Metal organic framework–organic polymer monolith stationary phases for capillary electrochromatography and nano-liquid chromatography

Analytica Chimica Acta 779 (2013) 96–103

Contents lists available at SciVerse ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Metal organic framework–organic polymer monolith stationary phases for capillary electrochromatography and nano-liquid chromatography Hsi-Ya Huang ∗ , Cheng-Lan Lin, Cheng-You Wu, Yi-Jie Cheng, Chia-Her Lin ∗ Department of Chemistry, Chung Yuan Christian University, 200 Chung Pei Road, Chung-Li 320, Taiwan

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

• Microwave-assisted

synthesis of hybrid metal organic framework (MOF)–polymer monolith. • MOF–polymer monolith was applied to CEC, nano-UHPLC and nano-LC–MS2 . • Excellent separation for isomers, aromatic acids and PAHs were achieved. • High sequence coverage for the nanoLC separation tryptic digested BSA peptides.

a r t i c l e

i n f o

Article history: Received 18 January 2013 Received in revised form 25 March 2013 Accepted 29 March 2013 Available online 16 April 2013 Keywords: Metal organic framework Organic polymer monolith Capillary electrochromatography Nano-liquid chromatography

a b s t r a c t In this study, metal organic framework (MOF)–organic polymer monoliths prepared via a 5-min microwave-assisted polymerization of ethylene dimethacrylate (EDMA), butyl methacrylate (BMA), and 2-acrylamido-2-methylpropane sulfonic acid (AMPS) with the addition of various weight percentages (30–60%) of porous MOF (MIL-101(Cr)) were developed as stationary phases for capillary electrochromatography (CEC) and nano-liquid chromatography (nano-LC). Powder X-ray diffraction (PXRD) patterns and nitrogen adsorption/desorption isotherms of these MOF–organic polymer monoliths showed the presence of the inherent characteristic peaks and the nano-sized pores of MIL-101(Cr), which confirmed an unaltered crystalline MIL-101(Cr) skeleton after synthesis; while energy dispersive spectrometer (EDS) and micro-FT-IR spectra suggested homogenous distribution of MIL-101(Cr) in the MIL-101(Cr)–poly(BMA–EDMA) monoliths. This hybrid MOF–polymer column demonstrated high permeability, with almost 800-fold increase compared to MOF packed column, and efficient separation of various analytes (xylene, chlorotoluene, cymene, aromatic acids, polycyclic aromatic hydrocarbons and trypsin digested BSA peptides) either in CEC or nano-LC. This work demonstrated high potentials for MOF–organic polymer monolith as stationary phase in miniaturized chromatography for the first time. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Nanoporous metal–organic frameworks (MOFs) are one type of novel crystalline materials constructed from clusters of metal ions which are connected by organic linkers [1–5]. The unparalleled

∗ Corresponding authors. Tel.: +886 3 2653319; fax: +886 3 2653399. E-mail addresses: [email protected] (H.-Y. Huang), [email protected] (C.-H. Lin). 0003-2670/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2013.03.071

properties of these highly ordered porous materials such as large surface areas (up to thousands m2 g−1 ), high porosity, tunable pore sizes (including meso- and micro-sized pores), as well as specific adsorption affinities make MOFs highly promising in a wide range of applications including gas storage and separation [6], catalysis [7], magnetism [8], luminescence [9], drug delivery [10] as well as chromatography [11]. Chromatography is a separation technique via differential partition of solute molecules between a stationary phase and a mobile phase. To serve as good stationary phase for small and large molecule separations, materials

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

generally have small mesopores as well as large through-pores (macropores) in their structure; in which the small mesopores provide adequate surface area required for solute retention (impact on resolution), while large through-pores allow the mobile phase to flow through the stationary phase (impact on column efficiency and hydraulic resistance) [12,13]. Since MOFs have plenty of meso- and micro-sized pores and high surface area, they meet the requirements of a good chromatographic stationary phase thus, the feasibility of various MOFs including MOF-508 [14], MIL47 [15–17], MIL-53 [18–20], MIL-101 [21–24], ZIF-8 [25], IRMOF [26], UIO-66 [27] and silica–MOF composite [28,29] as stationary phases in liquid chromatography (LC) or gas chromatography (GC) have been studied and results demonstrated remarkable success in small molecule (substituted aromatics) separation; using particulate packed columns with 3–6 mm I.D. for LC. Probably because of a relatively high back pressure produced, MOFs potential as stationary phase for capillary- or nano-scale liquid chromatography has not been explored so far. Recently, capillary or nano-liquid chromatography (nanoLC) and capillary electrochromatography (CEC) techniques have attracted great attention for their advantages such as low consumption of sample and mobile phase, high efficiency and resolution, high compatibility with mass spectrometry and an eco-friendly method due to less solvent wastage [30,31]. To accompany with the development of these miniaturized chromatographic techniques, porous monoliths that were synthesized by simple in situ preparation of organic polymer or silica-based monoliths with various specific functionalities, have been regarded as alternative stationary phases to conventional C18 packing materials. In contrast to conventional particulate packed columns, the in situ prepared monoliths are chemically attached onto the inner walls of the capillaries eliminating the need for retaining frits thus simplifying column preparation and reducing cost [32,33]. Our previous studies have demonstrated a very efficient strategy of organic polymer monolith preparation via room temperature ionic liquids (RTIL) as reaction media coupled with microwave-assisted heating for just a few minutes, which enhances the competitive advantage of this type of monolith as stationary phase in the future [34,35]. As mentioned earlier, large through-pores (macropores) and mesopores play different functions in the chromatographic behavior (column efficiency and resolution, respectively) [13], however, tailoring the porosity of organic polymer monoliths to contain both pore types is a challenge with only few studies demonstrated this possibility [36]. In order to combine the merits of both porous monoliths (macropores donor) and MOFs (mesopores donor), we develop a novel MOF–polymer monolith as an alternative organic–inorganic hybrid stationary phase. MIL-101(Cr), a nano-/meso-porous chromium terephthalate cage-type MOF, which carries excellent characteristics such as high surface area (∼4100 m2 g−1 ), large micropores ˚ and mesopores (29–34 A) ˚ [2], and excel(12 A˚ and 16 A˚ × 14.5 A) lent chemical and solvent stabilities [37], was used as MOF source in this study. Via in situ polymerization of common methacrylate monomers (butyl methacrylate (BMA), ethylene dimethacrylate (EDMA) and few amounts of 2-acrylamido-2-methylpropane sulfonic acid (AMPS)) with 1-hexyl-3-methylimidazolium tetrafluoroborate ([C6 min][BF4 ], a RTIL as reaction medium), and assisted by a 5-min microwave heating, MIL-101(Cr) was dispersed in the polymer monolith. Characterization methods employed were scanning electron microscopy (SEM), energy dispersive spectrometer (EDS), powder X-ray diffraction pattern (PXRD), FT-IR spectroscopy and nitrogen adsorption/desorption isotherm while column performance was assessed by CEC and nano-LC conditions. Several compounds including isomers of xylene, chlorotoluene, cymene and aromatic acids, polycyclic aromatic hydrocarbons (PAHs) and BSA digested peptides were used as test analytes.

97

To emphasize the merits of this MOF–polymer monolith, a MIL101(Cr) packed column and a neat poly(BMA–EDMA) column, fabricated in the same capillary dimensions, were also compared. 2. Experimental 2.1. Chemicals and reagents All chemicals and reagents used were at least of analytical grade. m-Xylene, o-xylene and naphthalene (Fluka, Steinhiem, Germany), p-xylene, m-chlorotoluene, and o-chlorotoluene (Alfa Aesar, Lancashire, UK), p-chlorotoluene, p-cymene, and phenanthrene (Acros, New Jersey, USA), m-cymene and o-cymene (TCI, Tokoyo, Japan), benzoic acid, terephthalaldehydic acid, isophthalic acid, and terephthalic acid (Merck, Darmstadt, Germany), pyrene and benzo[ghi]perylene (Aldrich, Steinhiem, Germany), benzo[a]pyrene (MP-ICN, Illkrich, France) were used as analytes. BMA and AMPS (Lancaster, Ward Hill, MA, USA), EDMA and 3trimethoxysilylpropyl methacrylate (MSMA) (Acros, New Jersey, USA), styrene and azobisisobutyronitrile (AIBN) (Showa, Tokoyo, Japan), divinylbenzene (DVB) and sodium 4-vinylbenzenesulfonate (VBSA) (Aldrich, Steinhiem, Germany) were used for the polymeric monolith preparation. Uncoated fused-silica capillaries with 100 ␮m I.D. and 375 ␮m O.D. were purchased from Reafine Chromatography Ltd. (Hebei, China). 2.2. Synthesis of MIL-101(Cr) MIL-101(Cr) was synthesized according to a procedure by Férey et al. [2]. Typically, Cr(NO3 )3 ·9H2 O (800 mg, 2.0 mmol), terephthalic acid (332 mg, 2.0 mmol), and HF (0.1 mL, 0.5 mmol) were mixed with ultrapure water (9.6 mL) in a Teflon-lined bomb. The Teflonlined bomb was then sealed and placed in an oven at 220 ◦ C for 8 h. After being cooled to room temperature, the resultant green crystalline solid was washed thoroughly with DMF and ethanol, and was collected by centrifugation at 6000 rpm for 5 min. The procedure was repeated at least three times to remove the unreacted terephthalic acid from MIL-101(Cr) pores. The green solid was then evacuated in vacuum at 150 ◦ C for 12 h to form dehydrated MIL-101(Cr), which was characterized by PXRD spectrometry, thermal gravimetric analysis (TGA), nitrogen adsorption/desorption isotherm and SEM. 2.3. Preparation of MIL-101(Cr)–poly(BMA–EDMA) monolithic column Prior to the preparation of a polymeric monolithic column or frit, the inner wall of a 100 ␮m ID capillary column was treated according to the procedure described in our previous article [34,35]. Capillary columns were filled with vinylization solution composed of MSMA mixed with [C6 mim][BF4 ] (50%, v/v). After both ends of the capillaries were sealed with epoxy adhesive, these were placed in a beaker containing RT water and heat-treated in a microwave oven (900 W, SAMPO RE-1002SM) for 5 min. This vinylization method that can reduce the time from 17 h (water bath) to 5 min (microwave heating) was reported in our previous study [34]. For the preparation of MIL-101(Cr)–poly(BMA–EDMA) monolithic column, 4 mg of MIL-101(Cr) (50 wt.% of monomer amount) was suspended in the mixture containing monomers (BMA (3.6 ␮L) and EDMA (5.4 ␮L)), and porogenic solvents ([C6 min][BF4 ] (38 ␮L) and water (3 ␮L)), AIBN (0.5 mg) and AMPS (0.5 mg). The mixture was used to fill the preconditioned capillary up to the desired length (25 cm or 40 cm for CEC and 35 cm for nanoLC) by N2 pressure injection followed by in situ polymerization via microwave-assisted heating for 5 min. The MOF–polymer monolithic column was then

98

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

washed with methanol and mobile phase by a LC pump. A detection window (for CEC use) was fabricated using a microtorch to remove the polyimide coating at the 25 cm (or 40 cm) position on the column where a polymer bed was absent. 2.4. Preparation of MIL-101(Cr) packed column A short and porous polystyrene-based monolith (∼2 mm in length) was used as frit to retain MIL-101 particulate packings in the capillary. The porous frit was made by mixing monomer solution (styrene (463 ␮L) and divinylbenzene (695 ␮L)), and porogenic solvent (cyclohexanol (1740 ␮L), N,N-dimethylacetamide (1740 ␮L), and water (187 ␮L)) followed by sonication (15 min), then this solution was used to fill the preconditioned capillary (see Section 2.3) to a total length of 2 mm (at one end) by syringe injection. When both ends of the capillary were sealed with adhesive resin, the capillary was submerged in a 70 ◦ C water-bath for 15 h. The pretreated column with polystyrene-based frit was cut into two pieces and the column that contained the ∼2 mm polystyrene-based frit was packed with MIL-101(Cr)–methanol slurry (15 mg mL−1 ) using a LC pump at 5000 psi up to the desired length (5- and 7-cm for nano-LC and CEC, respectively). The MIL-101(Cr) packed column for nano-LC (5 cm in length) was employed directly to the system, while the column for CEC (7 cm in length) was connected to a capillary (100 ␮m I.D. × 24-cm in length) with a polystyrene-based monolith outlet frit (∼2 mm) through an auxiliary PTFE tube prior to use (the frit setup is illustrated in the supplementary information, Fig. S1). 2.5. Apparatus CEC experiments were performed with a Hewlett-Packard 3D capillary electrophoresis system equipped with a 3D UV–vis detector (Waldbronn, Germany). Agilent Chemstation software was used for instrumental control and data analysis. Nano-LC experiments for all small analytes were carried out using Waters nanoAcquity UPLC system (MA, USA) with a binary solvent pump, sample manager, autosampler and TUV detector equipped with a 10 nL cell; while BSA peptide digests were separated in an UltiMate 3000 Nano-LC system (Dionex, Amsterdam, The Netherlands) coupled to the Amazon SL mass spectrometer (Bruker-Daltonik, Germany) equipped with a nanoelectrospray ionization source (Bruker). A LC pump (Model 260D, ISCO, Lincoln, NE, USA) was used for packing, washing and equilibrating the packed or polymeric monolithic columns. A scanning electron microscope model JSM-7600F (JEOL, Japan) was used for morphology observation. A X-ray diffractometry model X’Pert Pro MRD (Panalytical, Netherlands) was used for PXRD pattern recording. A FT-IR spectroscopy model FT/IR 4200 (Jasco, Japan) was used for infrared spectrum measurement. A BET nitrogen adsorption/desorption equipment model Micromeretics Tri-star 3000 (Norcross, GA, USA) was employed for surface area measurement. A thermogravimetry model TG/DTA 6200 (SII Nano Tech., Japan) was used for thermal decomposition temperature (Td ) measurement. 2.6. Operational condition for CEC and nano-LC The prepared stationary phases (MIL101(Cr)–poly(BMA–EDMA) and packed MIL-101(Cr)) were used in CEC and nano-LC systems. In CEC, these columns were equilibrated with the mobile phase under 10 kV applied voltage and 5 bar pressure at both ends of the column until a stable baseline was obtained. Mobile phases for CEC operation were prepared by mixing acetonitrile (ACN) with phosphate buffer (5 mM) in different volume ratios. Samples and standards were electrokinetically

injected into the capillary for 3 s at a voltage of 5 kV. CEC separations were carried out using an electrical voltage of 20 kV. For nano-LC experiments, a volumetric flow rate of 1 ␮L min−1 was used for isocratic elution. The mobile phases were 0.1% formic acid in water (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B). The injected volume was 0.5 ␮L. The detection wavelength and the column temperature in both CEC and nano-LC systems were maintained at 200 nm and 25 ◦ C, respectively. 3. Results and discussion In the fabrication of the MOF–polymer monolith stationary phase, poly(BMA–EDMA) monolith was used as the organic polymer source since it is a common polymeric stationary phase in CEC and capillary-LC, while MIL-101(Cr) was the chosen MOF for this hybrid preparation as well as for the packed capillary column preparation due to its good solvent stability, high resolution and reproducibility in the previous conventional LC separations [22–24]. 3.1. Characterization of MIL-101(Cr)–polymer monolith and MIL-101 packed columns Morphological differences of the different stationary phases prepared with varying amounts of MIL-101(Cr), with weight percentages of 100%, 50% and 0% for MIL-101(Cr) packed columns, MIL-101(Cr)–polymer and neat poly(BMA–EDMA) monoliths, respectively are evidently observed in their SEM images shown in Fig. 1. Variations as to their skeleton strongly correlates with the amount of MOF added wherein the large-sized through-pores characterize MIL-101(Cr)–polymer monolith (∼1.3 ␮m), mediumsized through-pores for poly(BMA–EDMA) monolith (∼1 ␮m), and small-sized ones for MIL-101(Cr) packed column (∼0.2 ␮m) (the through-pore sizes were estimated from the SEM images). Since MIL-101(Cr) packed column was the densest among the three columns, it was the least permeable and thus, exhibited high back pressures (see Section 3.2) which prompted us to use a shorter column length (for instance, 5- or 7-cm) for the chromatographic testing. On the other hand, a longer column (for instance, 25–40cm in length) was used to assess the chromatographic performance of MIL-101(Cr)–polymer monolith because of the increased column permeability contributed by the large through-pores from the polymer part (i.e. poly(BMA–EDMA)) which was absent in the MOF packed column. Several instruments were used to probe the structural characteristics of these monoliths including PXRD, nitrogen adsorption/desorption isotherms and FT-IR spectroscopy. PXRD patterns for MIL-101(Cr) in the MOF–polymer monolithic column containing 50% (wt.) MIL-101(Cr) (Fig. 2(a)) revealed the inherent characteristic peaks of crystalline 3D MIL-101 while the nitrogen adsorption/desorption isotherms for MIL-101(Cr)–polymer monolith and as-synthesized MIL-101(Cr) showed almost the same micro-sized (<2 nm) and meso-sized pores (>2 nm) (1.97-, 2.31-, 3.06-nm for MIL-101(Cr) and 1.76-, 1.99-, 2.92- nm for MIL101(Cr)–polymer) (Fig. 2(b) and (c), Table 1). Both PXRD spectra and BET data confirmed intact MIL-101 framework after the polymerization. FT-IR spectra indicated the characteristic absorption of alkyl C H stretch and carbonyl C O stretch (2800–3000 cm−1 and 1730 cm−1 , respectively, originated from the methacrylate-based polymer) as well as the aromatic C C bending (1600 cm−1 from 1,4-terephthalic acid, MOFs organic linker (Fig. 2(d)). Moreover, SEM–EDS mapping (Fig. S2) revealed homogeneous distribution of Cr atoms (from MIL-101(Cr)) as well as C and O atoms (from BMA and EDMA) in the MOF–polymer monolith. Also, in the micro-FTIR measurement (Fig. S3), nearly similar IR spectra were observed from 5 different sampling spots of the MOF–polymer. The above

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

99

Fig. 1. SEM micrographs of MIL-101(Cr) packed column (a and b), MIL-101(Cr)–polymer monolith (c and d) and neat polymer monolith (e and f). Magnifications were 0.8k (a, c, and e) and 50k (b, d and f).

results proved the homogeneous dispersion of MIL-101(Cr) in the MOF–polymer monolith. Moreover, this phenomenon could possibly be due to the attraction of MOFs unsaturated chromium ion with BMAs (or EDMAs) oxygen atom (from carbonyl) via coordination bonding which also enhanced the good dispersion of

MOFs in poly(BMA–EDMA) monolith. These observations conclude that the addition of MIL-101(Cr) (50 wt%) enhanced the crystallinity and the micro/meso porous nature, while the presence of poly(BMA–EDMA) provided large through-pores, to the produced MOF–polymer monolith.

Fig. 2. Powder XRD patterns (a), N2 adsorption/desorption isotherms (b), pore size distribution (c) and FT-IR spectra (d) of as synthesized MIL-101(Cr), MIL-101(Cr)–polymer monolith and neat polymer monolith.

100

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

(a) 50 40 30 20 10 0

Type of column

Surface area (m2 g−1 )

Pore size (nm)

Permeability, KF (cm2 )b

MIL-101(Cr) packed column MIL101(Cr)–polymer monolith Neat polymer monolith

2938a

1.97, 2.31, 3.06a

<2.8E−13

732

1.76, 1.99, 2.92

2.2E−10

None

1.2E−10

3

a

As synthesized MIL-101(Cr). b KF = (Fm ·  · L)/(p ·  · r2 )„ Fm : flow rate of mobile phase; : dynamic viscosity of mobile phase; p: pressure drop across the column; L: column length; r: column inner radius [45].

3.2. CEC applications of MOF packed column and MOF–polymer monolith MOF–polymeric monoliths prepared from BMA, EDMA and various amounts of MIL-101(Cr) (30–60 wt%) were developed as separation columns for CEC. In the fabrication of the MOF–polymer monoliths, the composition of the polymerization mixture for the organic polymer was kept constant (i.e. BMA–EDMA monomer ratio, porogenic solvent type (IL, [C6 min][BF4 ]), monomer–porogenic solvent ratio and charge monomer amount [35]). When the amount of MIL-101(Cr) was lower than 50 wt%, a baseline separation for the test analytes was not possible; while an increase of more than 50%, the MOF–monomer solution was hard to introduce into the capillary because of serious aggregation and precipitation. Therefore, the MOF–polymer monolith containing 50 wt% MIL-101(Cr) that provided the most stable baseline and excellent separation for most test analytes was used in the succeeding chromatographic separation. The retention behaviors of three xylene compounds (p-, m- and o-xylene) on the three different columns were compared. Their electrochromatograms derived from the same mobile phase (pH 6 phosphate buffer (5 mM) with 50% (v/v) acetonitrile) and operational conditions (except for the applied voltages, 15 kV for MOF packed column and 20 kV for MOF–polymer and poly(BMA–EDMA) monoliths) are shown in Fig. 3. Results indicated that both MIL101(Cr) packed column and MIL-101(Cr)–polymer monolith (7-cm and 25-cm effective lengths, respectively) provided baseline separations while no separation was achieved in neat poly(BMA–EDMA) monolith (25-cm effective length), which demonstrated the efficiency of MIL-101(Cr) as CEC stationary phase for xylene isomers separation. It is worth noting that all neutral xylenes were eluted in the MOF packed column (the elution order are the same with previous reports on GC and HPLC [21,22]) indicating a strong electroosmotic flow (EOF) produced in this neat MOF stationary phase. The EOF produced in the MIL-101(Cr) packed column was ascribed to the dissociation of free carboxylic acid from 1,4-terephthalic acid (the precursors used for MIL-101(Cr) construction) at the acidic mobile phase (pH 6). When the mobile phase’s pH was decreased to pH 4, no xylene or EOF signals were determined therefore, a neutral or basic mobile phase had to be used in order to produce enough EOF for the electrically driven CEC system. On the other hand, in the fabrication of the MOF–polymer monolith, a charge monomer whose purpose was to generate the EOF, was added to copolymerize with the neutral monomers making up the polymerization mixture, and is capable of producing adequate EOF in a wide range of mobile phase pH (for example, AMPS charge monomer used in the study produced a stable EOF from pH 2–12) allowing the use of a wide range of mobile phase pH

A b s o rb a n c e (m A U )

Table 1 Surface area, pore size and permeability of MIL-101(Cr) packed column, MIL101(Cr)–polymer monolith and neat polymer monolith.

(b) 50 40 30 20 10 0 (c)

1 2 3

2

4 T

8

10

12

1 2 3

2

4

6

8

10

12

8

10

12

T

200

1, 2, 3

100 0

6

2

4

6

Time (min) Fig. 3. CEC electrochromatograms of xylene isomers separated on MIL-101 packed column (a), MIL-101(Cr)–polymer monolith (b) and neat polymer monolith (c). Separation conditions: mobile phase, 5 mM phosphate (pH 6.0)/ACN = 50/50 (v/v); separation, 15 kV (a) and 20 kV (b and c); injection, 10 kV (a) and 5 kV (b and c) for 3 s; effective length, 7-cm (a) and 25-cm (b and c); analyte concentration, 250 ␮g mL−1 . T (thiourea), 1 (p-xylene), 2 (m-xylene), 3 (o-xylene).

without affecting the chromatographic behavior (Fig. S4). It can also be observed that the stronger EOF produced in neat polymeric column was due to the higher amount of AMPS molecules (5.9 wt%) attached to it while a lower AMPS amount (4.0 wt%) was used in MIL-101(Cr)–polymer preparation (Fig. 3(b) and (c)). These results showed that the MOF–polymer monolith synthesized within 10 min by microwave heating addressed not only the limitations posed by the complex fabrication of MOF capillary packed column requiring a retaining frit formed inside the capillary (∼15 h), as well as the time-consuming high-pressure packing of the MOF slurry (∼8 h), but also allowed the use of mobile phases regardless of pH conditions. To further evaluate the performance of the MOF–polymer monolith in the separation of isomers, chlorotoluene and cymene compounds were used as test analytes. Both MOF packed and MOF–polymer monolithic columns provided baseline separation for chlorotoluene isomers (Fig. S5), similar to xylene isomers separation; however, cymene compounds were indistinguishable from each other (Fig. 4(a)) as with other LC reports employing MOF packing material [18]. This scenario was changed when the effective length of the MOF–polymer monolith was increased from 25 cm to 40 cm wherein cymene isomers were efficiently separated (Fig. 4(b) and (c)), and this improved resolution was probably due to the presence of mesopores from the MOF and the existence of the large through-pores from the polymer monolith making possible the extended column length without back pressure limitation which was encountered in MOF packed column (7-cm effective length) (see Section 3.3). Moreover, decreased separation time of the analytes was observed in the MOF–polymer monolith with increased acetonitrile (ACN) content in the mobile phase and it exhibited a reversedphase separation mechanism for all isomeric analytes as evidenced by the linear behavior of the plots of their log k values when the ACN content was varied from 50 to 70%.

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

(a)

40

separation columns in nanoscale ultra-high pressure liquid chromatography (nano-UHPLC) however, at 5-cm length of the MOF packed column, the maximum pressure tolerance of the instrument was almost reached even at very low mobile phase flow rate of 50 nL min−1 (over 10,000 psi back pressure generated) which limits the application of MIL-101(Cr) packed column for nanoUHPLC application. On the other hand, this MOF packed column was successfully applied in CEC with even longer effective length (7 cm) showing that MOFs permeability (KF = 2.8 × 10−13 cm2 ) can only be tolerated with an electrically driven mobile phase flow as opposed to the pressure-driven flow in nano-UHPLC posing difficulty as to its usage. Contrastingly, better permeability was achieved when MOF–polymer monolith was tested in nanoUHPLC which can be operated at a 20-fold higher flow rate compared to MOF packed column (for example, the back pressures and the permeabilities (KF ) at 35-cm effective length of neat poly(BMA–EDMA) and MOF–polymer monolith were 4500- and 2400-psi and 1.2 × 10−10 - and 2.2 × 10−10 -cm2 , respectively at a flow rate of 1 ␮L min−1 which were similar to the results of previous methacrylate based monoliths [38,39]). In comparison to previous reports wherein nonporous particles embedded within polymer monolith resulted in a lower permeability, the permeability found in the MOF–polymer column, however, was increased most likely because MOF materials have plenty of micropores as well as very large pore volume in the structure, while both properties were absent in most nanoparticles [40,41]. To further emphasize the merits of MOF materials as a component in the fabrication of stationary phases for nano-UHPLC application, the chromatographic performance of MIL-101(Cr)–polymer and neat poly(BMA–EDMA) monoliths were compared in the separation of polycyclic aromatic hydrocarbons (PAHs) and aromatic acids test analytes. Elution order was the same for both stationary phases and their elution followed an increasing aromatic-ring number (Fig. S6). Comparison of the profiles in Fig. S6 indicated that the presence of MIL-101(Cr) in the polymeric monolith improved the peak symmetry as well as increased the resolution especially for PAHs with more aromatic rings (phenanthrene and pyrene, benzo[a]pyrene and benzo[ghi]perylene). This separation behavior could be ascribed to the high surface area, existence of bimodal pore sizes (meso- and nano-sized pores) (Table 1) and the presence of aromatic terephthalate moieties imparted by MIL-101(Cr) to the MOF–polymer hybrid column which enhanced the sieving effect, while the strong ␲–␲ interactions between PAHs and the MOFs organic linker, 1,4-terephthalate moieties (which is absent in

1, 2

30 20

3

10

Absorbance (mAU)

0 40

5

(b)

30

10

T

20

1, 2 3

10 0 40

5

(c)

30

10

15

1 2

T

20

3

10 0

5

10

15

Time (min)

20

25

Fig. 4. CEC electrochromatograms of cymene isomers separated on MIL-101 packed column (a), MIL-101(Cr)–polymer monolith (b and c). Separation conditions: mobile phase, 5 mM phosphate (pH 6.0)/ACN = 50/50 (v/v); separation, 15 kV (a), 20 kV (b) and 27 kV (c); injection, 10 kV (a and c) and 5 kV (b) for 3 s; effective length, 7-cm (a), 25-cm (b) and 40-cm (c); analyte concentration, 250 ␮g mL−1 . T (thiourea), 1 (m-cymene), 2 (p-cymene), 3 (o-cymene).

In summary, limitations as to the column preparation time, back pressures with longer columns and mobile phase conditions (e.g. solution pH) were overcame by employing MOF–polymer monolith stationary phase in CEC, as well as the significant enhancement of chromatographic performance contributed by the varied pore sizes (large through-pores and nano-sized pores) produced in one column. 3.3. Nano-UHPLC applications of MOF packed column and MOF–polymer monoliths To extend their application in chromatography, MOF packed column and MOF–polymer monolithic column were used as

60

(b)

Absorbance (mAU)

1

200

1

50 40

Absorbance (mAU)

(a)

2

30 20

3

4

10

101

0

2, 3, 4

150

100

50

0 0

5

10 Time (min)

15

20

0

5

10

15

20

Time (min)

Fig. 5. Nano-UHPLC chromatograms of aromatic acids separated on MIL-101(Cr)–polymer monolith (a) and neat polymer monolith (b). Separation conditions: mobile phase, 0.1% formic acid in water/0.1% formic acid in ACN = 20/80 (v/v); flow rate at 1 ␮L min−1 ; detection wavelength, 200 nm; injection volume, 500 nL; effective length, 35-cm; column temperature, 25 ◦ C; analyte concentration, 10 ␮g mL−1 . 1 (benzoic acid), 2 (terephthalaldehydic acid), 3 (isophthalic acid), 4 (terephthalic acid).

102

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

(a)

MIL-101(Cr)-poly(BMA-EDMA) monolith

250

benzoic acid terephthalaldehydic acid isophthalic acid terephthalic acid

HETP ( m)

200

150

100

50

0.5

1.0

1.5

(b)

Linear velocity, (mm/s)

80

MIL-101(Cr)-poly(BMA-EDMA) monolith poly(BMA-EDMA) monolith

3.4. Repeatability and reproducibility of MOF–polymer monoliths in CEC and UHPLC

HETP ( m)

70 60 50 40 30 0.0

the potential of this MOF–polymer monolith as nano-UHPLC stationary phase in the improvement of the ␲-electron rich solutes separation. van Deemter plots of plate height versus flow rate (aromatic acids and ethylbenzene were used as test probes in Fig. 6(a) and (b), respectively) showed that the efficiency (height equivalent to a theoretical plate, HETP) of MOF–polymer monolith did not diminish for all test aromatic acids (70–180-␮m for each) as flow rate is increased which could possibly allow for a reduction in the analysis time for these aromatic acids (Fig. 6(a)). Moreover, optimum efficiency (HETP around 43 ␮m) was reached for ethylbenzene in the same column at a flow rate of 0.3 ␮L min−1 (Fig. 6(b)). In contrast to previous reports on conventional MOF packed columns (5 cm × 4.6 mm I.D. with MIL-101(Cr) packing material) with 50–100-␮m HETP [22], these results emphasized that with MOF–polymer monolith a wide range of flow rates could be used without significant efficiency loss thus, better chromatographic performance could be obtained.

0.5

1.0

1.5

2.0

2.5

Linear velocity, (mm/s) Fig. 6. van Deemter curves of aromatic acids (a) and ethyl benzene (b) on MIL-101(Cr)–polymer monolith and/or poly(BMA–EDMA) monolith. Flow rate, 0.3–1 ␮L min−1 for (a) and 0.1–1 ␮L min−1 for (b). All other conditions are the same as in Fig. 5.

neat poly(BMA–EDMA) monolith), further improved the separation selectivity. In the nano-UHPLC separation of several aromatic acids (benzoic acid, terephthalic acid, isophthalic acid and 4carboxybenzaldehyde), these analytes were baseline separated on the MOF–polymer hybrid column whereas co-elution of analytes was observed in neat poly(BMA–EDMA) (Fig. 5). Usually, separation of these aromatic acids in HPLC used silica stationary phase modified by chemically bonding ␤-cyclodextrin while in CE a mobile phase with a multi-component microemulsion solution was needed to achieve good separation [42–44]. However, with our proposed MOF–polymer monolith all of these aromatic acids were efficiently separated without chemical modifications or use of complex mobile phase composition which demonstrated

The performance of the proposed MOF–polymer monolith used as stationary phase in CEC and nano-UHPLC system are shown in Table 2 and Table S2, respectively. In the CEC mode, intra-day and inter-day repeatability for retention time in the same column was less than 3.5% relative standard deviation (RSD) and peak area was less than 5.0% (n = 3). Column-to-column precision for retention time was less than 3.05% RSD and peak area was less than 5.43% (n = 3). Batch-to-batch precision for retention time was less than 4.33% and peak area was less than 5.20% (n = 3). These results reflect the good repeatability and reproducibility of MOF–polymer monolith in the analyses of all test analytes. 3.5. Nano-LC–MS2 separation of protein digests with MOF–polymer monolith as stationary phase Nowadays, various MS-based proteomics strategies have been developed to explore complex biological processes, disease biomarkers, drug deliveries, and so on. Among them “bottomup” MS-based proteomics is one generally applicable approach for protein identification, however, it encounters serious challenges because of the complexity in the separation of peptides. To examine the potential of MOF–polymer monolith in proteomics studies, this was used as stationary phase in the nano-LC separation of peptides from the trypsin digested BSA (50 fmol), coupled by mass spectrometer for detection (Fig. S7). Results verified using the BioTools 3.1 (Bruker Daltonik) and the Mascot database showed satisfactory sequence coverage of 64% with 46 matched peptides obtained from the MOF–monolith (Table S1), while a sequence coverage of 52% with 32 matched peptides and 74% with 51 matched

Table 2 Retention time, peak area repeatability and column efficiency of substituted aromatics on MIL-101(Cr)–polymer monolith in CEC. Analyte

p-Xylene m-Xylene o-Xylene p-Chlorotoluene m-Chlorotoluene o-Chlorotoluene m-Cymene p-Cymene o-Cymene

Intra-day (n = 3), %

Inter-day (n = 3), %

Column to column (n = 3), %

Batch to batch (n = 3), %

Time

Area

Time

Area

Time

Area

Time

Area

0.10 0.07 0.30 0.12 0.20 0.22 0.69 0.57 0.48

0.57 0.61 1.70 0.24 0.76 0.66 2.65 3.05 3.34

1.09 1.20 1.65 1.30 1.42 1.83 2.56 2.85 3.37

2.21 1.74 2.91 2.27 2.19 4.29 2.17 2.98 4.82

2.44 2.64 3.05 1.72 1.89 2.43 1.24 1.50 2.40

1.89 1.85 1.84 2.20 2.03 2.21 3.62 4.34 5.43

2.76 3.86 2.82 3.34 3.36 3.19 4.33 4.23 4.11

3.97 4.02 3.50 4.26 4.40 3.39 3.87 4.27 5.20

Column efficiency N (plates m−1 ) 37,000 32,000 22,000 37,000 32,000 24,000 52,000 39,000 22,000

H.-Y. Huang et al. / Analytica Chimica Acta 779 (2013) 96–103

peptides were attained in the separation of the same digests on neat poly(BMA–EDMA) monolith and C18 silica (3 ␮m) packed column (20 cm × 100 ␮m I.D.), respectively. These results demonstrated the feasibility of MOF–polymer monolith as stationary phase in the complex peptides separation. 4. Conclusion In this work, a novel MOF–polymer monolith prepared via a 5min microwave-assisted heating of methacrylate-based monomers (BMA and EDMA) and MIL-101(Cr) mixture was successfully developed as stationary phase for CEC, nano-UHPLC and nano-LC–MS2 . The inclusion of MIL-101(Cr) MOF provided very high surface areas, meso- and nano-sized pores and aromatic terephthalate moieties in the produced MOF–polymer monolith which caused high column permeability as well as increased separation performance for all test analytes (xylene, chlorotoluene, cymene, PAHs, aromatic acids and BSA trypsin digested peptides). This work demonstrated for the first time the potentiality of using MOF–polymer monolith composite as chromatographic stationary phases with the following merits: lower back pressure, higher separation performance and simpler preparation than capillary MOF packed column. These findings would provide great impact for future studies in the fields of MOF and chromatography. Acknowledgement This study was supported by both Grants NSC-100-2632M-033-001-MY3 and NSC-101-2113-M-033-002-MY3 from the National Science Council of Taiwan. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aca.2013.03.071. References [1] H. Li, M. Eddaoudi, M. O’Keeffe, O.M. Yaghi, Nature 402 (1999) 276–279. [2] G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I. Margiolaki, Science 309 (2005) 2040–2042. [3] M. Eddaoudi, D.B. Moler, H. Li, B. Chen, T.M. Reineke, M. O’Keeffe, O.M. Yaghi, Acc. Chem. Res. 34 (2001) 319–330. [4] G. Férey, C. Serre, Chem. Soc. Rev. 38 (2009) 1380–1399. [5] C. Janiak, J.K. Vieth, New J. Chem. 34 (2010) 2366–2388. [6] J.-R. Li, R.J. Kuppler, H.-C. Zhou, Chem. Soc. Rev. 38 (2009) 1477–1504. [7] J.S. Seo, D. Whang, H. Lee, S.I. Jun, J. Oh, Y.J. Jeon, K. Kim, Nature 404 (2000) 982–986. [8] W. Zhung, H. Sun, H. Xu, Z. Wang, S. Gao, L. Jin, Chem. Commun. 46 (2010) 4339–4341.

103

[9] M.D. Allendorf, C.A. Bauer, R.K. Bhakta, R.J.T. Houk, Chem. Soc. Rev. 38 (2009) 1330–1352. [10] P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T. Baati, J.F. Eubank, D. Heurtaux, P. Clayette, C. Kreuz, J.-S. Chang, Y.K. Hwang, V. Marsaud, P.-N. Bories, L. Cynober, S. Gil, G. Férey, P. Couvreur, R. Gref, Nat. Mater. 9 (2010) 172–178. [11] Z.-Y. Gu, C.-X. Yang, N. Chang, X.-P. Yan, Acc. Chem. Res. 5 (2012) 734–745. [12] F. Svec, J. Sep. Sci. 27 (2004) 747–766. [13] R. Skudas, B.A. Grimes, E. Machtejevas, V. Kudirkaite, O. Kornysova, T.P. Hennessy, D. Lubda, K.K. Unger, J. Chromatogr. A 1144 (2007) 72–84. [14] B. Chen, C. Liang, J. Yang, D.S. Contreras, Y.L. Clancy, E.B. Lobkovsky, O.M. Yaghi, S. Dai, Angew. Chem. Int. Ed. 45 (2006) 1390–1393. [15] L. Alaerts, C.E.A. Kirschhock, M. Maes, M.A. van der Veen, V. Finsy, A. Depla, J.A. Martens, G.V. Baron, P.A. Jacobs, J.F.M. Denayer, D.E. De Vos, Angew. Chem. Int. Ed. 46 (2007) 4293–4297. [16] V. Finsy, H. Verelst, L. Alaerts, D. De Vos, P.A. Jacobs, G.V. Baron, J.F.M. Denayer, J. Am. Chem. Soc. 130 (2008) 7110–7118. [17] L. Alaerts, M. Maes, P.A. Jacobs, J.F.M. Denayer, D.E. De Vos, Phys. Chem. Chem. Phys. 10 (2008) 2979–2985. [18] L. Alaerts, M. Maes, L. Giebeler, P.A. Jacobs, J.A. Martens, J.F.M. Denayer, C.E.A. Kirschhock, D.E. De Vos, J. Am. Chem. Soc. 130 (2008) 14170–14178. [19] C.-X. Yang, S.-S. Liu, H.-F. Wang, S.-W. Wang, X.-P. Yan, Analyst 137 (2012) 133–139. [20] S.-S. Liu, C.-X. Yang, S.-W. Wang, X.-P. Yan, Analyst 137 (2012) 816–818. [21] Z.-Y. Gu, X.-P. Yan, Angew. Chem. Int. Ed. 49 (2010) 1477–1480. [22] C.-X. Yang, X.-P. Yan, Anal. Chem. 83 (2011) 7144–7150. [23] C.-X. Yang, Y.-J. Chen, H.-F. Wang, X.-P. Yan, Chem. Eur. J. 17 (2011) 11734–11737. [24] Y.-Y. Fu, C.-X. Yang, X.-P. Yan, Langmuir 28 (2012) 6794–6802. [25] N. Chang, Z.-Y. Gu, X.-P. Yan, J. Am. Chem. Soc. 132 (2010) 13645–13647. [26] Z.-Y. Gu, J.-Q. Jiang, X.-P. Yan, Anal. Chem. 83 (2011) 5093–5100. [27] N. Chang, X.-P. Yan, J. Chromatogr. A 1257 (2012) 116–124. [28] R. Ameloot, A. Liekens, L. Alaerts, M. Maes, A. Galarneau, B. Coq, G. Desmet, B.F. Sels, J.F.M. Denayer, D.E. De Vos, Eur. J. Inorg. Chem. (2010) 3735–3738. [29] K. Tanaka, T. Muraoka, D. Hirayama, A. Ohnish, Chem. Commun. 48 (2012) 8577–8579. [30] C. Legido-Quigley, N.D. Marlin, V. Melin, A. Manz, N.W. Smith, Electrophoresis 24 (2003) 917–944. [31] G. D’Orazio, S. Rocchi, S. Fanali, J. Chromatogr. A 1255 (2012) 277–285. [32] M. Bedair, Z.E.I. Rassi, Electrophoresis 23 (2002) 2938–2948. [33] E.F. Hilder, F. Svec, J.M.J. Fréchet, J. Chromatogr. A 1044 (2003) 3–22. [34] Y.-H. Shih, B. Singco, W.-L. Liu, C.-H. Hsu, H.-Y. Huang, Green Chem. 13 (2011) 296–299. [35] B. Singco, C.-L. Lin, Y.-J. Cheng, Y.-H. Shih, H.-Y. Huang, Anal. Chim. Acta 746 (2012) 123–133. [36] C.-L. Lin, B. Singco, C.-Y. Wu, P.-Z. Liang, Y.-J. Cheng, H.-Y. Huang, J. Chromatogr. A 1272 (2013) 65–72. [37] D.-Y. Hong, Y.K. Hwang, C. Serre, G. Férey, J.-S. Chang, Adv. Funct. Mater. 19 (2009) 1537–1552. [38] J. Urban, P. Jandera, P. Schoenmakers, J. Chromatogr. A 1150 (2007) 279–289. [39] I. Nischang, O. Brüggemann, J. Chromatogr. A 1217 (2010) 5389–5397. [40] R.D. Arrua, A. Nordborg, P.R. Haddad, E.F. Hilder, J. Chromatogr. A 1273 (2013) 26–33. [41] W. Lei, L.-Y. Zhang, L. Wan, B.-F. Shi, Y.-Q. Wang, W.-B. Zhang, J. Chromatogr. A 1239 (2012) 64–71. [42] Y. Kawaguchi, M. Tanaka, N. Mitsugu, K. Funazo, T. Shono, Anal. Chem. 55 (1983) 1852–1857. [43] M. Zedda, J. Tuerk, T. Teutenberg, S. Peil, T.C. Schmidt, J. Chromatogr. A 1216 (2009) 8910–8917. [44] H.-Y. Huang, Y.-R. Lin, S.-H. Hsieh, Anal. Chim. Acta 632 (2009) 148–155. [45] J. Urban, P. Jandera, P. Langmaier, J. Sep. Sci. 34 (2011) 2054–2062.