Frederick F. Cantwell

Solvent Microextraction into a Single Drop

An analytical technique is described which combines solvent extraction with gas chromatographic (GC) analysis in a simple and inexpensive apparatus involving very little solvent consumption. A small drop (8 μL) of a water-immiscible organic solvent, containing an internal standard, is located at the end of a Teflon rod which is immersed in a stirred aqueous sample solution. After the solution has been stirred for a prescribed period of time, the probe is withdrawn from the aqueous solution, and the organic phase is sampled with a microsyringe and injected into the GC for quantification. The observed rate of solvent extraction is in good agreement with a convective-diffusive kinetic model. Analytically, the relative standard deviation of the method is 1.7% for a 5.00-min extraction of the analyte 4-methylacetophenone into n-octane.

Electrical double-layer models of ion-modified (ion-pair) reversed-phase liquid chromatography

Stoichiometric models of ion-modified reversed-phase liquid chromatography are based on chemical equilibria between ionic modifiers and analyte. These are briefly discussed. Non-stoichiometric models portray the ionic solute as being under the summed influence of all of the ions in the system. Chromatographic theories have been developed that are based on the Poisson-Boltzmann equation, which quantitates the summed influence of the ions in the system on the solute. These ideas and quantitative predictions are described and are critically discussed.

Influence of solvent uptake and swelling by poly(styrene–divinylbenzene) column packings on sample sorption rate and band broadening in reversed-phase liquid chromatography

Abstract Porous poly(styrene–divinylbenzene) (PS–DVB) HPLC packings give broad, tailed peaks for some types of solutes. This phenomenon is due to sorption of these solutes into the polymer matrix where they experience very slow, hindered diffusion. It is known that the presence of small amounts of tetrahydrofuran (THF) in the methanol (MeOH)–H 2 O mobile phase yields narrower and less tailed peaks for such solutes. In this study, Hamilton PRP-∞, a nonporous PS–DVB polymer, serves as a model for the matrix of porous polymers. The following measurements were made: sorption isotherms for MeOH and THF from aqueous solution; swelling of PRP-∞ as a function of activity of MeOH and THF in aqueous solution; and both sorption isotherms and sorption rate curves for the solute naphthalene from various solvent mixtures. The addition of 2% THF to 70:30 MeOH–H 2 O produced an additional 0.4% swelling, an 11% decrease in the sorption capacity and a 90% increase in the diffusion coefficient of naphthalene in PRP-∞. The increase in the diffusion coefficient, which is responsible for the improved elution peak shape of naphthalene, is shown to be caused by the decrease in sorption capacity, rather than by the additional swelling. The sorbed THF serves to fill and “block” the smallest, most highly hindered micropores in the polymer matrix.

Retention model for ion-pair chromatography based on double-layer ionic adsorption and exchange

Previous work has demonstrated the validity of an electrical double-layer model for sorption of sample ions onto low capacity ion exchangers. In the present work it is shown how this model can be used in describing the chromatographic retention of sample ions in so-called reversed-phase ion-pair chromatography. The pairing ion added to the mobile phase is sorbed onto the reversed-phase sorbent, creating an electrical double layer and a surface electrical potential psi(o). Sample counterions undergo both ion exchange for electrolyte ions in the diffuse part of the double layer and surface adsorption. The latter depends on the magnitude of psi(o), which can be calculated from the Stern-Gouy-Chapman theory. The model predicts virtually all of the phenomena that have been described in the literature on ion-pair chromatography.

Intra-particle sorption rate and liquid chromatographic bandbroadening in porous polymer packings III. Diffusion in the polymer matrix as the cause of slow sorption

In order to discover the physical cause of the slow intra-particle sorption rate of naphthalene in 10-μm spheres of the macroporous poly(styrene-divinylbenzene) (PS-DVB) polymeric HPLC sorbent Hamilton PRP-1, which has been shown to cause excessive bandbroadening of eluted peaks, the sorption-rate curve for naphthalene from methanol-water (85:15) was measured on PRP-1 using the shallow-bed technique. Sorption on PRP-1 follows a two-term theoretical rate equation for sorption on a biporous particle. From the (fast) first term it is found that 91 % of the naphthalene is sorbed on the walls of the large pores and that the diffusion coefficient in these large pores is 3×10−6 cm2/s. This is close to the free-solution diffusion coefficient, which demonstrates that large-pore diffusion is nearly unhindered. From the (slow) second term in the rate equation it is found that 9% of the naphthalene is sorbed into the polymer matrix of PRP-1, in which the effective diffusion coefficient is no larger than 10−12 cm2/s. It is clear from these results that the cause of the slow intra-particle rate, and therefore of excessive Chromatographic bandbroadening, is slow diffusion into the polymer matrix of PRP-1. To provide additional information on the PS-DVB polymer matrix, the sorption rate of naphthalene was also measured on Hamilton PRP-∞ which is a 19-μm diameter, spherical, nominally nonporous PS-DVB Chromatographic packing. The sorption is slow and follows the theoretical rate equation for hindered diffusion into a homogeneous sphere. The effective diffusion coefficient is (4±1)×10−9 cm2/s. Diffusion through the polymer matrices in PRP-1 and PRP-∞ could be either hindered diffusion through micropores in a rigid matrix or diffusion through a flexible polymer ‘gel’.

Intra-particle sorption rate and liquid chromatographic bandbroadening in porous polymer packings II. Slow sorption rate on a microparticle packing

Abstract The rate of sorption of naphthalene from solution in methanol-water (85:15) onto 10-μm diameter particles of the macroporous poly(styrene-divinylbenzene) sorbent Hamilton PRP-1 is measured by the shallow-bed technique. The sorbent bed is 3 mm in diameter by about 0.3 mm high and contains only 1 mg of PRP-1. This bed is located in the slider of a very low hold-up volume valve. Linear velocity of naphthalene solution through the bed is high enough to exceed 400 bed volumes per second and the valve is capable of accurately providing flow times as short as 0.04 s. The curve of moles sorbed versus time is fit by an empirical tri-exponential rate equation. The parameters of this rate equation are used in a recently developed mathematical model to predict the plate heights and peak shapes of naphthalene peaks that would be expected to elute from a 15-cm long HPLC column of PRP-1 at various linear velocities, if intra-particle processes were the only source of bandbroadening. The predicted peaks are close representations of the peaks that are experimentally observed to elute from such a PRP-1 column. This demonstrates that slow intra-particle processes are the major cause of bandbroadening of naphthalene on PRP-1.

Intra-particle sorption rate and liquid chromatographic bandbroadening in porous polymer packings I. Methodology and validation of the model

An approach is proposed by which an experimentally measured sorption-rate curve for a solute on a Chromatographic sorbent may be used to accurately predict the plate height and peak shape contributions of intra-particle sorption rate to the Chromatographic peak that will be obtained when the solute elutes from a liquid Chromatographic column of the sorbent. The sorption-rate curve is measured on a “shallow bed” of sorbent (e.g. ≈ 1 mm) under “infinite solution volume” conditions and is fit by an empirical tri-exponential equation. Intra-particle rate processes include the bandbroadening phenomena traditionally identified by the HS and HSM plate-height terms. The elution Chromatographic column is imagined to be composed of three “hypothetical columns” in series, and the hypothetical peak eluting from each is calculated from well-known equations. Numerical convolution of these hypothetical peaks in series yields the overall predicted elution peak contribution from intra-particle rate processes. The accuracy of this model is demonstrated by comparing predicted peaks with eluted peaks measured on a long column of very large diameter (0.36 mm) porous polymer sorbent. On this column, Chromatographic bandbroadening is due nearly exclusively to slow intra-particle processes. The proposed approach is not limited, either theoretically or in practice, to large particles or to porous polymers.

Role of the free metal ion species in soluble nickel removal by activated sludge.

review and therefore does not necessarily reflect the views of the Agency, and no official endorsement should be inferred, Mention of commercial products and company names does not constitute endorsement by the U.S. EPA. A Quality Control Evaluation Report associated with this work is available from the U.S. Environmental Protection Agency, Environmental Sciences Research Laboratory, Technical Planning and Review Office (TPRO), Research Triangle Park, North Carolina 27711.

Slow change in the electrical potential at glass and silica surfaces due to Na+ sorption in the hydrated layer

Abstract Controlled Pore Glass (CPG) is a high-SiO2 content, Vycor-type glass which resembles fused silica. In CPG-Oxine the ligand oxine is covalently bound to the CPG surface. The presence of SiO− groups on the CPG generates a negative electrical potential at the surface (ϕO) and also in the solution adjacent to the surface ϕX, where the immobilized oxine is to be found. The potential ϕX influences the extent of complexation of Ca2+ (from solution) by oxine so that the bound oxine serves as a probe of electrical potential near the surface. When the solution pH is raised there is a relatively rapid ionization of SiOH groups to SiO−. Then, slowly (i.e. tens of minutes), Na+ from solution diffuses into the hydrated gel layer on CPG. This reduces the negative charge, making ϕX less negative and, consequently, reduces the extent of complexation of Ca2+ by oxine. A linear relationship is predicted and experimentally observed between the logarithm of the sorbed Ca2+ concentration (mmol/g) and the potential ϕX (V). The potential ϕX is expected to correlate strongly with the zeta potential which controls the rate of electroosmotic flow at silicious surfaces.

Chain Unfolding in an ODS-Bonded Phase Caused by the Sorbed Tetra-n-butylammonium Ion

The simultaneous sorption of the tetra-n-butylammonium ion (TBA(+)) and butanol on the bonded phase sorbent Partisil-10 ODS-3 from an aqueous mobile phase, at the two different ionic strengths 0.50 and 0.050 mol/L, is studied by the column equilibration technique. When the TBA(+) concentration in the mobile phase is kept constant while the butanol concentration is varied, the plots of moles of TBA(+) sorbed versus moles of butanol sorbed decrease linearly. This indicates that butanol simply competes with TBA(+) for sorption space. In contrast, when the butanol concentration in the mobile phase is kept constant while the TBA(+) concentration is varied, the plots of moles of butanol sorbed versus moles of TBA(+) sorbed decrease nonlinearly. This indicates that, in addition to competing with butanol for space, sorbed TBA(+) also causes an unfolding of the originally collapsed C(18) chains. At 0.050 mol/L ionic strength, chain unfolding causes an increase in the total space that is available for sorption of butanol, a decrease in overlap between sorbed TBA(+) and butanol, and a decrease in the sorbent strength of the ODS phase for butanol. The last effect is likely due to a reduction in contact area between a sorbed butanol molecule and the C(18) chains. At 0.50 mol/L ionic strength, chain unfolding decreases both the TBA(+)/butanol overlap and the sorbent strength but does not increase the total space available for butanol sorption. A physicochemical model is developed which fits well to the experimental data.