IMPROVED HLB COPOLYMERS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001 ] This application claims the benefit of priority of U.S. Provisional Patent Application No. 62/549,771 filed August 24, 2017, the entirety of which is incorporated herein by reference.

BACKGROUND

[0002] An omnipresent desire within analytical chemistry is the development of a "universal" approach to sample preparation that is useful for a wide variety of matrices, from environmental and agrochemical samples, to biological samples, to pharmaceutical samples, to foods and beverages. These samples typically include both analytes of interest as well as unwanted or interfering analytes. Solid phase extraction (SPE) has long been used as a sample preparation technique, capable of matrix simplification by removing unwanted analytes, as well as trace enrichment for analytes of interest.

[0003] For SPE applications, typical adsorbents consist of silica, carbon or other adsorbents typically used as liquid chromatography (LC) and HPLC packing. These adsorbents are limited first in the unwanted analytes that can be removed from a sample as each may work well for some classes of analytes, but not as well for others. Additionally, these materials must be activated or pre-treated prior to use to allow for optimal performance, adding additional steps and time necessary for an analysis. Additionally, performance can be lost if the material is allowed to dry during use - a problem known as the overdrying effect.

[0004] Although the number and variety of analytes in any given sample has posed a great challenge, the discovery of hydrophilic-lipophilic balance (HLB) polymers for solid phase extraction (SPE) has increased the feasibility of the idea of a generic SPE sorbent. HLB technology is the most widely used SPE material due to the versatility and wide range of analytes encompassed. HLB resins are copolymers made up of a hydrophilic monomer and a lipophilic monomer. Although still considered a reversed-phase sorbent, the presence of the hydrophilic monomer increases the retention of polar analytes, allowing a single extraction to be effective for a wide range of analytes (e.g., acids, neutrals, bases, polar, and non-polar).

[0005] Currently, there are several HLB SPE products on the market. The most common HLB sorbent is a co-polymer containing /V-vinylpyrrolidone (VP) as the hydrophilic component and divinylbenzene (DVB) as the lipophilic monomer.

[0006] While widely used for HLB SPE applications, the nature of vinylpyrrolidone- divinylbenzene copolymers, however, limits their usefulness. These vinylpyrrolidone and divinylbenzene copolymers are not repeatable and result in retention and peak shape variability in chromatographic separation processes. The water solubility of vinylpyrrolidone results in the polymerization of the vinylpyrrolidone at the co-monomer droplet/water interface during the suspension polymerization process. The use of vinylpyrrolidone, which is also extensively used as a surface surfactant in suspension polymerization processes, leads to irreproducibility in the copolymer sequence. Therefore, an increase in the divinylbenzene (hydrophobic) surface in the deeper pores of the spherical, porous polymer leads to biased selectivity within the porous copolymer. This biased selectivity has limited the development of HLB adsorbents capable of effectively adsorbing analytes over a wide range of sizes and chemistries. A need exists for improved HLB adsorbents that can effectively, and simultaneously, remove analytes that vary greatly in both size and properties.

[0007] Additionally, conventional SPE products typically follow a traditional SPE workflow having 5 steps: conditioning, equilibration, loading, washing, and elution. The ability of a HLB sorbent to successfully extract a variety of analytes can reduce the total number of extractions that are required, however the process can still be quite time consuming. While some HLB adsorbents have been adapted to work in a simplified process, a need still exists for improved HLB adsorbents capable of extracting analytes without the need for conditioning and equilibrium steps. [0008] Accordingly, it would be useful to develop new polymeric adsorbents useful for SPE and other separation techniques. Ideally, the new polymers would be useful for efficient, simultaneous extraction of analytes of varying properties including acids, bases, neutral analytes, strongly polar, moderately polar, and non-polar analytes, and of varying size, including small molecules and large molecules, in a variety of solvents and over a wide pH range. Such new polymeric adsorbents should ideally be resistant to overdrying effects seen with many conventional adsorbents, thus allowing their use in a simplified workflow.

SUMMARY

[0009] Provided herein are improved hydrophilic-lipophilic balance polymeric adsorbents for solid phase extraction (SPE) including a divinylbenzene- acrylonitrile copolymer capable of simultaneously adsorbing a plurality of analytes. Such analytes may include acids, bases, neutrals, strongly polar, moderately polar, and non-polar analytes, and may include small molecules, large molecules, or a combination of both.

[0010] In some embodiments, the polymer adsorbents may be spherical particles. In some embodiments, the polymeric adsorbents may be coated on or bonded to a substrate, such as a particle or fiber. In other embodiments, the polymer adsorbents may be formed into a substrate.

[001 1 ] Also provided are methods for removing one or more solutes from a solution by contacting the solution with the provided polymeric adsorbent. The methods provided are useful for both nonpolar and polar solvents and are useful over a broad range of pH values. The methods provided can remove a plurality of different solutes having varying properties.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a pore plot overlay illustrating the pore structure of the HLB adsorbent described herein (HLB-67) and two conventional HLB products.

[0013] Figure 2 is a chart illustrating the batch to batch synthetic reproducibility of the HLB copolymers described herein. [0014] Figure 3A shows a density functional theory (DFT) plot for experimental HLB adsorbent HLB-65 on silica particles before and after washing with acetonitrile. Figure 3B shows a density functional theory (DFT) pore plot illustrating the uniform pore decrease using the silica coating process described herein.

[0015] Figure 4 shows a DFT overlay plot for three experimental HLB adsorbents, HLB-44, HLB-48, and HLB-49, having varying copolymer ratios.

[0016] Figure 5 is a scanning electron micrograph of an HLB polymer bonded to a SPME fiber.

[0017] Figure 6 compares the recoveries of components of a six-analyte mix using six experimental HLB adsorbents and two conventional HLB adsorbents. A traditional 5-step protocol that incorporated a dry step following conditioning and equilibration was used.

[0018] Figure 7 compares the recoveries of components of a six-analyte mix using four experimental HLB adsorbents and two conventional HLB adsorbents. A 3- step protocol was used; no conditioning or equilibration was performed.

[0019] Figure 8 shows the absolute recoveries of 20 analytes extracted from spiked serum under acidic conditions using experimental resins HLB-01 , HLB- 22, and HLB-26.

[0020] Figure 9A shows serum removal data of unextracted calf serum using experimental resins HLB-01 , HLB-22, HLB-26, and Conventional HLB A; Figure 9B shows the same with an expanded y-axis.

[0021 ] Figure 10 shows a comparison of absolute analyte recoveries from spiked matrix samples extracted using a standard 5-step protocol with experimental HLB-67 and Conventional HLB B.

[0022] Figure 1 1 shows a comparison of absolute analyte recoveries from spiked matrix samples extracted using a simplified 3-step protocol with experimental HLB-67 and Conventional HLB B. [0023] Figure 12 is a comparison of absolute recoveries from spiked matrix samples extracted using a 3-step protocol with experimental HLB-67 and Conventional HLB A.

[0024] Figure 13 shows the absolute recovery of six analytes using experimental HLB-01 to extract from a buffer solution.

[0025] Figure 14 shows a comparison of the absolute recovery of an acrylonitrile- containing HLB copolymer described herein, HLB-01 , with Conventional HLB A— a vinylpyrrolidone-containing HLB material.

[0026] Figure 15 shows the absolute recovery of a variety of analytes using silica coated with experimental adsorbent HLB-65.

[0027] Figure 16 shows a comparison of an HLB copolymer HLB-79 coated a silica substrate versus a conventional HLB adsorbent using a 3-step SPE protocol.

DETAILED DESCRIPTION

[0028] The HLB copolymers described herein, also referred to as adsorbents, offer improved separation and removal of a wide variety of analytes of varying properties, including acids, bases, neutral analytes, strongly polar, moderately polar, and non-polar analytes, and of varying size, including small molecules and large molecules, in a variety of solvents. Additionally, they are suitable for use without the need for conditioning and equilibrium steps within the SPE workflow. In addition to being useful for solid phase extraction (SPE) techniques, they are also suitable for solid phase microextraction (SPME) and can also be adapted for HPLC or UHLPC columns.

[0029] The HLB adsorbents for solid phase extraction (SPE) are formed by copolymerizing at least one hydrophobic monomer and at least one hydrophilic monomer. In a preferred embodiment, the hydrophobic monomer includes divinylbenzene and the hydrophilic monomer includes acrylonitrile. Other hydrophobic monomers that could be used in addition to divinylbenzene include monovinyl and polyvinyl monomers such as styrene, trivinylbenzene, divinyltoluene, divinylnaphthalene, divinylpyridine, vinylpyridine, ethylvinylbenzene and divinylxylene. Non-aromatic monomers such as 1 ,5- hexadiene, 2,5-dimethyl-hexadiene, 1 ,7-octadiene, trivinylcyclohexane, polyvinylidine chloride, polyvinyl chloride, and combinations thereof are included. Other hydrophilic monomers that can be used in addition to acrylonitrile include methyl methacrylate, ethyl methacrylate, ethylene glycol diacrylate, diethyleneglycol divinyl ether, 2-cyanoethyl methacrylate, dimethacrylic ester, vinylquinolines and combinations thereof. In a particularly preferred embodiment, the copolymer adsorbent includes only divinylbenzene and acrylonitrile.

[0030] Although not traditionally used in HLB copolymers, acrylonitrile is particularly preferred as the hydrophilic monomer in the HLB copolymers described herein, as it has been found that it offers many advantages over the traditional hydrophilic monomer used in HLB adsorbents, N-vinylpyrrolidone. With a log P of 0.25, it is more hydrophilic than N-vinylpyrrolidone (log P of 0.37). Acrylonitrile is also more effectively incorporated within the polymer across a broader range of monomer ratios, thus allowing for an ability to extract incredibly hydrophilic analytes (log P < 0) and very lipophilic analytes (log P > 6), simultaneously, as shown in Figure 8, discussed below.

[0031 ] Acrylonitrile is also a hygroscopic material, a fact that led to it being used in the development of the super sluper polymer by the U.S. F.D.A. in the 1960s - a polymer that was able to hold 400x its weight. The inventors have found that utilizing a hygroscopic monomer allows for improved performance of this material for SPE and SPME. This allows the material to perform well without the need for a pre-treatment step, commonly referred to as conditioning and equilibration. Therefore, the SPE procedure can be simplified from 5 total steps (conditioning, equilibration, sample loading, washing, and elution) to either 3 steps (sample loading, washing, and elution) or 1 step (sample loading/elution).

[0032] In a preferred embodiment, a porogen is added to the synthesis process to control the pore size distribution of the HLB copolymer. Porogens, formerly referred to as phase extenders, are solvents that are added to polymerization to affect the properties of the polymer being formed. Porogens can affect many properties of a polymer, including, but not limited to surface area, pore volume, pore size, porosity, and hydrophobic-hydrophilic properties. Particular porogens, or combinations of porogens are preferably added to tune the properties of the resulting polymer.

[0033] Suitable porogens include nonpolar or hydrophobic; and polar, or hydrophilic, solvents, and combinations of nonpolar and polar solvents. In a preferred embodiment, a combination of nonpolar and polar porogens are used. Preferred nonpolar porogens used in the preparation of the HLB copolymers described herein include toluene, xylenes, benzene, hexane, cyclohexane, pentane, heptane, octane, nonane, decane, dodecane, isooctane, dichloromethane, chloroform, carbon tetrachloride, and combinations thereof. Preferred polar porogens used in the preparation of the HLB copolymers described herein include benzyl alcohol, butanol, pentanol, hexanol, heptanol, 4- methyl-2-pentanol, isoamyl alcohol, dodecanol, ethyl acetate, 2-ethylhexanol, cyclohexanol, and combinations thereof. One skilled in the art would also be able to choose additional suitable nonpolar and polar porogens to tune the properties of the resulting polymers. In a particularly preferred embodiment, a combination of o-xylene and 4-methyl-2-pentanol are used for porogens.

[0034] In accordance with the methods described herein, the ratio of divinylbenzene and acrylonitrile monomers used in forming the polymeric adsorbent can vary greatly while still achieving suitable separation properties necessary for an HLB polymer.

[0035] In some embodiments, the ratio of hydrophobic monomer to hydrophilic monomer is in the range from 10:90 (wt. %) to 90:10 (wt. %). In further embodiments, ratio of hydrophobic monomer to hydrophilic monomer is in the range from 30:70 (wt. %) to 70:30(wt. %). In further embodiments the ratio of hydrophobic monomer to hydrophilic monomer is in the range from 15:85 (wt. %) to 85:15 (wt. %); from 20:80 (wt. %) to 80:20 (wt. %); from 25:75 (wt. %) to 75:25 (wt. %); from 30:70 (wt. %) to 70:30 (wt. %), from 35:65 (wt. %) to 65:35 (wt. %); from 40:60 (wt. %) to 60:40 (wt. %); from 55:45 (wt. %) to 45:55 (wt. %); or approximately 50:50 (wt. %). In a preferred embodiment, the ratio of hydrophobic monomer to hydrophilic monomer is about 63:37 (wt. %). [0036] In one preferred embodiment, the polymeric adsorbent is poly(divinylbenzene-co-acrylonitrile). In some embodiments, the ratio of divinylbenzene to acrylonitrile is in the range from 90:10 (wt. %) to 10:90 (wt. %). In further embodiments the ratio of divinylbenzene to acrylonitrile is in the range from 15:85 (wt. %) to 85:15 (wt. %); from 20:80 (wt. %) to 80:20 (wt. %); from 25:75 (wt. %) to 75:25 (wt. %); from 30:70 (wt. %) to 70:30 (wt. %), from 35:65 (wt. %) to 65:35 (wt. %); from 40:60 (wt. %) to 60:40 (wt. %). In one preferred embodiment, the ratio of divinylbenzene to acrylonitrile is about 75:25 (wt. %). In still further preferred embodiments, the ratio of divinylbenzene to acrylonitrile is about 63:37 (wt. %).

[0037] In certain preferred embodiments, the polymeric adsorbent disclosed herein is adapted to simultaneously adsorb a plurality of analytes including acids, bases, neutrals, strongly polar, moderately polar, and non-polar analytes. Exemplary analytes may include small molecules, large molecules, proteins, phospholipids, and pharmaceutically relevant molecules.

[0038] Some typical analytes of interest may include, but are not limited to compounds fitting into these broad classes: essential oils, food preservatives, vitamins, plasticizers, pesticides, steroids, hydrocarbons, pollutants, amines, alcohols, dyes, phenols, alkaloids, mycotoxins, amino acids, flavonoids, heterocyclic compounds, lipids, steroids, organic acids, terpenes, surfactants, carbohydrates, catecholamines, antibiotics, amino acids, nucleic acids, nucleic acid bases, proteins, and peptides.

[0039] Some non-limiting examples of pharmaceutical analytes of interest include: acetaminophen, amiloride, amitriptyline, atrazine, benzoylecgonine, buspirone, clarithromycin, 2,7-dihydroxynaphthalene, doxepin, haloperidol, hydroquinidine, imidacloprid, imipramine, imiquimod, loratadine, losartan, mesoridazine, methapyrilene, mianserin, mirtazapine, nefazodone, nevirapine, nizatidine, procainamide, propranolol, and p-toluamide.

[0040] In certain embodiments, the polymeric adsorbent is a porous spherical particle. When spherical particles are desired, the particle size is not particularly limited. Particles can be formed with favorable, reproducible particle size distributions over a wide range, allowing use of these HLB copolymers in a variety of applications, including, but not limited to SPE, SPME and HPLC.

[0041 ] In various embodiments, the HLB adsorbents may be porous spherical particles having a diameter in the range from about 10 nm to about 1 mm. In some embodiments, the porous spherical particles have diameters in the range from about 20 μιη to about 125 μιη. In certain embodiments, the porous spherical particle has a diameter in the range from about 30 μιη to about 85 μιη. For use in SPE, the spherical particles have a diameter in the range from about 40 to about 75 μιη. In a preferred embodiment for SPE, the spherical particles have a diameter of approximately 50 μιη. In some embodiments, the porous spherical particle has a diameter in the range from about 10 nm to about 10 μιη. For SPME, the spherical particles preferably have a diameter in the range from about 0.5 μιη to about 4 μιη; and more preferably, a diameter in the range from about 0.8 μιη to about 2 μιη. The use of monodispersed particle sizes is desirable for improved sample preparation processes.

[0042] In some embodiments, the porous spherical particles have a surface area in the range from about 10 m ² /g to 1000 m ² /g. In some embodiments, the porous spherical particles have a surface area in the range from about 350 m ² /g to about 675 m ² /g. In some embodiments, the surface area is about 350 m ² /g; in other embodiments, the surface area is about 375 m ² /g, in other embodiments, the surface area is about 400 m ² /g; in other embodiments, the surface area is about 425 m ² /g; in other embodiments, the surface area is about 450 m ² /g; in other embodiments, the surface area is about 475 m ² /g; in other embodiments, the surface area is about 500 m ² /g; in other embodiments, the surface area is about 525 m ² /g; in other embodiments, the surface area is about 550 m ² /g; in other embodiments, the surface area is about 575 m ² /g; in other embodiments, the surface area is about 600 m ² /g; in other embodiments, the surface area is about 625 m ² /g; in other embodiments, the surface area is about 650 m ² /g; in still other embodiment, the surface area is about 675 m ² /g; and in still other embodiments, the surface area is about 700 m ² /g.

[0043] In some embodiments, the porous spherical particles have an average pore diameter in the range from about 50 A to about 105 A. Moreover, the average pore diameter for the HLB adsorbents described herein may be about 50 A, about 55 A, about 60 A, about 65 A, about 70 A, about 75 A, about 80 A, about 85 A, about 90 A, about 95 A, about 100 A, about 105 A, or about 1 10 A.

[0044] In certain embodiments, the polymeric adsorbent is coated on a substrate. Suitable substrates may include carbon, silica, and metal, among other things. In some embodiments, the may be particulate, while in other embodiments, the substrate may be a fiber.

[0045] When the substrate is a fiber, the polymeric adsorbent may be coated directly onto the fiber. In some embodiments, an adhesive or a binder is used to adhere the polymeric adsorbent to the fiber. Suitable binders are known to those skilled in the art. Some non-limiting examples of adhesives that may be used in accordance with the HLP adsorbents described herein includes, polydimethylsiloxane (PDMS), epoxy resins, polysilazane, acrylic resins, phenol formaldehyde resins, polyvinyl acetate (PVA, white glue), yellow carpenter's glue (aliphatic resin), polyvinylpyrrolidone (PVP or PVA based), polyester resin, resorcinol resin, urea-resin glue (plastic resin) or combinations thereof. A particularly preferred adhesive is polydimethylsiloxane (PDMS). The use of PDMS as a binder to adhere adsorbents to a structure is disclosed in US Patents No. 5,559,445 and 7,815,864, for example, the entirety of the patents is incorporated herein by reference. In other embodiments, a binder, i.e., that does not exhibit cross-linking characteristics, can be used. Such binders may include, but are not limited to polyacrylonitrile (PAN), TEFLON (PTFE and similar fluorinated polymers/plastics), gelatin, cellulose, cellulose derivatives, polyvinylpyrrolidone, starch, sucrose and polyethylene glycol.

[0046] Notably, unlike conventional HLB coated fibers, the HLB fibers provided herein may be used without a conditioning step. As explained above, this simplification of the extraction process is very desirable from a time standpoint, allowing for analysis of many more samples in a shorter length of time.

[0047] Also provided are methods for removing one or more solutes from a solution by contacting the solution with a polymeric adsorbent as described herein whereby the solute is adsorbed onto the polymeric adsorbent. In some embodiments, the solute may then be eluted from the polymeric adsorbent.

[0048] In some embodiments of this method, the solution has a polar solvent. In a preferred embodiment, the polar solvent is an aqueous solvent. In another preferred embodiment, the polar solvent is an organic solvent.

[0049] In accordance with the methods provided herein, the HLB adsorbents are useful for removing at least 6 solutes, interfering analytes, or analytes of interest from a solution. In some embodiments, the HLB adsorbents are useful for removing at least 8 solutes from a solution. In further embodiments, the HLB adsorbents are useful for removing at least 10 solutes from a solution. In still further embodiments, the HLB adsorbents are useful for removing at least 15 solutes from a solution. In a preferred embodiment, the HLB adsorbents are useful for removing at least 20 solutes from a solution. In accordance with the methods provided herein the HLB adsorbents are useful for removing a several different types of solutes, interfering analytes, or analytes of interest from a solution. The solutes, interfering analytes, or analytes may include one or more proteins, one or more pharmaceutically relevant analytes, and one or more phospholipids.

[0050] In a preferred embodiment, the HLB adsorbents remove at least about 75% of the solutes are removed from the solution with a recovery of at least 50%. In another preferred embodiment, the HLB adsorbents remove at least about 80% of the solutes are removed from the solution with a recovery of at least 50%. In another preferred embodiment, the HLB adsorbents remove at least about 85% of the solutes are removed from the solution with a recovery of at least 50%. In another preferred embodiment, the HLB adsorbents remove at least about 90% of the solutes are removed from the solution with a recovery of at least 50%. In yet another preferred embodiment, the HLB adsorbents remove at least about 95% of the solutes are removed from the solution with a recovery of at least 50%.

[0051 ] In still another preferred embodiment, the HLB adsorbents remove at least about 75% of the solutes are removed from the solution with a recovery of at least 70%. In another preferred embodiment, the HLB adsorbents remove at least about 80% of the solutes are removed from the solution with a recovery of at least 70%. In another preferred embodiment, the HLB adsorbents remove at least about 85% of the solutes are removed from the solution with a recovery of at least 70%. In another preferred embodiment, the HLB adsorbents remove at least about 90% of the solutes are removed from the solution with a recovery of at least 70%. In another preferred embodiment, the HLB adsorbents remove at least about 95% of the solutes are removed from the solution with a recovery of at least 70%.

[0052] Also provided is a method for selective removal of interfering solutes from a solution comprising the step of contacting the solution with a polymeric HLB adsorbent as described herein, wherein the solutes are either adsorbed onto the polymer, or washed off the polymer in the wash step of the method, and wherein the solutes are selected from the group consisting of proteins, phospholipids, inorganic salts, combinations thereof, and wherein at least about 90% of the interfering solutes are removed.

[0053] Also provided are cartridges containing the HLB adsorbents described herein. In some embodiments, the cartridges include spherical particles of HLB adsorbent. In other embodiments, the cartridges include spherical silica particles coated with the HLB adsorbent.

[0054] The HLB materials described herein may be referred to as "polymers," "copolymers," "resins," and "adsorbents" interchangeably. Any of these terms may be substituted for another and the use of one term over another does not connote a difference in the composition or the function of the HLB material.

[0055] The HLB adsorbents described herein are useful for adsorbing analytes, either for the function of removal for purification or removal for concentrating. It is understood "removing" or "adsorbing" may be used interchangeably or substituted with other terms such as "extracting" without changing the meaning. It is further understood that the terms "analytes," "analytes of interest," "solutes," and "interfering solutes" may be used interchangeably as the substances that are adsorbed or removed by the HLB copolymers described herein. [0056] The procedures, experiments and examples provided herein are meant to be illustrative of the improved HLB copolymers provided herein and are not meant to limit the invention.

[0057] The following illustrates the synthesis, use, and properties of the HLB copolymers described herein. To illustrate the improved properties of these polymers, the new HLB copolymers described herein were tested against various conventional HLB adsorbents currently available, identified as Conventional HLB A, Conventional HLB B, Conventional HLB C, and so forth, and those labels are used consistently with the same materials throughout.

[0058] As described above, the HLB copolymers described herein are copolymers of divinylbenzene (DVB) and acrylonitrile (AN) made using a suspension or dispersion polymerization process.

[0059] Figure 1 shows a pore plot overlay illustrating the pore structure of one of the HLB copolymer adsorbents described herein, Conventional HLB A and Conventional HBL B materials. The DVB/AN copolymer was prepared using both hydrophobic and hydrophilic phase extenders, to achieve the necessary microporosity and mesoporosity, as shown in Table 1 .

Table 1 Copolymer surface area and porosity of conventional products and several HLB copolymers described herein [0060] Example: Synthesis of an Exemplary HLB copolymer. An organic phase was prepared by combining a mixture of technical grade divinylbenzene (100g), technical grade acrylonitrile (43g), benzoyl peroxide (6.9g), o-xylene (54g) and 4-methyl-2-pentanol (54g). An aqueous phase was prepared by combining: sodium dodecyl sulfate (1 .2g), culminal (1 .2g, high molecular weight methyl cellulose), boric acid (32g) and bulk water (794g). The organic phase was charged to the aqueous phase in a reactor. Agitation was applied to disperse the organic phase into fine droplets, and the reactor was heated to 80 °C for approximately 16 hours to initiate and complete polymerization. The result was a spherical, white powder with a yield of approximately 140g.

[0061 ] The HLB copolymers made by the process described herein were found to have good batch to batch synthetic reproducibility, as shown in Figure 2. As illustrated in Figure 2, the pore strucure was shown to be reproducible throughout the microporous and mesoporous ranges to one batch to the next.

[0062] HLB copolymer coated Silica The process of coating the silica is as follows. First, HPLC grade silica (3.0μιη, 300A), is vacuum dried overnight at ~\ 40 °C. The dried silica is charged to a rotary evaporator flask. Divinylbenzene (4.5g), acrylonitrile (1 .5g), benzoyl peroxide (0.18g) and dichloromethane (90g, bulk solvent) are pre-mixed, then the premixed solution is charged to the flask. The flask is rotated at 50 rpms for 2 hours at ambient temperature to allow for equilibration. The flask is heated stepwise to slowly evaporate the dichloromethane, as follows: (1 ) hold two hours at ambient conditions; (2) raise the temperature from ambient to 60 °C and hold for one hour; (3) hold for one hour at 60-65 °C; then (4) raise the temperature from 65 to 80 °C and hold 16 hours. The silica may also be encapped, for example with trimethylchlorosilane or oleic acid, prior to polymer coating.

[0063] For SPME, copolymer coated silica possesses a typical particle size range of 2 to 4 μιη, and possesses similar pore structure and surface chemistry characteristics to the silica hard template. However, it is noted that the copolymer coated silica can be in the range from about 30 nm to about 5 μιη, depending on the application of interest.

[0064] To investigate the suitability of coating the HLB copolymers described herien on silica, example HLB-65 was chosen. The silicas chosen for this material possessed pore diameters of: 100A, 200A and 300A. To ensure copolymerization of the DVB and ACN monomers on the surface of the silica, wash procedures were chosen to determine copolymer loss. No loss was detected, as shown in Figure 3A. Figure 3B illustrates the uniform pore decrease using the above-mentioned silica coating process. Figure 4 illustrates the coatings using varying copolymer ratios.

[0065] The effect of washing on analyte recovery was investigated using experimental HLB copolymer HLB-65. The testing used the procedures for analsys of twenty analytes described below. The results are summarized in Table 2, Table 3, and Table 4 below.

Table 2 Effect of washing on absolute recoveries obtained using experimental HLB-65.

Table 3 Effect of washing on matrix matched recoveries obtained using experimental HLB- 65. Number of Analytes (Relative Recoveries)

HLB-65 W HLB-65 sox HLB-65 W HLB-65 sox

Recovery

(Neutral) (Neutral) (Acidic) (Acidic)

(%) No dry down No dry down No dry down No dry down

>120% 3 3 3 1

80-120% 1 1 6 7 12

<80% 1 6 5 3

Table 4 Effect of washing on relative recoveries obtained using experimental HLB-65.

[0066] For illustration, Figure 5 is an scanning electron micrograph of a typical silica coated SPME fiber. The adhesive used for the bonding of the 3.0 urn polymer coated silica is the polydimethylsiloxane (PDMS) described herein.

[0067] Further evaluation of the HLB copolymers described herein was done as follows:

[0068] Analvte recovery evaluations: Six Analyte Preliminary Screen. A 1 mg/mL stock was prepared by solvating 50 mg of the following analytes in 50 mL of methanol: acetaminophen, procainamide, p-toluamide, 2,7- dihydroxynaphthalene, propranolol, and doxepin. A 10 μg/mL analytical stock was prepared by transferring 500 μ\- of 1 mg/mL stock into 49.5 mL of 20 mM potassium phosphate buffer. Reagents were prepared using volumetric flasks filled according to standard procedure.

[0069] For experimental resins or conventional comparisons, 3 mL SPE cartridge formats with 60 mg bed weights were utilized. Samples were processed utilizing the protocol described in Table 5.

SPE Protocol

Load: 1 mL sample in 20 mM phosphate buffer

Wash: 1 mL 5% methanol aq.

Elute: 1 mL 90:10 ACN: Methanol

Evaporate: 40 °C, 10 psi nitrogen stream

I Reconstitute: 1 mL 20 mM phosphate buffer

Table 5 3-step SPE protocol used for basic screening.

[0070] Drop rates were maintained at 1 drop per second or less. Analyte stocks were initially dissolved in methanol at 1 img/mL concentrations and were then diluted to 10 μg/ mL in 20 mM potassium phosphate buffer (pH 7) to produce the loading stock. After evaporation all samples were vortexed 30 seconds in 20 mM potassium phosphate buffer (pH 7) to reconstitute for analysis. Three aliquots of 10 μg/mL loading solution were analyzed and the average area count for each compound was used as a reference to calculate recovery of the respective compounds.

[0071 ] HPLC parameters. Analysis was performed using an Agilent 1 190 HPLC coupled with a UV detector. Method parameters can be found in Table 6.

HPLC Instrument Parameters

Column Discovery® C18, 5 cm x 4.6 mm x 5 μιη (Sigma-Aldrich)

Mobile phase A 20 mM potassium phosphate

Mobile phase B MeOH

Injection volume 25 μΙ_

Time (min) A % B %

Gradient

0.0 95 5

2.00 95 5

3.00 65 35

7.00 65 35

8.00 30 70

12.00 30 70

12.10 95 5

20.00 95 5

Wavelength 230 nm

Flow 1 .0 imL/min

Table 6 HPLC-UV parameters for analysis of 6 analyte preliminary screening.

[0072] Protocol I : this protocol utilized a 20 analyte mix consisting of the following: amiloride, amitriptyline, atrazine, benzoylecgonine, buspirone, clarithromycin, haloperidol, hydroquinidine, imidacloprid, imipramine, imiquimod, loratadine, mesoridazine, methapyrilene, mianserin, mirtazapine, nefazodone, nevirapine, nizatidine, and losartan. For acidic matrix extractions, newborn calf serum was spiked with analyte at a concentration of 200 ng/mL. The spiked serum was then diluted 1 :1 with a 0.4% formic acid solution. These solutions were also spiked with an internal standard mixture, allowing for final concentrations of 100 ng/mL analyte and 50 ng/mL internal standards. The SPE procedure is described in Table 7.

[0073] It is noted that all cartridges containing experimental materials were packed with soxhlet washed polyethylene frits in order to remove lipophilic interferences that would otherwise be co-extracted with the analytes. SPE Procedure

Load 1 mL sample onto a dry bed (60 mg bed weight).

Wash cartridge with 1 mL of 5% MeOH (aq.)

Elute with 1 mL of elution solvent.

Evaporate samples to dryness at 40 °C with 5 psi nitrogen stream.

Reconstitute in 1 mL of starting mobile phase.

Table 7 SPE procedure used for Protocol I.

[0074] Protocol O: Newborn calf serum was diluted 1 :1 with 4% phosphoric acid (final analyte concentration: 100 ng/mL; final internal standard concentration: 50 ng/mL). The previously described procedure was followed using 100% MeOH as the elution solvent. Sample eluates were dried down and reconstituted in the starting mobile phase. Samples were also prepared following a "5-step protocol", this simply involved adding a conditioning (1 mL MeOH) and equilibration (1 mL water) step prior to sample loading. The 20 analyte mix was used for this evaluation.

[0075] It is noted that all cartridges containing experimental materials were packed with soxhlet washed polyethylene frits in order to remove lipophilic interferences that would otherwise be co-extracted with the analytes.

[0076] Protocol P: Newborn calf serum was diluted 1 :1 with 4% phosphoric acid (final analyte concentration: 100 ng/mL; final internal standard concentration: 50 ng/mL). The procedure described in Table 5 was followed using 100% MeOH as the elution solvent. Sample eluates were directly analyzed via LC-MS/MS. An eight analyte mix containing azidothymidine, phenacetin, betamethasone, alprazolam, naproxen, propranolol, protriptyline, and amitriptyline was used for this evaluation.

[0077] It is noted that all cartridges containing experimental materials were packed with soxhlet washed polyethylene frits in order to remove lipophilic interferences that would otherwise be co-extracted with the analytes. [0078] LC-MS/MS Analysis: Analysis of Protocol I and Protocol O HLB protocol samples. Samples were analyzed using an Agilent 1290 Infinity UHPLC coupled with an Agilent 6460 triple quadrupole mass spectrometer. Table 8 and Table 9 display the specific UHPLC and MS source parameters. Table 10 and Table 1 1 display the compounds specific MS parameters for the analysis of analytes in positive and negative mode, respectively.

UHPLC Instrument Parameters

Column Ascentis® Express RP-Amide; 10 cm x 2.1 mm ; 2.7 μιη

(Sigma-Aldrich)

Column Temp. ( °C) 40

Mobile Phase A 0.1 % acetic acid, 5 imM ammonium acetate

Mobile Phase B 5 imM ammonium acetate in acetonitrile

Flow rate (imL/min) 0.4

Injection Volume (μί) 2.00

Time A % B %

Gradient

(min)

0.00 95.0 5.0

1 .00 95.0 5.0

7.00 57.0 43.0

8.00 57.0 43.0

10.00 38.5 61 .5

1 1 .00 95.0 5.0

14.00 95.0 5.0

Table 8 UHPLC instrument parameters used for analysis of in-house protocol and conventional 5-step protocol samples.

MS source parameters

Capillary Voltage (V) 3500

Gas Temp. ( °C) 350

Gas Flow (L/min) 8

Nebulizer (psi) 45

Sheath Gas Heater 350

Sheath Gas Flow 12

Cell Accelerator (V) 7

Polarity Positive and Negative (2 separate analyses)

AEMV 200

Table 9 MS source parameters used for analysis of in-house protocol and conventional 5- step protocol samples.

Table 1 1 Compound Parameters for negative mode analysis of in-house protocol and conventional 5-step protocol samples. [0079] Analysis of 3-step protocol samples. Samples were analyzed using an Agilent 1290 Infinity UHPLC coupled with an Agilent 6460 triple quadrupole mass spectrometer. Table 12 and Table 13 display the specific UHPLC and MS source parameters. Table 14 displays the compound specific MS parameters.

UHPLC Source Parameters

Column Titan™ C18; 5 cm x 2.1 mm; 1 .9 μηι (Sigma-Aldrich)

Column Temp. ( °C) 40

Mobile Phase A 0.1 % formic acid

Mobile Phase B 0.1 % formic acid in acetonitrile

Flow rate (imL/min) 0.5

Injection Volume (μί) 2.00

Time (min) % A % B

Gradient

0.30 70.0 30.0

1 .00 70.0 30.0

4.00 70.0 30.0

6.00 50.0 50.0

6.10 50.0 50.0

6.50 70.0 30.0

8.50 70.0 30.0

Table 1 2 UHPLC instrument parameters used for analysis of in-house protocol and conventional 3-step protocol samples.

MS source parameters

Capillary Voltage (V) 3500

Gas Temp. ( °C) 300

Gas Flow (L/min) 10

Nebulizer (psi) 20

Sheath Gas Heater 400

Sheath Gas Flow 12

Cell Accelerator (V) 7

Polarity Positive

AEMV 200

Dwell 30

Table 13 MS source parameters used for analysis of in-house protocol and Protocol P protocol samples.

Precursor Product Fragmentor Collision

Analyte

(m/z) (m/z) (V) Energy (V)

Betamethasone 393.01 373.1 100 1

Prednisolone 361 .01 147 95 25

Alprazolam^ 314.01 286 170 25

Alprazolam 309.01 281 160 25

Oxazepam 288.01 241 .9 1 15 17

Abacavir 287.01 191 1 10 17

Amitriptyline- fe 281 .01 105 1 10 21

Amitriptyline 278.2 58.2 1 13 24

Azidothymidine 268.01 126.9 1 10 9

Protriptyline- fe 267.01 155 123 24

Propranolol-^ 267.01 56.2 120 28

Protriptyline 264.2 155.1 123 24

Propranolol 260.2 1 16 125 16

Ketoprofen 255.29 104.9 1 15 24

Naproxen 231 .31 185 80 8

Phenacetin 180.1 65 125 44

Table 14 Compound Parameters for positive mode analysis of in-house protocol and conventional 3-step protocol samples.

[0080] Phospholipid monitoring. Samples were analyzed using an Agilent 1290 Infinity UHPLC coupled with an Agilent 6460 triple quadrupole mass spectrometer. Table 15 and Table 16 display the specific UHPLC and MS source parameters. Table 17 displays the compound specific MS parameters. UHPLC instrument parameters

Column Ascentis® Express C18; 5 cm x 2.1 mm ; 2.7 μιη

(Sigma-Aldrich)

Column Temp. ( °C) 40

Mobile Phase 5 mM ammonium formate in 90:10 (MeOH/water)

Flow rate (imL/min) 0.5

Injection Volume (μΙ_) 2.00

Table 15 UH PLC instrument parameters used for phospholipid monitoring.

Table 1 6 MS source parameters used for phospholipid monitoring.

Analyte Precursor Product Fragmentor Collision

(m/z) (m/z) (V) Energy (V)

Phospholipid 1 184 104 160 80

Phospholipid 2 496 184 160 80

Phospholipid 3 524 184 160 80

Phospholipid 4 704 184 160 80

Phospholipid 5 758 184 160 80

Phospholipid 6 786 184 160 80

Phospholipid 7 806 184 160 80

Table 1 7 Compound Parameters for phospholipid monitoring.

[0081 ] Preliminary Studies. The focus of preliminary studies was to screen a series of co-polymer resins for recovery of a six analyte mix out of buffer. The analytes of interest were selected due to their variety of hydrophobic character (Table 18). Although a simple screen, the results were useful in identifying potential resins to evaluate further.

Table 18 Analytes of interest utilized for the preliminary screen.

[0082] A total of eight co-polymer resins were prepared for preliminary screening. These resins were designed to study the role of copolymer ratios and pore structure on analyte recovery. Table 19 describes the co-polymer composition of each experimental resin. The o-xylene:MIBC (w:w) ratio was varied in several resins in order to vary the pore structure.

Table 19 Co-polymer composition of preliminarily screened experimental resins. DVB: divinyl benzene; AcryIN: acrylonitrile; VP: N-vinylpyrrolidone; MIBC: methyl isobutyl carbinol.

[0083] Several experimental resins were further analyzed to better understand their chemistries. These resins were evaluated for CHN composition (Table 20) and surface area and porosity (Table 21 ). Conventional HLB A, Conventional HLB B, and Conventional HLB C materials were also evaluated for comparison.

Resin ID N (%) C (%) H (%) Total (%)

Conventional

0.70 35.47 3.72 39.89 HLB A

Conventional

2.35 86.14 8.27 96.76 HLB B

Conventional

2.42 85.61 8.10 96.13 HLB C

HLB-96 2.00 87.56 7.84 97.40

HLB-98 1 .88 89.37 7.89 99.14

HLB-99 < 0.02 90.01 8.02 98.03

HLB-100 1 .05 88.10 8.13 97.28

HLB-01 9.22 82.75 7.35 99.32

HLB-03 8.82 82.75 7.37 98.94

HLB-05 0.97 87.49 7.92 96.38

Comparative

< 0.02 90.84 7.92 98.76 Sample

Table 20 CHN composition of conventional products and several experimental resins.

[0084] In terms of CHN composition it became apparent that Conventional HLB B and Conventional HLB C materials are very similar. However Conventional HLB A material greatly differs, with only 35.47% carbon.

Diameter Surface area Single point total Average pore

Resin ID

(Mm) (m ² /g) pore volume (mL/g) diameter (A)

Conventional 438 1 .04 95

HLB A

Conventional 716 1 .29 72

HLB B

Conventional 201 0.958 190

HLB C

HLB-96 38-75 423 1 .02 97

HLB-99 38-75 448 1 .17 105

HLB-100 38-75 667 1 .21 73

HLB-01 38-75 346 0.74 85

HLB-03 38-75 401 0.67 66

HLB-05 38-75 658 0.82 50

Table 21 Co-polymer surface area and porosity of conventional HLB products and several experimental HLB resins.

[0085] In terms of surface area and porosity, Conventional HLB A and Conventional HLB B also differ. Conventional HLB B has a greater surface area and a smaller pore diameter than the Conventional HLB A material.

[0086] Each experimental resin was evaluated for the ability to recover the six analytes of interest following two protocols. The first protocol was modeled after a conventional HLB method, however the bed was dried following conditioning and equilibration. The results of this experiment can be found in Figure 6. As expected, the Conventional HLB B product resulted in high recoveries for all 6 analytes, with the lowest recovery of 79% for the most lipophilic analyte, doxepin. Screening of the experimental resins revealed great differences. The highest performing resin at this stage was determined to be HLB-01 . Recoveries were greater than 80% for all six analytes.

[0087] The second protocol utilized a 3-step process: loading, washing, and elution. At the time of these experiments, the 3-step protocol was unknown, so extractions were performed at neutral pH. The results of this experiment can be found in Figure 7. Surprisingly, only 3 out of the 6 analytes were recovered using the Conventional HLB A product. These results may reveal a weakness with the product, or may simply be due to these initial extractions being performed at neutral pH. The best performing experimental resin was again found to be HLB- 01 , with greater than 80% recoveries and RSDs typically < 5% for all 6 analytes.

[0088] The preliminary screening revealed that the next level of experimentation should center around resin HLB-01 . This resin was prepared starting with a monomer ratio of 63:37 (DVB/AN) and a phase extender ratio of 150:150 (o- xylene/MIBC). This resin performed better than HLB-03, which only differed by the phase extender ratio, resulting in less macroporosity; and HLB-100, which differed in the hydrophilic monomer (substituting N-vinylpyrrolidone for acrylonitrile).

[0089] Extended Screening. 20 additional analytes were chosen in order to perform an extended screening to further optimize the experimental HLB resin. Table 22 displays all 20 analytes and their log P values. This mix represents a wider range of hydrophobic character. This variety should aid in finding an even more ideal HLB resin.

Table 22 20-analyte mix utilized for extended screening of experimental resins [0090] During preliminary screening, large changes were made in the polymer chemistry in order to identify conditions of interest. In order to further optimize the HLB resin, small changes were made to the chemistry during extended screening. This will allow for a "fine tuning" of the polymer chemistry. Table 23 contains the co-polymer compositions and the porosity characteristics of the additional resins: HLB-22 and HLB-26, compared with HLB-01 .

Table 23 Co-polymer compositions and porosity characteristics of additional experimental resins.

[0091 ] Figure 8 and Table 24 compare the recoveries of 20 analytes extracted out of spiked serum under acidic conditions for the three experimental co-polymeric resins. In this analysis, resin HLB-22 was found to have improved performance over resins HLB-01 and HLB-26. This resin has a starting monomer ratio of 75:25 (divinylbenzene/acrylonitrile).

Table 24 Number of analytes within each absolute recovery category for three experimental resins.

[0092] Each experimental resin was also evaluated for the ability to remove phospholipids from matrix samples. Figure 9 displays an overlay of MRM traces of unextracted calf serum with samples that had been extracted with various experimental resins and Conventional HLB A. All experimental resins were capable of removing >95% of phospholipids present. HLB-22 was found to result in 98.7% removal when acidic extraction conditions were utilized.

[0093] Next, a polymer coated silica was also investigated. A material was prepared that was successful in >99% phospholipid removal. After identifying the optimal polymer conditions, the next stage was to compare the performance of this experimental polymer against two conventional products: Conventional HLB A and Conventional HLB B.

[0094] The first comparison was performed using the 5-step protocol with the 20- analyte mixture detailed in Table 22. Figure 10 compares absolute analyte recoveries across the 20 analytes using experimental polymer HLB-67 and Conventional HLB B. When using the same extraction methodologies, experimental copolymer HLB-67 was able to extract all 20 analytes with recoveries >70%. For Conventional HLB B, 18/20 analytes had high absolute recoveries. This result confirms that the materials described herein have the potential to be used as a direct replacement for existing conventional 5-step methods.

[0095] Next, the potential of experimental resin HLB-67 to outperform Conventional HLB B was evaluated. This involved again extracting the 20- analyte mix from spiked serum under acidic conditions, however during this evaluation a simplified 3-step protocol was used. No conditioning or equilibration steps were included. The results of this evaluation can be seen in Figure 1 1 . Only 14/20 analytes were found to have absolute recoveries >70% when Conventional HLB B was used with a 3-step method. For our experimental resin HLB-67, 17/20 analytes had >70% recovery when a sample was loaded onto a dry bed. This suggests that the experimental resin has an advantage over the conventional product. Since the conditioning and equilibration steps are not required, a user can save time and steps using the 3-step, rather than 5-step, method. [0096] Experimental HLB resin HLB-67 was then tested against Conventional HLB A. For this, an eight-analyte mix was chosen. Table 25 lists the eight analytes and their log P values.

Table 25 Analytes used for the 3-step protocol evaluations.

[0097] The experimental HLB copolymeric resin HLB-67 was found to perform almost identically to Conventional HLB A (Figure 12). These results also demonstrate that the HLB material described herein is suitable for 3-step SPE methods.

[0098] The next comparison that was made with conventional HLB products was matrix removal. This was accomplished by monitoring seven different MRM transitions that represent different classes of phospholipids. Total peak areas were compared between extracted samples and unextracted serum to determine the average percentage of phospholipid removal. Table 26 compares matrix removal between an experimental resin, Conventional HLB A and Conventional HLB B. The experimental material was shown to remove a comparable amount of phospholipids from the sample as the conventional materials. Resin Used Average % Phospholipid Removal

Experimental HLB 98.8

Conventional HLB A 98.9

Conventional HLB B 98.7

Table 26 Comparison of phospholipid removal accomplished using an experimental resin and two conventional resins.

[0099] After establishing that the experimental copolymeric resins described herein are suitable as both a direct replacement and improvement over various conventional products, the final stage of evaluation was focused on determining the reproducibility of material preparation. A total of five small scale lots were prepared: 3 research-scale lots (HLB-22, HLB-67, and HLB-79) and 2 manufacturing lots. Each material was evaluated using the 20-analyte protocol outlined above: a 3-step method using 50:50 (ACN/MeOH) as the elution solvent. Table 27, Table 28 and Table 29 display the absolute, matrix matched, and relative recoveries of each material, respectively.

Table 27 Comparison of absolute recoveries out of spiked serum across different lots of experimental resin.

Table 28 Comparison of matrix matched recoveries out of spiked serum across different lots of experimental resin. Number of Analytes (Relative Recoveries)

R&D Lot 1 R&D Lot 2 R&D Lot 3 Manuf. Manuf.

Recovery (%)

HLB-22 HLB-67 HLB-79 Lot 1 Lot 2

>120% 1 0 0 0 0

80-120% 13 15 15 15 15

<80% 2 1 1 1 1

Table 29 Comparison of relative recoveries out of spiked serum across different lots of experimental resin.

[0100] Additional experiments:

[0101 ] Experimental HLB copolymer HLB-01 was evaluated for possible overdrying effects by measuring absolute of a six-analyte mixture using the following procedure. A 1 img/mL stock was prepared by solvating 50 mg of the following analytes in 50 mL of methanol: acetaminophen, procainamide, p- toluamide, 2,7 dihydroxynaphthalene, propranolol, and doxepin. A 10 μg/mL analytical stock was prepared by transferring 500 μ\- of 1 img/mL stock into 49.5 mL of 20 mM potassium phosphate buffer. Reagents were prepared using volumetric flasks filled according to standard procedure. Three mL SPE cartridge formats with 60 mg bed weights were utilized. Samples were processed utilizing the protocol described in Table 30.

5-step Protocol :

Conditioning : 1 imL of MeOH

Equilibration : 1 imL of 20 mM phosphate buffer

Dry: applied 1 0" Hg vacuum for 10 min

Load : 1 imL sample in 20 mM phosphate buffer

Wash : 1 mL 5% MeOH

Elute: 1 mL MeOH

Evaporate: 40 °C, 1 0 psi nitrogen stream

I Reconstitute: 1 mL 20 mM phosphate buffer

Table 30 5-step SPE protocol with dry step (used for Figure 13 data).

[01 02] Drop rates were maintained at 1 drop per second or less. Analyte stocks were initially dissolved in methanol at 1 img/mL concentrations and were then diluted to 1 0 μg/ mL in 20 mM potassium phosphate buffer (pH 7) to produce the loading stock. After evaporation all samples were vortexed 30 seconds in 20 mM potassium phosphate buffer (pH 7) to reconstitute for analysis. Three aliquots of 10 μg/mL loading solution were analyzed and the average area count for each compound was used as a reference to calculate recovery of the respective compounds.

[01 03] HPLC parameters: Analysis was performed using an Agilent 1 1 90 HPLC coupled with a UV detector. Method parameters can be found in Table 31 .

HPLC-UV instrument parameters

Column Discovery C18, 5 cm x 4.6 mm x 5 μιη

Mobile phase A 20 imM potassium phosphate

Mobile phase B MeOH

Injection volume 25 μΙ_

Time (min) A % B %

Gradient

0.0 95 5

2.00 95 5

3.00 65 35

7.00 65 35

8.00 30 70

12.00 30 70

12.10 95 5

20.00 95 5

Wavelength 230 nm

Flow 1 .0 imL/min

Table 31 HPLC-UV parameters for analysis of 6 analyte screening.

[0104] Figure 13 shows the absolute recovery of six analytes using an acrylonitrile- containing HLB copolymer described herein (HLB-01 ) to extract from a buffer solution. The material was dried following pre-treatment, showing that the overdrying effect is not an issue when working with the new HLB material. These analytes represent a log P range of 0.5-4.3.

[0105] HLB copolymer HLB-01 was then evaluated against a conventional vinylpyrrolidone-containing HLB, Conventional HLB A, by measuring the recovery of the same six analytes but using the 3-step process. The same solution preparation and HPLC analysis was followed as described above. The SPE procedure was slightly altered as shown in Table 32. 3-step Protocol:

Load: 1 mL sample in 20 mM phosphate buffer

Wash: 1 mL 5% MeOH

Elute: 1 mL 90:10 (ACN/MeOH)

Evaporate: 40 °C, 10 psi nitrogen stream

I Reconstitute: 1 mL 20 mM phosphate buffer

Table 32 3-step SPE protocol for Figure 14 data.

[0106] Figure 14 shows a comparison of the absolute recovery of acrylonitrile- containing HLB material (HLB-01 ) and a vinylpyrrolidone-containing conventional HLB material (Conventional HLB A) when a simplified 3-step protocol is utilized (no conditioning or equilibration prior to sample loading) to extract from spiked serum. This set of compounds in Figure 14 represents a log P range of 0.5-4.3.

[0107] Experimental HLB copolymer HLB-65 was coated on silica particles as described above. The HLB-coated silica particles were used as follows. A 20- analyte mixture including the following: amiloride, amitriptyline, atrazine, benzoylecgonine, buspirone, clarithromycin, haloperidol, hydroquinidine, imidacloprid, imipramine, imiquimod, loratadine, mesoridazine, methapyrilene, mianserin, mirtazapine, nefazodone, nevirapine, nizatidine, and losartan was utilized. A 20 mM ammonium acetate buffer, pH 7.4, was spiked with analyte at a concentration of 200 ng/mL. The spiked sample was then diluted 1 :1 with 20 mM ammonium acetate. These solutions were also spiked with an internal standard mixture, allowing for final concentrations of 100 ng/mL analyte and 50 ng/mL internal standards. The SPE procedure is described in Table 33. SPE Procedure

Load 1 mL sample (diluted 1 :1 with 20 mM ammonium acetate) onto a dry bed (60 mg bed weight).

Wash cartridge with 1 mL of 5% MeOH (aq.)

Elute with 1 mL of 50:50 (ACN/MeOH).

Evaporate samples to dryness at 40 °C with 10 psi nitrogen stream.

Reconstitute in 1 mL of starting mobile phase.

Table 33 SPE procedure used for evaluation of polymer coated silica (used for Figure 15 data).

[0108] LC-MS/MS Analysis: Samples were analyzed using an Agilent 1290 Infinity UHPLC coupled with an Agilent 6460 triple quadrupole mass spectrometer. Table 34 and Table 35 show the specific UHPLC and MS source parameters. Table 36 and Table 37 display the compounds specific MS parameters for the analysis of analytes in positive and negative mode, respectively.

UHPLC Instrument parameters

Column Ascentis Express RP-Amide; 10 cm x 2.1 mm ; 2.7 μιη

Column Temp. ( °C) 40

Mobile Phase A 0.1 % acetic acid, 5 mM ammonium acetate

Mobile Phase B 5 mM ammonium acetate in acetonitrile

Flow rate (imL/min) 0.4

Injection Volume (μί) 2.00

Time (min) A % B %

Gradient

0.00 95.0 5.0

1 .00 95.0 5.0

7.00 57.0 43.0

8.00 57.0 43.0

10.00 38.5 61 .5

1 1 .00 95.0 5.0

14.00 95.0 5.0

Table 34 UH PLC instrument parameters used for analysis of the 20-analyte mix (used for Figures 15 and 1 6 data). MS source parameters

Capillary Voltage (V) 3500

Gas Temp. ( °C) 350

Gas Flow (L/min) 8

Nebulizer (psi) 45

Sheath Gas Heater 350

Sheath Gas Flow 12

Cell Accelerator (V) 7

Polarity Positive and Negative (2 separate analyses)

AEMV 200

Table 35 MS source parameters used for analysis of the 20-analyte mix (used for Figure 15 and 1 6 data).

Table 37 Compound Parameters for negative mode analysis of the 20-analyte mix (used for Figure 15 and 1 6 data). [0109] The absolute recoveries of the analytes using silica coated with experimental HLB-65 is shown in Figure 15. Silica coated with this HLB copolymer extracted a variety of analytes from spiked serum samples using a traditional 5-step SPE procedure.

[01 10] Evaluation of Conventional HLB B and experimental HLB material HLB-79 (data shown in Figure 16). Table 38 displays the SPE procedure. The previously described LC-MS/MS method was used.

SPE Procedure

Load 1 imL sample (diluted 1 :1 with 2% phosphoric acid for

Conventional HLB B; 0.4% formic acid for experimental HLB-79) onto a dry bed (60 mg bed weight).

Wash cartridge with 1 mL of 5% MeOH (aq.)

Elute with 1 mL of MeOH (for Conventional HLB B); 50:50

MeOH:ACN (for HLB-79)

Evaporate samples to dryness at 40 °C with 10 psi nitrogen stream.

Reconstitute in 1 mL of starting mobile phase.

Table 38 SPE procedure used for evaluation of polymer coated silica (used for Figure 16 data).

[01 1 1 ] As shown in Figure 16, experimental HLB-79 showed a more consistent recovery for the 20-analyte mixture than vinylpyrrolidone-containing Conventional HLB B when a simplified 3-step protocol was utilized to extract analytes from spiked serum. This 20-analyte set of compounds represents a log P range of -0.3-6.1. Due to the increased polarity of acrylonitrile and the ability to incorporate within divinyl benzene, improved recoveries were obtained, particularly for analytes with lower log P values. 1 12] The HLB copolymers provided herein have been shown to be useful in a variety of SPE applications. Using the methods described herein, the improved HLB adsorbents are capable of consistently extracting at least 18/20 analytes with recoveries >70% when using either 5-step or 3-step methods. Advantageously, the improved HLB adsorbents provided herein are well-suited to a 3-step protocol, where conditioning and equilibration steps were not required. The improved HLB material was also found to remove >98% of phospholipids present within a serum sample, on average.