Molecularly imprinted chitosan-based adsorbents for the removal of salicylic acid and its molecular modeling to study the influence of intramolecular hydrogen bonding of template on molecular recognition of molecularly imprinted polymer

Abstract

The present study reports the synthesis of molecularly imprinted chitosan-based adsorbents using 4-hydroxy benzoic acid (4HBA) as a dummy template for the selective extraction and analysis of salicylic acid (SA) using suspension polymerization method. Dummy template strategy is useful when templates like salicylic acid exhibit intramolecular hydrogen bonding. The synthesized adsorbents were characterized by Fourier transform infrared spectroscopy (FTIR) and scanning electron micrograph (SEM). The adsorption capacity of the adsorbents was investigated as a function of time, concentration of analyte and mass of adsorbent employed and was found to be 500.96 µmol g⁻¹ for the chitosan-based 4-hydroxy benzoic acid molecularly imprinted polymer (4HBACTSMIP), 221.2 µmol g⁻¹ for the salicylic acid MIP (SACTSMIP), 220 µmol g⁻¹ for the non-imprinted polymer (NIP), 149.7 µmol g⁻¹ for chitosan (CTS) and 60.63 µmol g⁻¹ for the granular activated carbon (GAC). The specific binding capacity of 4HBACTSMIP for salicylic acid was found to be 280.96 µmol g⁻¹. The change in the Gibbs free energy (ΔG) for the template–monomer complexes formed between the templates and the functional monomer was computed using density functional theory (DFT) to interpret the nature of the interactions between them and to compare their stabilities. A systematic investigation of the molecular recognition abilities of the synthesized adsorbents was carried out using Freundlich (F) adsorption isotherm model. The binding parameters obtained from the F model demonstrate that 4HBACTSMIP exhibited higher specific molecular recognition ability towards salicylic acid. The adsorption ability of SACSTMIP was same as that of NIP due to intramolecular hydrogen bonding present in salicylic acid molecule leading to low imprinting effect. The results are supported by molecular modeling studies and the selectivity studies.

Keywords

Chitosan molecularly imprinted polymer dummy template salicylic acid molecular modeling 4-hydroxy benzoic acid density functional theory selectivity

Introduction

Contamination of aquatic media by organic pollutant is a serious environmental problem. The main causes for the contamination are the discharge of industrial waste, excretion of unmetabolised medication into wastewater collection systems, excretion of veterinary drugs by farm animals into fields run off and surface water, improper commercial disposal methods and the release of the personal care products (PPCPs) like cosmetics, perfumes, and sunscreens into water while bathing. Amongst various techniques, adsorption has been reported as the most efficient method for removal of residual pollutants from wastewater (Dutta et al., 2004; Guibal, 2004; Varma et al., 2004). A wide range of adsorbents have been used for removing organic pollutant from contaminated water such as cellulosic wastes and their carbonisation products, bituminous coal and commercial activated carbons. However, the high cost of the activation process limits their use in wastewater treatment (Beltran et al., 2010; Mourya and Inamdar, 2008).

Chitosan (CTS) is a cheap bio-sorbent and is the N-deacetylated product of chitin which is the second most abundant natural biopolymer after cellulose. Both chitin and CTS have unique combination of properties like biodegradability, biocompatibility and bioactivity. These biodegradable polymers with good physical, chemical and mechanical properties have gained more attention in the application of adsorption techniques. Combined with the advantage of abundant availability and low price, their use has been prevalent in the pollution control operations. Presence of variety of the functional groups like primary amino, acetamido, and hydroxyl groups on CTS is responsible for its high chemical reactivity and therefore supports its use for removal of pollutants. CTS can be modified by chemical or physical processes in order to improve the mechanical and chemical properties (Dutta et al., 2004; Mourya and Inamdar, 2008). Numerous CTS derivatives have been reported with good mechanical stability by crosslinking CTS with various crosslinking agents like with epichlorohydrin, glutaraldehyde or ethylene glycol dimethacrylate. Grafting new functional groups on the CTS backbone can lead to increase in the adsorption selectivity. One of the novel methods of improving the adsorption capacity of CTS-based adsorbents is by molecular imprinting technique which creates specific recognition sites for higher adsorption. Recently, stress has been placed on the synthesis of molecularly imprinted polymers (MIP) using CTS as functional monomer for enhancing the adsorption efficiency of CTS-based adsorbents. CTS because of its versatility can be easily modified into different forms which increase its adsorption efficiency. Molecularly imprinted membrane has been synthesised using CTS as functional monomer for selective recognition of flavonoids in aqueous media. A novel molecularly imprinted nanofibre was synthesised by using β-cyclodextrin as monomer for the selective recognition of flavonoids using a pore forming agent to enhance the binding capacity of the polymers. MIPs have been effectively used as sensor material for sensing trace quantity of emerging persistent organic pollutants (Ma et al., 2011, 2013, 2016).

MIP has selective recognition sites for the template molecule which are generated within the polymer by allowing the template molecule to react with the functional monomer in the presence of porogen to form a template–monomer complex (Pardeshi et al., 2012). The type of interaction involved between the template and the monomer may be either covalent or non-covalent which should result in the formation of highly stable prepolymerisation complex (template–monomer complex). On polymerisation, the crosslinking agent freezes the prepolymerisation complex within the polymeric matrix and extracting the template molecule using a suitable solvent from the prepolymerisation complex results in the vacant binding sites with shape and size complementary to the structure of template molecule. The selectivity of the MIP is directly proportional to the number of such selective binding site within the polymeric matrix. However, the template with strong intramolecular hydrogen bonding generally leads to very weak non-covalent interaction between the template and the monomer. In such cases ‘dummy template strategy’ can lead to the higher selectivity in MIPs. Molecularly imprinted CTS-based adsorbents are found to be efficient at adsorbing residual pollutants from wastewater (Bai et al., 2008; Jiang et al.,2013; Kyzas et al., 2008, 2013; Shinde et al., 2013; Yu et al., 2008).

One large class of chemicals that have received comparatively little attention is Pharmaceuticals and active ingredients in the PPCPs, which are extensively used throughout the world. PPCPs and their metabolites are ubiquitous and display persistence into the aquatic environment. SA is a prominent metabolite of aspirin (acetylsalicylic acid, ASA), which is widely used as an analgesic for the prevention of heart attack, in keratolytic and dermatitis treatment and also as a preservative in the food items. SA has been identified as one of the PPCPs (Huang et al., 2003; Meng et al., 2013; Shi et al., 2014). In the United States alone, approximately 20,000 million tablets of Aspirin are consumed annually. ASA has been reported as the world’s first synthetic drug possessing anti-inflammatory activity against rheumatism. At present, ASA has achieved clinical prominence as a globally relied therapeutic agent not only because of its traditional anti-inflammatory paradigm, but also for its more advanced life-saving perspective related to cardiovascular events such as heart attacks and stroke (Sreenivasan, 2007). Generally, aspirin can be easily hydrolysed to SA during storage. Therefore, excess concentration of SA is often detected in aspirin, which can cause harmful effect to stomach and intestine, may induce the mucosa bleeding and even cause death in serious case. Therefore, monitoring the concentration of SA is important for clinical and environmental applications (You et al., 2014; Zhang et al., 2001, 2002).

As per the definition of MIP, it could be possible to synthesise a corresponding MIP for every compound. However, literature survey reveals that only limited number of compounds have been used as templates for the synthesis of MIP. The stability of MIP depends upon the stability of prepolymerisation complex formed between template and functional monomer. Compounds like SA with strong intramolecular hydrogen bonding interaction generally exhibit weak intermolecular hydrogen bonding interaction with the functional monomer resulting in poor imprinting effect. This can lead to corresponding MIP with very low selectivity. However this can be overcome by using ‘dummy templates strategy’ for the synthesis of MIP. Dummy templates are the structural analogues of target analytes. For the present study, 4-hydroxy benzoic acid (4HBA) was used as dummy template for the selective enrichment and analysis of SA (Huang et al., 2003; Meng et al., 2013; Shi et al., 2014; Sreenivasan 2007; You et al., 2014).

Computational approaches, such as molecular modeling, have been used in the molecularly imprinted polymer area to guide synthesis and performance by the design of virtual libraries that screen the best possible functional monomers for the preparation of an MIP for a specific template and also to study the nature of interaction between template–monomer complex or prepolymerisation complex (Pardeshi et al., 2012).

The present work reports “dummy template strategy” for the synthesis of highly efficient CTS-based molecularly imprinted polymer for the removal of SA using CTS as a functional monomer, 4HBA as a dummy template and epichlorohydrin as a crosslinking agent by suspension polymerisation method. Suspension polymerisation method is the simplest and most widely employed method for the preparation of MIP in beaded form (Biffins et al., 2012). Molecular modeling studies were carried out to support the importance of the ‘dummy template approach’ in synthesising highly efficient CTS-based MIP for the removal of SA, an emerging contaminant. Binding capacities of the various reported polymers along with the porogen and dummy template used in polymerisation when compared with the present method in Table 1, proves the present method to be efficient. Binding capacity of the present method was compared with the common adsorbents like granular activated carbon (GAC) and CTS. To the best of our knowledge, for the first time, the computational approach has been used to demonstrate the importance of the ‘dummy template strategy’ for the synthesis of MIP for SA which exhibits strong intramolecular hydrogen bonding using an environmental friendly functional monomer CTS for the selective enrichment and analysis of SA, an emerging contaminant.

Table 1.

Re-binding capacities of different adsorbents for SA prepared using dummy template.

Dummy template	Monomer	Method	Solvent	Specific binding capacity (B_MIP – B_NIP) (µ mol g⁻¹) / (mg g⁻¹)	Reference
Benzoic acid	4 Vinyl pyridine	Bulk polymerisation	ACN	82.54/11.4	You et al., 2014
Benzoic acid	4 Vinyl pyridine	Bulk polymerisation	ACN	60.09/8.3	Shi et al., 2014
4HBA	4 Vinyl pyridine	Bulk polymerisation	ACN	36.20/5.0	Shi et al., 2014
4HBA	Chitosan (CTS)	Suspension polymerisation	Mineral oil	280.96/38.81	Present study

Materials and methods

Chemicals

CTS with >75% degree of deacetylation, epichlorohydrin, 4HBA, SA, 2,5 dihydroxy benzoic acid (2,5DBA), gallic acid (GA), granular activated carbon (GAC), sorbitan monooleate (Span 80), and formaldehyde were supplied by sigma Aldrich. Liquid paraffin and sucrose were obtained from Fisher scientific. Sodium hydroxide, acetic acid, Acetonitrile, and high performance liquid chromatography (HPLC) grade water were provided by Merck chemicals. All other chemicals were of analytical grade and supplied by Rankem.

Spectrophotometric analysis

Spectrophotometric analyses were performed using a Shimadzu (Japan) UV-visible spectrophotometer with matched 1 cm quartz cells. Calibration curves at 298 nm for salicylic acid adsorption experiments were prepared using standard solutions of salicylic acid in methanol and water (10:90 v/v).

Morphological characterization

4HBACTSMIP, SACTSMIP, NIP, CTS, and GAC were analysed using a scanning electron microscope (SEM, JEOL JSM-6360LV). The samples were coated with a thin gold film before observation in SEM.

The FTIR spectra of compounds were obtained using a Fourier transform infrared spectrometer (IR-Affinity-1, Shimadzu). The samples were ground with anhydrous KBr and the spectra were recorded between 4000 and 400 cm⁻¹, by averaging 45 scans for each spectrum.

Preparation of 4HBACTSMIP and SACTSMIP

CTS (0.6 g) was dissolved in 30 mL of 2% acetic acid solution (v/v) and mixed with 15 mL of methanol containing 193 mg of dummy template 4HBA. This mixture was stirred for 30 minutes; 50 mL of liquid paraffin containing 0.3 mL of span 80 and 2 g of sucrose were added to this mixture and stirred at 50℃ for 10 minutes. Further, 1 mL of formaldehyde solution (37%, w/w) as amino protective solute was added to this mixture and stirred for 30 minutes. After adjusting the pH to 9.0 with 2 M sodium hydroxide solution, 1 mL epichlorohydrin was added drop wise into the mixture for crosslinking. The crosslinking reaction was performed by agitating the above mixture at 70℃ for 3 hours with the pH maintained at 9.0 throughout the process. The prepared 4HBACTSMIP was washed with acetic acid: methanol (20:80 v/v) to remove the unreacted precursors. The dummy template was extracted using acetic acid: methanol (20:80 v/v) in a Soxhlet apparatus for 24 hours followed by methanol and finally digested with 0.1% hydrochloric acid to remove the amino protective solute. 4HBACTSMIP was thoroughly washed with deionized water and dried in a vacuum drier before storing in vacuum desiccator. SACTSMIP was prepared using the same method with SA as the template. The control polymers non-imprinted polymer (NIP) were also prepared using same procedure in the absence of template molecule 4HBA or SA (Wang et al., 2014).

Binding experiments

The binding experiments were carried out batch-wise in triplicate to study the recognition performance of 4HBACTSMIP, SACTSMIP, NIP, GAC, CTS in the binding solvent methanol and water (10:90 v/v) using the method given below.

Dose optimisation

The dose of each adsorbent was varied in the range 10–100 mg and equilibrated in 10 ml solutions of 4 mmol L⁻¹ SA using a water bath shaker for 2 hours and analysed for SA.

Time optimization

The contact time was optimised by equilibrating the optimised dose of each adsorbent for the fixed time period ranging from 10 minutes to 3 hours with 4 mmol L⁻¹ of SA solution. SA was analysed after each time interval.

Selectivity study

4HBA, GA, 2,5DBA and SA were used as structural analogues to verify the selectivity of MIPs for the template over other structurally similar compounds.

Figure 1.

SEM images of (a) 4HBACTSMIP, (b) NIP, (c) SACTSMIP, (d) CTS and (e) GAC.

4HBACTSMIP, SACTSMIP, and NIP were equilibrated with a mixture containing 5 mmol L⁻¹ solutions of SA and 4HBA, at optimized condition for the optimum time period. After equilibration, the equilibrium concentrations of SA and 4HBA were analysed using HPLC.

4HBACTSMIP, SACTSMIP, GAC, CTS, and NIP were equilibrated with a mixture containing 5 mmol L⁻¹ solution of SA, 4HBA, 2,5DBA, and GA at optimized condition for the optimum time period. The residual concentration of SA and 4HBA, 2,5DBA, and GA in the solution was analysed after equilibration, using HPLC.

HPLC and Chromatographic conditions

Waters HPLC equipped with a quaternary pump (2545 waters HPLC pump and photodiode array (PDA) detector (2998 Waters)), ODS-SP C18 Column (5 µm, 4.6 × 250 mm; Merck) was used for the selectivity studies. Chromatographic conditions were as follows: water/acetonitrile/acetic acid (70:30:0.2 v/v) as mobile phase, flow rate of 1 mL min⁻¹, injection volume of 10 µL, and UV detection at 281 nm. The concentrations were determined using peak area.

Binding isotherm

Re-binding experiments were carried out to study the binding capacity of the MIPs. A series of standard SA solutions ranging from 0.5 to 5 mmol L⁻¹ were prepared; 10 mL aliquots of each standard solution were mixed with the optimised dose of each adsorbent and shaken for the determined optimum time period. The solutions were then analysed for SA. The binding capacity, B (µmol g⁻¹) of each adsorbent was calculated using the following equation

Q = \frac{(C_{initial} - C_{final}) V_{solution} \times 1000}{W}

(1)

where C_inital and C_final are the initial and final concentration of the SA solution, respectively. V_solution is the volume of the SA solution and W is the weight of the adsorbent.

Kinetic study

For kinetic experiments, 50 mg of each adsorbent was suspended in 10 mL of 4 mmol L⁻¹ SA solution and continuously shaken at room temperature. The concentrations of SA at the time intervals of five minutes (5–40 minutes) were analysed by UV-visible spectrophotometer at 298 nm and then the adsorption capacity Q_t (µmol g⁻¹) at different contact times t was calculated.

Molecular modeling

Molecular modeling studies can prove to be useful for the selection of suitable grafting agent for the functionalisation of CTS and also for the selection of suitable monomer for MIP synthesis (Farrington et al., 2006). Molecular modeling studies include several steps, as follows (Pardeshi et al., 2012).

Building of models: The first step is building of molecular models using Chem 3D Ultra software. The 3D structures are drawn and Cartesian coordinates of stable conformers are generated to prepare input file for running the Gaussian simulations.

Geometry optimization: The B3LYP (Becke-Style 3-parameter density functional theory (DFT) using the Lee–Yang–Parr correlation function) with 6-31 G (d) basis set is used for geometry optimisations to obtain structures with minimum energy.

Frequency calculations: The harmonic vibrational frequencies confirmed the structures as minima with no imaginary frequencies and enabled the calculation of Gibbs free energies.

Results and discussion

SEM result

Scanning electron micrographs of 4HBACTSMIP, NIP, SACTSMIP, GAC, CTS are presented in Figure 1. It can be seen that the NIP presents a relatively smooth surface compared to the rough porous surface of 4HBACTSMIP. This must have resulted in the higher binding capacity of 4HBACTSMIP as compared to NIP. SEM images also indicated that the beads formed were different in size (polydispersed). Amongst GAC and CTS, GAC depicted relatively smooth surface.

Figure 2.

FTIR spectra of CTS, 4HBACTSMIPBE (before Extraction), 4 HBACTSMIPAE (after extraction) and NIP.

FTIR result

The FTIR spectra of CTS, 4HBACTSMIPAE (after extraction of 4HBA), 4HBACTSMIPBE (before extraction of 4HBA), and NIP are presented in Figure 2. The spectrum of CTS exhibits a broad peak around 2400 to 3400 cm⁻¹ due to the presence of intramolecular H bonding between –NH₂ and –OH groups present in CTS. The peak around 1650 cm⁻¹ can be attributed to the amide group in CTS. (Brugnerotto et al., 2001). The spectra of 4HBACTSMIPBE (before extraction of template 4HBA) show decrease in the intensity around 2400 to 3400 cm⁻¹ which can be attributed to the formation of intermolecular hydrogen bond between –OH group of CTS and carbonyl group of 4HBA, thus increasing the O–H bond length in CTS. These results are in accordance with the molecular modeling studies. The spectra of NIP and 4HBACTSMIPAE (after extraction of 4HBA) were seen to be almost similar. FTIR results depict that the dummy template 4HBA has been successfully imprinted in CTS via hydrogen bonding between carbonyl group of 4 HBA and –OH group in the functional monomer CTS.

Figure 3.

Effect of time on adsorption of salicylic acid onto 4HBACTSMIP, SACTSMIP, NIP, CTS and GAC.

Suspension polymerisation method

Generally in suspension polymerisation method, water is used as a continuous phase to suspend a droplet of prepolymerisation complex (template–monomer complex). MIPs prepared by suspension polymerisation method are polydispersed but shows enhanced affinity for template molecule in comparison to NIP. The presence of water as a continuous phase during polymerisation is expected to negatively affect the efficiency of imprinting process by disrupting the weak hydrogen bonding or dipolar interactions involved in imprinting process. To overcome this limitation, alternative solvent has been used as dispersing medium which preserves non-covalent interactions in the template–monomer complex (4HBA-CTS or SA-CTS). In the present method, mineral oil was employed as continuous phase instead of water. The added advantage of using mineral oil is that no stabiliser is needed during synthesis (Biffins et al., 2012).

Batch re-binding studies

All experiments were performed in triplicate for reasons of accuracy and precision.

Optimization of time

The adsorption experiment was carried using 20 mg of 4HBACTSMIP, in neutral pH at room temperature using 10 mL of 4 mmol SA solution, for contact time ranging from 5 minutes to 180 minutes. The results are shown in Figure 3. It was observed that the adsorption capacity of 4HBACTSMIP for SA increased with the increase in the contact time and reached adsorption equilibrium within 100 minutes. Similar trend was observed for SACTSMIP, NIP, CTS, and GAC. Therefore, all subsequent adsorption experiments were performed at optimum equilibrium time of 100 minutes.

Figure 4.

Effect of adsorbent dose on Binding capacity of 4HBACTSMIP, SACTSMIP, NIP, CTS and GAC.

Optimization of dose

It was observed that for the given SA concentration, increase in adsorbent dosage of 4HBACTSMIP, CTS, GAC, SACTSMIP, and NIP in each case leads to the increase in adsorption capacity as shown in Figure 4. The maximum binding was observed at the dose of 50 mg of adsorbents in most of the adsorbents and was used for the further experiments. The probable reason is that the increase in adsorbent dosage provides more functional group and active sites, thus leading to the increase in adsorption capacity (Liu et al., 2010).

Figure 5.

Effect of concentration on binding capacity of 4HBACTSMIP, SACTSMIP, NIP, GAC and CTS.

Adsorption isotherms and binding site characterization

It was observed that the binding capacity of 4HBACTSMIP increased with the increase in concentration of SA and reached adsorption equilibrium at higher concentration as shown in Figure 5. With increase in the SA concentration, more active sites of 4HBACTSMIP were involved in the adsorption process. Higher concentration enhances the mass transfer driving force, thus increasing the adsorption capacity at equilibrium. In addition, with increasing concentration, the number of collision between template molecule and adsorbent also increases which in turn enhances adsorption process (Jin et al., 2008; Umpleby et al., 2001). Similar trend was observed for SACTSMIP, NIP CTS, and GAC.

Figure 6.

Adsorption isotherm of 4HBACTSMIP, SACTSMIP, NIP, GAC, and CTS using Freundlich (F) adsorption isotherm model.

The purpose of the adsorption isotherms is to relate the adsorbate concentration in the bulk with the adsorbed amount at the interface. The concentration of SA bound to the polymers, i.e. the binding capacity (B), was plotted against the free concentration of SA in the solution .The isotherm results were analysed using the Langmuir and Freundlich isotherms. The experimental data were found to fit better into the Freundlich isotherm equation as compared to the Langmuir isotherm equation (Umpleby et al., 2001).

The Freundlich isotherm is an empirical equation employed to describe heterogeneous systems. The Freundlich equation is expressed as

Q_{e} {= K}_{F} C_{e}^{1 / n}

(2)

where K_F and n are Freundlich constants with K_F as the adsorption capacity of the adsorbent and n giving an indication of how favourable is the adsorption process. The magnitude of the exponent, 1/n, gives an indication of the favourability of adsorption (adsorption intensity). Values of n > 1 represent favourable adsorption condition. To determine the constants K_F and n, the linear form represented in equation (3) is used to produce a graph of ln(Q_e) against ln(Ce) (Figure 6)

\ln Q_{e} {= lnK}_{F} + l / n \ln Ce

(3)

Figure 7.

Pseudo first-order kinetic plot for 4HBACTSMIP, SACTSMIP, NIP, GAC, CTS.

Values of K_F and n are calculated from the intercept and slope of the plot, respectively (Figure 6). Freundlich equation can be used to describe a monolayer adsorption as well as a multilayer adsorption. The values of K_F and 1/n are listed in Table 2.

Table 2.

Fitting parameters of the Freundlich model for different adsorbents.

Adsorbents	1\n (adsorption intensity)	K_F (adsorbents relative adsorption Capacity) (µmol g⁻¹)	R²
CTS	0.448	15.84	0.989
GAC	0.441	18.41	0.990
NIP	0.468	173.78	0.982
SACTSMIP	0.477	276.69	0.980
4HBACTSMIP	0.497	467.75	0.994

Kinetic Study

Kinetic studies were performed for different adsorbents and the time dependent adsorption profiles are shown in Figure 3. SA adsorption on 4HBACTSMIP was quite rapid in the first 80 minutes, accounting for 84% of SA removal (414 µmol g⁻¹). Adsorption on SACTSMIP and NIP on the other hand, proceeded much more slowly, with 40% removal of SA achieved after 90 minutes contact time. Adsorbent saturation was reached at 90 minutes for 4HBACTSMIP, whereas NIP and SACTSMIP attained saturation in 45 minutes. The fast adsorption kinetics and high capacity of 4HBACTSMIP resulted from improved imprinting effect and porous structure. To better understand the adsorption behaviour of SA onto the 4HBACTSMIP, two conventional kinetic models were applied to analyse the kinetic data. The pseudo first-order and pseudo second-order kinetic equations are shown in equations (4) and (5), respectively (An et al., 2011; Hameed et al., 2008).

\ln (q_{e} - q_{t}) = \ln q_{e} - k_{1} t

(4)

t / q_{t} = 1 / k_{2} q_{e}^{2} + t / q_{e}

(5)

where k₁ (min⁻¹) and k₂ (g mg⁻¹min⁻¹) are the rate constants of pseudo first-order and pseudo second-order adsorption, respectively; q_e and q_t (mg g⁻¹) are the amounts of SA adsorbed at equilibrium and at time t, respectively. The pseudo first-order plot is shown in Figure 7.

Figure 8.

Structures of 4-hydroxy benzoic acid (4HBA), Gallic acid(GA), 2,5 Dihydroxy Benzoic Acid (2,5DBA) and Salicylic acid (SA) and the intramolecular hydrogen bonding in 2,5DBA and SA.

Kinetic parameters and correlation coefficients for the adsorption of SA by different adsorbents are summarised in Table 3. For many adsorption processes, the pseudo first-order kinetics was found to be suitable for the initial 20 to 30 minutes of interaction time and not for the whole range of contact times. As expected, the rate of adsorption of 4HBACTSMIP was faster than SACTSMIP and NIP by pseudo first-order modeling. The correlation coefficient (R²) for the pseudo first-order model exceeded 0.97 for 4HBACTSMIP and 0.96 for SACTSMIP and NIP which indicates accurate fitting of experimental data, whereas the pseudo second-order model did not fit well for the binding data of all the adsorbents. The experimental adsorption capacity (q_e,_exp) was in good accordance with the calculated adsorption capacity (q_e,_cal) obtained from the pseudo first-order model. This result indicates that the pseudo first-order kinetic model accurately fits the adsorption of SA on 4HBACTSMIP, NIP, and SACTSMIP. These data clearly indicate that 4HBACTSMIP is superior to the SACTSMIP and other adsorbents for the selective adsorption of SA from water.

Table 3.

Adsorption kinetic data of different adsorbents for pseudo first-order model.

Adsorbent	k₁	q_e(cal) (µmol g⁻¹)	q_e(exp) (µmol g⁻¹)	R²
GAC	0.018424	73.96	60.63 ± 4.25	0.948
CTS	0.04606	147.57	149.7 ± 4.47	0.948
SACSTMIP	0.062181	213.79	220 ± 5.24	0.964
NIP	0.059878	189.234	221.2 ± 3.26	0.975
4HBACSTMIP	0.06909	537.03	500.96 ± 8.87	0.979

Selectivity study

4HBACTSMIP, SACTSMIP, GAC, CTS, and NIP were equilibrated with the mixture of SA, 4HBA, 2,5DBA, and GA and the solutions were analysed by HPLC. The binding capacities of 4HBACTSMIP, SACTSMIP, GAC, CTS, and NIP are presented in the Table 4. It can be seen that, 4HBACTSMIP has highest binding capacity for 4HBA because of molecular imprinting effect. SA, 2,5DBA, and GA are structural analogues of 4HBA. The difference in their structures is just the number of hydroxyl groups and their relatively different positions with respect to carboxylic groups, different molecular sizes and the presence of intramolecular H-bonding. From Table 4 it is observed that the binding capacity of 4HBA is much higher than that of SA, 2,5DBA or GA, which indicates that the molecular recognition ability of the 4HBACTSMIP is high for HBA. The plausible reason is that the cavities in polymeric matrix of 4HBACTSMIP are complementary both in shape and the functional group arrangement to that of 4HBA. Because 4HBA fits the cavities best, it has the highest binding capacity amongst all the structural analogues. It can be speculated that the smaller binding capacity for GA may have resulted from the weaker interaction between GA and the binding sites within the 4HBACTSMIP because of the larger molecular volume of GA (64.46 Å³) as compared to the molecular volume of 4HBA (54.53 Å³) (Pardeshi et al., 2014). However, the higher binding of GA as compared to SA and 2,5DBA onto 4HBACTSMIP may be due to the absence of intramolecular hydrogen bonding in GA. The higher binding capacity of 4HBACTSMIP for 4HBA as compared to SA and 2,5DBA can be attributed to the presence of intramolecular H-bonding in SA and 2,5DBA as can be seen in Figure 8.

Table 4.

Binding capacities of SACTSMIP, 4HBACTSMIP, NIP, GAC, and CTS for SA and its structural analogues.

Adsorbent	Binding Capacity (µmol g⁻¹)
Adsorbent	4HBA	GA	SA	2,5DBA
4HBACTSMIP	546 ± 8.25	442 ± 7.89	401 ± 8.2	345 ± 8.54
SACTSMIP	298 ± 6.24	284 ± 6.39	202 ± 6.06	186 ± 7.35
NIP	297 ± 3.24	281 ± 5.74	199 ± 5.98	184 ± 5.52
CTS	264 ± 2.87	238 ± 3.89	176 ± 5.28	156 ± 4.68
GAC	251 ± 2.38	217 ± 2.78	164 ± 4.92	136 ± 4.08

It is also observed that the binding capacity of SA is less than that of GA although SA has two hydroxyl group less than GA. Least binding capacity was observed for 2,5DBA which has one hydroxyl group lesser than GA. Low binding capacity for 2,5DBA and SA may have resulted due to weak interaction between 2,5DBA or SA with the binding cavities in the 4 HBACTSMIP because of the intramolecular hydrogen bonding between the carboxylic group and the adjacent hydroxyl group in 2,5DBA and SA molecule. This clearly indicates that the intramolecular hydrogen bonding within the template molecule can adversely affect the imprinting efficiency. Amongst 2,5DBA and SA, SA had greater binding capacity due to the molecular volume being nearly same as 4HBA than 2,5DBA (61.56 Å³). Thus, it can be concluded that the 4HBACTSMIP was found to be much selective for 4HBA over its structural analogues.

SACTSMIP was synthesised to study the influence of intramolecular hydrogen bonding of templates on imprinting using SA as template which exhibits intramolecular hydrogen bond, and its molecular recognition ability was tested. Table 4 clearly indicates that the binding capacity of SA and other structural analogues (GA, 2,5DBA, and 4HBA) onto SACTSMIP is almost similar to that of NIP. Thus, it can be concluded that SACTSMIP had no molecular imprinting effect.

Chromatogram for 4HBA and SA before and after equilibration with NIP, SACTSMIP, and 4HBACTSMIP is presented in Figure 9. The chromatogram shows a decrease in peak intensity after equilibration. The decrease in peak intensity was found to be greater for 4HBACTSMIP and was nearly same for SACTSMIP and NIP for SA.

Figure 9.

Chromatogram for 4HBA and SA after equilibration with NIP, SACTSMIP, and 4HBACTSMIP.

Thus, it can be concluded that 4HBACTSMIP was found to be more selective for 4HBA over its structural analogues because of molecular imprinting effect. However, in case of SACTSMIP, the binding capacity of SA and all other structural analogues was found to be almost similar as that of NIP because of no molecular imprinting effect. The binding capacity of SA on 4HBACTSMIP was found to be greater than on SACTSMIP. Thus, the use of the dummy template 4HBA, the structural analogue of SA for the selective enrichment, and analysis of SA can be justified for the synthesis of MIP.

Amongst GAC and CTS, it was observed that CTS had greater binding capacity than GAC for all structural analogues which may be due to the presence of more active sites on CTS.

Molecular modeling to study the influence of intramolecular H-bond of 4HBA and SA on molecular recognition of molecularly imprinted polymer

Compounds like SA with strong intramolecular hydrogen bond exhibit weak intermolecular hydrogen bonding interactions with functional monomers in the imprinting process. Consequently, the corresponding MIPs exhibit no specific adsorption ability. In SA, a strong intramolecular hydrogen bond is formed between the adjacent –OH and –COOH groups on benzene ring resulting into the formation of six membered ring (Figure 8), as a result the intermolecular H-bonding interaction between SA and CTS is weakened, resulting into poor imprinting effect. But in 4HBA molecule the –COOH and –OH groups are at para position (Figure 8) so they are unable to form intramolecular H-bonding although it exhibits intermolecular H-bonding, not as strong as the intramolecular H-bond in solution (Meng et al., 2013; Shi et al., 2014; Sreenivasan 2007; You et al., 2014; Zhang et al., 2001, 2002). The strength of intermolecular H-bonding is related to the concentration of components. In the present study the concentration of 4HBA is in low concentration of 1 mmol. Therefore, the intermolecular H-bonding would not be very strong amongst the 4HBA molecules in the solution. When suitable excess amount of the functional monomer CTS is added to the polymerisation mixture, the intermolecular H-bonding among 4HBA will be weakened by large amount of CTS because the chance of interaction between CTS and 4HBA is increased. Thus, the stable 4HBA-CTS complex is formed through intermolecular H-bonding between CTS and 4HBA consequently leading to the formation of specific binding sites for 4HBA in 4HBACTSMIP with high imprinting effect. When the binding capacity of 4HBACTSMIP and SACTSMIP for SA was compared, it was observed that 4HBACTSMIP had greater binding capacity than SACTSMIP as SA had poor imprinting effect (Table 4). These results are in line with the molecular modeling studies.

Molecular modeling

The type of interaction between templates (4HBA, SA) and CTS was also confirmed with the help of DFT by calculating Gibbs free energy (ΔG) during the complex formation. The Gibbs free energy gains of the complexes were calculated using equation (6)

Δ G = G_{template - monomercomplex} - [G_{template} + G_{monomer}]

(6)

where ΔG is the change in Gibbs free energy on the formation of template–monomer complex, G_{template–monomer} complex is the Gibbs free energy of template–monomer complex, G_template is the Gibbs free energy of template, and G_monomer is the Gibbs free energy of monomer molecules.

Figure 10(a) to (e) (see online data supplement for supporting information) shows the optimised geometries of the template (SA and 4HBA), monomer (CTS), and template–monomer complexes. Hydrogen bonds can be formed between proton donor such as amide, amine or hydroxyl group and proton acceptor such as carbonyl group. Similar interactions were observed between proton donor (CTS) and proton acceptor (SA, 4HBA) to form template–monomer complex. Figure 10(d) and (e) shows that there is a strong interaction between CTS and 4HBA (1.858 Å) as compared with CTS and SA (1.897 Å). After H-bond formation with template molecule, an increase in the O–H bond length of CTS was observed in the case of CTS-4HBA complex (0.970 Å to 0.978 Å Figure 10(c) and (d)), while in case of CTS-SA complex, the O–H bond length of CTS changed from 0.970 to 0.977 (Figure 10(c) and (e)). Increase in the O-H bond length of CTS was observed to be greater for CTS-4HBA complex due to strong interaction between CTS and 4HBA (1.858 Å) as compared to CTS and SA complex (1.897 Å).

Figure 10.

(a) Optimized structure of SA with Intramolecular H-bonding. (b) Optimized structure of 4HBA. (c) Optimized structure of chitosan monomer. (d) Optimized structure of chitosan −4 HBA complex (template–monomer complex). (e) Optimized structure of chitosan-salicylic acid complex (template–monomer complex).

The nature of the template–monomer complex determines the quality and performance of the MIPs. Moreover, the quantity and quality of molecularly imprinted polymer recognition sites are directly related to the mechanisms and the extent of the template–monomer interactions present in the pre-polymerisation mixture. Thus, the strength of the prepolymerisation complexe determines the ability of the MIP to selectively re-bind the template. As a measure of the affinity between the template and the monomer, the changes in the Gibbs free energy (ΔG) during template–monomer complex formation were calculated and are presented in Table 5.

Table 5.

Gibbs free energy of template–monomer complex.

Template–monomer complex	ΔG (Kcal/mol)
CTS-4HBA	−18.959
CTS-SA	−13.397

CTS formed stable complex with 4HBA than with SA. A possible explanation for the stability of the CTS-4HBA complex is that SA forms strong intramolecular hydrogen bond; as a result, the intermolecular H-bonding interaction between SA and CTS gets weakened. But in 4HBA molecule, the –COOH and –OH groups are at para position and are therefore unable to form intramolecular H-bonding. This leads to the formation of stable CTS-4HBA complex through intermolecular H-bonding between CTS and 4HBA.

Comparative study

Comparison of the binding capacity of different adsorbents for SA can be seen in Table 6. It was observed that CTS had greater binding capacity than GAC which may be due to the presence of more number of active sites on CTS for SA uptake. 4HBACTSMIP has shown greater binding capacity than SACTSMIP for SA due to poor imprinting effect as discussed above. Binding capacity was found to be the highest for 4HBACTSMIP as compared to the other adsorbents used in this study.

Table 6.

Comparative study for binding of salicylic acid using different adsorbents.

Adsorbent	Binding capacity (µmol g⁻¹)
GAC	60.63 ± 4.23
Chitosan	149.7 ± 3.98
NIP	220 ± 2.97
SACTSMIP	221.2 ± 1.987
4 HBACTSMIP	500.96 ± 2.597

Conclusion

Present work demonstrates the preparation of an efficient 4HBACTSMIP for the uptake of SA from the aqueous solution with specific binding capacity of 38.8 mg g⁻¹. The use of CTS as the functional monomer complies with greener approach for the polymer preparation. Experimental and modeling results show that the molecular recognition ability of the MIP decreases if the template forms intramolecular hydrogen bonding, when hydrogen bonding is responsible for the molecular recognition of the imprinted polymer as in case of SACTSMIP. The equilibrium data fitted well to Freundlich equation and the adsorption process could be described by pseudo first-order model. The synthesized adsorbent can prove useful for the removal as well as recovery of SA from aqueous matrix.

Footnotes

Acknowledgements

Sincere thanks to Kavita Pande and Dr. Rita Dhodapkar for their continuous help and support.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental material

The online supporting information for Figure 10(a) to (e) is available at .

References

Gao

Dai

(2011) Efficient Removal of heavy metal ions from aqueous solution using salicylic type chelate adsorbents. Journal of Hazardous Material 192: 956–962.

Bai

Cao

Lan

(2008) Chitosan gel beads immobilised Cu (II) for selective Adsorption of amino acids. Journal of Biochemical and Biophysical Methods 70: 903–908.

Beltran

Borrull

Cormack

(2010) Molecularly-imprinted polymers: useful Sorbents for selective extractions. Trends in Analytical Chemistry 29(11): 1363–1375.

Biffins A, Dvorakova G and Falcimaigne-Cordin A (2012) Physical forms of MIP’s. In: Karsten Haupt (ed) Topics in Current Chemistry. Vol. 325. Berlin Heidelberg: Springer, pp.29–82.

Brugnerotto

Lizardi

Goycoolea

(2001) An infrared investigation in relation with chitin and chitosan characterization. Polymer 42: 3569–3580.

Dutta

Tripathi

(2004) Chitin and chitosan chemistry, properties and applications. Journal of Scientific and Industrial Research 63: 20–31.

Farrington

Magner

Regan

(2006) Predicting the performance of molecularly imprinted polymers: Selective extraction of caffeine by molecularly imprinted solid phase extraction. Analytical Chimica Acta 566: 60–68.

Guibal

(2004) Interactions of metal ions with chitosan based sorbents: A review. Separation Purification Technology 38: 43–74.

Hameed

Mahmoud

Ahmad

(2008) Equilibrium modeling and kinetic studies on the adsorption of basic dye by a low-cost adsorbent: Coconut (Cocos nucifera) bunch waste. Journal of Hazardous Material 158(1): 65–72.

10.

Huang

Kong

(2003) Molecular imprinting of nitrophenol and hydroxybenzoic acid isomers: Effect of molecular structure and acidity on imprinting. Journal of Molecular Recognition 16: 406–411.

11.

Jiang

Chen

Luo

(2013) Synthesis of novel nanocamposite Fe₃O₄ /ZrO₂/chitosan and its application for removal of nitrate and phosphate. Applied Surface Science 284: 942–949.

12.

Jin

Choi

Row

(2008) Adsorption isotherm of caffeine on molecular imprinted polymer. Korean Journal of Chemical Engineering 25(4): 816–818.

13.

Kyzas

Lazaridis

Bikiaris

(2013) Optimization of chitosan and β-cyclodextrin molecularly imprinted polymer synthesis for dye adsorption. Carbohydrate Polymers 91: 198–208.

14.

Kyzas

Siafaka

Pavlidou

(2008) Synthesis and adsorption application of succinyl-grafted chitosan for the simultaneous removal of zinc and cationic dye from binary hazardous mixtures. Chemical Engineering Journal 259: 438–448.

15.

Liu

Cao

Hua

(2010) Selective adsorption of uranyl ion on ion-imprinted chitosan / PVA cross-linked hydrogel. Hydrometallurgy 104(2): 150–155.

16.

Chen

Zheng

(2011) Preparation of molecularly imprinted CS membrane for recognizing naringin in aqueous media. Polymer Bulletin 66: 853–863.

17.

Liu

(2016) Ultrasensitive sensing of tris(2,3-dibromopropyl) isocyanurate based on the synergistic effect of amino and hydroxyl groups of a molecularly imprinted poly(o-aminophenol) film. New Journal of Chemistry 40: 1649–1654.

18.

Liu

Zhang

(2013) The cooperative utilization of imprinting, electro-spinning and pore-forming agent for β-cyclodextrin polymers to enhance the recognizing of naringin. RSC Advances 3: 25396–25402.

19.

Meng

Feng

Zhang

(2013) Optimization of surface imprinted layer attached poly (vinylidene fluoride) membrane for selective separation of salicylic acid from acetylsalicylic acid using central composite design. Chemical Engineering Journal 231: 132–145.

20.

Mourya

Inamdar

(2008) Chitosan-modifications and applications: Opportunities galore. Reactive and Functional Polymers 68: 1013–1051.

21.

Pardeshi

Dhodapkar

Kumar

(2012) Studies of the molecular recognition abilities of gallic acid-imprinted polymer prepared using a molecular imprinting technique. Adsorption Science & Technology 30(1): 23–34.

22.

Pardeshi

Dhodapkar

Kumar

(2014) Influence of porogens on the specific recognition of molecularly imprinted poly (acrylamide-co-ethylene glycol dimethacrylate). Composite Interface 21(1): 13–30.

23.

Pardeshi

Patrikar

Dhodapkar

(2012) Validation of computational approach to study monomer selectivity towards the template gallic acid for rational molecularly imprinted polymer design. Journal of Molecular Modeling 18(11): 4797–4810.

24.

Shi

You

(2014) Selective and simultaneous extraction and determination of hydroxybenzoic acids in aqueous solution by magnetic molecularly imprinted polymers. Chemical Engineering Journal 243: 485–493.

25.

Shinde

Pandey

Acharya

(2013) Chitosan-transition metal ions complexes for selective arsenic (V) preconcentration. Water Research 47: 3497–3506.

26.

Sreenivasan

(2007) Identification of salicylic acid using surface modified polyurethane film using an imprinted layer of polyaniline. Analytica Chimica Acta 583(2): 284–288.

27.

Umpleby

Baxter

Chen

(2001) Characterization of molecularly imprinted polymers with the Langmuir-Freundlich Isotherm. Analytical Chemistry 73: 4584–4591.

28.

Varma

Deshpande

Kemedy

(2004) Metal complexation by chitosan and its derivatives: A review. Carbohydrate Polymer 55: 77–93.

29.

Wang

(2014) Synthesis of chitosan molecularly imprinted polymers for solid-phase extraction of methandrostenolone. Carbohydrate Polymers 101: 517–23.

30.

You

Peng

Zhang

(2014) Preparation of magnetic dummy molecularly imprinted polymers for selective extraction and analysis of salicylic acid in Actinidia chinensis. Analytical and Bioanalytical Chemistry 406(3): 831–8399.

31.

Deng

(2008) Selective removal of perfluorooctane sulfonate from aqueous solution using chitosan-based molecularly imprinted polymer adsorbents. Water Research 42: 3089–3097.

32.

Zhang

Liu

Chen

(2001) Influence of intramolecular hydrogen bond of templates on molecular recognition of molecularly imprinted polymers. Analytica Chimica Acta 450: 53–61.

33.

Zhang

Liu

Wang

(2002) Molecular Recognition Properties of Salicylic Acid-Imprinted Polymer. Chromatographia 55: 447–451.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.00 MB