EP4631622A1

EP4631622A1 - Method and system for trapping particles in a trapping matrix

Info

Publication number: EP4631622A1
Application number: EP24169062.7A
Authority: EP
Inventors: Thomas Laurell; Michael GERLT
Original assignee: Acousort AB
Current assignee: Acousort AB
Priority date: 2024-04-08
Filing date: 2024-04-08
Publication date: 2025-10-15

Abstract

A method of trapping particles, in a sample of a disperse liquid, having a non-zero acoustic contrast factor relative to a suspending liquid, comprising: i. providing an acoustofluidic device comprising a substrate in which a cavity is formed, the cavity being at least partially filled with a trapping matrix comprising or consisting of a material having a non-zero acoustic contrast factor, the cavity having an inlet for introducing the sample into the cavity, and an ultrasound transducer in acoustic contact with the substrate, ii. Introducing the sample into the cavity so the sample perfuses the trapping matrix, and iii. actuating the ultrasound transducer at an actuation frequency which does not correspond to a resonance of the cavity filled with the sample of disperse liquid, and couples an acoustic wave into the substrate. A system is also provided.

Description

Technical field

The technology proposed herein relates generally to the field of acoustofluidics in which ultrasound transducers are used to actuate acoustic waves in liquids and suspensions for interacting with the liquids or different types of particles in the liquids and suspensions, for example for performing inter alia separation and sorting of the particles or mixing of liquids or suspensions. The technology proposed herein particularly relates to methods and systems for trapping particles in a trapping matrix using an ultrasound transducer.

Background

Acoustofluidics generally refer to using sound waves to affect liquids or particles in liquids or suspensions. Acoustofluidics used to affect particles is generally termed acoustophoresis. Acoustophoresis has been used inter alia for separating different types of cells in suspensions such as separating blood cells from plasma or for separating and collecting circulating tumor cells from blood. Generally, an acoustofluidic device comprises a cavity such as a flow channel fashioned in a substrate. The suspension is pumped through the flow channel under laminar flow conditions, or alternatively is stationary in the flow channel. An ultrasound transducer, particularly a piezoelectric element, is attached to the substrate and actuated to produce an ultrasonic vibration (displacements in the 10-100 nm range) in the substrate in the range of about 1-20 MHz. Provided that the dimensions, in particular height or width, of the flow channel is properly matched with the frequency of the ultrasonic vibration, an acoustic standing wave may be generated in the channel. This acoustic standing wave exerts a force, i.e. the acoustic radiation force, on the particles in the suspension dependent on the acoustic contrast of each individual particle as determined by the properties of each particle relative to those of the suspending liquid in the suspension, and thus particles will be forced to move, dependent on the acoustic contrast, towards or away from the pressure node(s) of the standing wave. Applications include, as stated above, separation, sorting, trapping and other manipulations of the particles. The ultrasonic vibrations transmitted into the device do not only affect particles dispersed in the liquid, but also in the liquid itself by causing acoustic streaming. Such streaming may be used for mixing.
As the acoustic radiation force exerted on the particles inter alia is dependent on the size of the particle, more particularly it is proportional to the volume of the particle, smaller particles are more difficult to affect and hence to separate, sort, and trap. Smaller particles move more slowly towards or away from the pressure node(s), depending on their acoustic contrast, and in many cases the forces exerted on small particles are so low that the particles cannot be separated, sorted, and/or trapped, especially in view of the fact that acoustic streaming exerts a drag force on the particles that scales with the radius of the particles and does typically counteract the motion induced by the acoustic radiation force. Hence there exists a critical particle radius at which the motion of the particles is dominated by the acoustic streaming. As an example, this critical particle radius typically lies around 1 µm for a polystyrene particle in water and an actuation frequency of 2 MHz. This is a problem for using acoustofluidics on biological nanoparticles. Such nanoparticles include extracellular vesicles which are formed as lipid bilayer-delimited particles that are released from cells. Extracellular vesicles may for example have diameters in the range of 30-1000 nm and face a growing interest e.g. for therapeutic purposes.
Methods for separating extracellular vesicles from samples of disperse liquids include ultracentrifugation, density gradient ultracentrifugation, size exclusion chromatography, ultrafiltration, capillary electrophoresis, asymmetric-flow field-flow fractionation, and affinity/immunoaffinity capture methods.
Noting the simplicity and efficiency, as compared to e.g. ultracentrifugation, by which acoustofluidics can separate whole blood into formed elements and plasma, it would be advantageous if acoustofluidics could be applied also to the separation of extracellular vesicles and other nanoparticles from samples.
Accordingly, there is a need for simple and efficient methods and systems for trapping and/or separating particles such as biological nanoparticles.

Objects

A primary object of the technology proposed herein is to provide a method of trapping particles in a trapping matrix using an ultrasound transducer.
A further object of the technology proposed herein to provide a system for trapping particles in a trapping matrix using an ultrasound transducer.

Summary

At least one of the abovementioned objects or at least one of the further objects which will become evident from the below description, are according to corresponding first and second aspect of the technology proposed herein achieved by a method of trapping particles in a sample of a disperse liquid, the disperse liquid comprising particles dispersed in a suspending liquid, the particles having a non-zero acoustic contrast factor relative to the suspending liquid, the method comprising the steps of:

i. providing an acoustofluidic device comprising:
1. a. a substrate in which a cavity is formed, the cavity being at least partially filled with a trapping matrix, the trapping matrix comprising or consisting of a material having a non-zero acoustic contrast factor relative to the suspending liquid, the cavity further having an inlet for introducing the sample of disperse liquid into the cavity, and
2. b. an ultrasound transducer in acoustic contact with the substrate,
ii. introducing the sample of the disperse liquid into the cavity so that the sample perfuses the trapping matrix, and
iii. actuating the ultrasound transducer at an actuation frequency which:
1. a. does not correspond to a resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with sample of the disperse liquid and optionally also the trapping matrix,
2. b. couples an acoustic wave into the substrate,

a) an acoustofluidic device comprising:
1. a. a substrate in which a cavity is formed, the cavity being at least partially filled with a trapping matrix, the trapping matrix comprising or consisting of a material having a non-zero acoustic contrast factor relative to the suspending liquid, the cavity further having an inlet for introducing the sample of disperse liquid into the cavity, and
2. b. an ultrasound transducer in acoustic contact with the substrate, and
b) a drive circuit connected to the ultrasound transducer, the drive circuit being configured to actuate the ultrasound transducer at an actuation frequency which:
1. a. does not correspond to a resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with the fluid and optionally also the trapping matrix, and,
2. b. couples an acoustic wave into the substrate.

Accordingly, the technology proposed herein is based on the realization by the present inventors that ultrasound can be used to cause the trapping matrix to trap particles that otherwise would be too small to trap in the direct sound field from the ultrasound transducer. It is believed that the acoustic wave that is coupled into the substrate from the ultrasound transducer generates secondary acoustic fields within the trapping matrix, which secondary acoustic field provide the trapping effect. As seen in the examples, this trapping effects provides trapping of both micro- and nanoparticles.
Most surprising is that the actuation frequency used does not correspond to a resonance of the cavity filled with the sample of disperse liquid, and that an acoustic wave is coupled into the substrate.
This is a major departure from prior art attempts.
Specifically, in Gupta & Feke, "Acoustically driven collection of suspended particles within porous media", Ultrasonics 35 (1997) 131-139, a chamber, e.g. a cavity, housing 3 mm diameter glass beads was actuated at the resonance frequency of the chamber width by an ultrasound transducer and reflector defining the chamber walls, to trap 325 mesh (particles radius of 2 to 15 µm) polystyrene divinyl benzene particles.
Further, whereas Habibi & Neild, "Sound wave activated nano-sieve (SWANS) for enrichment of nanoparticles", Lab Chip, 2019, 19, 3032, showed trapping of nanoparticles (500 nm, 190 nm, 100 nm) this was obtained using surface acoustic waves stated to have a frequency matching the eigen frequency of the beads, e.g. 72 MHz. The volume and capacity of the packed bed (14 nL/cm) was small and the stated flow rate of 100 nL/min low.
In contrast, as shown in the examples, the method and system according to the first and second aspects of the technology proposed herein uses completely different actuation frequencies and type of actuation to provide trapping of particles and nanoparticles.
The examples thus show that there is no need to use an actuation frequency matched to provide resonance in the cavity or matched to the eigen frequency of the trapping matrix. To the contrary, the most preferable actuation frequency is closer to an antiresonance frequency of the width or length dimension of the ultrasound transducer. Further, there is no need to use surface acoustic waves for trapping nanoparticles, rather such particles may be trapped with acoustic waves which are coupled into the substrate, e.g. bulk acoustic waves, which significantly simplifies construction of the acoustofluidic device and provides for improving throughput and capacity.
The advantages of the method and system according to the first and second aspects of the technology proposed herein are clearly shown when considering that the examples show trapping of 380 nm nanoparticles at a throughput of 50 µL/min, which is 500 times higher than Habibi & Neild, using a packed bed comprising 100 µm polystyrene beads packed in a 400 µm * 4 mm cross-section rectangular glass tube that is 10 mm long providing a packed bed volume of 16 µL, i.e. 100 times larger than Habibi & Neild.
In addition, the technology proposed herein is scalable as shown in the later example where a cavity having a cross section of 2 mm * 4 mm is used. This scalability is an effect of the use of acoustic waves coupled into the substrate and thereby provides for large scale trapping of particles, in particular nanoparticles. In comparison, surface acoustic waves limit upscaling due to the higher attenuation, and corresponding lesser penetration into the trapping matrix, of the higher frequencies required by surface acoustic waves. Instead, the method and system according to the first and second aspects of the technology proposed herein provides for significantly better scaling properties as the lower frequencies used to provide vibrations in the substrate penetrate much better into the trapping matrix. Additionally, in contrast to surface acoustic waves, which generally require two counter propagating wave generators, only a single ultrasound transducer is required. Further, the ultrasound transducer does not need to constitute a wall of the cavity, nor is a specific reflector required, thus simplifying construction of the acoustofluidic device.
It is specified that the method and system are for trapping particles. Worded differently, the method and system are for one or more of separating, concentrating, and retaining particles. Thus, the method may alternatively be considered a method of separating, concentrating, or retaining particles from a sample of disperse liquid.
The disperse liquid may be any liquid comprising particles dispersed in a suspending liquid. The suspending liquid may be any liquid or mixture of liquids.
The term sample refers to a portion or volume of the disperse liquid. The sample may have any volume. Typically, a sample may have a volume of 1 mL to 1000 mL, but the sample may be smaller, such as in the µL range or smaller, or larger such as in the liter range or m³ range or even larger.
The disperse liquid is preferably a biological liquid, such as a clinical biofluid, or a liquid comprising biological tissue. Alternatively, the liquid is a synthetic liquid. The disperse liquid may preferably comprise or be selected from the group consisting of undiluted or diluted whole blood plasma or serum, undiluted or diluted intracellular fluid, undiluted or diluted interstitial fluid, undiluted or diluted synovial fluid, undiluted or diluted peritoneal fluid, undiluted or diluted urine, undiluted or diluted bone marrow, undiluted or diluted cerebrospinal fluid, undiluted or diluted stroma, undiluted or diluted culture medium comprising bacterial or yeast cultures, undiluted or diluted milk, undiluted or diluted suspensions of animal cells, undiluted or diluted suspensions of dissociated cells from normal or cancerous tissue, undiluted or diluted suspensions of red blood cells, white blood cells, platelets, cancer cells, bacterial cells, stem cells, mononuclear cells, viruses, yeast cells, algae, pollen, extracellular vesicles such as microvesicles and exosomes, dust particles, silica particles, magnetic particles, and polymer particles.
The particles are preferably of biological origin. In other words, the particles may be naturally occurring particles or particles manufactured by biological organisms. The particles may for example be selected from the group consisting of lipoproteins, exosomes, viruses, magnetosomes, and ferritin. Additionally, the nanoparticles may comprise cell organelles.
Preferably the particles are nanoparticles as further discussed below.
The particles are dispersed in suspending liquid. In other words, the particles are mixed with the suspending liquid.
The particles have a non-zero acoustic contrast factor relative to the suspending liquid.
The acoustic contrast factor φ for a particle having the compressibility β_p and the density ρ_p suspended in a medium having the compressibility β_m and the density ρ_m can be expressed as: $Φ = \frac{5 ρ_{p} - 2 ρ_{m}}{2 ρ_{p} + ρ_{m}} - \frac{β_{p}}{β_{m}}$
Accordingly, the compressibility and or the density of the particles differ from those of the suspending liquid such that the acoustic contrast factor is non-zero. This means that the particles, when suspended in the suspending liquid and subjected to an acoustic field, will experience a force causing the particle to move and/or be retained in position relative to the suspending liquid. This effect provides for trapping the particles in the trapping media. The acoustic contrast factor of a particle, if zero, may be rendered non-zero for example by adjusting the density and/or compressibility of the suspending liquid and/or the particle.
It is specified that an acoustofluidic device is provided. In other words, an acoustofluidic device is used in the method.
The acoustofluidic device may in other words be termed acoustophoretic device, acoustophoresis chip, and acoustophoresis device-
The substrate may be made from a number of different materials including oxide, glass, metal, ceramics, and silicon. It is further contemplated within the context of the technology proposed herein that the substrate may be made from polymeric materials, in particular plastics such as cyclic olefin copolymer (COP), cyclic olefin polymers (COC), polycarbonate (PC), polypropylene (PP) poly(methyl methacrylate) (PMMA), polystyrene (PS), and Polyether Ether Ketone (PEEK).
The substrate may have different shapes, lengths, heights, and widths. Typically, the substrate has a bottom surface, an opposing top surface, two opposing side surfaces, and two opposing end surfaces. The length, height and width of the substrate are typically in the range of 10-100 mm (length) 0.2-3 mm (height), and 1-10 mm (width). The substrate may for example be a capillary. The substrate may also be larger, e.g. having a length of 100-1000 mm, height of 3-100 mm, and width of 10-100 mm. Even larger substrates are possible.
The substrate may be formed in one piece. Typically, however, the substrate is fashioned from two parts so that the cavity may be easily implemented as a trough or groove in one of the parts whereafter the other part is placed as a lid to seal the trough or groove to form the cavity.
The cavity may run along at least a part of the substrate and may be provided with inlets and outlets at its opposite ends. The cavity may in particular comprise a channel. The cavity may have a floor, a ceiling, and two opposing side walls. Typically, the cavity will have a rectangular or substantially rectangular cross section, although other shapes of cross section are possible. The width of the cavity is typically from 0.1 to 4 mm and the height 0.05 to 2 mm, depending on the size of any particle that is to pass through the cavity. Such a cavity may be termed a microfluidic cavity. The cavity may however be any size, in particular larger, such as having a width, height and/or length in the centimeter, decimeter or meter range, or even larger. As an example, a width of the cavity may be 4-90 mm and the height may be 2-90 mm, a length of 100-1000 mm.
It is to be understood that the sample of the disperse liquid may, or may not, flow through the cavity. Generally, a lower flow rate may lead to a more efficient trapping. Generally suitable flow rates depend on the dimensions of the cavity. Typically, the flow rate, if the sample is flowing through the cavity, may be 5 to 200 µL/minute, such as 100 µL/minute, or even more than 200 µL/minute. However, suitable flow rates may in any case be found for an acoustofluidic device by testing different flow rates and determining the efficiency of the trapping of the particles, whereafter a flow rate can be chosen that provides a sufficient desired efficiency of trapping.
The cavity may further have different dimensions at different positions along its length. The cavity may further branch into plural cavities, or plural cavities may join into one cavity, at different positions along its length.
The cavity is at least partially filled with the trapping matrix.
The cavity may alternatively be fully filled with the trapping matrix.
When the cavity is at least partially filled with the trapping matrix, then it is preferred that the width and height cross section of the cavity is filled with the trapping matrix whereas only a part of the length of the cavity is filled with the trapping matrix.
A retaining device may be provided in the cavity to segregate the trapping matrix into one or more parts of the cavity. The retaining device is configured to physically prevent the trapping matrix from passing through or passing by the retaining device. The retaining device may comprise one or more of a screen, grid or lattice having openings smaller than the trapping matrix or individual parts thereof but larger than the particles, a membrane having pores smaller than the trapping matrix or individual parts thereof but larger than the particles, a structure such as a plurality of protrusions from an inner wall of the cavity, or a plurality of bodies, being arranged so that the distances between the protrusions or bodies are smaller than the trapping matrix or parts thereof but larger than the particles, a plurality of retaining particles that are larger than the trapping matrix or parts thereof and packed so that the gaps between the retaining particles are smaller than the trapping matrix or parts thereof but larger than the particles, and a porous material having pores smaller than the trapping matrix or parts thereof but larger than the particles.
The trapping matrix may be any material having a nonzero acoustic contrast relative to the suspending liquid and having pores within, or interstices between parts thereof, through which the sample and particles may flow so as to perfuse the trapping matrix. The trapping matrix may comprise or consist of one or more of oxide, glass, metal, and polymer.
The trapping matrix may comprise or consist of a porous solid body, such as a porous mesh. A porous mesh may for example be a metal mesh such as an aluminium mesh.
The trapping matrix may alternatively comprise or consist of a plurality of parts, such as a plurality of trapping particles, trapping structures, or trapping fibers. The parts may be porous or non-porous. If the parts have pores, then the minimum diameter of the pores are preferably larger than the maximum diameter of the particles. The parts may be separate from each other, or may be attached to each other or to a common surface or support. When the parts are separate from each other they may be considered to form a packed bed when positioned in the cavity.
Trapping particles may comprise particles made of a polymeric material. The trapping particles may for example comprise particles made from oxide, glass, metal, polymer, or polystyrene. Trapping particles may have maximum dimensions of 10-500 µm, preferably 10-200 µm, more preferably 50-200 µm, such as 50-150 µm, for example 75-125 µm or 90-110 µm. Trapping particles encompasses trapping beads and trapping granules. Trapping structures may comprise pillars or protrusions. The pillars or protrusions are preferably monolithic. The pillars or protrusions may preferably be formed from, or attached to, an inner surface of the cavity. The pillars or protrusions may be arranged randomly or in arrays.
Trapping fibers may comprise balls or amounts of fibers. Trapping fibers may for example comprise glass wool, rock wool.
As for the particles, the trapping matrix comprising or consisting of a material having a non-zero acoustic contrast factor relative to the suspending liquid. This provides that an acoustic field applied to the cavity will affect the trapping matrix differently compared to how it affects the suspending liquid. It is believed that the sound field applied to the cavity generates secondary sound fields in and around the trapping matrix due to the changing acoustic contrast at the interface between suspending liquid and trapping matrix. It is believed that it is this secondary acoustic field that provides the trapping effect on the particles.
The trapping matrix may be configured to provide increased trapping efficiency of a first type of particle relative to a second type of particle. This may be obtained by selecting the trapping matrix to have at least one property configured to provide increased trapping efficiency of the first type of particle. The at least one property may be selected from the group consisting of:

minimum and maximum diameter,
shape,
minimum and maximum pore diameter,
minimum and maximum interstice diameter,
hydrophobicity,
affinity,
electrostatic charge, and
surface energy.

In particular, the trapping matrix may comprise or consist of beads such as size exclusion chromatography beads. Alternatively, the trapping matrix may comprise or consist of other types of resins or beads as used in chromatography, such as hydrophobic beads or resins, beads or resins with affinity ligands, etc.
This is advantageous in that it can simplify collecting the first type of particle after release. As an example, a trapping matrix having pores the minimum diameters of which are larger than the maximum diameter of the first type of particle, but smaller than the maximum diameter of the second type of particle, will allow the first type of particle to enter the pores of the trapping matrix whereas the second type of particle may only enter the interstices (if any) of the trapping matrix. When the ultrasound transducer is active, the acoustic attraction is expected to be higher on the first type of particle since the acoustic interaction force scales with the distance of the sound scatterer (trapping matrix) and the to be trapped particle, which would cause the first type of particle to be preferentially trapped. This may provide a higher trapping efficiency. When the ultrasound transducer is deactivated, the second type of particle may be easily and quickly released from the trapping matrix and washed away before the first type of particle finds its way out of the pores of the trapping matrix. Accordingly, there is a greater separation in time between the release and washing out of the different types of particles which simplifies collecting each type of particle separately. This separation time may be further increased by reactivating the transducer after most of the second type of particles has left the trapping matrix, but before most of the first type of particle has left the trapping matrix.
It is further contemplated that the trapping matrix may comprise two or more sections differing in at least one property listed above.
This is advantageous in that it provides for trapping different types of particles in the two or more sections.
Even where the trapping matrix has the same properties throughout, it has been observed that different types of particles may be more or less efficiently trapped in different sections of the trapping matrix. This difference in trapping efficiency could be further increased by utilizing multiple transducers actuated at different frequencies to form trapping regions that are more efficient for different particle sizes, depending on the frequency. Multiple transducers can also be used to release particles trapped in different section of the trapping matrix separately, e.g. subpopulations preferentially trapped in separate sections, or to stack the released particles into a smaller volume e.g. by turning off the transducers sequentially as this volume is passing or by pulling the particles together over fewer transducers than originally used to capture them.
As an alternative to using a retaining device, the trapping matrix may be configured to prevent it from leaving the cavity. In particular, the trapping matrix, or part of it, may be solidified, compounded, joined together, or rigidified, to prevent it from leaving the cavity.
In a preferred embodiment the trapping matrix, or at least a part thereof, is sintered, by heating, to prevent it from leaving the cavity. Accordingly, the trapping matrix, or part thereof, is heated to cause individual parts, preferably trapping particles, of the trapping matrix to attach to each other and thereby form a larger structure which cannot leave the cavity. Accordingly, one part of the trapping matrix, preferably an end part of the trapping matrix, more preferably the end part that is closest to the outlet of the cavity, may be hated to cause the individual parts, e.g. the individual trapping particles, to sinter together to form a larger structure. This larger, compounded, structure of trapping particles joined together is too large to leave the cavity through the outlet, and further acts a retaining device for preventing the other, non-sintered, trapping particles from leaving the cavity.
As an example, polystyrene trapping particles may be easily sintered by heating.
As a further alternative to using a retaining device or solidifying, compounding, joining together or rigidifying the trapping matrix, the trapping matrix or part thereof may be fixed to one or more inner surfaces of the cavity, for example using an adhesive.
The inlet is suitable for introducing the sample of disperse liquid into the cavity. The inlet may be a combined inlet and outlet. More than one inlet may be provided to the cavity.
The cavity may further have an outlet. The outlet is suitable for removing the sample of disperse liquid from the cavity. More than one outlet may be provided to the cavity.
The ultrasound transducer is preferably a bulk ultrasound transducer providing a bulk acoustic wave (BAW), or alternatively a thin film ultrasound transducer. The ultrasound transducer preferably comprises a piezoelectric or electrostrictive material. Suitable materials for ultrasound transducers include a crystalline material having a non-centrosymmetric crystal structure such as langasite (La3Ga5SiO14), gallium orthophosphate (GaPO4) and lithium niobate (LiNbO3), lithium tantalate (LiTaO3), or a ferroelectric ceramic with randomly oriented grains such as lead titanate (PbTiO3), potassium niobate (KNbO3), sodium tungstate (Na2WO3) lead zirconate titanate (PZT) with the formula (Pb[ZrxTi1-x]O3 with 0 ≤ x ≤ 1), as well as lead-free ceramics such as sodium potassium niobate ((K,Na)NbO3), bismuth ferrite (BiFeO3), sodium niobate (NaNbO3), barium titanate (BaTiO3), bismuth titanate (Bi4Ti3O12), and sodium bismuth titanate (NaBi(TiO3)2), for example. Further materials include lead magnesium niobate (PMN), lead magnesium niobate-lead titanate (PMN-PT), and lead lanthanum zirconate titanate (PLZT). Other possible materials include molybdenum disulfide which exhibits piezoelectricity also in monolayer form.
A thin film ultrasound transducer may preferably have a thickness that is less than 100 µm, preferably less than 10 µm, more preferably 5 µm or less, such as 0.01 to 5 µm, more preferably 3 µm or less, such as 0.5 to 3 µm, more preferably 2 µm or less, such as 0.5 to 2 µm, such as 0.5 to 1.5 µm, such as 1 µm or less than 1 µm.
The ultrasound transducer may for example comprise a piezoelectric or electrostrictive material selected from the group consisting of zinc oxide, aluminum nitride, scandium-doped aluminum nitride, cerium oxides, and lead-zirconate-titanate.
Preferably the ultrasound transducer comprises a layer of piezoelectric or electrostrictive material placed between a first electrode layer defining a first side of the ultrasound transducer and a second electrode layer defining a second side of the ultrasound transducer, and wherein the first side of the thin film ultrasound transducer is provided in acoustic contact with an outer surface of the substrate. The first and second electrode layers should be made of an electrically conductive material such as a metal, for example silver or gold.
Typically, the ultrasound transducer comprises a piezoelectric or electrostrictive material which is caused to vibrate by an electric drive signal. However, it is further contemplated within the context of the technology proposed herein that also other types of ultrasound transducer materials, such as magnetostrictive materials (which change dimension in magnetic fields) and thermoacoustic materials (which emit acoustic vibrations in response to temperature changes) can be used. It is further contemplated within the context of the technology proposed herein that the ultrasound transducer could employ an electrostatic function, whereby vibrations are caused by varying the electrostatic attraction and/or repulsion between electrodes of the ultrasound transducer.
The ultrasound transducer is in acoustic contact with the substrate. This encompasses that the ultrasound transducer is in direct physical contact with the substrate or in indirect physical contact via for example an acoustically conducting material. Such an acoustically conducting material may be a solid material such as glass or metal, or may be a liquid material such as a gel or a fluid.
Preferably the transducer is attached to the substrate, for example by an adhesive.
Preferably, the ultrasound transducer is in acoustic contact with an outer surface of the substrate. Preferably the ultrasound transducer is provided in acoustic contact with a majority, such as all, of one surface of the substrate, Alternatively, the ultrasound transducer is provided in contact with a part of the one surface.
The ultrasound transducer may generally have a thickness dimension, a width dimension, and a length dimension. Generally, the length dimension is the longest dimension, and the thickness dimension is the shortest dimension. Alternatively, or additionally, the thickness dimension is the dimension that is perpendicular to the surface of the substrate with which the ultrasound transducer is in acoustic contrast with. Further alternatively or additionally, the thickness dimension of the ultrasound transducer extends in the same direction as the height dimension of the cavity. Accordingly, the length dimension of the ultrasound transducer preferably extends in the same direction as the length dimension of the cavity, and the width dimension of the ultrasound transducer extends in the same direction as the width dimension of the cavity.
Accordingly, the length dimension of the cavity is the longest dimension of the cavity, and the height dimension of the cavity is the shortest dimension of the cavity.
Introducing the sample of disperse liquid into the cavity may comprise any of injecting, flowing, and pumping, the sample of disperse liquid into the cavity. The sample of disperse liquid may be introduced into the cavity by one or more of pumping, suction, capillary action and gravity flow. The sample of disperse liquid may be flowed through the cavity or may be positioned in the cavity and then allowed to remain stationary, i.e., stopped flow.
The steps are preferably performed in the order (i), (ii), (iii).
Step (ii) of introducing the sample of the disperse liquid into the cavity may be performed before, or at the same time as, step (iii) of actuating the ultrasound transducer. Further, step (iii) of actuating the ultrasound transducer may be performed before step (ii) of introducing the sample of the disperse liquid into the cavity.
The sample of disperse liquid is introduced into the cavity so that the sample perfuses the trapping matrix. Expressed differently, a sufficient amount of the disperse liquid is introduced into the cavity so that at least part of the sample perfuses the trapping matrix.
The sample perfuses the trapping matrix by entering into any pores within the trapping matrix, and/or into any interstices between parts of the trapping matrix.
The ultrasound transducer is actuated at the actuation frequency. Expressed differently, the ultrasound transducer driven, or energized, to emit ultrasound having a frequency corresponding to the actuation frequency. The actuation frequency is the frequency of the drive signal to the ultrasound transducer, i.e. the frequency at which the ultrasound transducer vibrates and emits ultrasound.
It is expressed that the actuation frequency does not correspond to resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with sample of the disperse liquid and optionally also the trapping matrix, and couples an acoustic wave into the substrate. Worded differently, the actuation frequency is a frequence that provide an acoustic wave in at least a part of the substrate, preferably the whole substrate, more preferably the whole substrate including the cavity filled with the sample of disperse liquid. Further, the actuation frequency does not correspond to a resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with the fluid and optionally also the trapping matrix. This means that the actuation frequency does not give rise to a one (one of length, width, and height), two (any combination of two of length, width, and height), or three dimensional (all of length, width, and height) standing wave in the cavity, wherein the dimensions are length, width, and height. Such a one, two, or three dimensional standing wave has the inner surfaces (e.g. side walls, floor and ceiling) of the cavity as boundaries or endpoints for the standing wave. For such resonances, the resonance frequencies are dependent on the dimensions of the cavity because in general, in order for a standing wave to form, the dimension of the cavity must correspond to nλ/2 where n is a positive integer and λ is the wavelength of the acoustic standing wave, this wavelength being inversely proportional to the frequency. In other words, such one or two-dimensional resonances are resonances where the dimensions of the cavity, e.g. width, height, and/or length, correspond to an integer number of half wavelengths of the ultrasound.
As noted above, this is a major difference to the prior art. Further, as seen in the example section, the trapping efficiency is actually degraded if the actuation frequency should correspond to such a resonance.
Generally, it is further preferred when the actuation frequency additionally does not correspond to an eigen frequency of a majority of individual parts of the trapping matrix. Thus, when the trapping matrix comprises a plurality of parts, such as a plurality of trapping particles or plurality of trapping structures, it is preferred that the actuation frequency does not correspond to an eigen frequency of a majority such parts. As seen in the examples, better trapping is seen at frequencies other than such eigen frequencies. This is also a significant difference from the prior art.
Although specified in the method and system according to the first and second aspects of the technology proposed herein, the actuation frequency may, as seen in the examples, alternatively be any frequency. Any frequency thus includes frequencies that a) corresponds to a resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with sample of the disperse liquid and optionally also the trapping matrix, and/or does not couple an acoustic wave into the substrate. In particular, the actuation frequency may be a frequency that corresponds to a resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with sample of the disperse liquid and optionally also the trapping matrix.
The actuation frequency is typically in the ultrasound range, preferably in the range of 0.1 to 20 MHz, more preferably in the range of 0.1 to 8 MHz, most preferably in the range of 0.1 to 5 MHz.
Suitable actuation frequencies may be determined by an impedance spectrometry analysis of the ultrasound transducer attached to the acoustofluidic device with the trapping matrix and with the sample in the cavity. The impedance spectrometry analysis of the ultrasound transducer comprises driving the ultrasound transducer using a range of frequencies and determining the impedance of the ultrasound transducer for each frequency to obtain an impedance spectrum. In such a spectrum a resonance frequency of the acoustofluidic device may be identified as a minimum. Consequently, an antiresonance frequency may be identified as a maximum. An admittance spectrum may be used instead of an impedance spectrum. In admittance spectrum a resonance frequency may be identified as a maximum whereas an antiresonance frequency may be identified as a minimum. As noted further below, suitable actuation frequencies may be advantageously found near or between resonance frequencies and antiresonance frequencies.
Alternatively, suitable actuation frequencies can be determined experimentally, by measuring the amount of particles that are trapped by the trapping matrix for different actuation frequencies.
The drive circuit may comprise a function generator electrically connected to the ultrasound transducer.
Connected to the ultrasound transducer means that the drive circuit is connected in such a way as to be able to drive the ultrasound transducer. Typically, the drive circuit is connected electrically to the ultrasound transducer so as to drive the ultrasound transducer using an electrical drive signal.
The drive circuit is configured to drive the ultrasound transducer at the actuation frequency described above.
Preferably, the actuation frequency corresponds to:

c. a frequency in a range extending between two neighbouring resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity, or preferably:
d. a frequency in a range extending between a resonance frequency and its neighbouring anti-resonance frequency, of the acoustofluidic device including the sample of disperse liquid in the cavity, or more preferably:
e. a frequency that is within 10% of a resonance frequency or an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity, or most preferably:
f. a frequency that is within 10% of an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity.

As regards (c), it was found in the examples that trapping of the particles is possible in a very wide range of frequencies, i.e. between two neighbouring resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity. Thus, in contrast to the prior art, a wide range of actuation frequencies are possible, and the wide range of actuation frequencies are based on the resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity, not the resonance frequencies of the cavity filled with the sample of disperse liquid.
As regards (d), it was found in the examples that the range between a resonance frequency and its neighbouring anti-resonance frequency, of the acoustofluidic device including the sample of disperse liquid in the cavity, provided even better trapping.
As regards (e), it was found in the examples that frequencies close to a resonance frequency or an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity provided very useful trapping, e.g. more than 15% trapping efficiency. The best trapping was obtained when the actuation frequency was close to an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity (f).
Additionally, the actuation frequency may correspond to a resonance frequency or an anti-resonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity.
Typically, resonance frequencies and anti-resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity are found within 10% of corresponding resonance frequencies and anti-resonance frequencies of the ultrasound transducer. As an example, when a substrate in the form of a mm-scaled glass capillary is combined with an ultrasound transducer in the form of a piezoelectric element, resonance frequencies and anti-resonance frequencies of the combination including the sample of disperse liquid in the cavity are generally found within 10% of resonance frequencies and anti-resonance frequencies of the transducer. This is illustrated in the examples where an inherent width resonance of the ultrasound transducer alone at 440 kHz corresponds to a resonance frequency of 447 kHz (about 2% higher) for the acoustofluidic device with the ultrasound transducer attached. Similarly, an inherent thickness resonance of the ultrasound transducer alone at 4.3 MHz corresponds to a resonance frequency of 4.7 MHz (about 10 % higher) for the acoustofluidic device with the ultrasound transducer attached.
Accordingly, the alternatives (c)-(f) may alternatively be expressed such that the references to resonance and anti-resonance frequencies apply to the ultrasound transducer instead of to the acoustofluidic device.
In other words, the actuation frequency may correspond to:

c. a frequency in a range extending between two neighbouring resonance frequencies of the ultrasound transducer, or preferably:
d. a frequency in a range extending between a resonance frequency and its neighbouring anti-resonance frequency, of the ultrasound transducer, or more preferably:
e. a frequency that is within 10% of a resonance frequency or an antiresonance frequency of the ultrasound transducer, or most preferably:
f. a frequency that is within 10% of an antiresonance frequency of the ultrasound transducer.

As an alternative to percentages, resonance frequencies and anti-resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity may be found within 30kHz (for transducer resonances below 1000 kHz) or within 0.4 MHz (for transducer resonances above 1MHz) of resonance frequencies and anti-resonance frequencies of the ultrasound transducer.
The deviation or shift in resonance frequency may differ depending on the mass of the substrate compared to the mass of the ultrasound transducer.
Further, as found in the examples, actuation frequencies that were within 10% of inherent length or width dimension, preferably width dimension, resonances or anti-resonances of the ultrasound transducer were generally preferred. This applied especially when the actuation frequency was within 10% of an antiresonance frequency of the width dimension of the ultrasound transducer.
Generally, it was found that, using an actuation frequence within 10% of a resonance frequency of the width or length dimension of the ultrasound transducer provided efficient trapping. Increasing the power supplied to the ultrasound transducer however provided more heating of the ultrasound transducer and thereby of the acoustofluidic device.
In contrast, using an actuation frequence within 10% of an antiresonance frequency of the width or length dimension of the ultrasound transducer provided less efficient trapping at the same power used for the resonance frequency, but this power input generated less heat. This allowed for obtaining even better trapping by using a higher power as compared to the resonance frequency while reaching the same temperature.
A length dimension of the ultrasound transducer here corresponds to, or is parallel to, a length dimension of the cavity.
A width dimension of the ultrasound transducer here corresponds to, or is parallel to, a width dimension of the cavity, and is therefore also transverse to the length dimension of the cavity.
Generally, the ultrasound transducer may have a first surface in acoustic contact with the substrate and an opposite second surface, a thickness defined as the perpendicular distance between the first and second surfaces, a length defined as the largest dimension of the first and second surfaces, and a width defined as the smallest dimension of the first and second surfaces.
The maximum diameter of the particles is preferably 10 µm or less, more preferably 2 µm or less, and the maximum diameter of the particles is preferably at least 30 nm, more preferably at least 80 nm, or alternatively the maximum diameter of the particles is preferably 30-380 nm, more preferably 80-380 nm. Sizes of particles may be measured according to ISO 13320-1, or may be measured based on specific surface area (SSA) [m²/g] which can be determined according to ISO 9277:2022. Preferably the diameters of particles may be measured by nanoparticle tracking analysis (NTA). In NTA Brownian motion is analyzed by video - individual particle positional changes are tracked in two dimensions from which the particle diffusion is determined. Knowing the diffusion constant (Dt), the particle hydrodynamic diameter can be then determined. Additionally, the diameter of the nanoparticles may be measured using dynamic light scattering (DLS) or microscopy. The maximum diameter of a particles is the diameter if the particles is spherical, and the length, i.e. the largest dimension, of the particle if the particle is non-spherical.
Correspondingly the minimum diameters of the pores, if the trapping matrix is porous, and/or the minimum diameter of interstices between parts of the trapping matrix, if the trapping matrix comprises individual parts, is preferably 100 µm or less in diameter, more preferably 20 µm or less, and preferably at least 300 nm, more preferably at least 800 nm, or alternatively preferably 300 nm to 3.8 µm, more preferably 800 nm to 3.8 µm.
Accordingly, it is generally preferred that maximum diameter of the particles is at the most 1/10 of the minimum diameter of the pores and/or interstices in the trapping matrix. Conversely it is generally preferred that the minimum diameter of the pores and/or interstices in the trapping matrix is at least 10 times the maximum diameter of the particles.
As seen in example 1, 100 µm trapping particles were used as a trapping matrix to trap 380 nm polystyrene nanoparticles. In this example the trapping particles were about 263 times larger than the nanoparticles. Accordingly, the minimum dimension of the parts of the trapping matrix, e.g. trapping particles of the trapping matrix, is preferably at least 50 times, more preferably at least 100 times, most preferably at least 150 times, larger than the maximum dimension of the particles.
Preferably the trapping matrix comprises or consist of one or more of: porous or nonporous beads, particles, granules or fibers, monolithic pillars, macroporous polymer monoliths, glass wool, aluminium and metal mesh.
These are suitable matrices for trapping the particles.
Preferably the cavity further comprises an outlet, and step ii comprises flowing the sample of the disperse liquid through the cavity.
Correspondingly, in the system, the cavity further comprises an outlet, and the system further comprises:

c) a sample container for holding the sample of the disperse liquid, and
d) a sample pump arranged to pump the sample of the disperse liquid from the sample container to the inlet for pumping the sample of the disperse liquid through the cavity.

This is advantageous in that it allows continuous trapping, and therefore collecting, of particles in the trapping matrix. This allows for trapping larger numbers of particles.
The outlet is preferably positioned so that trapping matrix is provided between the inlet and the outlet. The outlet may thus be provided a one end of the cavity and the inlet provided at another, opposite, end of the cavity. When the cavity is elongated, e.g. when the cavity is a capillary or tube, then the inlet and outlet are preferably provided at opposite ends of the capillary or tube.
Flowing the sample of the disperse liquid through the cavity encompasses that the sample of the disperse liquid is caused to flow into the cavity through the inlet, through the cavity to the outlet, and out of the cavity through the outlet. Flowing the sample of the disperse liquid through the cavity thus entails that the sample, or at least part of it, perfuses the trapping matrix. As above, various flow rates are possible depending on the dimension of the cavity and the desired trapping efficiency.
The sample container may be any container suitable for holding a liquid sample, but is preferably a test tube or other closed or closable container.
The sample pump may be any type of pump, such as a syringe pump or a peristaltic pump.
The sample pump may be arranged to pump the sample of the disperse liquid from the sample container to the inlet for pumping the sample of the disperse liquid through the cavity by comprising, or being connected to, pipes or tubing fluidly connecting the sample pump to the sample container and to the cavity.
The method may further comprise the step of
iv. introducing a wash buffer into the cavity so that the wash buffer perfuses the trapping matrix.
Correspondingly, the system may further comprise:

e) a buffer container for holding a wash buffer, and
f) a buffer pump arranged to pump the wash buffer from the buffer container to the inlet for pumping the wash buffer through the cavity.

This is advantageous in the wash buffer may be used to wash away any remnants of the sample of disperse liquid. Accordingly, the wash buffer may be used to wash the particles. Preferably step (iv) is performed concurrently with step (iii), i.e. while the ultrasound transducer is actuated as that keeps the particles trapped during the washing.
Introducing the wash buffer into the cavity may comprise any of injecting, flowing, and pumping, the wash buffer into the cavity. The wash buffer may be introduced into the cavity by one or more of pumping, suction, capillary action and gravity flow. The sample of disperse liquid may be flowed through the cavity or may be positioned in the cavity and then allowed to remain stationary, i.e., stopped flow.
The wash buffer may be any liquid different from the sample of disperse liquid. The wash buffer may for example comprise one or more of water (H₂O), a buffer system, sodium ions, a buffer system such as a phosphate or carbonate buffer system, and a surfactant. The wash buffer may further comprise a cell medium.
The wash buffer is introduced into the cavity. Accordingly, the wash buffer will flow from the inlet to the outlet though the cavity and thus perfuse the trapping matrix. The wash buffer may thus displace and/or dilute any remaining disperse liquid present in the cavity and/or in the trapping matrix.
The buffer container may be any container suitable for holding the wash buffer but is preferably a test tube or other closed or closable container.
The buffer pump may be any type of pump, such as a syringe pump or a peristaltic pump.
The buffer pump may be arranged to pump the wash buffer from the buffer container to the inlet for pumping the wash buffer through the cavity by comprising, or being connected to, pipes or tubing fluidly connecting the buffer pump to the buffer container and to the cavity.
The method may further comprise the steps of:

v. decreasing, or ceasing, actuating the ultrasound transducer so as to release particles trapped by the trapping matrix, and/or
vi. repeatedly and/or pulsingly actuating the ultrasound transducer, preferably at a resonance frequency of the transducer, so as to release particles trapped by the trapping matrix.

Accordingly, the drive circuit, or a control unit comprised by the system, may be configured to decrease, or cease, actuating the ultrasound transducer at the actuation frequency.
This is advantageous in that it allows the trapped particles to be released from the trapping matrix. Worded differently, decreasing, or ceasing, actuating the ultrasound transducer corresponds to decreasing or removing the actuation of the acoustofluidic device and thereby decreasing or removing the forces trapping the particles. Additionally, the amount of released particles may be increased by repeatedly, i.e. as pulses, actuating the ultrasound transducer so as to cause the particles to be released from the trapping matrix.
The method may be monitored by detecting the particles at the inlet to the cavity and/or in the cavity and/or at the outlet of the cavity. The monitoring may comprise monitoring the content of particles that are being trapped by the trapping particles.
The system may thus comprise:

g) a particle detector arranged to detect particles entering and/or exiting the cavity.

The particles may be detected at both the inlet and the outlet of the cavity, using a particle detector at each of the inlet and outlet.
The particle detector may for example comprise an optical scattering detector.
The system may generally comprise a control circuit configured to cause the system to carry out the method according to the first aspect of the technology proposed herein by controlling one or more of the drive circuit, the sample pump, the buffer pump, and the particle detector.

Brief description of the drawings and detailed description

A more complete understanding of the abovementioned and other features and advantages of the technology proposed herein will be apparent from the following detailed description of preferred embodiments in conjunction with the appended schematic drawings, wherein:

Fig. 1A: shows a top view of an embodiment of an acoustofluidic device for use in the method according to the first aspect of the technology proposed herein.
Fig. 1B: shows a cross section of the acoustofluidic device of Fig. 1A along line B-B'
Fig. 1C: shows a cross section of the acoustofluidic device of Fig. 1A along line C-C'.
Fig. 2A: shows, in top view and partial cut-away, a first alternative embodiment of a retaining device in the form of retaining structures for retaining the trapping particles in the cavity of the acoustofluidic device of Fig 1A-1C.
Fig. 2B: shows a cross section along line B-B' in Fig. 2A.
Fig. 3A: shows, in top view and partial cut-away, a second alternative embodiment of a retaining device in the form of retaining particles for retaining the trapping particles in the cavity of the acoustofluidic device of Fig 1A-1C.
Fig. 3B: shows a cross section along line B-B' in Fig. 3A.
Fig. 4A: shows, in top view and partial cut-away, a third alternative embodiment of a retaining device in the form of a porous block for retaining the trapping particles in the cavity of the acoustofluidic device of Fig 1A-1C.
Fig. 4B: shows a cross section along line B-B' in Fig. 4A.
Fig. 5A: shows how nanoparticles, in a sample being introduced into the cavity of the acoustofluidic device of Fig. 1A-1C, pass the trapping particles when the ultrasound transducer is not actuated.
Fig. 5B: shows how nanoparticles, in a sample being introduced into the acoustofluidic device of Fig. 1A-1C, are trapped in and among the trapping particles when the ultrasound transducer is actuated.
Fig. 5C: shows how nanoparticles, in a sample being introduced into the acoustofluidic device of Fig. 1A-1C, have accumulated in and among the trapping particles when the ultrasound transducer is actuated.
Fig. 5D: shows how nanoparticles are released from the trapping particles when the when the ultrasound transducer is not actuated and buffer is introduced into the cavity of the acoustofluidic device.
Fig. 6: shows an embodiment of a system according to the second aspect of the technology proposed herein for performing the method according to the first aspect of the technology proposed herein.
Fig. 7: shows an example of a 3D printed polymeric substrate for use in the method and system according to the first and second aspects of the technology proposed herein.
Fig. 8A: shows an embodiment of an acoustofluidic device having a trapping matrix with two sections for use in the method and system according to the first and second aspects of the technology proposed herein.
Fig. 8B: shows an embodiment of an acoustofluidic device having two separate sections of a trapping matrix.
Fig. 8C: shows a further acoustofluidic device which uses a trapping matrix in the form of a porous body.
Fig. 9: shows an embodiment of a method according to the first aspect of the technology proposed herein.
Fig. 10A: shows the fluorescence intensity measured at the outlet of a cavity in an experiment trapping 2 µm polystyrene particles using a method and system according to the first and second aspects of the technology proposed herein.
Fig. 10B: shows the fluorescence intensity measured at the outlet of a cavity in an experiment trapping 380 nm particles using a method and system according to the first and second aspects of the technology proposed herein.
Fig. 11A: shows the trapping efficiency for different frequencies used in the experiment shown in Fig. 10A and Example 1.
Fig. 11B: shows the input power for different frequencies used in the experiment shown in Fig. 10A and Example 1.
Fig. 11C: shows the peak-to-peak voltage over the transducer for different frequencies used in the experiment shown in Fig. 10A and Example 1.
Fig. 12A: shows the trapping efficiency for a wider range of frequencies used in the experiment shown in Fig. 10A and Example 1.
Fig. 12B: shows the input power for a wider range of frequencies used in the experiment shown in Fig. 10A and Example 1.
Fig. 13A: shows trapping efficiency for a wider range of frequencies including both width and thickness resonances in the ultrasound transducer when the experiment shown in Fig. 10A and Example 1
Fig. 13B: shows input power and current for a wider range of frequencies including both width and thickness resonances in the ultrasound transducer when the experiment shown in Fig. 10A and Example 1.
Fig. 14: shows the fluorescence intensity measured at the outlet of a 3D-printed polymer substrate cavity in an experiment trapping 2 µm particles using a method and system according to the first and second aspects of the technology proposed herein.
Fig. 15: shows trapping of 270 nm particles in an experiment similar to Example 2 and Fig. 10B.
Fig. 16: shows an impedance spectrum for a 1 mm thick ultrasound transducer used together with the substrate of example 2 and actuated over a range corresponding to a thickness resonance of the ultrasound transducer which also corresponded to a cavity resonance at 2 MHz.
Fig. 17: shows an admittance spectrum for the frequency range of 401 kHz to 494 kHz for the acoustofluidic device used in Example 1 and Fig. 10A.

Fig. 1A shows a top view of an embodiment of an acoustofluidic device 10 for use in the method according to the first aspect of the technology proposed herein. The acoustofluidic device 10 comprises a substrate in the form of a glass capillary 12 whose outer wall 14 enclose a cavity 16 having an inlet 18 and an outlet 20. Cavity 16 is partially filled by trapping particles, one of which is designated the reference numeral 30, which together form trapping matrix 32. The trapping particles are at least partially retained in the cavity 16 by a retaining device in form of a screen 40 arranged between the trapping matrix 32 and the outlet 20. An ultrasound transducer 50 is attached to the outer wall 14 on the underside of the glass capillary 12.
As seen in Fig. 1A, 1B, and 1C, the trapping matrix 32 extends between the inner walls 22, 24 and between the floor 26 and ceiling 28 of the cavity 16.
Fig. 2A shows, in top view and partial cut-away, a first alternative embodiment of a retaining device in the form of retaining structures 40' for retaining the trapping particles 30 in the cavity of the acoustofluidic device of Fig 1A-1C. The retaining structure 40' may be fabricated integrally with the substrate 12 or may be attached to the substrate 12 after fabrication of the latter. The retaining structures serve to retain the trapping particles 30 of the trapping matrix 32 so that they are not washed out of the cavity 16 when a sample 2 of a disperse fluid (see fig 5A-5D and Fig. 6) is introduced into the cavity 16. The retaining structures 40' are further advantageous in that they allow simple exchange of the trapping matrix 32 by simple backflushing of the cavity 16 before introduction of a fresh amount of trapping particles 30 to form a fresh trapping matrix 32.
As seen in Fig. 2B the retaining structure 40' extends from the floor 26 to the ceiling 28, however, it is not necessary as long as any gap between the retaining structure 40' is sufficiently small to prevent the escape of trapping particles 30.
Fig. 3A shows, in top view and partial cut-away, a second alternative embodiment of a retaining device in the form of retaining particles 40" for retaining the trapping particles in the cavity of the acoustofluidic device of Fig 1A-1C. The retaining particles 40" comprise particles that are larger than the trapping particles 30 as well as larger than at least the outlet of the cavity or tubing connected to the outlet. The retaining particles 40" are advantageous in that they do not require the provision of any fixed structure within the cavity 16 which decreases the cost of fabricating the acoustofluidic device 10.
As seen in Fig. 3B the retaining particles 40" do not need to extend the full distance between the floor 26 and ceiling 28 as long as any gap between a retaining particle 40" and the ceiling is sufficiently small to prevent escape of trapping particles 30.
Fig. 4A shows, in top view and partial cut-away, a third alternative embodiment of a retaining device in the form of a porous block 40‴ for retaining the trapping particles 30 in the cavity of the acoustofluidic device of Fig 1A-1C. The porous block 40‴ may be any type of porous solid material that has pores smaller than the trapping particles 30. The porous block provides a compromise between retaining structures and retaining particles since it does not need to be fabricated with, or attached to, the interior of the cavity while at the same time being easier to handle than a plurality of retaining particles.
As seen in Fig. 4B the porous block 40‴ may advantageously extend from the floor 26 to the ceiling 28 to prevent, by interference, its dislocation from the intended position in the cavity.
Fig. 5A shows how particles 4, in a sample 2 being introduced into the cavity of the acoustofluidic device 10 of Fig. 1A-1C, pass the trapping particles 30 when the ultrasound transducer 50 is not actuated. The arrow marked "sample" shows the direction of flow. As seen, in the absence of ultrasound energy in the cavity, the particles, which are significantly smaller than the trapping particles 30, slip past the trapping particles 30 while being suspended in the suspending liquid of the sample of disperse liquid 2. Alternatively, or additionally, the particles 4 may pass through any pores of the trapping particles 30 (not shown).
Fig. 5B shows how particles 4, in a sample 2 being introduced into the acoustofluidic device 10 of Fig. 1A-1C, are trapped in the pores of, and among, i.e. in the interstices of, the trapping particles 30 when the ultrasound transducer 50 is actuated as indicated by dashed curved lines W. The arrow marked "sample" shows the direction of flow. The ultrasound energy provided into the acoustofluidic device 10 is believed to cause secondary acoustic force fields in and among the trapping particles 30 which attract and retain the particles 4.
Fig. 5C shows how particles 4, in a sample 2 being introduced into the acoustofluidic 10 device of Fig. 1A-1C, have accumulated in the pores of, and among, i.e. in the interstices of, the trapping particles 30 when the ultrasound transducer 50 is actuated as indicated by dashed curved lines W. The arrow marked "sample" shows the direction of flow. As long as the ultrasound transducer 50 is actuated and sample 2 is introduced into the cavity 16, the particles 4 will accumulate in and among the trapping particles 30. This accumulation proceeds until the secondary acoustic force fields have become so attenuated by the presence of particles 4 among the trapping particles 30 that further nanoparticles 4 are no longer trapped and pass through the trapping matrix 32.
Fig. 5D shows how particles 4 are released from the trapping particles 30 when the ultrasound transducer 50 is not actuated and buffer 6 is introduced into the cavity 16 of the acoustofluidic device 10. The arrow marked "buffer" shows the direction of flow. The method and system according to the technology proposed herein thus allows trapping, and consequently separating, isolating, and/or concentrating, particles 4 form a sample 2 of a disperse liquid.
Fig. 6 shows an embodiment of a system 100 according to the second aspect of the technology proposed herein for performing the method according to the first aspect of the technology proposed herein.
The system 100, in addition to an acoustofluidic device 10' which differs from the acoustofluidic device 10 in Fig. 1A-1C merely by being shorter and by the trapping matrix 32 filling essentially all of the cavity 16, comprises an inlet connection 110 and an outlet connection 120, an inlet selector valve 130 and an outlet selector valve 140, and an inlet particle detector 150 and an outlet particle detector 160. Upstream of the acoustofluidic device 10 a sample container 170, for holding the sample 2, is connected to the inlet selector valve 130 via sample pump 172 and tubing 174. Likewise, a buffer container 180 is connected to the inlet selector valve 130 via buffer pump 182 and tubing 184. Downstream of the acoustofluidic device 10 a collection container 190 is connected to the outlet selector valve 140 via tubing 192. A waste container 200 for receiving waste 8 is also connected to the outlet selector valve 140 via tubing 202. A drive circuit 210 is connected for driving the ultrasound transducer 50 via leads 212 and 214. The drive circuit 21, the sample pump 172, the buffer pump 182, the inlet selector valve 130 and the inlet particle detector 150, as well as the outlet selector valve 140 and the outlet particle detector 160, are controlled by a control unit 220 via connections 222, 224, 226, 228, 230, 232, and 234.
The system 100 can thus, using the control unit 220, pump a sample 2 through the acoustofluidic device 10 by controlling sample pump 172 and inlet selector valve 130 while controlling the drive circuit 210 to actuate the ultrasound transducer 50 so as to trap and accumulate particles 4 from the sample 2 in the trapping matrix 32 within the cavity 16. Once the desired number of particles 4 has been trapped, as determinable by the inlet and/or outlet particle detectors 150 and 160, the control unit 220 may control inlet selector valve 130 and buffer pump 182 to pump buffer 6 into the acoustofluidic device 10. Preferably the ultrasound transducer 50 is actuated by the drive circuit 210 during a first phase of pumping the buffer 6 through the acoustofluidic device 10 so as to displace and wash away any remnants of the suspending liquid in the sample 2. In a second phase, while still pumping the buffer 6, the control unit 220 controls the drive circuit 210 so stop actuating the ultrasound transducer so that no acoustic energy is imparted on the acoustofluidic device 10, whereby the particles are released from the trapping matrix 32. The control unit 220 the further controls the outlet selector valve 140 to direct the liquid exiting the acoustofluidic device 10 from the waste container 200, which receives the sample 2 depleted of particles 4 as well as any mixtures of the sample 2 and the buffer 6 during the second phase, as waste liquid 8, into the collection container 190 where the buffer 6 carrying the released particles 4 is collected. In this way, as illustrated in the collection container 190, the particles 4 that were initially present in the sample 2 may become one or more of separated from the sample 2, washed using the buffer, and concentrated.
The control unit 220 may for example be configured to use measurements from the inlet and outlet particle detectors 150 and 160 to control as well as monitor the carrying out of the method. In particular the outlet particle detector 160 can be used to determine when the trapping capacity of the trapping matrix 32 has become exhausted so the particles 4 pass through the acoustofluidic device 10' without becoming trapped. This detection may be used as a signal to start the first phase of washing the trapping matrix 32 using the buffer 6.
Further, the outlet particle detector 160 can be used during the first phase of washing the packed bed determine when any non-bound particles or other remnants of the sample 2 have been displaced from the packed bed and only fresh buffer exits the outlet of the acoustofluidic device 10'. This detection may be used as a signal to enter the second phase in which the drive circuit 210 is deactivated and the outlet selector valve 140 controlled to lead the liquid exiting the acoustofluidic device 10' to the collection container 190 instead of the waste container 200.
Fig. 7 shows an example of a 3D printed polymeric substrate 12' for use in the method and system according to the first and second aspects of the technology proposed herein. The substrate 12' has an inlet an outer wall 14' enclosing a cavity 16' with an inlet 18' and an outlet 20'. As shown in the magnification, a retaining device in the form of a screen 40ʺʺ made up from individual pillars forming a comb-like structure has been formed integrally with an inner surface of the cavity 16' close to the outlet 20'.
Fig. 8A shows an embodiment of an acoustofluidic device 10‴ having a trapping matrix 32' with two sections for use in the method and system according to the first and second aspects of the technology proposed herein. The substrate 12 is identical with that shown in Fig. 1A-C, however, as noted, here the trapping matrix 32' comprises porous trapping particles 30 and 30' of two different sizes, where the larger trapping particles 30' are positioned upstream, i.e. closer to the inlet than the outlet. This allows different size particles 4 in the sample of disperse liquid 2 to become even more separated from each other as the large interstices between the larger trapping particles 30' will allow large particles to quickly pass through and into the section of the trapping matrix 32' made up by the smaller particles 30, whereas smaller particles will enter the pores of the larger particles 30' and thus be slowed down more, in addition to being trapped as the ultrasound is actuated.
Fig. 8B shows an embodiment of an acoustofluidic device 10ʺʺ having two separate sections of a trapping matrix 32" with a first section having larger trapping particles 30' and a second section having smaller trapping particles 30. Each section is provided with, or arranged adjacent, an ultrasound transducer 50' and 50'a (which are identical to ultrasound transducer 50 except slightly shorter). The acoustofluidic device 10ʺʺ shown in Fig. 8B is thus capable of selectively, by actuating either or both of the ultrasound transducers 50' and 50'a, trap different size particles in the corresponding different trapping particles 30 and 30' of the two sections of the trapping matrix 32". The acoustofluidic device 10ʺʺ is then further capable of selectively, by deactivating either or both of the ultrasound transducers 50' and 50'a, release the particles trapped in the corresponding section of the trapping matrix 32". A similar result can be achieved by connecting two or more acoustofluidic devices 10 as shown in Fig. 1A-1C in series and using different trapping matrices 32 in the different acoustofluidic devices 10 (not shown). Additionally, the two ultrasound transducers 50' and 50'a may be actuated at different frequencies to further affect or alternate trapping of particles. In this case the trapping matrix may be the same.
Fig. 8C shows a further acoustofluidic device 10ʺ‴ which is identical to acoustofluidic device 10 of Fig. 1A-1C except that it uses a trapping matrix 32‴ in the form of a porous body 30".
Fig. 9 shows an embodiment of a method according to the first aspect of the technology proposed herein. In the first step (i), reference numeral 1, an acoustofluidic device 10 is provided, the device comprising a substrate 12 in which a cavity 16 is formed, the cavity being at least partially filled with a trapping matrix 32, the cavity further having an inlet 18 for introducing a sample of disperse liquid 2 into the cavity, and an ultrasound transducer 50 in acoustic contact with the substrate. In the second step (ii), reference numeral 3, the sample of the disperse liquid 2 is introduced into the cavity 16 so that the sample perfuses the trapping matrix 32. In the third step (iii), reference numeral 5, the ultrasound transducer 50 is actuated at an actuation frequency.
As shown by reference numeral 7, the method may further comprise an optional step (iv) wherein a wash buffer is introduced into the cavity so that the wash buffer perfuses the trapping matrix. This provides for washing away any parts of the sample of disperse liquid that is not trapped by the trapping matrix.
Further, as shown by reference numeral 9, the method may further comprise an optional step (v) comprising decreasing, or ceasing, actuating the ultrasound transducer at the actuation frequency. This step releases trapped nanoparticles from the trapping particles 30 in the packed bed 32.

Example 1: Trapping of 2 µm particles in an acoustofluidic device comprising a packed bed using actuation close to a resonance of the acoustofluidic device

In this example the substrate was a glass capillary with 0.28 mm wall thickness defining a cavity in the form of a rectangular cross section channel with a width of 4 mm and a height of 0.4 mm (2540 Rectangle VitroTubes^™, Vitrocom, USA). A 10 x 4 x 0.5 mm ultrasound transducer was attached with epoxy adhesive to one of the flat sides of the capillary. A 10 mm long bed of 100 µm diameter polystyrene trapping particles was provided as a trapping matrix or packed bed in the cavity. A small portion of the trapping particles near the outlet of the capillary was sintered by heating to keep the trapping matrix in place.
One end (inlet) of the capillary was attached to a valve connected to two syringe pumps, one holding phosphate buffered saline buffer (PBS buffer) and the other one holding particles dissolved in PBS buffer. The other end of capillary (outlet) was connected to a particle detector.
A sample of a disperse liquid comprising particles, was also provided. The particles were green fluorescent polystyrene 2 µm particles. The concentration of particles was 0.01% v/v = 2.5*10⁷ particles / mL
The particles were detected using a fluorescence microscope positioned to detect particles exiting the outlet. The fluorescence microscope was equipped with a CoolLED, FITC filter set and IDS camera.
The flow rate of PBS, and the general flow rate, through the cavity was 50 µL/min.
Initial attempts to trap the particles using either an actuation frequency close to the eigen frequency of the trapping particles (7 MHz) or using an actuation frequency close to resonance in the height direction of the cavity (1.85 MHz), were unsuccessful as no significant change in particle concentration could be detected by the fluorescence microscope at the outlet.
An impedance spectrum was then obtained for actuation frequencies between 1.85 and 7 MHz. An Impedance minimum was observed at 4.7 MHz (close to the thickness resonance of the Piezo at 4.4 MHz). Trapping experiments at this frequency lead to acceptable results. In an attempt to increase performance, a broader spectrum was obtained for 0-10 Mhz. This spectrum revealed an impedance minimum at 460 kHz. When the ultrasound transducer was driven at this frequency, a more significant decrease in particle concentration compared to actuation at 4.7 MHz was measured by the particle detector at the outlet. Further, when the ultrasound was turned off after 1 minute, a significant increase in particle concentration was also detected. Accordingly, actuating the ultrasound transducer at 460 kHz somehow caused particles to become trapped, i.e. retained, in the packed bed of trapping particles, whereas deactivating the ultrasound transducer caused the trapped particles to be released from the trapping particles to follow the flow of sample out of the cavity.
The actuation frequency of 460 kHz was thus used to further explore the trapping efficiency and characteristics.
Accordingly, for the protocol defined by the steps below and illustrated in Fig. 10A, the ultrasound transducer was actuated using a drive signal of 26 V peak-to-peak and 0.16 A, at 460 kHz, providing a power input of 0.44 W. The power input into the ultrasound transducer led to a temperature of 30.5 °C in the acoustofluidic device.
The following steps were used to trap and release the particles:

Step 1. The cavity and associated tubing were flushed with PBS buffer for 2 minutes (Time = 0-2 minutes). During the flushing, a baseline fluorescence amplitude of about 5-10 intensity units was detected.
Step 2. At time t = 2 minutes, 150 µL of the sample of the disperse liquid was injected into the flow of PBS buffer. This corresponded to 3.75*10⁶ particles. The particles were trapped during the two minutes (Time = 2-4 minutes) while a gradual increase in fluorescence intensity, up to about 13 intensity units, was observed, induced by particles that were not trapped in the packed bed, i.e. the trapping matrix, but passed by.
Step 3. At time t = 4 min, the injection of sample was finished and a further 200 µL PBS buffer was allowed to pass through the cavity and associated tubing during about 2 minutes and 45 second. During this time, untrapped particles were washed from the trapping matrix and a gradual decrease of fluorescence was observed down towards the baseline value of 10.
Step 4. At time t = 6 minutes and 45 second, the actuation of the ultrasound transducer was stopped, i.e. the drive signal was turned off. This led to a high and narrow spike in fluorescence up to over 30 intensity units as the trapped particles were released and washed out of the trapping matrix by the PBS buffer until about time = 7 minutes and 45 seconds.

The trapping efficiency was measured by comparing the area of the wide peak formed during trapping and washing (between times 2 and 4 minutes), with the area of the high peak (between time 6 minutes 45 seconds and 7 minutes 45 second) formed during release. Here, the area of the wide and low peak provides a representation of particles lost, i.e. particles that were not trapped but instead passed through the cavity to be detected during the trapping phase (step 2) and the washing phase (step 3).
In contrast, the areas of the high and narrow peak provide a representation of the trapped particles which were released en masse as the drive signal was turned off. The former area was determined as 365 and the latter as 342, meaning that the trapping efficiency was 342/(342+365)=0.48, i.e. 48%.
Alternatively expressed, the efficiency was calculated as the area detected during step 4 divided by the area detected during steps 2 and 3 and 4.
In this experiment, the frequency of the drive signal at 460 kHz clearly did not correspond to an eigen frequency of the 100 µm polystyrene beads used as trapping particles (7 MHz), nor did it correspond to a resonance in the height direction of the cavity (1.85 MHz).
Further, 460 kHz was also far removed from any resonance in the width (4 mm) dimension of the cavity, the wavelengths λ and frequencies f of the closest possible resonances being given by the formulas λ = 2*w/n and f = speed of sound / λ in the table below:

Width n Λ F (water, 1481 m/s)

0.004 m 1 0.008 m 185 kHz

0.004 m 2 0.004 m 370 kHz

0.004 m 3 0.00267 m 555 kHz

0.004 m 4 0.002 m 741 kHz
Further investigation showed that the 460 kHz actuation frequency was positioned between a resonance at 447 kHz, and it neighboring antiresonance at 475 kHz, of the acoustofluidic device including the sample of disperse liquid in the cavity. Further, the 460 kHz actuation frequence, and also the resonance of the acoustofluidic device at 447 kHz, were close to an inherent resonance at 440 kHz in the width (4 mm) dimension of the ultrasound transducer.
Example 1 thus showed trapping of particles in a trapping matrix by using an actuation frequency that did not correspond to a resonance of the cavity.

Example 2: Trapping of 0.38 µm particles in an acoustofluidic device comprising a packed bed using actuation at a resonance of the acoustofluidic device

Example 1 was repeated but with the following differences:

The particles used for the test were green fluorescent polystyrene 0.38 µm particles, i.e. nanoparticles. The concentration of test particles was 0.002% v/v = 7*10⁸ particles / mL.
The flow rate was 20 µL/min and the ultrasound transducer was actuated using a drive signal of 25.4 V peak-to-peak and 0.136 A, at 460 kHz, providing a power input of 0.43 W. The temperature in the acoustofluidic device rose to 33 °C as a result of the power input into the ultrasound transducer.

The following steps were used, as also illustrated in Fig. 10B:

1. The cavity and associated tubing was flushed with PBS buffer for 2 minutes. During the flushing, a baseline fluorescence amplitude of about 0.1 was detected.
2. At time t = 2 minutes, 50 µL of the solution containing the particles were injected into the flow of PBS buffer. This corresponded to 3.5*10⁷ particles. The particles were trapped during two minutes while a gradual increase in fluorescence, up to about 1.9, was observed.
3. At time t = 4 min, the injection of nanoparticles was finished and a further 73 µL PBS buffer was allowed to flow through the cavity and associated tubing during about 3 minutes and 40 second. During this time, untrapped particles were washed from the trapping matrix and a gradual decrease of fluorescence was observed down towards the baseline value of about 0.1.
4. At time t = 7 minutes and 40 second, the drive signal was turned off. This led to a peak in fluorescence up to about 0.85 as the trapped particles were released and washed out of the trapping matrix by the PBS buffer.

Comparing the peaks formed during the trapping and washing (area 264) with the peak formed during release (area 26) yielded a trapping efficiency of 26/(26+264) = 0.09, i.e. 9%.

Example 3: Evaluating trapping efficiency, power, and Voltage for different drive signal frequencies near a width resonance of the ultrasound transducer

Example 1 was repeated several times for different frequencies, see Fig. 11A-11C. The temperature was constant at 35°C.
As seen in Fig. 11A, maximum trapping efficiency of about 27-28% was obtained at 490 and 500 kHz. Generally, Fig. 11A shows that there is obtained a good trapping efficiency for a wide range of frequencies, which, as before, are not associated with the cavity resonances or the eigen resonances of the trapping beads. Earlier investigation had shown that the 460 kHz actuation frequency used in Example 1 and 2 was close to an inherent resonance at 440 kHz in the width (4 mm) dimension of the transducer used. Accordingly, it was noted that, while actuation at frequencies close to an inherent frequency of the ultrasound transducer could provide efficient trapping, also frequencies removed from this resonance frequency, i.e. anti-resonance frequencies such as e.g. 490 and 500 kHz, could provide even more efficient trapping. The decreased trapping efficiency at 460 kHz as compared to Example 1 was believed to be due to the repeated use of the device possibly resulting in a reduced trapping capacity.
Fig. 11B shows the variation of input power for the different frequencies tested. It was found that the input power increased from 480 kHz towards 540 kHz. It was thus noted that higher input power could be applied to the system while maintaining the same temperature at the anti-resonance frequencies, i.e. frequencies removed from the inherent resonance in the width dimension of the ultrasound transducer at 440 kHz, see the input power values for frequencies of 500 kHz and above.
Fig. 11C shows that the peak-to-peak voltage with which the transducer is supplied can generally be increased while maintaining the same temperature of the transducer as the frequency is increased from the inherent resonance in the width dimension of the ultrasound transducer at 440 kHz, reaching a maximum at 500 to 520 kHz.
The findings in Fig. 11B and 11C thus indicated the possibility that the higher input power possible at anti-resonance frequencies could be utilized to further increase the trapping efficiency without resulting in significant heating of the substrate, transducer, and liquid.

Example 4: Evaluation of trapping efficiency for a wider frequency range

Example 1 was repeated several times for a wider range of different frequencies, see Fig. 12A-12B. The temperature was kept constant at 35°C.
As seen from Fig. 12A, trapping was possible for all tested frequencies. However, lesser trapping efficiency was observed for frequencies far removed from the width resonance frequency of the transducer alone at 440 kHz, such as for example at 1000 kHz. For such far removed frequencies as 1000 kHz, the highest power of all tested frequencies could be applied, see Fig. 12B. Despite the high input power, the trapping efficiency was lower than any frequency tested for frequencies close to the width resonance of the transducer. This power may however be supplied without pronounced increased heating of the substrate, transducer, and liquid because the transducer is not actuated at or near any of its inherent resonance frequencies. The thickness of the transducer, 0.5 mm, indicated that a thickness resonance should be found around 4400 kHz. combined with the substrate, the resonance shifted towards 4700 kHz. As seen in Fig. 12B, such higher frequencies, i.e. 4700-4900 kHz allowed much less power supplied for the same temperature, and thus it appeared that using a frequency near the thickness resonance allowed for less power supplied for the same temperature, and accordingly less efficient energy conversion.
From Fig. 12A and 12B it appeared that more power could be supplied at the actuation frequencies closer to the width resonance or antiresonance of the transducer. For the acoustofluidic device tested, a first resonance was found at 460 kHz. According to the requirements for resonance, a second and third resonance should be found at 1300 and 1900 kHz, respectively. Accordingly, frequencies at, or even better between, these resonance frequencies, such as e.g. 530 and 540 Hz, provided good trapping.
It was thus concluded that the best actuation frequencies were either close to or at a resonance frequency in a width or length dimension of the ultrasound transducer, or even more advantageously, close to or at an antiresonance frequency in a width or length dimension of the ultrasound transducer.
For the higher frequencies, e.g. 4700-4900 kHz, that are between the calculated thickness resonances of 4700 and 6800 kHz, corresponding to the first and second resonances of the acoustofluidic device associated with the thickness dimension of the transducer, Fig. 12A and 12B shows a less efficient trapping and that less power can be supplied for a constant temperature. Accordingly, actuation frequencies close to or at a resonance or antiresonance in the thickness dimension of the ultrasound transducer were less efficient. However, actuation frequencies at, or between, these resonance frequencies still provide trapping.

Example 5: Evaluation of trapping efficiency at constant voltage

Following example 4, it was theorized that the higher frequencies, i.e. at or between the resonances in the thickness dimension of the transducer, could nevertheless be advantages for some applications where the voltage of the drive signal cannot be amplified and should thus be chosen as low as possible.
The results are shown in Fig. 13A and 13B.
Fig. 13A thus shows that, indeed, for a constant low input voltage of 10 V peak-to-peak, the trapping efficiency was better at the higher frequencies close to the resonances and antiresonances in the thickness dimension of the transducer, e.g. 4.65 to 4.75 MHz, than the lower frequencies close to resonance or antiresonance in the width dimension, i.e. 0,44-0.48 MHz. As seen in Fig. 13B, the current and thus also power through the ultrasound transducer was higher for these higher frequencies, and that also led to a higher temperature in the substrate, transducer and liquid, i.e. 32°C, compared to the lower temperature of 22°C obtained when using the lower frequencies associated with resonance in the width dimension for the same constant voltage.
It should however be noted that, trapping efficiency at the higher frequencies was still much lower than the trapping efficiency at the same temperature for the lower frequencies, i.e. the frequencies associated with resonance in the width dimension of the transducer.
Accordingly, it was confirmed that it was for most cases more advantageous to use an actuation frequency close to or at a resonance or antiresonance in the width or length dimension of the ultrasound transducer.

Example 6: Trapping of particles using a 3D-printed polymer substrate

Example 1 was repeated using a 3D-printed polymer substrate as schematically shown in Fig. 7.
As shown in Fig. 14, also the 3D-printed polymer substrate provided trapping of the 2 µm particles during the time 2-4 minutes, whereafter non-trapped particles were washed out during the time interval 4- 7:45 minutes, and the trapped particles were released from 7:45 to 10 minutes.

Example 7: Trapping of 270 nm particles

Example 2 was repeated for 270 nm particles.
Fig 15 shows how the particles were trapped during the time interval 0:45 to 2:45, followed by washing of non-trapped particles during the time interval 2.45 to 5:10 minutes. The trapped particles were released during the time interval 5:10 to 5:40 minutes.
In further experiments using 270 nm particles trapping efficiencies up to 63% were obtained (graph not shown).

Example 8: Trapping of 380 nm nanoparticles using a larger volume packed bed

In this example the capillary of example 2 was replaced with a larger glass tube having a cross section width of 4 mm and a cross section height of 2 mm resulting in a 5 times larger packed bed volume. A similar result as in Example 2 was obtained (not shown)

Example 9: Attempts to trap at an actuation frequency corresponding to resonance in the cavity.

Further experiments were made using a 1 mm thick ultrasound transducer actuated over a range corresponding to a thickness resonance of the ultrasound transducer which also corresponded to a cavity resonance in the height direction at 2 MHz. As seen in Fig. 16, such actuation produced an impedance spectrum without any clear features, and which did not provide for clearly identifying the resonance in the thickness dimension of the ultrasound transducer. Particle trapping at this frequency was not efficient.
Accordingly, using an actuation frequency corresponding to a resonance of the cavity dimensions is detrimental for the trapping efficiency.

Example 10: Admittance spectrum showing resonance and antiresonance

During the performance of examples 1 and 2, impedance spectra and admittance spectra were recorded. Fig. 17 shows an exemplary admittance spectrum for the frequency range of 401 kHz to 494 kHz. The admittance spectrum clearly shows both a resonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity (the peak at 447 kHz) and the neighboring anti-resonance (the minimum at 475 kHz). In contrast, a resonance in the width dimension of the ultrasound transducer was found close by, at 440 kHz.
Thus, as described herein, whereas it is possible to trap particles with a wide range of actuation frequencies, it is more preferable to use an actuation frequency that is close to, or at, a resonance frequency (e.g. 447 kHz) of the acoustofluidic device including the sample of disperse liquid in the cavity, or its neighboring anti-resonance (e.g. 475 kHz).

Feasible modifications of the technology proposed herein

The technology proposed herein is not limited to the embodiments described above and shown in the drawings, which primarily have an illustrative and exemplifying purpose. This patent application is intended to cover all adjustments and variants of the preferred embodiments described herein, thus the present invention is defined by the wording of the appended claims and the equivalents thereof. Thus, the equipment may be modified in all kinds of ways within the scope of the appended claims.
Throughout this specification and the claims which follows, unless the context requires otherwise, the word "comprise", and variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated integer or steps or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

Claims

A method of trapping particles in a sample of a disperse liquid, the disperse liquid comprising particles dispersed in a suspending liquid, the particles having a non-zero acoustic contrast factor relative to the suspending liquid, the method comprising the steps of:
i. providing an acoustofluidic device comprising:
a. a substrate in which a cavity is formed, the cavity being at least partially filled with a trapping matrix, the trapping matrix comprising or consisting of a material having a non-zero acoustic contrast factor relative to the suspending liquid, the cavity further having an inlet for introducing the sample of disperse liquid into the cavity, and

b. an ultrasound transducer in acoustic contact with the substrate,

ii. introducing the sample of the disperse liquid into the cavity so that the sample perfuses the trapping matrix, and

iii. actuating the ultrasound transducer at an actuation frequency which:
a. does not correspond to resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with sample of the disperse liquid and optionally also the trapping matrix, and,

b. couples an acoustic wave into the substrate.
The method according to claim 1, wherein the actuation frequency corresponds to:
c. a frequency in a range extending between two neighbouring resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity, or preferably:

d. a frequency in a range extending between a resonance frequency and its neighbouring anti-resonance frequency, of the acoustofluidic device including the sample of disperse liquid in the cavity, or more preferably:

e. a frequency that is within 10% of a resonance frequency or an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity, or most preferably:

f. a frequency that is within 10% of an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity.
The method according to any preceding claim, wherein maximum diameter of the particles is 10 µm or less, preferably 2 µm or less, and wherein maximum diameter of the particles is at least 30 nm, preferably at least 80 nm, or wherein the maximum diameter of the particles is 30-380 nm, preferably 80-380 nm.
The method according to any preceding claim, wherein the trapping matrix comprises or consist of one or more of: porous or nonporous beads, particles, granules or fibers, monolithic pillars, macroporous polymer monoliths, glass wool, and metal mesh.
The method according to any preceding claim, wherein the cavity further comprises an outlet, and step ii comprises flowing the sample of the disperse liquid through the cavity.
The method according to any preceding claim, further comprising the step of
iv. introducing a wash buffer into the cavity so that the wash buffer perfuses the trapping matrix.
The method according to any preceding claim, further comprising the steps of:
v. decreasing, or ceasing, actuating the ultrasound transducer so as to release particles trapped by the trapping matrix, and/or

vi. repeatedly and/or pulsingly actuating the ultrasound transducer, preferably at a resonance frequency of the transducer, so as to release particles trapped by the trapping matrix.
A system for trapping particles in a sample of a disperse liquid, the disperse liquid comprising particles dispersed in a suspending liquid, the system comprising:
a) an acoustofluidic device comprising:
a. a substrate in which a cavity is formed, the cavity being at least partially filled with a trapping matrix, the trapping matrix comprising or consisting of a material having a non-zero acoustic contrast factor relative to the suspending liquid, the cavity further having an inlet for introducing the sample of disperse liquid into the cavity, and

b. an ultrasound transducer in acoustic contact with the substrate, and

b) a drive circuit connected to the ultrasound transducer, the drive circuit being configured to actuate the ultrasound transducer at an actuation frequency which:
a. does not correspond to a resonance of the cavity filled with the sample of disperse liquid, and/or is not resonant to the cavity when filled with the fluid and optionally also the trapping matrix, and,

b. couples an acoustic wave into the substrate
The system according to claim 8, wherein
c. a frequency in a range extending between two neighbouring resonance frequencies of the acoustofluidic device including the sample of disperse liquid in the cavity, or preferably:

d. a frequency in a range extending between a resonance frequency and its neighbouring anti-resonance frequency, of the acoustofluidic device including the sample of disperse liquid in the cavity, or more preferably:

e. a frequency that is within 10% of a resonance frequency or an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity, or most preferably:

f. a frequency that is within 10% of an antiresonance frequency of the acoustofluidic device including the sample of disperse liquid in the cavity.
The system according to any of the claims 8-9, wherein the minimum diameters of the pores, if the trapping matrix is porous, and/or the minimum diameter of interstices between parts of the trapping matrix, if the trapping matrix comprises individual parts, is 100 µm or less in diameter, preferably 20 µm or less, and at least 300 nm, preferably at least 800 nm, or alternatively 300 nm to 3.8 µm, preferably 800 nm to 3.8 µm.
The system according to any of the claims 8-10, wherein the trapping matrix comprises or consist of one or more of: porous or nonporous beads, particles, granules or fibers, monolithic pillars, macroporous polymer monoliths, glass wool, and metal mesh.
The system according to any of the claim 8-11, wherein the cavity further comprises an outlet and the system further comprises:
c) a sample container for holding the sample of the disperse liquid, and

d) a sample pump arranged to pump the sample of the disperse liquid from the sample container to the inlet for pumping the sample of the disperse liquid through the cavity.
The system according to any of the claim 12, wherein the system further comprises:
e) a buffer container for holding a wash buffer, and

f) a buffer pump arranged to pump the wash buffer from the buffer container to the inlet for pumping the wash buffer through the cavity.
The system according to any of claims 8-13, wherein the system further comprises:
g) one or more particle detectors arranged to detect particles entering and/or exiting the cavity.