Method, apparatus and system for controlling sound generation

Abstract

Broadly speaking, embodiments of the present techniques provide a method, apparatus, and system to use acoustic waves to exploit interactions between the liquid-air interface of arbitrary materials placed in open containers. The methods and apparatus enable new manipulation capabilities of such materials (i.e., depression / channel creation, droplet ejection, bubble creation and multi-sample manipulation, mixing and deposition), such as contactless bio-chemical handling or manufacturing techniques. An object is manufactured within an acoustic volume using an array of transducers which generates sound. The acoustic volume comprises a scattering object in the form of a container of material. The method comprises: sending control instructions to each transducer in the array of transducers to generate sound to create an acoustic focal point at a least one control point on or under a surface of the material in the container; and adjusting operating parameters for each control instruction to generate at least one of a depression on a surface of the material, generate a bubble under the surface of the material and a droplet which is ejected from the surface of the material. The operating parameters may include amplitude pressure for a continuous signal or pulsed signal generated by each transducer and timing between pulses for a pulsed signal.

Description

[0001] Method, Apparatus and System for Controlling Sound Generation

[0002] Field

[0003] The present techniques generally relate to methods and systems for ultrasound manipulation of liquids, particularly exploiting the liquid-air interface and the control of the liquid samples once they are ejected / levitated.

[0004] Background

[0005] Acoustophoresis is a non-contact and label-free (i.e., non-reliance on magnetic or chemical tags) technique, using controlled ultrasound to allow manipulation of particles and cell populations in a high-resolution and non-invasive manner [1-4], Its main advantages for additive manufacturing are its capability to avoid cross-contamination and printing objects on top of an existing object from any direction. However, traditional methods face limitations in terms of material diversity, 3D manipulation capabilities, and structural versatility, which limit their practical application.

[0006] A number of acoustophoretic studies have utilized surface acoustic waves for patterning of droplets, particles or cells for screening and analysis applications [5, 6], However, these approaches are typically in 2D, and their acoustic radiation force fields are often inadequate for 3D patterning. Habibi, et al. [7] have used ultrasound for 3D fabrication by inducing sonochemistry with acoustic waves to solidify and print polydimethylsiloxane. On the negative side, it only allows printing with one type of material and the whole structure is porous, as porosity is induced from solidification. Recent advances have integrated acoustophoresis with 3D printing to allow fabrication of complex 3D structures [2, 8-15], While these advancements have enhanced the structure and / or function of the 3D printed products, the modality of 3D printing is unchanged and multi-material printing and customized / localized porous structure printing remain a challenge.

[0007] Acoustic levitation is an acoustophoretic technique which manipulates particles / inks in mid-air which has the potential for 3D fabrication. Ezcurdia, et al.

[0016] utilized acoustic levitation technique to assemble 3D objects including beads and sticks with UV glues in empty working space or on acoustically translucent materials. However, printing sizable objects or on any acoustic scattering substrate poses a big challenge for all acoustic levitation studies as they scatter sound within the working space leading to distortion of the sound field, potentially causing objects to be dropped. In addition, relying on UV glues makes this approach unsuitable for certain applications (e.g., food, biochemistry). Finally, the existing attempts

[0015] rely on external mechanical attachments (i.e., PAT mounted on a robotic arm), unnecessarily increasing complexity and not fully leveraging the 3D manipulation capabilities of acoustic levitation.

[0008] The sound field control capability of phased arrays of transducers (PATs) allows for suspending and manipulating matters (solid or fluid) in a medium, such as in mid-air

[0017] , This acoustophoretic levitation technique is material-agnostic and thus has successfully been employed to levitate objects like expanded polystyrene (EPS) particles, electrical components, pieces of fabric, and liquid droplets. Hirayama, et al.

[0018] proposed a high-performance holographic algorithm using the boundary element method (BEM) technique, allowing levitation even if external, sound-scattering objects (e.g., the object being manufactured) are present in the workspace. Consequently, it allows 3D printing of particles / droplets on sizable complex objects / substrates. This allows incorporation of any materials for multi-material printing using droplets or particles as a manufacturing primitive. However, their approach does not exploit interactions of ultrasound with the liquid interface (e.g., channel creation), or the way such primitives are generated (e.g., droplet ejection).

[0009] The applicant has therefore identified the need for improved techniques for acoustic levitation.

[0010] Summary

[0011] In a first approach of the present techniques, there is provided a computer-implemented method for manufacturing an object within an acoustic volume using an array of transducers which generates sound, wherein the acoustic volume comprises a scattering object in the form of a container of material. The method comprises sending control instructions to each transducer in the array of transducers to generate sound to create an acoustic focal point at at least one control point on or under a surface of the material in the container; and adjusting operating parameters for each control instruction to generate at least one of a depression on a surface of the material, generate a bubble under the surface of the material and a droplet which is ejected from the surface of the material. The container of material may be one container or the container of material may comprise more than one container, such as for example, two or three containers. It will also be appreciated that there may be other scattering objects in the acoustic volume.

[0012] Such techniques manipulate the air / liquid interface and depend on a range of material properties (e.g., density, viscosity, surface tension) which could be analytically modelled or determined in a calibration process. The operating parameters may include amplitude pressure for a continuous signal or pulsed signal generated by each transducer and timing between pulses for a pulsed signal. More specifically, the operating parameters may include one or more of a maximum amplitude pressure for creating a surface depression, a surface depression depth at the maximum amplitude pressure, a restoration time for a depression to be refilled by surface tension; a timing between pulses of a pulsed signal which causes a droplet to be ejected; an amplitude pressure for a pulsed signal which causes a droplet to be ejected; a minimum amplitude pressure which allows controlled levitation of an ejected droplet; and a maximum amplitude pressure which does not burst a droplet. Pressures below the maximum amplitude pressure (atft) will cause increasing depressions, while pressure above it will cause a local disruption to the surface, mixing air into it and creating localized air bubbles. The restoration time (tr) must be considered when creating microchannels on the surface by quickly moving focus points along the microchannel length. The timing and / or amplitude pressure which cause a droplet to be ejected may be stored in the look-up table together with a height and / or volume of the ejected droplet. Determining control instructions may thus comprise obtaining operating parameters from one or more look-up tables which have been determined in a calibration process. This is particularly helpful when generating depressions and / or bubbles and / or ejecting one or more droplets from the surface of the material.

[0013] The method may comprise obtaining a static matrix (H) representing a contribution of each transducer in the array of transducers to each of a plurality of locations on the scattering object; defining at least one control point at a surface of the material within the acoustic volume; calculating, in real-time, a direct transmission matrix (F) which represents a direct contribution to the at least one control point from each transducer in the array of transducers; calculating, in real-time, a scattering transmission matrix (G) which represents a scattering contribution to the at least one control point from the plurality of locations on the scattering object; and determining, in real-time, an extended transmission matrix (F) which represents direct and scattered contributions from each transducer in the array of transducers to each of the multiple control points, wherein the extended transmission matrix (which may also be termed a complete transmission matrix) is determined using the static matrix, the direct transmission matrix and the scattering transmission matrix. The method may further comprise determining, using the extended transmission matrix, control instructions for each transducer in the array of transducers to generate an acoustic focal point at the least one control point, wherein the acoustic focal point is configured to generate a depression on the surface of the material, eject a droplet from the surface of the material and generate a bubble under the surface of the material.

[0014] As well as generating acoustic focal points to generate depressions, bubbles and ejected droplets, the method may comprise simultaneously or sequentially generating acoustic traps. There may be a plurality of acoustic focal points and / or acoustic traps which are simultaneously generated. The number of acoustic focal points and / or acoustic traps may vary depending on the application, and as an example may range between 1 and 16 depending on the application. The method may further comprise determining, using the extended transmission matrix, control instructions for each transducer in the array of transducers to generate an acoustic trap at the least one control point wherein the acoustic trap is configured to trap a target object to control the location of the target object within the acoustic volume. The target object may be a droplet which is ejected from the surface of the material. Alternatively, the target object may be a droplet which is injected into the acoustic volume using an injector. The target object may be deposited in a depression or trench formed in the material.

[0015] There is a two-step modelling of the extended transmission matrix, with a static matrix obtained before the direct and scattering transmission matrices are calculated in real-time. Once the extended transmission matrix is obtained, the next step may be to solve for the transducers’ activation (T) that generates acoustic trap points or acoustic focal points at target positions (i.e. control points). An acoustic focal point may be defined as a point of maximum acoustic pressure and an acoustic trap point may be defined as a region of low acoustic pressured surrounded by high acoustic pressure areas whereby a target object may be trapped in the region of low acoustic pressure. The transducers can then be controlled with the determined control instructions. Controlling the transducers when generating acoustic traps controls the location of the target object using the acoustic pressure created by the array of transducers. Controlling the transducers when generating acoustic focal points means controlling the generation of depression and / or bubbles and / or controlling ejection of a droplet from the surface of the material.

[0016] The extended transmission matrix E may be expressed as E = F + GH where F is the direct transmission matrix, G is the scattering transmission matrix and H is the static matrix. There may be L control points within the acoustic volume, N transducers and M points on the scattering object. Accordingly, the sizes of these matrices are L x N for E and F, L x M for G, and M x N for H. Given the fact that the inequality L « N « M is usually satisfied in acoustic levitation, the determination of H is more time-consuming than the other matrices. The matrices F and G depend on control point positions while the matrix H, the largest and most computationally expensive element in the claimed model, does not. Therefore, once the geometry of the set-up is determined, H remains constant and does not have to be computed every time the trapping positions are updated (i.e. the set-up-related part). On the other hand, F and G must be computed every time for interactive applications (i.e., the application-related part), but the computations of these are highly suitable for computing in parallel. Therefore, once the matrix H is pre-computed, the extended transmission matrix can be computed at a very high rate. This two-part modelling means that it is possible to calculate the extended transmission in real-time, i.e. as the target object is being controlled.

[0017] The static matrix may be obtained by calculating the static matrix in a set-up phase. For example, the method may comprise defining a plurality of locations on the scattering object (i.e. the container of material) within the acoustic volume; obtaining location information for each of the plurality of locations; obtaining position information for each transducer in the array of transducers; calculating, for each of the plurality of locations, a set of acoustic pressure contributions from each transducer in the array of transducers and storing each set of acoustic pressure contributions in the static matrix. The position information for each transducer may comprise position and normal of each transducer.

[0018] The plurality of locations may be a plurality of mesh elements. In other words, a surface of the scattering object may be covered in a plurality of mesh elements. In this example, the location information may comprise one or more of position, area and normal of each mesh. Each mesh element has a maximum length which may be less than A, A / 2, A / 4, or A / 8 where A is the wavelength of the sound being generated by each transducer. Each mesh element has a maximum length of A / 2 because this is the best-balanced mesh size between speed and accuracy.

[0019] The scattering object may change in location and / or shape over time, for example as the material in the container is ejected as droplets and / or by forming depressions and / or bubbles. Additional layers of material may also be added to the container which will change the shape of the scattering object. Extruders may be inserted and / or removed from the acoustic volume to extrude material and these extruders when present form a scattering object. Although the static matrix is fixed, multiple static matrices may be calculated, one for each time step. This calculation may be done in advance. The method may comprise using these multiple static matrices to determine the extended transmission matrix for a plurality of time steps. The extended transmission matrix may still be determined in real-time because the computational load of the static matrices is performed in the set-up phase, i.e. off-line.

[0020] Determining control instructions may comprise optimising phases (<p = [<p1, ..., <pw]T) of each transducer in the array of transducers to maximise trapping stiffness at each location of an acoustic trap and / or maximise pressure amplitude at each location of an acoustic focal point. The maximisation may be done using any suitable cost function, for example a cost function which combines a trapping stiffness metric which is based on the trapping stiffness (V2[ / ;) at each location of an acoustic trap and a pressure metric which is based on the pressure amplitude at each location of an acoustic focal point. The Laplacian of the Gor’kov potential at the point j (V2[ / ;) may be used as the trapping stiffness metric Uj'. A weighted negative squared amplitude pressure may be used as the pressure metric Uf. The trapping stiffness metric [ / / may be defined as: where V represents the volume of the target object; a> represents the angular frequency of the target object; c and p represent the speed of sound and density, and the subscripts 0 and p refer to the host medium (i.e., air) and the particle material, respectively, pj represents that acoustic pressure at the control point from the / th transducer and z is the principal axis. The pressure metric may be defined as

[0021] When calculating control instructions for both acoustic focal points and acoustic traps, a joint cost function may be used and an example is: where wsis the weight coefficient, Ux' here represents the mean value of [ / / and Uf, as appropriate (i.e., x = j or x = f), J is the number of traps, F is the number of focal points, [ / / and Ufare the metrics defined above, <p is the phase of each transducer in the array and U here represents the mean value of

[0022] Calculating the cost function may comprise sampling acoustic pressure for only two control points per acoustic trap and / or two control points per acoustic focal point. The two control points may be along a principal axis of the transducer array. The method may thus comprise defining each position of an acoustic trap; determining a principal axis of the array of transducers; sampling acoustic pressure values at two locations along the principal axis around each position of an acoustic trap; and maximising the trapping stiffness metric using a cost function. Similarly, the method may comprise defining each position of an acoustic focal point; determining a principal axis of the array of transducers; sampling acoustic pressure values at two locations along the principal axis around each position of an acoustic focal point; and maximising the pressure metric using a cost function. Such methods represent a simplified solver and it will be appreciated that the simplified solver may be used independently from or together with the two-part model of the transmission matrix described above.

[0023] Determining control instructions may comprise adjusting an amplitude pressure input for each transducer in the transducer array. Determining control instructions may comprise determining a target amplitude pressure at each control point (e.g. for an acoustic trap and / or acoustic focal point), simulating an acoustic pressure at each control point; comparing the simulated acoustic pressure with the target acoustic pressure; determining, using the comparison, whether the target acoustic pressure is feasible and when it is determined that the target acoustic pressure is feasible, adjusting the control instruction for an amplitude input for each transducer to generate the target amplitude pressure. Determining that the target acoustic pressure is feasible may comprise determining that the target acoustic pressure is feasible may comprise determining whether the target amplitude pressure is below the simulated amplitude pressure (atar< asim).

[0024] The method may comprise generating a depression on a surface of the material in the container. The method may comprise determining control instructions which determine an amplitude pressure for each transducer which generates a depression of a particular depth at each acoustic focal point. Each transducer may be controlled to apply the determined amplitude to create at least one trench by generating multiple depressions; and simultaneously curing the material to form a manufactured object having the at least one trench. As an alternative to creating many depressions at once, the method may comprise continuously controlling each transducer to apply the determined amplitude; controlling each transducer to move each acoustic focal point to create at least one trench; and curing the material when the trenches have been created. Once a layer with the desired pattern of trenches has been formed, more material may be inserted in the container and the process repeated.

[0025] Each formed trench may be filled with material which is extruded into the acoustic volume. The method may thus comprise generating control instructions for each transducer in the array of transducers to generate an acoustic trap to trap a target object (e.g. material to be inserted in the trench) to control the location of the target object within the acoustic volume whereby the target object is movable to be deposited in the at least one trench. The volume, location, and material properties of this target object (liquid) can be varied, and the liquid can be deposited in a contact-free way using acoustic levitation.

[0026] As an alternative (or in addition to) creating trenches using depressions, we can create a porous structure using bubble creation. The method may thus comprise determining control instructions by determining an amplitude pressure for each transducer which generates at least one bubble at each acoustic focal point. Each transducer may be controlled to apply the determined amplitude to create bubbles at multiple separate locations; and curing the material to form a manufactured object having a porous layer. As an alternative to creating many bubbles at once, the method may comprise continuously controlling each transducer to apply the determined amplitude to create at least one bubble; controlling each transducer to move each acoustic focal point to create a pattern of bubbles; and curing the material when the pattern of bubbles have been created to create a porous layer and then cure the material to form the manufactured object. Once a layer with the desired pattern of bubbles has been formed, more material may be inserted in the container and the process repeated. In this way, we can create dynamic porous structures by focusing sound continuously on the liquid surface. As explained above, the structure can be patterned in two ways, either by using a single focal point that is moved in the desired shape, or by creating a holographic pattern directly on the surface, or using a combination of both techniques. By varying the intensity of the focal pressure, while still keeping it continuously on, we can create bubbles inside the liquid while taking into account the bath height and liquid properties. Curing may be using UV curing to cure the liquid as we generate the bubbles, sealing them in. We can dynamically control the location, geometry, and pore density of our porous structures by controlling the applied sound field.

[0027] As an alternative (or in addition to) creating trenches using depressions and / or bubbles, we can use an acoustic focal point to eject a droplet from the material in the container. Determining control instructions may thus comprise determining a timing between pulses and an amplitude pressure for each pulse applied by each transducer which ejects at least one droplet from a surface of the material. In other words, it is possible to eject liquid from a surface by using a transducer array placed above it and pointed at the liquid. This method offers greater flexibility in design as the array can be fixed at a distance away from the liquid surface. The same array can be used for different purposes, such as ejecting liquid for contact-free fabrication, handling the ejected liquid by moving and depositing it, and creating channels in the same or different liquid.

[0028] The timing and the amplitude pressure may be determined to provide a droplet of a predetermined volume and at a predetermined height of ejection. To eject a droplet from the surface and cause it to jump to different heights, the acoustic focal point may thus be created on the liquid-air surface. By varying the pressure intensity, pulse speed, and duty cycle, we can control the droplet's trajectory and height.

[0029] Once the droplet has been ejected, an acoustic trap may be generated to trap and then move the droplet within the acoustic volume. The method may thus comprise generating control instructions for each transducer in the array of transducers to generate an acoustic trap to trap the ejected droplet and to control the location of the ejected droplet within the acoustic volume whereby the ejected droplet is movable to be deposited at a desired location. For example, we can create a levitation trap at the highest point of the droplet's trajectory to manipulate it for printing, fabrication, or pipetting. This method works with multiple locations, with each location being in the same or different liquids. The liquids can be in any container and filled to any depth.

[0030] The creation of depressions, trenches, bubbles and ejected droplets uses acoustic focal points as described above. The acoustic traps may be used without the acoustic focal points to manufacture an object using printing, e.g. additive manufacturing.

[0031] Thus according to another aspect of the present techniques, there is provided a computer-implemented method for manufacturing an object within an acoustic volume using an array of transducers which generates sound, wherein the acoustic volume comprises a scattering object. The method comprises obtaining a static matrix (H) representing a contribution of each transducer in the array of transducers to each of a plurality of locations on the scattering object; defining at least one control point at a surface of the material within the acoustic volume; calculating, in real-time, a direct transmission matrix (F) which represents a direct contribution to the at least one control point from each transducer in the array of transducers; calculating, in real-time, a scattering transmission matrix (G) which represents a scattering contribution to the at least one control point from the plurality of locations on the scattering object; and determining, in real-time, an extended transmission matrix (F) which represents direct and scattered contributions from each transducer in the array of transducers to each of the multiple control points, wherein the extended transmission matrix is determined using the static matrix, the direct transmission matrix and the scattering transmission matrix from E = F + GH. The method then comprises determining, using the extended transmission matrix, control instructions for each transducer in the array of transducers to generate at least two acoustic traps at the least one control point wherein each acoustic trap is configured to trap a droplet to be printed and to control the location of each droplet within the acoustic volume whereby the at least two droplets are moveable to be deposited at a desired location.

[0032] The features of the first aspect apply equally to the second aspect and are not repeated. Such a printing technique is different from the conventional way of printing multiple materials using multiple nozzles attached to a robotic arm. We do not require a robotic arm for multidirectional printing, which is limited by the attached mechanical structure of the arm. Instead, we can manipulate multiple droplets in mid-air, with no attached arms, independently. Our experiments have shown that we can manipulate up to 16 droplets, but it is possible to manipulate even more. Each droplet may be formed from a different material. By generating at least two traps, we can achieve drop-by-drop printing and manipulate multiple drops of different material properties, e.g. solids and / or liquids. This manipulation allows us to deposit drops on any surface, whether it is vertical, horizontal, inclined, or curved, without any contact. For example, this technique enables us to print hydrogels directly onto wounds for in-situ bioprinting.

[0033] The method may further comprise determining control instructions for each transducer in the array of transducers to control the location of each droplet within the acoustic volume whereby two droplets are mixed before being deposited. This unique method for mixing materials in mid-air, which eliminates contamination and enables us to mix materials just in time. Before combining two or more droplets, the method may further comprise processing each droplet differently, e.g. by heating to a specific temperature or curing, e.g. using UV light. This technique also allows us to create Janus particles. One droplet may be a crosslinker and one droplet may be a binding agent. This allows flexibility in structure fabrication.

[0034] In a related approach, there may be provided an apparatus comprising: an array of transducers for generating acoustic pressure; an acoustic volume which is defined by the acoustic pressure generated by the array of transducers; and a processor for carrying out the methods described above to manufacture an object. The apparatus may further comprise a container of material which is used at least in part to manufacture the object. The apparatus may further comprise an injector for injecting printing material which is used at least in part to manufacture the object.

[0035] In a related approach of the present techniques, there is provided a non-transitory data carrier carrying processor control code to implement any of the methods, processes and techniques described herein.

[0036] As will be appreciated by one skilled in the art, the present techniques may be embodied as a system, method or computer program product. Accordingly, present techniques may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[0037] Furthermore, the present techniques may take the form of a computer program product embodied in a computer readable medium having computer readable program code embodied thereon. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.

[0038] Computer program code for carrying out operations of the present techniques may be written in any combination of one or more programming languages, including object oriented programming languages and conventional procedural programming languages. Code components may be embodied as procedures, methods or the like, and may comprise subcomponents which may take the form of instructions or sequences of instructions at any of the levels of abstraction, from the direct machine instructions of a native instruction set to high- level compiled or interpreted language constructs.

[0039] Embodiments of the present techniques also provide a non-transitory data carrier carrying code which, when implemented on a processor, causes the processor to carry out any of the methods described herein.

[0040] The techniques further provide processor control code to implement the abovedescribed methods, for example on a general purpose computer system or on a digital signal processor (DSP). The techniques also provide a carrier carrying processor control code to, when running, implement any of the above methods, in particular on a non-transitory data carrier. The code may be provided on a carrier such as a disk, a microprocessor, CD- or DVD- ROM, programmed memory such as non-volatile memory (e.g. Flash) or read-only memory (firmware), or on a data carrier such as an optical or electrical signal carrier. Code (and / or data) to implement embodiments of the techniques described herein may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as C, or assembly code, code for setting up or controlling an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or code for a hardware description language such as Verilog (RTM) or VHDL (Very high speed integrated circuit Hardware Description Language). As the skilled person will appreciate, such code and / or data may be distributed between a plurality of coupled components in communication with one another. The techniques may comprise a controller which includes a microprocessor, working memory and program memory coupled to one or more of the components of the system.

[0041] It will also be clear to one of skill in the art that all or part of a logical method according to embodiments of the present techniques may suitably be embodied in a logic apparatus comprising logic elements to perform the steps of the above-described methods, and that such logic elements may comprise components such as logic gates in, for example a programmable logic array or application-specific integrated circuit. Such a logic arrangement may further be embodied in enabling elements for temporarily or permanently establishing logic structures in such an array or circuit using, for example, a virtual hardware descriptor language, which may be stored and transmitted using fixed or transmittable carrier media.

[0042] In an embodiment, the present techniques may be implemented using multiple processors or control circuits. The present techniques may be adapted to run on, or integrated into, the operating system of an apparatus.

[0043] In an embodiment, the present techniques may be realised in the form of a data carrier having functional data thereon, said functional data comprising functional computer data structures to, when loaded into a computer system or network and operated upon thereby, enable said computer system to perform all the steps of the above-described method.

[0044] Brief description of the drawings

[0045] Implementations of the present techniques will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0046] Figure 1 is a schematic block diagram of a system for generating and controlling sound according to the described techniques;

[0047] Figures 2a and 2b are example set-ups which could be used in the apparatus of Figure 1a;

[0048] Figure 3 shows a flowchart illustrating the steps for controlling sound generation in the systems of Figures 1a to 2b;

[0049] Figure 4 is a visual representation of the two-step scattering model used in Figure 3; Figure 5 shows sampling points which can be used in the method of Figure 2;

[0050] Figure 6 is a flowchart of the detail of the solving process using the sampling points of Figure 5;

[0051] Figure 7 is a flowchart showing the steps for adjusting amplitude of transducers in the arrays shown in Figures 1 to 2b;

[0052] Figures 8a and 8b show two alternative arrangements for printing using focal points and levitation traps as described in Figure 3;

[0053] Figures 9a to 9e show the steps in various calibration techniques which can be used to determine operational parameters required for using in the process of Figures 3, 6 and 7;

[0054] Figure 10a is a flowchart of a method for fabricating fluidic channels using sound waves combined with UV curing;

[0055] Figure 10b shows an example of a UV-curable resin in which an acoustophoretic focal point has been generated;

[0056] Figure 10c shows an example of a channel created in the UV-curable resin of Figure 10b using the method of Figure 10a;

[0057] Figure 11a illustrates a schematic set-up for droplet ejection from a liquid reservoir;

[0058] Figure 11b is a flowchart of the steps in this printing process of Figure 11a;

[0059] Figure 12a is a flow chart of the steps used in a targeted patterned cavitation bubblebased approach for the fabrication of porous materials;

[0060] Figure 12b is an example output from the method of Figure 12a;

[0061] Figure 13a is a flowchart of the process for controlling multi-material, multi-directional printing;

[0062] Figure 13b shows an example set-up for the process of Figure 13a;

[0063] Figure 13c is an example output from the method of Figure 13a;

[0064] Figure 14a shows two separate droplets being levitated using the process of Figure 14d;

[0065] Figure 14b shows the separate droplets of Figure 14a merged into a single droplet;

[0066] Figure 14c shows the merged droplet of Figure 14b on a surface; and

[0067] Figure 14d shows a method for mixing and printing merged droplets.

[0068] Figures 15a to 15e show the results of multi-material food fabrication printed using the process of Figure 12a;

[0069] Figure 16a shows an example set-up for a bio-printing process;

[0070] Figure 16b shows an acoustic trap in a sound field generated in the set-up of Figure 16a;

[0071] Figure 16c shows a partially printed object on its printing surface;

[0072] Figure 16d shows a fully printed object printed using the set-up in Figure 16a; Figure 17a shows a stereolithography (SLA) printing system which is similar to the one shown in Figure 8b;

[0073] Figures 17b and 17c illustrate a product generated using the system of Figure 17a; and Figure 17d shows the steps of a multi-modal process using the system of Figure 17a.

[0074] Detailed description of the drawings

[0075] Broadly speaking, embodiments of the present techniques provide a method, apparatus, and system to use acoustic waves to exploit interactions between the liquid-air interface of arbitrary materials placed in open containers. Our methods and apparatus enable new manipulation capabilities of such materials (i.e., depression / channel creation, droplet ejection, bubble creation and multi-sample manipulation, mixing and deposition), such as contactless bio-chemical handling or manufacturing techniques.

[0076] In the following, we first explain the basic acoustic setup used to support such manipulations. Second, we describe a two-step scattering model and solver used to allow controlled creation of focal points and levitation traps of variable amplitudes in the presence of the liquid, containers and external objects. Third, we describe calibration and control techniques for contact-less manipulation of samples, exploiting the properties of the air-liquid interface by leveraging two key techniques:

[0077] 1. Combined generation of two basic primitives: focus points (i.e., points of maximum acoustic pressure) and levitation traps (i.e., regions of low acoustic pressure, surrounded by high acoustic pressure areas with the ability to trap materials inside them).

[0078] 2. Simultaneous temporal and intensity modulation of the primitives (points and / or traps) generated.

[0079] Finally, we detail several example applications and embodiments.

[0080] Basic acoustic setup:

[0081] Figure 1 shows a system 100 for generating and controlling sound to provide acoustic levitation traps and / or focal points. The system 100 comprises a control apparatus 110 which may be any suitable computing device, e.g. a personal computer or computing device, a laptop, or a server or combination thereof. Figure 1 shows some of the components of the control apparatus and it will be appreciated that they may additionally be other standard components which are not shown. The control apparatus 100 comprises at least one processor 112 coupled to memory 114. The at least one processor 112 may comprise one or more of: a microprocessor, a microcontroller, and an integrated circuit. The memory 114 may comprise volatile memory, such as random access memory (RAM), for use as temporary memory, and / or non-volatile memory such as Flash, read only memory (ROM), or electrically erasable programmable ROM (EEPROM), for storing data, programs, or instructions, for example.

[0082] The control apparatus 100 typically comprises at least an input / output interface 116 for a user to input instructions and / or receive information. The at least one input / output interface 116 may take any appropriate form, e.g. a keyboard, a mouse, a touchpad or other input device for inputting instructions from the user and / or a display or other output device for providing the results and / or data generated during the method described below.

[0083] The processor 112 is also coupled to an array of transducers 122 to control sound generation from the array of transducers 122. The array of transducers 122 is located in an acoustic chamber 120 (which may also be termed an acoustic volume). Within the acoustic chamber 120, there is also at least one target object 124 whose movement within the acoustic chamber 120 is controlled by the generation of sound. A scattering object 126 is also located in the acoustic chamber 120 and the scattering object 126 affects the sound generation within the acoustic chamber 120. The scattering object 126 may be a container and / or liquid within the container or an external object such as the one shown. Material ejection mechanisms may also be included in the system but are omitted here for clarity.

[0084] As described in more detail below, computation within the control apparatus may be split into two stages. In a first stage a two-step scattering model 118 is used and part of this model known as matrix H may be generated in advance and may be stored in memory 114 as illustrated. The next stage is applied using a simplified solver 119. Combining both approaches it is possible to achieve over 10,000 updates per second.

[0085] Figures 2a and 2b illustrate different set-ups for the acoustic chamber. In Figure 2a, the acoustic chamber 220 is defined between upper and lower planar surfaces which are generally parallel to each other. An array of transducers 222 is mounted to the upper surface so that the array of transducers 222 generate sound towards the lower surface of the acoustic chamber as illustrated by the direct sound waves. In this example, the array is 16 x 16. A scattering object 226 (which may also be termed a physical object) is located on the lower planar surface and the direct sound is scattered from the scattering object 226 as illustrated by the scattered sound waves. In this example, the target object comprises four particles. The four particles are levitated and their movement within the acoustic chamber is controlled by the sound generated by the transducer array.

[0086] Figure 2b illustrates an arrangement in which the acoustic chamber 320 is also defined between upper and lower planar surfaces which are generally parallel to each other. In this example, there is an array of transducers 322 mounted on both the upper and lower surfaces so that the two arrays direct sound towards each other and into the acoustic chamber 320. In this example, the scattering object 326 is a sphere which is suspended within the acoustic chamber 320. As illustrated by the pressure distribution in chamber 320, the process described below can create levitation traps 328 in the presence of the sound-scattering physical object. Pmaxrepresents the maximum amplitude of the pressure in the sound field, with acoustic traps occurring at regions of low pressure between high pressure lobes.

[0087] Two-step scattering model and solver:

[0088] Figures 3, 6 and 7 are flowcharts of the steps carried out by the system to realise multipoint acoustic manipulation (i.e., focal points or levitation traps) with minimum loss of accuracy, even within a non-empty working volume (i.e., sound scattering objects, such as the liquid container or other physical objects are present). The method exploits a two-step scattering model shown in Figure 3 and a simplified acoustic modeller shown in Figure 6, to manipulate both solid and liquid materials. This two-step solver and model extends the capabilities in PCT / GB2023 / 051367 to the present applicant, by enabling creation of both levitation traps and focal points of variable intensity (i.e., trapping potential or acoustic pressure), using the same underlying technology (Phased Arrays of Transducers - PATs).

[0089] Linear model for traps and focal points in the presence of external objects:

[0090] Our solution relies on a linear model, represented as a transmission matrix F. The matrix F describes how complex activations of N transducers (T e Cw) contribute to the complex acoustic pressures at L points of interest in a sound field « e CL), using a linear system: < = FT, with L « N. Each element of this matrix (F( n) equals the pressure at the Z-th point of interest generated by the n-th transducer, when its activation is 1 (i.e., the maximum amplitude with a phase delay of 0 rad), and it can be approximated as a piston model if we consider only direct contributions (i.e., no sound-scattering objects present).

[0091] Boundary Element Methods can model situations where a sound-scattering object is present, by discretising the surface of such scattering objects into M mesh elements, with each element being small enough that acoustic pressure across a mesh can be considered constant. Acoustic pressure at some point of interest x can be represented as a boundary integral equation (i.e., Helmholtz-Kirchhoff integral equation) obtained via Green’s theorem. However, this approach is very demanding in terms of memory and computational power, and is usually considered incompatible with real-time applications.

[0092] As detailed below, we incorporate the capabilities and accuracy of BEM (i.e., scattering objects) while reducing the problem to a linear system, compatible with real-time computation. In a first step S200, a transmission matrix E that captures both the direct and scattering contributions of the transducers to target points is defined. The matrix E is defined by three matrices as E = F + GH. The first matrix F represents the contribution from the transducers to the points of interest (i.e., F is the conventional transmission matrix capturing only transducer contribution and may be termed a direct contribution matrix). The second matrix G represents the contribution from the scattering object to the points of interest (i.e. G may be termed a scattering contribution matrix), while matrix H represents the contribution from the transducers to the scattering object (i.e. H may be termed a static contribution matrix). As shown in Figure

[0093] 4, the full transmission matrix E (which may also be termed the complete or extended transmission matrix) can be represented as: where pl nis the pressure that the n-th transducer generates at Z-th point; p™ is the direct incident contribution from a transducer to a point, pmnis pressure that the n-th transducer generates at m-th element of the object mesh, smis the surface area of that element, xmand xtrepresent the positions of mesh elements (m) and points (Z), and n(xm) represents the normal of each mesh element.

[0094] Direct contributions p from a transducer n to other elements (point I or mesh element m) can be computed as: where <Pl ndenotes the complex phase propagation approximated as a spherical sound source, P( ndenotes the scalar directivity of the transducers approximated as a piston model (please note other directivity functions could be used, depending on the transducers used): where, Prefrepresents the transducer’s reference pressure at 1 m distance; r represents the transducer’s radius; 0(x;,xn) is the angle between the transducer’s normal and point Z; and Ji represents a Bessel function of the first kind.

[0095] The third matrix (H) in the system E = F + GH can be pre-computed offline at this point (in other words, the static contribution matrix can be built at step S206). The process comprises defining a plurality of locations on the scattering object, for example by obtaining the mesh information (e.g. position , xm, area smand normal n(xm) of each mesh element) on a reflecting surface of the scattering object S202. The process also comprises obtaining the location information (e.g. position xnand normal n(xn)) of each transducer S204. These steps are shown in parallel but it will be appreciated that they can be done sequentially in either order. The values of Matrix H (i.e. , pmn) can be computed by solving the following linear system ApW = b<-n\ where each element in vector b^ reuses the phase propagation and directivity functions described in Eq (c):

[0096] The matrix A may be termed an intermediate matrix and may have size M x M. The vector b^ may be termed an incident vector for the n-th transducer, where each element of the vector defines the incident pressure at that point as specified above. The values of Matrix H may be built separately for each transducer. Multiple static matrices may be calculated for different stages of the manufacturing process because the scattering object will change as the manufacturing (printing) process is carried out.

[0097] At step S208, there is an optional calibration process which is described in more detail below. This may be carried out before building the static matrix or at the same time.

[0098] When each static contribution matrix is complete, it can be used in real-time to obtain the full transmission matrix. By using these three matrices (F, G and H) each element of the matrix (El n) describes the pressure (pl n) with a transducer’s complex activation rn= 1, resulting in the same expression typically used in BEM formulations for acoustically rigid surfaces:

[0099] The first step is to obtain the locations of the points of interest at step S210. From the expressions above, the direct incident contribution from a transducer to a point of interest (Piji )> namely the elements of Matrix F, can be computed reusing the functions in Eq (c) as:

[0100] Scattering contributions from each mesh element to each point of interest, namely the elements of Matrix G, are computed from Green’s function (and its derivative) as follows: where tp(X{,xm) represents the angle between the target point x(and the normal of each mesh element n(xm). In other words, at step S212, the direct contribution matrix F and the scattering contribution matrix G are determined in real-time. Then at step S214, the full transmission matrix E = F + GH can be calculated, by combining these with the pre-computed matrix H. for the Transducers’ Phases for Acoustic 3D

[0101] Once it is known how to model the extended transmission matrix (E = F + GH), the next step is to solve for the transducers’ activation (T) that generates levitation traps or focal points at target positions in the presence of sound scattering objects and to determine how to control the transducers to achieve the desired traps or focal points as shown at step S216. The transducers can then be controlled with the determined control instructions, and the air pressure distributions they generate used for contactless manipulation. Phase-only optimisation is assumed at this step (i.e. the amplitudes of the transducers are always maximum), and amplitude control implemented at a later step. Thus, the goal of the present optimisation step is to find the optimum phases of the transducers (<p = that maximise trapping stiffnesses (V2t / ;) at every trap position, as well as acoustic pressure (pf) at every focal point.

[0102] We do this by using two metrics ([ / / and U / ) approximating stiffness and acoustic pressure, but other simplified metrics could be used instead. We first describe the general method to optimize for traps using the simplified metric [ / / , as a direct replication of PCT / GB2023 / 051367, and then describe the inclusion of the new metric allowing for simultaneous creation of focal points (using U / ).

[0103] The metric ([ / / ) is based on the Gor’kov potential at the point j (Uj) and may be termed a trapping stiffness metric. While Uj can be determined by the complex acoustic pressure (p7) at the trap position or focal point, its spatial derivatives, and constants Krand K2'.

[0104] The simplified metric ([ / / ) is instead computed as:

[0105] Where V represents the volume of the levitated particle; represents the angular frequency; c and p represent the speed of sound and density, and the subscripts 0 and p refer to the host medium (i.e., air) and the particle material, respectively. [ / / can be computed by sampling pressure values at only two points around each trap or focal point (i.e., L = 2 / , see Figure 5), located at the target position (xpypZj) and the position slightly above it (x7,y7,z7+ h) (e.g. in the present application, This simplified metric is suitable for experimental set-ups such as those shown in Figures 2a and 2b, in which the transducers face downward (i.e., —z direction) and sound-scattering objects are placed underneath or arrangements having transducers facing both downwards and upwards (i.e., -z direction and z direction) and sound-scattering objects are between the transducer arrays, but it can also be used for other arrangements to maintain a negative correlation between V2[ / 7and [ / / .

[0106] Computation of J levitation traps (as per PCT / GB2023 / 051367) can be done simply by minimizing the following expression:

[0107] Here represents the mean value of the simplified Gor’kov potential among the J levitation traps.

[0108] We extend the cost function (0(<p)) to allow simultaneous creation of J trapping points and F focal points as follows:

[0109] Here, U'frepresents a weighted negative squared amplitude pressure at each focal point, represented as and may be termed a pressure metric. here represents the mean value of , as appropriate While minimizing [ / / allows the creation of low-pressure points (e.g., a node of the standing wave) to create levitation traps, minimizing allows the creation of high-pressure points to create focal points.

[0110] The weight coefficient wswas fixed to 0.0001. The gradient of this cost function (VO(<p)) can be computed as:

[0111] Again, computing this gradient requires sampling pressure values at only two points per trap and / or focal point, allowing high-speed computation.

[0112] We minimize this extended cost function using a gradient descent method (suitable for parallel computation), with step size set to - iterations. It must be noted that other number of iterations or optimization algorithms (e.g., BFGS) could be used instead. Please note further processing steps could be applied to further increase the quality of the fields created, such as in minimizing the phase changes applied to each transducer in each PAT update. It will be appreciated that as an alternative to extending the cost function in equation (3) to simultaneously solve for focal points and levitation traps, it is also possible to minimize the cost function in equation (3) if only levitation traps are required. Similarly, it is also possible to minimize the following cost function in equation (3) if only focal points are required:

[0113] Figure 6 summarises the steps in implementing the simplified solver. In a first step S600, the desired locations of the levitation traps and / or focal points are determined. In a next step S602, the principal axis of the array of transducers is determined. In the example above, the principal axis is the z-axis but it will be appreciated that this is not always the case. It will also be appreciated that these steps can be carried out in any order or simultaneously.

[0114] Pressure values at two points around each trap are then sampled at step S604 and at step S606, pressure values at the focal point are sampled. These sampled values are used to calculate at steps S608 and S610 the simplified metrics (trapping stiffness metric and / or pressure metric). The calculated metrics are then jointly minimised at step S612 to minimise the extended cost function, e.g. as shown in equation (4), or individually using the simpler cost functions shown in equations (3) and (6) respectively to obtain the control instructions for the transducers. Any suitable optimisation algorithm can be used to minimise the cost function, e.g. as shown in equation (5), and it will be appreciated that the minimisation may iterate through steps S604 to S612. By minimising the trapping stiffness metric, we effectively maximise the trapping stiffnesses every trap position. Similarly, by minimizing the pressure metric, we maximise the acoustic pressure (pf) at every focal point.

[0115] Amplitude control

[0116] Figure 7 shows one method for achieving amplitude control by adjusting the amplitude input of the transducers. In a first step S700, the points of interest, e.g. the location(s) of the levitation traps and / or focal points are determined. At step S702, we determine a target amplitude pressure atarat each location. We simulate, at step S704, the acoustic pressure amplitude, asim, of the trap or focal point using the boundary element method, assuming the amplitude of the transducers is at their maximum capacity (i.e. , 100%).

[0117] Determining the asimfor a focal point is straightforward, as it is the amplitude pressure at the position where the focal point is supposed to be created. For a levitation trap, it is required to simulate a pressure amplitude along the z-axis in the vicinity of the trap. The simulated amplitude asimis determined by calculating the mean of the first peak amplitude values identified in the positive and negative z-directions. Henceforth, these two different definitions of pressure amplitude for levitation traps and focal points will be used. Then, at step S706, we compare the target amplitude pressure to the simulated acoustic pressure, for example by determining the amplitude input as: where ataris the target amplitude pressure at the point.

[0118] At step S708, there is a determination as to whether the target amplitude pressure is feasible by determining whether the target amplitude pressure is below the simulated amplitude pressure (atar< asim). As shown at step S710, when the target amplitude pressure is feasible, the amplitude pressure can be adjusted to the target amplitude pressure. However, when atar> asim, the maximum amplitude input is used and the final amplitude pressure we can get will be asim. In other words, as shown at step S712, there is no adjustment to the amplitude in this case.

[0119] Manufacturing Techniques using the method described above

[0120] The main component required to exploit the control techniques and applications described below (termed AcoustoFab systems) is an acoustic setup such as the one described in Figures 1 to 2b. Transducers in the PAT array operate at a consistent frequency of 40 kHz, are powered at 20 Vpp and allow independent control of the phase and amplitude of each transducer. Please note that other PAT devices, arrangements and voltages could also be used.

[0121] The system comprises other hardware components, such as an adequately prepared container for liquids (e.g., a petri dish), extruders to introduce materials into the system, and illumination devices (e.g., UV projectors, imaging devices). All these external objects (or the parts of them within the working volume of the PAT) comprise the geometry of the physical object, and determine the geometry of the mesh G used by our two-step model and solver.

[0122] Two such example setups comprising other hardware components are shown in Figures 8a and 8b. Figure 8a shows a PAT 822 directly above a conventional printing bed which comprises a container 824 supported on a substrate 826. There are two extruders 828 in this arrangement. Figure 8b shows a PAT 822 directly above a UV resin tank 830 in which there is a leveller 832. There is one extruder 828 in this arrangement as well as an ultraviolet (UV) projector 834. This strategic placement of the UV projector ensures that the ultrasound field permeates throughout the printing volume to effectively enhance and control the additive manufacturing processes. These two setups are simple examples, and the system can be integrated into other additive manufacturing solutions with other components (e.g., stereolithography, direct ink writing) for applications that benefit from material diversity, heterogeneity and / or personalized porous structure.

[0123] Using Figure 8b to further illustrate how such a manufacturing process could look like, this setup integrates an existing stereolithography (SLA) platform into which the AcoustoFab system. In this setup, a resin tank 830 with a leveller 832 inside and a UV projector 834 is installed to selectively cure and solidify the resin on the leveller. As an example, at the beginning, the leveller 832 is located just below the resin surface. After curing the first layer with 50-second UV illumination, the printing bed moves downward together with the solidified layer, and we repeat this process to print a 3D object layer-by-layer. We adapted this top-down approach even though the bottom-up one is commonly used in current commercial UV printers. This is because the top-down approach exposes the liquid surface to the PAT, allowing the creation of acoustic focal points and levitation traps above it. So AcoustoFab’s capabilities can be integrated into the SLA printing process for allowing multi-material fabrication, generating customizing porous structures, and fabricating fluidic channels.

[0124] A diversity of liquid materials (e.g., UV resin, water, agar) can be poured into the container or vat placed under the PAT and then be processed by using the system’s focusing capabilities. The PAT creates acoustic focal points onto the liquid surface at a certain level of acoustic pressure for a specific period. The system can be equipped with an extruder to pour a certain amount of the liquid into the container, so that we can print a 3D structure in a layer- by-layer manner. Our software controls the pressure amplitude and the period, enabling us to induce different phenomena, which are depression of the surface to create fluidic channels, ejection of a droplet, and generation of bubbles.

[0125] For applications where UV resin is used, the system is equipped with a UV illumination device, as shown for example in Figure 8b. This can be a UV lamp to cure the resin entirely or a UV projector to cure it selectively. We control the UV illumination synchronously with the PAT control, so that the created fluidic channels, bubbles, or deposited droplets can be cured instantly to keep their forms. We can add another device to the AcoustoFab system to induce chemical reactions, such as a laser to apply heat to cure thermos-curable materials or to melt thermoplastics.

[0126] Another important way of using an extruder in the AcoustoFab system is to extrude droplets to be acquired and picked up by acoustic traps generated by the PAT. The extruder(s) such as those shown in Figures 8a and 8b, allow the precise control of the droplet volume to be levitated. This is important for realizing stable levitation of liquid droplets because the minimum and maximum acoustic amplitudes that can be applied to the droplet depend on the droplet volume (as explained below, too high pressures can cause a droplet to burst, while to low pressures will lead to the drop not being levitated). The AcoustoFab system controls the pressure depending on the material and volume size of the droplet in real-time, ensuring the appropriate pressure is applied to each levitated droplet.

[0127] The deposited droplets can be used mainly for the rest of the capabilities of AcoustoFab: multi-material and multi-directional printing and liquid droplet mixing. For acoustophoretic printing, we deposit different materials from different directions (not limited to the traditional layer-by-layer manner) to construct a 3D object. While the translation of the droplets, our system keeps the optimum pressure values for stable manipulation. The main advantage of this approach is that it inherently avoids cross-contamination between nozzles or materials. The deposited droplets can be mixed in mid-air without any contact before the deposition. Our photo- or heat-inducing devices can be used to cure levitated droplets both before and after the deposition. Half-curing a droplet while it is levitated allows us to change its physical properties (e.g., surface tension) to realize more stable manipulation. We can then cure the deposited droplet again to completely solidify them.

[0128] In this system, the nozzles stay fixed unlike in other existing platforms which manoeuvre their nozzles on a 3D stage for deposition. When a nozzle ejects material, AcoustoFab’s platform immediately creates a levitation trap, which then holds, secures, and transports the levitated particle to its intended location. This approach makes it easier to add materials without adding dedicated 3D stages or robotic arms, which are structurally limited for three directional motions, and ensures the nozzle never touches other materials during the printing process, drastically reducing the risk of cross-contamination.

[0129] The core contribution in this disclosure (i.e. , allowing the creation of the setups above) lies in using variable time and amplitude control of focal points and levitation traps, for the manipulation of the liquid-air interface of an arbitrary substance, allowing functionalities such as fluid channel creation, single droplet ejection, localized porous structures, multi-material and multi-directional printing, and liquid droplet mixing, using a single platform that can integrate external, application-dependent components (e.g., such as containers, extruders, illumination devices or others).

[0130] The following sections provide a detailed description of this core contribution, structured in three main sections:

[0131] 1. We first describe a set of calibration procedures used to determine the operational parameters needed to manipulate the liquid-air interface.

[0132] 2. Second, we describe specific manipulation techniques enabled by our approach, by making use of the parameters identified.

[0133] 3. We describe more specific examples of applications exploiting the aforementioned parameters and techniques.

[0134] Calibration procedures

[0135] The techniques described later rely on the manipulation of the liquid-air interface, either to modify it or cause drops to eject from it. Such manipulations depend on a range of material properties (e.g., density, viscosity, surface tension) which could be analytically modelled. However, exploiting them in this way in a real device is challenging, due to device tolerances, and other potential modelling errors. We instead abstract the liquid-air interface as a simple spring-damper model dependent only on the properties of the material manipulated (m) and its container (c), and propose a set of calibration techniques to determine the operating parameters required to apply our techniques. Please note, all parameters derived from these calibration techniques are specific for each type of material and container (e.g., ath(m, c), dmax(m, c)), but we omit such dependencies from our description for simplicity.

[0136] Please also note the calibration techniques described in Figures 9a to 9e make use of simple look-up tables to record dependencies between such parameters and the pressure or timing applied. Other techniques, such as interpolation, regression models or Al models could be used to model such dependencies instead.

[0137] Finally, each calibration technique is presented to derive a specific set of parameters, keeping their descriptions as simple and focused as possible. As a result, some calibration techniques share some steps and they can be merged into a smaller set of techniques (i.e., at the expense of focus and simplicity).

[0138] Calibration technique 1 - Surface tension threshold:

[0139] Figure 9a illustrates this first calibration technique which is used to determine the relationship between the pressure applied and the surface tension of the material at hand. When acoustic pressure is applied, the surface of the liquid undergoes depression, and trenches are formed. More specifically Figure 9a shows how to determine the maximum amplitude pressure athwhich does not overcome the surface tension of the liquid material, as well as the depression dmaxsuch pressure exerts on the material. Pressures below this threshold will cause increasing depressions, while pressure above it will cause a local disruption to the surface, mixing air into it and creating localized air bubbles.

[0140] We use the PAT setup shown in Fig. 8a, with the container replaced by a clear container (preferably of circular shape). The container is open, and its surface much larger than the diameter of a focal point (i.e., ~14mm in our 40KHz setup), so that the pressure applied has a minimum impact on the overall pressure of the liquid material. The setup is equipped with a camera to record a liquid surface.

[0141] At step S900, the material to characterize (e.g. liquid) is poured into the container. There is enough material in the container so that interactions of the depression generated with the bottom plate can be neglected. In other words, the height of the liquid in the container is at least 2dmaxwhere dmaxis the maximum height of the depression. At step S902, the setup is adjusted to ensure a fixed known distance between the PAT and the liquid surface (e.g., 12 cm in our setup). The geometry of the container and any other elements under the PAT are modelled accordingly (i.e., object mesh information in S202, in Figure 3). At step S904, we use the method above to determine the control instructions for each transducer. At step S906, the control signals are used to generate a single focal point of a specific target amplitude a on the surface of the liquid. At step S908, the surface depression depth, d(a) is measured and recorded. This can be measured in various ways, such as by using a camera placed at the same level as the liquid surface (e.g., for clear / transparent materials), or using distance meters. Such methods can also be used to detect the transition from a depression (i.e., locally circular depression) to bubble generation. Other characteristics of the depression may also be optionally detected in this step. For example, the depression may be in the form of a trench, and thus the channel dimensions and curvature, whether circular or rectangular may also be recorded. The dimensions of and the time to form the trenches will vary depending on the liquid properties. For instance, it takes more time to form a depression in highly viscous UV curable resin than in water or less viscous fluid. The trench formation time (T) may be directly proportional to liquid viscosity and inversely proportional to the applied acoustic pressure.

[0142] We start from a target amplitude a = 0, and then gradually increase pressure (e.g., steps of 100 Pa in our examples) as indicated at step S910. Note here that the focal point is constantly active, even as we increase to the next pressure amplitude, as to avoid restorative forces from the liquid’s surface tension from causing disturbances in the material’s surface (i.e., if the focal point is turned off, surface tension will cause the depression to be refilled, and waves to propagate along the surface).

[0143] We observe the effects on the liquid surface and at step S912 detect whether a bubble has formed. If no bubble has formed, we loop back to step S908 to measure d(a) and optionally other characteristics. We continue increasing amplitude pressure until bubble formation on the liquid surface is detected, and at step S914 we record the maximum target amplitude before surface tension is overcome (atft) and the related surface depression depth at such pressure (dmax). The data from the previous steps can be used to generate a look-up table relating pressure applied and resulting depression.

[0144] The pressure at the liquid-air interface increases as the acoustic pressure amplitude increases or the viscosity of the solution decreases. However, the relationship between the size of the bubbles and the applied acoustic pressure is not directly proportional. Bubble size depends on various factors, such as the frequency and power of the ultrasound, the properties of the liquid, and the presence of dissolved gases. The complex relationship is described for example in “Effect of static pressure on acoustic energy radiated by cavitation bubbles in viscous liquids under ultrasound” by Yasui et al published in the Journal of the Acoustical Society of America in November 2011. The pore size of the bubbles can be defined as inversely proportional to the viscosity of the fluid medium; inversely proportional to the applied acoustic frequency; and / or directly proportional to the acoustic power. Other parameters which also determine material porosity are density of the medium, surface tension of the medium, temperature of the medium, transducer duty cycle / exposure and acoustic pressure velocity or medium velocity. These relationships can be derived for example from the literature, for example from “Sonochemistry-assisted synthesis and optical properties of mesoporous ZnS nanomaterials” published in the Journal of Materials Chemistry A and “Sonochemistry: A Greener Protocol for Nanoparticles Synthesis” published by SpringerLink. It will be appreciated that once bubbles are created using the method of Figure 9a, such parameters can also be captured and stored into look-up tables during this calibration step.

[0145] Calibration 2 - Surface restoration time:

[0146] Figure 9b illustrates a second technique which is used to determine the restoration time (tr), that is, the time elapsed after the pressure causing a depression on the surface is released, until the depression is re-filled due to the effects of surface tension. This restoration time must be considered when creating microchannels on the surface by quickly moving focus points along the microchannel length (see applications below).

[0147] The same setup and arrangement of the material used in Calibration 1 is used here. Thus steps S900 and S902 described in relation to Figure 9a are repeated. In this arrangement, there is a high-speed camera to detect the current height of the depression caused by the focal point.

[0148] In this case, we determine at step S916 control instructions to generate a single focal point at one of the target amplitudes a recorded during Calibration 1. This focal point is created at step S918 and at step S920 we extend the method above as to measure the depression at each point in time d(a, t).

[0149] At step S922, we remove the focal point (e.g., turn PAT off), and record the depth at a detected time at step S924. There is a determination at step S926 as to whether the depression created in step S918 has been filled. If not, the depth continues to be recorded at each time period. Over time, the liquid refills the depression. When the depression is filed, such restoration can be modelled as a simple spring-damper system, making it suitable at step S928 to record the frequency of the oscillations generated (i.e. , from lowest to highest point), and computing the restoration time tras the half-period of such oscillations (i.e., we compute this across two full oscillations, but a different number could be used instead).

[0150] Calibration 3 - Resonant frequency for droplet ejection:

[0151] Figure 9c illustrates a calibration which applies periodic pulses on the material surface to cause a local failure in the surface and cause ejection of single droplets of material (see the application described in Fig 11a). The pulses are tuned to the container’s resonant frequency (i.e., timing matches the time surface waves take to travel from the centre of the container to the edges and back to the centre. Circular containers with the focal point created at the centre of the surface are preferred, but any other containers can be used. The symmetry of the container and the material properties determine the feasibility / success of this technique (e.g., not applicable in very viscous materials, with high tr). The current calibration method is designed to determine the optimum timing to cause such ejection.

[0152] We reuse the setup used in Calibrations 1 and 2. Thus, steps S900 and S902 described in relation to Figure 9a are repeated. In this arrangement, a high-speed camera is used to detect the droplets ejected.

[0153] In this case, we determine at step S934 control instructions to generate a pulsed signal wherein each pulse generates a single focal point with target amplitude athfor a duration tr. At step S936, the focal points are generated using the pulsed signal. At step S938, the timing (T) between pulses is varied increasingly, (i.e. , increments of 0.1 ms, in our examples). At step S940, there is a determination as to whether the high-speed camera has detected a droplet. If not, the method loops back to increasing the timing until droplet ejection is detected by our imaging system.

[0154] When the droplet is detected, the timing which achieved the droplet is recorded at step S942. There is a determination as to whether the process is to be repeated at step S944. For example, the process may be repeated several times (e.g., 5 times), and each sequence of steps S936 to S942 might yield a slightly different timing (T’j). Once all iterations have been completed, the resonant frequency is computed from the average of the timings registered (fr= (E T;)-1) at step S946. Please note a short pause is required between steps, as to allow for the vibrations along the surface to be dampened and disappear.

[0155] Calibration 4 - Controlled droplet ejection:

[0156] Figure 9d shows a calibration technique which uses the frequency determined in Calibration 3, to provide a mapping between pressure applied and the height and volume of the droplet ejected:

[0157] The same setup used in Calibrations 1 to 3 is used here. Thus, steps S900 and S902 described in relation to Figure 9a are repeated. Like Calibration 3, a high-speed camera is used to detect the droplets ejected

[0158] In Figure 9d, we determine at step S944 control instructions to the PAT so that it generates pulses of a single focal point for a duration trat frequency fr(i.e., resonant frequency from Calibration 3). At step S946, the focal point is generated using the pulsed signal. At step S948, the target amplitude is increasingly varied between zero and the target amplitude athat fixed increments (e.g., 100Pa). At step S950, there is a determination as to whether the high-speed camera has detected a droplet. If not, the method loops back to increasing the amplitude until droplet ejection is detected by our imaging system. When the droplet is detected, the amplitude achieved by the droplet is recorded at step S952. Our imaging system also measures the maximum height reached by the droplet h(a, tr) and its volume v(a, tr). There is a determination as to whether the process is to be repeated at step S954. For example, the process may be repeated several times (e.g., 5 times), and at step S956 average values across the iterations may be taken to create a lookup table relating pressure amplitude with height and volume. Please note additional parameters can be derived from these (e.g., such as kinetic energy imparted on the droplet).

[0159] Calibration 5 - Pressure thresholds for droplet manipulation:

[0160] Figures 9a to 9d describe calibration techniques relating to the generation of focal points. Accurate control of pressure is also crucial to enable control of levitated particles, e.g. droplets ejected from an extruder or using Calibration 3 and 4. The droplet will simply fall if the pressure applied by the levitation trap is too weak, while it will be atomized (bust into an aerosol) if the pressure is too strong (i.e. , a(t) > amax). These values highly depend on the droplet volume and material.

[0161] Our levitation traps can be represented as standing waves, with nearly-zero pressure at the trapping point and a specific amplitude a (t) at the antinodes (i.e., A / 4 ~ 2mm below / above). We use such amplitude at the antinode (instead of trapping potential Uj or our simplified metric [ / / ) to characterize the weakest and strongest traps allowing successful manipulation of every given configuration (i.e., material, container, and droplet volume combination). Please note that for a specific setup (PAT, container and external objects), both pressure at the anti-node and potential are highly correlated. We use pressures to keep them consistent with other parameters derived along the previous calibration steps, but pressure at the antinode and potentials could be easily exchanged for this step (i.e., C / 7(t) or Uj'(t) could be used, instead of a(t)).

[0162] Figure 9e shows a method for determining the minimum pressure aminwhich allows controlled levitation of the droplet (i.e., for traps with a(t) > amin), as well as their atomization limit amax(i.e., droplets bursting if a(t) > amax), using the following steps:

[0163] We use the PAT setup shown in Fig. 8a, with the container replaced by an automatic liquid injector and a camera focusing on the injector’s outlet. At step S960, control instructions for each transducer are determined to create an acoustic trap at a particular point. At step S962, material is injected into the volume at the injection point. Such injector mechanism might involve the controlled droplet ejection mechanism addressed in Calibration 3&4 (in which case the container must be included in the setup), or a dedicated ejection mechanism such as the one shown in Fig 8a or 8b, or any other mechanism able to deliver the intended volumes of the materials of interest. A target volume v of a material of interest is injected, while creating an acoustic levitation trap with amplitude at the antinode a at the injection point in step S964. This trap might be a point below the ejector tip (e.g., 3 mm below the needle tip, for the setup in Fig 8a), or a point at height h(a, tr) above the container, matching the target volume v(a, tr) (i.e., for setups using the ejection mechanism in Calibration 3&4).

[0164] We start from the minimum a(t) value, and we increase a(t) at step S968 until finding the minimum pressure value required, aminat step S970 to trap the droplet. The minimum amplitude is recorded at step S972.

[0165] Once we find amin, we gradually increase the target pressure at step S974 until the droplet gets atomized by the high acoustic pressure as determined at step S978. This maximum amplitude is recorded at step S980 as amax.

[0166] Although not shown in Figure 9e, this process may be repeated as many times as necessary to find out the aminand amaxfor each configuration required (combination of material, container and droplet volume).

[0167] Manipulation techniques:

[0168] Our apparatus, two-step solver and calibration techniques above can enable seamless combinations of five key capabilities which are detailed below:

[0169] 1) Fluidic channel fabrication: this involves creating a depression / indentation in the liquid bath using targeted acoustic pressure below ath. This can be combined with methods to affect the nature of the liquid (e.g., UV light to solidify UV resins), resulting in a new approach to create permanent channels in the material in a contactless manner, which can then be combined with other manufacturing steps.

[0170] 2) On-demand, single-droplet ejection: The use of pulsed focal points of specific amplitude {a < ath), duration trand frequency fr, leads to controlled ejection of droplets of known volume v(a, tr) and at known heights h(a, tr). This allows contactless ejection and can be used for multi-material manipulation. For instance, in a pipetting system, our method avoids replacing plastic tips every time a new material is used, as well as reducing wasted material that might remain in the tips as they are disposed.

[0171] 3) Localized and targeted porous structures: The use of focal points at pressure amplitudes {a > ath), leads to the creation of bubbles in the liquid. This can be combined with methods to solidify the liquid to affect the local mechanical properties of the material, change its texture or its optical properties.

[0172] 4) Multi-material multi-directional printing: The combination of injection mechanisms (i.e., on-demand, single droplet ejection or conventional liquid ejection mechanisms), with the trapping of such droplets and their controlled deposition for printing in any desired pattern and direction, and on the surface of other objects (which might even be extended with micro-channels) enables contactless fabrication techniques for bioprinting, chemical mixing or food printing.

[0173] 5) Liquid droplet mixing: We developed a contactless mixing method using acoustophoresis to process materials separately in mid-air before mixing, which avoids contamination. This method allows for flexible structure fabrication by mixing a crosslinker and a binding agent just before deposition. The approach stands out by processing each material separately before mixing, ensuring high-quality results.

[0174] This application focuses on these five main capabilities of our platform, but our platform has many other fabrication possibilities, like picking and placing different solid or liquid components, sealing microfluidic channels with epoxy glue, and assembling resistors, batteries, and liquid metal ink droplets for building integrated circuits, which can have other applications for contactless fabrication.

[0175] Fluidic Channel Fabrication:

[0176] Figure 10a is a flowchart of a method for fabricating fluidic channels of specific depths d, combined with UV curing. In a first step S1000, we use a single PAT (e.g. 16 x 16 transducers) to create multiple acoustic focal points in a dish of UV curable resin. These multiple acoustic focal points may be created using the two-step scattering method and solver as described above. At step S1002, we set the amplitude at the focal points to the focal point amplitude (a) required to create the required depression depth (i.e., d(a)). The focal point amplitude may be determined using the look-up tables described above. When the focal points are close enough, they form a trench. It is noted that if the pressure is turned OFF, the fluid rim would retract inwardly and would eventually collapse to close the expanding depressions and trench (tr). Accordingly at step S1004, we simultaneously shine UV light when the focused acoustic pressure is ON to solidify the trenched resin and fabricate the channel.

[0177] As an alternative to create all the acoustic focal points at once in step S1000, a reduced set of points could be created and then moved them along the shape of the desired trench. In this case, each point along the trench must be traversed before liquid refills the depression created (tr), and the relationship between depth, time and amplitude (i.e., d(a, t), in Calibration 2) can be used to determine the traversing speeds of the focal points. In either case (using multiple static points, or a few moving points), this technique combines focussing sound waves on a fluid at a specific pressure depending on the fluid's height and properties with UV curing (or other techniques affecting the fluid’s mechanical properties) as trench formation occurs.

[0178] As shown at step S1006, we can optionally monitor progress of the channel formation by capturing the curvature dynamics and acoustic pressure-based trench / depression formation and extension using a high-speed camera. These can be compared to the known results from the calibration techniques to determine whether the trench is proceeding as expected, implementing closed-loop systems. Once the desired trench is formed, at step S1008, the PAT is turned off and the product can be output. Alternatively, as an optional approach, the bed on which the cured resin is located can be lowered, additional resin can be dropped using acoustic levitation, and steps S1000 to S1006 can be repeated to build up multiple layers each having a channel or trench fabricated therein. In each layer, we can create a different pattern determined by the holographic image from a phased array of transducers or a combination of transducers and metamaterials.

[0179] Figure 10b shows an example of a UV-curable resin 1002 in which an acoustophoretic focal point 1010 has been generated using the technique of Figure 10a. It will be appreciated that two such focal points may be generated but only one is shown for illustrative purposes. Merely as an example, the resin may be of 150 mL volume poured into a petri dish and transducers of 40 Hz acoustic frequency may be used to generate the two acoustic focal points which each 8 mm in diameter. Figure 10c shows that in this example, a 20 mm length and 8.6 mm wide fluid channel 1012 was created in the cured resin 1016. The channel was filled after production with a coloured water solution 1014 made by mixing pink food colour dye to make it easier to see. Alternatively, as we cure the resin, we can also inject a new liquid into the channel to modify the printed structure's properties. The volume, location, and material properties of this liquid can be varied, and the liquid can be deposited in a contact-free way using acoustic levitation.

[0180] In this example, the focal point size approximately equates to the wavelength (~8.6 mm). Here, the focal point size is directly proportional to the wavelength and the focal length, and is inversely proportional to the aperture size of the PAT. Thus, the technique of pressure focusing is scalable, and one can use transducers of 300 kHz (wavelength 1 mm) or 1 MHz (wavelength 0.3 mm) respectively to fabricate microfluidic channels (pm to mm) using the same principle. Thus, the current setup can be adapted to obtain thinner fluidic channels.

[0181] Thus, we accomplished unique texturing on a bath of UV-curable resin solution using localized acoustic pressure. We cured the silicone bath with a focused holographic pattern to demonstrate the real-time reconfigurable fabrication of fluidic channels. The proposed technique provides high reconfigurability and flexibility to design, re-shape and develop channels without sophisticated tools and machines. Since the proposed technique is material agnostic, any channel-building material can be used for fabrication, like Polydimethylsiloxane (PDMS), by utilizing the same technique as demonstrated for UV-curable resin. The proposed technique allows the simultaneous fabrication of multiple fluidic channel devices with variable materials in different positions within a fixed spatial volume, providing multi-dimensional control and greater degrees of freedom, which is not possible in other printers or cleanroom fabrication approaches. We can use a greater number of focal points to create more complex microfluidic channel shapes, for example, a T-junction can be created by generating four focal points at the terminus of each line of the T (i.e. , top, bottom, left and right). Furthermore, the present techniques can be adopted as a desktop-size fluidic channel fabrication tool, enabling versatile, “mask-less,” and cost-effective do-it-yourself manufacturing of fluidic channel-based devices.

[0182] Single Droplet Ejection or Mono Droplet Formation:

[0183] Acoustic droplet ejection technology has existed for more than a decade now. However, the most common configuration explored for ejecting or generating liquid droplet here has been using a single sound source immersed in the liquid to focus the acoustic pressure on the airliquid interface. Liquid ejection from a surface can be achieved using transducers coupled from below. However, this approach limits the design flexibility since the transducer has to be located close to the bottom of the container. Additionally, only a single transducer / speaker can be used, and there is no manipulation of the acoustic phase to enhance energy. This method also requires a shallow container with a small amount of liquid. The physical mechanism behind this shooting is the formation of cavitation bubbles which reach the liquid surface. The substantial internal and external pressure difference makes the bubbles rupture and produce a strong impact. The impact caused the surrounding liquid to eject from the liquid film due to atomization. The focused pressure creates a cavity and liquid droplet ejects from the surface.

[0184] Figure 11a illustrates a schematic set-up for droplet ejection, which is more flexible and versatile than the previously known systems described above. As shown in Figure 11a, there is a transducer array 1122 positioned above the liquid surface 1134 and directed towards it. This allows for greater design flexibility since the array can be located at any predetermined distance from the liquid surface. Additionally, the same array can be used for different purposes, such as ejecting liquid for contact-free fabrication purposes, handling the ejected liquid by moving and depositing it, and creating channels in the same or different liquids. As schematically shown the transducer array 1122 (also termed an external PAT board) is used to focus an acoustic field 1130 onto a liquid bath to eject a single droplet 1136. There is also an acoustically transparent mesh 1132 (i.e. a printing surface) above the top of the liquid bath 1134. The mono-droplet 1136 is ejected, and then deposited on the underside of the mesh (i.e. on the side of the mesh closest to the liquid).

[0185] Figure 11b is a flowchart of the steps in this printing process which completely removes the need for pipettes, syringes or needles, thereby truly allowing contactless printing from any liquid bath to any substrate. It is highly beneficial for the bio-chemical community where contamination and clogging in syringes are very serious issues. In a first step S1100, a series of acoustic pulses are generated at the centre of the container, to cause a droplet to be ejected from the liquid, with predetermined height and volume (i.e., using h(a, tr) and v(a, tr), from Calibration 4 to determine amplitude and height, and trand frfrom Calibration 3, to determine the duration and timing of the pulses). By varying these parameters according to the data in our look-up tables, we can control the droplet's size and height. For example, using a constant acoustic pulse duration of 20 ms and a water bath height of 7 mm, a single mono droplet is released from a liquid bath when acoustic waves were focused onto it, above a threshold pressure of 4500 Pa. It will be appreciated that water is just an example liquid and other materials can be used. Different pressures and pulses will be required and can be determined using the techniques described in the calibration 3 and 4.

[0186] At step S1111 , the ejected droplet is optionally, trapped, levitated and / or moved around using the sound field generated by the transducer array. This can be done using the techniques described above, namely the two-step method and solver together with using the information generated from calibration 5. This method works with multiple locations, with each location being in the same or different liquids. The ejected liquid droplet can then be deposited on a mesh or other suitable material at step S1112.

[0187] At step S1114, there is a determination as to whether more droplets are to be printed and if so, there is an optional step S1116 of repositioning the mesh. For example, the mesh can be controllably moved to pattern the deposited droplets in any required pattern, like the letter ‘II’ or ‘S’. The steps of ejecting the droplet and depositing the droplet may then be repeated. The movement of the mesh can be used as an alternative or in addition to moving the droplets before depositing them. Once there are no more droplets to be printed, the printed mesh is output at step S1118.

[0188] Optionally, the process can be monitored, for example using a high-speed camera. The monitoring can be used to determine the droplet formation time scale (few milliseconds). It was also observed that the droplet was rotating when ejected and stable when falling.

[0189] Our approach of utilizing targeted acoustic pressure focusing from air is unique and allows multiple degrees of freedom. Thus, our system can be considered to act as a contactless pipette where a micro-droplet can be generated from any liquid like oil, water, or resin and then it can be levitated and placed on any required operation. This contactless single droplet generation and manipulation has applications in numerous fields like micro-robotics, laboratory automation research, surface cleaning, droplet dispenser, additive manufacturing, liquid printing and patterning.

[0190] Localized and Targeted Porous Structure Fabrication:

[0191] 3D porous materials are used in numerous applications like tissue engineering, gas separation, sewage treatment, sensing etc. Our system uses focused ultrasound to induce localized microbubbles on a surface of liquid (e.g., resin) during a manufacturing process, creating tuneable porous structures that offer varying degrees of texture and bendability. In other words, we provide a tool providing a material-agnostic approach to pattern localized porosity of materials in any required shape, pore size and pore density by controlling acoustic pressure and its duty cycle.

[0192] We propose a targeted patterned cavitation bubble-based approach for the fabrication of porous materials. Figure 12a is a flow chart of the steps used in this method. In a first step S1200, we create a focal point on a surface of a liquid, e.g. a UV curable resin. The acoustic focal point is generated using the method described in relation to Figures 1 to 7 and the acoustic pressure at the acoustic focal point a is adjusted S1210 to generate at least one bubble using the calibration techniques of calibrations 1 and 2 (i.e. , a > ath). The at least one bubble is formed due to the acoustic force-based cavitation in the liquid bath caused due to inertial motion of the surrounding liquid. These are known as cavitation bubbles or inertial cavitation bubbles. The creation and size of the bubbles are regulated using acoustic pressure, the depth of the liquid bath and pulse length for the focused acoustic signal. These bubbles collapse to form daughter bubbles, which are chemically inactive. The repeated procedure of collapsing bubbles due to acoustic pressure produces multiple bubbles, which are patterned using acoustic traps.

[0193] At step S1214, there is a determination as to whether more bubbles are to be formed and if so, at step S1216, the acoustic focal point is then moved across the liquid surface. Repeatedly creating and developing multiple bubbles in a defined path allows a desired overall pattern to be created. For example, the pattern may be a straight line or bubble II shape on a very thin layer of resin. During the creation of the pattern, sound is continuously focussed on the liquid surface. As an alternative to creating a single focal point and moving it across the surface, multiple focal points can be created using a holographic pattern. A combination of the single focal points and the holographic pattern can be used. We can dynamically control the location, geometry, and pore density of our porous structures by controlling the applied sound field.

[0194] After the required pattern was achieved or simultaneously with generation of the pattern, the UV resin was hardened using UV light at step S1218. The resulting cured resin can then be output at step S1220 and an example of a printed output with a targeted and localized porous structure having a plurality of bubbles 1250 in the shape of the alphabet U, is shown in Figure 12b.

[0195] For printing UV polymer resin, the UV projector was used to project UV light with a designed pattern for crosslinking UV resin into a customized shape. Like in the other applications, multiple layers can be built up. For example, using an arrangement such as the one in Figure 8b, a leveller may be moved in the resin tank to move the crosslinked resin downwards after a layer has been crosslinked (cured) to allow the next layer of liquid resin on top of the printed structure for crosslinking. The resin was then crosslinked and printed in a layer-by-layer manner. Particles and droplets can be positioned in between the layers by acoustic levitation, and porous structures can be generated by acoustic focusing and are solidified by UV crosslinking. This process can be repeated in a layer-by-layer manner for hybrid 3D printing into a 3D-designated multi-material and porous structure.

[0196] Multi-material Multi-directional Printing:

[0197] This novel system leverages acoustophoresis to enhance the capabilities of 3D printing and presents an innovative solution to the material diversity problem by enabling multi-material printing through customized multi-syringe extrusion and acoustic levitation. Moreover, it addresses the structural complexity issue by inducing localized microbubbles (as explained in above section) within the material in a controlled manner for creating tuneable porous structures to add a dimension in display, texture, taste and functionality.

[0198] Figure 13a is a flowchart of the process for controlling multi-material, multi-directional printing. In a first step S1300, we create an acoustic trap within a printing volume and at step S1310, a droplet is injected into the volume and caught in the acoustic trap. It will be appreciated that these steps can be done simultaneously or in a different order. Merely as an example, we levitated UV-curable resin liquid droplets (~2 mm radius) using acoustic force trapping through standing waves. Figure 13b shows an example of a droplet 1310 in a volume 1316 defined between a transducer array 1322 and a deposit surface 1324.

[0199] At step S1314, it is determined whether the droplet is in the correct location for printing and if not, the levitated droplets were moved in the printing volume by controlling the acoustic field, i.e. by moving the acoustic trap point at step S1316. After the droplet was moved to the correct position, merely as an example approximately 1 cm above the target depositing position, the acoustic field was turned OFF at step S1318 to cause the droplet to deposit on the substrate just below its levitated position.

[0200] At step S1320, there is a determination as to whether more droplets are required to be printed and if so, the process is repeated. Once all droplets have been deposited, the printing liquid may optionally be cured at step S1322, e.g. if the liquid is a curable resin. The printed object can be output at step S1324. Merely as an example, a hybrid liquid pattern deposition of a letter II fabricated using the process above in shown in Figure 13c.

[0201] The proposed technology has numerous unique selling points over the existing printing or fabrication technologies. It can be used as a desktop printer to provide ease of fabrication with the same efficiency as conventional methods and is compatible with all materials which cannot be used in 3D printers or conventional methods. It can do fabrication on any surface; it doesn’t need to be flat or horizontal. Traditional fabrication approaches take ample time and resources for this multi-layer and multi-alignment fabrication, unlike our system, which allows multi-axis parallel control of multiple levitated liquid resin droplets. The process can be automated and programmed to produce multiple batches of the same or reconfigured device design for large-scale production.

[0202] Thus, the process facilitates sound manipulation for multi-material printing by acoustically levitating different materials, such as conductive inks and hydrogel, into the design in a contactless manner. This printing technique can revolutionize many microfabrication techniques like lithography, where the photoresist can be deposited on the substrate in any desired pattern with good resolution using our controlled acoustophoretic levitation and deposition.

[0203] Liquid Droplet Mixing

[0204] Figures 14a to 14d show how levitation of multiple droplets simultaneously enables chemical mixing in mid-air. We have used this technique to crosslink hydrogel droplets before deposition. By mixing a crosslinker and binding agent just before deposition, we can achieve flexibility in structure fabrication, which is not possible with conventional printers.

[0205] As shown in Figure 14d, at step S1400, a first acoustic trap is generated within the printed volume and a droplet of a first fluid, e.g. a hydrogel is injected into the printing volume and caught in the first acoustic trap at step S1402. Similarly, at step S1410, a second acoustic trap is generated within the printed volume and a droplet of a second fluid, e.g. a crosslinking solution is injected into the printing volume and caught in the second acoustic trap at step S1412. These steps may be done in any order, e.g. simultaneously or sequentially. The result of these steps is that there are two acoustic traps for levitating a hydrogel droplet and a droplet of crosslinking solution simultaneously and two such droplets are shown in Figure 14a.

[0206] Returning to Figure 14d, the next step S1414 is to merge the droplets by moving the two traps moved to the same position to mix two droplets into one. Figure 14b illustrates the two droplets of Figure 14a mixed into a single droplet. As soon as the two traps merged, a single, third acoustic trap is generated at the same position at step S1416. The third acoustic trap is used instead of the first and second traps to keep levitating the crosslinked hydrogel bead at step S1418. The merged droplet can then be printed in a similar manner to that described in relation to Figure 13a. For example, there is a determination at step S1420 whether the merged droplet is at the correct location. If not, the third acoustic trap is moved at step S1422. Once the droplet is in the correct location, the acoustic field may be deactivated at step S1424 to deposit and print the merged droplet on a substrate. Figure 14c illustrates an example of the merged droplet printed on a surface (in this case a vertical surface, e.g. one which is perpendicular to the array). The method may then loop back to print additional merged droplets. It will be appreciated that merging two droplets is illustrative and more droplets may be combined.

[0207] The levitated droplets may be cured (at least partially) before and after deposition. By half-curing droplets while levitated, we can alter their physical properties to achieve rock-solid stability in manipulation. And finally, our deposited droplets are completely solidified without a doubt. The droplets may be treated (e.g. cured or heated) before mixing. For instance, we can heat one material to a specific temperature while curing the other using UV light, without any cross contamination. This technique also allows us to create Janus particles (i.e., particles having two distinct faces with different physical or chemical properties).

[0208] Maximum potential was used for all traps involved in this process. The contactless nature of the acoustophoresis-based levitation also allows chemical mixing without contamination because droplets can be transported while confidently maintaining optimum pressure values for stable manipulation, ensuring zero contamination between nozzles or materials. For bioprinting hydrogels, mixing with crosslinking solution solidifies the hydrogel and enhances its degradation property. The crosslinking of hydrogel droplets before deposition allows printing crosslinked hydrogel beads for drug delivery applications. We mix droplets in mid-air before deposition with ease.

[0209] Applications

[0210] Example application of Localized and Targeted Porous Structure Fabrication - Food printing

[0211] The technique of Figure 12a can be used to 3D print food. The acoustic focusing can generate customized porous structure including drawings or words in jelly that tunes its visual display and unique texture. As examples of food printing, gelatine (Sigma-Aldrich G1890), agar (Mr. P Ingredients) and sodium alginate (Sigma-Aldrich W201502) can be used to make jelly. The food substance, e.g. the jelly, gelatin and agar, are loaded in a syringe (also termed extruder) and then extruded with custom-made syringe extruders in measured amounts into a container, e.g. a petri dish or a shaped vessel such as a chocolate cup. The customized vessel may itself have been 3D printed by a food printer (e.g. using Procusini 5.0). The customized pore structure was generated by applying acoustic focusing.

[0212] Figures 15a to 15e show the results of multi-material food fabrication done using the process of Figure 12a. As you can see, there are different results with customized size, shape, and porous structure in different substances. Figure 15a shows agar jelly with a customized butterfly drawn as a porous structure in the jelly. Figure 15b shows agar jelly with a customized porous structure and colours spelling the letters CHI. Figure 15c shows a gelatine jelly with a customized butterfly porous structure in a 3D printed white chocolate cup of butterfly shape. A 3D food printer (Procusini 5.0) has also been used to print the white chocolate cup to contain the jelly providing a more solid and stiffer texture compared to the jelly itself. Figure 15d shows two types of gelatines with customized porous structure in a 3D printed chocolate cup of yin yang.

[0213] Figure 15e shows a multi-material and porous cake with two layers of different gelatines. The bottom layer is red with a circular porous structure while the top layer is transparent. Droplets of custard cream sauce were deposited in between the gelatine layers and sprinkles were deposited on top of the structure. Acoustic levitation technique has been used to levitate and print sprinkles and other flavourings on the food. Polysaccharide jelly and sprinkles were printed both between the layers and on top of the dish. Thus, the current process has integrated diversity in the display, texture, taste and smell in the cake from multiple materials and customized porosity structure that can be customized to each individual.

[0214] Example application of Multi-material Multi-directional Printing - Bio-printing

[0215] Bioprinting involves layer-by-layer precise positioning of bioink consisting of biomaterials, and living cells to directly fabricate complex cellular tissue structures. As acoustophoresis has high resolution and cellular compatibility, it allows us to levitate bio droplets and position them to print into a 3D structure. Acoustic levitation provides a high degree of freedom for the movement of levitated droplets / particles. It allows printing on surfaces at different angles, surpassing the challenge of having to print on a flat surface for traditional 3D printing. This enables in situ bioprinting directly on a complex tissue structure for tissue regeneration. The contactless nature of the acoustophoresis-based levitation eliminates the risk of contamination and damage to the host tissue during in situ bioprinting.

[0216] Figures 16a to 16d illustrate an example of acoustic 3D printing hydrogel composite droplets using acoustophoresis-based levitation. In this example, a hydrogel composite ink composed of 3% alginate and 1% gelatine was used. Alginate has been widely used for 3D bioprinting and drug delivery applications due to its well-known biocompatibility and crosslinking ability and gelatine provides cell adhesive property and shear-thinning property. To make the hydrogel composite ink, weighted alginate powders were mixed with de-ionized water (DI water) under magnetic stirring (500 rpm) for 1 hour. Then weighted gelatine powders were added and mixed under magnetic stirring (200 rpm) for 1 hour. The inks were stored for at least 1 day before further tests to allow equilibration and were centrifuged (2000 rpm for 1 minute) to eliminate bubbles.

[0217] Figure 16a illustrates the printing device which comprises a transducer array (PAT) 1622 below which there is a printing volume. A deposit surface 1624 is below the PAT and as shown by the arrows, the deposit surface may be parallel or perpendicular to the PAT. The PAT is above the substrate emitting standing ultrasound waves to levitate and print hydrogel droplets which are injected into the printing volume by the injector 1612. Figure 16b is a visualization of the sound field forming an acoustic trap 1610 on top of the deposit surface which in this case has a wavy surface.

[0218] Figure 16c shows a hydrogel droplet above the wavy deposit surface structure and shows that several droplets have already been printed in a drop-on-demand manner. In this arrangement, hydrogel droplets were levitated and positioned accurately on top of and then dropped onto the structure. The droplets have connected and after crosslinking. Figure 16d illustrates the final ring structure which has been 3D printed using acoustophoresis-based levitation with hydrogel droplets on the non-planar (wavy) surface. The ring structure may have been printed with the deposit surface in a vertical position as shown in Figure 16a or in a horizontal position as shown in Figure 16c. When printing on a vertical surface, the droplets and printing surface have compatible rheological properties, whereby the droplets stay on the vertical surface and connect into the desired pattern, in this case a ring structure. It is also noted that the bottom surface of the printed structure matches the contacting deposit surface. In other words, the bottom surface of the printed structure is also wavy and compliant to the deposit structure proving its ability to integrate a printed structure on the receiving substrate.

[0219] Thus, using acoustophoresis, we can achieve drop-by-drop printing and manipulate multiple drops of different material properties. This manipulation allows us to deposit drops on any surface, whether it is vertical, horizontal, inclined, or curved. For example, this technique enables us to print hydrogels directly onto wounds for in-situ bio-printing. There is no contact with the drops which are being printed.

[0220] Although the above process shows one droplet being printed at a time, we can manipulate multiple droplets, up to 16 droplets at one time. Our printing technique is different from the conventional way of printing multiple materials using multiple nozzles attached to a robotic arm. We do not require a robotic arm for multi-directional printing and are thus not limited by the attached mechanical structure of the arm. Instead, we can manipulate multiple droplets in mid-air independently. Finally, as described in more detail below liquid drops may be combined before printing.

[0221] Example application combining multiple techniques - AcoustoFab Stereolithography (SLA) system

[0222] Figures 17a to 17d illustrate a system, method and output which combines various techniques described above.

[0223] Figure 17a shows a stereolithography (SLA) printing system which is similar to the one shown in Figure 8b. Figure 17b shows a PAT 1722 directly above a UV resin tank 1730 in which there is a printing bed 1732 (which may also be termed a leveller). The height of the printing bed 1732 is controlled, for example using a stepper motor 1736. The PAT 1722 generates an ultrasound field which is reflected by the printing bed and thus there is an acoustic volume defined between the printing bed (reflector) and the PAT. There is one extruder 1728 (which may also be termed an injector) in this arrangement which injects material 1742 which is to be levitated by the generated sound field and can be incorporated as additional material 1744 in the UV resin as explained below. There is an ultraviolet (UV) projector 1734 which emits UV light to cure the UV resin in the tank 1730. For selectively curing UV polymer resin into a customized shape, one example of a setup uses a UV projector (Keynote Photonics LC4500-UV / B). Porous zones can be created as explained above.

[0224] The positioning of PATs is crucial for achieving optimal material manipulation. For the system to function seamlessly, the PATs must situate itself in a position where it can proficiently create and control levitation traps. After examining various configurations, we adopted a single-sided setup, utilizing the additive manufacturing platform itself as a reflector, as shown in Figure 17a. This configuration leverages the flexibility of a single-sided setup while achieving levitation capabilities comparable to a setup in which there are PATs above and below the acoustic volume. A single-sided set-up allows for seamless integration with current additive manufacturing platforms without compromising their inherent features. It is worth noting that creating focal points is less demanding in terms of the PATs’ positioning, so we prioritize material manipulation.

[0225] Comparing the arrangement of Figure 17a with other arrangements, it is noted that the top-bottom arrangement of PATs is widely recognized for its superior levitation capabilities. However, its integration into an additive manufacturing system presents challenges because elements of the manufacturing system (e.g., a printing bed) can obstruct ultrasound waves from the bottom PAT, preventing the creation of levitation traps. This problem can be addressed by rotating the setup by 90° (i.e., side-by-side setup). However, this arrangement poses its challenge: generating an acoustic radiation force strong enough to counter gravity, given that this force is inherently stronger in the direction of sound propagation. Single-sided setups offer greater adaptability but have limited levitation capabilities in the absence of a sound reflector. The proposed set-up generates the desired results.

[0226] The ability to print multi-materials can greatly enhance the versatility, customization and functions of the printed products. Although there is only one extruder 1734 shown for simplicity in Figure 17a, it will be appreciated that there may be several extruders, each injecting different materials. Alternatively, the same extruder may be used to inject different materials. Liquid materials can be extruded and directly deposited on the substrate in a controlled manner. Furthermore, both soft and liquid materials can be extruded in droplet form, levitated in midair, and subsequently printed onto complex surfaces. Similarly, solid particles can be levitated and subsequently printed. The solidification techniques for liquid or soft materials encompass equilibration, thermal processes, and customized ultraviolet (UV) curing methods. We utilize sound to pick and place objects through acoustic trap-based control. Additionally, these control traps can be created anywhere in the 3D space within the sound field control volume. They do not have any attached structure like robot-arm, which allows in tandem parallel trapping and deposition of multiple material from all dimensions in space, improving both structural diversity, and speed of printing. This allows multiple materials and almost any ink irresponsive of its properties like thermoplasticity, conductivity, magnetic properties.

[0227] By contrast, most commercial 3D printing techniques can only print one material at a time on one structure. For example, fused-deposition modelling (FDM) printers allow only one filament to be extruded from one nozzle and printed and most FDM printers only have one nozzle. For the FDM printer that allows printing different materials by equipping different nozzles, the materials must be of the same type, thermoplastics. Layer by layer FDM printing prototyping has largely been employed for design verification and small production runs. However, most of these printers allow limited thermoplastic resins as permitted ink materials. Furthermore, robot-arm attached nozzle mechanism limits the volumetric deposition of ink to areas structurally limited by motion of arm, affecting both speed and efficiency of prints.

[0228] The UV resin tank 1730 can only hold one type of UV resin. However, as shown in Figure 17d, the system can be used with a diverse range of modalities for the deposition or curing of various materials. The process addresses the structural complexity issue by inducing localized microbubbles within the material in a controlled manner for creating tuneable porous structures to add a dimension in display, texture, taste and functionality.

[0229] Figures 17b and 17c illustrate a product generated using the system of Figure 17a. The product is a butterfly shape formed by curing the UV resin, examples of suitable resins are a transparent resin such as Liqcreate Clear Impact™ or a stretchable resin such as Liqcreate Flexible-X™. Microbubbles have been created to form a porous structure in the centre of the product to make this specific region opaque due to light-scattering. In this way, the “wings” of the butterfly may appear to glow in the dark. Multiple materials have been deposited using acoustic levitation. These multiple materials may include for example resin droplets loaded with functional fillers including photoluminescent and UV fluorescent powders. The product may be described as a light-manoeuvring device that incorporates light-scattering, photoluminescence, and fluorescence effects. As shown in Figure 17c some of these materials emit red fluorescence.

[0230] Figure 17d shows the steps of a multi-modal process using the system of Figure 17a. As shown at step S1700, in a first step the distance between the PAT and printing bed is set. Merely as an example, which is used to achieve the results shown in Figures 17b and 17c, the PAT and the UV resin tank are aligned vertically with a separation of 12 cm. The printing bed is attached to a stepper motor, and thus, its height can be controlled. At the beginning, the printing bed needs to be located just below the resin surface. At step S1702, there is an optional step of curing a base layer. For this example, we used transparent UV-curable resin (Liqcreate Clear Impact) for the base of the structure to avoid light absorption by the resin. The first layer may be cured with 50-second UV illumination. After curing, the printing bed moves downward together with the solidified layer.

[0231] At step S1704, there is a determination as to whether any material is to be deposited using the acoustic trap process described above. If material is to be deposited, at step S1710, one or more droplets are injected into the acoustic volume (i.e. between the PAT and the printing bed). When multiple droplets are injected, they may be combined as described above, e.g. to create resin droplets loaded with functional fillers including photoluminescent and UV fluorescent powders or they may be maintained separate. At step S1712, there is a determination as to whether the droplets (combined or separate) are in the correct location and if not, the acoustic trap point is moved at step S1714 until the droplets are in the correct location. When the droplets are in the correct location, the acoustic field may be turned off to deposit the droplets at step S1716.

[0232] At step S1718, there is a determination as to whether more material is to be deposited. If so, the process loops back to step S1710 to inject and trap more droplet(s). The steps of moving the acoustic trap points and depositing are then repeated (step S1712 to S1718). The techniques described above to use acoustic levitation are thus used to deposit at least one resin droplet in at least one cycle. If all the material which is desired to be printed on this layer has been deposited, there is then a determination as to whether any structures are to be created at step S1706. If no structures are to be created, the process moves to curing the layer at step S1728 (if this has not already been done).

[0233] The structures, e.g. channels or bubbles, may be generated as described in more detail as above. Thus in summary, an acoustic trap may be created on the surface of the resin or within the resin at step S1720. The acoustic field is then adjusted to create the desired structure(s) at step S1722. When creating channels, the resin may be cured whilst creating the structures, thus there is an optional curing step S1724. There is a determination as to whether there are more structures at step S1726 and if more structures are required, the process loops back to step S1720 to generate the required acoustic traps. If no more structures are required, the layer may be cured as shown at step S1728 (if this has not already been done).

[0234] Once the layer has been cured, there is a determination as to whether more layers are required at step S1730. If more layers are required, this process is repeated until the print of a target 3D structure finishes. Once the product is complete, it can be output. Figure 17d shows that when more layers are required, the process loops back to step S1704 to determine whether material needs to be deposited on the new layer. Clearly, the distance between the PAT and printing bed may also have been reset but is not shown for simplicity. The process in Figure 17d may be described as a top-down approach because it provides space above the liquid surface for the AcoustoFab setup to create acoustic traps and focal points during the printing process.

[0235] It will also be appreciated that although the determination regarding depositing material is made before determining whether to create any structures, the order of these determination could be reversed and subsequent changes to the process made. The determinations may also be done simultaneously and if both material is to be deposited and structures are to be generated, a decision as to which to do first may also be made. The necessary acoustic traps and focal points can then be created to allow for contactless deposition of material(s) and the generation of porous structures at any target locations between the layers, realizing hybrid 3D printing into a designated multi-material and porous structure. The overall shape of the product, e.g. the butterfly shape, may be obtained by the controlled curing of the resin.

[0236] Summary

[0237] As described above, we have achieved a multi-platform hybrid tool spanning a wide range of fabrication capabilities, from fabricating patterned microfluidic channels to localized and targeted porous structure contactless printing. We provide a thorough evaluation of the proposed tool (termed AcoustoFab), showcasing its potential in practical applications.

[0238] Our software controls the pressure amplitude and period to induce various phenomena such as depression of the surface to create fluidic channels, ejection of droplets, and generation of bubbles. The range of pressure applied is related to both the fluidic channel and bubble generation, and the ejection of droplets is a function of the applied acoustic pressure impulse function. We also demonstrate novel fabrication approaches of material agnostic printing and localized microbubbles for porous printing. The acoustic trapping parameters of acoustic pressure, acoustic pulse length, and fluid properties, which affect material roughness, accessibility, and flexibility, were optimized to handle different fabrication techniques. As described above, we successfully implemented a versatile system for fabricating microfluidic channels, patterned porous structures, and a multi-material printing system, highlighting the extraordinary properties of design reconfiguration, axis-free, multi-orientation, material agnostic levitation and deposition operation.

[0239] Five capabilities can be combined seamlessly in a single fabrication process. For example, we can induce fluidic channels and bubbles on the surface of UV resin and then deposit different materials onto those channels or bubbles. These combinations allow for multimaterial and multi-functional fabrication. This application integrates acoustic levitation and acoustic focusing techniques enabled by PAT and custom-made SLA technique for fabricating customized multi-material and porous structures. Acoustic levitation technique allows positioning and depositing a variety of droplets or particles and assembling them for multi- materials printing. Acoustic focusing generates bubbles in a controlled manner for creating porous structure. The custom-made SLA technique is capable of 3D printing UV resin in a layer-by-layer manner while allowing deposition of droplets and particles and porous generation using PAT. Our printing method generates complex, layer-by-layer structures with specific material distributions and precise porous patterns with great potential for various applications.

[0240] There are potential applications for biofabrication, such as cytoscribing, large-scale micropatterning, and high-throughput bioanalysis. We develop spatially localized patterned surfaces that can give localized topographical or structural features or other functionalities relevant to sensor or biocomposites or food engineering applications.

[0241] Those skilled in the art will appreciate that while the foregoing has described what is considered to be the best mode and where appropriate other modes of performing present techniques, the present techniques should not be limited to the specific configurations and methods disclosed in this description of the preferred embodiment. Those skilled in the art will recognise that present techniques have a broad range of applications, and that the embodiments may take a wide range of modifications without departing from any inventive concept as defined in the appended claims.

[0242] References

[0243] [1] Melde, K., Choi, E., Wu, Z., Palagi, S., Qiu, T. and Fischer, P. Acoustic Fabrication via the

[0244] Assembly and Fusion of Particles. (2018). Advanced Materials, 30, 3 1704507. https: / / doi.Org / 10.1002 / adma.201704507.

[0245] [2] Lu, L., Tang, X., Hu, S. and Pan, Y. Acoustic Field-Assisted Particle Patterning for Smart Polymer Composite Fabrication in Stereolithography. (2018). 3D Print Addit Manuf, 5, 2 151- 159. 10.1089 / 3dp.2017.0157.

[0246] [3] Ma, Z., Hoile, A. W., Melde, K., Qiu, T., Poeppel, K., Kadiri, V. M. and Fischer, P. Acoustic Holographic Cell Patterning in a Biocompatible Hydrogel. (2020). Advanced Materials, 32, 4 1904181. https: / / d0i.0rg / l 0.1002 / adma.201904181.

[0247] [4] Sriphutkiat, Y., Kasetsirikul, S., Ketpun, D. and Zhou, Y. Cell alignment and accumulation using acoustic nozzle for bioprinting. (2019). Sci Rep, 9, 1 17774. 10.1038 / s41598-019- 54330-8.

[0248] [5] Collins, D. J., Alan, T., Helmerson, K. and Neild, A. Surface acoustic waves for on-demand production of picoliter droplets and particle encapsulation. (2013). Lab on a Chip, 13, 16 3225- 3231. 10.1039 / C3LC50372K.

[0249] [6] Tanaka, H., Wada, Y., Mizuno, Y. and Nakamura, K. Effect of holed reflector on acoustic radiation force in noncontact ultrasonic dispensing of small droplets. (2016). Japanese Journal of Applied Physics, 55, 6 067302. 10.7567 / JJAP.55.067302. [7] Habibi, M., Foroughi, S., Karamzadeh, V. and Packirisamy, M. Direct sound printing. (2022). Nature Communications, 13, 1 1800. 10.1038 / s41467-022-29395-1.

[0250] [8] Foresti, D., Kroll, K. T., Amissah, R., Sillani, F., Homan, K. A., Poulikakos, D. and Lewis, J.

[0251] A. Acoustophoretic printing. (2018). Science Advances, 4, 8 eaat1659. doi: 10.1126 / sciadv.aat1659.

[0252] [9] Llewellyn-Jones, T. M., Drinkwater, B. W. and Trask, R. S. 3D printed components with ultrasonically arranged microscale structure. (2016). Smart Materials and Structures, 25, 2 02LT01 . 10.1088 / 0964-1726 / 25 / 2 / 02 LT01.

[0253]

[0010] Melchert, D. S., Collino, R. R., Ray, T. R., Dolinski, N. D., Friedrich, L., Begley, M. R. and

[0254] Gianola, D. S. Flexible Conductive Composites with Programmed Electrical Anisotropy Using Acoustophoresis. (2019). Advanced Materials Technologies, 4, 12 1900586. https: / / doi.Org / 10.1002 / admt.201900586.

[0255]

[0011] Collino, R. R., Ray, T. R., Fleming, R. C., Cornell, J. D., Compton, B. G. and Begley, M.

[0256] R. Deposition of ordered two-phase materials using microfluidic print nozzles with acoustic focusing. (2016). Extreme Mechanics Letters, 8 96-106. https: / / doi.Org / 10.1016 / j.eml.2016.04.003.

[0257]

[0012] Greenhall, J. and Raeymaekers, B. 3D Printing Macroscale Engineered Materials Using Ultrasound Directed Self-Assembly and Stereolithography. (2017). Advanced Materials Technologies, 2, 9 1700122. https: / / doi.org / 10.1002 / admt.201700122.

[0258]

[0013] Yunus, D. E., Sohrabi, S., He, R., Shi, W. and Liu, Y. Acoustic patterning for 3D embedded electrically conductive wire in stereolithography. (2017). Journal of Micromechanics and Microengineering, 27, 4 045016. 10.1088 / 1361-6439 / aa62b7.

[0259]

[0014] Collino, R. R., Ray, T. R., Friedrich, L. M., Cornell, J. D., Meinhart, C. D. and Begley, M. R. Scaling relationships for acoustic control of two-phase microstructures during direct-write printing. (2018). Materials Research Letters, 6, 3 191-198. 10.1080 / 21663831.2018.1431317.

[0260]

[0015] Li, X., Lim, K. M. and Zhai, W. A novel class of bioinspired composite via ultrasound- assisted directed self-assembly digital light 3D printing. (2022). Applied Materials Today, 26 101388. https: / / d0i.0rg / l 0.1016 / j.apmt.2O22.101388.

[0261]

[0016] Ezcurdia, I., Morales, R., Andrade, M. A. B. and Marzo, A. 2022. LeviPrint: Contactless Fabrication using Full Acoustic Trapping of Elongated Parts. ACM, City, 2022.

[0262]

[0017] Marzo, A. and Drinkwater, B. W. Holographic acoustic tweezers. (2019). Proceedings of the National Academy of Sciences, 116, 1 84-89. 10.1073 / pnas.1813047115.

[0263]

[0018] Hirayama, R., Christopoulos, G., Martinez Plasencia, D. and Subramanian, S. High-speed acoustic holography with arbitrary scattering objects. (2022). Science Advances, 8, 24 eabn7614. doi: 10.1126 / sciadv.abn7614.

[0019] Montano-Murillo, R., Hirayama, R. and Plasencia, D. M. OpenMPD: A Low-Level Presentation Engine for Multimodal Particle- Based Displays. (2023). ACM Trans. Graph., 42, 2 Article 24. 10.1145 / 3572896.

[0264]

[0020] Kim, H., Kim, M., Lee, W., & Ahn, J. Gerchberg-Saxton algorithm for fast and efficient atom rearrangement in optical tweezer traps. (2019) Optics express, 27, 3, 2184-2196. https: / / doi.org / 10.1364 / OE.27.002184.

Claims

CLAIMS1. A computer-implemented method for manufacturing an object within an acoustic volume using an array of transducers which generates sound, wherein the acoustic volume comprises a scattering object in the form of a container of material, the method comprising: sending control instructions to each transducer in the array of transducers to generate sound to create an acoustic focal point at a least one control point on or under a surface of the material in the container; and adjusting operating parameters for each control instruction to generate at least one of a depression on a surface of the material, generate a bubble under the surface of the material and a droplet which is ejected from the surface of the material.

2. The method of claim 1 , further comprising sending control instructions to each transducer in the array of transducers to generate sound to create an acoustic trap at at least one further control point wherein the acoustic trap is configured to trap a target object to control the location of the target object within the acoustic volume.

3. The method of claim 1 or claim 2, further comprising: obtaining a static matrix (H) representing a contribution of each transducer in the array of transducers to each of a plurality of locations on the scattering object; defining the at least one control point at or under a surface of the material within the acoustic volume; calculating, in real-time, a direct transmission matrix (F) which represents a direct contribution to the at least one control point from each transducer in the array of transducers; calculating, in real-time, a scattering transmission matrix (G) which represents a scattering contribution to the at least one control point from the plurality of locations on the scattering object; determining, in real-time, an extended transmission matrix (F) which represents direct and scattered contributions from each transducer in the array of transducers to each of the multiple control points, wherein the extended transmission matrix is determined using the static matrix, the direct transmission matrix and the scattering transmission matrix from E = F + GH, and determining, using the extended transmission matrix, the control instructions for each transducer in the array of transducers to generate the acoustic focal point at the at least one control point.

4. The method of claim 3, when dependent on claim 2, further comprising determining, using the extended transmission matrix, control instructions for each transducer in the array of transducers to generate an acoustic trap at the at least one control point wherein the acoustic trap is configured to trap a target object to control the location of the target object within the acoustic volume.

5. The method of claim 4, comprising determining control instruction by defining a cost function which combines a trapping stiffness metric which is based on the trapping stiffness at each location of an acoustic trap and a pressure metric which is based on the pressure amplitude at each location of an acoustic focal point and minimising the cost function by minimising the trapping stiffness metric and the pressure metric.

6. The method as claimed in any one of the preceding claims, further comprising determining control instructions by: determining a target amplitude pressure at each control point; simulating an acoustic pressure at each control point; comparing the simulated acoustic pressure with the target acoustic pressure; determining, using the comparison, whether the target acoustic pressure is feasible; and when it is determined that the target acoustic pressure is feasible, adjusting the control instruction for an amplitude input for each transducer to generate the target amplitude pressure.

7. The method as claimed in any one of the preceding claims, wherein the operating parameters include one or more of an amplitude pressure for creating a depression, a depression depth at the amplitude pressure, a restoration time for a depression to be refilled by surface tension; a timing between pulses of a pulsed signal which causes a droplet to be ejected; an amplitude pressure for a pulsed signal which causes a droplet to be ejected; a minimum pressure which allows controlled levitation of an ejected droplet; and a maximum amplitude pressure which does not burst a droplet.

8. The method as claimed in any one of the preceding claims, comprising obtaining the operating parameters from one or more look-up tables which have been determined in a calibration process.

9. The method according to any one of the preceding claims, comprising determining control instructions by determining an amplitude pressure for each transducer which generates a depression of a particular depth at each acoustic focal point.

10. The method of claim 9, further comprising: continuously controlling each transducer to apply the determined amplitude pressure to create at least one trench; and simultaneously curing the material to form a manufactured object having the at least one trench.

11. The method of claim 9, further comprising: continuously controlling each transducer to apply the determined amplitude pressure; updating the control signal to each transducer to move each acoustic focal point to create at least one trench; and curing the material when the trenches have been created.

12. The method of any one of claims 9 to 11 , further comprising: inserting more material into the container; and determining an amplitude pressure for each transducer which generates a depression of a particular depth at each acoustic focal point in the inserted material.

13. The method of any one of claims 10 to 12, comprising sending control instructions for each transducer in the array of transducers to generate an acoustic trap to trap a target object to control the location of the target object within the acoustic volume whereby the target object is movable to be deposited in the at least one trench.

14. The method according to any one of the preceding claims, comprising determining control instructions by determining an amplitude pressure for each transducer which generates at least one bubble at each acoustic focal point.

15. The method of claim 14, further comprising: continuously controlling each transducer to apply the determined amplitude pressure to create a plurality of bubbles at different locations; and curing the material to form a manufactured object with a porous layer.

16. The method of claim 14, further comprising: continuously controlling each transducer to apply the determined amplitude pressure to create at least one bubble; updating the control signal to each transducer to move each acoustic focal point to create a pattern of bubbles; andcuring the material to form a manufactured object with a porous layer.

17. The method of claim 15 or claim 16, further comprising: inserting more material into the container; and determining an amplitude pressure for each transducer to generate more bubbles in the inserted material.

18. The method according to any one of the preceding claims, comprising determining control instructions by determining a timing between pulses and an amplitude pressure for each pulse applied by each transducer which ejects at least one droplet from a surface of the material.

19. The method according to claim 18, wherein determining control instructions further comprises determining the timing and the amplitude pressure to provide a droplet of a predetermined volume and at a predetermined height.

20. The method of claim 18 or claim 19, comprising sending control instructions for each transducer in the array of transducers to generate an acoustic trap to trap the ejected droplet and to control the location of the ejected droplet within the acoustic volume whereby the ejected droplet is movable to be deposited at a desired location.

21. A computer-implemented method for manufacturing an object within an acoustic volume using an array of transducers which generates sound, wherein the acoustic volume comprises a scattering object, the method comprising: obtaining a static matrix (H) representing a contribution of each transducer in the array of transducers to each of a plurality of locations on the scattering object; defining at least one control point at a surface of the material within the acoustic volume; calculating, in real-time, a direct transmission matrix (F) which represents a direct contribution to the at least one control point from each transducer in the array of transducers; calculating, in real-time, a scattering transmission matrix (G) which represents a scattering contribution to the at least one control point from the plurality of locations on the scattering object; determining, in real-time, an extended transmission matrix (F) which represents direct and scattered contributions from each transducer in the array of transducers to each of the multiple control points, wherein the extended transmission matrix is determined using the staticmatrix, the direct transmission matrix and the scattering transmission matrix from E = F + GH, and determining, using the extended transmission matrix, control instructions for each transducer in the array of transducers to generate at least two acoustic traps at the least one control point wherein each acoustic trap is configured to trap a droplet to be printed and to control the location of each droplet within the acoustic volume whereby the at least two droplets are moveable to be deposited at a desired location.

22. The method of claim 21 , wherein each droplet is a different material.

23. The method of claim 21 or claim 22, further comprising determining, using the extended transmission matrix, control instructions for each transducer in the array of transducers to control the location of each droplet within the acoustic volume whereby two droplets are mixed before being deposited.

24. The method of claim 23, further comprising processing each droplet differently before mixing.

25. An apparatus comprising: an array of transducers for generating acoustic pressure; an acoustic volume which is defined by the acoustic pressure generated by the array of transducers; and a processor for carrying out the steps of any one of the preceding claims to manufacture an object.

Images