2015) Facile electrospun polyacrylonitrile/poly (acrylic acid) nanofibrous membranes for high efficiency particulate air filtration. Fiber Polym 16(3):629–633Google Scholar
  43. 43.
    Wang Z, Zhao C, Pan Z (2015) Porous bead-on-string poly (lactic acid) fibrous membranes for air filtration. J Colloid Interface Sci 441:121–129Google Scholar
  44. 44.
    Li X, Kong H, He J (2015) Study on highly filtration efficiency of electrospun polyvinyl alcohol micro-porous webs. Indian J Phys 89(2):175–179Google Scholar
  45. 45.
    Wang N, Zhu Z, Sheng J, Al-Deyab SS, Yu J, Ding B (2014) Superamphiphobic nanofibrous membranes for effective filtration of fine particles. J Colloid Interface Sci 428:41–48Google Scholar
  46. 46.
    Zhang S, Liu H, Yu J, Luo W, Ding B (2016) Microwave structured polyamide-6 nanofiber/net membrane with embedded poly (m-phenylene isophthalamide) staple fibers for effective ultrafine particle filtration. J Mater Chem A 4(16):6149–6157Google Scholar
  47. 47.
    Li P, Wang C, Zhang Y, Wei F (2014) Air filtration in the free molecular flow regime: a review of high-efficiency particulate air filters based on carbon nanotubes. Small 10(22):4543–4561Google Scholar
  48. 48.
    Wang H, Zheng G-F, Wang X, Sun D-H (2010) Study on the air filtration performance of nanofibrous membranes compared with conventional fibrous filters. In: 2010 IEEE 5th international conference on nano/micro engineered and molecular systems, 2010. IEEE, pp 387–390Google Scholar
  49. 49.
    Kim K, Lee C, Kim IW, Kim J (2009) Performance modification of a melt-blown filter medium via an additional nano-web layer prepared by electrospinning. Fiber Polym 10(1):60–64Google Scholar
  50. 50.
    Wang S-X, Yap CC, He J, Chen C, Wong SY, Li X (2016) Electrospinning: a facile technique for fabricating functional nanofibers for environmental applications. Nanotechnol Rev 5(1):51–73Google Scholar
  51. 51.
    Zhang Q, Welch J, Park H, Wu C-Y, Sigmund W, Marijnissen JC (2010) Improvement in nanofiber filtration by multiple thin layers of nanofiber mats. J Aerosol Sci 41(2):230–236Google Scholar
  52. 52.
    Yun KM, Hogan CJ Jr, Matsubayashi Y, Kawabe M, Iskandar F, Okuyama K (2007) Nanoparticle filtration by electrospun polymer fibers. Chem Eng Sci 62(17):4751–4759Google Scholar
  53. 53.
    Hung C-H, Leung WW-F (2011) Filtration of nano-aerosol using nanofiber filter under low Peclet number and transitional flow regime. Sep Purif Technol 79(1):34–42Google Scholar
  54. 54.
    Wang S, Zhao X, Yin X, Yu J, Ding B (2016) Electret polyvinylidene fluoride nanofibers hybridized by polytetrafluoroethylene nanoparticles for high-efficiency air filtration. ACS Appl Mater Interfaces 8(36):23985–23994Google Scholar
  55. 55.
    Bortolassi ACC, Nagarajan S, de Araújo Lima B, Guerra VG, Aguiar ML, Huon V, Soussan L, Cornu D, Miele P, Bechelany M (2019) Efficient nanoparticles removal and bactericidal action of electrospun nanofibers membranes for air filtration. Mater Sci Eng C 102:718–729Google Scholar
  56. 56.
    Liu C, Hsu P-C, Lee H-W, Ye M, Zheng G, Liu N, Li W, Cui Y (2015) Transparent air filter for high-efficiency PM 2.5 capture. Nat Commun 6:6205Google Scholar
  57. 57.
    Vitchuli N, Shi Q, Nowak J, McCord M, Bourham M, Zhang X (2010) Electrospun ultrathin nylon fibers for protective applications. J Appl Polym Sci 116(4):2181–2187Google Scholar
  58. 58.
    Kuo Y-Y, Bruno FC, Wang J (2014) Filtration performance against nanoparticles by electrospun nylon-6 media containing ultrathin nanofibers. Aerosol Sci Technol 48(12):1332–1344Google Scholar
  59. 59.
    Wang N, Yang Y, Al-Deyab SS, El-Newehy M, Yu J, Ding B (2015) Ultra-light 3D nanofibre-nets binary structured nylon 6–polyacrylonitrile membranes for efficient filtration of fine particulate matter. J Mater Chem A 3(47):23946–23954Google Scholar
  60. 60.
    Iijima S (1991) Helical microtubules of graphitic carbon. Nature 354(6348):56Google Scholar
  61. 61.
    Nasibulin AG, Kaskela A, Mustonen K, Anisimov AS, Ruiz V, Kivisto S, Rackauskas S, Timmermans MY, Pudas M, Aitchison B (2011) Multifunctional free-standing single-walled carbon nanotube films. ACS Nano 5(4):3214–3221Google Scholar
  62. 62.
    Yildiz O, Bradford PD (2013) Aligned carbon nanotube sheet high efficiency particulate air filters. Carbon 64:295–304Google Scholar
  63. 63.
    Li P, Zong Y, Zhang Y, Yang M, Zhang R, Li S, Wei F (2013) In situ fabrication of depth-type hierarchical CNT/quartz fiber filters for high efficiency filtration of sub-micron aerosols and high water repellency. Nanoscale 5(8):3367–3372Google Scholar
  64. 64.
    Zhao Y, Low Z-X, Feng S, Zhong Z, Wang Y, Yao Z (2017) Multifunctional hybrid porous filters with hierarchical structures for simultaneous removal of indoor VOCs, dusts and microorganisms. Nanoscale 9(17):5433–5444Google Scholar
  65. 65.
    Yang S, Zhu Z, Wei F, Yang X (2017) Carbon nanotubes/activated carbon fiber based air filter media for simultaneous removal of particulate matter and ozone. Build Environ 125:60–66Google Scholar
  66. 66.
    Park JH, Yoon KY, Na H, Kim YS, Hwang J, Kim J, Yoon YH (2011) Fabrication of a multi-walled carbon nanotube-deposited glass fiber air filter for the enhancement of nano and submicron aerosol particle filtration and additional antibacterial efficacy. Sci Total Environ 409(19):4132–4138Google Scholar
  67. 67.
    Karwa AN, Tatarchuk BJ (2012) Aerosol filtration enhancement using carbon nanostructures synthesized within a sintered nickel microfibrous matrix. Sep Purif Technol 87:84–94Google Scholar
  68. 68.
    Li P, Wang C, Li Z, Zong Y, Zhang Y, Yang X, Li S, Wei F (2014) Hierarchical carbon-nanotube/quartz-fiber films with gradient nanostructures for high efficiency and long service life air filters. RSC Adv 4(96):54115–54121Google Scholar
  69. 69.
    Wang C, Li P, Zong Y, Zhang Y, Li S, Wei F (2014) A high efficiency particulate air filter based on agglomerated carbon nanotube fluidized bed. Carbon 79:424–431Google Scholar
  70. 70.
    Schnelle KB Jr, Dunn RF, Ternes ME (2015) Air pollution control technology handbook. CRC press, Boca RatonGoogle Scholar
  71. 71.
    Colls J, Tiwary A (2017) Air pollution: measurement, modelling and mitigation. CRC Press, London/New YorkGoogle Scholar
  72. 72.
    Zhuang Y, Kim YJ, Lee TG, Biswas P (2000) Experimental and theoretical studies of ultra-fine particle behavior in electrostatic precipitators. J Electrost 48(3–4):245–260Google Scholar
  73. 73.
    Kim J-H, Lee H-S, Kim H-H, Ogata A (2010) Electrospray with electrostatic precipitator enhances fine particles collection efficiency. J Electrost 68(4):305–310Google Scholar
  74. 74.
    Dey L, Venkataraman C (2012) A wet electrostatic precipitator (WESP) for soft nanoparticle collection. Aerosol Sci Technol 46(7):750–759Google Scholar
  75. 75.
    Chen T-M, Tsai C-J, Yan S-Y, Li S-N (2014) An efficient wet electrostatic precipitator for removing nanoparticles, submicron and micron-sized particles. Sep Purif Technol 136:27–35Google Scholar
  76. 76.
    Huang S-H, Chen C-C (2003) Loading characteristics of a miniature wire-plate electrostatic precipitator. Aerosol Sci Technol 37(2):109–121Google Scholar
  77. 77.
    Jaworek A, Czech T, Rajch E, Lackowski M (2006) Laboratory studies of back-discharge in fly ash. J Electrost 64(5):326–337Google Scholar
  78. 78.
    Li Z, Liu Y, Xing Y, Tran T-M-P, Le T-C, Tsai C-J (2015) Novel wire-on-plate electrostatic precipitator (WOP-EP) for controlling fine particle and nanoparticle pollution. Environ Sci Technol 49(14):8683–8690Google Scholar
  79. 79.
    de Oliveira AE, Guerra VG (2018) Influence of particle concentration and residence time on the efficiency of nanoparticulate collection by electrostatic precipitation. J Electrost 96:1–9Google Scholar
  80. 80.
    de Oliveira AE, Guerra VG (2019) Effect of low gas velocity on the nanoparticle collection performance of an electrostatic precipitator. Sep Sci Technol 54(7):1211–1220Google Scholar
  81. 81.
    Naito M, Yokoyama T, Hosokawa K, Nogi K (2018) Nanoparticle technology handbook. Elsevier, AmsterdamGoogle Scholar
  82. 82.
    Boddu S, Gutti V, Meyer R, Ghosh T, Tompson R, Loyalka S (2011) Carbon nanoparticle generation, collection, and characterization using a spark generator and a thermophoretic deposition cell. Nucl Technol 173(3):318–326Google Scholar
  83. 83.
    Brown DP, Biswas P, Rubin SG (1994) Transport and deposition of particles in gas turbines: effects of convection, diffusion, thermophoresis, inertial impaction and coagulation. American Society of Mechanical Engineers, New YorkGoogle Scholar
  84. 84.
    Tsai C-J, Lin J-S, Aggarwal SG, Chen D-R (2004) Thermophoretic deposition of particles in laminar and turbulent tube flows. Aerosol Sci Technol 38(2):131–139Google Scholar
  85. 85.
    Tsai C-J, Lu H-C (1995) Design and evaluation of a plate-to-plate thermophoretic precipitator. Aerosol Sci Technol 22(2):172–180Google Scholar
  86. 86.
    Gonzalez D, Nasibulin AG, Baklanov AM, Shandakov SD, Brown DP, Queipo P, Kauppinen EI (2005) A new thermophoretic precipitator for collection of nanometer-sized aerosol particles. Aerosol Sci Technol 39(11):1064–1071Google Scholar
  87. 87.
    Ladino L, Stetzer O, Hattendorf B, Günther D, Croft B, Lohmann U (2011) Experimental study of collection efficiencies between submicron aerosols and cloud droplets. J Atmos Sci 68(9):1853–1864Google Scholar
  88. 88.
    Sparks LE, Pilat MJ (1970) Effect of diffusiophoresis on particle collection by wet scrubbers. Atmos Environ (1967) 4(6):651–660Google Scholar
  89. 89.
    Katoshevski D, Dodin Z, Ziskind G (2005) Aerosol clustering in oscillating flows: mathematical analysis. Atomization Sprays 15(4):401–412Google Scholar
  90. 90.
    Ruzal-Mendelevich M, Katoshevski D, Sher E (2016) Controlling nanoparticles emission with particle-grouping exhaust-pipe. Fuel 166:116–123Google Scholar
  91. 91.
    Zhao B, Li M, Wang L-Y, Katoshevski D, Chung T-S (2018) Particle grouping and agglomeration assisted by damper oscillation systems. Sep Purif Technol 207:12–19Google Scholar
  92. 92.
    Hoffmann TL, Koopmann GH (1996) Visualization of acoustic particle interaction and agglomeration: theory and experiments. J Acoust Soc Am 99(4):2130–2141Google Scholar
  93. 93.
    Noorpoor A, Sadighzadeh A, Habibnejad H (2013) Influence of acoustic waves on deposition and coagulation of fine particles. Int J Environ Res 7(1):131–138Google Scholar
  94. 94.
    Yuen W, Fu S, Kwan JK, Chao CY (2014) The use of nonlinear acoustics as an energy-efficient technique for aerosol removal. Aerosol Sci Technol 48(9):907–915Google Scholar
  95. 95.
    Zu K, Yao Y, Cai M, Zhao F, Cheng D (2017) Modeling and experimental study on acoustic agglomeration for dust particle removal. J Aerosol Sci 114:62–76Google Scholar
  96. 96.
    Cheng M, Lee P, Berner A, Shaw D (1983) Orthokinetic agglomeration in an intense acoustic field. J Colloid Interface Sci 91(1):176–187Google Scholar
  97. 97.
    Vu TV, Delgado-Saborit JM, Harrison RM (2015) Particle number size distributions from seven major sources and implications for source apportionment studies. Atmos Environ 122:114–132Google Scholar
  98. 98.
    Stanier CO, Khlystov AY, Pandis SN (2004) Ambient aerosol size distributions and number concentrations measured during the Pittsburgh Air Quality Study (PAQS). Atmos Environ 38(20):3275–3284Google Scholar
  99. 99.
    Woo K, Chen D, Pui D, McMurry P (2001) Measurement of Atlanta aerosol size distributions: observations of ultrafine particle events. Aerosol Sci Technol 34(1):75–87Google Scholar
  100. 100.
    von Bismarck-Osten C, Birmili W, Ketzel M, Massling A, Petäjä T, Weber S (2013) Characterization of parameters influencing the spatio-temporal variability of urban particle number size distributions in four European cities. Atmos Environ 77:415–429Google Scholar
  101. 101.
    Rawat VK, Buckley DT, Kimoto S, Lee M-H, Fukushima N, Hogan CJ Jr (2016) Two dimensional size–mass distribution function inversion from differential mobility analyzer–aerosol particle mass analyzer (DMA–APM) measurements. J Aerosol Sci 92:70–82Google Scholar
  102. 102.
    Fomin A, Poliak M, Rahinov I, Tsionsky V, Cheskis S (2013) Combined particle mass spectrometer–quartz crystal microbalance apparatus for in situ nanoparticle monitoring during flame assisted synthesis. Combust Flame 160(10):2131–2140Google Scholar
  103. 103.
    Marple V, Olson B, Romay F, Hudak G, Geerts SM, Lundgren D (2014) Second generation micro-orifice uniform deposit impactor, 120 MOUDI-II: design, evaluation, and application to long-term ambient sampling. Aerosol Sci Technol 48(4):427–433Google Scholar
  104. 104.
    Keskinen J, Pietarinen K, Lehtimäki M (1992) Electrical low pressure impactor. J Aerosol Sci 23(4):353–360Google Scholar
  105. 105.
    Murphy DM, Froyd KD, Bian H, Brock CA, Dibb JE, DiGangi JP, Diskin G, Dollner M, Kupc A, Scheuer EM (2019) The distribution of sea-salt aerosol in the global troposphere. Atmos Chem Phys 19(6):4093–4104Google Scholar
  106. 106.
    Stolzenburg MR, McMurry PH (1991) An ultrafine aerosol condensation nucleus counter. Aerosol Sci Technol 14(1):48–65Google Scholar
  107. 107.
    Mirme A, Noppel M, Peil I, Salm J, Tamm E, Tammet H (1984) Multi-channel electric aerosol spectrometer. In: International commission for cloud physics. 11th international conference on atmospheric aerosols, condensation and ICE nuclei, vol 2, pp 155–159 (SEE N85-32596 21-47)Google Scholar
  108. 108.
    Sioutas C (1999) Evaluation of the measurement performance of the scanning mobility particle sizer and aerodynamic particle sizer. Aerosol Sci Technol 30(1):84–92Google Scholar
  109. 109.
    Johnson T (2003) An engine exhaust particle sizer spectrometer for transient emission particle measurements, A117. In: 7th ETH-ETH conference on combusion generated particlesGoogle Scholar
  110. 110.
    Manigrasso M, Protano C, Martellucci S, Mattei V, Vitali M, Avino P (2019) Evaluation of the submicron particles distribution between mountain and urban site: contribution of the transportation for defining environmental and human health issues. Int J Environ Res Public Health 16(8):1339Google Scholar
  111. 111.
    Wang J, Pikridas M, Spielman SR, Pinterich T (2017) A fast integrated mobility spectrometer for rapid measurement of sub-micrometer aerosol size distribution, part I: design and model evaluation. J Aerosol Sci 108:44–55Google Scholar
  112. 112.
    Baquero T, Shukrallah S, Karolia R, Osammor O, Inkson B (2015) Quantification of airborne road-side pollution carbon nanoparticles. J Phys Conf Ser 644(1):012023. IOP PublishingGoogle Scholar
  113. 113.
    Dubtsov S, Ovchinnikova T, Valiulin S, Chen X, Manninen HE, Aalto PP, Petäjä T (2017) Laboratory verification of aerosol diffusion spectrometer and the application to ambient measurements of new particle formation. J Aerosol Sci 105:10–23Google Scholar
  114. 114.
    Asbach C, Fissan H, Stahlmecke B, Kuhlbusch T, Pui D (2009) Conceptual limitations and extensions of lung-deposited nanoparticle surface area monitor (NSAM). J Nanopart Res 11(1):101–109Google Scholar
  115. 115.
    Onasch T, Trimborn A, Fortner E, Jayne J, Kok G, Williams L, Davidovits P, Worsnop D (2012) Soot particle aerosol mass spectrometer: development, validation, and initial application. Aerosol Sci Technol 46(7):804–817Google Scholar
  116. 116.
    Wang W, Shao L, Guo M, Hou C, Xing J, Wu F (2017) Physicochemical properties of individual airborne particles in Beijing during pollution periods. Aerosol Air Qual Res 17:3209–3219Google Scholar
  117. 117.
    Liao B-X, Gong W-C, Li Z, Tsai C-J (2019) A mass correction method for the aerosol particle mass analyzer to measure the particle mass of sub-50 nm nanoparticles. Aerosol Science and Technology 53(9):1056–1066.Google Scholar
  118. 118.
    Su L, Ou Q, Cao LN, Du Q, Pui DY (2019) A new instrument prototype to measure the geometric surface area of nanoparticles with a time resolution of 1s. J Aerosol Sci 132:32–43Google Scholar
  119. 119.
    Scheckman JH, McMurry PH, Pratsinis SE (2009) Rapid characterization of agglomerate aerosols by in situ mass – mobility measurements. Langmuir 25(14):8248–8254Google Scholar
  120. 120.
    Olfert J, Collings N (2005) New method for particle mass classification—the Couette centrifugal particle mass analyzer. J Aerosol Sci 36(11):1338–1352Google Scholar
  121. 121.
    Johnson TJ, Olfert JS, Cabot R, Treacy C, Yurteri CU, Dickens C, McAughey J, Symonds JP (2015) Transient measurement of the effective particle density of cigarette smoke. J Aerosol Sci 87:63–74Google Scholar
  122. 122.
    Liu C-N, Awasthi A, Hung Y-H, Tsai C-J (2013) Collection efficiency and interstage loss of nanoparticles in micro-orifice-based cascade impactors. Atmos Environ 69:325–333Google Scholar
  123. 123.
    Tsai C-J, Liu C-N, Hung S-M, Chen S-C, Uang S-N, Cheng Y-S, Zhou Y (2012) Novel active personal nanoparticle sampler for the exposure assessment of nanoparticles in workplaces. Environ Sci Technol 46(8):4546–4552Google Scholar
  124. 124.
    Huang C-H, Chang C-S, Chang S-H, Tsai C-J, Shih T-S, Tang D-T (2005) Use of porous foam as the substrate of an impactor for respirable aerosol sampling. J Aerosol Sci 36(11):1373–1386Google Scholar
  125. 125.
    Chen S-C, Tsai C-J, Chen H-D, Huang C-Y, Roam G-D (2011) The influence of relative humidity on nanoparticle concentration and particle mass distribution measurements by the MOUDI. Aerosol Sci Technol 45(5):596–603Google Scholar
  126. 126.
    Olfert JS, Kulkarni P, Wang J (2008) Measuring aerosol size distributions with the fast integrated mobility spectrometer. J Aerosol Sci 39(11):940–956Google Scholar
  127. 127.
    Xu R (2015) Light scattering: a review of particle characterization applications. Particuology 18:11–21Google Scholar
  128. 128.
    Arakawa A, Mori T, Inoue T (2012) Particle counter. Google PatentsGoogle Scholar
  129. 129.
    Bauer P, Amenitsch H, Baumgartner B, Köberl G, Rentenberger C, Winkler P (2019) In-situ aerosol nanoparticle characterization by small angle X-ray scattering at ultra-low volume fraction. Nat Commun 10(1):1122Google Scholar
  130. 130.
    Wang J, Flagan RC, Seinfeld JH (2002) Diffusional losses in particle sampling systems containing bends and elbows. J Aerosol Sci 33(6):843–857Google Scholar
  131. 131.
    Hering SV, McMurry PH (1991) Optical counter response to monodisperse atmospheric aerosols. Atmos Environ Part A 25(2):463–468Google Scholar
  132. 132.
    Kuang C (2016) Condensation particle counter instrument handbook. DOE ARM Climate Research Facility, Washington, DCGoogle Scholar
  133. 133.
    Kuang C, Chen M, McMurry PH, Wang J (2012) Modification of laminar flow ultrafine condensation particle counters for the enhanced detection of 1 nm condensation nuclei. Aerosol Sci Technol 46(3):309–315Google Scholar
  134. 134.
    Kangasluoma J, Ahonen L, Attoui M, Vuollekoski H, Kulmala M, Petäjä T (2015) Sub-3 nm particle detection with commercial TSI 3772 and Airmodus A20 fine condensation particle counters. Aerosol Sci Technol 49(8):674–681Google Scholar
  135. 135.
    Barmpounis K, Ranjithkumar A, Schmidt-Ott A, Attoui M, Biskos G (2018) Enhancing the detection efficiency of condensation particle counters for sub-2 nm particles. J Aerosol Sci 117:44–53Google Scholar
  136. 136.
    Picard D, Attoui M, Sellegri K (2019) B3010: a boosted TSI 3010 condensation particle counter for airborne studies. Atmos Meas Tech 12(4):2531–2543Google Scholar
  137. 137.
    Iida K, Stolzenburg MR, McMurry PH (2009) Effect of working fluid on sub-2 nm particle detection with a laminar flow ultrafine condensation particle counter. Aerosol Sci Technol 43(1):81–96Google Scholar
  138. 138.
    Knutson E, Whitby K (1975) Aerosol classification by electric mobility: apparatus, theory, and applications. J Aerosol Sci 6(6):443–451Google Scholar
  139. 139.
    Intra P, Tippayawong N (2008) An overview of differential mobility analyzers for size classification of nanometer-sized aerosol particles. Songklanakarin J Sci Technol 30:243–256Google Scholar
  140. 140.
    Wang SC, Flagan RC (1990) Scanning electrical mobility spectrometer. Aerosol Sci Technol 13(2):230–240Google Scholar
  141. 141.
    Wang Y, Pinterich T, Wang J (2018) Rapid measurement of sub-micrometer aerosol size distribution using a fast integrated mobility spectrometer. J Aerosol Sci 121:12–20Google Scholar
  142. 142.
    Tröstl J, Tritscher T, Bischof OF, Horn H-G, Krinke T, Baltensperger U, Gysel M (2015) Fast and precise measurement in the sub-20 nm size range using a scanning mobility particle sizer. J Aerosol Sci 87:75–87Google Scholar
  143. 143.
    Johnson T, Caldow R, Pöcher A, Mirme A, Kittelson D (2004) A new electrical mobility particle sizer spectrometer for engine exhaust particle measurements. SAE technical paperGoogle Scholar
  144. 144.
    Asbach C, Kaminski H, Fissan H, Monz C, Dahmann D, Mülhopt S, Paur HR, Kiesling HJ, Herrmann F, Voetz M (2009) Comparison of four mobility particle sizers with different time resolution for stationary exposure measurements. J Nanopart Res 11(7):1593Google Scholar
  145. 145.
    Levin M, Gudmundsson A, Pagels J, Fierz M, Mølhave K, Löndahl J, Jensen K, Koponen I (2015) Limitations in the use of unipolar charging for electrical mobility sizing instruments: a study of the fast mobility particle sizer. Aerosol Sci Technol 49(8):556–565Google Scholar
  146. 146.
    Kulkarni P, Wang J (2006) New fast integrated mobility spectrometer for real-time measurement of aerosol size distribution—I: concept and theory. J Aerosol Sci 37(10):1303–1325Google Scholar
  147. 147.
    Flagan RC (2004) Opposed migration aerosol classifier (OMAC). Aerosol Sci Technol 38(9):890–899Google Scholar
  148. 148.
    Zhang M, Wexler AS (2006) Cross flow ion mobility spectrometry: theory and initial prototype testing. Int J Mass Spectrom 258(1–3):13–20Google Scholar
  149. 149.
    Ranjan M, Dhaniyala S (2007) Theory and design of a new miniature electrical-mobility aerosol spectrometer. J Aerosol Sci 38(9):950–963Google Scholar
  150. 150.
    Knutson EO (1999) History of diffusion batteries in aerosol measurements. Aerosol Sci Technol 31(2–3):83–128Google Scholar
  151. 151.
    Vosburgh DJ, Klein T, Sheehan M, Anthony TR, Peters TM (2013) Design and evaluation of a personal diffusion battery. Aerosol Sci Technol 47(4):435–443Google Scholar
  152. 152.
    Onischuk A, Valiulin S, Baklanov A, Moiseenko P, Mitrochenko V (2018) Determination of the aerosol particle size distribution by means of the diffusion battery: analytical inversion. Aerosol Sci Technol 52(8):841–853Google Scholar
  153. 153.
    Fierz M, Weimer S, Burtscher H (2009) Design and performance of an optimized electrical diffusion battery. J Aerosol Sci 40(2):152–163Google Scholar
  154. 154.
    Burtscher H, Scherrer L, Siegmann H (2001) The electrical diffusion battery for dynamic classification of nanoparticles. In: Proceedings of ETH conference on nanoparticle measurement. BUWAL, BernGoogle Scholar
  155. 155.
    Jimenez JL, Jayne JT, Shi Q, Kolb CE, Worsnop DR, Yourshaw I, Seinfeld JH, Flagan RC, Zhang X, Smith KA (2003) Ambient aerosol sampling using the aerodyne aerosol mass spectrometer. J Geophys Res Atmos 108(D7):8425.  https://doi.org/10.1029/2001JD001213Google Scholar
  156. 156.
    Canagaratna M, Jayne J, Jimenez J, Allan J, Alfarra M, Zhang Q, Onasch T, Drewnick F, Coe H, Middlebrook A (2007) Chemical and microphysical characterization of ambient aerosols with the aerodyne aerosol mass spectrometer. Mass Spectrom Rev 26(2):185–222Google Scholar
  157. 157.
    Su Y, Sipin MF, Furutani H, Prather KA (2004) Development and characterization of an aerosol time-of-flight mass spectrometer with increased detection efficiency. Anal Chem 76(3):712–719Google Scholar
  158. 158.
    Goldstein JI, Newbury DE, Michael JR, Ritchie NW, Scott JHJ, Joy DC (2017) Scanning electron microscopy and X-ray microanalysis. Springer, New YorkGoogle Scholar
  159. 159.
    Zhou W, Wang ZL (2007) Scanning microscopy for nanotechnology: techniques and applications. Springer science & business media, New YorkGoogle Scholar
  160. 160.
    Ferrante R, Boccuni F, Tombolini F, Iavicoli S (2019) Measurement techniques of exposure to nanomaterials in workplaces. In: Nanotechnology in eco-efficient construction, 2nd edn. Woodhead Publishing, Sawston, Cambridge, UK, pp 785–813Google Scholar
  161. 161.
    Gonzalez-Pech NI, Stebounova LV, Ustunol IB, Park JH, Renee Anthony T, Peters TM, Grassian VH (2019) Size, composition, morphology, and health implications of airborne incidental metal-containing nanoparticles. J Occup Environ Hyg 16(6):1–13Google Scholar
  162. 162.
    Ervik TK, Benker N, Weinbruch S, Thomassen Y, Ellingsen DG, Berlinger B (2019) Size distribution and single particle characterization of airborne particulate matter collected in a silicon carbide plant. Environ Sci Process Impacts 21(3):564–574Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2020

Authors and Affiliations

  1. 1.Department of ChemistryUniversity of CopenhagenCopenhagenDenmark

Section editors and affiliations

  • Michael Evan Goodsite
    • 1
  • Matthew S. Johnson
    • 2
    • 3
  • Ole Hertel
    • 4
  • Nana Rahbek Jørgensen
    • 5
  1. 1.School of Civil, Environmental & Mining Engineering, The Australian School of PetroleumThe University of AdelaideAdelaideAustralia
  2. 2.Department of ChemistryUniversity of CopenhagenCopenhagen ØDenmark
  3. 3.AirlabsCopenhagenDenmark
  4. 4.Department of Environmental ScienceAarhus UniversityRoskildeDenmark
  5. 5.Faculty of EngineeringUniversity of Southern DenmarkOdense MDenmark