Microparticle filling method and microparticle filling system

Abstract

To provide a technique for accurately aligning a tip of a nozzle with a micro reaction vessel when filling microparticles into the micro reaction vessel.SOLUTION: A microparticle filling method for filling microparticles into a container comprises: aligning a nozzle with the container using an image captured by an imaging device; and filling the microparticles from the nozzle into the container. The alignment comprises: aligning a tip of the nozzle with a horizontal center position of an opening of the container using (a) a horizontal differential distance between a reference position in the image captured by the imaging device and a horizontal center position of the tip of the nozzle and (b) coordinates at which the horizontal center position of the opening of the container is located at the reference position.SELECTED DRAWING: Figure 12A

Description

Technical Field

[0001] The present disclosure relates to a method for filling microparticles and a microparticle filling system.

Background Art

[0002] Generally, single-cell analysis is one of the analysis methods in life science and is a technique for analyzing and quantifying biomolecules for each single cell. Single-cell analysis is, in particular, a method for determining the nucleotide sequences of DNA and RNA (especially mRNA) for each single cell and measuring the number of molecules by using a DNA sequencer.

[0003] Specific methods of single-cell analysis include a method using a tube (a plastic reaction container of about 0.2 mL to 2 mL) or a titer plate (a plate-like assembly of a specified number of resin reaction containers such as 96 or 384) (Non-Patent Document 1), a method using a microchannel (Non-Patent Document 2), a method using an emulsion (Non-Patent Document 3), a method using a microwell (Non-Patent Document 4), a method using a through microwell (Non-Patent Document 5), and the like.

[0004] In any method, a large number of cells to be analyzed are separated into single cells, and a cell-specific DNA base sequence (hereinafter referred to as "barcode") is introduced into the nucleic acid extracted from each cell. Thereafter, the nucleic acid samples obtained from these multiple cells are combined, and DNA sequencing is performed. Single-cell analysis data can be obtained by separating the DNA sequence information for each barcode sequence information from the sequence data obtained by DNA sequencing.

[0005] In particular, in the methods described in Non-Patent Documents 3 to 5, a barcode is introduced after cell isolation and before nucleic acid amplification. Thereby, samples for a DNA sequencer can be prepared from a large number of cells with a small amount of reagents, and sequence data can be obtained.

[0006] The single-cell analysis method described in Non-Patent Document 5 uses a larger number of microparticles for analyzing one cell compared to the methods described in Non-Patent Documents 3 and 4 (in Non-Patent Documents 3 and 4, only one microparticle is used). Therefore, the number of DNA probes for nucleic acid capture used for analyzing one cell can be increased by about 100 times or more. Also, since the cell lysis solution passes through a porous structure composed of a large number of microparticles, nucleic acids from a single cell can be captured on the surface of the microparticles with high efficiency in a short time.

[0007] Thus, in order to use a large number of microparticles for analyzing one cell, it is required to fill a microreaction vessel with a large number of microparticles.

[0008] In Non-Patent Document 5, by utilizing the fact that there are through-holes in the bottom surface of the microreaction vessel, a microparticle suspension is dispensed into the microreaction vessel, and excess moisture is discharged through the through-holes, thereby filling the microreaction vessel with microparticles. Also, since the diameter of the microreaction vessel is as small as 50 - 100 μm, an inkjet device is used.

[0009] Patent Document 1 describes a technique in which microparticles are concentrated in advance at the tip of a nozzle to a concentration close to the closest packing density, then the nozzle is lowered, and only the concentrated microparticle solution is separated and filled into a microreaction vessel (container) by utilizing surface tension (wettability).

Prior Art Documents

Patent Documents

[0010]

Patent Document 1

Non-Patent Documents

[0011]

Non-Patent Document 1

[0012] The above-described method of filling microparticles into a microreaction vessel (container) using an inkjet device as in Non-Patent Document 5 has the following problems.

[0013] First, in order to fill the fine particles into the micro reaction vessels, after dispensing a solution in which the fine particles are suspended into the micro reaction vessels, it is necessary to discharge the excess solution from the micro reaction vessels. At this time, in order to discharge only the solution from the micro reaction vessels while leaving the fine particles, a membrane having a mesh finer than the fine particles must be used to prevent the fine particles from flowing out. However, when the diameter of the fine particles is as small as several μm or less, the size of the openings of the mesh must be smaller than the diameter of the fine particles. It is very difficult to produce a membrane with a mesh having small openings and high strength. When using a membrane having such a mesh of available size, the thickness of the membrane becomes on the order of mm in order to maintain the strength. As a result, due to a large pressure loss, the solution cannot be sucked quickly. Furthermore, in order to quickly remove the solution by applying a high pressure, it is necessary to use a membrane having a mesh with even higher strength, so it is impossible to achieve both the maintenance of the membrane strength and the improvement of the suction rate of the solution.

[0014] Second, in order to control the number of fine particles after filling, it is necessary to stabilize the fine particle density of the fine particle suspension and control the number of particles by the amount of the solution. However, when the specific gravity of the fine particles is larger than that of the solvent, not only does the fine particle density change due to the sedimentation of the fine particles, but the sedimentation rate also changes depending on the degree of microscopic aggregation of the fine particles, so it is difficult to maintain a stable fine particle density.

[0015] Furthermore, in order to stably eject the droplets, the fine particle density in the ejection nozzle must be stable and constant. When the physical properties of the solution become non-uniform, not only do the ejection amount and the ejection direction become unstable, but also the ejection itself stops frequently. When the ejection stops, the fine particle suspension must be prepared again and refilled into the inkjet head. Therefore, the time required to fill the fine particles into a large number of micro reaction vessels becomes longer by the amount of the replacement work.

[0016] The technique of Patent Document 1 was made to solve the problems in the method for filling fine particles using an inkjet device as described above.

[0017] Here, in the method described in Patent Document 1, in order to successfully fill the micro reaction vessel (container) with the high-concentration fine particle solution accumulated at the tip of the nozzle, it is necessary to first ensure that the tip of the nozzle descends into the interior of the micro reaction vessel 101.

[0018] Figure 1A is a cross-sectional schematic view (left side) of the state where the alignment between the tip of the nozzle 109 and the micro reaction vessel 101 is successful, and a photographed image from above the micro reaction vessel 101 in the state where the filling of the fine particles is successful (right side). As shown in Figure 1A, when the alignment between the tip of the nozzle 109 and the micro reaction vessel 101 is successful, the tip of the nozzle 109 descends into the interior of the micro reaction vessel 101, and fine particles (high-concentration fine particle solution) can be filled. In the photographed image on the right side of Figure 1A, the fine particles inside the micro reaction vessel 101 are shown in black. However, since the size of the micro reaction vessel 101 is as small as 50 to 100 μm, the first problem for successfully filling the fine particles is that it is necessary to accurately align the tip of the nozzle and the target micro reaction vessel to be filled in units of 1 μm.

[0019] Figure 1B is a cross-sectional schematic view (left side) of the state where the alignment between the tip of the nozzle 109 and the micro reaction vessel 101 fails, and a photographed image from above showing the state where the filling of the fine particles fails (right side). As shown in Figure 1B, when the alignment between the tip of the nozzle 109 and the micro reaction vessel 101 fails, the nozzle 109 cannot appropriately descend into the interior of the micro reaction vessel 101. As a result, as shown in the photographed image on the right side of Figure 1B, the fine particles are discharged outside the micro reaction vessel 101, resulting in filling defects.

[0020] Therefore, the present disclosure provides a technique for accurately aligning the tip of the nozzle and the micro reaction vessel in the filling of fine particles into the micro reaction vessel.

Means for Solving the Problems

[0021] To solve the above problems, the present disclosure provides a method for filling fine particles into a container, including aligning a nozzle and the container using an image captured by an imaging device, and filling the fine particles from the nozzle into the container. The alignment includes: (a) using a horizontal difference distance from a reference position of the image captured by the imaging device to a horizontal center position of the tip of the nozzle, and (b) using coordinates at which a horizontal center position of an opening of the container is disposed at the reference position, to align the tip of the nozzle with the horizontal center position of the opening of the container.

[0022] Further features related to the present disclosure will become apparent from the description herein and the accompanying drawings. Also, aspects of the present disclosure are achieved and realized by elements and combinations of various elements and the aspects of the following detailed description and the appended claims. The description herein is merely exemplary and does not limit the scope of the claims or applications of the present disclosure in any way.

Advantages of the Invention

[0023] According to the technology of the present disclosure, in filling fine particles into a micro reaction vessel, the tip of the nozzle can be accurately aligned with the micro reaction vessel. Other problems, configurations, and effects will be clarified by the description of the following embodiments.

Brief Description of the Drawings

[0024]

Figure 1A

Figure 1B

Figure 2A

Figure 2B

Figure 3A

Figure 3B

Figure 3C

Figure 4

Figure 5

Figure 6

Figure 7A

Figure 7B

Figure 8

Figure 9A

Figure 9B

Figure 10A

Figure 10B

Figure 11A

Figure 11B

Figure 12A

Figure 12B

Figure 13

Figure 14A

Figure 14B

Figure 15

Figure 16A

Figure 16B

Figure 16C

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21A

Figure 21B

Figure 21C

Figure 21D

Figure 21E

Figure 22A

Figure 22B

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27A

Figure 27B

Figure 27C

Figure 27D

Figure 28A

Figure 28B

Figure 29A

Figure 29B

Figure 29C

Figure 30A

Figure 30B

Figure 31A

Figure 31B

Figure 32A

Figure 32B

Figure 33A

Figure 33B

Figure 34A

Figure 34B

Figure 34C

Figure 35

Figure 36A

Figure 36B

Figure 37

Figure 38

Figure 39A

Figure 39B

Embodiments for Carrying Out the Invention

[0025] [Definition of Terms] First, the definition of the terms used in each embodiment of the present disclosure will be explained.

[0026] "Micro reaction vessel": A concave - shaped container formed on a planar substrate. Also called a microwell. Typically, the diameter is several μm to several hundred μm, and in particular, those with a diameter of several tens of μm. It also includes those in which through - holes for capturing cells are formed on the bottom surface as described in Non - Patent Document 5.

[0027] "Nozzle": It has a hollow structure through which a solution and microparticles can move, and the tip can be inserted into the micro reaction vessel.

[0028] "Alignment": Making the horizontal center position of the tip of the nozzle coincide with the horizontal center position of the micro reaction vessel.

[0029] "Fine particles": Particles with a specific gravity greater than that of the solution and a diameter ranging from several hundred nanometers to several hundred micrometers, or particles having magnetism, electric charge, or dielectric properties and a diameter ranging from several nanometers to several hundred micrometers.

[0030] "Buffer": A solution in which fine particles are suspended. The pH or salt concentration may be controlled to maintain the suspended state of the fine particles. Further, a surfactant may be mixed as necessary to control the wettability (contact angle) with the outer wall of the nozzle and the inner wall of the micro reaction tank.

[0031] "Contact": The solution component (buffer) of the fine particle solution sucked at the tip of the nozzle touches the inner wall (bottom surface and side wall surface) of the micro reaction tank. In particular, it refers to a state in which a contact angle can be defined between the solution and the inner wall.

[0032] "Filling": Supplying the required number (required fine particle concentration) of fine particles into the micro reaction tank.

[0033] "Concentration": Increasing the fine particle concentration (number of fine particles per unit volume (number density)) of the fine particle suspension and reaching a fine particle concentration that is 10% or more of the maximum concentration while maintaining the dispersed state of the fine particles.

[0034] "Sedimentation": Deviating from the fine particle concentration of the fine particle solution stored in the storage container, and the fine particle concentration increases in a part of the region, particularly at the tip of the nozzle.

[0035] "Judgment": Determining whether a region (high-concentration suspension) with an increased fine particle concentration is formed at the tip of the nozzle, and the fine particle concentration and the volume of the region contain a number of fine particles that occupy 10% or more of the volume of the micro reaction tank.

[0036] "High concentration": A state in which the fine particle concentration of the fine particles required for filling is reached. It refers to a fine particle concentration that is 10% or more of the maximum concentration while maintaining the dispersed state of the fine particles.

[0037] [Regarding the problems of the present disclosure] It has been described that the first problem to be solved by the technology of the present disclosure is to accurately align the tip of the nozzle with the micro reaction vessel. Here, the second problem to be solved by the technology of the present disclosure will be described. Even if the tip of the nozzle is successfully aligned with the micro reaction vessel and the tip of the nozzle can descend into the micro reaction vessel, the filling of the high-concentration fine particle solution into the micro reaction vessel does not necessarily succeed. As will be described later, the principle of fine particle filling of this method is the spread of a liquid using the Laplace pressure due to the surface tension (wettability) of the fine particle solution with respect to the nozzle and the micro reaction vessel as the driving force. If this spreading is not sufficient, there is a risk that almost no fine particle solution is filled or only partially filled.

[0038] FIG. 2A is a cross-sectional schematic view (upper side) for explaining the filling process when the filling of the high-concentration fine particle solution into the micro reaction vessel 101 is successful, and a photographed image (lower side) of the micro reaction vessel 101 when the filling of the fine particles is successful. As shown in FIG. 2A, when the high-concentration fine particle solution can be spread to the side wall of the micro reaction vessel 101, the filling of the fine particles is successful.

[0039] FIG. 2B is a cross-sectional schematic view (left side) for explaining the filling process when the filling of the high-concentration fine particle solution into the micro reaction vessel 101 fails, and a photographed image (right side) of the micro reaction vessel when the filling of the fine particles fails. As shown in FIG. 2B, when the fine particle solution cannot be spread to the side wall of the micro reaction vessel 101, partial non-filling (voids are generated) occurs and the filling fails. In the upper part of FIG. 2B, the fine particle solution does not reach the side wall of the micro reaction vessel 101 and is held only in the central part of the micro reaction vessel 101. In the lower part of FIG. 2B, the fine particle solution reaches only one side wall of the micro reaction vessel.

[0040] As described above, the second problem of the present disclosure is that it is necessary to increase the probability (yield) of successfully filling the micro reaction tank with the high-concentration fine particle solution accumulated at the tip of the nozzle. Only by solving the above first and second problems can the filling of fine particles into the inside of the micro reaction tank 101 as shown in FIGS. 1A and 2A be successful. Thus, in order to successfully fill the micro reaction tank with the fine particle solution, it is necessary to accurately align the tip of the nozzle with the micro reaction tank and increase the probability of successfully filling the micro reaction tank with the high-concentration fine particle solution accumulated at the tip of the nozzle.

[0041] [First Embodiment] <Fine Particle Filling System> FIG. 3A is a schematic diagram showing a fine particle filling system 100 according to the first embodiment. FIG. 3B is a top view showing an implementation example of the fine particle filling system 100. FIG. 3C is a side view showing an implementation example of the fine particle filling system 100. The fine particle filling system 100 is a device that automatically implements a method for aligning the positions of a micro reaction tank and a nozzle, which will be described later, and a fine particle filling method. As shown in FIG. 3A, the fine particle filling system 100 includes a first drive stage 106 (first drive mechanism), a first imaging device 107, a second drive stage 108 (second drive mechanism), a nozzle 109, a second imaging device 114, a drive mechanism 115, a sensor system 116, a dispensing mechanism 117, a plurality of storage containers 118, a stirring mechanism 119, and a control device 120.

[0042] The first drive stage 106 is fixed in position in the vertical direction and configured to be movable only in the horizontal direction. The planar substrate 102 is fixedly arranged on the first drive stage 106. The planar substrate 102 includes a plurality of micro reaction tanks (containers). The first drive stage 106 has a table for placing the planar substrate 102. The table has a space below (directly below) the planar substrate 102, and the space is configured to be able to hold a plurality of magnets 121.

[0043] The first imaging device 107 is arranged directly above the first drive stage 106 in the vertical direction. The first imaging device 107 is fixed in position in the horizontal direction. The first imaging device 107 is configured to be able to image the planar substrate 102 from above in a bird's-eye view. The first imaging device 107 is used to acquire the XY coordinates on the first drive stage 106 in the alignment method and the fine particle filling method described later.

[0044] The second drive stage 108 is fixed in position in the horizontal direction and is configured to be movable only in the vertical direction. The nozzle 109 is provided on the second drive stage 108. The nozzle 109 is configured to be able to suck the fine particle solution 110.

[0045] The drive mechanism 115 automatically controls the first drive stage 106 and the second drive stage 108 via the control board circuit. The first drive stage 106 and the second drive stage 108 are used for the alignment of the micro reaction tank 101 and the nozzle 109, the access of the nozzle 109 to various storage containers 118, etc.

[0046] The second imaging device 114 is used to observe the concentration status of the fine particle suspension in the nozzle 109 moved directly above the micro reaction tank. The second imaging device 114 may be a camera system that captures an image of the tip of the nozzle 109, or may be an optical system that irradiates light to the tip of the nozzle 109 and analyzes scattered light, absorbed light, or transmitted light.

[0047] The sensor system 116 detects the contact between the nozzle 109 and the micro reaction tank. The sensor system 116 typically has a pressure sensor, a load sensor, or a strain sensor. The sensor system 116 determines the contact between the nozzle 109 and the micro reaction tank by detecting, with the above sensor, the upward force applied to the tip of the nozzle 109 generated by the contact between the nozzle 109 and the micro reaction tank. Alternatively, the sensor system 116 may be configured to detect the contact by optical means such as a camera.

[0048] The dispensing mechanism 117 has, for example, a syringe pump and controls the suction and discharge of liquid to the nozzle 109. The dispensing mechanism 117 particularly controls the suction of the particulate solution 110 into the nozzle 109. The nozzle 109 is connected to the dispensing mechanism 117 by a pipe (for example, a PEEK pipe tube). As one of the components of the dispensing mechanism 117, an electromagnetic valve, typically a solenoid valve, can be provided and is used for pressure increase / decrease control by releasing atmospheric pressure inside the dispensing mechanism and realizing a closed system. The syringe pump in the dispensing mechanism 117 may be controlled, for example, as part of the drive mechanism 115.

[0049] The plurality of storage containers 118 include a storage container for storing the particulate solution, a storage container for storing the cleaning liquid, and a waste liquid container for collecting waste liquid. In particular, the storage container for storing the particulate solution is mounted on the stirring mechanism 119, stirred at an appropriate timing, and controlled to always maintain a good dispersion state.

[0050] The magnet 121 is typically a neodymium magnet. When the nozzle 109 is arranged directly above the micro reaction tank 101, magnetic particles can be moved in a short time by the magnetic field, and a highly concentrated suspension can be efficiently formed. The magnetization direction of the magnet is set to the direction in which the particles are desired to settle, that is, the direction perpendicular to the planar substrate, so that the particles can be efficiently settled.

[0051] The control device 120 controls the operations of the first imaging device 107, the second imaging device 114, the drive mechanism 115, the sensor system 116, the dispensing mechanism 117, and the stirring mechanism 119. Further, the control device 120 receives the detection signals of the first imaging device 107, the second imaging device 114, and the sensor system 116 and executes necessary arithmetic processing.

[0052] FIG. 4 is a schematic cross-sectional view of a planar substrate 102 having a plurality of microreactors 101. The planar substrate 102 is, for example, a chip for single-cell analysis, and has a structure in which a so-called microtiter plate is miniaturized. The planar substrate 102 has a plurality of microreactors 101 (containers). For the sake of simplicity of illustration, in FIG. 4, only three microreactors 101 are shown, but the number of microreactors 101 can be four or more. The microreactor 101 has an upper opening 103a and a bottom 103b, and a through hole 105 extending from the bottom 103b to the bottom surface 104 of the planar substrate 102 is formed. The shape of the microreactor 101 is shown here as an example of a straight cylindrical shape, but it may have a different shape as in the embodiments described later. The through hole 105 may have a straight cylindrical shape or a frustum shape.

[0053] FIG. 5 is a perspective view showing an implementation example of the sensor system 116. As shown in FIG. 5, the sensor system 116 has two ferrules 92, a block 93, two leaf springs 94 (elastic bodies), a fixing jig 95, and a strain sensor 96. The two ferrules 92 connect the nozzle 109 and the pipe. The ferrule 92 is fixed to the block 93 (for example, made of PMMA). One end portion in the longitudinal direction of the two leaf springs is screwed and fixed to the block 93. The other end portion in the longitudinal direction of the leaf spring 94 is fixed to a metal fixing jig. With such a configuration, when stress is applied to the nozzle 109, the stress is transmitted to the leaf spring 94 via the block 93, and the leaf spring 94 deforms.

[0054] A strain sensor 96 is fixed to each leaf spring 94 along the surface of the leaf spring 94. When the tip of the nozzle 109 contacts the bottom surface of the microreactor 101 and an upward force (stress) is applied to the nozzle 109, the leaf spring 94 is distorted, and the strain sensor 96 detects this amount of distortion. When the amount of distortion exceeds a predetermined value, it can be determined that the nozzle 109 has contacted the microreactor 101.

[0055] <Method for aligning the microreactor and the nozzle> With reference to FIGS. 6 to 12B, a method for aligning a micro reaction vessel and a nozzle according to the particle filling method of the first embodiment will be described. As described below, the alignment method of the present embodiment includes the following steps (i) to (v).

[0056] (i) Prepare a micro reaction vessel First, the user prepares a planar substrate 102 having a plurality of the above-described micro reaction vessels 101.

[0057] (ii) Obtain position information of the micro reaction vessel FIG. 6 is a plan view showing a state in which the planar substrate 102 is fixed on the first drive stage 106. As shown in FIG. 6, the user fixes the planar substrate 102 at a predetermined position on the first drive stage 106. Here, the lateral direction in the driving direction of the first drive stage 106 is defined as the X coordinate, and the longitudinal direction is defined as the Y coordinate. The initial position of the first drive stage 106 is set as the origin (0, 0) in the XY coordinates. Also, the driving direction of the second drive stage 108 is defined as the Z coordinate. The initial position of the second drive stage 108 is set as the origin (0) in the Z coordinate.

[0058] FIG. 7A is a plan view showing a state in which the planar substrate 102 is disposed directly below the first imaging device 107. As shown in FIG. 7A, the first drive stage 106 is driven to dispose the planar substrate 102 directly below the first imaging device 107. Then, the plurality of micro reaction vessels 101 are photographed using the first imaging device 107. The control device 120 performs image processing on the photographed image to obtain the in-image position information of each of the plurality of micro reaction vessels 101.

[0059] FIG. 7B is a photographed image for explaining a method for obtaining the in-image position information of the plurality of micro reaction vessels 101. First, as shown in FIG. 7B, the in-image position information (relative position with respect to the reference position) of the micro reaction vessel 101 is obtained from the reference position of the photographed image. The reference position of the photographed image can be, for example, the center position of the photographed image (left side in FIG. 7B), or the end portion of the photographed image (right side in FIG. 7B). Hereinafter, in the present disclosure, the reference position of the photographed image will be described as the center position.

[0060] Next, the in-image position information of the micro reaction tank 101 is converted into the XY coordinates of the first drive stage 106. At this time, the reference position of the captured image and the XY coordinates of the first drive stage 106 are set to be linearly interlocked. That is, it is set so that linear conversion can be performed using the coefficient for converting from a specific pixel position information in the image to the XY coordinate information of the first drive stage 106. By such a setting, the image information of the first imaging device 107 and the position information of the first drive stage 106 can be interlocked. Generally, the above coefficient is determined by the resolution of the first imaging device 107.

[0061] As described above, when the planar substrate 102 is disposed directly below the first imaging device 107, the horizontal center position of the micro reaction tank 101 can be obtained as the "coordinates at which the horizontal center position of the container opening is disposed at the reference position of the captured image" on the XY coordinates of the first drive stage 106. Specifically, the horizontal center position of the micro reaction tank 101 can be calculated by the following formula (1).

[0062]

Equation

[0063] Here, X wellNth_initial , Y wellNth_initial are the X coordinate and Y coordinate on the first drive stage where the Nth initially detected micro reaction tank 101 is disposed at the reference position of the captured image of the first imaging device 107. X stage_refer , Y stage_refer are the X coordinate and Y coordinate on the first drive stage used as a reference when the micro reaction tank 101 is imaged. Δx wellNth_image , Δy wellNth_image are the in-image X coordinate and Y coordinate (relative coordinates from the reference position) of the Nth micro reaction tank 101 on the captured image of the first imaging device 107. α is the coefficient for converting from the pixel position information in the captured image of the first imaging device 107 to the XY coordinate information of the first drive stage 106. For example, the reference position of the captured image is set as the center position, and X stage_refer , Y stage_refer are the current X coordinate and Y coordinate on the first drive stage at the time of imaging, and Δx wellNth_image, Δy wellNth_image When wellNth_image is set as the X coordinate and Y coordinate (relative coordinates from the reference position) within the image of the Nth micro reaction tank on the captured image, X wellNth_initial , Y wellNth_initial can be obtained.

[0064] Also, a certain specific micro reaction tank is determined in advance as the reference position, and when that micro reaction tank is arranged at the reference position (center position) of the captured image of the first imaging device 107, the X coordinate and Y coordinate on the first drive stage can be set as X stage_refer , Y stage_refer . At this time, Δx wellNth_image , Δy wellNth_image is set as the relative position from a certain specific micro reaction tank on the captured image. Similarly, with such a setting, X wellNth_initial , Y wellNth_initial can be obtained.

[0065] (iii) Define the differential distance between the first imaging device 107 and the nozzle 109 As described above, it is necessary to horizontally align the desired micro reaction tank 101 directly below the nozzle 109. Since the first imaging device 107 and the nozzle 109 are horizontally fixed in position, the "horizontal differential distance from the reference position of the captured image to the horizontal center position of the tip of the nozzle" between the first imaging device 107 and the nozzle 109 is basically a fixed value. This horizontal differential distance is determined in advance based on the layout information in mechanical design.

[0066] Fig. 8 is a top view for explaining the differential distance between the first imaging device 107 and the nozzle 109. As shown in Fig. 8, ΔX nozzule_initial , ΔY nozzule_initial are defined as the distance from the reference position of the captured image of the first imaging device 107 to the horizontal center position of the tip of the nozzle 109 as the differential distance in the X coordinate and Y coordinate directions on the first drive stage.

[0067]

Number

[0068] (iv) Correct the differential distance between the first imaging device 107 and the nozzle 109 The "horizontal differential distance from the reference position of the captured image to the horizontal center position of the tip of the nozzle" described above may shift slightly by several to several tens of micrometers due to misalignment of mechanical parts caused by temperature fluctuations, etc., or the degree of freedom of horizontal movement of the tip of the nozzle 109 in actual operation. The dimensional difference relationship between the micro reaction tank 101 and the tip of the nozzle 109 needs to be aligned to within several micrometers or less. Therefore, if this minute shift is not corrected, it will cause a factor for the occurrence of dropping failure (refer to FIG. 1B) during alignment. Therefore, the value of the differential distance is not a fixed value, and by correcting it at an appropriate timing, dropping failure can be prevented.

[0069] FIG. 9A is a plan view for explaining a method of correcting the differential distance between the first imaging device 107 and the nozzle 109. First, prepare a flat substrate 112 for the shape trace for providing a nozzle shape trace 111 having the same shape as the tip shape of the nozzle 109, and fix it on the first drive stage 106. Next, using the first drive stage 106, move the XY coordinates so that the flat substrate 112 for the shape trace is disposed directly below the first imaging device 107, and set the "location where the nozzle shape trace 111 on the flat substrate 112 for the shape trace should be provided" at the reference position of the captured image.

[0070] FIG. 9B is a captured image of the flat substrate 112 for the shape trace. As shown in FIG. 9B, X stamp_before , Y stamp_before are set as the X coordinate and Y coordinate on the first drive stage at the "location where the nozzle shape trace 111 on the flat substrate 112 for the shape trace should be provided".

[0071]

Number

[0072] FIG. 10A is a top view for explaining the movement of the first drive stage 106 for providing the nozzle shape trace 111. As shown in FIG. 10A, the first drive stage 106 is moved so that the nozzle 109 is disposed directly above the place where the nozzle shape trace 111 on the planar substrate 112 for shape trace should be provided, by adding the preset "horizontal direction difference distance from the reference position of the captured image to the horizontal direction center position of the tip of the nozzle" to the XY coordinates of the first drive stage 106. Specifically, the first drive stage 106 is moved to the X coordinate and Y coordinate calculated by the following formula (2).

[0073] [Number]

[0074] Here, X stamp_nozzule , Y stamp_nozzule are the X coordinate and Y coordinate on the first drive stage 106 where the nozzle 109 is disposed directly above the place where the nozzle shape trace 111 on the planar substrate 112 for shape trace should be provided.

[0075] FIG. 10B is a captured image of the second imaging device 114 when providing the nozzle shape trace 111. As shown in FIG. 10B, the second drive stage 108 is used to lower the nozzle 109 to the Z coordinate where it contacts the planar substrate 112 for shape trace, and the nozzle shape trace 111 is provided on the planar substrate 112 for shape trace.

[0076]

[0077] FIG. 11A is a top view for explaining the movement of the first drive stage 106 for photographing the nozzle shape trace 111. As shown in FIG. 11A, the first drive stage 106 is used to return the XY coordinates to the place where the nozzle shape trace 111 on the planar substrate 112 for shape trace should be provided, and the nozzle shape trace 111 is photographed.FIG. 11B is a photographed image of the nozzle shape trace 111. If the reference position of the photographed image coincides with the center position of the nozzle shape trace 111 in the image, no correction is necessary. However, if each position does not coincide, correction of the deviation amount for that portion is required. Using a conversion coefficient that converts the pixel deviation amount in this image into coordinate values on the drive stage from the pixel values of the image, the "horizontal direction difference distance from the reference position of the photographed image to the horizontal direction center position of the tip of the nozzle" between the first imaging device 107 and the nozzle 109, which is registered in advance, is corrected and added, so that correct correction can be performed. Specifically, the corrected coordinates of the difference distance between the first imaging device 107 and the nozzle 109 can be calculated by the following formula (3).

[0078] [Number]

[0079] Here, Δx stamp , Δy stamp are the X coordinate and Y coordinate (relative coordinates from the reference position) in the image of the newly formed nozzle shape trace 111 on the photographed image of the first imaging device 107. ΔX nozzule_modified , ΔY nozzule_modified are the corrected difference distances in the X coordinate and Y coordinate directions on the first drive stage from the reference position of the photographed image of the first imaging device 107 to the horizontal direction center position of the tip of the nozzle 109.

[0080] As a method for forming the nozzle shape trace 111, for example, it is possible to apply ink (such as an aqueous solution of amide black). The nozzle 109 sucks the aqueous solution of amide black, and a small amount of the aqueous solution of amide black is attached to a predetermined position of the flat substrate 112 for the shape trace, and the nozzle shape trace 111 can be provided. Alternatively, as the material of the flat substrate 112 for the shape trace, a resin having mechanical plasticity may be used. In this case, the nozzle shape trace 111 can be formed by pressing the tip of the nozzle 109 against the flat substrate 112 for the shape trace.

[0081] (v) Aligning the micro reaction tank and the nozzle By using the "coordinates at which the horizontal center position of the container opening is arranged at the reference position of the captured image" on the XY coordinates of the first drive stage 106 and the "horizontal distance difference from the reference position of the captured image to the horizontal center position of the tip of the nozzle" obtained in steps (ii) to (iv), the horizontal center positions of the micro reaction tank 101 and the nozzle 109 can be aligned.

[0082] Figure 12A is a top view for explaining the movement of the first drive stage 106 when aligning the horizontal center positions of the micro reaction tank 101 and the nozzle 109. As shown in Figure 12A, by moving the first drive stage 106 to the XY coordinates obtained by adding the above two variables, the horizontal center position of the micro reaction tank 101 can be moved to a position that coincides with the horizontal center position of the nozzle 109. For example, the target XY coordinates are calculated by the following formula (4).

[0083]

Equation

[0084] Here, X wellNth_initial , Y wellNth_initial are the X coordinate and Y coordinate on the first drive stage where the Nth micro reaction tank is arranged directly below the nozzle.

[0085] Figure 12B is a captured image of the second imaging device 114 in a state where the nozzle 109 and the micro reaction tank 101 are aligned. In a state where the above-described precise alignment is achieved, by moving the second drive stage 108 to the Z coordinate at which the nozzle 109 descends into the micro reaction tank 101, the tip of the nozzle 109 can be arranged directly above the micro reaction tank 101. Then, the nozzle 109 can be accurately lowered into the micro reaction tank 101, and the fine particle solution can be filled into the micro reaction tank 101 from inside the nozzle 109.

[0086] The nozzle 109 is typically a glass capillary and has a tapered shape towards the tip. It is only necessary that the nozzle 109 can be inserted into the interior of the micro reaction vessel 101 until the tip of the nozzle 109 contacts the bottom 103b of the micro reaction vessel 101, and the outer diameter of the nozzle 109 may have a constant shape.

[0087] <Method for filling fine particles> FIG. 13 is a schematic diagram for explaining a method for filling fine particles including the alignment method of the present embodiment. As shown in FIG. 13, the method for filling fine particles of the present embodiment includes basic steps (a) to (f) for filling the fine particle solution 110 into the interior of the micro reaction vessel 101.

[0088] (a) Suction of the fine particle solution 110 by the nozzle 109 Using the first drive stage 106, it moves to the XY coordinate position where the solution tube holding the fine particle solution 110 is arranged directly below the nozzle 109. Next, using the second drive stage 108, the nozzle 109 descends to the Z coordinate position where it contacts the fine particle solution 110, and sucks the fine particle solution 110 into the interior of the nozzle 109. Thereafter, the second drive stage 108 is moved to the origin coordinates, and this step is completed.

[0089] The solution tube may be a single PCR tube or an 8 - connected type PCR tube, or a 96 - well microplate or a 384 - well microplate. For these solution tubes, by using an appropriate automatic stirrer and performing a stirring process immediately before the solution suction in this step, a good dispersion state of the fine particles can be maintained.

[0090] The suction of the fine particle solution 110 by the nozzle 109 utilizes capillary action to suck the fine particle suspension by immersing the empty nozzle 109 into the solution. Alternatively, a syringe pump or the like is connected to the tip of the empty nozzle 109 on the side opposite to the tip where the fine particle suspension is dispensed, and the fine particle solution 110 can be discharged and sucked through a discharge and suction operation by a pressurization and depressurization operation.

[0091] (b) Alignment of the micro reaction vessel 101 and the nozzle 109 The first drive stage 106 is moved so that the horizontal center positions of the micro reaction vessel 101 and the nozzle 109 coincide with each other by the alignment method described above. Then, the second drive stage 108 is moved so that the tip of the nozzle 109 is disposed at a position about 100 to 300 μm away in the Z direction directly above the micro reaction vessel 101.

[0092] (c) Concentration (sedimentation) of fine particles The fine particles in the fine particle solution 110 are allowed to settle by gravity to the tip of the nozzle 109 and concentrated by leaving them for a predetermined time, and a high-concentration fine particle solution 113 having a predetermined volume and a predetermined concentration is obtained. Alternatively, when the fine particles are magnetic fine particles, the fine particles are moved to the tip of the nozzle 109 by applying a magnetic force from the outside and concentrated to obtain a high-concentration fine particle solution 113. The predetermined volume is set to a value including, for example, the number of fine particles that occupy 20% or more of the volume of the micro reaction vessel 101 when the high-concentration fine particle solution 113 is filled in the micro reaction vessel 101. The predetermined concentration (high concentration) is a fine particle concentration that becomes a concentration of 10% or more of the highest concentration while maintaining the dispersed state of the fine particles as described above.

[0093] From the time when the state of step (c) is constructed, for example, the tip of the nozzle 109 is photographed by the second imaging device 114, and the temporal and spatial changes of the fine particles in the nozzle 109 are observed. In this way, the volume of the high-concentration fine particle solution 113 in the nozzle 109 is estimated. At this time, when the height of the liquid surface of the high-concentration fine particle solution 113 in the nozzle 109 reaches a predetermined value, it can be determined that a high-concentration fine particle solution 113 having a predetermined volume has been obtained. At this time, since the second imaging device 114 is installed in the side direction with respect to the nozzle 109, it becomes easy to observe the fine particles in the nozzle 109.

[0094] Alternatively, by irradiating the tip of the nozzle 109 with light and measuring the amount or distribution of light such as scattered light, absorbed light, or transmitted light, it is also possible to determine whether a high-concentration fine particle solution 113 of a predetermined volume has been obtained. In this case, it is necessary to appropriately control the spot size and position of the light near the nozzle 109. Also, light may be irradiated at a position at a predetermined height of the nozzle 109, and based on the scattered light, absorbed light, or transmitted light at that position, it may be determined that a high-concentration fine particle solution 113 of a predetermined volume has been obtained.

[0095] (d) Lower the nozzle 109 into the micro reaction vessel 101 If a high-concentration fine particle solution 113 of a predetermined volume is obtained in step (c), lower the nozzle 109, and by the surface tension of the high-concentration fine particle solution 113 at the tip of the nozzle 109, bring the high-concentration fine particle solution 113 into contact with the bottom 103b or the side wall surface 103c of the micro reaction vessel 101. As a result, the high-concentration fine particle solution 113 wets and spreads within the micro reaction vessel 101. Then, the high-concentration fine particle solution 113 oozes out to at least a part of the side wall surface 103c of the micro reaction vessel 101, and the high-concentration fine particle solution 113 is in a state of being in contact with the bottom 103b and the side wall surface 103c of the micro reaction vessel 101.

[0096] (e) Pull up the nozzle 109 When the high-concentration fine particle solution 113 comes into contact with the bottom 103b and the side wall surface 103c of the micro reaction vessel 101 in step (d), pull up the nozzle 109 at a predetermined speed to fill the micro reaction vessel 101 with the high-concentration fine particle solution 113. The principle of this filling will be described later.

[0097] (f) Filling completion The state where the micro reaction vessel 101 is filled with the high-concentration fine particle solution 113 is defined as filling completion.

[0098] <Principle> FIG. 14A and 14B are cross-sectional schematic diagrams for explaining the principle of the fine particle filling method according to the present embodiment, in which the tip of the nozzle 109 and the internal space of the micro reaction tank 101 are shown enlarged. FIG. 14A shows a state immediately after the high-concentration fine particle solution 113 in the nozzle 109 contacts the bottom 103b of the micro reaction tank 101. As shown in FIG. 14A, in step (d), the nozzle 109 is lowered. When a part of the lower end of the nozzle 109 contacts the bottom 103b of the micro reaction tank 101, the high-concentration fine particle solution 113 wets and spreads (contacts) on the bottom 103b of the micro reaction tank 101.

[0099] FIG. 14B shows a state in which the high-concentration fine particle solution 113 also contacts the side wall surface 103c of the micro reaction tank 101. As shown in FIG. 14B, the high-concentration fine particle solution 113 oozes out until it contacts at least a part of the side wall surface 103c.

[0100] Here, let the radius of the bottom 103b of the micro reaction tank 101 be Rw, the radius of the outer wall 109a of the tip of the nozzle 109 be Rc, the contact angle of the high-concentration fine particle solution 113 with respect to the inner wall of the micro reaction tank 101 be θw, and the contact angle of the high-concentration fine particle solution 113 with respect to the nozzle outer wall 109a be θc.

[0101] When the suction pressure (ΔP) represented by the following formula (5) is positive, the high-concentration fine particle solution 113 can be drawn out from the nozzle 109. Therefore, in step (e), by pulling up the nozzle 109 at a predetermined speed while maintaining the state of ΔP>0, the micro reaction tank 101 can be filled with the high-concentration fine particle solution 113. The pulling-up speed of the nozzle 109 can be constant as long as the state of ΔP>0 can be maintained, or it may be pulled up while accelerating or decelerating.

[0102]

Equation

[0103] If the particle concentration and volume of the high-concentration particle solution 113 are set under appropriate conditions, the concentration can be set to 50% or more of the highest concentration while maintaining the dispersion state of the high-concentration particle solution 113, and the volume of the high-concentration particle solution 113 can be made equal to or greater than the volume of the micro reaction vessel 101. As a result, at least 50% or more of the volume of the micro reaction vessel 101 is filled with particles. If the particle concentration and volume of the high-concentration particle solution 113 are set under more appropriate conditions, 80% or more of the volume of the micro reaction vessel 101 can be occupied by particles in a dry state. By appropriately setting the concentration and volume in this way, there is no need to additionally discharge the high-concentration particle solution 113 filled in the micro reaction vessel 101.

[0104] The volume of the high-concentration particle solution 113 formed in the nozzle 109 can also be made larger than the volume of the micro reaction vessel 101. Assuming that the micro reaction vessel 101 is formed without error, the opening angle is formed at 90°, and there is no deposit on the inner wall (the shape in Fig. 4), when the pulled-up nozzle 109 in step (e) causes the infiltrated high-concentration particle solution 113 to reach the upper end of the micro reaction vessel 101, the contact angle θw becomes less than 90°. Substituting this contact angle θw into Equation (1) results in ΔP < 0, and the discharge of the high-concentration particle solution 113 from the nozzle 109 stops. Therefore, by appropriately setting the conditions, even if the volume of the high-concentration particle solution 113 is larger than the volume of the micro reaction vessel 101, the discharge stops at the upper end of the micro reaction vessel 101, so the high-concentration particle solution 113 does not overflow from the micro reaction vessel 101.

[0105] The micro reaction vessel 101 shown in Fig. 4 has a through-hole 105 for cell capture required for a single-cell analysis chip, but the through-hole 105 is not essential for the above particle filling principle. In fact, according to the above principle, it is possible to fill particles into a container without the through-hole 105. That is, the container filled with particles by the technology of this embodiment is not limited to the micro reaction vessel 101 of a single-cell analysis chip, and any container into which the nozzle 109 can be inserted is acceptable. The cross-sectional shape of the container filled with particles is not limited to a rectangle, and other shapes such as a trapezoid whose bottom is narrower than the opening can be adopted.

[0106] The method of filling the microparticles into a single microreactor 101 has been described above. When filling microparticles into a plurality of microreactors 101, it can be realized by repeating the above operations. Specifically, after completing the filling into the first microreactor according to the above steps (a) to (f), the above steps (a) to (f) are also carried out for the second microreactor, and the steps for filling microparticles can be repeated in the same way hereinafter.

[0107] The same operation is applicable when filling different types of microparticles into different microreactors 101 respectively. For example, in order to fabricate a single-cell analysis chip described in Non-Patent Document 5 and Patent Document 1, an example of individually filling a plurality of types of microparticles with different DNA probes fixed thereon into preset microreactors will be described.

[0108] First, prepare solution tubes of microparticle solutions 110 in which various microparticles are suspended respectively, and carry out step (a) for the solution tube containing the desired solution. Then, carry out steps (b) to (f) to complete the filling of the desired microparticle solution 110 into the desired microreactor 101.

[0109] Next, move the first drive stage 106 to the XY coordinate position where the waste liquid tube is arranged for the nozzle 109, and by driving the second drive stage 108 and operating a syringe pump or the like, discard the excess microparticle solution 110 in the nozzle 109 into the waste liquid tube.

[0110] Next, for cleaning the nozzle 109, move the nozzle 109 to the XY coordinate positions where a plurality of solution tubes containing a plurality of types of cleaning liquids are arranged. Then, by driving the second drive stage 108 and operating a syringe pump or the like, immerse the nozzle 109 and repeat the suction and discard of the cleaning liquid. Thereby, the microparticle solution remaining in the nozzle 109 and the attached microparticles are completely removed by cleaning. As the cleaning liquid, for example, pure water such as ion-exchanged water filtered through a filter and alcohol such as ethanol can be used.

[0111] The above operations are repeated to fill the following types of fine particles into the micro reaction vessels at the following positions.

[0112] <Operation of the fine particle filling system> Figure 15 is a flowchart showing the operation of the fine particle filling system 100.

[0113] In step S1, the control device 120 constructs the overall initial state of the fine particle filling system 100. Specifically, first, the user places the planar substrate 102 on the first drive stage 106 and arranges the storage container 118 for the fine particle solution 110 and the storage container 118 for the cleaning liquid at predetermined positions. Then, when the user inputs an operation start instruction, for example, from the input device of the control device 120, the control device 120 checks whether the planar substrate 102, the nozzle 109, and the storage container 118 are all located at predetermined initial positions. If any of these is not at the initial position, it is moved to the initial position by the drive mechanism 115.

[0114] In step S2, the control device 120 constructs the initial state for each micro reaction vessel 101. Specifically, the control device 120 drives the drive mechanism 115 and the dispensing mechanism 117 to suck the cleaning liquid into the nozzle 109 and discharge it into a waste liquid container (not shown) to clean the nozzle 109. Then, the control device 120 causes the nozzle 109 to suck the fine particle suspension in step (a) and move it directly above a predetermined micro reaction vessel 101 in step (b).

[0115] In step S3, step (c) is performed. The control device 120 determines whether the fine particles in the fine particle solution 110 in the nozzle 109 have settled to form a high-concentration fine particle solution 113 with a predetermined fine particle concentration and a predetermined volume. Here, since the high-concentration fine particle solution 113 is optically dense, it can be observed as a black region by the second imaging device 114 compared to the low-concentration region. The control device 120 calculates the size of the black region from the contrast of the image received from the second imaging device 114, and calculates the concentration and volume of the high-concentration fine particle solution 113 based on this. Alternatively, by detecting the scattering surface at the boundary, the size of the black region of the high-concentration fine particle solution 113 is calculated, and the concentration and volume are calculated in the same way. When the fine particles are magnetic fine particles, by utilizing the magnetic field of the magnet 121, after constructing the state of step S2, for example, by holding for 3 to 10 seconds, it can be determined that the concentration and volume of the high-concentration fine particle solution 113 have reached predetermined values. When gravity is utilized, this holding time is about 50 to 120 seconds.

[0116] In step S4, step (d) is performed. The control device 120 drives the drive mechanism 115 to lower the nozzle 109 and brings the tip of the nozzle 109 into contact with the bottom 103b of the micro reaction tank 101. Due to the contact, the high-concentration fine particle solution 113 at the tip of the nozzle 109 wets the bottom 103b and the side wall surface 103c of the micro reaction tank 101. In the present disclosure, it is defined that the nozzle 109 is in contact when the high-concentration fine particle solution 113 wets the inner wall of the micro reaction tank 101. This is because it is necessary and essential for the high-concentration fine particle solution 113 to wet both the inner wall of the micro reaction tank 101 and the outer wall 109a of the nozzle 109 to form an appropriate contact angle and for the high-concentration fine particle solution 113 in the nozzle 109 to be drawn into the micro reaction tank 101. However, in practice, since the descent of the nozzle 109 must be stopped somewhere, a sensor system 116 is provided to detect the contact.

[0117] After the contact between the nozzle 109 and the micro-reaction tank 101, for example, within 1 second, the interface of the particulate suspension enters between the nozzle 109 and the side wall surface 103c of the micro-reaction tank 101 and rises. Step (e) is performed in step S5. The control device 120 drives the drive mechanism 115 to pull up the nozzle 109 at a speed of, for example, about 1 to 130 μm / s. Thereby, the high-concentration particulate solution 113 in the nozzle 109 is moved into the micro-reaction tank 101 to fill the micro-reaction tank 101 with the particulates. Typically, since the volume in the high-concentration particulate solution 113 filled in the micro-reaction tank 101 is 1 nL or less, it evaporates within a few seconds and only the particulates remain in the micro-reaction tank 101. In the drying process, the particulates in the high-density state aggregate while attracting each other due to the effects of the surface tension of the solution and the interfacial interaction between the particulates, so that the particulates in a densely packed state remain in the micro-reaction tank.

[0118] In step S6, the control device 120 determines whether all the micro-reaction tanks 101 are filled with particulates.

[0119] If NO in step S6, the control device 120 drives the drive mechanism 115 and the dispensing mechanism 117 to discard the excess particulate solution 110 in the nozzle 109 into the waste liquid container for filling the next micro-reaction tank 101 with particulates.

[0120] Next, returning to step S2, the control device 120 drives the drive mechanism 115 and the dispensing mechanism 117 to suck and discharge two types of cleaning liquids to clean the nozzle 109. Next, the control device 120 drives the drive mechanism 115 and the dispensing mechanism 117 to suck a suspension of particulates different from the previously filled particulates (for example, particulates with a cell recognition tag sequence different from the previous one) into the nozzle 109 and move it above the micro-reaction tank 101 different from the previous one. Thereby, an initial state for the next micro-reaction tank 101 is constructed. Thereafter, steps S3 to S6 are executed in the same manner as above.

[0121] For the second and subsequent micro reaction vessels 101, steps S2 to S6 are similarly repeated. When the result in step S6 is YES, the control device 120 terminates the operation of the fine particle filling system 100.

[0122] In order for the fine particles to be filled in the micro reaction vessel 101, the fine particles need to be concentrated in step S4 (step (c)), and in the next steps S4 and S5 (steps (d) and (e)), the fine particle solution 110 gets wet and reaches the side wall surface 103c of the micro reaction vessel 101, and the contact angle between the inner wall of the micro reaction vessel 101 and the fine particle solution 110, and the contact angle between the outer wall of the nozzle 109 and the fine particle solution 110 need to satisfy the conditions of formula (1). In order to satisfy this condition, it is necessary to appropriately select the respective materials that determine the hydrophilicity of the inner wall of the micro reaction vessel 101 and the hydrophilicity of the outer wall of the nozzle 109a, and the inner diameter of the micro reaction vessel 101 and the outer diameter of the tip of the nozzle 109.

[0123] When the target fine particles are magnetic fine particles, it is necessary to shorten the time by means of a magnetic field in order to shorten the filling time. In that case, a magnet 121 can be arranged directly below the planar substrate 102 having the micro reaction vessel 101. Thereby, when the nozzle 109 approaches the micro reaction vessel 101 within a series of steps, the fine particles can be automatically moved to the tip of the nozzle 109 by the remote action of the magnetic field. On the other hand, when this configuration is adopted, it becomes practical to use photographing means based on the reflected image of the micro reaction vessel 101 from a top-down view from directly above. Therefore, the first imaging device 107 needs to be installed on the same vertical side as the nozzle 109. Under the limiting conditions of such a configuration, the technique of the present embodiment is particularly useful for realizing a short-time and accurate alignment method.

[0124] <Technical Effect> As described above, by using the fine particle filling method using the alignment method of the micro reaction vessel and the nozzle according to the first embodiment, the tip of the nozzle 109 is surely lowered into the micro reaction vessel, leading to an increase in the success probability of the fine particle filling method using the subsequent fine particle filling principle.

[0125] [Second Embodiment] In the first embodiment, a technique for successfully filling microparticles by accurately aligning a micro reaction vessel formed on a planar substrate and a nozzle was described. In the second embodiment, an example in which the above-described technology of the present disclosure is applied to the filling of microparticles into a micro reaction vessel provided in a chip (substrate) of a single cell analysis device will be described.

[0126] <Example of chip configuration> FIG. 16A is a top view showing a single cell analysis chip 122 (substrate). FIG. 16B is a cross-sectional view showing a part of the cross-section of the chip 122. FIG. 16C is a photographed image of the chip 122 by the first imaging device 107. On the chip 122, 10×10 (total 100) micro reaction vessels 123 are formed at equal intervals. The micro reaction vessel 123 is provided with a through-hole 124 for cell capture, and thus has a chip structure capable of isolating cells for single cell analysis. A chip 122 having such a structure can be molded, for example, using imprinting technology with dimethylpolysiloxane as a material. Alternatively, it can be molded by molding with a mold using an elastomer material. The through-hole 124 can be drilled by laser processing, but other particle beam processing or molding by a mold may also be used.

[0127] Various materials can be used as the material of the chip 122. For example, various resin materials (polycarbonate-based, polycycloolefin-based, polyolefin-based, polypropylene-based, polyethylene-based, acrylic-based, acrylonitrile-based, polystyrene-based resin, polydimethylsiloxane-based resin, etc.), metal materials, semiconductor materials such as silicon, oxide materials such as quartz and alumina, and amorphous materials such as glass can be mentioned. These materials can also be used in combination for processing or for controlling surface wettability.

[0128] Figures 16A and 16B show an example of the dimensions of chip 122. In the illustrated example, the micro reaction vessel 123 is cylindrical with a diameter of 75 μm and a depth of 70 μm. The center-to-center spacing between adjacent micro reaction vessels 123 is 105 μm. The pore diameter of the through-hole 124 is 3 μm at its smallest and its length is 30 μm. Note that the above sizes are an example of sizes suitable for a single cell analysis device, and as the size of the micro reaction vessel 123 to be filled with fine particles, either the diameter or the depth may be anywhere from several μm to several hundred μm. Also, the shape of the micro reaction vessel 123 does not necessarily have to be cylindrical. As long as there is an opening capable of filling fine particles, the micro reaction vessel 123 may have a polygonal prism shape such as a quadrangular prism, or a tapered shape such as a truncated cone or a quadrangular frustum.

[0129] In a single cell analysis device, the fine particles filled in the micro reaction vessel 123 of chip 122 are fine particles for performing gene expression analysis that measures the amount of mRNA in a single cell for each gene sequence. Depending on the analysis purpose, by changing the molecule fixed to the fine particles, it can be applied to various single cell analyses.

[0130] In this embodiment, magnetic fine particles (diameter 1 μm) with a large number of streptavidin molecules fixed on the surface will be used. For use in single cell analysis, in order to capture the mRNA to be measured on the surface of the fine particles and introduce a barcode sequence for cell identification into the nucleic acid sample, a DNA probe is fixed. At the 5' end of the DNA probe, biotin for strongly binding to streptavidin is fixed.

[0131] The sequence of the DNA probe has, in order from the 5' end, a common sequence for PCR, a tag sequence for molecular recognition, a tag sequence for cell identification (7 bases), and a continuous sequence of Ts for mRNA capture at the 3' end. The tag sequence for cell identification is 100 known sequences of 7 bases, and the tag sequence for molecular recognition is a random sequence of 7 bases.

[0132] The length and position of the tag sequences for cell identification may be different from the above. For example, three types of 7-base known sequences may be arranged in such a way that they sandwich two 3-base known sequences at two positions, or they may be arranged as random sequences instead of known sequences. Similarly, the tag sequences for molecular recognition can also have their length and arrangement changed.

[0133] The method of immobilizing DNA probes on microparticles is as follows. First, while taking care to avoid cross-contamination between DNA probes with different tag sequences for cell identification, the microparticles and DNA probes are mixed in 1x B&W buffer (5 mM Tris-HCl (pH 7.5), 0.5 mM EDTA, 1 M NaCl), and a binding reaction is carried out at room temperature for about 30 minutes. Then, after washing twice with 1x B&W buffer, the microparticle solution is obtained by resuspending in 10 mM Tris pH 8.0 buffer. As microparticle solutions, 100 types with different tag sequences for cell identification were prepared in 100 different storage tubes.

[0134] <Filling of the Chip with Microparticles Immobilized with DNA Probes> A method of filling the chip 122 with microparticles immobilized with DNA probes will be described.

[0135] The storage container 118 containing the microparticle solutions immobilized with the above 100 different cell identification tags is placed on the first drive stage 106. As storage containers for the washing liquid, two containers containing filter-filtered ion-exchanged water (pure water) and two containers containing the buffer solution in which the microparticles were resuspended are installed. The storage containers are, for example, PCR microtubes or 96-well microplates. The washing liquid may be 100% EtOH instead of the buffer solution in which the microparticles were resuspended.

[0136] Using the microparticle filling system 100, operations from step S1 to S6 are performed according to a control program. At this time, 20 chips each having 100 microreactors are installed. The microparticle filling system 100 determines the microparticle solution to be filled out of 100 types of microparticle solutions for each of the total 2000 microreactors. Then, the microparticle filling system 100 basically executes operations of predetermined steps (a) to (f) in the same manner as in the first embodiment. At this time, one type of each of 100 corresponding microparticle solutions is filled into the chip having 100 microreactors.

[0137] As the nozzle 109, a linear cylindrical one made of glass capillary with an outer diameter of 71 μm and an inner diameter of 54 μm was used. Also, 61 μm was adopted as the amount of a predetermined amount of high-concentration suspension to be determined in step (c). The pulling-up speed of the nozzle 109 in step (e) is increased at a speed of 8 μm / s.

[0138] FIG. 17 is a diagram showing a workflow of microparticle filling according to the second embodiment. FIG. 17 shows steps S2 to S5 described in the first embodiment in more detail. In the above-described configuration, microparticle filling was tried on the chip 122 according to the workflow of FIG. 17.

[0139] In step S2-0, the tip of the nozzle 109 is cleaned with a cleaning liquid. Then, a nozzle shape trace is formed, and the "horizontal direction difference distance from the reference position of the photographed image to the horizontal direction center position of the tip of the nozzle" is detected and corrected. The nozzle shape trace at this time was formed using ink.

[0140] In the next step S2-1, the particulate solution was aspirated from the solution tube No.i into the nozzle. In step S2-2, the micro reaction tank No.i in the chip No.j was moved directly below the first imaging device. In step S2-3, the coordinates at which the horizontal center position of the container opening is arranged at the reference position of the captured image of the micro reaction tank No.i were detected or corrected. In step S2-4, the tip of the nozzle 109 and the micro reaction tank No.i were aligned by the first drive stage, and the tip of the nozzle 109 was lowered directly above the micro reaction tank. Steps S3 to S5 were executed as in the above-described first embodiment. By attempting to continuously fill the micro reaction tanks to be filled with the same particulate solution in this way, the number of washings can be reduced so that washing is only required when changing the solution tube.

[0141] FIG. 18 is a diagram for explaining image processing in filling of particulates into the single cell analysis chip according to the second embodiment. At the upper left of FIG. 18, a captured image of the first imaging device 107 in step S2-3 is shown. At the lower left and upper right of FIG. 18, captured images of the second imaging device in steps S3 and S4 are shown, respectively. At the lower right of FIG. 18, a schematic diagram showing the formation region of the nozzle shape trace in step S2-0 is shown. As shown at the lower right of FIG. 18, the chip 122 was fixed by being attached to a cover glass, and the formation region of the nozzle shape trace was specified in an arbitrary region of the cover glass and executed.

[0142] FIG. 19 is a captured image of the chip 122 showing the result of particulate filling according to the second embodiment. As shown in FIG. 19, it can be seen that it is possible to accurately fill the particulates without displacement in each of the 100 micro reaction tanks of the chip 122.

[0143] <Regarding the concentration time of the particulate solution> Next, the results of comparing the concentration time of the particulate solution in the nozzle 109 when the magnet 121 is used and when it is not used will be described. First, a concentration of 1.0×10 9The microparticle solution at [X] particles / mL was aspirated into nozzle 109. As magnet 121, a rectangular neodymium magnet with a surface magnetic flux density of 540 mT, 2.8 mm square, and a length of 10 mm was used and placed directly below chip 122. The tip of nozzle 109 was stopped 220 μm directly above the microreaction vessel 123, and the distance from the surface of the rectangular neodymium magnet to the tip of nozzle 109 was stationary at 540 μm (magnetic flux density 400 mT). The waiting time for the high-concentration microparticle solution 113 to reach the target height of 61 μm was measured. At the same time, a microparticle solution with a concentration of 5.0×10 10 particles / mL (50 times concentrated when using the magnet) was allowed to settle naturally by gravity. The waiting time for the high-concentration microparticle solution 113 to reach the target height of 61 μm in this case was also measured. The results are shown in Fig. 20.

[0144] Fig. 20 is a graph showing the measurement results of the concentration time of the microparticle solution by magnetic force or gravity. As shown in Fig. 20, it was confirmed that the natural sedimentation by gravity took an average of 52 seconds, while the accelerated sedimentation by the magnetic field took an average of 5 seconds.

[0145] [Third Embodiment] In the third embodiment, a first image processing technique for detecting the horizontal position (position on the XY coordinates) on the first drive stage of the microreaction vessel required during alignment will be described. Specifically, during step (ii) in the first embodiment, an image processing technique for collectively detecting the horizontal positions of a plurality of microreaction vessels in a short time will be described.

[0146] Figures 21A to 21E are conceptual diagrams for explaining a first image processing technique in alignment according to the third embodiment. A planar substrate 102 provided with a plurality of micro reaction vessels 101 is moved directly below a first imaging device 107 installed at the upper part of a first drive stage 106. The first imaging device 107 captures an aerial image of a plurality of micro reaction vessel groups (Fig. 21A). For each individual micro reaction vessel included in this captured image, pattern matching is performed using an image of a micro reaction vessel (Fig. 21B) registered in advance as a reference image. The positions on the captured images of the micro reaction vessels determined to be images similar to the reference image by the pattern matching method are collectively detected (Fig. 21C). At this time, the pattern matching algorithm may be any method. For example, as shown in Fig. 21C, a method of performing pattern matching after binary conversion of the original image is an example.

[0147] Next, the positions on the captured images of the detected micro reaction vessels are calculated as the pixel deviation amounts from the reference position (center position: also a predetermined XY coordinate position on the first drive stage 106) of the captured images. Next, using a conversion coefficient for converting from the pixel values of the images to the coordinate values on the first drive stage, the positions of each micro reaction vessel are converted into position information as XY coordinates on the first drive stage 106. In this way, the horizontal positions of the micro reaction vessels can be collectively detected from the captured images (Fig. 21D). Specifically, it can be calculated using the formula (1) described in the first embodiment. For example, for the sake of convenience, the micro reaction vessel in Fig. 21A is set as number 1 as a specific micro reaction vessel in advance, and the micro reaction vessel numbered 1 is determined as the reference position. When the micro reaction vessel numbered 1 is arranged at the reference position (center position) of the captured image of the first imaging device 107, the X coordinate and Y coordinate on the first drive stage are set as X stage_refer , Y stage_refer , and Δx wellNth_image , Δy wellNth_image are set as the relative positions from the micro reaction vessel numbered 1 on the captured image, and thus the X wellNth_initial , Y wellNth_initial of each micro reaction vessel in the captured image can be obtained.

[0148] At this time, in order to improve the determination accuracy of the horizontal position of the micro reaction tank 101, if the dimension information of the arrangement grid in which a plurality of micro reaction tanks are arranged in advance is known, this information can be used. For example, when the micro reaction tanks are arranged in an array at equal intervals (such as a square grid or a triangular grid), by giving the XY coordinate position information on the first drive stage 106 of a certain reference point (for example, the four points at the four corners) as prior information, it is possible to theoretically predict the XY coordinate position information of each micro reaction tank on the first drive stage 106. In reality, the planar substrate 102 having the micro reaction tank 101 often has the characteristics of elastic deformation. When the planar substrate 102 is installed and fixed on the first drive stage 106, the planar substrate 102 often expands and contracts, resulting in a non-uniform interval arrangement. In such a case, by performing a homography transformation matrix using the homography transformation matrix calculated based on the position coordinates of the four points at the four corners with respect to the above XY coordinate position information, it is possible to theoretically predict including the influence of expansion and contraction.

[0149] FIG. 22A is a diagram showing the predicted values and detected values of the XY coordinate positions of each micro reaction tank on the first drive stage 106 in a certain chip. Using the theoretically predicted XY coordinate position information of the micro reaction tank obtained as described above as reference information, if the position information of the actual micro reaction tank detected by the above pattern matching method is included within a certain threshold from this position information, it can be determined that the position information of this detected micro reaction tank is a correct value. By using such reference information, the risk of false detection of the horizontal position of the micro reaction tank can be reduced.

[0150] Also, even when the horizontal position information of some of the micro reaction tanks cannot be locally obtained due to reasons such as unclear contrast of the captured image, this can be compensated by using the horizontal position information of the plurality of actual micro reaction tanks obtained above.

[0151] FIG. 22B is a diagram showing the detected values and complemented values of the XY coordinate positions of each micro reaction tank. As shown in FIG. 22B, when the position information of a certain micro reaction tank cannot be detected by the pattern matching method, it is also possible to complement the position information by linear interpolation using the position information of a plurality of correctly detected micro reaction tanks around it. This method can be used when the original micro reaction tanks are arranged at equal intervals.

[0152] When the contrast of the captured image is unclear and it is necessary to increase the probability of detecting the micro reaction tank in the captured image by the pattern matching method, it is also effective to capture a plurality of images while continuously changing the parameters of the imaging device at the same position to increase the probability of obtaining good contrast. Specifically, a plurality of images are acquired while continuously changing the analog gain value, digital gain value, or exposure time of the imaging device, and an attempt is made to detect the micro reaction tank in the captured image from these image groups by the above pattern matching method (FIG. 21E). The position information of the plurality of micro reaction tanks obtained from the plurality of images is combined and merged. When the position information of the micro reaction tanks overlaps, the position information with the highest determination score value in the pattern matching method is adopted to degenerate it into one-point information, so that the position information of the micro reaction tank can be obtained with high accuracy. At this time, it is also useful in the subsequent processing to simultaneously acquire the imaging condition information of the analog gain value, digital gain value, or exposure time of the imaging device with the highest determination score value in the pattern matching method.

[0153] In the above series of processes, when the number of micro reaction tanks is large and cannot fit within one field of view of the captured image, it is possible to perform the above series of processes in multiple fields of view and synthesize and obtain the horizontal positions of the obtained plurality of micro reaction tanks. Specifically, a plurality of images can be captured at equal interval pitches so that about 10 to 20 micro reaction tanks overlap in one field of view, and the processing can be performed so that there is no observation omission of the micro reaction tanks within the field of view (FIG. 21A).

[0154] FIG. 23 is a flowchart showing a first image processing technique according to the third embodiment. The first image processing technique is executed by the control device 120. In step S101, the XY coordinates on the first drive stage at the four corners and four points are acquired as prior information. In step S102, the area for imaging the chip is divided into N×M, and the imaging area is moved to the area where position detection is performed by the pattern matching method (repeated N×M times).

[0155] In step S103, the set value (such as the analog gain value) of the first imaging device is switched. In step S104, the pattern matching method is implemented on the captured image. In step S105, the XY coordinates on the first drive stage are converted from the detected image position of the micro reaction vessel. In step S106, it is determined whether the processing for all set values has ended. If YES in step S106, return to step S102. If NO in step S106, return to step S103.

[0156] In step S107, it is determined whether the position detection of the micro reaction vessels for all divided areas has ended. If YES in step S107, proceed to step S108. If NO in step S107, return to step S102.

[0157] In step S108, all the obtained XY coordinate information is merged and the duplicate information is degenerate to one point. In step S109, each detected value is assigned to the micro reaction vessel with the number i to be detected. In step S110, it is determined whether the detected value is within the threshold from the predicted value of the micro reaction vessel with the number to be detected. If YES in step S110, proceed to step S111. If NO in step S110, proceed to step S112.

[0158] In step S111, it is determined that the detection is successful. Then, the detected value is adopted as the XY coordinates on the first driving stage of the micro reaction tank numbered i. In step S112, it is determined that the detection fails. Then, the linear interpolation value from the detected values of the surrounding micro reaction tanks is adopted as the XY coordinates on the first driving stage of the micro reaction tank numbered i.

[0159] <Technical effect> As described above, in the third embodiment, the horizontal positions (positions on the XY coordinates) on the driving stages of a plurality of micro reaction tanks, which are necessary for alignment, can be collectively detected in a short time.

[0160] <Examples of the third embodiment> Hereinafter, specific examples of the third embodiment will be described. First, a planar substrate-like resin sheet provided with 400 micro reaction tanks per chip in 20 chips (chip numbers 1 to 20) was placed on the first driving stage. The micro reaction tanks were arranged at equal intervals in a three-way lattice pattern. As the first imaging device, a monochrome CCD camera provided above the resin sheet was used. The first driving stage 106 was moved so that chip number 1 having the micro reaction tank in the resin sheet was arranged directly below the monochrome CCD camera. Since 400 micro reaction tanks cannot fit within one field of view of the monochrome CCD camera, 400 per chip were accommodated by dividing into four parts. First, at division number 1 of chip number 1, 10 consecutive images were taken while changing the gain value of the monochrome CCD camera from a low value to a high value. For each captured image, the pattern matching method was performed to collectively detect a plurality of micro reaction tanks. At this time, the amount of deviation of each detected micro reaction tank from the reference position (center position) of the captured image was detected by image processing. At this time, the XY coordinates on the first driving stage 106 of the micro reaction tanks at the four corners of each chip number were acquired in advance, and the XY coordinates of the micro reaction tank (number 1) at the upper left corner of each chip were set as the reference coordinates.

[0161] Using the conversion coefficient (1 pixel = 1.1625 μm) for converting the relative position on the image of each micro-reaction vessel detected by the pattern matching method for the micro-reaction vessel numbered 1 and the coordinate values on the drive stage from the pixel values of the image, the XY coordinates on the first drive stage 106 of each micro-reaction vessel were calculated according to Equation (1).

[0162] Next, the first drive stage 106 was moved to the positions of split numbers 2 to 4 of chip number 1, and the same processing was performed for each. The obtained XY coordinates of a plurality of micro-reaction vessels were merged. When the position information of the micro-reaction vessels was obtained repeatedly, it was reduced to single-point information by adopting the position information with the highest determination score value in the pattern matching method. In this series of processes, when there was a micro-reaction vessel that could not be detected by the pattern matching method, linear interpolation was performed using the position information of the detected micro-reaction vessels around it. In this way, the XY coordinate position information of 400 micro-reaction vessels in chip number 1 was finally obtained. The same processing was also performed for chips numbered 2 to 20.

[0163] FIG. 24 is a diagram showing the result of obtaining the XY coordinate position information of each micro-reaction vessel for 20 chips. As shown in FIG. 24, finally, the XY coordinate position information of a total of 8000 micro-reaction vessels provided on the resin sheet could be obtained.

[0164] [Fourth Embodiment] In the fourth embodiment, a second image processing technique for detecting the horizontal position (position on the XY coordinates) on the drive stage of the micro-reaction vessel required for alignment will be described. Specifically, the image processing technique for precisely correcting the horizontal position of the micro-reaction vessel immediately before filling with fine particles, which is performed in step S2-3 in the second embodiment, will be described.

[0165] FIG. 25 is a conceptual diagram for explaining a second image processing technique according to the fourth embodiment. The method of this embodiment assumes that the horizontal positions of a plurality of micro reaction vessels 101 have been acquired in advance using the technique described in the third embodiment. Using the first drive stage 106, among the planar substrate 102 provided with a plurality of micro reaction vessel groups, it moves to the XY coordinates of the horizontal center position of a specific numbered micro reaction vessel as the target, directly below the first imaging device 107, specifically, to the reference position (center position) of the captured image, and moves so that the horizontal center position of a specific numbered micro reaction vessel as the target is arranged.

[0166] Next, the first imaging device 107 performs imaging with an overhead view, sets a certain threshold value centered on the reference position, restricts the image processing range, and processes it into an image restricted to only the micro reaction vessels with specific numbers (the left side of FIG. 25). As an example, this narrowing process is performed in a circular region. From the image with the range narrowed down, using the edge detection method, the center position on the captured image of a specific numbered micro reaction vessel is detected (the center of FIG. 25). Next, the position of the detected micro reaction vessel on the captured image is calculated as the pixel deviation amount from the reference position (center position) of the captured image. Next, using the conversion coefficient for converting from the pixel deviation amount to the XY coordinate values on the first drive stage 106, it is converted into the XY coordinate information of each micro reaction vessel on the first drive stage 106 (the right side of FIG. 25).

[0167] As described above, the horizontal position of a specific numbered micro reaction vessel can be detected and corrected from the captured image. Specifically, the horizontal position of a specific numbered micro reaction vessel can be calculated by the following formula (6).

[0168]

Equation

[0169] Here, X wellNth_final 、Y wellNth_final are the X coordinate and Y coordinate on the first drive stage where the Nth micro reaction vessel that has been precisely corrected is arranged at the reference position of the captured image of the first imaging device 107. ΔxwellNth_diff_image , Δy wellNth_diff_image is the X coordinate and Y coordinate (relative coordinates from the reference position) within the image of the N-th micro reaction vessel on the captured image of the first imaging device 107 detected by the processing of this embodiment.

[0170] By performing the processing of this embodiment immediately before filling the micro reaction vessel 101 with the particulate solution, the alignment accuracy can be improved. The reasons are as follows. If the time for filling the particulate solution into a plurality of micro reaction vessels is long, the relative arrangement of each component in the apparatus may vary due to environmental factors such as temperature fluctuations. For example, using the technique described in the third embodiment, the horizontal positions of a plurality of micro reaction vessels acquired in advance may vary by several to several tens of μm due to temperature fluctuations. Therefore, by performing the processing of this embodiment immediately before filling the particulate solution into the micro reaction vessel 101, the horizontal position of the micro reaction vessel can be precisely corrected immediately before, and the alignment accuracy can be improved.

[0171] As an example of the edge detection method in the processing of this embodiment, there is a circular edge detection method. Specifically, perform binary conversion processing on the image, emphasize the circular edge of the upper opening of the micro reaction vessel, then perform a process of filling the inside of the circle, and finally perform the circular edge detection method. This method is effective for a type of micro reaction vessel having a structure with one through-hole smaller than the diameter of the upper opening of the micro reaction vessel. If the above processing is not performed, edge detection may be misdetected due to the presence of the through-hole, but if the above processing is performed, it is possible to remove the influence of the through-hole and increase the success rate of edge detection.

[0172] Also, when acquiring an image with the imaging device in the processing of this embodiment, the imaging conditions such as the analog gain value, digital gain value, or exposure time of the imaging device corresponding to the horizontal position information of the micro reaction vessel acquired in advance by the technique of the third embodiment can be used to adjust the imaging conditions simultaneously. This leads to an increase in the probability of obtaining good contrast even when the image contrast of the imaging target is non-uniform, and can increase the success rate of edge detection.

[0173] <Technical effects> As described above, in the fourth embodiment, it is possible to precisely correct the horizontal position (position on the XY coordinates) on the drive stage of a specific micro reaction tank with a certain specific number, which is necessary during alignment.

[0174] <Verification of the effects of the fourth embodiment> Hereinafter, the effects of the fourth embodiment will be verified. The same apparatus configuration and measurement object as in the second and third embodiments were used. Using the horizontal position information of the micro reaction tank previously obtained by the technology of the third embodiment, the drive stage was moved to the XY coordinates of the horizontal position of a specific micro reaction tank and moved directly below the monochrome CCD camera. At this time, it was moved so that the horizontal center position of a specific micro reaction tank was arranged at the image reference position (center position) of the camera. With this arrangement, imaging was performed using the previously obtained imaging condition information, and an attempt was made to detect the center position on the captured image of a specific micro reaction tank in the captured image by the above-described circular edge detection method. At this time, the pixel deviation amount of each detected micro reaction tank from the reference position (center position) of the captured image was detected by image processing. The image processing calculates how many pixels the position of each micro reaction tank detected by the technology of the third embodiment is deviated from the center position of this image, and uses the conversion coefficient 1 pixel = 1.1625 μm for converting from the pixel deviation amount to the XY coordinate value on the first drive stage 106, and converts and corrects the XY coordinates on the first drive stage 106. In this way, correction processing was performed on a total of 8,000 micro reaction tanks using a planar substrate-shaped resin sheet provided with 20 chips of 400 micro reaction tanks per chip. This process was performed immediately before filling the fine particle solution.

[0175] FIG. 26 is a diagram showing the result of precisely correcting the XY coordinates of 400 micro reaction tanks in 20 chips using the second image processing technology. As shown in FIG. 26, it was confirmed that the deviation amounts in the X and Y coordinate directions of all the micro reaction tanks could be obtained and the horizontal positions of all were correctly corrected.

[0176] [Fifth embodiment] In the fifth embodiment, an example of the technique for forming the above-described nozzle shape trace will be described. More specifically, the tip of the nozzle 109 is brought into contact with a resin sheet 125 having mechanical plasticity, and a mechanical load is applied until the resin sheet is deformed, thereby forming a nozzle shape trace 111 identical to the tip of the nozzle 109 on the resin sheet. The formation of the nozzle shape trace is important in the correction step of the "horizontal direction difference distance from the reference position of the captured image to the horizontal direction center position of the tip of the nozzle" required during alignment.

[0177] FIG. 27A is a schematic cross-sectional view for explaining a method of forming a nozzle shape trace according to the fifth embodiment. First, a resin sheet 125 (a planar substrate for the shape trace) having mechanical plasticity is prepared. A mechanical load greater than a predetermined threshold is applied to the resin sheet 125. As a result, the location where the load is applied reaches the plastic deformation region beyond the elastic deformation region and consequently deforms. Utilizing this phenomenon, as shown in FIG. 27A, the tip of the nozzle 109 is pressed against the resin sheet 125 with a predetermined mechanical load. Thereby, a nozzle shape trace 111 having the same shape as the shape of the tip of the nozzle 109 can be formed on the surface of the resin sheet 125.

[0178] FIG. 27B is a top view of the nozzle shape trace 111. FIG. 27C is an actual measurement view of the nozzle shape trace 111. As shown in FIGS. 27B and 27C, generally, the tip of the nozzle 109 has a hollow circular shape, and by using this method, a nozzle shape trace 111 having a double circle of the outer circle diameter and the inner circle diameter of the tip of the nozzle 109 can be formed with good reproducibility.

[0179] FIG. 27D is a conceptual diagram for explaining image processing using the nozzle shape trace 111. As shown in FIG. 27D, both circles of the double circle of the nozzle shape trace 111 are detected, and using the position information of the centers of both circles, the position coordinates of the tip of the nozzle 109 on the captured image are accurately detected. Thereby, the "horizontal direction difference distance from the reference position of the captured image to the horizontal direction center position of the tip of the nozzle 109" can be accurately corrected. The specific correction process can be implemented by the same process as described in step (iv) of the first embodiment. When the nozzle shape trace 111 is formed, the inside of the nozzle may be in an empty state, or may be filled with pure water or a buffer.

[0180] At this time, as an imaging device for taking an overhead image from above for observing the nozzle shape trace 111, it can be configured to include a coaxial epi-illumination system and an optical system that acquires only the direct light reflected from the object. By adopting such a configuration, even if the resin sheet of the object is a transparent body, the shape trace can be clearly captured with high contrast. Also, by using such an optical system, the shape trace formed on the resin sheet can be adopted even if it is not necessarily a physically recessed shape. In the formed shape trace portion, the light reflection characteristics change due to being pressed against the tip of the nozzle 109, and the light irradiated from the above optical system cannot be reflected straight to the imaging device. Then, in this region, since the direct reflected light attenuates, it becomes a dark portion compared to the surroundings, and can be recognized as a shape trace of a dark portion on the captured image (FIG. 27C). Actually, as a result of measuring the height distribution of the shape trace formed on the surface of the transparent resin sheet using a white interference microscope, it is confirmed that the shape trace is deformed only to such an extent that it cannot be significantly recognized as a physical dent, and is in a state where it can be recognized as a contrast change in the coaxial epi-illumination system.

[0181] The resin sheet 125 may be made of the same material as the micro reaction tank, or may be separately provided as a material different from the micro reaction tank. In the case of the same material as the micro reaction tank, a nozzle shape trace can be formed at a position away from the region where the micro reaction tank is provided. The advantage of this method lies in that the shape reproducibility of the nozzle shape trace is good and the correction accuracy can be guaranteed to be high. In addition, there is no need to separately prepare components for correcting the position coordinates of the tip of the nozzle 109, and the components used in the fine particle filling method can be directly reused, making it possible to correct the position coordinates of the tip of the nozzle 109 simply, at low cost, and in a short time. Also, due to the above characteristics, it can be easily incorporated into a series of continuous loop sequences in the fine particle filling method.

[0182] Since this method is a method in which a mechanical load is applied to the tip of the nozzle 109, by detecting the mechanical load applied to the tip of the nozzle 109 with an appropriate threshold value and controlling, damage to the tip of the nozzle 109 can be prevented. For example, the strain signal obtained from the sensor system 116 using a leaf spring and a strain sensor can be converted into a load value applied to the tip of the nozzle 109. From this, by setting an appropriate threshold value for the signal on the sensor system 116, it is possible to realize a mechanical load with high reproducibility without damaging the tip of the nozzle 109 and form a shape trace with high reproducibility.

[0183] Various materials can be used as the material of the resin sheet 125. For example, various resin materials (such as polycarbonate-based, polycycloolefin-based, polyolefin-based, polypropylene-based, polyethylene-based, acrylic-based, acrylonitrile-based, polystyrene-based, polydimethylsiloxane-based resins, etc.) can be mentioned.

[0184] <Technical Effect> As described above, in the fifth embodiment, by using the resin sheet 125 having mechanical plasticity and bringing the tip of the nozzle 109 into contact with the resin sheet 125 and applying a mechanical load, a nozzle shape trace 111 identical to the tip of the nozzle 109 is formed on the resin sheet 125. Using this nozzle shape trace 111, it is possible to correct the "horizontal difference distance from the reference position of the captured image to the horizontal center position of the tip of the nozzle" required for alignment.

[0185] <Verification of the Effects of the Fifth Embodiment> Hereinafter, the effects of the fifth embodiment are verified. The same apparatus configuration as in the second to fourth embodiments was used. Specifically, during the correction step after cleaning the tip of the nozzle 109 in step S2-0 in the fine particle filling method shown in FIG. 17, the formation of the nozzle shape trace was performed a plurality of times. A nozzle shape trace was formed by pressing the tip of the nozzle (outer circle diameter 66 μm, inner circle diameter 42 μm) against a resin sheet made of the same material as the micro reaction tank and having mechanical plasticity of a transparent body with a mechanical load of 3.2 mN for 3 seconds.

[0186] FIG. 28A is a captured image showing an example of a plurality of formed nozzle shape traces (formed at 100 μm pitch intervals). For the nozzle shape trace captured in this way, the pixel deviation amount from the reference position (center position) of the captured image was detected by image processing. The image processing detected both the outer circle and the inner circle by the circular edge detection method, and obtained the average value of the center positions of each circle on the image. It was calculated how many pixels the average value of the center positions of each circle on the image was deviated from the center position of this image, and using the conversion coefficient (1 pixel = 1.1625 μm) for converting it into the XY coordinate values on the first drive stage 106, it was converted into the XY coordinate deviation amount on the first drive stage. In this way, the "horizontal difference distance from the reference position of the captured image to the horizontal center position of the tip of the nozzle" was calculated using Equation (3) and corrected.

[0187] FIG. 28B is a graph showing the result of the correction amount for the "horizontal differential distance from the reference position of the photographed image to the horizontal center position of the tip of the nozzle". In this way, by periodically correcting the "horizontal differential distance from the reference position of the photographed image to the horizontal center position of the tip of the nozzle" by a deviation amount of about several to 15 μm even during continuous operation of the fine particle filling, it was confirmed that accurate alignment can be achieved.

[0188] [Sixth Embodiment] In the second embodiment, an example of a simple cylindrical shape was described as the shape of the micro reaction tank 101. In the sixth embodiment, a technique will be described in which by changing the shape of the micro reaction tank, it is possible to relax the allowable alignment accuracy in aligning the tip of the nozzle 109 with the micro reaction tank.

[0189] FIG. 29A is a schematic cross-sectional view of a planar substrate 102 according to the sixth embodiment. As shown in FIG. 29A, the shape of the micro reaction tank 101 may be a shape in which the size of the upper opening 103a is larger than the size of the bottom 103b. For example, it may be an inverted frustum shape in which the diameter of the upper opening 103a is larger than the diameter of the bottom 103b. By using such an inverted frustum shape, the diameter of the upper opening 103a can be made larger than in the case of a simple cylindrical shape. In particular, when the volume of the micro reaction tank 101 is kept constant, the inverted frustum shape can make the diameter of the upper opening 103a larger than that of a simple cylindrical shape.

[0190] FIG. 29B is a schematic cross-sectional view showing the state immediately before lowering the tip of the nozzle 109 into the micro reaction tank 101. As shown in FIG. 29B, when filling the same volume of fine particles, using an inverted frustum-shaped micro reaction tank leads to an increase in the success probability of being able to lower the tip of the nozzle 109 into the micro reaction tank 101 compared to using a cylindrical micro reaction tank.

[0191] At this time, as a combination, the tip of the nozzle 109 can be left with a degree of freedom so as to be slightly movable in the horizontal direction. Since the tip of the nozzle 109 has movability, even when the tip of the nozzle 109 descends slightly eccentrically from the center position of the upper opening 103a of the micro reaction tank 101 having an inverted frustum shape into the interior, the tip of the nozzle 109 can reach the bottom 103b while contacting the sloped side wall surface 103c. At this time, in order to prevent the tip of the nozzle 109 from stopping at the side wall surface 103c during the descent and not reaching the bottom 103b, the diameter of the tip of the nozzle 109 can be set to a size equal to or smaller than the circular diameter of the bottom 103b.

[0192] The cross-sectional shape of such a structure in which the size of the upper opening 103a is larger than the size of the bottom is not limited to the inverted frustum shape, and may be other elliptical shapes or polygonal shapes. The important point is that the size of the upper opening 103a expands and the likelihood of the tip of the nozzle 109 descending into the interior of the micro reaction tank is increased as compared with the case where the side wall of the micro reaction tank is vertical, and any other shape may be used as long as such dimensional relationships can be realized.

[0193] As described above, the process in which the tip of the nozzle 109 contacts the bottom 103b of the micro reaction tank 101 and when the tip of the nozzle 109 is pulled up, the high-concentration fine particle solution 113 contacts the side wall surface 103c of the micro reaction tank 101 due to its wettability and the liquid advances is an important driving force. Therefore, if the tip of the nozzle 109 were temporarily linear, for example, when the shape of the micro reaction tank 101 is an inverted frustum shape, the gap between the tip of the nozzle 109 and the side wall surface 103c of the micro reaction tank becomes large near the opening of the micro reaction tank 101. As a result, the Laplace pressure when the high-concentration fine particle solution 113 advances decreases, and there is a risk of insufficient filling. Therefore, as an important point, when the shape of the micro reaction tank 101 is changed, it is desirable to change the shape of the tip of the nozzle 109 to a similar shape.

[0194] FIG. 29C is a schematic cross-sectional view showing an example in which the shape of the tip of the nozzle 109 is changed. As shown in FIG. 29C, when the micro reaction tank has an inverted frustum shape, the tip of the nozzle 109 can also have an inverted frustum shape at the same ratio.

[0195] <Technical effect> As described above, in the sixth embodiment, by changing the shape of the micro reaction tank, the allowable alignment accuracy in aligning the tip of the nozzle 109 with the micro reaction tank can be relaxed.

[0196] <Verification of the effect of the sixth embodiment> Hereinafter, the effect of the sixth embodiment is verified. The same apparatus configuration as in the second to fifth embodiments is used. Also, the technique for improving the alignment accuracy described in the second to fifth embodiments is used.

[0197] FIG. 30A is a photographed image of a chip having a micro reaction tank in the shape of an inverted frustum. The chip shown in FIG. 30A is provided with 20×20 (a total of 400) micro reaction tanks 101. Each micro reaction tank has an inverted frustum shape (taper angle 8°) with a circular diameter of 83 μm at the upper opening 103a, a circular diameter of 66.7 μm at the bottom 103b, and a depth of 65 μm.

[0198] FIG. 30B is a photographed image of the side surface of a nozzle with a tip shape of an inverted truncated cone. A glass capillary with a tip shape of an inverted truncated cone (taper angle 8°), an outer circle diameter of 64 μm, and an inner circle diameter of 42 μm at the tip of the nozzle 109 was used. FIG. 30B shows the state when attempting to fill fine particles. In the case of the tip of such a nozzle 109 and the micro reaction vessel 101, the difference between the diameter of the upper opening 103a and the outer circle diameter of the tip of the nozzle 109 is 19 μm, and the misalignment amount of the alignment accuracy can be allowed by this difference. On the other hand, when using the cylindrical micro reaction vessel 123 with a circular diameter of 75 μm and a depth of 70 μm shown in FIGS. 16A and 16B, and using a glass capillary with a tip shape of a cylinder, an outer circle diameter of 71 μm, and an inner circle diameter of 54 μm at the tip as the nozzle, the difference between the diameter of the upper opening 103a and the outer circle diameter of the tip of the nozzle 109 is 4 μm, and the allowable degree of misalignment amount of the alignment accuracy is small. Therefore, by adopting a structure in which the size of the upper opening 103a is larger than the size of the bottom 103b as the micro reaction vessel 101, the allowable accuracy during alignment can be relaxed.

[0199] [Seventh Embodiment] In the seventh embodiment, a technique capable of increasing the probability that the high-concentration fine particle solution 113 is filled into the micro reaction vessel 101 by adjusting the difference between the size of the micro reaction vessel 101 and the size of the tip of the nozzle 109 will be described.

[0200] In the micro reaction vessel 101, as described in the sixth embodiment, it was shown that the inverted truncated cone shape in which the upper opening diameter is larger than the bottom diameter is suitable in the alignment method. In this embodiment, in the case of an example where the tip shape of the nozzle 109 is a hollow inverted truncated cone shape and the micro reaction vessel 101 is an inverted truncated cone shape, the difference between the size of the bottom diameter of the micro reaction vessel 101 and the size of the tip of the nozzle 109 affects the probability that the high-concentration fine particle solution 113 is filled into the micro reaction vessel 101, and it is shown that the yield can be increased by adjusting this difference to 1 μm or less.

[0201] As described above, when the tip of the nozzle 109 contacts the bottom 103b of the micro-reaction tank 101 and the tip of the nozzle 109 is pulled up, the high-concentration fine particle solution 113 contacts the side wall surface 103c of the micro-reaction tank 101 due to its wettability, which is an important driving force for the process of the high-concentration fine particle solution 113 advancing. Therefore, when the bottom diameter of the micro-reaction tank 101 having an inverted frustum shape and the outer diameter of the circle at the tip of the nozzle 109 are as close as possible in size, the probability of the high-concentration fine particle solution 113 contacting the side wall surface 103c of the micro-reaction tank 101 increases, and the filling success rate increases.

[0202] When adjusting the size of the micro-reaction tank 101 and the size of the tip of the nozzle 109, the size of the smallest part at the tip of the nozzle 109 (for example, the outer diameter of the circle if it has the above-mentioned hollow inverted frustum shape) can be made the same or smaller than the size of the smallest part in the micro-reaction tank 101 (for example, the bottom diameter if it has the above-mentioned inverted frustum shape). However, when the material of the micro-reaction tank 101 has elastic deformation, even if the size of the smallest part at the tip of the nozzle 109 is larger than the size of the smallest part in the micro-reaction tank 101, it is possible to contact the bottom 103b of the micro-reaction tank 101 while deforming by applying an appropriate load.

[0203] <Technical Effect> As described above, in the seventh embodiment, by adjusting the difference between the size of the micro-reaction tank 101 and the size of the tip of the nozzle 109, the yield can be further increased.

[0204] <Verification of the Effect of the Seventh Embodiment> Next, the effects of the seventh embodiment are verified. The same apparatus configuration as in the second to sixth embodiments is used. Also, the technique for improving the alignment accuracy described in the second to sixth embodiments is employed. A resin sheet in the form of a planar substrate provided with eight chips shown in FIG. 30A is used. Each chip is provided with 400 micro reaction vessels. Each micro reaction vessel has an inverted frustum shape (taper angle 8°) with a circular opening diameter of 83 μm at the upper part 103a, a circular diameter of 66.7 μm at the bottom 103b, and a depth of 65 μm. As the nozzle, a glass capillary having an inverted frustum shape (taper angle 8°) at the tip is used. The filling success rate was determined when the difference between the circular diameter of the bottom 103b of the micro reaction vessel 101 and the circular outer diameter of the tip of the nozzle 109 was varied by varying the circular outer diameter of the tip of the nozzle 109 from 55 μm to 66.7 μm.

[0205] FIG. 31A is a table showing the results of the filling success rate when the difference between the circular diameter of the bottom 103b of the micro reaction vessel 101 and the circular outer diameter of the tip of the nozzle 109 is varied. The resin sheet is provided with eight chips each having 20×20 (total 400) micro reaction vessels 101. A plurality of filling trials were conducted for this micro reaction vessel 101. As a result, as shown in FIG. 31A, the filling success rate was greatly improved by reducing the difference. In particular, it was confirmed that the filling success rate can be increased to 99.2% by reducing the difference to 1 μm or less. On the other hand, it was confirmed that the number of filling failures increases when the difference between the circular diameter of the bottom 103b of the micro reaction vessel 101 and the circular outer diameter of the tip of the nozzle 109 is large.

[0206] FIG. 31B is a captured image of a chip in which the microparticle filling has failed. As shown in FIG. 31B, a plurality of micro reaction vessels in which the microparticles are not filled throughout the micro reaction vessels are formed.

[0207] Therefore, by reducing the difference between the circular diameter of the bottom 103b of the micro-reaction tank and the outer circular diameter of the tip of the nozzle 109, the filling success rate can be increased. When the micro-reaction tank has a simple cylindrical shape, since the circular diameter of the bottom 103b is equal to the diameter of the upper opening 103a, reducing the above difference will increase the required accuracy during alignment, and the filling success and alignment success will be in a trade-off relationship. Even in such a case, if the micro-reaction tank has an inverted frustum shape as in the sixth embodiment, it is possible to both reduce the difference between the circular diameter of the bottom 103b of the micro-reaction tank and the outer circular diameter of the tip of the nozzle 109 to increase the filling success rate and increase the diameter of the upper opening 103a to increase the likelihood of alignment and the alignment success rate.

[0208] [Eighth Embodiment] In the eighth embodiment, a technique will be described in which by increasing the pressure inside the tip of the nozzle 109, it is possible to increase the probability of filling the high-concentration fine particle solution 113 into the micro-reaction tank 101.

[0209] As described above, when the tip of the nozzle 109 contacts the bottom 103b of the micro-reaction tank and the tip of the nozzle 109 is pulled up, the high-concentration fine particle solution 113 contacts the side wall surface 103c of the micro-reaction tank 101 due to its wettability, which is an important driving force for the process of the high-concentration fine particle solution 113 advancing. Therefore, the Laplace pressure due to the surface tension determined among the three members of the micro-reaction tank 101, the nozzle 109, and the high-concentration fine particle solution 113 is the driving force for the high-concentration fine particle solution 113 to advance into the micro-reaction tank 101. Therefore, by further pressurizing the inside of the tip of the nozzle 109 with respect to this Laplace pressure, the above driving force can be increased.

[0210] Specifically, in the fine particle filling system 100 described in the second embodiment, with the electromagnetic valve closed and the entire system from the tip of the nozzle 109 to the syringe pump being a closed space, the internal pressure of the entire system is increased using the syringe pump. Thereby, the high-concentration fine particle solution 113 inside the tip of the nozzle 109 can be pressurized.

[0211] FIG. 32A is a schematic cross-sectional view showing a state in which the high-concentration fine particle solution 113 inside the nozzle 109 is pressurized. As shown in FIG. 32A, due to the pressurization, the high-concentration fine particle solution 113 is slightly extruded from the tip of the nozzle 109, wetting the tip of the nozzle 109. This also leads to assisting the process in which the high-concentration fine particle solution 113 advances from the tip of the nozzle 109 to the micro reaction tank 101. In this way, since the wetting inside the micro reaction tank 101 can be assisted by the Laplace pressure + auxiliary pressure, the probability that the solution wets and contacts the side wall of the micro reaction tank 101 can be improved. Note that at the same time, this pressurization needs to be controlled to a value that does not exceed the Laplace pressure at the tip of the nozzle 109 so that the high-concentration fine particle solution 113 stays at the tip of the nozzle 109. If the Laplace pressure is exceeded, the high-concentration fine particle solution 113 will be discharged as large droplets from the tip of the nozzle 109, making it difficult to shift to an appropriate filling state.

[0212] Appropriate pressure control can be implemented immediately before the tip of the nozzle 109 descends into the micro reaction tank 101. Specifically, it can be implemented between step (c) and step (d) described in the first embodiment. By adding pressure control during such steps, the driving force for the high-concentration fine particle solution 113 to advance into the micro reaction tank 101 can be assisted at an appropriate timing.

[0213] <Technical effect> As described above, in the eighth embodiment, by increasing the pressure inside the tip of the nozzle 109, the probability that the micro reaction tank 101 is filled with the high-concentration fine particle solution 113 can be increased.

[0214] <Verification of the effect of the eighth embodiment> The effects of the eighth embodiment are verified below. The same apparatus configuration as in the second to seventh embodiments is used. Also, the technique for improving the alignment accuracy described in the second to sixth embodiments is used. Furthermore, the same resin sheet and glass capillary as in the seventh embodiment are used. The pressure inside the tip of the nozzle 109 is changed using a syringe pump and pressurized by Δ0.5 kPa from atmospheric pressure. Under these conditions, the filling success rate when the pressure inside the tip of the nozzle 109 is increased is determined.

[0215] Figure 32B is a table showing the results of the filling success rate when the pressure inside the tip of the nozzle 109 is increased. As shown in Figure 32B, it was confirmed that by pressurizing the tip of the nozzle 109, the filling success rate increases from 87% (without pressurization) to 98% (with pressurization).

[0216] [Ninth Embodiment] In the seventh embodiment, a technique was described in which by reducing the difference (gap) between the bottom diameter of the microreaction vessel 101 and the outer circular diameter of the tip of the nozzle 109, the success rate of filling the high-concentration fine particle solution 113 into the microreaction vessel 101 is increased. In this embodiment, a technique for increasing the probability of filling the high-concentration fine particle solution 113 into the microreaction vessel 101 by reducing this gap is explained.

[0217] An elastic deformable resin material is used as the material of the microreaction vessel 101. Then, during fine particle filling, the tip of the nozzle 109 is brought into contact with the inside of the microreaction vessel 101, a load is applied to deform the microreaction vessel, and the gap between the side wall surface 103c of the microreaction vessel 101 and the tip of the nozzle 109 is reduced.

[0218] FIG. 33A is a schematic cross-sectional view showing how the micro reaction vessel 101 according to the ninth embodiment elastically deforms. As shown in FIG. 33A, in order for the micro reaction vessel 101 to deform in the shrinking direction, a mechanical load is applied to the bottom 103b of the micro reaction vessel 101 using the tip of the nozzle 109, and the micro reaction vessel 101 is deformed. In this example, as in the sixth embodiment, an example using a micro reaction vessel in the shape of an inverted frustum and a nozzle 109 in the shape of an inverted frustum is illustrated. In this case, there is a space where no member is installed directly below the micro reaction vessel, for example, a space of about 100 μm, and this deformed state can be realized. When such a minute gap is provided directly below the micro reaction vessel, it is also possible to be compatible with the method of installing the magnet 121 directly below the micro reaction vessel as described in the second embodiment and accumulating the high-concentration fine particle solution 113 at the tip of the nozzle 109 in a short time.

[0219] <Technical effect> As described above, in the ninth embodiment, a resin material that can be elastically deformed is used as the material of the micro reaction vessel 101. Then, the tip of the nozzle 109 is brought into contact with and loaded into the micro reaction vessel 101 to deform the micro reaction vessel, and the gap between the micro reaction vessel and the tip of the nozzle 109 is reduced. Thereby, the probability that the high-concentration fine particle solution 113 is filled into the micro reaction vessel 101 can be increased.

[0220] <Verification of the effect of the ninth embodiment> Hereinafter, the effect of the ninth embodiment is verified. The same apparatus configuration as in the second to seventh embodiments is used. In addition, the technique for improving the alignment accuracy described in the second to sixth embodiments is used. Furthermore, the same resin sheet and glass capillary as in the seventh embodiment are used. This resin sheet has elastic deformability. A 100-μm gap is provided directly below the micro reaction vessel, and the shape of the micro reaction vessel 101 elastically deforms by pushing the tip of the nozzle 109 into the bottom 103b of the micro reaction vessel. Under these conditions, the filling of fine particles was attempted.

[0221] FIG. 33B is a table showing the results of the filling success rate of the fine particles when the micro reactor 101 is elastically deformed. As shown in FIG. 33B, it was confirmed that by reducing the gap between the tip of the micro reactor 101 and the nozzle 109 due to elastic deformation, the filling success rate increased from 87% (without elastic deformation) to 100% (with elastic deformation).

[0222] [Tenth Embodiment] Even when the above-described first or second embodiment and the third to ninth embodiments are utilized, there are micro reactors in which fine particles become "unfilled" with a low probability of 1 to 2% or less. In the tenth embodiment, a technique for automatically determining and detecting the filling state of fine particles in the micro reactors and attempting automatic refilling for the detected "unfilled" micro reactors will be described.

[0223] FIGS. 34A to 34C are schematic diagrams for explaining the fine particle filling method according to the tenth embodiment. FIGS. 34A to 34C show steps additionally implemented after the series of flows described in the first or second embodiment. In the method of this embodiment, it includes a step of image processing an image obtained by photographing a plurality of micro reactors using the first drive stage 106 and the first imaging device 107.

[0224] The processing of this embodiment is started after attempting to fill fine particles in all the micro reactors while referring to a list regarding a predetermined number of micro reactors 101 to be filled that are paired with a predetermined number of fine particle solutions 110 registered in advance. When the processing of this embodiment is subdivided, it is composed of the following steps (I) to (III). (I) Move the first drive stage 106 to place a plurality of micro reactors 101 after filling attempts directly below the first imaging device 107 and photograph them. (II) Determine and detect "unfilled" micro reactors from the photographed image by image processing. (III) For the detected list of a plurality of "unfilled" micro reactors, attempt to refill with the same predetermined number of fine particle solutions again.

[0225] In step (I), first, all the micro reaction vessels after the filling trial to be targeted are photographed in an overhead view. At this time, using the first drive stage 106, a plurality of predetermined micro reaction vessel groups are moved to positions where they can be photographed by the first imaging device 107 in the XY coordinates on the first drive stage. At this time, if a plurality of micro reaction vessels do not fit within one field of view of the first imaging device 107, they are divided into a plurality of fields of view and information synthesis is performed in a subsequent process for processing. This process is essentially similar to the technology described in the third embodiment and can be implemented by the same process. In particular, since the XY coordinates on the first drive stage of a plurality of micro reaction vessels acquired in advance can be referred to, the coordinate positions at which all the micro reaction vessels can be photographed in an overhead view can be calculated from this coordinate information.

[0226] Step (II) of determining "unfilled" micro reaction vessels from the photographed image by image processing is implemented as follows. First, since the XY coordinate information on the first drive stage 106 of the micro reaction vessel 101 "to be filled" is held in advance by the technology described in the third embodiment, this coordinate information is used as reference information.

[0227] FIG. 34A is image information registered in advance as a "filled" state. FIG. 34B is a diagram for explaining image processing for detecting micro reaction vessels in a "filled" state from the photographed image. Next, from the photographed image, using the image information registered as a "filled" state as shown in FIG. 34A, a group of micro reaction vessels similar to the image information in a "filled" state as shown in FIG. 34B is detected by the pattern matching method. As the pattern matching method, the same method as in the third embodiment can be adopted. At this time, the micro reaction vessels detected by pattern matching and determined to be in a "filled" state can be acquired as the XY coordinate information on the first drive stage 106, similar to the third embodiment.

[0228] FIG. 34C is a diagram for explaining a method of determining the filling state of a micro reaction vessel "to be filled". As shown in FIG. 34C, the XY coordinate position information of the micro reaction vessel "(c-1) to be filled" is compared and collated with the XY coordinate position information of the micro reaction vessel detected as the "(c-2) filled" state. When the position information of (c-2) is included within a predetermined threshold value from the position information of (c-1), the micro reaction vessel "(c-1) to be filled" can be determined as "filled". On the other hand, when the position information of (c-2) is not included within a predetermined threshold value from the position information of (c-1), the micro reaction vessel "(c-1) to be filled" can be determined as "not filled". In this way, by using so-called negative detection using the pattern matching method, micro reaction vessels in the "not filled" state can be detected.

[0229] FIG. 35 is a diagram showing a workflow for detecting unfilled micro reaction vessels according to the tenth embodiment. The image processing technology of this embodiment is executed by the control device 120. In step S201, the coordinate positions at which all the micro reaction vessels can be photographed in an overhead view are calculated from the XY coordinate information on the first drive stage of a plurality of micro reaction vessels acquired in advance. In step S202, the area for imaging the chip is divided into N×M, and it is moved to the imaging area for position detection by the pattern matching method (repeated N×M times).

[0230] In step S203, the set value (analog gain value, etc.) of the first imaging device is switched. In step S204, the pattern matching method is performed on the captured image. In step S205, the conversion is performed from the position on the image of the detected micro reaction vessel to the XY coordinates on the first drive stage. In step S206, it is determined whether the processing for all the set values has been completed. If YES in step S206, it returns to step S202. If NO in step S206, it returns to step S203.

[0231] In step S207, it is determined whether the detection of the positions of the micro reaction vessels for all the divided regions has been completed. If YES in step S207, the process proceeds to step S208. If NO in step S207, the process returns to step S202.

[0232] In step S208, all the XY coordinate information of the micro reaction vessels detected as being in the "filled" state is merged to degenerate the duplicate information into one point. In step S209, the position information of the XY coordinates of the micro reaction vessel with the number i for which (c-1) "filling should be carried out" is compared with the position information of the XY coordinates of the micro reaction vessels detected as being in the "filled" state. In step S210, it is determined whether the position information in (c-2) is included within a predetermined threshold from the position information in (c-1). If YES in step S210, the process proceeds to step S211. If NO in step S210, the process proceeds to step S212.

[0233] In step S211, a "filled" determination is made. That is, the micro reaction vessel with the number i for which "filling should be carried out" is determined to be in the "filled" state. In step S212, an "unfilled" determination is made. That is, the micro reaction vessel with the number i for which "filling should be carried out" is determined to be "unfilled".

[0234] The advantages of this method are described below. The first advantage is that by increasing the determination threshold in the pattern matching with the image of the micro-reaction vessel in the "filled" state, the risk of misjudging the micro-reaction vessel in the "filled" state as "unfilled" (so-called false negative detection) can be reduced. The second advantage is that the risk of detection omission of the micro-reaction vessel in the "unfilled" state can be reduced. In order to use so-called positive detection, which directly detects similar images of the micro-reaction vessel in the "unfilled" state by the pattern matching method, it is necessary to prepare a plurality of reference images. However, as shown in FIG. 2B, there are a plurality of unfilled states in the micro-reaction vessel in the "unfilled" state, and various image patterns exist. Therefore, in order to correspond to such various image patterns, it is necessary to prepare a large number of learning images such as machine learning and deep learning. Also, it is not always possible to cover all image patterns. On the other hand, by detecting the micro-reaction vessel in the "unfilled" state by negative detection as in this method, these problems can be avoided at low cost.

[0235] Refer to the list of a plurality of "unfilled" micro-reaction vessels detected by this method, and for the micro-reaction vessel with a predetermined number, try to refill the fine particle solution with the same predetermined number again. In this way, by automatically trying to refill the unfilled micro-reaction vessel again, the yield can be further increased.

[0236] At the time of automatic refilling, it is also possible to try refilling by resetting the device control variables using the log information that was unfilled during the previous trial. As an example, when a predetermined condition was not satisfied during the previous trial (for example, when the volume of the high-concentration fine particle solution 113 did not fall within the target threshold, that is, when the volume of the high-concentration fine particle solution 113 was too small or too large and the filling trial was skipped), the volume of the fine particle solution 110 to be sucked into the nozzle 109 was increased or decreased with reference to the conditions during the previous trial. It is possible to increase the success rate at the time of refilling by operating feedback control such as this.

[0237] <Technical Effect> As described above, in the tenth embodiment, the yield can be further increased by automatically detecting unfilled microreactors and attempting automatic refilling.

[0238] <Verification of the Effects of the Tenth Embodiment> Hereinafter, the effects of the tenth embodiment are verified. The same apparatus configuration as in the second to eighth embodiments was used. Also, the same resin sheet and glass capillary as in the seventh embodiment were used. In addition, the techniques for improving the alignment accuracy described in the first to sixth embodiments and the techniques for improving the filling success rate described in the seventh to eighth embodiments were used in combination. Under these conditions, filling trials were performed on the fine particles, and for the microreactors detected as "unfilled", the filling of the fine particles was retried.

[0239] Figure 36A is a table showing the results of the filling success rate of the fine particles when the fine particles were refilled in the tenth embodiment. As shown in Figure 36A, it was confirmed that the filling success rate increased from 99.6% to 99.94% before and after the addition of the automatic refilling function.

[0240] Figure 36B is an overhead view image of the resin sheet provided with the microreactors that were successfully filled in this embodiment.

[0241] [Eleventh Embodiment] In the eleventh embodiment, a technique for a continuous filling method that can shorten the time for filling the microreactors with the fine particle solution will be described. This method is one of the modified examples of the workflow described in the second embodiment.

[0242] In the first and second embodiments, as shown in FIG. 13, it is intended to fill the entire amount of the high-concentration fine particle solution 113 formed at the tip of the nozzle 109 into the micro reaction vessel 101. That is, the total volume of all the fine particles contained in the high-concentration fine particle solution 113 is adjusted to be equal to or less than the total volume of the micro reaction vessel 101. In such an operation process, as in the workflow shown in FIG. 18, every time the micro reaction vessel 101 is filled with the high-concentration fine particle solution 113, it is preferable to perform the operation of sucking the fine particle solution again. This workflow is suitable for the purpose of precisely controlling the volume of the high-concentration fine particle solution 113 filled into the micro reaction vessel 101.

[0243] On the other hand, a filling method by a different operation process is also possible by using the configuration of the present disclosure. As a modification of such a workflow, a continuous filling method can be used. The continuous filling method is a method in which, assuming L is a natural number, the fine particle solution 110 capable of being filled L times is previously sucked into the nozzle 109, and after concentrating the high-concentration fine particle solution 113 capable of being filled L times at the tip of the nozzle 109, it is continuously filled into L micro reaction vessels.

[0244] FIG. 37 is a schematic diagram for explaining a fine particle filling method of the continuous filling method according to the eleventh embodiment. The basic flow is similar to FIG. 13, but it is different in that the filling of fine particles into a plurality of micro reaction vessels is continuously performed until L fillings are carried out.

[0245] FIG. 38 is a diagram showing a workflow of the continuous filling method according to the eleventh embodiment. In the workflow, as shown in FIG. 38, steps S2-1, S2-2, S2-3, and S3 are performed only at the first filling or after L fillings in the fine particle solution. In other cases, continuous filling can be realized by skipping steps S2-1, S2-2, S2-3, and S3. The determination of "at the first filling or after L fillings" can be processed, for example, by counting the number of fillings in the control program and using this number of fillings for determination.

[0246] In the above continuous filling method, when pulling up at a minute speed of about 1 to 130 μm / s in step (e) of FIG. 37, the volume of the fine particle body filled into the minute reaction tank can be controlled by controlling the amount of pulling up. At the completion of filling in step (f) of FIG. 13 or step (f) of FIG. 37, the nozzle is pulled up from the minute reaction tank at a speed significantly higher than that in step (e) of FIG. 37, for example, at a speed of 1 to several tens of mm / s. As a result, it has shifted from the state where the nozzle 109 and the high-concentration fine particle solution 113 are liquid-connected to a separated state. Therefore, by appropriately controlling the amount of pulling up in step (e) of FIG. 37, the volume of the high-concentration fine particle solution 113 to be left inside the minute reaction tank 101 can be controlled. Note that this control effect is not limited to the continuous filling method, and it can be similarly implemented in the filling methods in the first embodiment and the workflow of FIG. 18.

[0247] Since the filling amount of the fine particles can be controlled as described above, even when "at the first filling or after L times of filling" is YES, the step of waiting until the fine particles are concentrated in step S3 is not essential and can be skipped. Since a large amount of the fine particle solution is sucked, when magnetic force accelerated sedimentation is carried out, typically, a sufficient amount of the high-concentration fine particle solution for filling L times is concentrated at the tip of the nozzle 109 in 1 to 2 seconds. Therefore, since a sufficient amount of the high-concentration fine particle solution is supplied to the tip of the nozzle 109 while step S4 is being carried out, step S3 can always be skipped.

[0248] In the present embodiment, when step S2-3 of FIG. 38 is carried out, not only the coordinate detection and correction of one minute reaction tank are performed, but also the coordinate detection and correction are simultaneously performed on a plurality of minute reaction tanks photographed in the same image by image recognition, whereby the coordinate correction process of step S2-3 can be skipped. This image recognition method can be realized by expanding and applying the third image processing technique or the fourth image processing technique to a plurality of minute reaction tanks.

[0249] The effect of this continuous filling method is that by skipping steps S2-1, S2-2, S2-3, and S3 in FIG. 38, it becomes possible to shorten the time for filling the micro reaction tank with the particulate solution. Particularly in the case of filling a plurality of micro reaction tanks with the same type of particulate solution, this method leads to a significant reduction in time.

[0250] In this embodiment, in order to accumulate a large amount of high-concentration particulate solution capable of being filled L times at the tip of the nozzle 109, the viscous resistance of the high-concentration particulate solution as a fluid at the tip of the nozzle 109 is greater than that in the case of a high-concentration particulate solution capable of only one-time filling. According to the wetting principle described in FIG. 12, the pressure required for the high-concentration particulate solution to spread inside the micro reaction tank 101 has increased. Therefore, by simultaneously using the seventh and eighth embodiments, the probability of filling the high-concentration particulate solution 113 inside the micro reaction tank 101 can be increased.

[0251] <Technical Effect> As described above, in the eleventh embodiment, by using the continuous filling method, the time for filling the micro reaction tank with the particulate solution can be shortened.

[0252] <Verification of the Effect of the Eleventh Embodiment> Hereinafter, the effect of the eleventh embodiment is verified. The same apparatus configuration as in the second to eighth embodiments was used. Also, the same resin sheet and glass capillary as in the seventh embodiment were used. As the continuous filling method, the workflow of FIG. 38 was adopted. The number of continuous fills was set to 4 times, and under the condition that step S3 was always skipped, a total of 12 filling trials were carried out.

[0253] FIG. 39A is a graph showing the filling time per micro reaction tank when using the workflow of FIG. 18 in the second embodiment. In FIG. 39A, the typical filling time per micro reaction tank was about 40 seconds.

[0254] FIG. 39B is a graph showing the filling time per minute reaction tank when the continuous filling method of the 11th embodiment is used. As shown in FIG. 39B, when the continuous filling method is used, it was confirmed that when the solution is not being sucked, the filling time per minute reaction tank can be typically shortened to about 6 seconds.

[0255] [Modification Example] The present disclosure is not limited to the above-described embodiments and includes various modification examples. For example, the above-described embodiments have been described in detail for easy understanding of the present disclosure, and it is not necessary to include all the configurations described. Also, a part of one embodiment can be replaced with the configuration of another embodiment. Also, the configuration of another embodiment can be added to the configuration of one embodiment. Also, a part of the configuration of each embodiment can be added, deleted, or replaced with a part of the configuration of another embodiment.

Description of Reference Numerals

[0256] 100 Particle filling system 101 Minute reaction tank 102 Planar substrate 103a Upper opening 103b Bottom 103c Side wall surface 104 Bottom surface of the planar substrate 105 Through hole 106 First drive stage 107 First imaging device 108 Second drive stage 109 Nozzle 109a Outer wall of the nozzle 110 Particle solution 111 Nozzle shape trace 112 Planar substrate for shape trace 113 High-concentration particle solution 114 Second imaging device 115 Drive mechanism 116 Sensor system 117 Dispensing mechanism 118 Storage container 119 Stirring mechanism 120 Control device 121 Magnet 122 Chip for single cell analysis 123 Micro reaction tank of chip for single cell analysis 124 Through hole for cell capture 125 Resin sheet

Claims

1. A method for filling a container with fine particles, comprising: aligning a nozzle and a container using an image captured by an imaging device; filling the container with fine particles from the nozzle, wherein the aligning includes: using (a) a horizontal difference distance from a reference position of an image captured by the imaging device to a horizontal center position of the tip of the nozzle and (b) coordinates at which a horizontal center position of an opening of the container is disposed at the reference position to align the tip of the nozzle with the horizontal center position of the opening of the container.

2. The method for filling fine particles according to claim 1, wherein a horizontal position of the imaging device and a horizontal position of the nozzle are fixed, and the container is configured to be movable in the horizontal direction.

3. The aligning includes: forming a shape trace of the nozzle on a substrate horizontal to the nozzle; capturing the shape trace of the nozzle by the imaging device; further including correcting (a) a horizontal difference distance from a reference position of an image captured by the imaging device to a horizontal center position of the tip of the nozzle by analyzing an image of the shape trace of the nozzle captured by the imaging device.

4. The aligning includes: analyzing an image of the container captured by the imaging device and obtaining (b) coordinates at which a horizontal center position of an opening of the container is disposed at the reference position by a pattern matching method.

5. The aligning includes: analyzing an image of the container captured by the imaging device and obtaining (b) coordinates at which a horizontal center position of an opening of the container is disposed at the reference position by an edge detection method.

6. The fine particles have magnetism, wherein the filling of the fine particles includes: disposing a magnet below the container, thereby accelerating sedimentation of the fine particles at the tip of the nozzle by magnetic force and shortening a formation time of a high-concentration suspension with a predetermined fine particle concentration.

7. The method for filling fine particles according to claim 1, wherein the container has an inverted frustum shape in which a diameter of the opening is larger than a diameter of the bottom.

8. The tip of the nozzle is smaller than the opening of the container, and the difference between the diameter of the bottom of the container and the diameter of the tip of the nozzle is 1 μm or less. The method for filling fine particles according to claim 7.

9. Filling the fine particles The method for filling fine particles according to claim 1, including pressurizing the inside of the nozzle immediately before lowering the tip of the nozzle into the container, thereby wetting the tip of the nozzle with a high-concentration fine particle solution.

10. The container has a property of elastic deformation, a space is provided below the container, and when the nozzle is lowered, the container elastically deforms, so that the gap between the nozzle and the container becomes smaller. The method for filling fine particles according to claim 1.

11. Detecting from the image captured by the imaging device whether the filling of the fine particles into the container is completed or not, And further including refilling the unfilled container with the fine particles. The method for filling fine particles according to claim 1.

12. A fine particle filling system for filling one or more containers with fine particles, comprising: A nozzle; A dispensing mechanism for controlling the suction of the suspension of the fine particles into the nozzle; A first drive mechanism for horizontally moving a stage that supports the container; A second drive mechanism for vertically moving the nozzle; A control device for controlling the dispensing mechanism, the first drive mechanism, and the second drive mechanism; An imaging device for photographing the container from above, and The control device A process of causing the dispensing mechanism to suck the suspension into the nozzle; A process of concentrating the fine particles in the suspension to form a high-concentration suspension with a predetermined fine particle concentration at the tip of the nozzle; Processing the image captured by the imaging device, and using (a) the horizontal difference distance from the reference position of the image captured by the imaging device to the horizontal center position of the tip of the nozzle, and (b) the coordinates at which the horizontal center position of the opening of the container is arranged at the reference position, driving the first drive mechanism to align the tip of the nozzle with the horizontal center position of the opening of the container; A fine particle filling system that drives the second drive mechanism to fill the container with the high-concentration suspension.

13. The horizontal position of the imaging device and the horizontal position of the nozzle are fixed, and the stage is fixed in the vertical direction. The fine particle filling system according to claim 12.

14. Further comprising a substrate horizontal to the nozzle, The control device executes a process of forming a shape trace identical to the tip shape of the nozzle with respect to the horizontal substrate, the fine particle filling system according to claim 12.

15. The horizontal substrate has mechanical plasticity, The control device drives the first drive mechanism and the second drive mechanism to execute a process of forming the shape trace of the nozzle by pushing the tip of the nozzle into the horizontal substrate, the fine particle filling system according to claim 14.

16. The fine particles have magnetism, The fine particle filling system further comprises a magnet disposed below the container, In the process of forming the high-concentration suspension, the precipitation of the fine particles at the tip of the nozzle is accelerated by the magnetic force of the magnet, and the formation time of the high-concentration suspension is shortened, the fine particle filling system according to claim 12.

17. The container has an inverted frustum shape in which the diameter of the opening is larger than the diameter of the bottom, the fine particle filling system according to claim 12.

18. The tip of the nozzle is smaller than the opening of the container, and the difference between the diameter of the bottom of the container and the diameter of the tip of the nozzle is 1 μm or less, the fine particle filling system according to claim 17.

19. The container has the property of elastic deformation, and a space is provided below the container, In the process of filling the high-concentration suspension into the container, when the nozzle descends, the container elastically deforms, so that the gap between the nozzle and the container becomes smaller, the fine particle filling system according to claim 12.

Images