Penetration of different nanoparticles through melt-blown filter media used for respiratory protective devices

Abstract

The appearance of new dangers in the form of harmful aerosols containing nanoparticles has necessitated work on ensuring effective protection of humans’ respiratory systems. So far, there is no uniform method, which would make it possible to classify filtering devices, for evaluating the efficiency of respiratory protective devices against nanoaerosols. The authors assessed the morphology of various nanoaerosols of NaCl, titanium dioxide in two forms, anatase and rutile, and aluminosilicate based on observations using a scanning microscope. Then the efficiency of filtration of meltblown fabrics with surface densities of 30, 50, 70, and 90 g/m² against those nanoaerosols were tested. The experimental results were subjected to the theoretical analysis based on filtration theory. For all the fabric types, similar filtration efficiencies were obtained irrespective of the morphological structure of the nanoaerosol particles. The theoretical predictions are in good agreement with the experimental results. The conclusion is therefore drawn that equipment can be evaluated based on testing of filtration efficiency against any tested aerosols.

Keywords

Penetration nanoparticles meltblown filter media respiratory protective devices

For a number of years there has been rapid growth in the worldwide usage of nanotechnologies.¹ In surface engineering the application of nanocoatings produces significant changes in the properties of a material, such as adhesion, tribology, optics, and electrostatic properties.^2,3 In biology and medicine they have led to the development of tissue engineering.⁴ Nanoparticles are now also used in less complex technological processes, such as the production of nonscratch spectacle lenses; crack-resistant paints; anti-graffiti wall paints; dirt-resistant materials; self-cleaning windows; the production, processing, and packaging of foodstuffs; and in cosmetology.⁵ Nanoparticles are also commonly produced in such processes as combustion, cleaning, and welding. It should be noted, however, that in spite of the significant advantages of using this type of technology, the development of nanotechnology brings with it many not fully identified dangers, most of which involve diseases transmitted to the respiratory system.⁶ Hence the effective protection of the respiratory system against nanoaerosols has become an important subject of research in the context of investigating of the mechanisms of the filtration of nanoparticles by filter materials, as well as developing materials for use in respiratory protective devices.⁷

In order for it to be used by workers, respiratory protective equipment must be tested according to the relevant procedures. In the European Union the testing of filters and filtering half-masks is conducted in accordance with the methodology described in EN series standards (EN 143:2000 and EN 149:2001 + A1:2009)^8,9 harmonized with Directive 89/686/EEC.¹⁰ The efficiency of the protection provided by these materials is evaluated as an index of penetration with the use of two model aerosols: non-neutralized polydisperse NaCl and paraffin oil particles at a flow rate of 95 L min⁻¹. For NaCl aerosol the diameter of the particles varies from 40 to 1200 nm with an MMD of 600 nm. For polydisperse oil the particle size distribution is a log-normal distribution with a number median Stokes diameter of 400 nm and a GSD of 1.82. Although the given particle size distribution values differ significantly from the dimensions of nanoparticles encountered in workplaces, no new testing methodology that would enable the classification of filter equipment with respect to a range of nanosized particles has been established. Owing to the lack of regulations in this area, manufacturers of respiratory protective equipment are not able to provide users with information to properly select equipment appropriate to the level of hazard present in particular workplaces. In spite of the numerous publications available concerning the evaluation of the commercial respiratory protective equipment in common use in workplaces in the European Union and the United States,^{11, 12} the issue of correct selection of equipment has not been fully addressed.

In order to bring the methodology for filtration efficiency testing against nanoparticles closer to the actual conditions found in workplaces, it is necessary to establish all process dependencies. Numerous studies have been made on the dominant mechanisms of filtration over a wide range of nanoparticles, using the model aerosols of NaCl and silver particles. The authors have undertaken research to determine how the type of aerosol used for testing influences the filtering efficiency of the meltblown filter materials commonly used in respiratory protective equipment. This direction of experimentation is motivated by the conclusions drawn from research¹³ on the effect of polystyrene nanoparticles with positive and negative charges and with diameters of 20, 100, and 120 nm and monolayer carbon nanotubes suspended in the air on epithelial cells in rats, which demonstrated disturbance in their functioning. Those authors concluded that the physicochemical properties of the nanoparticles (i.e. their size, structure, shape and superficial charge) are especially significant factors affecting how they act on the barrier provided by the epithelial layer in the lung alveoli. Moreover, the need to carry out tests using aerosols with various morphological structures is also indicated by the results of studies¹⁴ on nanoparticles with diameters <100 nm collected in the vicinity of major freeways around Los Angeles. Such morphological forms as aggregates, spherical particles, irregularly shaped particles, and particles with numerous inclusions were observed in aerosol samples. It was suggested that in the case of many of these particles, a heterogenic internal mixing occurred. One of the most important conclusions from this research was that a reduction in the concentration of particles, for example due to a greater distance from the source of emission of the primary aerosol, does not reduce the intensity of the processes of collision and the joining of nanoparticles into larger conglomerates. Hence a fundamental objective of the present work was to establish if the proper evaluation of the efficiency of filter equipment requires the conduction of simulation tests using real nanoaerosols.

Experimental method

Tested materials

The tests were carried out on nonwoven filter fabrics produced by a meltblown method, at an experimental stand at the Central Institute for Labour Protection National Research Institute. The production of meltblown fabrics involved blowing off a molten polymer into elementary fibers with various thicknesses and lengths, which settle on the collecting device to form a compact porous nonwoven fabric. The pneumothermal technique enables the formation of filter fabrics used for respiratory protection with surface densities between 10 and 150 g/m². Since low-density fibers have insufficient strength, and those of a high density produce too great resistance to air flow, the following values of nominal densities were selected for use in the experiments: 30—70—50—90 g/m². ^.

Using this wide range of fabric densities made it possible to determine the relationship between penetration of the test aerosol sodium chloride and paraffin oil mist and the resistance to air flow (see Table 2). The deviation from the nominal density for each variant did not exceed ±10 %.

Table 1.

Properties of Malen P F401 polypropylene granulate

Property	Value
Density [g/cm³]	0.905–0.917
Flow rate [g/10 min]	2.4–3.3
Water content [%]	Max. 0.01
Softening point [°C]	145–155
Melting point [°C]	160–170
Limiting viscosity number [cm³/g]	190

Table 2.

Characteristics of meltblown filter media

Filter parameters	Area density (g/m²)
Filter parameters	30	50	70	90
Thickness (m)	0.0011	0.0018	0.0023	0.0030
Pressure drop at linear velocity of 0.15 m/s (Pa)⁹	101.35	138.83	161.07	208.10
Penetration of paraffin oil mist of 0.3 µm at linear velocity of 0.15 m/s (%)⁹	38.10	18.16	17.51	8.59
Packing density, α (−)	0.025	0.025	0.027	0.027
Mean fiber diameter (µm)	2.3	2.3	2.3	2.3

The filter materials were produced from a granulate of F 401 polypropylene, with basic properties given in Table 1.

The surface density of the filter material produced was dependent on the traverse speed of the receiving device where the material was formed. The surface density of the samples was calculated by weighing samples with an area of 500 cm². The standard deviation of the mean value was 10% for each thickness of the nonwoven fabric. The thicknesses of the respective variants of nonwoven fabrics are given in Table 2.

The filtration parameters for produced, nonwoven fabrics were set at the level of first-class protection in respect to European standards EN 143 and EN 149.^9,10 For these standards, the maximum value of the penetration of paraffin oil vapor is 20% at a breathing resistance up to 210 Pa.

Figures 1 and 2 show a polypropylene filter material with a surface density of 90 g/m². The fibers shown in the figures are twisted and form a three-dimensional structure. Their surface is smooth. The study of the porosity and dimensional distribution of the fibers indicates that the structure of the filtration materials tested was identical for all surface densities.

Figure 1.

Polypropylene filter material with a surface density of 90 g/m² at a magnification of 1000.

Figure 2.

Polypropylene filter material with a surface density of 90 g/m² at a magnification of 5000.

The dimensional distribution studies were performed using scanning electron microscopy. Fiber thickness was measured using image analysis software at a magnification of 5000 for 1000 measurements. Table 3 shows example results of the dimensional distribution and porosity of the tested filter material.

Table 3.

Dimensional distribution and porosity of filtering materials of surfaces densities 50 g/m² and 90 g/m²

Filtering material surface density [g/m²]	Min. fiber diameter [µm]	Max. fiber diameter [µm]	Mean fiber diameter [µm]	Standard deviation [µm]	Average size of pores [nm]	Total surface of pores [nm]
90	0.07	8.92	2.68	1.05	12789	3.127
50	0.09	10.21	2.73	1.10	13138	3.219

Test aerosols

The tests used four aerosols obtained from an aqueous solution or suspension of the following substances: sodium chloride, titanium dioxide in anatase form, titanium dioxide in rutile form, and aluminosilicate (marked in Figures 12 –15 as Al-Si), which contained nanoparticles over a particle size range from 7 to 300 nm. The aerosols differed in terms of the density of the material and the morphological structure of the particles. The lightest was sodium chloride with a density of 2.1 g/cm³, the aluminosilicate aerosol had a density of 2.8 g/cm³, and the heaviest was titanium dioxide at 4.2 g/cm³.

The sodium chloride was generated from 0.1% aqueous solution and the other aerosols from a suspension with analogous concentration. All aerosols were produced using a TSI atomizer, at an input air pressure of 2.5 bar.

Figures 3 –6 show the particle size distributions of the test aerosols used.

Figure 3.

Size distribution of NaCl aerosol.

Figure 4.

Size distribution of rutile aerosol.

Figure 5.

Size distribution of anatase aerosol.

Figure 6.

Size distribution of aluminosilicate aerosol.

Figures 7 –10 show atomic force microscope photographs of the nanoparticles of the tested aerosols, enabling their shape and dimensions to be determined. The aerosols were selected in such a way that they comprised nanoparticles of different shapes. This was to reflect the impact of the different tendencies the particles could exhibit in adhering to the surface of fibers, which depends on their nature and morphology.

Figure 7.

NaCl aerosol particle: (a) surface image, (b) 3-D image.

Figure 9.

Rutile aerosol particle: (e) surface image, (f) 3-D image.

Figure 8.

Anatase aerosol particle: (c) surface image, (d) 3-D image.

Figure 10.

Aluminosilicate aerosol particle: (g) surface image, (h) 3-D image.

The nanoparticles of the sodium chloride aerosol (Figure 7) have a cubic shape. The anatase particles (Figure 8) are oval, and the rutile particles are shaped like prisms with rhomboid bases, while the aluminosilicate particles are cylindrical.

The nanoaerosol particles analyzed, therefore, differ in morphology and shape but maintain the fractal dimension at a similar level: 3.

The aerosol particles were neutralized using an ionization neutralizer, but they exhibited different dielectric constants, which for sodium chloride was 5.9, for aluminum silicate 6.8, and for titanium dioxide about 100.

Experimental setup

Penetration tests with aerosols containing nanoparticles, for the produced filter materials and selected aerosols, were carried out using the equipment illustrated and explained in Figure 11.

Figure 11.

Schematic diagram of the experimental set-up: 1. Nanoaerosol generator, 2. Desiccant, 3. Electrostatic charge neutralizer, 4. Testing chamber, 5. Sample holder, 6. Electrostatic classifier of particles, 7. Condensation nanoparticle counter, 8. Personal computer, 9. Compressed air valves, 10. Flow meters, 11. High-efficiency filter, 12. Test sample.

Compressed air was pumped via filters, dryers, and flow meter¹⁰ to a particle generator¹. The generated aerosol was then passed to a desiccant² and a particle electrostatic charge neutralizer³. To obtain a flow rate of 90 l/min, corresponding to the ventilation of the lungs when the human body is under significant physical strain, the aerosol was mixed with an additional air stream through an additional flow meter¹⁰. After mixing, the dried and neutralized aerosol was passed to a chamber⁴ in which the tested filter material¹² was fastened. From the chamber the aerosol was discharged via highly effective industrial filters¹¹. The aerosol samples were collected from the chamber before and after passing through the tested filter and were passed to a TSI 3080 electrostatic particle classifier⁶ and a TSI 3775 condensation nanoparticle counter⁷. With these instruments an analysis of the size distribution and concentration of nanoparticles was made. The measurement range of the system enabled testing of the particles from 7 to 270 nm, divided into 90 measurement classes.

The procedure of the study of the penetration of nanoparticles through filter materials consisted of

the measurement of the particles concentration at the outlet of the empty test (chamber without filter sample),

the measurement of particles concentration in the air after the test chamber with the filter installed, and

a comparison of these two values to calculate the penetration.

The test time was set at seven minutes to obtain an average concentration and distribution of particles in three measurement cycles of 120 seconds each, with a 15-second pause between cycles to allow for the resetting of the electrostatic particle classifier.

The stability of the background had been checked and verified in previous studies involving repeated tenfold measurements of the background. The results confirmed its stability over time and that there were no changes in both the concentration and the dimensional distribution of the nanoparticles of the generated aerosol. A diagram of the measuring stand is shown in Figure 11.

Comparison of theoretical calculations with experimental data

For the experimental data obtained, a comparative analysis was performed using classic filtration theory as described in many research papers.^3,15,16

The equation for the filter penetration P, used to calculate the theoretical curve, had the classical form

P = \exp (- \frac{4 E α}{π \cdot d_{f} (1 - α)} \cdot L)

(1)

where E is the efficiency of a single fiber.

The filter surface density $ρ_{SF}$ was determined by weighing samples with known areas. The packing density of α, a filter of given thickness L, made of polymer of the density ρ was determined using the relationship from equation (2). The resulting value of α was 0.027 for materials with area densities of 70 and 90 g/m² and 0.025 for materials of densities 30 and 50 g/m².

α = \frac{ρ_{SF}}{L ρ}

(2)

Considering that the diameters of the tested nanoparticles ranged from 7 to 270 nm, it was assumed that the basic mechanisms of deposition of particles on the filter fibers were diffusion $E_{D}$ and direct interception E_R.

The efficiency of a single fiber based on the diffusion mechanism is found from an equation based on Peclet’s number, which is defined as follows:

Pe = \frac{d_{f} U_{0}}{D},

(3)

where U₀ is the linear velocity of the aerosol (U₀ = 0.15 m/s), d_f is the fiber diameter, and D is the coefficient of diffusion. The fiber diameter d_f = 2.3 µm was determined using an optical microscope connected to a digital camera and computer.

The coefficient of diffusion is defined as

D = \frac{{kTC}_{c}}{3 π η d_{p}},

(4)

where air viscosity η = 18.2 × 10⁻⁶ Ns/m², the Boltzmann constant k = 1.3806488 × 10⁻²³ J/K, and fluid absolute temperature T = 298 K. The particle diameter d_p was taken to be the central value of the particle counter’s measurement classes.

The Cunningham slip factor C_c was computed on the basis of the equation for d_p within a limit of 100 nm

C_{c} = 1 + \frac{λ}{d} [2.514 + 0.800 \exp (- 0.55 \frac{d}{λ})]

(5)

where λ is a gas mean free path: 68 × 10⁻⁹ m at room temperature.

The efficiency of a single fiber for the diffusion mechanism was calculated from the equation given by Davies¹⁶

E_{D} = 2 {Pe}^{- 2 / 3}

(6)

The efficiency of a single fiber for the mechanism of direct interception was found from

E_{R} = \frac{1}{2 Ku} [2 (1 + R) \ln (1 + R) - (1 + R) + (\frac{1}{1 + R})]

(7)

where R is a dimensionless parameter relating to direct interception, defined as

R = \frac{d_{p}}{d_{f}}

(8)

The Kuwabara hydrodynamic factor, Ku, was computed from

Ku = - \frac{\ln α}{2} - \frac{3}{4} + α - \frac{α^{2}}{4}

(9)

The efficiency of a single fiber based on the mechanisms of diffusion and direct interception, assuming that the two mechanisms are independent, is computed using

E = E_{R} + E_{D} + E_{DR},

(10)

where E_DR is defined as

E_{DR} = \frac{1.24 R^{2 / 3}}{(KuPe)^{1 / 2}}

(11)

The diffusion mechanisms, which affect nanoparticles most intensely, were included in the theoretical computations (based on the data in the literature³) on the model process of the filtration of nanoaerosol particles. This is a direct attachment mechanism, which exerts a significant effect on particles of diameters more than 200 nm. The inertia mechanism, which for particles of diameters up to 200 nm is negligible, was not included. For the high flow rate applied and the small mass of aerosol particles, the gravity mechanism was not included in the analysis, since gravity becomes significant only in respect to particles with diameters greater than 5 µm.

Because the nonwoven fabric used for the tests was produced without the application of an electrostatic charge and the charge of the aerosol particles was neutralized, the electrostatic field forces were not taken into consideration and, hence, neither was the mechanism of deposition involving electrostatic field forces.

Results and discussion

The results of the tests are presented in graphic form. The graphs in Figures 12 –15 show the calculated penetrations of an aerosol containing nanoparticles of the various tested materials depending on the diameter of particles for a selected filter fabric. It was found that there were no significant differences between the penetration results obtained for different aerosols in all of the investigated cases. This is confirmed by the theoretical curve obtained for particular thicknesses of filter material, which fits well with the experimental data and the scanning microscope photographs showing the shape of the aerosol particles (Figures 7 –10).

In Figures 12, 13, and 14, the theoretical curve marked as “Model” is in agreement with the experimental curves within a particle diameter range from 7 to 100 nm. At more than 100 nm, the theoretical curve moves slightly upward, which may be due to a high degree of non-uniformity in the structure of the experimentally obtained fabric and a slight excess electrostatic charge produced during the formation of the fabric. In the case of the fabric of density 90 g/m² (see Figure 15), the theoretical curve agrees with the experimental curves over the entire particle size range from 7 to 250 nm.

Figure 12.

Comparison of experimental data for change in penetration of aerosol nanoparticles through a filter material of area density 30 g/m² and theoretical calculations for the model adopted.

Figure 13.

Comparison of experimental data for change in penetration of aerosol nanoparticles through a filter material of area density 50 g/m² and theoretical calculations for the model adopted.

Figure 14.

Comparison of experimental data for change in penetration of aerosol nanoparticles through a filter material of area density 70 g/m² and theoretical calculations for the model adopted.

Figure 15.

Comparison of experimental data for change in penetration of aerosol nanoparticles through a filter material of density 90 g/m² and theoretical calculations for the model adopted.

For a fabric with density 30 g/m² (see Figure 12), the maximum value of penetration is at a level of 60% for particles in the range of 150–200 nm. For fabric with density 50 g/m² the maximum penetration value is around 45%; for density 70 g/m², it is at a level of 35%; and for density 90 g/m², it is 20%. In all cases there is a clear relationship between penetration and the thickness of the filter material; thus an increase in area density causes an increase in the resistance to air flow and a decrease in penetration, the two basic parameters that determine a material’s filtration properties. These relationships are confirmed by earlier research.¹⁷^–²⁰

In order to confirm the lack of differences between the experimental curves and the theoretical curve, a statistical analysis was performed. All data obtained for a given aerosol was included in the analysis and checked for statistically significant differences among the range of aerosols examined.

Both the analysis of variance ANOVA and the multiple comparisons tests (Scheffe’s and Dunnett's) indicated no statistically significant differences in penetration values for different variants of filter materials (probability >0.05) between the aerosols examined.

The scatter of the experimental data more than 100 nm resulted from the greater scatter of measures obtained in this range for nonwoven fabrics of surface densities 70, 50, and 30 g/m², as shown for the example aerosol, anatase, in Table 4. The mean values of penetration, given in Table 4, were calculated for 10 samples in each case.

Table 4.

Mean value of penetration of anatase nanoaerosol for tested filtering materials

Filtering material surface mass	70 g/m²		50 g/m²		30 g/m²		90 g/m²
Particle size [nm]	Medium penetration [%]	Standard deviation	Medium penetration [%]	Standard deviation	Medium penetration [%]	Standard deviation	Medium penetration [%]	Standard deviation
25	2.26	1.05	2.73	1.24	12.11	2.50	0.70	0.19
50	11.01	3.24	9.97	2.08	33.84	5.27	3.66	1.69
75	22.95	5.27	17.00	4.20	38.69	6.67	8.57	1.41
100	25.23	4.36	29.70	7.70	44.00	5.61	12.91	3.08
150	34.61	8.75	32.23	7.61	44.45	11.58	17.15	5.26
200	31.06	11.53	30.10	9.75	49.11	12.40	20.62	4.96
270	29.31	18.76	27.32	16.34	47.8	17.32	7.06	2.52

The results of the tests show that penetration is not dependent on the shape and size of the particles or on the density of the material from which they are made. This has also been confirmed by tests on the filtration efficiency of polypropylene fibers, carried out using spherical PSL particles and cubic MgO particles, in a diameter range of 50 to 300 nm and at linear velocities of 10 and 20 cm/s.²¹ It was found that, irrespective of their shape, both types of particles had analogous filtration efficiency, which was confirmed in the described experiments for a linear velocity of 15 cm/s.

Filter materials capture particles with sizes between 7 and 250 nm due to two principal mechanisms: diffusion and direct interception. The contribution of other mechanisms, such as gravitation and inertia, is negligibly small. The best fit of the theoretical curve to the experimental data was obtained for particles with diameters in the range from 7 to 100 nm. This applied to all tested nanoaerosols. The results agree with other experimental data for silver nanoparticles.²² This confirms that penetration is independent of the type of nanoaerosol used to evaluate the efficiency of filtration. The dominance of the diffusion mechanism in the filtration of nanoparticles has also been shown in tests of penetration by particles in the range from 4 nanometers to 10 micrometers²³ and in studies of the effect of the uniformity of the filter material for nanoparticle penetration.²⁴

Conclusions

The experiments show that, when evaluating the efficiency of filter materials used in respiratory protective devices against the hazard caused by nanoparticles, it is not significant which test aerosol is selected. Within a particle size range from 7 to 250 nm, the test aerosol may be any of those used in this research. In all cases, penetration was at an analogous level in the range from 7 to 100 nm and was in accordance with theoretical predictions. In this case the shape of particles with nanometric dimensions is not significant for penetration what is in accordance with classical filtration theory cited in this paper.

Footnotes

Acknowledgments

Publication based on results obtained under Operational Programme Innovative Economy (POIG) 2007–2013, Priority 1: “Research and development of modern technologies”; Submeasure 1.1.2: “Strategic programmes of scientific research and development”.

POIG Project 01.01.02-10-018/09 titled: “Innovative polymer and carbon materials for protection of respiratory system against nanoparticles, vapours and gases.”

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Teresko J. The history of nanotechnology, http://www.industryweek.com (accessed 1th August 2005).

Castaneda

Terrones

. Synthesis and structural characterization of novel flower-like titanium dioxide nanostructures. Physical 2007; 390(B): 143–146.

Honds

. Aerosol technology, properties, behaviour, and measurement of airborne particles, New York: Wiley Interscience, 1982.

Grass

Limbach

Athanassiou

. Exposure of aerosols and nanoparticle dispersions to in vitro cell cultures: a review on the dose relevance of size, mass, surface and concentration. J Aerosol Sci 2010; 41: 1123–1142.

Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR). The appropriateness of existing methodologies to assess the potential risks associated with engineered and adventitious products of nanotechnologies, SCENIHR/002/05, September 2005.

Yang

Peters

Williams

III . Inhaled nanoparticles: a current review. Int J Pharma 2008; 356: 239–247.

Zhang

Weich

Park

. Improvement in nanofiber filtration by multiple thin layers of nanofiber mats. J Aerosol Sci 2010; 41: 230–236.

Council Directive 89/686/EEC of 21 December 1989 on the approximation of the laws of the Member States relating to personal protective equipment, OJ N C 304, 4 December 1989, 29.

European Standard EN 143:2000, EN 143:2000/A1:2006. Respiratory protective devices—particle filters—requirements, testing, marking.

10.

European Standard EN 149:2001+A1:2009. Respiratory protective devices—filtering half masks to protect against particles—requirements, testing, marking.

11.

Rengasamy

King

Eimer

. Comparison of nanoparticle filtration performance of niosh-approved and CE-marked particulate filtering facepiece respirators. J Occup Environ Hyg 2009; 53: 117–128.

12.

Balazy

Toivola

Reponen

. Manikin-based performance evaluation of N95 filtering-facepiece respirators challenged with nanoparticles. Ann Occup Hyg 2006; 50: 259–269.

13.

Yacobi

Phuleria

Demaio

. Nanoparticle effects on rat alveolar epithelial cell monolayer barrier properties. Toxicol in Vitro 2007; 21: 1373–1381.

14.

Barone

Zhu

. The morphology of ultrafine particles on and near major freeways. Atmos Environ 2008; 42: 6749–6758.

15.

Kasper

Schollmeier

Meyer

. The collection efficiency of a particle-loaded single filter fiber. J Aerosol Sci 2009; 40: 993–1009.

16.

Davies

. Air filtration, New York: Academic Press, 1973.

17.

Hamada

Seto

Otani

. Influence of filter inhomogeneity on air filtration of nanoparticle. Aerosol Air Qual Res 2011; 11: 155–160.

18.

Brochocka

. Characteristics of melt-blown filter materials produced by simultaneous blowing of polymer melt from two extruders. Fibres Text East Eur 2001; 4(35): 66–69.

19.

Gradon L, Podgorski A, and Balazy A. Filtration of nanoparticles in the nanofibrous filters. Filtech Europa 2005—Conference Proceedings, Wiesbaden, Germany, 2005, 2: 178–185.

20.

Podgorski

Balazy

Gradon

. Application of nanofibers to improve the filtration efficiency of the most penetrating aerosol particles in fibrous filters. Chem Eng Sci 2006; 61: 6804–6815.

21.

Boskovic

Agranovski

Braddock

. Filtration of nanosized particles with different shape on oil coated fibers. J Aerosol Sci 2007; 38: 1220–1229.

22.

Kim

SCh

Harrington

Pui

DYH

. Experimental study of nanoparticles penetration through commercial filter media. J Nanopart Res 2007; 9: 117–125.

23.

Huang

Chen

Chang

. Penetration of 4.5 nm to 10 µm aerosol particles through fibrous filter. J Aerosol Sci 2007; 38: 719–727.

24.

Hamada

Seto

Otani

. Influence of filter inhomogeneity on air filtration of nanoparticles. Aerosol Air Qual Res 2011; 11: 155–160.