Method and device for sorting fibers in suspension in an aerosol through the combination of electrostatic and centrifugal forces
11396025 · 2022-07-26
Assignee
Inventors
Cpc classification
B03C3/06
PERFORMING OPERATIONS; TRANSPORTING
International classification
B03C3/06
PERFORMING OPERATIONS; TRANSPORTING
Abstract
The invention consists of a continuous sorting method and device which highlights the trajectory differences to which fibers of different form factors and particles charged under the joint influence of electrical forces and a centrifugal force could be subjected. Thus, according to the sorting method, the conditions exploit this difference in order to recover/collect the fibers separated from the non-fibrous particles present in the same initial aerosol or to sort fibers exhibiting different form factors.
Claims
1. A method for sorting micro- and nano-fibers in suspension in an aerosol containing fibers of different sizes and non-fibrous particles, comprising the following steps: a/charging of the fibers and particles in suspension in the aerosol, by unipolar ion diffusion; b/application of an electrical field between two electrically conductive surfaces of revolution, each being arranged with with a vertical axis and defining a space between the two surfaces or revolution; the electrical field being directed from an outer surface to an inner surface; c/introduction of an aerosol flow from an input on top between the two surfaces of revolution; a flow of air being non-turbulent in the space between the two surfaces of revolution; the air flow circulating from the input on top between the two surfaces of revolution to an output below, delimited by an output tube arranged between the two surfaces of revolution; c′/simultaneously with the step c/, rotation at a given velocity of the two surfaces of revolution and of the output tube; and d/recovery of a part of air flow charged with fibers and circulating inside the output tube; the fibers recovered in the part of air flow being separated from the non-fibrous particles initially present in the aerosol and ejected by centrifugal force out of the output tube.
2. The sorting method according to claim 1, further comprising a step d′/simultaneous with the step d/, whereby a part of air flow charged with non-fibrous particles and circulating inside the output tube is recovered.
3. The sorting method according to claim 2, wherein the two surfaces of revolution and the output tube are cylinders, whereby: the step c/is performed by introduction of the aerosol into an input slit arranged in the space between cylinders and by circulation of an axial flow of filtered air introduced on either side of the slit co-current with the aerosol flow; and the step d/is performed by recovery of the part of the air flow charged with fibers inside the output tube.
4. The sorting method according to claim 2, wherein the two surfaces of revolution are openwork discs with concave circular edge, arranged horizontally, defining the space between them; the input being a duct produced in an axial extension of an outer disc on top thereof; the output tube being composed of a first portion separating the space between discs into two, prolonged by a cylindrical portion along an axis of revolution; a method whereby: the step c/is performed by introduction of the aerosol into the input duct; and the step d/is performed by recovery of the part of air flow charged with fibers inside the output tube.
5. The sorting method according to claim 1, further comprising detection and measurement of concentration in terms of number of fibers in air.
6. The sorting method according to claim 1, further comprising a step d′/simultaneous with the step d/, whereby a part of air flow charged with non-fibrous particles and circulating inside the output tube is recovered separately from the part of air flow charged with fibers.
7. The sorting method according to claim 1, wherein the two surfaces of revolution and the output tube are cylinders, whereby: the step c/is performed by introduction of the aerosol into an input slit arranged in the space between cylinders and by circulation of an axial flow of filtered air introduced on either side of the slit co-current with the aerosol flow; and the step d/is performed by recovery of the part of the air flow charged with fibers inside the output tube.
8. The sorting method according to claim 1, wherein the two surfaces of revolution are openwork discs with concave circular edge, arranged horizontally, defining the space between them; the input being a duct produced in an axial extension of an outer disc on top thereof; the output tube being composed of a first portion separating the space between discs into two, prolonged by a cylindrical portion along an axis of revolution; a method whereby: the step c/is performed by introduction of the aerosol into the input duct; and the step d/is performed by recovery of the part of air flow charged with fibers inside the output tube.
9. A sorting device, comprising: two electrically conductive surfaces of revolution, each having a vertical axis and defining a space between the two surfaces; means for applying an electrical field between the two surfaces, the field being directed from an outer surface to an inner surface; means for introducing an aerosol flow of fibers in suspension in an aerosol likely to contain non-fibrous particles, from an input on top between the two surfaces of revolution; means for rotating at a given velocity the two surfaces of revolution and an output tube arranged coaxially below between the two surfaces of revolution; and means for recovering a part of an air flow charged with fibers, inside the output tube.
10. The device according to claim 9, further comprising means for recovering the part of the air flow charged with non-fibrous particles inside the output tube.
11. The device according to claim 10, further comprising means for recovering the part of the air flow charged with non-fibrous particles inside the output tube, separately from the part of air flow charged with fibers.
12. The device according to claim 9, wherein the two surfaces of revolution are coaxial cylinders, the input being a slit arranged in the space between cylinders, the output tube being a cylinder.
13. The device according to claim 9, wherein the two surfaces of revolution are two disks with concave circular edge, arranged horizontally coaxially to one another defining the space between them; the input being a duct produced in an axial extension of an outer disk on top thereof; the output tube being composed of a first portion separating the space between disks into two, prolonged by a cylindrical portion along an axis of revolution.
14. The device according to claim 13, further comprising means for detection and measurement of concentrations of short asbestos fibers (SAF) and/or of fine asbestos fibers (FAF).
Description
DETAILED DESCRIPTION
(1) Other advantages and features will become more apparent on reading the detailed description, given in an illustrative and nonlimiting manner, with reference to the following figures in which:
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19) Throughout the present application, the terms “vertical”, “bottom”, “top”, “low”, “high”, “below”, “above”, and “height” should be understood with reference to a separation device according to the invention arranged horizontally or vertically.
(20) Likewise, the terms “input”, “output”, “upstream” and “downstream” should be understood with reference to the direction of the flow of aerosol in a device according to the invention. Thus, the input designates a zone of the device through which the aerosol containing the fibers and the non-fibrous particles is introduced whereas that of output designates that through which the air flow charged only with fibers is discharged.
(21) For clarity, the same elements of the sorting devices according to the examples illustrated of the two alternatives are designated by the same numeric references.
(22)
(23) It is specified that previously, before the introduction of the aerosol into the device 1, the particles of the aerosol are charged negatively by unipolar ion diffusion.
(24) The sorting device 1 first of all comprises two parallel flat plates 2, 3, arranged horizontally. These plates 2, 3 are electrically conductive.
(25) At a longitudinal end of the plates 2, 3, there is arranged an input slit 4, in the middle of the space between plates, that is to say the middle of the slit 4 is at half the height h of the space between plates 2, 3. The slit 4 can for example be produced by two plates, also flat and mutually parallel, but over a height much lesser than the space between plates 2, 3.
(26) At the other longitudinal end, there is arranged a separation wall 5, also at the middle of the space between plates 2, 3. This wall 5 therefore delimits, with the plate on top 2, a channel 6, while it delimits, with the plate below 3, a channel 7.
(27) An electrical field E is generated, preferably uniform and preferably of constant intensity, between the plates 2, 3, the field E being directed from bottom to top. For this, for example, the bottom plate 3 is brought to the zero potential, whereas the top plate 2 is at the potential +U. In the context of the invention, it is perfectly possible to envisage the reverse, that is to say particles positively charged with an electrical field in the device equal to −U.
(28) A longitudinal flow of filtered air with non-turbulent flow is introduced from the side of the slit 4, into the space between plates 2, 3. The filtered air flow is separated into a flow q1 between the slit 4 and the plate on top 2 and a flow q2 between the slit 4 and the plate below 3.
(29) The aerosol is then introduced through the slit 4, at a flow rate q.sub.0.
(30) Thus, in its travel between the plates, any particle, including a fibrous one, will be subjected to these two antagonistic force fields, field of gravity g and electrical field E.
(31) Each particle, fibrous or not, will therefore be subjected to two opposing transverse velocities:
(32) an upward velocity due to the electrical field denoted w, such that w=Z*E, where Z is the electrical mobility of the particle, and a downward velocity due to the field of gravity denoted u, such that u=τ*g, where T is the relaxation time of the particle, and g is the Earth's field of gravity.
(33) The trajectory of a particle will therefore result from the composition of these two transverse velocities u and w on the one hand, of its longitudinal velocity v in the non-turbulent flow on the other hand.
(34) For a fixed geometry and flow rate, an appropriate value of the field E can therefore direct the fibers and the fine particles that are highly electrically mobile and not subject to gravity, into the top part of the space between plates 2, 3, and direct the non-fibrous particles into the bottom part 7 of this space, above all the large particles, which have little electrical mobility and are subject to gravity.
(35) It is therefore possible to recover, in the output channel 6, the fibers separated and borne by the air flow at the flow rate Q1.
(36) In parallel, it is possible to recover, in the output channel 7, the fibers exhibiting the lowest form factor or the non-fibrous particles borne by the air flow at the flow rate Q2.
(37) The sum of the input flow rates q.sub.0, q.sub.1 and q.sub.2 equals the sum of the output flow rates Q.sub.1 and Q.sub.2.
(38) Thus separated from the fibers, the large particles can no longer mask the count of the fibers for the asbestos fiber measuring application.
(39) The inventors have corroborated, by calculations presented hereinbelow, the separation between fibers and non-fibrous particles by the combined action of electrical force resulting from a field E created between flat plates, and the Earth's field of gravity g.
(40) In the calculations, the case of carbon fibers is considered, specifically those which were used in the experiments mentioned in the publication [3], of 3.74 μm diameter, charged by unipolar ion diffusion with a product Ni*t=1.9.10.sup.7 s/cm.sup.3. The advantage of using carbon fibers is that their electrical characteristics have been particularly well studied by the authors of the publication. Another advantage is also deliberately choosing conditions conducive to revealing the action of the field of gravity relative to the action of the electrical field.
(41) The trajectory of a particle is obtained by composing the velocities u, v, w, in which:
u=τ*g
w=Z*E,
i.e.
(42)
(43) in which
(44) τ represents the relaxation time of a particle, in seconds (s)
(45) g is the acceleration of gravity, in m/s.sup.2;
(46) Z is the electrical mobility of the particle, in m.sup.2/(V*s);
(47) E is the electrical field in V/m;
(48) Q is the air flow rate driving the particle in m.sup.3/s;
(49) is the width of the air flow circulation channel; and
(50) h is the air flow circulation height.
(51) By eliminating dt, in the equations (1) and (2), the following is obtained:
(52)
(53) In other words by performing the integration
(54)
(55) Hence the final equation (3) as follows:
(56)
(57) To calculate the relaxation time τ.sub.f of a fiber, the equation (4) is used:
(58)
(59) in which:
(60) ρ represents the density equal to 1.832.10.sup.3 kg/m.sup.3 for carbon fibers;
(61) d′=d*(1,5*ß).sup.1/3 and d is equal to 3.74 μm;
(62) η represents the viscosity of air equal to 1.81*10.sup.−5 Pa.Math.s;
(63) χ.sub.f is the form factor dependent on β; and
(64) β is the slenderness (ratio between fiber length and diameter).
(65) By taking into account the experimental data from the publication [3] and according to the equation (4), the table 1 below of fiber characteristics is obtained:
(66) TABLE-US-00001 ß χ.sub.f Z.sub.f in m.sup.2/(V*s ) τ.sub.f in s 10 1.269 7.97*10.sup.−8 3.77*10.sup.−4 20 1.541 11.27*10.sup.−8 4.93*10.sup.−4
It is specified that the experimental data used are valid for N.sub.i*t equal to 1.9*10.sup.13 ions*s/m.sup.3, where N.sub.i is the concentration of unipolar ions and t is the dwell time.
(67) To calculate the electrical field E which allows fibers of factor β equal to 20, to arrive at the top of the space between plates, i.e. closest to the top plate, with x=L, the equation (3) for
(68)
which gives:
(69)
(70) with Q representing the flow rate equal to 2 liters per min; l=5 cm, L=20 cm,
(71) τ.sub.f=4.93*10.sup.−4 s, Z.sub.f=11.27*10.sup.−8 m.sup.2/(V*s) and g=9.81 m/s.sup.2, an electrical field value E equal to 5.76*10.sup.4 V/m is obtained.
(72) By using this value in the equation (3) above, all the elements are there to find the trajectory of the fibers of factor β equal to 20.
(73) For the same value E, it is also possible to find all the elements to find the trajectory of the fibers of factor β equal to 10.
(74) It is possible to proceed and do the same calculations for a volume-equivalent sphere (the indices “se” hereinbelow corresponding to an equivalent sphere).
(75) Let d.sub.sc be the volume diameter of a sphere equivalent to a fiber of diameter d and of length l.sub.f, then the following relationship applies:
(76)
(77) Then, with β which is the ratio between fiber length lf and diameter d, the equation (4) applies:
d.sub.se=d*(1.5*ß).sup.1/3.
(78) For the calculation of the relationship time of the sphere, the equation (5) is used:
(79)
with χ.sub.se equal to 1.
(80) For the calculation of electrical mobility of the spheres, the publication [1] makes it possible to determine it for a product N.sub.i*t equal to 10.sup.13 ions*s/m.sup.3.
(81) It is possible to extrapolate to assume conditions calculated for the fibers, i.e. with N.sub.i*t equal to 1.9*10.sup.13 ions*s/m.sup.3.
(82) To do this, the expression (15.24) on page 325 of the publication [1] is used, which makes it possible to find a multiplying coefficient equal to 1.083.
(83) The table 2 below of characteristics of the equivalent spheres is therefore obtained:
(84) TABLE-US-00002 ß d.sub.se in μm χ.sub.se Z.sub.se m.sup.2/(V*s ) τ.sub.se in s 10 9.24 1 4.52*10.sup.−8 4.78*10.sup.−4 20 11.63 1 4.48*10.sup.−8 7.59*10.sup.−4
(85) The trajectory of these two types of particles, i.e. fibers and equivalent spheres, is illustrated in
(86) It emerges from this
(87)
(88)
(89) In the variant of
(90) In the variant of
(91) It is possible to envisage arranging devices according to the variants of
(92)
(93) In the variant of
(94) The device of
(95) The device of
(96)
(97) In this variant, the flat surfaces between which the electrical field is established are composed of the bottom face 210, of the disk 200, and of the top face of a disk 300 arranged coaxially horizontal one inside the other defining a space of constant thickness between them.
(98) Each of these two disks 200, 300 is openwork and has concave circular edge.
(99) The bottom face 210 of the bottom disk 200 is at least partly a filtering membrane.
(100) The aerosol of charged particles is, here, introduced through a duct 800 produced in the axial extension of the outer disk 300 on the top of the latter then circulates in the space between disks 200, 300.
(101) The separated fibers are collected on the membrane 210, the air transporting them being discharged through an output duct 600 produced in the axial extension above the filtering membrane 210. The output duct 600 can be coaxial to the input duct 800.
(102) Optionally, the non-fibrous particles that fall through gravity can be discharged by the air in an output duct 700 produced in the axial extension of the outer disk 300 below the latter.
(103)
(104) More specifically, the device 1′ comprises, first of all, two electrically conductive coaxial cylinders 2′, 3′.
(105) These cylinders 2′, 3′ are preferably arranged vertically. Such an arrangement makes it possible not to add force of gravity which could disturb the centrifugal force to which the particles are subjected.
(106) A cylindrical slit 4′ is arranged in the space between the cylinders 2′,3′.
(107) An output cylindrical tube 5′ is arranged between the two cylinders 2′, 3′.
(108) A uniform electrical field E is generated, of constant intensity between the cylinders 2′, 3′, the field E being directed radially from outside to inside. To do this, for example, the outer cylinder 3′ is brought to the zero potential, whereas the inner cylinder 2′ is at the potential +U.
(109) The cylinders 2′, 3′ and the cylindrical output tube 5′ are rotated at a rotation velocity {right arrow over (ω)}.
(110) Simultaneously, a longitudinal flow of filtered air with non-turbulent flow is introduced from the side of the slit 4′ and made to circulate in the space between cylinders 2′,3′. The filtered air flow rate is separated into a flow rate q1 between the slit 4′ and the inner cylinder 2′ and a flow rate q2 between the slit 4′ and the outer cylinder 3′.
(111) The aerosol is then introduced through the slit 4′, at a flow rate q.sub.0.
(112) The electrically charged fibers have a greater electrical mobility than the equivalent spherical particles, the electrical field E therefore tends to attract them toward the inner cylinder 2′ since their velocity according to the component w is greater than that according to the component u.
(113) Thus, the flow rate Q1 is fiber-enriched and non-fibrous particle-depleted.
(114) On the other hand, the non-fibrous particles are more subject to the centrifugal force than the fibers and therefore tend to move away from the inner cylinder to approach the outer cylinder 3′.
(115) For a fixed geometry and flow rate, an appropriate value of the field E and of the rotation velocity W, it is therefore possible to direct the fibers and the fine particles that are highly electrically mobile and not subject to the centrifugal force, into the internal part of the space between the cylinders 2′, 3′, and direct the non-fibrous particles into the outer part of this space, above all the large particles, that have little electrical mobility, and are subject to the centrifugal force.
(116) The fibers separated and borne by the air flow at the flow rate Q1 can therefore be recovered in the channel 6′ delimited by the interior of the output tube 5′ and the inner cylinder 2′.
(117) In parallel, the large, non-fibrous particles borne by the air flow at the flow rate Q2 can be recovered in the output channel 7′ delimited by the outside of the output tube 5′ and the outer cylinder 3′.
(118) The sum of the input flow rates q.sub.0, q.sub.1 and q.sub.2 equals the sum of the output flow rates Q.sub.1 and Q.sub.2.
(119)
(120) In this variant, the flat surfaces between which the electrical field is established are composed of the bottom faces 210′, 310′ of two disks 200′, 300′ arranged coaxially horizontal in one another defining a space of constant section between them.
(121) Each of these two disks 200′, 300′ has a concave circular edge and the outer disk 300′ is openwork.
(122) The bottom faces 210′, 310′ of the disks 200, 300 are solid.
(123) An output tube 500′ is produced in the axial extension of the bottom face 310′ of the outer disk 300′. This output tube 500′ is composed of a first portion 510′ separating the space between disks into two, prolonged by a cylindrical portion 520′ along the axis of revolution of the disks.
(124) The aerosol of charged particles is, here, introduced through a duct 800′ produced in the axial extension of the outer disk 300′ above the latter then circulates in the space between disks 200′, 300′.
(125) The disks 200′, 300′ and the output cylindrical tube 500′ are rotated at a rotation velocity W.
(126) The separated fibers are recovered with the air transporting them in the output duct 600′ delimited by the inside of the tube 500′.
(127) The non-fibrous particles subject to the forces, both centrifugal and of gravity, are discharged by the air into the output duct 700′ produced around the tube 500′.
(128)
(129)
(130)
(131) Other variants and enhancements can be made without in any way departing from the scope of the invention.
(132) Thus, if, in the embodiments illustrated, the flow rate Q1 is shown equal to that of Q2 equal to the total flow rate divided by two Q/2, it is perfectly possible to envisage having Q1 different from Q2 and from Q/2.
(133) The same goes for q.sub.0,q.sub.1 and q.sub.2 which can be different from one another and also different from q/3.
(134) Moreover, if, in all the examples illustrated, the flat surfaces of revolution are arranged coaxially one inside the other and define a space of constant thickness, it is perfectly possible to envisage implementing the invention with surfaces of revolution that are not parallel/coaxial and therefore with a space of variable thickness.
(135) The invention is not limited to the examples which have just been described; it is notably possible to combine with one another features of the examples illustrated within variants that are not illustrated.
REFERENCES CITED
(136) [1]: W. Hinds, “Aerosol Technology”, 2.sup.nd Edition, 1999. [2]: Zebel G., Hochrainer D., Boose C., “A sampling method with separated deposition of fibres and other particles”, J. Aerosol Sci., 8:205-213 (1977). [3]: Chen B. T., Yeh H. C., Hobbs C. H., “Size Classification of Carbon Fiber Aerosols”, Aerosol Sci. Technol., 19:109-120 (1993). [4]: Chen B. T., Yeh H. C., Johnson N. F., “Design and use of a virtual impactor and an electrical classifier for generation of test fiber aerosols with narrow size distributions”, J. Aerosol Sci., 27(1):83-94 (1996). [5]: Han R. J., Moss O. R., Wong B. A., “Airborne Fiber Separation by Electrophoresis and Dielectrophoresis: Theory and Design Considerations”, Aerosol Sci. Technol., 21:241-258 (1994). [6]: Griffiths W. D., “The selective separation of aerosol particles of different shapes”, J. Aerosol Sci., 18(6):761-763 (1987). [7]: Griffiths W. D., “The shape selective sampling of fibrous aerosols”, J. Aerosol Sci., 19(6):703-713 (1988). [8]: Baron P. A., Deye G. J., Fernback J., “Length separation of fibers”, Aerosol Sci. Technol., 21:179-192 (1994). [9]: Lilienfeld, P. “Rotational Electrodynamics of Airborne Fibers”, J. Aerosol Sci. 4: 315-322 (1985). [10]:Deye G. J., Gao P., Baron P. A., Fernback J., “Performance Evaluation of a Fiber Length Classifier”, Aerosol Sci. Technol., 30:420-437 (1999). [11]: Dubey P., Ghia U., Turkevich L. A., “Numerical investigation of sheath and aerosol flows in the flow combination section of a Baron fiber classifier”, Aerosol Sci. Technol., 48:896-905 (2014). [12]: Ehara, K., Hagwood, C., Coakley, K. J., “Novel method to Classify Aerosol Particles According to Their Mass-to-Charge Ratio”, Aerosol Particle Mass Analyzer, J. Aerosol Sci, 27:217-234 (1996). [13]: Olfert, J. S., Collings, N. “New Method for Particle Mass Classification: the Couette Centrifugal Particle Mass Analyser”, J. Aerosol Sci, 36:1338-1352 (2005).