WO2021064403A1

WO2021064403A1 - Acoustic particle manipulation

Info

Publication number: WO2021064403A1
Application number: PCT/GB2020/052404
Authority: WO
Inventors: Sriram Subramanian; Diego Martinez PLASENCIA; Ryuji HIRAYAMA
Original assignee: The University Of Sussex
Priority date: 2019-10-01
Filing date: 2020-10-01
Publication date: 2021-04-08
Also published as: GB201914174D0

Abstract

Disclosed herein is an apparatus comprising a set of acoustic sources controllable to generate an acoustic trap for suspending one or more particle(s) within the volume of the apparatus. The position of the acoustic trap within the volume of the apparatus is changed to cause a particle suspended within the acoustic trap to move around the volume of the apparatus. In particular, the acoustic field defining the acoustic trap is updated at such a rate that the particle is prevented from reaching a static equilibrium position within the acoustic trap. The apparatus may comprise a display apparatus. However, other non- display applications of such apparatus are also disclosed.

Description

ACOUSTIC PARTICLE MANIPULATION

The present invention relates generally to apparatuses and techniques for manipulating particles using acoustic waves (“acoustophoresis”). In some embodiments a display apparatus and corresponding methods for displaying content are provided. In particular, various embodiments relate to multimodal display apparatuses that may be capable of providing simultaneous visual, tactile and/or audio content. However, various other applications of the techniques presented herein are also contemplated. For instance, in other embodiments an apparatus and method for additive manufacturing (e.g. 3D printing) are provided.

BACKGROUND

The ability to manipulate particles in a more controlled manner may be important in various fields including, but not limited to, (micro) fabrication/manufacturing techniques, three- dimensional displays, and “lab-on-a-chip” devices. One approach for manipulating particles is by using acoustic levitation techniques, or “acoustic tweezers”, wherein a particle can be suspended within a medium using acoustic radiation pressure from acoustic waves in the medium, and the particle then caused to move, e.g., along acoustic pressure gradients.

Acoustic levitation is an example of the broader phenomenon of ‘acoustophoresis’, i.e. movement with sound. Other non-acoustic approaches for manipulating particles may rely on optical methods (e.g. optical tweezers), electric and/or magnetic forces, and so on.

The principles behind acoustic levitation are generally now well-understood. However, current implementations of acoustic levitation techniques suffer from various limitations in terms of their update speed and range of operations that may render them unsuitable, or at least undesirable, for some of the applications presented above.

In part for this reason acoustic levitation has not therefore been widely exploited, e.g., for three-dimensional display devices, and various other types of three-dimensional displays are more commonly used.

For example, holographic and lenslet displays provide visibility of apparently three- dimensional content. However, these systems typically rely on a two-dimensional display modulator, thus constraining the visibility of the content to the volume between the observer and the display surface (i.e. direct line of sight).

Other three-dimensional ‘volumetric’ display approaches have been developed based on light scattering, emitting or absorbing surfaces. For instance, volumetric display devices offering unconstrained three-dimensional visibility around the display have been implemented using various techniques including, but not limited to, rotating surfaces, plasmonics, air displays, and photophoretic traps.

However, these approaches are of course limited to displaying visual content and only offer a limited range of operating modes. Thus, it is desired to provide improved techniques for manipulating particles that can in turn be used for various applications that may benefit from an improved control of particles including, but not limited to, improved three-dimensional display devices, as well as improved manufacturing techniques, and any other such applications.

SUMMARY

According to a first aspect, there is provided an apparatus comprising: a set of one or more acoustic source(s) controllable to generate an acoustic field within a volume of the apparatus; and a control circuit that is configured to control the set of one or more acoustic source(s) to generate an acoustic field defining one or more acoustic trap(s) for suspending one or more particle(s) within the volume of the apparatus, wherein the control circuit is further configured to update the acoustic field to change the position of one or more of the acoustic trap(s) within the volume of the apparatus such that for at least a first acoustic trap of the one or more acoustic trap(s) the position of the first acoustic trap is moved to thereby cause a particle suspended within the first acoustic trap to move around the volume of the apparatus (a ‘movement event’), and wherein during a movement event wherein the position of the first acoustic trap is being moved to thereby cause a particle suspended within the first acoustic trap to move around the volume of the apparatus the control circuit is configured to update the acoustic field at such a rate that the particle is prevented from reaching a static equilibrium position within the first acoustic trap as the acoustic field is updated multiple times to change the position of the first acoustic trap within the volume of the apparatus.

According to a second aspect, there is provided a method of manipulating a particle within a volume, the method comprising: generating using a set of one or more acoustic source(s) an acoustic field defining one or more acoustic trap(s) for suspending one or more particle(s) within the volume; and updating the acoustic field to change the position of the acoustic trap(s) within the volume of the apparatus such that for at least a first acoustic trap of the one or more acoustic trap(s) the position of the first acoustic trap is moved to thereby cause a particle suspended within the first acoustic trap to move around the volume, wherein during a movement event wherein the position of the first acoustic trap is being moved to thereby cause the particle suspended within the first acoustic trap to move around the volume of the apparatus the acoustic field is updated at such a rate that the particle is prevented from reaching a static equilibrium position within the first acoustic trap as the acoustic field is updated multiple times to change the position of the first acoustic trap within the volume of the apparatus.

Thus, according to embodiments of the present invention a particle is acoustically suspended, or ‘levitated’, within the confines of a certain volume using a suitable acoustic trapping field, i.e. acoustic trap. By changing the position of the acoustic trap within the volume, e.g. by suitably updating and re-configuring the acoustic field defining the acoustic trap, it is thus possible to move the particle around the volume. Conventionally it has been thought that in order for a particle to be reliably levitated using such an acoustic trap, i.e. without dropping due to gravity, such that the particle remains suspended as the acoustic trap is being moved, the particle must be allowed to reach a static equilibrium position, e.g., at the center of the acoustic trap, each time the acoustic field is updated to move the position of the acoustic trap. Thus, in conventional acoustic levitation techniques, each time the acoustic field is updated to move the position of the acoustic trap, the particle is allowed to take up a static equilibrium position (e.g. at the center of the trap) before the trap is then moved again and the particle is allowed to stabilise at a new equilibrium position, and so on.

This means that in conventional acoustic levitation techniques the particle can only be moved in a series of discrete steps and the rate at which the acoustic trap (and hence the particle) can be moved is therefore limited based on the time required for the particle to reach equilibrium after each update. This approach can nonetheless work well for some applications where a slow (precise) control is desired.

However, the Applicants recognise for many applications it would be desirable to be able to move the particles at increased speeds. Further, increased particle speeds would open up new potential applications for acoustic levitation.

According to embodiments of the present invention the acoustic field is updated to change the position of the (first) acoustic trap at such a rate that the particle is prevented from reaching a static equilibrium in the acoustic trap as the acoustic trap is being moved around the volume.

Thus, in embodiments, the system comprises a control circuit that is configured to control the set of one or more acoustic source(s) to generate an acoustic field defining an acoustic trap for suspending a particle within the volume of the apparatus, wherein the control circuit is further configured to update the acoustic field to change the position of the acoustic trap within the volume of the apparatus and to thereby cause the particle to move around the volume of the apparatus, and wherein during a movement event where the particle is being moved around the volume of the apparatus the control circuit is configured to update the acoustic field at such a rate that the particle is prevented from reaching a static equilibrium position within the acoustic trap as the acoustic field is updated multiple times to change the position of the acoustic trap within the volume of the apparatus.

Correspondingly, the method comprises generating using a set of one or more acoustic source(s) an acoustic field defining an acoustic trap for suspending a particle within the volume; and updating the acoustic field to change the position of the acoustic trap within the volume of the apparatus and to thereby cause the particle to move around the volume, wherein during a movement event where the particle is being moved around the volume of the apparatus the acoustic field is updated at such a rate that the particle is prevented from reaching a static equilibrium position within the acoustic trap as the acoustic field is updated multiple times to change the position of the acoustic trap within the volume of the apparatus.

The Applicants have found that in this way it is possible to control the position of the acoustic trap to still keep the particle suspended whilst allowing a more continuous and faster movement of the particle. For instance, the effect of this is that according to embodiments of the present invention the particle may be substantially constantly kept in motion as the position of the acoustic trap is updated and moved around the volume of the apparatus. This means that the particle is able to retain its momentum and hence accumulate speed between updates.

By contrast, in the more conventional approach described above, because the particle stabilises at zero velocity at a fixed equilibrium position (e.g.) at the center of the acoustic trap after each update, the particle’s momentum is therefore lost after each update. By retaining/increasing momentum between updates, embodiments of the present invention thus allow much higher particle speeds to be achieved than in the more conventional approach.

The techniques described herein may also provide a smoother particle motion. For example, when the particle is allowed to equilibrate after each update, as in more conventional approaches, it will be appreciated that when the particle is allowed to equilibrate after each update, the particle will typically initially overshoot the equilibrium position and then oscillate about the equilibrium position until its motion is stabilised. This may therefore lead to uneven accelerations, which can be (and preferably are) avoided according to embodiments of the present invention.

In particular, in embodiments of the present invention this may be achieved by updating the acoustic field at such a rate to allow the acoustic trap to be repositioned before the particle has reached the static equilibrium position, e.g., in the center of the acoustic trap.

As the position of the acoustic trap is moved around the volume, the particle may therefore be caused to continue to accelerate towards its equilibrium position, e.g., at the center of the acoustic trap, but the center of the acoustic trap will keep moving as the acoustic field is updated. The particle is thereby prevented from reaching its static equilibrium position at the center of the acoustic trap, and is instead preferably caused to continue to accelerate towards the (updated) center of the acoustic trap.

In this way the speed at which the particle can be moved around the volume may be significantly increased since there is no longer any requirement to allow the particle to equilibrate each time the acoustic trap is moved, which not only allows the acoustic trap to be moved more rapidly, but also allows the particle to be continuously accelerated as the acoustic trap is moved. Thus, it is possible to provide a faster and more continuous movement of the particle.

For instance, embodiments of the present invention allow particle speeds (e.g. in air) of greater than 1 m/s. In preferred embodiments, the particle speed in air may be greater than 2 m/s, greater than 5 m/s, or even greater than 8 m/s. However, it is believed that even greater speeds could be achieved using the techniques described herein. The maximum particle speed may depend, among other things, on the properties of the medium within the volume within which the particle is moved, the rate at which the acoustic field used to generate the acoustic trap is updated, and so on. The maximum particle speed may also depend on the arrangement of the acoustic source(s) and may be anisotropic. For example, the maximum vertical particle speed may be higher than the maximum horizontal particle speed, or vice versa, depending on the arrangement of the acoustic source(s). Regardless, it will be appreciated that the achievable particle speeds according to embodiments of the present invention are significantly higher than those used for conventional acoustic levitation systems, which are typically limited to maximum particle speeds significantly below 1 m/s.

Correspondingly, the techniques described herein also allow for an improved control over decelerations of the particle, for instance, when the particle approaches a corner or a part of the path where lower speeds are required. In this case, the position of the trap may be kept at a certain distance from the particle as to apply substantially optimal (e.g. higher) deceleration forces.

Thus, in embodiments, the particle is kept away from the position of the (e.g.) center of the acoustic trap such that an acceleration (or deceleration) force is preferably continually provided to the particle as the acoustic trap is moved. That is, whilst the acoustic trap is being moved, the particle may be maintained in a dynamic state where the particle is kept (at least) a certain distance from the equilibrium position (e.g. the center) of the acoustic trap.

That is, in embodiments, the control circuit is configured to update the position of (i.e. to move) the acoustic trap at such a rate that the particle is kept at least at a certain distance from the equilibrium position within the acoustic trap whilst the acoustic trap is being moved around the volume of the apparatus. As the acoustic field is updated multiple times to move the position of the acoustic trap, the position of the acoustic trap after each update may thus be selected so as to keep the particle a certain distance from the equilibrium position (e.g. at the center of the acoustic trap), and the control circuit can be configured to re-compute the desired acoustic field with each update in order control the position of the acoustic trap accordingly. In this way, the particle may be kept at a substantially optimal distance from the equilibrium position, e.g. such that a substantially constant acceleration (or deceleration) force is provided to the particle. The control circuit may thus be configured to determine where the acoustic trap should be positioned in order to prevent the particle reaching its static equilibrium and thereby maintain the particle in motion (e.g. according to optimum accelerations/decelerations required), and to update the acoustic field accordingly.

Although the particle is prevented from reaching static equilibrium as it is moved around the volume, it will of course be appreciated that the particle may still reach a static equilibrium position in the acoustic trap between movement events, i.e. when the acoustic trap is stationary (and remains stationary for a substantial period of time, e.g. over multiple updates of the acoustic field). For example, after the particle has been moved to a desired position within the volume of the apparatus, the particle may then be held in the acoustic trap at that position, and allowed to reach static equilibrium, until it is desired to move the particle again (another movement event), at which point the acoustic trap is then moved in the manner described above such that particle is kept in motion.

Another case where the particle might (temporarily) reach the equilibrium position is when changing between acceleration and deceleration events (or vice versa), e.g. when causing the particle to turn around, or turn a corner. For instance, when it is desired to start decelerating a particle, the particle may be allowed to initially go past the equilibrium position. In this case, however, the particle does not stabilise in the equilibrium position but merely passes through it. However, during a movement event the particle may be moved along a selected path from a starting (first) position within the volume to a selected end (second) position without reaching static equilibrium. The selected path may thus be programmed into the control circuit, and the control circuit then configured to update the acoustic field over time to progressively move the acoustic trap to cause the particle to follow the selected path.

It will be appreciated that ‘moving’ the acoustic trap generally means updating and adjusting the form of the associated acoustic field such that the position of the acoustic trap is changed. The position of the acoustic trap may thus in effect be constantly re-calculated over time with each update of the acoustic field. The acoustic field is therefore typically updated multiple times during a single movement event to move the acoustic trap appropriately to cause the particle to move along the selected path. In embodiments, the particle is therefore prevented from reaching static equilibrium over multiple cycles of updating the acoustic field.

For example, in embodiments, the acoustic field may be updated with an update rate of at least 2500 updates/second (i.e. a cycle time of 400 microseconds). It has been found that relatively higher update rates may allow for the optimal (i.e. highest) particle speeds to be achieved, and may also allow for an extended range of operations. Thus, in preferred embodiments, update rates of at least 5000 updates/second (a cycle time of 200 microseconds) may be used. In some embodiments, an update rate of at least 10000 updates/second (a cycle time of 100 microseconds), or at least 20000 updates/second (a cycle time of 50 microseconds), may be used. For example, update rates of 40000 updates/second (a cycle time of 25 microseconds) have been found to work well for a single particle trap. For more complex (e.g. multi-particle) trapping fields, it may be necessary to use slower update rates, e.g. depending on the complexity of the field and/or the available processing resource. For example, update rates of about 16000 updates/second (a cycle time of about 66 microseconds) have been demonstrated with a multi-particle trapping field.

It will be appreciated that relatively faster update rates may help ensure that the acoustic trap can be moved sufficiently rapidly to prevent the particle reaching static equilibrium, in the manner described above. Nonetheless, it will be appreciated that slower update rates could in principle be used, e.g. depending on the medium and the particle, whilst still achieving the desired effects.

It will be appreciated that this means that the particle may typically be kept in motion and prevented from reaching a static equilibrium position for time periods significantly greater (e.g. orders of magnitude higher) than this cycle time. For instance, in embodiments, the particle may be kept in motion and prevented from reaching a static equilibrium position for time periods greater than 1 second, or greater than 5 seconds, or 10 seconds, or even longer, e.g. depending on the desired application.

Thus, in embodiments, the acoustic field may be updated to change the position of the acoustic trap at such a rate that the particle is moved substantially continuously (without stopping) along a selected path from a first point within the volume to a second point. This is in contrast to a more conventional approach, as discussed above, where the particle is allowed to re-equilibrate after each update and hence moves in a series of discrete steps. The approach described herein wherein a particle is manipulated using an acoustic trapping field is generally scalable to any volume, as desired, e.g. by providing suitable acoustic source(s). However, depending on the application, a typical apparatus volume might be of the order 5 cm x 5 cm x 5 cm, or 10 cm x 10 cm x 10 cm. Of course the apparatus volume need not be cuboidal and any other suitable arrangements are possible in this regard.

During a movement event, when the particle is being moved along a selected path within the volume of the apparatus, the particle may therefore in embodiments travel over a distance of greater than about: (i) 1 cm; (ii) 5 cm; or (iii) 10 cm without reaching static equilibrium, e.g. depending on the selected path and the desired application.

The form of the acoustic trap (or acoustic traps) defined by the acoustic field generated by the acoustic source(s) may be selected, as desired, to achieve the desired acoustic levitation of the particle. For instance, in preferred embodiments, the acoustic trap that is used to levitate and manipulate the particle may comprise a single “twin trap”, for which the desired acoustic field can readily (and rapidly) be calculated, and re-calculated, in order to update the position of the acoustic trap (i.e. to move the acoustic trap). However, various other trap morphologies exist, including vortex traps, bottle beams, and so on, and in principle the trap may comprise any form of acoustic trap that may suitably be used for the purposes of the present invention.

In general the form of the acoustic field that is generated by the acoustic source(s) may be controlled, i.e. to define and move the acoustic trap(s), by suitably controlling a spatial distribution of phase delays and/or amplitudes for the acoustic waves generated by the set of one or more acoustic source(s). For example, the set of one or more acoustic source(s) may comprise one or more array(s) of acoustic sources (e.g. an array of transducers) which may be controlled according to a certain spatial distribution of phase and/or amplitude values to define the desired acoustic field. (However, in preferred embodiments the acoustic trap that is used for levitating and moving the particle is defined through phase modulation (alone), for reasons that will be explained below.)

In preferred embodiments the form of the acoustic field required to define the acoustic trap(s) at the desired position(s) is computed using a local control circuit, for example, at hardware level on a suitable Field Programmable Gate Array (FPGA) or GPU. Preferably, the calculation of the required acoustic trapping field is therefore embedded into hardware associated with the apparatus. This may help facilitate re-calculating the acoustic field at the desired update rates. In that case, the desired acoustic field can thus be rapidly and easily reprogrammed in order to move the acoustic trap(s) and this may therefore help to reduce any processing lag associated with generating and moving the acoustic trap e.g. that might occur when using a remote personal computer (PC) to control the generating of the acoustic trap as is typically the case in more conventional systems. It will be appreciated that in more conventional systems this may typically be less of a problem since the update speed of the device is in any case limited by the requirement for the particles to reach static equilibrium and the devices are therefore typically designed for relatively slow (precise) applications.

According to preferred embodiments of the present invention, the speed at which the acoustic trap(s) can be moved may thus be essentially limited only by the rate at which the acoustic field generated by the acoustic source(s) can be updated. The particle speed is then determined by the update rate for the acoustic field (i.e. the speed at which the acoustic trap can be moved) and also the acceleration forces provided by the acoustic field (e.g. due to acoustic pressure gradients causing the particle to accelerate towards the equilibrium position (e.g. center) of the acoustic trap).

In preferred embodiments the set of acoustic source(s) may comprise one or more array(s) of transducers. In that case, the acoustic trapping field may be defined by controlling the (relative) phase values introduced at different positions within the array of transducers. The acoustic field may also be controlled by controlling the (relative) amplitude values at different positions within the array of transducers. Typical commercially available transducers may be capable of switching at rate 40 kHz (i.e. for 40 kHz ultrasonic operation). Operating at 40 kHz has been found to provide a good balance between using relatively low-cost commercially available transducers whilst still being able to provide the desired range of operations of the apparatus, as will be explained further below. However, faster (or slower) switching devices are also available and may also of course be used, as desired, e.g. depending on the desired operating frequency. For example, in embodiments, an operating frequency of greater than 40 kHz, such as about 80 kHz may be used, to further enhance the range of operations. Similarly, the present invention is not limited to the use of such phased arrays of transducers and other suitable arrangements for generating the desired acoustic fields would of course be possible.

In embodiments the operating frequency of the acoustic source(s) is preferably at least 10 kHz. Preferably the operating frequency of the acoustic source(s) is at least 20 kHz, and most preferably at least 40 kHz, or 80 kHz, or even higher. In general the operating frequency of the acoustic source(s) may be selected as desired. However, working at higher operating frequencies may facilitate improved update rates (since the rate at which the acoustic field can be updated may otherwise be limited by the operating frequency of the acoustic source(s)). Further, using higher operating frequencies may help shift certain harmonic artefacts resulting from switching/modulation of the acoustic field outside the audible range, or at least towards the upper end of the audible range.

The set of acoustic source(s) may generally be disposed around the volume of the apparatus in any suitable fashion, as desired, e.g. so long as the set of acoustic source(s) is capable of generating the required acoustic fields. For instance, in embodiments, the apparatus volume is defined between two vertically spaced-apart arrays of transducers. However, other arrangements would be possible.

Because the particle can be moved at higher speeds this opens up various (novel) possibilities for using the apparatus.

For instance, utilising the techniques described herein, it is now possible to move a particle at such speeds that “persistence-of-vision” phenomena can be exploited, e.g. in order to provide a volumetric (three-dimensional) display device. By suitably illuminating the particle as the particle is moved around the volume of the display device at such speeds, it is thus possible to generate semi-persistent, or even fully rasterised, visual content. For example, if the particle is moved along the selected path to trace a desired shape within a time period of less than about 0.1 seconds, the human eye will then perceive the shape. This is possible using the techniques described herein.

Thus, the particle may be moved around the volume, at such speeds, along a selected path in order to trace out, or raster, a desired pattern to generate visual content. The movement of the particle may thus be controlled to generate visual content (e.g. an image), as desired. According to a first main application, the apparatus may therefore comprise a display device.

It will be appreciated that using the techniques described herein the particle can be manipulated in three dimensions around the volume of the apparatus. Thus, preferred embodiments provide a volumetric (three-dimensional) display device. Correspondingly, there is also provided a method of generating a (volumetric) display, using the techniques described herein. However, the apparatus is of course also capable of generating two- dimensional content, e.g. by constraining the particle motion within a plane.

In such embodiments, where the apparatus comprises a display device, the apparatus may further comprise one or more light source(s) for illuminating the particle. The light source(s) may comprise any suitable light source. For example, the light source may comprise a set of one or more LED(s), or an LED light tile. In embodiments, the light source may provide a substantially uniform illumination. Alternatively, a directional light source (or a set of directional light sources) may be used. The light source(s) may be controlled by the control circuit in order to control the visual output. For example, the colour (e.g. RGB) and/or timing of the light source(s) may be controlled over time, e.g. according to a desired light sequence, to selectively illuminate the particle in order to control the visual display output.

The adjustment of the acoustic field causing the movement of the particle, preferably in conjunction with suitable illumination of the particle, can thus be used to control the visual content that is generated by the volumetric display device. However, the acoustophoretic operating principles used to control the movement of the particle also naturally open up the possibility for providing audio/tactile content using the same acoustic source(s).

That is, according to embodiments of the present invention a single set of one or more acoustic source(s) (e.g. arrays of transducers) is used to provide a range of different types of outputs from the apparatus. In preferred embodiments, the display apparatus may therefore comprise a “multimodal” display apparatus that is capable of providing a combination of visual, audio and/or tactile content. Such a display apparatus may find utility in a range of applications including for providing digital signage, haptic interfaces, consumer electronic devices, and so on.

In embodiments the multimodal display apparatus may be operated to alternately generate one of visual, audio or tactile content (only). However, a benefit of the techniques described herein is that a multimodal display apparatus can be realised that is capable of simultaneously providing any of visual, audio and/or tactile content.

For instance, in preferred embodiments, and as mentioned above, the acoustic trap that is used for suspending and moving the particle is defined by a phase modulation of the acoustic waves generated by the set of acoustic source(s) (only). The visual content output can thus be controlled by moving the position of the acoustic trap in the manner described above by adjusting a spatial distribution of phase values associated with the set of acoustic source(s) (e.g. by adjusting a spatial distribution of relative phase values across an array of transducers) accordingly.

This means that the amplitude of the acoustic field is then free to be adjusted as desired, and the amplitude of the acoustic field (which is the same acoustic field used to define the acoustic trap that controls the visual content) can therefore be modulated to provide simultaneous audio content, for example. In particular, because the acoustic trap used for controlling the movement of the particle is defined by phase modulation (only), the amplitude of the acoustic field can be changed independently of this and without substantially changing the shape (or position) of the trap.

It will be appreciated that operation at a frequency of 40 kHz covers approximately the entire audible spectrum. Correspondingly, operation above 40 kHz, e.g. at 80 kHz, can cover the entire audible spectrum and beyond. Thus, the relatively higher update rates preferably utilised by embodiments of the invention also facilitate the generation of essentially arbitrary audio content, as desired, using the same set of acoustic source(s) that are used to control the visual content, e.g. through suitable sideband modulation of the acoustic field.

Embodiments also allow the possibility for the generation of time-multiplexed acoustic fields. For instance, the set of acoustic source(s) may be operated in a time-multiplexed fashion wherein two or more different acoustic fields are generated in a temporally interleaved manner. For example, a first portion of the duty cycle may be used for generating a first acoustic field defining the acoustic trap that controls the motion of a particle, in the manner described above. A second (or further) portion of the duty cycle may be used for generating a second (or further) acoustic field (or acoustic fields) that can be temporally interleaved with the first acoustic field, e.g. and then used for any desired purpose. Again, it will be appreciated that operating with relatively higher update rates (preferably greater than 10000 updates/second, and more preferably greater than 20000 updates/second, or 40000 updates/second) for the acoustic field may help to ensure that the particle can be effectively levitated even when only a portion of the duty cycle is used for this purpose.

The ability to temporally interleave different acoustic fields in this manner opens up a range of further operating possibilities.

For example, first and second acoustic traps may be temporally interleaved with one another to create two (or more) temporally modulated acoustic traps the spatial positions of which can be independently controlled. In this way the apparatus can be extended to simultaneously trap two (or more) different particles. Thus, whilst embodiments are described for ease of reference to levitating a single particle it will be appreciated that the apparatus may also be used to simultaneously levitate a number of particles.

Thus, in some preferred embodiments, only a single particle is trapped in the display device at any time. This may help simplify the calculations of the required field and thus facilitate the faster update rates described herein. However, it is also contemplated that more than one particle may be trapped in the display device at any time, e.g. in order to generate larger and/or more complex visual content. As discussed above, this may be achieved using a corresponding set of time-multiplexed acoustic traps. This could also be achieved, and in other embodiments is achieved, by generating a more complex trapping field defining a plurality of acoustic traps for simultaneously levitating a corresponding plurality of particles. It will be appreciated that this may generally involve more complex processing by the control circuit to maintain the trapping field (e.g. compared to time-multiplexing the acoustic fields). However, both approaches may work well, e.g. depending on the application.

This of course applies more generally to the present invention according to any of its aspects and embodiments, regardless of whether the techniques are used as part of a display apparatus. Thus, whilst various embodiments are described herein for ease of understanding in relation to manipulation of a single particle (using a single acoustic trap), in general the apparatuses and methods described herein may be used to simultaneously suspend more than one particle (such as two, three, four, five, six, or more particles). Thus, any references herein to an apparatus or method that is used to manipulate a particle may generally be understood to refer to one or more particle(s). Similarly, any references to an acoustic trap may in embodiments refer to one or more acoustic trap(s) for suspending a one or more particle(s). Thus, generally, the acoustic field may define one or more acoustic trap(s) for suspending one or more particle(s). Each acoustic trap, or trapping region, may be used to suspend a corresponding particle. However, it is also contemplated that less than all of the acoustic traps, where multiple acoustic traps are provided, may be used to suspend particles. That is, there may be a greater number of acoustic traps than particles being suspended. Equally, it would in principle also be possible to suspend multiple particles in a single acoustic trap, e.g. depending on the form of the trapping field.

As another example, the second acoustic field may be used to provide tactile content. For instance, the second acoustic field may be used to create a second acoustic trap that is used to provide a tactile sensation. For example, the second acoustic field may define a focussing point that a user can then interact with in order to receive tactile feedback. For instance, in embodiments a relatively low frequency (e.g. about 250 Hz) side band amplitude modulation may be applied to the second acoustic trap to provide the tactile content. However it will be appreciated that various other approaches for generating tactile content are known that may suitably be used. Preferably the focussing point is defined away from the position of the first acoustic trap (and hence the particle) to prevent interference with the visual content.

When both audio and tactile content is to be provided this could be done by multiplexing the audio and tactile modulation signals in a similar manner as described above such that a portion of the duty cycle is used for audio content and another portion of the duty cycle for tactile content. However, preferably, the signals for the audio and tactile content are combined into a single signal that is used to modulate the amplitude of the acoustic field to simultaneously provide both the audio and tactile content. This advantageously removes any artefacts associated with switching between audio and tactile content.

It will be appreciated that the display apparatus may therefore be capable of generating any combination of visual, audio and/or tactile content using the same set of acoustic source(s), e.g. through appropriate control of the acoustic trapping field (e.g. phase modulation) to generate the desired visual content in conjunction with time-multiplexing of multiple trapping fields and/or amplitude modulation for simultaneous generation of audio and/or tactile content.

Thus in embodiments, the control circuit may be configured to control the set of one or more acoustic source(s) to generate first and second temporally interleaved acoustic fields, wherein the first acoustic field defines the acoustic trap for suspending and moving the particle within the volume of the apparatus and wherein the second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output. Correspondingly, the method may involve generating first and second temporally interleaved acoustic fields, wherein the first acoustic field defines the acoustic trap for suspending and moving the particle within the volume of the apparatus and wherein the second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output.

It is believed that this method of time-multiplexing acoustic fields may also be novel and inventive in its own right.

Thus, from a further aspect there is provided an apparatus comprising: a set of one or more acoustic source(s) controllable to generate an acoustic field within a volume of the apparatus; and a control circuit that is configured to control the set of one or more acoustic source(s) to generate first and second temporally interleaved acoustic fields, wherein the first acoustic field defines a first acoustic trap for suspending a particle within the volume of the apparatus; and wherein the second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output.

There is also provided a method comprising: generating first and second temporally interleaved acoustic fields, wherein the first acoustic field defines the acoustic trap for suspending and moving the particle within the volume of the apparatus and wherein the second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output.

When generating a set of time-multiplexed fields it will be appreciated that the multiplexing rate may introduce artefacts, e.g. at harmonics of the multiplexing frequency. Operating at relatively higher update rates (preferably above 20000 updates/second, or even 40000 updates/second, or higher) may help to shift harmonics outside of, or at least to the upper part of, the audible range thereby enhancing the quality of the audio content generated in this way, e.g. by removing certain artefacts associated with the switching of the acoustic source(s).

The volume of the apparatus may be filled with any suitable medium. In preferred embodiments, the volume is filled with the ambient medium, which is typically air, but in some embodiments could also be, e.g., water. However, it would also be possible to fill the volume of the apparatus with a different medium, if desired, to potentially further control the output. The choice of medium may impact the achievable speeds and the required update rates to keep the particle from reaching static equilibrium, e.g. due to viscous drag effects.

The particle may generally be selected based on its size and/or density to ensure that the particle can be appropriately levitated by the acoustic fields. When the device is used to provide a volumetric display, the particle may further be selected based on its ability to couple appropriately to the light.

As a general guide, the size of the particle should be (and in preferred embodiments therefore is) less than about a half of the operating wavelength with which the acoustic field that provides the acoustic trap is generated. For operational frequencies of 40 kHz, this suggests an upper limit of about 4 mm in air. Typically, the particle may therefore have a size of between about 1 to 2 mm (about a quarter of the operating wavelength at 40 kHz). However, the techniques described herein are also scalable to larger particle sizes, if desired, especially when operating at lower frequencies. In general the size (and shape) of the particle may impact the particle speed, e.g. due to differences in weight and drag effects.

The particle should also be light enough to be effectively levitated. Various suitable materials are known in this regard. For example, in one preferred embodiment of a display device, the particle may comprise an expanded polystyrene (EPS) bead. This beneficially has an appropriate density/size for acoustic levitation and also provides a suitable coupling to the light source(s). In particular such a particle may provide an approximation of a Lambertian surface having substantially the same brightness regardless of the viewing direction. It is also contemplated that the particle may be coated with a suitable reflective material. This may be used to provide high reflection in all directions, or to provide a directional reflectance, if desired.

However, it will be appreciated that the particle may take other forms, as desired. For instance, as another example the particle may comprise cubic zirconia, or other suitable such materials. It will be also be appreciated that the particle need not be solid. For example, the particle may equally comprise a droplet of liquid, such as water or a molten thermoplastic material.

Indeed, it will be appreciated that the present invention may find utility for various other highspeed, non-contact applications where it is desired to manipulate particles of various forms.

For example in another preferred application, the apparatus comprises an additive manufacturing apparatus, e.g. in the form of a three-dimensional (3D) printer. Additive manufacturing (e.g. 3D printing) is generally carried out through a selective deposition and/or solidification of material with blocks of typically liquid or semi-liquid material being moved into a desired position and then deposited and solidified accordingly in order to build up a three-dimensional output, a ‘model’, usually in plural build layers.

In that case, the particle that is levitated may comprise a particle (e.g. a droplet) of the material that is being used for the additive manufacturing process. For example, in some embodiments, the particle may comprise a droplet of suitable thermoplastic material that is being used by the additive manufacturing apparatus. Other examples of materials that may suitably be used for such purposes, and which may therefore be used according to various embodiments, include various other polymers, plastics, metals, stereolithographic materials, waxes, and so on, depending on the desired application, e.g. on the model that is being manufactured. Indeed, a general advantage of such additive manufacturing techniques is that they can be readily used to manufacture a wide range of parts, or components for a range of different applications. For example, such techniques may be used for manufacturing models over a range of characteristic length scales. The model that is built may thus take a wide range of forms ranging from items of clothing, medical equipment, prosthetics, circuit boards, etc., and the present techniques can be used for any such application in which additive manufacturing may find utility.

For instance, the use of acoustic levitation in this context provides a useful non-contact approach for handling the materials that are used to build the model. In this context, the improved particle speeds that can be achieved according to the present invention may further allow for an increased print rate.

Thus, from another aspect there is provided an additive manufacturing apparatus comprising an apparatus as described herein in conjunction with one or more material supply(supplies) that are configured to introduce particles of material into the volume of the apparatus. Correspondingly, there is provided a method of additive manufacturing utilising the techniques described herein, wherein the particle comprises an ink droplet received at a first position from a first ink source, and wherein the method comprises moving the particle from the first position to a second position, and depositing the particle onto a substrate in order to build an output model.

The additive manufacturing apparatus may thus comprise one or more material supply(supplies). Each material supply preferably comprises a capillary that is used to introduce particles (e.g. droplets) of material into the volume of the apparatus. For instance, a single particle is preferably introduced from the capillary into the acoustic trap at a first position. The position of the acoustic trap is then updated to thereby move the particle from its initial (first) position to a desired position on the substrate where the particle can then be deposited onto a substrate (i.e. by removing the acoustic levitation field) in order to build the model (the printed output).

This approach can also be used to safely handle multiple different particles from different sources (e.g. which may be different coloured materials, or different types of material), e.g. in order to be able to manufacture more complex models. Again, this could be done either using more complex acoustic fields defining a plurality of acoustic trapping regions for suspending a corresponding plurality of particles or by using a plurality of time-multiplexed acoustic traps.

Another benefit of using the present invention in the context of an additive manufacturing apparatus is that because the position and movement of the particles can be controlled through phase modulation of the acoustic field alone, the amplitude is free (as discussed above). The amplitude of the acoustic field can then be used, e.g., to modify the shape of the particles as they are moved into position. This allows new functionality that goes beyond that of traditional three-dimensional printers. For instance, in embodiments a particle may be flattened or otherwise re-shaped before it is deposited onto the substrate.

This may be achieved by moving the particle into position, and then allowing particle to reach equilibrium within the acoustic trap, before adjusting the amplitude to change the shape of the particle as desired. However, the present techniques also allow this adjustment to be performed as the particle is being moved into position, e.g. to provide an increased print rate. For instance, the control circuit generally has knowledge of where the particle is to be deposited, and can calculate how long it will take to move the particle to that position, and so the control circuit is able to calculate the required acoustic fields to ensure that the particle is appropriately shaped when it arrives at the specified position on the substrate.

Again, this is achievable using the high update rates described herein. For example, by suitably adjusting the amplitude between updates it is possible to continue to change the dynamics around the liquid to create different oscillation modes and shapes.

When used as an additive manufacturing apparatus it will be appreciated that the model that is being built (i.e. ‘printed’) may impact the acoustic fields, e.g. if the model is also positioned within the volume of the apparatus. This may be mitigated by simply arranging the acoustic source(s) such that the model does not interfere with the acoustic fields. For example, the particles may be dropped out of the volume of the apparatus onto the model which is located outside of the volume of the apparatus. However, this need not be the case, and the model could also be located within the volume of the apparatus. In that case, the control circuit may take into account the presence of the model when generating the acoustic fields. For instance, the control circuit will generally know the expected form of the model at any moment in time based on the blueprint for the model, and can thus calculate the acoustic field accordingly to take this into account.

Indeed, an advantage of the present approach when used in this context is that the apparatus can be implemented within existing additive manufacturing pipelines by suitably programming the control circuit based on the blueprints (layer by layer) for the model. For example, the control circuit may be arranged to control the movement of particles according to a suitable “G-code” containing the instructions for controlling the apparatus to print a certain model.

The methods in accordance with the technology described herein may be implemented at least partially using software e.g. computer programs. It will thus be seen that when viewed from further embodiments the technology described herein comprises computer software specifically adapted to carry out the methods herein described when installed on a data processor, a computer program element comprising computer software code portions for performing the methods herein described when the program element is run on a data processor, and a computer program comprising code adapted to perform all the steps of a method or of the methods herein described when the program is run on a data processor. The data processor may be a microprocessor system, a programmable FPGA, etc.. Thus, whilst in preferred embodiments the methods are implemented on an FPGA, various other arrangements would of course be possible.

Various other applications for the present invention are also contemplated. For instance, the embodiments described above may also find utility for other fabrication techniques, such as semiconductor fabrication, for microfluidic experiments, “lab-on-a-chip” applications, and any other similar areas where it may be desired to manipulate particles.

BRIEF DESCRIPTION OF THE DRAWINGS

Various embodiments will now be described, by way of example only, and with reference to the accompanying drawings in which: FIGURE 1 shows an example of a multimodal acoustic trap display (MATD) device according to an embodiment;

FIGURE 2 shows examples of visual content generated according to the MATD device according to various embodiments;

FIGURES 3A, 3B and 3C illustrate details of transducer control for operating the MATD device;

FIGURES 4A and 4B illustrate the generation of audio content using the MATD device according to various embodiments;

FIGURES 5A and 5B illustrate the generation of tactile content using the MATD device according to various embodiments;

FIGURES 6A-6D shows a spectral analysis of the audio response in the MATD device; FIGURES 7A-7C show the results of testing of the particle speeds in the MATD device; FIGURE 8 summarises the speed testing in FIGURES 7A-7C;

FIGURE 9 shows how the maximum particle speed may vary with the rate at which the acoustic field is updated;

FIGURE 10A illustrates the acoustic pressure forces due to a twin trap potential;

FIGURE 10B the operating principles underlying embodiments of the present invention wherein a particle is kept in a dynamic state;

FIGURE 11 shows how liquid droplets can be manipulated and re-shaped using acoustic pressure forces according to embodiments; and

FIGURE 12 shows an example of an additive manufacturing apparatus (e.g. a three- dimensional (3D) printer) according to an embodiment.

DETAILED DESCRIPTION

The concepts described herein relate generally to novel approaches for manipulating particles using acoustic fields. To illustrate the concepts described herein a first preferred implementation will now be described with respect to a multimodal acoustic trap display (MATD) device, i.e. a mid-air volumetric display that is capable of simultaneously delivering any combination of visual, audio, and tactile content, using acoustic wave manipulation as the single operating principle. However, it will be appreciated that the concepts described herein may generally find utility in a range of different applications and are not limited to providing such display devices. The MATD device according to the present embodiment acoustically traps a particle and illuminates the particle as the particle is quickly scanned through the display volume, thereby creating a three-dimensional volumetric display by exploiting “persistence-of-vision” (POV) effects. For example, the particle may be selectively illuminated with red, green and blue light to control the colour of the particle as it is moved along a selected path to trace out a desired shape and thereby provide semi-persistent visual content.

As will be explained further below, the MATD device can also simultaneously provide tactile stimulation and audio content through the use of secondary acoustic traps and/or amplitude modulation of the acoustic field. Indeed, the control techniques described in this application (e.g. phase optimization, particle dynamics prediction, illumination control) also allow substantially optimal simultaneous integration of the visual, audio and tactile modalities supported. The MATD device can thus be operated in a range of different modes, e.g. wherein any combination of visual, audio and tactile content is provided.

FIGURE 1 thus shows an example of a multimodal acoustic trap display (MATD) device 10 according to an embodiment. The MATD device 10 of FIGURE 1 comprises two opposed arrays of acoustic transducers 14 that are spaced-apart in the vertical direction of the device. A display volume is therefore defined between the two arrays within which an acoustic field can be generated.

In particular the arrays of acoustic transducers 14 are used to generate a first acoustic trapping field that is used to levitate a particle 18. Although only a single particle 18 is shown in FIGURE 1, it will be appreciated that a more complex acoustic trapping field could also be used in order to simultaneously levitate a plurality of particles. Thus, it will be understood that the MATD device 10 may control a plurality of particles. However, for ease of explanation, various embodiments will now be described in relation to a single particle 18.

The MATD device 10 also includes an illumination module 16 comprising a set of high intensity RGB LEDs. In FIGURE 1 a single illumination module 16 is provided towards the top right of the MATD device 10. However, it will be appreciated that this need not be the case and any suitable arrangement of light sources may be used to illuminate the particle.

The device shown in FIGURE 1 is thus operable to acoustically trap a particle 18 and selectively illuminate it with a sequence of red, green, and blue light to control its colour as the particle 18 quickly scans through the display volume. The particle 18 can then be moved around the volume along a selected path in order to trace out a desired pattern.

It will be appreciated that the MATD device 10 shown in FIGURE 1 is based on an “acoustic tweezers” arrangement, wherein ultrasonic radiation forces are used to trap the particle 18 and suspend it in mid-air. Such acoustic levitation is now well-understood and has been demonstrated in various media including air and water, and for particle sizes ranging from the micrometre to the centimetre scale.

For spherical particles significantly smaller than the wavelength and operating in the far-field regime (i.e. as is the case for MATD device 10 shown in FIGURE 1), the forces exerted on the particle are determined by the gradient of the Gor’kov potential. Several different trap morphologies have been demonstrated to date, including twin traps, vortex traps, and bottle beams, which can all now be analytically computed with efficiency.

The device shown in FIGURE 1 exploits this by analytically computing a single twin trap at a hardware level on an FPGA (Field Programmable Gate Array) 12. This allows for position and amplitude updates of the trap within the volume of the device at a rate limited only by the transducer frequency. In contrast, Spatial Light Modulators are limited to update rates of hundreds of updates/second, while galvanometers are usually limited to about 20 kHz. Existing acoustic modulators are limited to hundreds of Hz and displacement speeds well below 1 m/s. By contrast, the MATD device shown in FIGURE 1 enables update rates of up to 40000 updates/second and particle displacement speeds of up to 8.75 m/s and 3.75 m/s in the vertical and horizontal directions respectively.

This means that the particle 18 can be scanned around the display volume at such speeds that POV effects can be exploited in order to generate semi-persistent visual content. For instance, the particle 18 may be caused to trace out a certain image in order to generate desired visual content.

The hardware-embedded computation of the twin trap provides controlled and fast levitation of the scanning particle 18, which can be appropriately synchronized with the illumination module 16 to generate the desired visual content. For instance, this allows for a POV display with accurate control of the perceived colour (gamma corrected 2.2), able to deliver either two-dimensional or three-dimensional vector contents by POV, or even fully visual rasterized content (e.g. by repeatedly scanning the particle 18 along the same path).

FIGURE 2 thus illustrates some examples of visual content that has been generated using a MATD device 10 of the type shown in FIGURE 1 and a single scanning particle 18 which in this case was provided in the form of a 1 mm radius, white, spherical expanded polystyrene (EPS) bead.

The EPS bead used to generate the visual content in FIGURE 2 provides a good approximation to a Lambertian surface. Such particle then allows for predictable models of acoustic trapping forces, as well as a simple analytical model to describe perceived colour under controlled illumination. However, of course it will be appreciated that other particles may suitably be used within such an MATD device 10 in order to generate visual content and this is merely one example.

In particular, the visual content shown in FIGURE 2 was generated using a MATD device 10 of the type shown in FIGURE 1 comprising two 16 x 16 arrays of Murata MA40S4S transducers (40 kHz operating frequency, 1 cm diameter (-1.2 l), 12 Vpp, delivering an acoustic pressure of approximately 1.98 Pa at a distance of 1 m). A Waveshare CoreEP4CE6 FPGA was arranged to receive updates from a CPU (3D position, RGB colour, phase and amplitude), using 10 bits to encode each XYZ position (0.25mm resolution), 24 bits for colour (RGB) and 8 bits for the amplitude and phase of the trap, requiring 18 bytes for each update (9 bytes per array of transducers). Communication was implemented using a UART protocol at 12Mbps allowing for 40000 updates per second.

The transducers were driven at the specified voltage using a 12Vpp square wave signal at 40 kHz. Phase delays were implemented by temporal shifting of the 40 kHz square wave (see FIGURE 3A), while amplitude control was implemented by reducing the duty cycle of the square wave (i.e. reduce duration of the high period, as in the lower row in FIGURE 3A). As shown in FIGURE 3B, the complex amplitude of the transducers did not vary linearly with duty cycle (i.e. a control signal with 25% duty cycle does not result in half the amplitude of a control signal using 50% duty cycle).

As shown in FIGURE 3C the transducers have a substantially sinusoidal response due to the narrowband response of the transducers, and no harmonics are introduced from the square wave used to drive them. The duty cycle can then be matched to effective amplitude as in equation (1), with overall behaviour as shown in FIGURE 3B.

This function can be stored in the FPGA 12 as a look-up table mapping the amplitude to duty cycle for efficient computation of the updates at the desired rate (40000 updates/second in this example, although higher (or lower) operating frequencies may of course also be used). This resulted in a modulator providing 64 levels of phase (resolution TT/32 radians) and 32 levels of amplitude resolution.

In the present embodiment the computation of the acoustic trapping field is thus embedded into the FPGA 12. For a focus point at position p and with phase f_r, the phase of each transducer (</>_t) was discretised as in equation (2):

where k represents the wave number for the frequency used {k-2 p/l = 726.4 rad/m), p_t represents the position of each transducer and d represents the Euclidean distance function.

Twin traps were computed by combining a high intensity focus point (as in equation (2)) and a suitable levitation signature. For example, the levitation signature may be implemented by adding a phase delay of p radians to the transducers in the top array producing traps maximizing vertical forces. Transducer positions and discretized phase delays relative to distance were stored in two look-up tables in the FPGA 12, simplifying the computation of the focus point and levitation signature.

The resulting perceived luminance of the particle (e.g. a point within the visual content) for an observer around the MATD can be analytically approximated from the definition of the Bidirectional Reflectance Distribution Function (BRDF) as shown in equation (3), and it only depends on the angle a between the observer, the particle and the light. As mentioned above, the white and diffuse surface of the EPS particle used in this example allows its BRDF to be approximated as a Lambertian surface. The small diameter of the particle compared to the distance to the light source also means that the incoming illuminance is almost constant across the illuminated surface of the particle, as well as a constant incoming direction (i.e. light source approximated as a directional light). Similarly, the large distance to the observer (compared to the particle diameter) means that the direction of the rays from the particle to the observer is substantially parallel. The perceived luminance is then the summation of the luminances scattered towards the observer direction from each fraction of the sphere illuminated by the source and visible to the observer, as in equation (3):

where, dE_t represents the differential of incoming illuminance hitting the particle; dL represents the differential in luminance towards the observer at each point of the particle’s surface; dS represents the differential of surface and Q and f represent spherical coordinates.

Finally, incoming illuminance (amount of perceived radiant energy emitted per unit area and unit time) may need to be corrected for the ratio of time per second that the particle will be actually present across each discretized part of the visual content. For example, non-linear human response to luminance (e.g. Steven’s power law) may need to be considered. This may be done, and was done in the present examples, using a Gamma correction method ( y=2.2 ), similar to the one used in CRT monitors, to correct for these effects.

Thus, in this way, by appropriately moving the particle around the display volume, it is possible to generate using the MATD device 10 a wide range of visual content, e.g. as illustrated in FIGURE 2. The acoustic field that is used to levitate the particle 18 can also be used for providing simultaneous audio content. For instance, this may be achieved through a suitable side-band modulation of the acoustic field.

Audible sound can thus be generated by sampling the intended 40 kHz audio signal (e.g. from a file), and then using this to modulate the amplitude of the transducers in the arrays 14. In embodiments, a single side band modulation method (modulation index a= 0.2) is used, resulting in audible sound of >60 dB (i.e. in the level of a conventional human conversation). However, other arrangements would of course be possible.

In the present embodiments the amplitude is modulated at the same time that the particle is being levitated, in order to create audible sound at the levitation point. More specifically an upper sideband modulation is used (see equation (4)), which avoids harmonics distortion and allows for simultaneous levitation and audible sound. The modulated signal was computed as:

where g(t ) represents the audio signal required to be created at time t, g(t ) represents a Hilbert transform of g t) and a represents the modulation index. The signal was sampled at 40 kHz and the resulting amplitude (A_sss, from equation (4)) sent to the FPGA together with the remaining required parameters for the current update (i.e. position, colour and phase).

Further, the MATD device 10 can also be used to generate tactile content, e.g. in the form of mid-air tactile feedback at controlled locations (e.g. user’s hand), by using a secondary focusing trap 10 and custom multiplexing policy. For instance, the hardware can provide individual phase and amplitude updates at 40 kHz and then time multiplexing to simultaneously create several levitation traps.

In order to generate visual and tactile content the MATD device 10 may thus use two time- multiplexed traps: a primary twin trap for levitating the particle 18 and generating the visual content in the manner described above; and a secondary focus point 19 for providing tactile content. The MATD device 10 can then be operated in various different operating modes.

For example, in a single trap mode, only the primary twin trap is present (100% duty cycle, 40000 updates per second), and loaded with an EPS particle of ~1mm radius, as described above. This levitation trap is used to scan the volume which, synchronized with the illumination module, provides the visual component of the display.

However, the MATD device 10 is also operable in a dual trap mode, e.g. for cases where tactile feedback needs to be delivered (e.g. only in the presence of the user’s hand). In this case, the primary trap can be setup as above, but is time-multiplexed with a secondary trap which creates the tactile stimulation.

Two main parameters need to be considered for this multiplexing: amplitude multiplexing and position multiplexing.

Amplitude multiplexing relates to the recreation of tactile textures, which involves a modulation frequency which can be detected by skin’s lamellar corpuscles. For example, a suitable modulation frequency may be of the order 250 Hz. One approach for generating both audio and tactile content would be to multiplex between the amplitude of the tactile signal (250 Hz) and any audio signals (multiple frequencies), at the expense of limiting the frequency of each individual signal and this may in principle be done.

However, in the present embodiment, when it is desired to provide simultaneous audio and tactile content, the 40 kHz multi-frequency audio signal is combined with the tactile modulation signal (250Hz) into a single signal, thus maintaining the sampling frequency of the individual signals (i.e. avoiding such amplitude multiplexing) and reducing losses in audio quality. Sampling at 40 kHz encodes most of the audible spectrum (44.1 kHz), and the high power transducer array produces audible sound even from a relatively small modulation index (a = 0.2), while still modulating particle positions and tactile points at the 40kHz rate.

The MATD device 10 generally supports two different modes to create audio content: a scatter mode providing non-directional sound but compatible with simultaneous visual and tactile content (FIGURE 4A); and a directional mode implemented by using a secondary trap to steer the sound on the direction of the user but not allowing simultaneous tactile points (i.e. only visual content and directional audio) (FIGURE 4B). The scatter mode uses the trapped particle as a scattering media implicitly providing spatialized audio (i.e. sound coming from the content displayed). Such directional cues may be relatively weak (most sound coming from the centre of our working volume). Accordingly, the directional mode instead uses a secondary trap to steer sound towards the user, resulting into a stronger directional component and higher sound levels. However, the use of directional audio currently comes at the expense of not simultaneously delivering tactile feedback (simultaneous visual, tactile and directional audio would require multiplexing of three traps, one for each modality).

FIGURE 4A and FIGURE 4B also show the sound levels around the display during horizontal and vertical scans around the MATD volume. In particular, these figures show audible levels of sound at all points around the display (74±12 dB for the non-directional scatter mode and 72±13 dB for the directional mode). Points of higher intensity can be found at some points around the MATD, which are to be expected as a result of constructive interference. In the directional case, high pressure levels of 103 dB can be observed around the intended targeted point, which then continue to propagate forwards along the direction between each transducer array and the focussing point. In all cases, the inclusion simultaneous tactile and audio information results in only a small reduction on the intensity of audible sound (66±11 dB and 63±12 dB for the non-directional and the directional methods).

It will be appreciated that the primary trap that is used for levitating the particle 18 should be kept spatially apart from the secondary trap 19 used for providing the tactile content. This is referred to herein as position multiplexing. Unlike the amplitude multiplexing discussed above, position multiplexing only affects the phases of the transducers, and it cannot be avoided in such dual trap scenarios.

The overall duty cycle may thus be split between generating the primary and secondary traps. For instance, in an example, 75% of the duty cycle (i.e. three contiguous updates, providing an update rate of 30000 updates/second when the overall update rate is 40000 updates/second) is used to update the primary trap and 25% of the duty cycle is used for updating the secondary trap (an update rate of 10000 updates/second when the overall update rate is 40000 updates/second).

This high frequency change of location between the primary and secondary traps (i.e. 10000 changes between the primary and secondary traps per second) may introduce sudden changes in the transducers phases, which might force the transducers to operate at sub- optimal frequencies. To alleviate this, the phase of the next update ( f_r , in equation (2)) may be set to the value that minimizes the summation of absolute phase differences between the current transducer phase distribution and the previous one.

Thus it has been found that well-differentiated tactile feedback can be delivered using only 25% of the duty cycle for the tactile content (and thus leaving 75% of the duty cycle still available for positioning the primary trap) and in that case the generation of the tactile content results only in a relatively small loss of scanning speed. For instance, the tactile content may be generated using a 250 Hz modulation frequency, avoiding the 2 kHz-5 kHz primary range of human audible perception (to minimize parasitic noise), but remaining well within the optimum perceptual threshold of skin lamellar corpuscles for vibration. The 10000 updates/second update rate for tactile stimulation is sufficient for spatiotemporal multiplexing strategies to maximize fidelity of mid-air tactile content. It has been found that this allows accurate positioning and focusing of the tactile points and sound pressure levels of >150 dB, well above the threshold of 72 dB levels required for tactile stimulation. FIGURE 5A shows how the sound pressure level generated by the MATD device 10 may vary across the device when delivering tactile content. In the first condition, only the tactile content was delivered (i.e. the array created a tactile point during the 25% duty cycle allocated for the secondary trap, and no output was produced by the array during the remaining 75% percent of the time). The second condition represents the case where both visual and tactile content is presented. In the third condition a combined signal (i.e. audio with a 2 kHz, combined with 250 Hz signal) is used to represent the case where visual, tactile and audio content is presented.

FIGURE 5B shows the effects that a user hand could have (i.e. due to hands occluding part of the transducers or to scattering on the user’s hand), on the sound levels. In particular FIGURE 5B shows the results of the measured acoustic field both in the presence and absence of a silicone hand. When the silicone hand was present, the tactile point was created on the surface of the bottom part of the index’s fingertip. In all three conditions (visual only; visual and tactile; and multimodal), a horizontal and vertical plane of 10x10cm was scanned, measuring SPL levels at a resolution of 1 mm.

The results presented above show that the device provided accurate positioning and focussing of the acoustic pressure around the central point (where tactile feedback is presented) in all three cases and both in the presence and absence of the hand. Vertical scans show a repeated pattern of lobes, consistent with the interference of the acoustic radiation emitted from the top and bottom arrays. Some differences can be found between the tactile only condition (first column) and the other two cases, as a result of the effects of the primary trap (visual content). However, the effects around the tactile point are small, the sharpness of the tactile point is maintained and there is very little variation across all three cases. Maximum pressure levels are found at the centre of the tactile points, and are always well above the thresholds of 78 dB required for perceivable tactile feedback. It is noted that the presence of a second-high pressure area to the bottom left of the images shown in FIGURE 5 in the second and third conditions is the result of the primary trap used to deliver the visual content.

FIGURES 6A-6D shows the results of a spectral analysis of the audio response in the MATD device 10 in particular to explore the effects of amplitude and position multiplexing, as discussed above. In particular, FIGURE 6A shows the signals used as input, which respectively comprise an acoustic chirp signal (left), a 250 Hz modulation signal for generating tactile sensations (center), and a combined signal representing a combination of these two signals in the frequency domain (right). This may represent the case when the primary trap is used to trap a particle (visual and audio feedback), while the second trap is used to create tactile feedback on the user’s skin.

FIGURE 6B shows in the left hand panel the output from the system when only audio content is generated (i.e. using only the primary trap). The center panel of FIGURE 6B then shows the output when the audio content is multiplexed with tactile content using amplitude multiplexing with a multiplexing rate of 20000 updates/second (i.e. an overall update rate of 40000 updates/second, with a 50% duty cycle for each mode). The right hand panel of FIGURE 6B then shows the output when using a combined signal with an update rate of 40000 updates/second. It can be seen that the amplitude multiplexing introduces various harmonic artefacts and the use of a combined signal thus produces better quality audio. The use of position multiplexing (i.e. focusing the acoustic power at the location of the levitation trap for 75 ps, and then refocusing it at the location of the tactile trap for 25 ps) cannot be avoided if simultaneous tactile and audio-visual content is to be delivered. Such position multiplexing may introduces frequency aliasing at the (e.g. 10 kHz) multiplexing rate, as well as harmonic frequencies, as a result of acoustic pressure being focalised at different locations. FIGURE 6C shows how the approach according to the present embodiment using position multiplexing with combined 40 kHz signal may help to reduce audible artefacts when compared to the use of both amplitude and position multiplexing, particularly for harmonics and how this approach minimizes the artefacts present in the human primary auditory range (i.e. 2 kHz - 5 kHz).

These results also illustrate the benefits of using high update rates for an MATD modulator (i.e. beyond enabling higher particle speeds). For instance, a multiplexing rate of 10 kHz may create aliasing effects also at harmonic frequencies (i.e. 20 kHz, etc.). However, a modulator with a lower multiplexing rate would create artefacts at many more frequencies, spread across the auditory range (e.g. a modulator at 10 kHz would require a multiplexing rate of 2.5 kHz, introducing artefacts around 2.5 kHz, 5 kHz, 7.5 kHz, etc.). It is also worth noting that the aliasing effects are related to the multiplexing schedule used (e.g. the relative portions of the duty cycle used for levitation and tactile content), which in turn is related to the power constraints of the acoustic source. Increased transducer power, allowing for effective levitation at a 50% duty cycle (50% for levitation, 50% for tactile feedback) would avoid most of these artefacts, by shifting them around a primary 20 kHz frequency. FIGURE 6D shows a test performed using such configuration (50% duty cycle), with reduced artefacts compared to the use of a combined signal. Again, it can be seen that using the combined signal provides a better audio quality.

Thus, it has been found that the techniques described herein allow for high scanning speeds and accelerations, well above optical or acoustic setups demonstrated to date. The most critical display parameters are summarized in Table 1, according to the MATD’s various modes of operation: single particle with no amplitude modulation (visual content only), single particle with minimum amplitude (worst case displaying visual and audio content), and time multiplexed dual trap with minimum amplitude (worst case delivering all visual, audio and tactile content). It will be appreciated that trapping forces and achievable speeds and accelerations vary with the direction of motion of the particle (i.e. highest in the vertical direction). Table 1 provides maximum displacement parameters along the horizontal direction (i.e. worst case with weaker trapping forces), as conservative reference values that allow content reproduction independently of the particle direction.

Table 1: Main parameters MATD . , , Visual and Visual, audio and

Visual only ,. .. audio tactile

Highest speed recorded (v_max) 3.75 m/s 3.375 m/s 2.5 m/s Highest acceleration recorded (a_max ) 141 m/s² 122 m/s² 62 m/s² Highest speed for corner features 0.75 m/s 0.5 m/s 0.375 m/s

(y corner)

Highest image framerate until now 12.5Hz 10.0Hz 10.0Hz Colour 24bpp 24bpp 24bpp

The parameters set out in Table 1 can thus be used to compute and plan paths to create POV content visible to the naked eye.

Human eyes can integrate different light stimuli under a single percept (i.e. a single shape/geometry) during reduced periods of time (0.1s usually accepted as a conservative estimation, even in bright environments), and thus, the particle needs to be scanned the content in less than this time (0.1s). The parameters above allow feasible paths (particle speed, acceleration and curvature within the limits identified) to be determined, which can be revealed in less than 0.1s exploiting only a fraction of the display’s capabilities. This then allows the generation of visual content as described above and shown in relation to FIGURE 2.

In particular The MATD device 10 has demonstrated the possibility to manipulate particles by retaining them in a dynamic equilibrium (rather than a static one, as most other levitation approaches), enabling the high accelerations and speeds observed.

In general the trapping forces are dependent on direction due to the type of levitation trap we use. For instance, the trap geometry shown in FIGURE 1 maximizes vertical trapping forces, while forces along the horizontal plane are weaker, which affects the accelerations and speeds that can be imparted on the particle in each of these directions.

An exploration of the speeds that can be achieved with the MATD device 10 will now be described. During this testing, the MATD device 10 was characterised under three sets of experimental conditions, to reflect both optimistic and pessimistic scenarios, as follows: (i) optimistic single trap mode (OSTm), with only the main trap and fixed maximum amplitude ( _SSB=1); (ii) pessimistic single trap mode (PST m), with only the main trap and minimum amplitude {A_SSB=0.83, equivalent to using the silent section of an audio file); and (iii) pessimistic dual trap mode (PDTm), with both traps (75% duty cycle for the primary trap;

25% for the secondary trap) and minimum amplitude (A_SSB=0.83). The location of the secondary (tactile) trap was fixed, horizontally placed at the edge of the array and at a height equal to the centre of the array.

Particularly, for a chosen particle (in this example ~2 mm), tests can be performed characterizing maximum displacement speeds for each of the three experimental conditions (OSTm, PSTm and PDTm) for particle motion along three directions: along the vertical axis Y (both in the upwards and downwards directions) and the horizontal axis X. It will be appreciated that for the MATD device 10 set-up shown in FIGURE 1 , the X and Z axes are equivalent (e.g. 90-degree rotation).

To test the achievable particle speeds, linear paths of 10 cm were used, with the particle starting at 5 cm to the left and stopping at 5 cm to the right of the centre of the MATD device 10 (i.e. or 5 cm above/below the centre, for the vertical tests). In particular, the particle started at rest and was constantly accelerated to reach maximum speed at the centre of the array. The particle was then constantly decelerated until brought back to rest at a position 10 cm away from the starting position. The results of this testing is summarised in FIGURES 7A-7C which show the results for the maximum linear speeds (i ™_*), particle-to-trap distances, and accelerations, obtained for each condition (OSTm, PSTm and PDTm), for particles travelling along the horizontal direction (FIGURE 7A), as well as travelling up/down in the vertical direction (FIGURE 7B and FIGURE 7C respectively). In the velocity plots the solid black lines represent the speed of the levitation trap, while the other lines show examples of actual particle velocities as captured during experimental tests.

As expected, maximum displacement speeds are influenced by the mode of operation used. While the decrease in maximum speed is small when audio is included (OSTm vs. PSTm), the effect is much larger when tactile effects are introduced as the acoustic power is split between two traps (i.e. time multiplexing for the PDTm mode). Also, linear speeds are much higher along the vertical axis (particularly when going downwards, due to the effect of gravity), when compared to horizontal displacements. This is because the setup shown in FIGURE 1 with vertically-spaced top and bottom arrays and the twin traps used create trapping forces around the levitation trap that are much stronger along the vertical direction, allowing for higher accelerations.

The paths observed in this testing show expected correlations between particle velocities (top), particle to trap distances (middle) and accelerations (bottom). Points of zero Dr (i.e. no net force being applied to the particle) correspond with maximum/minimum points in each velocity plot (i.e. derivative equal to zero), and the sign of Tip aligns with the monotonicity of velocity plots, increasing when delta p is negative or decreasing otherwise. Similar correlations can be observed between Dr (middle) and acceleration plots (bottom). Accelerations remain positive when D p is negative and vice-versa (i.e. trap as a restorative force), and prominent features in both plots match well (e.g. maximum, minimum, roots).

FIGURE 8 summarises the speed testing in FIGURES 7A-7C and illustrates the maximum linear horizontal speeds and accelerations for each of the three modes (OSTm, PSTm and PDTm). Note that FIGURE 8 plots the speed of the acoustic trap, rather than observed particle trajectories. FIGURE 8 also shows the maximum linear speeds achievable by particles following circular paths of increasing radius for each of the three modes.

It can be seen from the plots shown in FIGURES 7A-7C that the particle almost always remains at a few millimetres from the place where the actual levitation trap was placed (Tip), thereby being subject to high acceleration rates. This observation is important to understand the behaviour of the MATD device 10 in comparison to other levitators.

FIGURE 9 is a plot showing how the maximum achieved particle speeds (in the vertical direction) varies as a function of the rate at which the acoustic trapping field is updated. In particular, FIGURE 9 illustrates the benefits of the relatively higher update rates used according to the present embodiments, e.g. higher update rates allow for higher particle speeds. FIGURE 9 shows that the maximum particle speeds can be achieved for update rates greater than about 5000 updates/second. FIGURE 9 also shows that the PDTm mode is not supported at update rates less than about 2500 updates/second. The operating principle behind the MATD device 10 of the present embodiment will now be described with reference to FIGURES 10A and 10B. For instance, the behaviour of the particle can be understood in terms of the derivative of the Gor’kov potential at the points around the trap. For instance, the distribution of trapping forces for a specific acoustic trap depends on its specific point-spread function and the levitation signature used. FIGURE 10A thus shows the distribution of acoustic pressure around a vertical twin trap for a top- bottom setup, as in FIGURE 1.

FIGURE 10A also shows the distribution of forces for a particle located at a specific horizontal and vertical distance from the centre of the trap. In particular, FIGURE 10A shows how such forces evolve for points around a trap, as analytically derived considering our particular trap (twin trap), particle (radius ~1 mm, density ~19 kg/m³, speed of sound in EPS 900 m/s), setup (top and bottom arrays of 16x16 transducers, each modelled using a piston model) and assuming 346 m/s and 1.18 kg/m³ as the speed and density of air.

As shown in FIGURE 10A, a particle placed exactly at the centre of the levitation trap (Dr=0) receives a zero net force contribution, making it stable at that position, but also providing no acceleration. This is satisfactory for levitators designed for precise (but slow) particle manipulation. Also, such levitators usually operate at much lower update rates (i.e. hundreds of hertz), so when the position of the trap is moved, the particle has enough time to transition to the new trap location.

As the particle approaches the centre of the trap, the acceleration received will decrease. If the duration of each update is long enough, the particle will go past the centre of the trap and start receiving negative forces (decelerating), getting engaged in a oscillatory motion until it stabilizes (nearly) at the centre of the trap. As such, modulators with a slow update rate can result in uneven accelerations of the particle or make it difficult for the particle to retain its momentum (accumulate speed) between updates.

The particles manipulated by the MATD device 10 of the present embodiments however do not reach such a static equilibrium after each update. Instead, the particle remains at a distance from the centre of the levitation trap ( Dr ), so as to receive force and hence be accelerated.

The dynamic behaviour of the particle is shown in FIGURE 10B. In particular, as shown in FIGURE 10B, at a first time, to, the particle is located at the edge of the trap. The particle will thus start to accelerate towards the center of the trap, and this is shown in the middle panel, which represents the situation at a second time, t1. However, before the particle can reach the center of the trap, the position of the trap is moved (to the right in FIGURE 10B). This has the effect that the particle is effectively shifted back to the edge of the trap. The particle is thus accelerated again towards the center of the trap, and so on. In this way the particle can be kept in motion as the acoustic trap is moved in order to achieve the higher particle speeds described above.

As shown in FIGURE 10A, the restorative forces along the horizontal axis (Fx) peak at distances of nearly +3.5 mm from the centre of the trap, closely matching the distances at which the particles were detected during the horizontal speed testing described above. A similar behaviour can be observed for the vertical tests. In these cases, the peaks of the restorative forces along the vertical direction (Fy) are at distances +1.5 mm, again matching the observed displacements.

The fact that the trap and the particle did not always remain at those peak distances (i.e.

+3.5mm and +1.5mm) indicates that even higher speeds should be achievable for both horizontal and vertical displacements. This, however may require a more complex control mechanism to determine the location of the levitation trap, accurately predicting the current location of the particle at each point in time (considering the acoustic force along with drag, gravity and centrifugal forces) and positioning the trap accordingly (e.g. 3.5 mm ahead of the particle for maximum horizontal acceleration). Other factors, such as the temporal changes in complex amplitude (and hence force) related to the simultaneous creation of audible sound; or the multiplexing and interference from the secondary trap should also be considered for such a model.

As described above, the creation of visual content for the MATD is achieved through the definition of closed and smooth parametric curves, illuminated with varying RGB colours at different points of the path. For content to be visible by the naked eye, such closed curves typically need to be traversed by the particle in less than 0.1 s, which becomes a constraint influencing the particle manipulation required, that is, the speeds and accelerations that need to be imparted at each point along the curve to reveal it within 0.1 s.

However, whilst maximum linear displacement speeds ( v_max ) are a relevant constraint to plan/design such paths, other parameters (i.e. maximum particle acceleration, feasible radius of curvature, maximum speed at corner features, etc.) are also relevant. Testing of the maximum speed at which the particle could execute a complete change of direction {y come_t), such as those required to render corners or sharp features, was thus also performed, and the results are reported in Table 1.

The use of models accurately predicting the dynamics of the particle (i.e. in terms of acoustic forces, drag, gravity and centrifugal forces, but also considering interference from secondary traps and transient effects in the transducers’ phase updates) would allow for better exploitation of the observed maximum speeds and accelerations, enabling larger and more complex visual content. Alternatively, they could instead allow for a more efficient use of the acoustic pressure, providing similar speeds and accelerations to the ones provided by the MATD device 10, but allocating a lower duty cycle for the primary trap. This power could then be dedicated for stronger tactile content or to support more simultaneous traps (e.g. the three traps required for the simultaneous visual, tactile and directional audio scenario).

Thus, the results above demonstrate an approach for creating volumetric POV displays with simultaneous delivery of audio content and tactile feedback, exceeding the capabilities of alternative optical approaches. Polarization based photophoretic approaches could potentially match the potential for particle manipulation (i.e. speeds and accelerations) demonstrated in this study, but they would still not be able to engage with sound and touch. The MATD device demonstrated hence provides a volumetric display providing a full sensorial reproduction of virtual content.

Whilst an example has been described above which utilises relatively simple LED illumination, it will be appreciated that more advanced illumination approaches (e.g. using galvanometers or beam steering mechanisms) would potentially allow for focused light and brighter displays. The use of several illumination modules around the display would allow for more control on the visual properties of the content displayed.

For instance, combining a denser illumination array (e.g. a ring of light sources) and the predictable light scattering pattern of our particle, the final scattered field from the particle can be computed as the linear combination of the scattered field from each light source.

This could be used, for instance, to create visual content approximating various material properties (e.g. make content look metallic or matte), simulating different lighting conditions or even delivering different contents in different viewing directions.

The MATD device of the present embodiment thus allows for a volumetric display where, for the first time, users can simultaneously see visual content in mid-air from any point around the display volume and receive auditory and tactile feedback from that volume. This offers clear competitive advantages in the fields of interactive experiences and/or scientific visualization. The present embodiment also shows particle manipulation capabilities that are superior to other approaches demonstrated to date, offering opportunities for non-contact, high-speed manipulation of matter, with applications in computational fabrication and biomedicine.

Thus, beyond enabling simultaneous visual, tactile and audio content for multimodal display purposes, the techniques described herein for enabling positioning and amplitude modulation of acoustic traps at the sound-field frequency rate (e.g. 40 kHz), also offer opportunities for non-contact, high-speed manipulation of matter, with applications in computational fabrication, biomedicine, and interesting experimental setups for chemistry or lab-on-a-chip applications.

Indeed, it will be appreciated that the techniques described herein may generally be used for manipulating any suitable particles. For instance, the techniques described herein may also be used for manipulating liquid droplets. In this case, amplitude modulation may be used to re-shaped (e.g. flatten) the droplet, as shown in FIGURE 11.

The ability of manipulate liquid particles may find utility, for example, for additive manufacturing (e.g. 3D printing). A second main application of the techniques described herein is thus shown in FIGURE 12 which illustrates an additive manufacturing apparatus 80 according to an embodiment. The additive manufacturing apparatus 80, similarly to the MATD device 10 shown in FIGURE 1 , includes a pair of spaced-apart transducer arrays 84 and an FPGA 82 for controlling the acoustic field that is generated in the volume of the additive manufacturing apparatus 80.

Material to be printed is introduced into the volume of the additive manufacturing apparatus 80 through a suitable extruder capillary 86 on a particle by particle basis. Each particle can then be levitated and then deposited onto a substrate in order to build the printed model 89.

In FIGURE 12 the model 89 is built inside the volume of the additive manufacturing apparatus 80, and the acoustic field should be calculated to take this into account. However, other arrangements would of course be possible. In embodiments the particles may be manipulated to control their shape as they are deposited onto the model 89, e.g. as shown in FIGURE 11. In general, the devices described herein may be used in any medium, depending on the application. For instance, whilst the testing described above has been performed in air, it will be appreciated that the techniques described herein may also be extended to other media, including water.

Also, as mentioned above, although various embodiments have been described with respect to a single particle, it will be appreciated that the techniques described herein can readily be extended to multi-particle embodiments. For instance, in the content of a MATD device, this may allow for the creation of larger and/or more complex content. It will be appreciated that multiple particles may be suspended either using a more complex acoustic trapping field (defining a plurality of trapping regions) or through multiplexing of multiple acoustic traps.

Thus, although the present invention has been described with reference to preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the accompanying claims.

Claims

144351/01 Claims:

1. An apparatus comprising: a set of one or more acoustic source(s) controllable to generate an acoustic field within a volume of the apparatus; and a control circuit that is configured to control the set of one or more acoustic source(s) to generate an acoustic field defining one or more acoustic trap(s) for suspending one or more particle(s) within the volume of the apparatus, wherein the control circuit is further configured to update the acoustic field to change the position of one or more of the acoustic trap(s) within the volume of the apparatus such that for at least a first acoustic trap of the one or more acoustic trap(s) the position of the first acoustic trap is moved to thereby cause a particle suspended within the first acoustic trap to move around the volume of the apparatus, and wherein during a movement event wherein the position of the first acoustic trap is being moved to thereby cause a particle suspended within the first acoustic trap to move around the volume of the apparatus the control circuit is configured to update the acoustic field at such a rate that the particle is prevented from reaching a static equilibrium position within the first acoustic trap as the acoustic field is updated multiple times to change the position of the first acoustic trap within the volume of the apparatus.

2. The apparatus of claim 1 , wherein the control circuit is configured to move the first acoustic trap at such a rate that the particle is kept at least at a certain distance from its equilibrium position within the first acoustic trap whilst the first acoustic trap is being moved around the volume of the apparatus.

3. The apparatus of claim 1 or 2, wherein the set of one or more acoustic source(s) comprises one or more array(s) of transducers.

4. The apparatus of any of claims 1 , 2 or 3, wherein the control circuit comprises an FPGA.

5. The apparatus of any preceding claim, being a display apparatus, wherein the movement of the particle around the volume generates visual content, the apparatus further comprising one or more light source(s) for illuminating the particle to control the visual content.

6. The apparatus of any preceding claim, being a display apparatus, wherein the position of the first acoustic trap is defined by a phase distribution of the acoustic field.

7. The apparatus of claim 6, wherein the control circuit is configured to modulate the amplitude of the acoustic field to generate an audio output.

8. The apparatus of any preceding claim, wherein the control circuit is configured to control the set of one or more acoustic source(s) to generate an acoustic field defining a plurality of acoustic trapping regions for simultaneously suspending a corresponding plurality of particles within the volume of the apparatus.

9. The apparatus of any preceding claim, wherein the control circuit is configured to control the set of one or more acoustic source(s) to generate first and second temporally interleaved acoustic fields, wherein the first acoustic field defines the first acoustic trap for suspending and moving the particle within the volume of the apparatus and wherein the second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output.

10. The apparatus of any preceding claim wherein the control circuit is configured to update the acoustic field(s) at a rate of at least 5000 updates/second.

11. An apparatus comprising: a set of one or more acoustic source(s) controllable to generate an acoustic field within a volume of the apparatus; and a control circuit that is configured to control the set of one or more acoustic source(s) to generate first and second temporally interleaved acoustic fields, wherein the first acoustic field defines a first acoustic trap for suspending a particle within the volume of the apparatus; and wherein the second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output.

12. The apparatus of any one of claims 1 to 4, being an additive manufacturing device, comprising at least a first material supply, and wherein the control circuit is configured to update the acoustic field to change the position of the first acoustic trap within the volume of the apparatus and to thereby move a particle received from the first material supply onto a target substrate.

13. The apparatus of claim 12, wherein the set of one or more acoustic source(s) are configured to generate first and second temporally interleaved acoustic fields, wherein the first acoustic field defines the first acoustic trap for moving a first type of particle from a first material supply, and wherein the second acoustic field defines a second acoustic trap for moving a second type of particle from a second material supply, and wherein the control circuit is configured to update the first and second acoustic fields to change the position of the acoustic traps and to thereby move the particles around the volume of the apparatus.

14. The apparatus of claim 12 or 13, wherein the position of the first acoustic trap is defined by a phase distribution of the acoustic field and wherein the control circuit is configured to modulate the amplitude of the acoustic field to control a shape of a particle suspended in the first acoustic trap.

15. A method of manipulating a particle within a volume, the method comprising: generating using a set of one or more acoustic source(s) an acoustic field defining one or more acoustic trap(s) for suspending one or more particle(s) within the volume; and updating the acoustic field to change the position of the acoustic trap(s) within the volume of the apparatus such that for at least a first acoustic trap of the one or more acoustic trap(s) the position of the first acoustic trap is moved to thereby cause a particle suspended within the first acoustic trap to move around the volume, wherein during a movement event wherein the position of the first acoustic trap is being moved to thereby cause the particle suspended within the first acoustic trap to move around the volume of the apparatus the acoustic field is updated at such a rate that the particle is prevented from reaching a static equilibrium position within the first acoustic trap as the acoustic field is updated multiple times to change the position of the first acoustic trap within the volume of the apparatus

16. The method of claim 15, wherein the method comprises moving the particle around the volume to generate visual content.

17. The method of claim 15 or 16, further comprising illuminating the particle as it is moved around the volume.

18. The method of any one of claims 15 to 16, comprising modulating the amplitude of the acoustic field to generate audio content.

19. The method of any one of claims 15 to 18, comprising generating using the set of one or more acoustic source(s) two or more temporally interleaved acoustic fields wherein a first acoustic field is the acoustic field defining the first acoustic trap for suspending and moving the particle within the volume of the apparatus and wherein a second acoustic field defines a second acoustic trap for suspending a second particle and/or for generating a tactile output.

20. A method of additive manufacturing comprising the method of claim 15, wherein the particle comprises a particle received at a first position from a first material supply, and wherein the method comprises moving the particle from the first position to a second position, and depositing the particle onto a substrate in order to build an output model.

21. The method of claim 20, wherein the position of the first acoustic trap is defined by a phase distribution of the acoustic field, the method further comprising modulating the amplitude of the acoustic field to control a shape of the particle suspended in the first acoustic trap.

22. The method of any one of claims 15 to 21, wherein the generated acoustic field defines a plurality of acoustic trapping regions for simultaneously suspending a corresponding plurality of particles within the volume.

23. The method of any one of claims 15 to 22, wherein the acoustic field is updated at a rate of at least 5000 updates/second.