Analysis device and analysis method

Abstract

The present invention addresses the problem of providing an analysis method capable of analyzing the state of a target object in detail in a short time without requiring complicated operations or devices. An analysis method for solving the abovementioned problem comprises: a step of mixing and causing an interaction between a target object and one or more types of luminescent probe having a luminescence behavior that changes in response to an interaction with the target object; a step of exciting the one or more types of luminescent probe in a state of interaction with the target object to generate two or more types of luminescence, and simultaneously detecting the two or more types of luminescence; and a step of analyzing the two or more types of detected luminescence to identify the state of the target object.

Description

Analysis device and analysis method

[0001] The present invention relates to an analysis device and an analysis method.

[0002] 1-1. Benefits of Applying Data-Driven Manufacturing Traditional "manufacturing" is generally accomplished by skilled professionals, known as experts or artisans, who use their experience and learning to build a foundation of things and accomplishments over many years, using this foundation to address challenges in a deductive, so-called "forward" manner. This is essentially a "personal" approach that relies on the sense of the leader, and while it may work by chance, in most cases it involves repeated trial and error. In addition to deduction and induction, there is also a method of conducting research and development and technological development called abduction (hypothesis testing), which is recommended in academia and industry as the most superior method. Students at universities learn this in laboratories before setting out into society.

[0003] On the other hand, data-driven approaches are completely different from the aforementioned deduction and abduction. They are based on inductive learning from data and direct deriving of solutions from the data. For this reason, they are also referred to as "inverse problem solving" or "backward." The most familiar example of successful inverse problem solving is e-commerce (electronic commerce) that utilizes big data, and its power and impact are truly astounding. Applications of the same concept to "manufacturing" are referred to as development DX and production DX. While these approaches are particularly hailed as groundbreaking methods in the industrial world, they are merely a topic of discussion in actual development and production settings. In reality, traditional methods are still applied, leading to derision such as the "wall of legacy."

[0004] Meanwhile, a NEDO project focused on chemical substances and chemical materials. Eighteen companies formed a research consortium, with the participation of the National Institute of Advanced Industrial Science and Technology (AIST), to demonstrate chemical manufacturing DX in Japan. This project, called the "Ultra-Advanced Materials Ultra-Rapid Development Platform Technology Project" (abbreviated "Ultra-Ultra Project"). Over a six-year period from 2016 to 2022, each of the 18 companies brought a different model material to the project, and the AIST also focused on one material. This project pursued an unprecedented and challenging goal: to reduce development time or prototype count to 1 / 20th of the conventional timeframe by combining measurement, calculation, and process, particularly by utilizing data-driven development. It was estimated that development could be completed in 1 / 23 of the conventional development time for all 19 themes, on average. This result is no longer coincidental, but rather demonstrates the necessity of this approach. The minutes of the project's post-evaluation have also been made public. As this large-scale demonstration experiment makes clear, the importance of data-driven manufacturing goes beyond the aforementioned e-commerce, and it is predicted that such methods will dominate the industrial world in the near future.

[0005] 1-2. Data Necessary for Data-Driven Manufacturing. The objects of industrial manufacturing include a wide variety of products, including buildings, houses, ships, automobiles, robots, printing equipment, video equipment, and cameras. Manufacturing also encompasses the components that make up these products, which are designed and processed to create products. In manufacturing where the final product is a machine, this manufacturing process has traditionally been done manually, meaning that the scale (size) to be controlled is typically in the centimeter to millimeter range. Furthermore, when manufacturing these products, the size of components and other components has been rapidly measured and digitized. This method is used on a large scale and is also known as factory automation. It is relatively easy to generate data in manufacturing sites and use this data for data-driven development and data-driven production, and it is already being applied worldwide.

[0006] Meanwhile, the targets of the aforementioned ultra-ultra project were chemical substances and chemical materials (compounds), including micron-sized fillers and fibers, high-molecular polymers ranging from submicron to several tens of nanometers in size, and even substances ranging from Å to several nanometers in size, such as low-molecular-weight functional substances and low-molecular-weight additives. When using these in combination depending on the application, or when using data-driven methods, unless precise measurements can be made of the parts that are the source of functionality, it is impossible to find a data-driven solution.

[0007] 1-3. Problems with Data-Driven Manufacturing in Complex Systems Taking chemical composite materials as an example, it is extremely difficult to distinguish whether the desired properties are due to micron-sized fillers, the entanglement of submicron-sized polymer chains, or minute bonds and interactions in the Å to nm range, such as hydrogen bonds or van der Waals bonds between low-molecular-weight compounds present as additives and other existing substances. For this reason, traditionally, each expert would carefully conduct scientific measurements in their area of ​​expertise, then deductively understand the phenomena, and combine these multiple understandings of phenomena to conduct forward-looking research and development.

[0008] Furthermore, in recent years, synthetic biology, which uses biologically-related substances such as microorganisms and enzymes (collectively referred to as "bio"), has become increasingly popular. Even more so than with the chemical materials mentioned above, it is extremely difficult to achieve the desired results and outcomes through deductive understanding of phenomena alone. When attempting to achieve digital transformation (DX) in manufacturing, including bio, the complexity of the systems often makes it difficult to determine which parts of the target object control its function when solving data-driven inverse problems. Therefore, the problem is that machine learning is performed on data sets that do not contain "correct answer data," which can be said to limit and inhibit its adoption in industry.

[0009] 1-4. Data acquisition methods other than measurement On the other hand, an analysis of the activities of the Ultra-Ultra Project based on its report materials shows that all 18 companies and 19 themes were able to achieve their goal of reducing development times to less than 1 / 20 of the conventional development period, which is thought to be due to the fact that they solved inverse problems based on multi-scale simulations (four companies only solved forward problems). The problems with simulation include the need for physical models, the fact that model selection is subjective, and the lack of versatility of the models, which requires the selection of adaptation laws according to the target phenomenon and the adjustment of simulation parameters. As a result, it takes time to select adaptation measures and adjust simulation parameters, making it difficult to popularize it.

[0010] 1-5. Advantages and Disadvantages of Real Measurement To overcome the above challenges, it is possible to measure objects in real time and at multiple scales. This approach can largely overcome the weaknesses of simulation. However, data-driven measurement requires the rapid generation of large amounts of data. Without this rapid generation, data acquisition becomes a rate-limiting factor, making it impossible to call it digital transformation. If the fundamental premise of DX is to use digital technology to bring about transformation, it would be counterproductive if data acquisition, the source of that transformation, becomes a rate-limiting factor. While we speak of "multi-scale measurement," as mentioned above, technologies already exist for rapid and reliable measurement at the centimeter and millimeter scales. These measurement results are utilized as data in various industries. Therefore, although μm-scale measurement takes some time, requires advanced equipment, and is costly, there are no major obstacles that prevent data-driven measurement.

[0011] As mentioned above, the success of "monozukuri DX" requires the rapid, large-scale, and easy acquisition of data on various sizes and scales of target objects. However, when target objects are chemical molecules, biological molecules, or even microbial metabolites, almost all of the key phenomena governing their functions are driven by intermolecular interactions such as hydrogen bonds and van der Waals bonds. The sizes that determine these interactions are on the nm or sub-nm (≒ angstrom) scale. For example, when the distance between the hydrogen and oxygen atoms forming a hydrogen bond is 0.18 nm compared to 0.20 nm, the former hydrogen bond is overwhelmingly stronger. Therefore, without sub-nm measurement capabilities, data-driven development is impossible for chemical and biological manufacturing. Furthermore, examples of intermolecular interactions and bonds, in order of strength, are covalent bonds (coordinate bonds), ionic bonds, metallic bonds, hydrogen bonds, and van der Waals bonds. When performing data-driven analyses involving these elements, measurement data on the nm to sub-nm scale is required. For example, when converting data into the processing of biomaterials (in this case, leather manufacturing), the above-mentioned chemical bonds play a major role in maintaining the structure of collagen proteins and in all processes, such as tanning, dyeing, fat-adding, and finishing. Therefore, the range of applications for measurements at this scale is extremely wide, from cutting-edge materials to familiar materials.

[0012] 1-6. Direct and indirect measurements One of the reasons why the above-mentioned nm-scale measurements are difficult is that they are in a range that is no longer visible even with optical microscopes or electron microscopes. It should be noted that even small molecules themselves can be observed with AFMs (atomic force microscopes). However, pre-processing is complicated, and the measurements themselves require careful and time-consuming work. Therefore, they are not suitable for data-driven applications due to the volume of data that can be generated.

[0013] 2-1. The Relationship Between Data-Driven R&D / Production and a Circular Economy: The super-smart society envisioned for 2030 will fundamentally rethink the fundamental industrial structure of "mass production / mass consumption" and assume a "providing only what is needed, to the people who need it, when they need it, and only what is needed." This means that manufacturing will also require "making only what is needed," and technological development will require speed and timeliness when it is needed. In a circular economy, not only will it be necessary to "avoid wasteful production," but it will also impose restrictions on greenhouse gas emissions, particularly carbon dioxide emissions. Therefore, it will be essential to convert conventional petroleum-based plastics and chemicals produced through petroleum cracking to natural and biological sources as much as possible. In other words, from the perspective of producing "only what is needed" and "without waste," the manufacturing process will utilize process informatics, a method for solving inverse problems. As for the timeliness of "when it is needed," the deductive approach of the past cannot be applied, so inverse problem solving methods such as materials informatics and bioinformatics will be applied. In order to build a recycling-oriented society, data-driven methods such as informatics will become an essential requirement.

[0014] 2-2. Challenges when data-driven approaches become mainstream Weaning chemical substances from dependence on petroleum is a symbolic initiative toward a circular society, but compared to the technical fields of physics, electricity, and machinery, chemistry and biochemistry lack logic and rely heavily on the experience and know-how of engineers and researchers. In other words, this is the biggest challenge when it comes to implementing data-driven approaches. As mentioned above, in order to adopt a data-driven approach, it is necessary to quickly, easily, accurately, and in large quantities to digitize the phenomena occurring in complex systems.

[0015] As mentioned above, in chemistry and biochemistry in particular, the mechanisms that govern function are rooted in intermolecular interactions, making it a major challenge to generate data at the Å to nm scale that meets these demands. When applying synthetic biology techniques using microorganisms and bio-related substances to the production of chemicals, submicron-scale fungi and bacteria, as well as the submillimeter-scale microorganisms (a collective term for protozoans and metazoans) that prey on them, are directly or indirectly involved in chemical synthesis (i.e., metabolism). Therefore, in order to generate data from these organisms as well, some kind of measurement must be performed and used as descriptors or explanatory variables. In other words, in these fields, data acquisition that is completely different from that required for mechanical manufacturing is essential for realizing a circular economy, which can be seen as a current hindrance.

[0016] 2-3. Blind Spots in Sustainability in Manufacturing In recent years, with a view to moving away from dependence on petroleum, synthetic biology techniques using biological reactions and microorganisms have been attracting attention as new manufacturing methods for a recycling-oriented society, but there is a major blind spot. Around 30 years ago, green chemistry, a chemical technology that performs synthetic reactions without using organic solvents, was in the spotlight, but it has now completely died down, and the number of papers published on the subject has plummeted. The reason for this is that even if the synthetic reaction is made solvent-free, the chemicals must still be extracted (called "extraction") and then highly purified (called "purification") to improve their quality, which requires large amounts of organic solvents, and so it was concluded that this method is ultimately almost ineffective.

[0017] The same can be said for synthetic biology. Despite much fanfare, it is impossible to produce useful organic compounds using living organisms alone. In fact, biochemical reactions often take place at ultra-dilute concentrations at liquid / liquid interfaces, which means that larger amounts of organic solvents are required than in conventional synthesis processes. If the organic solvents used in this process are incinerated, carbon dioxide is of course released into the atmosphere. Therefore, there will be an ever-increasing demand for the purification of organic substances using activated sludge, a method long used in the chemical industry.

[0018] Figure 1 shows a schematic diagram of a conventional activated sludge process, which consists of an aeration tank using activated sludge and a settling tank connected together. Biological treatment, one type of wastewater treatment method, uses microorganisms to remove contaminants from wastewater. A typical example is the "conventional activated sludge process" shown in Figure 1. The "conventional activated sludge process" involves cultivating microorganisms 51 (activated sludge) at high concentrations (approximately 4000 mg / L) in pools or tanks to feed on contaminants in the wastewater. Microorganisms 51 can be classified into those that utilize oxygen (aerobic microorganisms) and those that do not (anaerobic bacteria). The conventional activated sludge process uses aerobic microorganisms. Aerobic microorganisms grow by absorbing oxygen from the water and decomposing organic matter 50 in the wastewater. Therefore, equipment for constantly blowing air into the tank (aeration) is required. Furthermore, as the microorganisms 51 grow, suspended matter called flocs (activated sludge) is produced, and more types of microorganisms then remove contaminants from the wastewater. In the aeration tank, while the wastewater is purified, microorganisms continue to grow. This makes it necessary to separate the grown microorganisms from the purified water. Good microorganisms 51 (activated sludge) have the property of sinking to the bottom, taking with them the surrounding suspended solids, when left standing. Therefore, it is common to draw the water from the aeration tank into another tank (sedimentation tank) and leave it standing, and then release only the supernatant water into a river or the sea after sterilization. Also, it is common to return some of the settled microorganisms 51 (activated sludge) to the aeration tank to maintain the concentration of microorganisms 51 (activated sludge), and to dispose of the excess activated sludge as waste.

[0019] Among the microorganisms in activated sludge, bacteria contribute most to wastewater purification, gathering together to form activated sludge flocs (aggregations of bacteria). Representative examples of bacteria that form flocs include the Zoogloea, Bacillus, Pseudomonas, and Flavobacterium genera. On the other hand, protozoa are microorganisms in activated sludge whose shapes can be seen under a microscope at around 100x magnification; they eat the flocs to reduce their volume and make them easier to sink, and they also eat floating bacteria to purify the treated water. Examples of these include ciliates such as Vorticella, Epistylis, Aspidisca, and Litonotus, sarcozoans such as Arcella and Amoeba, and flagellates such as Entosiphon and Peranema. Furthermore, multicellular metazoans can also be observed at about 100x magnification, and representative examples include rotifers, caterpillars, earthworms, tardigrades, and nematodes.

[0020] The above wastewater treatment methods are almost identical all over the world, but because the purification mechanism is biologically derived as described above, the activity of waste organic matter treatment is prone to large fluctuations depending on factors such as temperature, oxygen concentration, the amount and type of waste organic matter that serves as food for fungi and microorganisms, and the distribution of microbial species in activated sludge. Seasonal fluctuations are also common. In response to these factors, most management is carried out based on the experience and knowledge of the staff and experts in charge, and as demand increases in the future, it will become extremely important to digitalize this purification process.

[0021] In a broader sense, activated sludge treatment itself can be considered a form of synthetic biology, in the sense that it uses fungi and microorganisms to convert waste organic matter into useful chemicals that can be released into rivers or oceans. This purification process DX can also be applied to process informatics based on synthetic biology for the production of almost any other advanced chemical product. This is an area where new data-driven methods are expected to be applied in both the "disposal" and "production" aspects.

[0022] 2-4. DX in Food Processing While it's unclear whether it can be considered a manufacturing industry, food processing also faces numerous challenges. For example, in response to the globalization of food production and distribution, the revised Food Sanitation Act, passed in June 2018, made the implementation of HACCP mandatory in Japan starting June 1, 2020. After a one-year grace period, full HACCP implementation will be required for all food-related businesses from June 2021. HACCP breaks down manufacturing processes into smaller steps and manages risk at each step, preventing the shipment of problematic products. In the unlikely event of a food accident, the cause can be quickly identified. Because this law requires non-large-scale manufacturers to monitor every process, managing all processes with advanced analytical equipment is cost-prohibitive, and the wide range of items to be monitored presents significant challenges.

[0023] Traditional methods have primarily focused on random sampling inspections from packaging to shipping. The HACCP system, on the other hand, is a hygiene management methodology that ensures product safety by predicting hazards, such as microbial contamination and the inclusion of foreign matter, at each stage from raw material receipt to processing and shipping, and by continuously and continuously monitoring and recording particularly critical processes that contribute to the prevention of hazards. This approach makes it possible to prevent the shipment of more problematic products than the traditional random sampling inspection of final products. However, the cost and effort (man-hours) involved in testing and analysis, as well as the need for large-scale modifications to manufacturing processes, have already become significant issues. Because HACCP is a system and regulation that has only just begun in Japan, it is not fully understood by those outside the industry as a major challenge. However, it is clear that this issue requires measures to be implemented from various angles, beyond the food industry.

[0024] 2-5. Challenges of data acquisition and data management in the primary industry Quality assurance is an important task in food processing and manufacturing, and various methods have been implemented to date. However, what must be considered problematic is that quality assurance evaluation items are limited to managing conventions and process conditions (for example, heating at 100°C for 2 minutes, or annealing at room temperature for 1 hour after molding), and there are many cases where essential analysis of quality is not carried out.

[0025] Needless to say, this type of quality assurance system was created based on the premise of "mass production / mass consumption," which has been the norm in the manufacturing industry until now. With on-demand manufacturing that meets the needs of a super-smart society, sampling inspections are no longer sufficient, and a system that can individually and easily maintain and guarantee quality is required.

[0026] When it comes to manufacturing "what is needed, to the people who need it, when they need it, in the amount they need," traceability during manufacturing will likely become more important than the final characteristics or quality of the product. Furthermore, the fact that strictly controlled raw materials are produced through a highly transparent manufacturing process is itself a quality indicator, which in turn leads to consumer trust. Therefore, it is important to easily record and digitize the traceability of the manufacturing process. As on-demand products become widely distributed throughout the world, preparing for product liability will become a major issue.

[0027] While traceability, as mentioned earlier, is one way of preparing for this, it will also be necessary to record the identity and condition of each substance—raw materials used, intermediates removed during processing, and final products—in other words, wide-area ID technology. Currently, many products record IDs using QR codes (registered trademark) or barcodes. However, if IDs are required even for raw materials, it will be essential to devise and popularize methods that allow primary industry workers to easily assign and acquire IDs. Furthermore, if all raw materials, intermediates, and final products are produced on demand, storing and managing these IDs, as well as linking data and tracing back to the upstream manufacturing process when problems arise, will become major challenges. Current human-centered management is insufficient to address these issues, and the effective use of artificial intelligence (AI) is a prerequisite.

[0028] Furthermore, as with the AFM (atomic force microscope) mentioned above, when data can no longer be obtained from something that can no longer be measured directly, indirect measurement is the only way. When technical constraints make direct measurement impossible, humans have a history of using indirect measurement media or chips as intermediates to convey information that humans can interpret. For example, indirect media have been used in fields such as X-ray film, toner, and photosensitive materials.

[0029] The following technologies have been proposed as methods for indirectly measuring and aggregating various physical properties and characteristics. For example, Patent Document 1 describes measuring various physical quantities, such as power consumption and air flow rate, and monitoring air leaks in systems powered by a bus duct system. Patent Document 2 describes a device for easily imaging various physical or chemical phenomena, such as the two-dimensional distribution of solution pH, pressure, magnetic field, or temperature. Patent Document 3 describes a measurement information transmission device and a multipoint measurement information collection system that efficiently collect physical quantities measured at multiple points. Patent Document 4 describes a method for measuring strain due to nanometer-scale displacement using light intensity and polarization direction.

[0030] JP 2004-152084 A JP 2005-337806 A JP 2004-173123 A JP 2016-156762 A

[0031] However, the devices and systems for indirect measurement described in the above patent documents have the problem that it is difficult to shorten the measurement time for minute samples. In particular, when measuring minute samples, high accuracy is required, making it difficult to shorten the measurement time and further difficult to obtain a large amount of data in a short period of time. In addition, there is a problem that general-purpose devices have limitations in terms of resolution, etc., and expensive dedicated measuring devices are required.

[0032] One method for measuring molecular interactions and binding, albeit relatively, reliably and simply, is the use of "luminescent probes." Conventional luminescent probes are used to qualitatively convey information to humans as optical signals through interactions and binding with samples. If the principles of these luminescent probes can be used to increase the amount of data by arraying them, and if machine learning can be used to perform regression analysis using a simple device or chip, then indirect measurement could potentially be used for data-driven analysis.

[0033] Furthermore, if machine learning or deep learning is used, the raw data measured does not need to be easy to understand for humans. Furthermore, if the amount of data is large, statistical mathematical processing can overcome the accuracy issue and convert it into data or signals that can be used for data-driven applications. This shift in judgment from humans to machines is required in data-driven research and development and manufacturing.

[0034] The present invention has been made in consideration of the above-mentioned techniques and problems, and specifically aims to provide an analysis method and an analysis device that can analyze the state of an object in detail in a short time and without requiring complicated operations or devices.

[0035] - Multidimensional data AI technology assumes a finite-dimensional linear space for input and output. It can also map a variety of data (organisms, cells, DNA, reaction pathways, molecular structures, etc.) into a finite-dimensional linear space. Mapping methods include multi-scale data (directed graphs), discrete structures (graphs), embedding into continuous spaces, and submodular functions.

[0036] - Regarding sufficient data volume: Complex systems manufacturing requires large amounts of diverse data. The reason for this is that a highly flexible model, including deep learning, is required to explain complex phenomena. Furthermore, in practice, it is believed that at least 10 times the number of degrees of freedom is required to make statistically meaningful inferences based on the model's degrees of freedom. Another reason diverse data is required is that it is not easy to predict and define in advance which data will be useful for the target variable. Unexpected explanatory variables may be influencing the data. Therefore, it is meaningful to collect diverse quantities with non-trivial relationships with the target variable as candidates for explanatory variables.

[0037] As mentioned above, collecting large amounts of diverse data requires fast, inexpensive data collection methods. Collecting data takes time and money. Therefore, a fast, inexpensive method of collecting data is necessary. Deep learning became possible because it was possible to collect large amounts of data cheaply from the internet. Here, there is data that is known to be deductively important. A priori knowledge can be reflected in the structure (architecture) of the network. In addition, deductive models and data-driven models can be combined. The approach of improving performance by using a deductive model as a first approximation and data-driven models as a second approximation can be said to be a universal approach.

[0038] <Organic EL elements as ultra-sensitive sensors> (1) Unique characteristics of organic EL elements As one of the methods for obtaining realistic and inductive data as described above, we, the inventors, first considered organic EL elements, the essence of which we are familiar with through our previous research and development. We decided to use blue phosphorescent light-emitting elements, which are said to be the most difficult organic EL-related technology and have not yet been put to practical use in electronic displays such as televisions and smartphones.

[0039] The main reason why blue phosphorescence has not yet been put to practical use in electronic display devices (green and red have been in practical use for over 10 years) is the short emission lifetime of the device. After many years of research into the cause of this, we have discovered that one reason is, of course, that blue phosphorescent materials must be aromatic or complex compounds with a wide band gap. However, we have also discovered that a more important reason is that blue phosphorescent materials are hundreds of times more sensitive than green or red phosphorescent materials to slight aggregation or changes in their state in the surrounding host material, as well as to trace amounts of water vapor and gas components that may be mixed in from outside the device over time.

[0040] We understand all of the issues, and are the only company in the world to have successfully used blue phosphorescence in lighting devices (Symphos OLED-010K), so we have a thorough understanding of its special characteristics and peculiarities.

[0041] (2) Reverse Thinking The primary reason why organic EL blue phosphorescent elements experience a decrease in luminance, i.e., a shortened light-emitting element lifespan, is not due to the decomposition of the luminescent material itself. It is due to the direct influence of changes in the environment surrounding the luminescent material on the phenomenon of luminescence. Specifically, the purity of the host material surrounding the luminescent material has a much greater impact than green phosphorescence or red phosphorescence. Furthermore, the lower the glass transition point of the host material, the greater the change in state of the host over time. However, it is also known that blue phosphorescence is much more sensitive to this influence than other types of luminescence, and is therefore more susceptible to adverse effects.

[0042] It is also possible to fabricate organic EL elements using a coating method that uses organic solvents to form the light-emitting layer. In this case, it has been confirmed that the lifespan characteristics of blue phosphorescent elements can vary significantly depending on the manufacturer or lot, even when the same coating solvent is used. Even if the specifications show the same purity and refinement, the performance of the fabricated elements can vary significantly, posing a major challenge for practical use. Furthermore, even with regard to the ultrapure water used before fabricating the elements, the luminescence lifespan characteristics of the elements vary even after impurities below the detection limit have been removed. Therefore, it has been found that in addition to organic substances, trace amounts of metal ions and other substances can also affect luminescence.

[0043] Although this effect is enhanced by blue phosphorescence, it is natural that a similar phenomenon occurs, albeit imperceptibly to the human eye, in green or red phosphorescent elements, or in organic EL elements that use a fluorescent rather than phosphorescent method. In other words, we thought that organic EL elements, which are sensitive to substances and their environment, could be used as sensor devices rather than light-emitting devices.

[0044] (3) Organic EL element as a sensor device Based on this concept, the present inventors investigated organic EL devices as ultra-high sensitivity sensors. Conventional organic EL elements are formed by laminating films using a vacuum deposition method, and are fabricated in a completely controlled environment, such as a high vacuum or an inert atmosphere, to completely isolate them from the atmosphere. Therefore, it is impossible to expose organic EL elements for research and development, as well as commercially produced organic EL devices, to the atmosphere during film formation.

[0045] Meanwhile, we have been developing technology for phosphorescent organic EL elements that can be fabricated under atmospheric pressure. During this manufacturing process, it is possible to add substances to be measured into or around the organic EL element. Therefore, it is possible to use this element as a sensor device. If we can add minute amounts of various substances to be measured into the organic EL element in this way, it should be possible to precisely obtain multidimensional real data, as mentioned above. We have actually confirmed that such data can be obtained with organic EL elements fabricated using a coating process.

[0046] (4) Benefits of Simple OLED However, the high-performance coating-type organic EL phosphorescent elements we have developed usually have a laminated structure of four organic layers, and we realized through actual studies that it is not realistic to create such a precise thin-film laminated element for the purpose of sensing. As a way to overcome this, we focused on simple-OLEDs, which we recently developed, which have one or two organic layers and can be fabricated in air.

[0047] This method can also be used with an inkjet system, making it possible to easily create sensor parts of any size and shape, from a few tens of microns to even larger sensors. Furthermore, thanks to the characteristics of inkjet technology, the sensor part can be formed as a multi-point element, numbering in the tens of thousands to millions, in a single manufacturing process. In this way, not only can the number of dimensions of data be increased, but the sheer volume of data can also be used to create revolutionary sensor devices that could truly be called disruptive innovations.

[0048] Although a voltage is usually applied to an organic EL element using a direct current, it is easy to change this application method to an alternating current, and even to apply a voltage in the form of a sine wave, triangular wave, square wave, etc. By changing the voltage application method in this way, it becomes possible to achieve even more multidimensionality.

[0049] This ability to achieve various multidimensionality is another major feature, along with high sensitivity. The reason this technique allows for multidimensional data generation is that organic EL devices are field-excited light-emitting devices. Their fundamental mechanism stems from the recombination of holes and electrons. During this process, all electrical changes, such as carrier trapping and charge charging and discharging within the compound, affect the recombination position and width of the recombination region, resulting in multiple interferences of the emitted light. Furthermore, it can be considered that the emission phenomena unique to organic EL devices, such as total reflection at the cathode and organic layer interface, affect the emission brightness, emission wavelength, emission spectrum waveform, and current characteristics and impedance changes during emission.

[0050] These considerations were only possible to understand and realize after numerous failures and successes in the light-emitting phenomenon and extending the lifespan of organic EL elements. We believe that our ability to correctly understand this phenomenon and handle it as data for data drive is an advanced invention that could not have been achieved by anyone other than us, the inventors. The sensing mechanism of the simple-OLED sensor is explained below using the schematic diagram shown in Figure 14.

[0051] In an organic EL (OLED), as shown in Figure 14, holes are injected from the anode, which is a transparent electrode, and electrons are simultaneously injected from the cathode, which is a reflective electrode. These recombine on luminescent molecules (also known as fluorescent molecules, phosphorescent molecules, or dopants) in the light-emitting layer and, when deactivated to their ground state, emit light with a wavelength corresponding to their band gap. Organic EL elements have an organic layer thickness of approximately 100 nm to 300 nm, due to the constraints of surface smoothness of the transparent electrode (typically a metal oxide thin film layer such as ITO or IZO) and the need for an electric field excitation mechanism that requires the film to be as thin as possible. This applies to organic EL elements formed by vacuum deposition, coating, or inkjet deposition.

[0052] In the vacuum deposition method, it is virtually impossible to return the pressure to atmospheric pressure midway and add the analyte to the element, but this is possible with the coating method and inkjet method, which form organic layers under atmospheric pressure. In particular, the simple-OLED that we developed, which has one or two organic thin film layers, makes it easy to add the analyte, and is more suitable for use as a sensor than other methods.

[0053] Two representative examples of analyte addition methods are described using Figures 15A to 15D. As shown in Figure 15A, a Simple-OLED is constructed by laminating a substrate, a transparent electrode, an insolubilized film (e.g., an organic thin film such as poly-TPD) formed by a large-area coating method such as spin coating or die coating, and a general-purpose polymer layer such as polystyrene. Then, as shown in Figure 15B, the analyte liquid from which data is to be acquired is applied to the polystyrene surface using a dropper, pipette, or dipping method, allowing a small amount of liquid to be impregnated. Excess liquid is then blown off using an air gun or similar to form a layer. Next, as shown in Figure 15C, an ink containing a dissolved emissive dopant and host is ejected using an inkjet method. This blends the polystyrene containing the analyte components with the host and emissive dopant, which then dries to form an organic EL layer containing trace amounts of analyte components. A metal cathode is then deposited on top of this by evaporation, sputtering, transfer, or other methods, completing the Simple-OLED device, as shown in Figure 15D.

[0054] By applying a current and voltage to the organic EL element, a luminescence phenomenon is observed, which is output as a signal reflecting the interaction between the sample component and the organic EL host or dopant (luminescent molecules). In other words, by measuring the luminescence, a signal on the order of nm that is linked to the interaction can be obtained.

[0055] Another method for adding analytes is described using Figures 16A to 16D. In this method, as shown in Figure 16A, a laminate is prepared, consisting of a substrate, a transparent electrode, an insolubilized film (e.g., an organic thin film such as poly-TPD) formed by a large-area coating method such as spin coating or die coating, and a general-purpose polymer layer such as polystyrene. Then, as shown in Figure 16B, an ink containing a dissolved luminescent dopant and host is first ejected using an inkjet method to form an organic EL layer. Next, as shown in Figure 16C, the substance to be measured is made into an ink and applied. To make the ink of the substance to be measured, the analyte may be added directly to an ink solution of the host and dopant for the organic EL. Alternatively, by utilizing the high-precision position control characteristics of inkjet printing, the ink may be positioned so that it is deposited over the entire surface from an ink tank and head separate from the ink for the organic EL. Then, as shown in Figure 16D, a metal cathode is deposited by evaporation, sputtering, transfer, or other methods to complete the simple-OLED device.

[0056] In this case, although it takes time to turn the sample into ink, the minute / precise ejection characteristics of inkjet printing make it possible to add a continuous amount of sample to the organic EL layer of each organic EL element, thereby enabling generation of signals with less noise or higher precision.

[0057] Next, we will explain the mechanism by which trace amounts of added sample components can be sensed using a simple-OLED. Figures 17A and 17B show photographs of the light-emitting state of an organic EL element. It is known that organic EL elements change color depending on the angle from which a human observes them. This is due to optical interference, and is similar to the principle behind why soap bubbles appear rainbow-colored, as shown in Figure 18. As shown in Figure 18, light is reflected and refracted at the interface of the soap bubble film, resulting in multiple optical interferences.

[0058] On the other hand, in organic EL devices, the organic layer between the anode and cathode is a thin film approximately 100 nm to 300 nm thick. Designing the layer so that recombination occurs near the center of the layer improves luminous efficiency and the lifespan of the light-emitting element. In particular, for phosphorescent elements that use phosphorescent compounds as luminescent dopants, it is considered preferable to achieve recombination as close to the center of the emissive layer as possible (a means of avoiding triplet-triplet annihilation, or TTA). This phenomenon is also observed in thermally activated delayed fluorescence (TADF) systems. In such organic EL elements where recombination occurs within the emissive layer rather than at the interface, the shift in carrier recombination position due to the interaction between trace components of the sample and the emissive layer materials (host, dopant, polymer, etc.) of the organic EL is further enhanced. This expands the dynamic range of signal acquisition.

[0059] This disclosure has the following two features: Feature 1) By applying various actions to the sample itself, the physicochemical properties of the sample are converted into optical information. In addition, by changing the type and degree of action, the types of optical information can be increased. Feature 2) By utilizing inkjet printing, large amounts of information can be obtained at high speed. These features function effectively when using AI to identify and distinguish substances.

[0060] First, Figures 19A to 19D are diagrams to explain classification using machine learning. Figure 19A shows a normal distribution of two classes. This represents a case where the distribution is simple and can be separated with a single broken line (number of parameters = 2). Figure 19B shows a complex distribution of two classes. Four broken lines are required for separation (number of parameters = 8). Figure 19C shows a normal distribution of four classes. Five broken lines are required for separation (number of parameters = 10). Figure 19D shows an example where separation is not possible in two dimensions, but is possible when expanded to three dimensions.

[0061] In machine learning classification, compared to when the distribution is simple and the number of target substances is small ( FIG. 19A ), (i) the more complex the data distribution (e.g., FIG. 19B ) and (ii) the more target substances (= the more classification classes) ( FIG. 19C ), the more parameters are required to describe the surface separating the classes (broken lines in the example shown). Therefore, the amount of data required increases. Even if classes cannot be separated in low dimensions, they may be separated by increasing the type of data (= dimensions) (e.g., FIG. 19D ). Higher dimensions require more parameters to describe the separating surface, resulting in a larger amount of data. Thus, increasing the amount and type (dimension) of data is necessary to achieve more accurate classification. The issues of quantity and type can be resolved by Features 1 and 2 of the present disclosure.

[0062] It is possible to reduce the estimation error of the class mean by assuming that the data independently follow the same probability distribution and taking the average of the data in the same class. This is equivalent to improving the accuracy of cluster center estimation in K-means, and can be said to be the most simple form of machine learning. It is similar to improving the position accuracy of the red line in Figure 19A.

[0063] Machine learning is divided into (i) the learning phase and (ii) the inference phase. The learning phase is a preparation stage, in which the data used is used to estimate parameters (for the separating surface in the above example). In the inference phase, all collected data is used for inference.

[0064] That is, the present disclosure makes it possible to realize the following: 1. The analytical device "generates a large number of signals for a single specimen simultaneously and quickly." 2. The majority of the signals generated in large quantities are "used to improve the quality of the data," and the data used for machine learning, deep learning, etc. is "an analytical method that uses a portion of the signals generated" from the device. 3. The analytical device is characterized by the use of artificial intelligence (AI) to improve the quality of the data. 4. The signals generated by the analytical device and the data whose quality has been improved using the majority of those signals are "used to solve inverse problems using machine learning," rather than being used by humans to understand phenomena deductively. 5. Since the focus is on application to chemical or biochemical manufacturing, the characteristics of the obtained signals should be "sensitive to sizes on the order of nm or Å," suitable (suitable) for describing interactions between substances. Since such sizes are difficult to observe directly in a quick and easy manner, they are converted into light and the light is measured to obtain a large amount of signals resulting from that size.

[0065] According to one embodiment of the present invention, an analysis method and an analysis device are provided that are capable of analyzing the state of an object in detail in a short time and without requiring complicated work or devices.

[0066] FIG. 1 is a schematic diagram illustrating the standard activated sludge process. FIG. 2 is a flowchart of an analytical method according to one embodiment of the present invention. FIG. 3 is a diagram illustrating the structure of a luminescent probe (multiple luminescent compounds) used in an analytical method according to one embodiment of the present invention. FIG. 4 is a diagram illustrating a method for producing a luminescent probe (multiple luminescent compounds) used in an analytical method according to one embodiment of the present invention. FIGS. 5A to 5D are diagrams illustrating the mechanism by which the luminescent probe (multiple luminescent compounds) used in an analytical method according to one embodiment of the present invention exerts its effect. FIG. 6 is a schematic diagram illustrating an analytical device used in a second aspect of an analytical method according to one embodiment of the present invention. FIG. 7 shows the results of principal component analysis in an example. FIG. 8A shows a linear discriminant analysis model plot when 95 brands of beverages of seven types were analyzed using luminescent probes 1 to 15, and FIG. 8B shows the confusion matrix. FIG. 9A shows a linear discriminant analysis model plot when 95 brands of beverages of seven types were analyzed using luminescent probes 1 and 17 to 31, and FIG. 9B shows the confusion matrix. Figure 10A shows a linear discriminant analysis model plot when 95 brands of beverages of seven types were analyzed using luminescent probes 1 to 15 and 17 to 31, and Figure 10B shows the resulting confusion matrix. Figure 11A shows a linear discriminant analysis model plot when 95 brands of beverages of seven types were analyzed using luminescent probes 1 and 32 to 45, and Figure 11B shows the resulting confusion matrix. Figure 12A shows a linear discriminant analysis model plot when 95 brands of beverages of seven types were analyzed using luminescent probes 46 to 60, and Figure 12B shows the resulting confusion matrix. Figure 13A shows a linear discriminant analysis model plot when 95 brands of beverages of seven types were analyzed after changing the explanatory variables used to create the discriminant model, and Figure 13B shows the resulting confusion matrix. Figure 14 is a diagram illustrating the light emission principle of an organic EL element. Figures 15A to 15D are schematic diagrams illustrating the addition of a sample to an organic EL element. Figures 16A to 16D are schematic diagrams illustrating the addition of a specimen to an organic EL element. Figures 17A and 17B are photographs of the light-emitting state of an organic EL element. Figure 18 is a diagram illustrating the reflection and refraction of light at the interface of a soap bubble film. Figures 19A to 19D are diagrams illustrating classification by machine learning. Figure 20 is a schematic diagram of various display elements.1 is a cross-sectional view showing an example of a microarray device; 2 is a plan view showing an example of a microarray device; 3 is a perspective view showing an example of a microarray device; 4 is a cross-sectional view showing an example of a microarray device; 5 is a plan view showing an example of a microarray device; 6 is a cross-sectional view showing an example of a microarray device; 7 is a cross-sectional view showing an example of a microarray device; 8 is a cross-sectional view showing an example of a microarray device; 9 is a cross-sectional view showing an example of a microarray device; 10 is a cross-sectional view showing an example of a microarray device; 11 is a plan view showing an example of a microarray device; 12 is a perspective view showing an example of a microarray device; 13 is a diagram for explaining a method for producing an OLED element; 14 is a diagram for explaining a method for producing an OLED element; 15 is a schematic diagram showing the mechanism of a multidimensional, rapid, large-volume data generation method using simple-OLED; 16 is a schematic diagram of sensing elements produced in Examples 4 and 5; 17 is a diagram showing a dataset obtained in Example 4; 18 is a diagram showing the results of discriminant analysis in Example 4; 19 is a diagram showing a dataset obtained in Example 5; 20 is a diagram showing the results of discriminant analysis in Example 5;

[0067] <Challenges in data-driven manufacturing (or research and development)> (Changes in the rate-limiting process (bottleneck)) As mentioned above, the most common means of generating large amounts of "essentially deductive data that corresponds to the substance itself or its state of existence" are microtiter plates (hereafter referred to as "microplates"), which are widely used in biochemical research and development, automatic dispensers that inject or drop chemical solutions or specimens into the plate's experimental areas (wells), and plate readers that automatically read color and light. However, although this method has been widely used in biochemistry, bioengineering, etc. for over 20 years, it is rarely used in the chemical industry, other manufacturing industries, or the food processing industry.

[0068] This is due to the fact that sample preparation and processing methods differ depending on the industry. For example, while water is the common solvent in biochemistry, organic solvents are used in the chemical industry, high temperatures of over 80°C are often used in food processing, and the volumes and scales of the objects being evaluated and inspected vary. In reality, data acquisition methods that are effectively used in the biochemistry field are rarely used in other industrial fields.

[0069] (1) Rate Limitations of Inductive Methods In other words, in many industrial and academic fields, particularly those dealing with liquids and fluids, the need to rapidly measure a large number of samples is a bottleneck in moving to a data-driven approach, i.e., an inductive approach. Meanwhile, in analytical equipment, technological innovation and widespread use of autosamplers have made it possible to perform precise analysis day and night. While this is one way to resolve the bottleneck, given the need for high-precision analysis, it is still insufficient in terms of obtaining large amounts of data in a short time, and it is not easy to secure a sufficient amount of data to apply inductive methods.

[0070] So how can we eliminate this bottleneck? While one could say that the rate-limiting factor in deductive R&D, which has been the mainstream up until now, was "eliminating trade-offs," in data-driven, or inductive, R&D, the quantity of data is a necessary condition, so we can think of the rate-limiting factor as having shifted to "eliminating bottlenecks." In other words, it is difficult to collect the amount of data required for data-driven R&D within the scope of human operations, and so it is necessary to eliminate human labor and adopt machine-based automation for everything from sample creation to data generation, data import, and data analysis.

[0071] (2) Dry-type high throughput and wet-type high throughput However, such large-scale automation usually requires advanced mechatronics technology that makes full use of robotic hands and control devices, as well as corresponding large-scale equipment, and robots are required depending on the application, such as daily experiments and processing. This creates inconveniences in terms of both cost and the time required to build the robots, and ultimately remains an ideal that has not been realized.

[0072] On the other hand, as a groundbreaking method to overcome these problems, computational science, which uses large-scale calculations to predict the deductive actions that humans think about, without humans having to perform experiments, manufacturing, or processing in the first place, and computer simulations that utilize this have come into the spotlight, and methods have also been adopted in which computer calculations actually perform human experimental actions.

[0073] In the fields of biochemistry, medicine, and bioengineering, traditional human-conducted experiments are called "wet," while computer-generated experimental predictions and results are called "dry," and they are used appropriately. Of course, similar methods are already widely used in the fields of chemistry and physics, with quantum chemistry calculations and dynamics calculations being prime examples, and by linking these with human experiments, they have achieved great results and effectiveness in a variety of fields.

[0074] However, even in development and production activities that appear to be sped up in this way, the "wet" part handled by humans is ultimately the rate-limiting factor, and to realize the super-smart society of the future, this rate-limiting factor must be overcome. In other words, in addition to the fields of biochemistry and medicine, the plate reader mentioned above needs to be used in an inexpensive, easy-to-use, data-driven way to quickly obtain the necessary and sufficient amount of high-quality data. This is essential to align manufacturing and research and development with the major inflection points of the SDGs, the super-smart society, and the Web 3.0 era. Realizing this is the greatest benefit, and we are entering an era in which methodologies themselves become inventions.

[0075] (3) Digital transformation in testing and diagnosis As mentioned above, it can be said that almost all testing and diagnosis to date have been activities that seek a definitive diagnosis of an object. In other words, devices have been developed to derive scientifically based analytical results that humans (experts) can confirm, and conversely, it can be said that society has created specialists who can use these devices to make precise definitive diagnoses and judgments.

[0076] However, just as pharmaceuticals have evolved from low-molecular-weight compounds to biologics such as antibody drugs and nucleic acid drugs such as mRNA, materials are also shifting from single elements such as metal plates to composite materials. Substances that can no longer be handled by definitive diagnosis or rigorous analysis have become indispensable in the society in which we live. In other words, a major change is that conventional analytical and diagnostic equipment can no longer handle things that humans want to use conveniently and safely. We should consider that now is the time for new non-definitive diagnostic technologies and methodologies to address this.

[0077] When the number of dimensions exceeds three, analysis becomes suddenly difficult because humans are no longer able to keep track of coordinates in their heads, but with machine learning and deep learning using AI, it is possible to perform calculations and analysis even in ultra-multiple dimensions of hundreds or thousands, although the number of dimensions that can be used is not infinite.In other words, we believe that digital transformation in testing and diagnosis is about creating a system that can generate large amounts of multidimensional data that can fully utilize these characteristics with as little human intervention as possible.

[0078] (4) Enabling high-throughput data acquisition When using machine learning or deep learning techniques, i.e., informatics, to perform non-definitive diagnoses or tests on items that cannot be definitively diagnosed or tested as described above, the results will be obtained in the form of clustering rather than quantitative values. However, as the amount of data increases, the clustering matrix will also become finer, and if it can be made finer and finer, the output will be essentially a clustering result similar to a quantitative value. In other words, performing this type of high-resolution clustering is the basic approach to digital transformation in testing and diagnosis.

[0079] To achieve this, it is necessary to miniaturize the biochemical microplates mentioned above and increase the number of wells. Additionally, it is also necessary to have manipulators that can precisely position specimens and reagents in the micro-wells (experimental fields), and devices that can accurately and quickly measure the color, light, and electromagnetic wave information of the micro-wells.

[0080] In addition to these mechanical and equipment requirements, scientific ingenuity is also required to amplify signals so that samples and reagents supplied to the microwells can sensitively output color, light, or electromagnetic waves as a result of their interactions. In response to these diverse new requirements, we, the inventors, devised a means to rationally resolve all of them, and have completed the technology of this disclosure. A representative example of the technology of this disclosure is described below.

[0081] <High-Throughput Multiwell with Built-in Light Source> (Light Source Corresponding to the Microplate Reaction Field (Microwell)) When acquiring data using a microplate with a large number of microwells for the purpose of non-confirmatory diagnosis or non-confirmatory testing as described above, for example, suppose that milk is selected as the sample. In this case, in addition to the milk, each of the numerous microwells must be filled with a reagent (also referred to as a "contrast agent" in this specification) that can interact with some component of the milk to generate color, light, or electromagnetic waves. Furthermore, when the sample is milk, it does not contain any component that emits light itself. Furthermore, the color itself is basically white, and chromaticity variation between batches is small. Therefore, a light source is required for sensitive detection. Furthermore, since data will be acquired in multiple dimensions, it is advantageous to have a wide variety of light sources.

[0082] On the other hand, providing such a multi-wavelength light source requires a large-scale device, which limits the place where it can be used and also causes inconvenience in terms of cost.

[0083] (1) Ideas from electronic displays As mentioned above, by shifting the focus of information provision from the human eye to measuring instruments and computers, it becomes possible to use large amounts of data, which can be said to be the fuel for data-driven development. Electronic displays are a prime example of providing stimuli in the form of optical information to the human eye. In order to generate all wavelengths of light that humans can perceive, electronic displays almost always emit the three primary colors of light - blue, green, and red - as tiny pixels.

[0084] Electronic displays are made by adjusting this light emission to wavelengths that are sensitive to the rods and cones of the human eye. By manipulating the wavelength so that it emits light at a wavelength that is appropriate for the contrast agent rather than the human eye, we realized that it could be used as a versatile on-demand light source for data science.

[0085] Figure 20 shows schematic diagrams of a liquid crystal display element, an organic EL element, and a plasma display element. Naturally, the self-luminous plasma display and organic EL display shown in Figure 20 can serve as excitation light sources. Furthermore, a non-emissive liquid crystal display can also serve as an excitation light source, as the light ultimately emitted from a white-light-emitting backlight via a color filter. Therefore, by understanding the structure and principles of such electronic displays and adjusting the wavelength to match the contrast agent rather than the conventional BGR, we came up with the idea that they could be used as multi-wavelength, multiple-planar excitation light sources (or viewing light sources).

[0086] It is not easy to imagine that the combination of consumer electronic displays and biochemistry microplates could be effectively applied to the digital transformation of manufacturing. However, based on the logic described above, it is easy to understand the feasibility and effectiveness of this idea.

[0087] (2) Ideas from high-quality printing Although electronic displays are currently mass-produced and the cost has been reduced significantly, manufacturing costs become a major issue when it comes to improving them to perform the data science mentioned above. So, how can we achieve two-dimensional arrangement of ultra-multi-wells more cheaply and easily? The answer lies in high-quality digital printing.

[0088] Digital printing machines that use the electrophotographic method used in copiers to print on demand (even one sheet at a time) with quality comparable to gravure printing are already available on the market (for example, https: / / www.konicaminolta.jp / business / products / graphic / ondemand_print / iq-501 / index.html and https: / / www.konicaminolta.jp / business / products / graphic / ondemand_print / iq-501 / index.html).

[0089] The printer's built-in Intelligent Quality Optimizer, called the IQ-501, is a unit that optimizes quality through automatic measurement, eliminating the need for manual labor and expertise in typical printing processes, such as adjusting color, density, and register position, as well as eliminating waste and redoing work due to human error and loss. By adapting this technology, it is entirely possible to read color information from pixels measuring several tens of microns across a large A2-sized area in one go and digitize it. Furthermore, if it can be used in data science, it will enable the rapid measurement and processing of unprecedented "big data linked to the science of materials."

[0090] In particular, the light source side is based on the present invention in that an electronic display is used, and for example, the following sheet-like light source can be suitably used in the present invention.

[0091] <Device Form for Carrying Out the Invention> The microarray device is characterized by having a plurality of microdot light-emitting portions arranged two-dimensionally and a plurality of microwell structures arranged two-dimensionally. This feature is a technical feature common to or corresponding to each of the following embodiments.

[0092] One embodiment of the microarray device includes an organic EL element, and the organic EL element has a plurality of the microdot light-emitting portions arranged two-dimensionally.

[0093] One embodiment of the microarray device comprises an organic EL element, the organic EL element having an anode, a receiving layer, a cathode, a sealing member, a circuit, and a power receiving portion, and the receiving layer having an insulating portion and a plurality of the microdot light-emitting portions arranged two-dimensionally.

[0094] One embodiment of the microarray device has two or more types of microdot light-emitting units with different maximum emission wavelengths, which increases wavelength selectivity and explanatory variables, improving the accuracy of data analysis.

[0095] In one embodiment of the microarray device, the plurality of microdot light-emitting portions include at least two or more of microdot light-emitting portions having a maximum emission wavelength of less than 380 nm, microdot light-emitting portions having a maximum emission wavelength of 380 nm or more but less than 500 nm, and microdot light-emitting portions having a maximum emission wavelength of 500 nm or more, thereby further improving the accuracy of data analysis.

[0096] In one embodiment of the microarray device, the plurality of microdot light-emitting portions include at least a microdot light-emitting portion having a maximum emission wavelength of less than 380 nm, a microdot light-emitting portion having a maximum emission wavelength of 380 nm or more but less than 500 nm, and a microdot light-emitting portion having a maximum emission wavelength of 500 nm or more, thereby further improving the accuracy of data analysis.

[0097] In one embodiment of the microarray device, the in-plane coordinates of the microdot light-emitting elements and the in-plane coordinates of the microwell structure elements at least partially overlap. This positional relationship allows simultaneous acquisition of absorption spectrum data and emission spectrum data of the sample during detection. Furthermore, this positional relationship shortens the optical path length, allowing the microwell structure elements and microdot light-emitting elements to be arranged at a higher density.

[0098] In one embodiment of the microarray device, the number of microdot light-emitting units is greater than the number of microwell structures, and multiple microdot light-emitting units are located directly below each microwell structure. This positional relationship makes it possible to irradiate a single microwell structure with excitation light of different wavelengths without using a spectroscopic system. Irradiating a single microwell structure with excitation light of different wavelengths can increase the number of explanatory variables and improve the accuracy of data analysis.

[0099] In one embodiment of the microarray device, the number of the microdot light-emitting units is the same as the number of the microwell structures, and the in-plane center coordinates of each of the microdot light-emitting units are the same as the in-plane center coordinates of each of the microwell structures. By achieving such a positional relationship, excitation emission efficiency is maximized, thereby improving the accuracy of data analysis.

[0100] One embodiment of the microarray device includes a receiving layer having an insulating portion and a plurality of the microdot light emitting portions arranged two-dimensionally, wherein the material of the insulating portion and the material of the microdot light emitting portion exist in the receiving layer with a concentration gradient in the depth direction and in the in-plane direction.

[0101] In one embodiment of the microarray device, at least one of the plurality of microwell structures includes a contrast agent on its surface. This improves the sensitivity of data acquisition. Furthermore, including a contrast agent on the surface enables the generated data to be multidimensional. Specifically, the generated data changes due to interactions such as energy transfer between the specimen and the contrast agent, allowing for the generation of more multidimensional data on the specimen's molecular structure, mixture ratio, density, environment, and the like.

[0102] In one embodiment of the microarray device, at least one of the microwell structures contains a contrast agent on its surface. The contrast agent uses electromagnetic waves irradiated from the microdot light-emitting unit as excitation light to exhibit one or more of the following luminescence mechanisms: fluorescence, phosphorescence, excimer luminescence, exciplex luminescence, thermally activated delayed fluorescence, excited-state intramolecular proton luminescence, triplet triplet annihilation luminescence, twisted intramolecular charge transfer luminescence, and aggregation organic luminescence. These luminescence mechanisms are suitable for multidimensionalizing generated data.

[0103] In one embodiment of the microarray device, at least one of the microwell structures contains a contrast agent on its surface, and the contrast agent is an oligonucleotide. Oligonucleotides are preferred as contrast agents because they can be adapted to the shape of the specimen. Oligonucleotides are also preferred because, compared to general compounds, the molecular group distance and quantitative ratio can be controlled by block polymerization.

[0104] The method for manufacturing a microarray device is characterized in that the microwell structures are formed by an inkjet method. This allows the microwell structures to be formed easily and quickly. The inkjet method is also suitable for forming microwell structures of any shape or minute size.

[0105] In one embodiment of the method for manufacturing the microarray device, the microdot light-emitting portions are formed by an inkjet method. This allows for the easy and rapid formation of a large number of microdot light-emitting portions. The inkjet method is also suitable for forming microdot light-emitting portions of any shape or minute size.

[0106] The test kit is a test kit including a microarray device, and is characterized by including the microarray device and one or more of a contrast agent, a specimen, a dripping device, a cover glass, a curing agent, a lens, and a microscope.

[0107] The inspection system is an inspection system including a microarray device, and is characterized by including the microarray device, an imaging device, and an analysis device.

[0108] In one embodiment of the testing system, the analysis device extracts RGB data or hyperspectral data from the digital image data generated using the microarray device and the imaging device, thereby enabling rapid and highly accurate analysis of even large amounts of data or complex data.

[0109] In one embodiment of the testing system, the analyzer uses artificial intelligence to evaluate the condition of the specimen, thereby enabling rapid and highly accurate analysis of even large amounts of data or complex data.

[0110] The testing method is characterized by using the microarray device.

[0111] [Microarray Device] The microarray device is characterized by having a plurality of microdot light-emitting sections arranged two-dimensionally and a plurality of microwell structure sections arranged two-dimensionally.

[0112] FIG. 21 is a cross-sectional view showing an example of the microarray device. The microarray device 110 shown in FIG. 21 includes a sheet-like light source unit 10 located on the lower surface of a support 31 and a sheet-like multiwell unit 20 located on the upper surface of the support 31. The light source unit 10 shown in FIG. 21 is an organic light-emitting diode (OLED). The organic EL element includes an anode 11, a receiving layer 19, a cathode 14, a sealing member 15, a barrier material 16, a circuit 17, and a power receiving section 18. The receiving layer 19 includes an insulating section 12 and a plurality of microdot light-emitting sections 13 arranged two-dimensionally. The multiwell unit 20 includes a cured layer 21 and a plurality of microwell structure sections 22 arranged two-dimensionally.

[0113] Figure 22 is a plan view showing an example of the microarray device, and Figure 23 is a perspective view showing an example of the microarray device.

[0114] In the microarray device 110, electromagnetic waves are irradiated from the microdot light emitter 13, which serves as a light source, onto the microwell structure 22, which serves as a reaction site. These electromagnetic waves act as excitation light, causing the specimen present in the microwell structure 22 (or a contrast agent that may be present together with the specimen) to emit light such as fluorescence and generate electromagnetic waves. The electromagnetic waves thus generated from the microwell structure 22 become scientific data regarding the specimen, substance, or object.

[0115] Furthermore, the microarray device 110 has multiple microwell structures 22 that serve as reaction fields, making it possible to generate a large amount of data at once. Furthermore, by having multiple microdot emitters 13 that serve as light sources, it is possible to freely adjust the number of microdot emitters 13 for one microwell structure 22 and the positional relationship between the microwell structure 22 and the microdot emitters 13. This allows the amount of electromagnetic waves irradiated to each microwell structure 22 and the irradiation angle to be changed depending on the microwell structure 22, making it possible to generate a variety of data in a single test. Furthermore, the microarray device 110 is easy to handle, making it possible to quickly generate a large amount of data.

[0116] Some examples of the positional relationship between the in-plane coordinates of the microdot light-emitting portion 13 and the in-plane coordinates of the microwell structure portion 22 will be described.

[0117] For example, the in-plane coordinates of the microdot light emitters 13 and the microwell structure 22 can be positioned so as to at least partially overlap. By adopting such a positional relationship, it becomes possible to simultaneously capture absorption spectrum data and emission spectrum data of the sample during detection. Furthermore, by adopting such a positional relationship, the optical path length can be shortened, making it possible to provide the microwell structure 13 and the microdot light emitters 22 at a higher density.

[0118] 21, it is also possible to arrange the positional relationship such that the number of microdot light-emitting units 13 and the number of microwell structures 22 match, and the in-plane center coordinates of each microdot light-emitting unit 13 match the in-plane center coordinates of each microwell structure 22. By arranging the positional relationship in this way, the excitation light emission efficiency is maximized, thereby improving the accuracy of data analysis.

[0119] On the other hand, as shown in the cross-sectional view of Figure 24 and the plan view of Figure 25, the microdot light emitters 13 and the microwell structure 22 can also be positioned alternately. Even in this case, the electromagnetic waves emitted from the microdot light emitters 13 are irradiated onto the microwell structure 22. By adopting such a positional relationship, it is possible to prevent direct detection of the light emitted from the microdot light emitters 13, which could become noise. Since this positional relationship improves the signal-to-noise ratio, it is particularly effective when the analyte (specimen) is to distinguish slight differences in structure or substituents of molecules with the same skeleton, or when the luminescence signal intensity of the analyte (specimen) is weak.

[0120] 26, the number of microdot light emitters 13 can be made greater than the number of microwell structures 22, and a positional relationship can be established in which multiple microdot light emitters 13 are located directly below each microwell structure 22. This positional relationship makes it possible to irradiate one microwell structure 22 with excitation light of different wavelengths without using a spectroscopic system. Irradiating one microwell structure 22 with excitation light of different wavelengths can increase the number of explanatory variables, thereby improving the accuracy of data analysis.

[0121] Next, details of the microarray device 110 shown in Fig. 21 will be described. Note that the microarray device 110 shown in Fig. 21 is of a type having organic EL elements as light source units 10, but the present invention is not limited to this.

[0122] The support, gas barrier film, electrode, demand layer, insulating part, and other preferred embodiments applied to the organic EL element can be those described in known documents. For example, those described in JP 2022-99677 A, JP 2022-116901 A, and WO 2022 / 163199 can be similarly utilized.

[0123] (Support) The support 31 can be made of a material that can transmit the electromagnetic waves irradiated from the microdot light-emitting section 13. If the multiwell unit 20 is placed directly on the light source unit 10, the support 31 is not necessary. Furthermore, if the microdot light-emitting section 13 and the microwell structure section 22 are formed in the same layer or the same unit, the support 31 is also not necessary.

[0124] The support 31 may be provided with a filter that converts the wavelength of the electromagnetic waves emitted from the microdot light-emitting section 13, or a filter that converts the color of light emitted from the microdot light-emitting section 13 into multiple colors using a fluorescent material.

[0125] 21 has a light source unit 10 which is an organic EL element. The light source unit 10 which is an organic EL element has an anode 11, a receiving layer 19, a cathode 14, a sealing member 15, a barrier material 16, a circuit 17, and a power receiving section 18. The receiving layer 19 has an insulating section 12 and a microdot light-emitting section 13.

[0126] The thickness of the light source unit 10 can be, for example, about 0.2 μm.

[0127] (Electrode) A metal, alloy, electrically conductive compound, or mixture thereof having a large work function is preferably used as the anode 11. A metal, alloy, electrically conductive compound, or mixture thereof having a small work function (referred to as an electron injecting metal), or mixture thereof is preferably used as the cathode 14.

[0128] These electrodes (anode 11 and cathode 14) can be obtained by depositing an electrode material as a thin film by, for example, a method such as vapor deposition or sputtering, or a method such as coating and drying a precursor solution. The electrodes (anode 11 and cathode 14) may be provided as a flat film with a uniform thickness, or may be provided in a desired pattern shape. In the case of a pattern shape, they can be formed by, for example, a photolithography method, or an inkjet method for coating a precursor solution. Alternatively, when high pattern precision is not required, they can also be formed by vapor deposition or sputtering of an electrode material using a mask of a desired shape.

[0129] 21, the anode 11 is on the microwell structure side, but the cathode 14 may also be on the microwell structure side. In either case, the electrode on the microwell structure side is preferably transparent or semi-transparent so as to transmit the electromagnetic waves irradiated from the microdot light-emitting unit 13.

[0130] (Receptor Layer) The receptor layer 19 has an insulating portion 12 and a plurality of secondary arranged microdot light-emitting portions 13. The receptor layer 19 is formed by being laminated on at least one of the electrodes (anode 11 or cathode 14). It is usually preferable that the receptor layer 19 be formed as a continuous layer.

[0131] The receiving layer 19 is formed so that its entirety is in contact with one of the electrodes. The region where the receiving layer 19 is formed may be, for example, a region that covers the entire surface of the electrode, or a region that covers a specific region of the electrode. Note that the insulating portion 12 is preferably made of at least one type of insulating polymer.

[0132] (Microdot Light Emitting Section) A plurality of microdot light emitting sections 13 are arranged two-dimensionally.

[0133] The constituent materials of the microdot light emitting portion 13 include a light emitting compound and a host compound.

[0134] The microdot light-emitting portions 13 are preferably formed by an inkjet method, which allows for easy and rapid formation of a large number of microdot light-emitting portions 13. The inkjet method is also suitable for forming microdot light-emitting portions 13 of any shape or minute size.

[0135] Regarding preferred embodiments of the inkjet method, those described in JP-A-2022-99677, JP-A-2022-116901, and WO 2022 / 163199 can be similarly utilized.

[0136] The microdot light emitting portion 13 may be present in a coffee ring shape.

[0137] (Sealing Member) For example, a sealing adhesive can be used for the sealing member 15. Examples of sealing adhesives include photo-curable and heat-curable adhesives having a reactive vinyl group such as acrylic acid oligomers and methacrylic acid oligomers, moisture-curable adhesives such as 2-cyanoacrylate esters, heat- and chemical-curable (two-part mixed) adhesives such as epoxy-based adhesives, hot-melt polyamide, polyester, polyolefin, and cationic curable ultraviolet-curable epoxy resin adhesives.

[0138] The sealing adhesive is preferably one that can be cured at temperatures ranging from room temperature to 80° C., since the organic EL element may be deteriorated by heat treatment.

[0139] The sealing adhesive may contain a desiccant. The sealing adhesive may be applied using a dispenser or by printing such as screen printing.

[0140] (Barrier Material) A resin film having a gas barrier film or the like can be used as the barrier material 16. The gas barrier film is the same as that which can be provided on the support described above.

[0141] (Circuit) The circuit 17 is provided to pass electricity from the power receiving unit 18 to the anode 11 and the cathode 14, and its configuration is not particularly limited.

[0142] (Power Receiving Unit) The power receiving unit 18 is not particularly limited, but may be one that can receive power via near field communication (NFC), one that can receive power from a battery, one that can be connected to a power supply cable, or a combination of these, etc. From the viewpoint of miniaturizing the microarray device, it is preferable that the power receiving unit 18 be one that can receive power via NFC.

[0143] (Multiwell Unit) The multiwell unit 20 has a hardened layer 21 and a plurality of microwell structures 22 arranged two-dimensionally.

[0144] The thickness of the multiwell unit 20 can be, for example, 1 to 100 μm.

[0145] The multiwell unit 20 can be formed, for example, by applying the material of the hardened layer 21 to an area other than the area that will become the microwell structure 22 on the support 31 .

[0146] Although there are no particular limitations on the method for applying the material for the cured layer 21, it is preferable to use an inkjet method. This allows the multiwell unit 20 and the microwell structure 22 to be formed easily and quickly. The inkjet method is also suitable for forming microwell structures 22 of any shape or minute size.

[0147] In the case of the inkjet method, the inkjet device or the like can be formed under the same conditions as those for forming the microdot light emitting portion 13, for example.

[0148] The material of the curable layer 21 may be composed of, for example, an actinic radiation polymerizable compound and a polymerization initiator. Depending on the type of specimen, the insulating polymer that can be used to form the above-mentioned receiving layer may also be used as the material of the curable layer 21.

[0149] The term "actinic ray polymerizable compound" refers to a compound that crosslinks or polymerizes when irradiated with actinic rays. Examples of actinic rays include electron beams, ultraviolet rays, α rays, γ rays, and X-rays. Among actinic rays, ultraviolet rays or electron beams are preferred. Examples of actinic ray polymerizable compounds include cationically polymerizable compounds, radically polymerizable compounds, and mixtures thereof. Among actinic ray polymerizable compounds, radically polymerizable compounds are preferred. The actinic ray polymerizable compound may be any of a monomer, a polymerizable oligomer, a prepolymer, and a mixture thereof.

[0150] The radical polymerizable compound is a compound having an ethylenically unsaturated double bond group in the molecule. The radical polymerizable compound may be a monofunctional or polyfunctional compound. Examples of the radical polymerizable compound include various acrylates and methacrylates, which are unsaturated carboxylic acid ester compounds.

[0151] The cationically polymerizable compound is a compound having a cationically polymerizable group in the molecule. Examples of the cationically polymerizable compound include epoxy compounds, vinyl ether compounds, and oxetane compounds.

[0152] The polymerization initiator may be any one that can initiate polymerization of the actinic ray-polymerizable compound. For example, when the actinic ray-curable ink contains a radically polymerizable compound, the polymerization initiator may be a photoradical initiator. When the actinic ray-curable ink contains a cationically polymerizable compound, the polymerization initiator may be a photocationic initiator (photoacid generator). Note that when the actinic ray-curable ink can be sufficiently cured without a polymerization initiator, such as when the actinic ray-curable ink is cured by irradiation with electron beams, a polymerization initiator is not necessary. Known polymerization initiators can be used.

[0153] The material of the cured layer 21 applied on the support 31 is cured by irradiation with actinic rays. The actinic rays to be irradiated can be selected from, for example, electron beams, ultraviolet rays, α rays, γ rays, and X-rays. Irradiation conditions such as irradiation time and illuminance can be appropriately selected depending on the material and thickness of the cured layer 21.

[0154] It is preferable that at least one of the multiple microwell structures 22 contains a contrast agent on its surface. This can improve the sensitivity of data acquisition. Furthermore, by including a contrast agent on the surface, it becomes possible to generate multidimensional data. Specifically, the generated data changes due to interactions such as energy transfer between the specimen and the contrast agent, so that data such as the molecular structure, mixture amount ratio, density, and environment of the specimen can be generated in a more multidimensional manner.

[0155] Multiple contrast agents may be used, with different microwell structures 22 containing different contrast agents, thereby enabling more multidimensional generation of data regarding the specimen.

[0156] The contrast agent is preferably applied to the microwell structure 22 by, for example, dispersing or dissolving the contrast agent in a solvent to prepare a contrast agent-containing liquid, and then applying the contrast agent to the microwell structure 22 by an inkjet method. The inkjet method allows for accurate application of the contrast agent to the fine microwell structure 22. The components of the contrast agent will be described later.

[0157] Preferred embodiments other than the above-described embodiment and modifications will be described below.

[0158] The light source unit, which is an organic EL element, may have an electron transport layer, a hole blocking layer, an electron injection layer, a hole transport layer, an electron blocking layer, a hole injection layer, etc. In this case, the layer configuration including the electrodes and these layers can be, for example, as follows.

[0159] (i) anode / acceptor layer / cathode (ii) anode / acceptor layer / electron transport layer / cathode (iii) anode / hole transport layer / acceptor layer / cathode (iv) anode / hole transport layer / acceptor layer / electron transport layer / cathode (v) anode / hole transport layer / acceptor layer / electron transport layer / electron injection layer / cathode (vi) anode / hole injection layer / hole transport layer / electron blocking layer / acceptor layer / hole blocking layer / electron transport layer / cathode

[0160] For each layer such as the electron transport layer, those described in, for example, JP-A-2022-99677, JP-A-2022-116901, and WO 2022 / 163199 can be similarly utilized.

[0161] Microwell structure 22 may be formed in receptive layer 19, as shown in the cross-sectional views of Figures 27 and 28. Such microwell structure 22 can be formed by a method of forming it at the same time as forming receptive layer 19, a method of forming it by drilling holes in support 31 and the electrode, a method of forming it using a porous material, or the like.

[0162] The microdot light-emitting units 13 are not limited to light-emitting units of organic EL elements, but may be light-emitting units of inorganic light-emitting diodes or plasma displays. Fig. 29A is a cross-sectional view of a microarray device 110 in which the light source units 10 are inorganic light-emitting diodes.

[0163] The microdot light-emitting section 13 may be one that generates near-infrared light, infrared light, or ultraviolet light. The microdot light-emitting section 13 may also be made of PVK, phosphite, acene, a Pt complex thermal deactivator, a metal oxide, or a composite metal oxide.

[0164] The plurality of microdot light-emitting elements 13 are preferably two or more types of microdot light-emitting elements 13 with different maximum emission wavelengths, which increases wavelength selectivity and explanatory variables, improving the accuracy of data analysis.

[0165] The plurality of microdot light-emitting units 13 preferably include at least two or more of the following: microdot light-emitting units with a maximum emission wavelength of less than 380 nm; microdot light-emitting units with a maximum emission wavelength of 380 nm or more but less than 500 nm; and microdot light-emitting units with a maximum emission wavelength of 500 nm or more, and more preferably include at least all of them. This allows the microarray device 110 to generate the following data, thereby further improving the accuracy of data analysis.

[0166] Electromagnetic waves with wavelengths less than 380 nm contribute to generating data related to the absorption of, for example, excitation light sources and molecular structures. Examples of electromagnetic waves in this wavelength range include ultraviolet light with wavelengths of 10 nm or more and less than 380 nm. Ultraviolet light includes far ultraviolet light with wavelengths of 10 nm or more and less than 200 nm, and near ultraviolet light with wavelengths of 200 nm or more and less than 380 nm.

[0167] Electromagnetic waves with a wavelength of 380 nm or more and less than 500 nm contribute little as an excitation light source for generating emission data of a single compound, for example, but contribute to generating data that more emphasizes the interaction and excitation and absorption data of cleavage products of conjugated compounds.

[0168] Electromagnetic waves with wavelengths of 500 nm or more contribute to the generation of data on the absorption of a single compound and data on aggregates and decomposition products. Examples of electromagnetic waves in this wavelength range include green to red visible light with a maximum emission wavelength of 500 nm or more and less than 780 nm, and infrared light with a maximum emission wavelength of 780 nm or more and less than 100 μm.

[0169] The microdot light emitters 13 may be heat-generating so that the temperature of the microwell structure 22 can be controlled.

[0170] In the present invention, for example, one microwell structure can be provided with one or more microdot light-emitting units, which allows the size, shape, emission wavelength, etc. of the microdot light-emitting units to be adjusted on demand to suit the specimen or contrast agent.

[0171] In conventional microarray plate analysis, the light source wavelength, light intensity, temperature, etc., were not controlled for each individual well. However, this microarray device can also control the light source wavelength, light intensity, temperature, etc., for each individual well. This makes it possible to intentionally induce chemical or biochemical reactions by exceeding the activation energy of samples or reagents for the purpose of acquiring data in industrial or food processing applications.

[0172] The arrangement density of the microwell structures 22 is 5 / cm ２ This allows the microarray device 110 to be miniaturized. The arrangement density of the microwell structures 22 is, for example, 5 to 10,000 / cm. ２ It can be said that:

[0173] The shape and position of the microwell structure 22 can be freely changed depending on the purpose, such as specimen structure / immobilization, centrifugation, light path, reaction path, drying promotion, drying suppression, etc.

[0174] The shape of the microwell structure 22 is not limited to a perfect circle, square, hemisphere, square bottom, etc. The shape can be designed to suppress drying of the sample, support, degas, etc., a flow path shape for acquiring reaction time series data or sieving data, a shape suited to light irradiation / removal, etc.

[0175] The size of the microwell structure 22 can be, for example, several tens of μm depending on the inkjet resolution. Any size larger than that can be freely designed using the input image.

[0176] The method for forming the microwell structure 22 is not particularly limited, and it can be formed by inkjet lamination of a curable resin, 3D printer lamination of a curable resin, laser processing of a resin, photolithography of a curable resin, etc. The microwell structure 22 can also be formed by a method in which a solvent is ejected into a resin by inkjet to create holes. Among these, inkjet lamination is preferred because it combines fineness, on-demand properties, and waste-free properties.

[0177] The shape of the microwell structures 22 may be different for each microwell structure 22, as shown in the cross-sectional view of FIG. 30, the plan view of FIG. 31, and the perspective view of FIG.

[0178] By utilizing such existing electronic displays as flat light sources, they can be used as light sources for at least the 16 types of multi-well devices mentioned above. However, because they are light sources for generating electromagnetic waves from the dyes, fluorescent materials, or contrast agents that serve as sensors, they do not require a long light-emitting life like conventional electronic displays, and there is almost no need for low power consumption.

[0179] In other words, among the electronic display methods known to date, even those that have not been put to practical use due to their lifespan or power consumption can be used for this purpose, and conversely, it goes without saying that a method that can easily form a point light source is preferable in this invention.

[0180] For example, such examples are shown in Figures 33 and 34. Examples include sheet-shaped OLEDs manufactured by a highly simplified OLED element manufacturing method (for example, SID Symposium Digest of Technical Papers, 52(1), pp. 297-300). This sheet-shaped OLED can also handle multi-dots, with one dot per well. Furthermore, it becomes possible to easily and conveniently create on-demand point light sources that emit the wavelengths most suitable for generating electromagnetic waves from the color developer, phosphor, contrast agent, etc. present in each well, using an inkjet method. Therefore, this is the most preferred method for manufacturing a light source sheet in the present invention.

[0181] As a modification of the above, the light source unit 10 shown in FIG. 29A may be an organic EL, and one light-emitting section 13 may be arranged for one microwell structure 22 as shown in FIG. 29A.

[0182] As a modification of the above, the light source unit 10 shown in FIG. 26 may be an organic EL, and as shown in FIG. 26, a plurality of light-emitting sections 13 may be arranged for one microwell structure 22.

[0183] As a modification of the above, when the light source unit 10 is an organic EL, one light-emitting section 13 may be arranged for all microwell structures 22, as shown in FIG. 29B.

[0184] As a modification of the above, when the light source unit 10 is an organic EL, one light-emitting section 13 may be arranged for one area formed by a plurality of microwell structures 22, as shown in FIG. 29C.

[0185] 29D, light source unit 10 may be a microarray device having a light-emitting section 71, a light guide plate 72, a color filter 73, and a diffusion plate 74. In this case, one light-emitting section may be arranged for one microwell structure 22, or one light-emitting section may be arranged for multiple microwell structures 22. Furthermore, light-emitting sections that emit the same color may be arranged over the entire surface, or light-emitting sections that emit different colors may be arranged in different areas.

[0186] As a variation of the above, light source unit 10 shown in Fig. 29A may be a plasma display or a micro LED. In this case, as shown in Fig. 29A, one light-emitting unit 13 may be arranged for one microwell structure 22, or one light-emitting unit 13 may be arranged for multiple microwell structures 22. Furthermore, light-emitting units 13 that emit the same color may be arranged over the entire surface, or light-emitting units 13 that emit different colors in different areas may be arranged.

[0187] (3) Advantages of a Multiwell Sheet with a Built-in Light Source As described above, the present invention makes it possible to convert the wells, which are the reaction sites, into ultra-multiwells. Furthermore, it is possible to equip the multiwells with light sources, and to adjust the size, shape, and emission wavelength of the light source on demand to suit the intended use and the specimen, developer, or contrast agent in the wells. Furthermore, the substrate for the multiwells and light source can be either glass or plastic. In either case, it is possible to create an ultra-thin multiwell sheet with a built-in light source that is less than 1 mm thick.

[0188] For example, this multi-well sheet can be used as an ID sheet for traceability in primary industries such as milk and fruit juice. It can also be used to understand the status and record reactions in industrial manufacturing processes such as wine brewing, soy sauce brewing, fake meat production, paint manufacturing, and ink manufacturing. In such applications, the ease of archiving is also a key feature. Multi-well sheets, in particular, have a built-in light source, eliminating the need for an external light source and enabling data acquisition, making them versatile and offering the major advantage of being applicable to almost all primary and secondary industries.

[0189] (4) Data Acquisition Method In the present invention, the means for acquiring multidimensional and large amounts of data are color, light, and electromagnetic waves, and therefore the detection method will be to use a spectrophotometer, a fluorescence clock, a photodetector according to wavelength, an electromagnetic wave sensor, etc.

[0190] Furthermore, the simplest and most convenient method of acquiring data is to convert color, light, or electromagnetic waves into an electrical signal using a photoelectric conversion sensor such as a CCD or CMOS sensor in synchronization with the light emitted from a sheet-shaped excitation light source, which is simple and preferable.

[0191] This can be a photo taken with a digital camera on a smartphone, and data can be obtained from analyzing the image data, so this type of use is thought to be suitable regardless of the situation in which it is used.

[0192] (5) Remote data processing using the cloud While the data converted into electrical signals or image data is generally stored directly on a data server or data lake, cloud servers are now available and are suitable for use. In particular, remote data processing using the cloud is effective when the location and time of data acquisition differ from that of data processing.

[0193] (6) Compatibility with the Digital Garden City-State Vision Until now, digital manufacturing and its representative advanced tools, such as materials informatics and process informatics, have targeted industrial products, i.e., the secondary industry. As mentioned above, in the future, these will need to be applied to primary industries such as food and agricultural products, as well as processing industries such as food and beverages that use these products, and digital transformation will need to be brought about in all industries.

[0194] In particular, to realize the "Digital Garden City Nation Vision" advocated by the Japanese government in recent years, it is practically difficult to apply conventional, heavy, and large-scale advanced analytical equipment that accurately performs definitive diagnoses to regional factories and agricultural processing sites. This invention makes it possible to acquire useful data related to multidimensional and large amounts of substances on-site using lightweight, ultrathin sheet-like multi-wells, which can then be processed remotely via the cloud. This is an important and innovative method that is consistent with the concept of comfortable and safe living not only for Japan but for all of humanity, and with the SDGs, such as decarbonization. We believe it is an excellent invention that can be developed for a wide range of applications.

[0195] This invention relates to manufacturing DX, which aims to achieve data-driven research and development and data-driven production. It identifies the problems and bottlenecks specific to manufacturing in inductive inverse problem solving, which has traditionally been an excellent technique with some notable effects, and achieves rapid and simple measurements below the molecular scale by combining chemistry and optics, thereby solving the conventional data acquisition rate-limiting problem, as well as the analytical method and analytical device required for this.

[0196] The specific technical concept of a multidimensional, rapid, large amount of data generation method using simple-OLED (organic EL element) will be explained below using figures. Figure 35A shows a schematic diagram of the appearance of a simple-OLED. As shown in Figure 35A, in an organic EL element, a specific carrier recombination occurs as a hole (h ＋ ) and electrons (e － ) are carriers. When a trace component of a specimen is mixed into such an organic EL element, it almost certainly causes a change in the mobility of either holes or electrons. Therefore, it is possible to generate a signal by utilizing the optical interference of the luminescence phenomenon resulting from the interaction (on the order of nm or Å) between the trace component of the specimen and the substance in the organic EL element.

[0197] Figures 35B to 35D show the mechanism by which this interaction state is obtained as an optical signal. Figure 35B is a schematic cross-sectional view of a simple-OLED, and Figure 35C is a deformed schematic view of the area enclosed by the dashed line in Figure 35B (mainly the organic layer portion). The organic layer (light-emitting layer) in Figure 35C contains a binder, a dopant, and a host, but does not contain an analyte component. Furthermore, the type and ratio of the host and dopant in this simple-OLED element are optimized to some extent so that recombination occurs approximately in the center of the organic layer. If the organic layer is 200 nm thick, the recombination center of the recombination position (light-emitting site) would be approximately 100 nm from the cathode, which is exactly in the center of the organic layer.

[0198] A trace component of the analyte is introduced into the device, as shown in Figure 35D. A simple-OLED device is fabricated using the methods shown in Figures 15C or 16C. This results in an interaction between the analyte component and the host, dopant, or both, or the binder polymer. This causes a change in the velocity of the hole or electron flow, shifting the position of the light-emitting site. This change is observed through interference between the light emitted directly from the transparent anode and the light emitted after hitting the reflective cathode, shifted in phase by twice the optical path length (see Figure 18). In other words, this observed light is obtained as a signal describing the interaction caused by the analyte. The difference between observed light 1 (not containing the analyte, shown in Figure 35C) and observed light 2 (containing the analyte, shown in Figure 35D) is what enables the light emitted from this simple-OLED to produce a signal capable of describing interactions on the nm or Å scale. This is the sensing mechanism.

[0199] Furthermore, by using the inkjet method to create the elements, it is possible to form a large number of elements (more than tens of thousands) on a single substrate, making it possible to rapidly acquire large amounts of data. Another advantage of the inkjet method is that the shape of the organic EL element can be freely designed to suit the characteristics of the device to be detected, from circular to triangular, square, or irregular. Furthermore, the following are possible using this sensor device, making it suitable for use as an unprecedented, futuristic sensor device.

[0200] (i) When a host and a dopant are used in the ink liquid ejected from the inkjet head, changing the type of host or the type of dopant changes the state of interaction with the trace components of the specimen. Therefore, by arranging multiple of these in an element, it becomes possible to obtain multidimensional data.

[0201] (ii) By aligning the organic layer of two-dimensionally arranged organic EL elements, the specimen can be converted into ink and overwritten onto the organic layer, as shown in Figures 16A to 16D. Furthermore, the amount of specimen added to each organic EL element can be determined at the discharge resolution level of an inkjet head (a few picoliters), enabling ultra-precise titration-like addition.

[0202] (iii) In addition to the test sample, a substance that can intentionally change the carrier recombination position of a simple-OLED can be made into ink and overwritten as in (ii), thereby emphasizing the change in light. In addition, since a regression curve can be intentionally created, it is also possible to eliminate inconvenient data such as outliers.

[0203] (iv) Two-dimensional inkjet printing makes it possible to easily create organic EL elements of, for example, 1,000 dots x 1,000 dots (= 1 million dots). Therefore, it is not necessary to use all of the data as data for machine learning or truth learning, and the majority of the data can be used to improve the quality of the data (for example, (ii) and (iii) are examples, and it also becomes possible to take statistical averages).

[0204] (v) By using an element that does not contain a specimen as a comparison element and subtracting data from that of an element that does contain a specimen, light-related data is always output as a "relative value" rather than being used as is. This makes it possible to cancel out performance variations between the organic EL elements that serve as sensors and defects in the elements themselves.

[0205] This is not limited to simple-OLEDs, but can be used in a similar manner with any element that can be arrayed two-dimensionally, and the same is true when the above-mentioned DNA-type luminescent sensors are arrayed.

[0206] We recognize that this discovery is completely new and an extraordinary invention that could only be noticed by those who understand the light emission and quenching mechanisms of organic electroluminescence, but we cannot deny the possibility that there are examples in numerous papers and patents in which it has been used as a sensor device. However, these were applications as sensors, and are of a fundamentally different nature from the present invention, and therefore should be distinguished from the present invention.

[0207] <Specific Embodiments> The present invention will be described in detail below based on embodiments, but the present invention is not limited to these embodiments.

[0208] (1) Analysis Method The analysis method of this embodiment is a method for analyzing the state of an object, and includes the steps of mixing the object and one or more types of luminescent probes whose luminescence behavior changes upon interaction with the object and allowing them to interact (hereinafter also referred to as the "interaction step"); exciting the luminescent probes that are in a state of interaction with the object to produce two or more types of luminescence and simultaneously detecting the two or more types of luminescence (hereinafter also referred to as the "light detection step"); and analyzing the two or more types of luminescence detected to identify the state of the object (hereinafter also referred to as the "analysis step"). This analysis method is capable of detecting nanometer-sized chemical structures and physical structures, as well as changes therein. Note that the two or more types of luminescence analyzed by this analysis method may be two or more types of light emitted by a single luminescent probe, or may be two or more types of light emitted by different types of luminescent probes.

[0209] The analysis method of this embodiment uses an indirect and simple means of detecting light, and therefore does not require complicated equipment or complicated processes to acquire data. Furthermore, this method simultaneously detects two or more types of light emitted by a luminescent probe that has interacted with a target object, making it possible to acquire large amounts of data about a single target object all at once. Furthermore, by making full use of machine learning and deep learning during analysis, it is possible to analyze the target object in detail based on subtle differences that are imperceptible to humans.

[0210] Here, the object that can be analyzed by the analysis method of this embodiment is preferably one that changes state due to a chemical change or a nanometer-sized physical structural change, etc. The object may be in a liquid, solid, or gaseous state as long as it is capable of interacting with the luminescent probe.

[0211] Furthermore, the object to be analyzed in this embodiment may be a substance whose structure is known or a substance whose structure is unknown. The object may also be a mixture of various compounds. The field to which the object belongs is not particularly limited, and the object may be a substance, compound, or composition belonging to any field, such as the medical field, industrial field, food field, or water quality control field. Examples of objects belonging to the medical field include proteins, antibodies, antibody-attached beads, tumor markers, etc. On the other hand, examples of objects belonging to the industrial field include metal nanoparticles, carbon nanotubes, magnetic fluids, nanosilica, crystalline zirconia, etc. Examples of objects belonging to the food field include agricultural products and processed products thereof. Examples of objects belonging to the water quality control field include water and microorganisms, etc.

[0212] Furthermore, the analytical method may be performed, for example, for the prototyping, development, or research of industrial products, or may be performed in the process of testing a product production line (in a test plant) or the process of producing a prototype (in a pilot plant). Furthermore, it may be performed in the process of manufacturing a product or the process of inspecting the quality of a product. It may also be performed for the purpose of the manufacturing, preservation, or quality control process of agricultural products and their processed products, or for water quality management. It may also be performed for various analyses in the medical field.

[0213] Here, the luminescent probe may be any compound whose luminescence behavior changes upon interaction with a target substance. In this specification, "the luminescence behavior of a luminescent probe changes upon interaction with a target substance" refers to the luminescent probe interacting with a specific group or structure of the target substance, causing a change in its electronic state (chemical interaction) or a change in its three-dimensional structure (physical interaction) compared to before the interaction, resulting in changes in its emission wavelength, emission lifetime, emission intensity, etc. Such luminescent probes may be inorganic or organic. The luminescence mechanism is also not limited, and may be, for example, a compound that emits light upon irradiation with excitation light, or, as shown in a specific embodiment (second embodiment) described below, a compound that emits light due to carrier movement upon application of a voltage. Furthermore, the luminescent probe may have a nucleic acid structure or the like, as shown in a specific embodiment (first embodiment) described below, in order to enhance its interactivity with the target substance.

[0214] Furthermore, in the interaction step, the method for causing the object and the luminescent probe to interact is not particularly limited, and for example, they may be simply mixed, or energy such as heating may be added to cause or promote the interaction.

[0215] In the light detection step, the method for exciting the luminescent probes interacting with the target substance to produce two or more types of luminescence is appropriately selected depending on the type and number of luminescent probes used, the luminescence method of the luminescent probes, etc. For example, different types of luminescent probes are arranged on a plane, and the luminescent probes are made to emit light in a predetermined manner while interacting with the target substance, thereby making it possible to detect the light from each luminescent probe. In this case, arranging a plurality of different types of luminescent probes in a matrix on a plane makes it easier to recognize the luminescence from each luminescent probe. Furthermore, by detecting the luminescence state of the luminescent probes before they interact with the target substance, the amount of data regarding the interaction between the target substance and the luminescent probes increases.

[0216] Furthermore, in the light detection step, the method for simultaneously detecting two or more types of light emitted by the luminescent probe is not particularly limited, and may be, for example, detection using a spectrophotometer or a spectral radiance measuring device. Images may also be captured using a CCD camera or CMOS camera, and chromaticity, brightness, and the like may be determined from the image or video in the analysis step described below. Furthermore, the light detection step may continuously or intermittently detect changes over time in the light emitted by the luminescent probe. Obtaining changes over time in light increases the amount of data regarding the interaction between the target object and the luminescent probe.

[0217] In the analysis step, the method for analyzing the light information acquired in the light detection step is not particularly limited, and for example, standard data for an object in an ideal state may be created in advance, and the standard data may be compared with the data detected in the light detection step to identify the state of the object. Furthermore, the light information acquired in the light detection step may be analyzed using a pre-constructed trained model or the like.

[0218] The following two modes are given as specific examples of methods for carrying out the analysis method of this embodiment: However, the analysis method of this embodiment is not limited to these methods.

[0219] (1) First Aspect One of the features of the analysis method of the first aspect is the use of a molecular luminescent probe having a light-emitting moiety that emits light upon excitation by irradiation with excitation light. A flow diagram of the analysis method of this aspect is shown in Figure 2. This aspect at least involves the steps of preparing one or more types of luminescent probes that emit light upon excitation by irradiation with excitation light and whose luminescence behavior changes upon interaction with the target (S101, the luminescent probe preparation step described above), allowing the target and the luminescent probe to interact (S103, the interaction step described above), simultaneously detecting two or more types of luminescence emitted by the luminescent probe that has interacted with the target (S104, the light detection step described above), and analyzing the two or more types of luminescence detected to indirectly identify the state of the target (S105, the analysis step described above).

[0220] In the following description, an example will be given in which, in addition to performing the luminescent probe preparation step S101, a step of exciting a luminescent probe that is not interacting with the target and acquiring luminescence information (S102, hereinafter also referred to as a "blank information acquisition step") is further performed between the luminescent probe preparation step S101 and the interaction step S103. Performing the blank information step S102 makes it possible to compare the light information detected in the light detection step S104 with the blank information acquired in the blank information acquisition step S102, thereby further increasing the amount of data related to the target. However, the blank information acquisition step S102 does not necessarily have to be performed. Furthermore, the method may further include steps other than those described above, provided that the purpose and effect of this embodiment are not impaired. Each step will be described below.

[0221] (Luminescent probe preparation step) In the luminescent probe preparation step, one or more types of luminescent probes (compounds) are prepared that are excited to emit light when irradiated with excitation light, and whose luminescence behavior changes upon interaction with a target substance. The luminescent probes prepared in this embodiment are compounds that have one or more luminescent moieties that are excited to emit light when irradiated with excitation light. The luminescent probe may have a plurality of the above-mentioned luminescent moieties. Note that when one luminescent probe (molecule) has a plurality of luminescent moieties, the types of the plurality of luminescent moieties may be the same or different.

[0222] Furthermore, the light emitted by a luminescent probe excited by irradiation with excitation light may be light emitted by a single luminescent moiety in the luminescent probe alone, or may be light emitted by the interaction of multiple luminescent moieties in the luminescent probe. Examples of light emitted by luminescent probes include fluorescence, phosphorescence, excimer luminescence, exciplex luminescence, thermally activated delayed fluorescence, excited-state intramolecular proton luminescence, triplet triplet annihilation luminescence, twisted intramolecular charge transfer luminescence, and aggregation-induced luminescence.

[0223] The size of each light-emitting portion is preferably 100 nm or less, and more preferably 0.1 nm or more and 100 nm or less. When the size of the light-emitting portion is 100 nm or less, it becomes easier to detect changes in the nanometer size of the object. The size of the light-emitting portion can be determined using atomic / molecular scale structure modeling / visualization software (e.g., Winmostar (developed by CrossAbility)). Specifically, when the group functioning as the light-emitting portion is considered to be an independent molecule, the length of the cylinder with the smallest diameter in which the molecule is inscribed can be determined as the size of the structure.

[0224] In cases where one luminescent probe emits multiple types of light, only one type of luminescent probe may be prepared for one target. However, from the perspective of obtaining a large amount of data about the target, it is preferable to prepare multiple types of luminescent probes for one target. More specifically, it is preferable to prepare two or more types of luminescent probes for one target, and it is even more preferable to prepare four or more types of luminescent probes. In this case, it is also preferable that the sizes of the luminescent moieties possessed by the multiple luminescent probes are different from each other. Luminescent moieties with different sizes differ in their molecular structure, atomic type, three-dimensional structure, etc. Therefore, by allowing multiple luminescent probes with luminescent moieties of different sizes to interact with the target, a large amount of data about the target can be obtained.

[0225] Here, the luminescent probe may be a compound capable of producing multiple types of luminescence from a single luminescent probe (hereinafter also referred to as a "multiple luminescent compound"), or a compound capable of producing a single luminescence (hereinafter also referred to as a "single luminescent compound"). These will be described in detail below.

[0226] -Multiple-Emitting Compound (i) Structure A multiple-emitting compound may be any compound capable of emitting multiple types of light in response to a single excitation light. Examples of light emitted by a multiple-emitting compound include fluorescence, phosphorescence, excimer luminescence, exciplex luminescence, thermally activated delayed fluorescence, excited-state intramolecular proton luminescence, triplet triplet annihilation luminescence, twisted intramolecular charge transfer luminescence, and aggregation-induced luminescence. Typically, the compound exhibits two or more types of luminescence selected from the group consisting of these.

[0227] When the multiple light-emitting compounds interact with an object, the structure and electronic state of the light-emitting moieties in the multiple light-emitting compounds change, resulting in complex light-emitting behavior different from that of the multiple light-emitting compounds alone. For example, multiple light-emitting compounds that exhibit three different types of light emission in response to a single excitation light (fluorescence, phosphorescence, and excimer emission) are allowed to interact with an object. As a result, the process by which fluorescence, phosphorescence, and excimer emission occur changes due to the interaction between the object and the multiple light-emitting compounds, and the wavelength and lifetime of each light change. Therefore, complex and large amounts of data can be obtained that combine these lights depending on the structure and state of the object. This complex and large amount of data makes it possible to understand the structure and state of the object in great detail.

[0228] Examples of such multiple light-emitting compounds include compounds having a nucleic acid structure and one or more light-emitting moieties (hereinafter also referred to as "chromophores" or "luminophores") bound to the backbone of the nucleic acid structure. The following description will use this compound as an example. In this specification, the term "nucleic acid structure" refers not only to structures derived from DNA or RNA, but also to structures derived from one or more compounds selected from the group consisting of phosphorothioate oligodeoxynucleotides, 2'-O-(2-methoxy)ethyl-modified nucleic acids, siRNA, cross-linked nucleic acids, peptide nucleic acids, aTNA, SNA, GNA, LNA, and morpholino antisense nucleic acids.

[0229] The multiple luminescent compounds may have luminescent moieties evenly bound to the main chain of the nucleic acid structure. Alternatively, as shown in the schematic diagram of Figure 3, the multiple luminescent compounds may have a region in which the luminescent moieties are bound to the main chain of the nucleic acid structure and contribute to the luminescence of the luminescent probe (hereinafter also referred to as the "signal-generating portion"), and a region in which the luminescent moieties are not bound to the main chain of the nucleic acid structure and do not contribute to the luminescence of the luminescent probe (herein also referred to as the "base portion"). The signal-generating portion is preferably located on the tip side of the multiple luminescent compounds, i.e., on the side that is more likely to come into contact with the target object.

[0230] Examples of multiple luminescent compounds include structures (or molecules) having a main chain with one or more structural units containing a pentose- or hexose-derived sugar structure and a phosphate ester bond bound to the sugar structure, and one or more chromophores or luminophores bound to the sugar structure. In the signal-generating portion of the multiple luminescent compound, it is preferable that 50% or more of the sugar structures bound to the chromophores or luminophores are in the β form. Generally, DNA has a structure in which bases are bound to a main chain (deoxyribose) containing a phosphate ester bond and a deoxyribose-derived structure. However, in natural DNA, all deoxyribose in the main chain is in the β form. Therefore, if 50% or more of the sugar structures bound to the chromophores or luminophores are in the β form, it can be said that the structure is highly similar to naturally occurring substances (targets) such as DNA and RNA. When mixed with a target, such multiple luminescent compounds are less likely to cause steric hindrance and can penetrate into the target or conform to the shape of the target. This allows for more detailed analysis of the target.

[0231] Furthermore, in the signal generating unit of the multiple light-emitting compounds, it is more preferable that 80% or more of the sugar structures to which the chromophores or luminophores are bound are in the β form, and it is even more preferable that all of them are in the β form. Whether the sugar structures to which the chromophores or luminophores are bound are in the β form or the α form can be confirmed by NMR analysis, X-ray crystal structure analysis, or the like. The specific structure of the signal generating unit of the multiple light-emitting compounds is described below.

[0232] The main chain of the signal generating part having the pentose- or hexose-derived sugar structure of the multiple light-emitting compound may have one or more structural units including a pentose- or hexose-derived sugar structure and a phosphate ester bond bonded to the sugar structure. The main chain may include only one of the structural units, or may include multiple structural units. That is, the main chain may have one of the sugar structures and one of the phosphate ester bonds bonded to the sugar structure, or may have a structure including the sugar structure and the phosphate ester bond alternately. Typically, both ends of the main chain of the multiple light-emitting compound are sugar structures, so that the number of sugar structures is one more than the number of phosphate ester bonds. Note that when the main chain includes multiple structural units, the multiple structural units may be the same or different from each other.

[0233] The number of the structural units contained in the main chain of the signal generating portion of the multiple light-emitting compound is selected appropriately depending on the type of target object, etc., but is preferably 2 or more and 6 or less. As the amount of the structural units increases, the multiple light-emitting compounds tend to act specifically on the target object. However, in this embodiment, it is preferable to cause the multiple light-emitting compounds to interact with various positions on the target object to obtain a large amount of data. Therefore, it is preferable that the multiple light-emitting compounds and the target object have appropriate (not excessive) specificity, and it is preferable that the number of the structural units is 6 or less.

[0234] The main chain of the signal generating portion of the multiple light-emitting compounds may partially contain a structure other than the above-mentioned pentose- or hexose-derived sugar structure and structural unit containing a phosphate ester bond, as long as the purpose and effect of this embodiment are not impaired. Furthermore, the structures at both ends of the main chain are not particularly limited and may be various structures, such as an OH group or an alkoxy group.

[0235] Examples of the pentose include ribose, deoxyribose, and xylose. Specific examples of the hexose include allose, glucose, and mannose. Among these, a sugar structure derived from ribose or deoxyribose is particularly preferred, since the main chain of the multiple luminescent compounds has a structure similar to that of DNA or RNA, making it easier to interact with DNA or RNA.

[0236] When the structural unit contains a structure derived from ribose or deoxyribose, the phosphate ester bond is preferably bonded to the 3-carbon and 5-carbon of the ribose or deoxyribose. Furthermore, the chromophore or luminophore described below is preferably bonded to the 1-position of the ribose or deoxyribose. In other words, the multiple light-emitting compound of this embodiment preferably contains a structure represented by the following general formula (1a) or (1b): In the general formulae (1a) and (1b), Y represents a chromophore or luminophore as described below.

[0237] The light-emitting moiety (chromophore or luminophore) possessed by the signal generating unit of the multiple light-emitting compounds may have a structure that emits a predetermined type of light by itself in response to a single excitation light, or that emits a predetermined light through the action of multiple chromophores or luminophores. It is preferable that the chromophore or luminophore is bound to the sugar structure of the main chain so that the sugar structure is in a β-form. In this specification, the term "chromophore" refers to a structure that absorbs light with a wavelength of 300 nm or more, and the term "luminophore" refers to a structure that absorbs light with a wavelength of 300 nm or more and emits light.

[0238] The number of chromophores or luminophores possessed by the signal-generating portion of the multiple light-emitting compound may be only one, as long as the multiple light-emitting compound is capable of exhibiting multiple types of light emission. However, from the viewpoint of facilitating the multiple light-emitting compound to exhibit multiple types of light emission, two or more chromophores or luminophores are preferred, and three to six are even more preferred. When the signal-generating portion of the multiple light-emitting compound has multiple chromophores or luminophores, the types may be only one or two or more. Here, in the multiple light-emitting compound, one chromophore or luminophore is typically bound to one sugar structure in the main chain. Therefore, when the multiple light-emitting compound has two or more chromophores or luminophores, it is preferable that the main chain also has two or more sugar structures. In other words, it is preferable that the number of chromophores or luminophores in the multiple light-emitting compound is the same as or less than the number of sugar structures (or peptide structures) in the main chain of the signal-generating portion.

[0239] In addition, when the number of chromophores or luminophores in the signal generating portion in the multiple light-emitting compound is less than the number of sugar structures (or peptide structures) in the main chain, some sugar structures will not have a chromophore or luminophore bound to them. Other atomic groups may not be bound to sugar structures to which a chromophore or luminophore is not bound. However, natural nucleobases may be bound to the extent that the purpose and effect of this embodiment are not impaired. In this specification, natural nucleobases refer to adenine, guanine, cytosine, thymine, and uracil. However, the total number of natural nucleobases bound to the main chain in the multiple light-emitting compound is preferably 50% or less, and more preferably 25% or less, of the total number of structural units constituting the main chain structure of the region (signal generating portion) that mainly controls the interaction. This description will be given using the multiple light-emitting compound shown in Figure 3 as an example, which has a signal generating portion (the portion that interacts with the target substance) and a base (the portion that contributes little to the interaction), and the signal generating portion is composed of the first 10 sugar chains from the tip. In this case, the number of natural nucleobases bound to the sugar chain in the first ten sugar chains (signal generating unit) from the tip is preferably 50% or less, more preferably 25% or less, of the total number of sugar structures contained in the sugar chain. When the number of natural nucleobases in the signal generating unit is 50% or less, association between multiple light-emitting compounds is suppressed, and the interaction between the target substance and multiple light-emitting compounds tends to be dominant. However, when the target substance has a large molecular size such as a protein or liposome, the signal generating unit may extend from the tip to the sugar chains beyond the tenth sugar chain. In this case, a structure in which multiple natural bases are linked to the sugar chains beyond the tenth sugar chain from the tip is also preferable from the viewpoint of overall affinity. In this case, the number of natural bases bound to the signal generating unit is preferably about 10 to 50 bases in length, more preferably between 15 and 30 bases. Furthermore, in the signal generating unit, it is preferable that the natural nucleobases and unnatural nucleobases are bound so that the sugar structure is in a β-form, as described above. The structure of the base is not limited to the above and may have various structures.

[0240] Here, examples of chromophores or luminophores that emit fluorescence include structures derived from fluorescein, rhodamine, boron dipyrromethene, etc. Examples of chromophores or luminophores that emit phosphorescence include structures derived from iridium complexes, platinum complexes, etc. Examples of chromophores or luminophores that emit excimer emission include structures derived from pyrene, anthracene, perylene, etc. Examples of chromophores or luminophores that emit exciplex emission include structures derived from pyrene-dimethylaniline, etc. Examples of chromophores or luminophores that emit thermally activated delayed fluorescence include structures derived from 4CzIPN, DABNA, etc. Examples of chromophores or luminophores that emit excited-state intramolecular proton emission include structures derived from hydroxyphenylbenzoxazole, etc. Examples of chromophores or luminophores that emit triplet triplet annihilation emission include structures derived from 9,10-diphenylanthracene, rubrene, etc. Examples of chromophores or lumophores that emit twisted intramolecular charge transfer luminescence include structures derived from diaminoanthracene, diaminonaphthalene, etc. Examples of chromophores or lumophores that emit aggregate organic luminescence include structures derived from tetraphenylethene, hexaphenylsilole, etc.

[0241] Furthermore, compounds used as light-emitting materials or hosts, electron transport materials, hole transport materials, or light-emitting materials in organic electroluminescence (EL) can also be suitably used as the chromophore or luminophore materials. Specific examples of such compounds include those described in "State-of-the-art Organic EL" (CMC Publishing), "Organic EL Material Technology" (CMC Publishing), "All About Organic EL" (Nippon Jitsugyo Publishing), and "Various Dye Materials Pioneering the Future" (Kagaku Dojinsha).

[0242] The reasons why various organic EL compounds are suitable as chromophores or luminophores for the signal generation components of the above-mentioned multiple light-emitting compounds are as follows. For example, electron transport materials contain electron-accepting aromatic compounds that readily form anion radicals, thus exhibiting strong interactions with electron-rich compounds in the target. Conversely, hole transport materials contain electron-donating aromatic compounds that readily form cation radicals, thus exhibiting strong interactions with electron-deficient compounds in the target. Organic EL light-emitting materials possess both of these properties and also have high luminescence quantum yields, thereby producing strong luminescence signals. Phosphorescent and delayed fluorescent materials can also be used. These emit light with a time delay of nanoseconds to microseconds compared to conventional fluorescent emission. Therefore, as sensing materials, the time factor can be included in the number of dimensions. This makes them suitable for multidimensional data generation for use in machine learning and deep learning.

[0243] Representative examples of organic EL materials are listed below. These are molecular groups that can be Y in the above general formulas (1a) and (1b). Furthermore, Y may be formed via a linking group or the like.

[0244]

[0245]

[0246]

[0247] Further, examples of chromophores or luminophores other than those mentioned above include the following:

[0248] Furthermore, in this embodiment, the multiple light-emitting compounds may contain structures derived from non-light-emitting monomers as well as structures that exert various functions as sites for controlling the interaction between the multiple light-emitting compounds and the target object. For example, structures derived from multiple light-emitting compounds may coexist in a molecular chain. Examples of representative functions and monomer structures that realize them are shown below.

[0249] A1) Compounds Capturing Metal Ions Representative examples include compounds having a ligand structure that forms a chelate, and coordination types include N,N coordination, N,O coordination, O,O coordination, N,S coordination, O,S coordination, S,S coordination, N,Se coordination, O,Se coordination, S,Se coordination, Se,Se coordination, etc. Specific examples include 2,2'-bipyridine, phenanthroline, 1,8-diaminonaphthalene, amino acids, cryptands, crown ethers, 8-hydroxyquinoline, 3-mercaptopropanol, 3-mercaptopropionic acid, thiocatechol, salicylaldehyde, acetoacetic esters, β-diketones, etc.

[0250] A2) Compounds whose molecular shape changes depending on the wavelength of irradiated light. Representative examples include azobenzene, stilbene, fulgide, diarylethene, spiropyran, spirooxazine, dihydropyrene, phenoxyquinone, biindenylidenedione, cobalt complexes, and imidazole dimers.

[0251] A3) Compounds that act as Lewis acids Representative examples include Group 13 elements and transition metals, and specific examples include triarylborane, trialkylborane, trialkoxyborane, trifluoroborane, trichloroborane, and tribromoborane.

[0252] A4) Compounds that act as Lewis bases Representative examples include Group 15 and 16 elements, and specific examples include arylamines, alkylamines, arylphosphines, alkylphosphines, aryl ethers, alkyl ethers, aryl sulfides, and alkyl sulfides.

[0253] A5) Compounds that interact with peptides or proteins Representative examples include nucleic acid bases, transition metal complexes, and oxoacids, and specific examples include adenine, thymine, guanine, thymine, cytosine, zinc complexes, copper complexes, nickel complexes, cobalt complexes, tungstic acid, molybdic acid, and phosphate.

[0254] A6) Compounds that form hydrophobic-hydrophobic interactions Representative examples include straight-chain alkanes, straight-chain alkenes, straight-chain alkynes, branched alkanes, branched alkenes, branched alkynes, aromatic rings, heteroaromatic rings, and the like.

[0255] A7) Compounds that form dipole-dipole or quadrupole-quadrupole interactions Representative examples include alkyl halides, aryl halides, nitriles, nitroarenes, anilines, anisoles, heterocyclic compounds, and the like.

[0256] In this embodiment, the light-emitting compound preferably contains, as a chromophore or luminophore, at least one structure selected from a structure that emits fluorescence, a structure that emits excimer emission, and a structure that emits exciplex emission. Among these, it is preferable that the light-emitting compound contains at least a structure that emits fluorescence. When the light-emitting compound emits fluorescence, this has the advantage of being easy to analyze using various measuring devices.

[0257] Furthermore, the multiple light-emitting compounds preferably exhibit multiple types of light emission when irradiated with light having a wavelength of 300 to 400 nm. If the multiple light-emitting compounds exhibit multiple types of light emission when irradiated with light having such wavelengths, a special light source is not required when analyzing the target object, and the target object is less likely to be damaged.

[0258] However, when an LED or an organic EL element is used as an excitation light source, excitation in the visible light region is advantageous. In such a case, the absorption wavelength of the above-mentioned multiple light-emitting compounds is effective in the range of 400 to 700 nm, and such a dye can also be used.

[0259] The molecular weight of the multiple light-emitting compounds (when the multiple light-emitting compounds have a signal-generating portion and a base portion, the molecular weight of the signal-generating portion) is appropriately selected depending on the types of chromophores and luminophores possessed by the multiple light-emitting compounds, the length of the main chain, etc. Generally, a molecular weight of 500 or more and 10,000 or less is preferable, and a molecular weight of 500 or more and 4,000 or less is more preferable. When the molecular weight of the multiple light-emitting compounds is 10,000 or less, the specificity for the target substance is appropriately low, making it possible to cause the compounds to react nonspecifically with multiple locations on the target substance.

[0260] In the above description, the structural unit constituting the main chain of the nucleic acid structure of the signal generating part has a structural unit containing a sugar structure derived from pentose or hexose, such as DNA or RNA, and a phosphate ester bond, but as mentioned above, it is not limited to these. Representative examples of other structural units include peptide nucleic acid type structural units such as those shown below. Like DNA / RNA, peptide nucleic acids can be comprehensively synthesized using commercially available automated synthesizers (peptide synthesizers). Because peptide nucleic acids have no charge and no electrostatic repulsion, they can form stronger aggregates with target substances. Furthermore, they are resistant to enzymes such as nucleases and proteases, making them suitable for use in cells. Furthermore, they can be synthesized on a relatively large scale. However, due to their nonionic structure, they may aggregate in water, reducing their solubility. Therefore, when the signal generating unit has a structure derived from peptide nucleic acids, it is preferable to appropriately select the type of base linked to the signal generating unit and the solvent.

[0261] (ii) Manufacturing Method: The manufacturing method for the above-mentioned multiple light-emitting compound is appropriately selected depending on the type of nucleic acid structure in the multiple light-emitting compound. For example, multiple light-emitting compounds having the above-mentioned sugar structure can be manufactured by the following method. A monomer is prepared by binding the above-mentioned chromophore or luminophore and a phosphate ester to a pentose or hexose. The monomer can be synthesized by polymerizing the desired sequence using the phosphoramidite method using a DNA / RNA synthesizer or the like. According to this method, as shown in the schematic diagram of FIG. 4, for example, multiple types of monomers (three types in FIG. 4) with different types of chromophores or luminophores (represented by A, B, and C in FIG. 4) can be prepared, and the sequence order of the monomers can be changed to allow the desired number of bonds (three bonds in FIG. 4). In other words, a wide variety of multiple light-emitting compounds can be synthesized from multiple types of monomers with different types of chromophores or luminophores. In the example shown in FIG. 4, 27 different multiple light-emitting compounds can be synthesized. By changing the type of monomer used and the number of monomers bonded, a very large number of multiple light-emitting compounds can be synthesized.

[0262] In addition, when performing the phosphoramidite method, a portion of the monomer, such as a hydroxyl group of a sugar structure (e.g., a hydroxyl group bonded to the carbon-3 position of ribose or deoxyribose) or a hydroxyl group derived from phosphoric acid, is typically supported on a particulate carrier (also referred to herein as "carrier particles") and polymerized. The carrier particles are, for example, porous glass, porous silica gel, or polystyrene. Porous glass includes porous metal oxides such as silica gel and alumina. After synthesis of the multiple luminescent compounds (polymerization of the monomers), the carrier may be removed to obtain only the multiple luminescent compounds, or the multiple luminescent compounds may be used in a state supported on carrier particles. Furthermore, the multiple luminescent compounds may be mixed with a liquid to form an ink.

[0263] (iii) Effects As described above, the multiple luminescent compounds have a main chain structure similar to that of substances found in nature (e.g., DNA and RNA). Therefore, they can easily interact with various objects, and the state of the object can be grasped in detail using the multiple luminescent compounds. Furthermore, the multiple luminescent compounds exhibit multiple types of luminescence when irradiated with light of specific wavelengths. Therefore, it is possible to obtain a large amount of complex luminescence data depending on the state of the object, and to analyze the object in great detail.

[0264] The mechanism by which this effect manifests is shown schematically in Figures 5A to 5D. First, to facilitate understanding, the multi-luminescent compound is defined as a molecule having four sugar structures and four luminophores, as shown in Figure 5A. All four R's may be pyrene (Py in Figure 5), dimethylaminobiphenyl (N in Figure 5), or a mixture of these. Furthermore, one or two of the four R's may be hydrogen atoms. This type of molecular structure allows the construction of a composite multi-luminescent compound.

[0265] This embodiment is characterized by the fact that the target liquid, dispersion, or gaseous substance causes complex interactions with multiple light-emitting compounds or their carriers, generating multidimensional and large amounts of data as light or color signals. Figures 5B to 5D show this concept in its simplest and most concrete form, and there are three main aims envisioned.

[0266] For example, assuming that all of the R's are pyrenes, when a specimen is brought into contact with a liquid or medium containing the multiple light-emitting compounds and excited with ultraviolet light, if there is an object (such as an aliphatic compound) that is contained in the specimen and inserted between the pyrenes present in the multiple light-emitting compounds and has a wide band gap that does not quench the fluorescence of pyrene, then monomeric emission of pyrene as shown in the left diagram of Figure 5B will be observed from the multiple light-emitting compounds.

[0267] If the sample does not contain any components that interact with pyrene, excimer emission of pyrene is observed, as shown in the center of Figure 5B. If the sample contains a target substance (fluorescent substance) that interacts with pyrene and has an energy level close to the band gap of pyrene, exciplex emission from pyrene and the target substance (fluorescent substance) is observed, as shown in the right of Figure 5B.

[0268] Furthermore, when the specimen contains metal ions such as sodium ions or calcium ions, they form chelates with the phosphate groups present in the main chains of the multiple luminescent compounds. As shown in Figure 5D, the intermolecular distance between pyrene molecules changes depending on the size of the metal ions, and the emission color (emission spectrum) of the excimer emission itself also changes.

[0269] Furthermore, when R in the structure shown in Figure 5A is dimethylaminobiphenyl (N), the proximity of an acid and a base causes a change in the emission color (spectrum) of the fluorescent dye (N), as shown in Figure 5C. This is different from the on / off switching of acid-base ion pairs, such as those formed between mineral acids (such as sulfuric acid and nitric acid) and alkali metals. In this case, the proximity distance to N changes continuously depending on the acidity (ease of proton donation) of the substance contained in the analyte. Therefore, using this emission phenomenon as a signal leads to an expansion of the dynamic range. Furthermore, because N is a Lewis base, it also interacts with Lewis acidic substances (e.g., triarylboranes, trialkylaluminums, and tetraalkoxytitaniums) in addition to protic acidic substances. Therefore, specific emission color changes occur when such substances are present in the analyte.

[0270] Next, when every other R in the multiple luminescent compound is pyrene (Py) and dimethylaminobiphenyl (N), exciplex luminescence between pyrene and dimethylaminobiphenyl is observed if there is no interaction between the analyte and pyrene or dimethylaminobiphenyl, whereas complex luminescence due to mixed exciplexes between the analyte component and pyrene and / or dimethylaminobiphenyl is observed if they interact, as in the case described above.

[0271] Furthermore, when one or two of the four inner R's are hydrogen atoms, excimer or exciplex emission does not occur from the luminescent dye molecule itself, or the contribution of such emission is small, and monomer emission is observed almost entirely. However, since the steric hindrance of the hydrogen atoms is small, the interaction between the target substance in the sample and the R's of the luminophore is enhanced, and the change in the luminescence signal is enhanced.

[0272] 5A is not a normal fluorescent substance, but a phosphorescent compound or a thermally excited delayed fluorescent compound, light emission occurs several tens of nanoseconds to several microseconds after excitation. Therefore, the dynamic range is expanded by the factor of "time" rather than the color of the emitted light, and such a phenomenon can be applied to the present invention in addition to the mechanisms shown in FIGS. 5B to 5D.

[0273] Such intermolecular interactions and the resulting subtle changes in luminescence color and spectrum, as well as the resulting microsecond delay in luminescence, and even the extremely subtle changes in luminescence phenomena caused by metal chelate formation due to the main chain structure, have until now been inapplicable as analytical information for human understanding. However, by utilizing artificial intelligence (AI), which has become widely available in recent years, and the machine learning and informatics that utilize it, these diverse luminescence phenomena beyond human comprehension can be converted into state-description data corresponding to the target analyte. This new concept is the fundamental concept of the present invention and will be extremely useful as a new method for describing the state of complex analytes in various research and development and production processes, as well as cell culture and wastewater / sludge treatment.

[0274] Single luminescent compounds On the other hand, single luminescent compounds are compounds that emit a single light in response to a single excitation light, and their luminescent state changes depending on the interaction with the target. Examples of single luminescent compounds include substances that hydrogen bond with the target and substances that interact with the target through π-π interaction. Single luminescent compounds may be any of inorganic substances, organic substances, metal complexes, etc. Examples of inorganic substances that can be used as single luminescent compounds include YAG phosphors, quantum dot phosphors, and VO ２ Examples of organic substances that can be used as the single light-emitting compound include fluorescent materials such as coumarin and perylene, and examples of metal complexes that can be used as the single light-emitting compound include phosphorescent materials such as Eu (europium) complexes and Ir (iridium) complexes.

[0275] In this embodiment, by combining two or more types of such single compounds, it becomes possible to obtain two or more types of light emitted by the luminescent probe in the light detection step described below.

[0276] (Blank Information Acquisition Step) In the blank information acquisition step of the analysis method of this embodiment, the luminescent probes prepared in the luminescent probe preparation step are excited and information regarding their luminescence (blank information) is acquired. When multiple luminescent probes are used, blank information may be acquired for each individual luminescent probe, but it is preferable in terms of efficiency and the like to simultaneously acquire blank information for multiple luminescent probes. For example, a plate having multiple storage compartments for allowing the target and the luminescent probes to interact may be prepared, with multiple types of luminescent probes individually placed in each storage compartment. Use of this plate allows multiple luminescent probes to be simultaneously irradiated with excitation light, and the light emitted by these probes to be simultaneously detected. Furthermore, use of this plate allows an additional target to be placed in the location (storage compartment) where the luminescent probes were placed in this step in the interaction step described below, making it easy to cause each luminescent probe to interact with the target.

[0277] The plurality of storage sections of the plate are preferably arranged at intervals. The plate may be flat or may have projections and recesses depending on the shape of the storage sections. The material, size, shape, etc. of the plate are appropriately selected depending on the purpose of analysis, the type of luminescent probe and the target object, etc.

[0278] The spacing between the multiple storage units is selected appropriately depending on the size of the storage units and the types of luminescent probes and targets. When the luminescent probes or targets are dropped onto the plate using an inkjet device or the like, the plate does not need to be provided with marks (relief structures or markings) indicating the position of each storage unit. On the other hand, if the plate has marks (relief structures or markings) indicating the position of each storage unit, it becomes easier to accurately place the luminescent probes or targets in the desired positions (storage units).

[0279] Furthermore, if each storage section is formed in a concave shape or if a partition section is arranged around each storage section, the luminescent probes and targets placed in adjacent storage sections are less likely to mix, making it easier to perform more accurate analysis. Furthermore, for example, if a water-repellent treatment section is arranged around the storage section, the luminescent probes and targets in adjacent storage sections are less likely to mix, making it easier to perform more accurate analysis. In this embodiment, a plate with multiple wells arranged in a regular pattern is used. In a plate with such wells, the wells (storage sections) are physically separated by partitions, making it harder for the luminescent probes and targets in adjacent storage sections to mix, making it easier to perform accurate analysis.

[0280] Here, the number of the storage sections in one plate is appropriately selected depending on the type of luminescent probe, the target object, etc. The number of storage sections is not particularly limited, but the more storage sections there are, the more multidimensional data can be acquired and the more precise the analysis can be performed.

[0281] Furthermore, the method for disposing the luminescent probe in each storage section is not particularly limited and may be selected appropriately depending on the type and physical properties of the luminescent probe. Examples of methods for disposing the luminescent probe include application using an inkjet device, application using a dispenser, disposing a carrier that carries the luminescent probe, and directly fixing the luminescent probe to the storage section. Among these, the inkjet method is particularly preferred. With the inkjet method, it is possible to efficiently dispose liquid luminescent probes in a large number of regions (storage sections) and form the storage sections. This makes it possible to acquire a large amount of data.

[0282] In this embodiment, different types of luminescent probes are placed in different storage compartments, but multiple types of luminescent probes may also be placed in one storage compartment.

[0283] Furthermore, there are no particular limitations on the method for obtaining the blank information from a plate in which luminescent probes are placed in the storage section. For example, specific excitation light (single-wavelength excitation light) may be irradiated, and the intensity and wavelength of light emitted by each luminescent probe (luminescence information) may be measured using a spectrophotometer or the like, and this may be used as blank information. Furthermore, when specific excitation light is irradiated, the change over time in the spectral distribution of light emitted by the luminescent probes or the change in chromaticity over time may be measured continuously or intermittently using a spectrophotometer or the like, and this may be used as blank information.

[0284] Furthermore, when acquiring the change in chromaticity of the light emitted by the luminescent probe, excitation light of a single wavelength may be irradiated for only a short time, and an image of the light emitted by the luminescent probe in response to this may be acquired using a CCD camera, CMOS camera, etc. By identifying the chromaticity, etc. from the acquired image, data on the change in chromaticity over time may be acquired.

[0285] The wavelength of the excitation light is appropriately selected depending on the type of luminescent probe and the target object, and can be, for example, light with a wavelength of 300 nm or more and 700 nm or less. Such light is unlikely to affect the target object. Furthermore, no special light source is required.

[0286] (Interaction step) In the interaction step, the target substance and the luminescent probe are mixed and allowed to interact with each other. In the blank information acquisition step described above, when a plate having multiple storage sections is used, the target substance can be placed in each storage section (storage section containing a luminescent probe) of the plate from which the blank information was acquired. Note that targets may also be placed in storage sections in which the luminescent probe is not placed. Furthermore, when the plate described above has multiple storage sections, different targets may be placed in some of the storage sections, and multiple targets may be analyzed on a single plate. On the other hand, the same target substance may be placed in all of the storage sections.

[0287] The method for disposing the target object is not particularly limited and can be selected appropriately depending on the type and properties of the target object. This method can be the same as the method for disposing the luminescent probe described above. Furthermore, the method for causing the target object and the luminescent probe to interact is not particularly limited, and the target object and the luminescent probe may be simply mixed, but heating or other treatments may also be performed as necessary.

[0288] (Luminescence information acquisition process) In the luminescence information process, excitation light is irradiated onto the object and the luminescent probes in a state where they are interacting with each other, and two or more types of light emitted by the luminescent probes are simultaneously detected. The method for simultaneously detecting two or more types of light is not particularly limited, and for example, specific excitation light (excitation light of a single wavelength) may be irradiated, and the intensity and wavelength of light emitted by each luminescent probe (luminescence information) may be detected using a spectrophotometer or the like. Furthermore, when specific excitation light is irradiated, changes over time in the spectral distribution of light emitted by the luminescent probes or changes in chromaticity over time may be detected continuously or intermittently using a spectrophotometer or the like.

[0289] Furthermore, when detecting changes in the chromaticity of light emitted by a luminescent probe, excitation light of a single wavelength may be irradiated for only a short time, and an image of the light emitted by the luminescent probe in response to this may be captured using a CCD camera, CMOS camera, etc. By identifying the chromaticity, etc. from the captured image, data relating to changes in chromaticity over time can be obtained.

[0290] (Analysis Step) The analysis method in this step is appropriately selected depending on the purpose, the type of information related to the luminescence acquired in the light detection step, and the like. For example, the data detected in the light detection step may be used as the analysis data as is. Alternatively, the difference between the blank information acquired in the blank information acquisition step and the information related to the luminescence acquired in the light detection step may be used as the analysis data. In this step, for example, standard data for an object in an ideal state may be prepared in advance, and the state of the object may be identified by comparing the standard data with the analysis data. According to this method, even when the object is composed of multiple components or when multiple parameters are involved (e.g., food quality or taste), the state of the object can be identified without analyzing each component of the object by creating standard data for when the object is in a good state and when the object is in a bad state.

[0291] Here, when performing this analysis, the standard data and the analytical data may be simply compared. Alternatively, the comparison result between the standard data and the analytical data may be converted into a distance matrix and analyzed using a heat map (unweighted). Furthermore, the distance matrix may be subjected to principal component analysis (also known as PCA, which weights with emphasis on anisotropy) or DL ​​analysis (weights with emphasis on isotropy).

[0292] On the other hand, the analysis data may be analyzed using a pre-constructed trained model. The trained model can be created, for example, by a machine learning process described below, but the trained model used is not limited to one created by the machine learning process described below. Using a trained model enables more appropriate analysis of the target object.

[0293] When referencing a trained model, the analysis data described above is applied to the trained model. This makes it possible to determine (predict) from accumulated data whether the object has a desired structure, how much of a predetermined structure the object contains, whether the object is in good condition, etc. The prediction results may be obtained, for example, as classification, regression, clustering, anomaly detection (outlier detection), etc.

[0294] (Machine Learning Step) The analysis method of the present embodiment may further include a learning step of performing machine learning on the analysis data to generate a trained model.

[0295] For example, in the machine learning process, multiple prediction models may be constructed based on the differences between the analysis data and the blank information, and the results of the multiple prediction models may be combined to create a trained model that can predict information about the target object (e.g., structure, quantity, etc.).

[0296] When the structure and quantity of the target object are known in advance, the above-mentioned prediction model can be constructed by performing machine learning using the characteristics of the analysis data as explanatory variables and the structure and quantity of the target object as objective variables. As explanatory variables, numerical values ​​representing the characteristics of the above-mentioned analysis data and numerical values ​​calculated from them can be used. When the blank information and the luminescence information acquired in the light detection process are spectral distributions, the intensity of light for each wavelength can be used as the explanatory variable. On the other hand, the objective variable can be selected appropriately depending on the purpose of the analysis, and is not limited to the structure and quantity of the target object, and any other variable related to the target object may also be used.

[0297] The machine learning performed in this process may be either supervised learning or unsupervised learning. Note that supervised learning is a learning method that learns the "relationship between input and output" from training data with correct answer labels. Unsupervised learning is a learning method that learns the "structure of a data group" from training data without correct answer labels.

[0298] Machine learning may be reinforcement learning, deep learning, or deep reinforcement learning. Reinforcement learning is a learning method that learns an "optimal sequence of actions" through trial and error. Deep learning is a learning method that learns features contained in data in a stepwise and deeper manner (at a deeper level) from a large amount of data. Deep reinforcement learning is a learning method that combines reinforcement learning and deep learning.

[0299] Common analytical methods (algorithms) can be applied to machine learning, such as a predictive model constructed by an analytical method selected from linear regression (multiple regression analysis, partial least squares (PLS) regression, LASSO regression, Ridge regression, principal component regression (PCR)), random forest, decision tree, support vector machine (SVM), support vector regression (SVR), neural network, discriminant analysis, and the like.

[0300] (Modifications) In the above, an analysis was described in which a plate having a plurality of storage compartments for allowing the object and the luminescent probe to interact with each other is used, and the object and the luminescent probe are placed in the storage compartments. However, instead of the luminescent probe, a carrier carrying the above-mentioned luminescent probe, or ink in which the luminescent probe is dissolved or dispersed in a liquid, may be used. Furthermore, a cell for measuring fluorescence photometric activity may be used instead of the plate. In this case, the blank information acquisition step and the light detection step may be performed using a known fluorescence measurement device. Furthermore, when using the above-mentioned ink, a desired substrate (e.g., paper) may be used instead of the plate.

[0301] -Analysis device The above-mentioned analysis method can be performed by an analysis device having a holding section (plate) including a plurality of storage sections for storing the object and the above-mentioned luminescent probes, an excitation section (light source) for exciting the luminescent section arranged in the storage section, a detection section for detecting two or more types of light emitted by the luminescent probes, and an analysis section for analyzing the detected light.

[0302] The analysis device of this embodiment may further include components other than the holding unit, excitation unit, detection means, and analysis unit. For example, it may further include an inkjet printing unit. Note that the plate is the same as that described in the analysis method above, so a detailed description thereof will be omitted here.

[0303] On the other hand, the excitation unit can be, for example, an excitation light source for irradiating excitation light, examples of which include a supercontinuum light source (a broadband pulse light source that utilizes the nonlinear effect of optical fiber to emit strong light with a coherent phase over an extremely wide wavelength range, also known as an "SC light source"), an LED, etc. These light sources can increase the amount of light, making it easier to clarify the emitted light information.

[0304] The detection unit is configured to simultaneously detect two or more types of light emitted by the luminescent probe, and the type of detection unit is appropriately selected depending on the information of the two types of light detected by the detection unit. For example, the detection unit may be a spectrophotometer, an ultraviolet-visible light absorptiometer, or a fluorescent fingerprint measurement device. It may also be a circular dichroism spectrometer, a high-performance liquid chromatograph (HPLC), or the like. It may also be a CCD camera, a CMOS camera, or the like. The detection unit may also be a combination of two or more of these.

[0305] The analysis unit that analyzes the detected light can be a general information processing device, such as a personal computer. Such an analysis unit can be a general computer (general-purpose computer) equipped with a storage means, such as a hard disk drive (HDD), a solid-state drive (SSD), or a read-only memory (ROM), for storing programs and data, and a central processing unit (CPU) for executing programs and performing calculations. The computer may further include input means, such as a keyboard or a mouse, and output means, such as a monitor or a printer.

[0306] (2) Second Aspect One of the features of the analysis method of this aspect is the use of a specific analysis device. An example of an analysis device 100 used in the analysis method of this aspect is shown in FIG. 6 . The analysis device 100 includes a pair of electrodes 1 and 2 (excitation unit), a receptive layer (retaining unit) 3 sandwiched between the pair of electrodes 1 and 2 for holding the target object 5 and the luminescent probe 6 in a predetermined position, and a detection unit (not shown) for detecting light emitted by the luminescent probe 6. In the analysis device 100, a voltage is applied between the pair of electrodes 1 and 2 to supply carriers (holes and electrons) to the luminescent probe in the receptive layer 3, causing the luminescent probe to emit light. In the analysis device 100, one or both of the electrodes 1 and 2 are configured to transmit light. This makes it possible to detect the light emitted by the luminescent probe 6. Furthermore, in the analysis device 100, two or more types of light can be generated by holding different luminescent probes 6 in the holding unit 4 of the receptive layer 3. That is, in the analysis device 100 , a region (light-emitting site) where the object 5 and the luminescent probe 6 interact is formed on the surface or inside of the receptive layer 3 .

[0307] In an analysis method using the analysis device 100, first, one or more types of luminescent probes 6 are prepared, which emit light when a voltage is applied and whose luminescence behavior changes upon interaction with the target object (luminescent probe preparation step). A separate receptor layer (retaining section) 3 is prepared, and the target object 5 and luminescent probe 6 are respectively accommodated in the accommodation section 4 of the receptor layer 3 to form the above-mentioned luminescent site, and electrodes 1 and 2 are placed on both sides of the receptor layer 3 (the above-mentioned interaction step). Then, a voltage is applied between the pair of electrodes 1 and 2 to cause the luminescent probe 6 to emit light, and two or more types of luminescence are simultaneously detected by a detection section (not shown) (the above-mentioned light detection step). The two or more types of luminescence detected are then analyzed to indirectly identify the state of the target object (analysis step). Each step of the method is described below.

[0308] (Luminescent probe preparation step) In the luminescent probe preparation step, one or more types of luminescent probes are prepared that emit light upon application of voltage and whose luminescence behavior changes upon interaction with the target substance. Known fluorescent compounds, delayed fluorescent compounds, phosphorescent compounds, etc. can be used as the luminescent probe. Multiple types of luminescent probes may be used in combination, for example, different phosphorescent compounds or a phosphorescent compound and a fluorescent compound. This allows for the production of any luminescent color.

[0309] In this step, multiple luminescent probes with different luminescent colors may be prepared. The color emitted by the luminescent probe is determined by applying the results of measurement using a spectroradiometer CS-1000 (manufactured by Konica Minolta) to CIE chromaticity coordinates in Figure 3.16 on page 108 of "New Color Science Handbook" (edited by the Color Science Association of Japan, University of Tokyo Press, 1985).

[0310] In this step, only two types of luminescent probes may be prepared for one object, but from the viewpoint of obtaining a large amount of data about the object, it is preferable to prepare a large number of luminescent probes for one object. More specifically, it is preferable to prepare two or more types of luminescent probes for one object, and it is more preferable to prepare four or more types of luminescent probes.

[0311] Fluorescent Compound In the present invention, the term "fluorescent compound" refers to a compound that emits fluorescence other than delayed fluorescence. For example, those described in JP-A-2022-99677, JP-A-2022-116901, and WO 2022 / 163199 can be similarly utilized.

[0312] Phosphorescent Compound The term "phosphorescent compound" in the present invention refers to a compound that emits phosphorescence. For example, those described in JP-A-2022-99677, JP-A-2022-116901, and WO 2022163199 can be similarly utilized.

[0313] Examples of phosphorescent compounds that can be suitably used in this embodiment are shown below. The phosphorescent compounds described in the Examples below (e.g., blue phosphorescent probe Ep1, red phosphorescent probe Ep2, etc.) can also be suitably used in this embodiment.

[0314]

[0315]

[0316] Delayed Fluorescent Compound In this specification, the term "delayed fluorescent compound" refers to a compound that emits delayed fluorescence. For example, those described in JP-A-2022-99677, JP-A-2022-116901, and WO 2022 / 163199 can be similarly utilized.

[0317] A thermally activated delayed fluorescent compound will be given below as an example, but the present embodiment is not limited thereto.

[0318]

[0319]

[0320]

[0321] Triplet-triplet annihilation delayed fluorescent compound The term "triplet-triplet annihilation delayed fluorescent compound" as used herein refers to a compound that emits triplet-triplet annihilation delayed fluorescence (TTA delayed fluorescence), and those described in, for example, JP-A Nos. 2022-99677, 2022-116901, and WO 2022 / 163199 can be similarly utilized.

[0322] Examples of host compounds that can be suitably used in this embodiment are shown below. Note that the host compounds described in the examples below can also be suitably used in this embodiment.

[0323]

[0324]

[0325]

[0326]

[0327]

[0328] (Interaction Step) In the interaction step, a desired receptor layer 3 (holding portion) is prepared, and the target object 5 and the above-described luminescent probe 6 are placed at predetermined positions on the receptor layer 3 .

[0329] In this step, the target object 5 and the luminescent probe 6 may be placed at predetermined positions on the receptive layer 3, and then a pair of electrodes 1 and 2 may be placed on either side of them. However, it is preferable to first place the electrode 1 and the support substrate on one side of the receptive layer 3, then place the target object 5 and the luminescent probe 6 at predetermined positions on the receptive layer 3 to form the light-emitting site described above, and then place the other electrode 2 on the receptive layer 3.

[0330] Here, the receiving layer 3 is preferably made of at least one insulating polymer. The "insulating" property of the insulating polymer means that the electrical resistivity is 1×10 ６ The electrical resistivity of the insulating polymer is preferably 1×10 ８ Ω·m or more, and more preferably 1×10 １０ The electrical resistivity of the insulating polymer alone is 1×10 ６ By ensuring that the resistance is Ω·m or more, when the object 5 or the light-emitting polymer 6 is placed at a plurality of positions, it is possible to prevent leakage current from flowing between a plurality of light-emitting sites.

[0331] The type of insulating polymer is not particularly limited as long as it can hold the target object 5 and the light-emitting polymer 6. Examples include nonionic polymers such as polystyrene, polymethyl methacrylate, polyvinyl alcohol, polyacrylamide, polyvinylpyrrolidone, polyvinyl polypyrrolidone, polyethylene glycol, polymethyl vinyl ether, and polyisopropyl acrylamide; cationic polymers such as sodium polyacrylate, sodium polystyrene sulfonate, sodium polyisopropylene sulfonate, polynaphthalene sulfonic acid condensate salt, and polyethyleneimine xanthate salt; anionic polymers such as dimethylaminomethyl (meth)acrylate quaternary salt, dimethyldiallylammonium chloride, polyamidine, polyvinyl imidazoline, dicyandiamide condensate, epichlorohydrin dimethylamine condensate, and polyethyleneimine; and amphoteric polymers such as dimethylaminoethyl (meth)acrylate quaternary salt acrylic acid copolymer and Hoffmann degradation product of polyacrylamide. Polystyrene and polymethyl methacrylate are preferred.

[0332] The weight average molecular weight of the insulating polymer is not particularly limited, but is preferably 5×10 ３ or more, more preferably 10×10 ３ More preferably, it is 1000×10 ３ or less, and more preferably 400×10 ３ When the weight average molecular weight is within this range, the target object 5 and the luminescent probe 6 placed in the storage section 4 of the receptive layer 3 are less likely to diffuse within the receptive layer 3 .

[0333] Receptor layer 3 may contain other materials or additives. Examples of additives include halogen elements such as bromine, iodine, and chlorine, halogenated compounds, alkali metals such as Pd, Ca, and Na, alkaline earth metals, and transition metal compounds, complexes, and salts.

[0334] In this step, the target object 5 and the luminescent probe 6 are respectively placed in a plurality of storage sections 4 of the receptive layer 3. It is preferable that adjacent storage sections 4 (4A, 4B in FIG. 6) are set with a gap between them. This makes it possible to suppress current leakage occurring between the adjacent storage sections 4A, 4B. Although only two storage sections 4A, 4B are shown in FIG. 6, it is preferable to place a large number of storage sections 4 in one receptive layer 3. For example, if 10 storage sections 4 are formed vertically and 10 horizontally in a 1 cm square with a gap between them, the receptive layer 3 will have the storage sections 4 spaced 1 cm apart. ２ The receiving layer 3 has 100 or more receiving portions 4. The positions of the receiving portions 4 may be set so that the receiving portions 4 are arranged in a matrix when viewed from above.

[0335] Either the object 5 or the light-emitting probe 6 can be placed in the storage section 4 first. The object 5 and the light-emitting probe 6 can be placed by, for example, a coating method. If the object 5 and the light-emitting probe 6 are liquid, they may be applied as is, but if they are solid or have a high viscosity, they may be dissolved or dispersed in a liquid before application. There are no particular restrictions on the application method, but a non-contact method is desirable from the perspective of minimizing damage to the receptor layer, and ejection from an inkjet head is preferred. The volume of the droplets of the object when dropped is preferably within the range of 1 pL to 100 μL.

[0336] Here, one luminescent probe and one host material may be placed in one luminescent site, and either the type of luminescent probe or the type of host material may be changed between adjacent luminescent sites. Alternatively, multiple types of luminescent probes 6 may be placed in one storage section 4 to form multiple luminescent sites. More specifically, multiple luminescent probes 6 may be combined to adjust the luminescent color or luminescent intensity of each storage section 4 to be different. Furthermore, the concentration of the target may be adjusted to be different for each storage section 4. By adjusting the luminescent color, intensity, and target concentration of each storage section 4 to be different, it becomes easier to obtain a large amount of data about the target 5. It is also preferable to combine multiple luminescent probes 6 with different luminescent colors to make the luminescent color of the storage section 4 white. There are no particular limitations on the combination of luminescent probes that produce white, but examples include combinations of blue and orange, or blue, green, and red. In this case, white is defined as a value of 1000 cd / m when the 2-degree viewing angle front luminance is measured using the method described below. ２ It is preferable that the chromaticity in the CIE 1931 color system at 2000 kJ / s is within the range of x=0.39±0.09, y=0.38±0.08.

[0337] In addition to the target object 5 and the luminescent probe 6, the container 4 may further contain a substance that changes the carrier recombination position, or an additive that promotes their interaction or assists in the light emission of the luminescent probe 6. The substance that changes the carrier recombination position may contain a substance that changes the carrier recombination position. Specifically, the following three groups of substances can be used: (1) the electron transport materials and hole transport materials typically mentioned above; (2) ionic materials that affect the movement of electrons and holes by possessing a charge; and (3) aromatic and aliphatic compounds that interact with the luminescent material and host material. The use of such substances increases the variety of data and enhances changes in light. For example, by controlling the amount of the substance dispensed to form luminescent sites so that adjacent luminescent sites are different by a certain amount, a regression curve can be intentionally created. This makes it possible to eliminate inconvenient data such as outliers, and furthermore, to treat the entire regression curve as a single data set.

[0338] There are no particular limitations on the method for causing the object and the luminescent probe to interact within the receiving layer 3 (storage section 4), and the object 5 and the luminescent probe 6 may simply be applied to the same area, or heating or other treatments may be performed as necessary.

[0339] Of the pair of electrodes 1 and 2 arranged on both sides of the receptor layer 4, one is an anode and the other is a cathode, and at least one of them is optically transparent. For the anode, a metal, alloy, electrically conductive compound, or mixture thereof having a large work function (4 eV or more, preferably 4.5 V or more) is preferably used as the electrode material. Specific examples of such electrode materials include metals such as Au, CuI, indium tin oxide (ITO), and SnO. ２ , ZnO, and other conductive transparent materials. ２ O ３ -ZnO) or the like.

[0340] A conductive polymer may also be used for the anode. Examples of conductive polymers include PEDOT:PSS, polypyrrole, polyaniline, polythiophene, polythienylene vinylene, polyazulene, polyisothianaphthene, polycarbazole, polyacetylene, polyphenylene, polyphenylene vinylene, polyacene, polyphenylacetylene, polydiacetylene, polynaphthalene, and derivatives thereof. These electrode materials may be used alone or in combination of two or more materials. It is also possible to form an electrode by stacking two or more layers of each material.

[0341] On the other hand, for the cathode, a metal (called an electron injecting metal), an alloy, an electrically conductive compound, or a mixture thereof having a small work function (5 eV or less) is used as an electrode material. Specific examples of such electrode materials include sodium, sodium-potassium alloy, magnesium, lithium, silver, a magnesium / copper mixture, a magnesium / silver mixture, a magnesium / aluminum mixture, a magnesium / indium mixture, and aluminum / aluminum oxide (Al２ O ３ ) mixtures, indium, lithium / aluminum mixtures, aluminum, rare earth metals, and the like.

[0342] Among these, from the viewpoint of electron injection property and durability against oxidation, etc., mixtures of an electron injection metal and a second metal which is a metal having a larger work function value and is more stable than the electron injection metal, such as magnesium / silver mixtures, magnesium / aluminum mixtures, magnesium / indium mixtures, and aluminum / aluminum oxide (Al ２ O ３ ) mixtures, lithium / aluminum mixtures, aluminum, etc. are preferred.

[0343] The anode and cathode can be prepared by a known method, and may be a film formed by a method such as vapor deposition or sputtering, or may be a film formed by a printing method, etc. Furthermore, they may be patterned into a desired shape.

[0344] A transparent or opaque support substrate may be disposed on the outer side of the pair of electrodes 1 and 2. However, it is preferable that a transparent support substrate is disposed on the light extraction side. Examples of transparent support substrates include known glass, quartz, and transparent resin films. A gas barrier film or the like may be disposed on the surface of these substrates as needed. A hue-improving filter such as a color filter may also be disposed on the light extraction side.

[0345] (Light Detection Step) In the light detection step, a voltage is applied between the pair of electrodes 1 and 2 to cause the luminescent probes 6 in the receptive layer 3 to emit light. Then, a detection unit (not shown) detects two or more types of light emitted by the luminescent probes.

[0346] Furthermore, the voltage applied between the pair of electrodes 1 and 2 is not particularly limited and may be either a DC voltage or an AC voltage, but an AC voltage is preferred. For example, an AC voltage can be applied by connecting the pair of electrodes 1 and 2 to an AC power supply. The AC power supply preferably has a frequency of 1 Hz to 100 MHz, an offset voltage of 0 to 20 V, and an amplitude voltage of approximately 0 to 20 V. Examples of usable AC power supplies include the 6243 DC VOLTAGE CURRENT SOURCE / MONITOR (manufactured by ADCM) and the FG300 Function Generator (manufactured by YOKOGAWA). It is preferable to also measure the current during the voltage application, and it is preferable to connect a current measurement device capable of measuring currents of approximately 1 nA to 100 mA. An example of a current measurement device is the SourceMeter 2400 (manufactured by Keithley Instruments).

[0347] Furthermore, the method for detecting two or more types of light by the detection unit is not particularly limited, and can be performed using a spectral radiance measurement device, etc. For example, the radiance or emission spectrum of the luminescent probe in the area where the object 5 and the luminescent probe 6 are placed may be measured.

[0348] The detection unit may also be an imaging device, which may capture an image of the light emitted by the luminescent probe when a voltage is applied.

[0349] (Analysis Step) The analysis method in this step is selected as appropriate depending on the purpose, the type of information related to the light emission obtained in the light detection step, etc. For example, the data detected in the light detection step may be used as is as analysis data. As described above in Figures 35A to 35D, in this embodiment, the difference between the light emitted when the target object is included and the light emitted when the target object is not included may be obtained and analyzed as a relative value.

[0350] In the analysis step, the discrimination rate may be calculated from, for example, the voltage application conditions applied by the power supply unit, the current characteristics, and the emission spectrum identified by a spectral radiance measurement device, etc. The discrimination can be performed by a discrimination rate calculator, etc., which calculates the discrimination rate based on the number of voltage supplies and the voltage supply time.

[0351] The discrimination algorithm used by the discrimination rate calculator is preferably any one of principal component analysis (PCA), cluster analysis (hierarchical method, k-means method, normal mixture method), linear discriminant analysis (LDA), partial least squares method (PLS regression), decision tree, random forest, boosted trap forest, neural network, neural boosting, K nearest neighbor method, naive Bayes, support vector machine (SVM), nominal logistic (multiple logit), generalized regression (ridge, lasso), or a combination thereof. These will be explained below.

[0352] Analysis methods: Statistical analysis and machine learning analysis can be performed using the statistical analysis software JMP16.2 or JMPpro16.2 from SAS Institute Japan, Inc., using principal component analysis (PCA), cluster analysis (hierarchical method, k-means method, normal mixed method), linear discriminant analysis (LDA), partial least squares (PLS regression), or a combination of these. Other methods that may be used include decision trees, random forests, boosted trap forests, neural networks, K-nearest neighbors, naive Bayes, support vector machines (SVM), nominal logistic (multiple logit), generalized regression (ridge, lasso), etc.

[0353] However, when the number of samples is approximately 50 or less, the number of explanatory variables is 100 or more, and the variables are spectral data, from the viewpoint of preventing overfitting and robustness of the results, PCA, which is a linear regression using dimensionally reduced principal components, PLS regression, and LDA using principal components (the wide data method in JMP) are more preferable. However, depending on the data set, other analysis methods such as those mentioned above may also be effective.

[0354] Regarding Linear Discriminant Analysis (LDA), JMP 16.2 supports both the stepwise method, which allows for the selection of explanatory variables, and the wide-angle method, which reduces the dimensions to principal components. The wide-angle method is preferable from the perspective of reducing calculation time and preventing overfitting, as mentioned above.

[0355] Regarding evaluation of discrimination performance, for evaluation of discrimination performance, all sample data is divided into training data and validation data at an arbitrary ratio, and linear discriminant analysis (LDA) is performed to calculate the discrimination rate (correct answer rate, hit rate), misclassification rate, and entropy R-squared. For calculation of the validation results using validation data, it is preferable to use the hold-out method using an arbitrary or random set of validation data, or k-fold cross-validation in which all sample data is divided into k parts and cross-validated using the k combinations. Furthermore, for the results of k-fold cross-validation, it is preferable to display the average value of the k parts, the one with the best statistics (smallest error from the average) from among the k discriminant models, or the one with the smallest difference in discrimination rate or entropy R-squared between training and validation.

[0356] (Others) In the analytical device 100 used in the above-described analytical method, organic functional layers such as a hole injection layer, a hole transport layer, an electron transport layer, and an electron injection layer may be further provided in the receptor layer 3. Typical configuration examples of the above-described analytical device include, but are not limited to, the following configuration examples.

[0357] (i) anode / receptor layer (receptor section) / cathode (ii) anode / receptor layer (receptor section / electron transport layer) / cathode (iii) anode / receptor layer (hole transport layer / receptor section) / cathode (iv) anode / receptor layer (hole transport layer / receptor section / electron transport layer) / cathode (v) anode / receptor layer (hole transport layer / receptor section / electron transport layer / electron injection layer) / cathode (vi) anode / receptor layer (hole injection layer / hole transport layer / electron blocking layer / receptor section / hole blocking layer / electron transport layer) / cathode

[0358] Note that any of the hole injection layer, hole transport layer, electron blocking layer, detection region, hole blocking layer, and electron transport layer may be present between the external receptor layer and the electrode (anode or cathode) rather than inside the receptor layer. The "electron transport layer" according to this embodiment is a layer having the function of transporting electrons, and in a broad sense, the electron injection layer and the hole blocking layer are also included in the electron transport layer. It may also be composed of multiple layers. The "hole transport layer" according to this embodiment is a layer having the function of transporting holes, and in a broad sense, the hole injection layer and the electron blocking layer are also included in the hole transport layer. It may also be composed of multiple layers. These are similar to known layers.

[0359] 1. Example 1 1-1. Synthesis of Luminescent Probes 1 to 16 All reactions were carried out in oven-dried glassware under a nitrogen atmosphere unless otherwise noted. All chemical products were purchased from Aldrich, TCI, or Kanto Chemical and used as is without further purification.

[0360] (1) Synthesis of Monomer 1 Based on the reaction formula below, Monomer 1 having a main chain containing a phosphate ester and a luminophore bonded to the main chain was synthesized via intermediates 1 to 6.

[0361] Synthesis of Intermediate 1: Thymidine (15.0 g, 61.9 mmol) and imidazole (16.9 g, 248 mmol) were dissolved in DMF (124 mL), and tert-butyldimethylsilyl chloride (19.6 g, 130 mmol) was added and stirred at room temperature for 17 hours. Water was added to the reaction mixture, and the mixture was subjected to separation and extraction with ethyl acetate. The resulting organic phase was dried over magnesium sulfate, and the solvent was evaporated to obtain the desired Intermediate 1 (28.3 g, 97%) as a colorless solid.

[0362] Synthesis of Intermediate 2 Intermediate 1 (28.3 g, 60.1 mmol) and ammonium sulfate (12.7 g, 96.2 mmol) were dissolved in hexamethyldisilazane (314 mL, 1.50 mol) and heated under reflux for 3 hours. Water was then added to the reaction mixture, and the mixture was subjected to liquid separation and extraction with ethyl acetate. The resulting organic phase was dried over magnesium sulfate, and the solvent was distilled off to obtain a crude product. The crude product was purified by silica gel column chromatography to obtain the desired Intermediate 2 as a brown liquid (13.2 g, 64%).

[0363] Synthesis of Intermediate 4 A mixture of intermediate 2 (10.1 g, 29.3 mmol), 1-bromopyrene (8.24 g, 29.3 mmol), tris(dibenzylideneacetone)dipalladium(0) (671 mg, 733 μmol), tri-tert-butylphosphonium tetrafluoroborate (850 mg, 2.93 mmol), dicyclohexylmethylamine (9.35 mL, 44.0 mmol), and 1,4-dioxane (100 mL) was heated at 90°C for 1 hour. Water was added to quench the reaction, and separation and extraction were performed with ethyl acetate. The resulting organic phase was dried over magnesium sulfate, and the solvent was distilled off to obtain a crude product containing intermediate 3, which was used directly in the next reaction.

[0364] To the crude product containing intermediate 3, 100 mL of THF, 1 M tetrabutylammonium fluoride THF solution (117 mL, 117 mmol), and acetic acid (6.74 mL, 117 mmol) were added and stirred at 40°C for 2 hours. Water was added to quench the reaction, and separation and extraction were performed with ethyl acetate. The resulting organic phase was dried over magnesium sulfate, and the solvent was evaporated to obtain a crude product. This was purified by silica gel column chromatography to obtain the desired intermediate 4 as a light brown solid (5.87 g, 63%).

[0365] Synthesis of Intermediate 5: A solution of sodium triacetylborate (11.8 g, 55.8 mmol) and acetic acid (7.87 mL, 138 mmol) in 93 mL of acetonitrile was cooled to 0 °C. A solution of Intermediate 4 (5.87 g, 18.6 mmol) in THF (62 mL) was then added dropwise. After the addition, the mixture was warmed to room temperature and stirred for 15 minutes, after which water was added to terminate the reaction. Separation and extraction with ethyl acetate was performed, and the resulting organic phase was dried over magnesium sulfate. The solvent was then distilled off to obtain a crude product. Purification by silica gel column chromatography and reverse-phase HPLC afforded the desired Intermediate 5 as a colorless solid (3.44 g, 58%).

[0366] Synthesis of Intermediate 6 A mixture of Intermediate 5 (3.44 g, 10.8 mmol), 4,4'-dimethoxytrityl chloride (4.40 g, 13.0 mmol), ethyldiisopropylamine (2.82 mL, 16.2 mmol), and dehydrated pyridine (54 mL) was stirred at room temperature for 4 hours. The reaction was then stopped by adding methanol. The solvent was distilled off, and the resulting crude product was purified by silica gel column chromatography to obtain the desired Intermediate 6 as a colorless viscous solid (5.71 g, 85%).

[0367] Synthesis of Monomer 1 2-Cyanoethyldiisopropylchlorophosphoramidite (3.08 mL, 13.8 mmol) was added dropwise to a mixture of Intermediate 6 (5.71 g, 9.20 mmol), ethyldiisopropylamine (6.42 mL, 36.8 mmol), and dehydrated dichloromethane (92 mL) at 0°C. The mixture was warmed to room temperature and stirred for 3 hours, after which the solvent was distilled off to obtain a crude product. Purification by silica gel column chromatography afforded the target Monomer 1 as a colorless solid (4.64 g, 61%).

[0368] (2) Preparation of Monomer 2 Monomer 2 was a reagent having the structure shown below, purchased from Glen Research (Sterling, Virginia).

[0369] (3) Measurement of the size of the light-emitting moiety of Monomer 1 and Monomer 2 For Monomer 1 and Monomer 2, the size of the structure represented by R in the following general formula (light-emitting moiety that contributes to light emission) was calculated by the following method. When the structure represented by R is a hydrogen atom, the size was set to 0 because this moiety cannot be a light-emitting moiety.

[0370] -Method for measuring the size of the light-emitting portion The size of the structure indicated by R in the above compound was determined by treating the structure indicated by R as an independent molecule using Winmostar (developed by CrossAbility), an atomic / molecular scale structure modeling and visualization software. The length of the cylinder with the smallest diameter in which the molecule was inscribed was then calculated as the size of the structure. The results are shown in Table 1 below.

[0371]

[0372] (4) Synthesis of Luminescent Probes 1-16 Sixteen types of oligonucleotides (Seq 1-16) with mixed sequences of Monomer 1 and Monomer 2 were synthesized according to standard methods, as shown in Table 2 below. DNA synthesis reagents were purchased from Glen Research (Sterling, Virginia). All oligonucleotides were synthesized using a standard protocol for phosphoramidite-based coupling techniques on a Nippon Techno Service NTS T-series DNA / RNA synthesizer. Each luminescent probe carrier obtained by automated synthesis was reacted with ammonium water at room temperature for 2 hours, and then excised from the particulate carrier. The solvent was dried using a centrifugal dryer, and then ultrapure water was added to obtain first components 1-16 containing each of the luminescent probes 1-16. It was confirmed that the luminescent probes 1-15 emitted fluorescence and excimer luminescence when exposed to specific excitation light (light with a wavelength of 350 nm). Luminescent probe 16 did not exhibit fluorescence or excimer luminescence. The proportion of β-deoxyribose in the table was calculated by {(number of β-deoxyribose in the luminescent probe) / (number of deoxyribose in the luminescent probe)}×100[%].

[0373] A 96-well microplate was prepared, in which wells with an opening diameter of 7 mm were arranged in 12 columns and 8 rows at 9 mm intervals. 100 μl of each of the above-mentioned luminescent probes 1 to 16 was placed in the 96-well microplate using an automatic dispenser (NichiMart CUBE, manufactured by NICHIRYO Corporation) to form multiple reaction fields.

[0374] The 96-well microplate on which the above-mentioned luminescent probes were placed was irradiated with excitation light (wavelength 350 nm), and the fluorescence spectrum was obtained as first signal information (blank information).

[0375] 1-3. Interaction Step After the blank information acquisition step, 20 μl of each of three types of soft drinks (subjects 1 to 3) was placed on the 96-well microplate in the same manner as above.

[0376] 1-4. Light Detection Step The 96-well microplate on which the above-mentioned luminescent probes and the target substance were placed was irradiated with excitation light (wavelength 350 nm), and the fluorescence spectrum was obtained as second signal information.

[0377] 1-5. Analysis step The first signal information acquired in the blank information acquisition step was subtracted from the second signal information acquired in the light detection step to calculate analytical data. Then, principal component analysis was performed using the analytical data as explanatory variables. As shown in Figure 7, separation by type of second component (target substance) was possible in the two-dimensional space where principal components 1 and 2 were plotted.

[0378] 1-6. Others Pigment inks were prepared as follows using the aqueous dispersions of each luminescent probe carrier obtained by the above-mentioned automated synthesis as pigment dispersions, and it was confirmed that the same analytical methods as those described above could be used.

[0379] [Production of ink] Ink raw materials Triethylene glycol monobutyl ether (TEGmBE, manufactured by Tokyo Chemical Industry Co., Ltd.) 5 parts by mass 2-pyrrolidinone (manufactured by BASF) 8 parts by mass Purified glycerin (manufactured by Kao Corporation) 2 parts by mass Surfynol 440 (nonionic surfactant manufactured by AIRPRODUCTS) (2,4,7,9-tetramethyl-5-decyne-4,7-diol) 0.5 parts by mass 1,2-hexanediol (manufactured by Degussa) 1 part by mass Triethanolamine (manufactured by Konishi Co., Ltd.) 0.8 parts by mass Megafac 444 (nonionic surfactant manufactured by DIC Corporation) (perfluoroalkylethylene oxide adduct) 0.2 parts by mass Polyurethane A 0.5 parts by mass (solid content) AQUACER 552 1 part by mass (solid content) Pure water Balance (Here, the "balance" in the amount of pure water to be added means that the amount of pure water is adjusted and added so that the total parts by mass of the blended raw materials becomes 100 parts by mass.)

[0380] The above components were placed in a 100 ml plastic container in the above ratio and stirred for 1 hour, after which 20 parts of the aqueous dispersion of the support of the luminescent probe 1 was added and stirred for another 1 hour to prepare a pigment ink composition.

[0381] [Inkjet Ejection Evaluation Test] An ejection evaluation was performed using the ink composition prepared by the above method. A commercially available thermal jet inkjet printer (Photosmart D5360, manufactured by Hewlett-Packard) was used for the ejection evaluation. In addition, printing was performed on photo paper (glossy) (HP Advance Photo Paper, manufactured by Hewlett-Packard), which is paper specifically for inkjet printing. As a result, it was confirmed that the ink composition prepared by the above method could be ejected by inkjet.

[0382] [Storage] It was also confirmed that ink-jet ejection was possible as described above even after storage at room temperature for one month.

[0383] [Discussion] Based on the results of the above analysis process and the results of the inkjet discharge evaluation test, it was found that by using a luminescent probe with a nanometer-order luminescent element whose luminescence behavior changes chemically and / or physically upon interaction with the target substance, and by combining multiple luminescent elements of different sizes, it was possible to separate the data for each type of soft drink. In other words, the analysis using the luminescent probe made it possible to obtain changes in such physicochemical phenomena (information on the state of the target substance) in a short time and in a simple manner.

[0384] 2. Example 2 2-1. Synthesis of luminescent probes 1 to 75 (1) Synthesis of monomer 3 Based on the reaction formula below, monomer 3 having a main chain containing a phosphate ester and a luminophore bound to the main chain was synthesized via intermediates 7 to 10.

[0385] Synthesis of Intermediate 8 A mixture of the above-mentioned Intermediate 2 (2.00 g, 5.80 mmol), 1,4-dibromobenzene (8.24 g, 29.0 mmol), tris(dibenzylideneacetone)dipalladium(0) (133 mg, 145 μmol), tri-tert-butylphosphonium tetrafluoroborate (168 mg, 580 μmol), dicyclohexylmethylamine (1.85 mL, 8.70 mmol), and 1,4-dioxane (29 mL) was heated at 90°C for 3 hours. Water was added to quench the reaction, and separation and extraction were performed with ethyl acetate. The resulting organic phase was dried over magnesium sulfate, and the solvent was distilled off to obtain a crude product containing Intermediate 7, which was used directly in the next reaction.

[0386] To the crude product containing intermediate 7, 29 mL of THF, 1 M tetrabutylammonium fluoride THF solution (23.2 mL, 23.2 mmol), and acetic acid (1.32 mL, 23.2 mmol) were added, and the mixture was stirred at 40°C for 2 hours. Water was added to quench the reaction, and separation and extraction were performed with ethyl acetate. The resulting organic phase was dried over magnesium sulfate, and the solvent was evaporated to obtain a crude product. This crude product was purified by silica gel column chromatography to obtain the desired intermediate 8 as a yellow-brown oil (1.08 g, 69%).

[0387] Synthesis of Intermediate 9 Sodium triacetylborate (2.52 g, 11.9 mmol) and acetic acid (1.82 mL, 31.8 mmol) were dissolved in 20 mL of acetonitrile. The solution was cooled to 0°C, and a solution of Intermediate 8 (1.08 g, 3.98 mmol) in THF (13 mL) was added dropwise. After the dropwise addition, the temperature was raised to room temperature and the mixture was stirred for 2 hours and 30 minutes. Water was then added to quench the reaction. Separation and extraction was performed with ethyl acetate, and the resulting organic phase was dried over magnesium sulfate. The solvent was then evaporated to obtain a crude product. After washing with heptane, the product was purified by recrystallization from ethyl acetate to obtain the desired Intermediate 9 as a colorless solid (481 mg, 44%).

[0388] Synthesis of Intermediate 10 A mixture of intermediate 9 (1.00 g, 3.66 mmol), N,N-dimethyl-4v(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)aniline (905 mg, 3.66 mmol), bis(dibenzylideneacetone)palladium(0) (106 mg, 184 μmol), ethyldiisopropylamine 2-dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl (175 mg, 367 μmol), potassium phosphate (2.33 g, 367 μmol), N,N-dimethylformamide (33 mL), and water (4 mL) was stirred at 90° C. for 1 hour, and then water was added to terminate the reaction. The mixture was subjected to liquid separation and extraction with dichloromethane. The resulting organic phase was dried over magnesium sulfate, and the solvent was distilled off to obtain a crude product. The crude product was purified by silica gel column chromatography to obtain the target intermediate 10 as a colorless solid (1.10 g, 96%).

[0389] Synthesis of Intermediate 11 A mixture of Intermediate 10 (1.10 g, 3.52 mmol), 4,4'-dimethoxytrityl chloride (1.32 g, 3.90 mmol), ethyldiisopropylamine (0.92 mL, 5.29 mmol), and dehydrated pyridine (17.5 mL) was stirred at room temperature for 4 hours, and then methanol was added to quench the reaction. The solvent was distilled off, and the resulting crude product was purified by silica gel column chromatography to obtain the desired Intermediate 11 as a yellow viscous solid (2.89 g, 82%).

[0390] Synthesis of Monomer 3 2-Cyanoethyldiisopropylchlorophosphoramidite (468 μL, 2.10 mmol) was added dropwise to a mixture of Intermediate 11 (1.23 g, 2.00 mmol), ethyldiisopropylamine (1.39 mL, 8.00 mmol), and dehydrated dichloromethane (80 mL) at room temperature. After stirring for 2 hours, the solvent was distilled off to obtain a crude product. Purification by silica gel column chromatography yielded the target Monomer 3 as a yellow viscous solid (1.53 g, 94%).

[0391] (2) Preparation of Monomer 4 Monomer 4 was purchased from Glen Research (Sterling, Virginia) as a reagent having the structure shown below. Thymidine contained in Monomer 4 is a type of natural nucleic acid base.

[0392] (3) Synthesis of Monomer 5 Monomer 5 was synthesized according to a non-patent document (J. Am. Chem. Soc. 1996, 118, 7671-7678.) Monomer 5 is a structural isomer of the above-mentioned Monomer 1, and is a monomer having an α-form sugar structure.

[0393] (4) Measurement of the size of the light-emitting moiety of Monomers 3 to 5 For Monomers 3 to 5, the size of the structure represented by R in the following general formula (light-emitting moiety that contributes to light emission) was calculated in the same manner as described above. The results are shown in Table 3. Table 3 also shows the results for Monomer 1 and Monomer 2. Note that when the structure represented by R is a hydrogen atom, the size was set to 0 because that moiety cannot be a light-emitting moiety.

[0394]

[0395] (5) Synthesis of Luminescent Probes 1 to 60 (5-1) Preparation of Luminescent Probes 1 to 16 Luminescent probes 1 to 16 were prepared in the same manner as in Example 1 above.

[0396] (5-2) Synthesis of Luminescent Probes 1, and 17 to 31 Sixteen types of mixed sequence oligonucleotides (Seq 1, and 17 to 31) were synthesized using the same method as used to synthesize the above-mentioned luminescent probes 1 to 16, as shown in Table 4 below, with the above-mentioned Monomer 1 and aniline-containing Monomer 3. It was confirmed that the luminescent probes 1 and 17 to 31 each emit fluorescence and exciplex emission when exposed to specific excitation light (light with a wavelength of 350 nm).

[0397]

[0398] (5-3) Synthesis of Luminescent Probes 1 and 32 to 45 Fifteen types of mixed sequence oligonucleotides (Seq 1 and Seq 32 to 45) of the above-mentioned Monomer 1 and thymidine-containing Monomer 4 were synthesized by a method similar to that used to synthesize the above-mentioned Luminescent Probes 1 to 16, as shown in Table 5 below. Furthermore, it was confirmed that Luminescent Probes 1 and 32 to 45 each emit fluorescence and excimer luminescence when exposed to specific excitation light (light with a wavelength of 350 nm).

[0399]

[0400] (5-4) Synthesis of Luminescent Probes 46 to 60 Fifteen types of mixed sequence oligonucleotides (Seq 46 to 60) of Monomer 5 containing the above-mentioned α-sugar structure and Monomer 2 were synthesized using a method similar to that used to synthesize the above-mentioned luminescent probes 1 to 16, as shown in Table 6 below. It was confirmed that each of the luminescent probes 46 to 60 emitted fluorescence and excimer luminescence when exposed to specific excitation light (light with a wavelength of 350 nm).

[0401]

[0402] 2-2. Analysis using luminescent probes (1) Analysis using luminescent probes 1 to 15 Blank information acquisition process Multiple 96-well microplates were prepared, each with wells having an opening diameter of 7 mm and arranged in 12 columns and 8 rows at 9 mm intervals. 100 μl of each of the above luminescent probes 1 to 15 was placed in each of the 96-well microplates using an automatic dispenser (NichiMart CUBE, manufactured by NICHIRYO Corporation), one for each target substance, to form multiple reaction fields.

[0403] The 96-well microplate on which the above-mentioned luminescent probes were placed was irradiated with excitation light (wavelength 350 nm), and the fluorescence spectrum was obtained as first signal information (blank information).

[0404] Interaction step: After the blank information acquisition step, 20 μl of 95 brands of beverages from 7 categories (category I: 12 brands, type II: 23 brands, type III: 6 brands, type IV: 10 brands, type V: 20 brands, type VI: 13 brands, type VII: 11 brands, all of which are the target substances) was placed on the 96-well microplate using the same method as above.

[0405] Light Detection Step The 96-well microplate on which the above-mentioned luminescent probes and the target substance were placed was irradiated with excitation light (wavelength 350 nm), and the fluorescence spectrum was obtained as second signal information.

[0406] Analysis step: The first signal information acquired in the blank information acquisition step was subtracted from the second signal information acquired in the light detection step to calculate analysis data. Then, the analysis data was used as explanatory variables, and the beverage type data was used as the objective variable for learning, creating a discriminant model using linear discriminant analysis (LDA). The resulting linear discriminant analysis model plot is shown in Figure 8A. Then, six-fold cross-validation was used to calculate the accuracy rate and create a confusion matrix, quantifying the generalization performance of the discriminant model. The confusion matrix is ​​shown in Figure 8B.

[0407] Results As shown in FIG. 8B, the above discrimination model was able to classify the types of 95 brands of beverages with an accuracy of approximately 65%.

[0408] (2) Analysis using luminescent probes 1 and 17 to 31 The above-mentioned blank information acquisition process through analysis process was carried out in the same manner using luminescent probes 1 and 17 to 31 (16 types). The obtained linear discriminant analysis model plot is shown in Figure 9A. Then, the accuracy rate was calculated and a confusion matrix was created by 6-fold cross-validation to quantify the generalization performance of the discriminant model. The confusion matrix is ​​shown in Figure 9B.

[0409] Results As shown in FIG. 9B, the above discrimination model was able to classify the types of 95 brands of beverages with an accuracy of approximately 63%.

[0410] (3) Analysis using luminescent probes 1 to 15 and 17 to 31 The above-mentioned blank information acquisition step through analysis step were carried out in the same manner using luminescent probes 1 to 15 and 17 to 31 (30 types). The linear discriminant analysis model plot obtained in this manner is shown in Figure 10A. Then, six-fold cross-validation was performed to calculate the accuracy rate and create a confusion matrix, and the generalization performance of the discriminant model was quantified. The confusion matrix is ​​shown in Figure 10B.

[0411] Results As shown in FIG. 10B, the above-described discrimination model was able to classify the types of 95 brands of beverages with approximately 90% accuracy.

[0412] (4) Analysis using luminescent probes 1 and 32 to 45 The above-mentioned blank information acquisition process through analysis process were carried out in the same manner using luminescent probes 1 and 32 to 45 (15 types). The linear discriminant analysis model plot obtained in this manner is shown in Figure 11A. Then, six-fold cross-validation was performed to calculate the accuracy rate and create a confusion matrix, and the generalization performance of the discriminant model was quantified. The confusion matrix is ​​shown in Figure 11B.

[0413] Results As shown in FIG. 11B, the above discriminant model was able to classify the 95 brands by type with an accuracy of approximately 54%.

[0414] (5) Analysis using luminescent probes 46 to 60 The above-mentioned blank information acquisition process through analysis process was carried out in the same manner using luminescent probes 46 to 60 (15 types). The linear discriminant analysis model plot obtained in this manner is shown in Figure 12A. Then, six-fold cross-validation was performed to calculate the accuracy rate and create a confusion matrix, and the generalization performance of the discriminant model was quantified. The confusion matrix is ​​shown in Figure 12B.

[0415] As shown in Figure 12B, the above discrimination model was able to classify the 95 brands by type with an accuracy of approximately 45%. The reason for this lower accuracy compared to other analyses is that the proportion of β-sugar structures in the luminescent probes was low.

[0416] (6) Analysis when explanatory variables were randomly permuted To clarify the effects of the present invention, a model was created by randomly permuting explanatory variables as a negative control experiment. Specifically, the above-mentioned measurement and analysis steps were performed using luminescent probes 1 to 15 (15 types), and the obtained explanatory variables were randomly permuted to create a discriminant model. The resulting linear discriminant analysis model plot is shown in Figure 13A. Then, six-fold cross-validation was performed to calculate the accuracy rate and create a confusion matrix, and the generalization performance of the discriminant model was quantified. The confusion matrix is ​​shown in Figure 13B.

[0417] Results: As shown in Figure 13B, the discriminant model in which the explanatory variables were randomly swapped was able to classify the types of 95 brands of beverages with an accuracy of approximately 25%. This result indicates that the above-mentioned accuracy was not achieved by chance when using the discriminant model with the above-mentioned luminescent probe and the correct setting of explanatory variables and target variables. This result shows that the discriminant model in which the above-mentioned luminescent probe and the correct setting of explanatory variables and target variables are used can accurately analyze various compounds.

[0418] Furthermore, based on the above results, it was found that by using a luminescent probe with a nanometer-order luminescent element whose luminescence behavior changes chemically and / or physically upon interaction with the target substance, and further by combining multiple luminescent elements of different sizes, it was possible to separate the data for each type of soft drink. In other words, such changes in physicochemical phenomena (information on the state of the target substance) could be obtained in a short time and in a simple manner by analysis using the luminescent probe.

[0419] 3. Example 3 The following monomers X1 to X13 were prepared by applying the monomer synthesis methods described in Examples 1 and 2. The above-mentioned analytical method was also carried out using oligomers made from these monomers, and it was possible to distinguish them in the same manner as above.

[0420]

[0421]

[0422] (Measurement of the size of the light-emitting moiety of monomers X1 to X13) For monomers X1 to X13, the size of the structure represented by R in the following general formula (the light-emitting moiety that contributes to light emission) was calculated in the same manner as above. The results are shown in Table 3.

[0423]

[0424] 4. Example 4 <Preparation of Sensing Plate> A 30 mm x 30 mm x 0.7 mm glass substrate was coated with a 100 nm thick indium tin oxide (ITO) film, which was then patterned. The substrate was then ultrasonically cleaned with isopropyl alcohol, dried with dry nitrogen gas, and subjected to UV ozone cleaning for 5 minutes to obtain a transparent support substrate with an ITO transparent electrode (anode). A 1.0% solution of insulating polymer polystyrene (manufactured by ACROS ORGANICS, molecular weight = 260,000) diluted with n-propyl acetate as a solvent was spin-coated at 500 rpm for 30 seconds, followed by drying at 120°C for 30 minutes to form a 50 nm thick ink-receiving layer.

[0425] <Setting the test specimen (object) on the sensing plate> Three types of commercially available drinking water (drinking water A, drinking water B, and drinking water C) were prepared as test specimens. The sensing plate was immersed in a petri dish containing each liquid for 1 minute, and then the liquid was removed with an air gun and the plate was dried.

[0426] <Setting of luminescent compound (luminescent probe) on sensing plate> A luminescent compound (luminescent probe: blue phosphorescent compound Ep1 described below) was mixed with n-propyl acetate as a solvent at a concentration of 10 mg / mL, heated with ultrasound for 30 minutes, and then filtered through a 0.2 μm filter to remove aggregated components, thereby preparing a luminescent ink. Next, a solution (luminescent ink) containing phosphorescent compound Ep1, having the structure shown below, was dropped onto the ink-receiving layer using an inkjet printing method under the following conditions to form a detection area, and then dried at 120°C for 30 minutes to evaporate the solvent.

[0427] The inkjet printing conditions were as follows: an IJCS-1 manufactured by Konica Minolta was used as the inkjet printing system, and a KM512 manufactured by Konica Minolta was used as the inkjet head. The number of ejection shots was 2, the distance between the ejection nozzles from the head was 140 μm pitch, and printing was performed at a head scan speed of 90 mm / sec.

[0428] Next, this was attached to a vacuum deposition apparatus, and the vacuum chamber was filled with 4 × 10 －４ After reducing the pressure to 100 Pa, an electron injection layer and a cathode were formed under the following conditions: The electron injection layer was formed by vapor-depositing potassium fluoride at a deposition rate of 0.1 Å / sec to a thickness of 2.0 nm, and the cathode was formed by vapor-depositing Al at a deposition rate of 4 Å / sec to a thickness of 100 nm.

[0429] A sensing element (sensing device) was fabricated through the above steps. Figure 36(a) is a plan view of the sensing element, and Figure 36(b) is a schematic cross-sectional view taken along line A-A'. Note that the reference numerals in the figure are the same as those in Figure 6 and the like, and the reference numerals not in the above figures include reference numeral 8 representing the substrate and reference numeral 9 representing the electrode region (the region where anode 1 and cathode 2 overlap in a plan view). Here, electrode 1 represents the anode, and electrode 2 represents the cathode.

[0430] The fabricated sensing element had four detection regions (2 x 2 mm) within a 30 x 30 mm area, with the detection regions (inkjet regions) the same size as the electrodes. Because the polystyrene receptive layer 3 is insoluble in the analyte, trace amounts of solid analyte components were present on the polystyrene surface after immersion and drying. When the luminescent compound was then applied by inkjet printing, the inkjet ink solvent dissolved the polystyrene, forming a dispersion with the analyte and reaching the lower electrode. As a result, as shown in Figure 36(b), a detection region was formed in which the luminescent compound and analyte were dispersed at the molecular level. (Note that analytes present on the receptive layer outside the detection region are not shown.) In this example, two sensing elements were fabricated for each analyte type, providing eight measurement points for each analyte (a total of 24 measurement points for the three types). Furthermore, a voltage could be applied to the anode and cathode via wiring (not shown).

[0431] <Data Acquisition> The spectral radiance spectrum [W sr －１ ・m －２ ・nm －１ ] and the current density [mA / cm ] for each driving voltage. ２ The driving voltage was varied from 1 to 10 V in 1 V increments, and the current was measured at each voltage. A spectroradiometer CS-2000 (manufactured by Konica Minolta) was used to measure the luminance. A 6243 DC VOLTAGE CURRENT SOURCE / MONITOR (manufactured by ADCMT) was used to measure the current.

[0432] The obtained data set is shown in Figure 37. Eight rows were formed for each type of specimen, for a total of 24 rows. The columns contained 401 spectral data points from 380 to 780 nm in 1 nm increments, and 10 current density data points from 1 to 10 V in 1 V increments. The spectral data was normalized by setting the value at the wavelength with the highest radiance to 1. The voltage was also normalized by setting the current density value at 10 V to 1, thereby reducing the influence of variations in the device.

[0433] <Discriminant analysis> A discriminant analysis was performed based on the above dataset. Linear discriminant analysis (LDA) was performed using JMP 16.2, a statistical analysis software from SAS Institute Japan, Ltd. Details of discriminant analysis are provided, for example, in "Multivariate Data Utilization Techniques with JMP, Kaibundo Publishing Co., Ltd.", pp. 173-193 (Chapter 6).

[0434] Canonical function 1 and canonical function 2 were determined from the measurement data, and 24 discriminant scores corresponding to canonical axis 1 and 24 discriminant scores corresponding to canonical axis 2 were determined. Next, a discriminant plot (plot with canonical axes 1 and 2 as axes) was created using 42 discriminant scores corresponding to rows 2 to 8, rows 10 to 16, and rows 18 to 24 of the data set. Then, the remaining three discriminant scores for verification corresponding to rows 1, 9, and 17 were determined from the Mahalanobis distance to determine which group in the discriminant plot was closest to, and the number of correct answers was counted.

[0435] Next, the combinations were changed from 1 and 3 to row 8, from 9 and 11 to row 16, and from 17 and 19 to row 24, and a discrimination plot consisting of 42 corresponding discrimination scores was created. Then, the remaining three discrimination scores for verification corresponding to rows 2, 10, and 18 were determined from the Mahalanobis distance to determine which group in the discrimination plot was closest to, and the number of correct answers was counted.

[0436] Similarly, the combinations were changed eight times (discriminant analysis was performed eight times), and the number of times that three types of samples (drinking water A, drinking water B, drinking water C) gave correct answers was counted over 24 tests. The results are shown in Figure 38. The number of correct answers out of 24 tests was 16, and the discrimination rate (number of correct answers / number of tests) was 66.7%. This result is significantly higher than the chance of 33.3%, demonstrating the effectiveness of this sensing system.

[0437] The JMP software was used in the following steps: 1. Load the following table (dataset) from the "File" menu. 2. Select rows 1, 9, and 17 of the table, right-click, and select "Exclude." 3. Select "Multivariate" - "Discriminant Analysis" from the "Analysis" menu. 4. Specify "Column 1" as "X, Category," "Remaining Columns" as "Y, Covariate," and specify "Linear Landscape Data" as the discriminant method. After selecting "Validate Model with Excluded Rows" in the advanced options, click the "OK" button to perform the calculation. 5. Subtract 3 from the "Number of Misclassifications" for "Exclusion" in "Score Summary" to obtain the number of correct answers. 6. Changed step 2 above to rows 2, 10, and 18 (changing the combination), and then performed steps 2 through 5. This was repeated a total of eight times.

[0438] Furthermore, when the above-mentioned analytical method was carried out using oligomers using the above-mentioned compounds P-1 to P-11 instead of the phosphorescent compound Ep1, discrimination was possible in the same manner as above.

[0439]

[0440] 5. Example 5 <Preparation of Sensing Plate> A 30 mm × 30 mm × 0.7 mm glass substrate was coated with indium tin oxide (ITO) to a thickness of 100 nm, and the substrate was then patterned. The substrate was then ultrasonically cleaned with isopropyl alcohol, dried with dry nitrogen gas, and cleaned with UV ozone for 5 minutes, yielding a transparent support substrate provided with an ITO transparent electrode (anode).

[0441] On the formed anode, a solution of polystyrene (manufactured by ACROS ORGANICS, molecular weight = 260,000) as an insulating polymer diluted to 1.0% with n-propyl acetate as a solvent was spin-coated at 500 rpm for 30 seconds, and then dried at 120°C for 30 minutes to form an ink-receiving layer with a thickness of 50 nm.

[0442] <Setting the specimen (object) on the sensing plate> As the specimen (object), aqueous solutions of Mg ions with concentrations of 10 ppb, 100 ppb, 1 ppm, 10 ppm, 100 ppm, and 1000 ppm were prepared. The sensing plate was immersed in a petri dish containing each solution for 1 minute, after which the solution was removed with an air gun and the plate was dried.

[0443] <Setting of luminescent compound (luminescent probe) on sensing plate> A luminescent compound (luminescent probe, the same blue phosphorescent compound Ep1 as in Example 4) was mixed with n-propyl acetate as a solvent at a concentration of 10 mg / mL, heated with ultrasound for 30 minutes, and then filtered through a 0.2 μm filter to remove aggregated components, thereby preparing a luminescent ink. Next, a solution containing the phosphorescent compound Ep1 (luminescent ink) was dropped onto the ink-receiving layer using an inkjet printing method under the following conditions to form a detection area, and then dried at 120° C. for 30 minutes to evaporate the solvent.

[0444] The inkjet printing conditions were as follows: an IJCS-1 manufactured by Konica Minolta was used as the inkjet printing system, and a KM512 manufactured by Konica Minolta was used as the inkjet head. The number of ejection shots was 2, the distance between the ejection nozzles from the head was 140 μm pitch, and printing was performed at a head scan speed of 90 mm / sec.

[0445] Next, this was attached to a vacuum deposition apparatus, and the vacuum chamber was filled with 4 × 10 －４ After reducing the pressure to 100 Pa, an electron injection layer and a cathode were formed under the following conditions: The electron injection layer was formed by vapor-depositing potassium fluoride at a deposition rate of 0.1 Å / sec to a thickness of 2.0 nm, and the cathode was formed by vapor-depositing Al at a deposition rate of 4 Å / sec to a thickness of 100 nm.

[0446] By the above steps, a sensing element (sensing device) was fabricated as shown in Fig. 36. Fig. 36(a) is a plan view of the sensing element, and Fig. 36(b) is a schematic cross-sectional view taken along line AA'.

[0447] The fabricated sensing element had four detection regions (2 x 2 mm) within a 30 x 30 mm area, with the detection regions (inkjet regions) the same size as the electrodes. Because the polystyrene receptive layer 3 is insoluble in the analyte, trace amounts of the analyte's solid components are present on the polystyrene surface after immersion and drying. When the luminescent compound (luminescent probe) is then applied by inkjet, the inkjet ink solvent dissolves the polystyrene, forming a dispersion with the analyte and reaching the lower electrode. As a result, as shown in Figure 36(b), a detection region is formed in which the luminescent compound and analyte are dispersed at the molecular level. (Note that analytes present on the receptive layer outside the detection region are not shown.) In this example, two sensing elements 1 were fabricated for each analyte type, providing eight measurement points for each analyte (a total of 24 measurement points for the three types). Furthermore, a voltage could be applied to the anode and cathode via wiring (not shown).

[0448] The spectral radiance spectrum [W sr －１ ・m －２ ・nm －１ ] and the current density [mA / cm ] for each driving voltage. ２ The driving voltage was varied from 1 to 10 V in 1 V increments, and the current was measured at each voltage. A spectroradiometer CS-2000 (manufactured by Konica Minolta) was used to measure the luminance. A 6243 DC VOLTAGE CURRENT SOURCE / MONITOR (manufactured by ADCMT) was used to measure the current.

[0449] The obtained data set is shown in Figure 39. Eight rows were formed for each type of specimen, for a total of 24 rows. The columns contained 401 spectral data points at 1 nm intervals from 380 to 780 nm, and six current density data points at 1 V intervals from 6 to 12 V. The spectral data was normalized by setting the value at the wavelength with the highest radiance to 1. The voltage was also normalized by setting the current density value at 10 V to 1, thereby reducing the influence of variations in the device.

[0450] <Discriminant analysis> A discriminant analysis was performed based on the above dataset. Linear discriminant analysis (LDA) was performed using JMP 16.2, a statistical analysis software from SAS Institute Japan, Ltd. Details of discriminant analysis are provided, for example, in "Multivariate Data Utilization Techniques with JMP, Kaibundo Publishing Co., Ltd.", pp. 173-193 (Chapter 6).

[0451] Principal component analysis was performed on the measurement data, and mapping was performed using principal component 1 and principal component 2. Furthermore, correlation analysis was performed on the principal component mapping for each ion concentration with principal components 1 and 2. The results are shown in Figure 40.

[0452] The correlation with ion concentration on the principal component mapping was particularly pronounced for principal component 1, and when the correlation coefficient for the relationship between the common logarithm of the Mg ion concentration and principal component 1 was calculated by dividing their covariance by the product of their respective standard deviations, the correlation coefficient was -0.88. Thus, it became clear that the signal data obtained from this sensing system correlated strongly with the Mg ion concentration in the sample over an extremely wide range, from 10 ppb to 1000 ppm, demonstrating the effectiveness of this sensing system.

[0453] The JMP software was used in the following steps: 1. Load the following table (dataset) from the "File" menu. 2. Select rows 1, 9, and 17 of the table, right-click, and select "Exclude." 3. Select "Multivariate" - "Discriminant Analysis" from the "Analysis" menu. 4. Specify "Column 1" as "X, Category," "Remaining Columns" as "Y, Covariate," and specify "Linear Landscape Data" as the discriminant method. After selecting "Validate Model with Excluded Rows" in the advanced options, click the "OK" button to perform the calculation. 5. Subtract 3 from the "Number of Misclassifications" for "Exclusion" in "Score Summary" to obtain the number of correct answers. 6. Changed step 2 above to rows 2, 10, and 18 (changing the combination), and then performed steps 2 through 5. This was repeated a total of eight times.

[0454] This application claims priority based on Japanese Patent Application No. 2023-121744, filed on July 26, 2023. The contents of the specification and drawings of that application are incorporated herein by reference in their entirety.

[0455] "This invention provides the most fundamental technology for transforming mass production into on-demand production and long-term development into immediate development, in order to realize a recycling-oriented society. Its purpose is to describe, record, and evaluate with high sensitivity the state of substances related to traceability and ID in raw materials and manufacturing, such as in the manufacturing, processing, and quality assurance of chemicals, materials, bio-related substances, food, beverages, etc., which are based on the nm and Å scales, as well as in the control of wastewater and sewage treatment, and in agricultural crops, dairy products, and their processed products."

[0456] REFERENCE SIGNS LIST 1, 2 Electrode 3 Receptor layer 4, 4A, 4B Storage section 5 Target object 6 Luminescent probe 8 Substrate 9 Electrode region 10 Light source unit 11 Anode 12 Insulator 13 Light-emitting section 14 Cathode 15 Sealing member 16 Barrier material 17 Circuit 18 Power-receiving section 19 Receptor layer 20 Multiwell unit 21 Hardened layer 22 Microwell structure 31 Support 50 Organic matter 51 Microorganism (activated sludge) 100 Analysis device 110 Microarray device

Claims

1. An analysis method comprising the steps of: mixing an object and one or more types of luminescent probes whose luminescence behavior changes upon interaction with the object, and allowing them to interact; exciting the one or more types of luminescent probes that are in a state of interaction with the object to produce two or more types of luminescence, and simultaneously detecting the two or more types of luminescence; and analyzing the two or more types of luminescence detected, to identify the state of the object.

2. The analysis method according to claim 1, wherein the one or more types of luminescent probes have a luminescent portion that is excited by irradiation with excitation light and emits light, and the size of the luminescent portion is 100 nm or less.

3. The analysis method according to claim 2, wherein the step of allowing the object to interact with the one or more types of luminescent probes is a step of allowing the object to interact with two or more types of luminescent probes, and the sizes of the luminescent moieties possessed by the two or more types of luminescent probes are different from one another.

4. The analytical method according to claim 1, wherein the step of mixing the target substance with the luminescent probe and allowing them to interact with each other is carried out on the surface or inside of a receptive layer of an organic EL element having a pair of electrodes and a receptive layer sandwiched between them, and light is emitted from a light-emitting site formed on the surface or inside of the receptive layer after the step of mixing the target substance with the luminescent probe and allowing them to interact with each other.

5. The analytical method according to claim 4, wherein the receptive layer after the step of mixing and allowing the object and the luminescent probe to interact has a plurality of the luminescent sites, and at least one of the luminescent sites contains the luminescent probe and a host material, and at least one of the luminescent probe and the host material is of a different type from the other luminescent sites.

6. The analytical method according to claim 4, wherein in the step of mixing and allowing the object and the luminescent probe to interact with each other, a plurality of luminescent sites each having a different content of the object are formed in the receptive layer.

7. The analytical method according to claim 4, wherein in the step of mixing and interacting the object with the luminescent probe, the luminescent site further contains a substance that changes a carrier recombination position.

8. The analytical method according to claim 4, wherein in the step of mixing and allowing the object and the luminescent probe to interact with each other, an ink liquid containing at least one of the luminescent probe, a host material, the object, and a substance that changes a carrier recombination position is prepared and ejected from an inkjet head.

9. The receptor layer after the step of mixing the object and the luminescent probe and allowing them to interact has a plurality of the luminescent sites, and the receptor layer has the luminescent sites within 1 cm ２ The analysis method according to claim 8 , wherein the number of sites per region is 100 or more.

10. The analysis method according to claim 4, further comprising the step of preparing an organic EL element not containing the object as a comparison element, and in the step of identifying the state of the object, a difference between the light emission detected by the comparison element and the light emission detected by the organic EL element is obtained and analyzed as a relative value.

11. The analysis method according to claim 1, wherein the step of identifying the state of the object is a step of analyzing the two or more types of light emission using a pre-constructed trained model.

12. An analytical device having: a holding section including one or more storage sections for storing an object and one or more types of luminescent probes whose luminescence behavior changes upon interaction with the object; an excitation section for exciting the one or more types of luminescent probes placed in the storage sections; and a detection section for simultaneously detecting two or more types of light emitted by the one or more types of luminescent probes.

13. The analytical device according to claim 12, wherein the excitation section has a light source for exciting the luminescent probe, and the holding section is a plate having the container section.

14. The analysis device according to claim 12, wherein the excitation section has a pair of electrodes, and the holding section is sandwiched between the pair of electrodes.

15. The analysis device according to claim 12, further comprising an analysis unit that analyzes information on the two or more types of light detected by the detection unit using a pre-constructed trained model, and analyzes a state of the object.

Images