Hydrophilic Molecularly Imprinted ResorcinolâFormaldehyde

Subscriber access provided by NEW YORK MED COLL

Article

Hydrophilic Molecularly Imprinted Resorcinol#Formaldehyde#Melamine Resin Prepared in Water with Excellent Molecular Recognition in Aqueous Matrices Tianwei Lv, Hongyuan Yan, Jiankun Cao, and Shiru Liang Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b03253 • Publication Date (Web): 06 Oct 2015 Downloaded from http://pubs.acs.org on October 10, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Hydrophilic Molecularly Imprinted Resorcinol‒Formaldehyde‒Melamine Resin Prepared in Water with Excellent Molecular Recognition in Aqueous Matrices Tianwei Lv1, Hongyuan Yan∗1,2, Jiankun Cao1, and Shiru Liang1 1

Key Laboratory of Pharmaceutical Quality Control of Hebei Province, College of Pharmacy, Hebei University, Baoding, 071002, China 2

Key Laboratory of Medicinal Chemistry and Molecular Diagnosis, Ministry of Education, Hebei University, Baoding, 071002, China

ABSTRACT:

Hydrophilic

molecularly

imprinted

resorcinol‒formaldehyde‒

melamine resin (MIRFM) is synthesized in water and shows excellent molecular recognition in aqueous matrices. The double functional monomers resorcinol and melamine, and crosslinker formaldehyde are all hydrophilic, and then the hydrophilic groups like hydroxyls, imino groups, and ether linkages can be introduced into MIRFM, which make the material compatible with aqueous samples. The general principle is demonstrated by the synthesis of MIRFM using sulfanilamide as a dummy template for the selective recognition to sulfonamides (SAs) in milk samples. Resorcinol and melamine can interact with the template mainly by hydrogen bond and π‒π interaction, which make MIRFM and the analytes have strong affinity. Besides, melamine can improve the rigidity of MIRFM and accelerate the polymerization process, so there is no need to add base or acid as a catalyst, which guarantees the success of molecular imprinting. MIRFM shows higher recovery and improved purification effect for SAs compared with silica, HLB, C18, and SCX. Due to its excellent hydrophilicity and specificity, MIRFM is promising to be applied in

∗

Corresponding author. Tel.: +86-312-5079788. Fax: +86-312-5971107.

E-mail address: [email protected]. 1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

biological, environmental, and clinical fields. INTRODUCTION Nowadays, sample preparation is still the bottleneck of analytical chemistry with the complexity, diversity, and uncertainty of samples.1 The exploration of adsorbents with excellent adsorption performance, purification efficiency, and selectivity is extremely needed. Molecularly imprinted polymers (MIPs), which have high affinity, selectivity, and stability, are suitable for selective separation of micro or trace level of analytes in complex samples.2‒5 Based on MIPs, the studies on selective adsorbents, bio- and chemosensors, plastic antibodies, drug delivery, and biomimetic catalyst develop rapidly.6‒10 But until now, most reported MIPs are prepared in organic or organic-rich solvents and they show poor molecular recognition in aqueous solutions,11‒13 which limits their application in food, biological, and environmental fields.14‒16 Although some publications have reported about MIPs prepared in water (such as in situ polymerization,17 Pickering emulsion polymerization,18 and inverse emulsion suspension polymerization19), another problem appears that the non-specific adsorption is similar to or higher than the specific recognition,20,21 which indicating the low efficiency of molecular imprinting. So, the development of hydrophilic MIPs with excellent molecular recognition in aqueous matrices remains a challenge. Hydrophilic resins are one common group of adsorption materials, on which there are different kinds of functional groups that can interact with analytes, including many hydrophilic groups.22 Resorcinol‒formaldehyde resin has abundant hydroxyls and it has been used as adsorbent for the separation of metal ions.23 Melamine‒formaldehyde resin has the characteristics of high mechanical strength, good stability, and rich amino groups.24 Resorcinol‒formaldehyde‒melamine resin (RFM) reported recently has two functional monomers, resorcinol and melamine, and 2


Page 2 of 27

Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


combines the advantages of resorcinol‒formaldehyde and melamine‒formaldehyde resins.25 Because resorcinol, melamine, and formaldehyde are all hydrophilic and there are abundant polar groups on RFM like hydroxyls, imino groups, ether linkages and so on,26‒29 it has the potential to be applied as adsorbent to separate target molecules in aqueous matrices.30 Nevertheless, the lack of special selectivity is a critical problem with RFM, which would lead to a poor extraction and purification efficiency. Herein, the synthesis of hydrophilic MIRFM in aqueous solution is presented with resorcinol and melamine as double functional monomers and formaldehyde as crosslinker, which introduce abundant hydroxyls, imino groups, and ether linkages into the material. MIRFM prepared with sulfanilamide as dummy template shows good compatibility with water and excellent molecular recognition to the analytes when it is applied to the extraction of SAs in milk. The separation and purification effect of MIRFM is outstanding in comparison with four commercial adsorbents (silica, HLB, C18, and SCX). The development of MIRFM overcomes MIPs’ incompatibility with water by employing the hydrophilic functional monomers and crosslinker, and the material shows good specificity for target molecules in aqueous matrices. EXPERIMENTAL SECTION Materials. Resorcinol was purchased from Guangfu Chemical Co. Ltd. (Tianjin, China). Melamine, formaldehyde (37%), and trifluoroacetic acid (TFA) were obtained from Kermel Chemical Co. Ltd. (Tianjin, China). Sulfanilamide, sulfadimidine sodium (SM2-Na), sulfachloropyridazine sodium (SCP-Na), sulfamethoxazole (SMZ), acyclovir, florfenicol, and thiamphenicol were obtained from Aladdin Chemical Co. Ltd. (Shanghai, China). Lead acetate (Pb(AC)2) was purchased from Tianda Chemical 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Co. Ltd. (Tianjin, China). Ultrapure water was filtered with a 0.45-µm membrane before use. Instrumentation and Conditions. High-performance liquid chromatography (HPLC) analysis was performed on an LC-20A system equipped with one LC-20AT solvent delivery unit and one SPD 20A UV‒Vis detector (Shimadzu, Kyoto, Japan). An N2000 chromatography workstation (Zhedazhineng, Hangzhou, China) was used for data acquisition. A Venusil XBP C18 analytical column (250 mm × 4.6 mm, 5 µm) was purchased from Bonna-Agela Technologies Co. Ltd. (Tianjin, China). The detection wavelength of the detector was set at 270 nm. The mobile phase was water (containing 0.5‰ TFA)‒methanol (72:28, v/v) with a flow rate of 1.0 mL/min. The morphology evaluation was carried out by Phenom Pro scanning electron microscope (SEM) (Phenom, Einthoven, Netherlands) and the Fourier transform infrared spectra (FTIR) were obtained on a Vertex70 Fourier transform infrared spectrometer (Bruker, Karlsruhe, Germany) in a range of 500‒4000 cm−1. Synthesis of MIRFM. Twenty-three kinds of MIRFMs were prepared according to the synthesis scheme (Table S1 in Supporting Information). MIRFM 22 was synthesized as follows: Resorcinol (30 mmol) and formaldehyde (60 mmol) were added into ultrapure water (30 mL) in three-necked bottle A and stirred for 1 h at 40 °C; meanwhile, melamine (10 mmol) and formaldehyde (30 mmol) were added into ultrapure water (10 mL) in three-necked bottle B and stirred at 80 °C until the solution was clear. When the solution in three-necked bottle B decreased to 40 °C, it was mixed into three-necked bottle A. Then sulfanilamide (0.8 mmol) was added into the bottle at 40 °C and stirred until the template dissolved, after which acetonitrile (10 mL) was added and the solution was stirred to self-assemble for 30 min. After it was heated at 80 °C for 24 h under a static condition, the product was filtered and the solid 4


Page 4 of 27

Page 5 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was retained. Methanol‒water (4:1, v/v) was used to remove template with the assistance of ultrasound and then the mixture was centrifuged, which was repeated for 8 times. The solid was air-dried at 60 °C for 5 h, and MIRFM was obtained. Nonimprinted resorcinol‒formaldehyde‒melamine resin (NIRFM) was synthesized in an identical way except the addition of the template. Procedure of MIRFM‒Solid-Phase Extraction (MIRFM‒SPE). MIRFM (30 mg) was packed into an empty cartridge (60 mm × 10 mm) between two polyethylene frits. The cartridge was firstly preconditioned with 2.0 mL of methanol and then 2.0 mL of water (pH 5), after which 0.5 mL of milk sample (pH 5) was loaded into the cartridge and washed with 0.25 mL of water. Finally, the cartridge was eluted by 0.7 mL of methanol‒water‒acetic acid (4:1:0.25, v/v/v), and the eluate was injected to HPLC for analysis. RESULTS AND DISCUSSION Characteristics of MIRFM. Figure 1 shows the SEM images of molecularly imprinted melamine‒formaldehyde resin (MIMF, A), molecularly imprinted resorcinol‒formaldehyde resin (MIRF, B), and MIRFM (C and D). MIRFM are monodispersed and intact microspheres and the particle size is mainly in the range of 3‒6 µm. It can be seen from the comparison of A, B, and C that MIMF has some bites on the surface of the microspheres and for MIRF the aggregation of the particles is severe. The particle size varies a lot for both MIMF and MIRF. Under the same preparation condition, MIRFM has a regular spherical shape and a narrow size distribution, which are favorable to application as SPE adsorbent. When packed into the cartridge, microspheres with a uniform particle size can construct a homogeneous solid layer to guarantee adequate contact between the adsorbent and the sample solution.31 A wide size distribution would result in heterogeneous pathways for the 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

sample solution to go through, and the surface bites and aggregation of particles may lead to some gaps existing in the solid layer. Therefore, MIRFM which has a better morphology is more suitable to be employed as the adsorbent of SPE. The purposes of FTIR are to verify that the double functional monomers, resorcinol and melamine, are incorporated into MIRFM, and that the material has the hydrophilic groups like hydroxyls, imino and amino groups, and ether linkages. The two strong peaks at 1560 and 1330 cm−1 appearing in the spectra of MIMF and MIRFM represent the aromatic stretching vibration and breathing mode of C‒N in the triazine ring of melamine, respectively (Figure 2). The two peaks at 1610 and 1475 cm−1 which not observed in the spectra of MIMF are attributed to the C=C stretching vibrations in the aromatic ring of resorcinol. From the comparison of the three spectra, the characteristic peaks of MIRF and MIMF can all be found in MIRFM, which demonstrates the successful incorporation of the two functional monomers into the material. For the hydrophilic groups, the strong and broad peak at 3423 cm−1 is associated with the stretching vibrations of O‒H and N‒H, and the out-of-plane bending vibration of N‒H corresponds to the peak at 809 cm−1. Moreover, the peak at 1100 cm−1 is attributed to the stretching vibration of C‒O‒C. These characteristic peaks indicate the existence of the hydrophilic groups in MIRFM. All the above imply the successful introduction of the functional monomers and the hydrophilicity of MIRFM. Static adsorption was conducted with MIRFM and NIRFM to evaluate molecular imprinting efficiency. The adsorption amounts for the three SAs increase with the increasing concentration and the values of MIRFM are higher than those of NIRFM (Figure 3A, B, and C). The adsorption amounts of MIRFM for SM2-Na and SMZ are about twice, and for SCP-Na three times as those of NIRFM, which indicates the high 6


Page 6 of 27

Page 7 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


efficiency of molecular imprinting in aqueous solution. In the molecular imprinting process, the specific functional groups are generated in the imprinted pores after the elution of the template. When the analytes (structural analogues of the template) go through MIRFM, they can be captured into the imprinted pores and the certain groups in the molecules would interact with the imprinted sites. This explains the phenomenon that MIRFM has higher adsorption amounts for the three SAs than NIRFM. In competitive adsorption, MIRFM has much higher adsorption amounts for the three SAs in water than those for the other three drugs, which have different molecular structures from the template (Figure 3D). This is because the imprinted pores have special selectivity for the structural analogues of the template, and thus the interaction between MIRFM and SAs is stronger than that between MIRFM and other analytes. It is evident that MIRFM has excellent specificity for the target analytes, which also implies that MIRFM has good molecular recognition in aqueous matrices. Besides, the adsorption amount for SCP-Na is the highest, and the values for SM2-Na and SMZ are similar. The reason may be that the chlorine atom in SCP-Na can interact with the hydrogen atom of MIRFM by hydrogen bond, which strengthens the interaction between SCP-Na and MIRFM. To demonstrate the hydrophilicity of MIRFM, the dispersion stability of C18, HLB, and MIRFM in water was studied. After dispersion in water (1 mg/mL) by ultrasonic, the dispersed mixtures were allowed to stand still for 2 h at 25 °C. From Figure S1A in Supporting Information, it can be seen that the dispersion of C18 is heterogeneous because it is highly hydrophobic, while HLB and MIRFM can disperse homogeneously in water after ultrasonic immediately. After standing still for 2 h, for C18 and HLB most particles precipitate onto the bottom of the vials, but MIRFM 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

keeps dispersing well in water (Figure S1B). Although HLB is a hydrophiliclipophilic balance material, its compatibility with water is obviously lower than that of MIRFM, which indicates the excellent hydrophilicity of MIRFM. Preparation of MIRFM. After MIRFMs were obtained, their molecular recognition in aqueous solution was evaluated by performing SPE. The materials were packed into cartridges and 1.0 mL of SAs aqueous standard solution (SM2-Na, SCPNa, and SMZ, 5.00 µg/mL) was loaded into each cartridge. Then the effluents were collected for HPLC analysis to detect the loss of the analytes. The packed amount of the materials was 50 mg in Figure 4A, B, and C, and 30 mg in Figure 4D and E. Molar Ratio of Resorcinol to Melamine. The molar ratio of resorcinol to melamine is the most important factor in preparing MIRFM because the double functional monomers determine the composition, affinity, rigidity, and polymerization rate. Melamine which is a weak base can accelerate the polymerization process, so the formation of MIRFM can happen without adding base or acid, which ensures the success of molecular imprinting.25 Figure 5 shows that resorcinol can interact with the template by hydrogen bond and π‒π interaction because of the hydroxyls and benzene ring of the monomer. Imino groups and some reserved amino groups in melamine can interact with the template by hydrogen bond, and the conjugate rings of the template and melamine can generate π‒π interaction. From Figure 4A, it can be found that resorcinol is the dominant monomer for the affinity between MIRFM and the target molecules. The reason may be that the oxygen atoms in resorcinol have greater electronegativity and smaller atom diameter than the nitrogen atoms in melamine, and therefore it is easier for the oxygen atoms to generate hydrogen bonds with the hydrogen atoms in the target molecules.

8


Page 8 of 27

Page 9 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


From Table S1, by synthesizing MIRFM 1, 2, 3, 4, and 5, the effect of this factor was investigated. The loss rates of these five materials for the three SAs are shown in Figure 4A. It can been seen if the molar ratio of resorcinol to melamine is increased, the loss rates of the three analytes all decrease until the value reaches 30:10, and when it is 30:10 and 40:0, MIRFM can adsorb the analytes completely. But if the value is 40:0, the material would be MIRF, which is sticky and liable to shrink. After the sample solution was added into the SPE cartridge, MIRF would be very dense, which made it hard for the solution to go through, and thus the procedure was timeconsuming. Besides, the particle size distribution of MIRF is very wide. When the molar ratio of resorcinol to melamine is 30:10, the loss rates of the three SAs are 0, and the permeability of the material is also good. Because the triazine ring in melamine molecule is highly rigid and the bonding of melamine and formaldehyde is robust, the mechanical strength is improved and the shrinking and swelling of the material are also relieved effectively.32,33 Therefore, 30:10 is chosen to be the molar ratio of resorcinol to melamine. Molar Ratios of Formaldehyde to Resorcinol and Formaldehyde to Melamine. The functions of formaldehyde can be found in Figure 5. Firstly, formaldehyde and the two monomers react to form hydroxymethyl monomers, and then the condensation of hydroxymethyl groups happens to generate methylene or methylene ether bridged clusters, which can act as the nucleuses with many active groups on them. The nucleuses then grow to microspheres by further copolycondensation, which is similar to the Stöber method. 34 Besides, the ether linkages formed from formaldehyde are beneficial to the hydrophilicity of MIRFM. Formaldehyde can crosslink the monomers by copolycondensation to form threedimensional resin networks. 1 mol of resorcinol can react with 1‒3 mol of 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

formaldehyde, and different hydroxymethyl resorcinols have an effect on the characteristics of MIRFM. The amount of formaldehyde can change the crosslinking degree of the material, which is vital for the adsorbent to interact with analytes. In order to investigate the impact of the molar ratio of formaldehyde to resorcinol, MIRFM 6, 7, 4, 8, and 9 were prepared and compared, and the molar ratios were 1, 1.5, 2, 2.5, and 3, respectively. From Figure 4B, it can be seen that the loss rates of these five materials for the three SAs have the same trend which goes down till the molar ratio of 2 and then rises. Also, this factor can affect the permeability of MIRFM, and the permeability of the material is the best when the molar ratio is 2. Therefore, the molar ratio of formaldehyde to resorcinol is chosen to be 2. The molar ratio of formaldehyde to melamine was also investigated in this work. 1 mol of melamine can react with 1‒6 mol of formaldehyde and MIRFM 10, 11, 4, 12, and 13 were synthesized in the molar ratios of 1, 2, 3, 4.5, and 6, respectively. From the result in Figure 4C, the trends of the loss rates for the three SAs are corresponding, which decrease to the value of 3 and then go up. Besides, the amount of formaldehyde has an influence on the rigidity of the material by changing the crosslinking degree, and the mechanical strength of the material increases with more crosslinker added.35 When the value is 3, the loss rates for the analytes are the lowest, and MIRFM has excellent rigidity and permeability. So, the molar ratio of formaldehyde to melamine is optimized as 3. Amount of Template. The template sulfanilamide, which has a similar molecular structure to those of SAs, was used in the preparation of MIRFM to avoid the effect of template leakage on quantitative analysis. Figure 5 shows the possible interactions between the template and MIRFM, which is the foundation of the successful molecular imprinting in water. From that, it can be known that hydrogen bond and 10


Page 10 of 27

Page 11 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


π‒π interaction are the main interactions. For molecularly imprinted materials, the amount of template is a key parameter to molecular recognition. MIRFM 14, 15, 4, 16, 17, 18, and 19 were prepared using 0, 0.2, 0.4, 0.8, 1.6, 4.0, and 10.0 mmol of the template, respectively. Figure 4D shows that the loss of the three analytes is obvious for NIRFM, because there is no template added in the preparation of NIRFM and its affinity for the SAs is weak. When template amount is 0.2 mmol, the loss of the analytes is also detected, which implies the deficiency of template. As template amount is increased, the loss rates of the three SAs decrease firstly and rise when template amount is over 0.8 mmol. The reason may be that the template can’t be eluted completely if template amount is too large. The residual sulfanilamide takes up many interaction sites, so the occupied sites can’t interact with the analytes in the loading process, which leads to a low adsorption amount. The result indicates that MIRFM with 0.8 mmol of the template is with the lowest loss rate, so 0.8 mmol is chosen as the amount of template. Volume of Porogen. Acetonitrile acted as porogen in the synthesis of MIRFM to increase the specific surface area of the obtained material, and then improve the adsorption performance of MIRFM. The volume of porogen was investigated and MIRFM 20, 21, 16, 22, and 23 were prepared with 0, 2, 5, 10, 15 mL of acetonitrile, respectively. Figure 4E shows the loss rates of the five materials for the three SAs, which go down until the volume of acetonitrile reaches 10 mL, and the loss for SMZ is detected when 15 mL of acetonitrile is used. With the addition of porogen, larger surface of MIRFM would be exposed, so more imprinted sites appear on the surfaces or on the walls inside the pores, which makes MIRFM attract more analytes. But the value of specific surface area would decrease if macropores and through-holes are generated with a large amount of porogen, which results in a low extraction efficiency. 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

This may be the cause for the loss of SMZ when 15 mL of acetonitrile is employed. Therefore, 10 mL of acetonitrile is used as porogen. From all the above, the five parameters including the molar ratios of resorcinol to melamine, formaldehyde to resorcinol, formaldehyde to melamine, the amount of template, and the volume of porogen have been optimized and the optimal material is MIRFM 22, which is employed for further study. Optimization of MIRFM‒SPE. The factors that affected the extraction efficiency were optimized, including the sample pH, washing solvent, and elution solvent. The concentration of each drug in spiked milk samples was 5.00 µg/g. SAs are one group of amphoteric compounds and their molecular forms are influenced by the sample pH, which has an effect on the recovery of SAs in SPE. In this work, the impact of the sample pH was evaluated from pH 2 to 11 and the pH was adjusted with 1 M HCl solution or 1 M NaOH solution. The pKa1 and pKa2 for SM2Na is 2.6 and 7.4, respectively, 1.9 and 5.5 for SCP-Na, and 1.7 and 5.6 for SMZ. When the sample pH is in the range between pKa1 and pKa2, the main existing form for the three SAs is neutral molecule,36 which can achieve high recovery in the SPE (Figure S2A, pH 3‒5). The main interactions between the analytes and MIRFM are hydrogen bond and π‒π interaction, and when the existing form of the analytes is neutral molecule the interactions are the strongest. If the sample pH is lower than the pKa1 of the SAs, the amino group in SAs would be protonated and in this situation the nitrogen atom doesn’t have two lone pair electrons, so it can’t participate in the formation of hydrogen bond. When the pH is higher than the pKa2 of the SAs, the hydrogen atom of the sulfamide part in SAs would be lost and it can’t form hydrogen bond, either. The hydrogen bonds between MIRFM and the SAs would reduce in the two cases, and therefore the affinity weakens. In the range of pH 3‒5, the molecular 12


Page 12 of 27

Page 13 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


stability of the SAs is affected a little by the change of pH value. Because at pH 5 the highest recovery of SM2-Na and SCP-Na is obtained and for SMZ the recovery just declines for 0.8% from pH 4 to 5, the value of sample pH is optimized as 5. Washing solvent plays an important role in removing the co-adsorbed interferents from the adsorbent and five kinds of washing solvents were investigated including water‒acetonitrile (9:1, v/v), water‒methanol (9:1, v/v), water, water‒acetone (9:1, v/v), and acetone. After 0.5 mL of milk sample (pH 5) was loaded, 0.3 mL of different washing solvents were added and the effluents were collected for HPLC analysis. From Figure S2B, among the five washing solvents, the loss rates of the three SAs by water are the lowest and interferents can be also washed away. So water was selected and its volume (0.1, 0.2, 0.25, 0.3, and 0.4 mL) was then investigated. From the result, when 0.1 and 0.2 mL of water are used, the interference is not eliminated completely. Then more water can achieve excellent purification, but the loss of the analytes increases if the volume of water is more than 0.25 mL. Finally, 0.25 mL of water is selected as washing solvent with a low loss of the analytes and a satisfactory purification effect. A suitable elution solvent should be utilized to elute the analytes from MIRFM and six kinds of elution solvents were evaluated in this work including methanol, methanol‒water‒ammonia (4:1:0.25, v/v/v), methanol‒water‒acetic acid (4:1:0.25, v/v/v), methanol‒water (4:1, v/v), acetonitrile‒water (4:1, v/v), and acetonitrile. Among them, methanol‒water‒acetic acid (4:1:0.25, v/v/v) can obtain the highest recovery for all the three SAs (Figure S2C). Water is added into methanol to adjust the polarity of the solvent, and then it would have a better affinity to the SAs. On the one hand, acetic acid protonates the SAs to weaken the adsorption between MIRFM and the analytes, and on the other hand, hydrogen bond can be formed between acetic 13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

acid and MIRFM, or acetic acid and the SAs.37 Thus, the hydrogen bonds between MIRFM and the analytes would be broken. Figure S2C shows that the recovery of the three SAs increases until the volume of methanol‒water‒acetic acid (4:1:0.25, v/v/v) reaches 0.7 mL and then the recovery almost remains unchanged. Hence, the optimal elution solvent is 0.7 mL of methanol‒water‒acetic acid (4:1:0.25, v/v/v). Validation of MIRFM‒SPE‒HPLC Method. The MIRFM‒SPE‒HPLC method was validated by linearity, limit of detection, accuracy, precision, and repeatability under the optimum conditions. Milk samples were spiked with the three SAs to obtain seven increasing spiked concentrations in a range of 0.05‒5.00 µg/g, and these spiked samples were pretreated with MIRFM‒SPE under the optimized condition, in which three replicates were performed for each concentration. HPLC‒UV detector was employed for analysis and calibration curves for the three SAs were constructed with the areas of chromatographic peaks measured at the seven spiked concentrations. The result is listed in Table S2, which demonstrates that good linearity is obtained in the whole concentration range with correlation coefficient (r) ≥0.9997. The limit of detection (LOD) and limit of quantification (LOQ) are 0.005‒0.007 µg/g and 0.016‒0.024 µg/g, which are based on the signal to noise ratios of 3 and 10, respectively. Recovery experiment was performed to evaluate the accuracy of the method with spiked milk samples at three levels of the SAs (0.10, 1.00, and 5.00 µg/g). Table S3 indicates that the average recovery of the three SAs is in the range of 91.5%‒100.3% with relative standard deviation (RSD) ≤6.8%, which implies that the method is reliable and applicable. The precision of the method was assessed by performing five replicates of spiked samples (5.00 µg/g) on the same day (n=3) and three consecutive days. The intra-day and inter-day precisions expressed as RSDs are in the range of 1.6%‒2.9% and 2.4%‒4.7%, respectively. Moreover, blank milk 14


Page 14 of 27

Page 15 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


samples were pretreated and analyzed, and there was no interference at the retention time of the three SAs. The result further confirms the purification effect and practicability of the MIRFM‒SPE‒HPLC method. Comparison of MIRFM with Commercial Adsorbents. MIRFM was compared with four commercial adsorbents including silica, HLB, C18, and SCX by performing SPE with the same adsorbent amount (30 mg) and loading 0.5 mL of spiked milk sample (5.00 µg/g). For MIRFM, the procedure of SPE was carried out under the optimal conditions, and the SPE procedure for the four commercial adsorbents was conducted according to the reported papers.38‒41 The result is shown in Figure S2D, which indicates that MIRFM can achieve higher recovery for the three SAs than the commercial adsorbents. Because MIRFM can interact with SAs by hydrogen bond and π‒π interaction, it has strong affinity for the analytes, which contributes to the high recovery. Besides, MIRFM can achieve a better purification effect because of its specificity for SAs. Analysis of Milk Samples. To evaluate the molecular recognition of MIRFM in aqueous matrices and the practicability of MIRFM‒SPE‒HPLC method, seven kinds of milk samples purchased from local markets in Baoding city were pretreated and analyzed. As shown in Figure 6, the comparison of the chromatograms of spiked milk sample (5.00 µg/g) before and after MIRFM‒SPE (A and B) indicates that the great interference around SM2-Na is eliminated efficiently and there is no interferences were observed around SCP-Na and SMZ after MIRFM‒SPE. SMZ was detected in one of the seven milk samples with a concentration of 160 ng/g (C), which was higher than the maximum residue limit (MRL) for total SAs in milk established in China (100 ng/g). The result implies that the molecular recognition ability of MIRFM is

15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

excellent in aqueous matrices and the presented MIRFM‒SPE‒HPLC method can be applied to the isolation and determination of SAs in milk. CONCLUSIONS Hydrophilic MIRFM is synthesized in water with resorcinol and melamine as double functional monomers and formaldehyde as a crosslinker, which introduce abundant hydrophilic groups such as hydroxyls, imino groups, and ether linkages into MIRFM and make the material well compatible with aqueous solutions. MIRFM with sulfanilamide as dummy template presents excellent molecular recognition to SAs in aqueous matrices and has advantages over silica, HLB, C18, and SCX in the extraction and purification of SAs in milk. Because MIRFM is one kind of material with excellent hydrophilicity and specificity, it has the potential to be applied in many fields, such as biological, environmental, and clinical sciences. ASSOCIATED CONTENT Supporting Information Experimental: Binding assays of MIRFM, preparation of milk samples. Figure S1: Dispersion stability of C18, HLB, and MIRFM in water. Figure S2: Optimization of MIRFM‒SPE and comparison of different adsorbents. Table S1: Synthesis scheme of hydrophilic MIRFMs. Table S2: Parameters of MIRFM‒SPE‒HPLC method. Table S3: Recovery of the three SAs in spiked milk samples. AUTHOR INFORMATION Corresponding Author ∗

Tel.: +86-312-5079788. Fax: +86-312-5971107. E-mail: [email protected].

Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS 16


Page 16 of 27

Page 17 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The project is sponsored by the National Natural Science Foundation of China (21575033), the Natural Science Foundation of Hebei Province (B2015201132), and the Natural Science Foundation of Education Department of Hebei Province (ZD2015036). REFERENCES (1) Martín-Esteban, A. Trends Anal. Chem. 2013, 45, 169‒181. (2) Xu, S.; Li, J.; Chen, L. J. Mater. Chem. 2011, 21, 4346‒4351. (3) Zhang, W.; He, X.; Chen, Y.; Li, W.; Zhang, Y. Biosens. Bioelectron. 2012, 31, 84‒89. (4) Chianella, I.; Guerreiro, A.; Moczko, E.; Caygill, J. S.; Piletska, E. V.; De Vargas Sansalvador, I. M. P.; Whitcombe, M. J.; Piletsky, S. A. Anal. Chem. 2013, 85, 8462‒8468. (5) Yan, H.; Liu, S.; Gao, M.; Sun, N. J. Chromatogr. A 2013, 1294, 10‒16. (6) Yao, G.; Liang, R.; Huang, C.; Wang, Y.; Qiu, J. Anal. Chem. 2013, 85, 11944‒11951. (7) Poma, A.; Guerreiro, A.; Whitcombe, M. J.; Piletska, E. V.; Turner, A. P.; Piletsky, S. A. Adv. Funct. Mater. 2013, 23, 2821‒2827. (8) Orozco, J.; Cortés, A.; Cheng, G.; Sattayasamitsathit, S.; Gao, W.; Feng X.; Shen, Y.; Wang, J. J. Am. Chem. Soc. 2013, 135, 5336‒5339. (9) Yan, H.; Yang, C.; Sun, Y.; Row, K. H. J. Chromatogr. A 2014, 1361, 53‒59. (10) Piletska, E. V.; Abd, B. H.; Krakowiak, A. S.; Parmar, A.; Pink, D. L.; Wall, K. S.; Wharton, L.; Moczko, E.; Whitcombe, M. J.; Karim, K.; Piletsky, S. A. Analyst 2015, 140, 3113‒3120. (11) Golsefidi, M. A.; Es’haghi, Z.; Sarafraz-Yazdi, A. J. Chromatogr. A 2012, 1229, 24‒29. 17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(12) Wan, W.; Biyikal, M.; Wagner, R.; Sellergren, B.; Rurack, K. Angew. Chem. Int. Ed. 2013, 52, 7023‒7027. (13) Fan, J.; Tian, Z.; Tong, S.; Zhang, X.; Xie, Y.; Xu, R.; Qin, Y.; Li, L.; Zhu, J.; Ouyang, X. Food Chem. 2013, 141, 3578‒3585. (14) Guo, Z.; Guo, T.; Guo, M. Anal. Chim. Acta 2008, 612, 136‒143. (15) Hu, X.; Xie, L.; Guo, J.; Li, H.; Jiang, X.; Zhang, Y.; Shi, S. Food Chem. 2015, 179, 206‒212. (16) Niu, H.; Yang, Y.; Zhang, H. Biosens. Bioelectron. 2015, 74, 440‒446. (17) Piletsky, S. A.; Matuschewski, H.; Schedler, U.; Wilpert, A.; Piletska, E. V.; Thiele, T. A.; Ulbricht, M. Macromolecules 2000, 33, 3092‒3098. (18) Shen, X.; Ye, L. Chem. Commun. 2011, 47, 10359‒10361. (19) Luo, X.; Zhan, Y.; Huang, Y.; Yang, L.; Tu, X.; Luo, S. J. Hazard. Mater. 2011, 187, 274‒282. (20) Ma, Y.; Pan, G.; Zhang, Y.; Guo, X.; Zhang, H. Angew. Chem. Int. Ed. 2013, 52, 1511‒1514. (21) Zhao, M.; Zhang, C.; Zhang, Y.; Guo, X.; Yan, H.; Zhang, H. Chem. Commun. 2014, 50, 2208‒2210. (22) Liu, J.; Qiao, S.; Liu, H.; Chen, J.; Orpe, A.; Zhao, D.; Lu, G. Angew. Chem. Int. Ed. 2011, 50, 5947‒5951. (23) Dwivedi, C.; Kumar, A.; Juby, K. A.; Kumar, M.; Wattal, P. K.; Bajaj, P. N. Chem. Eng. J. 2012, 200, 491‒498. (24) Wu, Y.; Li, Y.; Xu, J.; Wu, D. J. Mater. Chem. B 2014, 2, 5837‒5846. (25) Zhou, H.; Xu, S.; Su, H.; Wang, M.; Qiao, W.; Ling, L.; Long, D. Chem. Commun. 2013, 49, 3763‒3765. (26) Fang, G.; Lau, H. F.; Law, W. S.; Li, S. F. Y. Food Chem. 2012, 134, 2473‒2480. 18


Page 18 of 27

Page 19 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(27) Lv, Z.; Sun, Q.; Meng, X.; Xiao, F. J. Mater. Chem. A 2013, 1, 8630‒8635. (28) Tokudome, Y.; Nakane, K.; Takahashi, M. Carbon 2014, 77, 1104‒1110. (29) Mitome, T.; Iwai, Y.; Uchida, Y.; Egashira, Y.; Matsuura, M.; Maekawa, K.; Nishiyama, N. J. Mater. Chem. A 2014, 2, 10104‒10108. (30) Augusto, F.; Hantao, L. W.; Mogollon, N. G.; Braga, S. C. Trends Anal. Chem. 2013, 43, 14‒23. (31) Hu, Y.; Pan, J.; Zhang, K.; Lian, H.; Li, G. Trends Anal. Chem. 2013, 43, 37‒52. (32) Jiang, S.; Hou, H.; Greiner, A.; Agarwal, S. ACS Appl. Mater. Interfaces 2012, 4, 2597‒2603. (33) Shen, Z.; Wang, T.; Liu, M. Langmuir 2014, 30, 10772‒10778. (34) Li, W.; Zhao, D. Adv. Mater. 2013, 25, 142‒149. (35) Benavides, S.; Villalobos-Carvajal, R.; Reyes, J. E. J. Food Eng. 2012, 110, 232‒239. (36) Díaz-Álvarez, M.; Barahona, F.; Turiel, E.; Martín-Esteban, A. J. Chromatogr. A 2014, 1357, 158‒164. (37) Zhang, Y. P.; Cui, J.; Lynd, L. R.; Kuang, L. R. Biomacromolecules 2006, 7, 644‒648. (38) Chang, H.; Hu, J.; Asami, M.; Kunikane, S. J. Chromatogr. A 2008, 1190, 390‒393. (39) Xie, W.; Han, C.; Hou, J.; Wang, F.; Qian, Y.; Xi, J. J. Sep. Sci. 2012, 35, 3447‒3454. (40) Shaaban, H.; Górecki, T. J. Sep. Sci. 2013, 36, 252‒261. (41) Dubreil-Chéneau, E.; Pirotais, Y.; Verdon, E.; Hurtaud-Pessel, D. J. Chromatogr. A 2014, 1339, 128‒136.

19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

List of Figures Figure 1. SEM images of MIMF (A), MIRF (B), and MIRFM (C and D). Figure 2. FTIR spectra of MIRFM, MIRF, and MIMF. Figure 3. Binding isotherms of SM2-Na (A), SCP-Na (B), and SMZ (C) in static adsorption, and six analytes (D) in competitive adsorption. Figure 4. Factors affecting molecular recognition of MIRFM. (A: molar ratio of resorcinol to melamine; B: molar ratio of formaldehyde to resorcinol; C: molar ratio of formaldehyde to melamine; D: amount of template; E: volume of porogen.) Figure 5. Schematic illustration of MIRFM’s synthesis and molecular recognition. Figure 6. Chromatograms of spiked milk sample before and after MIRFM‒SPE (A and B), and real milk sample detected SMZ after MIRFM‒SPE (C).

20


Page 20 of 27

Page 21 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


A

B

C

D

Figure 1. SEM images of MIMF (A), MIRF (B), and MIRFM (C and D).

21



150

MIRF

Transmittance (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 27

100 VO-H

MIRFM

50

VC=C

VC-O-C

VO-H, VN-H VC-N

MIMF

0 4000

γN-H

VO-H, VN-H

3500

3000

2500

2000

1500

1000

Wavenumber (cm-1) Figure 2. FTIR spectra of MIRFM, MIRF, and MIMF.

22


500

Page 23 of 27

1200

A

800

400

B

MIRFM NIRFM

2000

Adsorption amount (µg/g)


MIRFM NIRFM

1600

1200

800

400

0

0 0

20

40

60

80

0

100

20

2400

1200

MIRFM NIRFM

40

60

80

100

Concentration of SCP-Na (µg/mL)

Concentration of SM2-Na (µg/mL)

C

SCP-Na SMZ SM2-Na

2000



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


800

400

D

Acyclovir Florfenicol Thiamphenicol

1600 1200 800 400 0

0 0

20

40

60

80

0

100

20

40

60

80

100

Concentration of six analytes (µ g/mL)

Concentration of SMZ (µg/mL)

Figure 3. Binding isotherms of SM2-Na (A), SCP-Na (B), and SMZ (C) in static adsorption, and six analytes (D) in competitive adsorption.

23



A

B 15

SM2-Na SCP-Na SMZ

60

SM2-Na SCP-Na SMZ

40

Loss rate (%)

Loss rate (%)

10

20

0

5

0 0:40

10:30

20:20

30:10

1

40:0

1.5

C 80

D

SM2-Na

2.5

3

SM2-Na SCP-Na SMZ

30

SCP-Na SMZ

Loss rate (%)

60

2

n1(formaldehyde)/n(resorcinol)

n(resorcinol)/n(melamine)

40

20

0

20

10

0

1

2

3

6

4.5

0

0.2

0.4

0.8

1.6

4.0

10.0

Amount of template (mmol)

n2(formaldehyde)/n(melamine)

E SM2-Na SCP-Na SMZ

8

6

Loss rate (%)

Loss rate (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 27

4

2

0 0

5

2

10

15

Volume of porogen (mL)

Figure 4. Factors affecting molecular recognition of MIRFM. (A: molar ratio of resorcinol to melamine; B: molar ratio of formaldehyde to resorcinol; C: molar ratio of formaldehyde to melamine; D: amount of template; E: volume of porogen.) 24


Page 25 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


H N N O

NH N N OH

O N H O O

O O

S

N

N

molecular recognition

H O H

H N

N

O H HO

H N

H OH

N

HO

NH N

NH O

Figure 5. Schematic illustration of MIRFM’s synthesis and molecular recognition.

25



SM2-Na

50

SMZ SCP-Na

40

Voltage (mV)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 27

30

A 20

10

B SMZ

C

0 0

5

10

15

Time (min) Figure 6. Chromatograms of spiked milk sample before and after MIRFM‒SPE (A and B), and real milk sample detected SMZ after MIRFM‒SPE (C).

26


Page 27 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For TOC Only

resorcinol sulfanilamide formaldehyde

+

water

melamine

remove of template molecular recognition

27


Hydrophilic Molecularly Imprinted ResorcinolâFormaldehyde

Recommend Documents

Hydrophilic Molecularly Imprinted ResorcinolâFormaldehyde