Hierarchical polymeric architectures through molecular imprinting in liquid crystalline environments

European Polymer Journal 106 (2018) 223–231

Contents lists available at ScienceDirect

European Polymer Journal journal homepage: www.elsevier.com/locate/europolj

Hierarchical polymeric architectures through molecular imprinting in liquid crystalline environments Natacha Ndizeye, Subramanian Suriyanarayanan, Ian A. Nicholls

T

⁎

Bioorganic & Biophysical Chemistry Laboratory, Linnaeus University Centre for Biomaterials Chemistry, Department of Chemistry & Biomedical Sciences, Linnaeus University, SE-391 82 Kalmar, Sweden

A R T I C LE I N FO

A B S T R A C T

Keywords: Bupivacaine Liquid crystalline medium Molecularly imprinted polymer Nanostructured polymer films Piezoelectric sensor Quartz crystal microbalance

The use of liquid crystalline (LC) media as sacrificial templates during the polymer synthesis has been explored. The LC-media introduce morphological features into resultant polymers which when used together with molecular imprinting can produce materials with hierarchical architectures. Bupivacaine (1) imprinted co-polymers of 2-hydroxyethylmethacrylate (HEMA) (2a) and 1,4-divinylbenzene (DVB) (3a) were synthesized using photochemical initiation in lyotrophic liquid crystalline phases of AOT (5) in water/p-xylene and Triton X-100 (6) /water systems. SEM studies revealed the impact of the LC-media on polymer morphology, with polymer brushlike structures, with bristles of ≈30 nm diameter. The polymer morphology reflects that of the hexagonal phase of the LC medium. The rebinding characteristics of polymer films were evaluated quartz crystal microbalance (QCM, under FIA conditions). The influence of the presence of imprinting-derived recognition sites in AOT (5) in water/p-xylene polymer film induced brush-like features which provided a 25-fold enhancement of sensor sensitivity. This chemosensor was shown to be selective for the local anesthetic template, bupivacaine, through studies using the structural analogues ropivacaine and mepivacaine.

1. Introduction Materials facilitating Ångström- or nano-scale events such as in chemical catalysis or molecular recognition, e.g. in biosensors and biomaterials, require architectures that present appropriate chemical functionalities for interaction while possessing structural (morphological) features for regulating access to the material surface [1]. Surfacebased sensing technologies are a particular challenge due to two factors, the need for recognition events to take place close to the transducer surface, and the impact of this limited volume. Accordingly, the development of strategies for maximizing the number of recognition sites in close proximity to transducer surfaces has the potential to impact upon sensor technologies based upon quartz crystal microbalance (QCM), surface plasmon resonance (SPR), total internal reflectance fluorescence spectroscopy (TIRF) and electrochemical sensing [2–7]. In these contexts, sensor surfaces based upon thin polymer film coatingbased sensor surfaces have risen in prominence due to their mechanical and chemical stabilities, range of polymer functionalities available and the capacity to regulate film thickness by using electrochemical or INIFERTER-based synthesis strategies [8–11]. One strategy for introducing recognition sites into thin polymer films is the use of the molecularly imprint technique (MIP) [11–14]. ⁎

This has attracting significant attention due to the relative ease with which the ligand-recognition characteristics of the material can be the templating process, in conjunction with the general advantages of thin polymer films referred to above. In spite of these developments, there is an ever-growing need for more sensitive methods for the detection of toxins, biomarkers, chemical warfare agents and illicit drugs [15–18]. This is driving the development of molecularly imprinted thin film materials with morphological characteristics that can facilitate the access of analyte to recognition sites in close proximity to the transducer surface. To date these have included the use of ultra-thin films [19–21] and the creation of hierarchical thin-film polymer architectures where nanostructuring is combined with molecular imprinting. Common to those examples using the latter approach is the use of sacrificial materials, either organic or inorganic, to guide the formation of structural features, voids, on the nanometer-micrometer scales that allow efficient mass-transfer to the Ångström–nanometer scale recognition sites [22–26]. In efforts to develop molecularly imprinted materials with even higher solvent accessible surface areas, we have examined the use of a liquid crystalline (LC) medium (Triton X-100/water) for the synthesis of bupivacaine-selective molecularly imprinted 3-aminophenylboronic acid-p-phenylenediamine co-polymer (MIP) films [27]. This

Corresponding author. E-mail address: [email protected] (I.A. Nicholls).

https://doi.org/10.1016/j.eurpolymj.2018.07.036 Received 23 March 2018; Received in revised form 20 July 2018; Accepted 23 July 2018 Available online 24 July 2018 0014-3057/ © 2018 Published by Elsevier Ltd.

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

2. Materials and methods

electrochemical synthesis was performed under cyclic voltammetric conditions on gold-coated quartz (Au/quartz) resonators. It was apparent from SEM studies that the polymerization had taken place in the hexagonal phase, as witnessed by the presence of brush-like bundles of polymer fibrils on the 30–100 nm scale. QCM studies demonstrated the presence of template-selective binding sites, and a ≈250% enhancement in sensitivity relative to thin films of the same polymer. These results have prompted a more detailed study of scope of this method for preparing fibrous polymers in general, and to examine its use in molecular imprinting for generating a broader range of hierarchical material architectures. In the present study, we have commenced a broader exploration of the scope of polymerization in LC-media through a study using a series of polymer systems commonly used in molecular imprinting protocols: HEMA-DVB, HEMA-EGDMA and MAA-BAP. Preparation of these polymers in thin films in the presence and absence of surfactant, has been undertaken and the materials characterized by SEM, zeta-potential and QCM studies under flow injection analysis conditions.

2.1. Chemicals Bupivacaine hydrochloride (1), 2-hydroxyethyl methacrylate (HEMA, 2a), methacrylic acid (MAA, 2b), 1,4-divinylbenzene (DVB, 3a), ethylene glycol dimethylmethacrylate (EGDMA, 3b), 1,4-bis(acryloyl)piperazine (BAP, 3c), 2,2′-azobis(2-methylpropionitrile) (AIBN, 4a), 2,2′-azobis(2-methylpropionamidine) dihydrochloride (ABAH, 4b), dioctyl sulfosuccinate sodium salt (AOT, 5), Triton X-100 (TX-100, 6), and toluene were purchased from Sigma-Aldrich Inc., Sweden (see Scheme 1). The water used was a Milli-Q gradient water filtration system (Millipore, MA, USA) was used to purify distilled water to ultrapure grade with resistance values of ≤18.2 MΩ Ultrapure waterMilli Q water. Mepivacaine (7) and ropivacaine (8) were from AstraZeneca R&D, Sweden. [3H]-bupivacaine hydrochloride (1 Ci/mmol) was from Moravek Biochemicals Inc., USA. Bupivacaine free-base was prepared from bupivacaine hydrochloride (1 g) by partitioning between dichloromethane (75 mL) and NaOH (aq) (2 M, 75 mL). The organic phase was washed three times with 20 mL of NaCl (aq) (1M, pH 8.0, 20 mL). The organic phase was

CH 3 H N

O

O N

H2C

H2C

OH

OH

O O CH 3

CH 3

(1)

CH 3

(2a)

CH 3

(2b)

O H2C

H2C

N

O O

CH 2

CH 3

(3a)

CH 2

(3b)

CH 2

(3c)

H3C

CH 3

CH 3

N

N

NH 2

H2N

N H3C

N

H2C

O

NH H3C

HC

O

O

CH 3

H3C

CH

CH 3

N CH 3

NH

(4b)

(4a)

O

H

CH 3

O

O

H3C O

H3C

CH 3

O

H3C O

H3C

S

O

–+

O Na

n

H3C H3C

CH 3

6

O

5 CH 3 H N

CH 3 H N

N N O O

CH 3

CH 3

CH 3

CH 3

8

7 Scheme 1. Structural formulas of template functional monomer, crosslinker and initiator employed for synthesis of molecularly imprinted polymers: Bupivacaine (1), 2-hydroxyethyl methacrylate (HEMA, 2a) methacrylic acid (MAA, 2b), 1,4-divinylbenzene (DVB, 3a), ethylene glycol dimethacrylate (EGDMA, 3b), 1,4 bis (acryloyl)piperazine (BAP, 3c) and 2,2′-azobis(2-methylpropionitrile) (AIBN, 4a) and 2,2′-azobis(2-methylpropionamidine) dihydrochloride (ABAH, 4b). Chemical structure of surfactant molecules used for the preparation of liquid crystalline medium: dioctyl sulfosuccinate sodium salt (AOT) (5) and TX-100 (6). Chemical structures of bupivacaine analogues: mepivacaine (7) and ropivacaine (8). 224

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

then dried over anhydrous MgSO4 for 15 min, before filtering and the filtrate evaporated to dryness under reduced pressure. Prior to use, HEMA was passed through a column of activated basic aluminum oxide then fractional distillation was carried out before storage at 4 °C until use. MAA was vacuum distilled and stored at 4 °C. EGDMA (100 mL) was extracted three times with NaOH (aq) (0.1 M, 75 mL) and saturated with NaCl (25 mL). It was then dried over anhydrous MgSO4 for 15 min and stored at 4 °C. Prior to use, EGDMA was filtered through aluminum oxide. Acetonitrile, toluene and DVB were distilled then dried over 4 Å molecular sieves before storage at −8 °C. AIBN and ABAH were recrystallized with methanol and acetone–water mixture (1:1, v/v), respectively.

prepared using bupivacaine freebase as the template and a template–functional monomer–cross-linker of 1:12:55. The pre-polymerization mixture was dissolved in LC phase for 10 min then saturated with nitrogen for 5 min to purge dissolved oxygen. The mixture was stirred till a clear homogenous phase was obtained. In each case, the volume of the aqueous or organic phase was 1.7 times the total volume of the mixture of template, functional monomer and cross-linker. For the bulk synthesis of HEMA-DVB and HEMA-EGDMA polymer monoliths, AIBN initiator was added to the LC phase and irradiated under UV-light for 6 h to initiate the polymerization reaction. In the case of MAA-BAP polymer monoliths, ABAH was used as initiator. The resultant bulk polymers were crushed, ground with a mortar and pestle and rinsed with acetone to remove template, unreacted monomers and surfactants (LC phase). Fine particles of the polymer were recovered by sedimentation from acetone and allowed to dry in a fume hood. Polymer (1 g) was placed in a falcon tube and washed with a series of solvents to remove residual template, monomers and surfactant [30]. Subsequently, the particles were collected by centrifugation, washed in acetone, air-dried and stored in vacuum desiccators. Surface polymerizations were carried out on silicon dioxide (SiO2) coated Au-quartz resonators for synthesis of thin polymer films for sensor applications. The piranha solution (caution: the piranha solution reacts violently with organic compounds and contact with the skin or eyes is dangerous) (1:3, v/v; H2O2:H2SO4) was used to clean SiO2 coated Auquartz resonators for 1 min then SiO2 coated Au-quartz resonators were sonicated in 0.5 M NaOH for 15 min and were immersed in a fresh solution containing, 72 μL 3-(trimethoxysilyl) propyl methacrylate and 7.2 μL triethylamine in 0.36 mL of toluene. After 12 h the silanized substrates were soaked in a series of 1 mL volumes of solvents (toluene, THF and acetone) each for 30 min, before being dried under a stream of N2. Polymer films were grown on these silanized surfaces following a

2.2. Preparation of liquid crystalline phases Lyotrophic liquid crystalline phases of TX-100 and AOT(-p-xylene) in water were prepared using an established protocol [28]. LC TX-100 was prepared by mixing TX-100 in water (42%, v/v). After an initial equilibration (5 min), the mixture was heated to 45 °C until a homogenous phase was obtained before cooling to room temperature (18 °C) to yield the LC phase. The LC phase of AOT used in this study can be described as AOT(a)-p-xylene/H2O(b), where a and b denote the concentration of AOT in p-xylene and molar ratio of water to AOT, respectively. To prepare the LC phase, a 1.5 M AOT solution in p-xylene was prepared [29]. The molar ratio of water to AOT was then adjusted to 15:1. The mixture was stirred continuously for 12 h to afford a homogenous phase of the viscous liquid at 18 °C. 2.3. Polymer synthesis Bupivacaine

freebase

molecularly

imprinted

polymers

were

Scheme 2. Synthesis of polymer films from liquid crystalline medium. 225

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

sputtering unit (LEICA EM SCD 500) prior to being inserted into the SEM instrument. The pressure of the measurement chamber was maintained at 2 × 10−5 mbar. A 3 kV potential was applied to the electron gun to generate the electron beam used to scan the sample surface.

strategy summarized in Scheme 2. Briefly, an aliquot of the LC medium containing the monomers and initiator, and template in the case of molecularly imprinted materials, was dispensed on the substrate and placed under the UV lamp at 354 nm for 2 h. The polymer film was then washed thoroughly with acetone, 10 mM NaOH and water to remove the unreacted monomers and bupivacaine template, dried under N2 and stored in a vacuum desiccator.

2.4.5. Brunauer-Emmett-Teller (BET) BET analyses were performed to determine the pore size distribution and surface areas of polymer samples. Nitrogen adsorption-desorption measurements were conducted on a Micromeritics Tristar 3000 analyzer. Polymer particles were first dried and degassed at 50 °C for 16 h.

2.4. Physical characterization of synthesized polymers 2.4.1. Optical microscopy (OPM) A polarizing microscope (Leitz Ortholux Pol BK) with cross-polarizers equipped with a hot stage (Mettler FP 82) was used. A drop of the sample sandwiched between glass microscope slide and coverslips was placed on the hot stage. The sample was heated to 45 °C to isotropic phase and observations of optical topographic properties recorded during the cooling process.

2.4.6. Zeta-potential measurements Zeta-potentials of the polymer particles were estimated by dynamic light scattering measurements using a Delsa Nano C series instrument (Beckman Coulter, Fullerton, CA). About 50 mg of the polymer samples were suspended in 1 mL of acetonitrile and 0.5 mL of this suspension was loaded in the sample vial (0.9 mL). The scattered light of the laser beam (30 mW dual laser diode), operating at 658 nm, was received at 165° angle. The zeta-potential studies were performed by the electrophoretic movement of the charged polymer particles in acetonitrile, wherein the scattering of the laser beam was detected at 30° angle at 27 °C. CONTIN algorithm and Delsa Nano 2.21 software, procured from the manufacturer was used to analyze. All the measurements were repeated three times.

2.4.2. Fourier transform infrared spectroscopy (FT-IR) FT-IR studies were performed using a Perkin Elmer Spectrum 1000 FT-IR Spectrometer/Microscope. Polymer samples (5 mg) were mixed with 95 mg of potassium bromide. The mixture was ground manually using a pestle and a mortar. The resultant fine powder was transferred into a KBr disk, which was then placed into the sample holder on the microscope. The samples were recorded within 600–4000 cm−1 with 16 scans and 4 cm−1 resolution.

2.4.7. Quartz crystal microbalance (QCM) Sensor performance of the imprinted and reference polymer films grown on SiO2 coated Au-quartz surfaces were evaluated by piezoelectric microgravimetry using quartz crystal microbalance (QCM) system (Attana A200, Attana AB, Stockholm) under flow injection analysis (FIA) conditions and was controlled by the Attester software supplied by the maker. The polymer film substrates (10 MHz AT-cut quartz resonator from Attana AB, Stockholm, Sweden) were loaded in the flow cell holder supplied by the manufacturer. Phosphate buffer saline solution (0.01 M) containing 150 mM NaCl at pH 8.5 was used as carrier solution. The QCM instrument, equipped with a dual peristaltic

2.4.3. Elemental analysis (CHN) Carbon, hydrogen and nitrogen elemental analyses were performed using Perkin-Elmer 2400 Series CHNS/O Elemental Analyzer. Duplicate samples were dried overnight under vacuum at 50 °C prior to analysis. 2.4.4. Scanning electron microscopy (SEM) SEM studies were performed on a Leo 1550 Gemini Field Emission SEM (Angstrom Scientific Inc. Ramsey, New Jersey, USA). The polymer particles were first deposited on black carbon tape, attached to alumina stubs and coated with a thin layer of platinum using a platinum

P2

P1

100 m 100 m

P4

P3

100 m

100 m

Fig. 1. Polarizing optical microscope images showing textures of TX100/water and AOT(p-xylene)/water micelle system comprising the pre-polymerization mixtures (HEMA-DVB and bupivacaine) for the synthesis of HEMA-DVB polymer system referred in Table 1. 226

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

pump propel the carrier buffer to flow over the polymer films at a controlled flow rate. The setup was allowed to equilibrate until minimal change in the resonant frequency (< 0.5 Hz over 400 s) was observed. The analyte, or test solution, prepared using the carrier buffer was introduced into the flow cell through a 6-port injection valve provided in the instrument.

Table 1 Polymer systems synthesized and characterized in this study.

3. Results and discussion Recent efforts to enhance the performance of molecularly imprinted materials in applications dependent upon mass transfer properties, e.g. sensing, separation and catalysis, have focused on the incorporation material architectures that facilitate analyte or reactant access to recognition sites. Top-down, e.g. lithography, and bottom-up strategies have both been explored, with the use of liquid crystalline phases for producing polymers with high surface areas being of particular interest due to the scalability of the process and relatively inexpensive materials that are required [27,31]. To explore the scope of this method we have investigated two common liquid crystalline phases, TX-100/water and AOT(p-xylene)/water systems, and their use in the synthesis of a series of copolymer systems commonly used in conjunction with molecular imprinting; HEMA-DVB, HEMA-EGDMA, and MAA-BAP. The local anesthetic bupivacaine has been used as a template in the study of many aspects for the molecular imprinting process [32,33] and was selected for use here so as to allow for benchmarking with other systems. In the preparation of polymerization reaction mixtures, TX-100/ water and AOT(p-xylene)/water systems were first prepared. In each case the presence of hexagonal phases, cylindrical micelles, was observed using optical microscope equipped with a cross-polarizer. The retention of hexagonal phase character after inclusion of functional and cross-linking monomers and template was confirmed by optical microscopy, though in the absence of initiator. Polymerization reactions were initiated by UV irradiation. For comparative purposes, a series of polymers was prepared in parallel using conventionally used solvents. The porosity and morphological features of the polymers were studied by SEM and BET to determine the influence of the LC media relative to conventional solvents. Further characterization was undertaken using infrared spectroscopy and zeta-potential measurements. A piezoelectric microgravimetry (QCM) was used to determine the impact of LC media on the imprinting of bupivacaine MIP thin films. When preparing the polymers, optical microscopy was used to investigate the phase behavior of the pre-polymerization mixtures. The TX-100/water and AOT(p-xylene)/water systems showed bifringence patterns representative for hexagonal liquid crystals with broken focal conic textures (Fig. 1-SI). Transition from the hexagonal to the isotropic phase starts at 30 and 35 °C for both systems, respectively. In the case of the HEMA-DVB system, the optical texture of the LC medium was undisturbed by the presence of the monomers and template (Fig. 1) though the phase transition occurs at a lower temperature, 15 °C. In the cases of HEMA-EGDMA and MAA-BAP, hexagonal phases were maintained below 15 and 2 °C, respectively (Figs. 2, 3-SI). Thus, the prepolymerization mixture components were entrapped in the hydrophobic domains of the hexagonal structures below these temperatures, thereafter UV-irradiation was used to initiate polymerization. In parallel, a series of non-imprinted reference polymers was synthesized in the absence of the template (Table 1). To examine the impact of the surfactants on polymer structure, the polymer systems were prepared using solvents commonly used in conjunction with their use in molecular imprinting protocols: HEMA-DVB (toluene, P5) [34], HEMAEGDMA (acetonitrile, P11) [35] and MAA-BAP (water, P16) [36]. In all cases, the polymer monoliths were crushed, ground and sieved into a 63–125 Å size fraction. The particles were then washed with series of organic and aqueous solution at different pH conditions [30] to remove the micelle phases and unreacted components. Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.eurpolymj.2018.07.036.

Polymer systems

DVB-HEMA

EGDMA-HEMA

BAP-MAA

TX-100 MIP TX-100 REF AOT MIP AOT REF Solventa REF Solventa MIP

P1 P2 P3 P4 P5 P6

P7 P8 P9 P10 P11 –

P12 P13 P14 P15 P16 P17 (film)

a

DVB-HEMA (toluene), EGDMA-HEMA (acetonitrile), BAP-MAA (water).

The structural characteristics of the synthesized polymer particles were examined with FT-IR spectroscopy, Fig. 4-SI. Importantly, no evidence of the presence of residual LC media was found. In particular, the absence of distinct spectral bands at 1045 and 1230 cm−1, for ν(CeOeC) and νb(SO3), respectively, corresponding to the TX-100/ water and AOT(p-xylene)/water in the polymer scaffolds. In other words, the LC medium acts as a sacrificial structure during the polymerization process. Furthermore, the absence of strong bands around 1550 cm−1 for the ν(CeN) mode in the various imprinted polymers confirms that bupivacaine template has been removed. In each case, the similar spectral features of the polymers synthesized in LC media and conventional solvents demonstrated the presence of the corresponding monomer functionalities. For example, in the case of the HEMA-DVB systems, prominent bands at 690, 820, 1450, 1604, 1710 and 2926 cm−1 account for δ(CeH), δ(OeH), δ(CeOH), ν(C]C), ν(C]O), and ν(CeH) vibrational modes from the DVB and HEMA moieties. Similarly, the spectra of the LC-mediated synthesis-derived HEMAEGDMA and MAA-BAP reflected those of the corresponding polymers prepared with conventional solvents (Fig. 5-SI and Fig. 6-SI). Elemental analysis results (Table 1-SI) revealed comparable compositions for the imprinted and reference polymers prepared in the LC media. The compositions of the polymers prepared in LC media were even comparable with polymers synthesized in conventional solvents. Together with the FT-IR data this indicates that the monomers are incorporated into the polymer as anticipated and that the presence of the LC media does not impact dramatically on polymerization. Importantly, neither of the techniques show significant differences between the compositions of the imprinted and non-imprinted polymers. The morphology of the polymer monoliths was investigated by SEM, Fig. 2, Fig. 7-SI and Fig. 8-SI. In each case disorganized corrugated assembly with apparent nanostructural patterns in all the polymer systems (P1-P6) clearly indicates collapse of polymerizable components packed hexagonal conic structures when initiated with UV light. That way the solubility of functional monomer, cross-linker and molecular template drives the polymerization reaction either in oil or water phase. For the normal micelle (TX-100/water), since the HEMA-DVB and bupivacaine are highly soluble in organic medium, the reaction proceeds in the oil phase of either spherical or hexagonal micelles resulting in highly mesoporous structure. In this case the morphology is constructed with the agglomeration of polymer microparticles while the surface area and porosity have been drastically enhanced compared to the polymer prepared in the reverse micelle system. In the reverse micelle system AOT(p-xylene)/water, the polymerizable component is the voluminous oil phase and the presence of water phase has minimal effect on the polymerization reaction and permeability features. Hence the P3-P4 system exhibit relatively smoother surface than the one synthesized in toluene (P5-P6) and TX-100/water micelle system (P1-P2). Other polymer systems, P7-P11 (Fig. 7-SI) and P12-P16 (Fig. 8-SI) also display the effect of LC medium with the surface area reduced to a relatively microporous structure (P7-P11) or condensed to less porous monoliths (P12-P16). In both cases the solubility of the pre-polymerization mixture either in oil or water plays the major role and not the bupivacaine template. Absence of significant differences within the imprinted and REF polymers for a particular medium rules out the 227

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

P1

P2

P3

P4

P5

P6

Fig. 2. Surface topography of the HEMA-DVB (P1-P6) polymer system visualized using SEM.

parameters (Table 2). Hexagonal lattice structures of LC medium do show profound impact in terms of permeability features on other polymer systems (P7-P11 and P12-P16). Figs. 10, 11-SI shows the N2 adsorption and desorption isotherm for P7-P11 and P12-P16 polymers with H3 type hysteresis characteristic of nearly microporous structures or with very wide distribution of pore size. In this case the presence of LC medium reduces the surface area and pore size distribution. Figs. 12 and 13-SI show the decreasing trend in surface area, pore volume and its distribution compared to the polymers prepared otherwise without LC medium. Here again the difference between MIP and REF polymers is insignificant and not manifested by the LC phase evidencing no contribution from bupivacaine template on permeability characteristics of the polymer. On the whole, the LC medium affects the morphological features in the nanometer regime in terms of pore volume and surface area of the polymer without disturbing the molecular imprinting strategy. The pre-polymerization mixture is trapped within the oil or water hexagonal phases of the micelle that resemble a microreactor like condition for the polymerization process. That way the presence of small amount of the molecular template has negligible effect on the polymerization reaction and the morphological features of the polymer thus prepared. Liquid crystalline mediated polymer synthesis significantly affects the physico-chemical parameters such as zeta-potential and particle size that are crucial for polymer-based drug delivery, catalysis,

effect of molecular template on the polymer morphology. BET measurements compliment the SEM studies revealing the variation in permeability parameters such as surface area, the pore volume, the cavity size and its distribution of the cavity sizes. Shape and size of the adsorption and desorption isotherms can be correlated to the porosity and surface area of the polymers. In addition, the volume of gas adsorbed on a nonporous or macroporous material is very low when compared with a porous one. Shape of the N2 adsorption and desorption isotherm for P1-P6 polymer reveals mesoporous polymer scaffolds with H2, H3 or H4 type hysteresis (Fig. 9-SI). Polymers synthesized in organic medium (P5 and P6) reveal H4 type hysteresis with limited amount of mesopores limited by micropores, which is again confirmed with the pore distribution plot in Fig. 3. This architecture has been enhanced to larger pore size (P1-P2, H2 type hysteresis, Fig. 9-SI) or shrunken to near microporous structure (P3-P4, H3 type hysteresis, Fig. 9-SI) [37] with the presence of TX-100/water or AOT(p-xylene)/ water LC medium, respectively. Increase and decrease in the surface area observed for the polymers (Table 2) P1-P2 and P3-P4, respectively, substantiates the effect of LC medium on the polymer porosity. Also, this trend is confirmed with the pore distribution plot showing a much broader distribution of pore size reduced to one tenth of the polymer synthesized otherwise without the LC medium (P5 and P6, Fig. 3). Within the MIP and REF polymers synthesized in LC medium no significant differences have been noticed in terms of permeability 228

European Polymer Journal 106 (2018) 223–231 P1 P2

0.08 3

Pore volume (cm /g)

-Resonant frequency change , Hz

N. Ndizeye et al.

0.04

0.00

0

500

1000

1500

2000

Pore diameter (A)

3

Pore volume (cm /g)

0.5 P3 P4

0.02

0

500

1000

1500

Pore diameter (A)

3

0

1

2

3

4

5

quartz resonators coated with SiO2 layers were functionalized, initially, with acrylate group terminated silane monolayers as anchoring group. Polymer films were grown on top of this layer as mentioned in Section 2.3. Fig. 4 shows resonant frequency vs concentration (FIA curve) for bupivacaine binding on P1-P6 films. Slope of the calibration plot gives sensitivity of the chemosensor, which is higher for MIP than the REF polymer in all the cases. Within the MIP films the TX-100/water LC mediated polymer films (P1) exhibit higher sensitivity which is four and seven times greater than that of the one synthesized in AOT(p-xylene)/ water (P3) and toluene (P6) respectively. This clearly validates the effect of nanostructural morphology to enhance the analyte binding with recognition sites. To demonstrate the polymer imprinting in AOT(p-xylene)/water reverse micelle system we have chosen MAA-BAP system owing to its appreciable solubility in water phase than oil. That way the pre-polymerization mixture can compactly be packed in hexagonal phase of water phase of AOT(p-xylene)/water LC medium affording nanostructural patterns with the molecular imprinted recognition sites. For that polymer films were prepared on SiO2 coated Au-quartz resonator surface as shown in Section 2.3. SEM measurements reveal the topology of both the MIP and REF films prepared in the AOT(p-xylene)/water LC medium (Fig. 5). It shows constitute organized, densely packed nanowires or brush-like structures with thickness 39.9 ± 7.8 nm (P14) and 44.6 ± 6.7 nm (P15) compared to the films prepared in TX100-water (P12-P13) and water (P16-P17). Resonant frequency versus time curve for the repetitive injections of bupivacaine, and analogues (7 and 8) under FIA conditions on MAABAP polymer film prepared in AOT(p-xylene)/water is shown in Fig. 6. The resonant frequency change due to injection of bupivacaine analyte is significantly higher than that of the interferants 7 and 8 (Inset in Fig. 6). Sensitivity the nanostructured MIP film (P14) prepared from LC medium was four times higher than that of REF surface prepared in AOT(p-xylene)/water (P15) with linearity over the concentration range 0.1–2 mM (3–60 μg/mL). However, the polymer film prepared with TX100-water LC medium has failed to show any improvement in sensitivity owing to the lack of nanostructural morphology (Table 3). In this case the pre-polymerization mixture dispersed within the water phase of LC medium and not in the micelle phase. The cross-reactivity study was performed using bupivacaine closely related analogs mepivacaine and ropivacaine. The sensitivity values (determined from the FIA calibration plots) for the MIP-analyte complexes reveal the high affinity and selectivity (Table 4) of the bupivacaine (1) MIP chemosensor for. The greater surface area afforded by synthesis of the polymer films in the AOT(p-xylene)/water LC medium was anticipated to

2000

0.4

Pore volume (cm /g)

50

Fig. 4. FIA calibration plot for bupivacaine binding calculated from QCM measurements on HEMA-DVB polymer films of SiO2 coated Au-quartz resonators. FIA conditions: carrier buffer solution was a phosphate buffer (0.01 M) containing 150 mM NaCl at pH = 8.5, with a flow rate of flow rate of 25 µL/min and the volume of analyte injected was 75 µL.

0.04

P6 P5

0.3 0.2 0.1 0.0

100

Bupivacaine concentration , mM

0.4

0.00

P1 film P2 film P3 film P4 film P5 film P6 film

150

0

500

1000

1500

2000

Pore diameter (A) Fig. 3. Desorption pore volume plots determined from BET analysis for HEMADVB (P1-P6) polymer system. Table 2 Polymer particle physical characterization determined by BET and zeta-potential measurements of polymer particles P1-P6. Polymer system

Surface area (m2/g)

Pore volume (cm3/g)

Pore diameter (Å)

Zeta-potential (mV)

P1 P2 P3 P4 P5 P6

216.3 254.2 57.4 88.7 115.0 111.0

0.3 0.5 0.0 0.3 0.3 0.4

67.1 89.8 42.9 60.7 106.0 147.2

0.0 673.9 11.5 680.5 −162.9 71.0

microfluidics and ligand recognition. The micelle medium either decreases or increases the zeta-potential both in the presence and absence of the template, for the vinyl polymer system (HEMA-DVB, P1-P6). However, for the acrylate and acrylamide systems (P7-P11 and P12P16) the zeta-potential tends to decrease when synthesized in LC medium irrespective of the bupivacaine template. That way, the ligand binding may be enhanced owing the reduction in surface charge when compared with polymer synthesized in conventional solvents. With this inference, we have executed the molecular imprinting method to employ LC-mediated polymers as recognition elements in chemosensor for the detection of bupivacaine. For that HEMA-DVB polymer films were synthesized and evaluated the bupivacaine interactions using piezoelectric microgravimetry under FIA conditions. Au229

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

P13 (TX-100/REF)

P12 (TX-100/MIP)

P14 (AOT/MIP)

P15 (AOT/REF)

P17 (water/REF)

P16 (water/MIP)

Table 3 Sensitivities and correlation coefficients of the bupivacaine imprinted (MIP) and non-imprinted (REF) films prepared in AOT (5)-water/p-xylene medium or water.

0 -50 Mepivacaine

-100 -150

Bupivacaine

-200 500 seconds

-Resonant frequency change, Hz

Resonant frequency change (Hz)

Fig. 5. Scanning electron micrograph of MAA-BAP films synthesized on silanized SiO2 coated Au-quartz surfaces.

Ropivacaine

250 200

Bupivacaine Ropivacaine Mepivacaine

150 100

Chemosensor

Sensitivity Hz/mM

Correlation coefficient for sensitivity

P12 P13 P14 P15 P16 P17

8.23 ± 0.28 4.50 ± 0.23 116.03 ± 7.1 30.21 ± 2.58 15.03 ± 0.85 4.54 ± 0.37

0.9988 0.9959 0.9997 0.9978 0.9989 0.9799

(TX-100/MIP) (TX-100/REF) (AOT/MIP) (AOT/REF) (water/MIP) (water/REF)

50 0

1

2

3

4

5

Concentration of the analyte , mM

Table 4 Sensitivity of bupivacaine template MIP films prepared in AOT-water reverse micelle (P15) towards structural analogues of bupivacaine.

Time (Sec) Fig. 6. Resonant frequency versus time curve for the repetitive injections of bupivacaine, and analogues (7 and 8) under FIA conditions on MAA-BAP film SiO2 coated Au-quartz surfaces. Inset is the FIA calibration plot of for MIPanalyte complex. The carrier solution was a phosphate buffer solution (0.01 M) containing 150 mM NaCl at pH = 8.5, with a flow rate of flow rate of 25 µL/ min. An injection volume of 75 µL was used for the analytes at (a) 2, (b) 1, (c) 0.7, (d) 0.5, (e) 0.3 and (f) 0.1 mM concentration.

230

Analyte

Sensitivity Hz/mM

Correlation coefficient for sensitivity

Bupivacaine (1) Mepivacaine (7) Ropivacaine (8)

116.03 ± 7.1 15.73 ± 1.37 20.58 ± 1.57

0.9997 0.9998 0.9986

European Polymer Journal 106 (2018) 223–231

N. Ndizeye et al.

increase the number of accessible ligand binding cavities in close proximity to the sensor surface. The hexagonal cylindrical structures present in the LC medium comprising the polymerizable components afford imprinted polymers with nanofibrils-like structures. LC media have even been claimed to enhance the polymerization process due to improved solubility of monomers and radicals [38,39].

[15] S. Wu, N. Duan, H. Gu, L. Hao, H. Ye, W. Gong, Z. Wang, A Review of the methods for detection of staphylococcus aureus enterotoxins, Toxins 8 (2016) 176. [16] S. Chen, L. Dong, M. Yan, Z. Dai, C. Sun, X. Li, Rapid and sensitive biomarker detection using molecular imprinting polymer hydrogel and surface-enhanced Raman scattering, R. Soc. Open Sci. 5 (2018) 171488. [17] L. Malosse, P. Buvat, D. Ades, A. Siove, Detection of degradation products of chemical warfare agents by highly porous molecularly imprinted microspheres, Analyst 133 (2008) 588–595. [18] J. Sanchez-Gonzalez, R. Salgueiro-Fernandez, P. Cabarcos, A.M. Bermejo, P. Bermejo-Barrera, A. Moreda-Pineiro, Cannabinoids assessment in plasma and urine by high performance liquid chromatography-tandem mass spectrometry after molecularly imprinted polymer microsolid-phase extraction, Anal. Bioanal. Chem. 409 (2017) 1207–1220. [19] L. Elmlund, S. Suriyanarayanan, J.G. Wiklander, T. Aastrup, I.A. Nicholls, Biotin selective polymer nano-films, J. Nanobiotechnology 12 (2014) 8. [20] B.-C. Iacob, E. Bodoki, C. Farcau, L. Barbu-Tudoran, R. Oprean, Study of the molecular recognition mechanism of an ultrathin MIP film-based chiral electrochemical sensor, Electrochim. Acta. 217 (2016) 195–202. [21] N. Gao, J. Dong, M. Liu, B. Ning, C. Cheng, C. Guo, C. Zhou, Y. Peng, J. Bai, Z. Gao, Development of molecularly imprinted polymer films used for detection of profenofos based on a quartz crystal microbalance sensor, Analyst 137 (2012) 1252–1258. [22] S. Suriyanarayanan, L. Petrone, T. Ederth, I.A. Nicholls, Biotinyl moiety-selective polymer films with highly ordered macropores, Chem. Commun. 49 (2013) 5274–5276. [23] C. Lu, T. Ben, S. Xu, S. Qiu, Electrochemical synthesis of a microporous conductive polymer based on a metal-organic framework thin film, Angew. Chem. Int. Ed. 53 (2014) 6454–6458. [24] S. Suriyanarayanan, S. Mandal, K. Ramanujam, I.A. Nicholls, Electrochemically synthesized molecularly imprinted polythiophene nanostructures as recognition elements for an aspirin-chemosensor, Sens. Actuators, B 253 (2017) 428–436. [25] C. Wang, D. Chen, X. Jiao, Lyotropic liquid crystal directed synthesis of nanostructured materials, Sci. Technol. Adv. Mater. 10 (2009) 023001. [26] K. Booker, C.I. Holdsworth, C.M. Doherty, A.J. Hill, M.C. Bowyer, A. McCluskey, Ionic liquids as porogens for molecularly imprinted polymers: propranolol, a model study, Org. Biomol. Chem. 12 (2014) 7201–7210. [27] S. Suriyanarayanan, H. Nawaz, N. Ndizeye, I. Nicholls, Hierarchical thin film architectures for enhanced sensor performance: liquid crystal-mediated electrochemical synthesis of nanostructured imprinted polymer films for the selective recognition of bupivacaine, Biosensors 4 (2014) 90. [28] S.V. Ahir, P.G. Petrov, E.M. Terentjev, Rheology at the phase transition boundary: 2. Hexagonal phase of triton X100 surfactant solution, Langmuir 18 (2002) 9140–9148. [29] L. Huang, Z. Wang, H. Wang, X. Cheng, A. Mitra, Y. Yan, Polyaniline nanowires by electropolymerization from liquid crystalline phases, J. Mater. Chem. 12 (2002) 388–391. [30] J.G. Karlsson, L.I. Andersson, I.A. Nicholls, Probing the molecular basis for ligandselective recognition in molecularly imprinted polymers selective for the local anaesthetic bupivacaine, Anal. Chim. Acta. 435 (2001) 57–64. [31] C.G. Goltner, M. Antonietti, Mesoporous materials by templating of liquid crystalline phases, Adv. Mater. 9 (1997) 431–436. [32] L. Schweitz, L.I. Andersson, S. Nilsson, Capillary electrochromatography with molecular imprint-based selectivity for enantiomer separation of local anaesthetics, J. Chromatogr. A 792 (1997) 401–409. [33] K. Golker, B.C.G. Karlsson, G.D. Olsson, A.M. Rosengren, I.A. Nicholls, Influence of composition and morphology on template recognition in molecularly imprinted polymers, Macromolecules 46 (2013) 1408–1414. [34] M. Azenha, E. Schillinger, E. Sanmartin, M.T. Regueiras, F. Silva, B. Sellergren, Vapor-phase testing of the memory-effects in benzene- and toluene-imprinted polymers conditioned at elevated temperature, Anal. Chim. Acta. 802 (2013) 40–45. [35] L.I. Andersson, Selective solid-phase extraction of bio- and environmental samples using molecularly imprinted polymers, Bioseparation 10 (2001) 353–364. [36] J.P. Rosengren-Holmberg, J. Andersson, J.R. Smith, C. Alexander, M.R. Alexander, G. Tovar, K.N. Ekdahl, I.A. Nicholls, Heparin molecularly imprinted surfaces for the attenuation of complement activation in blood, Biomater. Sci. 3 (2015) 1208–1217. [37] S.S. Chang, B. Clair, J. Ruelle, J. Beauchene, F. Di Renzo, F. Quignard, G.J. Zhao, H. Yamamoto, J. Gril, Mesoporosity as a new parameter for understanding tension stress generation in trees, J. Exp. Bot. 60 (2009) 3023–3030. [38] R.A. Mackay, J. Texter, Electrochemistry in colloids and dispersions, J. Chem. Educ. 70 (1993) A199. [39] J.-B. Raoof, R. Ojani, S.R. Hosseini, Electrochemical fabrication of novel Pt/poly (m-toluidine)/Triton X-100 composite catalyst at the surface of carbon nanotube paste electrode and its application for methanol oxidation, Int. J. Hydrogen Energy 36 (2011) 52–63.

4. Conclusion Control of the morphology of cross-linked polymer resins can be achieved by the LC-mediated imprinting technique. We have showed the synthesis of HEMA-DVB, HEMA-EGDMA and MAA-BAP based polymers with molecular imprints or cavities in this study using LC media. Self-assembly of pre-polymerization monomers within the hexagonal phases of LC medium affords unique morphological and permeability features that governs analyte diffusion through microcavities and its binding with imprinted sites. Apparently, the binding and hence the resolution of separation of structurally related chemical compounds can be improved significantly. These implications have been demonstrated by utilizing LC- mediated molecularly imprinted polymer recognition films for selective determination of bupivacaine from its structurally related compounds using QCM technique. Polymer films imprinted with bupivacaine binding sites prepared either of normal and reverse micelle systems exhibit enormous sensitivity for bupivacaine over its structural analogues. References [1] S. Zhang, Fabrication of novel biomaterials through molecular self-assembly, Nat. Biotechnol. (2003) 1171–1178. [2] L. Daikhin, M. Urbakh, Effect of surface film structure on the quartz crystal microbalance response in liquids, Langmuir 12 (1996) 6354–6360. [3] Z. Zhang, D.-F. Lu, Q. Liu, Z.-M. Qi, L. Yang, J. Liu, Wavelength-interrogated surface plasmon resonance sensor with mesoporous-silica-film-enhanced sensitivity to small molecules, Analyst 137 (2012) 4822–4828. [4] K. Hotta, A. Yamaguchi, N. Teramae, Nanoporous waveguide sensor with optimized nanoarchitectures for highly sensitive label-free biosensing, Am. Chem. Soc. Nano. 6 (2012) 1541–1547. [5] A. Berrier, P. Offermans, R. Cools, B. van Megen, W. Knoben, G. Vecchi, J.G. Rivas, M. Crego-Calama, S.H. Brongersma, Enhancing the gas sensitivity of surface plasmon resonance with a nanoporous silica matrix, Sens. Actuators, B 160 (2011) 181–188. [6] G. Gaur, D.S. Koktysh, S.M. Weiss, Immobilization of quantum dots in nanostructured porous silicon films: characterizations and signal amplification for dualmode optical biosensing, Adv. Funct. Mater. 23 (2013) 3604–3614. [7] X. Niu, Y. Li, J. Tang, Y. Hu, H. Zhao, M. Lan, Electrochemical sensing interfaces with tunable porosity for nonenzymatic glucose detection: a Cu foam case, Biosens. Bioelectron. 51 (2014) 22–28. [8] M. Ates, A review study of (bio)sensor systems based on conducting polymers, Mater. Sci. Eng., C 33 (2013) 1853–1859. [9] Y. Anraku, Y. Takahashi, H. Kitano, M. Hakari, Recognition of sugars on surfacebound cap-shaped gold particles modified with a polymer brush, Colloids Surf., B 57 (2007) 61–68. [10] R. Gutierrez-Climente, A. Gomez-Caballero, M. Halhalli, B. Sellergren, M.A. Goicolea, R.J. Barrio, Iniferter-mediated grafting of molecularly imprinted polymers on porous silica beads for the enantiomeric resolution of drugs, J. Mol. Recognit. 29 (2016) 106–114. [11] S. Suriyanarayanan, H.H. Lee, B. Liedberg, T. Aastrup, I.A. Nicholls, Protein-resistant hyperbranched polyethyleneimine brush surfaces, J. Colloid Interface Sci. 396 (2013) 307–315. [12] M.J. Whitcombe, I. Chianella, L. Larcombe, S.A. Piletsky, J. Noble, R. Porter, A. Horgan, The rational development of molecularly imprinted polymer-based sensors for protein detection, Chem. Soc. Rev. 40 (2011) 1547–1571. [13] A.G. Mayes, M.J. Whitcombe, Synthetic strategies for the generation of molecularly imprinted organic polymers, Adv. Drug Deliv. Rev. 57 (2005) 1742–1778. [14] M.J. Whitcombe, N. Kirsch, I.A. Nicholls, Molecular imprinting science and technology: a survey of the literature for the years 2004–2011, J. Mol. Recognit. 27 (2014) 297–401.

231