METHOD OF CHROMATOGRAPHY ON A MULTICAPILLARY TRIM

专利摘要:
Chromatography process in which a gaseous, liquid or supercritical mobile phase containing species to be separated is circulated through a lining, said lining being characterized in that: it comprises a plurality of capillary ducts passing through the lining between one face; said upstream by which the mobile phase enters the lining and a so-called downstream face through which the mobile phase leaves the lining - the material of the walls comprises a first population of related pores, ensuring passages from one conduit to the other allowing the molecular diffusion to take place between adjacent conduits, pores having a mean diameter (dore) greater than 2 times the molecular diameter of the molecules to be separated - the diameter of the conduits is less than 50 μm
公开号:FR3026313A1
申请号:FR1459176
申请日:2014-09-29
公开日:2016-04-01
发明作者:Francois Parmentier
申请人:Francois Parmentier；
IPC主号:

专利说明:

[0001] FIELD OF THE INVENTION The present invention relates to optimizing the morphology and porosity of multicapillary packings in operation. BACKGROUND OF THE INVENTION Chromatography is a particular technique, which has its own advantages and constraints and is thus different from other related techniques by the use of solid packings and fluids, such as adsorption and heterogeneous catalysis. . In adsorption it is sought to retain a compound of a fluid effluent on the surface of which it is adsorbed via an isotherm, or on which it reacts. We seek to purify the fluid. We need specific high surfaces. High capacity beds are needed The effectiveness of the packing is not critical (number of theoretical trays) and it is preferred to use pellet beds of 1 to 2 mm in diameter. Effectiveness has only a negligible influence on the dimensioning of the bed insofar as it will only play on the stiffness of the percolation front, which is good as soon as one reaches about 20 theoretical plates. It is then necessary to regenerate the adsorbent by a combination of means, temperature or chemical reaction, which eliminates the adsorbed or combined impurities. The operation is therefore sequential but the cycle times are in days or weeks. We size on the mass of the bed. The pressure drops are low. In catalysis, it is sought to perform a chemical reaction on the surface of the solid. It is desired that the reagents remain an optimal time in contact with the solid. These are again adsorption forces and chemical reactions. We are interested in living time criteria. The reasoning in number of theoretical plateaux is inoperative. The regularity of the packing is one factor among others and is secondary to the catalytic selectivity. We do not try to separate molecules. The losses are low. In chromatography, several components present in a sequentially admitted fluid charge are separated by a short time interval in minutes, propagating it from an entry point to an exit point of a solid column under the effect. of an eluent fluid. The separation obtained can be achieved by a very wide variety of forces that compete with the driving effect of the eluent, sharing, adsorption, steric interactions, ionic interactions, etc. This method offers a high separating power, each component behaves differently. To emphasize this separating power, the column must have a high number of theoretical plates, for example 1000. This also means that the diffusion resistances must be minimized, and therefore that the diffusion distances are short, and that the column must be long. These combined factors make chromatography a technique which calls for an excellent regularity of the flow and therefore of the packing, and a small characteristic dimension thereof, leading to loss of charges which become rapidly critical with particulate solids. These are the problems that must be solved in chromatography. A multicapillary monolith has been described in patent application VVO 2011/114017. This invention describes a monolith consisting mainly of silica or alumina whose walls are porous so as to allow rapid equilibration of the composition between adjacent conduits and its application to chromatography. In particular, there is described a process for obtaining this monolith, characterized in that precursor threads are coated in a porous matrix, and in that the fibers are destroyed in order to leave only the matrix. It is mentioned as structural materials of soils or gels of silica or alumina.
[0002] This document and the following do not describe, however, the hydrodynamic, diffusional, molecular and structural conditions that make it possible to take maximum advantage of a porous multicapillary packing. In fluid bed chromatography, fluid streams are remixed by convection and form a continuum. In multicapillary packing the fluid streams are convectively independent and only communicate by molecular diffusion. The porous structure of the walls therefore represents a fundamental character of the effectiveness of these packings. This results in the need to characterize and optimize the nature and distribution of the porosity of the material separating the conduits.
[0003] The following publications: K Nakanishi, Phase separation in silica sol-gel system containing polyacrylic acid, Journal of non-crystalline Solids 139 (1992, 1-13 and 14-24, K. Nakanishi, Phase separation in Gelling Silica-Organic Polymer Solution: Systems Containing Poly (sodium styrenesulphonate), J. Am., Ceram, Soc., 74 (10) 2518-2530-30 (1991), deal with materials having two families of pores and consist in producing a monolithic silica packing comprising two families of pores, on the one hand interconnected macropores through which can flow a relatively free liquid, and on the other hand a family of mesopores or micropores creating the specific surface, and therefore the activity vis-à- The following publication: Deen, WM Hindered Transport of Large Molecules in Liquid Filled Pores, AIChE Journal, 33, 1409-1425 describes a relationship between the diffusivity in pores and the radius of molecules in molecules. liquid phases. on the other hand, the relationships linking the average free path of a molecule in the gas phase and the diffusional regime of this molecule in a porous matrix. In particular, it is known that when the pore size becomes close to the average free path of a molecule, the diffusion goes into the Knudsen regime and is slowed down. There is therefore a need to define the pore size and the sizing and operational parameters of a multicapillary packing so as to obtain optimum efficiency for a given solute and hydrodynamic walking conditions. BRIEF DESCRIPTION OF THE INVENTION An object of the invention is to define the properties of a multicapillary packing that makes it possible to obtain optimum efficiency for a given solute and hydrodynamic walking conditions. For this purpose, the invention proposes a chromatography process in which a gaseous, liquid or supercritical mobile phase containing species to be separated is circulated through a packing comprising a stationary phase, said packing being characterized in that: it comprises a plurality of capillary ducts passing through the lining between a so-called upstream face through which the mobile phase enters the lining and a so-called downstream face through which the mobile phase leaves the lining, - the material of the walls of said ducts comprises a pore network related, said pores forming passages from one conduit to another allowing the molecular diffusion to occur between adjacent conduits, said pores having an average diameter greater than twice the molecular diameter of the species to be separated, - the average diameter ducts is less than 50 μm. In a particularly advantageous manner, the ratio, called the "relative height of dispersion", of the theoretical plateau height (Hd, p) due to the inhomogeneities of the packing on the total theoretical plateau height (H) of the packing is less than 0, 66, preferably less than 0.3 and more preferably less than 0.1. In said process: the species to be separated have a Rh molecular radius in the elution solvent, a molecular diffusivity Do in the elution solvent, a molecular diffusivity D in or on the stationary phase, a partition coefficient K between the stationary phase 4 and the eluting solvent, a retention factor k 'in the chromatographic column, and - the lining comprises conduits of average diameter d, separated by walls of average thickness d, the irregularity of which is defined by a standard deviation of the diameter d, brought back to its average SigmaD and by a standard deviation of the thickness d, brought back to its mean SigmaE, - the porous material constituting the walls has a porous volume fraction P, a volume fraction of phase stationary f VolStat or a specific adsorption surface S, a tortuosity T, and the network of related pores has a diameter dpore, - the mobile phase flows with the average speed v, in the ducts and - the theoretical plateau height (Hd ,, p) due to the inhomogeneities of the packing is defined by the relation: vo * (FKD * SigmaD2 + FKE * SigmaE2) * (d, + de) 2 Hdisp = 0.778 * FDif f * FDil * (1 k ') * 2 Advantageously the ratio called "relative dispersion height", the theoretical plateau height (Hd ,, p) due to the inhomogeneities of the packing on the total theoretical plateau height (H ) of the packing is calculated at the optimum efficiency of the packing given by the Van Deemter curve. Advantageously, the conduits have an average diameter of less than 30 μm, and preferably less than 10 μm. The lining advantageously comprises at least a part of which: the capillary ducts are substantially straight and parallel to one another; the ducts have a substantially uniform cross-section with respect to each other; the section of each duct is regular over its entire length, all the ducts pass through said part from one side to the other. The pore network of the packing has a mean diameter greater than 5 times the molecular diameter of the species to be separated and preferably greater than 10 times the molecular diameter of the species to be separated. According to one embodiment of the invention, the mobile phase is in the condensed state and said pore network has an average pore diameter greater than 2 nanometers, preferably greater than 10 nanometers, and even more preferentially greater than 100 nanometers. nanometers. According to another embodiment of the invention, the mobile phase is in the gaseous state and the pores have a diameter greater than the average free path of the molecules.
[0004] BRIEF DESCRIPTION OF THE DRAWINGS Other characteristics and advantages of the invention will emerge from the detailed description which follows, with reference to the appended drawings, in which: FIG. 1 is a sectional view of a basic view of a lining; multicapillary cylindrical in a direction parallel to its major axis. FIG. 2 is a sectional view of a principle view of a cylindrical multicapillary packing in a direction perpendicular to its major axis; FIG. 3 shows the physical diagram used to simulate the behavior of the multicapillary packings. FIG. 4 schematically shows the numerical scheme used to simulate the behavior of the multicapillary packings. FIG. 5 shows the relationship between NEPT and the length of the lining for porous multi-capillary monoliths and non-porous multi-capillary monoliths. FIG. 6 shows the appearance of transient diffusion phenomena. FIG. 7 shows the same diffusive phenomenon projected on a set of adjacent ducts. FIG. 8 shows the correlation between Hdisp and Dfactor. FIG. 9 exemplifies the nature of the numbers characterizing a porous medium. FIG. 10 schematizes the characteristic sizes of the molecules interacting with the walls in a porous medium. FIG. 11 compares the pressure drop of multicapillary and particulate packings. FIG. 12 shows a Van Deemter curve obtained by simulation.
[0005] DETAILED DESCRIPTION OF THE INVENTION One is interested in the optimization parameters of a separation of a given chemical body, molecule or biomolecule, in a multicapillary packing comprising, between the walls of the conduits, a porous solid including at least one population of pores related to which the molecular diffusion of the species to be separated can take place. These pores provide a free passage between contiguous ducts.
[0006] Chromatography is a particular method of molecular separation characterized in that it carries out a separation of a mixture of chemical substances under the contradictory action of a dynamic entrainment of these species by a current of an eluting phase. retention of these species by a stationary phase. Preferably this process is continued until complete elution of the separated species out of the stationary phase. According to the invention, the chromatographic method is advantageously characterized by its behavior in the linear regime, ie for a brief injection in the form of a pulse of products to be separated. Under these conditions the dilution of the species is large and the partition coefficient of the species to be separated with the stationary phase does not depend on the concentration. It goes without saying that such columns may be used in preparative processes.
[0007] According to the invention, such a chromatographic separation or such a chromatographic process will be characterized in that it comprises at the optimum of efficiency at least 300 theoretical stages, and preferably at least 1000 theoretical stages. This distinguishes it from membrane separations, catalytic processes and adsorption or ion exchange separations in particular.
[0008] Chromatography is applied in liquid, gas and supercritical phases. The present invention is a chromatography process in which a gaseous, liquid or supercritical mobile phase containing species to be separated is circulated through a lining, said lining being characterized in that: it comprises a plurality of through capillary ducts; the lining between a so-called upstream face through which the mobile phase enters the lining and a so-called downstream face through which the mobile phase leaves the lining - the wall material comprises a first population of related pores, ensuring passages of a duct to the other allowing the molecular diffusion to take place, between adjacent ducts, pores having a mean diameter (dpo re 1 greater than 2 times the molecular diameter of the molecules to be separated - the diameter of the ducts is less than 50 pm Advantageously the ratio , called the "relative height of dispersion", of the theoretical plateau height (Hd ,, p) due to inhomogeneity s of the packing on the total theoretical plate height (H) is less than 0.66.
[0009] Advantageously, the theoretical plateau height (Hd, p) due to the inhomogeneities of the packing is defined by the relation Hdisp = 0.778 * FDif f * FDil * (1 + k ') * 2 It will be noted that this formula includes static terms connecting the morphological structure of the packing (ex dc, de), diffusional terms at the molecular level (ex Do, Ds, C) with hydrodynamic terms which describe its behavior in operation with respect to a fluid flow (ex ve). It therefore describes a field of hydrodynamic walking conditions and in operation. This formula applies for species to be separated characterized by a Rh molecular radius in the elution solvent, a molecular diffusivity Do in the elution solvent, a molecular diffusivity D, in or on the stationary phase, a partition coefficient K between the stationary phase and the elution solvent, a retention factor k 'in the chromatographic column, for a lining comprising conduits of average diameter d, separated by walls of average thickness d, the irregularity of which is characterized by a standard deviation of the diameter deduced from its average SigmaD and by a standard deviation of the thickness d, brought back to its average SigmaE, packing of which the porous material constituting the walls is characterized by a porous volume fraction P, a volume fraction of phase stationary fvoistat or a specific adsorption surface S, a tortuosity T, and comprising at least one population of related pores of diameter d t for a mobile phase of viscosity u. flowing with the average speed v, in the conduit. Advantageously, the ratio called "relative dispersion height", of the theoretical plateau height (Hd, p) due to the inhomogeneities of the packing on the total theoretical plateau height (H) is calculated at the optimum of the effectiveness of the defined packing. by the Van Deemter curve.
[0010] Advantageously, the material of the walls provides continuity in the condensed phase linking the conduits to each other. These conduits will have a diameter suitable for chromatographic separation, less than 500 μm, preferably less than 250 μm. However, very preferably for the embodiment of the invention, these ducts will have a diameter less than 50 μm, preferably less than 30 μm, and even more preferably less than 10 μm. vo * (FKD * SigmaD2 + FKE * SigmaE2) * (d, + de) 2 Indeed, an essential differentiating factor between the multicapillary packing and the particulate packing with respect to a chromatographic process lies in the loss of charge. weaker of the first. At optimum efficiency this also means that a multicapillary packing operating under a loss of load identical to that of a particulate packing will present a number of theoretical plates 3 to 4 times higher, and a productivity (in flow / unit surface area of its section) also 3 to 4 times higher. These advantages become relevant when the pressure drop of the bed becomes a sensitive operating parameter requiring special tools to be imposed.
[0011] It therefore has the low limit for condensed phase chromatography and more preferably in the liquid phase the simple head of the stationary phase bed itself allowing a gravity flow. For multicapillary packing this effective limit is between a conduit diameter of less than 50 μm for fluids commonly used, and preferably less than 30 μm.
[0012] Indeed the chromatography is carried out in a simple way in gravity devices, or the weight of the fluid column on the packing causes its flow. The upper limit of the diameter of the capillaries will be obtained when the flow of the fluid at the speed allowing the optimum of the effectiveness of the packing will cause a loss of load equal to the weight of the column of fluid considered on the height of the bed.
[0013] We know that for a multicapillary packing with the optimum of efficiency: Vo * dc r = VR Do The law of Poiseuille is written 32 * i2 * LG * vc 3 i32 *, u * Do * VR dmax P * 9 The table below exemplifies dmax for various liquids common in chromatography. Solvent! DO (m2 / s) JR JR I kg / m3 dmax (pm Water 0.001 1E-09 5 1000 25.3869139 Hexane 0.00031 3E-09 5 659 28.4754613 AP = The pressure caused by a fluid height LG s It follows that: Methanol 0.00055 2E-09 5 791 28.3363475 Chloroform 0.00057 2E-09 5 1480 23.2716977 VR is generally between 2 and 5. dmax is the maximum diameter of the ducts allowing a natural equilibration of a gravity chromatography Another limit of the multicapillary packing has been discovered by the inventor It has been discovered that when the walls of the ducts are porous the effectiveness of the chromatographic process measured by its number of Plateaux Theoretical or NPT linearly increases with the length of the column instead of being limited by the imperfections of independent conduits not communicating by diffusion.
[0014] It turns out after study, and this constitutes an essential point of the present invention, that this consideration applies essentially to the species to be separated and not to the elution solvent. According to the invention this consideration leads to selecting for a given separation a packing having a porosity adapted to the desired separation.
[0015] The pores of the material constituting the walls of the conduits will therefore be adapted to the size of the molecules to be separated: for condensed phase separation, a related population of these pores advantageously has an average diameter greater than twice the molecular diameter of the molecules to be separated, and preferably between 2 and 1000 times the molecular diameter of the molecules to be separated. These pores provide a connected passage from one conduit to another allowing molecular diffusion to occur between adjacent conduits. Advantageously, these pores have a diameter greater than the mean free path of the molecules for application in gas phase chromatography. Advantageously, molecules of molecular weight between 0 and 1000 g / mol will be separated with communicating pore sizes of the material of the walls between the ducts of between 4 and 30 nm, molecules of molecular weight of 10000 g / mol will be separated with As the communicating pore sizes of the wall material between the ducts between 10 and 100 nm, the molecular weight molecules of 100,000 g / mol will be separated with communicating pore sizes of the wall material between the ducts of between 30 to 300 nm. According to the invention the conduits are essentially free to the circulation of a fluid.
[0016] By free flow of a fluid is meant that the pressure drop of a fluid through a conduit is less than 3 times the pressure drop in said completely empty conduit of solid material. According to the invention, the ducts are advantageously empty of solid matter.
[0017] In particular, in the case where the ducts contain solid material, its pore volume will advantageously be greater than 95% by volume, advantageously greater than 98% by volume. The inventor has discovered that the diffusivity of the molecules to be separated through the walls of a multicapillary packing makes it possible to considerably increase the effectiveness of this packing in terms of separating power and to make this efficiency directly proportional to the length of this packing. . However, the effectiveness of the packing obtained is not equal to that of a single capillary, and the recorded loss of efficiency results in an additional theoretical plateau height which is added to the theoretical plateau height of the single capillary.
[0018] The inventor has demonstrated that this theoretical plateau height is dependent on several factors intimately related to the geometry of the conduits and the porous solid that separates them. Computer simulations were performed and showed that the efficiency of a chromatographic process in a porous monolith was constant and independent of its length. However, it is demonstrated that the efficiency of a non-porous solid wall multicapillary packing whose stationary phase thickness is constant and the diameter of the randomly variable capillaries must tend towards a maximum given by the formula: 1 N maxD = FKD * SigmaD2 It is also demonstrated that the efficiency of a non-porous wall-packed multicapillary packing whose diameters are equal and the randomly variable stationary phase thickness must tend towards a given maximum by the formula: 1 NmaxE FKE * SigmaE2 An analysis of the phenomenon of Molecular diffusion between the ducts makes it possible to understand the phenomenon.
[0019] It is considered that the same species injected into each conduit are assigned a fictitious conduit number allowing them to be tracked independently. A chromatographic process in a porous multicapillary packing is subjected to a diffusional remixing phenomenon between adjacent ducts.
[0020] The diffusion of an injected solute in quantity Q at time to = 0 at a central point of a two-dimensional homogeneous diffusive continuous medium obeys a Gaussian law of the type: R2 x = * exp (4 * * D * t) 4 * D * t Concentration x has the shape of a bell curve that spreads and collapses over time by diffusion of matter in the infinite medium.
[0021] It is thus conceivable that the individual conduits separated by a medium open to molecular diffusion receive material from the ducts which surround them. It can be said that this contribution makes use of a population of surrounding ducts that is becoming larger with time. According to this analogy, the term in 2 * D * t is homogeneous to a mean surface covered by diffusion diffusion phenomenon after a time t. Dividing this area by the unit area of an elementary cell of the packing having the mean section of a conduit and its stationary phase, i.e. the section of the packing divided by the number of ducts, a characteristic quantity of the number of conduits in interactions at time t is obtained.
[0022] This number increases linearly with time. On the other hand, it can be calculated analytically that the maximum number of theoretical plates that can produce a set of independent solid-walled capillaries of variable diameters according to a Gaussian probability law having a constant thickness of stationary phase on their wall is given by the formula involving the factor defined by the name FKD in this text and in the claims: 1 NmaxD FKD * SigmaD2 With (2 + 3 * k ' 2 FKD = 1+ k' SigmaD is defined in this text and in the claims as being the relative standard deviation (standard deviation / mean) of the hydraulic pipe diameter k 'is calculated as the ratio between the quantity of solute or species to be separated in the stationary phase and the quantity of solute in the mobile phase at equilibrium .
[0023] We classically measure k 'by the relation tR - to k' = K Vm K is the partition coefficient of the chemical species considered between the stationary phase and the mobile phase. In the case of an adsorption partition on a solid stationary phase, the concentration in the stationary phase can be conveniently calculated on the basis of a ratio of the adsorbed component mass to the volume of the solid phase. This makes it possible to preserve the definition of K. It is also possible to calculate analytically that the maximum number of theoretical plates that can produce a set of independent solid-walled capillaries of constant diameters having a variable stationary phase thickness according to a Gaussian probability law on their wall is given by the formula involving the factor defined by the name FKE in this text: NmaxE FKE * SigmaE2 With FKE * PB = - ko)) (fVOIStat * K + P) * of * (2 * dc + de) / Co = (2 * d, + 2 * of) PB - (2 * d, + de) These formulas take into account the porosity P, the wall thickness and the stationary phase volume fraction. SigmaE is defined in this text and in the claims as the relative standard deviation (standard deviation / mean) of the stationary phase thickness surrounding the conduit. It is noted that a set of full-walled capillaries will be very poor in terms of chromatographic performance, limited in practice to a few hundred theoretical plates. This results in the need to confer on the walls of the ducts a porous structure allowing the molecular diffusion of take place between adjacent channels. It is known that if we analyze the behavior of N capillaries whose ducts have a randomly variable diameter according to a normal distribution, the behavior obtained is that of a new normal distribution whose variance is the elementary variance divided by N. is the variance of a multiple draw of the same Gaussian random variable. The behavior of this multiple draw is expressed by the law: k '= ta In the case of the partition chromatography Vs 1 Nm' = FK * Sigma2 Molecular diffusion resulting in a number of conduits interacting linearly increasing with time, N linearly increases with time, and the efficiency N, "linearly increases with time and therefore with the length of the lining, which agrees with a constant partial height of dispersion in time.
[0024] The number N can be correlated with the Gaussian spread by the formula: NOC (dc + de) 2 The number d, is calculated assuming all the volume contained in the walls of the ducts of the lining distributed uniformly and concentrically on the periphery of the ducts. . The number d is twice the thickness of this layer.
[0025] The contribution of the inhomogeneity of the ducts to the efficiency, or theoretical partial plateau height, linked to this phenomenon, therefore remains constant over time for a constant mobile phase velocity. L * FKD * SigmaD2 L * FKD * SigmaD2 * (d, +) 2 Hdis D = P lvmaxD = 2 * Dreel * tR L * FKE * SigmaE2 L * FKE * SigmaE2 * (d, + de) 2 Hdis EPN max The efficiency of a chromatographic packing is measured by the theoretical plateau height. This can be written in the context of linear chromatography as the contribution of several terms or partial heights, including a term relating to a virtual packing of capillaries of diameters strictly equal to and equal to the average of the ducts of the actual packing. , having a uniform wall thickness, and a term of inter-duct interaction or inhomogenite-related dispersion thereof. : H = Hcap Hdisp Hdisp = HdispD HdispE The theoretical plateau height H'p of an average single capillary is easily accessible for any material, documented, and its theory is known. It can also be computed by computer. The total theoretical plateau height is easily measurable experimentally using the LMH half-height width of the chromatographic peak from an apparatus, according to the formula: NPT = 5.54 * (t, / LMH) 2 2 * Dreel * tR And H = LG / NPT It will be necessary for a good operation of a multicapillary packing that the plateau height related to the dispersion term HdISp does not become very high compared to the term of global efficiency H, that is to say that is, it does not represent a fraction too high of the total theoretical plateau height. This is characterized by the value of Relative Dispersion Height given by the following formula: Relative Dispersion Height = H According to the invention, the operating parameters and the morphology of the packing should be such that the Relative Dispersion Height does not exceed 0 , 66 for the species to be separated. Advantageously, for a better effectiveness of the packing, this Relative Dispersion Height will not exceed 0.5, and even more preferably will be less than 0.3. The characteristics of such packing will be fully exploited when the Relative Dispersion Height is less than 0.1. The dispersion height can be known from the following formula: H = Hdisp + Htheo Knowing H by a measure and calculating Htheo as the theoretical plateau height of a single capillary column with characteristics identical to a representative mean capillary column If the actual packing, or if the structural details do not allow it, by calculating Htheo as the theoretical plateau height of a packing characteristic having uniform average solids representative of the actual packing, Hd is deduced as their difference. Such a column may comprise a central duct of diameter equal to the arithmetic average of the diameter of the ducts, and on its periphery a thickness of a stationary phase identical to that of the actual lining and having a thickness equal to the arithmetic mean of its thickness in the lining. Htheo can be calculated analytically or obtained by a computer simulation. In order to know Htheo of a multicapillary packing having a real wall structure, it will be preferable to use a computer simulation comprising all the morphological, geometrical and constituent, physical and physicochemical details of said wall and the lining, as well as the parameters of the flowing fluids and the thermodynamic Hdisp of the species to be separated. Software such as COMSOL multiphysics makes it easy to achieve such performance. The input data of such a simulation is essentially - the porous fractions filled by mobile phases in the wall, the tortuosity and the average pore size and pore size distribution of these porous fractions as well as the molecular diffusivity of the species. to be separated measured in these phases under the conditions of the chromatographic separation. When these are not available experimentally, we can estimate them by the method of VVilke and Chang. The porous fractions filled with organic gel or a liquid organic stationary phase in the wall, as well as the molecular diffusivity of the species to be separated, measured in these gels if appropriate under the conditions of the chromatographic separation. - The geometry of the wall including details such as the position and the dimensions of the zones filled by the organic gel, the organic liquid and the mobile phase and any dead zones or filled by fluids or substrates other than the mobile phases, organic liquid and organic gel, as well as the molecular diffusivity of the species to be separated measured therein under the conditions of the chromatographic separation. - The partition coefficients of the species to be separated between the different phases present in the concentration range encountered during the chromatographic process. The pressure drop applied to the packing and the composition of the eluent fluid as well as its viscosity under the Htheo chromatographic separation conditions for a non-retained compound can advantageously be calculated in a preliminary manner by the formula: LG NTHMax = dc + * P * 1 , 6 Htheo = LG / NTHMax We deduce Hdisp relative Relative Dispersion Height = = 1 H NTHMax * H The relative dispersion height can also be advantageously calculated using the following formulas. The medium in which the diffusion of the species to be separated occurs is not a homogeneous medium. It contains free conduits in which flows a fluid, or live volume LG containing a living mass, and a stagnant volume present in the pores comprising the eluent phase and the stationary phase. If we consider the previous Gaussian, several correction factors must be taken into account.
[0026] The diffusion coefficient of the species to be separated in the actual medium of the packing can be modeled by the following formula corresponding to a parallel conduction in the center of the conduits and in the walls of the packing associated in series with a conduction in a thickness of the material of the walls: d + d Dreel Fdif f dc of c _L e DC 'DE With dc dc * Do + [(dc + de) * (dc + de) - dc dc] * DE (de + de) * (de + de) And DE = Do * P * C * (P + fvotstat) + Ds * fvotstat * C * K * (P + fvotstat) T * (P + fvotstat * K) T * (P + fvotstat * K) The effect of the living mass and the inert mass is expressed by the following correction coefficient defined by the name FDil in this text and in the claims: 2 dC (c1 + ((de + de) 2 * (P + fvotstat * K)) DC = FDil = The final formula is written: ConstPropD L FKD SigmaD2 (dc + de) 2 Hdisp 2 * FDiff FDil * tR ConstPropE L FKE SigmaE2 (dc + de) 2 2 * FDiff FDil * tR We can ask: ConstPropD = ConstPropE = ConstProp Either with: DFac tor = 2 * FDiff * FDil * (1+ k ') vo (FKD SigmaD2 + FKE SigmaE2) (dc + de) 2 for vo = - to Hdisp = ConstProp DFactor The value of the constant ConstProp is found to be 0.778 by a Computer simulation study.
[0027] All the quantities present in these formulas are well known to those skilled in the art and accessible to the measurement, commonly characterized or easily deduced from the usual characterizations. Porosity is the proportion of related pore volume of the material constituting the walls of the packing, it is a characteristic published in the data sheets of the commercial materials. The porosity is measurable by mercury porosimetry for pores larger than 50 nm, by nitrogen adsorption for pores smaller than 50 nm. The pore size is derived from these same techniques.
[0028] Tortuosity represents the spatial path that a molecule has to traverse to get from one point to another in the porous material by moving away from the straight line. It is a commonly accepted and documented value. In particular, many laws have been proposed to relate porosity to tortuosity.
[0029] See "Tortuosity-porosity relationship in the porous media flow," Maciej Matyka, Arzhang Khalili, Zbigniew Koza, 22/01/2008. Without pretending to be exhaustive one can cite the following law, which will be the one used for the definitions of this text when a direct measurement will not be available: T = 1- p * ln (P) In which we can take for p a value 0.80.
[0030] Factor C measures the reduction of solute molecular diffusivity related to the pore size of the wall material. This quantity is calculated differently in dense condensed phase (supercritical or liquid) and diluted phase (gas). In the gas phase the diffusion becomes impeded when the diffusive flow enters in flow of Knudsen. This occurs when the average free path of the molecules becomes on the order of or greater than the pore diameter. The diffusivity of Knudsen is written: CIPore K * Na, * TK ri * MA DKA * 3 When the diffusivity of Knudsen and the molecular diffusivity are in competition, one writes: 1 1 1 - a * yA DAe DKA DAB With NB a = 1 + - NA In general, we simplify this formula by: 1 1 1 DAe DKA DAB The coefficient C deduces from this CDAe DKA DAB DAB + DKA In condensed phase C is calculated differently. Many correlations are available in the literature. We will quote this one as definition for the formulas of the present text (Deen, 1987): C = Kp * Kr With Kp = (1 - To) 2 And Kr = 1 - 2,104 * To + 2,089 * To2 - 0,948 * To3 To = 0 Rh is the molecular radius of the species molecule to be separated as a sphere and the radius of the pores.
[0031] Kp is a factor that accounts for a difference in the equilibrium concentration between the pores and the infinite medium. Kr takes into account the steric genes of the molecules to be separated in the pore volume. It is thus found that the group C becomes less than 0.1 for a ratio Δ of 0.5, corresponding to a pore size less than 2 times the diameter of the species molecule to be separated. One loses an order of magnitude on the effective diffusivity, which becomes prohibitively low, and the efficiency of the separation becomes bad. The following table calculates the C ratio for different molecules and pore sizes.
[0032] Rh25 molecule rh m ro (n / Kp Kr C organic 0.15 0.3 1 0.00 0.04 0.00 organic 0.15 0.6 0.5 0.25 0.35 0.09 organic 0, 15 1 0.3 0.49 0.53 0.26 organic 0.15 2 0.15 0.72 0.73 0.53 organic 0.15 4 0.075 0.86 0.85 0.73 organic 0.15 0.05 0.90 0.90 0.81 organic 0.15 10 0.03 0.94 0.94 0.88 protein 1.5 6 0.5 0.25 0.35 0.09 protein 1, 0.3 0.49 0.53 0.26 protein 1.5 0.1 0.81 0.81 0.66 protein 1.5 100 0.03 0.94 0.94 0.88 macromolecule 5 0.33 0.44 0.50 0.22 macromolecule 5 100 0.10 0.81 0.81 0.66 macromolecule 5 300 0.03 0.93 0.93 0.87 As seen previously, the effective diffusivity of a molecule in a porous medium is linked to several factors: 1. To the fraction of volume of the porous medium open to the diffusion of molecules 2. To the tortuosity of the medium, ie to the length that the molecule must actually to join two points, bypassing the obstacles formed by the walls of the pores.It is expressed in the form of rap. port between the distance in a straight line and the distance actually traveled. 3. At the actual diffusivity of the molecules in the pores of the medium.
[0033] It is considered that the effective diffusivity is written using the grouping in the formula giving FDiff: Deff = Do * P * C / T. We can indifferently use Deff / OD or P * C / T in said formula. The grouping of capacity valid for a partition chromatography is used in the formula giving FDil: fVolStat * K Any equivalent term representing the capacity of retention of a solute by the stationary phase and in particular in the case of an adsorption can be used. surfacic substitute the group: * Kacis The realization of packings with the required characteristics will preferably be made respecting the following item 7, advantageously items 1,2,3,5, 6 and 7 simultaneously, and even more advantageously the all of the following 9 items simultaneously: Preferably, ducts of circular, hexagonal or square sections will be made. Preferably these ducts will be stacked at the vertices of a constant triangular or square mesh. The ducts will extend straight and elbow-free along the entire length of the lining. The openings of the ducts will be open and open on each side of the lining. The variability (or standard deviation) on the diameter of the conduits will be less than 15% of their average diameter, preferably 5% of their average diameter, and more preferably 2.0% thereof. The wall material will have a porosity greater than 15% by volume, and preferably greater than 40% by volume. This magnitude affects both the porosity factor P and the tortuosity factor T. A packing having a high pore volume ratio will be less tortuous. It will therefore be all the more effective. Preferably, packings having a porosity of approximately 60% by volume will be used. The pores of the material will be adapted to the size of the molecules to be separated: for a condensed phase separation, a population of these pores advantageously has an average diameter greater than 2 times the molecular diameter of the molecules to be separated, and preferably between 2 and 1000 times. times the molecular diameter of the molecules to be separated. These pores provide a connected passage from one conduit to another allowing molecular diffusion to occur between adjacent conduits. Advantageously, these pores have a diameter greater than the mean free path of the molecules for application in gas phase chromatography. The walls will be made to give their thickness, their topology and their homogeneity a great regularity. This means that the thickness d, will have a standard deviation better than 30% of its average value between the ducts and for the same duct. Preferably, this standard deviation will be better than 10% of its mean value. This may be done, for example, by a coating procedure by continuous dipping of a regular circular yarn. The fibers are then removed as described in patent application VVO 2011/114017 for example. 9. Advantageously, in the walls of the lining, a first connected population of large diameter (macroporous or mesoporous) addressing pores will be produced, allowing rapid diffusion of the molecules into the porosity and between the ducts, and a second population of pores. functional contiguous and communicating with those of the first population, which will be responsible for the activity of the separation and will bring specific surface, and capacity. The packing according to the invention is characterized in that it develops more than 300 theoretical plates at optimum efficiency, preferably more than 1000 theoretical plates, and even more advantageously more than 10000 theoretical plates. Indeed porous commercial monoliths are intended for a different use of chromatography, in general filtration or membrane separations, and have diameters of ducts greater than 0.5 mm and wall thicknesses of the order of a millimeter . The packing walls resulting from a sintering of ceramics at high temperatures are generally non-porous.
[0034] The maximum theoretical maximum number of trays NTHMax of a porous monolith according to the invention will be calculated as first given by the relation: LG NTHMax = c + * P * 1.6 d To take full advantage of the packings according to the invention invention, it will be characterized in that it develops more than 100000 trays.
[0035] The packings according to the invention may develop more than 500,000 trays. Among the methods that can be used to produce these multicapillary packings, mention will be made in particular of the methods described in application VVO 2011/114017. Recall that these methods comprise a method of manufacturing a monolithic porous material comprising substantially straight and parallel capillary conduits between them, characterized in that it comprises the steps of: providing a bundle of fibers known as precursors of the conduits of which the diameter is equal to that of the capillary ducts, - formation of a matrix around the fibers, - removal of the fibers so as to form said capillary ducts. Advantageously, the lining comprises at least a portion of which: the capillary ducts are substantially rectilinear and parallel to one another; the ducts have a diameter that is substantially uniform with respect to each other; the section of each duct is regular over its entire length; all the ducts pass through said part from one side to the other. By substantially uniform diameter is understood in the present text that the standard deviation on the diameter of the ducts does not differ by more than 15% of their average diameter, Advantageously not more than 5% and even more preferably not 2% of their average diameter. By substantially uniform section is meant that the section of a conduit does not vary by more than a factor 3 between two parts of the same conduit. Advantageously, a porous material is selected from the walls having a first population of interconnected pores and with the conduits of diameters such that they allow effective diffusivity of the species to be separated in the walls of the packing at least equal to 10% of their diffusivity. in non-porous free medium. Advantageously, these pores have a mean diameter of between 1 and 2000 nanometers. In order not to lose excessively in specific surface area, it will be advantageous to operate in a range of pore size depending on the application. For condensed phase separation, this population advantageously has an average pore diameter greater than 2 times, preferably greater than 5 times and even more preferentially greater than 10 times the molecular diameter of the molecules to be separated. For condensed phase separation, this population advantageously has an average pore diameter of less than 1000 times, more preferably less than 100 times and even more preferably less than 30 times the molecular diameter of the molecules to be separated. In order not to excessively lose specific surface area and therefore capacity, it will advantageously operate in a range of pore size depending on the application. Advantageously, the average diameter of the communicating pores of the material are greater than 2 times and less than 1000 times, more preferably greater than 2 times and less than 100 times and even more preferably greater than 2 times and less than 30 times the molecular diameter of the molecules to to separate.
[0036] The molecules of organic chemistry are divided into small molecules with a molecular diameter of about 0.3 nanometers, large intermediate-sized molecules such as proteins, and large macromolecules. Advantageously, molecules of molecular weight between 0 and 1000 g / mol will be separated with communicating pore sizes of the material of the walls between the ducts of between 4 and 30 nm, molecules of molecular weight of 10000 g / mol will be separated with As the communicating pore sizes of the wall material between the ducts between 10 and 100 nm, the molecular weight molecules of 100,000 g / mol will be separated with communicating pore sizes of the wall material between the ducts of between 30 to 300 nm. Advantageously, the optimum communicating pore size (in nm) of the material of the walls between the lining ducts will be expressed as a function of the molecular weight MW (in kg / mole) of the species to be separated by the law: ro = 10 * VMW Advantageously these pores have a greater diameter than the average free path of the molecules for application in gas chromatography. Advantageously, this material will contain a second population of pores in open contact with the first population. Advantageously this second population will have a pore diameter smaller than that of the first population.
[0037] Advantageously, the pore volume of the first population will represent more than 40%, preferably more than 50% and even more preferably more than 60% of the pore volume of the porous material of the packing, a pore volume calculated excluding the volume of the ducts. that is, considering the volume of the lining external to the walls of the ducts.
[0038] Advantageously, the pore volume of the second population will represent between 10% and 60% of the pore volume of the porous material of the packing. Advantageously, bimodal monolithic silica gels will be used, the first population being macroporous and the second population mesoporous. Advantageously, this monolith is made of a silica gel.
[0039] Advantageously the first population consists of macropores, and the second population of mesopores. The following publications provide non-limiting examples of methodologies for controlling the pore size of silica gels. These publications are quoted for illustrative purposes and do not constitute an exclusive basis for setting the state of the art. "Shrinkage during drying of silica gel" by DM smith and garlic, Journal of non-crystalline solids, 188, (1995), 191-206, "Pore structure evolution in silica gel during aging / drying Part I, Temporal and thermal aging" Pamela J. Davis, Journal of Non-crystalline solids, 142, (1992), 189-196, "Pore structure evolution in silica gel during aging / drying Part II, Effect of Pore fluids Pamela J. Davis, Journal of non-crystalline solids, 142 (1992), 197-207, "Pore structure evolution in silica gel during aging / drying Part II, Effects of Surface Tension" Ravindra Deshpande, Journal of Non-crystalline solids, 144, (1992), 32-44. In a particularly advantageous manner, the pore size of the lining according to the invention can be perfectly controlled by using a binder charge process for its manufacture. In such a process, a powdery filler is agglomerated by the action of an organic or inorganic binder. The charge may advantageously consist of a porous stationary phase for chromatography such as silica, alumina, cellulose, etc.
[0040] The binder may be a soil, a sol gel process, a clay, a ceramic, a polymer, etc. These products are commercially available with pore sizes and porous characteristics (tortuosity, porous volume or porosity, etc ...) in a wide range and perfectly defined.
[0041] Especially at the Silicycle Company are silicas of defined pore size of 40 Angstroms, 60 Angstroms, 80 Angstroms 90, Angstroms, 100 Angstroms, 120 Angstroms, 150 Angstroms, 300 Angstroms, 1000 Angstroms. These silicas for chromatography are on their catalog available online on their website with morphologies (spherical, irregular) and various particle sizes (from 1.8 to 500 μm).
[0042] These silicas can be milled and sieved to obtain any desired particle size, between 0.2 μm and 500 μm, for example. Their multicapillary packing shaping can be done using a binder, around fibers assembled in a bundle. to form a matrix. The fibers are then removed leaving their imprint in the form of channels in the matrix.
[0043] Advantageously this packing will be performed by a bimodal silica gel for separations of macromolecules: examples of which can be used can be found in the following document: N. Ishizuka, Designing Monolithic Double Pore Silica for High Speed Liquid Chromatography, Journal of Chromatography A, 797 ( 1998), 133-137 Advantageously this packing is made of a monolithic organic polymeric gel. The nature of the gel and its degree of crosslinking will be chosen to select the desired pore size. This pore size and porosity are well documented in commercial literature, scientific literature and patent literature. Polymeric gels include, in particular, but not limited to, Polystyrene Divinylbenzene polymers containing from 2 to 8% of DVB, cyclodextrin grafted silicas, cellulose and its derivatives, in particular its esters, carbamate or amylase, polyholosides, polytriphenyl methyl methacrylate, polypyridyl-2 diphenyl methyl methacrylate, polyacrylates, polyesters, polyvinyl alcohols, crosslinked agaroses and their substituted derivatives, crosslinked dextrans. According to one embodiment, the organic gel is a copolymer of styrene and divinylbenzene. The copolymers of styrene and divinyl benzene exhibit high diffusivity and permeability with respect to molecules dissolved in a solvent. In order to increase this diffusivity, the rate of styrene is decreased. This decrease has the effect of varying the pore size of the material. The weight of divinylbenzene in styrene can be varied between 20% and 2%, a high rate reducing the pore size.
[0044] In particular, these organic gels can be made from monofunctional monomer mixtures and multifunctional monomers polymerized in a porogenic medium. The multifunctional monomers crosslink the polymer obtained. These monomers can be acrylates, methacrylates, acrylamides, methacrylamides, vinylpyrrolidones, vinylacetates, acrylic acid, methacrylic acid, vinyl sulfonic acid, etc. The level of monofunctional monomer can vary between 2% and 98% by weight of the total monomers. Advantageously it is between 2% and 40% by weight of the total monomers. The bi or multifunctional monomers may be monomers based on benzene, naphthalene, pyridine, alkyl ethylene, glycol, etc. having two or more vinyl functional groups. The level of bi or multifunctional monomer may vary between 100% and 2% by weight of the total monomers. Advantageously, said content is between 98% and 60% by weight of the total monomers.
[0045] The porogen is any material or product that can be removed after the polymerization to generate porosity. It can be an organic solvent, water, a decomposable polymer, etc. The choice of porogen and its quantity determines the average size and the size distribution of the pores obtained.
[0046] Advantageously, the volume of porogen is between 20% and 500% of the volume of the monomers or oligomers constituting the organic gel. Even more advantageously, the volume of porogen is between 40% and 300% of the volume of the monomers or oligomers constituting the organic gel. The pore sizes obtained by varying these conditions are well documented and are part of the state of the art. In particular, the following document is consulted for the synthesis of crosslinked agaroses: "Agar derivatives for chromatography, electrophoresis and gel-bound derivatives" Jerker Porath, JC Janson, T. Laas, J. Chromatogr., 60 (1971), 167-177 Crosslinked dextrans and agaroses are known under the trade names Sephadex and Sepharose (a trademark of GE Healthcare). The size of their pores is perfectly mastered and determined by their degree of crosslinking. Figure 1 is a sectional view of a cylindrical multicapillary packing 3 in a direction perpendicular to its major axis. It comprises a porous mass 2 and empty capillary ducts 1 where the fluid passing through the lining 3 can circulate freely. The porous mass is advantageously monolithic. In the case described, the capillary ducts are straight, parallel, and spaced regularly. The different ducts have morphologies and diameters as identical as possible. Each duct passes through the material, that is to say, it advantageously has its ends open on each side 4 and 5 of the cylindrical packing, allowing the flow of fluid from the inlet side to the outlet side. FIG. 2 is a view from above of a face 5 of the cylindrical packing seen in the direction 6. The openings of the individual capillary ducts 1 can be distinguished in the mass 2. FIG. 3 represents the computer-simulated physical diagram. The conduits distributed on a square or hexagonal mesh exchange material with their neighbors by molecular diffusion through their walls.
[0047] Figure 4 shows the computer simulated numerical scheme. The multicapillary column is discretized into slices or cells along the axis of the fluid flow (arrow D1). Inside each slice the section of each conduit is discretized into cylindrical symmetry cells (arrow D2). The porous material is characterized by a porosity or void volume fraction, a stationary phase fraction, a tortuosity and a correction coefficient C of the diffusivity calculated on the basis of an average pore size. Ordinary differential equations (ODEs) describing the material balance of each eluted or eluent compound in each cell are posed and solved sequentially by an explicit Euler numerical integrator with a small time step.
[0048] Figure 5 shows a first result of the simulations. The curves are plotted with the length of the column (in pm) on the abscissa and the number of theoretical plates on the ordinate. The diamond-bounded curve that capped at a theoretical number of plates independent of the length represents the behavior of an independent capillary bundle with solid walls. Said curve represents a packing consisting of channels of randomly variable diameter according to Gaussian statistical law around an average of 10 μm with a standard deviation of 0.5 μm, for non-porous walls. The diameter of the ducts varies statistically according to a normal distribution. The line delimited by squares represents the behavior of the same column with porous walls allowing the molecules of the substances to be separated to diffuse between the ducts. Efficiency increases linearly with length. Diffusion levels differences in behavior between conduits. This straight line represents the same beam as the lower curve with porous walls having 55% of pore volume, a wall thickness of 2 μm and a pore size ten times greater than the molecular diameter of the species to be separated. FIG. 6 represents the concentration profile resulting from the diffusion of a component present initially in a pipe after a time t. The matter diffuses in the other conduits. The bell curve Gaussian type obtained spreads and collapses over time. As a result, the behavior of each duct becomes a cumulative average of contributions from an increasing number of ducts as time passes. The independent behavior of each duct is leveled and permanently becomes the result of an average over a large sample of adjacent ducts. Figure 7 shows this behavior. A central duct diffuses into the peripheral ducts and receives material from them, the quantity of material being all the greater as the color attributed to the duct is dark.
[0049] FIG. 8 represents the correlation obtained between the theoretical plateau height attributable to the dispersion phenomenon Hdisp and the DFactor, both expressed in micrometres. The correlation is linear and excellent throughout the field of operating conditions numerically explored.
[0050] FIG. 9 represents the different characterization variables of a porous material according to the invention. The material seen in section has solid parts 6 separated by a connected network of pores 7 of diameter 10. The volume fraction of vacuum 7 is the porosity P. The average ratio between the path 9 actually traveled by a diffusing molecule and the optimal path 11 is the tortuosity T of the material. It is assumed in this figure that the flow of fluid flows from the top to the bottom of the figure. FIG. 10 schematically represents the ratios between the diameter of the condensed phase diffusing molecules and the diameter of the pores extending through a porous material 12. The large molecules 14 whose diameter is of the order of magnitude the pore diameter is sterically hindered during diffusion. On the other hand, the molecular diffusivity of small species 13 whose diameter is an order of magnitude below that of the pores is not significantly affected. FIG. 11 shows the pressure drop of an aqueous solvent in two columns of the same length, one filled with a spherical particulate stationary phase (45 ° square points), the other with a multi-capillary monolith (points horizontal squares).
[0051] The abscissa carries the diameter of the particles or conduits in pm, the ordinate the pressure drop in bar. The pressure drop is calculated at the optimum efficiency of the column in both cases. It is found that the pressure drop is a lower order of magnitude with capillary ducts. The difference becomes significant in practice for a characteristic diameter of 30 μm.
[0052] Figure 12 shows a Van Deemter curve obtained by simulation. This curve is obtained for a 10 μm diameter capillary surrounded by a porous stationary phase film of 55% porous volume and 2 μm thick. The abscissa axis is the velocity of the mobile phase in the duct in pm / s, the ordinate axis represents the theoretical plateau height in pm. The Van Deemter curve shows that the height of a theoretical plate has a minimum corresponding to the optimum efficiency of the column. In the present text, the molecular diameter will be calculated in two ways depending on the molecular weight and characteristics of the substance under consideration. For substances with a gaseous phase or for which the coordinates of the critical point can be calculated, the covolume, term b of the Van der Vals equation, divided by 4 and the Avogadro number, shall be calculated. the diameter of a sphere of equivalent volume. It is known that the covolume b is equal to four times the molecular volume. The covolume is easily accessible from the critical coordinates of the body. For macromolecules, biological molecules (proteins, etc.) and molecules that do not have a gas phase, the hydrodynamic diameter measured by dynamic light scattering will be used. In the present text, the average diameter and the standard deviation of the diameter of the ducts are calculated by image analysis on a slice of the packing parallel to the ducts. When the ducts have a variability or standard deviation on their upper diameter along the lining at their standard deviation measured on a slice of the packing perpendicular to the ducts, the average diameter is measured on a multiplicity of slices carried along the lining, in order to know the volume of each channel and knowing this one its average diameter on the basis of a constant length equal to that of the packing.
[0053] In the present text, the average thickness and the standard deviation of the thickness of the duct walls are calculated by image analysis obtained by scanning electron microscopy on a wafer of the lining parallel to the ducts. When the ducts have a variability or standard deviation on the upper wall thickness along the lining at their standard deviation measured on a wafer of the lining perpendicular to the ducts, the average wall thickness is measured on a multiplicity of wafers made along of the lining, in order to know the volume of each wall and knowing it its average thickness on the basis of a constant length equal to that of the lining. In the present text, the mean of a set of values of a variable X is its arithmetic mean E [X]. The standard deviation is defined as the square root of the arithmetic mean of (X-E [X]) 2. By distribution, is meant in the present text a set of values of the variable X. Example 1 A thread of a mixture of polymethyl methacrylate (PMMA), is produced with a diameter of 0.3 mm. This wire is cut into straight needles 220 mm long. 200 g of silica gel for the 4 nm pore size chromatography (SiliCycle ref R10030 A) is milled to a mean particle diameter of about 10 μm. The powder is gradually suspended in 500 ml of a mixture of 200 ml of silica sol H530 with 30% dry matter and 300 ml of demineralised water.
[0054] Once the suspension is complete, the PMMA wire is dried under a stream of dry air at 80 ° C. The needles are then cut with an exact length of 200 mm by releasing each side and arranged in a square housing of 3.0 mm side and 200 mm long carved in a sheet of 20x10x200 mm 316L stainless steel, and a flat cover in a sheet of 20x10x200 mm PTFE (Teflon brand registered by DuPont de Nemours). The needles are arranged parallel to each other and regularly in seven successive layers forming a square section in the lower stainless steel housing. A 5 ml mixture of Ludox HS30 (Grace brand, sol of amorphous silica particles 7 nm in diameter, with a specific surface area of 360 m 2 / g, 30% by weight of silica contained) and 0.1 ml of acid acetic acid is prepared. The PMMA needle beam is impregnated with this mixture. The liquid must fill the entire packing, which must be immersed in it. Both parts, stainless steel and teflon are screwed against each other.
[0055] The mixture is maintained at 90 ° C until complete freezing of the soil. The upper PTFE cover holding the bundle is removed from the gel, the ends of the gel are released, and the bundle in its stainless steel housing is gradually raised with a temperature rise of 1 ° C / min up to 500 ° C in a furnace . They are kept 5 hours at this temperature.
[0056] A cover manufactured in a sheet of 20x10x200 mm of 316L stainless steel is replaced to close the packing in substitution of the PTFE half-shell. The monolith is washed with percolated water percolated by the free ducts. The monolith thus obtained can be used directly for the liquid chromatography of molecular weight molecules from 100 g / mol to 200 g / mol.
[0057] Embodiment 2 A yarn of a blend of polymethyl methacrylate (PMMA) is produced with a diameter of 0.3 mm. This wire is cut into straight needles 220 mm long. 200 g of silica gel for 30 nm pore size chromatography (SiliCycle ref S10070 M) is milled to an average particle diameter of about 10 μm. A silica sol of 40 nm particle size and 70 m 2 / g of dry material is obtained by adding deionized sodium silicate (by passage over a cation exchanger), so as to achieve pH 9 and 90 ° C a regular silica deposit on Grace TM 50 sol in dilute solution.
[0058] The powder is gradually suspended in 500 ml of a mixture of 200 ml of silica sol at 50% solids and 40 nm particle size obtained above and 300 ml of demineralized water. Once the suspension is complete, the PMMA wire is dried under a stream of dry air at 80 ° C. The needles are then cut with an exact length of 200 mm by releasing each side and arranged in a square housing of 3.0 mm side and 200 mm long carved in a sheet of 20x10x200 mm 316L stainless steel, and a flat cover in a sheet of 20x10x200 mm PTFE (Teflon brand registered by DuPont de Nemours). The needles are arranged parallel to each other and regularly in seven successive layers forming a square section in the lower stainless steel housing. A mixture of 5 ml of the sol of 40 nm to 50% dry matter and 0.1 ml of acetic acid is prepared. The PMMA needle beam is impregnated with this mixture. The liquid must fill the entire packing, which must be immersed in it. Both parts, stainless steel and teflon are screwed against each other. The mixture is maintained at 90 ° C until complete freezing of the soil. The upper PTFE cover holding the bundle is removed from the gel, the ends of the gel are released, and the bundle in its stainless steel housing is gradually raised with a temperature rise of 1 ° C / min up to 500 ° C in a furnace . They are kept 5 hours at this temperature. A cover manufactured in a sheet of 20x10x200 mm of 316L stainless steel is replaced to close the packing in substitution of the PTFE half-shell. The monolith is washed with percolated water percolated by the free ducts.
[0059] The monolith thus obtained can be used directly for the liquid chromatography of molecular weight molecules of about 1000 g / mol. Example 3 A polydioxanone wire is produced with a diameter of 0.25 mm. The wire is cut into rectilinear sections 15 cm long, covered by dipping with a thin layer of a mixture containing 8 g of hydroxyethyl methacrylate, 32 g of divinylbenzene, 445 mg of azobisisobutyronitrile, 120 g of dodecanol and a volume of Aerosil 380 equivalent to 9/10 of the volume of the solution. The sections are polymerized for 24 hours under nitrogen at 70 ° C. They are then cut into lengths of 120 mm and assembled into a bundle of about 4 mm diameter in a glass tube 100 mm long and 4 mm in diameter previously prepared.
[0060] A mixture containing 16 g of hydroxyethyl methacrylate, 64 g of divinylbenzene, 890 mg of azobisisobutyronitrile and 120 g of dodecanol is prepared in parallel and degassed under nitrogen for 20 minutes. The previously prepared bundle is filled with this mixture and heated to 70 ° C for 24 hours. The mixture polymerizes. The composite thus produced is released by cutting the sections of wire on either side of the glass tube flush with its ends, perpendicular to said sections. The monolith thus produced is washed by percolating THF for 30 minutes and dried in an oven at 90 ° C.
[0061] The polydioxanone son are dissolved in 90 ° C sodium hydroxide percolated through the packing for 1 hour, then the packing is washed with distilled water until neutral. The monolith thus obtained can be used directly for the liquid chromatography of molecules of molecular weight from 500 g / mol to 5000 g / mol. Exemplary embodiment 4 A polydioxanone wire is produced with a diameter of 0.25 mm. The wire is cut into rectilinear sections 15 cm long, covered by dipping with a thin layer of a mixture containing 8 g of hydroxyethyl methacrylate, 32 g of divinylbenzene, 445 mg of azobisisobutyronitrile, 120 g of dodecanol and a volume of Aerosil 380 equivalent to 9/10 of the volume of the solution. The sections are polymerized 24 hours under nitrogen at 70 ° C. They are then cut into lengths of 120 mm and assembled into a bundle of about 4 mm diameter in a glass tube 100 mm long and 4 mm in diameter previously prepared. A mixture containing 8 g of hydroxyethyl methacrylate, 32 g of divinylbenzene, 445 mg of azobisisobutyronitrile and 120 g of dodecanol is prepared in parallel and degassed under nitrogen for 20 minutes. The previously prepared bundle is filled with this mixture and heated to 70 ° C for 24 hours. The mixture polymerizes. The composite thus produced is released by cutting the sections of wire on either side of the glass tube flush with its ends, perpendicular to said sections.
[0062] The monolith thus produced is washed by percolating THF for 30 minutes and dried in an oven at 90 ° C. The polydioxanone son are dissolved in 90 ° C sodium hydroxide percolated through the packing for 1 hour, then the packing is washed with distilled water until neutral.
[0063] One liter of agarose beads are dissolved in one liter of deionized water at 95 ° C.
[0064] The porosity of the monolith is impregnated with this solution by soaking and drainage at 95 ° C of the core of the pipes and cooling. The occluded and frozen agarose is washed with distilled water. One liter of 1N NaOH solution containing 20 ml of epichlorohydrin and 5 g of NaBH4 is prepared. We fill the ducts of the monolith. The monolith is heated to 60 ° C for one hour. The crosslinked gel obtained is washed with hot water until neutral. The monolith thus obtained can be used as a basis for liquid chromatography of proteins.
[0065] Glossary C: coefficient of correction of the diffusion in free medium related to the size of the pores and to the diameter of the molecules to be separated Dréel: molecular diffusivity of a species to be separated in the real medium of packing m2 / s Do: molecular diffusivity in medium free in the mobile phase of a species to be separated m2 / s D,: molecular diffusivity in free medium in the stationary phase of a species to be separated m2 / s DKA Knudsen diffusivity, m2 / s DAB molecular diffusivity M2 / S DAe : average diffusivity m2 / sd: mean diameter of the ducts, meter of: average thickness of the wall separating the ducts, deore meter: pore diameter, m dp-, ': maximum diameter of the ducts allowing a natural equilibrium of a gravitational chromatography, m fvolstat stationary phase volume fraction in the lining wall (m3 / m3) H: overall theoretical plateau height, meter H'p: theoretical plateau height of a single ca pillar of average diameter, meter Hdise: height of theoretical chromatographic plate related to the irregularities of the diameter of the conduits, meter HdispE: height of theoretical plate chromatographic related to the irregularities of the walls of the conduits, Hdiep meter: height of theoretical plate related to the irregularities of the capillaries, meter k ': column retention factor (elution time = minimum retention time * (1 + k')) K: partition coefficient of the species to be separated between the stationary phase and the eluent phase, (mole / m3) / (mole / m3) Kads: partition coefficient in adsorption of the species to be separated, (mole / m2) / (mole / m3) L: distance traveled by the peak eluted in the chromatograph during time t, meter LG: lining length, meter NPT: number of theoretical plates of the column. MA molar mass of component A, kg / mole M: molecular mass of the component to be separated, kg / mole NrnaxD,: maximum number of trays of a chromatographic separation linked to ducts of irregular diameter NmaxD, NmaxE maximum number of trays a chromatographic separation linked to irregular wall ducts N: equivalent number of ducts in interaction Nay: number of Avogadro P: porosity of the porous material constituting the walls of the packing, volume fraction Q: quantity of material injected into the capillary, mol R : distance from the diffusive point, Rh meter: molecular radius of the molecules of the species to be separated, nanometer ro: pore radius, nanometer SigmaD: relative standard deviation (standard deviation / mean) of the hydraulic diameter of the conduits. SigmaE: Relative standard deviation (standard deviation / mean) of the wall thickness. S: specific surface area of the porous material, m2 / m3 t: time since injection, second to: retention time of a non-retained compound, second tR: retention time of a retained R compound, second T: tortuosity TK : absolute temperature, Kelvin vo: elution rate of an unrestrained compound, m / sy, velocity of the mobile phase in the duct, m / s VmoiEi molar volume of the eluent phase, m3 / mol Vmoistat molar volume of the stationary phase, m3 / mole Vs stationary phase volume in the column (m3) Vm mobile phase volume in the column (m3) X: molar fraction, mole / m2 K: Boltzmann constant, MKSA: viscosity of the mobile phase (Pa. $) VVO REFERENCES 2011/114017 K Nakanishi, Phase separation in silica sol-gel system containing polyacrylic acid, Journal of non-crystalline Solids 139 (1992, 1-13 and 14-24 K. Nakanishi, Phase separation in Gelling Silica -Organic Polymer Solution: Systems Containing Poly (sodium styrenesulfonate), J. Am. Ceram, Soc., 74 (10) 2518-2 530-30 (1991) Deen, WM Hindered transport of large molecules in liquid-filled pores, AICh.E Journal, 33,1409-1425 Tortuosity-porosity relation in the media flow, Maciej Matyka, Arzhang Khalili, Zbigniew Koza, 22/01/2008 Shrinkage during drying of silica gel, DM smith et al, Journal of non-crystalline solids, 188, (1995), 191-206, Pore structure evolution in silica gel during aging / drying Part I, Temporal and thermal aging , Pamela J. Davis, Journal of Non-crystalline solids, 142, (1992), 189-196, Pore structure evolution in silica gel during aging / drying Part II, Effect of Pore fluids, Pamela J. Davis, Journal of non-crystalline solids , 142, (1992), 197-207, Pore structure evolution in silica gel during aging / drying Part II, Effects of Surface Voltage, Ravindra Deshpande, Journal of Non-crystalline solids, 144, (1992), 32-44, N. Ishizuka, Designing Monolithic Double Pore Silica for High Speed Liquid Chromatography, Journal of Chromatography A, 797 (1998), 133-137 Agar For the purposes of this paper, see Jerker Porath, J. C. Janson, T. Laas, J. Chromatogr., 60 (1971), 167-177.

权利要求:
Claims (10)
[0001]
REVENDICATIONS1. A chromatography process in which a gaseous, liquid or supercritical mobile phase containing species to be separated is circulated through a packing comprising a stationary phase, said packing being characterized in that: it comprises a plurality of capillary ducts passing through the packing between a so-called upstream face through which the mobile phase enters the lining and a so-called downstream face through which the mobile phase leaves the lining, - the material of the walls of said ducts comprises a network of related pores, said pores forming passages a conduit to the other allowing the molecular diffusion to occur between adjacent conduits, said pores having a mean diameter (dp ') greater than twice the molecular diameter of the species to be separated, - the average diameter of the conduits is less than 50 pm.
[0002]
2. Chromatographic process according to claim 1, wherein the ratio, called "relative height of dispersion", of the theoretical plateau height (Hd, sp) due to inhomogeneities of the lining on the total theoretical plateau height (H) of the lining. is less than 0.66.
[0003]
3. Chromatographic method according to claim 2, in which: the species to be separated have an Rh molecular radius in the elution solvent, a molecular diffusivity Do in the elution solvent, a Ds molecular diffusivity in or on the stationary phase. , a partition coefficient K between the stationary phase and the elution solvent, a retention factor k 'in the chromatographic column, and - the packing comprises conduits of average diameter d, separated by walls of average thickness d, whose irregularity is defined by a standard deviation of the diameter d, brought back to its mean SigmaD and by a standard deviation of the thickness d, brought back to its average SigmaE, - the porous material constituting the walls has a porous volume fraction P, a volume fraction of stationary phase fvoistat or an adsorption specific surface S, a tortuosity T, and the network of related pores has a pore diameter, - the phase m obile flows with the mean velocity v in the ducts and the theoretical plateau height (Hdisp) due to the inhomogeneities of the packing is defined by the relation: (FKD * SigmaD2 + FKE * SigmaE2) *. (dc + de) 2 Hdisp = 0.778 * * FDiff * FDil * (1 + 1e) * 2 where: vo is the elution rate of a compound not retained by the stationary phase FKD = (2 + 3 * 1 + k 'FKE - k0 * PB k0)) (fVo / Stat * K + * of * (2 * dc + of) k0-, 42 (2 * d, + 2 * of) * d, * D0 + [(d, + of) * (dc + de) - d, * ci,] * DE DC = (d, + de) * (d, + of) Do * P * C * (P + fvoistat) Ds * fvoistat * C * K * (P + fvoistat) DE - T * (P + fvoistat * K) T * (P + fvoistat * K) FDil = (ce + ((d, + de) 2 - ce.) * (P + fvoistat * K)) C is a coefficient of correction of the diffusivity in free medium related to the size of the pores and to the diameter of the molecules to be separated.
[0004]
4. Method according to one of claims 1 to 3 characterized in that the ratio said "relative dispersion height" of the theoretical plateau height (Hdisp) due to inhomogeneities of the packing on the total theoretical plateau height (H ) of the packing is calculated at the optimum packing efficiency given by the VanDeemter curve.
[0005]
5. Process according to any one of claims 1 to 4, characterized in that the ducts have an average diameter of less than 30 μm, and preferably less than 10 μm.
[0006]
6. A method according to any one of claims 1 to 5, characterized in that the lining comprises at least a portion of which: - the capillary ducts are substantially straight and parallel to each other PB - (2 * d, + de) dc + The ducts have a section that is substantially uniform with respect to each other, the section of each duct is regular over its entire length, and all of the ducts pass through said portion from one side to the other.
[0007]
7. Method according to one of claims 2 to 4, characterized in that the relative dispersion height is less than 0.3 and preferably less than 0.1.
[0008]
8. Method according to one of claims 1 to 7, characterized in that said pore network of the packing has a mean diameter (dp, ') greater than 5 times the molecular diameter of the species to be separated and preferably greater than 10 times the molecular diameter of the species to be separated.
[0009]
9. Process according to any one of the preceding claims, in which the mobile phase is in the condensed state and said pore network has an average pore diameter greater than 2 nanometers, preferably greater than 10 nanometers, and even more preferentially greater than 100 nm.
[0010]
10. A process according to any one of the preceding claims, wherein the mobile phase is in the gaseous state and the pores have a diameter greater than the mean free path of the molecules.

类似技术:

---

公开号 | 公开日 | 专利标题

---

FR3026313A1|2016-04-01|METHOD OF CHROMATOGRAPHY ON A MULTICAPILLARY TRIM

---

Cheong et al.2013|Molecular imprinted polymers for separation science: A review of reviews

---

Avramescu et al.2003|Preparation of mixed matrix adsorber membranes for protein recovery

---

Avramescu et al.2003|Mixed-matrix membrane adsorbers for protein separation

---

Urban2016|Current trends in the development of porous polymer monoliths for the separation of small molecules

---

Malkin et al.2010|Submicrometer plate heights for capillaries packed with silica colloidal crystals

---

JP6858701B2|2021-04-14|Organic gel or liquid chromatography method

---

Wang et al.2019|Construction of β-cyclodextrin covalent organic framework-modified chiral stationary phase for chiral separation

---

Billen et al.2007|Understanding and design of existing and future chromatographic support formats

---

EP2547440A2|2013-01-23|Multicapillary monolith

---

Barth et al.2018|Monolayer microgel composite membranes with tunable permeability

---

EP2755751B1|2019-05-15|Packing for a chromatography column and production process

---

AU2011302277A1|2013-05-02|Chromatography membranes for the purification of chiral compounds

---

Hara et al.2017|Chromatographic properties of minimal aspect ratio monolithic silica columns

---

Wu et al.2014|Fabrication and evaluation of artemisinin-imprinted composite membranes by developing a surface functional monomer-directing prepolymerization system

---

Reich et al.2019|Transport under confinement: Hindrance factors for diffusion in core-shell and fully porous particles with different mesopore space morphologies

---

Renkecz et al.2012|In situ synthesis of molecularly imprinted nanoparticles in porous support membranes using high‐viscosity polymerization solvents

---

Sugiyama et al.2012|Lithium isotope separation with displacement chromatography using crown ether resin immobilized on porous silica beads

---

Low et al.2012|Interaction of isothermal phase inversion and membrane formulation for pathogens detection in water

---

Smook et al.2020|Friends, foes, and favorites: Relative interactions determine how polymer brushes absorb vapors of binary solvents

---

Wang et al.2020|Scalable manufacturing of hierarchical biphasic bicontinuous structures via vaporization-induced phase separation |

---

Zhang et al.2017|Preparation of novel three-dimensionally ordered macroporous molecularly imprinted microspheres and its recognition for proteins

---

Zhang et al.2022|The fabrication of porous hollow polysulfone microspheres with PEG as a porogen for methylene blue adsorption

---

Wang et al.2003|Determination of the competitive adsorption isotherms of nadolol enantiomers by an improved h-root method

---

FR3041547B1|2019-09-20|METHOD OF CHROMATOGRAPHY ON A POROUS TRIM MADE BY STRETCHING

同族专利:

---

公开号 | 公开日

---

EP3200889A1|2017-08-09|

---

FR3026313B1|2018-07-13|

---

US20170259190A1|2017-09-14|

---

WO2016050797A1|2016-04-07|

引用文献:

---

公开号 | 申请日 | 公开日 | 申请人 | 专利标题

---

WO2011114017A2|2010-03-15|2011-09-22|Parmentier Francois|Multicapillary monolith|

---

WO2013064754A1|2011-09-15|2013-05-10|Parmentier Francois|Multi-capillary monolith made from amorphous silica and/or activated alumina|FR3049874A1|2016-04-06|2017-10-13|Francois Parmentier|CHROMATOGRAPHY PROCESS|

---

WO2019073124A1|2017-10-12|2019-04-18|Parmentier Francois|Chromatography method|

---

FR3026312B1|2014-09-29|2018-07-13|Francois Parmentier|PROCESS FOR CHROMATOGRAPHY ON A GEL OR ORGANIC LIQUID|

---

CN106922168B|2014-11-11|2019-09-03|巴斯夫欧洲公司|The storage container of integral type formed body including porosu solid|

---

FR3041547B1|2015-09-29|2019-09-20|Francois Parmentier|METHOD OF CHROMATOGRAPHY ON A POROUS TRIM MADE BY STRETCHING|

法律状态:
2015-09-25| PLFP| Fee payment|Year of fee payment: 2 |

---

2016-04-01| PLSC| Search report ready|Effective date: 20160401 |

---

2016-09-30| PLFP| Fee payment|Year of fee payment: 3 |

---

2017-09-26| PLFP| Fee payment|Year of fee payment: 4 |

---

2019-09-26| PLFP| Fee payment|Year of fee payment: 6 |

---

2021-02-24| PLFP| Fee payment|Year of fee payment: 7 |

---

2021-09-30| PLFP| Fee payment|Year of fee payment: 8 |

优先权:

---

申请号 | 申请日 | 专利标题

---

FR1459176A|FR3026313B1|2014-09-29|2014-09-29|METHOD OF CHROMATOGRAPHY ON A MULTICAPILLARY TRIM|FR1459176A| FR3026313B1|2014-09-29|2014-09-29|METHOD OF CHROMATOGRAPHY ON A MULTICAPILLARY TRIM|

---

EP15770952.8A| EP3200889A1|2014-09-29|2015-09-29|Multicapillary packing chromatography method|

---

PCT/EP2015/072474| WO2016050797A1|2014-09-29|2015-09-29|Multicapillary packing chromatography method|

---

US15/514,918| US20170259190A1|2014-09-29|2015-09-29|Multicapillary packing chromatography method|