What's New ?

Our e-journal " Pharmaceutical Reviews " (ISSN 1918-5561 ) is now indexed in directory of open access journal (DOAJ) .

Monolithic Columns

By Zahid Zaheer - 10/01/2008

in

Login or register to post comments

Average:

Average: 3.7 (3 votes)

Investigations into new separation media continue to be a major aspect in separation sciences.

This is due to the variety of new challenges in separating very complex sample mixtures. Emerging technology is enabling very high separation efficiency and speed of analysis, surpassing the conventional particle-packed columns in high performance liquid chromatography (HPLC). These included capillary electrochromatography,¹ultrahigh pressure HPLC (UHPLC),²^{, 3}and the use of monolithic silica columns.⁴Monolithic columns attracted attention because of their potential high performance under common operating conditions that rivals that of packed columns without high pressure requirements.

Conventional HPLC columns are packed with particles of micrometer size. Packing a high-efficiency column requires skills as well as the appropriate packing material with suitable properties. Columns having one-piece network structures are thought to be desirable to avoid difficulties in packing columns, which can change as new surface modifications are introduced.⁵Monolithic columns were reported first with organic polymers,⁶^{, 7}followed by silica columns prepared in capillaries.⁸There have been numerous reports on polymer- and silica-based monolithic columns for HPLC. ⁹^-11

Silica Based Monoliths

One of the great features of monolithic silica columns is their rod-like structure consisting of a skeleton and through-pores. The columns can offer a variable external porosity and through-pore size/skeleton size ratios that are impossible to achieve with particle-packed columns.These characteristics give monolithic columns very high permeability that allows their operation at pressures much lower than traditional HPLC. The preparation of silica-based monoliths make use of sol-gel processing under controlled conditions allowing desirable characteristics.

Sol-gel Process

The most common approach to fabricate silica-based monolithic columns is the acid-catalyzed sol-gel process.^{10,12,13,14,15}This process consists of hydrolysis and condensation reactions of metal alkoxides under acid conditions .¹⁶The commonly used metal alkoxides for silica-based monoliths are alkoxysilanes such as tetramethoxysilane (TMOS) and tetraethoxysilane (TEOS). Polycondensation then occurs with the linkage of silanol groups to form cyclic oligomers and eventually cast a silicate network.

Additives

The pore size and the mechanical properties of gels can be varied with the addition of polyethylene glycol (PEG) to the sol. PEG is a porogen which acts as a through-pore template and solubilizer of the silane reagent. This has been done by Nakanishi¹⁷, Judenstein¹⁸and Martin¹⁹who claimed that high concentrations of PEG weaken the solid matrix whereas a small concentration of PEG strengthens the matrix. The pore size of macroporous silica aerogel can be controlled by varying the concentration of water soluble polymer. Narrow and more uniform pore size distribution is observed with the addition of glycerol which acts as a drying additive since it prevents further reaction of water. Mesopores are formed in the silica skeletons by a treatment with ammonia, introduced after the formation of a network structure of silica skeletons. Ammonia canalso be generated by the hydrolysis of urea, which can be added in an initial reaction mixture.

Rinsing and Aging

Once the gel forms with a desired shape, rinsing in H₂O/EtOH will increase the permeability of the solid portion of the silica gel by a dissolution–reprecipitation process. Aging in a siloxane solution increases the stiffness and strength of the alcogel by adding new monomers to the silica network and by improving the degree of siloxane crosslinking; conversely this step will reduce the permeability²⁰. Einarsrud et al. ²¹^{, 22}have reported strengthening of the silica gels aged in TEOS, water, and ethanol solutions. Gels are washed with a 20% water–ethanol solution for 24 h at 50-60 ºC, then an aging solution (70%TEOS/ethanol, v/v) is used for 6–72 h at 50-70 ºC followed by washing with ethanol and heptane. Data from small angle neutron scattering shows only a slight increase in the volume fractal dimension of the porous gel network. The same group demonstrated that washing in a water solution increases the permeability of the gels by dissolution–reprecipitation.²³Silylation removes Si–OH surface groups by promoting silica polycondensation resulting in a decrease of pore size.

Drying

Drying of the gelis a critical step. Drying is governed by capillary pressure. During drying, shrinkage of the gel occurs due to capillary stress. Itis the gradient in capillary pressure within the pores that leads to mechanical damage; the capillary tension developed during the drying may reach up to 15,000-30,000 psi (1000-2000 bar)²⁴ with consequent shrinkage and cracking. Silica gel may decrease in volume by as much as a factor of 10 as it dries.

The extent of shrinkage is governed by the balance between capillary pressure P_c,and modulus of the solid matrix:

P_c = -(2 γ _(LV) cosθ / r_h)

Where γ _(LV)is the surface tension of the pore liquid at the liquid vapor interface, θ is the

contact angle of the liquid, and r_his the hydraulic pore radius.

Monolith in a Tube or in a Capillary

Monolithic silica columns can be prepared either in a test tube or in a fused-silica capillary column. In the case of preparation in a test tube, silica network structures undergo shrinkage during the reaction and a subsequent aging process, typically up to 70% of the mold size,^{26 -28}while in a capillary the silica skeletons are covalently attached to the capillary wall so that no void can be formed. The problem of shrinkage of skeletons, however, has not been completely solved. Currently, a monolithic silica structure can be formed in a capillary of up to 100 μm i.d., starting from TMOS, while successful preparation of up to 530 μm i.d. capillary columns was reported, starting from a mixture of TMOS and methyltrimethoxysilane (MTMS).²⁹

For preparation in a test tube, the resulting silica monolith are heat-treated and the resulting rods are encapsulated with Polyaryletheretherketone (PEEK) resin to fabricate a column that can be used under a pressure of up to 20 MPa. This encapsulation is known as the cladding process. The columns showed a slight decrease in efficiency at high pressure.³⁰A high temperature treatment is commonly carried out at above 600°C for preparation in a test tube, and up to 330°C for those in a capillary so as not to damage the polyimide coating of the fused silica capillary tube. A chemical modification of the silica surface is then carried out on-column by traditional silane chemistry. In the case of monolithic silica prepared in a test tube, batch modification prior to the cladding process is possible. Commercial octadecylsilylated (ODS)monolithic columns possess nearly maximum surface coverage with ODS groups, and are further endcapped. An improvement is still needed, because tailing of the peaks for amines has been reported.³¹

Advantages and Limitations of Monolithic Silica Columns

The monolithic silica columns currently available consist of silica skeletons of 0.5 – 2 μm diameter and through-pores of 1 – 8 μm. Those prepared in a test tube possess co- continuous spongy structures in contrast to polymer monolithic columns or silica particles commonly possessing corpuscular structures.^9,32,33Through-pores of a monolithic silica column are relatively large compared to those of a column packed with particles with the (through-pore size)/(skeleton size) ratio in a range of 1 – 4, much larger than that in a particle packed column, 0.25 – 0.4.³³The skeleton size and through-pore size can be varied independently to some extent. The presence of mesopores of 10 – 30 nm in silica skeletons was reported. The exclusion limit is not as clear as for a column packed with particles, while the presence of a break corresponding to a molecular weight of 20000 – 30000 in a molecular weight elution volume curve for styrene standard in size exclusion chromatography (SEC) was observed.³⁴This is similar to silica particles having ca. 10 nm pores.

The monolithic column possesses much larger through-pores than a particle-packed column. ³⁵Those prepared in a capillary column possess even higher porosity. High porosity leads to a high permeability or a low pressure drop, and a small skeleton size at a similar through-pore size can lead to higher column efficiency than what could be expected from the pressure drop. An additional advantage of a monolithic silica column is increased mechanical stability provided by the integrated network structure, which allows elution at mobile phase linear velocities greater than 10 mm/s. Particle- packed columns often show problems in the permeability and/or in the stability of their packed bed at such linear velocities A high porosity leads to a high permeability of a monolithic silica column, but the column is inevitably accompanied by a low phase ratio (stationary phase / mobile phase). The amount of silica existing in a monolithic column is less than in a particle-packed column, resulting in a smaller surface area for the monolithic silica, which in turn results in a short retention. The k values found with monolithic silica-C18 columns are then smaller than those with particle-packed columns by a factor of 2 – 5, depending on the total porosity, 80 – 85% for a conventional size column and 90 – 95% for a capillary type column.⁵The sample loading capacity, however, has been reported to be not as small as expected from the phase ratio.³⁶The loading capacity depends on the separation conditions, because the mobile phase can contribute to the capacity. For a second column in 2D-HPLC, small retentivity is a clear disadvantage, because large volume injections need highly retentive columns for maintaining the column efficiency. Other possible disadvantages of monolithic silica columns include small internal porosity, resulting in a small range for size exclusion mode elution, the labor-intensive preparation of individual columns with possible reproducibility problems, and limited availability.⁵

Chromatographic Properties of Monolithic Silica Columns

The high external porosity, uniformity of through-pores, and large (through-pore size)/(skeleton size) ratios can lead to a much higher permeability of monolithic silica columns than that of a column packed with particles of similar column efficiency.³⁶

Pemeability is defined by Equations (2)-(4). In Equations (2) and (3), u stands for the linear velocity of the mobile phase, η the solvent viscosity, L the column length, ΔP the column pressure drop, and t₀the column dead time. The relation between the particle diameter (d_p) and the specific permeability for a particle-packed column is described by the Kozeny–Carman equation (Equation (4), ε: interstitial porosity). The equation indicates that the size of, and the cross section occupied by the through-pores affect the permeability of a column. The external porosity is commonly 0.4 for a particle-packed column, while flow through porosities of 0.6 – 0.8 are found for monolithic silica columns that possess large through-pores compared to the skeleton size.⁵A Chromolith™ column provided by Merck shows specific permeability, K = 8 × 10^{– 14 m 2}, greater than that of a column packed with 5 μm particles by a factor of about two, and a plate height, H, of about 10 μm or smaller.³⁰Capillary columns with large through-pores show up to a 30-times higher permeability, ³⁷K = 1.3 × 10^–1²m². High permeability is an important feature of monolithic silica columns, particularly for high peak-capacity (number of peaks resolved per unit time) applications. While columns with a small domain size showed a high column efficiency and high pressure drop, those with a large domain size have shown low pressure drop, and are suitable for fast separation or for use in a long capillary column. Chromatographic efficiency can be defined in terms of plate height by equation (5) and(6).³⁸

Where H is plate height, D_mis the diffusion coefficient of a solute in the mobile phase; C_xis a coefficient for the contribution of each term; d_pis the particle diameter. A, B and C are the coefficients for eddy diffusion, longitudinal diffusion and resistance to mass transfer respectively. Equation 6 is typically known as the Knox equation. Alternatively,one can use the van Deemterequation.

The column efficiency at the nearly optimum linear velocity for the monolithic silica columns are similar to a column packed with 3 μm particles. However, the efficiency at high linear velocity is higher, and the pressure drop much smaller, with monolithic columns.³⁹The properties are presumably provided by small-sized silica skeletons, contributing to a small C-term in Eqs. (5) and(6).

Monolithic columns with a small domain size, especially those with high porosity prepared in the capillary, do not produce the performance expected from a small domain size. This is partly due to the increased inhomogeneity of the network structure of a monolithic silica column. The size and the distribution of through-pores might also be a factor. It has been suggested that an increase in the homogeneity of monolithic silica can increase the performance by several times.⁴⁰The silica skeletons prepared in a test tube are smoother and more homogeneous than those prepared in a capillary.^12,36The inhomogeneity and large-sized through-pores seem to explain the lower performance of monolithic silica than a particle-packed column at a high-speed range, particularly for capillary columns. Improvements of preparation conditions to achieve more homogeneous structures with smaller domains, particularly smaller through-pores, are in progress. Small columns, especially the capillary type, are sensitive to extra-column band broadening. Injection and detection as well as line connection must be carried out carefully so as not to reduce the performance of small-sized columns by minimizing extra-column effects.³⁰In a sense, each monolithiccolumn is unique, or produced as the product of a separate batch,because the columns are prepared one by one by a process including monolith formation, columnfabrication, and chemical modification.

Increase in Separation Speed by Using Monolithic Silica Columns

High speed separation is desirable when one has many samples to be analyzed in a limited period of time. An increase in the flow rate is possible for monolithic silica columns, because they show higher permeability and higher stability against fast flow. Such an example is shown in the separation of pharmaceuticals, such as atenolol, pindolol, and metoprolol, in less than 1 min by using a flow rate of 9 mL/min for 4.6 mm i.d. column with a length of 10 cm.⁴¹Many examplescan be found for an increase in the separation speed by using the high flow rate. However, an increase in the flow rate meansthe consumption of a large amount of solvents, unless smaller sized columns are used. Fast separations can also be achieved using columns of higher performance, such ascolumns packed with 1.7 – 2 μm particles.⁴²

Summary

At present monolithic columns commonly provide a plate height, H, of 8 – 20 μm at a linear velocity of 1 – 10 mm/s. The performance of monolithic silica columns available at present is similar to a column packed with 3 μm particles, but lower than the most advanced particle-packed columns, namely sub 2 μm particles columns, especially at high linear velocity. Although monolithic columns cannot provide the performance of columns packed with ≤ 2 μm particles at high-speed separations, the application of longmonolithic capillary columns under gradient conditions is attractive. Monolithic silicacolumns that can provide high efficiency at high-speed separation, however, are yet to be developed.

References

(1) Dittmann, M. M.; Rozing, G. P. Journal of chromatogr. A 1996, 744, 63-74.

(2) MacNair, J. E.; Lewis, K. C.; jorgenson, J. W. Anal. Chem. 1997, 69, 983-989.

(3) MacNair, J. E.; Patel, K. D.; Jorgenson, J. W. Anal. Chem. 1999, 71, 700-708.

(4) Minakuchi, H.; Nakanishi, K.; Soga, n.; Ishizuka, N.; Tanaka, N. Anal. Chem.1996, 68, 3498-3501.

(5) Kobayashi, H.; Ikegami, T.; Kimura, H.; Hara, T.; Tokuda, D.; Tanaka, N. Anal. Sci2006, 22, 491-501.

(6) Scev, F.; Frechet, J. M. J. Anal. Chem. 1992, 64, 820.

(7) Hjerten, S.; Liao, J.-L.; R, Z. Jouranl of chromatogr. A 1989, 473, 273-275.

(8) Field, M. S. Anal. Chem. 1996, 68, 2709-2715.

(9) Svec, F. J. Sep. Sci 2004, 27, 1419-1430.

(10) Tanaka, N.; Kobayashi, H.; Nakanishi, K.; Minakuchi, H.; Ishizuka, N. Anal. Chem. 2001, 73, 420A-429A.

(11) Tanaka, N.; Koyabashi, H.; Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Hosoya, K.; Ikegami, T. J. Chromatogr. A 2002, 965, 35-49.

(12) Hatsis, P.; Lucy, C. A. Analyst (Camb.) 2002,, 127,, 451-454. (13) Miyabe, K.; Guiochon, G. J. Sep. Sci 2004, 27, 853-873.

(14) Minakuchi, H.; Nakanishi, K.; Soga, N.; Ishizuka, N.; Tanaka, N. Anal. Chem. 1996, 68, 3498-3501.

(15) Li, W.; Fries, D. P.; Malik, A. J. Chromatogr., A 2004, 1044, 23-52.

(16) L., C.; T., M.; Anspach, J.; Colon, H. In Advances in Chromatography, 2003, 42,43-106.

(17) Nakanishi, K.; Minakuchi, H.; Nakanishi, K.; Tanaka, N. J. Sol-gel Sci. Technol.1998, 13.

(18) Judenstein, P.; Titman, J.; Stamm, M.; Schmidt, H. Chem. Mater. 1994, 6,127.

(19) Martin, J.; Hostickaa, B.; Lattimer, C.; Norris, P. M. J. Non-Cryst. Solids 2001, 228, 222-229.

(20) Kirkbir, F.; Murata, H.; Meyers, D.; Ray Chaudhuri, S. J. Non-Cryst. Solids 1998, 225, 14-18.

(21) Einarsrud, M.-A. J. Non-Cryst. Solids 1998, 225, 1-7.

(22) Einarsrud, M.-A.; Kirkedelen, M. B.; Nilsen, E.; Mortensen, K.; Samseth, J. J. Non-Cryst. Solids 1998, 231,10-16.

(23) Einarsrud, M.-A.; Nilsen, E.; Rigacci, A.; Pajonk, G. M.; Buathier, S.; Valette, D.; Durant, M.; Chevalier, B.; P., N.; Ehrburger-Dolle, F. J. Non-Cryst. Solids 2001, 285, 1-7.

(24) Scherer, G. W.; Smith, D. M. J. Non-Cryst. Solids 1995, 189, 197-211. (25) Siouffi, A. M. J. Chromatogr. A 2003, 1000, 801-808.

(26) Minakuchi, H.; Nakanishi, K.; Soga, N.; Tanaka, N. Anal. Chem. 1996, 68, 3498- 3501.

(27) Minakuchi, H.; Nakanishi, K.; Soga, N.; Tanaka, N. J. Chromatogr., A 1997, 797,121.

(28) Minakuchi, H.; Nakanishi, K.; Soga, N.; Tanaka, N. J. Chromatogr., A 1997, 762,135-146.

(29) Motokawa, M.; Shintani, Y.; Sawada, T.; Minakuchi, H.; Nakanishi, K. 29th International Symposium on High Performance Liquid Phase Separations and Related Techniques 2005, P4-2.

(30) Ikegami, T.; Dicks, E.; Kobayashi, H.; morisaka, H.; Tokuda, D.; Cabrera, K.; Hosoya, K.; Tanaka, N. J. Sep. Sci 2004, 27, 1292-1302.

(31) McCalley, D. V. J. Chromatogr., A 2002, 965, 51-64.

(32) Tanaka, N.; Kimata, K.; Araki, T.; Tsuchiya, H.; Hashizume, K. J. Chromatogr., A 1991, 544, 317-344.

(33) Unger, K. K. Porous Silica in Journal of Chromatography Library 1979, 16, 169.

(34) Al-Bokari, M.; Cherrak, D.; Guiochon, G. J. Chromatogr., A 2002, 975, 275-284.

(35) Motokawa, M.; Kobayashi, H.; Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Jinnai, H.; Hosoya, K.; Ikegami, T.; Tanaka, N. J. Chromatogr., A 2002, 961, 53-63.

(36) Leinweber, F. C.; Tallarek, U. J. Chromatogr., A 2003, 1006, 207-228.

(37) Ishizuka, N.; Kobayashi, H.; Minakuchi, H.; Nakanishi, K.; Hirao, K.; Hosoya, K.; Ikegami, T.; Tanaka, N. J. Chromatogr., A 2002, 960, 85-96.

(38) Kennedy, G. J.; Knox, J. H. J. Chromatogr. Sci. 1975, 13, 25.

(39) Tanaka, N.; Nagayama, H.; Kobayashi, H.; Ikegami, T.; Hosoya, K.; Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Cabrera, K.; Lubda, D. J. High Resolut. Chromatogr. 2000, 23, 111-116.

(40) Gzil, P.; Vervoort, N.; Baron, G. V.; Desmet, G. Anal. Chem. 2004, 76, 6707- 6718.

(41) Cabrera, K.; Lubda, D.; Eggenweiler, H.-M.; Minakuchi, H.; Nakanishi, K. J. High Resolut. Chromatogr. 2000, 23, 93-99.

(42) Novakova, L.; Matysova, L.; Solich, P. Talanta 2006, 68, 908-918.