Three-layer microfluidic sorting chip for reflecting enhanced focused acoustic field and segmented pulse control method

Abstract

The application relates to the technical field of micro-fluidic chip control, and particularly discloses a three-layer micro-fluidic sorting chip for reflection-enhanced focused acoustic field and a segmented pulse control method. The three-layer micro-fluidic sorting chip adopts a laminated integrated structure, a focused interdigital transducer and an arc-shaped acoustic wave reflector are prepared on a lithium niobate piezoelectric substrate, and three layers of PDMS micro-fluidic channels are bonded through a coupling sealing interface layer; the upper and lower channels are separated by a slope hole array interlayer with an interlaced micro-hole array, and non-contact layered sorting is realized based on the size difference of cells and the response difference of acoustic radiation force; the segmented pulse control method dynamically regulates acoustic field parameters based on a clogging index, and cyclically executes a start calibration, stable sorting and anti-clogging recovery stage. The three-layer micro-fluidic sorting chip and the segmented pulse control method can effectively improve the acoustic energy utilization rate and the sorting stability, reduce the clogging risk of the hole array, and can be applied to label-free cell continuous sorting.

Description

Technical Field

[0001] This invention relates to the field of microfluidic chip control technology, specifically to a three-layer microfluidic sorting chip with reflection-enhanced focusing sound field and a segmented pulse control method. Background Technology

[0002] Microfluidic sorting technology can complete sample transport, cell manipulation, focusing, separation, and detection within microscale channels, offering advantages such as low sample consumption, high structural integration, and ease of integration into online observation and detection systems. Based on the driving mechanism, microfluidic cell sorting methods can be divided into passive and active types: passive sorting relies on differences in cell size, density, morphology, inertia, or pore size for separation, resulting in simple structures and low energy consumption. However, under conditions of high sample concentration, small particle size differences, or continuous operation, it is prone to decreased sorting stability and channel blockage. Active sorting utilizes external fields such as acoustic, electric, magnetic, or optical fields to apply forces to cells, achieving higher precision cell manipulation. Among these, acoustic microfluidics technology has broad application prospects due to its non-contact, label-free, and biocompatible characteristics.

[0003] However, existing acoustic microfluidic sorting chips still have the following shortcomings in terms of structural design: First, conventional focusing interdigital transducers lack an effective acoustic energy convergence and recovery mechanism at the chip structure level. The acoustic energy is severely dispersed during propagation, making it difficult to efficiently concentrate in the sorting area within a limited space, resulting in low overall acoustic energy utilization of the chip. Second, the geometric design of the aperture array and narrow sorting area lacks orderly guidance for cell migration paths. Cells are prone to retention and blockage when passing through the microporous structure, affecting the stability and reliability of the chip under continuous operation. Third, existing chip structures mostly focus on single-layer flow channels or simple planar layouts, lacking flow channel hierarchical structures and guiding structures adapted to the acoustic field focusing area. This makes it difficult to achieve synergistic coupling between the acoustic field action path and the cell sorting path at the structural level, limiting further improvement in sorting performance. Summary of the Invention

[0004] To optimize existing acoustic microfluidic sorting systems, this invention provides a three-layer microfluidic sorting chip with enhanced reflection and focusing sound field, and a segmented pulse control method.

[0005] The present invention adopts the following technical solution: This invention provides a three-layer microfluidic sorting chip with enhanced reflection and focused sound field, comprising a lithium niobate piezoelectric substrate layer, a microfluidic channel layer and a top capping layer stacked sequentially from bottom to top; The lithium niobate piezoelectric substrate layer includes a lithium niobate piezoelectric substrate, and a focusing interdigital transducer and an acoustic wave reflector are disposed on the upper surface of the lithium niobate piezoelectric substrate; the focusing interdigital transducer is an arc-shaped focusing structure, and the acoustic wave reflector is an arc-shaped reflective grating array; The microfluidic channel layer is an integral structure composed of three PDMS sublayers bonded together by alignment. From bottom to top, it includes a lower collection channel layer, a ramp pore array partition layer, and an upper sorting channel layer. The upper sorting channel layer is sequentially connected along the main fluid flow direction, including a sample inlet branch, a mixing section, a horizontal sorting section, a ramp sorting section, and a target cell outlet branch. A buffer inlet branch is located near the sample inlet branch, and one end of the buffer inlet branch is connected to the mixing section. The ramp sorting section is inclined relative to the main flow direction of the horizontal sorting section in the horizontal plane. On the slope hole array partition, a micropore array is provided corresponding to the slope sorting section position, and the micropore array contains a number of micropores. The lower collection channel layer is provided with a collection channel and a background cell outlet; the collection channel is located below the micropore array and is fluidly connected to the upper sorting channel layer only through the micropore array, and is used to collect small background cells that have settled through the micropores; Except for the micropore-connected regions, the remaining areas between the layers are in a liquid-tight isolation state; The top capping layer is sealed to the upper surface of the upper sorting channel layer and has a sample inlet interface, a buffer inlet interface, a target cell outlet interface, and a background cell outlet interface.

[0006] This invention addresses the shortcomings of existing acoustic microfluidic sorting chips at the structural level, such as low acoustic energy utilization, susceptibility to cell retention and blockage in the aperture array, and insufficient design of flow channel layers and guiding structures. It proposes a chip structure design concept of "reflection-enhanced focusing acoustic field coupled with ramp guidance three-layer flow channel." At the acoustic field architecture level, an arc-shaped focusing interdigital transducer is used as the main driving source, and a coaxial arc-shaped acoustic wave reflector is configured outside its acoustic propagation path. The principle of coherent reflection is used to converge the divergent sound waves to the core sorting area. A structured reflective grating array enhances the effective sound pressure intensity at the micro-aperture array, strengthening the acoustic radiation force control efficiency without increasing the driving power. This structurally solves the problems of acoustic energy dispersion and high sorting threshold power in conventional focusing transducers. At the flow channel structure level, a microfluidic architecture with upper-layer sorting and lower-layer collection is constructed. The inclined slope sorting section forms a flow field guiding effect, guiding background cells to migrate orderly to the staggered micropore array. Combined with the difference in acoustic radiation force response of cells of different sizes, efficient stratified sorting is achieved, with target cells retained in the upper layer and background cells settling in the lower layer, reducing the risk of retention and blockage at the pore edge.

[0007] Furthermore, the lithium niobate piezoelectric substrate layer is 128° YX-cut with a thickness of 450 μm to 550 μm; the electrode material of the focusing interdigital transducer is a Cr / Au composite layer, wherein the Cr adhesion layer is 5 nm to 10 nm thick and the Au conductive layer is 80 nm to 150 nm thick; the finger width and adjacent finger spacing of the focusing interdigital transducer are both 85 μm to 90 μm, with a total of 18 to 22 pairs of fingers, a focusing curvature radius of 2 mm to 4 mm, an effective aperture of 1 mm to 3 mm, and a center operating frequency of 10 MHz to 11 MHz after bonding the microfluidic channel layer. MHz; the acoustic reflector is composed of 14 to 16 parallel gratings, the width of a single grating is 1 / 12 to 3 / 12 of the surface acoustic wave wavelength, and the center distance between adjacent gratings is 1 / 4 to 3 / 4 of the surface acoustic wave wavelength; the acoustic wave propagation axis of the focusing interdigital transducer forms an angle of 30° to 45° with the center line of the ramp sorting section in the horizontal plane, and the sound field focus falls on the micropore array, so that the sound field action direction is coupled with the cell migration path.

[0008] Furthermore, the lithium niobate piezoelectric substrate layer and the microfluidic channel layer are bonded and sealed through a coupling sealing interface layer; the thickness of the coupling sealing interface layer is 5 μm to 10 μm, and it is only distributed in the solid support area and non-fluid area on the outer periphery of the channel, without intruding into the internal space of the upper sorting channel layer, the lower collection channel layer and the micropore array; positioning coupling bosses are provided between the layers for precise positioning during the bonding process; the coupling sealing interface layer, the lithium niobate piezoelectric substrate layer and the microfluidic channel layer are all liquid-tightly connected through oxygen plasma activated bonding.

[0009] Furthermore, the upper sorting channel layer has a channel height of 30 μm and a main channel width of 100 μm; the sample inlet branch and the buffer inlet branch are arranged in a bifurcated manner upstream of the channel, and the two merge to form the mixing section; the length of the mixing section is 0.8 mm to 1.2 mm, the length of the horizontal sorting section is 1.0 mm to 1.4 mm, the length of the ramp sorting section is 0.8 mm to 1.2 mm, and it is inclined at 15° to 17° relative to the mainstream direction of the horizontal sorting section in the horizontal plane.

[0010] Furthermore, the thickness of the inclined hole array partition is 14 μm to 16 μm; the micropore array is arranged in 9 to 11 rows, each row containing 4 to 6 circular micropores with a single pore diameter of 8 μm to 10 μm; the center distance between adjacent micropores is 14 μm to 18 μm, and the micropores in adjacent rows are staggered by 7 μm to 9 μm along the channel width direction.

[0011] Furthermore, the height of the collection channel is 25 μm to 35 μm and the width is 90 μm to 110 μm, with its upstream end starting from the starting end of the micropore array; the collection channel, the micropore array, and the ramp sorting section form an overlapping connected area on the horizontal projection, and the overlapping connected area covers the entire micropore array area.

[0012] Furthermore, it also includes a segmented pulse control unit, which is electrically connected to the focusing interdigital transducer and is used to dynamically adjust the acoustic field parameters according to the blockage index B of the micropore array region to realize multi-stage closed-loop control of the sorting process; The blockage index B is obtained by weighted summation of the normalized value of the number of stagnant objects BN, the normalized value of the local gray scale change BG, the normalized value of the flow rate attenuation rate BQ, and the normalized value of the differential pressure increase rate BP. The calculation formula is: B = 0.35BN + 0.25BG + 0.2BQ + 0.2BP. The value range of the blockage index B is 0~1. The larger the value, the higher the blockage risk of the micropore array (701) region.

[0013] The present invention also provides a segmented pulse acoustic field control method applied to the three-layer microfluidic sorting chip, wherein the control process is executed cyclically in the following stages: The control-focusing interdigital transducer starts with low sound pressure and performs frequency scanning within a preset frequency range. Based on the focusing response deviation, sound field response intensity, and micro-pore array retention state, the optimal operating frequency is determined, and then the sound pressure is gradually increased to a stable sorting range; the lower collection channel layer, the ramp hole array partition layer, and the upper sorting channel layer are all included. A pulsed focused sound field is output to the focused interdigital transducer to maintain a stable sorting state in which target cells are retained in the slope sorting section and background cells enter the collection channel through the micropore array; When the blockage index B of the micropore array is detected to be ≥0.60 and lasts for at least two monitoring cycles, the system switches to an intermittent sound field with alternating strong and weak pulses. Strong pulses are used to loosen the residue at the orifice, and weak pulses are used in conjunction with fluid flushing to remove the residue, thereby clearing the micropore array. When the blockage index B < 0.30 and lasts for ≥ 3 s, switch back to the acoustic field parameters of the stable sorting stage and resume continuous sorting.

[0014] Furthermore, the frequency scanning range is 10 MHz to 11 MHz; the focal equivalent peak sound pressure of the focused sound field is 0.2 MPa to 0.4 MPa, the pulse duty cycle is 40% to 80%, and the pulse period is 0.05 s to 1 s.

[0015] Furthermore, the intermittent sound field consists of a strong pulse release phase and a weak pulse flushing phase: The strong pulse release phase is characterized by: a focal equivalent peak sound pressure of 0.45 MPa to 0.50 MPa, a pulse duty cycle of 20% to 40%, a pulse period of 0.05 s to 0.10 s, and a duration of 1 s to 2 s. The weak pulse flushing sub-stage is as follows: the focal equivalent peak sound pressure is 0 MPa to 0.20 MPa, the pulse duty cycle is 0% to 20%, the pulse period is 0.20 s to 0.50 s, and the duration is 1 s to 3 s.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention utilizes a three-layer microfluidic structure consisting of an upper sorting channel, a ramp-hole array partition, and a lower collection channel, combined with a focused sound field enhanced by an acoustic reflector, to achieve a synergistic improvement in sorting efficiency and acoustic energy utilization efficiency. On one hand, the layered sorting mechanism based on differences in cell size and acoustic radiation force response, coupled with the geometric guidance of the ramp-hole array, guides background cells to migrate orderly into the micropore region and enter the lower collection channel, reducing disordered retention at the pore edges. Compared to a horizontal pore array structure, this significantly improves the continuity of background cell passage and the stability of target cell retention. On the other hand, an arc-shaped acoustic reflector is added to the side of the focusing interdigital transducer, which coherently reflects and refocuses divergent sound waves into the ramp-hole array region. This ensures that the peak sound field energy is precisely applied to the core sorting region, achieving synchronous coupling of acoustic radiation force with vertical layering, ramp guidance, and micropore passage processes. This improves effective acoustic energy utilization under the same driving power and reduces the driving power required to reach the sorting threshold.

[0017] The segmented pulse acoustic field control method proposed in this invention balances the start-up stability, separation efficiency, and long-term anti-clogging capability of the sorting process through multi-stage dynamic adjustment of acoustic field parameters. Unlike the constant parameter drive mode of a fixed continuous acoustic field, this method can sequentially execute working stages such as start-up calibration, stable sorting, and anti-clogging recovery according to the sorting process: In the start-up stage, the optimal operating point is determined by frequency scanning to ensure initial focusing accuracy and sorting stability; in the stable sorting stage, efficient layered sorting is maintained by pulsed acoustic fields; when the micropore array is detected to be clogging-prone, it automatically switches to intermittent acoustic field disturbances with alternating strong and weak pulses. The synergistic effect of strong pulses loosening retained materials and weak pulses combined with fluid flushing effectively reduces the risk of pore array clogging and improves the stability of continuous chip operation. At the same time, the pulsed drive and segmented control mechanism can reduce the long-term acoustic exposure dose of cells and inhibit local temperature rise in the channels, which is beneficial to maintaining the biological activity of sorted cells.

[0018] This invention establishes a MATLAB / Simulink-based equivalent simulation verification system, which can quantitatively predict sound field distribution, cell forces and motion trajectories, blockage index evolution, and the effectiveness of control strategies before chip fabrication and biological experiments. The simulation model covers core modules such as sound field equivalence, cell motion, blockage evolution, and segmented control. It can intuitively compare sorting efficiency, outlet purity, blockage rate, and activity retention levels under different structural schemes and control logics, providing a theoretical basis for chip structure parameter optimization, sound field modulation strategy calibration, and experimental condition selection. This effectively reduces physical prototyping and experimental trial-and-error costs, improving R&D efficiency and scheme feasibility. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the three-layer microfluidic chip structure of the present invention.

[0021] Figure 2 This is a schematic diagram of the upper sorting channel layer structure of the three-layer microfluidic chip of the present invention.

[0022] Figure 3 This is a schematic diagram of the ramp hole array separator structure of the three-layer microfluidic chip of the present invention.

[0023] Figure 4 This is a schematic diagram of the lower collection channel layer structure of the three-layer microfluidic chip of the present invention.

[0024] Figure 5 The curves show the normalized vertical position changes of cells under different sound field control methods.

[0025] Figure 6 The curves show the blockage index variation of the aperture array under different sound field control methods.

[0026] Figure labeling: 1-Lithium niobate piezoelectric substrate layer; 2-Focusing interdigital transducer; 3-Acoustic reflector; 4-Coupling sealing interface layer; 5-Microfluidic channel layer; 6-Lower collection channel layer; 601-Collection channel; 602-Background cell outlet; 7-Sloping pore array partition; 701-Micropore array; 7011-Micropore; 702-Outlet; 8-Upper sorting channel layer; 801-Sample inlet branch; 802-Buffer inlet branch; 803-Mixing section; 804-Horizontal sorting section; 805-Sloping sorting section; 806-Target cell outlet branch; 9-Top capping layer; 901-Sample inlet interface; 902-Buffer inlet interface; 903-Target cell outlet interface; 904-Background cell outlet interface. Detailed Implementation

[0027] The specific embodiments of the present invention are described in detail below, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Unless otherwise specified, the experimental methods described in the embodiments of the present invention are conventional methods, and the materials and reagents used in the following embodiments are commercially available unless otherwise specified.

[0028] Example 1: A three-layer microfluidic sorting chip with reflection-enhanced focused sound field and a segmented pulse control method.

[0029] This embodiment uses the label-free continuous sorting of A549 tumor cells and erythrocytes as an application scenario to provide a complete explanation of the chip structure, fabrication process, sorting operation, and segmented pulse acoustic field control process.

[0030] I. Chip Structure and Core Parameters The three-layer microfluidic sorting chip, from bottom to top, includes a lithium niobate piezoelectric substrate layer 1, a microfluidic channel layer 5, and a top capping layer 9 (see structure). Figures 1-4 ).

[0031] 1. Lithium niobate piezoelectric substrate 1 The lithium niobate piezoelectric substrate 1 includes a lithium niobate piezoelectric substrate, on which a focusing interdigital transducer 2 and an acoustic wave reflector 3 are integrally formed, together constituting an acoustic field driving unit. The focusing interdigital transducer 2 has an arc-shaped focusing structure, with a finger width and adjacent finger spacing of 87 μm, a total of 20 pairs of fingers, a focusing curvature radius of 3.0 mm, an effective aperture of 2.0 mm, and a center operating frequency of 10 MHz to 11 MHz after loading. It is driven by a signal generator and a power amplifier (an external supporting driving device for the chip, electrically connected to the focusing interdigital transducer through electrode leads). The acoustic reflector 3 is an arc-shaped reflective grating array. Its curvature center coincides with the sound field focus of the focusing interdigital transducer 2. It consists of 15 parallel gratings. The width of a single grating is 1 / 6 of the surface acoustic wave wavelength, and the center distance between adjacent gratings is 1 / 2 of the surface acoustic wave wavelength. It can coherently reflect the divergent sound waves emitted by the transducer and refocus them into the focal area, thereby improving the acoustic energy utilization rate of the sorting area.

[0032] 2. Microfluidic channel layer 5 The microfluidic channel layer 5 consists of a lower collection channel layer 6, a ramp hole array partition layer 7, and an upper sorting channel layer 8 from bottom to top. The three-layer structure is aligned and bonded after being activated by oxygen plasma. Except for the connected regions of the micropore array, the remaining regions between the layers are in a liquid-tight isolation state.

[0033] The upper sorting channel layer 8 is located at the top of the microfluidic channel layer 5, with a channel height of 30 μm and a main channel width of 100 μm. Along the main fluid flow direction, it is sequentially divided into a sample inlet branch 801, a buffer inlet branch 802, a mixing section 803, a horizontal sorting section 804, a ramp sorting section 805, and a target cell outlet branch 806. The sample inlet branch 801 and the buffer inlet branch 802 are arranged in a bifurcated pattern upstream of the channel, merging to form a 1.0 mm long mixing section 803, used for uniform mixing of the sample and buffer. Downstream of the mixing section 803 is a 1.2 mm long linear horizontal sorting section 804, used for cell pre-focusing and stable transport. Downstream of the horizontal sorting section 804 is a 1.0 mm long linear horizontal sorting section 804. The slope sorting section 805, which is inclined at 16° relative to the mainstream direction of the horizontal sorting section in the horizontal plane, is used to guide the background cells to migrate to the slope hole array area; the downstream of the slope sorting section 805 is connected to the target cell outlet branch 806, which is used to output the large-sized target cells that have been sorted and retained in the upper layer.

[0034] The ramp-hole array partition 7 is a 15μm thick PDMS solid partition, horizontally sandwiched between the upper sorting channel layer 8 and the lower collection channel layer 6, forming a physical isolation structure between the two channels. The ramp-hole array partition 7 is provided with a micropore array 701 penetrating the partition. The micropore array 701 consists of 10 rows, each row containing 5 circular micropores 7011, each with a pore diameter of 9μm. The center-to-center distance between adjacent micropores 7011 is 16μm, and adjacent rows of micropores 7011 are staggered by 8μm along the channel width direction. The effective length of the entire micropore array 701 in the main flow direction is 153μm, and the effective width in the channel width direction is 81μm. The ramp-hole array partition 7 is provided with an outlet 702 in the main flow direction.

[0035] The lower collection channel layer 6 is located at the bottom of the microfluidic channel layer 5, and has a collection channel 601 and a background cell outlet 602. The collection channel 601 is 30 μm high and 100 μm wide, and its upstream end starts from the starting end of the micropore array 701. The background cell outlet 602 is connected to the outlet 702. The lower collection channel layer 6 has no independent sample inlet, but is fluidly connected to the upper sorting channel layer 8 only through the micropores 7011, which is used to collect small-sized background cells that have settled through the micropores. The collection channel 601, the micropore array 701, and the ramp sorting section 805 form an overlapping connected area in the horizontal projection, and are basically aligned with the acoustic field focal area of ​​the focusing interdigital transducer 2.

[0036] The lithium niobate piezoelectric substrate 1 and the microfluidic channel layer 5 are connected by a coupling sealing interface layer 4. The coupling sealing interface layer 4 is a PDMS flexible film with a thickness of 5 μm to 10 μm. Its lower surface is bonded tightly to the upper surface of the lithium niobate piezoelectric substrate 1 using an oxygen plasma activated bonding process. The upper surface is bonded to the lower surface of the microfluidic channel layer 5 using the same process. The coupling sealing interface layer 4 is only distributed in the physical support area and non-fluid region around the channel, without intruding into the internal space of the upper sorting channel layer 8, the lower collection channel layer 6, and the micropore array 701. This ensures efficient coupling and transmission of acoustic waves from the piezoelectric substrate to the liquid within the channel, while also achieving liquid sealing at the bottom of the chip, eliminating the risk of fluid leakage. Positioning coupling protrusions between the layers are used for precise positioning during the bonding process, ensuring the alignment accuracy between the acoustic field focus and the ramp hole array region.

[0037] 3. Top capping layer 9 The top capping layer 9 is bonded to the upper surface of the upper sorting channel layer 8 to provide a top liquid seal for the chip. Corresponding to the sample inlet branch 801, buffer inlet branch 802, and target cell outlet branch 806 of the upper sorting channel layer 8, the top capping layer 9 has sample inlet interface 901, buffer inlet interface 902, and target cell outlet interface 903, respectively. It also has a background cell outlet interface 904 connected to background cell outlets 602 and 702. Each interface is used to connect to external devices such as precision injection pumps and cell collection containers.

[0038] II. Chip Fabrication Process This embodiment employs SU-8 lithography, PDMS layering, alignment bonding, and piezoelectric substrate integration to fabricate a three-layer microfluidic sorting chip. The three-layer microfluidic channels are formed by separately molding the upper sorting channel layer 8, the ramp hole array separator layer 7, and the lower collection channel layer 6, and then aligning and bonding them together, rather than being cast as a single monolithic process.

[0039] 1. Microstructure mold preparation SU-8 molds were fabricated on a silicon wafer, consisting of an upper sorting channel layer 8, a ramp hole array separator layer 7, and a lower collecting channel layer 6.

[0040] Upper sorting channel layer 8 mold: SU-8 negative photoresist is spin-coated, and a channel protrusion structure with a height of 30 μm is formed by photolithography. It includes a sample inlet branch 801, a buffer inlet branch 802, a mixing section 803, a horizontal sorting section 804, a ramp sorting section 805, and a target cell outlet branch 806. The target cells are A549 cells. The ramp sorting section 805 is a 1 mm long, inclined guide section with a 16° angle relative to the mainstream direction in the horizontal projection plane. This angle is directly defined by the mask planar pattern.

[0041] The sloping hole array separator 7 mold uses a solid PDMS separator with a molding thickness of 15 μm. Micropores 7011 are achieved through mold boss transfer: cylindrical SU-8 bosses, complementary to the micropore array 701, are fabricated on the sloping hole array separator 7 mold. The bosses have a diameter of 9 μm, a center-to-center distance of 16 μm, and a height consistent with the thickness of the sloping hole array separator 7. After PDMS demolding, circular micropores 7011 penetrating the separator are formed at corresponding positions, allowing the upper sorting channel layer 8 and the lower collection channel layer 6 to be connected only through the micropore array 701.

[0042] The lower collection channel layer 6 mold is also formed with SU-8 to form a channel protrusion with a height of 30 μm. It overlaps with the ramp sorting section 805 and the micropore array 701 in the horizontal projection. Downstream, it is connected to the background cell outlet 602, and the background cells are red blood cells.

[0043] 2. PDMS channel molding PDMS prepolymer and curing agent were mixed at a mass ratio of 10:1 and stirred for 5 min until homogeneous. The mixture was then degassed under vacuum at -0.09 MPa for 25 min until no visible bubbles remained. The degassed PDMS was poured into molds for the upper sorting channel, lower collection channel, and ramp hole array partition layer, respectively. After pouring, the mixture was degassed again under vacuum for 10 min, then cured at 95℃ for 1 h. After cooling and demolding, the mixture was cleaned with clean tape, rinsed sequentially with deionized water and anhydrous ethanol, and dried with nitrogen. If necessary, it was dried at 55℃ for 15 min to obtain the upper sorting channel layer 8, ramp hole array partition layer 7, and lower collection channel layer 6. The permeability of the micropores 7011 and any debris or collapse were examined under a microscope. If debris was present, it was gently rinsed with deionized water or buffer solution; mechanical scraping was prohibited.

[0044] 3. Layer alignment and bonding Alignment marks are pre-placed on the outer periphery of the upper sorting channel layer 8, the ramp hole array partition layer 7, and the lower collection channel layer 6. Under a microscope, the micropore array 701, the ramp sorting section 805, and the collection channel 601 are aligned in the horizontal plane. After the bonding surfaces are activated by oxygen plasma (power 10 W, time 60 s), the lower collection channel layer 6, the ramp hole array partition layer 7, and the upper sorting channel layer 8 are bonded together in sequence, and placed at 70℃ for 20 min to enhance the bonding strength. After bonding, the upper sorting channel layer 8 and the lower collection channel layer 6 are connected only by the micropore array 701, with the remaining areas liquid-tightly isolated.

[0045] In accordance with the liquid-tight design requirements of the stacked structure, the three-layer PDMS structure after bonding must meet the following requirements: the upper sorting channel layer 8 and the lower collection channel layer 6 are only fluidly connected through the micropore array 701, and the remaining areas are completely isolated from the circumferentially bonded sealing area by the PDMS solid to prevent undesigned liquid cross-contamination; the thickness of the ramp pore array partition layer 7 and the arrangement of the micropores 7011 must be precisely matched with the geometric position of the upper sorting channel layer 8 and the lower collection channel layer 6 to take into account cell guidance, micropore connectivity and liquid-tight isolation effects.

[0046] 4. Fabrication of the sound field driving unit A focusing interdigital transducer 2 and an acoustic wave reflector 3 were fabricated on a 128° YX-cut, 500 μm thick lithium niobate piezoelectric substrate 1. To achieve efficient acoustic energy utilization and sorting path coupling, the focusing interdigital transducer 2 and the acoustic wave reflector 3 were fabricated according to the following design parameters:

[0047] The focusing interdigital transducer employs an arc-shaped focusing structure with the ramp-shaped aperture array region as the common focal point. The electrode material is a Cr / Au composite layer, with a Cr adhesion layer thickness of 5 nm to 10 nm and an Au conductive layer thickness of 80 nm to 150 nm. The finger width and the spacing between adjacent fingers are both 87 μm, with 20 pairs of fingers, a focusing curvature radius of 3.0 mm, and an effective aperture of 2.0 mm. After bonding the PDMS channel layer, the theoretically calculated center frequency is 10 MHz to 11 MHz. Its acoustic wave propagation axis forms an angle of 30° to 45° with the center line of the ramp sorting section 805 in the horizontal plane. The focal point of the focused acoustic field falls in the middle and lower reaches of the micropore array 701, causing the acoustic field action area to intersect with the cell migration path, thereby achieving coupling of acoustic radiation force, fluid shear force, and the direction of cell migration towards the micropore array, improving the sorting and control accuracy.

[0048] The acoustic wave reflector 3 is an arc-shaped reflective grating array, with the center of curvature coinciding with the focal point of the sound field. It is made of the same material as the transducer, consisting of 15 gratings, each grating being 1 / 6 the wavelength wide and the distance between adjacent centers being 1 / 2 the wavelength. The acoustic wave reflector 3 can coherently reflect divergent sound waves, causing them to refocus onto the micro-aperture array 701. This secondary focusing effect enhances the utilization rate of acoustic energy and reduces the driving power required to reach the sorting threshold.

[0049] The focusing interdigitated transducer and acoustic reflector can be fabricated using standard micro / nano fabrication processes. Specifically, they can be fabricated in-house under cleanroom conditions or outsourced to a third-party fabrication platform with micro / nano fabrication capabilities, following the same layout and process parameters. The specific steps are as follows:

[0050] S1. Substrate Cleaning: A 128° YX-cut lithium niobate piezoelectric substrate with a thickness of 500 μm was selected as the acoustic field driving substrate. The substrate surface was ultrasonically cleaned sequentially with acetone, isopropanol, and deionized water for 7±3 min each step. After cleaning, the substrate was dried with nitrogen and baked on a hot plate at 110±10℃ for 7±3 min to remove residual moisture.

[0051] S2. Photoresist Coating: Place the cleaned lithium niobate substrate on a spin coater and spin coat positive photoresist onto its surface. The spin coating speed is 4000±1000 rpm and the spin coating time is 45±15 s, forming a photoresist layer with a thickness of 1 μm to 2 μm. Then, soft bake at 100±10℃ for 90±30 s.

[0052] S3. Mask Exposure: Ultraviolet exposure is performed using a pre-designed electrode mask, which includes the electrode pattern of the focusing interdigital transducer 2 and the grating pattern of the acoustic reflector 3. The focusing interdigital transducer 2 adopts an arc-shaped focusing structure with the micro-aperture array 701 as the common focus, and the finger width and finger spacing are both 87 μm. The acoustic reflector 3 is an arc-shaped reflective grating array, and its curvature center coincides with the sound field focus.

[0053] S4. Development: Place the exposed substrate in a developer solution that matches the photoresist used and develop for 60±30 s until the electrode area pattern is fully opened. After development, rinse with deionized water and dry with nitrogen gas. Then check under a microscope whether the interdigital electrode gap, reflective grid pattern and lead pads are intact.

[0054] S5. Metal Deposition: A Cr / Au composite metal layer is deposited on the developed substrate surface using electron beam evaporation or magnetron sputtering. Cr serves as an adhesion layer with a thickness of 5 nm to 10 nm, while Au serves as a conductive layer with a thickness of 80 nm to 150 nm. The substrate temperature is controlled during deposition to prevent excessive hardening of the photoresist, which could affect subsequent stripping.

[0055] S6. Stripping: Immerse the metal-deposited substrate in acetone for stripping for 20±10 min. If necessary, supplement with low-power ultrasonic treatment to remove the metal film in non-electrode areas along with the photoresist, leaving only the focused interdigital transducer, acoustic reflector, and their lead pad patterns. After stripping, clean with isopropanol and deionized water sequentially, and then dry with nitrogen gas.

[0056] S7. Electrode Inspection and Testing: After completing the above electrode preparation, use an impedance analyzer or network analyzer to test the frequency response of the focused interdigital transducer. The test frequency range is 10 MHz to 11 MHz. After the PDMS microfluidic channel layer is bonded to the lithium niobate substrate, test the frequency response again after loading PDMS to determine the actual operating center frequency and driving voltage range.

[0057] 5. Chip integration and bonding The lithium niobate piezoelectric substrate 1 and the microfluidic channel layer 5 are integrated and bonded through the coupling sealing interface layer 4. This layer is the core structure in the stacked structure that realizes acoustic energy transmission and bottom liquid sealing. It is a PDMS flexible film with controlled thickness, which is only distributed in the solid support area and non-fluid area on the outer periphery of the channel and does not invade the interior of the flow channel.

[0058] Preparation of coupling sealing interface layer 4: PDMS prepolymer and curing agent (PDMS prepolymer is addition-type polydimethylsiloxane prepolymer, curing agent is SYLGARD 184 Silicone Elastomer Curing Agent) were weighed at a mass ratio of 10:1 and placed in a clean centrifuge tube or beaker. The mixture was stirred in the same direction for 5 min until homogeneous. The mixture was then placed in a vacuum dryer or vacuum degassing tank and degassed at -0.09 MPa for 25 min until no visible bubbles remained. The degassed PDMS mixture was spin-coated onto the non-electrode lead area and the corresponding area of ​​the microfluidic channel layer 5 on the surface of the lithium niobate piezoelectric substrate 1 at a spin-coating speed of 3000 rpm / min for 45 ± 15 s to form a PDMS film with a thickness of 5 μm to 10 μm. Pre-curing was then carried out at 70 °C for 10 min to maintain the PDMS film in a semi-cured, adhered state.

[0059] After pre-curing, the surface of the PDMS film and the bottom surface of the microfluidic channel layer 5 are subjected to oxygen plasma activation treatment. The specific steps are as follows: First, the surface of the PDMS film and the bottom surface of the microfluidic channel layer 5 are purged with nitrogen to remove surface dust; then, the surfaces to be bonded are placed face up into the oxygen plasma cleaning chamber; after evacuating to the equipment's working pressure, oxygen is introduced at a flow rate of 35±15 sccm; treatment is carried out at a radio frequency power of 10 W for 60 s; after treatment, the film is removed within 1 min, and the surface of the PDMS film is aligned and bonded to the bottom surface of the microfluidic channel layer 5, ensuring that the micropore array 701, the ramp sorting section 805, and the collection channel 601 are aligned in the horizontal plane; a uniform pressure of 10 kPa is maintained for 3±2 min, followed by heat treatment at 70℃ for 20 min to enhance bonding strength and form a liquid-tight sealing interface. The coupling sealing interface layer 4 does not enter the upper sorting channel layer 8, the lower collection channel layer 6, or the interior of the micropore array 701 to avoid clogging the fluid channels.

[0060] III. Cell sorting operation procedure 1. Sample preparation and injection control The sorting material was a mixed suspension of A549 cells and erythrocytes, with a ratio of 1:1000 and an A549 cell concentration of 1×10⁻⁶. 5 cells / mL, red blood cell concentration 1×10 8 cells / mL.

[0061] The suspension medium used was PBS (pH 7.3±0.1, osmotic pressure 300±20 mOsm / kg, dynamic viscosity 1.1±0.2 mPa·s at 22±2℃) to reduce non-specific cell adhesion on the PDMS channels and the surface of the microporous array 701.

[0062] The mixed cell suspension and buffer solution were injected into the sample inlet port 901 and the buffer inlet port 902 respectively using a precision injection pump, with equal flow rates of 0.045 μL / min for both sample and buffer solutions.

[0063] 2. Sound field driving and parameter calibration A 10 MHz to 11 MHz sine wave signal is generated by a signal generator, amplified by a power amplifier, and then applied to a focusing interdigital transducer 2 to generate a focused sound field, so that the focused area of ​​the sound field coincides with the micro-hole array 701.

[0064] To ensure accurate and controllable sound pressure parameters during the sorting process, sound pressure calibration must be completed beforehand. The sound pressure intensity adjustment range is 0.1 MPa to 0.5 MPa, defined as the equivalent peak sound pressure at the liquid focal point within the 701 channel of the microporous array, not the sound pressure on the substrate surface. The calibration method is as follows: under the same operating conditions as the chip (frequency, driving voltage, liquid medium, geometric position), the sound pressure at the liquid focal point is directly measured using a miniature hydrophone, or the normal vibration displacement of the substrate surface is measured using a laser Doppler vibrometer. The equivalent sound pressure inside the channel is then calculated using an acoustic-structure coupling finite element model, thereby establishing a calibration relationship between the driving voltage (or input power) and the focal point sound pressure. During the sorting process, the driving voltage is adjusted according to this calibration relationship to maintain the equivalent peak sound pressure at the focal point within the target range. Under stable sorting conditions, the default setting for the equivalent peak sound pressure at the focal point is 0.3 MPa.

[0065] 3. Closed-loop control of the sorting process The microfluidic chip of this invention is externally configured with a segmented pulse control unit, which is electrically connected to the focusing interdigital transducer 2. This unit can dynamically adjust the acoustic field parameters based on the clogging status of the micropore array 701, achieving closed-loop control of the sorting process. The segmented pulse control unit uses a clogging index B to achieve multi-stage acoustic field parameter adjustment and operating state switching. The value of B ranges from 0 to 1, comprehensively reflecting the degree of cell retention, micropore occlusion, flow rate attenuation, and pressure difference increase in the micropore array 701; a higher value indicates a higher risk of clogging.

[0066] Using the slope micro-orifice array region as the monitoring ROI, after the sorting stabilizes, a baseline image I0 and a baseline flow rate are acquired. Q 0 and reference pressure difference Δ P 0. During operation, calculate the following characteristic quantities within the sliding time window T:

[0067] Number of stranded objects Nr Cells or aggregates that remain continuously within the ROI for ≥2 s and have a centroid velocity <5 μm / s are considered to be objects stuck at the pore opening if they continuously overlap with the edge of the micropore. Local grayscale change Δ GThe average grayscale difference or change in occlusion area between the current image and I0 is used to characterize the degree of cell aggregation, retention, and occlusion at the wellhead in the well array region. Flow attenuation rate R Q Calculate using the following formula: , in, Q t For the outbound flow within the current time window, when Q t ≥ Q At 0 o'clock, R Q Set to 0. The flow rate attenuation refers to the change in outlet flow rate relative to the reference flow rate, under the condition that the set flow rate of the injection pump remains constant. Q The percentage of 0 decreases;

[0068] Pressure difference rise rate R P Calculate using the following formula: , Where, Δ P t Δ represents the pressure difference between the upstream and downstream sides within the current time window. P max This is the pre-calibrated pressure differential under severe blockage conditions. When Δ P t Less than or equal to Δ P At 0 o'clock, R P Take 0; when Δ P t ≥Δ P max hour, Take 1.

[0069] After normalization, each blockage characteristic is denoted as follows: BN , BG , BQ and BP ,in: , , , , N ref in, N ref To determine the number of objects trapped under a pre-defined clogging condition, it is preferable to take 30% of the total number of micropores in the pore array; Δ G refThis is a pre-calibrated reference value for local grayscale changes under clogging conditions.

[0070] Congestion index B is calculated using a weighted summation method: B=0.35 BN +0.25 BG +0.2 BQ +0.2 BP .

[0071] When no flow or differential pressure sensor is configured, only existing features are enabled, and the corresponding weights are proportionally normalized to a sum of 1. State switching logic: B < 0.30, low-clogging state, maintaining stable sorting parameters; 0.30 ≤ B < 0.60, controllable retention state, continuing monitoring; B ≥ 0.60 and persists for at least two monitoring cycles, entering the anti-clogging recovery phase, switching acoustic field parameters; during the recovery phase, if B < 0.30 and persists for ≥ 3 seconds, the recovery phase can be exited, returning to stable sorting.

[0072] 4. Sorting and collecting Due to their large size and strong acoustic radiation response, A549 cells remain stably in the upper sorting channel under the influence of the acoustic field and are eventually collected from the target cell outlet interface 903. Red blood cells, with their small size and weak acoustic radiation response, enter the lower collection channel layer 6 through the micropore array 701 under the guidance of gravity and the ramp, and are discharged from the background cell outlet 602, thus achieving non-contact continuous sorting.

[0073] IV. Segmented Pulse Sound Field Control Process The control process is executed cyclically in four stages, with the specific parameters and logic as follows: 1. Start the calibration phase The transducer starts with low sound pressure, with an equivalent peak sound pressure of 0.1 MPa to 0.2 MPa at the focal point. Frequency scanning is performed in the range of 10 MHz to 11 MHz with a step size of 0.02 MHz and a dwell time of 3 s at each frequency point.

[0074] During the dwell time at each frequency point, image sequences of the slope aperture array region are acquired, and the lateral / vertical displacement of cells or tracer particles, focusing response intensity, and aperture array dwell time are recorded. The equivalent sound pressure response at the focal point is obtained by combining the results of hydrophone or laser Doppler vibration calibration.

[0075] The optimal operating point is determined according to the following quantitative criteria: Distance deviation d: The distance between the center of the focused response and the geometric center of the ramp hole array; Sound field response index M: The average displacement of cells or tracer particles in the predetermined sorting direction per unit time; Blockage Index B: A pore array patency index.

[0076] Candidate operating points must simultaneously satisfy d ≥ 20 μm, B < 0.30, and no obvious cell aggregation. Among the candidate points, the frequency with the most stable focal equivalent sound pressure is selected as the optimal operating frequency f0; if multiple candidate points exist, the frequency closest to 10 MHz to 11 MHz with the smallest d is preferred.

[0077] After determining f0, maintain this frequency and gradually increase the sound pressure from 0.1 MPa to 0.2 MPa to 0.2 MPa in the stable sorting stage, and then enter the stable sorting stage.

[0078] 2. Stable sorting stage A pulsed focused sound field is output to the FIDT with the following parameters: sound pressure level 0.2 MPa to 0.4 MPa, pulse duty cycle 40%–80%, and pulse period 0.05 s to 1 s. This stage maintains a stable sorting state where the target cells are held in the upper layer and the background cells pass through the lower layer.

[0079] 3. Anti-congestion recovery phase When the blockage index B ≥ 0.60 and persists for at least two monitoring cycles, the system automatically enters the anti-blockage recovery phase. This phase clears the blockage through intermittent sound field disturbances with alternating strong and weak pulses. Each recovery cycle consists of a strong pulse loosening sub-phase and a weak pulse flushing sub-phase.

[0080] Strong pulse detachment: The equivalent peak sound pressure at the focal point is 0.45 MPa to 0.50 MPa, the pulse duty cycle is 20% to 40%, the pulse period is 0.05 s to 0.10 s, and the duration is 1 s to 2 s. The transient high sound pressure disturbance causes the cells or aggregates trapped at the orifice to detach.

[0081] Weak pulse flushing: The focal equivalent peak sound pressure is 0 MPa to 0.20 MPa, the duty cycle is 0% to 20%, the pulse period is 0.20 s to 0.50 s, and the duration is 1 s to 3 s. Under continuous injection conditions by the syringe pump, fluid flushing is used to remove the detached object and reduce the average acoustic exposure dose.

[0082] After each recovery cycle, B is recalculated: if B < 0.30 and remains so for ≥ 3 s, the recovery phase is exited, and the stable sorting parameters are switched back; if 0.30 ≤ B < 0.60, weak pulse flushing continues while monitoring is maintained; if B ≥ 0.60, the recovery cycle is repeated. The sound pressure amplitude, duty cycle, and cycle adjustment do not exceed the calibrated operating range, and the sound pressure does not exceed 0.50 MPa.

[0083] 4. Resumption of stable sorting stage When B < 0.30 and remains at ≥ 3 s, the system automatically switches back to the sound field parameters of the stable sorting stage and resumes continuous sorting; if B is in the range of 0.30 to 0.60, the current control is maintained and monitoring continues until the recovery conditions are met.

[0084] V. Performance Verification Results An equivalent simulation model was built using MATLAB / Simulink to pre-validate the synergistic effect of reflector enhancement, ramp aperture array, and segmented pulse acoustic field control. The model includes six modules: acoustic field equivalence, reflector enhancement, cell force and motion, blockage index evaluation, segmented pulse control, and sorting state evaluation.

[0085] 1. Sound field equivalent model The acoustic field generated by the focused interdigital transducer in the ramp micro-aperture array region is equivalent to a time-varying acoustic pressure field. p(x , y , z , t) With the sound field v a (x , y , z , t) Under reflectorless conditions, the focused sound field is denoted as... p 0 (x,y,z,t) Under the condition of setting up a sound wave reflector, the sound pressure field enhanced by the reflector is calculated using the following equivalent formula:

[0086] , in, P R This represents the equivalent sound pressure field enhanced by the reflector. K R The equivalent reflection enhancement coefficient of the reflector. τ To delay the propagation of reflected sound waves, ( x' , y' ) represents the mirror position relative to the geometric center line of the reflector. K R The values ​​were obtained through finite element method sound field calculations, laser Doppler vibration measurements, or hydrophone calibration, and ranged from 0.30 to 0.60. (This is from a contrast scheme without a reflector.) K R Take 0.

[0087] In the simulation, the equivalent peak sound pressure at the focal point is 0.1 MPa to 0.5 MPa, 0.2 MPa to 0.4 MPa during the stable sorting stage, and no more than 0.5 MPa during the anti-blockage recovery stage.

[0088] 2. Cellular acoustic radiation force and motion model A549 cells and red blood cells are respectively represented as spherical particles suspended in a liquid medium. The acoustic radiation force of the cells in the sound field is calculated as follows:

[0089] , in, F ac For sound radiation force, U ac Let be the acoustic potential energy. For a cell with radius 'a', U ac Established based on the square terms of sound pressure and sound velocity, its magnitude is related to the cell radius *a* and the fluid density. ρ f Fluid compressibility K f Cell density ρ P Cell compressibility coefficient K P And it is related to the local sound field intensity.

[0090] Under low Reynolds number conditions in microchannels, the cell inertia term is ignored, and the cell velocity is calculated using the following formula: , in, r The location of the cell's center of mass. u ( r , t Let α be the fluid velocity within the channel, and μ be the fluid dynamic viscosity. F g The resultant force of gravity and buoyancy. F wall This is a wall repulsion correction term when cells approach the channel wall or the edge of the micropores. If the cell centroid enters the upper A549 cell exit region, the target cell is determined to have entered the target exit; if the erythrocyte centroid passes through the micropore array and enters the lower collection channel, the erythrocyte is determined to have entered the erythrocyte exit path through the sorting well array.

[0091] 3. Congestion Index Evolution Model The slope micro-orifice array region was used as the ROI for clogging monitoring. The clogging index B was calculated as a normalized weighted average of the number of objects lodged in the image, local grayscale changes, flow rate attenuation rate, and differential pressure increase rate.

[0092] , in, B N This is the normalized value for the number of stranded objects. B G This is a normalized value for local grayscale changes. B Q This is the normalized value of the flow attenuation rate.B P This is the normalized value of the differential pressure rise rate. The normalization method for each item is consistent with the calculation method of the blockage index B in the segmented pulse control unit.

[0093] In the simulation, a stagnant object is defined as a cell or cell aggregate that remains continuously within the pore array ROI for at least 2 seconds and has a centroid velocity of less than 5 μm / s. When this object continuously overlaps with the edge region of the micropores, it is considered a pore-dwelling stagnant object. The blockage rate C is calculated using the following formula:

[0094] , in, N c This represents the number of monitoring cycles during a 120-second simulation period when an orifice stagnation event or flow attenuation event occurs. N t This represents the total number of monitoring cycles within 120 seconds. A flow rate decay event is defined as the decrease in outlet flow rate relative to the reference flow rate, under the condition that the set flow rate of the syringe pump remains constant. Q 0 Events with a decline of more than 10%.

[0095] 4. Basic Input Parameters The simulation input parameters are as follows: upper sorting channel height 30 μm, width 100 μm; lower collection channel height 30 μm; ramp sorting section length 1 mm; micropore diameter 9 μm, pore spacing 16 μm; operating frequency 10 MHz to 11 MHz; average linear velocity of the main channel flow 0.5 mm / s. The fluid medium is PBS buffer with a density of 1000 kg / m³. 3 Dynamic viscosity 1.0 mPa·s, sound velocity 1500 m / s.

[0096] A549 cells have an equivalent radius of 8 μm to 10 μm and a density of 1050 kg / m³. 3 Up to 1100 kg / m 3 Red blood cell equivalent radius 3 μm to 4 μm, density 1080 kg / m³ 3 Up to 1120 kg / m 3 .

[0097] Each simulation session lasts 120 seconds, with a time step ranging from 0.01 to 0.05 seconds. Each scheme is simulated at least three times, with at least 1000 A549 cells in each simulation, and red blood cells configured at a ratio of 1:1000. To reduce computational load, red blood cells can be modeled using equivalent particle packs, with each particle pack corresponding to a fixed number of red blood cells. Sorting efficiency and purity are calculated based on the particle pack weights.

[0098] 5. Set up the comparison plan Under conditions where channel height, width, micropore diameter and spacing, cell concentration, injection flow rate, simulation duration, and total input acoustic power range are consistent, the following three control schemes are set up for comparison with the synergistic scheme of the present invention, with only the structure or control factors to be compared being changed: Reflector-free FIDT chip solution: Retains the focusing interdigital transducer, upper and lower channels, and ramp aperture array, while eliminating the acoustic reflector, thus increasing the reflection enhancement coefficient. K R Take 0.

[0099] A slope-free horizontal micro-aperture array chip solution: The transducer and reflector are retained, and the slope micro-aperture array is replaced with a horizontal micro-aperture array (the aperture diameter is 9 μm and the aperture spacing is 16 μm unchanged), thus eliminating the guiding effect of the slope.

[0100] Fixed continuous sound field scheme: Retain the reflector, upper and lower channels and ramp hole array, but do not enable segmented pulse control, and the transducer continuously outputs the equivalent average sound pressure level of the sorting stage.

[0101] The collaborative solution of this invention employs reflector enhancement, ramp hole array, and segmented pulse sound field control simultaneously, and switches between stable sorting and anti-clogging recovery stages based on the blockage index B.

[0102] 6. Calculation method for sorting indicators The sorting efficiency η is defined as the ratio of the number of A549 cells exiting the upper A549 cell layer to the total number of A549 cells entering the cell layer: , in, N A,in To input the total number of A549 cells, N A,out This represents the number of A549 cells entering the A549 cell exit.

[0103] The target effluent purity P is defined as the proportion of A549 cells in the A549 cell effluent to the total number of cells in that effluent. , in, N R,out The number of red blood cells that mistakenly enter the A549 cell exit.

[0104] The 120-second blockage index B is the B value calculated according to the blockage index formula after 120 seconds of simulation. The blockage rate C is calculated as the proportion of monitoring cycles in which orifice stagnation events or flow attenuation events occur within 120 seconds out of the total monitoring cycles.

[0105] Activity retention prediction value V PThis is a simulation prediction index based on acoustic and thermal exposure risk, which is not equivalent to measured cell viability, and is calculated using the following formula: Vp =100% × [1-min( H ,1)], , in, H To normalize the acoustic heat exposure risk index, E ac This represents the cumulative acoustic exposure experienced by the cells during the simulation period. E ref To preset the reference sound exposure, Δ T Δ represents the maximum local temperature rise in the channel during the simulation. T ref The preset reference temperature rise is used; α and β are the weights of the acoustic exposure term and the temperature rise term, respectively, with α set to 0.6 and β set to 0.4.

[0106] 7. Simulation Results Under the unified simulation conditions described above, the simulation prediction results of the four schemes show that, under the same channel size, aperture array parameters, injection speed, cell input ratio and sound field operating range, the collaborative scheme of this invention, compared with the scheme without reflectors, the scheme without ramps and horizontal aperture arrays and the fixed continuous sound field scheme, can improve the proportion of target cells entering the upper target outlet and the purity of the target outlet, and reduce the 120 s blockage index and blockage rate.

[0107] Figure 5 The normalized vertical position change curves of cells under different sound field control methods are shown. Position 0 represents cells remaining in the upper sorting channel, and position 1 represents cells entering the lower collection channel, used to quantify the vertical movement trends of A549 cells and erythrocytes. Under the action of a fixed continuous sound field, the normalized position of A549 cells eventually stabilizes in the range of 0.35 to 0.38, showing a tendency to shift towards the lower layer of the channel; the normalized position of erythrocytes eventually reaches 0.75, with limited migration to the lower layer, resulting in insufficient overall sorting efficiency. Under the action of a reflector-free segmented pulse sound field, the normalized position of A549 cells remains at around 0.30, maintaining stability in the upper layer better than in the fixed continuous sound field; the normalized position of erythrocytes rises to about 0.85, with a significantly enhanced tendency to pass through the lower layer, demonstrating the improving effect of segmented pulse control on sorting performance. Under the enhanced segmented pulse sound field effect of the reflector, the normalized position of A549 cells was the lowest, eventually stabilizing at 0.25, indicating the best retention effect in the upper channel; the normalized position of red blood cells was the highest, eventually approaching 0.94, indicating the most thorough passage through the lower collection channel.

[0108] Figure 6The curves show the blockage index variation of the aperture array under different sound field control methods. The blockage index is positively correlated with the degree of cell retention, aggregation, and blockage in the aperture array region. The lower the value, the better the channel patency and the stronger the continuous sorting stability. Under the action of a fixed continuous sound field, the blockage index continuously increases with the running time, eventually approaching 0.95, indicating that long-term continuous sound field drive easily causes aperture array blockage and poor continuous operation stability. Under the action of a reflectorless segmented pulse sound field, the blockage index rises to about 0.60 and then tends to stabilize, without the continuous deterioration trend shown under the fixed continuous sound field. This indicates that the segmented pulse control successfully curbs the further development of the blockage trend after B reaches the trigger threshold. The shape of this curve is consistent with the setting of "entering the recovery stage after B≥0.60" in the segmented pulse control logic, that is, the control intervention effectively prevents the blockage index from continuing to rise. For the reflector-enhanced segmented pulse scheme, the blockage index remained stable at 0.27 throughout the process, consistently below the recovery trigger threshold of 0.60. This indicates that the synergistic effect of reflector enhancement and segmented pulse control has excellent anti-blockage capability under the simulation conditions, and can maintain channel patency without triggering the recovery phase.

[0109] In summary, the three-layer microfluidic sorting chip based on reflector-enhanced focused sound field and its segmented pulse sound field control method proposed in this invention can achieve non-contact, high-precision, and high-activity sorting of A549 cells and erythrocytes. It has the combined advantages of simple structure, strong manufacturability, and stable operation, and has broad application prospects in the fields of biomedical detection and cell analysis.

[0110] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0111] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A three-layer microfluidic sorting chip with reflection-enhanced focused sound field, characterized in that, It includes a lithium niobate piezoelectric substrate layer (1), a microfluidic channel layer (5) and a top capping layer (9) stacked sequentially from bottom to top; The lithium niobate piezoelectric substrate (1) includes a lithium niobate piezoelectric substrate, and a focusing interdigital transducer (2) and an acoustic reflector (3) are disposed on the upper surface of the lithium niobate piezoelectric substrate; the focusing interdigital transducer (2) is an arc-shaped focusing structure, and the acoustic reflector (3) is an arc-shaped reflective grating array; The microfluidic channel layer (5) is an integral structure composed of three PDMS sublayers bonded together by alignment. From bottom to top, it includes a lower collection channel layer (6), a ramp pore array partition layer (7), and an upper sorting channel layer (8). The upper sorting channel layer (8) is connected in sequence along the main fluid flow direction by a sample inlet branch (801), a mixing section (803), a horizontal sorting section (804), a ramp sorting section (805), and a target cell outlet branch (806). A buffer inlet branch (802) is provided near the sample inlet branch (801). One end of the buffer inlet branch (802) is connected to the mixing section (803). The ramp sorting section (805) is inclined relative to the main flow direction of the horizontal sorting section (804) in the horizontal plane. On the slope hole array partition (7), a micropore array (701) is provided at the position corresponding to the slope sorting section (805), the micropore array (701) includes a plurality of micropores (7011). The lower collection channel layer (6) is provided with a collection channel (601) and a background cell outlet (602); the collection channel (601) is located below the micropore array (701) and is fluidly connected to the upper sorting channel layer (8) only through the micropore array (701), and is used to collect small background cells that have settled through the micropores; Except for the micropore (7011) connected region, the remaining regions between layers are in a liquid-tight isolation state; The top capping layer (9) is sealed to the upper surface of the upper sorting channel layer (8) and has a sample inlet interface (901), a buffer inlet interface (902), a target cell outlet interface (903) and a background cell outlet interface (904).

2. The three-layer microfluidic sorting chip according to claim 1, characterized in that, The lithium niobate piezoelectric substrate (1) is 128° YX cut with a thickness of 450 μm to 550 μm; the electrode material of the focusing interdigital transducer (2) is a Cr / Au composite layer, wherein the Cr adhesion layer is 5 nm to 10 nm thick and the Au conductive layer is 80 nm to 150 nm thick; the finger width and the spacing between adjacent fingers of the focusing interdigital transducer (2) are both 85 μm to 90 μm, there are 18 to 22 pairs of fingers, the focusing curvature radius is 2 mm to 4 mm, the effective aperture is 1 mm to 3 mm, and the center operating frequency after bonding the microfluidic channel layer is 10 MHz to 11 MHz. MHz; the acoustic reflector (3) is composed of 14 to 16 parallel gratings, the width of a single grating is 1 / 12 to 3 / 12 of the surface acoustic wave wavelength, and the center distance between adjacent gratings is 1 / 4 to 3 / 4 of the surface acoustic wave wavelength; the acoustic wave propagation axis of the focusing interdigital transducer (2) forms an angle of 30° to 45° with the center line of the ramp sorting section (805) in the horizontal plane, and the sound field focus falls on the micropore array (701), so that the sound field action direction is coupled with the cell migration path.

3. The three-layer microfluidic sorting chip according to claim 1, characterized in that, The lithium niobate piezoelectric substrate layer (1) and the microfluidic channel layer (5) are bonded and sealed by a coupling sealing interface layer (4); the thickness of the coupling sealing interface layer (4) is 5 μm to 10 μm, and it is only distributed in the solid support area and non-fluid area on the outer periphery of the channel, without intruding into the internal space of the upper sorting channel layer (8), the lower collection channel layer (6) and the micropore array (701); positioning coupling bosses are provided between the layers for precise positioning during the bonding process; the coupling sealing interface layer (4) and the lithium niobate piezoelectric substrate layer (1) and the microfluidic channel layer (5) are all liquid-tightly connected by oxygen plasma activation bonding.

4. The three-layer microfluidic sorting chip according to claim 1, characterized in that, The upper sorting channel layer (8) has a channel height of 30 μm and a main channel width of 100 μm. The sample inlet branch (801) and the buffer inlet branch (802) are arranged in a bifurcated manner at the upstream of the channel, and the two merge to form the mixing section (803). The mixing section (803) has a length of 0.8 mm to 1.2 mm, the horizontal sorting section (804) has a length of 1.0 mm to 1.4 mm, and the slope sorting section (805) has a length of 0.8 mm to 1.2 mm, and is inclined at 15° to 17° relative to the mainstream direction of the horizontal sorting section (804) in the horizontal plane.

5. The three-layer microfluidic sorting chip according to claim 1, characterized in that, The thickness of the inclined hole array partition (7) is 14 μm to 16 μm; the micropore array (701) is arranged in 9 to 11 rows, each row containing 4 to 6 circular micropores (7011), with a single pore diameter of 8 μm to 10 μm; the center distance between adjacent micropores (7011) is 14 μm to 18 μm, and the adjacent rows of micropores are staggered by 7 μm to 9 μm along the channel width direction.

6. The three-layer microfluidic sorting chip according to claim 1, characterized in that, The height of the collection channel (601) is 25 μm to 35 μm and the width is 90 μm to 110 μm. Its upstream end starts from the starting end of the micropore array (701). The collection channel (601), the micropore array (701) and the ramp sorting section (805) form an overlapping connected area on the horizontal projection. The overlapping connected area covers the entire area of ​​the micropore array (701).

7. The three-layer microfluidic sorting chip according to claim 1, characterized in that, It also includes a segmented pulse control unit, which is electrically connected to the focused interdigital transducer (2) and is used to dynamically adjust the acoustic field parameters according to the blockage index B of the micro-pore array (701) region to realize multi-stage closed-loop control of the sorting process; The blockage index B is obtained by weighted summation of the normalized value of the number of stagnant objects BN, the normalized value of the local gray scale change BG, the normalized value of the flow rate attenuation rate BQ, and the normalized value of the differential pressure increase rate BP. The calculation formula is: B = 0.35BN + 0.25BG + 0.2BQ + 0.2BP. The value range of the blockage index B is 0~1. The larger the value, the higher the blockage risk of the micropore array (701) region.

8. A segmented pulse acoustic field control method applied to the three-layer microfluidic sorting chip according to any one of claims 1 to 9, characterized in that, The control process is executed cyclically in the following stages: The control focused interdigital transducer (2) is started with low sound pressure and frequency scanning is performed within the preset frequency range. The optimal working frequency is determined based on the focusing response deviation, sound field response intensity and the stagnation state of the micro-pore array (701). Then the sound pressure is gradually increased to the stable sorting range. The lower collection channel layer (6), the inclined hole array partition layer (7), and the upper sorting channel layer (8); A pulsed focused sound field is output to the focused interdigital transducer (2) to maintain a stable sorting state in which target cells are retained in the slope sorting section (805) and background cells enter the collection channel (601) through the micropore array (701); When the blockage index B of the micropore array (701) is detected to be ≥0.60 and lasts for at least two monitoring cycles, the system switches to an intermittent sound field with alternating strong and weak pulses. The strong pulses loosen the residue at the orifice, and the weak pulses flush out the residue with fluid, thereby clearing the micropore array (701). When the blockage index B < 0.30 and lasts for ≥ 3 s, switch back to the acoustic field parameters of the stable sorting stage and resume continuous sorting.

9. The segmented pulse sound field control method according to claim 8, characterized in that, The frequency scanning range is 10MHz~11MHz; the focal equivalent peak sound pressure of the focused sound field is 0.2MPa~0.4MPa, the pulse duty cycle is 40%~80%, and the pulse period is 0.05s~1s.

10. The segmented pulse sound field control method according to claim 8, characterized in that, The intermittent sound field consists of a strong pulse release phase and a weak pulse flushing phase: The strong pulse release phase is characterized by: a focal equivalent peak sound pressure of 0.45 MPa to 0.50 MPa, a pulse duty cycle of 20% to 40%, a pulse period of 0.05 s to 0.10 s, and a duration of 1 s to 2 s. The weak pulse flushing sub-stage is as follows: the focal equivalent peak sound pressure is 0 MPa to 0.20 MPa, the pulse duty cycle is 0% to 20%, the pulse period is 0.20 s to 0.50 s, and the duration is 1 s to 3 s.

Images