Subscriber access provided by United Arab Emirates University | Libraries Deanship
Article
Functionalized Cycloolefin Polymer Capillaries for Open Tubular Ion Chromatography Weixiong Huang, Sasikarn Seetasang, Mohammadmehdi Azizi, and Purnendu K. Dasgupta Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03669 • Publication Date (Web): 10 Nov 2016 Downloaded from http://pubs.acs.org on November 12, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Functionalized Cycloolefin Polymer Capillaries for Open Tubular Ion Chromatography Weixiong Huang, Sasikarn Seetasang, Mohammadmehdi Azizi and Purnendu K. Dasgupta* Department of Chemistry and Biochemistry, The University of Texas at Arlington, Arlington, Texas 76019-0065, United States
* E-mail:
[email protected]. Fax: 817-272-3808 1 Environment ACS Paragon Plus
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ABSTRACT We describe novel cycloolefin polymer (COP) based open tubular capillary ion exchange columns. COP capillaries (19-28 µm ) were successfully sulfonated at room temperature by a cocktail of ClSO3H (85-95% w/w) and HOAc or H2SO4. The cation exchange capacity is controlled by the sulfonation time and the sulfonation solution composition and can be as high as 300 peq/mm2. Following sulfonation, the capillaries were coated with 65 nm diameter anion exchanger (AEX) latex nanoparticles that attach electrostatically. The typical anion exchange capacities were ~20 peq/mm2. The chromatographic behavior of the AEX latex coated COP capillaries depend greatly on the degree of sulfonation. When the base is heavily sulfonated, neutrals elute after the anions. The position of the water dip varies with the degree of sulfonation, elution order is normal (water dip appear before anions) only with lightly sulfonated columns. On silica (-SiOH) or polymethylmethacrylate (-COOH) surfaces, AEX Latex attachment is not stable over long periods in significant concentrations of strong base (e.g. ≥10 mM NaOH). Latex attachment on sulfonated COP surfaces are much stronger, several types show sufficient binding to be used over long periods at practical eluent concentrations, paving the way for suppressed hydroxide eluent IC discussed in a companion paper. Another interesting feature of COP capillaries lies in their flexibility. If softened at modestly elevated temperatures (e.g., boiling water), they can be coiled down to 5 mM hydroxide over long periods. Carrying out multi-step chemistry in small bore capillaries is difficult and overall yield of usable columns are poor. In contrast, coating with a dilute latex nanoparticle suspension simply involves its passage through an oppositely charged capillary. Also, columns based on latex particles of already well-characterized selectivities will facilitate the transition from current packed columns to OTIC. The monolayer of nm-size latex particles in an OT column permits efficient stationary phase mass transfer, retaining the advantages of the pioneering approach11 to making efficient electrostatically bound latex columns. Compared to packed columns, the probability of a dislodged latex particle to be recaptured is much lower in an OT column. The attachment strength, which is expected to increase with increasing acidity of the capillary wall functional group, should be as high as possible. We wished therefore for an inert thermoplastic material capable of extrusion into capillaries and functionalization with sulfonic acid groups. Cyclic olefin polymers (COPs) and copolymers (COCs) (see Ref. (12) for similarities and differences between the two) fit the first criterion; both have been used in fabricating microchips. Like other polyolefins, COPs/COCs exhibit high chemical resistance and low water 4 Environment ACS Paragon Plus
Page 4 of 26
Page 5 of 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
absorption. Polyolefins are very unreactive but they can be sulfonated with chlorosulfonic acid (ClSO3H).13 Although this chemistry may not be widely known, it was exploited to make the first hollow fiber suppressors for IC.14 Between COP and COCs, COPs display greater ductility. We chose a particular COP that has among the lowest glass transition temperatures and also has the added benefits of very low fluorescence, usable transparency in capillary scale thicknesses down to 290 nm, with an additional narrow transparent window centered at ~250 nm. Herein we describe optimized sulfonation, resulting capacities, behavior of such capillaries with or without latex coating, usability with hydroxide eluents, and also demonstrate the effect of coiling the capillary. EXPERIMENTAL SECTION Materials and Methods. Chlorosulfonic acid (www.acros.com) was diluted with glacial acetic acid (HOAc); 85:15 to 95:5 (w/w) ratios were studied. In some experiments, a portion of the glacial acetic acid content was replaced by conc. H2SO4 but no significant difference in sulfonation kinetics was observed. Various AEX latex suspensions were obtained as a gift from Thermo Fisher Scientific (Dionex), Sunnyvale, CA. CAUTION: Chlorosulfonic acid is an extremely aggressive reagent that reacts explosively with water generating H2SO4 and HCl fumes. It can cause severe burns. Any initial transfer of ClSO3H must be carried out with extreme caution in a fume hood and only in small amounts at a time. Capillary. Zeonex 330R was purchased from the manufacturer (www.zeonex.com). The pellets were melted in vacuum and degassed over 48 hours and allowed to cool to form ingots ~46 x 10 x 8 cm in size. Cylinders (7.35 cm dia. x 20 cm long) were machined from the ingots and a 3.97 mm (5/32 in.) concentric hole was drilled to complete a preform that has the same aspect ratio as the o.d./i.d. of the intended extruded capillaries (370 x 20 m). Extrusion was conducted in two batches and resulted in 360375 m o.d. capillaries with average inner diameters of 19 and 28 m. As one may 5 Environment ACS Paragon Plus
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
surmise, the bore uniformity ( 2 m over 1 m) and occasionally, concentricity, of these capillaries are not as good as commercial fused silica or PEEK capillaries. The procedure for coiling capillaries is described in the Supporting Information (SI). Sulfonation procedure. Dry N2 was passed through the capillary (0.2-1 m) at 60-100 psi for 5-10 min to dry it thoroughly. A glass vial containing 1 mL of the sulfonation reagent, covered with two layers of PTFE tape with Parafilm atop, was put inside a pneumatically pressurizable Plexiglas cylinder. One end of the capillary is affixed to the Plexiglas reservoir top such that when the top is put back on the reservoir, the capillary tip punctures the PTFE/Parafilm seal and reaches the vial bottom (Figure 1). Depending on the bore and the length of the capillary, 10-40 psi is applied to pump the reagent for 2-15 h; the waste is collected in HOAc for disposal. After the desired sulfonation period, the COP capillary was pneumatically flushed with HOAc (30 min, 80 psi) and then rinsed with deionized water by a capillary HPLC pump at relatively high flow rate (1.5/610 µL/min for 19/28 µm i.d. capillaries for > 100 h to thoroughly clean the column. Latex Coating. Latex suspensions as received (nominally 10% w/v) was diluted 10x with 1.0 mM LiOH or water. Rather than unidirectional pumping, a vial was used at both ends and automated alternate pumping and venting (every 15 min) was used to move the suspension in the capillary back and forth for at least 2 h to allow the latex particles sufficient time to bind to the sulfonated wall. The column was then thoroughly washed with water before further testing. Unless otherwise stated, AS18 latex (courtesy ThermoFisher Scientific) was used with 19 m i.d. columns. Procedures for cation and anion exchange capacity determination respectively before and after latex coating is described in the SI. Chromatographic Characterization of Columns. The chromatographic setup for testing columns was largely the same as that used previously;5 minor differences in components and configuration is described in the SI (Figure S1 and accompanying
6 Environment ACS Paragon Plus
Page 6 of 26
Page 7 of 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
text). LabVIEWTM programs for controlling the system to measure the column capacity and conducting chromatography are given in Table S1 and S2, respectively.
RESULTS AND DISCUSSION Sulfonation. To our knowledge, there is no report on systematic characterization of sulfonation of any polyolefin, not just COPs, by chlorosulfonic acid. The present COP capillaries could not be heated safely beyond~100 C; on the other hand at lower temperatures (e.g., 60 C), concentrated H2SO4 was ineffective in sulfonating COP. In contrast, pure chlorosulfonic acid attacked it at room temperature at an uncontrollable rate. Inert diluents like CHCl3 or CH2Cl2 have been recommended for sulfonation with ClSO3H13 but they dissolve the COP. Both HOAc and H2SO4 were found to be usable as diluents but below ~80 wt% ClSO3H, no sulfonation occurred at room temperature in 24 h. 85-95 wt % ClSO3H: HOAc was studied in detail; except as stated, 90±2: 10±2 (%w/w) ClSO3H: HOAc was used. Sulfonation renders COP pale yellow that slowly darkens to brown; allowing one to qualitatively determine occurrence of sulfonation. Chemistry and Degree of Sulfonation. COP’s are produced by ring-opening metathesis polymerization of a cyclic monomer with at least one unsaturated linkage (e.g., norbornene) followed by hydrogenation.15 In principle, a linear chain of saturated aliphatic hydrocarbon rings is produced; however, significant optical absorption below 300 nm suggests presence of residual olefinic linkages; these are likely attacked first with ring opening and forming -chloro -alkanesulfonic acids.16 Chlorosulfonic acid attack on the saturated ring replaces an H atom (reactivity: tertiary>secondary> primary H) with a SO3H group, liberating HCl.Error! Bookmark not defined. At a fixed reaction temperature and ClSO3H concentration, degree of sulfonation was solely controlled by the sulfonation time. Figure 2 shows the degree of sulfonation observed with time using a sulfonation solution of 95.1% ClSO3H: 4.9% HOAc. The 7 Environment ACS Paragon Plus
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
capillary tip was dipped into the sulfonating reagent solution pneumatically pumped through the capillary. Over time, the capillary tips reacting with the reagent rendered it from yellow to brown. The sequence of sulfonation of the capillaries is listed in the figure as A, B, C, etc. Between the first set of 2 h (A: 3.2 ± 2.7 peq/mm2) and 4 h sulfonation (23.9 ± 1.1 peq/mm2), the capacity increased dramatically (B: 66.6 ± 1.1 and 143.8 ± 4.1 peq/mm2 for 2 and 4 h sulfonation respectively) but only showed modest further increases in extending sulfonation times to 6 to 8 h (C, D); with10 h sulfonation (E) the capacity was less than that with 8 h sulfonation, all of the differences being statistically significant. (Note that the surface area being used for computing capacity per unit area is the nominal inner surface area of a smooth cylinder.) As a benchmark, theoretically computed surface area occupied by a sulfonic acid group is 0.18-0.21 nm2 depending on its orientation,17 from which one can compute a maximum monolayer coverage of 89 peq/mm2 (the maximum capacity of a thoroughly sulfonated macroporous 300 m2/g surface area cation exchange resin is ~1 meq/g, translating to ~3.3 peq/mm2). It would thus be clear that for all but the 2 h sulfonation in the first sequence the reaction continues well beyond a monolayer coverage. It is also important to note that the stated capacities were obtained only after 100+ h of washing at high flow rates (vide supra). Our present belief is that while the capacity dramatically increases between 2 and 4 h sulfonation due to attack on newly exposed underlayers, further sulfonation is accompanied by ring opening and result in dangling polymer chains that can be washed off over a period of time. Even with pure water as the influent, heavily sulfonated columns always exhibit greater detector noise (vide infra) that diminish only with continued washing with some decrease in the measured capacity. A second unanticipated and somewhat inexplicable finding is that the sulfonation reaction is autocatalytic. Some entity is formed in the sulfonating reagent in contact with the COP. The color of the reagent slowly turns from clear to yellow to dark brown (Figure S2), accompanied by an increase in viscosity (flow rate decreases at constant 8 Environment ACS Paragon Plus
Page 8 of 26
Page 9 of 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
pressure). Strangely (and as the data in the sequential experiment demonstrates, compare F vs. A and G vs. B, in Fig 2, run for the same periods with an increasingly darker reagent), the partially spent reagent is much more effective for sulfonation. Sulfonation for 4 h with a reagent that has been in contact with some COP over 24 h (light yellow in color) provided a capacity of 99 peq/mm2 compared to 46 peq/mm2 obtained under identical conditions with a fresh sulfonating reagent. To reduce overall time and achieve reproducible sulfonation, our current practice is to immerse small sections (~2 cm each) of two COP capillaries in 1 g of the fresh sulfonation reagent for 48 h before use. Beyond autocatalysis, the sulfonation behavior is exclusively dependent on the ClSO3H content, substituting H2SO4 for HOAc made no significant difference. Highly Sulfonated Columns. Ion Exclusion Behavior, Long term Stability and Baseline Noise. By sulfonating COP, we have routinely attained columns with CEX capacity >200 peq/mm2, even exceeding 300 peq/mm2 (see Figure S3), compared to ~1 peq/mm2 –COOH CEX capacities on PMMA columns after a strong base hydrolytic step.2 From a standpoint of analytical ion chromatography, high capacity sulfonated columns are not especially useful, however. With the possible exception of some dedicated amino acid analyzers,18 most cation exchange separations in IC utilize – COOH rather than –SO3H based stationary phases. However, highly sulfonated COP columns, much like highly sulfonated gel-type resins, show ion exclusion behavior, not previously reported for an open capillary format. Figure 3a shows a highly repeatable separation of water and ethanol on a highly sulfonated column. When such highly sulfonated columns are coated with AEX latex the underlying sulfonated layer remain accessible; ion exchange and neutral retention behavior (through partition into the water phase associated with the sulfonate sites) proceed simultaneously leading to the sample water being retained substantially more than some anionic analytes (Figure 3b) or co-eluting with some strongly retained anion (Figure 3c). Such mixed but 9 Environment ACS Paragon Plus
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
independent retention modes may offer intriguing possibilities for simultaneous separations of anions from neutrals with appropriate detectors, as well as better detection of early eluters that are masked by the water dip by deliberately moving the latter. Very high cation exchange capacity columns (>200 peq/mm2) do show some decrease in capacity upon prolonged washing with water; the effluent baseline conductance decreases with increasing flow rate indicating conductive material washing off, possibly these are dangling open chains. Both the baseline and the baseline noise reach a stable low value only after extensive washing (Figure 4). Latex coating of a poorly washed or unwashed capillary only increases the baseline noise, apparently the latex can attach to dangling chains and both can wash off. Baseline noise of latex coated AEX columns tend to be significantly higher when the foundation CEX capacities are high rather than low (1-3 peq/mm2) columns. Table S3 shows the noise levels observed with an AS18 latex coated 250 peq/mm2 column with different eluents, different flow rates and with detector placed at 0.2 and 1.0 m from the eluent entrance. The data indicate that the noise is less dependent on the absolute baseline value than the flow rate (increased flow rate presumably causing increasing dislodgment of particles/dangling chains) and is thus cumulative with length. Most chromatography was therefore conducted with lightly sulfonated capillaries that were nevertheless thoroughly washed before coating. Anion Exchange Capacity and Underlying Cation Exchange Capacity. In going from -COOH based PMMA to –SO3H functionalized COP, our goal was to make a stronger, base-stable, attachment of a positively charged latex to the fully ionized strong acid site. The size of the typical AEX latex particle being in the range of 60-200 nm, multilayer coverage of similarly charged latex beads is unlikely;19 at least, this has never been reported. Coating ~1 peq/mm2 CEX capacity PMMA columns2 with AS18 latex results in AEX capacities of ~10 peq/mm2. With the same sulfonation conditions (90:10 wt% 10 Environment ACS Paragon Plus
Page 10 of 26
Page 11 of 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
ClSO3H: HOAC), the CEX capacity increased by 650% from 0.4 to 3 peq/mm2 by increasing sulfonation time from 2 to 6 h. But upon coating these columns with the same AS18 latex, the AEX capacity increased only by 5 h sulfonation time with 90:10 wt% ClSO3H: HOAc), the AEX capacities were statistically indistinguishable (28.20.8 peq/mm2). For a CEX capacity around ~1 peq/mm2, an AEX capacity of ~20 peq/mm2 is reproducibly realized with AS18 latex coating, regardless of column i.d. (Figure S4). Hydroxide Eluent Tolerance and Comparison of Eluent Strength vs. Packed Columns. For the same latex coating, the AEX capacity/unit area for the sulfonated COP columns is thus ~2-3x that of the corresponding PMMA columns. None of the latexes studied perceptibly wash off in 5 mM hydroxide (Table S4), while none are stable under the same conditions on PMMA. The strength of latex attachment also varies with the type of the latex (Figure S4), some are stable in 50 mM hydroxide, others are not (Table S4). Note that the capacity of latex-based commercial packed AEX columns depend not only on the diameter and capacity of the individual latex particles but also on the substrate surface area. While a commercial AS18 column has 7x the capacity of an AS5A column of the same volume, the capacity of an AS5A latex coated sulfonated COP column (CEX capacity 3 peq/mm2) is essentially the same as that of an AS18 column. Preliminary results indicate the stability of the AEX capacity may also be dependent on the underlying CEX capacity at the very low end (0- 3 peq/mm2), increasing with the CEX capacity. This is logical as any dislodged AEX latex particle will have a greater probability of being recaptured with increasing CEX sites. However a further increases in CEX capacities (that generally involves more than monolayer coverage) do not result in any better stability. Indeed, past a certain extent of sulfonation, stability is actually compromised as the underlayer itself becomes unstable. Rather than the loss of the latex particles altogether, an alternative and plausible explanation for the loss of capacity of the columns as suggested by an anonymous 11 Environment ACS Paragon Plus
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
reviewer involves the “poisoning” of the anion exchange sites by washed off sulfonated material. Several columns with CEX capacity in the 1-3 peq/mm2 range showed excellent base stability, AS5A coated columns showed no discernible loss of capacity (30.51.8 peq/mm2) over nearly 100 h of 50 mM KOH flowing through it, with intermittent capacity measurements. To put this in perspective of current commercial columns, a consideration of the applicable phase ratio is appropriate. In chromatography the phase ratio is generally defined as the volume of the mobile phase divided by the volume of the stationary phase; in an OT column the former is the volume of the bore.20 In the case of an ion exchange column, a more meaningful definition of the phase ratio iex with any specific eluent in the column would be the ratio of the number of ionic equivalents present in the mobile phase per unit length of the column to the ion exchange capacity of the stationary phase on the wall represented in the same length. Obviously, the higher this number, the less is the retention. Let us compare the situations for an AS5A column in the packed vs.19 or 28 µm bore OT formats. The data in Table S5 will indicate that the same eluent in an OT column represents a 5-50x greater iex depending on the latex; i.e., a proportionally lower eluent strength will be needed for the OT columns to obtain the same retention factors. Chromatographic Performance. Due to the difficulty of fabricating small enough suppressors, with rare exceptions in larger bore columns,21,22 OTIC has largely been conducted in the nonsuppressed mode with the obligatory use of dilute, relatively low conductivity eluents. In a companion paper we describe the fabrication of an appropriate suppressor and suppressed conductometric OTIC,23 but nonsuppressed chromatography can still be attractive. Because the present columns have much greater capacity than corresponding PMMA columns,5 higher eluent concentrations and/or shorter column lengths (requiring very low pressures) can be used. Figure 6 illustrates operation with 22-32 cm 19 m columns (active length 20-30 cm) at applied pressures 12 Environment ACS Paragon Plus
Page 12 of 26
Page 13 of 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
of 6.6 to 19 psi over these lengths, attaining flow velocities of 0.22-0.47 cm/s (flow rates of 39-80 nL/min), substantially above the Van Deemter optimum (0.12 cm/s, see below). The standard 5 anion (F-, Cl-, NO2-, Br-, and NO3-) separation was possible within 4 min using 1.0 mM Na-Benzoate (NaBz) as the eluent. The same ions could be baseline separated in 1.6 min (the actual separation window being
