Anal. Chem. 1996, 68, 100-105
Temperature-Responsive Chromatography Using Poly(N-isopropylacrylamide)-Modified Silica Hideko Kanazawa,* Kazuo Yamamoto, and Yoshikazu Matsushima
Kyoritsu College of Pharmacy, 1-5-30 Shiba-koen, Minato, Tokyo 105, Japan Nobuharu Takai
Faculty of Science and Engineering, Tokyo Denki University, Ishizuka, Hatoyama-cho, Hiki-gun, Saitama 350-03, Japan
Downloaded via UNIV OF CALIFORNIA SANTA BARBARA on July 12, 2018 at 16:13:36 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Akihiko Kikuchi, Yasuhisa Sakurai, and Teruo Okano*
Institute of Biomedical Engineering, Tokyo Women’s Medical College, 8-1 Kawada-cho, Shinjuku, Tokyo 162, Japan
A new concept in chromatography is proposed that utilizes a temperature-responsive surface with a constant aqueous mobile phase. The surface of the silica stationary phase in high-performance liquid chromatography (HPLC) has been modified with temperature-responsive polymers to exhibit temperature-controlled hydrophilic/hydrophobic changes. Poly(N-isopropylacrylamide) (PIPAAm) was grafted onto (aminopropyl)silica using an activated esteramine coupling method. These grafted silica surfaces show hydrophilic properties at lower temperatures which, as temperature increases, transform to hydrophobic surface properties. The elution profile of five mixed steroids on an HPLC column packed with this material depends largely on the temperature of the aqueous mobile phase. Retention times increase with increasing temperature without any change in the eluent. Changes in the retention times of hydrophobic steroids were larger than those for hydrophilic steroids. The temperature-responsive interaction between PIPAAm-modified silica and these steroids is proposed to result from changes in the surface properties of the HPLC stationary phase by the transition of hydrophilic/hydrophobic surface-grafted IPAAm polymers. We demonstrate a novel and useful new chromatography system in which surface properties and the resulting function of the HPLC stationary phase are controlled by external temperature changes. This method should be effective in biological and biomedical separations of peptides and proteins using only aqueous mobile phases. High-performance liquid chromatography (HPLC) has become the most widely used analytical technique for pharmaceutical compounds and biochemicals presently. Reversed-phase chromatography (RPC) using chemically modified stationary phases is generally faster and easier than other modes of HPLC and, consequently, has achieved wide popularity. The retention and selectivity in RPC are controlled primarily by changing the polarity of the mobile phase. Use of organic solvents is necessary to prevent excessively long retention times with conventional reversedphase columns. 100 Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
The aim of this work is to develop of a new chromatography system in which surface properties and functions of the HPLC stationary phase are controlled by external temperature changes without changing the mobile phase composition. Isocratic elution by an aqueous mobile phase alone is the basis for separation of peptides and proteins. There are many synthetic polymers with molecular conformations sensitive to environmental factors such as pH,1 electric field,2 chemical species,3 and temperature.4 The various properties of aqueous polymer solutions have been studied by many investigators.4-7 Among those, the phase transition of aqueous poly(N-isopropylacrylamide) (PIPAAm) solution is quite sensitive to, reversible, and reproducible in thermal stimulation, in contrast to that of other polymers.4,6 A lower critical solution temperature (LCST)5 of PIPAAm was observed around 32 °C in aqueous solutions. Polymer chains of IPAAm are hydrated and solubilized in water below the LCST and dehydrated and precipitated above the LCST. These transition phenomena of aqueous PIPAAm solution are mainly due to the conformational changes of the polymer chain arising from hydration changes of isopropyl side groups. Temperature-responsive properties of N-isopropylacrylamide (IPAAm) polymer and polymer gels have been applied in various fields.8-12 Previously, we have prepared PIPAAm beads packed into a glass column.13 HPLC elution behavior of antipyretics and (1) Kopecek, J.; Vacik, J.; Lim, D. J. Polym. Sci. A1 1971, 9, 2801. (2) Tanaka, T.; Nishino, I.; Sun, S.-T.; Ueno-Nishio, S. Science 1981, 218, 467469. (3) Ishihara, K.; Muramoto, N.; Shinohara, I. J. Appl. Polym. Sci. 1984, 29, 211. (4) Bae, Y. H.; Okano, T.; Kim, S. W. J. Polym. Sci. Polym. Phys. 1990, 28, 923-936. (5) Heskins, M.; Guillet, J. E.; James, E. J. Macromol. Sci. Chem. 1968, A2, 1441-1445. (6) Fujishige, S.; Kubota, K.; Ando, I. J. Phys. Chem. 1989, 93, 3311-3313. (7) Fujishige, S. Polym. J. 1987, 19, 297-300. (8) Okano, T.; Bae, Y. H.; Jacobs, H.; Kim, S. W. J. Controlled Release 1990, 11, 255-265. (9) Yoshida, R.; Sakai, K.; Okano, T.; Sakurai, Y. J. Biomater. Sci. Polym. Ed. 1991, 3, 155-162. (10) Takei, Y. G.; Aoki, T.; Sanui, K.; Ogata, N.; Sakurai, Y.; Okano, T. Bioconjugate Chem. 1993, 4, 341-346. (11) Matsukata, M.; Takei, Y. G.; Aoki, T.; Sanui, K.; Ogata, N.; Sakurai, Y.; Okano, T. J. Biochem. 1994, 116, 682-686. (12) Okahata, Y.; Noguchi, H.; Seki, T. Macromolecules 1986, 19, 493-494. 0003-2700/96/0368-0100$12.00/0
© 1995 American Chemical Society
proteins on the column with an aqueous mobile phase has been reported. Although the retention of proteins was altered in response to changes in mobile phase temperature, the temperature-induced shrinkage of the beads sometimes produced voids and channels in the column, leading to a loss of efficiency. Gewehr et al. examined gel permeation chromatography using porous glass beads modified with the IPAAm polymers.14 Grafted PIPAAm was used to control pore size by changing column temperature. Hosoya et al.15 also reported a surface-selective modification procedure for the incorporation of PIPAAm into porous polymer beads. They performed size exclusion chromatography of dextran with water as a mobile phase. However, these reports were concerned mainly with temperature-responsive coilglobule conformation changes in PIPAAm molecules to control pore size. Only slight changes in solute retention were observed on their column. This may be due to the scant introduction of PIPAAm molecules on the surface as well as inside surfaces of the porous beads to alter the solute diffusivity through the pores. We have prepared PIPAAm with a carboxyl group at one end (semitelechelic PIPAAm),16 which exhibits soluble/insoluble reversible changes in response to temperature changes in aqueous media.17 We have already succeeded in demonstrating the hydrophilic-hydrophobic surface property changes of PIPAAmimmobilized surfaces by temperature changes. On the PIPAAmmodified surfaces, controlled attachment-detachment modulation of cultured cells was achieved by temperature changes.18,19 We reported that IPAAm polymer-grafted surfaces exhibit hydrophilic properties at lower temperatures and hydrophobic properties with elevated temperatures using dynamic contact angle measurements.17 Carboxyl semitelechelic PIPAAm was introduced into aminated surface via amide bond formation. At 20 °C (below PIPAAm LCST), the PIPAAm-grafted surface showed a water contact angle of 51°, while above the LCST, the contact angle of the PIPAAm-grafted surface was 88°, which was comparable to the value observed on polystyrene surface.17 Temperatureresponsive PIPAAm-grafted surfaces maintaining a free mobile end demonstrated more rapid and significant temperature responses than multipoint-grafted PIPAAm surfaces. These features were suggested to be due to more effective restricted conformational freedom for the PIPAAm graft chains, which influence polymer dehydration and hydrogen bonding with water molecules. Furthermore, PIPAAm was introduced as graft chains into crosslinked PIPAAm hydrogel. This graft-type hydrogel clearly showed rapid deswelling change in response to temperature change. This rapid shrinking of the graft-type PIPAAm gel is due to the immediate dehydration of the free mobile grafted chains in the gel matrix, followed by subsequent hydrophobic interactions between dehydrated grafted chains.20 (13) Kanazawa, H.; Nagata, Y.; Matsushima, Y.; Takai, N.; Okano, T.; Sakurai, Y. Proceedings of the First International Conference on Intelligent Materials; Oiso, Japan, March 23-25, 1992; Technomic: Lancaster, 1993; pp 415418. (14) Gewehr, M.; Nakamura, K.; Ise, N.; Kitano, H. Macromol. Chem. 1992, 193, 249-256. (15) Hosoya, K.; Sawada, E.; Kimata, K.; Araki, T.; Tanaka, N.; Fre`chet, J. M. J. Macromolecules 1994, 27, 3973-3976. (16) Okano, T.; Katayama, M.; Shinohara, I. J. Appl. Polym. Sci. 1978, 22, 369377. (17) Takei, Y. G.; Aoki, T.; Sanui, K.; Ogata, N.; Sakurai, Y.; Okano, T. Macromolecules 1994, 27, 6163-6166. (18) Yamada, N.; Okano, T.; Sakai, H.; Karikusa, F.; Sawasaki, Y.; Sakurai, Y. Makromol. Chem., Rapid Commun. 1990, 11, 571-576. (19) Okano, T.; Yamada, N.; Sakai, H.; Sakurai, Y. J. Biomed. Mater. Res. 1993, 27, 1243-1251.
Figure 1. Temperature dependence for optical transmittance of 0.5 wt % PIPAAm solutions containing various NaCl concentrations: O, 1.0 M; b, 0.5 M; 0, 0.1 M; 9, 0.05 M; 4, 0.01 M; and 2, 0 M.
Scheme 1. Synthesis of PIPAAm with a Carboxyl Group at One End
In this study, IPAAm polymer grafted to (aminopropyl)silica by an activated ester-amine coupling was used as an HPLC packing material. A new thermoresponsive chromatography concept is demonstrated using the steroids with a variety of hydrophobicities. EXPERIMENTAL SECTION Materials. N-Isopropylacrylamide (IPAAm; Kodak, Rochester, NY) was purified by recrystallization from a toluene-hexane mixture and dried at room temperature in vacuo. 3-Mercaptopropionic acid (MPA; Wako Pure Chemicals, Osaka, Japan) was distilled under reduced pressure, and the fraction boiling at 95 °C/5 mmHg was used. 2,2′-Azobis[isobutyronitrile] (AIBN), N,Ndimethylformamide (DMF), and ethyl acetate (EtOAc) were obtained from Wako Pure Chemicals and purified by conventional methods. (Aminopropyl)silica (average diameter of 5 µm, pore size 120 Å) was purchased from Nishio Kogyo (Tokyo, Japan). Sulfosuccinimidyl 4-O-(4,4′-dimethoxytrityl)butyrate (s-SDTB) was obtained from Pierce (Rockford, IL). Cortisone acetate, hydrocortisone, and hydrocortisone acetate were from Wako Pure Chemicals, and other steroids were from Sigma Chemicals (St. Louis, Mo.). 1-Ethyl-3-[3-(dimethylamino)propyl]carbodiimide hydrochloride (WSC) was obtained from Dojindo Laboratories (Kumamoto, Japan). Milli-Q grade water was used for preparation of sample solutions. Other reagents and solvents were commercially obtained and used without further purification. (20) Yoshida, R.; Uchida, K.; Kaneko, Y.; Sakai, K.; Kikuchi, A.; Sakurai, Y.; Okano, T. Nature 1995, 374, 240-242.
Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
101
Figure 2. Chromatograms of a mixture of five steroids and benzene with water as mobile phase at (a) 5, (b) 25, (c) 35, and (d) 50 °C. Peaks: 1, benzene; 2, hydrocortisone; 3, prednisolone; 4, dexamethasone; 5, hydrocortisone acetate; and 6, testosterone.
Polymerization Procedure. Semitelechelic IPAAm polymers were prepared by radical polymerization of IPAAm in DMF.21 IPAAm (0.44 mol) was dissolved in DMF. AIBN (3.9 mmol) and MPA (12 mmol) were used as initiator and chain transfer agent, respectively. The reaction mixture was degassed by subjecting it to freeze-thaw cycles, and the ampule containing the mixture was sealed under reduced pressure. The reaction was then performed at 70 °C for 12 h. After evaporation of the solvent, the reactant was poured into diethyl ether to precipitate the polymer. The polymer was further purified by repeated precipitation from DMF into diethyl ether. The molecular weight of the polymer was determined by end group titration with 0.01 M NaOH using phenolphthalein as an indicator. Transmittance Measurements. The LCST of IPAAm polymers was determined by measuring the optical transmittance of polymer aqueous solutions. The optical transmittance of polymer solutions (5 mg/mL) was measured at 500 nm at various temperatures using a spectrophotometer (Shimadzu, UV-240). The temperature of the observation cell was controlled with a LAUDA RC20 water bath, with a deviation of (0.02 °C. Modification of (Aminopropyl)silica with IPAAm Polymer. Carboxyl groups terminating each IPAAm polymer were activated with N-hydroxysuccinimide using dicyclohexylcarbodiimide (DCC) in dry ethyl acetate at 0 °C for 2 h, followed at 25 °C for 12 h in mole ratios of 1:2.5:2.5, respectively. After the precipitated dicyclohexylurea (DCU) was removed by filtration, the mixture (21) Takei, Y. G.; Aoki, T.; Sanui, K.; Ogata, N.; Okano, T.; Sakurai, Y. Bioconjugate Chem. 1993, 4, 42-46.
102
Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
was concentrated by evaporation. The activated polymers were recovered by precipitation from dry diethyl ether. The presence of the succinimidyl group at the end of the polymer was confirmed by the appearance of a new peak in the IR spectrum at 1780 cm-1, corresponding to the CdO stretching, and an ultraviolet absorption at 260 nm in NH4OH, corresponding to the succinimidyl anion.22 The protocol used to prepare the PIPAAm-modified silica was based on the method of Bu¨ckmann et al.23 Modification of (aminopropyl)silica with the IPAAm polymer was performed as follows: To a solution of 2 g of activated IPAAm polymer in 50 mL of dioxane was added 6 g of (aminopropyl)silica. The solution allowed the condensation reaction to proceed between active esters and amino groups at 25 °C for 12 h with gentle shaking. This process was repeated three times. The resulting PIPAAmmodified silica was washed with 500 mL of water and 100 mL of methanol consecutively and dried under vacuum at room temperature. Quantification of Amino Groups on Supports. The amount of amino groups on silica supports was determined by spectrophotometric method using s-SDTB.24 The dry support (∼1 mg) was immersed into 5 mL of 50 mM sodium bicarbonate buffer (pH 8.5) in a centrifuge tube. One milliliter of 0.1 mM s-SDTB solution was added to the tube, and the support was resuspended. The suspension was vigorously shaken at 25 °C for 30 min. After (22) Miron, T.; Wilchek, M. Anal. Biochem. 1982, 126, 433-435. (23) Bu ¨ ckmann, A. F.; Morr, M.; Johansson, G. Makromol. Chem. 1981, 182, 1379-1384. (24) Gaur, R. K.; Gupta, K. C. Anal. Biochem. 1989, 180, 253-258.
Scheme 2. Modification of (Aminopropyl)silica with Semitelechelic PIPAAm
Figure 3. Retention times of the five steroids (a) on the temperatureresponsive column and (b) on unmodified silica column: 2, benzene; O, hydrocortisone; 0, prednisolone; 9, dexamethasone; b, hydrocortisone acetate; and 4, testosterone.
removal of the supernatant, the support was washed with water four times. A weighed support (W, mg) was treated with perchloric acid solution to liberate dimethoxytrityl cation. Absorbance at 498 nm (∆498) was measured on the spectrophotometer. The amount of the amino groups on the support (Q, µmol/g of solid support) is then calculated from the following equation:
Q ) 14.3∆498V/W
where V expresses the volume of perchloric acid solution. Temperature-Responsive Elution Changes for Steroids. The polymer-grafted silica support was packed into a stainlesssteel column (150 mm × 4.6 mm i.d.). The column was connected to an HPLC system (HITACHI Model L-6200 intelligent pump; L-4000 UV monitor, D-2500 data processor). Pure water was used as the mobile phase. The elution behaviors of drug samples were recorded at a flow rate of 1 mL/min at various temperatures within a deviation of (0.02 °C. The column temperature was controlled by a LAUDA RC20 water bath. RESULTS AND DISCUSSION PIPAAm and Their Temperature-Responsive Solubility Changes. Carboxyl semitelechelic polymers of IPAAm were synthesized by telomerization using MPA as a chain transfer agent at 70 °C in DMF. Scheme 1 shows the synthesis of IPAAm polymer. Polymer molecular weights were controlled by adjusting the relative concentration of monomers to MPA in the starting mixture for the telomerization.21 The molecular weight of IPAAm polymers estimated by an end group titration was 4400. Figure 1 shows the temperature dependence for optical transmittance of IPAAm polymer solutions of various concentrations in NaCl. In pure water, the transition temperature (LCST) was observed at 32 °C. The LCST decreased with increasing concentrations of NaCl, while the sharp soluble-insoluble changes were maintained. The LCST remarkably shifted to 20 °C in 1 M NaCl solution. As PIPAAm is a nonionic compound, an electrostatic interaction should not affect the LCST. The lowering of the LCST by the addition of salt should, therefore, be due to acceleration of dehydration, i.e., salting-out.
Modification of (Aminopropyl)silica with IPAAm Polymer. Scheme 2 summarizes the coupling of carboxyl-terminated IPAAm polymers to (aminopropyl)silica surfaces. The amount of amino groups on the silica support estimated by a spectrophotometric method using s-SDTB was 210 µmol/g of solid support. Similarly, the amount of grafted polymer chains on the silica support, estimated from the residual amount of amino groups on the silica support, was 154 µmol/g of solid support, indicating that 73% of amino groups were reacted with PIPAAm molecules. Temperature-Responsive HPLC. The effect of column temperature on the elution behavior of steroids with different hydrophobicity was examined. Figure 2 shows the elution behavior of five steroids and benzene. At temperatures lower than the LCST (32 °C), four of the steroids are not resolved. Excellent resolution of the steroids was achieved at 50 °C. Additionally, retention times of the steroids increased with increasing temperature. The elution profile was strongly affected by temperature. Changes in the retention times for hydrophobic steroids such as testosterone were larger than those of their hydrophilic counterparts and benzene. The retention times of five steroids and benzene on columns packed with supports modified with PIPAAm chains were measured in aqueous mobile phases over a variety of column temperatures. The retention times of the five steroids Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
103
Table 1. Effects of Column Temperature and NaCl in Mobile Phase on Capacity Factors of Steroids 5 °Ca
benzene hydrocortisone prednisolone dexamethasone hydrocortisone acetate testosterone a
25 °Ca
50 °Ca
H2O NaClb H2O
NaCl b
H2O
NaClb
0.42 0.75 0.95 1.17 1.23 2.81
0.92 2.25 3.10 4.49 6.72 11.96
1.11 3.04 4.02 5.89 8.42 15.01
1.37 3.73 4.91 7.65 13.16 21.77
0.06 1.13 1.46 1.86 2.28 4.89
0.52 1.13 1.45 1.81 2.14 4.28
Column temperture. b 1 M NaCl was used as mobile phase.
Figure 4. van’t Hoff plots of the five steroids (a) on the temperatureresponsive column and (b) on unmodified silica column: b, hydrocortisone; O, hydrocortisone acetate; 0, prednisolone; 9, dexamethasone; and 2, testosterone.
largely depended on temperature, as shown in Figure 3a. On the temperature-responsive column, the increase in the retention times with increasing temperature clearly demonstrated a reversed tendency compared with ordinary HPLC columns between 25 °C and 35 °C, which should reflect the phase transition of the PIPAAm. Figure 3b shows the retention times of the steroids on the column of the unmodified support plotted against temperature. The retention times of steroids were decreased at higher temperature, and therefore, poor separation was observed. In general, the adsorption of molecules on the surface and the viscosity of mobile phases were decreased, while the solubility was increased at an elevated temperature. Hence, the retention times should decrease with increasing temperature for a normal chromatographic process. In typical cases, retention times in reversedphase chromatography decrease by 1-2% for a temperature increase of 1 °C.25 However, in the PIPAAm-modified columns, the opposite behavior of retarded retention times was observed with increasing temperature in spite of decreased back pressure of column. It is noteworthy that the retention of steroids shows large changes above the LCST of PIPAAm. This implies that the transition of hydrophilic/hydrophobic surface properties at LCST causes this anomalous retention behavior of the steroids. Figure 4 shows the van’t Hoff plots for five steroids on both a PIPAAm-grafted silica packed column and an unmodified silica column. Discontinuities in the plots are clearly shown on PIPAAm-grafted silica column at the temperature around the PIPAAm phase transition, while opposite and smaller capacity factors with temperature changes were observed on unmodified silica column. A retarded elution of steroids was enhanced with more hydrophobic steroids; therefore, the hydrophobicity of the substances also has an influence on the increasing hydrophobic interaction between dehydrated PIPAAm-grafted surfaces and steroids. The results strongly supported the above discussion of the hydrophobic interaction between hydrophobized PIPAAmsurface and steroids becoming larger with increasing temperature. The retention of steroids was increased at lower temperature on the temperature-responsive column, in contrast to that on unmodified silica. There might be interactions between the solute molecules and PIPAAm chains even at lower temperatures, such as hydrophobic interaction with the backbone and/or the isopropyl group of the side chain of PIPAAm. (25) Lindsay, S. High performance liquid chromatography, 2nd ed.; John Wiley & Sons: Chichester, 1992; p 302.
104 Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
Figure 5. Chromatograms of a mixture of five steroids and benzene with 1 M NaCl aqueous solution as mobile phase at (a) 5, (b) 25, and (c) 50 °C. Peak numbers correspond to the steroids indicated in Figure 2.
The data shown above indicate that the driving force for retention in this system is the hydrophobic interactions between the solute molecules and polymer chains on the surface. At temperatures higher than the LCST, the PIPAAm-grafted surface exhibited a hydrophobic property, and the sensitivity to hydrophobicity of the solutes was remarkably increased. Effects of salt addition in the mobile phase were then examined. Little difference in the PIPAAm phase transition point was observed between pure water and 0.1 M NaCl, as can be seen in Figure 1. However, if a mobile phase with concentration higher than 1 M NaCl is used, the transition temperature decreases to 20 °C. Large elution differences were noted between the lower and the higher LCSTs induced by salt. The effects of column temperature and NaCl addition in mobile phase on the capacity factors of steroids are summarized in Table 1. As shown in Figures 2 and 5, the retention times of steroids at 25 °C were much larger with 1 M NaCl aqueous solution than with pure water as mobile phases. In Figure 5a, some resolution was obtained even at 5 °C. Comparing Figure 5c with Figure 2d, the retention time of testosterone (peak 6) changed from 34 min to 48 min on changing the mobile phases from pure water to 1 M NaCl aqueous solution at 50 °C. This should result from a lowering of the LCST by changing the mobile phase from pure water to 1 M NaCl aqueous solution. This suggests that we can control LCST and
the hydrophobicity of the surface of the stationary phases, and thus, elution time by controlling the salt concentration of mobile phases. As described above, the column packed with PIPAAm-modified silica showed drastic changes in retention of solutes with small changes in column temperatures. There should be interactions between the solutes and polymer chains of the surface on the stationary phase in our system. The temperature-responsive interaction between PIPAAm-modified silica and the steroids should be due to changes in the surface properties of the PIPAAmgrafted stationary phase by the reversible transition of hydrophilic/hydrophobic PIPAAm-grafted surface properties. In isocratic elution of samples containing solutes with a wide range of polarity, it is sometimes difficult to achieve the desired resolution in a reasonable time. It may be necessary to use gradient elution, where volumes of an organic solvent, composition of mobile phase, or other properties of the solvent (e.g., pH or ionic strength) are changed during the separation. On HPLC columns packed with thermoresponsive polymermodified silica, gradient elution-like effect can also be achieved with a single mobile phase by controlling external temperature. In the case of separating of peptides, proteins, and other biological molecules or maintaining viable cells, it is frequently necessary to avoid the use of organic solvents in the mobile phase, as these cause sample denaturation. Total avoidance of organic solvents is also advantageous for environmental reasons. HPLC with this new packing material is applicable for the separation of peptides and drugs with aqueous solutions as mobile phases,
purification of peptide drugs from synthesis using recombinant methods, and analysis of bioactive peptides in body fluids. CONCLUSION PIPAAm-modified silica exhibits temperature-controlled hydrophilic/hydrophobic surface property changes in aqueous system. Using the column packed with PIPAAm-modified silica, separation of steroids was carried out by changing the temperature. With increasing temperature, increased interaction between solutes and PIPAAm-grafted surfaces of the stationary phases was observed. Temperature-dependent resolution of steroids was achieved using only water as a mobile phase. This paper describes the first successful temperature-responsive chromatography, in which surface property and function of the stationary phase are controlled by external temperature change. This system would be highly useful for controlling the function and properties of HPLC stationary phases simply by changing temperature in aqueous solvent. Such temperature-responsive chromatography is a new concept in LC with a high potential and versatility. ACKNOWLEDGMENT The authors are grateful to Dr. David W. Grainger, Colorado State University, for his valuable comments and discussion throughout this work. Received for review April 11, 1995. Accepted October 16, 1995.X AC950359J X
Abstract published in Advance ACS Abstracts, November 15, 1995.
Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
105
