Figure 3 SEM (a-c) and TEM (e-g) images of MTCFs (a, e), BSA-MTCFs@SIP (b, f), MTCFs@SIP@CBMA (c, g); SEM images of MTCFs@NIP@CBMA (d); EDS analysis of MTCFs@SIP@CBMA: Full spectrum(h), C element (i), N element (g), O element (k), Fe(l)
Composition analysis
To confirm the composition of the polymer shell and prove the synthesis of MTCFs@SIP@CBMA, we systematically characterized the products at all steps. The FT-IR spectra of MTCFs, BSA-MTCFs@SIP, BSA-MTCFs@SIP@CBMA, MTCFs@SIP@CBMA, and MTCFs@NIP@CBMA are shown in Figure 4a. Compared with the infrared curves of MTCFs, the absorption peaks of BSA-MTCFs@SIP, BSA-MTCFs@SIP@CBMA, MTCFs@SIP@CBMA, and MTCFs@NIP@CBMA at 1460 cm-1 were attributed to the aromatic ring on dopamine, which proved the successful coating of imprinted polymer. BSA-MTCFs@SIP@CBMA, MTCFs@SIP@CBMA, and MTCFs@NIP@CBMA emerged absorption peaks at 1734 cm-1 attributable to -CO-O on CBMA, indicating that CBMA anti-protein adsorption segments were successfully grafted on the surface of BSA-MTCFs@SIP. Furthermore, the infrared spectra of BSA-MTCFs@SIP and BSA-MTCFs@SIP@CBMA both exhibited absorption peaks at 2568 cm-1. This band can be assigned to -SH on BSA, confirming the successful imprinting of BSA on the surface of fibers. After elution, the absorption peak of -SH in the infrared spectrum of MTCFs@SIP@CBMA disappeared. It was basically consistent with the absorption peak of MTCFs@NIP@CBMA, indicating that BSA was completely eluted.
The composition of organic and inorganic components in the nanofibers and the thermal stability of the nanofibers were analyzed by thermogravimetric analysis. The measured TGA curve is presented in Figure 4b. It was clear to see that all the samples had no obvious weight loss before 200℃, indicating superior thermal stability of nanofibers. As the temperature increased, MTCFs, BSA-MTCFs@SIP, and BSA-MTCFs@SIP@CBMA lost weight at different rates. MTCFs began to exhibit significant weight loss at 328 ℃ and was completely degraded at 369 ℃. The weight loss rate was 86.18%, which was caused by the combustion of carbon under oxygen. The degradation of BSA-MTCFs@SIP began at 268℃ and was complete at 387 ℃, with a weight loss rate of 89.11%. The reason why BSA-MTCFs@SIP demonstrated lower pyrolysis temperature and higher weight loss rate than MTCFs was that the coating of PDA on its surface. Compared with BSA-MTCFs@SIP, BSA-MTCFs@SIP@CBMA began to degrade at 237 ℃ and completely degraded at 390 ℃. The weight loss rate was 91.39%. The reincrease in weight loss rate confirmed the successful grafting of CBMA polymer segments on the surface of BSA-MTCFs@SIP.
The specific composition of the magnetic nanoparticles loaded in the tube of carbon nanofiber was analyzed by XRD. As shown in Figure 4c, three diffraction peaks appeared in MTCFs at diffraction angles 2θ of 44.67°, 65.02° and 82.33°, corresponding to (110), (200) and (211) crystal planes, respectively. According to the PDF card, the diffraction peak data was consistent with that of Fe (PDF 87-0721) with structure of body-centered cubic. Therefore, it could be inferred that the magnetic nanoparticles present in the carbon nanofiber tube were Fe nanoparticles. Additionally, the XRD diffraction peak positions of BSA-MTCFs@SIP and MTCFs@SIP@CBMA are basically consistent with that of MTFs. This meant that the coating process of PDA and the subsequent grafting of the CBMA anti-protein adsorption segments had no effect on the crystal form of Fe.
The magnetic properties of MTCFs, BSA-MTCFs@SIP, and MTCFs@SIP@CBMA were measured by VSM. The results are shown in Figure 4d. It can be seen that all the samples possessed magnetic properties. The specific saturation magnetization of MTFs was 8.29 emu/g, while the specific saturation magnetization of BSA-MTCFs@SIP and MTCFs@SIP@CBMA were 7.25 emu/g and 6.21 emu/g, respectively. The decrease of magnetic susceptibility was due to the fact that the magnetic responsiveness of the materials was weakened to some extent by both the coating of PDA and the grafting of CBMA polymer chains. Fortunately, this magnetic performance was sufficient to achieve rapid separation of the samples from the solution. Thus, the separation efficiency was improved.
Figure 4 FT-IR spectra of products at each stage (a); TGA curves of MTCFs, BSA-MTCFs@SIP and BSA-MTCFs@SIP@CBMA (b); XRD (c) and VSM (d) curves of MTCFs, BSA-MTCFs@SIP and MTCFs@SIP@CBMA
Characterization of specific surface area and pore performance
High specific surface area could increase the number of effective imprinted sites, thus improving the adsorption capacity. Therefore, nitrogen adsorption apparatus was applied to measure the specific surface area and pore properties of the materials. The nitrogen adsorption-desorption curves and pore size distribution curves of the samples are given in Figure 5. According to the nomenclature of the International Union of Pure and Applied Chemistry, the adsorption and desorption curves of MTCFs, BSA-MTCFs@SIP, BSA-MTCFs@SIP@CBMA, and MTCFs@SIP@CBMA all belonged to type IV. The hysteresis loop type was H4, indicating the slit-like pores in materials. It was worth noting that the specific surface area of the fibers decreased gradually as the procedure of PDA coating and CBMA polymer grafting progressed (see Table 1). This was because part of the pores of the fiber were blocked by the coating of the polymers. After elution, the specific surface area of MTCFs@SIP@CBMA increased compared with BSA-MTCFs@SIP@CBMA, confirming the formation of imprinted sites on the surface of fibers. From the pore size distribution curve, it was found that the pore size of fibers was mainly divided into two sections, 4-40 nm and 40-100 nm. The pores in the range of 4-40 nm were attributed to the small pores on the tube wall of fibers and the imprinted cavities left by the eluted templates. This part of the pores allowed BSA molecules to diffuse into the inner cavity of the tube. The range of 40-100 nm corresponded to the diameter of the cavity in the tube. The large-scale cavity caused an osmotic pressure difference between the solution inside and outside the tube, which in turn gave MTCFs@SIP@CBMA self-driven adsorption performance. The nitrogen adsorption-desorption curve type and pore size distribution of MTCFs@NIP@CBMA were consistent with those of MTCFs@SIP@CBMA (see Figure S1). Nevertheless, due to the absence of imprinting sites on its surface, the specific surface area (170.53 m2/g) and pore volume (0.02 cm3/g) of MTCFs@NIP@CBMA were lower than those of MTCFs@SIP@CBMA.
Figure 5 N2 adsorption-desorption curves and pore size distribution curves of MTCFs (a, b), BSA-MTCFs@SIP (c, d), BSA-MTCFs@SIP@CBMA (e, f) and MTCFs@SIP@CBMA (g, h)
Table 1 Specific surface area and pore performance data of samples