High Throughput Chemical Speciation

FIELD OF THE INVENTION

[0001] This invention relates to methods, systems and apparatus for separating and characterizing multi-component samples, and in particular a high throughput method, system and apparatus for the multi-dimensional separation and characterization of multi- component samples. This invention also relates to the characterization of one or more multi-component samples by multi-dimensional separation of the sample and characterization of the relative acidity (or acid number) and liydrophobicity of the sample.

BACKGROUND

[0002] Unblended crude oils contain a variety of non-purely hydrocarbon impurities, or "species", for example acids, sulphur compounds and nitrogen compounds. Different species cause a range of different problems in refineries. Because virtually all modern refineries use feedstocks which are blends of different crudes, rather than pure crudes, the effect of the varying species in the crudes can be difficult to predict and to manage. This is because once the feedstock is blended, a particular species may migrate to a different fraction (i.e. a particular "cut" of the feedstock having a particular boiling point range). An understanding of the distribution of the various species would provide extremely useful operating information for the refinery. Such information (generally known as "speciation") is, however, extremely difficult and time-consuming to obtain by traditional methods.

[0003] Multi-dimensional high-performance liquid chromatography systems are known in the art. See for example, Murphy et al. , Effect of Sampling Rate on Resolution in Comprehensive Two-Dimensional Liquid Chromatography. Anal. Chem. 70, 1585- 1594 (1998); and Murphy et al., One- and Two-Dimentional Chromatographic Analysis of Alcohol Ethoxylates. Anal. Chem. 70, 4353-4360 (1998). In general, high throughput multi-dimensional chromatography is known, see US Patent 6,866,786 and the references cited therein.

[0004] Although the methods and systems disclosed to date in the art have proven to be useful for characterizing a variety of sample types, e.g., biological and non-biological polymer samples, they generally suffer from inefficiencies with respect to overall sample

throughput, and/or with respect to complicated control and/or operation schemes and systems. Accordingly, there remains a need in the art for improved methods and systems for effecting multi-dimensional separation for characterization of complex multi-component samples.

SUMMARY

[0005] The invention provides apparatus and systems, as well as methods of using the same, which are useful in the testing of complex materials for the species or types of molecules that are present in the complex samples, and preferably including the quantity of such species. The systems and methods test multi-component samples for two or more properties, preferably in a high throughput manner. In particular, the high throughput system balances throughput with selectivity by using a high selectivity system for the first dimension and a high resolution system for the second dimension, with the throughput for the first and second dimension subsystems being matched to provide a useful effect. In further embodiments, additional tests can be performed, such a test on species functionalities. In other aspects, at least two properties of a particular species are mapped in at least two dimensions

[0006] Therefore, in general, in one aspect, the invention provides a high throughput system for characterizing a multi-component sample, comprising: a) a first dimension subsystem adapted to receive and separate in parallel an array of multi-component samples into two or more arrays of corresponding components, wherein the array of multi-component samples comprises at least 5 different samples, and the first dimension subsystem also being adapted to provide information about a first relative property of the two or more arrays of corresponding components; b) a second dimension subsystem adapted to receive the two or more arrays of corresponding components and separate the corresponding components and the second dimension subsystem also being adapted to provide information about a second relative property of the two or more subcomponents; and c) a processing subsystem adapted to receive information about the first relative property and information about the second relative property, the processing subsystem mapping the first relative property versus the second relative property to provide a characterization of the multi-component samples in the array of multi- component samples. This aspect can be advantageously used in a high throughput method with the first dimension subsystem being a high selectivity subsystem that

separates in parallel targeted components. Preferably the first dimension subsystem is an extraction, such as a liquid-liquid, liquid-gas, liquid-solid or solid-gas extraction, with a solid-phase extraction being especially preferred. In the first dimension, a series of eluents are used to separate the multi-component samples into corresponding components, with the components being separated into corresponding sub-components in the second dimension. The second dimension subsystem is a high resolution system and can be a number of different system types, with HPLC being especially preferred. [0007] Thus, in another aspect, the method is a high throughput method, comprising: a) providing an array of multi-component samples, the array comprising at least 5 samples, b) separating each sample in the array into two or more components by a parallel technique, the parallel technique substantially delineating targeted components from the samples, the components being different from each other, and determining a first relative property of the components of each sample, c) thereafter separating each of the two or more components of each sample in the array into two or more subcomponents by a second relative property using a high resolution technique and determining a second relative property of each of the two or more sub-components, and d) characterizing each sample in the array by mapping the first relative property versus the second relative property to create at least a two dimensional characterization. [0008] The mapping of the first dimension versus the second dimension provides a characterization of the multi-component sample by providing a two dimension layout particular to the sample. The mapping advantageously provides a view of the sample that can be combined with other mappings to characterize a broader complex sample. [0009] Another aspect of this invention is directed toward a single sample method and system for particular property measurements. In this aspect, the invention is directed toward a method comprising a) providing a multi-component sample, b) separating the multi-component sample into two or more components by relative acid strength or relative base strength and determining a relative acid or base strength for each of the two or more components, c) separating each of the two or more components into two or more sub-components by relative hydrophobicity and determining a relative hydrophobicity for each of the two or more sub-components, and d) characterizing the multi-component sample by mapping the relative acid or base strength versus the relative hydrophobicity to create at least a two dimensional view of the multi-component sample.

[0010] The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims.

DESCRIPTION OF DRAWINGS

[0011] FIG. 1 is a flow chart illustrating a variety of methods in accord with this invention;

[0012] FIG 2 is a diagram showing relative resolution and selectivity of separation systems useful in the invention;

[0013] FIG. 3 is a diagram showing data reduction of the information about the first and second properties into the form of a mapping fingerprint;

[0014] FIG 4. is a diagram showing the mappings of the invention split into zones for application to response factors; and

[0015] FIG. 5 is a diagram showing graphing of the mappings of the invention in combination to show the characterization of a broader sample comprised of a series of multi-component samples.

DETAILED DESCRIPTION

[0016] In the present invention, methods, systems and apparatus are disclosed for characterization of single multi-component samples and/or for characterization of a library comprising a plurality of multi-component polymer samples. This invention combines, in a high throughput workflow, a first dimension separation that is preferably a high selectivity separation with a second dimension separation that is preferably a high resolution separation. The first dimension separation generally delineates one component from the remainder of the sample or from other components, e.g., dividing the sample into at least two groups wherein the first group is substantially enriched in the targeted components relative to the initial sample and the second group is depleted of the targeted components. Thus, in some embodiments, the first dimension separation targets components for removal or separation from the initial sample in a high selectivity process. This first dimension is also preferably performed in parallel. Preferably, the characterization methods, systems and apparatus can be applied for fingerprinting multi- component samples by acid strength and hydrophobicity — for analysis in an analytical

laboratory, or for analysis in an on-line, near real time process monitoring or process control system.

[0017] The invention is described in further detail below with reference to the figures, in which like items are numbered the same in the several figures. Generally, the terminology used in this application will be as follows: a multi-component sample is derived from a complex sample. The multi-component sample is comprised of two or three or four or more components. The components themselves also typically have a variety of sub-components. Thus, the hierarchy is complex samples are made up of multi-component samples, which are made up of components, which are made up of subcomponents.

[0018] A multi-component samples characterization system of the present invention comprises a first dimension subsystem and a second dimension subsystem, and optionally, third dimension and/or fourth dimension and/or additional dimension subsystems. The first dimension subsystem is generally a high selectivity system, and is preferably performed in a high throughput manner. A high selectivity system means a method and apparatus that has high affinity for particularly selected or desired components or sub-components of a multi-component sample, as discussed herein. The first dimension subsystem separates the multi-component sample into at least two components. In more preferred embodiments, the first dimension subsystem separates the multi-component sample into at least three components and in even more preferred embodiments, the first dimension subsystem separates the multi-component sample into at least four components. During or after such separation, information about a first relative property of the components can be derived or produced by the first dimension subsystem. The second dimension subsystem is generally a high resolution system. A high resolution system means a method and apparatus that separates the components or sub-components of a multi-component sample with high precision, as discussed herein. The second dimension subsystem further separates the at least two components of the multi-component sample into two or more sub-components. In more preferred embodiments, the second dimension subsystem separates the at least three components of the multi-component sample into two or more sub-components, into three or more sub-components or into four or more sub-components; and in even more preferred embodiments, the second dimension subsystem separates the at least four components of the multi-component sample into two or more sub-components, into three or more sub-

components or into four or more sub-components. During or after such separation, information about a second relative property of the sub-components can be derived or produced by the dimension subsystem.

[0019] Generally, the first relative property of the multi-component sample is a compilation of a property that is measured or determined on the components during or after the separation in the first dimension subsystem. Also generally, the second property of the multi-component sample is a complication of a property measurement of the sub-components during or after the separation in the second dimension subsystem, with the property measurement on the sub-component(s) being compiled into a second property of the relevant component. The first and second property measurements are relative properties, meaning that the property determination is relative to the other components or sub-components tested in the system. Thus, the compilation of the property measurement for the multi-component sample is relative to other multi- component samples that are tested in the system. The property measurements can be correlated to other, known measurement techniques, such as mass spectrometry, as discussed herein. The first and second property determinations are mapped or laid out or plotted versus each other, as in an x and y plot. This mapping provides a broader relative characterization of the multi-component fluid, which can be considered a fingerprint of the multi-component fluid because of its unique characterization of the sample. [0020] The system can be run in a high throughput methodology and thus can be a high throughput system, hi general, in one embodiment, the invention is a high throughput method, comprising: a) providing an array of multi-component samples, said array comprising at least 5 samples, b) separating each sample in the array into two or more components by a parallel technique, wherein the parallel technique has high selectivity for desired components or the parallel technique substantially delineating targeted components from the samples, the components being different from each other, and determining a first relative property of the components of each sample, c) thereafter separating each of the two or more components of each sample in the array into two or more sub-components by a second relative property using a high resolution technique and deteπnining a second relative property of each of the two or more sub-components, and

d) characterizing the each sample in the array by plotting the first relative property versus the second relative property to create at least a two dimensional characterization. In step b), substantially delineating further comprises dividing the sample into at least two groups wherein the first group is substantially enriched in the components that are targeted relative to the initial sample and the second group is depleted of the targeted components. Components can be targeted in a number of ways, for example, by using a certain eluent. Moreover, step b) is unique in that it is adapted to handle a specified volume of sample, for example, it has the capacity to separate no more than 1000 mg and no less than 100 μg of targeted components. [0021] In other embodiments, there will be at 3 or more components separated from each sample, 4 or more components separated from each sample or 5 or more components separated from each sample. Other high throughput aspects of the invention can be used, such as parallel processing of the components or sub-components in the second dimension subsystem. In another embodiment, the invention is a high throughput system for characterizing a multi-component sample, comprising: a) a first dimension subsystem adapted to receive and separate in parallel an array of multi-component samples into two or more arrays of corresponding components, wherein the array of multi-component samples comprises at least 5 different samples, and the first dimension subsystem also being adapted to provide information about a first relative property of the two or more arrays of corresponding components; b) a second dimension subsystem adapted to receive the two or more arrays of corresponding components and separate the corresponding components and the second dimension subsystem also being adapted to provide information about a second relative property of the two or more sub-components; and c) a processing subsystem adapted to receive information about the first relative property and information about the second relative property, the processing subsystem laying out the first relative property versus the second relative property to provide a characterization of the multi-component samples in the array of multi-component samples.

[0022] The subsystems are adapted to provide information about the first and second relative properties of the components or sub-components in that one of skill in the art can determine or derive the relative property as a result of the chemistry and/or processes involved or the subsystem is configured to output the information (e.g., using a computer

and appropriate software). Moreover, the properties of one sample, component or subcomponent are relative to the other samples, components or sub-components. [0023] In one preferred embodiment, the first relative property determination is a determination of the acid strength of the multi-component sample. The acid strength can be any one of a number of different measurements, observations or processes. In some embodiments, the acid strength is defined in terms of the equilibrium constant K _a and pK _a = -log K _a . Thus, the pKa value of acids can be related to the acid strength as is the pH. As those of skill in the art appreciate, a substance that has a pH below 7 is considered acidic, with lower numbers indicating stronger acids. Measurement of the pH and pK _a values of a material are within the skill of those of skill in the art. In other embodiments, the acid strength can be the ionic strength of the concentration of another acid (e.g., eluent) that is capable of solubilizing the sample. In still other embodiments, as this application involves separation by acid strength, the acid strength can be the time of elution or the order of elution as the result of a separation, hi some embodiments, the pK _a is preferred because of its more absolute or descriptive nature. However, in a high throughput separation method, the elution time or the order of elution are preferred because of their ability to be determined as the separation occurs. [0024] hi a related manner known to those of skill in the art, the acid strength can also be a measure of the acid number or total acid number in the multi-component sample. The acid number is used to quantify the amount of acid present in the multi-component sample and can be expressed as a quantity of base, for example expressed in milligrams of potassium hydroxide that is required to neutralize the acidic constituents in 1 g of sample (or mgKOH/g). In another preferred embodiment, the second property determination is determination of the hydrophobicity of the multi-component sample. The hydrophobicity of the components of the multi-component sample are measured separately and then combined to provide an overall determination of the hydrophobicity. Hydrophobicity is the measure of the component's or sub-component's aversion to water, and can be related to a number of other properties of the multi-component sample. Such other properties include the carbon number, which is the approximate number of carbon atoms in a chain of the molecules that predominate in the components. The relationship is that the more hydrophobic the sample, then the larger the carbon number. Another other property is the boiling range of the components of the sample, which is the temperature range in which the molecules in the component (or sub-component) will

vaporize. Here, the relationship is that the more hydrophobic the component or subcomponent, then the higher the boiling range. [0025] Thus, the invention herein in one embodiment is a method, comprising: a) providing a multi-component sample, b) separating two or more components of the sample by relative acid strength and determining a relative acid strength for each of the two or more components, c) separating each of the two or more components by relative hydrophobicity and deteraiining a relative hydrophobicity for each of the two or more components or subcomponents thereof, and d) characterizing the multi-component sample by laying out the relative acid strength versus the relative hydrophobicity for the two or more components to create at least a two dimensional view of the multi-component sample.

[0026] In this embodiment, the relative hydrophobicity can be more quantitative if references exist for the components. Some standards do exist (for example total acid number or total base number) and these can be used for calibration. However, in many embodiments, no references exist, which makes the measurement a relative one. The invention can be carried out in an embodiment comprising a system for carrying out the method, and thus the invention herein, in another embodiment, is a system for characterizing the acid strength versus hydrophobicity of a multi-component sample, comprising: a) a first dimension subsystem adapted to receive and separate a multi-component sample into two or more components based on the acid strength of the components and the first dimension subsystem also being adapted to provide information about the acid strength of the components; b) a second dimension subsystem adapted to receive the two or more components and separate the components on the basis of hydrophobicity and the second dimension subsystem also being adapted to provide information about the hydrophobicity of components; and c) a processing subsystem adapted to receive information about the acid strength and the hydrophobicity of the components, the processing subsystem laying out the acid strength versus the hydrophobicity as a characterization of the multi-component sample. [0027] The speciation technique is a result of a unique combination of multiple separation steps and provides multidimensional molecular distribution information of

high-enough quality (resolution, sensitivity) at previously unachievable speed. The speciation is well applicable to a large number of diverse petroleum samples with purpose of the quick pre-assesment of their corrosivity and acid-related refinery behavior, which is based on determination of the acid species concentration and strength, and optionally functionality or composition. Generally, the multi-component samples can be tested for acids, bases, sulphur compounds or sulfur containing compounds (e.g., sulfidic, sulfonic, etc.) , nitrogen compounds or nitrogen containing compounds (e.g., amines, nitro-containing molecules), oxygen containing compounds (e.g., ketones, aldehydes, ethers, etc.), metal containing compounds (e.g., Ni, V, Fe, etc.) or hydrocarbon compounds.

[0028] Even more specifically, and in the most preferred embodiment, the system and method can be used for acid speciation, which is generally determining the species of acids in a complex sample. The system and method have four steps: 1) sample preparation: the acids are extracted from the multi-component sample, together with other polars in a liquid-liquid extraction, using a blend of organic and/or inorganic bases, organic solvents and water. Alternatively a simple solid phase extraction process based on silica-filled 96-well microtiter plate can be used for extraction of polars, under the conditions of flash or normal-phase chromatography; 2) first dimension separation and first property determination: the acid strength separation is performed by selective retention of the acids on a 96-well solid phase extraction plate filled with an ion exchange resin having selectivity for acids and subsequent multi-step elution using several buffers of increased acidity (various concentration of an acid in an aqueous mixture), which separates acid components in the multi-component sample by acid strength and wherein the repeated multi-step elutions provide information about the acid strength (e.g., order of elution or time of elution) that is inputted into the processing system for mapping or layout; 3) second dimension separation and second property determination: each effluent from the first dimension (i.e., component) is then separated by hydrophobicity by rapid serial or parallel high performance liquid chromatography (HPLC) with UV- Vis detection (e.g., 200-380nm absorbance), using a reversed-phase column with a mixture of solutions used as the mobile phase (for example, water/tetrahydrofuran/cyclohexane); and 4) layout or mapping and sample characterization: after background subtraction, the HPLC traces corresponding to individual components (e.g., solid phase extraction elutions) for each multi-component

sample are recombined into a two dimensional acid distribution map (the layout), in which the x-y-z coordinates are made of the HPLC retention time, the solid phase extraction elution order and the HPLC detection signal, and optionally the HPLC traces are integrated into several zones and each peak area is converted into the acid abundance values (mgKOH/g equivalents) by multiplying with zone-specific response factors. [0029] Now referring to Figure 1 , there is shown a block diagram of a method in accord with the invention herein. The method begins with a supply of multi-component samples 10. The multi-component samples can come from any available source, but typically comes from a source where measurement of the samples provides some added value, such as refinery feed streams or pharmaceuticals, as discussed below. The multi- component samples can be liquid or solid or a combination thereof at room temperature and pressure. Generally, the multi-component samples being characterized are fractions of portions of other complex materials, such as a refinery feedstock, including a crude oil, a synthetic crude, a biocomponent, an intermediate stream (such as a residue, gas oil, vacuum gas oil, naphtha or cracked stock), and blends of one or more of these materials (such as a blend of one or more crude oils or a blend of one or more crude oils with one or more synthetic crudes). Other complex materials include pharmaceuticals or biomaterials, foods (such as vegetable oils, olive oils, other oils), spirits (such as wine, beer or other alcoholic beverages). In preferred embodiments, the multi-component samples are libraries of fractions of crude oil, with the libraries of multi-component samples provided on a common substrate. The libraries of multi-components samples can be prepared in rapid-serial fashion or parallel using, for example, a micro-fluidic fraction separation device and method. The multi-component samples can be untreated, or pretreated with one or more steps selected from the group consisting of non- chromatographic separation, dilution, mixing and redissolution. [0030] The multi-component samples are typically placed into an array format 20, although this step is optional as the system and method are also operative on single samples. Arraying the multi-component samples, however, provides several benefits, including the ability to take several samples from a single broader sample and place them in relation to each other in the array. This relationship in the array allows for ease in sample tracking, parallel processing under substantially similar conditions, and high throughput treatment of the array. If the multi-component samples in the array are solid at room temperatures and pressures, the samples can be heated or otherwise put into a

form to flow. The samples can be arrayed manually or in an automated fashion using a liquid or solid handling robot, such as those available from Symyx Technologies, Inc., Santa Clara, CA. Arrays of multi-component samples are provided, wherein there are at least 5 different multi-component samples in the array. In other embodiments, there are at least 10 different multi-component samples in the array, at least 20 different multi- component samples in the array, at least 40 different multi-component samples in the array, or at least 80 different multi-component samples in the array. The array may also contain blanks or references to be used as standards or relative comparisons. A plurality of multi-component samples comprises 2 or more samples that are physically or temporally separated from each other - for example, by residing in different sample containers, by having a membrane, wall or other partitioning material positioned between samples, by being partitioned {e.g., in-line) with an intervening fluid, by being temporally separated in a flow process line {e.g., as sampled for process control purposes), or otherwise. Eight samples can be employed, for example, in connection with experiments having seven fractions of one broader complex sample and one control sample, with the different fractions being representative of high, medium and/or low fractions of the varied factor (e.g., varied boiling or varied acid number). Moreover, eight samples corresponds to the number of samples in a typical column of a 96-well microtitre plate, allowing for fractions from a broader complex sample to be grouped, handled and processed in parallel and/or together. Higher numbers of samples can be investigated, according to the systems and methods of the invention, to provide additional insights into larger compositional and/or process space. In some cases, for example, the plurality of samples can be 15 or more samples, preferably 20 or more samples, more preferably 40 or more samples and even more preferably 80 or more samples. Such numbers can be loosely associated with standard configurations of standard sample containers {e.g., 96-well microtiter plate-type formats). Moreover, even larger numbers of samples can be characterized according to the methods of the present invention for larger scale endeavors. Hence, the number of samples can be 150 or more, 400 or more, 500 or more, 750 or more, 1,000 or more, 1,500 or more, 2,000 or more, 5,000 or more and 10,000 or more multi-component samples. As such, the number of samples can range from about 2 samples to about 10,000 samples, and preferably from about 8 samples to about 10,000 samples. In many applications, however, the number of samples can range from about 80 samples to about 1500 samples. In some cases, in

which processing of samples using typical 96-well microtiter-plate formatting is convenient or otherwise desirable, the number of samples can be 96*N, where N is an integer ranging from about 1 to about 100. For many applications, N can suitably range from 1 to about 20, and in some cases, from 1 to about 5.

[0031] The plurality of samples can be a combinatorial library of samples. A library of samples comprises of two or more different samples, and can be in an array format as spatially separated samples - preferably on a common substrate, or temporally separated - for example, in a flow system. Candidate samples (i.e., members) within a library may differ in a definable and typically predefined way, including with regard to chemical structure, processing (e.g., synthesis) history, mixtures of interacting components, purity, etc. The samples can be spatially separated, preferably at an exposed surface of the substrate, such that the samples in the array are separately addressable for sampling into the characterization system and subsequent characterization thereof. The two or more different samples can reside in sample containers formed as wells in a surface of the substrate. The number of samples included within the library can generally be the same as the number of samples included within the plurality of samples, as discussed above. In general, however, not all of the samples within a library of samples need to be different samples. When process conditions are to be evaluated, the libraries may contain only one type of sample. Specifically, a different multi-component sample can be included within at least about 50%, at least 75%, at least 80%, at least 90%, at least 95%, at least 98% and most specifically at least 99% of the samples included in the sample library. In some cases, all of the multi-component samples in a library of multi- component samples will be different from each other.

[0032] The substrate can be a structure having a rigid or semi-rigid surface on which or into which the array of multi-component samples can be formed or deposited. The substrate can be of any suitable material, and preferably consists essentially of materials that are inert with respect to the polymer samples of interest. Certain materials will, therefore, be less desirably employed as a substrate material for certain process conditions. Stainless steel, silicon, including polycrystalline silicon, single-crystal silicon, sputtered silicon, and silica (SiO ₂ ) in any of its forms (quartz, glass, etc.) are preferred substrate materials. Other known materials (e.g., silicon nitride, silicon carbide, metal oxides (e.g., alumina), mixed metal oxides, metal halides (e.g., magnesium chloride), minerals, zeolites, and ceramics) may also be suitable for a

substrate material in some applications. Organic and inorganic polymers may also be suitably employed in some applications of the invention. Exemplary polymeric materials that can be suitable as a substrate material in particular applications include polyimides such as Kapton™, polypropylene, polytetrafluoroethylene (PTFE) and/or polyether etherketone (PEEK), among others. The substrate material is also preferably selected for suitability in connection with known fabrication techniques. As to form, the sample containers formed in, at or on a substrate can be preferably, but are not necessarily, arranged in a substantially flat, substantially planar surface of the substrate. The sample containers can be formed in a surface of the substrate as dimples, wells, raised regions, trenches, or the like. Non-conventional substrate-based sample containers, such as relatively flat surfaces having surface-modified regions (e.g., selectively wettable regions) can also be employed. The overall size and/or shape of the substrate is not limiting to the invention. The size and shape can be chosen, however, to be compatible with commercial availability, existing fabrication techniques, and/or with known or later- developed automation techniques, including automated sampling and automated substrate-handling devices. The substrate is also preferably sized to be portable by humans. The substrate can be thermally insulated, particularly for high-temperature and/or low-temperature applications. In preferred embodiments, the substrate is designed such that the individually addressable regions of the substrate can act as vessels for preparing a multi-component sample for further processing. Glass-lined, 96-well, 384-well and 1536-well micro titer-type plates, fabricated from stainless steel and/or aluminum, are preferred substrates for a library of multi-component samples. The choice of an appropriate specific substrate material and/or form for certain applications will be apparent to those of skill in the art in view of the guidance provided herein. [0033] Continuing with Figure 1, another optional step 30 is the removal of molecules that may interfere with the property determinations and either disposal or use of the interfering molecules in another process 40. The removal step 30 can also be considered a sample preparation step and take a number of forms and use a number of different processes, as discussed in detail below. For example, the removal step can be a liquid- liquid extraction, solid phase extraction, chromatography or some other separation. Also, removal step 30 can take place manually or in an automated methodology, in serial sample processing or parallel processing. If a removal step 30 is performed, the multi- component sample can be taken directly from this step and used in or injected into the

first dimension subsystem. In a preferred embodiment, the removal step is employed to separate polar molecules in the multi-component sample from non-polar molecules. In these preferred embodiments, a liquid-liquid extraction or solid phase extraction is employed. Generally, for liquid-liquid extraction, the multi-component samples are contacted with a solution comprising a combination of polar and non-polar solvents in the array format the samples are provided in from step 20. Solutions are dispensed automatedly, in parallel or otherwise (e.g, manually or with a automated dispenser), and the array is shaken or stirred or blended to allow for intimate contact, and then allowed to rest for separation of the extraction solutions. The polar and non-polar solvents are chosen for their compatibility with the materials being extracted, and in preferred embodiments are chosen to extract polar molecules from non-polar molecules using a mixture of solutions containing solvents, buffers, etc. The solutions can be made according to the knowledge of one of skill in the art, and ingredients in the LLE solutions include water, amines, alcohols, hydrocarbons, (e.g., chlorinated hydrocarbons, ethers, ketones), organic acids, inorganic acids, organic bases, inorganic bases, organic salts, inorganic salts etc. If a removal process 30 is employed, in preferred embodiments, the process is checked to determine that a fair representation of multi-components samples is removed, without leaving behind molecules that should be detected in the first and/or second dimension subsystems. One check of this removal process is the correlation to the TAN (or total base number (TBN)), as discussed below for a broader complex sample.

[0034] As also shown in Figure 1, there is a first dimension subsystem 50 that performs high selectivity separation and first property determination. The first dimension subsystem is capable of separating two or more components of each multi- component sample or array of multi-components with high selectivity so that the components are different from each other. In this context, components or subcomponents are considered different from each other if they have different distribution profiles; thus, the components or sub-components may have the same type of molecules present, but the molecules would be present in different quantities, at least on a relative basis. In other words, a high selectivity system means a method and apparatus that separates the components or sub-components of a multi-component sample with an ability to differentiate molecules based on their molecular characteristics (size, chemical

composition, functionality) via specific selective interactions between the system and the molecules.

[0035] As those of skill in the art will appreciate, selectivity relates to the relative distance on the response curve between two peaks from a defined start of the separation, as shown in Figure 2. A high selectivity refers to a large distance between the peaks and a low selectivity refers to a small distance between the peaks, relative to the start of the separation. Efficiency is inversely proportional to the width of the peaks, with a narrower peak width representing a higher efficiency and a larger peak width representing a lower efficiency, as shown in Figure 2. The high efficiency can be achieved with a high number of repetitions of the interaction equilibria. The resulting resolution is the ability to differentiate two response peaks on the curve. Providing that there is sufficient selectivity, a satisfactory separation can be achieved without high efficiency, as opposed to the high efficiency system that differentiates molecules primarily via a high number of repetitions of the interaction (e.g., high number of theoretical plates) rather than relying on the selectivity of the interaction itself. Figure 2 demonstrates this by showing that the resolution for the desired type of molecule is preserved by either a combination of high selectivity and low efficiency or low selectivity and high efficiency.

[0036] The first dimension subsystem can use a number of different techniques to achieve the parallel separation, including parallel liquid-solid extraction, parallel liquid- liquid extraction, parallel gas-liquid extraction, parallel gas-solid extractions, parallel precipitation-dissolution, parallel filtration-dialysis, parallel electrical field separation, parallel electrophoresis and parallel chromatography. Details of each of these techniques are known to those of skill in the art, but they are generally based on an equilibrium between two phases (like liquid versus solid) or differential migration through a separation medium (like a capillary filled with a buffer or a column filled with a chromatographic material). For example, capillary electrophoresis separation is one where a sample is injected into a capillary column containing a separation buffer and an electrical charge is applies to force analytes to migrate to one of the electrodes at a speed dictated by their electrophoretic mobility. Parallel capillary electrophoresis separation devices are know (for example, cePRO 9600™ System for pK _a Determination available from CombiSep, Inc., Ames, Iowa). Also for example, parallel or multi-channel capillary chromatography is a technique in which molecules are injected into a mobile

phase that drives them through a stationary phase (or separation medium or chromatographic packing material) and the separation is achieved based on interactions between the analytes and the chromatographic system (including both the stationary and mobile phases). To enhance separation, gradient elution using a mobile phase with a changing composition is typically applied. Parallel or multi-channel HPLC systems are commercially available from Eksigent, Inc., Dublin CA (e.g., the ExpressLC 800). Solid phase extraction is a separation where a solid material (e.g., ion exchange beads) having an affinity for a certain type of molecule is placed in the sample, the sample is washed and subsequently the molecules that attach or associate with the solid material is eluted by contact with liquid. Commercially available parallel systems are cited below. [0037] Each of these techniques provides information about the resulting effluent, which is the information for the first property determination. For example, a pH gradient in a parallel HPLC will provide the relative pH of the effluents by giving the order of elution based on pH, which would be used as the relative acid strength and can be correlated to the carbon number and compiled to total acid number of a complex sample. Those separations that take place in a column (e.g., electrophoresis) can be coupled to a detection device, such as a mass or concentration detector. Selection of one of these techniques depends on the particular type or character of multi-component samples that are to be characterized by the system or method as well as the desired throughput of the system. The throughput for the first dimension subsystem is generally at least about 48 multi-component samples per day, more specifically at least about 96 multi-component samples per day, and even more specifically at least about 144 multi-component samples per day. The selection of the first dimension subsystem can also depend on the format, and a preferred embodiment is a parallel format for its ability to keep samples together and increase the throughput.

[0038] hi a preferred embodiment, a solid phase extraction is used wherein the solid phase is an ion exchange resin having an affinity for acids or bases. These ion exchange resins selectively retain desired components (e.g., acids or bases or acidic functionalities or basic functionalities). After washing, the components of the multi-component sample are eluted from the solid phase with a solution selective for the first property, hi embodiments where acid strength is determined, the ion exchange resin has selectivity for acids and the acid components are eluted using buffers of different acidity, hi embodiments where base strength is determined, the ion exchange resin has selectivity

for bases and the basic components are eluted using buffers of different basicity. Typically, the order of elution is determined by using buffers of increasing strength, for example increasing acidity, with buffers having lower acidity being used before buffers having higher acidity. In this manner, components are separated by the strength of the acid in which they elute, proving the information for the first property determination of acid strength. The number of elusions can be at least two (providing two or more components), at least three (providing three or more components), at least four (providing four or more components) or at least five (providing five or more components). When preformed in parallel, each elution of an array of multi-component samples, produces an corresponding array of components. Thus, two elusions provides two arrays of corresponding components, three elusions provides three arrays of corresponding components, four elusions provides four arrays of corresponding components and five elusions provides five arrays of corresponding components. The solid phase extraction resin can be a adsorbent (e.g., silica, alumina, activated carbon, etc.), a hydrophobic medium (e.g., alkyl silica, cross-linked polystyrene, etc.), an ion exchange medium, either cation exchange or anion exchange (e.g., polystyrene resin or silica gel modified with a sulfonic group or quaternary amine, etc.). Parallel solid phase extraction ion exchange systems are readily available, such as Oasis MAX 96-Well Plates from Waters, Inc., Milford, MA; Strata SAX Plates available Phenomenex, USA, Torrance, CA; SAX plates from Supelco and available from Sigma- Aldrich; MultiPROBE® II HT PLUS from Perkin-Elmers, Inc., Wellesley, MA; SPE 215 System from Gilson, Inc., Middleton, WI; and Speedisk system from JT Baker, Phillipsburg, NJ. [0039] In some embodiments, the same type of acid buffer solution is used for each elution, but is more dilute or concentrated (e.g., same acid with more or less water added). In some embodiments, a different type of acid buffer solution is used for two or more of the elusions (e.g., a different acid or the same acid for two elutions, but a different acid for a third elution). The acid buffer solution is selected based on the targeted components of the first component as effluent as well as the presence of the buffer solution with the components in the second dimension subsystem, hi those embodiments where the buffer solution is not separated from the components, the buffer will be chosen to not substantially interfere with the second property determination in the second dimension subsystem. For example, with an acid buffer solution, the acid would be chosen to be either much smaller than the acids in the components or much larger

than the acids so that when the components are processed in the second dimension subsystem, the acid used to elute the components can be removed from the second property determination. Also, the acid buffer could be chosen to be not detected or low signal by the detector used in the second dimension subsystem. In preferred embodiments, the solution(s) used for elution is an acid buffer system comprising a blend of acid in a buffering solution. In preferred embodiments, the elution buffer solution is chosen with a combination of either a strong acid and a weak base in water or a weak acid and a strong base in water. Typically, the elution buffer solution is an aqueous medium comprising amines, carboxylic acids, hydrocarbons and alcohols, such as formic acid, triethylamine, iso-octane and isopropanol. Generally, the acid can be selected from the group consisting of organic acids and inorganic acids. More specifically, the acid can be selected from the group consisting of carboxylic acids having one, two or three carbon atoms (e.g., formic acid or acetic acid); hydrochloric acid; sulfuric acid; nitric acid; and phosphoric acid.

[0040] In these preferred embodiments, information about the first relative property is determined in connection with the first dimension subsystem by recording the order of elution. The order of elution of each component of a multi-component sample is relative to the other components. This order of elution is recorded and inputted into the processing subsystem for mapping or layout. For example, as those of skill will be able to perform, bar coding can be used to set an order of elution for a series of components or arrays of components. Thus, the input of the information regarding acid strength to the processing subsystem can be automated or manual from the first dimension subsystem.

[0041] Referring back to Figure 1, the components or arrays of components move from the first dimension subsystem 50 to the second dimension subsystem 60. Also the information about the first property (e.g., acid strength) moves from the first dimension subsystem 50 to the processing subsystem 70. The second dimension subsystem is a high resolution system having a throughput that approximately matches the throughput of the first dimension subsystem. Thus, the second dimension subsystem is adapted to received the components or arrays of components, separate them and provide information about a second property at the rate of about at least about 96 components per day, more specifically at least about 192 components per day, and even more specifically at least about 288 components per day. The second dimension separation can be

performed using a number different subsystems, including, microfluidic chromatography, capillary chromatography, solid phase extraction, gas chromatography or mass spectrometry. The second dimension subsystem is preferably a high- performance liquid chromatography subsystem, although other subsystems can be used such as electrophoresis, electrochromatography, solid phase elution or other types of separation and characterization subsystems. Each of these techniques is known in the art, for example as discussed in US Patent 6,866,786 and US Patent 5,670,054, each of which is incorporated herein by reference.

[0042] A detection system to provide information about the second property of the sub-components, components and ultimately the multi-component samples is part of the second dimension subsystem to detect a property of the two or more subcomponents separated from the components injected into the second dimension subsystem. Detection in the second dimension can generally be effected serially {e.g., components are injected serially into a mobile-phase eluent or parallel columns have mobile-phases that are sent serially to a single detector) or in parallel. In parallel second-dimension detection embodiments, each of the two or more chromatographic columns can have its own dedicated detector, such that detection of subcomponents derived from different components occurs substantially simultaneously as compared between different analysis channels of the second dimension. For any given component (or portion thereof) of the first-dimension, however, once injected into a particular analysis channel of the second dimension, detection of properties of the second-dimension separated subcomponents is effected serially within that analysis channel. The second dimension detector can be a mass detector and/or concentration detector. Optical detector detectors are preferred in some embodiments. An optical detector can be advantageous applied, particularly in highly parallel systems, and/or in systems designed to be effective for nano-scale and/or micro-scale analysis (e.g., lab-on-a-chip applications). An optical detector, such as a light-scattering detector, UV-vis absorbance, fluorescence, evaporative light scattering, chemiluminescence or other optical detector, can be applied directly to samples that can be detected by the optical detector, hi some cases, however, the detectability of the separated subcomponents can be developed, for example, by treating second-dimension separated subcomponents to change an optical property thereof before or after separation, but before detection with an optical detector (e.g., derivatization or complexing). A number of different treating agents can be used, including complexation agents (e.g., iron

salts, copper salts, multifunctional acids) or derivatization agents (e.g., o- phthaldialdebyde, 1-pyrenyldiazomethane and 5-(bromomethyl)fluorsecein). The derivatization or complexation agent can also be added to the mobile phase of the second dimension subsystem, discussed below.

[0043] The number of parallel second-dimension chromatographic columns, and associated second-dimension mobile phases is not of crucial significance, but is preferably four or more second-dimension chromatographic columns, and more preferably eight or more second-dimension chromatographic columns. Higher or lower numbers can also be employed, as described for example in US Patents 6,461,515, 6,776,902 and 6,491,816 and the references cited therein. Exemplary preferred second dimension HPLC subsystems include reverse phase chromatography subsystems, mobile-phase compositional gradient elution chromatography subsystems, or mobile- phase temperature gradient elution chromatography subsystems. Mobile-phase elution gradients of the second dimension preferably comprise a substantially universal co- solvent system, such as a water-tetrahydrofuran-hexane system. The column choice for the HPLC channel(s) is within the skill of one of ordinary skill in the art, with reversed- phase column being particularly preferred. In addition to these disclosures, serial injection is not necessary, and as such a injection port for each column may be employed, advantageously with liquid handling robot that an automatically inject the components into the mobile-phase of the second dimension subsystem. In preferred embodiments, a UV-vis absorbance detector is used on each channel of a parallel capillary HPLC subsystem, with UV-vis and flurosecent spectroscopy providing flexibility (especially in combination with pre-derivatization) and compatibility with a high throughput system. Such systems can be built in accord with this disclosure or purchased commercially, such as from ExpressLC 100 or ExpressLC 800 available from Eksigent, Inc., Dublin CA. Other commercially available systems that can be used in the second dimension subsystem include known micro-fluidic devices (e.g., Brio cartridge and the Veloce LC system, available from Nanostream, Inc., Pasadena, CA) and known single-channel rapid serial HPLC devices with autosampler (e.g., Alliance available from Waters and the 1100 Series available from Agilent). Selection of appropriate equipment for the systems and methods of this invention are within the scope of one of skill in the art in view of this disclosure.

[0044] In general, the method of operation of the second dimension subsystem is receipt of the components from the first dimension subsystem. The components are injected into an HLPC system, separated into sub-components and a relative second property of the sub-components is detected by the detector. Information about the relative second property of the sub-components is sent to the processor subsystem, and used for mapping or layout of a two dimensional characterization of the multi- component samples. The solvent gradient starts with a solvent combination that is targeted toward strong retention of targeted components and ends with a solvent combination that is targeted toward substantially-complete elution of the targeted components. The mobile-phase in the HPLC column is preferable operated in a reverse- phase method using a mobile phase gradient of increasing hydrophobicity. The mobile- phase is preferable a mixture of solutions, including ingredients selected from the group consisting of water, tetrahydrofuran, cyclohexane, isooctane, acetonitrile, isopropanol, carboxylic acids having one, two or three carbon atoms (e.g., formic acid or acetic acid), hydrochloric acid, sulfuric acid, nitric acid, phosphoric acid and combinations thereof. Those of skill in the art can chose appropriate mobile-phases in accord with the specific application. To keep the throughput at a rate that matches the first dimension subsystem, the second dimension subsystem can be optimized by considering column geometry (e.g., shorter and wider columns), bead selection for the columns and solvent gradient ramp rates. These optimizations are within the skill of one of ordinary skill in the art in view of this disclosure.

[0045] Additional dimensions can be added to the first and second dimension subsystems, with each additional dimension providing information about at least one additional property of the multi-component samples (e.g., components and/or subcomponents), hi one preferred embodiment, a third dimension is assess the chemical functionality of the components or sub-components. Chemical functionality refers to the functionalities on the molecules in the components or sub-components; for example with regard to acid speciation, chemical functionality refers to the type of acids that are present (for example, aliphatic acids, aromatic acids, etc.). A third dimension can be added as part of an additional subsystem (e.g., a third dimension subsystem) or can be accomplished with modification of the first or second dimension subsystems. For example, sorbent affinity, chemiluminescence (e.g., chemistry specific detection), dying, solid-phase extraction and thin layer chromatography can be used to determine a third

property, such as chemical functionality. In one preferred embodiment, the second dimension subsystem uses a UV- vis absorbance detector, where the wavelength of detection is used as a third dimension property determination, e.g., correlated to the chemical functionality. In another preferred embodiment, the second dimension subsystem is modified to detect chemical functionality using a UV-vis absorbance detector with derivatization (e.g., a fourth dimension). More specifically, additional information on the chemical character of the acids can be obtained considering the whole spectrum analysis from the second dimension detector, and optional mobile phase additive acid derivatization strategies (e.g., iron complexation, hydrazine reaction, etc.). A fourth dimension subsystem or fifth dimension subsystem can also be added, with the additional dimensions preferably not impeding the throughput of the overall system and method.

[0046] Information for the first and second properties are sent to and/or received by processing subsystem 70, in which a map or layout of the first and second properties is prepared to characterize the multi-component sample. The processing subsystem 70 also can be adapted to process the information or data relating to the first and second properties, including data reduction, background elimination, data selection and data mining, hi more detail, the raw data from the second dimension subsystem is a matrix of detector responses over time at a variety of wavelengths. The detector responses are typically processed to focus on a selected wavelength (e.g., 330nm, normalized to 530nm signal), in preferred embodiments, to reduce the data and possibly eliminate from consideration the response from the elution solution used in the first dimension. The selected HPLC traces for individual components are then corresponded to the information on the first property (in preferred embodiments namely the order of the elution according to acid strength) to create a two-dimension map of that characterizes the multi-component sample, i.e., a fingerprint of the sample. This is shown in Figure 3, wherein there is shown HPLC traces 100, 101, 102 and 103. The traces 100, 101, 102 and 103 are arranged according to the first property (e.g., order of elution) from the first dimension subsystem. A portion of the traces are focused upon, as shown by box 120, and the data is converted into a two-dimensional mapping 150, with the detector response providing the different shading seen in mapping 150. [0047] The raw data from the HPLC can be further processed by eliminating background effects, for example, background HPLC traces can be obtained for a non-

acidic oil control sample (e.g., low acidity distillation fraction or non-polar solvent such as dodecane or cyclohexane) and processed through the first and second dimension subsystems. This eliminates or minimizes an effect of interferences introduced to the final response by the first dimension subsystem (e.g., formic acid from the eluent). After the background removal, the response from the HPLC detector is integrated and split into several individual zones in a 2x2, 2x3, 3x3, 4x3, or 4x4 or other grid, representing the number of components separated and the number of sub-components. The layout or mapping can be split into 4 or more zones, 8 or more zones or 12 or more zones. This is shown in Figure 4, where there are four components (split along the property 1 axis) and three sub-components (split along the property 2 axis), and wherein each number represents a different zone (e.g., zone 1, zone 2, ... zone 12). hi preferred embodiments, map is of the acid sub-components that differ in hydrophobicity (and relatedly carbon number or molecular size or boiling range) along the property 2 axis versus the acid strength (e.g., order of elution based on the acidity of the solid-phase eluent) along the property 1 axis.

[0048] hi addition, in preferred embodiments, the intensities in different zones in the map or layout are weighted by response factors. These response factors take into account the fact that different types of components or sub-components (e.g., acids) may have different responses in the chosen detector in the second dimension subsystem, with a sub-component giving a disproportionate response compared to other sub-components. For example, different acids give different signals, which can be correlated to a more consistent basis, for example only, in other words, a strong acid may give a much larger response because its acidity instead of its quantity. The disproportionate responses can be brought back into proportion by multiplying each zone by a response factor. Thus, looking at Figure 4, zone 1 would have a different response factor than zone 9, etc. The response factors useful for each zone can be determined by those of skill in the art by operating the system with a set of validation standards. Validation standards are typically known samples for which the expected response of the system is known, hi a preferred embodiment, the validation standards are acid mixtures that are characterized by a known system (e.g., mass spectrometry or acid-base titration) and run through the first and second dimension subsystems. The detector responses are then plotted in a least square fit parameter, representing a measure of the contribution of individual zone signals to the total acidity for the validation samples. Generally, the stronger and larger

the acid, the weaker appears to be their detector response, and thus these sub-components would have larger response factors. The total acidity can then be calculated by summing the abundance of acids from all zones, proportioned by the response factors. Other statistical processing can also be used, such as determining standard deviations for each zone, model fitting, etc.

[0049] In a further embodiment, each multi-component sample is just one piece of a broader complex sample (e.g., a boiling range fraction) and the methodology is repeated for each piece of the broader complex sample, preferably in parallel. In this fashion, the property determinations for the multi-component samples are combined to characterize the broader sample. In other words, the properties from the first and second dimension subsystems for multi-component samples can be collected and correlated to produce a plot of the properties of a broader, complex sample (such as a crude oil or wine or pharmaceutical). In the case of refinery feedstocks, for example, the acidity abundance property from the mapping or layout can be summed to yield the total acidity and correlated to the tolal acidity number (TAN) as measured by standard ASTM techniques (e.g., ASTM D 664 and/or ASTM D974). This is shown in Figure 5. As shown in figure 5, there multiple mappings 200, 201, 202, 203, 204 and 205, from the property determinations at are collected into a single plot 210. When the property 2 in the plot 210 is the acid abundance (detector response from hydrophobic separation in the second dimension subsystem). The integrated area underneath the envelope curve 220 represents the total acidity of the broader complex sample (e.g., crude oil), which can be correlated to the TAN from the ASTM standards. Thus, for example, the acid speciation results for a single broader complex sample across a given property (e.g., the hydrophobicity or boiling range) can be composed together to give a complete picture of the acid distribution relevant to the refinery process.

[0050] The multi-dimensional system of the invention is preferably operated under the control of one or more microprocessors (not shown), preferably configured with software effective for operating the hardware (sampling systems, injection valves, mobile-phase pumps, detection systems) and for effecting sample tracking and acquiring data, etc. Such suitable software is commercially available, for example, from liquid chromatography systems manufacturers, such as Renaissance Software (Symyx Technologies, Inc., Santa Clara, CA), Millenium software (Waters), Eksigent's Control Software and/or from software manufacturers, such as Lab View brand (National

Instruments) software. The software can, if necessary, be modified to incorporate functionality for driving the aforementioned hardware and data tracking and acquisition needs for the first and second dimension subsystems.

[0051] In a preferred embodiment, for example, with reference to Figure 1, Lab View software can be modified to (i) integrate with Impressionist™ robotic-control software (Symyx Technolgies, Inc., Santa Clara, CA) used for controlling the robotic pipette (Cavro Instruments, Inc.) hardware for serially withdrawing aliquot samples 100 from a library of multi-component or component samples, and for injecting such samples into a loading port of a first or second-dimension injector, such integration including tracking of timing of injection as an intiation point for the two-dimensional analysis methodologies programmed into the Lab View software; and (ii) to control second dimension HPLC analysis operations, including mobile phase pumps to control mobile phase flow rates, and if desired, temperature control of the mobile phase and/or column, and if desired, mobile phase source selection valves (not shown) for providing mobile phase gradients for gradient elution chromatography.

[0052] The following example illustrates the principles and advantages of the invention.

EXAMPLE

[0053] This example uses fractions of crude oil, which can be any crude oil. An array of 96 crude oil fractionated samples is provided in a substrate having 96 wells glass lined with glass vials.

Liquid-Liquid Extraction (LLE)

[0054] The samples are being targeted to be approximately 500 mg each, and each is weighed using either manual weighing or a Bohdan automated weighing station. ImI per

500 mg of sample of a hydrocarbon solvent is added to each well and the array is mixed well on a shaker plate. The array can be heated to ensure that any solid samples are flowable. 4ml of extractant solution is added to each vial, with the extractant solution composed of 80 parts isopropanol and 20 parts of IOOM triethylamine in water. The vials are rigorously shaken for at least one hour, and then centrifuged for at least 5 minutes. An aliquot of each vial is taken from the aqueous portion of the extract solution.

First Dimension Separation

[0055] The first dimension separation is a solid phase extraction using a commercially available 96-well plate having an ion-exchange resin with high affinity for acids, specifically the Oasis MAX from Waters (60 mg of ion-exchange resin per well, 30 micron resin particle size) fitted with a Speedisk® pressure processor and manifold air pressure of 30 psi. Each well of the solid-phase extraction plate is preconditioned with isopropanol and water. The aliquot of sample from the aqueous fraction of the LLE step is loaded in steps into each well of the 96-well SPE plate, either manually with a parallel pipette or with an automated liquid handling robot. The extracts are pushed through the SPE plate slowly with no or minimum pressure applied. After loading is completed, isopropanol followed by a mixture of isopropanol and water are pushed slowly through the SPE under the same conditions.

[0056] Elution is performed in parallel for all 96 wells and in stages starting with the weakest acidic eluent and ending with the strongest acidic eluent. hi this example, there are four elusions. In the first elution, 0.5 niL of 0.178 M formic acid is loaded onto each well and pushed through at the pressure adjusted to maintain a steady flow of about 0.1 mL/min across the plate. Up to 0.5 niL of the effluent is collected into a separate 96-well microtiter plate and cooled. In the second elution, 0.5 mL of 0.356 M formic acid is loaded onto each well and pushed through at the pressure adjusted to maintain the same flow. Again, up to 0.5 mL of the effluent is collected into a separate 96-well microtiter plate, and cooled, hi the third elution, 0.5 mL of 0.890 M formic acid is loaded onto each well in four equal portions and pushed through at the pressure adjusted to maintain the same flow rate. Again, up to 0.5 mL of the effluent is collected into a separate 96- well microtiter plate, and cooled, hi the fourth elution, 0.5 mL of 1.425 M formic acid is loaded onto each well in four equal portions and pushed through at the pressure adjusted to maintain the same flow rate. Again, up to 0.5 mL of the effluent is collected into a separate 96-well microtiter plate, and cooled. The order of elution is recorded by coding each sample with a sample name that reflects its order of elution (e.g., using a processing system readable bar code). Second Dimension Separation

[0057] The four plates with effluents from the anion-exchange separation are put on the platform of a robotic liquid handling robot associated with a parallel capillary HPLC device, equipped with a UV- vis absorbance detector at the end of each of eight columns. Specifically, an 8-channel capillary HPLC ExpressLC™-800 from Eksigent, with robotic

autosampler HTS Pal from CTC Analytics / Leap Technologies is useful. There is an injection port and a sample valve holding two sample loops for each column, and each column is a reverse-phase column packed with 3.5 micron octyl-silica particles. Commercially available software runs the robot for injection, pumping and data collection from the detector. A gradient elution is used starting with a MobilePhase-A and moving to MobilePhase-B. MobilePhase-A is a mixture of 4 ingredients, including a majority of water and THF and a minority of two hydrocarbon solvents. MobilePhase-B is a mixture of 4 ingredients, including a minority of water and THF and a majority of two hydrocarbon solvents.

[0058] An aliquot of each of the effluent sample is injected onto the mobile phase of a capillary HPLC column. After injection a gradient elution program is executed, starting with 26 μL/min of 25 equivalents of MobilePhase-A and 1 equivalent of MobilePhase-B and holding it for about 15 seconds, followed by changing the mobile phase flow rate and composition to 37 μL/min of 1 equivalent of MobilePhase-A and 35 equivalents of MobilePhase-B in about 90 seconds, holding that composition for about 3 min, and subsequently returning back to original flow rate and composition. Each sample can take about 5 minutes, with 8 samples being run simultaneously, requiring about 1 hour per plate or all 384 samples in 4 hours. In addition, the robotic autosampler fills the injection loop with solvent between samples to clean the column between samples. Each HPLC separation produces a trace of time vs. detector response at UV- Vis absorbance wavelengths in the range of 200-380 nm. Data Reduction and Mapping

[0059] For each HPLC trace, a 10 nm bandwidth is extracted from the raw data at an appropriate wavelength and baseline-corrected using the signal at 530 nm at 50 nm bandwidth. Background profiles are prepared by performing both first and second dimension separations as discussed above using samples that contain no fractioned crude oil. These background profiles are then extracted from each individual sample trace. The four background-extracted traces per fractionated crude oil sample, representing the order of elution from the first dimension, are re-composed into two dimensional maps (as shown in Figure 3), in which the HPLC elution time makes X-axis, the order of the SPE elution makes makes the Y-axis, and the HPLC detector response makes the Z-axis. The background-extracted traces are also integrated into three zones of elution times, yielding peak areas per zone. The peak areas are then multiplied by the zone-specific response

factors, yielding the acidity abundance values per each zone in equivalents of mg KOH per g of fractionated crude oil sample. The acidity abundance values for each zone (4 SPE effluents x 3 HPLC elution time ranges) are summed together, yielding the total acidity value of petroleum sample, which can then compared to the TAN of the crude oil sample prepared by ASTM standards.

[0060] A number of embodiments of the invention have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the invention.