Method of forming an aggregate of objects

Abstract

A method of forming an aggregate of objects in a channel including a liquid, the method including: a) providing objects in at least a region of the channel, and b) forming an aggregate of the objects by submitting them to a modulated pulsed acoustic field, wherein the modulated pulsed acoustic field applied at step b) is modulated in amplitude.

Claims

1. A method of forming an aggregate of objects in a channel comprising a liquid, said method comprising: a) providing objects in at least a region of the channel, and b) forming an aggregate of said objects by submitting them to a modulated pulsed acoustic field, wherein the modulated pulsed acoustic field applied at step b) is modulated in amplitude.

2. A method according to claim 1, wherein the modulated pulsed acoustic field comprises a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency, and said groups being separated between each other by a period having a non-zero duration wherein no acoustic wave is applied.

3. A method according to claim 1, wherein the modulated pulsed acoustic field comprises a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency, and said groups being separated between each other by a period having a non-zero duration wherein an acoustic wave is applied, said acoustic wave having at least one extremum of the absolute value of its amplitude that is different from the highest amplitude of the acoustic wave pulses belonging to the group just preceding said period.

4. A method according to claim 1, wherein a standing acoustic wave is created along a transverse dimension of the channel at step b) during the application of the modulated pulsed acoustic field.

5. A method according to claim 4, wherein said transverse dimension is the thickness of the channel.

6. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency, and each of said groups of acoustic wave pulses comprising 10 or more acoustic wave pulses.

7. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a group, the same amplitude and frequency, at least one group of acoustic wave pulses lasting a duration t.sub.1 that is greater than or equal to 0.01 ms.

8. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency and step b) comprising submitting the objects to at least 100 groups of acoustic wave pulses.

9. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency and at least one period separating two successive groups of acoustic wave pulses having a duration t.sub.2 that is greater than or equal to 0.05 ms.

10. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency and at least one period separating two successive groups of acoustic wave pulses having a duration t.sub.2 that is less than or equal to 0.5 s.

11. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency and wherein at least a couple of consecutive group of acoustic wave pulses of duration t.sub.1 and period separating two successive groups of acoustic wave pulses of duration t.sub.2 have a pulse mode factor $P_{\mathrm{mf}}=\frac{t_1}{t_1+t_2}$ that is greater than or equal to 0.01.

12. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency and wherein at least a couple of consecutive group of acoustic wave pulses of duration t.sub.1 and period separating two successive groups of acoustic wave pulses of duration t.sub.2 have a pulse mode factor $P_{\mathrm{mf}}=\frac{t_1}{t_1+t_2}$ that is less than or equal to 0.95.

13. A method according to claim 1, at least two couples of consecutive groups of acoustic wave pulses of duration t.sub.1 and period separating two successive groups of acoustic wave pulses of duration t.sub.2 having a different pulse mode factor $P_{\mathrm{mf}}=\frac{t_1}{t_1+t_2}.$

14. A method according to claim 1, all the couples of consecutive groups of acoustic wave pulses of duration t.sub.1 and period separating two successive groups of acoustic wave pulses of duration t.sub.2 having substantially the same pulse mode factor $P_{\mathrm{mf}}=\frac{t_1}{t_1+t_2}.$

15. A method according to claim 1, the modulated pulsed acoustic field comprising a repetition of a plurality of groups of acoustic wave pulses, said pulses having, among a given group, the same amplitude and frequency and said groups of acoustic wave pulses being periodically spaced between each other by a period having a non-zero duration.

16. A method according to claim 1, wherein the created aggregate is a 2D-aggregate.

17. A method according to claim 1, wherein the channel comprises at least a first and a second wall defining an inner volume wherein said liquid is present and wherein the created aggregate is not in contact with said walls during step b).

18. A method according to claim 1 wherein the channel has, over at least a portion of its length, a ratio width/thickness and/or length/thickness that is greater than or equal to 10.

19. A method according to a claim 1, wherein the upper and/or lower surface of the created aggregate has a temporal average speed of growth during step b) that is less than or equal to 0.1 mm.sup.2/s.

20. A method according to claim 1, wherein, during step b), at least one feature of the aggregate is measured and at least one feature of the modulated pulsed acoustic field is modified in function of this measure.

21. A method according to claim 20, the measured feature being the instantaneous temporal speed of growth of the upper and/or lower surface of the created aggregate.

22. A method according to claim 1, wherein the formed aggregate of objects has a size of 1 nm or more.

23. A method according to claim 1, the objects being mono or poly disperse biological cells.

24. A method according to claim 1, the average size of the objects being less than or equal to 100 m.

25. A method according to claim 1, wherein the Reynolds number of the flow is 10 or less during step b).

26. A method according to claim 1, the liquid not be flowing during step b).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be better understood from a reading of the detailed description below, of non-limiting examples for its implementation, and from examination of the attached drawings, in which:

(2) FIG. 1 shows a device for carrying out of a method according to the invention,

(3) FIG. 2 shows an aggregate obtained according to a method of the invention,

(4) FIG. 3 is a view according to III-III of the channel used in FIG. 1,

(5) FIGS. 4 and 5 show examples of modulated pulsed acoustic fields for carrying out a method according to the invention,

(6) FIGS. 6 to 8 show other embodiments of devices for carrying out of methods according to the invention,

(7) FIG. 9A is a photograph of a 3D aggregate of 15 m latex particles produced in continuous mode of operation,

(8) FIG. 9B is a graph showing the evolution of the normalized surface area of the aggregate shown in FIG. 9A,

(9) FIG. 9C is a graph showing the evolution of the instantaneous temporal speed of growth of the normalized surface area of the aggregate shown in FIG. 9A,

(10) FIG. 10A is a photograph of a 2D aggregate of 15 m latex particles produced by a pulsed mode method (i.e. using a method according to the invention) at 250 pulses and repetition frequency of 5 kHz,

(11) FIG. 10B is a graph showing the evolution of the normalized surface area of the aggregate shown in FIG. 10A,

(12) FIG. 10C is a graph showing the evolution of the instantaneous temporal speed of growth of the normalized surface area of the aggregate shown in FIG. 10A,

(13) FIG. 11A is a graph showing the evolution of the normalized surface area of an aggregate of 10 m latex particles produced in continuous mode of operation (i.e. submitting objects to only one group of pulses having substantially the same amplitude and frequency),

(14) FIG. 11B is a graph showing the evolution of the instantaneous temporal speed of growth of the normalized surface area of an aggregate of 10 m latex particles produced in continuous mode of operation,

(15) FIG. 12A is a photograph of a 2D aggregate of 10 m latex particles produced by a method according to the invention (pulsed mode at 100 pulses and repetition frequency of 4 kHz),

(16) FIG. 12B is a graph showing the evolution of the normalized surface area of the aggregate shown in FIG. 12A,

(17) FIG. 12C is a graph showing the evolution of the instantaneous temporal speed of growth of the normalized surface area of the aggregate shown in FIG. 12A,

(18) FIGS. 13A and 13B respectively show cell aggregate surface area measurements produced at low (510.sup.5 cells/ml) and high (310.sup.6 cells/ml) cell concentrations in continuous and pulsed mode (50-200 pulses),

(19) FIGS. 14A and 14B respectively show representative Caco2 cell aggregates (concentration 310.sup.6 cells/ml) produced in continuous and pulsed mode (50 pulses and repetition frequency of 4 kHz),

(20) FIGS. 15A and 15B respectively show the evolutions in continuous and pulsed modes of the normalized surface area and of the instantaneous temporal speed of growth of the normalized surface area of an aggregate of Caco2 cells (310.sup.6/ml),

(21) FIGS. 16A and 16B respectively show 2-D aggregates of 15 m particles obtained in continuous mode at different voltages: a) 6 Vp-p and b) 4.5 Vp-p.

DETAILED DESCRIPTION

(22) FIG. 1 shows a device 1 which may be used in the methods according to the invention. The device 1 comprises a channel 2 extending along a longitudinal axis X.

(23) The channel 2 may, as mentioned above, be a microchannel.

(24) The channel 2 may present a cross-section that is rectangular. In the example described, the length/thickness ratio of the channel 2 is greater than 10.

(25) The channel 2 has bottom and top walls 3 and 4. As shown in FIG. 1 for example, a carrier liquid L and a plurality of objects O are present in the channel 2.

(26) Objects O may be mono or polydisperse, said objects O may be biological cells and liquid L may be a biological liquid such as, e.g. blood.

(27) The injection of objects O in the channel 2 can be controlled in frequency and in flow rate so as to enable the device 1 to operate continuously in order to process large volumes of objects.

(28) The device 1 is further provided with an acoustic field generator 10 which is, as shown, fastened to the top wall 4 of the channel 2. The acoustic field generator 10 enables formation, at step b), of an aggregate 20 of objects O by submitting them to a pulsed acoustic field modulated in amplitude. The formed aggregate 20 has as shown in FIG. 3 upper and lower surfaces S.sub.u and S.sub.l.

(29) As shown in FIG. 1, the aggregate 20 is in levitation around the pressure node of the waves generated by the acoustic field generator 10. The modulated pulsed acoustic field produced by the acoustic field generator 10 may allow the formation of a standing acoustic wave along the thickness of the channel 2 (Z-axis). The aggregate 20 is as shown in FIG. 1 not in contact with the walls 3 and 4 during step b).

(30) The expression acoustic levitation is employed when acoustic manipulation seeks to place objects in an equilibrium position against gravity. The equilibrium position depends on the acoustic properties of the objects and the suspending liquid, the acoustic power and the position and number of nodes of the acoustic waves.

(31) The acoustic field generator 10 may be supplied by a signal from a generator D which e.g. comprise a wave generator connected to an amplifier. The generator D may supply the acoustic field generator 10 with groups of sine-shaped voltage pulses. In a variant, the acoustic field generator 10 may be supplied by the generator D with groups of triangular-shaped or square-shaped voltage pulses.

(32) As explained above, an aggregate of objects may be more compact than a layer of objects. FIG. 2 shows an upper view of an aggregate 20 obtained at the end of step b) according to the invention. The aggregate 20 comprises a set of objects O that are in contact with each other, e.g. at least 50% of the objects O constituting said aggregate 20 can be in contact with each other.

(33) The invention may enable the formation of 2D and/or 3D aggregates. The definition of such 2D and 3D aggregates is given below.

(34) The aggregate 20 comprises a succession 20.sub.1 of objects O when moving along the Y axis which corresponds to a displacement along the width of the channel 2 but has a thickness formed of at least one object. In this case, the aggregate 20 is a 2D-aggregate.

(35) In an embodiment, the aggregate also comprises a succession of objects O when moving along the thickness of the channel 2. The aggregate is thus a 3D-aggregate.

(36) As shown in FIG. 3, a layer of a gel 11 acting as an acoustic impedance adapter may be present between the acoustic field generator 10 and the top wall 4 of the channel 2.

(37) FIG. 4 shows an example of a temporal evolution of a signal corresponding to a pulsed acoustic field modulated in amplitude which may be used in step b) according to the invention.

(38) The modulated pulsed acoustic field may as shown comprise a repetition of a plurality of groups 50 of acoustic wave pulses 51, said pulses 51 having, among a given group 50, substantially the same amplitude and frequency, and said groups 50 being separated between each other by a period 60 having a non-zero duration wherein no acoustic wave is applied.

(39) Each of the groups 50 of acoustic wave pulses 51 may comprise 10 or more, preferably 25 or more, preferably 50 or more, preferably 100 or more, acoustic wave pulses 51.

(40) The totality of the groups 50 of acoustic wave pulses 51 may last a duration t.sub.1 that is greater than or equal to 0.01 ms, preferably to 0.025 ms.

(41) The totality of the periods 60 separating two successive groups 50 of acoustic wave pulses 51 may have a duration t.sub.2 that is greater than or equal to 0.05 ms, preferably 0.1 ms, more preferably to 0.2 ms.

(42) The totality of the periods 60 separating two successive groups 50 of acoustic wave pulses 51 may have a duration t.sub.2 that is less than or equal to 0.5 s, preferably to 0.1 s, more preferably to 0.01 s, more preferably to 0.005 s.

(43) The present disclosure also encompasses the use of variants of modulated pulsed acoustic field wherein an acoustic wave is applied during a period separating two groups of pulses.

(44) Such an embodiment is shown in FIG. 5 wherein two successive groups 50 of acoustic wave pulses 51 are separated by a period 60 wherein an acoustic wave is applied. The acoustic wave applied during the period 60 has as shown a highest amplitude that is less than the highest amplitude of the acoustic wave pulses 51 of the groups 50.

(45) As shown in FIG. 5, the temporal average of the absolute value of the amplitude of the acoustic wave applied during the period 60, said temporal average corresponding to the horizontal line drawn, is less than 50% and for example approximately equal to 25% of the highest amplitude of pulses 51 in the groups 50.

(46) FIG. 6 shows a variant of a device 1 for carrying out a method according to the invention.

(47) In this embodiment, an aggregate 20 is created by submitting objects O to a modulated pulsed acoustic field produced by an acoustic field generator 10.

(48) A device 100 is present and allows the measurement of at least one feature of the aggregate 20. For example, the device 100 may allow the measurement of the instantaneous temporal speed of growth of the lower surface S.sub.l area of the aggregate 20.

(49) Said device e.g. comprises a microscope which allow image capturing in the direction of sound propagation (negative z-axis) connected to a standard PC which is equipped with any suitable software allowing the estimation of the instantaneous temporal speed of growth of the lower surface S.sub.l area of the aggregate 20 (e.g. Cell-D image acquisition and processing software (Soft Imaging System, SIS, GmbH) or Image J software).

(50) This measured feature is then transmitted to a device T controlling the modification of at least one feature of the modulated pulsed acoustic field in function of said measured feature. The device T e.g. controls the modification of the amplitude and/or frequency of the pulses in the groups and/or duration of the periods separating two successive groups.

EXAMPLES

(51) Materials and Methods

(52) Ultrasound Trap

(53) The in-house constructed traps 1 employed are shown in FIGS. 7A (top view of the first trap employed), 7B (side view of the device of FIG. 7A), 7C (top view of the second trap employed) and 7D (side view of the device of FIG. 7C).

(54) These traps have four layers; a transducer 10 (PZ26 Ferroperm, Kvistgard, Denmark) nominally resonant in the thickness mode at 3 MHz and mounted in a radially symmetric housing, a steel layer 14 coupling the ultrasound to a one half wavelength (/2 or 0.25 mm depth, where is the wavelength of sound in water at 3 MHz) aqueous layer and an acoustic reflector 4, e.g. made of quartz, that provided optical access from above.

(55) A resonant cavity 5 is defined by a polyethylene terephtalate (Mylar) or polyimide spacer 5a.

(56) It is also possible to use an acoustic reflector made of a plastic material and/or to substitute the steel layer 14 by a layer made of quartz or a plastic material.

(57) The size of the cylindrical steel body was 35 mm. The resonant cavity 5 may have a diameter comprised between 1 mm and 20 mm. The disc transducer (12 mm diameter) 10 was driven at 2.13 MHz.

(58) The trap 1 shown in FIGS. 7A and 7B has one sample inlet 6 and one sample outlet 7. The traps 1 are further provided with connectors 8 for connection to a voltage generator. Each of the connectors 8 is electrically connected to a face of transducer 10. One of the connectors 8 is as shown in contact with the steel layer 14, electrical contact with the face of the transducer 10 fastened to the layer 14 is ensured by a layer of conductive epoxy (not shown).

(59) The trap 1 shown in FIGS. 7C and 7D does not comprise any inlet or outlet.

(60) We note that the nominal resonance frequency of the transducer 10 (3 MHz) is different than the nominal resonance frequency of the resonator (2.13 MHz) due to the steel-coupling layer 14. The acoustic reflector 4 may have a thickness of 0.5 mm, 1 mm or 2 mm and may be made of glass, in particular quartz glass.

(61) The piezoceramic transducer 10 was driven by a 100 MHz dual channel arbitrary wave generator (5062 Tabor Electronics, Israel) and the signal was amplified by a dual differential wide band 100 MHz amplifier (9250 Tabor Electronics, Israel). The signal was visualized with a digital storage oscilloscope (IDS 8064 60 MHz ISOTECh, HananIsrael).

(62) In experiments with continuous and pulsed ultrasound, amplitudes of 5 to 15 V.sub.p-p were employed. Concerning the pulsed acoustic fields used, the number of pulses among each group varied between 25-250 and the repetition frequency of the groups from 1 to 5 kHz.

(63) Optical System

(64) A fast, high-resolution XM10 (Soft Imaging System, SIS, GmbH) mounted on an Olympus BX51M reflection epi-fluorescence microscope allowed observation in the direction of sound propagation (negative z-axis) (see FIG. 8).

(65) Images were captured by a standard PC equipped with the Cell-D image acquisition and processing software (Soft Imaging System, SIS, GmbH).

(66) Polystyrene Latex Beads Suspensions

(67) We used 15 and 10 m diameter polystyrene latex particles (density 1,056 kg/m.sup.3) supplied as a 10% suspension of solids by Micromod (Rostock, Germany). Aliquots were diluted here to 20 ml to give about 0.15% solids in deionized water.

(68) Cell Culture

(69) Caco2 (human epithelial colorectal adenocarcinoma) cells were obtained from the School of Pharmacy and Pharmaceutical Sciences, Trinity College Dublin, Ireland.

(70) Cells were maintained as a replicative culture at 37 C. under an atmosphere of 95% air and 5% CO.sub.2. Caco2 cells were cultured in Modified Eagle Medium (MEM) supplemented with 100 g/ml penicillin-streptomycin solution, 0.05 g/l sodium pyruvate and 10% foetal calf serum (FCS). At confluence, cells were treated with Trypsin/EDTA solution (1), for 10 min to release cells from the culture flask surface.

(71) Cells were filtered through a 40 m Nitex cell strainer (FALCON) to ensure a single cell suspension. Cells were then pelleted by centrifugation at 2000 rpm for 5 min, resuspended in fresh serum-free MEM medium (containing all of the aforementioned supplements), counted and finally diluted to concentrations of 510.sup.5 cells/ml and 310.sup.6 cells/ml.

(72) Experimental Procedure

(73) A Gilson minipuls3 peristaltic pump (Gilson, Inc. Middleton, USA) may be used to pump the sample into the ultrasound trap.

(74) The microscope was pre-focused on the central plane of the trap at the axial region. When a pump was used, the pump was switched off and the ultrasound exposure was immediately initiated. The aggregation process was observed using 5, 10 and 20 objectives.

(75) The pulsed ultrasound approach used consisted in generating groups of pulses (one pulse is one period) at a certain repetition frequency (see e.g. FIG. 4). For instance, at 2 MHz, one pulse lasts for 500 ns; and typically 100 periods may be used, lasting t.sub.1=0.05 ms, separated by t.sub.2=1 ms repetition time.

(76) This means that for one experiment lasting 1 min, each on-off cycle lasts t.sub.cycle=1.05 ms, leading to approximately 57,143 on-off cycles during the experiment. The total time at which the force is on will then be 2.86 seconds, i.e. 5% of the total time. For the conditions reported above the pulse mode factor P.sub.mf, defined as

(77) $P_{\mathrm{mf}}=\frac{t_1}{t_1+t_2},$
is equal to 0.05.

(78) One of the motivations for employing pulsed ultrasound, other than to control the aggregation process, is to avoid transducer heating; a minimum number of pulses fulfill this requirement.

(79) In addition, a minimum number of pulses allow us to increase the acoustic force by increasing the voltage, thereby reducing the risk of transducer damage. In turn, the risk of liquid heating is also reduced, keeping the thermodynamic and physicochemical properties of the suspension as stable as possible.

(80) By using pulsed mode ultrasound it is possible to modify the aggregation velocity by controlling the number of pulses and the repetition frequency. In fact, inventors assume that in continuous mode, particles and cells have the maximum transversal velocity at the nodal plane, while in pulsed mode they slow down in a controlled way.

(81) Experimental Results

(82) The aggregation process starts when isolated particles in levitation converge towards a specific point, where the acoustic energy is maximal.

(83) In the resonator used, levitation of cancer cells was obtained at all frequencies in the range of 2.1 and 2.4 MHz but the optimal aggregation started when the frequency was 2.13 MHz, at which frequency, the aggregation process occurred very fast.

(84) On the other hand, when the frequency was close but not at the optimal resonance frequency, several small aggregates were formed at different positions in the levitation plane. For instance, for cancer cells, operation at the aggregation frequency (2.13 MHz) and at 10Vp-p amplitude, allowed the aggregate to reach its maximum size within 10 s, as we will demonstrate below.

(85) Pulsed Acoustic Field Force

(86) Initially, particles or cells were injected in the resonator where they homogeneously filled the chamber of volume 45 l. The flow was then stopped, particles settled down and then acoustic field was turned on.

(87) The number of pulses and the repetition frequency were varied while keeping the other parameters constant so as to investigate the effect of a modulated pulsed acoustic field for aggregate formation.

(88) The transducer vibration has the maximum amplitude at the resonance frequency (the transducer vibration amplitude is a few nm and the vibration velocity a few cm/s) but the energy is not uniformly distributed and has a maximum intensity at the center.

(89) The Bernoulli force generated by this effect, as well as by other imperfections of the resonator, acts on particles driving them towards the maximum energy zone, thus leading to the generation of a single aggregate. The fact that in pulsed mode a single aggregate was also observed indicates that the resonance frequency does not change with respect to continuous mode.

(90) In all experiments we aimed at forming aggregates of similar sizes (maximum diameter 2 mm).

(91) Particle Aggregates

(92) An estimation of the transversal acoustic force responsible for aggregation was made, by measuring (using the software: http://www.cabrillo.edu/dbrown/tracker/) the migration velocity of a particle at the focusing plane in continuous and in pulsed mode at the resonance frequency of 2.13 MHz.

(93) In continuous mode, single 10 m and 15 m latex particles had average transversal velocities of u.sub.Tr10=210.sup.5 m/s (at 13.8 V.sub.p-p) and u.sub.Tr15=1.310.sup.5 m/s (at 16 V.sub.p-p) respectively. The respective average transversal forces are F.sub.Tr=1.810.sup.12 N and 210.sup.11 N; these values are as indicated before, at least two orders of magnitude smaller than the primary radiation force.

(94) FIG. 9A shows a 3-D aggregate of 15 m latex particles formed in continuous mode. The normalized particle aggregate surface area grows monotonically with time showing surface area fluctuations (see FIG. 9B). The aggregation process was completed in 18 s. The voltage used was 16 Vp-p.

(95) FIG. 9C (time derivative of the normalized surface area) shows the instantaneous temporal speed of growth of the normalized surface area, the strongest peak appears by 7 s. The resultant aggregate formed is 3-D. The voltage used was 16 Vp-p.

(96) When pulsed ultrasound was employed for 15 m particles, at 250 pulses and 5 kHz repetition frequency, the average particle velocity was u.sub.Tr15 pulsed3.510.sup.6 m/s, corresponding to a transversal acoustic force F.sub.Tr210.sup.13 N; the latter being much smaller than that obtained in continuous mode. The voltage used was 16 Vp-p.

(97) A 2-D aggregate was formed (see FIG. 10A). The surface area increased linearly with time (see FIG. 10B), and the fluctuations were much smaller than those observed in continuous mode. The instantaneous temporal speed of growth of the normalized surface area of the aggregate shows very small fluctuations (close to zero) (see FIG. 10C).

(98) 3-D and 2-D aggregates of 10 m latex particles, in continuous mode and in pulsed mode at 100 pulses and 4 kHz repetition frequency were also generated. The voltage used in these experiments was 13.8 Vp-p.

(99) The particle velocity in pulsed mode was of the order of u.sub.Tr10 pulsed410.sup.6 m/s, much smaller than in continuous mode, corresponding to a transversal force F.sub.Tr410.sup.13 N.

(100) The aggregation in pulsed mode was ended after 300 s when the aggregate was about 2 mm in diameter. FIGS. 11A and 11B show the evolutions of normalized aggregate surface area and of the instantaneous temporal speed of growth of the normalized aggregate surface area.

(101) Features analogous to those of 15 m particles are observed. Nevertheless, the normalized aggregate surface area growth in continuous mode (see FIG. 11A) showed an initial rapid increase (for the first 4 s), and then decreased to finally becoming linear.

(102) The instantaneous temporal speed of growth of the normalized aggregate surface area, peaked within one second during the aggregation process, followed by a smooth decrease; these features were more pronounced for 10 m than for 15 m particles (see FIG. 11B).

(103) In pulsed mode, a 2-D aggregate was obtained (see FIG. 12A), the growth of the aggregate surface area was linear (see FIG. 12B) and the instantaneous temporal speeds of growth of the normalized aggregate surface area were negligible compared to those in continuous mode (see FIG. 12C).

(104) Cell Aggregates

(105) For cell aggregation, different experimental parameters were varied including the cell concentration, number of pulses, repetition time and voltage.

(106) The typical cell size was 20 m and the typical velocities were 223 m/s in continuous mode and 147 m/s in pulsed mode, corresponding to acoustic forces of 1210.sup.11 N and 5.510.sup.11 N respectively.

(107) The histograms in FIGS. 13A and 13B show all the parameters and the conditions where 3-D and 2-D aggregates were generated. The voltage used for experiments in pulsed regime was 10 Vp-p. Unless otherwise specified in the figures, the pulse repetition time (period separating two groups of pulses) used in the pulsed mode experiments was 250 s.

(108) FIGS. 13A and 13B respectively show cell aggregate surface area measurements produced at low (510.sup.5 cells/ml) and high (310.sup.6 cells/ml) cell concentrations in continuous and pulsed mode (50-200 pulses). The final cell aggregate architecture (2D or 3D) and the pulse repetition time are also indicated.

(109) Representative 3-D and 2-D cell aggregates are shown in FIGS. 14A and 14B under continuous and pulsed ultrasound.

(110) The evolutions of the normalized aggregate surface area and of the instantaneous temporal speed of growth of the normalized aggregate surface area are depicted in FIGS. 15A and 15B respectively. The voltage used was 10 Vp-p.

(111) In continuous mode, the cell normalized aggregate surface area followed the same pattern as that for 10 m particle aggregates; a rapid increase of the surface area in the first second followed by a linear growth.

(112) In contrast, in pulsed mode the growth was slower and followed a linear pattern. This holds true for all the pulsed ultrasound conditions used here. The speeds of growth in continuous mode showed an initial peak and then decreased to converge with the curves obtained for the different pulsed ultrasound conditions; peaks in pulsed mode curves were rarely observed and were still much smaller than those observed in continuous mode.

(113) Discussion

(114) Initially, we demonstrated how pulsed ultrasound can control particle and cell aggregation. As demonstrated before, the speeds of aggregate growth are reduced when pulsed ultrasound is employed.

(115) The pulse mode factor for 15 and 10 m particles: P.sub.mf15=0.19 and P.sub.mf10=0.16 indicates that the acoustic force was applied during the 19% and 16% of the total experimental time respectively.

(116) By considering that particles have a constant transversal velocity and by taking into account that the drag force is proportional to the velocity, we can compare P.sub.mf to the velocity ratio: u.sub.pulsed/u.sub.continuous, being 0.24 for 15 m and 0.2 for 10 m particles, not very different from their respective pulsed mode factor.

(117) This result suggests that the inertial effects, i.e. the time needed for reaching the terminal velocity and the relaxation of the maximal amplitude related to the Q-factor, do not significantly influence the aggregation process in pulsed mode.

(118) In the opposite event the values for the Velocity ratio would be smaller than the P.sub.mf.

(119) The fact that in pulsed mode the aggregation is slower than in continuous mode does not directly imply that lower growth rates entail a higher probability of obtaining 2-D than 3-D aggregates; in fact, the initial particle or cell concentration needs to be taken into consideration.

(120) It is worth noting that in continuous mode during the aggregation process, almost all particles reached the focusing plane before aggregating (only a few particles joined the aggregate from the bottom of the resonator, and their contribution to the 3D structure was negligible), whereas in pulsed mode, all particles started aggregating at the focusing plane.

(121) Our experiments showed that when the aggregate was growing, the momentum transmitted by incoming particles gave rise to the rearrangements of particles and cells, indicating that the pressure distribution in the aggregate was modified.

(122) When the pressure was high enough, particles or layers started to overlap, mostly in the central part of the aggregate, leading to surface area growth rate fluctuations as depicted in FIG. 9C and FIG. 12C.

(123) As in those experiments only few particles joined the aggregate from elsewhere other than from the levitation plane, it is possible to conclude that the overlapping of the particle and cell layers were generated by the increasing pressure of incoming particles, thus determining the transition from 2-D to 3-D aggregates.

(124) Experiments with particles clearly showed layer overlapping (see FIG. 9A), while aggregates of cells showed rather darker layers (see FIG. 14A). The aggregation process showed that the fluctuations in the aggregate surface area coincided with the overlapping of layers. These fluctuations disappeared in pulsed ultrasound where only 2-D aggregates were formed. In FIG. 10C (pulsed mode, 15 m particles), one big fluctuation was observed, but still smaller than those during continuous mode. In fact, in this case we observed, a small aggregate reaching the main aggregate hence generating a great pressure and a strong particle rearrangement; the aggregate was nevertheless 2-D (see FIG. 10A).

(125) This new method is developed with the aim to generate controlled cell constructs.

(126) When cells were employed, even though the surface area fluctuations were visible, the process did not show rapid fluctuations. It is important to note that cells are more elastic than particles, thus initially, compaction of the aggregates occurred with increasing pressure prior to the formation of cell layers.

(127) The curve of the speed of growth of the aggregate surface area in continuous ultrasound is depicted in FIG. 15A. The curve shows a maximum analogous to that shown for 10 m particles, suggesting that a maximum in the instantaneous temporal speed of growth of the surface area could indicate the formation of a 3D structure. When cells are employed, the aggregate does not grow linearly with time. The elastic properties of the cells as well as their polydispersity could account for this effect.

(128) Finally, the histograms in FIGS. 13A and 13B indicate that the area of the cell aggregates is a function of the number of pulses employed for both concentrations studied.

(129) 2D constructs of different sizes can be formed with suspensions of very different initial cell concentrations; for example, 510.sup.5 cells/ml at 200 pulses generates smaller aggregates than 310.sup.6 cells/ml at 50 pulses. Regardless of the initial concentration, the pulsed ultrasound technique allows 3D or 2D aggregates to be formed.

(130) Thus, pulsed ultrasound introduces new parameters for controlling the aggregation process.

(131) The possibility of generating 2-D aggregates in continuous mode by reducing the amplitude of the acoustic force, i.e. by reducing the voltage, was investigated. Experiments with 15 m latex particles (0.1% solids in water) were performed.

(132) By reducing the voltage to 9 Vp-p, 6 Vp-p and 4.5 Vp-p, an increase in the duration of the aggregation time was observed, as expected. At 9 Vp-p the aggregate was still 3-D. At 6 Vp-p the aggregate was mostly 2-D (see FIG. 16A), as it was difficult to distinguish whether all particles were at the same plane at the centre of the aggregate. At 4.5 Vp-p the aggregate was 2-D but at the center a concave shape showed that the aggregate was not in plane (see FIG. 16B). We also observed that the aggregation time was double the aggregation time in pulsed mode.

(133) At 6 Vp-p and 4.5 Vp-p the aggregate looks like 2-D but, at the centre, particles are unfocused, and the aggregate is not in plane. At higher voltages the aggregate is 3-D.

(134) Finally, by comparing the 2-D aggregates obtained at reduced voltage and in pulsed mode (see FIG. 10A) we can see that in pulsed mode the aggregate is plane and regular.

(135) Based on these observations, it can be considered that when reducing the voltage in continuous mode to control aggregation, it is difficult to get an accurate voltage to generate the optimum or required 2-D aggregate.

(136) The expression comprising alone should be understood as comprising at least one.

(137) The expression comprised between . . . and . . . should be understood with the end points included.