A novel polyamidine-grafted carboxymethylcellulose: Synthesis, characterization and flocculation performance test

2.1 Materials and instruments

The following reagents were used in this experiment: carboxymethyl cellulose (CMC;Henan Weidao Chemical Products Co., Ltd.), ceric ammonium nitrate (CAN; Sinopharm Chemical Reagent Co., Ltd.), N-vinyl formamide (NVF; Sigma-Aldrich Co., Ltd.), acrylonitrile (AN; Sinopharm Chemical Reagent Co., Ltd.), 2,2′-Azobis (2-methylpropionamidine) dihydrochloride (AIBA, Aladdin Reagents Co., Ltd.), hydrochloric acid (Sichuan Xilong Chemical Co., Ltd.), anhydrous ethanol (Tianjin Fuyu Fine Chemical Co., Ltd.), polyacrylamide (PAM; Sinopharm Chemical Reagent Co., Ltd.), sodium hydroxide (Tianjin Hongyan Reagent Factory). The instruments used for the sample analysis were listed as follows: EA (Vario Macro Cube, Germany Element, Inc.), TGA-DTG (Discovery, U.S. TA Instruments), FT-IR spectrometer (Vertex 7.0, Germany Brukern Optics), SEM (EVO MA10, Germany Zeiss, Inc.), NMR (liquid-phase AV 400 MHz and solid-phase AV 600 MHz, Germany Bruker Optics, Inc.). The ¹H-NMR spectrum was recorded with dimethyl sulfoxide (DMSO) as the solvent, and ¹H MAS and ¹³C CP/MAS NMR were experimentally performed using a 4 mm MAS probe and a spinning rate of 12 kHz at a resonance frequency of 600.1 MHz and 150.9 MHz, respectively. TGA-DTG was constructed within 50-600°C at a heating rate of 10°C/min under a certain constant N₂ gas flow, and both mass of the substance and the thermal decomposition were determined as a function of temperature.

The procedure and mechanism of CMC-g-PAMD synthesis were described in Figure 1. Copolymerization and amidinization can proceed smoothly in water under a CO₂ atmosphere. 1 g CMC and 50 mL deionized water were added into a 250-mL three-neck flask equipped with blender, condenser and CO₂ inlet pipe, and the mixture was stirred under a thermostat water bath for 1 h at 40–60°C. Then the NVF, AN and initiator (CAN:AIBA = 1:1)

Figure 1

The procedure and mechanism of grafting copolymerization (a); amidinization (b).

were injected into the flask and kept stirring at 40-60°C for 5 h. Afterwards the concentrated hydrochloric acid was dropped in the reactants at desired times during the copolymerization, and kept the amidinization of copolymers at 80-100°C for 3 h. Later the copolymer samples were extracted using anhydrous ethanol, and the precipitated crude products were obtained after vacuum drying.

Finally the new flocculant (CMC-g-PAMD) samples could be obtained after removing the impurities inside the crude products by the following process: firstly the crude products were washed with acetone; secondly the samples was smashed into powder after drying under vacuum; finally the “Varma method” was used to reflux the crude samples in a Soxhlet extractor for 12 h, using a mixed solvent of glacial acetic acid and ethylene glycol (1:1 v/v), which could remove the unreacted PAMD and monomers from the crude products effectively.

2.2 Evaluation of the grafting efficiency (GE)

GE was calculated according to the following equation:

where M_c denotes the mass of CMC (g), M_p represents the mass of the product purified by Soxhlet extractor (g).

2.3 Flocculation performance evaluation

A certain amount of the flocculant and 50 ml of coalmine wastewater were put into a 50 ml beaker and stirred for 1 min, and the transmittance of supernatant was measured with spectrophotometer at the wavelength of 610 nm. The coalmine wastewater sample was taken from Shanxi Ximing Coal Group Co., Ltd., and the detailed information was given below: pH = 7.90, turbid degree = 4800 NTU, COD (chemical oxygen demand) = 89 mg/L, SS (suspended substance) = 1120 mg/L, iron ion = 8.93 mg/L, manganese ion = 7.75 mg/L.

3.1 EA

The EA results of AN, NVF, CMC, and the new flocculant (CMC-g-PAMD) were given in Table 1. It was apparent from the table that the C, H, O and N contents of the new flocculant were among those of AN, NVF, CMC and the prepared sample. Compared with CMC, a significant increase of nitrogen percentage inside the prepared sample can be observed, suggesting the successful grafting of AN and NVF onto the backbone of CMC. It can be concluded from Table 1 that PAMD was actually grafted onto CMC during the copolymerization, and produced the new flocculant with specific features and performances.

3.2 TG-DTG

The TGA-DTG curves of CMC and the new flocculant (CMC-g-PAMD) under a nitrogen atmosphere were shown in Figure 2. From the curves we can clearly see the loss of humidity, phase transition and degradation processes. For CMC, three weight loss stages were observed: the first stage of weight loss from room temperature to 180°C was supposed to be derived from water loss; the next weight loss stage within 200–400°C was due to broken of hydroxyl group; and the last weight loss stage at 400–530°C was attributed to the rupture of the main chains of CMC and key bonds. In the case of CMC-g-PAMD, the weight loss during the first region (46–190°C) was related to moisture removal; a sudden large weight loss occurred within 200-360°C as the temperature rose, indicating the thermal decomposition of CMC; and the final weight loss stage (360–520°C) could be explained as the degradation of PAMD chains. As shown in Figure 2, the new flocculant had more residue (37%) than that of CMC (29%) after high temperature heating, and performed a better thermal stability, which illustrated the successful grafting of PAMD onto CMC.

Figure 2

(a) TG-DTG curves of the prepared sample; (b) FT-IR of CMC, NVF, AN and the product.

3.3 FT-IR

FT-IR was used to characterize the functional groups of product (CMC-g-PAMD), and the results were shown in Figure 2b. From the figure we can witness that the broad peak of CMC occurred at 3311 cm^-1, which was assigned to the stretching vibrations of O–H (30). The peak at 1017 cm^-1 represented the functional groups of . For the NVF spectrum in Figure 2b, we can observe that thebands at 3262 and 1382 cm^-1 were corresponding to N–H and C-N stretching vibration peaks in amide group respectively, and a stretching vibration peak of C=O appeared at 1669 cm^-1 (31). For the AN spectrum in Figure 2b, the peak at 2228 cm^-1 (AN) confirmed the existence of C≡N (32). Obviously, compared with CMC, NVF and AN, there are some differences in the infrared spectrum of the product. From the infrared spectrum of the product, the O-H stretching band of CMC backbone and the N-H stretching band of the NVF chain overlapped each other, leading to a combined peak of O-H and N-H groups for the prepared flocculant at 3393 cm^-1. Meanwhile, the peak of product at 2243 cm^-1 was mainly due to the characteristic absorption of C≡N, confirming the participation of AN in the graft copolymerization. Compared with peak at 1669 cm^-1 (NVF), the peak at 1676 cm^-1 of the new flocculant was ascribed to the C=O stretching vibrations (29), and the peak area decrease could also verify the hydrolysis of C=O in NVF. Moreover, the stretching vibrations of C–N and C=N were observed at 1454 cm⁻¹ and 1152 cm⁻¹, respectively, demonstrating the occurrence and development of amidinization reaction as expected (33).

3.4 SEM

Figure 3 showed the SEM micrographs of the new flocculant (Figures 3a and 3b) and CMC (Figures 3c and 3d). It was obvious that the morphological characteristics of the product varied significantly after the reaction (Figure 3b vs. Figure 3d), and the original crystal form of CMC (Figure 3b vs. Figure 3d) turned into the amorphous state of the product (Figures 3a and 3b). It can be revealed from the Figure 3b that the new flocculant was uniform fine powder and these small granules showed a net-like structure under greater magnification. Since the smaller particles of flocculant samples tended to present higher diffusion rate in aqueous solution, the net structure of the product may promote the net capture effect remarkably, resulting in an outstanding flocculation performance accordingly.

Figure 3

SEM photographs of (a) CMC-g-PAMD (×50000); (b) CMC-g-PAMD (×10000); (c) CMC (×5000); (d) CMC (×10000).

3.5 NMR

The ¹³C NMR and ¹H NMR spectra of the new flocculant were experimentally conducted, and the results were shown in Figure 4. It can be revealed from the figure that the highest chemical shift (δ 3.7) in the ¹H NMR spectrum (Figure 4a) could be assigned to —NH₂ group. The peaks at chemical shifts (δ 3.2) can be ascribed to H2, H5 protons on the CMC backbone, and the chemical shift (δ 3.4) could be described as the protons of H7 on the carboxymethyl. The peak at chemical shifts of δ 2.1 can be corresponding to H1, H3 and H4; δ 1.3, δ 1.4, δ 2.0 and δ 2.5 ascribed to H10, H11, H6 and H9 respectively. For the ¹³C NMR spectrum (Figure 4b), the peaks at δ 108.6 and δ 76.0 were assigned to C1 and C10 respectively, and the peak at δ 121.6 ascribed to C3 and C4. Meanwhile the peak at δ 201.4 attributed to C8 and C13, and the peak at δ 44.1 ascribed to C2, C5, C7, C9 and C12.

Figure 4

(a) ¹H-NMR of the new flocculant; (b) ¹³C-NMR of the new flocculant.

4.1 Effect of initiator dosage

Since initiator is crucial for graft copolymerization, the proper type and dosage of initiator can significantly

promote the reaction and should be investigated intensively. Zhao et al. (34) studied different initiators for the copolymerization of NVF, AN and starch, and the optimal initiator for copolymer synthesis was determined as a mixture of CAN + AIBN (m:m = 1:1). Hence we used the initiator in this work, and further investigated its optimal addition amount for the graft copolymerization of the new flocculant. Different initiator dosage from 1 g/L to 5 g/L was experimentally studied, and results were shown in Figure 5a. It can be seen from the figure that the transmittance of wastewater samples climbed gradually with increasing initiator dosage at initial stage, and the flocculation effect reached its peak at the optimal initiator dosage of 4 g/L; afterwards the transmittance tended to drop. This is because the radical polymerization can proceed efficiently when the initiator concentrationis sufficiently high, and the initiator concentration of 4 g/L was proved to have the most excellent flocculation performance. Moreover, the double bond (C=C) activity of AN was obviously higher than that of NVF in the free radical polymerization due to its relatively small steric hindrance, so the excessive initiator concentration could enhance the homopolymerization of AN and hinder the copolymerization between NVF and AN, resulting in the reduced amidine content and poor flocculation effect. Similarly, as shown in Figure 5b, the product gained the largest GE of 156% under the initiator dosage of 4 g/L, meanwhile the over high initiator addition would impel the premature termination of polymerization chains, causing the GE decrease of the CMC-g-PAMD.

Figure 5

Effect of initiator dosage on transmittance (a) and GE (b); effect of polymerization temperature on transmittance (c) and GE (d); effect of amidinization temperature on transmittance (e) and GE (f).

4.2 Effect of copolymerization temperature

In Figures 5c and 5d, the influence of copolymerization temperature was presented. From the figure we can witness that both GE and transmittance showed the same trend: increase first and then decrease. Before 50°C, the reaction rate ascended steadily as graft copolymerization temperature extended. When the temperature exceeded 50°C, the side reactions including homopolymerization was accelerated remarkably, which can greatly affect the copolymerization and flocculation ability, leading to the evident decline of GE and transmittance accordingly. Thus, 50°C was determined as the optimal copolymerization temperature.

4.3 Effect of amidinization temperature

The amidinization temperature has great impact on the molecular diffusion, GE and flocculation performance, and a few experiments were carried out to explore the optimal temperature. As shown in Figures 5e and 5f, when the amidine temperature was below 90°C, the GE and flocculation performance of the CMC-g-PAMD ascended greatly as temperature extended. This was largely due to the increase of both molecular diffusion and amide hydrolysis (to produce amino group) rates, which could promote the driving force for the copolymerization. Whereas, both GE and flocculation efficiency tended to decline when the amidinization temperature exceeds 90°C. This could be because an excessively high temperature might cause a cross-linking reaction of the amino group generated by hydrolysis, consuming the amino group as the raw material for amidinization reaction, and thus the amidine content of the product could be reduced and the flocculation effect was deteriorated evidently. Additionally, high amidinization temperature may also cause the chain breaking of the copolymerized products, resulting in an obvious decrease of GE. Hence 90°C was selected as the optimal amidinization temperature.

4.4 Effect of acidification time

The acidification time is defined as the actual copolymerization time after the addition of hydrochloric acid. As shown in Figures 6a and 6b, we can clearly observe the optimal acidification time was confirmed to be 3 h, which demonstrated an overwhelming flocculation performance than any other acidification time. More or less acid treat time was unfavorable to the transmittance and GE, and both copolymerization efficiency and flocculation ability dropped significantly when the acid treat time exceeded 3 h. The reason was because NVF has extremely poor water solubility, and the ammonium salt was formed and the NVF solubility can be improved significantly after the treatment of hydrochloric acid, causing the stimulating effect on the copolymerization and flocculation. However, the polysaccharide in the product tended to hydrolyze apparently when the acidification time exceeded 3 h, leading to the decrease of GE and flocculation performance of the prepared samples.

Figure 6

Effect of acidification time on transmittance (a) and GE (b); effect of flocculant dosage on transmittance (c); flocculation ability comparison (d).

4.5 Effect of flocculant dosage

Figure 6c showed the impact of dosage of the synthesized flocculant on its transmittance under the optimal operation conditions. As show in Figure 6c, the flocculation performance increased steadily with the rise of the flocculant dosage at initial stage, afterwards the transmittance tended to decline obviously, and the experimental results indicated that the transmittance was over 95% under the optimal flocculant dosage of 60 mg/L. This could be explained as follows: since the new flocculant has dual effects on neutralization and flocculation, it is more prone to collect the suspended particles inside coalmine wastewater with increasing adding amount (35). Meanwhile the flocculation effect tended to decline when the flocculant dosage exceeded its optimal amount, which was not only because the over-neutralized electric charges of wastewater can make the like charges repelled each other, but also the impurity particle surface would be covered by superfluous copolymer chains, leading to the obstacle for bridging flocculation (36).

Furthermore, the flocculation efficiency between the product and conventional PAM were experimentally compared, and the results were shown in Figure 6d. It can be concluded from the figure that the new flocculant behaved a dominate treatment effect for coalmine waste water than PAM under its optimal operation conditions, which would have a great potential and promising application for waste water treatment in the future.