polyhedral oligomeric silsesquioxane as a cross-linker for preparation of inorganic−organic hybrid...

Polyhedral Oligomeric Silsesquioxane as aCross-linker for Preparation of Inorganic-OrganicHybrid Monolithic Columns

Minghuo Wu,†,‡ Ren’an Wu,*,† Ruibing Li,†,‡ Hongqiang Qin,†,‡ Jing Dong,† Zhenbin Zhang,†,‡ andHanfa Zou*,†

CAS Key Laboratory of Separation Sciences for Analytical Chemistry, National Chromatographic R & A Center, DalianInstitute of Chemical Physics, Chinese Academy of Sciences (CAS), Dalian 116023, China, and Graduate School ofChinese Academy of Sciences, Beijing 100049, China

An inorganic-organic hybrid monolithic capillary col-umn was synthesized via thermal free radical copolym-erization within the confines of a capillary using apolyhedral oligomeric silsesquioxane (POSS) reagentas the inorganic-organic hybrid cross-linker and asynthesized long carbon chain quaternary ammoniummethacrylate of N-(2-(methacryloyloxy)ethyl)-dimethy-loctadecylammonium bromide (MDOAB) as the organicmonomer. The preparation process was as simple aspure organic polymer-based monolithic columns in-stead of using POSS as the nanosized inorganic-organichybrid blocks (cross-linker) of the monolithic matrix.The pore properties and permeability could be tunedby the composition of the polymerization mixture. Thecharacterization and evaluation results indicated thatthe synthesized MDOA-POSS hybrid monolith pos-sessed the merits of organic polymer-based monolithsand silica-based monoliths with good mechanical andpH (pH 1-11) stabilities, which may be attributed tothe incorporation of the rigid nanosized silica core ofPOSS. Column efficiencies of 223 000 and 50 000N/m were observed in capillary electro-driven chroma-tography (CEC) and µ-HPLC, respectively. Peptides andstandard proteins were baseline separated by thishybrid column in CEC and µ-HPLC, respectively, aswell. The separation of bovine serum albumin (BSA)tryptic digest was also attempted to show its potentialapplication in proteome analysis.

Monolithic columns can be described as the integratedcontinuous porous separation media for separation sciences. Inthe past decade, monolithic columns, the novel state-of-the-artstationary phases, have been given comprehensive attention andapplied widely in microscale chromatographic separations, suchas capillary liquid chromatography (CLC), capillary electro-drivenchromatography (CEC), and microfluidic devices. Microscalemonolithic columns are usually prepared within the confines of

capillaries, and with no need of supporting frits. Monolithiccolumns possess advantages that include easy preparation,versatility in surface modification, great permeability, and goodpeak capacity. These unique merits have made monolithiccolumns the attractive alternative to the packed and open-tubularcolumns in the analytical separation sciences.1-4 In recent years,with the rapid development of nanoscale chromatographic separa-tion systems coupled to mass spectrometry, the use of capillarymonolithic columns have emerged as a promising choice for theseparation of complex biological samples to provide a lowerbackpressure drop, better column stability, and better resolutionand sensitivity.5-7

On the basis of the chemical nature of monoliths, monolithiccolumns can be mainly classified into organic polymer-based andsilica-based monolithic columns.8,9 However, the mechanical andsolvent instability of polymer-based monoliths, and the pHsensitivity of silica-based monoliths are the inherent drawbacksfor polymer-based and silica-based monoliths, respectively.10-12

Recently, the emerging organic-inorganic hybrid monolithiccolumns, incorporating organic moieties into inorganic (usuallysilica) monolithic matrices via the co-condensation of organofunc-tional trialkoxysilanes [(RO)3Si-R′: where R′ represents theorganofunctional group] and conventional tetra-alkoxysilanes(i.e., TMOS or TEOS) by the sol-gel method, seem to be thepromising choices for polymer- or silica-based monolithiccolumns.10-14 Organic-inorganic hybrid monolithic columns aresupposed to combine the merits of the organic polymer and

* To whom correspondence should be addressed. Tel.: +86-411-84379610.Fax: +86-411-84379620. E-mail: [email protected] (H.Z.). Tel.: +86-411-84379576. Fax: +86-411-84379620. E-mail: [email protected] (R.W.).

† Chinese Academy of Sciences.‡ Graduate School of Chinese Academy of Sciences.

(1) Svec, F. J. Sep. Sci. 2004, 27, 1419–1430.(2) Zou, H. F.; Huang, X. D.; Ye, M. L.; Luo, Q. Z. J. Chromatogr. A 2002,

954, 5–32.(3) Tanaka, N.; Kobayashi, H.; Ishizuka, N.; Minakuchi, H.; Nakanishi, K.;

Hosoya, K.; Ikegami, T. J. Chromatogr. A 2002, 965, 35–49.(4) Vlakh, E. G.; Tennikova, T. B. J. Sep. Sci. 2007, 30, 2801–2813.(5) Wu, R. A.; Hu, L. G.; Wang, F. J.; Ye, M. L.; Zou, H. J. Chromatogr. A 2008,

1184, 369–392.(6) Kasicka, V. Electrophoresis 2008, 29, 179–206.(7) Sandra, K.; Moshir, M.; D’Hondt, F.; Verleysen, K.; Kas, K.; Sandra, P.

J. Chromatogr. B 2008, 866, 48–63.(8) Gusev, I.; Huang, X.; Horvath, C. J. Chromatogr. A 1999, 855, 273–290.(9) Guiochon, G. J. Chromatogr. A 2007, 1168, 101–168.

(10) Hayes, J. D.; Malik, A. Anal. Chem. 2000, 72, 4090–4099.(11) Colon, H.; Zhang, X.; Murphy, J. K.; Rivera, J. G.; Colon, L. A. Chem.

Commun. 2005, 2826–2828.(12) Yan, L. J.; Zhang, Q. H.; Zhang, H.; Zhang, L. Y.; Li, T.; Feng, Y. Q.; Zhang,

L. H.; Zhang, W. B.; Zhang, Y. K. J. Chromatogr. A 2004, 1046, 255–261.(13) Xu, L.; Lee, H. K. J. Chromatogr. A 2008, 1195, 78–84.

Anal. Chem. 2010, 82, 5447–5454

10.1021/ac1003147 2010 American Chemical Society 5447Analytical Chemistry, Vol. 82, No. 13, July 1, 2010Published on Web 05/28/2010

inorganic silica-based monoliths, such as easy fabrication, widepH range tolerance, good mechanical stability, and high perme-ability. Since Hayes and Malik10 prepared an organic-inorganicporous capillary monolithic column using N-octadecyldimethyl[3-(trimethoxysilyl)propyl] ammonium chloride as the organofunc-tional alkoxysilane for CEC, a number of organic-inorganic hybridcolumns have been reported by using different organofunctionalalkoxysilanes including phenyltriethoxysilane, 3-aminopropyltri-ethoxysilane, C8-triethoxysilane, and methyltrimethoxysilane.11-19

Bridged silane monomers such as 1,2-bis(trimethoxysilyl)ethaneand 1,2-bis(triethoxysilyl)ethane have also been used to preparehybrid monoliths.20 Dulay et al.21 also prepared an organic-silicahybrid monolithic column, named the photopolymerized sol-gel(PSG) monolith, using methacryloxypropyl trimethoxysilane (MPT-MS) as the monomer via polycondensation and photoinitiatedpolymerization. Different from the above-mentioned organic-in-organic hybrid monolithic columns, we have previously developeda “one-pot” approach to prepare the hydrophilic and hydrophobicorganic-inorganic hybrid monolithic capillary columns via the insitu co-condensation and copolymerization between the organicpolymerization precursors and the inorganic alkoxysilanes, whichcan be developed as a versatile method to synthesize theorganic-silica hybrid monoliths by using a variety of organicmonomers.22 Nevertheless, the use of the alkoxysilanes in theabove-mentioned approaches would probably result in the residualsilanol groups on monolith surface, which would possibly causethe peak tailing, broadening, or nonspecific adsorption in prac-tice.23

Polyhedral oligomeric silsesquioxane (POSS) is a type ofcagelike silsesquioxane, which embodies a truly inorganic-organichybrid architecture containing an inner inorganic framework madeup of silicon and oxygen.24-26 It refers to the structures with theempirical formula Rn(SiO1.5)n, where R represents a range oforganofunctional groups, while n is an even integer g4. POSSchemical reagents are thought to be the smallest silica particleswith sizes of 1-3 nm, which can be easily incorporated intocommon polymers via copolymerization, grafting, or blending.Using POSS reagents as the monomers in copolymerizationprocesses is convenient with no dramatic change in reaction

conditions, as long as the POSS monomers are soluble in themonomer mixture.25 The POSS reagents also offer a uniqueopportunity to prepare truly molecularly dispersed nanocom-posites, which can be used as rigid hard building blocks inpolymer for various hybrid materials.27 The interest in POSSmaterials is based on the facts that the rigid silicon and oxygenframework could greatly enhance the mechanical and thermalstability of the resulted POSS-containing nanohybrid poly-mers.24,28

In this work, we applied a POSS reagent of POSS-methacrylsubstituted (POSS-MA) as the cross-linker to prepare an inorganic-organic hybrid monolithic column, for the first time to our bestknowledge, via the copolymerization with a functional monomerof N-(2-(methacryloyloxy)ethyl)-dimethyloctadecylammonium bro-mide (MDOAB) in a toluene-dodecanol porogen system. Theresulting inorganic-organic hybrid monolithic capillary column(MDOA-POSS) was systematically investigated, which exhibitedgood mechanical stability and good pH stability. This approachwould represent a versatile method to prepare the inorganic-organic hybrid monolithic column by using a variety of organicfunctional monomers for copolymerization.

EXPERIMENTAL DETAILSMaterials. POSS-methacryl substituted (cage mixture, N )

8,10,12, POSS-MA) was purchased from Acros (NJ, USA). 2,2-(Dimethylamino)ethyl methacrylate (DEMA) was purchased fromNanjing Chemlin Chemical Industry Co., Ltd. (Nanjin, China).γ-Methacryloxypropyltrimethoxysilane (γ-MAPS) and trifluoro-acetic acid (TFA) were purchased from Sigma (St Louis, MO,USA). Ribonuclease B from bovine pancreas, bovine serumalbumin (BSA), cytochrome C from bovine heart, enolase fromyeast, insulin, and ovalbumin were all purchased from Sigma (StLouis, MO, USA). Lysozyme from chicken egg white was obtainedfrom Sino-American Biotechnology Co. (Beijing, China). Azobi-sisobutyronitrile (AIBN) was purchased from Shanghai ChemicalPlant (Shanghai, China) and recrystallized in ethanol before use.A fused-silica capillary with 75 µm i.d. and 375 µm o.d. waspurchased from Reafine Chromatography Ltd. (Hebei, China).Dithiothreitol (DTT), iodoacetamide, and the protease inhibitorscocktail were all purchased from Sino-American BiotechnologyCo. (Beijing, China). Daisogel ODS-AQ (5 µm, 120 Å pore) waspurchased from Daiso (Osaka, Japan). HPLC-grade acetonitrile(ACN) from Merck (Darmstadt, Germany) was used for thepreparation of mobile phases. Water used in all experiments wasdoubly distilled and purified by a Milli-Q system (Millipore Inc.,MA, USA). Other chemical reagents were of analytical grade.

Synthesis of N-(2-(Methacryloyloxy)ethyl)dimethy-loctadecylammonium Bromide (MDOAB). The synthesis ofMDOAB was similar to that of allyldimethyldodecylammoniumbromide (ADDAB) as in a previous work.22 DEMA (4 mL, 23.8mmol) was added dropwise to a solution of 1-bromooctadecane(9 mL, 26.5 mmol) in ethanol (20 mL), which was stirred for 10min at room temperature. Then, the obtained solution was heatedto 55 °C and stirred for 36 h. After removal of the solvent by arotary evaporator under vacuum, the resulting yellow liquid was

(14) Roux, R.; Puy, G.; Demesmay, C.; Rocca, J. L. J. Sep. Sci. 2007, 30, 3035–3042.

(15) Yan, L. J.; Zhang, Q. H.; Zhang, W. B.; Feng, Y. Q.; Zhang, L. H.; Li, T.;Zhang, Y. K. Electrophoresis 2005, 26, 2935–2941.

(16) Yan, L. J.; Zhang, Q. H.; Feng, Y. Q.; Zhang, W. B.; Li, T.; Zhang, L. H.;Zhang, Y. K. J. Chromatogr. A 2006, 1121, 92–98.

(17) Tian, Y.; Zhang, L. F.; Zeng, Z. R.; Li, H. B. Electrophoresis 2008, 29, 960–970.

(18) Constantin, S.; Freitag, R. J. Sol-Gel Sci. Technol. 2003, 28, 71–80.(19) Kanamori, K.; Yonezawa, H.; Nakanishi, K.; Hirao, K.; Jinnai, H. J. Sep. Sci.

2004, 27, 874–886.(20) Hutanu, D. Synthesis and characterization of novel stationary phases for

small scale liquid chromatographic separations of proteins and nanopar-ticles. Ph.D. Thesis, Oregon State University, 2008.

(21) Dulay, M. T.; Quirino, J. P.; Bennett, B. D.; Kato, M.; Zare, R. N. Anal.Chem. 2001, 73, 3921–3926.

(22) Wu, M. H.; Wu, R. A.; Wang, F. J.; Ren, L. B.; Dong, J.; Liu, Z.; Zou, H. F.Anal. Chem. 2009, 81, 3529–3536.

(23) Allen, D.; Rassi, Z. E. Analyst 2003, 128, 1249–1256.(24) Li, G. Z.; Wang, L. C.; Toghiani, H.; Daulton, T. L.; Koyama, K.; Pittman,

C. U. Macromolecules 2001, 34, 8686–8693.(25) Li, G.; Wang, L.; Ni, H., C. U. P., Jr J. Inorg. Organomet. Polym Mater. 2001,

11, 123–154.(26) Pielichowski, K.; Njuguna, J.; Janowski, B.; Pielichowski, J. Adv. Polym. Sci.

2006, 201, 225–296.

(27) Livage, J. Curr. Opin. Solid State Mater. Sci. 1997, 2, 132–138.(28) Tanaka, K.; Inafuku, K.; Adach, S.; Chujo, Y. Macromolecules 2009, 42,

3489–3492.

5448 Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

recrystallized in 40 mL ethyl acetate/hexane (20/80, v/v). Afterincubated at 4 °C for 1 h, followed by filtration and vacuum drying,9.2 g white solid product (MDOAB) was obtained with 95% yield.The product was characterized by matrix assisted laser desorptionionization time-of-flight (MALDI-TOF), m/z at 410.33 [M+].

Preparation of the MDOA-POSS Hybrid MonolithicColumn. Before the preparation of the MDOA-POSS hybridmonolithic column, the fused-silica capillary was respectivelyrinsed by 1.0 M NaOH for 6 h, water for 30 min, 1.0 M HCl for6 h, and water for 30 min, and was dried by nitrogen stream atroom temperature. A 50% γ-MAPS methanol (v/v) solution wasused to introduce the CdC bonds onto the inner surface ofcapillary to anchor monolithic matrix as previously reported.29

First, the solution was injected into the capillary, and then thecapillary was submerged in a water bath at 50 °C overnight withthe both ends sealed with silicone rubbers. After that, the capillarywas rinsed by methanol to flush out the residuals and dried bynitrogen stream again for further use.

The polymerization mixture consisting of monomer MDOAB,cross-linker POSS-MA, initiator AIBN, and binary porogen oftoluene-dodecanol were ultrasonicated in a bath sonicator (90W) for 15 min to degas. The obtained homogeneous solution wasmanually introduced into the γ-MAPS pretreated capillary to anappropriate length with a syringe. By sealing the both ends ofthe capillary with silicone rubbers, the capillary was incubated at55 °C for 12 h. Finally, the prepared monolith capillary columnwas washed by methanol to remove unreacted monomers andporogens. The structures of POSS-MA and MDOAB and theschematic synthesis of the MDOA-POSS hybrid monolith areillustrated in Figure 1.

Preparation of BSA Tryptic Digest. The tryptic digestprocedure of BSA was according to that previously reported byus with minor modifications.22 BSA (2 mg) was dissolved in 1mL of denaturing buffer containing 8 M urea. After the additionof 10 µL DTT solution, the mixture was incubated at 60 °C for1 h to reduce the disulfide bonds of the protein. After that, 3.7mg IAA was added, and the mixture was incubated at roomtemperature in the dark for 40 min. Finally, the mixture wasdiluted 8-fold with 100 mM ammonium bicarbonate buffer (pH8.2) and digested at 37 °C for 20 h with trypsin at the enzyme-to-substrate ratio of of 1:40 (w/w). After digestion, the pH valueof the obtained BSA tryptic digest solution was adjusted to 2-3by 10% TFA aqueous solution. Followed by a solid-phase extraction(SPE) of the BSA tryptic digest with a homemade C18 cartridge,

the collected peptides elution was dried under vacuum anddissolved into a 0.1% formic acid aqueous solution (1 mL), whichwas stored in a -4 °C freezer before µ-HPLC-MS/MS analysis.

Capillary Electrochromatography. All CEC experimentswere carried out on a CE instrument-P/ACE MDQ System(Beckman, Fullerton, CA, USA) equipped a UV detector withtemperature at 25 °C and detection wavelength at 214 nm. Adetection window was made by removing the polyimide coatingof a fused-silica capillary with a razor blade in the empty sectionof the capillary at the edge of the hybrid monolithic continuousbed. The total length of the prepared capillary column was 31 cmwith an effective length of 21 cm. The monolithic column wasfirst preconditioned by running buffer for at least 30 min with amanual syringe pump, and then, equilibrated on an instrumentby applying a low voltage (10 kV, ramping time for 10 min) untila stable current was obtained. All data obtained were based onthree runs. The retention factor (k′) was defined as (tr - t0)/t0,where tr and t0 represent the retention times of an analyte andan unretained compound (thiourea was used as the void timemarker here), respectively.

Micro-HPLC System. A µ-HPLC system consists of an agilentpump, a UV detector (K-2501, Knauer), a chromatographyworkstation (Cailu, Beijing), and a injection valve (model 7125,Rheodyne) fitted with a T-joint using a fused-silica capillary (50µm i.d. and 375 µm o.d., length of 95 cm) as a splitter. The splitratio was controlled at ca. 1/400. The outlet of the hybridmonolithic column was connected to a fused-silica capillary (50µm i.d. and 375 µm o.d.,) with a Teflon tube. The detection windowwas made by burning off a 2 mm polyimide coating at a position6 cm from the separation monolithic column outlet.

A µ-HPLC system (ThermoFinnigan, San Jose, CA) for proteintryptic digest separation consists of a degasser, a quaternarySurveyor MS pump, and a six-port/two-position valve. A capillaryof 50 µm i.d. was used for splitting, and the flow rate after splittingwas adjusted to ca. 200 nL/min.

Mass Spectrometry Detection. The LTQ linear ion trap massspectrometer was equipped with a nanospray ion source. Thetemperature of the ion transfer capillary was set at 200 °C. Thespray voltage was set at 1.8 kV, and the normalized collisionenergy was set at 35.0%. One microscan was set for each MS andMS/MS scan. All MS and MS/MS spectra were acquired in thedata dependent mode. The mass spectrometer was set that onefull MS scan was followed by six MS/MS scans on the six mostintense ions. The dynamic exclusion function was set as follows:repeat count 2, repeat duration 30 s, and exclusion duration 90 s.System control and data collection were done by Xcalibur software

(29) Dong, X. L.; Dong, J.; Ou, J. J.; Zhu, Y.; Zou, H. F. Electrophoresis 2006,27, 2518–2525.

Figure 1. Schematic preparation of MDOA-POSS hybrid monolithic column using POSS-MA as cross-linker and MDOAB as monomer.

5449Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

version 1.4 (Thermo). The scan range was set from m/z 400 tom/z 1600.

Data Analysis. The acquired MS/MS spectra was searchedon the database using the MASCOT (version 2.2.04) proteinidentification platform (Matrix Science, London, UK) and the MS/MS spectra of pull-down were searched against IPI_bovineBOVIN_3.32 (32 946 sequences; 16 109 453 residues). Cysteineresidues were searched as fixed modification of 57.0215 Da, andmethionine residues as variable modification of 15.9949 Da.Peptides were searched using fully tryptic cleavage constraintsand up to two internal cleavages sites were allowed for trypticdigestion. The mass tolerances were 2 Da for parent masses and1 Da for fragment masses.

RESULTS AND DISCUSSIONPreparation and Characterization of the MDOA-POSS

Hybrid Monolithic Column. The preparation conditions of theMDOA-POSS hybrid monolithic column have been investigated.Considering the cross-linker of POSS-MA can be well-dissolvedin toluene, a binary porogenic solvent of toluene-dodecanol,which is a broadly used porogen in the preparation of organicpolymer-based monolithic columns,30-33 was utilized for thishybrid monolithic column. The amounts of toluene and dodecanolused in the formation of the MDOA-POSS hybrid monolithiccolumn were listed in Table 1. As can be seen from columns A-Cin Table 1, the permeability of the formed monoliths decreasedas the amount of dodecanol (toluene) in the porogenic solventdecreased (increased), while the homogeneity of the generatedmonoliths increased conversely. This phenomenon is similar tothat in the preparation of polymer-based monolithic columns byusing toluene-dodecanol as the porogenic solvent,30,31 where thetoluene was used as the good solvent for hydrophobic monomersand dodecanol used as the solvent for polar monomers. The ratiochange of toluene to dodecanol affected the solubility of themonomer of MDOAB and the cross-linkers of POSS-MA, andconsequently influenced the phase separation in the formation ofthe hybrid inorganic-organic monolith. The ratio of toluene tododecanol used for the preparation of the hybrid monolith was90/250. As shown in Table 1, the MDOA-POSS hybrid monolithiccolumns (columns E and F) could be synthesized with this rationalratio, though the amount of the organic monomer of MDOABwas increased from 5 to 12 mg. Interestingly, with the increase

of the amount of MDOAB in the polymerization mixture, theelectroosmotic flow (EOF) of the synthesized MDOA-POSShybrid monolithic column was decreased from 1.71 to 1.51 and0.93 for 5, 8, and 12 mg of MDOAB, respectively. This is actuallyin conflict with the general thought that increasing the functionalmonomer in copolymerization would increase the incorporatedmonomer and consequently would generate the increased EOFfor CEC. In the comparison of retention factors (k′) amongcolumns B, E, and F, it was also observed that the k′ for benzeneand butylbenzene of column E and F were much less than that ofcolumn B. These results indicated that the amount of the MDOABcopolymerized into the hybrid monolith decreased when theMDOAB content high than 5 mg in the prepolymerization mixture.The factors which affects of the polymerization mixture of cationicmonomers were complicated including the distance between thequaternary ammonium groups and the radical position, chargedensity, ionic strength, hydrophobic interactions, etc.34 In thisexperiment, we observed that the polymerization time increasedwith the increase of the MDOAB content in the polymerizationmixture. The cationic group of MDOAB results in a repulsioneffect between each other which would prevent the formation ofhigher molecular weight polymer. Additionally, due to thehydrophobic interaction, the long chain of the MDOAB monomertends to approach to the POSS-MA, which would hinder thecopolymerization between MDOAB and POSS-MA. Additionally,the incompatibility of the reactivity ratios of MDOAB to POSS-MA might also impact the copolymerization process.

On the basis of these observations, a polymerization mixtureconsisting of 90 µL toluene, 250 µL dodecanol, 5 mg MDOAB, 50mg POSS-MA, and 2 mg AIBN was chosen and polymerized at55 °C to obtain the homogeneous and permeable MDOA-POSShybrid monolithic column and column B was used for furtherexperiment.

The cross-section morphology of the prepared MDOA-POSShybrid monolithic column (column B) was characterized by SEM,and the obtained SEM images were shown in Figure 2 with twodifferent magnifications. As shown in Figure 2A, a uniformMDOA-POSS monolithic matrix with large through-pores wasobtained within the capillary. In Figure 2B, it can be seen thatthe formed organic-POSS monolithic matrix was attached tothe inner capillary wall well. This was because of not only thesuccessful pretreatment of capillary by γ-MAPS but also thecopolymerization happened among γ-MAPS, MDOAB, andPOSS-MA. If looking at the aggregated polymer clusters of thesynthesized MDOA-POSS hybrid monolith, one can tell that theobtained MDOA-POSS hybrid monolith was a kind of organicmonolith.

(30) Li, Y.; Zhang, J.; Xiang, R.; Yang, Y. H.; Horvath, C. J. Sep. Sci. 2004, 27,1467–1474.

(31) Ou, J. J.; Dong, J.; Tian, T. J.; Hu, J. W.; Ye, M. L.; Zou, H. F. J. Biochem.Biophys. Meth. 2007, 70, 71–76.

(32) Li, Y.; Gu, B. H.; Tolley, H. D.; Lee, M. L. J. Chromatogr. A 2009, 1216,5525–5532.

(33) Lubbad, S. H.; Buchmeiser, M. R. J. Sep. Sci. 2009, 32, 2521–2529. (34) Losada, R.; Wandrey, C. Macromolecules 2009, 42, 3285–3293.

Table 1. Composition of Polymerization Mixture for the Preparation of MDOA-POSS Hybrid Monolithic Columnsa

col toluene (µL) dodecanol (µL) MDOAB (mg) permeability homogeneity EOF (cm2/(V s) × 10-4) k′ benzene k′ butylbenzene

A 80 260 5 good poorB 90 250 5 good good 1.71 0.383 1.017C 100 240 5 poor goodE 90 250 8 good good 1.51 0.329 0.881F 90 250 12 good good 0.93 0.344 0.908

a Other preparation conditions: polymerization temperature, 55 °C; POSS, 50 mg; AIBN, 2 mg.


To examine the mechanical stability of the obtained MDOA-POSS hybrid monolithic column, a column (length, 25.5 cm) wasconnected to a HPLC pump (Agilent) with the flow rate rangedfrom 0.1 to 2.2 µL/min (in splitter mode) using the water andACN as the mobile phase, respectively. The measured backpres-sure was linearly increased as the flow rate was increased (Seethe Supporting Information, Figure S1), which indicated that theMDOA-POSS hybrid monolith possessed good mechanicalstability. Using Darcy’s Law35 of permeability B0 = FηL/(πr2∆P),where F is the flow rate of the mobile phase (m3/s), η is theviscosity of the mobile phase (Pa · s), L is the effective lengthof column (m), r is the inner radius of the column (m), and∆P is the pressure drop of the column (Pa), the permeabilityof the MDOA-POSS hybrid monolithic column was calculatedas 7.19 × 10-14 m2 and 5.35 × 10-14 m2 for water (η = 0.89 cP)and ACN (η = 0.38 cP), respectively, which indicated the goodpermeability of the prepared monolithic column. As the mobilephase changed from water to ACN, the change of columnpermeability was ca. 25.6% which was less than the estimatedchanges of ca. 36.1% for poly(butyl methacrylate-co-ethylenedimethacrylate) monolith.36 However, the change of the per-mibility was large than that of poly[hydroxyethyl acrylate-co-poly(ethylene glycol) diacrylate] monoliths which showednearly no swelling or shrinking in different polarity solvents.37

This result indicates the acceptable solvent swelling of theprepared monolithic column by using the POSS as the cross-linker. By changing the mobile phase back to the previous one,the column permeability could be recovered within 30 min,which means that the swelling and the shrinking of this hybridmonolith was reversible.

To investigate the pH stability, the obtained MDOA-POSShybrid monolithic column was flushed for more than 100 h by50% ACN with different extreme pH values at the flow rate of0.2-0.3 µL/min. The chromatographic performances includingthe theoretical plate number N and the retention factor k′ of theflushed column were monitored by CEC at several flush intervals.The obtained results were illustrated in Figure S2 (of theSupporting Information), which describes the performance traceof columns flushed by 50% ACN solution at extreme low and high

pH values. As seen in Figure S2A, both theoretical plate numberN and retention factors (k′) were remained above 90% afterflushing by 0.1 M HCl containing 50% ACN solution (pH ) 1.1)for over 100 h, which indicats that the MDOA-POSS hybridmonolithic column has good pH stability at a low pH value. Incontrast, the obtained hybrid monolithic column was also flushedby a basic solution at a high pH value to test its pH stability atbasic conditions. Figure S2B shows the similar performance traceof N and k′ under high pH conditions. The result shows that theretention factors (k′) of hydrophobic compounds remained about75% after flushing by a 10 mM phosphate solution containing 50%ACN at pH 11 for 35 h and, then, stayed at nearly the same levelfor over 100 h. At this high pH value, N was stable at the initialperiod of 35 h. After that, N increased first and then decreasedas the flush time increased to ca. 120 h. Even higher pH valuesof 12 and 13 have also been used for the pH stability test. It wasfound that N could remain constant for 24 h under such extremepH conditions. After flushing for 24 h at pH 13, k′ was decreasedremarkably, while the slight detachment of the monolith from theinner wall of the capillary was also observed by an opticalmicroscope. This may be due to the fact that the linkage betweenthe hybrid monolith and capillary, the Si-O-Si-C bonds formedbetween γ-MAPS and the capillary wall, was destroyed under suchstrong basic ambience.

CEC and µ-HPLC Performance. In CEC, EOF is the basicrequirement for driving mobile phases through a capillary column.To prepare the hybrid monolithic column for CEC with a sufficientEOF, a synthesized quaternary ammonium bromide monomer ofMDOAB, which acted as the main source to generate the EOFfrom cathode to anode, was used for the hybrid monolithic matrixfabrication. The examined relationship between the EOF and thepH value of the running buffer is presented in Figure S3 (of theSupporting Information). It can be seen that the MDOA-POSShybrid monolithic column could provide strong EOF (greater than1.98 × 10-4 cm2/(V s), from cathode to anode) in CEC in awide range of pH values (from 2 to 10) of running buffer. Thestrong EOF maintained at high pH condition was due to thestrong ionization of the quaternary ammonium group ofMDOAB, which indicates the successful in situ copolymeriza-tion of MDOAB and POSS-MA. The obtained strong EOF ofthe MDOA-POSS hybrid monolithic column would be apotential advantage for the separation of positively chargedanalytes in CEC by using a high pH value of the running buffer.

(35) Stanelle, R. D.; Sander, L. C.; Marcus, R. K. J. Chromatogr. A 2005, 1100,68–75.

(36) Gu, C.; Lin, L.; Chen, X.; Jia, J.; Wu, D.; Fang, N. J. Sep. Sci. 2007, 30,2866–2873.

(37) Li, Y. Y.; Tolley, H. D.; Lee, M. L. Anal. Chem. 2009, 81, 9416–9424.

Figure 2. SEM images of a MDOA-POSS hybrid monolithic column. Magnification: 2000× for A and 5000× for B.


As shown in Figure 1, the synthesized monomer of MDOABhas a methacrylate group and a quaternary ammonium cation anda long aliphatic carbon chain (C18). After the in situ copolymer-ization with POSS-MA, the long aliphatic carbon chain ofMDOAB played as the hydrophobic functional group for chro-matographic separation. In this work, the alkylbenzenes were usedto investigate the chromatographic performance of the obtainedMDOA-POSS hybrid monolithic column in both CEC andµ-HPLC. Figure 3 shows the resultant chromatograms of theseneutral aromatic compounds with good peak shapes on aMDOA-POSS hybrid monolithic column with the running bufferof 80% ACN at pH 3 for CEC and 0.1% TFA solution containing70% ACN for µ-HPLC. The analytes were all eluted in the orderof thiourea < benzene < toluene < ethylbenzene < propylbenzene< butylbenzene, which is corresponding to the hydrophobicitiesof these analytes from low to high. This result indicated thereversed-phase separation of these compounds on the hybridmonolithic column, which was further examined by investigatingthe change of retention factors (k′) over the ACN content in themobile phase in CEC. As shown in Figure S4 (of the SupportingInformation), the logarithms of the retention factors of alkylben-zenes decreased linearly with the increase of the ACN content inmobile phase, which confirmed the typical reversed-phase chro-matographic property of the MDOA-POSS hybrid monolithicstationary phase toward the hydrophobic solutes.10

The column efficiencies of the obtained MDOA-POSS hybridmonolithic column in CEC and µ-HPLC were also evaluated. InCEC, the applied voltage was changed from 5 to 29 kV, and therelationship between the flow velocity and the plate height forthiourea, benzene and toluene were demonstrated in Figure 4.The column efficiency in the range of 166 000-187 000 N/m foralkylbenzenes and 223 000 N/m for thiourea was observed. Thecolumn remained at high efficiency with the linear velocity rangingfrom 1.0 to 1.6 mm/s in CEC. The relationship between the flowvelocity and the plate height evaluated in µ-HPLC mode was also

shown in Figure 4, and the lowest plate height of ca. 20 µm wasobtained.

The run-to-run reproducibility was evaluated on a singlecapillary monolithic column in CEC. The relative standard devia-tions (RSDs) for EOF and retention time of analytes (thiourea,benzene, and toluene) on the capillary monolithic column wereless than 4.2% for 5 runs in CEC. Both column-to-column andbatch-to-batch reproducibilities were also evaluated in term of theRSDs of EOF and retention times of analytes, which were lessthan 6.7% (n ) 3) and 9.2% (n ) 3), respectively. These resultsindicated that the reproducibility of these prepared MDOA-POSShybrid monolithic columns was acceptable.

Application of MDOA-POSS Hybrid Monolithic Column.The MDOA-POSS hybrid monolithic column was used for theseparation of peptides in CEC. The separation mechanism ofnegatively charged compounds on this MDOA-POSS hybridmonolithic column is the combination of electrophoretic mobility,strong anion exchange (SAX) interaction, and hydrophobicinteraction. As shown in Figure 5, five peptides were separatedand four of them were eluted before the void time which wasmainly due to the electrophoretic mobility. The peptides separatedhere should be negatively charged at pH 6.5, so the direction ofelectrophoretic migration was same to EOF which would lead toshort separation time especially for the acidic peptides such asGly-Gly-Asp-Ala (peak 1). It also indicated that the hydrophobicinteraction were relatively weak since they were highly polarsolutes and showed weak retention on hydrophobic reversedphase stationary phase. The SAX interaction was due to theexistence of quaternary ammonium groups on the matrix surface.The effect of ionic strength in the mobile phase was alsoinvestigated. Figure 5 shows that the higher salt concentration(Figure 5A) could improve the peak shape since the SAXinteraction could be suppressed by higher salt concentration tosome content. The suppression of anion exchange interaction wasfavored for quicker electrophoretic migration which could leadto shorter separation time.

This hybrid monolithic column was also used for the separationof standard proteins in µ-HPLC mode. Figure 6 (The drift of thebaseline was due to the gradient elution) shows that sevenstandard proteins including ribonuclease B, cytochrome C, insulin,lysozyme, BSA, enolase, and ovalbumin were successfully sepa-

Figure 3. Separation of alkylbenzens on MDOA-POSS hybridmonolithic column. Experimental conditions for CEC: mobile phase,10 mM Na2HPO4-citric acid buffer containing 80% ACN at pH 3;separation voltage, -20 kV; injection, -5 kV for 1 s; column effectivelength, 21 cm; total length, 31 cm; detection wavelength, 214 nm.Experimental conditions for µ-HPLC: mobile phase, 0.1% TFA buffercontaining 70% ACN; column effective length, 34 cm; outlet todetection window, 6 cm; injection volume, 4 µL in split mode; detectionwavelength, 214 nm. Analytes: (0) thiourea; (1) benzene; (2) toluene;(3) ethylbenzene; (4) propylbenzene; (5) butylbenzene.

Figure 4. Relationship of the plate height on the linear velocity ofthe mobile phase on the MDOA-POSS hybrid monolithic column.Experimental conditions for CEC and µ-HPLC are the same as thosein Figure 3 except the flow velocity.


rated in 17 min with good peak shapes. Since the separationmechanism is a mixed mode containing SAX and hydrophobicinteraction, the separation efficiency of proteins as well as peptideswould not be comparable to that of pure reversed phasedseparation in µ-HPLC, so the fabrication of a pure reversed phasehybrid monolithic column with nonionic monomers such as octyl-or dodecyl-methacrylate was our following work. However, theseseparation results indicated that this hybrid monolith column wascapable of separating both small molecules and big biomolecules.

Besides the separation of alkylbenzenes, peptide mixtures, andstandard proteins, the separation of peptide mixtures of BSAtryptic digest was also attempted on this obtained MDOA-POSShybrid monolithic column by µLC-MS/MS for further investigatingthe potential use in the analysis of complex samples. One

picomolar BSA tryptic digest was manually loaded onto a capillarymonolithic column with a capillary (21 cm × 75 µm i.d.) for theµHPLC-MS/MS in RP mode with a gradient elution from 2 to 35%ACN within 40 min, and the chromatogram was shown in Figure7. On the basis of the database search of the obtained chromato-gram of BSA tryptic digest, 53 unique peptides were positivelyidentified and the protein coverage was 71% (RSD ) 6.4%, n ) 3).These results were comparable to that obtained from a particulate-packed commercial column (5 µm, 120 Å pore, 12 cm × 75 µmi.d.), where 59 unique peptides (RSD ) 4.1%, n ) 3) and 76%protein coverage (RSD ) 3.1%, n ) 3) for 1 pmol BSA trypticdigest using the same µ-HPLC MS/MS analysis conditions (achromatogram of the particle packed column was shown in theSupporting Information, Figure S5). The extract peaks (K.TC-VADESHAGCEK.S, ions score 34) between the particle packedcolumn and MDOA-POSS hybrid monolithic column werecompared (Supporting Information, Figure S6). The peak widthsat the 0.613 height were 0.184 and 0.188 min for the MDOA-POSShybrid monolithic column and particle packed column, respec-tively. These results also indicated that this MDOA-POSS hybridmonolithic column was capable of separating some complexsamples.

CONCLUSIONSThe monolithic column is being considered as the new

generation of column for the chromatographic separation sciences,and the difficulties in the preparation of the desired monolithiccolumn, either organic polymer-based or silica-based monolithiccolumns, has motivated the development of new methods forcolumn preparation. Here, we prepared a novel inorganic-organichybrid monolithic capillary column by the in situ thermal initiatedfree radical copolymerization within the confine of a capillary usinga POSS regent (POSS-MA) as the inorganic-organic hybridcross-linker and a synthesized aliphatic long chain methacrylatequaternary ammonium salt as monomer. To the best of ourknowledge, this is the first time to use a POSS reagent as the

Figure 5. Separation of peptide mixture on MDOA-POSS hybridmonolithic column by CEC. CEC conditions: mobile phase, 20 mMNa2HPO4-citric acid buffer for A and 10 mM Na2HPO4-citric acidbuffer for B containing 35% ACN at pH 6.5; separation voltage, -20kV; injection, -5 for 5 s. Other CEC conditions are the same as thosein Figure 3. Analytes: (1) Gly-Gly-Asp-Ala; (2) Gly-Ala-Ala; (3) Ser-Ser-Glu-Ala-Asn-Leu-Arg; (4) Ala-Thr-Val-Leu-Asn-Tyr-Leu-Pro; (5)Leu-Tyr-Leu. Other peaks are unknown impurities.

Figure 6. Separation of standard proteins on MDOA-POSS hybridmonolithic columns in µ-HPLC. Conditions: column length, 34/40 cm;mobile phase, (A) 0.1% TFA in water, (B) 0.1% TFA in ACN; gradient,15 to 45% B in 10 min; inject volume, 10 µL in split mode; injectedsample amount of these analytes after split were in the range of 4-8ng; flow rate 300 nL/min; detection wavelength, 220 nm. Analytes:(1) ribonuclease B, (2) cytochrome C, (3) insulin, (4) lysozyme, (5)BSA, (6) enolase (7) ovalbumin. Other peaks are unknown impurities.

Figure 7. Base peak chromatogram of µ-HPLC-MS/MS analysis ofBSA tryptic digest. Experimental conditions: MDOA-POSS hybridmonolithic column, 40 cm × 75 µm i.d.; flow rate, 200 nL/min; sampleamount, 1 pmol; mobile phase (A) 0.1% formic acid in water, (B) 0.1%formic acid in ACN; the separation gradient was buffer B from 2 to35% in 40 min and from 35 to 80% in 5 min. After flushing with 80%buffer B for 5 min, the separation system was equilibrated with 2%buffer B for 25 min.


inorganic-organic hybrid cross-linker in the preparation of hybridmonolithic column. The whole preparation process was simpleand very similar to that of organic polymer-based monolithiccolumns instead of the use of the inorganic-organic hybrid POSSreagent as cross-linker. The pore structure and the permeabilityof the synthesized inorganic-organic hybrid monolithic columnare tunable by changing the polymerization conditions. After thecharacterization of the resultant MDOA-POSS hybrid monolithiccolumns, it was found that the hybrid monolithic column exhibitedgood mechanical stability and good pH stability. The monomerof MDOAB incorporated in the hybrid monolithic matrix couldprovide not only a strong EOF in a wide pH range (pH 2-10),but also the high hydrophobicity for separation of small neutralanalytes in RP mode both in CEC and µ-HPLC. The columnefficiencies of 223 000 and 50 000 N/m for thiourea were observedin CEC and µ-HPLC, respectively. The separation of peptides,standard proteins, and protein tryptic digest showed its potentialin analysis of big biomolecules and complex samples.

The success of using the inorganic-organic hybrid POSSreagent as the cross-linker does provide an interesting clue forthe preparation of hybrid monolithic column by using othermultifunctional hybrid materials or reagents as the basic nanob-locks for monoliths. In fact, POSS functionalized with various

reactive organic groups can be incorporated into any existingpolymer system through either grafting or copolymerization.25

Additionally, a variety of POSS reagents have been commercialavailable, which would be a great advantage to prepare variousinorganic-organic hybrid monoliths by copolymerization withunrestricted organic monomers for different application purposes.

ACKNOWLEDGMENTThis work was supported by the National Natural Science

Foundation of China (No. 20735004) and the State Key BasicResearch Development Program of China (No. 2005CB522701)to H.Z.; the National Natural Sciences Foundation of China (No.20875089), the National High Technology Research and Develop-ment Program of China (No. 2008AA02Z211), and the HundredTalent Program of the Chinese Academy of Sciences to R.W.

SUPPORTING INFORMATION AVAILABLEFigures S1-S6. This material is available free of charge via

the Internet at http://pubs.acs.org.

Received for review October 27, 2009. Accepted May 11,2010.

AC1003147


polyhedral oligomeric silsesquioxane as a cross-linker for preparation of inorganic−organic hybrid...

Documents