silica gel-based monoliths prepared by the sol–gel method: facts and figures

Journal of Chromatography A, 1000 (2003) 801–818

www.elsevier.com/ locate/chroma

Review

S ilica gel-based monoliths prepared by the sol–gel method:facts and figures

*A.-M. Siouffi´UMR 6516,Universite d’ Aix Marseille 3, F-13397 Marseille cedex 20, France

Abstract

There is a great deal of interest in continuous beds as stationary phases for both HPLC and CEC. There are various waysto prepare monoliths, by polymerization of organic species or by polymerization of silicon alkoxides. The former method hasrecently been reviewed, while silica based monoliths are now commercially available. The purpose of this paper is to dealwith the problems associated with silica based monoliths. The most important problem is obviously the cracking and theshrinkage of the bed during drying. The second problem is monolith cladding. Much literature has been published but nodefinitive solution is available and thus a wide research area remains open. Monoliths are a compromise between loadability,permeability and mass transfer kinetics. Due to the better mass transfer properties of a monolithic skeleton over distinctparticles, high flow rates and high speed separations are possible. 2003 Elsevier Science B.V. All rights reserved.

Keywords: Reviews; Monolithic columns; Sol–gel; Columns; Stationary phases, LC; Stationary phases, CEC; Continuousbeds

Contents

8021 . Introduction ............................................................................................................................................................................8022 . Organic polymers ....................................................................................................................................................................8033 . Silica based monoliths .............................................................................................................................................................8043 .1. Sol–gel based monoliths..................................................................................................................................................8043 .1.1. Starting material..................................................................................................................................................

3 .1.2. Sol preparation and hydrolysis ............................................................................................................................. 8053 .1.3. Additives ............................................................................................................................................................ 8073 .1.4. Aging ................................................................................................................................................................. 8083 .1.5. Drying................................................................................................................................................................ 808

3 .2. Monolith information ...................................................................................................................................................... 8103 .3. Cladding ........................................................................................................................................................................ 811

4 . Particulate-alkoxide gels .......................................................................................................................................................... 8115 . Bonding.................................................................................................................................................................................. 8126 . Performances .......................................................................................................................................................................... 8127 . Conclusion.............................................................................................................................................................................. 815Acknowledgements...................................................................................................................................................................... 815References .................................................................................................................................................................................. 815

*Tel.: 133-4-9128-8345; fax:133-4-9128-8765.E-mail address: [email protected](A.-M. Siouffi).

0021-9673/03/$ – see front matter 2003 Elsevier Science B.V. All rights reserved.doi:10.1016/S0021-9673(03)00510-7

mailto:[email protected]

802 A.-M. Siouffi / J. Chromatogr. A 1000 (2003) 801–818

1 . Introduction often responsible for bubble formation during analy-sis. The presence of bubbles causes a drop in current

Column technology in HPLC has now reached a and the run must be interrupted. Problems are similarhigh standard and high reproducibility. Column in planar chromatography which may turn into planarperformance is expressed as the number of theoret- electrochromatography[4]. Preparation of a continu-ical plates per unit time,N /t, and the number of ous bed is obviously a convenient way to solve thetheoretical plates per unit pressure drop, i.e. the problem. The use of monolithic columns as anseparation impedance,DP 5 (DP/N)(t /N)(1 /h)5 alternative to packed capillaries has already been0

2h f, where DP is the pressure drop,N the plate reported in GC[5]. In the late 1990s there was greatcount, t the retention time of the inert solute,h the interest in monoliths devoted to either LC or CEC0

viscosity of the mobile phase,h the reduced plate and several extensive reviews appeared on mono-height andf the column resistance factor.E allows lithic columns[6–10].for comparison of different chromatographic sup- Four approaches have been utilized to prepareports of different geometry. For conventional HPLC continuous beds:columns with reduced plate height,h52, the small- (i) polymerization of an organic monomer withest value forE is 3000–4000. According to Darcy’s additives,law, the pressure drop is inversely proportional to the (ii) formation of a silica based network using asquare of the particle diameter. Consequently, small sol–gel process,particles have been used in a short column for fast (iii) fusing the porous particulate packing material inseparations and better column efficiency, based on a capillary by a sintering process, andsmaller eddy diffusion and shorter diffusion path (iv) organic hybrid materials.length. Increasing the flow rate to speed up the At present, the first two quoted procedures are theseparation process is limited by the resulting increas- most popular. Organic monoliths have been success-ing back pressure which might be associated with a fully produced on a laboratory scale but are not yetloss in efficiency. To reach higher and higher widely commercially available. Silica based mono-efficiencies one may use ultra high pressures as did liths seem to give more practical results and theJorgenson [1] or turn to electrochromatography number of applications is continuously increasing.(CEC) [2]. An increase in column performance can The fourth approach is being thoroughly studied byalso be achieved by increasing the column per- organic chemists and looks promising. Since themeability. Knox and Bristow[3] recognized the above quoted reviews[6–10] provide a lot ofpotential advantages of monolithic columns more information, the purpose of this paper is mainly tothan 30 years ago. Porous solid columns can exhibit focus on some aspects of silica based monoliths fromhigh performance if they have small size skeletons standpoint of sol–gel chemistry and give a shortand relatively large through-pores. overview of the variety of procedures. It is mainly

Most columns utilized in CEC or in micro HPLC devoted to those who want to be involved in a topicare prepared by packing fused-silica capillaries of which seems as rich as the bonding of alkyl moieties20–100mm I.D. with porous or non-porous function- on silica particles.alized particles of 2–10-mm diameter. The perform-ance and stability of such particulate packings great-ly depend on the retaining frits at column end 2 . Organic polymerskeeping the packing in place. These frits must retainthe packing particles and possess a porous structure Hjerten and Liao[11] were the first to report aso as to allow uniform mobile phase flow through the monolith from a compressed polyacrylamide gel withwhole cross-section. It is difficult to reproducibly the purpose of separating proteins. The idea was notprepare a highly permeable and mechanically strong totally new since Hansen and Sievers[12] hadend frit by sintering. Heat generated in the sintering published a preliminary paper in 1974.process partially destroys the stationary phase. End In their well documented review, Zou et al.[9]frits reduce the column separation efficiency and are distinguished: rods based on polystyrenes; rods based

A.-M. Siouffi / J. Chromatogr. A 1000 (2003) 801–818 803

on polymethacrylates; rods based on poly- This method of monolith preparation producesacrylamides; and monolithic molecularly imprinted phases with excellent pH stability but they maypolymers. shrink or swell when exposed to different mobile

In a recent review[13], Oberacher and Huber phases; another shortcoming is the mechanicalexamined the use of monolithic capillary columns stability of the packing with pressure. Due to con-prepared by copolymerization of styrene and di- vective flow these phases are well suited for proteinvinylbenzene inside a 200-mm I.D. fused-silica capil- and peptide analysis and a review on the topic haslary to analyse nucleic acids by HPLC–MS. appeared[33] where 200 000 plates per meter were

The polymer based monoliths are almost exclu- claimed.sively prepared by radical polymerization (thermal or Commercially available monolithic columns ofUV). For example, in a single one step procedure, a polymer based supports can be obtained from BIAsolution of acrylic monomers including piperazine as (Slovenia). They are highly cross-linked porous rigidcross linking agent was polymerized in a fused-silica monolithic polyglycidylmethacrylate styrene–di-capillary pretreated with 3-(trimethoxysilyl)propyl vinylbenzene polymers[34–38] and applicationsmethacrylate [14]. Preparation of a compressed have been published especially as regards preparativecontinuous bed for conventional and micro HPLC in purposes. Bio-Rad Laboratories and Sartoriusnormal-phase mode has been extensively studied by produce acrylamido based polymeric columnsFrechet and colleagues[15–25].They devised a rigid [39,40].polyacrylamide-based monolithic column containingbutyl methacrylate for separation of proteins byhydrophobic interaction chromatography[18]. They 3 . Silica based monolithsalso used a monolithic poly(styrene-co-divinyl ben-zene) [20]. Molded rigid monolithic columns con- Sol–gel technology has been utilized to createfined in untreated silica capillaries were prepared by surface bonded coatings. Engelhardt and Cunat-Wal-copolymerisation of mixtures of butyl methacrylate, ter[41] used sol–gel chemistry to create an olefinicethylene dimethacrylate, and 2-acrylamido-2-methyl- sublayer for subsequent polymerization with acryl-1 propanesulfonic acid in the presence of a porogenic amide to prepare polyamide coated open tubularsolvent [19]. The same group attached a chiral columns for capillary electrophoresis. Guo andmoiety to the surface[22,23]. The goal is to prepare Colon[42] bound retentive alkyl ligands in opena stationary phase for electrochromatography with tubular columns for LC and CEC through thisfurther application to separations on a chip[25]. The technique. Hayes and Malik[43] filled a capillaryreader is referred to the review published by Svec et with an alkoxide solution and performed hydrolysis.al. [22]. Recent papers on the preparation of meth- Condensation of tetrahydroxysilane with surfaceacrylate monolithic columns for capillary liquid silanols of the inner walls and derivatized fused-chromatography[26] or electrochromatography[27] silica surface with Ucon (a polyethylene oxide)have appeared. Hoegger and Freitag[28] treated the yields a claimed uniform coating. Historically, pa-capillary walls with 3-(trimethoxysilyl)propyl meth- pers on polymer based monoliths appeared beforeacrylate before performing in situ the those on silica based monoliths, but more researchmethacrylamide polymerization. Zhang and El Rassi has been carried out on inorganic matrix. The[29] devised a polyacrylamide monolithic phase with successful preparation of silica-based rod columnsbonded dodecyl ligands and sulfonic acid groups to was reported by Nakanishi and Soga in 1991[44,45].separate moderately polar solutes by CEC; peptide By combining the sol–gel reaction with phase sepa-analysis in capillary LC–MS was performed on ration and a subsequent solvent exchange treatment,surface alkylated polystyrene monolithic columns by double-pore silica gel (macropores and mesopores)Huang et al.[30]. An alternative approach is the use monoliths were prepared. The real breakthrough wasof ring opening metathesis polymerization (ROMP) the mastering of both the macropores and theas pioneered by Novak and Grubbs[31] and exten- mesopores. Silica rods thus prepared were releasedsively used by Buchmeiser and colleagues[7,32]. on the market by Merck. In situ derivatization was


performed with octadecyldimethyl-(N,N-diethyl and higher surface area than TEOS. Methyltriethoxyaminosilane) followed by end capping with hexa- silane (MTES) was advocated by Colon et al.[55]methyl disilazane. The problem of contact with the for the more flexible network produced from itswall had to be overcome and Merck produced a gelling. A mixture of TMOS and MTES was alsoPEEK column. The invention of monolithic silica promoted by Harreld et al.[56] and Ishizuka et al.based columns can be regarded as a major tech-[57]. Adding methyltrimethoxysilane to TMOS ornological change in column technology. dimethyldiethoxysilane to TEOS[58] increases the

hydrophobicity of the aerogel and shifts the pore size3 .1. Sol–gel based monoliths distribution towards larger pore radii[59]. Functional

groups on silicon atoms which impart the hydro-The basic sol–gel process involves the sequential phobic properties to the solids are atoms showing

hydrolysis and polycondensation of alkoxy silicon high electronegativity such as F (Fig. 1). One of thederivatives (e.g. tetraethyl orthosilicate (TEOS), or sol–gel precursors may be a mixture of alkoxy-tetramethyl orthosilicate (TMOS)), in aqueous acid silanes, one of them being a functionalized moietyor base with a mutual cosolvent. Non-silica aerogels such as aminopropyltriethoxysilane[60], (3-(2-are notably weak and fragile in monolithic form, aminoethylamino)propyltrimethoxysilane (EDAS)however the synthesis of high porosity monolithic [61] or N-octyltriethoxysilane as proposed by Rod-alumina aerogels has been recently described[46] riguez and Colon[62] or Constantin and Freitag[63]and hydrolysis of aluminium tri-sec-butoxide in who performed a copolymerization process. Guo etacetone was studied by NMR[47]. The first step is al. used a mixture of TEOS and polydimethylsilox-the formation of a sol. A sol is a colloidal suspension ane (M 54200) [64]. According to Mansur et al.w

of solid species in a liquid. The sol is converted into [60], a combination of different silane agents duringa gel through polycondensation of the sol leading to synthesis of the gels can satisfactorily be used toa wet structure. A gel is a biphasic medium from the produce materials containing specifically designedisotropic and progressive densification. It is a porous chemical functionalities. Einarsrud et al. studied thenetwork in a liquid. Hydrogels are formed from hydrolysis of polyethoxydisiloxane[65]. To prepareaqueous solutions whereas alcogels are formed from monolithic sol–gel columns with surface bondedalcoholic solutions. The process of gelation starts ligands, Malik and Hayes[66] utilized two sol–gelwith aggregation of particles or polymers into fractal precursors and a deactivation reagentclusters, then the clusters interpenetrate to some (phenyldimethylsilane) to produce a monolith.extent and finally link together to form an infinitenetwork [48]. Xerogels are dried by evaporation of

the liquid, and aerogels are usually obtained byremoval of solvents in hypercritical conditions. Insilica aerogels there is less than 2% silicon dioxideand 98% air.

3 .1.1. Starting materialSol–gel precursors are mainly silicon alkoxides

which can be obtained in a high degree of puritywhereas potassium silicate is very difficult to purify.Tetramethoxysilane (TMOS) undergoes a more rapidhydrolysis than tetraethoxysilane (TEOS) and wasextensively used by a Japanese team to producemonoliths [49–53]. Wagh et al.[54] compared theaerogels obtained from three different precursors: Fig. 1. Donor acceptor characteristics of alkoxysilanes. Arrow onTEOS, TMOS and PEDS (polyethoxydisiloxane) and left: increase of electron donor (acid) properties; arrow on right:claimed that TMOS yields narrow and uniform pores increase of electron acceptor (basic) properties.


3 .1.2. Sol preparation and hydrolysis drolysis of the alkoxide. By increasing water con-Numerous procedures are described in the litera- centration, chemical reactions are accelerated and

ture and there are many different methods (Table 1). gelation times decrease.It must be pointed out that most of these procedures Hydrolysis is performed with a catalyst. Threeare for other purposes than preparing monoliths for procedures are proposed: acid catalysis, basechromatography. In any case the starting silicon catalysis and two-step catalysis. It is generallyderivative is dissolved in a solvent, an aqueous agreed that under acid catalysis, entangled linear orsolution (acid or base) reacts and gelation is pro- randomly branched chains are formed in silica solsduced within a certain period of time. TEOS, TMOS whereas under base catalysis, it is easy to form aor other silicon alkoxides are not soluble in water. network of uniform particles in the sol. AcidThey are dissolved in alcohol to produce an homoge- catalysis is performed with HCl, H SO , HNO , HF,2 4 3

neous solution and water is then added. In their oxalic acid, formic acid, and acetic acid. A typicalpreparation of column frits, Zhuang and Huang[67] volume ratio is TEOS:C H OH:H O:acid2 5 2

used methylene chloride to dissolve MTES. Dai et al. 1:3:4:0.002. Gelation times are generally longer[68] promoted ionic liquids. Einarsrud et al.[65] when the pH of the sol is low. Kirkbir et al.[75]dissolved the acid catalyst in ethyl acetoacetate and observed that HF catalysis yields the highest poreadded this mixture to polyethoxydisiloxane. Hydrol- volume and pore diameter but the gel is weak. Theyysis of silicon alkoxides is a versatile technique also promoted double acid catalysis[76] with HFwhich can produce different materials according to and either HCl, HNO or H SO . Hydrolysis of3 2 4

the different parameters and to the acid catalysis or TMOS at 70–808C in an open vessel acceleratesbase catalysis reaction. It may produce silica spheres condensation and reduces the amount of liquid byor silica gels depending upon the experimental expelling excess methanol through distillation[77].conditions. Acid-catalyzed hydrolysis can be stimulated by

The critical point is the ratio Si:H O, i.e. the sonication[78]. Klemperer et al.[79] studied the2

proportions of TMOS (or TEOS) and water which products of the hydrolysis of Si(OMe) (3 mol) in4

yields different products. With concentrated am- methanol (14 mol) by water (1.8 mol) containingmonia, silica colloids (Stoeber colloids)[69] can hydrochloric acid (0.05 mol) after a period of 5 h.easily be prepared by the procedure described by The mixture of polysilicate species was then es-Nakanishi and Takammiya[70] and Osseo-Asare and terified with diazomethane to convert hydroxyl

29Arriagada[71]. Unger et al.[72] produced monodis- groups. Analysis with capillary GC and Si NMRpersed silica microspheres by hydrolysis in an al- evidenced a great number of species: (Si O(OMe) ),2 6

kaline solution. Karmakar et al.[73] observed that (Si O (OMe ) ), (Si 0 (OMe) ), (Si 0 (OMe) ),3 2 3 8 4 4 8 4 3 10

any acid–water mixture irrespective of the type of (Si 0 (OMe) ), (Si 0 (OMe) ), (Si 0 (OMe) ),5 5 10 5 5 10 5 4 12

acid (formic acid, acetic acid, propanoic acid, penta- (Si O (OMe) ), (Si 0 (OMe) ), (Si O (OMe) ),6 6 12 6 5 14 7 7 14

noic acid, hydrochloric acid, nitric acid, sulphuric (Si O (OMe) ), (Si O (OMe) ), and7 6 16 8 8 16

acid, orthophosphoric acid) can produce silica micro- (Si O (OMe) ).8 7 18

spheres from TEOS with a water:TEOS ratio in the Base catalysis usually involves dilute ammonia22range 1.0–1.5 for strong acids and 1.5–4.0 for weak 10M [80]. Under base catalysis it is easy to form

acids. They argued from FTIR spectra that four a network of uniform particles in the sol, and themembered siloxane rings (Si O ) are formed in the resulting pore volume is quite large. Under base4 12

heavier liquid. Nicolaon and Teichner[74] carried catalysis condensation kinetics are faster than hy-out a study of the influences of water and TMOS drolysis kinetics (Fig. 2). The two-step procedureconcentration when producing aerogels exhibiting was proposed by Brinker et al.[81] and used byhigh porosity and concluded that the water quantity others[82–84]. In a typical procedure[80] TMOS,should be 2–5-fold the stoichiometric proportion ethanol, H O and HCl are mixed in the volume ratio2

22with a TMOS concentration in the alcohol of|5– 4:4:1:1310 and this first solution is then mixed10%. The molar ratio of H O:Si(OR) in the sol with H O and ammonia in the volume ratio 9:1:432 4 2

22should be at least 2:1 to approach complete hy- 10 . The addition of NH OH as a second catalyst to4


T able 1Sol–gel preparation

Silicon derivative Solvent Water Catalyst Porogen Reference Remark

TEOS EtOHa1 3 4 HCl 0.02

1 3 4 HNO 0.023

1 3 4 H SO 0.02 [75]2 4

1 3 4 Oxalic acid 0.021 3 4 HF 0.02

TMOS MeOHa1 0.5–3 2–20 CH CO H 0.06–1M [74]3 2

NH OH 0.02–1M4

TMOS Ionic liquid HCO H [68]2a1 1 2

MTES CH Cl TFA [67] Addition of silica gel2 2a0.75 2 0.1

23TEOS EtOH HCl 12310 MC TEOS8

b107 ml 168 ml 60 ml [55]

TEOS/C TEOS HCl 0.11M [63]8b500 ml /282 ml 200 ml 93 ml 11 ml

26TEOS EtOH HCl 4.8310 M PEG [88]b1 g 1 ml 0.86ml

22TMOS EtOH HCl 10 M [59]a1 4 1

22TMOS MeOH NH OH 10 M [80]4a1 4 4

23TMOS MeOH NH OH 3.7310 M [92] Addition of TMOS4a1 14 4

PEDS Ethyl acetate HF 21 N [65]b1 1 0.02%

TMOS CH CO H 0.01M PEO [87] TMOS addition in3 2

45 ml then washing with CH CO /PEO3 2

NH OH4

TEOS EtOH NH OH 0.18M [93] Other silicon4a derivatives

TEOS EtOH NH OH (27%), HCl 1M [82] Double catalysis4b1 20 2.8/3.2

22 21TMOS EtOH HCl 10 M /NH OH 10 M [83] Double catalysis4b4 4 1

22TMOS CH CO H 10 M PEO [57] 9 g Urea3 2b0.45

a Molar ratio.b Volume ratio.


and then transforms into a viscoelastic gel. Analcoholic solution of triethoxysilane H(Si(OEt) )3

yields gelation without any catalyst reagent. Thehydrolysis does not induce the cleavage of Si–Hbonds[86].

The aggregation process yields mass fractal struc-tures. The fractal dimension is of paramount impor-tance. A small fractal dimension indicates a highlyporous sol–gel network; high values are indicative ofmicropores.

3 .1.3. AdditivesThe pore size and the mechanical properties of

gels can be varied with the addition of polyethyleneglycol (PEG) to the sol. PEG is a porogen which actsas a through-pore template and solubilizer of the

Fig. 2. Scheme of relative reaction kinetics of alkoxysilanessilane reagent. This has been done by Nakanishi etversus pH.al. [87], Judenstein et al.[88] and by Martin et al.

a sol initially catalyzed by HCl can increase the rate [89] who claimed that high concentrations of PEGof condensation reactions and reduce the gelation weaken the solid matrix whereas a small concen-time. A two-step procedure involving urea was tration of PEG strengthens the matrix. The pore sizerecently proposed by Ishizuka et al.[85]. Gelation of macroporous silica aerogel can be controlled bycan take place on wide time scale: seconds, minutes, varying the concentration of water soluble polymer.hours, days, or months. Gelation times can be Narrow and more uniform pore size distribution wasfollowed by measurement of viscosity (Fig. 3); the observed with the addition of glycerol which acts asgel exhibits a Newtonian viscosity in the initial state a drying additive since it prevents further reaction of

Fig. 3. TEOS gelation (double catalysis) followed by viscoelastic measurement (Contraves rheometer).


water [90]. Use of a surfactant in water–ethanol same group demonstrated that washing in a watersolution produces surfactant templated aerogels[91]. solution increases the permeability of the gels bySince the hydrophobic properties of TMOS based dissolution–reprecipitation[65] (Ostwald ripening).monoliths are increased by incorporating MTES as a Silylation removes Si–OH surface groups by pro-

´synthesis component[92], Alie et al. [93–95]consid- moting silica polycondensation resulting in a de-ered the incorporation of other silanes to be equiva- crease of pore size.lent to the incorporation of additives. Differences in We can notice that in many papers the gel is agedreactivity are obvious (Fig. 1). Consequently, the without indication of the procedure. Subsequentnucleation mechanism is related to the difference in drying is performed carefully to prevent cracking.reactivity between the main reagent (TMOS, TEOS,or TPOS) and the additive. They examined the 3 .1.5. Dryingbehavior of 3-(2-aminoethylamino)propyltrimethoxy- Drying of the gel is the critical step. Drying issilane (EDAS), 3-aminopropyltriethoxysilane (AES), governed by capillary pressure. During drying,3-aminopropyltrimethoxysilane (AMS), pro- shrinkage of the gel occurs due to capillary pressure.pyltrimethoxysilane (PMS) and 3(2-amino- It is the gradient in capillary pressure within theethylamino)propyltriethoxysilane (EDAES). The pores that leads to mechanical damage; the capillarymain reagent is either TMOS, TEOS, or TPOS. A tension developed during the drying may reach 100–nucleation mechanism by the additive takes place. 200 MPa[99] with consequent shrinkage and crack-When the additive contains methoxy groups, it reacts ing. Silica gel may decrease in volume by as muchfirst to form particles to which the main reagent as a factor 10 as it dries. The extent of shrinkage iscondenses in a later stage. When both the additive governed by the balance between capillary pressureand the main reagent contain an ethoxy group, there P , and modulus of the solid matrix:cis no nucleation mechanism by the additive. Amine

P 5 22g cosu /ror alkyl groups only influence gelation time. Addi- c (LV) h

tion of silica spheres (aerosil) in the solution beforewhereg is the surface tension of the pore liquid(LV)gelation strongly affects the aggregation mechanism;at the liquid vapor interface,u is the contact angle oftwo fractal structures coexist.Table 1 lists somethe liquid, andr is the hydraulic pore radius:htypical procedures. The main feature is the use of

very similar catalysts, e.g. HCl in acid catalysis and r 5 2V /Sh p pammonia in base catalysis.

whereV and S are pore volume and surface area,p p

3 .1.4. Aging respectively. They are critical parameters.Washing in H O/EtOH increases the liquid per- Three phases are present: liquid, solid and gas.2

meability of the solid part of the gel by a dissolu- When liquid evaporates from the pores of a gel andtion–reprecipitation process for silica. Aging in a when the contact angle,u, between the liquid and thesiloxane solution increases the stiffness and strengthnetwork is ,908, concave liquid–vapor menisciof the alcogel by adding new monomers to the silica form at the exterior surface of the body so that thenetwork and by improving the degree of siloxane liquid goes into tension and the solid network of thecross linking; conversely this step will reduce the gel is compressed. Wheng .g the liquid(SG) (SL)

permeability [96]. Einarsrud et al. [97,98] have ascends the pore by capillary rise, the solid /gasreported strengthening of the silica gels aged in interface is replaced by a solid / liquid interface, andTEOS, water, and ethanol solutions. Gels are washeda concave meniscus is observed. Ifu 5 0, or →0, thewith a 20% water–ethanol solution for 24 h at 608C, capillary liquid is stressed, and the solid network isthen an aging solution (70%TEOS:ethanol, v /v) is compressed. Wheng .g a convex meniscus is(SL) (SG)

used for 6–72 h at 708C followed by washing with observed, ifu 5p or →p, there is no compressionethanol and heptane. Data from small angle neutron of the gel (Fig. 4). The small pore size can inducescattering show only a slight increase in the volume fracture during drying due to enormous capillaryfractal dimension of the porous gel network. The forces.


negligible vapor pressure and can stabilize thenetwork. They are further removed using a solvent.

Efforts have been directed towards understandingdrying phenomena above the critical temperaturesand pressures but much less towards understand thebehavior of gels dried below the critical conditions.By drying at ambient pressure, the surface tensionbetween liquid and vapor cannot be avoided. Stresswithin the gel is proportional to the viscosity of thepore liquid and the drying rate and inversely propor-tional to the permeability of the wet gel. Theimportant parameters are: the pristine gel strength,the pore size of the wet gel, and the solvent used indrying. The small pore size can induce fractureduring drying due to enormous capillary forces. Thepore liquid is under enormous tension when the pore

˚size is smaller than 200 A. On the other hand, when˚the pore size is larger than 200 A, the shrinkage will

be less and cracking will be less likely to occur[59].Conversely in some cases small pore size gels (40A) are easier to dry than larger pore size gels, whichis explained by a theory of cavitation[99]. Manipu-lation of pore size distribution can be done throughthe drying solvent.

Kirkbir et al. [96] observed that a thresholdpressure exists above which the shrinkage is negli-gible. Below the threshold point, the capillary pres-sure overcomes the strength and the structure col-lapses. Above the threshold point, the strength isalways higher and the shrinkage becomes negligible.

Fig. 4. Formation of menisci during drying: concave or convex The threshold point depends on sol composition.meniscus on wetting angle.g( ), surface tension (liquid /vapor);LV Shrinkage is negligible for wet gels dried in iso-g( ), surface tension (solid /gas);g( ), surface tension (solid /SG SL

butanol and 2-pentanol at chamber pressures as lowliquid); u, wetting angle.as 1.8 MPa. The surface tension of isobutanol is twotimes higher than that of ethanol. As a result the gelswill experience twice the capillary pressure in iso-

There are many drying methods: replacing the butanol. However shrinkage is negligible. We foundalcohol by washing with liquid CO and drying that shrinkage is negligible for wet gels dried in2

above the critical point; replacing the alcohol with 2-pentanol at 1.8 MPa and 3008C. This can bewater followed by washing with, for example, ace- explained by silica solubility (Table 2), and structur-tone and liquid CO and supercritical drying at 358C al modification since the process may induce an2

and 8.5 MPa in an autoclave[100]; changing the increase in the network connectivity by formation ofpore fluid (solvent exchange) and drying; drying at new siloxane bonds. The variation in silica solubilityambient pressure[97]. may not be sufficient to explain the variation in

The surface of the solid network of silica gels is shrinkage with type of alcohols. Moreover,n-al-not modified by supercritical drying in CO since no cohols promote higherS , higher porosity, and2 p

surface tension exists and the extent of dissolution smaller pore diameter. Dieudonne et al.[80] claimedredeposition is not significant. Ionic liquids exhibit that thermal treatment of silica gels under alcohol in


T able 2 the silica structure stiffens to form a rigid structure toSilica solubility in n-alkanols a point at which the mercury can intrude into theMethanol (mg/ l) 1890 shrunken pores. Inverse size exclusion chromatog-Ethanol (mg/ l) 164 raphy (ISEC) is more informative since pore sizePropanol (mg/ l) 8 distribution is estimated from a calibration curve

drawn with molecular probes; the theory is described208 2058in Ref. [105]. ISEC allows the pore size distribution

Pore liquid g( ) dyne/cm P g( ) PLV v LV c to be examined. Al-Bokari et al.[106] determinedEthanol 22.4 3.2 2.5 0.7 the porosities of monolithic columns (from Merck);Isobutanol 22.6 1.8 7.2 1.9 the porosity of a conventional column is|0.622-Pentanol 24.0 1.0 6.1 1.6

whereas the observed porosity on the monolithicIsooctane 18.8 1.0 3.8 1.0column is |0.85. The internal porosity of a mono-

g( ), surface tension at different temperatures;P , estimatedLV c lithic column appears smaller than that of a conven-capillary pressure;P , vapor pressure.vtional column but the authors mentioned that it maydepend on the definition.

an autoclave induces textural transformations of the To obtain good chromatographic performances thesolid network on a nanoscopic scale. ratio (through-pore/skeleton size) is of primary

Smith et al.[101] developed a model to predict gel importance.shrinkage from which it turns out that the number of The bulk structure is studied mainly by scatteringprocess variables which can be utilized to control techniques.shrinkage is rather limited (pore fluid, drying rate, Fractal dimension is determined through the analy-and initial density), which modifies the gel structure. sis of SAXS data. The fractal dimension (usuallyAlaoui et al. [102] densified aerogels by isostatic found around 2) is the signature of the linkagecompression but making chromatographic columns between the polymeric clusters. The evolution withwith this technique looks difficult. Sglavo et al.[103] time of the SAXS intensity provides information ondried a gel under relative humidity for 150 days! the kinetics of aggregation of TEOS derived silica

sols; it gives the scattering exponent and the average3 .2. Monolith information radius of giration of the clusters. Ultra small angle

X-ray scattering and atomic force microscopy can be29The bulk density is measured from the weight to used to check densification. Si MAS NMR can be

volume ratio. Main parameters to control are: po- used to identify and characterize the early stages ofrosity, pore radius and specific surface. condensation reactions. These methods (SAX, NMR)

The surface area is measured by N adsorption. of investigation are difficult to apply to the study of2

The Kelvin equation for cylindrical pores establishes large monoliths and to the study of chromatographica relationship between pore radius and relative columns.pressure (P/Po) utilized in the N adsorption–de- Scanning Electron Microscopy (SEM) or field2

sorption technique. Estimates of specific pore vol- emission gun scanning electron microscopy are usedume are obtained from the amount of nitrogen to study the morphology of monoliths. The availableadsorbed by the sample in the range 0.994,P/Po, figures in the literature usually display a cross0.999. The total run time needed to measure a well section of the monolith which does not provideequilibrated isotherm increases with mesoporosity. information on the homogeneity of the whole

The mean pore radius,r, is calculated through skeleton. Tanaka et al.[10] recently displayed bothr 5 2V /S (BET). cross section and longitudinal SEM pictures ofp p

Mercury intrusion analyses can be performed, but monoliths. We display a SEM picture (Fig. 5) of aporosimetry data are affected by isostatic compres- longitudinal section of a Chromolith� exhibiting thesion of the monolith at pressures up to|80 MPa, a homogeneity of the skeleton size.

´problem discussed by Alie et al.[104]. Compression The gelation process can be studied by followingof the monolith causes constriction of the pores until the time evolution of the viscoelastic properties


octadecyldimethyl(3-(trimethoxysilyl)propyl) ammo-nium chloride, phenyldimethylsilane is the deactiva-tion reagent and trifluoroacetic acid is the selectedcatalyst. The fused-silica capillary is hydrothermallytreated prior to filling with the sol solution. Accord-ing to the method, the products of hydrolysis canundergo polycondensation: (i) between hydrolyzedproducts of the same original precursors; (ii) be-tween hydrolyzed products of two different originalprecursors; (iii) between the hydrolyzed products ofeither precursor with the silanol groups of the silicasurface.

Some authors display cross-sectional views ofmonoliths. In their frit making procedure, Zhang andHuang[67] showed that packing particles are tightlyfixed onto the capillary wall to form a mechanicallystable layer.

The commercial breakthrough came with silicarods from Merck [107–112] that turned toChromolith�. The monolith is prepared according tothe Nakanishi and Soga procedure[44,45] whichenables control of the bimodal pore structure (2-mmthrough-pores and 13-nm mesopores). The silicasorbent is then encased in a PEEK plastic cover andFig. 5. Two SEM pictures of Chromolith (longitudinal section),the silica sorbent is derivatized in situ. The PEEKperformed in the Centre Pluridisciplinaire de Micoscopie elec-

ˆtronique de Saint Jerome, Marseille. (a) Core of the monolith plastic cover is shrink wrapped onto the silica rods to(20-mm scale); (b) PEEK cladding (bottom of the picture). ensure that there is no void space between the silica

and the PEEK material.through dynamic oscillatory measurement with acontrolled stress rheometer (Fig. 3).

Aerogels obtained by alcohol supercritical drying 4 . Particulate-alkoxide gelsare hydrophobic as are those from TMOS–MTES;hydrophobic properties are determined from contact Horvath and colleagues[113] packed fused-silicaangle measurements. capillaries with porous C silica microspheres18

The three-dimensional structure is studied by laser which they subjected to thermal treatment at 3608Cscanning confocal microscopy. to produce a column with porous silica based

monolithic packing. After sintering, the monolithic3 .3. Cladding packing was reoctadecylated in situ. It is noteworthy

that Horvath and colleagues also produced a mono-It is necessary to encase the rod. Possible loss of lith prepared from silanized fused-silica capillaries of

contact with the wall and the subsequent by-pass of 75-mm I.D. by in situ copolymerization of di-eluent are problems which have to be overcome. vinylbenzene either with styrene or vinylbenzylMalik and Hayes[66] patented a method for the chloride in the presence of a suitable porogen[114].fabrication of monolithic beds embedded in a fused- Some investigators have synthesized gels by dis-silica capillary. The reagent system involves two persing SiO particles in alkoxide solutions. Toki et2

sol–gel precursors, a deactivation reagent, one or al.[115,116]prepared 15 cm315 cm30.5 cm glassmore solvents and a catalyst. The first sol–gel plates by sintering particulate-alkoxide gels dried atprecursor is a polymethoxysilane, the second isN- 60 8C for 10 days and found that porosity increased


with increasing SiO particle content. Dulay et al. 3-aminopropyltriethoxysilane side groups. Removal2

[117] prepared a monolith by immobilization of C of the aromatic cores created framework cavities in18

particles in a sol–gel matrix prepared in situ in a which the aminopropyl groups were anchored. Chencapillary tube with subsequent drying at high tem- et al.[125] grafted L-phenylalaninamide and otherperature. chiral selectors on the surface of the monolith

This procedure is very difficult to reproduce. Tang through available silanols. Kang et al.[126] used aet al. [118,119] filled a capillary tube with a CO coating procedure to prepare a Chirasil beta DEX.2

slurry of C silica particles, treated the material with The bonding of alkyl ligands must be performed18

propylsulfonic acid to modify the surface and bound after monolith preparation.the particles together using a sol–gel formationprocess (the alkoxide was in fact a mixture of TMOSand ethyltrimethoxysilane). Zhang and Huang[67] 6 . Performancesprepared an on column frit from a sol made withmethyltriethoxysilane, water, methylene chloride and Monoliths can be used in HPLC or CEC. We cantrifluoroacetic acid in which they suspended silica distinguish between commercially available columnsparticles. and those only described in papers. In Chromoliths,

To control the porosity of the organic based there are large through-pores which are comparablemonolithic polymers, solvents are used as porogenic to the interstitial voids of a particle packed columnagents. Use of silica beads as sacrificial material for and mesopores of the same size and volume as inthe templating of porosity eliminates the need for porous particles. Models of liquid flow throughporogenic solvent. Chirica and Remcho[120] pre- monolith have been devised by Liapis and colleaguespared such monolithic columns with 3-(trimethox- [127–129].ysilyl)propyl methacrylate which reacts on the As far as silica based monoliths are concerned,silanol groups of the wall while the methacrylate preliminary chromatographic data on silica rods weregroup is the anchor for the butyl methacrylate (or given by Minakuchi et al.[52], Ishizuka et al.[130],another monomer) to be synthesized in a polymeri- Cabrera and Lubda[131] who emphazised thezation reaction. separation of test solutes at very high flow rates.

A hybrid procedure was devised by the group of They insisted upon separation impedance and lowZare [121–123]. Methacryloxypropyltrimethoxy- pressure drop, allowing fast separations.silane (MPTMS) contains both methacrylate and For conventional HPLC column with reducedalkoxysilane groups, and is used to prepare a photo- plate heighth52.0 at a minimum andf5700–polymerized sol–gel in a single step reaction. A 1000, the smallest value for separation impedanceEmixture of MPTMS, hydrochloric acid and water is is 3000–4000. Great differences with monolithicstirred in the dark, toluene, which will act as a rods were demonstrated by Minakuchi et al.[52]porogen, is then added, and finally the photoinitiator. since a 7-fold decrease inE was observed. Impe-The solution is filled into a capillary and irradiated. dances as low as 300–700 were observed.Fig. 6It has been shown that the hydrochloric acid con- compares the separation impedance of conventionalcentration is critical for the rigidity of the columns columns versus Chromolith. Recent results withand the ratio of monomer solution to porogen fused-silica capillaries[136] are impressive since acontrols the through-pore size and the surface area value of 100 was obtained. The coupling of manyand thus the separation capabilities. columns in series is possible.

Bidlingmaier et al.[132] displayed the Van Deem-ter curve of a Chromolith prototype column; they

5 . Bonding obtained a 13-mm H at a linear velocity of 2.8 mmmin21 21s corresponding to a flow rate of 3 ml min .

210 2During the sol–gel polymerization process, Davis Permeability is 1.0310 cm which is six timesand Katz[124] incorporated imprint molecules con- lower than that observed with a conventional col-sisting of aromatic rings carrying one, two or three umn.


Fig. 6. Comparison of the separation impedance of conventional columns versus Chromolith.m, Chromolith; ♦ , 3534.6 mm column, 3mm; *, 5034.6 mm column, 3mm; j, 5034.0 mm column, 3mm (courtesy of Merck KgA).

Mc Calley [133] compared conventional micropar- monolith gave similar values ofH for benzene tomin

ticulate packed columns and a monolith from Merck the 5-mm phase.H remains practically constantmin21(Chromolith). He used the Dorsey Foley equation to over the range 2–5 ml min due to the favorable

determine column efficiency. He found that the diffusion characteristics of the monolith; as a conse-

T able 3HPLC performances of C monolithic silica columns18

21H (mm) U (mm s ) Column diameter (mm) Solute Ref.min opt

13 2.8 4.6 Alkylbenzenes [132]10 2.6 4.6 Benzene [133]9 1.5 4.6 Benzene [134]7.5–13 0.8–0.5 4.6 Amylbenzene [135]8–10 1–2 4.6–7.0 Hexylbenzene [10]

12 0 0.05 Alkylbenzenes [136]

Fig. 7. Separation of alkyl benzenes at different flow rates, Chromolith C column (courtesy of Merck KgA).18


21Fig. 8. Separation of some psychotropic drugs on Chromolith C . Chromolith Performance column, 10034.6 mm, flow rate 3.2 ml min .1824Mobile phase: phosphate buffer 10M, pH 3.2:acetonitrile (65:35) v /v. Elution order: 9-OH risperidone, risperidone, reduced haloperidol,

haloperidol, desmethyl cyamemazine, cyamemazine, desmethyllevomepromazine, levomepromazine, desmethyl chlorpromazine, chlor-promazine.


21quence at high flow rate (5 ml min ),H is almost medium.d can be approximated by regression ofdisp

3.5 times lower than on the 5-mm phase. These data theC term of the Van Deemter equation with a set ofare in accordance with data from Bidlingmaier but solutes (a homologous series is well suited). Theare different from those published by Sinz and hydraulic permeability can be characterized by anCabrera [134] and those published by Kele and equivalent particle diameterd which has beenperm

Guiochon [135]. With two columns of different found to be|15 mm.diameter (4.6–7.0 mm), Tanaka et al.[10] observedsimilar plate heights (Table 3). In capillary HPLC,Ishizuka et al.[136] reported 100 000 plates on a

7 . Conclusion130 cm long column,H is 12 mm and themin

separation impedance is as low as 100–200.Since the pioneering work of Tswett, liquid chro-Band spreading does not increase with velocity

matography columns are filled with particles thewhen high interparticle Peclet number is observed.diameter of which has decreased with the years toKele and Guiochon extensively studied the perform-withstand the high pressures required to operate theances of six Chromoliths which they found verycolumns at optimum velocities.similar, thus emphasizing the reproducibility of the

The advent of continuous beds represents a newmanufacturing process. They observed a very lowCgeneration and will be the major topic in future.term when the solute is unretained and a higher valueConsidering silica based monoliths it will lead to(C52 to 3) for retained solutes. Ishizuka et al.[136]continuous improvements. Much information can bealso observed the dependence of plate height onobtained from papers dealing with the fabrication ofretention factor. Values ofA andC terms increase assilica aerogels; it is interesting to watch papers stillthe solute is more retained. They performed mea-appearing on C bonded silica particles!18surements of the peak efficiencies from the peak

width at half height. Measured permeabilities lie in210 2the range 7.1–9.2310 cm which is more than

twice as low as for conventional columns. In this A cknowledgementsmode, a tremendous increase in flow rate is possible,as demonstrated byFig. 7 which shows the sepa- I am indebted to Dr D. Lubda (Merck) for reading

21ration of alkyl benzenes at 6 ml min . Flow rates the manuscript and for valuable suggestions.21up to 9 ml min are possible but careful attention

must be paid to the instrument and especially thedetector. Due to the reduction of back pressure it is

R eferencespossible to use high percentages of buffer in thebinary mobile phase to perform fast analysis of

[1] J . Jorgenson, lecture presented at HPLC 1998, Saint Louis,solutes exhibiting a wide range of retention withoutMI, USA; L. Tolley, J.W. Jorgenson, M.A. Moseley, Anal.

performing a gradient.Fig. 8 shows the separation of Chem. 73 (2001) 2985.some psychotropic drugs together with the main [2] M .M. Dittmann, G.P. Rozing, J. Chromatogr. A 744 (1996)

63.metabolites. New possibilities are offered in prepara-[3] J .H. Knox, P.A. Bristow, Chromatographia 10 (1977) 279.tive liquid chromatography: a separation of dia-[4] D . Nurok, M.C. Frost, D.M. Chenowieth, J. Chromatogr. Astereomers has been published[137].

903 (2000) 211.The A term is quite large compared to convention- [5] F .D. Hieman, R.E. Sievers, G.G. Hess, W.D. Ross, Anal.

al columns. Leinweber et al.[138] assumed an Chem. 45 (1973) 1126.[6] F . Svec, E.C. Peters, D. Sykora, J.M.J. Frechet, J. Chroma-independent contribution from mechanical and non-

togr. A 887 (2000) 3.mechanical dispersion mechanisms. They evidenced[7] M .R. Buchmeiser, J. Chromatogr. A 918 (2001) 233.diffusion in mesopores and calculated an equivalent[8] N . Tanaka, H. Koyabashi, K. Nakanishi, H. Minakuchi, N.

dispersion particle diameterd . d can be re-disp disp Ishizuka, Anal. Chem. 74 (2001) 421A.garded as an equivalent length scale which solute [9] H . Zou, X. Huang, M. Ye, Q. Luo, J. Chromatogr. A 954molecules travel in stagnant zones of the porous (2002) 5.


[10] N . Tanaka, H. Koyabashi, N. Ishizuka, H. Minakuchi, K. [42] Y . Guo, L.A. Colon, Anal. Chem. 67 (1995) 2511.[43] J .D. Hayes, A. Malik, J. Chromatogr. B 695 (1997) 3.Nakanishi, K. Hosoya, T. Ikegami, J. Chromatogr. A 965[44] K . Nakanishi, N. Soga, J. Am. Ceram. Soc. 74 (1991) 2518.(2002) 35.[45] K . Nakanishi, N. Soga, J. Non-Cryst. Solids 139 (1992) 1.[11] S . Hjerten, J.-L. Liao, J. Chromatogr. 457 (1988) 333.[46] J .F. Poco, J.H. Satcher, L.W. Hrubesh, J. Non-Cryst. Solids[12] L .C. Hansen, R.E. Sievers, J. Chromatogr. 99 (1974) 23.

288 (2001) 57.[13] H . Oberacher, C.G. Huber, TrAC 21 (2002) 166.[47] S . Kureti, W. Weiweiler, J. Non-Cryst. Solids 303 (2002)[14] A . Maruska, C. Ericson, A. Vegvari, S. Hjerten, J. Chroma-

253.togr. A 837 (1999) 25.[48] C .J. Brinker, G.W. Scherrer, Sol-Gel Science. The Physics[15] F . Svec, J.M. Frechet, Anal. Chem. 64 (1992) 820.

and Chemistry of Sol-Gel Processing, Academic Press, New[16] Q .C. Wang, F. Svec, J.M. Frechet, J. Chromatogr. A 669York, 1990.(1994) 230.

[49] K . Nakanishi, H. Minakuchi, N. Soga, N. Tanaka, J. Sol-Gel.[17] M . Petro, F. Svec, J.M. Frechet, J. Chromatogr. A 752Sci. Technol. 13 (1998) 163.(1996) 59.

[50] H . Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N.[18] S . Xie, F. Svec, J.M. Frechet, J. Chromatogr. A 775 (1997)

Tanaka, Anal. Chem. 68 (1996) 3498.65.

[51] K . Nakanishi, H. Minakuchi, N. Soga, N. Tanaka, J. Sol-Gel[19] E .C. Peters, M. Petro, F. Svec, J.M. Frechet, Anal. Chem. 70 Sci. Technol. 8 (1997) 547.

(1998) 2288. [52] H . Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N.[20] S . Xie, R.W. Allington, F. Svec, J.M. Frechet, J. Chromatogr. Tanaka, J. Chromatogr. A 762 (1997) 121.

A 865 (1999) 169. [53] N . Ishizuka, H. Minakuchi, K. Nakanishi, N. Soga, K.[21] D . Sykora, F. Svec, J.M. Frechet, J. Chromatogr. A 862 Hosoya, N. Tanaka, J. Chromatogr. A 797 (1998) 133.

(1999) 297. [54] P .B. Wagh, R. Begag, G.M. Pajonk, A. Venkasteswara Rao,[22] F . Svec, E.C. Peters, D. Sykora, J.M. Frechet, J. Chromatogr. D. Haranath, Mater. Chem. Phys. 57 (1999) 214.

A 887 (2000) 3. [55] L .A. Colon, Y. Guo, A. Fermier, Anal. Chem. 68 (1996)[23] M . Lammerhofer, E.C. Peters, C. Yu, F. Svec, J.M. Frechet, 2753.

W. Lindner, Anal. Chem. 72 (2000) 4614. [56] J .H. Harreld, T. Ebina, N. Tsubo, G. Stucky, J. Non-Cryst.[24] M . Lammerhofer, E.C. Peters, F. Svec, J.M. Frechet, Anal. Solids 298 (2002) 241.

Chem. 72 (2000) 4263. [57] N . Ishizuka, H. Minakuchi, K. Nakanishi, N. Soga, K.[25] C . Yu, F. Svec, J.M. Frechet, Electrophoresis 21 (2000) 120. Hosoya, N. Tanaka, Anal. Chem. 72 (2000) 1275.[26] P . Coufal, M. Cihak, J. Suchankova, E. Tesarova, Z. [58] B . Cao, C. Zhu, J. Non-Cryst. Solids 246 (1999) 34.

Bosakova, K. Stulik, J. Chromatogr. A 946 (2002) 99. [59] A . Venkateswara Rao, D. Haranath, Micropor. Mesopor.[27] R . Wu, H. Zou, M. Ye, Z. Lei, J. Ni, Anal. Chem. 73 (2001) Mater. 30 (1999) 267.

4918. [60] H .S. Mansur, W.L. Vasconcellos, R.S. Lenza, R.L. Orefice,[28] D . Hoegger, R. Freitag, J. Chromatogr. A 914 (2001) 211. E.F. Reis, Z.P. Lobato, J. Non-Cryst. Solids 273 (2000) 109.[29] M . Zhang, Z. El Rassi, Electrophoresis 22 (2001) 2593. [61] C . Alie, R. Pirard, A.J. Lecloux, J.P. Pirard, J. Non-Cryst.[30] X . Huang, S. Zhang, G.A. Schultz, J.D. Henion, Anal. Chem. Solids 246 (1999) 216.

74 (2002) 2336. [62] S .A. Rodriguez, L.A. Colon, Anal. Chim. Acta 397 (1999)[31] B .M. Novak, R.H. Grubbs, J. Am. Chem. Soc. 110 (1988) 207.

7542. [63] S . Constantin, R. Freitag, J. Chromatogr. A 887 (2000) 253.[32] B . Mayr, R. Tessadri, E. Post, M. Buchmeiser, Anal. Chem. [64] L . Guo, J. Hycon Lee, G. Beaucage, J. Non-Cryst. Solids

73 (2001) 4071, and references therein. 243 (1999) 61.[33] D . Josic, A. Buchacher, A. Jungbauer, J. Chromatogr. B 752 [65] M .-A. Einarsrud, E. Nilsen, A. Rigacci, G.M. Pajonk, S.

(2001) 191. Buathier, D. Valette, M. Durant, B. Chevalier, P. Nitz, F.[34] A . Podgornik, M. Barut, A. Strancar, D. Josicacute, T. Ehrburger-Dolle, J. Non-Cryst. Solids 285 (2001) 1.

Koloini, Anal. Chem. 72 (2000) 5693. [66] A . Malik, J.D. Hayes, PCT/USO1/04271, August 16th,[35] M . Vodopivec, A. Podgornik, A. Berovic, A. Strancar, J. 2001.

Chromatogr. Sci. 38 (2000) 489. [67] X . Zhang, S. Huang, J. Chromatogr. A 910 (2001) 13.[36] K . Branovicacute, A. Buchacher, M. Barut, S. Jaksa, M. [68] S . Dai, Y.H. Ju, H.J. Gao, J.S. Lin, S.J. Pennycook, C.E.

Zigon, A. Strancar, J. Chromatogr. A 903 (2000) 21. Barnes, Chem. Commun. (2000) 243.[37] M . Merhar, A. Podgornik, M. Barut, S. Jaksa, M. Zigon, A. ¨[69] W . Stober, A. Fink, E. Bohn, J. Colloid. Interf. Sci. 26

Strancar, J. Liq. Chromatogr. Relat. Tech. 24 (2001) 2429. (1968) 251.[38] A . Strancar, A. Podgornik, M. Barut, R. Necina, Adv. [70] K . Nakanishi, Y. Takammiya, J. Ceram. Soc. Jpn. 96 (1988)

Biochem. Eng. Biotechnol. 76 (2002) 49. 719.[39] A .K. Hunter, G. Carta, J. Chromatogr. A 897 (2000) 65. [71] K . Osseo-Asare, F.J. Arriagada, Colloids Surf. 50 (1990)

321.[40] A .K. Hunter, G. Carta, J. Chromatogr. A 930 (2001) 79.[72] K .K. Unger, J. Schick-Kalb, B. Straube, Colloid Polym. Sci.[41] H . Engelhardt, M.A. Cunat-Walter, J. Chromatogr. A 716

253 (1975) 658.(1995) 27.


[73] B . Karmakar, G. De, D. Ganguli, J. Non-Cryst. Solids 272 [100] G .W. Scherer, J. Non-Cryst. Solids 215 (1997) 155.(2000) 11. [101] D .M. Smith, D. Stein, J.M. Anderson, W. Ackermn, J.

[74] G .A. Nicolaon, S.J. Teichner, Bull. Soc. Chim. (1968) 1906. Non-Cryst. Solids 186 (1995) 104.[75] F . Kirkbir, H. Murata, D. Meyers, S. Ray Chaudhuri, A. [102] A .H. Alaoui, T. Woignier, J. Phalippou, G.W. Scherer, J.

Sarkkar, J. Sol-Gel Sci. Technol. 6 (1996) 203. Sol-Gel Sci. Technol. 13 (1998) 365.[76] H . Murata, D.E. Meyers, F. Kirkbir, S. Ray Chaudhuri, A. ´[103] V .M. Sglavo, S. Dire, M. Ferrari, J. Mater. Res. 14 (1999)

Sarkar, in: J.D. Mackenzie (Ed.), Sol-Gel Optics, Vol. III, 2100.SPIE – The International Society for Optical Engineering, ´[104] C . Alie, R. Pirard, J.-P. Pirard, J. Non-Cryst. Solids 292Washington, 1994, Vol. 2288, p. 709. (2001) 138.

[77] T .R. Bryans, V.L. Brawner, E.L. Quitevis, J. Sol-Gel Sci. [105] P . DePhilipps, A.M. Lenhoff, J. Chromatogr. A 883 (2000)Technol. 17 (2000) 211. 39.

[78] D .R. Vollet, D.A. Donatti, A. Ibanez Ruiz, J. Non-Cryst. [106] M . Al-Bokari, D. Cherrak, G. Guiochon, J. Chromatogr. ASolids 288 (2001) 81. 975 (2002) 275.

[79] V . Klemperer, V.V. Mainz, S.D. Ramamurthi, F.S. Rosenberg, [107] K . Cabrera, G. Wieland, D. Lubda, K. Nakanishi, N. Soga,in: C.J. Brinker, D.E. Clark, D.R. Ulrich (Eds.), Better

H. Minakuchi, K.K. Unger, TrAC 17 (1998) 50.Ceramics Through Chemistry, Vol. III, North Holland, New

[108] K . Cabrera, D. Lubda, H. Minakuchi, K. Nakanishi, J. HighYork, 1988.

Resolut. Chromatogr. Commun. 23 (2000) 81.[80] P . Dieudonne, A. Hafidi Alaoui, P. Delord, J. Phalippou, J.[109] K . Nakanishi, K. Cabrera, D. Lubda, J. High Resolut.Non-Cryst. Solids 262 (2000) 155.

Chromatogr. 23 (2000) 111.[81] C .J. Brinker, K.D. Keefer, D.W. Schaefer, R.A. Assink, B.D.[110] P . Zoellner, A. Leitner, D. Lubda, K. Cabrera, W. Lindner,Kay, C.S. Ashley, J. Non-Cryst. Solids 63 (1984) 45.

Chromatographia 52 (2000) 818.[82] G . Wu, J. Wang, J. Shen, T. Yang, Q. Zhang, B. Zhou, Z.[111] D . Lubda, K. Cabrera, W. Kraas, C. Schaefer, D. Cunnin-Deng, F. Bin, D. Zhou, F. Zhang, J. Non-Cryst. Solids 275

gham, LC?GC Int. 14 (2001) 730.(2000) 169.[112] D . Lubda, K. Cabrera, H. Minakuchi, K. Nakanishi, J.[83] P . Mezza, J. Phalippou, R. Sempere, J. Non-Cryst. Solids 243

Sol-Gel Sci. Technol. 23 (2002) 185.(1999) 75.´[113] R . Asiae, X. Huang, D. Farnan, Cs. Horvath, J. Chromatogr.[84] A . Hafidi Alaoui, T. Woignier, J. Phalippou, G.W. Scherer, J.

A 806 (1998) 251.Sol-Gel Sci. Technol. 13 (1998) 365.´[114] I . Gusev, X. Huang, Cs. Horvath, J. Chromatogr. A 855[85] N . Ishizuka, H. Minakuchi, K. Nakanishi, K. Hirao, N.

(1999) 273.Tanaka, Colloids Surfaces A 187–188 (2001) 273.[115] M . Toki, S. Miyashita, T. Takeuchi, S. Kanbe, A. Kochi, J.[86] M . Pauthe, F. Despetis, J. Phalippou, J. Non-Cryst. Solids

Non-Cryst. Solids 100 (1988) 479.155 (1993) 110.[116] M . Toki, T. Takeuchi, S. Miyashita, S. Kanabe, J. Mater.[87] K . Nakanishi, H. Minakuchi, N. Soga, N. Tanaka, J. Sol-Gel

Sci. 27 (1992) 2857.Sci. Technol. 13 (1998) 163.[117] M .T. Dulay, R.P. Kulkarni, R.N. Zare, Anal. Chem. 70[88] P . Judenstein, J. Titman, M. Stamm, H. Schmidt, Chem.

(1998) 5103.Mater. 6 (1994) 127.[118] Q . Tang, B. Xin, M.L. Lee, J. Chromatogr. A 837 (1999)[89] J . Martin, B. Hosticka, C. Lattimer, P.M. Norris, J. Non-

35.Cryst. Solids 288 (2001) 222.[119] Q . Tang, N. Wu, M.L. Lee, J. Microcol. Sep. 12 (2000) 6.[90] A . Venkateswara Rao, M.M. Kulkarni, Mater. Chem. Phys.[120] G .S. Chirica, V.T. Remcho, J. Chromatogr. A 924 (2001)77 (2003) 819.

223.[91] M .T. Anderson, P.S. Sawyer, T. Rieker, Micropor. Mesopor.[121] M .T. Dulay, J.P. Quirino, B.D. Bennett, M. Kato, R.N.Mater. 20 (1998) 53.

Zare, Anal. Chem. 73 (2001) 3921.[92] A . Venkateswara Rao, M.M. Kulkarni, Mater. Res. Bull. 306[122] M . Kato, M.T. Dulay, B.D. Bennett, J.P. Quirino, R.N.(2002) 1667.

Zare, J. Chromatogr. A 924 (2001) 187.´[93] C . Alie, F. Ferauche, R. Pirard, A.-J. Lecloux, J.-P. Pirard, J.[123] M . Kato, K. Sakai-Kato, T. Toyo’oka, M.T. Dulay, J.P.Non-Cryst. Solids 289 (2001) 88.

Quirino, B.B. Bennett, R.N. Zare, J. Chromatogr. A 961´[94] C . Alie, R. Pirard, A.-J. Lecloux, J.-P. Pirard, J. Non-Cryst.(2002) 45.Solids 285 (2001) 135.

[124] M .E. Davies, A. Katz, Nature 403 (2000) 286.´[95] C . Alie, R. Pirard, J.-P. Pirard, J. Non-Cryst. Solids 301[125] Z . Chen, K. Uchiyama, T. Hobo, J. Chromatogr. A 942(2002) 304.

(2002) 83.[96] F . Kirkbir, H. Murata, D. Meyers, S. Ray Chaudhuri, J.[126] J . Kang, D. Wistuba, V. Schurig, Electrophoresis 23 (2002)Non-Cryst. Solids 225 (1998) 14.

1116.[97] M .-A. Einarsrud, J. Non-Cryst. Solids 225 (1998) 1.[127] J .J. Meyers, A.I. Liapis, J. Chromatogr. A 711 (1995) 197.[98] M .-A. Einarsrud, M.B. Kirkedelen, E. Nilsen, K. Mortensen,[128] J .J. Meyers, A.I. Liapis, J. Chromatogr. A 852 (1999) 3.J. Samseth, J. Non-Cryst. Solids 231 (1998) 10.

[99] G .W. Scherer, D.M. Smith, J. Non-Cryst. Solids 189 (1995) [129] A .I. Liapis, J.J. Meyers, O.K. Crosser, J. Chromatogr. A197. 865 (1999) 13.


[130] N . Ishizuka, H. Minakuchi, K. Nakanishi, N. Soga, K. [134] K . Sinz, K. Cabrera, Int. Labmate 25 (7) (2001) 17.Hosoya, N. Tanaka, J. High. Resolut. Chromatogr. 21 [135] M . Kele, G. Guiochon, J. Chromatogr. A 960 (2002) 19.(1998) 477. [136] N . Ishizuka, H. Koyabashi, H. Minakuchi, K. Nakanishi, K.

[131] K . Cabrera, D. Lubda, Poster presented at HPLC 1999, Hirao, K. Hosoya, T. Ikegami, N. Tanaka, J. Chromatogr. AGrenada, Spain. 960 (2002) 85.

[132] B . Bidlingmaier, K.K. Unger, N. Von Doehren, J. Chroma- [137] M . Schulte, J. Dingenen, J. Chromatogr. A 923 (2001) 17.togr. A 832 (1999) 11. [138] F .C. Leinweber, D. Lubda, K. Cabrera, U. Tallarek, Anal.

[133] D .V. Mc Calley, J. Chromatogr. A 965 (2002) 51. Chem. 74 (2002) 2470.

silica gel-based monoliths prepared by the sol–gel method: facts and figures

Documents