A novel ionic liquid-based monolithic column and its application in the efficient separation of proteins and small molecules by high-performance liquid chromatography

Abstract

A novel skeleton porous polymer-based monolith chromatography column has successfully been prepared using an in situ free radical polymerization technique. A 50 mm × 4.6 mm i.d. stainless steel chromatographic column used dodecanol as porogen and ionic liquid (IL), 1-dodecene (C12) and trimethylol propane triacrylate (TMPTA) as monomers, and ethylene dimethacrylate as crosslinker. The effect of variables such as temperature and porogen solvent content on the porous structure was studied in detail. The polymer-based monolith obtained was characterized by scanning electron microscopy, infrared spectroscopy, mercury intrusion porosimetry, and nitrogen adsorption. The results indicated that the monolithic column had a porous structure, good mechanical stability, high permeability (6.77 × 10⁻¹⁴ m²), and a high specific surface area (155.62 m² g⁻¹). The liquid chromatographic performance of the monolith was evaluated in the separation of lysozyme from egg white and in the separation of a variety of mixtures of small molecules, such as amines and benzene analogues. The column showed good repeatability and reproducibility, and the column-to-column (n = 7) and batch-to-batch (n = 5) reproducibility was 2.85 and 3.15%, respectively.

---

1. Introduction

Polymeric monoliths were introduced about 20 years ago as materials facilitating rapid mass transport driven by convection though large pores in the monolith.¹ As fourth-generation chromatographic sorbents, monolithic columns possess a unique structure and exhibit unusual characteristics, making them suitable as a continuous porous separation medium.² During recent decades monolithic columns have developed rapidly due to their low back pressure, high permeability, rapid mass transfer, simple preparation and easy modification.³ As a result, monolithic columns have become an excellent tool in analytical techniques, not only in separation techniques, such as ion-exchange, hydrophobic interaction, size exclusion, and affinity chromatography, but also in sample preparation, including solid-phase and microextraction.^4,5

As a separation medium for high-performance liquid chromatography (HPLC), monolithic columns normally have a matrix which is either inorganic silica-based, or organic polymer-based derived from polystyrenes, polymethacrylate esters or polyacrylamides.^6–8 Silica- and polymer-based matrices have different properties. Those based on silica allow rapid separation of small molecules, but the preparation process is complex and difficult to control,⁹ and a further disadvantage is that the Si–O linkage is prone to hydrolysis, giving a relatively narrow pH range (2–8) in use.

Organic polymer-based monoliths, on the other hand, can readily be post-modified and are stable over a wide pH range (1–14).¹⁰ In particular, polymeric materials provide an excellent stationary phase in HPLC for the rapid separation of large molecules, such as proteins, nucleic acids and peptides.¹¹ Nevertheless, they have a number of disadvantages, such as poor reproducibility and resolution, and they have a non-uniform structure due to inadequate solubility of the monomers and porogens. To overcome these problems, we have therefore developed an alternative basis using ionic liquids (ILs) as a co-monomer.

Ionic liquids are a type of non-volatile organic molten salts which are non-molecular, and are stable and chemically inert. At temperatures below 100 °C they are liquid,^12–14 and they have attracted wide interest due to their unique properties, including low volatility, tunable viscosity and good biocompatibility.^15,16 ILs are of wide interest in a variety of applicational areas, not only in analytical chemistry but also more widely.^17,18 Based on the properties of ILs,¹⁹ a new IL-based monolithic material has now been synthesized.

In the present study a novel HPLC monolithic column was prepared by in situ free-radical polymerization using 1-vinyl-3-butylimidazolium chloride as co-monomer. The effect of the variables influencing the porous structure have been studied in detail, and the optimized monoliths were assessed by separating lysine (Lys) from egg white and aromatic compounds.

2. Experimental

2.1 Materials

1-Dodecene (C12), 1-vinylimidazole and 1-chlorobutane were purchased from Shanghai Aladdin Co. Trimethylol propane triacrylate (TMPTA), ethylene dimethacrylate (EDMA) and azo-bis-isobutyronitrile (AIBN) were obtained from the Tianjin Chemistry Reagent Factory, and dodecanol, polyethylene glycol (PEG-200) and hexadecanol from the Shanghai Chemical Plant. The aromatic compounds studied were provided by the National Institute for the Control of Pharmaceutical and Biological Products of China (Beijing). HPLC-grade methanol and potassium bromide (KBr) were supplied by Kermel Chemical Reagent Co Ltd. (Tianjin), and the stainless steel columns (i.d. 50 × 4.6 mm) were purchased from Beijing Xinyu Instrument Co Ltd. Lys was obtained from Sigma Chemical Co. (St Louis, MO, USA). Triple-distilled water was used for all the experiments, and all liquids and solutions were filtered through a 0.45 μm membrane before use.

2.2 Instrumental

An 1100 system from Agilent Technologies (USA) was used in the chromatographic studies. The HPLC system consisted of a quaternary pump with an online vacuum degasser, an autosampler with variable injection capacity from 0.1 to 100 μL, and a UV detector. All the sample solutions injected into the chromatographic system were filtered through a millipore membrane (0.45 μm) to remove particles and large aggregates. The morphology of the monolithic columns was studied using a Hitachi S-3400 scanning electron microscope (Hitachi High Technologies, Tokyo), and the FT-IR spectra were recorded on an FTIR-8400S IR spectrometer (Shimadzu, Kyoto, Japan) over the range 400–4000 cm⁻¹.

2.3 Preparation of the polymer-based monolithic columns

2.3.1 Synthesis of 1-vinyl-3-butylimidazolium chloride ionic liquids (VBC-ILs). 1-Chlorobutane (14.81 g, 160 mmol, 16.6 mL) was added dropwise to 1-vinylimidazole (8.00 g, 85 mmol, 7.7 mL). The mixture was heated at 70 °C for 24 h under stirring. Phase separation occurred, and the viscous yellow liquid obtained was washed with ethyl acetate. The product was filtered and dried in a vacuum oven to constant weight. The synthesis of VBC-ILs is illustrated in Scheme 1.

Scheme 1 Synthesis of VBC-ILs.

2.3.2 Preparation of IL-based monolithic columns. The monolithic columns were directly synthesized by in situ polymerization using a pre-polymerization solution consisting of functional monomers, cross-linker, initiator, and porogen, after degassing in a bath sonicator for 15 min. The compositions are listed in Table 1, where IL, C12, and TMPTA were used as common monomers; EDMA as crosslinking agent; dodecanol, polyethylene glycol (PEG-200) and hexadecanol as porogens, and AIBN as initiator. The homogeneous solution obtained was manually injected into a clean stainless steel column (50 × 4.6 mm i.d.) which then was sealed at both ends with closed column heads. The polymerization was initiated in a 60 °C water bath for 24 h. The resulting monolithic column was washed online with methanol in conjunction with HPLC to remove unreacted monomers, porogen and other soluble compounds present in the polymeric rod. The polymerization scheme is illustrated in Scheme 2.

Table 1 Composition of the pre-polymerization mixtures for preparation of the monoliths

Column^a	IL (mL)	C12 (mL)	TMPTA (mL)	EDMA (mL)	Dodecanol (mL)	PEG-200 (mL)	Hexadecanol (mL)	Polymerization temperature (°C)	Mechanical/physical properties	Pressure^b (bar)	Permeability (K, 10^−14 m^2)
a All columns were initiated by incorporation of 0.01 g AIBN. b Pressure was obtained using methanol as the mobile phase at 1.0 mL min^−1.
A	0.1	0.2	0.4	0.8	2.0	—	—	60	Proper hard	4	6.77
B	—	0.2	0.4	0.8	2.0	—	—	60	Hard	9	3.01
C	0.1	—	0.4	0.8	2.0	—	—	60	Granulous	10	2.71
D	0.1	0.2	0.4	0.5	2.0	—	—	60	Brittle	3	9.03
E	0.1	0.2	0.4	1.1	2.0	—	—	60	Hard	13	2.08
F	0.1	0.2	0.4	0.8	1.6	—	—	60	Quite hard	15	1.81
G	0.1	0.2	0.4	0.8	2.1	—	—	60	Soft	3	6.92
H	0.1	0.2	0.4	0.8	2.3	—	—	60	Soft	2	7.38
I	0.1	0.2	0.4	0.8	2.4	—	—	60	Quite soft	<2	—
J	0.1	0.2	0.4	0.8	—	2.0	—	60	Immiscible	—	—
K	0.1	0.2	0.4	0.8	—	—	2.0	60	Quite hard	—	0.91
L	0.1	0.2	0.4	0.8	2.0	—	—	40	—	—	—
M	0.1	0.2	0.4	0.8	2.0	—	—	50	Quite soft	<1	—
N	0.1	0.2	0.4	0.8	2.0	—	—	70	Quite hard	>20	0.65

---

Scheme 2 The polymerization scheme.

2.4 Characterization methods

2.4.1 Instrumental analytical methods. The preparation conditions have a considerable influence on the structure of monolithic columns. In order to obtain a poly(IL-co-C12-co-TMPTA-co-EDMA) monolithic column of suitable structure, SEM was used to investigate its morphology under different conditions. Porous properties of the monoliths were investigated by mercury intrusion porosimetry, and the specific surface area was calculated from nitrogen adsorption–desorption isotherms. The chemical groups in the monoliths were studied by Fourier transform infrared spectroscopy (FT-IR).

2.4.2 Calculation methods. The column permeability K was calculated according to eqn (1):where F is the mobile phase flow rate, η the dynamic viscosity of the eluent, L the column length, ΔP the pressure drop across the column, and r is the inner radius of the column. The dynamic viscosity of the methanol mobile phase was 0.580 × 10⁻³ kg m⁻¹ s⁻¹.

2.5 The separation of Lys and egg white

Lysine (abbreviated as Lys or K)¹ is an α-amino acid with structure HO₂C·CH(NH₂)·(CH₂)₄NH₂. It is an essential amino acid in humans.

The IL-based monolithic column was used to separate Lys from egg white by ionic interaction, in the following manner. Chicken egg white was separated from fresh eggs and diluted to 50% (v/v) using phosphate buffer (50 mmol, pH 7.0). The diluted egg white was homogenized in an ice-bath and centrifuged at 4 °C and 10 000 rpm for 10 min.

The following conditions were used for the chromatographic separation. The UV detector was used at 280 nm, the injection volume was 5.0 μL, the gradient was 0–3 min and an aqueous solution of 0.02 mol L⁻¹ Na₂HPO₄ (adjusted to pH 12 with aqueous NaOH solution) was used as the mobile phase; from 3.01–10 min, water was used as the mobile phase.

3. Results and discussion

3.1 Optimization of preparation conditions

3.1.1 The influence of porogens on monolith properties. The conditions and results shown in Table 1 indicate that columns based on PEG200 (column J) or hexadecanol (column K) as porogen had unsatisfactory properties, whereas dodecanol was a good solvent for the functional monomers and was chosen as the porogen solvent for further optimization.

In order to investigate the influence of dodecanol content of the pre-polymerization solution on the preparation of the poly(IL-co-C12-co-TMPTA -co-EDMA) monolith, a range of dodecanol contents were considered, listed in Table 1. The results showed that with an increasing proportion of dodecanol, the column permeability increased, but the hardness and back pressure decreased (columns A, F–I). On the other hand, when the proportion of dodecanol was increased to 2.4 mL (column I), the mechanical properties of the column became unacceptably poor. When the amount of dodecanol was 2.0 mL (column A), the monolithic column exhibited good permeability, moderate hardness, and low back pressure.

3.1.2 The influence of temperature on monolith properties. It is well known than the temperature of preparation affects the properties of the resulting polymeric monolith. Polymerization was therefore performed at four different temperatures (40, 50, 60 and 70 °C), and it was found that with increasing temperature the permeability of the monolith and its mechanical properties deteriorated. However, polymerization did not take place when the temperature was 40 or 50 °C. Bearing in mind the effect of temperature on back pressure and mechanical strength, 60 °C was therefore chosen for the subsequent experiments.

3.2 Characterization of the monoliths

3.2.1 SEM figures of monoliths. SEM was used to characterize the cross-sectional morphology of the resulting monolithic columns, as shown in Fig. 1(A–E), using the different compositions listed in Table 1(A–E). The results showed that different compositions of functional monomers affected the main structure. Fig. 1(A), obtained from column A, showed a more interconnected and uniformly porous structure than the other monoliths, formed by the accumulation of spherical particles. SEM photographs of the monolithic columns poly(C12-co-TMPTA-co-EDMA) (Table 1, column B) and poly(IL-co-TMPTA-co-EDMA) (column C) are shown in Fig. 1(B) and (C), respectively, both formed by accumulation of small globules. Increasing amounts of EDMA would increase crosslinking, the density of the pores would increase and the pore diameter would decrease. Compared to column A, column D (Fig. 1(D)) showed a looser pore structure, and column E (Fig. 1(E)) gave a relatively denser structure. These results indicated that a combination of IL, C12, TMPTA, EDMA and dodecanol could lead to a more porous and uniform structure of high permeability.

Fig. 1 Scanning electron microphotographs of monoliths.

The performance of the three monoliths, columns A, B and C, was confirmed by the liquid chromatographic separation of small molecules.

The peak band broadening from B and peak tailing from C indicate the effect of the composition of the monomer on the structure and the resulting chromatographic performance. After optimization and comparison, column A was adopted for the subsequent experiments.

3.2.2 FT-IR characterization of IL-based monolith. The groups present on the monolith were confirmed by FT-IR. As shown in Fig. 2, the spectrum at 2957–2901 cm⁻¹ was due to the C–H bonds. The characteristic peak of C=O double bonds was seen at 1750 cm⁻¹, and in addition the C–H bond stretching, vibration around the imidazole ring at 1169 cm⁻¹, could be observed, confirming the presence of IL on the IL-based monolith.

Fig. 2 The FT-IR spectrum of the groups present on column A.

3.2.3 Permeability and mechanical strength of the monoliths. Permeability (K) is an important parameter of HPLC columns. High permeability results in low back pressure and low mass transfer resistance. The permeability of the polymer-based monolithic columns was determined by pumping methanol through column A. Using eqn (1), the calculated permeability values were shown in Table 1. The results demonstrated that different preparations resulted in different permeabilities. After assessment of the back pressure and hardness of the monolithic columns, the IL-based column (column A), with a permeability of 6.77 × 10⁻¹⁴ m² was selected as optimal.

Fig. 3 shows the back pressure in Column A at different flow rates using methanol and water, respectively, as mobile phase. Although the flow rate was increased to 5 mL min⁻¹ using water as the mobile phase, the maximum pressure was 43 bar. In addition, the good linear relationship (r² > 0.999) between back pressure and flow rate confirmed its good mechanical stability.

Fig. 3 Back pressure in column A at different flow rates using methanol and water, respectively, as the mobile phase.

3.2.4 Pore size distribution in the monolith. The measurement of pore size distribution and specific surface area of the monolith was carried out by mercury intrusion porosimetry and the nitrogen adsorption–desorption isotherm, respectively. Fig. 4 shows the result obtained by mercury intrusion porosimetry. The total intrusion volume, average pore diameter and porosity were 1.88 mL g⁻¹, 1.31 μm, and 70.36%, respectively. By determination of specific surface area, the single point surface area at P/P₀ = 0.2999 was 154.74 m² g⁻¹ and the BET surface area was 155.62 m² g⁻¹. These results confirmed the large surface area of the IL-based monolithic column.

Fig. 4 Measurement of the pore size distribution of the monolith by mercury intrusion porosimetry.

3.3 Chromatographic behavior of the IL-based monolithic column

3.3.1 Separation of Lys from egg white. The chromatogram in Fig. 5 shows the successful separation of Lys from egg white. The mechanism was as follows. When Na₂HPO₄ aqueous solution (adjusted to pH 12 using aqueous NaOH solution) was used as the mobile phase at 0–3 min, the pH value of the mobile phase was higher than the pI (approximately 11) of Lys, and being negatively charged, the Lys was attracted by the positively charged monolith – thus the Lys was retained by the monolithic column. When water was used as the mobile phase at 3.01–10 min, the pH value of the mobile phase was lower than the pI of Lys, and the Lys became positive charged and was repulsed by the positively charged monolith, and the Lys was thus eluted. The IL-based monolithic column was positively charged due to the present of imidazolium groups, and the monolith could therefore be used to separate proteins by ionic interaction, as shown in the section below.

Fig. 5 Chromatogram of the separation of Lys from egg white.

In addition, the Lys content was assayed by ultraviolet spectrometry and its purity was calculated after vacuum freeze-drying as 92.1%.

3.3.2 The effect of the mobile phase in the separation of small molecules. Five mixed compounds were separated on the IL-based monolith using different ratios of methanol–water. Fig. 6 shows that the retention times of the five mixed compounds increased with decreasing methanol content, and were (a) 80%, (b) 75%, and (c) 70%, respectively. These analytes were eluted in the sequence aniline, p-xylene, naphthalene, diphenylamine and triphenylamine, in the order of their polarity from high to low, presenting a typical reversed-phase liquid chromatographic mode. Considering both the analysis time and the resolution, a ratio of methanol–water of 75 : 25 v/v, and a flow rate of 1.0 mL min⁻¹, was selected as providing optimal chromatographic conditions.

Fig. 6 Effect of methanol content in the mobile phase on the chromatographic separation. Conditions: monolithic column, 50 × 4.6 mm i.d.; flow rate: 1.0 mL min⁻¹; mobile phase: (a) methanol/water (80/20, v/v); (b) methanol/water (75/25, v/v); (c) methanol/water (70/30, v/v); UV detection wavelength: 254 nm; peak identification: (1) aniline, (2) p-xylene, (3) naphthalene, (4) diphenylamine, and (5) triphenylamine.

The chromatographic behavior of the IL-based monolithic column in the separation of aromatic compounds is shown in Fig. 7, in which the chromatogram (a) shows the baseline separation of four compounds with a methanol–water mobile phase (75 : 25, v/v). The analytes were eluted in the following order: aniline, p-xylene, naphthalene, diphenylamine and triphenylamine, which corresponded to the hydrophobicity of the five analytes from low to high. The result indicated a typical reversed-phase separation mode.

Fig. 7 (a) Chromatographic behavior of the IL-based monolith in the separation of five compounds with the mobile phase methanol/water (75/25, v/v). Analytes: (1) aniline, (2) p-xylene, (3) naphthalene, (4) diphenylamine, and (5) triphenylamine. (b). Retention factors (k) of each sample on the IL-based monolith determined at different methanol content in the mobile phase. Flow rate, 1.0 mL min⁻¹; detection wavelength, 254 nm. (c). Baseline separation of benzene, naphthalene, biphenyl and anthracene, with a mobile phase of methanol–water (68 : 32, v/v). Flow rate as above. (d). The retention factors of the four analytes on the IL-based monolith.

The retention factor (k) of each sample on the IL-based monolith was determined at different methanol contents in the mobile phase, and the results are shown in Fig. 7(b), confirming a typical reversed-phase liquid chromatographic mechanism. A typical separation of neutral aromatic compounds is shown in Fig. 7(c), with a baseline separation of four compounds with a mobile phase of methanol–water (68 : 32, v/v). The four compounds were eluted in accordance with their polarities from high to low in the order benzene, naphthalene, biphenyl and anthracene, and the column efficiencies for the four compounds were about 5940–9249 theoretical plates per meter. The retention factors (k) of the four analytes on the IL-based monolith are presented in Fig. 7(d), which also shows that the retention factor of each sample decreased following the increase in methanol content in the mobile phase, confirming the typical reversed-phase chromatographic mechanism of the separation.

3.4 Reproducibility

The reproducibility of the monolithic column was calculated from the percentage relative standard deviations of the retention factors of the test compounds on Column A. The average run-to-run reproducibility (n = 5) of benzene was 1.07%, while the average day-to-day reproducibility (n = 3) was 1.75%. This demonstrates the stability of the monolithic columns. In addition, the column-to-column and batch-to-batch reproducibility were also investigated using the same or different batches of polymerization mixture, respectively. The column-to-column (n = 7) and batch-to-batch (n = 3) reproducibility were 2.85 and 3.15%, respectively. This confirmed the intrinsic reproducibility and stability of the monolithic columns.

4. Conclusions

In this study a porous uniform poly-IL-based monolithic column with high specific surface area was successfully prepared using IL as co-monomer by in situ free radical polymerization, and was successfully used to separate Lys from egg white. In addition the monolithic column exhibited good performance in reversed-phase liquid chromatographic separation. The monolithic column offers a potentially useful alternative for the efficient separation of proteins and small molecules.

Acknowledgements

We are grateful for the financial support provided by the National Natural Science Foundation of China (21175031), the National Science Foundation of Hebei Province (B2012201052, B2013201082), and the National Science Foundation of Hebei University (2013-247).

References

1. G. Guiochon, J. Chromatogr. A, 2007, 1168, 101.
2. J. J. Ou, H. Lin, Z. B. Zhang, G. Huang, J. Dong and H. F. Zou, Electrophoresis, 2013, 34, 126.
3. H. F. Han, Q. Wang, X. Liu and S. X. Jiang, J. Chromatogr. A, 2012, 1246, 9.
4. W. J. Li, X. Zhou, S. S. Tong and Q. Jia, Talanta, 2013, 105, 386.
5. Z. H. Liu, Y. B. Peng, T. T. Wang, G. X. Yuan, Q. X. Zhang, J. L. Guo and Z. J. Zhang, J. Sep. Sci., 2013, 36, 262.
6. M. M. Zheng, G. D. Ruan and Y. Q. Feng, J. Chromatogr. A, 2009, 1216, 7739.
7. Y. J. Ueki, T. Umemura, J. X. Li, T. Odake and K. Tsunoda, Anal. Chem., 2009, 76, 7007.
8. F. Maya and F. Svec, Polymer, 2014, 55, 340.
9. J. Chen, P. F. Zhang and J. Li, J. Chromatogr. A, 2011, 1218, 3699.
10. Y. Tian, R. Feng, L. P. Liao, H. L. Liu, H. Chen and Z. R. Zeng, Electrophoresis, 2008, 29, 3153.
11. J. Liu, F. J. Wang, H. Lin, J. Zhu, Y. Y. Bian, K. Cheng and H. F. Zou, Anal. Chem., 2013, 85, 2847.
12. H. M. Albishri and D. A. El-Hady, Talanta, 2014, 118, 129.
13. T. Zhu, W. T. Bi and K. H. Row, J. Appl. Polym. Sci., 2010, 118, 3425.
14. Y. Wang, Q. L. Deng, G. Z. Fang, M. F. Pan, Y. Yu and S. Wang, Anal. Chim. Acta, 2012, 712, 1.
15. C. C. Liu, Q. L. Deng, G. Z. Fang, H. L. Liu and J. H. Wu, Anal. Chim. Acta, 2013, 804, 313.
16. X. J. Huang, L. L. Chen, D. X. Yuan and S. S. Bi, J. Chromatogr. A, 2012, 1248, 67.
17. H. J. Zheng, Q. W. Liu and Q. Jia, J. Chromatogr. A, 2014, 1343, 47.
18. B. Singco, C. L. Lin, Y. J. Cheng, Y. H. Shih and H. Y. Huang, Anal. Chim. Acta, 2012, 746, 123.
19. Y. Y. Fan, S. H. Liu and Q. L. Xie, Talanta, 2014, 119, 291.

---