Synthesis, Characterization and Antibacterial Activity of Novel β-cyclodextrin Polyurethane Materials

The advanced water treatment taken by organic micro-pollution or microbiological pollution water resource has been a hot issue of public concern. In this article, novel quaternary ammonium salt functionalized β-cyclodextrin polyurethane (QAS-β-CDPU) microparticles were successfully constructed for above-mentioned pollutants removal. By adjusting the proportion of the chlorine-containing monomers, we studied the effect of different quaternization degree (0%, 20%, 40%, 60%) on solvent resistance and thermal stability. Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) are recognized as significant human bacterial pathogens, which were used as model bacteria to investigate the antibacterial activity by contact-killing tests. Methylene blue (MB) served as a toxic organic model to quantify the dye wastewater remediation efficiency. Research shows that the reaction of β-cyclodextrin and diisocyanate can form a cyclodextrin-based carbamate network structure. Since both the cyclodextrin cavity and the carbamate network may remove the dye, the obtained polymer has a double advantage. Besides, the presence of quaternary ammonium groups enables QAS-β-CDPU microparticles to possess good bactericidal properties. When the quaternization degree is 60%, sterilization rates of QAS-β-CDPU microparticles are over 96% for S. aureus and E. coli. On basis of this simple, green and economical synthetic route, QAS-β-CDPU microparticles have the potential to become an ideal multifunctional material in water purification. Novel quaternary ammonium salt functionalized β-cyclodextrin polyurethane (QAS-β-CDPU) microparticles were successfully constructed for removing bacteria and dyes from wastewater.


Introduction
Water is the natural breeding ground for most pathogens. The levels of bacteria are treated as key indicators for water pollution. In countries like India, 80% of the diseases are caused by bacteria contamination in drinking water [1]. Water pollution with pathogenic bacteria pose a severe global threat to public health [2,3]. According to the 2017 World Health Organization (WHO) report, more than 25% of the global population lack access to safe water. Unclean water is more likely to contain a large variety of pathogens, such as Escherichia coli, Salmonella enteritidis, Vibrio cholerae and Shigella, which greatly increases the risk of waterborne diseases in humans [4]. Although the traditional disinfection methods are highly effective for microbial control in water treatment, for example, free chlorine, chloramine and ozone disinfection, it is easy to generate harmful disinfection by-products (DBPs), even numerous carcinogens [5]. In addition, the resistance of some pathogens, especially Cryptosporidium and Giardia, to conventional chemical disinfectants requires ever higher disinfectant dosage, which in turn exacerbates DBPs formation [6]. Therefore, the quest for novel water purifying materials without DBPs generation is of great significance.
Quaternary ammonium salt is deemed as a new type of cationic bactericide due to excellent cell membrane penetration, low toxicity, good environmental stability and superior biological activity [7]. Furthermore, it is inert to general redox agent and acids or bases, avoiding the formation of carcinogenic DBPs [8]. Unfortunately, the cationic structure endows quaternary ammonium salt with high water solubility and uncontrollable diffusion, leading to short-lived dwell on the target surface and increased dose [9,10]. Contact active surfaces have drawn extensive attention for their non-release and long-term activity. In these surfaces, the 1 3 antibacterial agents are covalently attached to polymer backbones to avoid releasing. Though this strategy it can not only achieves a long-term activity in an environmentally friendly way, but also avoid the drawback of rapid biocidal diffusion, obtaining satisfactory results in improving antibacterial efficiency and reducing residual toxicity [4]. Hence, modification to gain a long-acting slow-release antibacterial polymer is vital for widespread application of quaternary ammonium salt.
Recently, natural polymers like polysaccharides and their derivatives are considered as preferred low-cost adsorbents for wastewater treatment, attributing to their unique structure, physical and chemical properties, chemical stability, high reactivity, and good selectivity towards organics and metals [11,12]. Crini [13] reported a series of cheap polysaccharide-based dye adsorbents, one of the important polysaccharide derivatives is β-cyclodextrin (β-CD) [14]. β-CD is a kind of cyclic oligosaccharide produced from bacterial enzymatic hydrolysis of starch, featuring a hydrophilic exterior and a hydrophobic inner cavity [15]. The special structure can enable them to selectively combine various organic, inorganic and biological guest molecules into their cavities, and form stable host-guest inclusion complexes in their hydrophobic cavities [16]. Thus, β-CD may be favorable for removing pollutants in wastewater. However, the hydrophilic exterior makes β-CD readily dispersible in aqueous media, so it is difficult to adapt traditional separation methods to separate itself from the solution after the adsorption process, which may increase the costs considerably for industrial applications and/or cause secondary pollution [17]. In recent years, the emergence of cyclodextrin-based carbamate network structure from the reaction of cyclodextrin with diisocyanate link [hexamethylene diisocyanate (HMDI), toluene diisocyanate (TDI) or diphenylmethane diisocyanate (MDI)] provides an effective pathway to solve the above problems [18]. In fact, these bifunctional linkers help decrease the water solubility of cyclodextrin [18]. Moreover, since both the cyclodextrin cavity and the carbamate network can remove the dye, the obtained polymer has a synergistic advantage [19].
Polyurethane (PU) is a most popular polymer material with excellent mechanical properties and good water resistance [19]. PU foam material has been widely applied in various fields as a carrier on account of its network structure, small bulk density, large surface area, and high adsorption efficiency [19]. Though there were lots of reports on quaternary ammonium salt-polyurethane-based antibacterial polymers [20] and cyclodextrin-polyurethane-based adsorption materials [21], quaternized cyclodextrin polyurethane has not yet been heard so far.
Herein, we take β-CD and diisocyanate as raw material to prepare cyclodextrin polyurethane matrix via internal emulsification. With help of quaternization reaction, it further synthesizes novel QAS-β-CDPU microparticles for removing bacteria and dyes from wastewater. S. aureus and E. coli, microorganisms related to human health, act as model bacteria to evaluate antibacterial property of the polymer surface through contact-killing tests. MB, a toxic and carcinogenic cationic dye, behaves as a model dye to assess dye removal efficiency.

Preparation of QAS-β-CDPU Microparticles
In brief, 4.0 g of β-CD was dissolved in 40 mL of DMF, followed by adding 6.66 g of IPDI, 0.6 g of hydrophilic chainextender DMPA and 0.52 g of HEMA, then stirring constantly for 5 min. Afterward, 0.08 g of catalyst DY-12 was slowly added to the dispersion and kept at 80 °C for 3.5 h. After the reaction was completed, the resulting mixture was rapidly cooled down to 50 °C. The pH of the solution was adjusted to neutral using 0.1 g of TEA and incubated for 5 min. 40 g of deionized water was poured into the cooled prepolymer under mechanical stirring to obtain vinyl-containing cyclodextrin-based polyurethane. 35 g of the above yielded precipitate was treated with deionized water in a ratio of 1:1.2. Subsequently, 0.97 g of the chlorine-containing vinyl monomer ClHPMA dispersed in 20 g of chloroform was injected by grafting method at one time, bubbling nitrogen through the mixture for 15-20 min, and then stirring at room temperature for 2 h. After confirming that the reaction system was uniformly agitated, the initiator KPS was added dropwise at 75 °C and reacted for 24 h, the obtained product was denoted as Cl-β-CDPU. Finally, 0.75 g of TEA was added and stirred at ambient temperature for 24 h. The resultant sediment was rinsed repeatedly with ethanol and vacuum dried at 50 °C for 24 h to gain pale yellow fine powder, which was recorded as QAS-β-CDPU-20% microparticles. As control experiments, by adjusting the mass ratio of the chlorine-containing monomers, the products with the quaternization degree of 0%, 40%, and 60% were obtained, which were named as QAS-β-CDPU-0% microparticles (namely, β-CDPU), QAS -β-CDPU-40%

Preparation of Dye Solution
The dye stock solution (100 mg L −1 ) was made with deionized water and methylene blue, then can be diluted with water to any desired concentration. A standard curve was generated by plotting the absorbance of various dye samples with known concentration. Thermogravimetric analysis (TGA): The thermal stability of the polymer material was studied by using TGA Q500 (TA instrument, USA). The sample was heated from 25 °C to 700 °C at a heating rate of 10 °C/min and flow rate of 30 mL min −1 in N 2 atmosphere.

Characterization
The Brunauer-Emmett-Teller (BET): The BET surface area and porosity of the samples were determined on accelerated surface area and porosimetry system (ASAP 2020, Global Spec.Inc., US). BET surface areas of the solids were measured by nitrogen adsorption at − 196 °C using Surface Area Analyser 2010, ASAP apparatus. The samples were outgassed for 6 h at 120 °C.
Ultraviolet visible spectrum (UV-Vis): The MB concentration in liquid supernatant was monitored by an ultraviolet-visible spectrophotometer (Perkin Elmer, LAMBDA 1050) at 664 nm.
Gel Rate (G): The G of samples were investigated to confirm the occurrence of cross-linking reaction. Firstly, the dried sample was accurately weighed and represented as W i . Afterwards, the sample was immersed in diethylene oxide for 48 h at room temperature with stirring. Finally, the sample was separated and dried at 60 °C for 2 h before reweighing (W f ). G was calculated by the following formula: Water Absorption: The water absorption of powder samples were measured by gravimetric method. To be specific, after vacuum drying to constant weight at 60 °C, powder sample (W i ) was put in a weighing tea bag (W j ). Then it was soaked in distilled water for 24 h at room temperature. Taking out the tea bag and hanging until no water drips (about 10 min), it was weighed on a balance again (W). The water absorption rate (M) was calculated according to Eq. (2). The empty tea bag was used as the blank control group to repeat the above experiment, and the gain in weight of empty tea bag was subtracted from the experiment result. To guarantee the reliability of data, the measurement result is averaged from three repetitive tests.

Antibacterial Activity Assay
The gram-positive bacterium S. aureus and gram-negative bacterium E. coli were chosen as the models to estimate antibacterial activity through contact-killing experiments. Briefly, the glass slide was coated with a thin and uniform layer of silica gel, followed by evenly coating the antibacterial sample powder on the silica gel. After drying, 18 μM S. aureus or E. coli liquid slowly droped into it while covered with another clean glass slide. Afterwards, we placed it on the biological sterile incubate for 30 min at 37 °C in an incubator. Whereafter, the two slides were separated with tweezers, put in a centrifuge tube and cautiously added water (10 mL). After manual shaking for a while, the centrifuge tube was agitated ultrasonically for 60 s. This process was repeated twice, whereupon the slide was taken out. 100 μM of the supernatant was collected to seed and incubated for overnight. Approximately 24 h later, the number of colonies was observed to study its antibacterial activity. For comparison, the blank group was treated in the same way as steps.

Adsorption Experiments
Adsorption isotherm experiments were carried out by adding 10 mg sample to a series of 50 mL centrifuge tubes with MB solutions (10 mL, 5-40 mg L −1 ) at pH 7. Then, the centrifuge tubes were sealed and separately shaken at different temperatures (30, 40, 50 °C) for 24 h. The supernatant liquid was separated from sample by centrifugation. Finally, the residual MB concentration in the supernatant liquid was recorded on an ultraviolet-visible spectrophotometer at 664 nm. Adsorption kinetics tests were performed by adding 10 mg sample to 50 mL centrifuge tubes with MB solutions (10 mL, 10 mg L −1 ) at pH 7. Subsequently, the centrifuge tubes were sealed and stirred at 30 °C. The supernatant liquid was taken at time intervals specified in advance and measured the MB concentration similarly to quantify the adsorption rate.
Pseudo-first-order and pseudo-second-order models were applied to evaluate the adsorption kinetics. The linear equations of these models are expressed as follows: In Eq. (3), q e1 and q t were separately the adsorption capacity (mg g −1 ) at equilibrium and time t; k 1 was the rate constant of the pseudo-first-order model (min −1 ). In Eq. (4), q e2 and q t were the adsorption capacity (mg g −1 ) at equilibrium and time t, respectively; k 2 was the rate constant of the pseudo-second-order model (g mg −1 min −1 ).
The Langmuir and Freundlich isotherm equation are separately defined below: where Ce was the equilibrium concentration of MB in the solution (mg L −1 ), q e was the amount of MB adsorbed in equilibrium (mg g −1 ), q max was the theoretical maximum adsorption capacity (mg g −1 ), K L and K F was Langmuir equilibrium constant and Freundlich constant, respectively. In the Freundlich equation, n is an index of adsorption intensity.
Separation factor, a dimensionless equilibrium parameter, is proposed to describe the essential characteristics of the Langmuir isotherm, specifically as follows: In the formula, K L is the Langmuir constant (L mg −1 ), C 0 is the initial concentration of the dye (mg L −1 ). This parameter indicates that the shape of the isotherm is unfavorable (R L > 1), favorable (0 < R L < 1), linear (R L = 1) or irreversible (R L = 0) [22].
The following Eqs. (8,9) are used to deduce the changes in thermodynamic parameters, such as the standard Gibbs free energy change (ΔG 0 ), standard enthalpy change (ΔH 0 ) and standard entropy change (ΔS 0 ), etc. [23,24]: where K L (L mg −1 ) is the Langmuir isotherm constant, R (8.314 J·mol −1 ·K −1 ) is the universal gas constant, and T (K) is the absolute temperature in Kelvin.

Desorption and Reusability Experiments
In the desorption test, ethanol was selected to desorb the dye in QAS-β-CDPU-20% microparticles for investigating the reusability. In general, the adsorbent (10 mg) was added to 10 mL of dye solution (40 mg L −1 ). After stirring at room temperature for 24 h, the reaction mixture was subject to centrifugal separation to obtain precipitate and discard the supernatant. 10 mL of ethanol was added to the precipitate and continuously stirred for 24 h. Similarly, the suspension was centrifuged and collected the sediment for next adsorbent recirculation. The dye concentration in the supernatant was measured by an ultraviolet-visible spectrophotometer, and each experiment above was conducted five consecutive adsorption-desorption cycles.

FTIR Analysis
To identify possible product structures, FTIR spectra of β-CD, β-CDPU, Cl-β-CDPU and QAS-β-CDPU-20% microparticles are shown in Fig. 2. The spectrum of β-CD exhibited two peaks at 3400 cm −1 and 2922 cm −1 , which are fitted with the stretching vibration of -OH and -CH 2 -. In addition, two peaks located at 1031 cm −1 and 952 cm −1 are assigned to the antisymmetric glycosidic ν a (C-O-C) vibrations and α-1,4-glycosidic bond stretching vibration [19,25], respectively. Compared with β-CD, the infrared spectrum of β-CDPU has apparent changes. The peak at 1155 cm −1 and 1440 cm −1 are severally corresponded to the stretching vibration of C-C/C-O and methylene. The peak at 1660 cm −1 and 1550 cm −1 are attributed to stretching vibration of C=O and amide group [26]. The strong absorption peak around 1718 cm −1 is dominated by the stretching vibration of carbonyl of carbamate [27]. The disappearance of characteristic isocyanate peak at 2280 cm −1 , indicating the completion of polymerization reaction [28]. Furthermore, it is clear to see stretching peaks of C-Cl at 857 cm −1 in the spectrum of Cl-β-CDPU [29,30], which is the evidence of the polymerization. It is observed that the spectrum of QASβ-CDPU-20% microparticles displayed five characteristic bands around 842 cm −1 , 1148 cm −1 , 1460 cm −1 , 1718 cm −1 and 2988 cm −1 . The first one is in accordance with characteristic peak of C-Cl; the second one and the fourth one are attributable to the stretching vibration of C-C/C-O and the stretching vibration of carbonyl of carbamate; meanwhile, the rest of bands are assigned to characteristic peaks of quaternary ammonium salt [30]. The appearance of above peaks demonstrates the successful synthesis of QAS-β-CDPU-20% microparticles.

EDS and Mapping Analysis
EDS and mapping analysis (Fig. 3) reflected that C, O and Cl elements were uniformly distributed in the polymer material. Normally, there was supposed to be N element, rather than Pt element. This is mainly because synthetic polymer material is not conductive. To facilitate realistic testing, the surface of sample was sprayed with Pt to increase its conductivity. Unfortunately, since the content of N element is very low, the vast existence of Pt further influences the observation of N element. It is worth noting that the presence of Cl element indicates successful quaternarization for β-CDPU [31]. Figure 4 depicts the solubility of QAS-β-CDPU-20% microparticles in solvents. For pure β-CD (Fig. 4a), it is easily dissolved in water. By comparison, QAS-β-CDPU-20% microparticles are almost insoluble in water, HCl and NaOH (Fig. 4b-d), showing excellent solvent resistance. Additionally, the gel rate of QAS-β-CDPU microparticles with different quaternization degrees is shown in Fig. 5a. As the degree of quaternization rises, the gel rate of QAS-β-CDPU microparticles is a corresponding increase, which also has better solvent resistance. In fact, when the quaternization degree  is higher, the amount of chlorine-containing monomers in the synthesis process is higher, causing rapid polymerization of β-CDPU and further cross-linking of materials. As a result, the gel content of QAS-β-CDPU microparticles and the solvent resistance behavior reach maximum level. With regard to water absorption rate of QAS-β-CDPU microparticles, it gradually decreases with the increase of quaternization degree (Fig. 5b). The distance between the crosslinked points reduces along with the increase in cross-linking density. Thereby, the pore between the crosslinked points lose the ability to expand. Low volume porosity certainly results in low water absorption [32]. Besides, the swelling ability also relies on the presence of hydrophobic functional groups in its three-dimensional network. The formation of long chain alkyl groups on the backbone by means of quaternization utilizing TEA can improve hydrophobicity of QASβ-CDPU microparticles. The hydrophobicity increases with increasing concentration of TEA [33].

SEM Analysis
To gain an insight into the structure properties of β-CD, β-CDPU and QAS-β-CDPU-20% microparticles, SEM analysis was conducted in Fig. 6. β-CD exhibits a thick flake structure with a clear edge, which is roughly rectangular in shape. On the contrary, QAS-β-CDPU-20% microparticles have almost no regular crystal structure [34]. An even more fluffy appearance and rough surface for QAS-β-CDPU-20% microparticles were observed in comparison to β-CDPU. As is apparent, the quaternary ammonium salt modification can bring about further structural changes.

XRD Analysis
XRD analysis is able to characterize the phase purity and the structural effects related to the long-range order in polymer materials [35]. Figure 7 describes XRD patterns of β-CD, β-CDPU and QAS-β-CDPU-20% microparticles. For β-CDPU and QAS-β-CDPU-20% microparticles, there are a broad band around 15° ~ 20°, suggesting that they are amorphous in nature [19]. In contrast, the XRD pattern of pure β-CD presents many narrow peaks in range of 5-55°, confirming that it has good crystalline character. These results match well with the SEM images in Fig. 6. Since β-CD entered into PU backbone portion, its original crystal structure and internal hydrogen bonds were damaged [36]. Namely, the existence of PU backbone impeded the noncovalent interaction and filling efficiency of β-CD [37]. Therefore, the main specific diffraction peaks of β-CD disappeared in XRD pattern of β-CDPU.
XRD curves of β-CDPU and QAS-β-CDPU-20% microparticles have a very similar shape, the difference is that the peak intensity of QAS-β-CDPU-20% microparticles is weaker than β-CDPU. To some degree, this reflects a decrease in crystallinity due to the introduction of quaternary ammonium groups in QAS-β-CDPU-20% microparticles. Specifically, electrostatic repulsion by the positively charged trimethylammonium group in QAS-β-CDPU-20% microparticles may hinder the formation of inter-molecular and extra-molecular hydrogen bonds in β-CDPU backbone [38,39]. Figure 8 shows TGA curves of QAS-β-CDPU microparticles with different quaternization degrees under a nitrogen atmosphere. All samples underwent about 5.5-8% weight loss at temperatures lower than 200 °C, which was caused by evaporation of water adsorbed around the polymer skeleton (for instance, free water, physically adsorbed water, and bound water). As temperature rose, β-CDPU lost significant weight in the range of 260-320 °C due to the destruction of β-CD [21]. When the temperature reached up to 300 °C, the polyurethane moiety would decompose. It can be seen from Fig. 8 that β-CDPU was rapidly decomposed at 260 °C. Samples after quaternization appeared a notable weight loss at a higher temperature of 300-310 °C. It reflects that the thermal stability of QAS-β-CDPU microparticles is better than β-CDPU [40].

BET Analysis
Surface area is an important index to reflect adsorption capability of the adsorbent. Based on the International Union of Pure and Applied Chemistry (IUPAC) classification, N 2 adsorption-desorption isotherms of β-CDPU, QAS-β-CDPU-20%, QAS-β-CDPU-40% and QAS-β-CDPU-60% microparticles are analogous to the type II with H 3 hysteresis loop (Fig. 9). The specific area BET results in this study was measured using BET-type N 2 adsorption. By calculation and analysis (Table 1), the BET surface area and a total pore volume of QAS-β-CDPU-20% (20.57 m 2 g −1 , 4.727 cm 3 g −1 ) are larger than β-CDPU (15.56 m 2 g −1 , 3.575 cm 3 g −1 ), and both are superior to raw material β-CD (0.63 m 2 g −1 , 0.014 cm 3 g −1 ), QAS-β-CDPU-40% (0.258 m 2 g −1 , 0.00022 cm 3 g −1 ) and QAS-β-CDPU-60% (0.187 m 2 g −1 , 0.00027 cm 3 g −1 ). Owing to the amorphous structure, β-CDPU and QAS-β-CDPU-20% microparticles are more likely to form a microporous structure than the crystal structure of β-CD, resulting in a higher specific surface area [41]. In addition, the specific surface area was appearing to increase slightly after quaternary amine salt modification. Lima et al. also observed a slight increase in the BET surface area of quaternized coconut shell fibers [42]. Thamilarasi et al. found that the BET surface area of quaternized palm peel was lower than its raw material [43]. From our research,  it is clearly that an appropriate amount of quaternary ammonium salt modification will improve the BET surface area of the material. However, unfavorable effect will occur to adsorption performance at higher quaternization degree. The discrepancy between the results is probably attributable to the different properties and composition of the materials concerned [44]. Of course, the specific surface area of β-CD or QAS-β-CDPU-20% microparticles is nothing compared to commercial activated carbon (CAC, 1100 m 2 g −1 ) and commercial synthetic zeolite (450 m 2 g −1 ) [45,46]. Although the surface area of β-CD-based polymers varied with different raw materials and preparation methods, a similar small surface area (0.11-8.01 m 2 g −1 ) was seen in other studies [47]. Considering that the adsorption mechanism of β-CD to organic molecules is quite different from porous materials, surface area is one of critical factors for QAS-β-CDPU microparticles to adsorb MB [43].

Antibacterial Activity
As is well-known, bacterial or fungal infections may lead to severe complications. Inadequate treatment for infected wounds can cause cellulitis (cell inflammation) and even fatal sepsis. Wound infection is likely the most frequent reason for substantial morbidity and mortality in extensive burns, trauma, and surgical operations. In our research, S. aureus and E. coli, the most representative gram-positive bacteria and gram-negative bacteria, were selected as model bacteria to study antibacterial properties of QAS-β-CDPU microparticles [48]. As summarized in Figs. 10 and 11, all strains involving S. aureus and E. coli are well killed in the range of 63-98% when attaching the quaternary ammonium part to the cyclodextrin-polyurethane network. This definitely implies that QAS-β-CDPU microparticles have certain antibacterial ability. It can be noted that the number of S. aureus and E. coli decreased significantly with the increase of the quaternization concentration. QAS-β-CDPU with the highest quaternization concentration displayed almost 100% bacterial reduction against S. aureus and E. coli. Experiments show that the excellent bactericidal performance of QAS-β-CDPU microparticles was associated with quaternary ammonium salt groups on the surface. Refer to the relevant literature, the antibacterial activity of quaternary ammonium compounds benefits from its hydrophobicity [49,50]. The hydrophobic alkyl chain of the quaternary ammonium salt penetrates the bacterial membrane, causing the destruction of the membrane and the death of the bacteria [51]. The outstanding antibacterial properties of QAS-β-CDPU microparticles are conducive to the actual water treatment process, which can greatly reduce the use of disinfectants in the disinfection process, thereby saving costs and inhibiting by-products formation.

Adsorption Kinetics
In adsorption research, adsorption kinetics are of great importance because they can describe the adsorption rate   and provide valuable information for understanding the adsorption reaction mechanism [52]. Figure 12 depicts the adsorption kinetics of methylene blue by QAS-β-CDPU microparticles. Pseudo-first-order and pseudo-second-order models were applied to evaluate the adsorption kinetics. Table 2 exhibits the adsorption kinetic parameters of QAS-β-CDPU microparticles with different quaternization degrees. Compared to the pseudo-first-order kinetic model, the pseudo-second-order kinetic model is more consistent with the experimental data. More precisely, the correlation coefficient (R 2 ) obtained from the pseudo-second-order kinetic model is greater than 0.99, and the degree of agreement between the calculated and experimental q e value is higher. Figure 12c represents a linear graph of t/q t versus t for MB adsorption on the prepared sample, confirming that chemical adsorption controls the adsorption rate. During the chemical adsorption process, the adsorbate was adsorbed on QAS-β-CDPU microparticles in the form of chemical bonds and further searched for adsorption sites, maximizing their coordination number with the surface [53]. Moreover, QASβ-CDPU microparticles with 20% quaternization degree has a larger adsorption capacity for MB than β-CDPU. This may be attributed to the optimal specific surface area and more active sites for adsorption. Notably, the adsorption effect was gradually weakened accompanied by a further increase in the proportion of quaternary amine salts. As a matter of fact, the quaternary ammonium salt is a cationic compound and MB is a cationic dye. With the continuous increase of quaternary ammonium salt, the positive charge of the adsorbent continues to increase, generating a strong repulsive force for the same charges and then reducing adsorption effect.

Adsorption Isotherms
The adsorption isotherm can help us understand the distribution of adsorbed molecules between the liquid and solid phases and supply the most critical parameters for constructing an ideal adsorption model. It has been successfully used to describe the adsorption process [54]. On basis of Langmuir and Freundlich isotherm model, Fig. 13 reveals adsorption isotherm on QAS-β-CDPU-20% microparticles at 303.15 K, 313.15 K and 323.15 K. The Langmuir model suggests that there is no interaction between adsorbed molecules, but just confined to a single layer on adsorbent surface. For Freundlich isotherm equation, it is a non-uniform surface adsorption based empirical model. The model assumes that the strong binding sites are preferentially occupied, and the binding strength decreases as the degree of site occupation increases [55]. These models are widely adopted for investigating the relationship between adsorption capacity of adsorbent and equilibrium adsorbent concentration in an aqueous solution.
The calculated fitting parameters are shown in Table 3. Overall, the Langmuir isotherm model provides a better fit to  the experimental data, implying that the adsorption of MB is monolayer or homogeneous adsorption [53]. The adsorption process is influenced by properties of dyes and adsorbents.
The calculated values of R L (Table 3) are all in the range of 0-1, reflecting the favorable adsorption of MB by QASβ-CDPU-20% microparticles. The introduction of β-CD into QAS-β-CDPU-20% microparticles enable the cavity structure to capture organic molecules via the inclusion of host and guest. Additionally, the cyclodextrin-based carbamate network structure formed by the reaction of β-CD and diisocyanate also contributes to adsorb dyes, so it has a dual advantage. The adsorption capacity increases with the increase of temperature, which illustrates high temperature is favorable for adsorption test.

Adsorption Thermodynamics
The thermodynamics theory supposes that entropy change is a driving force in an isolated system where energy can be neither gained nor lost [56,57]. For further exploring the adsorption process, the Eqs. (8,9) are used to deduce the changes in thermodynamic parameters, such as the standard Gibbs free energy change (ΔG 0 ), standard enthalpy change (ΔH 0 ) and standard entropy change (ΔS 0 ), etc. [23,24]: Table 4 lists the ΔG 0 , ΔH 0 and ΔS 0 when MB was adsorbed on QAS-β-CDPU-20% microparticles. The positive value of ΔH 0 indicates that adsorption process is essentially endothermic [58]. This finding is in agreement with the result that the adsorption capacity of QAS-β-CDPU-20% microparticles for MB increases with increasing temperature. The negative ΔG 0 value and the positive ΔS 0 value demonstrate that the adsorption process is spontaneous.

Desorption and Reusability
The recycling capacity of the adsorbent is the key element for its practical application. In the recycling experiment, QAS-β-CDPU-20% microparticles were washed with ethanol for 12 h under a shaker, and the effect of five consecutive adsorption-desorption cycles was studied (Fig. 14).
Since MB might not completely desorb at elution step, the adsorption capacity decreased slowly in each subsequent cycle. Nevertheless, it can still achieve acceptable performance after five cycles. As it turned out, QAS-β-CDPU-20% microparticles can be recycled and reused for MB treatment.

Conclusion
In this paper, we successfully synthesized a series of QASβ-CDPU microparticles for bacteria and dyes removal. S. aureus and E. coli were regarded as model bacteria to assess antibacterial activity of polymer surface via contact sterilization experiments. QAS-β-CDPU microparticles with 60% quaternization degree has best antibacterial properties, above 96% sterilization ratio for S. aureus and E. coli. Meanwhile, QAS-β-CDPU microparticles show superb adsorption performance for toxic carcinogenic MB as well. The related adsorption process was discussed in-depth, which is accord with pseudo-second-order kinetic model and Langmuir isotherm model. Furthermore, QAS-β-CDPU microparticles are insoluble in acid and alkali solvents, displaying excellent solvent resistance and stability. In view of abovementioned advantages, QAS-β-CDPU microparticles can be used as a multifunctional adsorbent for removing pathogens and chemical contaminates in wastewater disposal. It will be helpful for developing technologies related to field application and recycling. Author Contributions YC and JY designed and performed the experiments, drafted and wrote the manuscript. YZ and WC performed the data interpretation. ZW and LW supervised the project, performed the data interpretation and revised the manuscript. All authors reviewed the manuscript. We all agree to accountable for all aspects of the works.