Reactive toughening of polyvinyl alcohol hydrogel and its wastewater treatment performance by immobilization of microorganisms

Abstract

In order to improve the hydraulic impact resistance of a polyvinyl alcohol (PVA) hydrogel as a microorganism immobilization material and meet the requirements of long-time aeration of sewage treatment, toughened PVA hydrogel beads were prepared by co-crosslinking with glycerol through the boric acid (H₃BO₃) – chemical crosslinking method. It was found that glycerol could increase the consumption of H₃BO₃ and decelerate the crosslinking reaction of PVA. Crosslinked structures of borate–PVA monodiol complex (abbreviation as BP) and PVA–borate–PVA didiol complex (abbreviation as BP₂) were formed, and the proportion of BP₂ increased by the introduction of glycerol. Moreover, the pores of the core and surface layers exhibited a similar size and the structure of the PVA/glycerol hydrogel was relatively uniform. With increasing glycerol content, the shear storage modulus (G′) and the effective network density (ν_e) increased first, then decreased, and reached a maximum in the presence of 1.5 wt% glycerol, indicating the formation of a dense network structure of the gel, resulting in an improvement in the tensile properties and crushing strength of the gel beads. PVA/1.5 wt% glycerol immobilized with and without a microorganism exhibited excellent hydrogel stability during long-term wastewater treatment process. The reactive toughening mechanism of glycerol on the PVA hydrogel was explored. The value of the oxygen uptake rate (OUR) and COD removal rate of the PVA hydrogel immobilized with activated sludge had no obvious difference with addition of glycerol, and a high microbial activity can be maintained.

---

1. Introduction

Microbial cell immobilization techniques have been extended to wastewater treatment.^1,2 Compared with the conventional free cell method, using activated sludge immobilization technology for such systems not only offers a high cell concentration in the reaction tank for increasing biodegradation efficiency, but is also advantageous for the separation of liquid and solids in the settling tank.³ It has been used to degrade many toxic substances, like phenol,⁴ 4-chlorophenol,⁵ 2,4-dichlorophenol,⁶ quinoline,⁷ phthalic acid esters.⁸ And it's also widely applied in wastewater treatment,^9–13 like nitrogen removal, heavy metal removal, coke plant wastewater treatment, and so on.

Crosslinked poly(vinyl alcohol) (PVA) hydrogels with hydrophilic three-dimensional polymeric networks, are insoluble in water due to the presence of chemical or physical crosslinks, and exhibit good biocompatibility, non-toxicity, high elastic modulus and low cost, attracting great attention in microorganism immobilization for wastewater treatment.^14,15 However, in sewage aeration tank, the strong shear force produced by oxygen aeration and fluid turbulence made the hydrogel beads easy to be broken and damaged, which required the PVA hydrogel to possess high hydraulic impact resistance in order to meet the requirements of long-time running of sewage treatment.¹⁶ Takei et al.¹⁷ prepared PVA hydrogel immobilized with microorganism through the sodium sulfate – chemical crosslinking method, resulting in the improvement of the mechanical strength of the hydrogel. Li et al.¹⁸ used acrylamide and N,N′-methylene double acrylamide to modify PVA hydrogel by copolymerization method, and the product was not easy to break.

For PVA hydrogel beads prepared through the boric acid (H₃BO₃) – chemical crosslinking method,^19–21 its surface crosslinked rapidly when contacting with the curing agent solution, forming a dense crosslinked shell, and hindering the curing agent solution to further disperse into core layer of the hydrogel, resulting in the low crosslinking degree and loose structure of the core layer. And thus the inhomogeneous crosslinking porous structure formed, leading to low mechanical strength of PVA hydrogel beads.

Glycerol is a simple polyol compound with low toxicity and a reactive molecule that undergoes all the usual reactions of alcohols.^22,23 The two terminal primary hydroxyl groups are more reactive than the internal secondary hydroxyl group.²⁴ In particular, H₃BO₃ could react with glycerol and yield glyceroborate, which was highly presented in the patent literature.^25,26 In this work, the crosslinking reaction rate of the shell layer and core layer of the PVA hydrogel beads was controlled through the competition reaction of glycerol and PVA with H₃BO₃, and the homogeneous crosslinking network structure was tried to be achieved. As a result the mechanical strength of PVA hydrogel can be expected to be improved. The effect of glycerol on crosslinking reactive kinetics and crosslinking molecules structure of PVA was investigated. The network structure and toughening mechanism were explored. Furthermore, the PVA hydrogel was applied in the wastewater treatment by immobilizing with microorganism, and the hydraulic impact resistance and waste water treatment efficiency were investigated.

2. Experimental

2.1 Materials

PVA with an average molecular weight (M_n) of 74 800 g mol⁻¹ was supplied by Sichuan Vinylon Co. (China). Sodium alginate (Alg) was purchased from Kelong Chemical Co. (Chengdu, China). Boric acid (H₃BO₃) and glycerol were purchased from Bodi Chemical Co. (Tianjin, China). Other chemical agents were all of analytical purity and used as received. The wastewater and activated sludge were collected from Chengdu Drainage Co. (Chengdu, China) on the day for preparation of PVA hydrogel beads.

2.2 Preparation of PVA hydrogel beads

PVA (10 g), Alg (1 g) and glycerol (0–2.5 g) were dissolved in distilled water (100 mL) by magnetic stirring at 95 °C. The solution was cooled to room temperature. The sludge (10 g) was added in the meantime for the sample immobilized with microorganisms. The mixture was then dropped into the saturated H₃BO₃ and CaCl₂ solution (3.0 wt%), keeping for 1 h to form spherical beads. The formed beads were finally soaked and washed with distilled water, and stored in wastewater at room temperature. The diameter of hydrogel beads ranged from 4–6 mm.

2.3 Measurements

2.3.1 ¹¹B NMR analysis. ¹¹B NMR measurements of H₃BO₃, PVA/H₃BO₃ and PVA/glycerol/H₃BO₃ were performed with a Varian Inova-400 NMR spectrometer (Chesterfield of MI, USA). The samples were dissolved in D₂O, and the test was operated at room temperature and a frequency of 400 MHz.

2.3.2 Mechanical property.
The tensile property. The tensile property of the samples of PVA hydrogel was measured with a 4302 material testing machine from Instron Co. (USA) according to ISO 527-1993. Samples with a dumbbell shape and size 150 × 10 × 4 mm³ were prepared. The tensile speed and temperature were 20 mm min⁻¹ and 23 °C, respectively.
The hydraulic impact property. PVA hydrogel beads were put into the distilled water, stirring at room temperature with stirring speed 2000 rpm. The retention rate of gel beads without damage (the hydrogel beads could keep completely spherical shape and there was no obvious decrease for the diameter of beads ranging from 4–6 mm) as a function of time was calculated with the following equation:²⁷

Retention rate (%) = (N₀ − N_t)/N₀ × 100% (2.1)

where N₀ was initial number of gel beads; N_t was the number of gel beads at time t.

2.3.3 Rheological property. The viscoelasticity properties of PVA hydrogel were performed on Rheometer System Gemini 200 of Malvern Instrument Co. (UK) with parallel plates with diameter of 25 mm and a plate-to-plate distance of 1–2 mm. Both the strain and the frequency sweep experiments were performed at room temperature. In strain sweep measurements, the shear storage modulus (G′) and loss modulus (G′′) were recorded at the strain of 0.005–100% and the frequency of 1 Hz. In the frequency sweep experiments, G′ and G′′ were measured in the linear viscoelastic regime, for frequencies ranging from 0.1 to 100 Hz, at a maximum strain, γ, of 0.1%. This γ value was determined by preliminary strain sweep experiments, in which G′ and G′′ were measured as a function of strain at a fixed frequency value of 1 Hz to check if the deformation imposed on the gel structure during the rheological experiment was entirely reversible.

2.3.4 Scanning electron microscope analysis (SEM). The fractured surface morphology of PVA hydrogel beads was observed with a JEOL JSM-5900LV scanning electron microscope (SEM) (Japan) at an acceleration voltage of 20 kV. The hydrogel beads were freezing dried and cryogenically fractured in liquid nitrogen. Then the samples were sputter-coated with gold for 2–3 min.

2.3.5 Oxygen uptake rate analysis (OUR). 100 mL distilled water was put into a flask, and aerated to make the dissolved oxygen saturated by air pump. Stop aeration and five grams of PVA hydrogel beads immobilized with sludge were added. The concentration of the dissolved oxygen (DO) variation with time was measured with AZ 8403 DO meter (China).

2.3.6 Chemical oxygen demand (COD) removal rate. COD value of the wastewater mainly depends on the composition and concentration of organic contaminants, and the absorbance of the organics has good correlation with COD.^28,29 The wastewater solution with different concentrations was prepared, and the UV absorbance of the solution at 254 nm was measured with U3010 UV-visible spectrophotometer (Japan). The standard curve of UV absorbance-wastewater concentration was thus obtained.

In this work, 20 g PVA hydrogel beads immobilized with sludge (1.65 g) were added in a reactor containing 200 mL wastewater. After 6 hours' aeration, the sample of 20 mL wastewater solution was taken out for UV measurement. The absorbance at 254 nm was measured, and the corresponding wastewater concentration can be obtained by the standard curve. The COD removal rate was then calculated with the following equation:

COD removal rate (%) = (C_t − C₀)/C₀ × 100% (2.2)

where C₀ was the original wastewater concentration before treatment, and C_t was the wastewater concentration after treatment. The original COD of wastewater was 324 mg L⁻¹.

2.3.7 Statistical analysis. The quantitative results were obtained from triplicate samples and the data was expressed as mean ± SD (n = 3). Statistical analysis was performed using one-way analysis of variance, followed by post hoc Student's t-test. A value of P < 0.05 was considered to be statistically significant.

3. Results and discussion

3.1 Effect of glycerol on crosslinking reaction kinetics of PVA

During the crosslinking reaction of PVA/glycerol system by using H₃BO₃ as the crosslinking agent, both the hydroxyl groups on the molecule of PVA and glycerol could react with H₃BO₃. Therefore, the effect of glycerol on the crosslinking reaction kinetics of PVA hydrogel was investigated. Fig. 1 illustrated the molar concentration of H₃BO₃versus crosslinking time of PVA hydrogels with different content of glycerol. It can be seen that in the first 20 minutes of reaction, the molar concentration of H₃BO₃ decreased sharply, and reached equilibrium in about 60 minutes. With increasing content of glycerol, the molar equilibrium concentration of H₃BO₃ decreased first, reached minimum in presence of 1.5 wt% glycerol, and increased afterwards, indicating that glycerol could increase consumption of H₃BO₃ during the crosslinking reaction of PVA hydrogel. Under low content of glycerol, with increasing glycerol content in the PVA solution, the increment of glycerol volume fraction in the multi-phase system raised the free energy, resulting in higher interface diffusion coefficient of PVA system, which helped to the crosslinking reaction, thus the consumption of H₃BO₃ increased. However, under excessive addition of glycerol, the increment of glyceroborate volume fraction yielded by reaction of H₃BO₃ with glycerol reduced free energy, resulting in lower interface diffusion coefficient, which hindered the crosslinking reaction, and so the consumption of H₃BO₃ decreased.^30–32

Fig. 1 The molar concentration of H₃BO₃versus crosslinking time of PVA hydrogel with various content of glycerol. Data points represent mean ± SD (n = 3), P = 0.04.

The crosslinking reaction kinetics of PVA composite hydrogel was analyzed by assuming that the crosslinking reaction process met the first order kinetic equation:

dB_t/dt = k(B_t − B_e) (3.1)

where t was crosslinking reaction time; B_t, molar concentration of H₃BO₃ at time t; B_e, molar concentration of H₃BO₃ when crosslinking reaction reached the equilibrium, and dB_t/dt, crosslinking reaction rate; k, the crosslinking reaction rate constant. The crosslinking reaction equation can be obtained by integral of the above equation:

B_t = B₀ − (B₀ − B_e)/e^kt (3.2)

Another form of the equation can be written as:

lg(B_t − B_e) = −kt/2.303 + lg(B₀ − B_e) (3.3)

By plotting the graph of lg(B_t − B_e) versus time t, as shown in Fig. 2, the crosslinking reaction rate constant k can be obtained from the slope of the curve, which exhibited good linear relationship. The crosslinking reaction rate constant k of PVA hydrogel beads with varying content of glycerol was listed in Table 1. It can be seen that with increasing content of glycerol, the crosslinking reaction rate constant k of PVA hydrogel decreased, indicating deceleration effect of glycerol on the crosslinking reaction.

Fig. 2 Crosslinking reaction kinetics of PVA hydrogel with various content of glycerol. Data points represent mean ± SD (n = 3), P = 0.03.

Table 1 Crosslinking reaction rate constant PVA with H₃BO₃

Sample	Reaction rate constant k (h^−1)	Equilibrium H3BO3 molar concentration (mmol L^−1)
PVA	1.40	39.26
PVA/0.5 wt% glycerol	1.30	39.02
PVA/1.0 wt% glycerol	1.21	35.82
PVA/1.5 wt% glycerol	1.16	35.31
PVA/2.0 wt% glycerol	1.07	37.86
PVA/2.5 wt% glycerol	1.06	37.68

---

3.2 Effect of glycerol on crosslinking molecular structure of PVA hydrogel

H₃BO₃ in water dissociated into borate ion (H₄BO₄⁻) and hydrogen ion (H⁺), while H₄BO₄⁻ can react with cis-ortho-di-hydroxyl groups on the molecular chain of polyol to yield single and bi-crosslinking products.^33,34 PVA with linear hydrocarbon chains containing a lot of 1,3-type hydroxyl groups reacted with H₃BO₃, resulting in the formation of crosslinking structure of borate–PVA monodiol complex (abbreviation as BP) and PVA–borate–PVA didiol complex (abbreviation as BP₂), as shown in Fig. 1s†.³⁵ Meanwhile, only crosslinking caused by formation of BP₂ made PVA generate a three dimensional crosslinking network structure, resulting in the formation of hydrogel. Glycerol with three hydroxyl groups on its molecule reacts with H₃BO₃ to form an organic ligand. Compared with PVA, glycerol as a small molecule compound has higher reactive activity when reacting with H₃BO₃. For PVA/glycerol system, H₃BO₃ first reacted with glycerol, and yielded glyceroborate (abbreviation as BG₂), and then the ester exchange reaction occurred between BG₂ and PVA, resulting in the formation of crosslinking structure of BP, BP₂ and borate–PVA–glycerol (abbreviation as BPG), as shown in Fig. 2s.†

The ¹¹B NMR spectrums of H₃BO₃, PVA/H₃BO₃ and PVA/glycerol/H₃BO₃ were shown in Fig. 3. It can be seen that for pure H₃BO₃, the peak at 4.38 ppm was attributed to the chemical shift of H₃BO₃/H₄BO₄⁻. For PVA/H₃BO₃ system, three peaks can be observed. The chemical shift of H₃BO₃/H₄BO₄⁻ appeared at 5.59 ppm, which was higher than that of pure H₃BO₃, and the peak became broader. It was perhaps mainly due to the consumption of H₃BO₃ by PVA and relatively low molar concentration of H₃BO₃/H₄BO₄⁻. The chemical shifts of BP₂ and BP exhibited at 1.71 ppm and 2.37 ppm, respectively. For PVA/glycerol/H₃BO₃ system, the peaks at 6.01 ppm, 1.69 ppm and 2.42 ppm belonged to the chemical shifts of H₃BO₃/H₄BO₄⁻, BP₂ and BP respectively. A new peak appearing at 4.21 ppm was attributed to the chemical shift of BG₂. The chemical shifts of BP₂ and BP in PVA/H₃BO₃ and PVA/glycerol/H₃BO₃ had partial overlap. The Peakfit software was used to separate them, and the proportion of integration area of each peak in the total peak area was calculated, and listed in Table 2. It can be seen that compared with PVA/H₃BO₃ system, the proportion of BP₂ in the two crosslinking structures was obviously higher for PVA/glycerol/H₃BO₃ system, indicating that glycerol could improve the crosslinking degree of PVA hydrogel.

Fig. 3 ¹¹B NMR spectrum of H₃BO₃ (a), PVA/H₃BO₃ (b) and PVA/glycerol/H₃BO₃ (c) in D₂O solution.

Table 2 ¹¹B NMR results of H₃BO₃, PVA/H₃BO₃ and PVA/glycerol/H₃BO₃

Samples	Chemical shift (ppm)	Type of B	Area (%)
H3BO3	4.38	H3BO3/H4BO4^−	100
PVA/H3BO3	5.59	H3BO3/H4BO4^−	80.97
	2.37	BP	12.01
	1.71	BP2	7.02
PVA/glycerol/H3BO3	6.01	H3BO3/H4BO4^−	28.70
	4.21	BG2	64.51
	2.42	BP	3.66
	1.69	BP2	3.13

---

The observation of the internal morphology of PVA hydrogel and PVA/glycerol hydrogel was carried out. Fig. 4 showed SEM images of the cross section, core layer and surface layer of cross section of PVA hydrogel and PVA/glycerol hydrogel. It can be seen that, for PVA hydrogel, the morphology of core layer and surface layer had obvious differences: the pores were dense and small in the surface layer, whereas the pores were loose and large in the core layer. For PVA/glycerol hydrogel, the pores of core layer and surface layer of exhibited similar size and relatively uniform structure.

Fig. 4 SEM image of the cross section of PVA hydrogel and PVA/glycerol hydrogel.

The pore size distribution for the PVA hydrogel and PVA/glycerol hydrogel was summarized in Fig. 5. It can be clearly observed that for PVA hydrogel, the diameter of pore on the surface layer ranged from 0–12 μm, and focused on 2–4 μm, while the pore size distribution of core layer ranged from 0–21 μm, and focused on 0–6 μm. However, for PVA/glycerol hydrogel, the pore on the surface layer had larger and uniform size, and the diameter of pore ranged from 0–9 μm, and focused on 3–6 μm. Meanwhile, the pore size distribution of core layer ranged from 0–10 μm, and focused on 4–6 μm. This indicated that introduction of glycerol could control pore size of surface layer and core layer of PVA hydrogel beads, and make the porous structure more uniform.

Fig. 5 Pore size distribution of PVA hydrogel and PVA/glycerol hydrogel (a) surface layer of PVA hydrogel (b) core layer of the PVA hydrogel (c) surface layer of PVA/glycerol hydrogel (d) core layer of the PVA/glycerol hydrogel.

Based on the above analysis results, the crosslinking reaction mechanism of PVA system can be deduced:

For PVA in absence of glycerol, the crosslinking reaction of PVA with H₃BO₃ started on the surface layer of the liquid drops of PVA solution and the formed dense crosslinking networks may hinder H₄BO₄⁻ to penetrate into core layer of PVA liquid drops, resulting in a relatively low crosslinking degree and inhomogeneous network structure.

For PVA in presence of glycerol, the crosslinking process included the following two steps:

Step I: glycerol and PVA on the surface layer of the liquid drops reacted with H₄BO₄⁻ to yield glyceroborate and PVA–borate crosslinking networks, respectively. Besides, the reactive rate of the former is higher than that of the latter. H₄BO₄⁻ and glyceroborate can percolate into the core of the drops.

Step II: ester exchange reaction of glyceroborate and PVA in the core layer of the drops promoted the formation of crosslinking networks in the core layer. This path of reaction made the crosslinking reaction rate constant of PVA/H₃BO₃ decrease, resulting in a relatively uniform porous structure of the PVA hydrogel beads.

3.3 Effect of glycerol on crosslinking network structure of PVA hydrogel

Fig. 6 showed the strain dependence at the frequency of 1 Hz of the shear storage modulus (G′), and loss modulus (G′′) for PVA hydrogel with different content of glycerol. At very low strain amplitudes, the loss modulus of the hydrogel was lower than the storage modulus, which was consistent with the existence of a network structure. In addition, it can be noted that, for small strain amplitudes, G′ was independent of the strain amplitude, which indicated that the deformation imposed on the network structure was entirely reversible.³⁶ At higher strain amplitudes, G′ was a decreasing function of the strain amplitude and the deformation was no longer reversible.

Fig. 6 Strain dependence of G′ and G′′ for PVA hydrogel with various content of glycerol.

The frequency dependence of G′ was plotted in Fig. 7 for PVA hydrogel with different content of glycerol. It can be seen that G′ does not depend on the test frequency in the range between 0.1 and 100 Hz for all samples, indicating that the elastic behavior of these samples predominated over their viscous behavior, and a perfect network formed. With increasing content of glycerol, G′ increased first, reached maximum in presence of 1.5 wt% glycerol and decreased afterwards.

Fig. 7 Frequency dependence of G′ for PVA hydrogel with various content of glycerol.

Equilibrium water content (EWC), volume fraction of polymer in the hydrogel at equilibrium swelling, ϕ₁, and volume fraction of the crosslinking polymer in the relaxed state, ϕ₂, can be calculated as follows:^37,38

where W_d was the weight of the dry gel, W_r was the weight of the relaxed gel and W_e was the weight of gel in the equilibrium swollen state, ρ_p and ρ_s were the densities of polymer matrix and water, respectively. For a blend of PVA and Alg as the matrix, the density of the polymer matrix (ρ_p) was the same for all the hydrogels and calculated as follows:where W_PVA and W_Alg were the weight fractions of PVA and Alg, respectively, and ρ_PVA and ρ_Alg were the densities of PVA and Alg, respectively.

The effective network density (ν_e) of the PVA hydrogel was determined from the following equation based on the rubber elasticity theory:

G = RTν_eϕ₁^⅓ϕ₂^⅔ (3.8)

where R was the gas constant, T was the temperature. The average molecular mass between crosslinks, M_c, was calculated as follows:

The value of G′, ν_e and M_c of PVA hydrogel with different content of glycerol was calculated and listed in Table 3. It can be seen that, with increasing glycerol content, G′ and ν_e increased first, decreased afterwards, and reached maximum in presence of 1.5 wt% glycerol, at this point M_c reached minimum indicating that a relatively low content of glycerol could promote the formation of relatively uniform and dense network structure in PVA hydrogel, which was coincident with the formation of BP₂ crosslinking structure. Excessive addition of glycerol led to the retarding of PVA crosslinking reaction and decrease of crosslinking density of the whole system.

Table 3 Network parameters of PVA hydrogel with various content of glycerol

Sample	ρ p (g cm^−3)	ϕ 1	ϕ 2	G′ (kPa)	ν e (mol m^−3)	M c (kg mol^−1)
PVA	1.219	0.1325	1	7.13	5.55	2.20 × 10^2
PVA/0.5 wt% glycerol	1.219	0.1222	1	10.3	8.23	1.48 × 10^2
PVA/1.0 wt% glycerol	1.219	0.1308	1	24.9	19.46	0.63 × 10^2
PVA/1.5 wt% glycerol	1.219	0.1213	1	31.7	25.41	0.48 × 10^2
PVA/2.0 wt% glycerol	1.219	0.1252	1	13.4	10.63	1.15 × 10^2
PVA/2.5 wt% glycerol	1.219	0.1262	1	11.3	8.94	1.36 × 10^2

---

Fig. 8 showed SEM images of cross section of PVA hydrogel beads with and without glycerol. It can be seen that PVA gel beads in absence of glycerol presented closed pores. By addition of 1.5 wt% glycerol, many large open pores formed, so as to provide channels for mass transfer. By addition of 2.5 wt% glycerol, a lot of small dense pores appeared on the wall of large pores.

Fig. 8 SEM image of the cross section of PVA hydrogel beads with different glycerol content (magnification: 2000×).

3.4 Effect of glycerol on mechanical properties of PVA hydrogel

The tensile mechanical property of PVA hydrogel with varying content of glycerol was depicted in Fig. 9. It can be seen that with increasing content of glycerol, the tensile strength and the elongation at break of PVA hydrogel increased dramatically, and reached maximum at 1.5 wt% glycerol, and then decreased, which was coincident with the ν_e tendency.

Fig. 9 Mechanical properties of PVA hydrogel with various content of glycerol. Data points represent mean ± SD (n = 3), P = 0.04.

Table 4 showed the effect of glycerol content on the hydraulic impact resistance of PVA hydrogel beads. It can be seen that addition of glycerol improved the crushing strength of the gel beads. With increasing content of glycerol, the retention rate of gel beads increased first, and reached the maximum at 1.5 wt% glycerol, and then decreased. These results indicated that proper content of glycerol could toughen and strengthen PVA hydrogel.

Table 4 Crushing strength of PVA hydrogel beads with various content of glycerol

Sample	Time (h)
	Retention rate (%)
	4	5	6	7	8	8.5	9	10	10.5
PVA	48	0	0	0	0	0	0	0	0
PVA/0.5 wt% glycerol	64	32	0	0	0	0	0	0	0
PVA/1.0 wt% glycerol	100	100	100	100	100	60	28	0	0
PVA/1.5 wt% glycerol	100	100	100	100	100	100	96	64	0
PVA/2.0 wt% glycerol	100	100	100	100	100	100	68	0	0
PVA/2.5 wt% glycerol	100	100	100	84	42	0	0	0	0
PVA/3.0 wt% glycerol	100	100	56	0	0	0	0	0	0

---

Furthermore, the long-term hydraulic stability of PVA/1.5 wt% glycerol hydrogel beads and PVA/1.5 wt% glycerol hydrogel beads immobilized with microorganism in the process of wastewater treatment was investigated. Fig. 3s† showed photos of the hydrogel samples before and after long-term aeration. It can be seen that after 11–22 months aeration test, the surface of the hydrogel beads was not damaged, and the shape and size of the beads were kept well, indicating that PVA/glycerol hydrogel had excellent long-term hydraulic stability.

Based on the above discussion of crosslinking reaction kinetics, the crosslinking molecular structure, porous crosslinking network structure and mechanical properties of PVA/glycerol hydrogel, the reactive toughening mechanism of such hydrogel was deduced: introduction of glycerol changed the crosslinking reaction pathway of PVA by slowing down the crosslinking reaction of PVA with H₃BO₃ on the surface layer of the liquid drop of PVA solution and promoting the crosslinking reaction in its core layer, resulting in the formation of more regular and uniform porous structure of the hydrogel. In the meantime, the proportion of BP₂ crosslinking structure and the effective network density increased, resulting in the improvement of mechanical strength and modulus, and the hydraulic impact resistance of PVA hydrogel beads.

3.5 Effect of glycerol on the microbial activity of PVA hydrogel immobilized with microorganism

The PVA hydrogel immobilized with activated sludge was applied in waste water treatment. OUR was an important parameter to characterize the microbial activity of activated sludge in the wastewater treatment process, which can be calculated with the following equation:³⁹where OUR was the oxygen concentration consumed per minute; DO₁ was the dissolved oxygen concentration at time t₁; DO₂ was the dissolved oxygen concentration at time t₂; t was the measuring time.

The effect of glycerol on DO variation of PVA immobilized beads with time in the saturated aerated distilled water was depicted in Fig. 10. The value of OUR was calculated and fluctuated between 0.103 and 0.113.

Fig. 10 ρ[DO]–t curve of PVA hydrogel beads with various content of glycerol. Data points represent mean ± SD (n = 3), P = 0.04.

The effect of glycerol content on the COD removal rate of PVA hydrogel was investigated, as shown in Fig. 11. The COD removal rate of PVA hydrogel with different addition of glycerol fluctuated and reached up to 80–85%, implying that the formed large pores of PVA hydrogel beads were beneficial to supply channels for mass transfer of microorganism, and the high microbial activity of the gel beads can be kept well. All the PVA hydrogel beads with porous structure had good permeability, providing channels for mass transfer, while the microbial reaction mainly controlled the waste water treatment efficiency. Therefore, the microbial activity and COD removal rate had no significant change by addition of glycerol.

Fig. 11 Effect of glycerol content on COD removal rate of PVA hydrogel beads. Column with error bars represent mean ± SD (n = 3), P = 0.02.

For the COD removal of PVA/glycerol hydrogel, the dosage was 20 g PVA gel per 200 mL, which meant 100 g L⁻¹ (glycerol concentration between 0.5 and 2.5%). Thus, the possible carbon release in 6 hours' aeration was estimated as (10 to 15)/100 × 6 × 100 × ((0.5–2.5%)/40%) = 0.5–6 mg L⁻¹.^40,41 The data were significantly lower than that under normal wastewater treatment process and then the carbon release process can be ignored.

4. Conclusions

The toughening PVA hydrogel beads were prepared by introduction of glycerol through H₃BO₃ – chemical crosslinking method. It was found that addition of glycerol could increase consumption of the H₃BO₃ and decelerate the crosslinking reaction PVA hydrogel. The ¹¹B NMR analysis results showed that compared with PVA/H₃BO₃ system, the proportion of BP₂ in the two crosslinking structures was obviously higher for PVA/glycerol/H₃BO₃ system. Besides, the introduction of glycerol could control pore size of surface layer and core layer of PVA hydrogel beads, and make the porous structure more uniform. The crosslinking reaction mechanism of PVA system with addition of glycerol was deduced. With increasing glycerol content, G′ and ν_e increased first, decreased afterwards, and reached maximum in presence of 1.5 wt% glycerol, at this point M_c reached minimum indicating the formation of a relatively uniform and dense network structure of PVA hydrogel, resulting in an improvement of tensile property and the hydraulic impact resistance of PVA hydrogel beads. Compared with pure PVA gel beads, many large open pores formed for PVA/glycerol hydrogel beads, so as to provide channels for mass transfer. Moreover, PVA/1.5 wt% glycerol immobilized with and without microorganism exhibited excellent hydrogel stability during long-term wastewater treatment process. The reactive toughening mechanism of glycerol on PVA hydrogel was also explored. More importantly, the value of OUR and COD removal rate of the PVA hydrogel fluctuated with addition of glycerol implying that the high microbial activity of the gel beads can be kept well.

References

1. S. Shi, Y. Qu, Q. Ma, X. Zhang, J. Zhou and F. Ma, Bioresour. Technol., 2015, 190, 159–166.
2. I. Cruz, Y. Bashan, G. Hernàndez-Carmona and L. E. de-Bashan, Appl. Microbiol. Biotechnol., 2013, 97, 9847–9858.
3. S. A. Covarrubias, L. E. de-Bashan, M. Moreno and Y. Bashan, Appl. Microbiol. Biotechnol., 2012, 93, 2669–2680.
4. G. Bayramoglu, A. Akbulut and M. Y. Arica, J. Hazard. Mater., 2013, 244, 528–536.
5. M. Murcia, M. Gómez, J. Bastida, A. Hidalgo, M. Montiel and S. Ortega, Environ. Technol., 2014, 35, 1866–1873.
6. R. Xu, Q. Zhou, F. Li and B. Zhang, Chem. Eng. J., 2013, 222, 321–329.
7. Y.-M. He, F.-T. Song and Q.-H. Fan, Stereoselective Formation of Amines, 2013, vol. 343, pp. 145–190.
8. J. Wang, P. Liu and Y. Qian, Environ. Int., 1997, 23, 775–782.
9. M. Herrero and D. Stuckey, Chemosphere, 2014, 140, 119–128.
10. I. Kim, M. Lee and S. Wang, J. Environ. Manage., 2014, 146, 568–574.
11. S. Shi, Y. Qu, F. Ma and J. Zhou, Bioresour. Technol., 2014, 166, 79–86.
12. B. Hui, Y. Zhang and L. Ye, Chem. Eng. J., 2014, 235, 207–214.
13. L.-S. Zhang, W.-Z. Wu and J.-L. Wang, J. Environ. Sci., 2007, 19, 1293–1297.
14. G. Zhou, Z. Wang, W. Li, Q. Yao and D. Zhang, Mater. Lett., 2015, 156, 205–208.
15. C.-W. Lin, C.-H. Wu, W.-T. Huang and S.-L. Tsai, Fuel, 2015, 144, 1–8.
16. Y. Zhang and L. Ye, Composites, Part B, 2014, 56, 749–755.
17. T. Takei, K. Ikeda, H. Ijima and K. Kawakami, Process Biochem., 2011, 46, 566–571.
18. J. Li, Z. Suo and J. J. Vlassak, J. Mater. Chem. B, 2014, 2, 6708–6713.
19. S. Hashimoto and K. Furukawa, Biotechnol. Bioeng., 1987, 30, 52–59.
20. Z.-E. Long, Y. Huang, Z. Cai, W. Cong and F. Ouyang, Process Biochem., 2004, 39, 2129–2133.
21. R. Dave and D. Madamwar, Process Biochem., 2006, 41, 951–955.
22. M. Gupta and N. Kumar, Renewable Sustainable Energy Rev., 2012, 16, 4551–4556.
23. H. Tan, A. A. Aziz and M. Aroua, Renewable Sustainable Energy Rev., 2013, 27, 118–127.
24. R. Christoph, B. Schmidt, U. Steinberner, W. Dilla and R. Karinen, Ullmann's encyclopedia of industrial chemistry, 2006.
25. C. Mukhopadhyay, A. Datta and R. J. Butcher, Tetrahedron Lett., 2009, 50, 4246–4250.
26. A. Z. Halimehjani, H. Gholami and M. R. Saidi, Green Chem. Lett. Rev., 2012, 5, 1–5.
27. X. Li, A. Hu and L. Ye, J. Polym. Environ., 2011, 19, 398–404.
28. M. Mrkva, Water Res., 1983, 17, 231–235.
29. D. Georgiou, P. Melidis, A. Aivasidis and K. Gimouhopoulos, Dyes Pigm., 2002, 52, 69–78.
30. D. Zhang, G. Zhou, X. Zhang, Y. Wang and G. Li, Environ. Technol., 2015, 36, 395–403.
31. P. L. Ritger and N. A. Peppas, J. Controlled Release, 1987, 5, 23–36.
32. R. G. Petschek and E. M. Terentjev, Phys. Rev. A, 1992, 45, 930.
33. S. W. Sinton, Macromolecules, 1987, 20, 2430–2441.
34. E. T. Wise and S. G. Weber, Macromolecules, 1995, 28, 8321–8327.
35. C. Cencetti, D. Bellini, A. Pavesio, D. Senigaglia, C. Passariello, A. Virga and P. Matricardi, Carbohydr. Polym., 2012, 90, 1362–1370.
36. Z. Lian and L. Ye, J. Appl. Polym. Sci., 2013, 128, 3325–3329.
37. W. C. Jao, H. C. Chen, C. H. Lin and M. C. Yang, Polym. Adv. Technol., 2009, 20, 680–688.
38. X. Li, W. Wu, J. Wang and Y. Duan, Carbohydr. Polym., 2006, 66, 473–479.
39. L. Zhang, W. Wu and J. Wang, J. Environ. Sci., 2007, 19, 1293–1297.
40. C.-W. Du, J.-M. Zhou and A. Shaviv, J. Polym. Environ., 2006, 14, 223–230.
41. D. Zhang, X. Zhang, Y. Wang, G. Zhou and G. Li, Environ. Technol., 2015, 36, 909–919.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra20495j

---