Patent Application
20250052673
The present invention relates to a solution for non-destructive quantification of one or more chemical substances in a matrix comprising coating and bulk material in a sample, for example coated seeds, using Infrared Spectroscopy data of the sample and multivariate data analysis.
In order to ensure seed health during storage, seeding, imbibition and in the early phase of plant growth, seeds are usually coated with biological, physical and chemical agents. Formulations of these individual agents or mixtures thereof are often denoted “seed dressings” or “seed coatings”. The applied agents (more generally referred to as chemical substance(s) of interest or active ingredient(s)) may be pesticides, fertilizer, nutrients (including inoculants), biologics, or mixtures thereof. Other compounds may be coated on seeds to provide a specific function which may be, but is not limited to, identification, coloration, drying, flowability, etc.
For all the agents that may be coated on seeds the determination of the actual loading may be necessary for quality, regulatory, efficiency or cost reasons. Several methods are available to quantify or semi-quantify the afore mentioned agents on seeds, e.g. visual inspection, photometric methods, HPLC, GC etc., As for all analytical tools, these described methods strongly differ in their physical nature (read out parameter), accuracy, ease of handling, time demand, steps, etc. If complex mixtures are coated on seeds, the current standard to achieve high accuracy for agent quantification is always based on chromatographic separation (ref. HPLC, UPLC, GC). These methods are also called reference method for analysis. Any chromatographic method for the quantification of actives on seeds will require a solvent extraction step in order to transfer the agent into a measurable form. The extraction process itself is characterized by a couple of disadvantages e.g.: a) high effort in hands-on work and resources, b) requires solvents or water-based extraction mixtures, c) need to consider physico-chemical stability of agent in extraction medium, d) not suitable if non-reproducible or highly time-dependent partitioning of active in the extraction media. Furthermore, currently employed HPLC methods for active analysis often cover one agent (or one active ingredient) in a multi-agent recipe. This is mainly due to the effort of sample preparation, individual partitioning, stability properties of the agent and associated cost.
The use of spectroscopic means for identification and quantification of chemical substances is known for many years. Spectroscopic methods have been described as a non-destructive method (spectroscopic surface measurement—IR either mid-IR or near-IR) with little or no sample preparation for analysis of one pesticide on coated seeds. Pigeon et al. describe such a method wherein spectra of a coated seed sample are acquired from 400 to 2500 nm in reflection mode. Quantification requires calibrations consisting in measuring spectra of reference samples of coated seeds with pesticide of interest; said reference samples are selected based on the repartition of spectra in a multidimension space using Principal Component Analysis (PCA). Said reference samples are analyzed by a reference method (chromatography); NIR Spectra for a reference sample are acquired seed by seed. Collected data are used to establish a predictive model connecting spectral data to reference analytical results (calibration equations) using the Modified Partial Least Square regression method (MPLS). Calibration equations are validated using further samples of known compositions (validation samples and related validation data). The validated calibration equations can be used to analyze unknown samples of the coated seeds; calibration is specific for seed and pesticide. [O. Pigeon et al. “NEAR INFRARED SPECTROSCOPY (NIR): A NON-DESTRUCTIVE, RAPID AND ACCURATE METHOD TO DETERMINE IMIDACLOPRID ON SUGAR BEET PELLETED SEEDS, Proceedings of the 64th IIRB Congress, June 209.1, Bruges (B); O. Pigeon et al., Study of the Quality of Seed Treatments with Plant Protection Products using Near Infrared Spectroscopy. PhD thesis, Faculté Universitaire des Sciences Agronomiques de Gembloux, Belgium (2003)].
However, the method of Pigeon et al. is limited to the quantification of only one active ingredient on coated seeds.
U.S. Pat. No. 5,900,944A describes a solution for indirect quantitative spectral analysis of pesticides with fluorescent pigments, wherein pigment is quantified on cotton seeds coated with a mixture of Gaucho® (imidacloprid) and a fluorescent pigment. The spectral analysis includes the pesticide analysis device emitting white light incident on a seed sample and detecting the transmitted light therefrom, wherein the detected light is in a range of 400-700 nm (for detection of the selected pigment). An analysis of the spectral data is performed to determine absorbance and color saturation data by the pesticide analysis device. Quantification is achieved using linear relationships between color saturation and quantity of pigment or pigment, obtained by conducting series of spectral tests on seeds samples having differing known ratios of pesticide to pigment.
Using a pigment in seed coating for the purpose of quantification is not desirable. Quantification of multiple active ingredients with such linear calibration method is not possible.
Influence of size and form of the seeds, of background, multiple active ingredients and coating components beyond active ingredients are not considered.
Considering seed dressings are very often complex mixtures that are applied as a very thin layer (usually few nm to μm) on the seeds, based on volume, most of the spectral information of a sample is related to the seed itself. In addition, the complex mixture of components of the seed dressing will cause a superimposing of the individual compound specific spectra, most challenging for simple data evaluation. As the seeds have a certain variation in size and shape and the treatment is often not 100% equally distributed on the seeds, the spectroscopic method must also be able to handle these inhomogeneities.
There is therefore a need for a solution which is able to distinguish the spectral information originating from the seeds, or more generally originating from the matrix comprising the bulk material and the coating ingredients other that the chemical substance(s) of interest, and the spectral signature of the actual chemical substance(s) of interest. Such solution should be a facile and fast, non-destructive method for full quantification of all active ingredients using data collected in one sample measurement run, solving the problems mentioned above.
The object underlying the invention is achieved by the combination of features according to independent claims. Exemplary embodiments of the invention can be gathered from dependent claims.
The invention is described in more detail without any division within the subject matter of the invention (process and system). The explanations below are intended to be applicable analogously to all the subject matters of the invention, in either context (process or system).
In the present invention the problem mentioned above was overcome by a computer-implemented method for non-destructive quantification of one or more chemical substances of interest on coated bulk material in a matrix, said matrix being defined by its coating components and said bulk material, said method comprising the following steps:
The term “loading value” as used herein means weight of active agent or volume of seed treatment composition (also referred to as formulated active agent) per unit coated bulk material. Weight or volume per unit for coated seeds may be defined for 1 seed, 60.000 seeds, kg, dt as needed.
In an example of the invention, spectroscopic data for a sample is acquired by RAMAN, mid-infrared or near-infrared spectroscopy, preferred mid-IR or near-IR, typically by acquiring one or more spectra representative of a sample in a corresponding spectroscopic apparatus in reflection mode. The term “infrared” as used herein is intended to encompass the spectral range above approximately 200 cm−1 and includes both the near IR and the (mid-)IR. The term “-mid-IR”, as used herein, refers to the spectral range from about 200 cm−1 to 4000 cm−1, while the term “near-IR” refers to the spectral range from about 4000 cm−1 to 12000 cm−1 and the term “RAMAN” refers to the spectral range from about 0 nm to 2500 nm
In an example more than one spectrum is acquired for the sample. Such plurality of spectra may be obtained by acquiring spectra from different subsets of a sample, stirring, rotating the sample in the spectroscopic apparatus or a combination thereof, so that more than one spectrum of the sample is representative of the sample. Spectra may be averaged before further treatment or each spectrum can be used for quantification and results can be averaged for the sample loading value.
The term “chemical substance to be quantified” or “chemical substance of interest” as used herein refers to either one active agent of interest in a coating formulation or several active agents in a combination formulation for coating (i.e. CropStar® which comprises Imidacloprid and Thiodicarb). In case a combination formulation is used, seed loading for the combination formulation is quantified. In case a seed treatment package (also called applied plant protection product) comprising multiple seed dressings—single substance dressing, combination formulations or a combination thereof—is used, the loading value for the respective seed dressings can be obtained with the method of the invention. For example, for seeds treated with a standard seed treatment package comprising the combination formulations CropStar® (i.e. Imidacloprid+Thiodicarb), and Derosal Plus® (i. e. Carbendazim+Thiram), loading values for CropStar® and for Derosal Plus® can be obtained with the solution of the invention.
The term “matrix” as used herein is given a broad meaning and comprises all elements other as the chemical substance to be quantified, in particular coating and/or treatment material as well as the bulk material, on which the treatment and/or the coating is applied. The coating is typically applied in a continuous solid phase and comprises one or more binder compounds and vacancies, voids or spaces occupied by the active agent(s) and filler. The term matrix may include what may be viewed as a matrix system, a reservoir system or a microencapsulated system comprising the bulk material. In general a ‘matrix system’ consists of one or more active ingredient(s) and filler uniformly dispersed within a polymer. A “reservoir system” consists of a separate active agent phase, active agent particles physically dispersed within a surrounding, rate limiting polymeric phase. Microencapsulation includes the coating of small particles. The term microencapsulation has not only been applied to coated particles but also to dispersions in a solid matrix. Without being limited to the specific encapsulating system (matrix, reservoir or microencapsulated) the term matrix is meant to be inclusive of the above listed systems.
In an example, the present invention is used for samples of coated seeds. This preferred embodiment is explained in somewhat greater detail below, but without any intention of restricting the invention to this embodiment.
In an example the more than one spectrum is acquired for a sample made of a plurality of seeds.
The term “seed” as used herein refers to the ripened ovule of gymnosperms and angiosperms, which contains an embryo surrounded by a protective cover. In particular, the term covers cereal kernels. The protective cover can comprise the seed coat (testa). Some seeds comprise a pericarp or fruit coat around the seed coat. In particular when this layer is closely adhered to the seed, as in cereal kernels, it is in some cases referred to as a caryopsis or an achene. As used herein, the term “seed coat” includes a caryopsis or an achene. In practical terms, the term “seed” includes but is not restricted to anything that can be planted in agriculture to produce plants, including pelleted seeds, true seeds, plant seedlings, rootstock, plant cuttings and plant parts such as a tuber or bulb.
The seed may be of the order of Monocotyledoneae or Dicotyledoneae. Virtually the sample may be of any plant seed, such as cereals, vegetables, ornamentals, and fruits. Particular plant seeds are selected from the group of corn (sweet and field), soybean, wheat, barley, oats, rice, maize, potato, sugar cane, sugar beet, cotton, sunflower, alfalfa, sorghum, rapeseed, Brassica spp., tomato, pepper, cucumber, melon, watermelon, onion, lettuce, spinach, leek, bean, carrot, tobacco and flower seed, for example, pansy, impatiens, petunia and geranium. Crop plants can be plants which can be obtained by conventional breeding and optimization methods or by biotechnological and genetic engineering methods or combinations of these methods, including the transgenic plants and including the plant varieties which can or cannot be protected by varietal property rights.
A wide range of materials and/chemical substances is used in seed surface treatments or coatings.
The term “surface treatment” as herein refers to a selective modification of the outer surface of the seed, on which a coating can be applied, typically without substantially modifying inner parts of the seed.
The term “coating” as used herein refers broadly to applying material to a surface of a seed, for instance as a layer of a material around a seed. Coating includes but is not limited to formulated active agent, film coating, pelleting, and encrusting. Pellets obtained with pelleting are also known as seed pills. The coating is preferably applied over substantially the entire surface of the seed, such as over 90% or more of the surface area of the seed, to form a layer. However, the coating may be complete or partial, for instance over more than 20%, or more than 50% of the surface area of the seed.
The terms “seed treatment composition”, “seed coating composition” or “formulated active agent” refer to a composition to be used for treatment or coating of seed comprising active agents, possibly after combination of the composition with other compositions, such as Plant Protective Products (PPP) formulations and/or diluents such as water. Hence, the term includes as well as coating formulation which are not yet mixed with PPP formulations and/or not yet applied as treatment or coating. The application of one or more active agent(s) as a dust, slurry or the like is a well-known practice in the art and is considered a coating within the meaning of the term used herein.
In an embodiment the solution of the invention is used for treated seeds, seeds coated with one or more film layers or seeds after treatment and coating process, means treated and coated seeds.
Typically, the one or more chemical substances to be quantified in a non-destructive manner in such seed samples are biological active agents.
The biologically active agent may be an agent selected to protect the seed from pests, bacteria, fungi, nematodes, birds other animals or selected to promote growth, such as an antimicrobial agent, bactericide, pesticide, acaricide, insecticide, fungicide, nematicide, molluscicides, repellent, ovicide, rodenticide, herbicide, herbicidal safener, microbiological, fertilizer, phytotonic, sterilant, safeners, semiochemical, plant defense modulator, plant growth regulator, nutrients, soil conditioning agents selected to promote germination and/or growth.
The biologically active agent may be employed as a mixture comprising one or more active agents. Suitable pesticides include those listed herein and those listed in The Pesticide Manual, 9th Ed., Editor, Charles Worthing, published by the British Crop Protection Council or can be found on the Internet (e.g. http://www.alanwood.net/pesticides). The classification is based on the current IRAC Mode of Action Classification Scheme at the time of filing of this patent application.
A combination of active agents may be coated in separate layers or alternatively may be combined in one or more coating layers.
All named mixing partners can, if their functional groups enable this, optionally form salts with suitable bases or acids, in particular in the form of biologically acceptable salts such as sodium, potassium, ammonium and the like.
Examples for herbicides are:
Examples for plant growth regulators are:
Examples of active compounds which may be mentioned as fungicide which are known from the literature are the following (compounds are either described by “common name” in accordance with the International Organization for Standardization (ISO) or by chemical name or by a customary code number), and always comprise all applicable forms such as acids, salts, ester, or modifications such as isomers, like stereoisomers and optical isomers. As an example at least one applicable form and/or modifications can be mentioned.
The active ingredients specified herein by their Common Name are known and described, for example, in The Pesticide Manual (16th Ed.British Crop Protection Council) or can be searched in the internet (e.g. www.alanwood.net/pesticides).
Where a compound (A) or a compound (B) can be present in tautomeric form, such a compound is understood herein above and herein below also to include, where applicable, corresponding tautomeric forms, even when these are not specifically mentioned in each case.
All named mixing partners of the classes (1) to (15) can, if their functional groups enable this, optionally form salts with suitable bases or acids.
Following groups of compounds are, for example, to be considered as safeners:
The active ingredients specified herein by their “common name” are known and described, for example, in the Pesticide Manual (“The Pesticide Manual”, 14th Ed., British Crop Protection Council 2006) or can be searched in the internet (e.g. http://www.aluiwoodiet/pesticides).
Further active ingredients with unknown or uncertain mode of action, for example Afidopyropen, Afoxolaner, Azadirachtin, Benclothiaz, Benzoximate, Bifenazate, Broflanilide, Bromopropylate, Chinomethionat, Cryolite, Cyclaniliprole, Cycloxaprid, Cyhalodiamide Dicloromezotiaz, Dicofol, Diflovidazin, Flometoquin, Fluazaindolizine, Fluensulfone, Flufenerim, Flufenoxystrobin, Flufiprole, Fluhexafon, Fluopyram, Fluralaner, Fluxametamide, Fufenozide, Guadipyr, Heptafluthrin, Imidaclothiz, Iprodione, Lotilaner, Meperfluthrin, Paichongding, Pyflubumide, Pyridalyl, Pyrifluquinazon, Pyriminostrobin, Sarolaner, Tetramethylfluthrin, Tetraniliprole, Tetrachlorantraniliprole, Tioxazafen, Thiofluoximate, Triflumezopyrim and Iodomethane; furthermore products based on Bacillus firmus (including but not limited to strain CNCM I-1582, such as, for example,VOTiVO TM, BioNem) or one of the following known active compounds: 1-{2-fluoro-4-methyl-5-[(2,2,2-trifluorethyl)sulfinyl]phenyl}-3-(trifluoromethyl)-1H-1,2,4-triazol-5-amine (known from WO2006/043635), {1′-[(2E)-3-(4-chlorophenyl)prop-2-en-1-yl]-5-fluorospiro[indole-3,4′-piperidin]-1(2H)-yl}(2-chloropyridin-4-yl)methanone (known from WO2003/106457), 2-chloro-N-[2-{1-[(2E)-3-(4-chlorophenyl)prop-2-en-1-yl]piperidin-4-yl}-4-(trifluoromethyl)phenyl]isonicotinamide (known from WO2006/003494), 3-(2,5-dimethylphenyl)-4-hydroxy-8-methoxy-1,8-diazaspiro[4.5]dec-3-en-2-one (known from WO2009/049851), 3-(2,5-dimethylphenyl)-8-methoxy-2-oxo-1,8-diazaspiro[4.5]dec-3-en-4-yl ethyl carbonate (known from WO2009/049851), 4-(but-2-yn-1-yloxy)-6-(3,5-dimethylpiperidin-1-yl)-5-fluoropyrimidine (known from WO2004/099160), 4-(but-2-yn-1-yloxy)-6-(3-chlorophenyl)pyrimidine (known from WO2003/076415), PF1364 (CAS-Reg.No. 1204776-60-2), methyl 2-[2-({[3-bromo-1-(3-chloropyridin-2-yl)-1H-pyrazol-5-yl]carbonyl}amino)-5-chloro-3-methylbenzoyl]-2-methylhydrazinecarboxylate (known from WO2005/085216), methyl 2-[2-({[3-bromo-1-(3-chloropyridin-2-yl)-1H-pyrazol-5-yl]carbonyl}amino)-5-cyano-3-methylbenzoyl]-2-ethylhydrazinecarboxylate (known from WO2005/085216), methyl 2-[2-({[3-bromo-1-(3-chloropyridin-2-yl)-1H-pyrazol-5-yl]carbonyl}amino)-5-cyano-3-methylbenzoyl]-2-methylhydrazinecarboxylate (known from WO2005/085216), methyl 2-[3,5-dibromo-2-({[3-bromo-1-(3-chloropyridin-2-yl)-1H-pyrazol-5-yl]carbonyl}amino)benzoyl]-2-ethylhydrazinecarboxylate (known from WO2005/085216), N-[2-(5-amino-1,3,4-thiadiazol-2-yl)-4-chloro-6-methylphenyl]-3-bromo-1-(3-chloropyridin-2-yl)-1H-pyrazole-5-carboxamide (known from CN102057925), 8-chloro-N-[(2-chloro-5-methoxyphenyl)sulfonyl]-6-(trifluoromethyl)imidazo[1,2-a]pyridine-2-carboxamide (known from WO2009/080250), N-[(2E)-1-[(6-chloropyridin-3-yl)methyl]pyridin-2(1H)-ylidene]-2,2,2-trifluoroacetamide (known from WO2012/029672), 1-[(2-chloro-1,3-thiazol-5-yl)methyl]-4-oxo-3-phenyl-4H-pyrido[1,2-a]pyrimidin-1-ium-2-olate (known from WO2009/099929), 1-[(6-chloropyridin-3-yl)methyl]-4-oxo-3-phenyl-4H-pyrido[1,2-a]pyrimidin-1-ium-2-olate (known from WO2009/099929), 4-(3-{2,6-dichloro-4-[(3,3-dichloroprop-2-en-1-yl)oxy]phenoxy}propoxy)-2-methoxy-6-(trifluoromethyl)pyrimidine (known from CN101337940), N-[2-(tert-butylcarbamoyl)-4-chloro-6-methylphenyl]-1-(3-chloropyridin-2-yl)-3-(fluoromethoxy)-1H-pyrazole-5-carboxamide (known from WO2008/134969), butyl [2-(2,4-dichlorophenyl)-3-oxo-4-oxaspiro[4.5]dec-1-en-1-yl]carbonate (known from CN 102060818), 3E)-3-[1-[(6-chloro-3-pyridyl)methyl]-2-pyridylidene]-1,1,1-trifluoro-propan-2-one (known from WO2013/144213), N-(methylsulfonyl)-6-[2-(pyridin-3-yl)-1,3-thiazol-5-yl]pyridine-2-carboxamide (known from WO2012/000896), N-[3-(benzylcarbamoyl)-4-chlorophenyl]-1-methyl-3-(pentafluoroethyl)-4-(trifluoromethyl)-1H-pyrazole-5-carboxamide (known from WO2010/051926), 5-bromo-4-chloro-N-[4-chloro-2-methyl-6-(methylcarbamoyl)phenyl]-2-(3-chloro-2-pyridyl)pyrazole-3-carboxamido (known from CN103232431), ), Tioxazafen, 4-[5-(3,5-dichlorophenyl)-4,5-dihydro-5-(trifluoromethyl)-3-isoxazolyl]-2-methyl-N-(cis-1-oxido-3-thietanyl)-benzamide, 4-[5-(3,5-dichlorophenyl)-4,5-dihydro-5-(trifluoromethyl)-3-isoxazolyl]-2-methyl-N-(trans-1-oxido-3-thietanyl)-benzamide and 4-[(5S)-5-(3,5-dichlorophenyl)-4,5-dihydro-5-(trifluoromethyl)-3-isoxazolyl]-2-methyl-N-(cis-1-oxido-3-thietanyl)benzamide (known from WO 2013050317 A1), N-[3-chloro-1-(3-pyridinyl)-1H-pyrazol-4-yl]-N-ethyl-3-[(3,3,3-trifluoropropyl)sulfinyl]-propanamide, (+)-N-[3-chloro-1-(3-pyridinyl)-1H-pyrazol-4-yl]-N-ethyl-3-[(3,3,3-trifluoropropyl)sulfinyl]-propanamide and (−)-N-[3-chloro-1-(3-pyridinyl)-1H-pyrazol-4-yl]-N-ethyl-3-[(3,3,3-trifluoropropyl)sulfinyl]-propanamide (known from WO 2013162715 A2, WO 2013162716 A2, US 20140213448 A1), 5-[[(2E)-3-chloro-2-propen-1-yl]amino]-1-[2,6-dichloro-4-(trifluoromethyl)phenyl]-4-[(trifluoromethyl)sulfinyl]-1H-pyrazole-3-carbonitrile (known from CN 101337937 A), 3-bromo-N-[4-chloro-2-methyl-6-[(methylamino)thioxomethyl]phenyl]-1-(3-chloro-2-pyridinyl)-1H-pyrazole-5-carboxamide, (Liudaibenjiaxuanan, known from CN 103109816 A); N-[4-chloro-2-[[(1,1-dimethylethyl)amino]carbonyl]-6-methylphenyl]-1-(3-chloro-2-pyridinyl)-3-(fluoromethoxy)-1H-Pyrazole-5-carboxamide (known from WO 2012034403 A1), N-[2-(5-amino-1,3,4-thiadiazol-2-yl)-4-chloro-6-methylphenyl]-3-bromo-1-(3-chloro-2-pyridinyl)-1H-pyrazole-5-carboxamide (known from WO 2011085575 A1), 4-[3-[2,6-dichloro-4-[(3,3-dichloro-2-propen-1-yl)oxy]phenoxy]propoxy]-2-methoxy-6-(trifluoromethyl)-pyrimidine (known from CN 101337940 A); (2E)- and 2(Z)-2-[2-(4-cyanophenyl)-1-[3-(trifluoromethyl)phenyl]ethylidene]-N-[4-(difluoromethoxy)phenyl]-hydrazinecarboxamide (known from CN 101715774 A); 3-(2,2-dichloroethenyl)-2,2-dimethyl-4-(1H-benzimidazol-2-yl)phenyl-cyclopropanecarboxylic acid ester (known from CN 103524422 A); (4aS)-7-chloro-2,5-dihydro-2-[[(methoxycarbonyl)[4-[(trifluoromethyl)thio]phenyl]amino]carbonyl]-indeno[1,2-e][1,3,4]oxadiazine-4a(3H)-carboxylic acid methyl ester (known from CN 102391261 A).
Preferred active compounds are selected from the group comprising SDH-Inhibitors, nAChR-Agonists (including neonicotinoides), chlorotica including PDS inhibitors (HRAC F1) and HPPD inhibitors (HRAC F2) and thiadiazole carboxamides/host defence inducers.
Typical bactericidal agents include streptomycin, penicillins, tetracyclines, ampicillin, and oxolinic acid.
Coating compositions are often packaged, stored and/or transported and only thereafter combined with formulations such as plant enhancing agents, for example growth regulators and/or plant stimulants.
The loading of active agent to be added in the seed coating composition typically ranges from about 0.001 to about 10% of the weight of the seed and preferably from about 0.01 to 2.0%. However, for a particular situation the amounts may be greater or less. For example, a fungicide can be included in the seed coating composition in an amount of 0.0001-10 wt. %, based on the total weight of/kg of the coated seed.
Biological treatment may also be used to enhance seed performance and help in the control of harmful organisms.
In an example, the matrix may also comprise biocontrol agents such as bacteria of the genera Rhizobium, Bacillus, Pseudomonas, and Serratia, fungi of the genera Trichoderma, Glomus, and Gliocladium and mycorrhizal fungi.
The above summarized lists are necessarily not exhaustive, new active agents and or active ingredients are continuously developed and can be incorporated in a seed treatment or coating composition.
Seed treatment or coating compositions and therefore the matrix in the sense of the present invention typically also comprise one or more binders, one or more filler (s), and an effective amount of one or more active agents.
The binder serves as a matrix for the one or more active agents and is preferably present in the seed coating in an amount adequate to ensure adherence of the entire matrix and all active agents to the seed and/or prevent or reduce the levels of phytotoxicity caused by the one or more active agent(s). The specific binder(s) will depend on the properties of the one or more active agents.
The binder component of the coating is composed preferably of an adhesive polymer that may be natural or synthetic and shows no negative, adverse effects to environment, seeds or human safety. The binder may be selected from polyvinyl acetates, polyvinyl acetate copolymers (-ethylene), polyvinyl alcohols, polyvinyl alcohol copolymers, celluloses, including ethylcelluloses and methylcelluloses, hydroxymethylcelluloses, hydroxypropylcellulose, hydroxymethylpropylcelluloses, polyvinylpyrolidones, dextrins, maltodextrins, polysaccharides, fats, oils, proteins, gum arabics, shellacs, vinylidene chloride, vinylidene chloride copolymers, calcium lignosulfonates, acrylic copolymers, starches, polyacrylates, polyvinylacrylates, polyurethans, zeins, gelatin, carboxymethylcellulose, chitosan, polyethylene oxide, polystyrenes, polybutadienes, acrylimide polymers and copolymers, polyhydroxyethyl acrylate, methylacrylimide monomers, alginate, ethylcellulose, polychloroprene and syrups, combinations of polyvinyl alcohol and sucrose or mixtures thereof. Preferred binders include polymers and copolymers of vinyl acetate, methyl cellulose, polyvinyl alcohol, vinylidene chloride, acrylic, cellulose, polyvinylpyrrolidone and polysaccharide or combination thereof. Particularly used classes of polymers include polymers and copolymers of vinylidene chloride and vinyl acetateethylene copolymers. Further binders may be molasses, granulated sugar, alginates, karaya gum, jaguar gum, tragacanth gum, polysaccharide gum, mucilage or combination thereof.
The amount of binder in the coating may be in the range of about 0.01 to 15% of the weight of the seed. A preferred range will be about 0.1 to 10.0% of the weight of the seed.
The matrix of the present invention may comprise fillers for seed coatings, which may be absorbent or inert fillers. Fillers are known in the art and may include wood-flours, clays, activated carbon, sugars, diatomaceous earth, cereal flours, fine-grain inorganic solids, calcium carbonate and the like. Clays and inorganic solids which may be used include calcium bentonite, kaolin, china clay, talcum, perlite, vermiculite, mica, silicas, quartz powder, montmorillonite, other state of the art gloss components and mixtures thereof. Sugars which may be used include dextrin and maltodextrin. Cereal flours include wheat flour, oat flour and barley flour. Preferred fillers include diatomaceous earth, perlite, silica and calcium carbonates and mixtures thereof. One skilled in the art will appreciate that this is a non-exhaustive list of materials and that other recognized filler materials may be used depending on the seed to be coated and the one or more active agent(s) used in the coating.
If present, the filler is typically chosen so that it will provide a proper microclimate for the seed, for example the filler is used to increase the loading rate of the active ingredient and to adjust the control-release of the active ingredient. A filler aids in the production or process of coating the seed. The effect varies, because in some instances formulated active agent compounds will comprise a filler. The amount of filler used may vary, but generally the weight of the filler components will be in the range of about 0.005 to 70% of the seed weight, more preferably about 0.01 to 50% and most preferably about 0.1 to 15%.
In an example the coated seed comprises one or more binder(s) in an amount from about 0.01 to about 15% of the weight of the seed, a filler in an amount of up to about 70% of the weight of the seed, one or more active agents in an amount from about 0.005 to about 50% of the weight of the seed.
In some embodiments the matrix may comprise a plasticizer. Plasticizers are typically used to make the film that is formed by the coating layer more flexible, improve adhesion, spreadability and improve the speed during processing. The improved film flexibility is important to minimize chipping, breakage or flaking during handling or sowing processes. Many plasticizers may be used however, most common plasticizers include polyethylene glycol, glycerol, butylbenzylphthalate, glycol benzoates, propylene glycol, polyglycols and related compounds.
Conventional means of coating may be used for carrying out the coating; various coating machines are available to one skilled in the art. Three well known techniques include the use of drum coaters, fluidized bed techniques or spouted beds. After coating the dried seeds are optionally sized by transfer to a sizing machine before or after coating.
Film-forming compositions for enveloping coated seeds are well known in the art, and a film overcoating can be optionally applied to the coated seeds. The film overcoat protects the coating layers and optionally allows for easy identification of the treated seeds. In general, additives are dissolved or dispersed in a liquid adhesive, usually a polymer into or with which seeds are dipped or sprayed before drying. Alternatively, a powder adhesive can be used. Various materials are suitable for overcoating including but not limited to, methyl cellulose, hydroxypropylmethylcellulose, dextrin, gums, waxes, vegetable or paraffin oils; water soluble or water disperse polysaccharides and their derivatives such as alginates, starch, and cellulose; and synthetic polymers such as polyethylene oxide, polyvinyl alcohol and polyvinylpyrrolidone and their copolymers and related polymers and mixtures of these. Further materials may be added to the overcoat including optionally plasticizers, colorants, brighteners, wetting agents and surface active agents such as, dispersants, emulsifiers and flow agents including for example, calcium stearate, talcum and vermiculite.
The matrix may further comprise antifoaming agents, antiseptics, thickening agents, dispersants, anti-freeze agents, adhesive agents, and the like. Additives and said further material are also together referred to as functional agents. The skilled person will appreciate that the above components are listed as examples and are not intended to be an exhaustive list of components that can be used in one or more seed coating layers. However, the examples listed above show the extreme variability of a matrix in the sense of the present invention and one skilled in the art will recognize that a facile and useable method for quantification of one or more chemical substances in such complex matrices should be able to cope with such complexity.
The solution of the present invention makes use of a multivariate correlation model trained for computing a loading value of the chemical substance(s) of interest based on a signature spectrum relevant for the chemical substance(s) at stake in consideration of matrix influences.
The term “matrix influences” as used herein comprises information related to the sample in relation with:
Further “matrix influences” information may relate to physical property parameters of seeds such as bulk flowing properties, dust binding strength or dust value, for example as measured using a Heubach dust-meter device according to Euroseeds reference method “Assessment of free-floating dust and abrasion particles of treated seeds as a parameter of the quality of treated seeds”.
Most relevant matrix influences on the spectroscopic evaluation were found to be seed variety, hybrids or traits, seed size, shape and coating composition, coating components and their dosage.
The skilled person understands that features of various embodiments may be combined with each other. For the correlation of the results of the spectroscopic measurements to the loading of the chemical substance of interest on the seeds, the actual amount of the chemical substance of interest is very important. This actual amount is measured for training and validation samples using one of the reference quantification methods established in the field, in particular, using chromatographic methods such as High Pressure Liquid Chromatography (HPLC), Ultra Performance Liquid Chromatography (UPLC) or Gas Chromatography (GC), gravimetric methods based on the applied product masses, or mass spectrometry.
Correlating a loading value with high error in the reference quantification method may lead to misinterpretation of the results. The method therefore requires accurate calibration depending on the matrix and the chemical substance to be quantified. A multivariate calibration was developed which comprises conducting spectrum pre-treating steps and a multivariate data analysis using a multivariate correlation model trained for computing a loading value of the chemical substance(s) based on a signature spectrum relevant for the chemical substance(s) at stake in consideration of matrix influences.
For simultaneous analysis of multiple active agents, individual calibration for each chemical substance to be quantified or for each combination formulation to be quantified, as the case may be, is used. Validations were designed and carefully developed to prove cross-insensitivity between the chemical substance/combination formulation to be quantified and the matrix components. In other words, several, individual calibrations are used if more than one chemical substance or combination formulation is to be quantified.
A further object of the present invention is therefore a computer-implemented method for provision of a calibration for a chemical substance to be quantified in a matrix, said matrix being defined by its further coating components and said bulk material, according to Claim 10. Exemplary embodiments of the method can be gathered from Claims 11 to 12.
Said calibration comprises conducting spectrum pre-treating steps and a multivariate data analysis using a multivariate correlation model for the correlation of a near-infrared, infrared or a Raman spectrum signature relevant for the chemical substance in the matrix and a reference loading value for the chemical substance in said matrix in consideration of matrix influences Said method comprises the following steps:
In an example and for the selection of the training data the matrix influences taken into consideration for the setting of the design of experiment (DoE) may comprise the following samples groups:
Multiple-active compositions might show variations of the active agents within the specification. This might lead to a change in the ratio of the active agents and possibly to an influence on the readout of the spectroscopic methods.
The seed itself may bring a strong background information into the spectra. As there are many hybrids or varieties available that could be used for seed treatment, it is preferred to clarify a possible influence of this variation.
In an example, ten other varieties of seeds with high market share can be measured both with and without treatment.
The size or weight of the seed itself may bring variation into the spectra. It is preferred to clarify a possible influence of this variation.
Drying agents and/or components of gloss finishing may bring a strong background information into the spectra. In an example, to check a possible influence of use of talcum alone and/or in a gloss finishing, respective samples, in which talcum comprising gloss finishing is used, can be considered in the training samples and in DoE reference samples.
The method for DoE can be chosen to ensure independent variation if the matrix influences. For the Design of Experiment method, the solution for “Calibration design” in the software OPUS from the company Bruker Corporate can be used to determine the optimal training loading distribution for the components without collinearity for a certain range. Random independent loading values are provided for the individual components.
In an example, the training examples can also comprise reference samples coated/treated with pure active agent of interest or pure other matrix components. Signals in the spectra of the coated/treated seeds can thus be correlated to the individual active agents, enabling a robust interpretation of the data.
In an example the training samples can also comprise pure seed sample without treatment/coating.
In an example normalizing the plurality of spectrum of the training data set can be achieved by way of at least first derivation, second derivation, straight line subtraction, offset correction, standard normal variate (SNV), detrend (DET), standard normal variate and detrend (SNVD), Minimum-Maximum normalization (MIN/MAX), multiplicative scatter correction (MSC), weighed multiplicative scatter correction (WMSC) or a combination thereof.
Preferred methods of pre-treatment are MCS and/or MIN/MAX, optionally in combination with first derivation.
In an example a spectral range of interest for one or more active agent(s) may be predefined or determined by way of variance analysis of the spectral signatures of the reference samples spectra or by way of comparison between spectra of samples with formulated active agent(s) and spectra of samples with pure active agent. Adequate method for variance analysis may be selected from the group comprising Principle Component Analysis (PCA), Singular Value Decomposition (SVD), Multivariate Curve Resolution (MCR) and others. Preferred method is PCA.
“Multivariate data analysis” as used herein comprises a set of statistical techniques used for analysis of data such as Partial Least Square Regression (PLS), Multi Linear Regression (MLR), Support Vector Machine Regression (SVM) and others. Method PLS was chosen because it is commonly used in the field of spectroscopy in chemical analysis.
In an example, optimization of step e) comprises varying the normalizing method or the combination of normalizing methods mentioned above and reiterating training steps (c and d) mentioned above based on the newly normalized spectrum signatures.
In a further example, optimization of step e) may also comprise varying the spectral range of interest.
In a further example, optimization of step e) may also comprise reducing the complexity of the multivariate correlation model for example by way of eliminating matrix influences. Such elimination may be achieved by way of excluding spectral ranges from the multivariate calibration models which have no information or information not relevant for the chemical substance of interest.
In a further example, optimization of step e) may also comprise reducing the complexity of the multivariate correlation model for example by way of reducing the number of independent main components being used in the multivariate correlation model. Such reduction may be achieved by interpretation of the independent main components and influence thereof on the loading prediction, selection of these independent main components in the calibration method neglecting the ones only showing non-relevant information like noise.
In a further example, optimization of step e) may comprise modifying the training data by eliminating samples showing weak correlation between the readout of the multivariate calibration model and the reference method (outlier elimination). Such elimination can be justified for example by identification of errors in the reference method or low spectral quality of the sample spectrum.
In an embodiment the multivariate correlation model is validated using a method comprising the following steps:
In an example the method of the invention further comprises validating the computed loading value of the chemical substance(s) by conducting the following steps:
Preferred value for comparability is the mahanalobis distance, describing the difference of the computed spectral signature of the sample and the distribution of the spectral signatures of the whole set of calibration samples is computed.
Use of validation samples which are representative for samples collected from a production plant can ensure that the method is robust in the common loading ranges of interest and in the common variation range for the sample matrix.
Use of validation samples selected by design of experiment to be at the boundaries of the variation room can be used to validate the robustness of the method outside of the common production ranges.
A further aspect of the invention is to determine both the accuracy and the robustness of the spectroscopic methods of the invention. As the accuracy of the existing method HPLC, including of sample preparation per extraction, is not easy to determine and not available in many cases it was agreed to aim for a correlation between the spectroscopic methods and HPLC in a region acceptable in the field. Acceptable in the field of seed treatment is an accuracy from 0.1 to 10%.
The solution of the invention may be used not only for active quantification in these complex films but also for quantification of any functional compound of the seed treatment or coating. The thoroughly developed multivariate data analysis method of the invention, applied to the acquired infrared spectra, was shown to be able to cope with the spectral diversity of a multicomponent seed treatment and/or coating and enable the quantification of the loading of single chemical substance/active agent in a single agent or in a combination formulation at the required high accuracy (<10% standard deviation).
Surprisingly, it was found, that using specific ranges of interest of the spectrum and carefully selecting the main matrix influences for the multi-variate data approach, eliminating redundant or least-influencing matrix influences, minimizes the risk of false active quantification. In this way, even broader natural and man-made variations of the sample matrix were shown to have nearly no influence on the accuracy of the readout.
The here described approach may be used with RAMAN, Mid-Infrared (MIR) and Near-Infrared (NIR) spectroscopy or a combination thereof as far as data for all three spectral ranges are available. Preferred is acquiring one or more spectrum representative of a sample using MIR- or NIR-spectroscopy. Most preferred is using NIR spectroscopy. NIR-technology was found to be superior to the MIR for the following reasons:
A further object of the present invention is a system for non-destructive quantification of one or more chemical substances of interest coated on bulk material in a matrix, said matrix being defined by its further coating components and said bulk material, according to independent Claim 13. Exemplary embodiments of the method and apparatus according to the invention can be gathered from the Claims dependent on Claim 13.
The system of the invention comprises:
In an embodiment, the one or more instructions when, executed by the one or more processors, further cause performance of:
In an embodiment, the one or more instructions when, executed by the one or more processors, further cause performance of:
In an embodiment, the one or more instructions when, executed by the one or more processors, further cause performance of:
In an embodiment the system of the invention may further comprise one or more features selected from:
The system of the invention may comprises or be connected to a user interface in particular for the selection of the chemical substance(s) of interest and display of the computed results A further object of the invention is a computer program element for conducting a non-destructive quantification of one or more chemical substances of interest coated on bulk material in a matrix, said matrix being defined by its further coating components and said bulk material, which when executed by a processor is configured to carry out the steps of the method described above.
A further object of the invention is a computer program element for provision of a calibration for a chemical substance to be quantified in a matrix, said matrix being defined by its further coating components and said bulk material, which when executed by a processor is configured to carry out the steps of the method described above.
A further object of the invention is a non-transitory computer readable medium having stored one or more of the computer program elements mentioned above.
It is clear to the person skilled in the art that spectrums can be analysed with the methods of the invention independently of the acquisition system provided specifications, for example resolution, signal-to-noise ratio, wavelength accuracy, is good enough for acceptable accuracy.
It is also clear to the person skilled in the art that the solutions of the present inventions are primarily usable for analysis of seed coatings but may be application to the analysis of other coated bulk material.
The use of terms “a” and “an” and “the” and similar referents in the context of describing the invention (especially in the context of the following claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. The terms “comprising”, “including”, “having”, and “containing” are to be construed as open-ended terms (i.e. meaning “including but not limited to”) unless otherwise noted.
The use of any and all examples, or exemplary language (e.g., “such as”) provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specifications should be constructed as indicating any non-claimed element as essential to the practice of the invention.
Preferred embodiments of this invention are described, including the best mode known to the inventors for carrying out the invention. Variations of those preferred embodiments can become apparent to those of ordinary skilled artisans to employ such variations as appropriate, and the inventors intend for the inventions to be practiced otherwise than specifically described herein. Accordingly, this invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context. The claims are to be construed to include alternative embodiments to the extent permitted by the prior art.
The invention will now be further illustrated by the following non-limiting drawings and examples.
Exemplary embodiments will be described in the following with reference to the following drawings:
And wherein the following steps are conducted for the validation of the computed calibration:
It is preferred that for each validated calibration identifiers for both training data and validation data set for the active agenet are saved together with confirmation of validation for later use in the value for comparability for a new sample.
The solution of the invention was used for the following examples without being limited thereto. Multi-active quantifications using near-IR and mid-IR(ATR—attenuated total reflection) are exemplarily described for soybean, maize, cotton and oilseed rape using the method of the invention. Spectral scans were collected from calibration samples and obtained data was subsequently used for identifying the correlation consequently establishing the calibration. Additionally, independent validation samples, which were not presented to the spectrometer as part of the calibration, proved the high quality of the calibration.
For NIR analysis a Bruker Tango-R equipped with a rotation sample cup was used. Acquisition parameters were as follows: FT interferometer, Tungsten source, InGaAs diode detector, spectral range: 11.500-4.000 cm−1, 64 scans.
The resolution for the NIR measurement was set to 8 cm−1. To improve the signal to noise ratio 8, 16, 32, 64, 128 and 256 acquisition for a single sample were tested, and the optimal number of resolution vs time was found to be 64 acquisitions. In addition, 8 physical repetitions per sample, i.e. new sub-sample aliquots were presented to the spectrometer and averaged. The examples described hereafter were obtained by averaging all 8 physical repetitions even though it was found that a minimum of 2 repetitions may be sufficient to achieve a reasonable accuracy that will be sufficient for process control requirements (data not shown).
The treated seeds were analyzed by NIR in the following sequence. 100 g of a sample of seeds was filled into the sample cup of the NIR spectrometer. The sampling cup was placed on the analyzer and the measurement started by entering the sample description and initiating the actual measurement using the predefined acquisition data as said. After completed measurement, the seeds were removed and the sampling cup cleaned with a soft tissue.
For mid-IR analysis (ATR) a Bruker Alpha II was used. Acquisition parameters were as follows: IR source, DTGS detector, spectral range: 350-8.000 cm−1, resolution 4 cm−1, circa 64 scans per acquisition. A number of specific spectral ranges were identified and selected for calibration and quantification of said actives, predominantly containing information which can be used to quantify respective actives of interest in the seed treatment. The specified regions were determined from 1st derivative and amplitude studies using the acquired spectra of seed and relevant reference samples proving the relevance of said spectral regions for active quantification, while excluding spectral ranges which are dominated by matrix effects. Specified spectral regions are described for each example separately.
Background signal correction, i.e. subtraction of spectrum of non-treated seeds was not done for any part of this method instead the following normalization was used to ensure high quality, comparable data sets. Said normalization was vector normalization. A spectrum is normalized by calculating a mean value for the y values (spectral absorbances at the different wavelengths) and then subtracting this value from the spectrum. Then the sum of the squares of all y values is calculated and then the corresponding spectrum is divided by the square root of this sum. In this manner, baseline and pathlength effects were corrected.
For all validation studies, deviations of NIR versus HPLC in percent were calculated for all samples using following equation:
HPLC/UPLC reference analytics were done using standard liquid chromatography instruments equipped with UV/VIS detection. Analysis were done using certified analytical references and dedicated qualified methods using commonly applied liquid chromatography routines.
A NIR analysis routine for soybean was established and validated using a standard seed treatment package (also called applied plant protection product) comprising the combination formulations CropStar®, i.e. Imidacloprid and Thiodicarb, and Derosal Plus® containing Carbendazim and Thiram. The concentrations of active agents in the applied respective seeds dressings were as follows, ref. Table 1:
The entire seed treatment application recipe for soybeans (also denoted seed treatment dressing) is characterized by its active agents or its combination of active agents and ratios thereof, means said insecticides and fungicides, and additional components comprising sticker & colorant, for example Peridiam 306, talcum gloss, micro-nutrients, inoculants and root enhancer, ref Table 2. Calibration samples and validation samples were calculated per DoE. The corresponding seed treatment dressings were applied in 1 kg scale using a pilot scale batch treater. For high precision all components were weighed in each dressing. Densities are available to the public, therefore not mentioned here.
It is noted that identifiers (variety x, validation sample VSx, etc) for the samples are merely for orientation within an example. In other words, variety 1 of example 1 is not the same as variety 1 of example x, and VS1 of example 1 is not the same as VS1 of example x, unless explicitly mentioned.
Table 2 describes the compositions of the samples prepared for calibration and validation respectively in order to evaluate the influence of active agent and matrix variations on the quantification of the active agents by NIR; sub-categories of variations as depicted in Table 2, column B were assigned to the samples for better overview. This step is also referred to as “study of the sample”. Further details on the additional components are described in Table 3.
Depending on the actual ratio of the active agents in combination formulations, e.g. Cropstar® and Derosal Plus® the spectral signature was found to be subjected to changes, thus NIR-readout. Product specifications of Cropstar® allow variation of the loading ratios of Imidacloprid and Thiodicarb between batches within a range from 2.9-3.2 to 1 (weight, Thiodicarb/weight, Imidacloprid).
Similarly, the product specifications of Derosal Plus® allow variations of the loading ratios of Carbendazim and Thiram between batches within a range from 2.9-3.2 to 1 (w/w).
For this reason, several commercial batches of combination formulation within the said ranges were included into the validation samples.
The soybean seed itself significantly contributes to the acquired spectral signature. To check the sensitivity/insensitivity of the method of the invention related to seed varieties, samples issued from 10 different varieties were considered in the samples for calibration and for validation.
The method was found to be tolerant toward spectral signature variations related to soybean seed varieties.
Influence of Thousand-Seed-Weight (TSW), Typically Measured in [g/1000 Seeds]
Because of changes of available surface area, variation of the TSW was expected to influence the loading of active agent(s) on each seed. This parameter was taken into consideration in the samples for calibration and in the samples for validation.
Four common micronutrients formulations were considered in the validation samples to check sensitivity of the NIR-method of the invention when using commonly applied micronutrient products on treated seeds. The calibration was found to be insensitive to micronutrients in the seed treatment dressing.
Common Talcum Gloss formulations were considered in the validation samples check sensitivity of described NIR-method when using different talcum drying and/or gloss enhancing products for the corresponding treatments. The method was found to be insensitive to these components.
Common inoculants formulations were used in the validation samples to check sensitivity of the described NIR-method when using different inoculants for the corresponding treatments. The method was found to be insensitive to these components.
Common root enhancers formulations were used in the validation samples to check sensitivity of the described NIR-method when using different root enhancers for the corresponding treatments. The method was found to be insensitive to these components.
In case wherein the validation shows the method being sensitive to further components, it may be advantageous to consider introducing variations of these components in the calibration samples.
It is to be noted that when quantifying seed loading with CropStar®, matrix variations related to the Derosal Plus® seed dressing are important variables that should be considered in both calibration and validation samples.
Spectra for the calibration samples were acquired using the acquisition parameters mentioned above.
The spectra were evaluated by using the multivariate evaluation method “Quant2” in the OPUS software of Bruker.
For Cropstar®, the spectra pretreatment with Multiple Scatter Correction was found to be most suitable. The spectral ranges of 9000 to 8200 cm−1, 6104 to 5444 cm−1 and 5076 to 4400 cm−1 have been used.
For Derosal Plus®, vector normalization and 1′ derivative was found to be most suitable. The spectral ranges of 7236 to 7112 cm−1, 6104 to 5444 cm−1 and 5076 to 4400 cm−1 have been used.
Upon spectral pretreatment and calibration based on the first 5 main components, quantifications of CropStar® and Derosal Plus® based on NIR data (y-axis) show an excellent correlation with the values obtained by HPLC reference analytics (x-axis), ref.
The optimal number of main components, i. e. matrix influences considered for adequate calibration, was determined by using the first minimum of RMSECV (root mean standard error of cross validation, y-axis) over number of main components (x-axis) (see
R2 (coefficient of determination) was found to be 99.3 and the standard deviation 0.15 ml/kg.
R2 was found to be 99.6 and the standard deviation 0.10 ml/kg.
The performance of the determined calibration was evaluated with the validation samples, ref. Table 2. The identified fluctuations in NIR-readout provided a measure for the robustness and accuracy of the calibration, ref. Table 4.
A NIR analysis routine for maize seeds was established and validated using a complex seed treatment package comprising the formulations 1) Poncho® (Clothanidine), 2) Dermacor® (Chlorantraniliprole), 3) Maxim Advanced (Metalaxyl co-applied with Tiabendazim) and 4) Derosal Plus® (Carbendazim and Thiram). The concentrations of the active agents in the respectively applied seed treatment dressings were as follows, Table 5:
The study of the sample for maize seeds was established by design of experiment (DoE) as exemplarily described in example 1, Table 2. Samples as summarized in Table 7 were prepared in 1 kg scale using a pilot scale batch treater. For high precision all products were weighed in. Densities are available to the public, therefore not mentioned here. The following sub-categories were considered for better overview: Batch variation of each dressing, hybrids variation (varieties), Thousand-Seed-Weight (TSW) variation, Talcum Gloss variation.
NIR-spectra for the samples were acquired using the acquisition parameters mentioned above.
The spectra acquired for the calibration samples were evaluated by using the multivariate evaluation method “Quant2” in the OPUS software of Bruker. Characteristics of the calibration are summarized in Table 6.
Using these parameters, the quantification of the respective dressings Poncho®, Dermacor® and Derosal Plus® on the treated maize seeds calculated based on NIR data showed a strong correlation with the quantification obtained using the HPLC reference analytics, ref.
The performance of the established calibrations was evaluated using all validation samples established in the DoE. Identified fluctuations in NIR-readout provided a measure for the robustness and accuracy of the calibration, ref. Table 7.
Deviations of NIR-readouts vs reference analysis HPLC were found to be low proving the validity of the method despite changes in the matrix.
A second NIR analysis routine was established and validated for maize seeds treated with a standard seed treatment package comprising Poncho®, Poncho Votivo®, Allegiance FL, Proline, Fluoxastrobin ST and Acceleron® B360. The concentrations of the actives in the applied seed dressings were as follows, Table 8:
Not quantified active agents are considered components of the matrix. Influences on the spectral were evaluated in the calibration and in the validation samples.
The entire seed treatment sample matrix for maize was established by design of experiment (DoE) as exemplarily described in example 1. Insecticides, fungicides and, if required additional products. E.g. Acceleron© E007 SAT, were applied using the sticker Peridiam Precise 1006 and a green pigment Color Coat Green (BASF). The treatment was applied in 1 kg scale using a pilot scale batch treater. For high precision all products were weighed in.
In order to evaluate the influence of variations on the quantification by NIR, sub-categories were established and used. Considered variables were categorized as followed: Batch variation of active products, Hybrids variation (varieties), Thousand-Seed-Weight (TSW) variation, variation of polymer, pigment and drying agent.
NIR-Spectra of 70 samples were acquired using the acquisition parameters mentioned above. The spectra were evaluated by using the multivariate evaluation method “Quant2” in the OPUS software of Bruker. Characteristics of the established calibrations are summarized in Table 9.
Using the calibrations of Table 9, the quantifications of Clothianidine, Prothioconazole, Fluoxastrobin in the seed treatment showed a strong correlation between NIR data versus HPLC reference analytics, ref.
In addition, or alternatively to HPLC as reference analytics, the weigh-ins of all active agents containing seed dressings may be used for referencing. These weights of applied seed dressings are effortlessly accessible and showed a similarly high correlation, hence, allowing use of gravimetric referencing for NIR-based quantification of multiple active agents, ref. Table 10.
The performance of the calibrations mentioned above was evaluated on 68 independent validation samples. Ten different corn varieties have been selected und used for preparing three different samples per hybrids, giving 30 validation samples for this variation. Identified relative deviations in NIR-readout versus HPLC-reference provided a measure for the robustness and accuracy of the calibration, ref Table 11. In general, deviations of NIR vs HPLC-reverence were found to be low proving the insensitivity of the method against changes in the matrix.
A general feasibility for NIR analysis of treated oil seed rape was carried out using a standard seed treatment package comprising the seed dressings Scenic Gold® and Buteo Start®. The concentrations of the active agents in the applied seed dressings were as follows:
This seed treatment was applied in 2 kg scale using a pilot scale batch treater. For high precision all products were weighed in. Densities are available to the public, therefore not mentioned here.
The reduced number of samples described above were not sufficient to develop a calibration that is suitable to compensate the variations in routine analytics for seed treatment but were used to evaluate the general usability of the NIR method of the invention for oil seed rape.
Spectra for the calibration samples were acquired using the acquisition parameters mentioned above. The spectra were evaluated by using the multivariate evaluation method “Quant2” in the OPUS software of Bruker.
For Scenic Gold®, the spectra pretreatment with vector normalization was found to be most suitable. The spectral ranges of 5948 to 5344 cm−1 and 4988 to 4120 cm−1 have been used. For Buteo Start®, vector normalization was found to be most suitable. Spectral ranges of 6140 to 5352 cm−1 and 5028 to 4140 cm−1 have been used.
The calibration was rated by using cross validation.
Upon spectral pretreatment and use of the first 2 main components of the method both products Scenic Gold® and Buteo Start® within the seed treatment showed an excellent correlation for NIR data versus HPLC reference analytics, ref.
Validation will confirm the usability of the method the invention compared to HPLC as reference analysis
Based on all samples described in Example 1, ref. Table 2, a mid-infrared FT-ATR analysis routine for treated soybean seeds was established and validated using a standard seed treatment package comprising CropStar® (Imidacloprid and Thiodicarb) and Derosal Plus® (Carbendazim and Thiram).
The description of calibrations and validation samples and seed treatment procedure as used are described in example 1.
The treated soybean seeds were analyzed by mid-IR in the following sequence. A single seed was placed on the ATR crystal and fixed with a clamping mechanism. The spectral acquisition was conducted using the predefined acquisition parameters mentioned above (e.g. resolution). After completed measurement, the seed was removed from the ATR crystal and this latter was cleaned with a soft tissue for next seed measurement. Repetitive measurement with a number of different seeds, treated in the same batch, were carried out to obtain a valid averaged readout spectrum. Number of repetitions is given in calibration section below.
Spectral differences for non-treated and treated seeds were found to be well-defined and different loadings are reflected by amplitude changing at specific spectral ranges (
The spectra show a very low level of noise. There is low impact of the soy seed as matrix component in the spectra due to the small immersion depth of the IR radiation (on a small μm scale). The influence of the active agents can clearly be seen by the difference between the sample with no active agent (signal almost zero between 1800 and 1200 cm−1) and the other calibration samples. Also. the variations in concentrations of active agents in the seed dressings can be seen in the varying heights of the absorptions for the corresponding samples. This is an excellent starting point for a multivariate calibration.
Workflow and tools used for data handling were the same for NIR and mid-IR.
The calibration was conducted using the acquisition parameters mentioned above. Of each sample, 25 single seeds were selected. Each of these seeds was measured one time, so 25 spectra per sample were acquired. These 25 single spectra were averaged in 3 groups (i.e. 2×8 averaged spectra and 1×9 averaged spectra) to obtain 3 derived averaged spectra.
The spectra were evaluated by using the multivariate evaluation method “Quant2” in the OPUS software of Bruker.
For Cropstar®, the spectra pretreatment with Mix/Max normalization was found to be most suitable. The spectral range of 1791 to 617 cm−1 has been used. Five main components have been found to be optimal.
For Derosal Plus®, vector normalization was found to be most suitable. Spectral ranges of 1675.9 to 1198.2 cm−1 and 848.2 to 609.4 cm−1 have been used. Three main components have been found to be optimal.
The calibrations were rated by using cross validation.
Using the spectral pretreatment of the method both products CropStar® and Derosal Plus® of the seed treatment show an excellent correlation, ref.
R2 was found to be 97.3 and the standard deviation 0.30 ml/kg.
R2 was found to be 96.8 and the standard deviation 0.17 ml/kg.
Using the established mid-IR calibration, the performance of said calibration was evaluated using all validation samples (ref. Table 2). The identified deviations in IR-readout vs reference analytics provided a measure for the robustness and accuracy of the calibration, ref. Table 14.
Another NIR analysis routine was established and validated for Cotton seeds treated with a standard seed treatment package comprising Poncho Votivo®, Allegiance FL, Gaucho®, Acceleron® D-612, Acceleron® DX-109 and Acceleron® D-510. The concentrations of the active agent in the applied seed dressings were as follows, Table 15:
Not quantified active agents are considered components of the matrix. Influences on the spectral were evaluated in the calibration and in the validation samples.
The entire seed treatment sample matrix for cotton was established by design of experiment (DoE) as exemplarily described in example 1. Insecticides, fungicides and, if required additional products. for example, Acceleron© E007 SAT, Secure 661, E-522 and a blue pigment were applied to seed samples. The treatment was applied in 1 kg scale using a pilot scale batch treater. For high precision all products were weighed in.
In order to evaluate the influence of variations on the quantification by NIR sub-categories were established and used. Considered variables were categorized as followed: Batch variation of active products, hybrid variation (varieties), Thousand-Seed-Weight (TSW) variation, variation of polymer, pigment and drying agent.
NIR-Spectra of 115 samples were acquired using the acquisition parameters mentioned above.
The spectra were evaluated by using the multivariate evaluation method “Quant2” in the OPUS software of Bruker. Characteristics of the established calibrations are summarized in Table 16.
Using the calibrations of Table 16, the quantifications of Clothianidin and Imidacloprid in the seed treatment showed a strong correlation between NIR data versus HPLC reference analytics.
In addition, or alternatively to HPLC as reference analytics, the weigh-ins of all active agents containing seed dressings may be used for referencing. These weights of applied seed dressings are effortlessly accessible and showed a similarly high correlation, hence, allowing use of gravimetric referencing for NIR-based quantification of multiple active agents, ref. Table 17.
The performance of the calibrations mentioned above was evaluated on 40 independent validation samples. Four different cotton varieties were selected und used for preparing 6 different samples per hybrids, giving 24 validation samples for this variation. Poncho® was not included in all validation samples to reflect the treatment offerings. Identified relative deviations in NIR-readout versus HPLC-reference provided a measure for the robustness and accuracy of the calibration, ref Table 18. In general, deviations of NIR vs HPLC-reverence were found to be low proving the insensitivity of the method against changes in the matrix.
| Number | Date | Country | Kind |
|---|---|---|---|
| 21214876.1 | Dec 2021 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2022/085107 | 12/9/2022 | WO |
