Dynamic surface analysis for acoustic droplet ejection

Abstract

A system for ejecting droplets from a sample surface using acoustic energy includes a processor, a transducer configured to receive electrical energy and emit corresponding acoustic energy, and further configured to receive acoustic energy and generate corresponding electrical energy, a transmitter communicating with the processor and transducer, the transmitter configured to receive at least one signal from the processor and transmit corresponding electrical energy to the transducer, and a receiver communicating with the processor and transducer, the receiver configured to receive electrical energy generated by the transducer and transmit at least one corresponding signal to the processor, wherein the transducer is configured to emit a first acoustic signal, a second acoustic signal, and a third acoustic signal.

Description

Background Art

[0001] (Cross - reference to related applications) This application claims the priority and benefit of U.S. Provisional No. 63 / 448,075, filed Feb. 24, 2023, which is hereby incorporated by reference in its entirety.

[0002] (Background) Acoustic droplet ejection (ADE) is a technique that uses acoustic energy to move a liquid without any physical contact. Acoustic energy (e.g., in the form of ultrasonic pulses) is emitted from a transducer towards a volume of liquid (hereinafter, “sample”). In some embodiments, the beam converges on or near the upper surface of the sample, and the acoustic energy is transmitted to a portion of the sample, thereby moving this portion upwardly away from the rest of the sample (e.g., as a droplet). The sample may be contained within a well of a container (e.g., a 96 - or 384 - well microplate).

Summary of the Invention

Means for Solving the Problems

[0003] (Abstract) According to one embodiment, a system for ejecting droplets from the surface of a sample using acoustic energy includes a processor, a transducer configured to receive electrical energy and emit corresponding acoustic energy, and further configured to receive acoustic energy and generate corresponding electrical energy, a transmitter communicating with the processor and transducer, the transmitter configured to receive at least one signal from the processor and transmit corresponding electrical energy to the transducer, and a receiver communicating with the processor and transducer, the receiver configured to receive electrical energy generated by the transducer and transmit at least one corresponding signal to the processor, wherein the transducer is configured to emit a first acoustic signal, a second acoustic signal, and a third acoustic signal, the first acoustic signal being determined to cause a surface ripple phase during which droplets are formed on the surface of the sample, the second acoustic signal being selected to acquire information from the surface of the sample during the surface ripple phase, and the third acoustic signal being selected to acquire further information from the surface of the sample during the surface ripple phase. The second and third acoustic signals may be part of a plurality of information-acquiring pings, and the transducer is configured to emit information-acquiring pings at intervals of approximately 50 to 100 μS. The plurality of information-acquiring pings may include approximately 4 to 2,000 pings. The processor may be configured to receive information from the surface of the sample corresponding to the reflection of the second and third acoustic signals, and the processor is configured, at least in part, to determine at least one characteristic of the surface ridge phase based on this information. The at least one characteristic of the surface ridge phase may include at least one of droplet size, droplet linear velocity, droplet angular velocity, droplet radial displacement, droplet detachment position, surface position, surface velocity, surface acceleration, surface ridge formation rate, or surface ridge relaxation rate.The transducer may further be configured to emit a fourth acoustic signal, which is determined to cause a second surface elevation phase during which a second droplet is ejected from the surface of the sample, and the processor may further be configured to determine the fourth acoustic signal according to at least one characteristic of the surface elevation phase. The first and fourth acoustic signals may have at least one of different powers, frequencies, or durations.

[0004] According to one embodiment, a system for ejecting droplets from the surface of a sample using acoustic energy includes a processor, a transducer configured to receive electrical energy and emit corresponding acoustic energy, and further configured to receive acoustic energy and generate corresponding electrical energy, a transmitter communicating with the processor and transducer, the transmitter configured to receive at least one signal from the processor and transmit corresponding electrical energy to the transducer, and a receiver communicating with the processor and transducer, the receiver configured to receive electrical energy generated by the transducer and transmit at least one corresponding signal to the processor, wherein the processor is configured to cause the transducer to emit a droplet formation signal, thereby causing a surface elevation phase in which droplets are ejected from the surface of the signal, and to further emit a plurality of pings at least one before, during, or after the surface elevation phase, the processor is further configured to receive information from the receiver corresponding to the reflections of the plurality of pings, and the processor is further configured to determine at least one characteristic of the surface elevation phase according to the information received from the receiver corresponding to the reflections of the plurality of pings. The processor may further configure the transducer to emit a second droplet formation signal, thereby inducing a second surface elevation phase in which a second droplet is ejected from the surface of the sample, and the processor may further configure the processor to determine the second droplet formation signal based at least in part on information corresponding to the reflection of multiple pings received from the receiver. The second droplet formation signal may be determined according to a machine learning model. The droplet formation signal and the second droplet formation signal may be determined such that the first droplet and the second droplet have at least one of different volumes or separation velocities. At least one characteristic of the surface elevation phase may include at least one of droplet size, droplet linear velocity, droplet angular velocity, droplet radial displacement, droplet detachment position, surface position, surface velocity, surface acceleration, surface elevation formation rate, or surface elevation relaxation rate.The transducer may be configured to emit multiple pings at intervals of approximately 50 to 100 μS. These multiple pings may include approximately 4 to 2,000 pings.

[0005] According to one embodiment, a method for implementing any of the aforementioned system embodiments is provided. [Brief explanation of the drawing]

[0006] [Figure 1] Figure 1 shows a representation of the ADE system, including a cross-sectional view of a container plate containing multiple containers (or wells) for holding multiple individual samples, a receiving plate, a transducer assembly, and a block diagram of the electronic circuitry.

[0007] [Figure 2] Figure 2 shows a block diagram of the transducer assembly.

[0008] [Figure 3] Figure 3 shows a representation of the movement of the transducer assembly relative to the container plate when performing ADE on multiple samples.

[0009] [Figure 4] Figure 4 shows a top view of a container plate with multiple containers.

[0010] [Figure 5] Figure 5 shows a top view of multiple containers within a container plate, and a flowchart illustrating the sequence for sequentially performing ADE on each container.

[0011] [Figure 6] Figure 6 shows the container for holding the sample.

[0012] [Figure 7] Figure 7 shows the reflected signal and its corresponding envelope, which is received in the transceiver assembly in response to the emitted signal.

[0013] [Figure 8] Figures 8A, 8B, 8C, and 8D illustrate sequences of transmitted and reflected acoustic signals according to an embodiment.

[0014] [Figure 9A] Figure 9A illustrates the intensity and timing of signals reflected from the surface of a sample during ADE.

[0015] [Figure 9B] Figure 9B illustrates the intensity and timing of signals reflected from the surface of a sample during ADE, and certain ADE parameters may be adjusted according to an embodiment.

[0016] [Figure 10A] Figure 10A illustrates a method according to an embodiment by which data may be selected from a set of data such as the set shown in Figure 9A or 9B to generate a graph shown in Figure 11A.

[0017] [Figure 10B] Figure 10B illustrates a method according to an embodiment by which data may be selected from a set of data such as the set shown in Figure 9A or 9B to generate a graph shown in Figure 11B.

[0018] [Figure 10C] Figure 10C illustrates a method according to an embodiment by which data may be selected from a set of data such as the set shown in Figure 9A or 9B to generate a graph shown in Figure 11C.

[0019] [Figure 10D] Figure 10D illustrates a method according to an embodiment by which data may be selected from a set of data such as the set shown in Figure 9A or 9B to generate a graph shown in Figure 11D.

[0020] [Figure 11A]Figures 11A, 11B, 11C, and 11D are graphs showing information collected about ejected droplets according to the embodiment. [Figure 11B] Figures 11A, 11B, 11C, and 11D are graphs showing information collected about ejected droplets according to the embodiment. [Figure 11C] Figures 11A, 11B, 11C, and 11D are graphs showing information collected about ejected droplets according to the embodiment. [Figure 11D] Figures 11A, 11B, 11C, and 11D are graphs showing information collected about ejected droplets according to the embodiment.

[0021] [Figure 12A] Figure 12A illustrates the training of a machine learning model according to an embodiment.

[0022] [Figure 12B] Figure 12B illustrates the use of a trained machine learning model within an ADE system according to an embodiment. [Modes for carrying out the invention]

[0023] The above description of a technique of this application, as well as the following detailed description, will be better understood when read in conjunction with the accompanying drawings. For illustrative purposes, a technique is shown in the drawings. However, it should be understood that the claims are not limited to the arrangements and fixtures shown in the accompanying drawings. Furthermore, the appearance shown in the drawings is one of many decorative appearances that may be adopted to achieve the described function of this system.

[0024] (Detailed explanation) Runtime adjustments to injection parameters may be useful to achieve improved performance in ADE (e.g., improved control of droplet location or volume).

[0025] In conventional systems, user-selected calibrations may be used to influence ADE settings such as the power and frequency for droplet ejection and the volume of the droplets formed. Developing calibration profiles and programming instruments is relatively slow and expensive, potentially requiring thousands of man-hours of work to develop calibrations and adjust individual systems throughout the instrument's life cycle. Furthermore, such calibration profiles can be relatively static and may lose accuracy / availability if system changes such as drift occur over time. Additionally, adjusting ADE parameters during a given transport may require interrupting the droplet ejection stream. This can disrupt the fluid surface behavior, which is set according to the operating ejection frequency, leading to disordered behavior within the system and degraded performance. Pausing the droplet stream can also delay the ADE process and negatively impact the workflow.

[0026] The techniques described herein enable intra-row acoustic measurements between injection rows, which will provide useful information for adjusting ADE settings. This can result in improved throughput and performance.

[0027] The techniques described herein enable real-time monitoring and analysis of the ADE system by collecting performance data during the droplet injection process. The information can be collected in such a way that it enables fast and effective adjustment of ADE operating parameters without excessive delay. According to the techniques described herein, performance data may be collected before, during, and / or after droplet injection. Such data may be used according to machine learning algorithms and conventional heuristic models. This can facilitate the effective adjustment of ADE system operating parameters. Such parameters may include tone burst power, injection frequency, droplet volume, and / or tone burst configuration.

[0028] Figure 1 depicts an exemplary ADE system 100, including a cross-sectional view of a container plate 120 (e.g., a microplate) containing multiple containers 122 (e.g., wells of a microplate) for holding multiple individual samples 101, and a receiving plate 130 containing multiple receiving wells for receiving the injected liquid 102 from the samples 101, as well as a block diagram of electronic equipment 140. The ADE system 100 further includes a transducer assembly 110, a binding fluid 160, an X / Y / Z motor 150, and a temperature sensor (not shown). Figure 2 further shows the transducer assembly 110, including a transducer 112 and an acoustic lens 113. The ADE system 100 can characterize both the containers 122 and the samples 101 and inject the liquid. The sample 101 is the liquid of interest, held in a particular container 122. While this disclosure focuses on a container, such as a microplate well, the techniques described herein can be used in conjunction with other containers, such as tubes, flasks, and beakers, and any samples contained therein, as will be recognized.

[0029] To eject the liquid 102 from the sample 101, the transducer 112 generates acoustic energy (e.g., ultrasonic energy), which is focused into a beam 170 by the acoustic lens 113. In the figure, the beam 170 is shown in two dimensions, but it should be understood that the beam 170 is three-dimensional. Furthermore, the beam 170 is shown as an equilateral triangle, but in practice, the beam 170 can have different shapes. By adapting the configuration of the transducer assembly 110, it may be possible to vary the height of the focal point of the acoustic energy beam 170. For example, the focal length of the beam 170 may be changeable by adapting the transducer assembly 110. Such a transducer assembly 110 is described in U.S. Patent Application No. 16 / 369,780 (U.S. Publication No. 2019 / 0302063) (which is incorporated herein by reference in its entirety). Furthermore, the height of the focal point can also be changed by moving the transducer assembly 110 along the z-axis (i.e., the vertical dimension between the container 122 and the transducer assembly 110).

[0030] As depicted in Figure 1, the beam 170 is focused on the upper surface of the sample 101, which may be at the interface between the sample 101 and the air above it. First, the beam 170 passes through the bonding liquid 160, the bottom wall 123 of the container 122, and then through the depth of the sample 101 to reach the surface 103 of the sample 101.

[0031] The electronic network 140 includes a processor 143, a motor controller 142, a signal transmission network 144, a signal receiving network 145, and a temperature sensor network 141. Although shown as separate components for illustrative purposes, some parts of the electronic device 140 may be combined or integrated. Furthermore, some of the components shown may include multiple different subcomponents that are not specifically shown. For example, the processor 143 may include multiple processors, either located together on a single chip or distributed across different locations.

[0032] The processor 143 generates or controls an analog electrical signal (an electronic transmission signal such as a radio frequency signal) in the signal transmission network 144, which is communicated to the transducer 112. The transducer 112 then vibrates in response to the analog signal (amplitude and frequency) so that a corresponding acoustic signal is emitted. The transducer assembly 110 may also receive an acoustic signal (for example, an acoustic signal reflected from the container 120 and / or sample 101 in response to an emitted acoustic signal) and vibrate in response to it. This vibration may generate an analog electrical signal (an electronically received signal), which is then communicated to the signal receiving network 145. The processor 143 may receive information from the signal receiving network 145 corresponding to the received acoustic signal in the form of an electronically received signal. The information in the electronically received signal will be analyzed by the processor 143.

[0033] The processor 143 can also communicate with the motor controller 142 to control the position of the transducer assembly 110. The motor controller 142 controls one or more of the X / Y / Z motors 150 to move the transducer assembly 110 relative to the container plate 120. The X / Y / Z motors 150 may provide movement of the transducer assembly 110 in three dimensions (e.g., along the X, Y, and Z axes) or in one or two dimensions (e.g., along only the X and Y horizontal dimensions and not along the vertical Z dimension). As shown, the X / Y / Z motors 150 are coupled (directly or indirectly) to the transducer assembly 110, but these or other motors may be coupled (directly or indirectly) to the container plate 120 and / or the receiving plate 130 to control relative movement between the transducer assembly 110, the container plate 120, and / or the receiving plate 130. The processor 143 can control one or more motors 150 to position the transducer assembly 110 directly beneath a given container 122 in the container plate 120 (for example, the transducer assembly 110 is centered on the center of a given container 122), and then move the transducer assembly 110 directly beneath another given container 122 in the container plate 120.

[0034] In some embodiments, the ADE system 100 may include a temperature sensor (not shown) which may be located in the bonding fluid 160, in the area between the container plate 120 and the receiving plate 130, or elsewhere. A temperature sensor network 141 receives signals (e.g., electrical or wireless) from the temperature sensor and communicates with a processor 143 so that the temperatures (e.g., bonding fluid 160, container 122, sample 101, air temperature) can be measured.

[0035] In some embodiments, the transducer assembly 110 may have a cylindrical shape. In some embodiments, instead of using a single transducer 112 to both transmit and receive acoustic signals, the transducer assembly 110 may include separate transmitter and receiver transducers, such as those disclosed in U.S. Patent No. 10,787,670 (which is incorporated herein by reference as a whole). According to one technique, the receiving transducer can substantially surround the transmitting transducer and acoustic lens.

[0036] Figure 3 shows a representation of the movement of the transducer assembly 110 relative to the container plate 120 when performing ADE on multiple samples 101. The transducer assembly 110 moves along the x-axis from container 122 to container 122. The transducer assembly 110 can also move along the y-axis to additional containers 122 (not shown), as further described with respect to Figure 5. For each container 122, the transducer assembly 110 may be centered directly below the container 122. The transducer assembly 110 may move vertically along the z-axis and emit and receive acoustic signals at different z-positions directly below the container 122. The transducer assembly 110 can be positioned along the z-axis to focus the beam 170 onto the surface 103 of the sample 101 and eject the ejected liquid 102 (ADE). As further described below, the transducer assembly 110 is positioned along the z-axis and focuses the beam 170 to a predetermined height relative to the container plate 120, allowing it to break bubbles without performing ADE, for example, at the interface between the container 122 and the sample 101.

[0037] Figure 4 shows a top view of a container plate 120 having multiple containers 122. The container plate 120 shown is a 384-well microplate (e.g., a polypropylene microplate, designated as 384-PP). Figure 5 shows a top view of the multiple container wells 122 and an exemplary pattern (a meandering pattern) for performing ADE and / or other ultrasonic techniques (e.g., bubble breaking) on ​​each container well 122 and the sample 101 therein, as described with respect to Figure 3. In this embodiment, a motor 150 moves the transducer assembly 110 along the x and y axes, positioning it beneath the various container wells 122. Any other suitable pattern may be used (e.g., a raster pattern). It may also be possible to move the container plate 120 or a combination of the container plate 120 and the transducer assembly 110 to achieve a similar effect.

[0038] Figure 6 shows a container 122 (or well) holding sample 101, which has a surface 103 (i.e., an upper surface or free surface). The bottom wall 123 of the container 122 defines the bottom upper interface (TB) 124 and the bottom bottom interface (BB) 125.

[0039] Figure 7 shows the reflected signal and its corresponding envelope, where the reflected signal is received in the transducer assembly 110 in response to the emitted signal, which is emitted by the transducer assembly 110. BB indicates the reflection from BB125. TB indicates the reflection from TB124. SR indicates the reflection from surface 103 of sample 101. The total time over which the reflection appears is called the time of flight or TOF. For each reflection BB, TB, and SR in Figure 7, there is a different, individual TOF.

[0040] Figures 8A, 8B, 8C, and 8D illustrate sequences of acoustic signal transmission and reflection according to embodiments. These figures are depicted as sequences. In Figure 8A, the surface bulge formation phase is initiated by the transmission of a first acoustic signal from a transducer toward the surface 103 of sample 101. The first acoustic signal (and other acoustic signals depicted) may be transmitted by a transducer assembly (not shown in Figures 8A-8D), such as transducer assembly 110. The emitted acoustic energy beam may be substantially focused toward the surface 103. In Figure 8A, the surface 103 may be substantially equilibrium prior to the transmission of the first acoustic signal. The first acoustic signal may be a droplet formation signal, sometimes referred to as a tone burst, but not necessarily. The droplet formation signal may include multiple frequencies and may have a relatively long duration. Examples of such frequencies and durations are disclosed in U.S. Patents 10,112,212 and 10,325,768 (which are incorporated herein by reference as a whole). The droplet formation signal may be sufficient to cause droplets to be ejected from surface 103, but droplet ejection is not required. The droplet formation signal may be sufficient to cause droplets to form, but the droplets do not necessarily separate from surface 103. Surface 103 does not need to be in equilibrium, but is shown as such for illustrative purposes. The energy of the first acoustic signal may be such that surface ridges begin to form within surface 103, for example, as shown in Figure 8B. If the energy of the first acoustic signal is sufficient (or the amount of energy transferred to surface 103 is sufficient), droplets may be ejected. The formation of surface ridges may be the start of a surface ridge phase, which may continue until surface 103 returns to equilibrium. The droplets may or may not be ejected between the surface raised phases.

[0041] In Figure 8B, a second acoustic signal is transmitted toward surface 103, and a portion of the energy of the second acoustic signal is reflected toward a transducer assembly (not shown). The second acoustic signal has less energy than the first acoustic signal. Examples of such energies are disclosed in U.S. Patents 10,112,212 and 10,325,768. The second acoustic signal may have a relatively shorter duration than the first acoustic signal. The second acoustic signal may have a different frequency than the first acoustic signal. For example, the second acoustic signal may not substantially affect the surface ridge formation and behavior. Instead, the second acoustic signal may be a "ping" and may serve to collect information about one or more characteristics of the sample during the surface ridge phase. The second acoustic signal may be reflected by different characteristics in the sample, and the intensity of these reflections (e.g., the measured amplitude of the reflections) may be measured and recorded along with associated time-of-flight information. For example, an ADE system (e.g., system 100) may determine the time of flight between surface 103 and a transducer (e.g., transducer 112) and the intensity of the reflection at surface 103. The time of flight represents the distance from transducer 112. For example, relatively strong reflection (a peak in the reflected signal) may exist from surface 103. This then may allow determining the height of surface 103 at a given location at a given time during the surface elevation phase.

[0042] In addition to time-of-flight and intensity information, reflections from the ping may also include frequency and / or associated phase information. Such information may be extracted, for example, by performing a Fast Fourier Transform (FFT) on the time-domain surface reflections received by the transducer. This FFT returns a power spectrum across a frequency range of approximately 5–20 MHz, which can be used, for example, to extract additional information about the fluid surface geometry and dynamics.

[0043] In addition to reflections from surface 103, Ping's reflections may also indicate reflections from other features in the sample. As will be further discussed, the ADE system may be able to determine characteristics of the surface and droplet ejection based on information determined from one or more Ping's reflections. Such characteristics may include one or more of the following: droplet size, droplet linear velocity, droplet angular velocity, droplet radial displacement, droplet detachment position, droplet surface position, droplet surface velocity, surface acceleration, surface bulge formation rate, or surface bulge relaxation rate.

[0044] In Figure 8C, a third acoustic signal is transmitted toward surface 103, and a portion of the energy of the third acoustic signal is reflected back toward the transducer assembly (not shown). The third acoustic signal may have less energy than the first acoustic signal or similar energy to the second acoustic signal. For example, the third acoustic signal may not substantially affect surface ridge formation and behavior. Instead, the third acoustic signal may be similar to the second acoustic signal, or it may be a ping, and may primarily serve to gather information about one or more properties of surface 103 (and / or droplet formation) during the surface ridge phase. Similar to the second acoustic signal, the ADE system may determine the time of flight between surface 103 and the transducer, the reflected signal intensity, frequency information, and / or phase information related to the frequency information. The third acoustic signal may optionally differ from the second acoustic signal. For example, the third acoustic signal may be varied or adapted from the second acoustic signal depending on the information received in the reflection of the second acoustic signal.

[0045] As shown in Figure 8C, the surface ridge continues to grow within surface 103, and droplets begin to form. The reflection of a third acoustic signal may be processed, and information may be determined about the changing properties of the surface ridge. Additional relatively low-energy acoustic signals (e.g., pings) may be emitted, as well as the second and third acoustic signals. Each of these signals may be emitted at any time at periodic intervals (e.g., 50-100 μS, e.g., 60 μS). For example, in one embodiment, the surface ridge phase is 0.96 mS, and it may be possible to transmit 16 pings at 60 μS intervals. In an embodiment, the duration of the entire reflection from a ping may be about 30-40 μS. Subsequent pings may be emitted after the transducer has received the entire reflection. For example, depending on the duration of the surface ridge phase and the rate at which pings are transmitted, it may be possible to transmit up to 2,000 pings or potentially many more. Second, third, and possible subsequent pings may be used to collect data about surface 103 during the surface elevation phase.

[0046] As shown in Figure 8D, the surface bulging phase ends, and surface 103 returns to equilibrium. Surface 103 does not necessarily have to be in equilibrium, but is shown as such for illustrative purposes. A fourth acoustic signal (e.g., a droplet formation signal or tone burst) is transmitted toward surface 103. The fourth acoustic signal may be similar to the first acoustic signal in that it has enough energy to cause surface bulging (and possibly droplet ejection). However, the fourth acoustic signal may differ from the first acoustic signal based on information determined by the ADE system from the reflection of the second acoustic signal and the reflection of the third acoustic signal (and possibly additional lower-energy acoustic signals). For example, the characteristics of the fourth acoustic signal may be determined at least in part by the characteristics of surface 103 or other aspects of the sample, which are determined in conjunction with the low-energy ping. For example, the first and fourth acoustic signals may have different power, frequency, associated phase, and / or duration from the previous droplet formation signal or tone burst. As a result of different signals, it may be possible to eject droplets with different characteristics at different times during a sequence, such as droplets with different droplet volumes and / or separation rates.

[0047] Figure 9A illustrates the measurements taken during ADE in response to lower energy acoustic signals (e.g., pings), such as the second acoustic signal discussed in the context of Figure 8B and the third acoustic signal discussed in the context of Figure 8C. Each row of pixels shows information determined from low-energy acoustic pings, starting from the top and advancing downwards, transmitted before, during, and after the surface bulge phase. Each row of pixels represents information gathered from a single ping, which is transmitted continuously at 60 μS intervals. Thus, the Y-axis represents the time as the ping is transmitted continuously in sequence downwards along the Y-axis. The X-axis shows the time of flight with respect to the ping. In one particular row where a spike exists in time of flight, it can be seen that a droplet is formed and / or ejected, and the signal extends to the right edge of the graph. Brighter pixels (Z-axis) indicate more reflected energy received from a given ping at a given time in time of flight in the ADE system. Yellow pixels indicate higher reflected energy, and blue pixels indicate lower reflected energy (for example, the darkest blue indicates the lowest reflected energy). As shown, the surface of the surface ridge (indicated by the last significant reflection from a given ping) rises to the point of droplet formation and / or ejection, and then begins to sink as the surface returns to equilibrium. As discussed above, the reflected signal may contain more information, such as frequency and / or associated phase information (with respect to one or more frequencies). This information is not shown in Figure 9A or 9B, but it should be understood that such information can be determined from the reflected signal, stored, and further processed.

[0048] In Figure 9A, the timeline along the Y-axis is discontinuous, resulting in a discontinuous plane towards the top of the graph. This discontinuous plane corresponds to the time when the system emits a power signal (e.g., the first acoustic signal described in Figure 8A) and forms surface ridges. With respect to this interval, it may be impossible to transmit pings because transducers may be used to transmit higher energy signals.

[0049] Figure 9B is similar to Figure 9A, but the measurement conditions are different. Figures 9A and 9B show two embodiments of a method in which a pinch can be used to measure the properties of a sample, particularly surface 103. The data collected from the pinch can be used to develop models, such as machine learning models, which can be used to improve the performance of the instrument.

[0050] Figure 10A illustrates an embodiment of a method by which data captured by the ADE system from Ping reflections can be heuristically assessed. Figure 10A shows a set of data, such as that shown in Figure 9A or 9B, with reflections of two Pings (Ping 6 and Ping 9 along the Y-axis) highlighted. For each reflection, points are designated as point A for the Ping 6 reflection and point B for the Ping 9 reflection. The designations of points A and B in this case are arbitrary and for illustrative purposes only. Points may also be designated for other Ping reflections or other times of flight in a given Ping reflection. Points may be designated before, during, or after surface ridge formation. The graph in the lower left corner of Figure 10A corresponds to the Ping 6 reflection, with amplitude on the Y-axis (corresponding to the Z-axis of the graph above (as depicted in U.S. Provisional Patent No. 63 / 448,075, filed February 24, 2023, incorporated herein by reference)) and time of flight on the X-axis. Point A is again depicted. The graph in the lower right corner of Figure 10A corresponds to the Ping 9 reflection, with the amplitude on the Y-axis (corresponding to the Z-axis of the graph above (as depicted in U.S. Provisional Patent No. 63 / 448,075, filed February 24, 2023, incorporated herein by reference)) and the time of flight on the X-axis.

[0051] The ratio of A to B may be determined not only with respect to one surface elevation phase, but also with respect to thousands, millions, or any number of surface elevation phases deemed reasonable. For example, one or more parameters of the droplet formation signal (e.g., power, duration, frequency, and / or associated phase information) may be varied in a given droplet formation signal, and the A / B ratio may be determined. Furthermore, characteristics of the sample itself, such as the sample composition and / or surface height, may be varied. Again, the designation of A and B may be arbitrary.

[0052] Figure 10B is similar to Figure 10A, except that points C and D are specified here. The ratio of C to D may be determined not only with respect to one surface elevation phase, but also with respect to thousands, millions, or any number of surface elevation phases deemed reasonable. For example, one or more parameters of the droplet formation signal (e.g., power, duration, frequency, and / or associated phase information) may be varied in a given droplet formation signal, and the ratio of C / D may be determined. Furthermore, characteristics of the sample itself, such as the sample composition and / or surface height, may be varied. Again, the designation of C and D may be arbitrary.

[0053] Figure 10C is similar to Figure 10A, except that points E and F are specified here. In this example, points E and F represent data that respond to the same ping, rather than different pings, as shown in Figures 10A and 10B. The ratio of E to F may be determined not only with respect to one surface elevation phase, but also with respect to thousands, millions, or any number of surface elevation phases deemed reasonable. For example, one or more parameters of the droplet formation signal (e.g., power, duration, frequency, and / or associated phase information) may be varied in a given droplet formation signal, and the E / F ratio may be determined. Furthermore, characteristics of the sample itself, such as the sample composition and / or surface height, may be varied. Again, the designation of E and F may be arbitrary.

[0054] Figure 10D is similar to Figure 10A, except that points G and H are specified here. The ratio of G to H may be determined not only with respect to one surface elevation phase, but also with respect to thousands, millions, or any number of surface elevation phases deemed reasonable. For example, one or more parameters of the droplet formation signal (e.g., power, duration, frequency, and / or associated phase information) may be varied in a given droplet formation signal, and the G / hour ratio may be determined. Furthermore, characteristics of the sample itself, such as the sample composition and / or surface height, may be varied. Again, the designation of G and H may be arbitrary.

[0055] With respect to some or each of these surface elevation phases, the properties of the sample surface and / or any ejected droplet may be measured using one or more different systems. For example, a phase Doppler interferometer may measure properties of an ejected droplet, including horizontal velocity on the X-axis substantially parallel to the sample surface, horizontal velocity on the Y-axis substantially parallel to the sample surface, vertical velocity on the Z-axis substantially perpendicular to the sample surface, droplet location (e.g., droplet trajectory, including radial displacement of the droplet), and / or droplet angular velocity. Such a phase Doppler interferometer may employ three positioned lasers and corresponding photodetector arrays. In another embodiment, measurements from reflected ping in an ADE system may measure additional properties such as vertical (Z-axis) droplet detachment position, surface velocity, surface acceleration, surface elevation formation rate, and / or surface elevation relaxation rate. In another embodiment, measurements from a camera (e.g., a visible-wavelength camera) may measure characteristics of ejected droplets and / or surfaces, such as the shape of surface ridges formed by acoustic pressure waves, the shape and trajectory of formed droplets, the presence of formed secondary satellite droplets, and the overall fluid surface behavior in harmony with the tone burst ejection frequency. Figures 11A, 11B, 11C, and 11D show graphs illustrating the correlation between regions of surface ridge phase images, similar to Figures 9A and 9B. Each data point corresponds to a given surface ridge phase evaluation. The data processing (and similar data compilation) associated with the generation in Figures 11A-11D may be performed by a processor such as processor 143 in the ADE system 100. Alternatively, the data processing may be performed by other processors, either in combination with or separately from processor 143. For example, as discussed below, data may be collected from other systems such as a phase Doppler interferometer system and a camera. These other systems may generate data that is stored and processed separately from the ADE system 100. Such separate processing may also receive data generated by the ADE system 100, including acoustic transmission and reception data. Together, this data from different systems may be used to train a machine learning model using one or more algorithms.Once a machine learning model for the ADE system is developed through training, the final model may be stored on the ADE system 100. The model may be accessed or utilized by the processor 143. For example, the processor 143 may be input information associated with the transmission and reception of a first acoustic signal. Based on this information, the machine learning model may suggest different transmission parameters for a subsequent droplet formation signal.

[0056] The graphs in Figures 11A–11D show the correlation of droplet ejection dynamics. For Figure 11A, the X-axis represents the A / B ratio, as defined in Figure 10A. For Figure 11B, the X-axis represents the C / D ratio, as defined in Figure 10B. For Figure 11C, the X-axis represents the D / E ratio, as defined in Figure 10C. For Figure 11D, the X-axis represents the G / H ratio, as defined in Figure 10A. For each graph, the color of the points represents the Z-axis, which is the average power emitted from the RF-pulser over the duration of the droplet formation signal. Blue indicates lower power, while red indicates higher power. The colors are shown in U.S. Provisional Patent No. 63 / 448,075, filed February 24, 2023, which is incorporated herein by reference. More specifically, the power range is indicated by blue, green, orange, and red as power increases. In Figure 11A, the horizontal droplet velocity along the X-axis, which is above and approximately parallel to the sample (i.e., not the X-axis in Figure 11A), is plotted against the A / B ratio. The horizontal droplet velocity along the X-axis for each droplet was measured by a phase Doppler interferometer and expressed in meters / second. In Figure 11B, the horizontal droplet velocity along the Y-axis, which is above and approximately parallel to the sample (i.e., not the Y-axis in Figure 11B), is plotted against the C / D ratio. The horizontal droplet velocity along the Y-axis for each droplet was measured by a phase Doppler interferometer and expressed in meters / second. In Figure 11C, the vertical droplet velocity along the Z-axis, which is above and approximately perpendicular to the sample (i.e., not the Z-axis in Figure 11C), is plotted against the E / F ratio. The vertical droplet velocity along the Z-axis for each droplet was measured by a phase Doppler interferometer and expressed in meters / second. In Figure 11D, the magnitude of the droplet placement error is plotted against the G / H ratio. The ratios in Figures 11A-D may include, but are not limited to, the Ping and time-of-flight constraints shown in Figure 10. The magnitude of the droplet placement error for each droplet was measured by a phase Doppler interferometer and expressed in meters per second. The magnitude of the droplet placement error may be related to the droplet trajectory and may be determined by velocity, for example, as discussed above in the context of Figures 11A, 11B, and 11C.

[0057] As shown, based on the conditions and droplet formation signal, it provides indications of how a given droplet will progress and whether the ejection event will be successful (in terms of droplet location), using a relatively strong correlation (e.g., R). 2 >0.8) may exist.

[0058] Instead of determining the relationship between a limited number of points and measured characteristics, machine learning may be used to model the complex relationships between a large number (e.g., thousands) of different specific data components of the acoustic droplet profile, such as the collected data discussed above in the context of Figures 9A and 9B, and data collected from other systems, including phase Doppler interferometer systems and / or cameras. Such modeling may provide relatively accurate predictions regarding droplet trajectories, for example. The trajectory may be determined by assessing a vertical velocity vector (perpendicular to the transducer plane) and two horizontal orthogonal velocity vectors to determine the magnitude and direction of axial displacement. As the ADE progresses, for example, during a droplet sequence, data from Ping can be collected and processed to improve performance and adapt aspects of the droplet formation signal (e.g., power, frequency, or duration). For example, by assessing data from Ping during a series of droplet ejections, it may be determined that the droplets exhibit undesirable large radial displacements. The ADE system may adjust the droplet formation signal parameters to reduce the degree of radial displacement of the droplets. In another embodiment, by assessing data from Ping during a series of droplet injections, it may be determined that the droplets exhibit undesirable low vertical velocities. The ADE system may adjust the droplet formation signal parameters to increase the vertical velocity of the droplets.

[0059] Figure 12A illustrates a system for training a machine learning model according to an embodiment disclosed herein. During training, the machine learning model 1240 may receive data from an ADE system 1210 (e.g., similar to or identical to ADE system 100), a phase Doppler interferometer 1220, and / or a camera 1230. The ADE system 1210 operates according to the principles discussed herein, and data is generated. This data may include droplet formation signal data (e.g., parameters relating to one or more droplet formation signals), ping transmission data (e.g., parameters relating to one or more pings), and / or ping reflection data. Some or all of this data may be provided to the machine learning model 1240. The machine learning model 1240 may be trained in a system separate from the ADE system 1210, the phase Doppler interferometer 1220, and / or the camera 1230. The phase Doppler interferometer 1220 may collect information corresponding to droplets being ejected during the operation of the ADE system. The phase Doppler interferometer 1230 may provide ejected droplet data to the machine learning model 1240. The camera 1220 may collect information (e.g., visible wavelength information) from the sample surface during the operation of the ADE system 1210. Surface data may be provided to the machine learning model 1240.

[0060] The training of the machine learning model 1240 may be implemented in a separate database and / or processor from that in the data generation system. The machine learning model 1240 may implement one or more convolutional neural networks (CNNs). The CNN may, in effect, scan through the acoustic data and extract some or all of the features of the data generated by the ADE system 1210. CNNs are known to be used in image classification models, in which case they may extract lines, curves, shapes, or colors from images to build a feature set that defines the object they are trying to classify. The machine learning model 1240 may implement a CNN presented with acoustic data to extract features, as well as image features. Examples of such features include contours of acoustic signal density profiles, curvature of the leading or trailing edges of non-zero signal regions, or perpendicularity of pre- and post-extrusion regions.

[0061] The machine learning model 1240 may further include a high-density neural network (DNN), which may receive features extracted by a CNN. The DNN may adjust the weights and biases associated with the extracted features over an iterative process to reduce errors. During model training, the feature set extracted from the acoustic signal profile may correspond to several attributes measured from any combination of systems 1210, 1220, and 1230. For example, if the model is trained to predict droplet vertical velocity as measured within system 1220, thousands of images may be presented to the model during training and "tagged" with vertical velocity values. Each time an image is fed into the model, the CNN may classify it into its features and then adjust the DNN's weights and biases to reduce the overall model error, for example, represented by the mean squared error (MSE). Each subsequent image and vertical velocity tag fed into the model may result in a similar process without impairing the ability to predict previous events. This may be achieved through loss reduction algorithms that allow the model to converge to an optimal state or minimize the MSE, enabling it to confidently predict events across a sample population without favoring any particular event type. This may be referred to as model generalization.

[0062] Figure 12B illustrates a system for implementing a machine learning model 180 within an ADE system 100 according to an embodiment disclosed herein. Once the model is sufficiently trained, the trained model 180 may be stored within the ADE system 100. The trained model 180 may be implemented in software and / or hardware. The trained model 180 may be implemented in software executed by the processor 143 and / or in hardware incorporated by the processor 143. For clarity, Figure 12B shows the trained model 180 outside the processor 143, but this is for illustrative purposes only.

[0063] The processor 143 may initiate an interaction with the transducer 112 according to the principles discussed herein. The processor 143 may transmit information causing the transducer 112 to emit a specific droplet formation signal and ping into the sample. The processor 143 may also receive information reflected from the ping by the sample, which is received from the sample in the transducer 112. The processor 143 may provide the trained model 180 with data (e.g., power, duration, and / or frequency) corresponding to the droplet formation signal and ping. The processor 143 may also process the ping reflection information received from the transducer 112 and provide some or all of the processed data to the trained model 180.

[0064] The trained model 180 may recognize a pattern from the input data and, in response, provide the processor 143 with new droplet formation signal data (e.g., power, duration, and / or frequency). The processor 143 may then generate droplet formation signal information from the droplet formation signal data received from the trained model 180 and provide this droplet formation signal information to the transducer 112 to emit a new droplet formation signal.

[0065] By utilizing the trained model 180, it may be possible to detect changes in droplet velocity and size without using systems such as a phase Doppler interferometer 1220 or camera 1230. For example, the reflection of the ping signal may have a width or amplitude profile that has changed over a certain period of time following droplet ejection, indicating that the droplet is moving towards our target slightly smaller and at a lower velocity. In response to this information, it may be possible to adjust the power of the droplet formation signal to shift the droplet back to a moderately stable state and sustain droplet transport. This can be done in real time without interrupting the droplet line. This may increase the ADE speed while improving performance when processing plates. In another embodiment, surface stability may be evaluated and considered using the trained model 180. During droplet ejection, the droplet formation signal may not be synchronized with the movement of the sample surface. In such cases, relatively high droplet location errors may occur. By using the trained model 180, it may be possible to adjust the rate or timing at which droplet formation signals are emitted by transducer 112, thereby better synchronizing surface movement such as ADE and observed oscillations.

[0066] Figure 11A is a graph showing the ratio of vertical droplet velocity to the amplitude of the acoustic signal. The color of a given point represents the power applied to generate the droplet, with blue representing low power and red representing high power. The colors are shown in US provisional number [number missing]. Figure 11A illustrates a relatively strong correlation between the acoustic signal and the vertical velocity of the droplet. Figure 11B illustrates a similar chart, with the Y-axis representing the magnitude of the velocity vector. The data processing associated with Figures 10A, 10B, 11A, and 11B may be performed, for example, by the ADE system 100 in processor 143.

[0067] Many of the embodiments described herein may be implemented on or in conjunction with a computer storage product, which may include a non-transient computer-readable medium (which may also be referred to as a non-transient processor-readable medium) having instructions or computer code thereon, for performing various computer implementation operations, as will be understood. These embodiments may include those with a processor 143 (or a relevant part of such an embodiment). The medium may include one or more distinctly different mediums. The code may be executed on one or more processors, such as the processor 143 (which may themselves include multiple processors). The computer-readable medium (or processor-readable medium) is non-transient in the sense that it does not itself contain transient propagating signals (e.g., propagating electromagnetic waves that carry information on a transmission medium such as space or cable). The medium and the computer code (which may also be referred to as code) may be designed and constructed for a specific purpose or for a combination of purposes. Examples of non-transient computer-readable media include, but are not limited to, magnetic storage media such as hard disks, floppy disks, and magnetic tapes; optical storage media such as compact discs / digital video discs (CD / DVDs), compact disc-read-only memory (CD-ROMs), and holographic devices; magneto-optical storage media such as optical discs; carrier signal processing modules and application-specific integrated circuits (ASICs); programmable logic devices (PLDs); read-only memory (ROMs); and random access memory (RAM) devices, and other hardware devices specifically configured to store and execute program code. Other embodiments described herein relate to computer program products, which may include, for example, instructions and / or computer code discussed herein. The computer-readable media may include a machine learning model 1240 that is being trained or will be trained. The computer-readable media may include a trained machine learning model 180.

[0068] Some embodiments and / or methods described herein can be implemented by software (running on hardware), hardware (e.g., processor 143), or a combination thereof. Hardware modules may include, for example, general-purpose processors, field-programmable gate arrays (FPGAs), and / or application-specific integrated circuits (ASICs). Software modules (running on hardware) may be C, C++, Java®, Ruby, Visual Basic TM , and / or other object-oriented, procedural, or other programming languages ​​and development tools can be represented in a variety of software languages ​​(e.g., computer code). Embodiments of computer code include, but are not limited to, microcode or microinstructions, machine instructions such as those produced by a compiler, code used to produce web services, and files containing high-level instructions executed by a computer using an interpreter. For example, embodiments may be implemented using imperative programming languages ​​(e.g., C, Fortran, etc.), functional programming languages ​​(e.g., Haskell, Erlang, etc.), logic programming languages ​​(e.g., Prolog), object-oriented programming languages ​​(e.g., Java®, C++, etc.), interpreted languages ​​(Java® Script, Typescript, Perl), or other preferred programming languages ​​and / or development tools. Additional embodiments of computer code include, but are not limited to, control signals, encrypted code, and compressed code.

[0069] It will be understood by those skilled in the art that various modifications may be made without departing from the scope of the novel techniques disclosed herein, and that equivalents may be substituted. In addition, many modifications may be made without departing from the scope to adapt specific situations or materials to the teachings of the novel techniques. Thus, it is intended that the novel techniques are not limited to the specific techniques disclosed, but rather include all techniques that fall within the scope of the appended claims.

Claims

1. A system for ejecting droplets from the surface of a sample using acoustic energy, wherein the system is Processor and A transducer configured to receive electrical energy and emit corresponding acoustic energy, wherein the transducer is further configured to receive acoustic energy and generate corresponding electrical energy. A transmitter that communicates with the processor and transducer, the transmitter being configured to receive at least one signal from the processor and transmit the corresponding electrical energy to the transducer, A receiver that communicates with the processor and transducer, the receiver being configured to receive the electrical energy generated by the transducer and to transmit at least one corresponding signal to the processor. Equipped with, The transducer is configured to emit a first acoustic signal, a second acoustic signal, and a third acoustic signal, wherein the first acoustic signal is determined to cause a surface ridge phase, during which droplets are formed on the surface of the sample; the second acoustic signal is selected to acquire information from the surface of the sample during the surface ridge phase; and the third acoustic signal is selected to acquire further information from the surface of the sample during the surface ridge phase.

2. The system according to claim 1, wherein the second acoustic signal and the third acoustic signal are part of a plurality of information acquisition pings, and the transducer is configured to emit the information acquisition pings at intervals of about 50 to 100 μS.

3. The system according to claim 1 or 2, wherein the plurality of information acquisition pings comprises approximately 4 to 2,000 pings.

4. The system according to any one of claims 1 to 3, wherein the processor is configured to receive information corresponding to the reflection of the second and third acoustic signals from the surface of the sample, and the processor is configured to determine at least one characteristic of the surface raised phase based at least partially on the information.

5. The system according to any one of claims 1 to 4, wherein the at least one characteristic of the surface raised phase is at least one of the following: droplet size, droplet linear velocity, droplet angular velocity, droplet radial displacement, droplet detachment position, surface position, surface velocity, surface acceleration, surface raised formation rate, or surface raised relaxation rate.

6. The system according to any one of claims 1 to 5, wherein the transducer is further configured to emit a fourth acoustic signal, the fourth acoustic signal being determined to cause a second surface elevation phase, a second droplet being ejected from the surface of the sample during the second surface elevation phase, and the processor is further configured to determine the fourth acoustic signal according to at least one characteristic of the surface elevation phase.

7. The system according to any one of claims 1 to 6, wherein the first acoustic signal and the fourth acoustic signal each have at least one of different powers, frequencies, or durations.

8. A system for ejecting droplets from the surface of a sample using acoustic energy, wherein the system is Processor and A transducer configured to receive electrical energy and emit corresponding acoustic energy, wherein the transducer is further configured to receive acoustic energy and generate corresponding electrical energy. A transmitter that communicates with the processor and transducer, the transmitter being configured to receive at least one signal from the processor and transmit the corresponding electrical energy to the transducer, A receiver that communicates with the processor and transducer, the receiver being configured to receive the electrical energy generated by the transducer and to transmit at least one corresponding signal to the processor. Equipped with, The processor is configured to cause the transducer to emit a droplet formation signal and induce a surface elevation phase, wherein droplets are ejected from the surface of the signal during the surface elevation phase, and to emit a plurality of pings at least one of before, during, or after the surface elevation phase. The processor is further configured to receive information from the receiver corresponding to the reflection of the plurality of pings. The system further comprises a processor configured to determine at least one characteristic of the surface raised phase according to the information corresponding to the reflections of the plurality of pings received from the receiver.

9. The system according to claim 8, wherein the processor is further configured to cause the transducer to emit a second droplet formation signal to induce a second surface elevation phase, during which a second droplet is ejected from the surface of the sample, and the processor is further configured to determine the second droplet formation signal based at least partially on the information corresponding to the reflection of the plurality of pings received from the receiver.

10. The system according to any one of claims 8 to 9, wherein the second droplet formation signal is determined according to a machine learning model.

11. The system according to any one of claims 8 to 9, wherein the droplet formation signal and the second droplet formation signal are determined such that the first droplet and the second droplet have at least one of different volumes or separation rates.

12. The system according to claim 8, wherein the at least one characteristic of the surface raised phase is at least one of the following: droplet size, droplet linear velocity, droplet angular velocity, droplet radial displacement, droplet detachment position, surface position, surface velocity, surface acceleration, surface raised formation rate, or surface raised relaxation rate.

13. The system according to claim 8, wherein the transducer is configured to emit the plurality of pings at intervals of approximately 50 to 100 μS.

14. The system according to claim 8, wherein the plurality of pings comprises approximately 4 to 2,000 pings.

15. A method for implementing any one of claims 1 to 14.

Images