Silica Monolith Nested in Sponge (SiMNS): A Composite Monolith as a

1 day ago - ... detection limits (MDLs) of 0.02–1.3 ng/L and method quantification limits (MQLs) of 0.05–4.3 ng/L. Successful identification and q...
0 downloads 0 Views 3MB Size
Download PDF
Article Cite This: Anal. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/ac

Silica Monolith Nested in Sponge (SiMNS): A Composite Monolith as a New Solid Phase Extraction Material for Environmental Analysis Zhongshan Liu, Ping Jiang, Guang Huang, Xiaowen Yan, and Xing-Fang Li* Division of Analytical and Environmental Toxicology, Department of Laboratory Medicine and Pathology, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta T6G 2G3, Canada

Anal. Chem. Downloaded from pubs.acs.org by WEBSTER UNIV on 02/13/19. For personal use only.

S Supporting Information *

ABSTRACT: We report a new material of a composite silica monolith nested in sponge (SiMNS) and demonstrate an application in the trace analysis of environmental contaminants in water. SiMNS is prepared through sponge absorption of a hydrolyzed mixture of siloxanes and in situ gel formation within the pores. Images obtained using scanning electron microscopy show that the silica and sponge skeletons are mutually nested in SiMNS. This nested composite structure of SiMNS enhances the mechanical flexibility of the material, allowing for reproducible production of desirable sizes and shapes for solid phase extraction (SPE) cartridges without the need to use frits. Functionalization of SiMNS provides appropriate SPE options for selective and efficient extraction of specific contaminants. SPE cartridges packed with functionalized SiMNS−SO3Na have high extraction capacity, good stability in the pH range of 2 to 11, and efficient enrichment of dipeptides in water. Extraction of six dipeptides from water using these new SiMNS−SO3Na SPE cartridges followed by HPLC-MS/MS analysis results in improved method detection limits (MDLs) of 0.02−1.3 ng/L and method quantification limits (MQLs) of 0.05−4.3 ng/L. Successful identification and quantification of three dipeptides, Tyr−Gly, Phe−Gly, and Tyr− Ala, from raw water demonstrates a useful application of the new SPE materials for environmental analysis of trace contaminants. On the basis of this work, a range of functionalized SiMNS materials can be produced and tailored for various environmental and exposomic analyses.

S

samples.5,6,17,22,23 However, the application of silica-monolith SPE cartridges in environmental analysis is limited. The main reason is that analysis of environmental samples, such as water, generally requires large-size SPE cartridges or columns for processing large-volume water samples. However, the preparation of large silica-monolith SPE cartridges is extremely difficult because of the shrinkage and brittleness of the silica monolith.24 For instance, to obtain a silica monolith with a suitable diameter for packing inside the cartridge, Nema et al. had to optimize a mold size by repetitive experiments.3 Because of the shrinkage of the silica monolith, the mold had to be slightly bigger than the internal diameter of the empty cartridge. The preparation conditions required stringent control and are difficult to repeat. The primary objective of this research is to develop analytically useful monolithic materials that maintain the advantages of silica monoliths and overcome the problems of producing large-size monolithic SPE cartridges for environmental analyses. We hypothesize that new nested monolithsponge materials can have tunable surfaces and the unique mechanical flexibility necessary for the production of various

olid phase extraction (SPE) is one of the most widely used techniques for sample preparation because of its ability to concentrate analytes at trace and ultratrace levels for quantification and to remove matrix interference.1−3 A desirable SPE cartridge should provide efficient retention and selectivity and have a high capacity for target analytes. Existing SPE cartridges are mostly prepared by packing particle sorbents.4 Particle sizes of 30−105 μm are usually utilized to balance the diffusional mass transfer of analytes and the back pressure of cartridges.3,5 However, large-size particles often create nonuniform packing and interparticle voids that are detrimental to extraction performance.6 The use of silica monoliths as chromatographic stationary phases has gained popularity in microscale liquid chromatographic separations over the past two decades.7−13 Compared with particle-based materials, silica monoliths possess the features of uniform through-pores and large surface areas; thus, capillary monolithic columns enable rapid and efficient separation at relatively low back pressures. Importantly, silica monoliths can be tuned with various surface properties, such as reversed phase,14 hydrophilic interactions, 15,16 ion exchange,17,18 boronate affinity,19 chirality, and molecularimprinting recognition sites. 20,21 In addition to their applications in chromatographic separations, silica monoliths have been fabricated in narrow capillaries, tips, and smallsyringe SPE cartridges for selective extraction of biological © XXXX American Chemical Society

Received: December 11, 2018 Accepted: January 28, 2019

A

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry Scheme 1. Preparation of SiMNSa

a

(i) Hydrolysis. (ii) Gel formation. (iii) Adjustment of nanopore size.

(2.7 g) were added to an aqueous solution of acetic acid (0.01 M, 30 mL). The mixture was stirred in an ice-water bath for 1 h and then absorbed by the sponge. In step ii, the saturated sponge was maintained in the incubator at 55 °C for 12 h, and in step iii, the temperature was increased to 80 °C for 8 h. The resulting composite SiMNS was washed with water and cut into cylindrical membranes with diameters of 15 mm and thicknesses of ∼10 mm. As a control, a pure silica monolith (PSM) was prepared using the same steps (i−iii) without incorporation of the sponge. Surface Functionalization of SiMNS. Eight SiMNS membranes were placed into a 50 mL flask with methanol (20 mL), VTMS (3.8 mL), and TEA (1.75 mL). After reaction under reflux for 12 h, the SiMNS membranes were washed with methanol (30 mL) three times. Then, a solution of MPS (1.0 g) in methanol/water (24 mL; 3:1, v/v) and AIBN (0.1 g) were added. The thiol−ene click reaction of MPS with the vinyl groups on the SiMNS surface was carried out at 60 °C for 5 h. The sulfonate-functionalized monolith, denoted as SiMNS−SO3Na, was obtained after washing with water (30 mL) five times. Characterization. The morphology study was carried out on a field-emission scanning electron microscope (FESEM, Zeiss, Oberkochen, Germany). For thermogravimetric analysis (TGA), samples were heated at 10 °C/min in air using a Discovery TGA instrument (TA Instruments, Waters). The macropore diameter of the SiMNS was determined by a mercury-intrusion porosimeter (MIP; Quantachrome Instruments, Boynton Beach, FL). Nitrogen adsorption−desorption measurements were performed on an Autosorb iQ (Quantachrome Instruments). Samples were outgassed under vacuum at 100 °C for 4 h before measurement. The surface area was calculated using the Brunauer−Emmett−Teller (BET) method. The nanopore size was determined by the NLDFT approach. FT-IR spectra were collected on a Nicolet iS50 FTIR spectrometer with attenuated-total-reflection mode (Thermo Fisher Scientific, Waltham, MA). SPE Method. The SiMNS−SO3Na membranes (13 mm in diameter and 3 mm thick) were packed into empty syringe cartridges. The SPE process was performed using a Supelco vacuum manifold. Briefly, SiMNS−SO3Na SPE cartridges were sequentially rinsed with methanol (2 mL) and acidic water with FA (0.25%, v/v; 4 mL) prior to use. Standard solutions containing seven dipeptides in Optima water (450 mL, each at 6 μg/L) passed through the prepared cartridges at a flow rate

sizes of SPE cartridges for environmental analyses. Our approach of synthesizing the new nested monolith sponge material utilizes the melamine-formaldehyde (MF) sponge as a skeleton to stabilize the silica-monolith sorbents, yielding a new type of silica monolith nested in sponge (SiMNS). This approach is built on the previous success of using the surfacemodified MF sponge for oil−water separation.25−27 The high porosity of the MF sponge offers large-volume loading of oil or water, while the flexible framework keeps its shape and stability even after repeated compression tests. Taking advantage of its high porosity and mechanical flexibility, we have filled the MF sponge with a hydrolyzed mixture of tetramethoxysilane (TMOS) and vinyltrimethoxysilane (VTMS). The gel formation within the sponge pores results in the generation of the new SiMNS material. We report here the synthesis of SiMNS, surface functionalization of SiMNS, construction of SPE cartridges with the functionalized SiMNS materials, and an application in water analysis. Using the new SPE with functionalized SiMNS materials, we demonstrate the selective extraction of seven peptides and two pharmaceuticals, acetaminophen (AAP) and 4,4′-sulfonyldiphenol (BPS), and subsequent analysis using high-performance liquid chromatography (HPLC) and electrospray-ionization tandem mass spectrometry (ESI-MS/MS).

■

EXPERIMENTAL SECTION Chemicals and Materials. TMOS, VTMS, urea, polyethylene glycol (PEG, Mn = 10 000), 3-mercapto-1-propanesulfonic acid sodium salt (MPS), α,α′-azoisobutyronitrile (AIBN), triethylamine (TEA), formic acid (FA), Tyr−Gly, Tyr−Ala, Gly−Ala, Phe−Gly, Tyr−Val, AAP, and BPS were purchased from Sigma-Aldrich (St. Louis, MO). 3-Iodo-Tyr− Ala (3-I-Tyr−Ala) and 3,5-di-iodo-Tyr−Ala (3,5-dI-Tyr−Ala) were obtained from the Chinese Peptide Company (Hangzhou, China). Optima water, methanol, acetic acid, and acetonitrile (ACN) were purchased from Fisher Scientific (Fair Lawn, NJ). Sep-Pak C18 cartridges (3 mL, 200 mg of sorbent), Oasis HLB cartridges (6 mL, 200 mg of sorbent), and MCX cartridges (6 mL, 150 mg of sorbent) were obtained from Waters (Milford, MA). Empty syringe cartridges (6 mL) were purchased from Agilent Technologies (Santa Clara, CA). MF sponges (RioRand) were purchased from Amazon. Preparation of SiMNS. Scheme 1 shows the three steps required to prepare SiMNS. In step i, solutions of TMOS (10.8 mL) and VTMS (3.6 mL) and solids of PEG (3.0 g) and urea B

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry Table 1. Adsorption−Breakthrough Data for Tyr−Val on SiMNS−SO3Na SPE Cartridges SiMNS−SO3Na SPE cartridge

SiMNS−SO3Na membrane weight (g)a

breakthrough volume (L)b

adsorption capacity (mg/g)b

maximum adsorption capacity (mg/g)c

untreated HCl (pH 2) treated NaOH (pH 11) treated NaOH−HCl treated

0.101 0.090 0.092 0.106

0.10 0.12 d 0.14

27 27 d 30

57 65 41 70

a

The SiMNS−SO3Na membrane weights were controlled at 90−110 mg during manual production and packing. bThe breakthrough volumes and adsorption capacities were determined at 10% Cfiltrate/Cstock. cThe maximum adsorption capacities were determined at 100% Cfiltrate/Cstock. dNot calculated because Cfiltrate/Cstock was always larger than 10% (Figure 4).

concentration levels (1, 3, 5, 10, and 20 ng/L) into the filtered raw water samples (450 mL, with 0.25% FA, v/v). Additionally, triplicate unspiked raw water samples were analyzed as the authentic samples. After SPE, the eluent was reconstituted to a final volume of 1 mL and analyzed using the HPLC-MS/MS (MRM) method as follows. HPLC-MS/MS (MRM) Method. HPLC separations were performed on an Agilent 1290 series LC system equipped with a binary pump, an autosampler with temperature control, and a Luna C18(2) column (100 × 2.0 mm i.d., 3 μm particles; Phenomenex, Torrance, CA). The autosampler was kept at 4 °C, and the injection volume of each sample was 20 μL. Mobile phases A and B were water (0.1% FA) and ACN (0.1% FA), respectively. The flow rate was set at 170 μL/min, with a gradient elution increasing from 5 to 70% mobile phase of B over 15 min. A triple-quadrupole ion-trap tandem mass spectrometer (SCIEX QTRAP 5500) was coupled with the HPLC to perform MS/MS (MRM) quantification of the peptides. The MRM transition ions of the seven dipeptides, AAP, and BPS are described in Table S1. The MS parameters were optimized as follows: the ion-spray voltage was 5500 V, the source temperature was 500 °C, gas 1 was at 45 arbitrary units, gas 2 was at 40 arbitrary units, the curtain gas was at 30 arbitrary units, and the accumulation time for each ion pair was 50 ms. System control and data collection were done with Analyst software (version 1.5.2, AB SCIEX, Framingham, MA).

of ∼2−3 mL/min. After the cartridges were washed with Optima water (2 mL), the dipeptides were eluted with ammonium hydroxide solution (5 wt % in methanol, 10 mL). The eluent was concentrated to 100 μL under a gentle nitrogen stream (20−50 KPa) for ∼2.5 h and then reconstituted with Optima water to 2 mL. The dipeptide concentrations in standard solutions and in the eluents were determined using HPLC-MS/MS with the multiple reaction monitoring (MRM) method described in the following section. The recovery was calculated using the following equation: recovery =

CeluentVeluent × 100% C0V0

(1)

where C0 (μg/L) and Celuent (μg/L) are the dipeptide concentrations in the solutions and the eluents, respectively. V0 (L) and Veluent (L) are the corresponding volumes. The adsorption capacity of the SiMNS−SO3Na SPE cartridge was evaluated using breakthrough experiments. A stock solution of Tyr−Val (Cstock = 28 mg/L) flowed continuously through the cartridge after equilibration with water (4 mL). The filtrates were collected, each at 10 mL. The Tyr−Val concentrations (Cfiltrate, mg/L) in the filtrates were quantified using a UV−vis spectrophotometer at 223 nm. The breakthrough curve was obtained by plotting Cfiltrate/Cstock versus the loading volume. To test the stability of the SiMNS−SO3Na SPE cartridges, three cartridges were treated with 10 mL of aqueous solutions of HCl (pH 2), NaOH (pH 11), and NaOH (pH 11) followed by HCl (pH 2). Breakthrough curves of these three treated cartridges were obtained using the same procedure described above. The adsorption capacities (mg/g, loading content of Tyr−Val per gram of SiMNS−SO3Na membrane) were calculated by the following equation: adsorption capacity =

∑ (Cstock − Cfiltrate)Vfiltrate m

■

RESULTS AND DISCUSSION Unique Features of SiMNS. Scheme 1 describes the procedure for the preparation of SiMNS. In step i, TMOS and VTMS were hydrolyzed in an ice-water bath and then absorbed by the MF sponge. The gel formation occurred within the sponge pores. In steps ii and iii, the formed silicasponge material was incubated at 55 °C for 12 h and then at 80 °C for 8 h. The synthesis process is simple and suitable for the preparation of large size monoliths. To characterize the SiMNS features, we examined the morphological characteristics of the sponge and SiMNS using SEM. In the sponge, the macropore size ranged from 100 to 200 μm, and the skeleton size was ∼5 μm (Figure 1a). The high size ratio of the macropores to the skeleton generated high porosity and offered space to inlay a monolithic silica matrix. Figure 1b clearly shows the uniform silica monolith throughout the three-dimensional pores of the sponge. The sponge skeleton (marked by arrows in Figures 1b and S1) was embedded through the whole silica monolith. This composite network structure was also supported by the energy dispersive X-ray spectroscopy (EDX)-mapping images (Figure 1c,d). The silicon signal was derived from the silica monolith, while the nitrogen distribution was derived from the sponge skeleton (Figure S2). In short, the silica and the sponge skeleton were mutually nested, similar to reinforced concrete.

(2)

where Vfiltrate (0.01 L) is the volume collected of each filtrate, and m (g) is the SiMNS−SO3Na membrane weight. On the basis of the breakthrough curves (Figure 4), we determined the breakthrough volumes and the adsorption capacities at 10% Cfiltrate/Cstock, as well as the maximum adsorption capacities at 100% Cfiltrate/Cstock (Table 1). Application for the Analysis of Raw Water. Raw water samples from the North Saskatchewan River were collected on March 29, 2018. The water samples were filtered using 1.5 μm glass-microfiber filters (Whatman) followed by 0.45 μm nylonmembrane disk filters and then stored at 4 °C before analysis. The filtrations were necessary to remove particles that could cause blockage of the SPE cartridge during extraction. To evaluate the SiMNS−SO3Na SPE cartridges for the extraction of analytes at trace levels, we spiked dipeptides at five different C

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 1. SEM images of (a) MF sponge and (b) SiMNS. (c,d) EDXmapping images of SiMNS: (c) silicon and (d) nitrogen. The sponge skeleton in SiMNS is marked by arrows in (b). Additional SEM images are shown in Figure S1 in the Supporting Information.

Figure 3. Photos of (a) SiMNS with different shapes and (b) Oasis HLB and SiMNS-packed cartridges. (c) Preparation of SiMNS− SO3Na: (i) treatment with VTMS to improve vinyl group density and (ii) reaction with MPS via thiol−ene click reaction.

To some extent, the flexibility of the sponge prevents the SiMNS from cracking when different sizes are made. Because the sponge is a pure organic polymer, we used TGA to determine its mass fraction in SiMNS. It was assumed that the residual should only be inorganic silica oxide. As shown in Figure S3, a weight loss of 15% occurred for the PSM, arising from vinyl groups in the starting monomer VTMS. The percentage of the organic moiety in SiMNS was increased to 23% after incorporation of the sponge. Therefore, the sponge accounted for about 9.4% of the total weight of SiMNS (for the detailed calculation, see the Supporting Information). In other words, SiMNS was mainly composed of the silica

monolith, allowing it to maintain a rigid structure while providing enhanced mechanical flexibility. The porosity of SiMNS after incorporation of the sponge was investigated using MIP and nitrogen adsorption− desorption measurements. As shown in Figure 2a, the macropore diameters of SiMNS ranged from 1 to 5 μm, which are larger than the macropore diameters of the PSM (1−2 μm). Silica monoliths usually demonstrate a hierarchically porous structure, and the nanopores in the skeleton contributed to a total surface area of 774 m2/g for the PSM (Figure S4a). In comparison, the surface area for SiMNS declined to 570 m2/g (Figure 2b). This decrease in surface area is attributable to the incorporation of the sponge, as there

Figure 2. (a) Macropore diameters of PSM and SiMNS as determined by MIP. (b) Nitrogen adsorption−desorption isotherm and (c) nanopore sizes of SiMNS. (d) Porosity and surface area comparisons between SiMNS and particle sorbents packed in several commercial SPE cartridges. D

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 4. Breakthrough curves for the adsorption of the Tyr−Val solution (28 mg/L) using an untreated SiMNS−SO3Na cartridge and three cartridges treated with 10 mL aqueous solutions of HCl (pH 2), NaOH (pH 11), and NaOH (pH 11) followed by HCl (pH 2).

are no nanopores present in the sponge. According to Figure 2c, the nanopore size of SiMNS was 4.8 nm, similar to that of the PSM (Figure S4b). We then compared the porosity of SiMNS with those of the particle sorbents packed in several commercial SPE cartridges (Figure 2d). SiMNS has a higher surface area than most of the particle sorbents, and the nanopore size in SiMNS is close to those of the Bond Elut C18, Inert Sep C18, and Empore C18 sorbents. SiMNS SPE Cartridges. Taking advantage of the excellent mechanical flexibility, a large-size bulk SiMNS can be cut into desired shapes and sizes, such as solid or hollow cylinders and films (Figure 3a). The SiMNS membrane at a thickness of 1 mm maintains its high mechanical stability. To perfectly fit SiMNS inside a syringe cartridge commonly used in SPE, a SiMNS membrane was made with a diameter of 13.0 mm and thickness of 3.0 mm using a hole punch. Compared with particle-packed cartridges, such as Oasis HLB, the SiMNSpacked cartridge was easier to prepare and did not require frits (Figure 3b). This new process dramatically simplifies the production of a large number of SiMNS SPE cartridges. We can also tune the surface properties of SiMNS for specific applications. To demonstrate this potential, we functionalized SiMNS with a short-chain sulfonic group for extraction of small peptides in water. Small hydrophilic peptides are poorly removed during water-treatment processes and can serve as precursors of disinfection byproducts (DBPs) of toxicological relevance.28−31 We previously observed low extraction efficiencies for the peptides using commercial cartridges, impeding the detection and identification of these DBP precursors in water samples.32 On the basis of interactions between the free amine groups of the peptides and sulfonate groups, we have functionalized SiMNS with MPS via the thiol−ene click reaction (SiMNS−SO3Na, Figure 3c). The surface modification was confirmed by FT-IR (Figure S5). For pristine SiMNS, the peak at 1412 cm−1 was ascribed to the in-plane bending vibration of C−H (CH2), whereas the peak at 1603 cm−1 represented the stretching vibration of the CC bonds in the vinyl groups. For SiMNS−SO3Na, new absorption bands at 2853 and 2925 cm−1 corresponded to C−

Figure 5. (a) Recovery of dipeptides on Sep-Pak C18, Oasis HLB, Oasis MCX, and SiMNS−SO3Na SPE cartridges. (b) Recovery of phenols on Oasis MCX, SiMNS−SO3Na, and CD-SiMNS cartridges. For the chemical structures of the dipeptides, see Figure S6; the experimental conditions are listed in Table S2 in the Supporting Information. Error bars represent standard deviations of the means of triplicate extractions.

H (−CH2−)-stretching vibrations, derived from the MPS moiety. After developing SiMNS−SO3Na SPE cartridges, we examined their repeatability from cartridge to cartridge and from batch to batch using the relative standard deviation (RSD) of Phe−Gly recoveries. The RSD values of the recoveries obtained from the cartridge-to-cartridge (n = 8) and batch-to-batch (n = 5) experiments were 2 and 3%, respectively, demonstrating the reproducible production of the SiMNS SPE cartridges. To evaluate the adsorption capacity of the SiMNS−SO3Na SPE cartridges, we used Tyr−Val (28 mg/L) as a probe and obtained its adsorption breakthrough curves, as shown in Figure 4. The adsorption capacity for the untreated SiMNS− SO3Na SPE cartridge was 27 mg/g, determined at 10% Cfiltrate/ Cstock in the breakthrough curve. The maximum adsorption capacity was 57 mg/g at 100% Cfiltrate/Cstock (Table 1). The HCl-treated SiMNS−SO3Na SPE cartridge exhibited a similar breakthrough volume and adsorption capacity to those of the untreated cartridge. The NaOH-treated cartridge had a poor E

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 6. Selected ion chromatograms of dipeptides in unspiked and spiked raw water after extraction using SiMNS−SO3Na cartridges. The concentrations of dipeptides spiked in the raw-water samples are 5−20 ng/L.

Table 2. Performance of the SiMNS−SO3Na SPE−HPLC-MS/MS Method for the Six Dipeptides dipeptide

retention time (min)

LODa (μg/L)

LOQa (μg/L)

MDLb (ng/L)

MQLb (ng/L)

unspiked raw waterc (ng/L)

Tyr−Gly Phe−Gly Tyr−Val Tyr−Ala 3-I-Tyr−Ala 3,5-dI-Tyr−Ala

2.6 4.1 4.0 2.5 4.6 5.7

0.05 0.04 0.02 0.05 0.03 0.01

0.09 0.06 0.03 0.09 0.06 0.02

1.3 0.07 0.4 1.3 0.07 0.02

4.3 0.2 1.1 4.3 0.3 0.05

5.8 ± 0.8 1.2 ± 0.5 d 7.3 ± 2.8 d d

a The limits of detection (LODs) and limits of quantification (LOQs) were calculated for the HPLC-MS/MS method (without SPE). The average (Sblank) and standard deviation (σblank) of peak area of the blank were calculated through triplicate analysis of Optima water (blank). The LOD was determined as the concentration of the standard that gives a peak area equal to Sblank + 3σblank. The LOQ was determined as the concentration of the standard that gives a peak area equal to Sblank + 10σblank. bThe MDL and MQL were obtained from SiMNS−SO3Na SPE−HPLC-MS/MS analysis of raw water samples containing each dipeptide at 1, 3, 5, 10, and 20 ng/L. The MDL and MQL were calculated as 3 and 10 times the standard deviation of the method blank signal divided by the slope, respectively. cThe concentrations of detected dipeptides in unspiked raw water samples were calculated as the intercepts divided by the slopes (Table S3), and standard deviations were determined through propagation of uncertainty. dNot detected.

provided recoveries ranging from 64 to 97% for the peptides tested. We further examined the selectivity of our SiMNS−SO3Na for the extraction of small peptides and compared the recoveries with those obtained using the MCX cartridge. Because most of the test peptides contained tyrosine (Tyr) and phenylalanine (Phe), we chose AAP and BPS as interfering compounds. Figure 5b shows that our SiMNS−SO3Na did not suffer interference from AAP and BPS, as the recoveries of the interfering compounds were as low as 0.03 and 3.6%. In contrast, AAP and BPS are well-retained on MCX cartridges, with recoveries of 55% for AAP and 72% for BPS, demonstrating possible interference in the analysis of small peptides. The difference in selectivity of SiMNS−SO3Na and MCX is due to their surface chemistry. The benzene rings and pyrrolidone moieties in the MCX polymeric surface provide π−π and hydrogen-bonding interactions with phenols (Figure S7). For the SiMNS−SO3Na, short-chain sulfonate groups effectively avoid retentions of these interfering compounds. These results demonstrate that SiMNS−SO3Na has high selectivity for small dipeptides. This high selectivity is needed

loading capability, because under strongly basic conditions, SiMNS−SO3Na remains instead of transforming into SiMNS− SO3H, which is required for the interaction with Tyr−Val. After the cartridge was washed with water and HCl (pH 2) solution, the adsorption capability of the NaOH−HCl-treated cartridge were regenerated. These results demonstrated that SiMNS−SO3Na SPE cartridges are stable at pH 2 to 11. For the enrichment of peptides, acidic solution preconditioning is required for achieving optimum extraction of the analytes. Environmental Application of SiMNS SPE Cartridges. We demonstrated an application of the SiMNS−SO3Na SPE cartridges in the analysis of small peptides, as shown in Figure 5a. These new cartridges can provide highly efficient extractions of dipeptides and halogenated dipeptides with recoveries of 100% for five of the seven peptides, 59% for Tyr− Gly, and 29% for Gly−Ala. The reproducibility of these cartridges is excellent, as demonstrated by the small error bars (2−9%, n = 3). In comparison, commercial C18 and Oasis HLB cartridges showed recoveries lower than 5% for most dipeptides (Figure 5a). The performance of the new SiMNS− SO3Na was comparable to that of the MCX cartridge, which F

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry to selectively concentrate the peptides at low levels in real water samples. To demonstrate the application of the SiMNS−SO3Na SPEs for trace analysis, we have performed additional extractions of raw-water samples spiked with 1−20 ng/L Tyr−Gly, Phe−Gly, Tyr−Val, 3-I-Tyr−Ala, and 3,5-dI-Tyr−Ala and 5−20 ng/L Tyr−Ala. The extracts were analyzed using HPLC-MS/MS, as shown in Figure 6. As summarized in Table S3, the relationship of peak area versus concentration tested is linear. Table 2 shows that the MDLs are 0.02−1.3 ng/L, and the MQLs range from 0.05 to 4.3 ng/L. Additionally, Tyr−Gly, Phe−Gly, and Tyr−Ala were detected in unspiked raw water (Figure 6), and their concentrations were determined to be 1.2 ± 0.5 ng/L for Phe−Gly, 7 ± 3 ng/L for Tyr−Ala, and 6 ± 0.8 ng/L (estimated) for Tyr−Gly from triplicate extractions and HPLC-MS/MS analysis. To our knowledge, this is the first time a sponge was used to stabilize silica-monolith sorbents. We further demonstrated that the method described in this work can be easily used to produce other SiMNS cartridges for different applications. For example, to selectively extract BPS, we replaced the mixture in Scheme 1 with another prepolymerization mixture composed of VTMS, TMOS, and a β-cyclodextrin (CD) derivative20 to produce a composite monolith denoted as CD-SiMNS (Figure S8). The CD-SiMNS cartridges were able to extract BPS with 100% recovery without extracting AAP or dipeptides (Figures 5b and S9). In addition to its superior performance, we also evaluated the cost to produce SiMNS cartridges. It costs approximately US$16.00 to produce a batch of 20 cartridges (Table S4). We can significantly reduce this cost when the production is scaled up to large numbers. As such, the production of a variety of SiMNS cartridges for different applications is both feasible and cost-effective.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel.: 1-780-492-5094. ORCID

Xing-Fang Li: 0000-0003-1844-7700 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Financial support from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the University of Alberta is acknowledged.

■

REFERENCES

(1) Pichon, V. J. Chromatogr. A 2000, 885, 195−215. (2) Płotka-Wasylka, J.; Szczepańska, N.; de la Guardia, M.; Namieśnik, J. TrAC, Trends Anal. Chem. 2015, 73, 19−38. (3) Nema, T.; Chan, E. C. Y.; Ho, P. C. Talanta 2010, 82, 488−494. (4) Huck, C. W.; Bonn, G. K. J. Chromatogr. A 2000, 885, 51−72. (5) Namera, A.; Saito, T. TrAC, Trends Anal. Chem. 2013, 45, 182− 196. (6) Nema, T.; Chan, E. C. Y.; Ho, P. C. J. Sep. Sci. 2011, 34, 1041− 1046. (7) Nakanishi, K.; Tanaka, N. Acc. Chem. Res. 2007, 40, 863−873. (8) Guiochon, G. J. Chromatogr. A 2007, 1168, 101−168. (9) Núñez, O.; Nakanishi, K.; Tanaka, N. J. Chromatogr. A 2008, 1191, 231−252. (10) Ou, J.; Liu, Z.; Wang, H.; Lin, H.; Dong, J.; Zou, H. Electrophoresis 2015, 36, 62−75. (11) Tanaka, N.; McCalley, D. V. Anal. Chem. 2016, 88, 279−298. (12) Svec, F.; Lv, Y. Anal. Chem. 2015, 87, 250−273. (13) Hara, T.; Futagami, S.; De Malsche, W.; Eeltink, S.; Terryn, H.; Baron, G. V.; Desmet, G. Anal. Chem. 2017, 89, 10948−10956. (14) Wu, C.; Liang, Y.; Yang, K.; Min, Y.; Liang, Z.; Zhang, L.; Zhang, Y. Anal. Chem. 2016, 88, 1521−1525. (15) Horie, K.; Kamakura, T.; Ikegami, T.; Wakabayashi, M.; Kato, T.; Tanaka, N.; Ishihama, Y. Anal. Chem. 2014, 86, 3817−3824. (16) Lin, H.; Ou, J.; Zhang, Z.; Dong, J.; Wu, M.; Zou, H. Anal. Chem. 2012, 84, 2721−2728. (17) Dong, M.; Wu, M.; Wang, F.; Qin, H.; Han, G.; Dong, J.; Wu, R. a.; Ye, M.; Liu, Z.; Zou, H. Anal. Chem. 2010, 82, 2907−2915. (18) Chen, M.; Zhang, J.; Zhang, Z.; Yuan, B.; Yu, Q.; Feng, Y. J. Chromatogr. A 2013, 1284, 118−125. (19) Lin, Z.; Pang, J.; Yang, H.; Cai, Z.; Zhang, L.; Chen, G. Chem. Commun. 2011, 47, 9675−9677. (20) Zhang, Z.; Wu, M.; Wu, R.; Dong, J.; Ou, J.; Zou, H. Anal. Chem. 2011, 83, 3616−3622. (21) Ou, J.; Li, X.; Feng, S.; Dong, J.; Dong, X.; Kong, L.; Ye, M.; Zou, H. Anal. Chem. 2007, 79, 639−646. (22) Liu, J.; Wang, F. J.; Lin, H.; Zhu, J.; Bian, Y. Y.; Cheng, K.; Zou, H. F. Anal. Chem. 2013, 85, 2847−2852. (23) Miyazaki, S.; Morisato, K.; Ishizuka, N.; Minakuchi, H.; Shintani, Y.; Furuno, M.; Nakanishi, K. J. Chromatogr. A 2004, 1043, 19−25. (24) Hayase, G.; Kanamori, K.; Abe, K.; Yano, H.; Maeno, A.; Kaji, H.; Nakanishi, K. ACS Appl. Mater. Interfaces 2014, 6, 9466−9471. (25) Ruan, C.; Ai, K.; Li, X.; Lu, L. Angew. Chem., Int. Ed. 2014, 53, 5556−5560. (26) Nguyen, D. D.; Tai, N.; Lee, S.; Kuo, W. Energy Environ. Sci. 2012, 5, 7908−7912. (27) Pham, V. H.; Dickerson, J. H. ACS Appl. Mater. Interfaces 2014, 6, 14181−14188.

■

CONCLUSIONS In summary, we have demonstrated a simple and efficient method to construct SiMNS SPE cartridges with various properties for specific applications. These new composite SiMNS monoliths not only maintain high surface areas and tunable surface properties but also provide enhanced mechanical flexibility compared with pure silica monoliths. The ability to produce large-size monoliths with flexible shaping property broadens their applications for SPE. As demonstrated in this paper, we can take advantage of the tunability of SiMNS to produce a variety of SiMNS SPE cartridges for the analysis of different compounds and broaden their applications for the analysis of environmental, food, and biological samples.

■

SPE conditions, calibration-curve parameters, and material-cost analysis (PDF)

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.8b05707. Additional experimental procedures, SEM images, chemical structure of the MF sponge, TGA analysis, nitrogen adsorption−desorption isotherm, nanopore size of the PSM, FT-IR spectra, dipeptide chemical structures, Oasis MCX polymeric surface chemistry, CD-SiMNS preparation, recovery of dipeptides on the CD-SiMNS cartridge, MRM transition ions, detailed G

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (28) Dotson, A.; Westerhoff, P. J. - Am. Water Works Assoc. 2009, 101, 101−115. (29) How, Z. T.; Linge, K. L.; Busetti, F.; Joll, C. A. Environ. Sci. Technol. 2017, 51, 4870−4876. (30) Shah, A. D.; Mitch, W. A. Environ. Sci. Technol. 2012, 46, 119− 131. (31) Huang, G.; Jiang, P.; Li, X.-F. Anal. Chem. 2017, 89, 4204− 4209. (32) Huang, G.; Jiang, P.; Jmaiff Blackstock, L. K.; Tian, D.; Li, X.-F. Environ. Sci. Technol. 2018, 52, 4218−4226.

H

DOI: 10.1021/acs.analchem.8b05707 Anal. Chem. XXXX, XXX, XXX−XXX