BCCP (1.0 g) calcined with six different temperatures were put into a 50-ml flask with pure NH₄⁺-N solution and shaken at 150 r min^–1 for 300 min at 25°C. Samples were collected at 5, 10, 20, 40, 60, 90, 120, 150, 180, 240, and 300 min. The samples were filtrated by 0.45-μm RC membrane to determine NH₄⁺-N concentration.

The adsorption capacity during the adsorption period was calculated by Equation (1):

where q_t is the amount of NH₄⁺-N during the adsorption time (mg kg^–1); C₀ and C_t (mg L^–1) are the initial NH₄⁺-N concentrations and different time residual concentration, respectively. V is the volume of reaction system (L), and m is the mass of adsorbent (g).

The data coming from adsorption batch experiment were fitted with pseudo-first- and pseudo-second-order models and intraparticle model, the expressions as following Equations (2)–(4):

where q_e (mg⋅g^–1) is the adsorbed amounts of NH₄⁺-N by the BCCP at equilibrium time, and q_t is the adsorbed amount at a given time interval (t). K₁ and K₂ are the rate constants for the pseudo-first- and pseudo-second-order models, respectively. Kp is the intraparticle diffusion rate constant (mg⋅g^–1 min^1/2), and C (mg g^–1) is a constant that reflects the boundary layer effect. A plot of q_t against t^1/2 gave a linear relationship from which the Kp value was determined from the slope and C as the intercept.

Freundlich and Langmuir equations were used to fit the adsorption isotherms of BCCP with different calcined temperatures, and the equation is given by

where q_e (mg⋅g^–1) is the amount of NH₄⁺-N adsorbed by the BCCP at equilibrium time, and q_max (mg g^–1) and K_L (L mg^–1) are Langmuir constants that indicate the maximum adsorption and relative binding energy of BCCP, respectively. K_F and n are Freundlich constants that measure the relative NH₄⁺-N adsorption capacity and adsorption intensity of BCCP, respectively, while C_e (mg L^–1) denotes the equilibrium concentration of NH₄⁺-N remaining in solution after adsorption is complete.

HXN-1was made into gel particles by the method of 2.4 and named SA-B and SA/PVA-B and the blank gel particles were named SA-C and SA/PVA-C. SA and PVA as base material to immobilize Ochrobactrum sp. was named SA/PVA-B, and as control group without adding Ochrobactrum sp. was named SA/PVA-C. The aforementioned gel particles were put into 100 ml of 50 mg L^–1 NH₄⁺-N medium, placed in a shaking incubator at 25°C, 180 r min^–1 for 7 days, and NH₄⁺-N concentration was measured daily. The medium characteristics were demonstrated in Huang et al. (2017). Free bacteria (FB) were used as a control group.

Six kinds of gel particles were, respectively, put into 100 ml of 50 mg L^–1 NH₄⁺-N solution with different salinities (0, 5, 15, 25, and 35‰) under 25°C condition. Also, the same six kinds of gel particles were with different temperatures (15, 20, 25, 30, and 35°C) with 5‰ salinity. All of these were placed in a shaking incubator and shocked with 180 r min^–1 for 10 days and NH₄⁺-N concentration was measured daily. Free bacteria were used as a control group.

NH₄⁺-N adsorption performances from aqueous solution by six BCCPs with different temperatures were conducted, and the results are shown in Figure 2. The adsorption equilibrium time among six experimental groups was 180 min, and the NH₄⁺-N removal efficiency by BCCP adsorption was about 29.4∼34.5%. The removal efficiency at 400°C group was better than that of the other four groups with the adsorption capacity of 0.473 mg g^–1. However, the value decreased first and then increased with the increase of calcined temperature. The same phenomenon existed in the research of Chen et al. (2018) who found that the surface area and pore volume of bentonite increased to 56.09 and 0.0611 cm³ g^–1 when the calcined temperature was 400°C, respectively, but sharply declined to 30.53 and 0.051 cm³ g^–1 at 800°C. Yan et al. (2018) prepared a kind of porous diatomite microsphere by spray drying method and the methylene blue adsorption capacity and removal efficiency demonstrated the maximum values when the calcined temperature was 600°C, and decreased when the temperature rose. Ojeda-López et al. (2021) found that adsorbent/adsorbate interactions for CO₂, CH₄, and N₂ were inversely proportional to calcined temperature (CMF-600 > CMF-700 > CMF-800) by the mean of the isosteric enthalpy of adsorption measurements.

This was because the pore volume and Brunauer–Emmett–Teller (BET) surface area reached maximum at some temperature and decreased further with the increase of calcined temperature (Kar and Equeenuddin, 2019). The organic compounds presenting in biochar or clay would condense on the surface of the particles, and clog the pores to decrease specific surface area after cooling with the increase of calcined temperature (Atkinson et al., 2010). Also, four forms of water existed in clay minerals (i.e., surface adsorbed water, pore adsorbed water, crystalline water combined with octahedral cations at the edge of pore, and cationic structural water combined with octahedral layer), and high temperature led to the adsorbed water, pore water, and bound water in the material lost when the temperature was less than 600°C, and the carbon in biochar and clay was oxidized. Meanwhile, the decomposition of NaHCO₃ increased the pores and adsorption in the green body. The decrease of adsorption capacity from 500°C might be due to the fission of C400 biochar fired at 400°C with the temperature rising to 500°C, and the forming of ash adsorbed in the pores of BCCP and reduced its adsorption performance for NH₄⁺-N. When the temperature exceeded 600°C, the water in BCCP evaporated and decomposed violently, and the pore structure was deformed, the porosity decreased, and the adsorption capacity decreased.

Many materials were reported to adsorb NH₄⁺-N, such as slag, biochar, and coal slag balls. The NH₄⁺-N adsorption behavior of slag was found in either neutral or alkaline conditions with 3.1 mg g^–1 sorption capacity (Zhang et al., 2013). Vu et al. (2017) prepared biochar using corncob and the highest adsorption capacity was 22.6 mg g^–1. However, Kong et al. (2021) reported that the biochar prepared from distilled spirit achieved lees adsorption capacity (5.92 mg g^–1). Wang et al. (2016) prepared coal slag balls using modified coal slag and organic binder (PVA) and the NH₄⁺-N adsorption capacity was only 0.09 mg g^–1. The higher adsorption capacity of biochar depends on the large specific surface area and abundant adsorption sites (Li et al., 2018, 2019), and composition was also a key factor affecting the adsorption capacity. In this study, the high proportion of inorganic clay in BCCP resulted in small adsorption capacity.

Adsorption kinetics could be fitted by first-order kinetic model, second-order kinetic model, and intraparticle model, and all of them could well fit the NH₄⁺-N adsorption process by BCCPs with different calcined temperature (Table 3). Comparing with first-order kinetic and intraparticle diffusion models, the second-order kinetic model was more suitable for describing the NH₄⁺-N adsorption process by BCCPs, which was reflected by chemical adsorption processes including ion exchange among chemical bonds and adsorption process. Si–O–Si, –OH functional group on BCCP was involved in the reaction between chemical bonds during NH₄⁺-N adsorption process (Figure 3). Yan et al. (2018) used porous diatomite microsphere to adsorb methylene blue and also found that the adsorption process followed the pseudo-second-order kinetic model.

The adsorption equilibrium isotherm could reflect the distribution of adsorbate molecules in liquid and solid phases under an equilibrium state (Huang et al., 2020). Both Langmuir isotherm and Freundlich isotherm models could better fit NH₄⁺-N adsorption by BCCP prepared with different temperatures (Table 4). However, Freundlich model with R² = 0.941∼0.988 was more suitable than Langmuir model (R² = 0.880∼0.946), and the conclusion was coincident with Yan et al. (2018). Adsorption site energy distribution characteristic and curvature in the isotherm could be responded by n in Freundlich model (Huang et al., 2020). The value of n was between 1 and 10 in this study, which proved that all adsorption processes of BCCPs with different calcined temperatures were preferential adsorption.

The FTIR is an essential technique to qualitatively determine characteristic functional groups of the adsorbents (Figure 3). The peak of IR curved at 1,030 cm^–1 exhibited the introducing of Si–O–Si bonds on the BCCPs (Liu et al., 2012). The bands appearing below 1,100 cm^–1 might be attributed to Si–O stretching, Si–O–Si bending, Si–O–Al bending, and Si–O–Mg bending vibrations (Chen et al., 2017). For the otherwise typical bands, the intensity of the OH stretch at approximately 3,420 cm^–1 in the spectrum of clay was considerably larger than in the FTIR spectrum of the BCCP. The intensity of the CO₃^2– stretch at approximately 1,440 cm^–1 in the spectrum of BCCPs appeared depending on the addition of NaHCO₃.

BCCP, as a carrier for microbial immobilization, is an effective method to ensure that the efficient flora continue its degradation efficiency in natural water and avoid the risk of free bacteria being dispersed to reduce its pollutant degradation performance (Huang et al., 2020). For maintaining its degradation efficiency, immobilization method becomes the main control factor restricting pollutant transformation in microbial immobilization process. In this study, immobilization method was studied first and the results are shown in Figure 4A. The immobilization of Ochrobactrum sp. exhibited preferable nitrogen removal capacities when ammonium chloride was used as the sole nitrogen source. During the initial stage of the experiment (the first 24 h), the gel particles (SA-C and SA/PVA-C groups) adsorbed NH₄⁺-N from solution with high efficiency and their adsorption efficiencies were 22.59 and 29.59%, respectively, which were 10∼19% higher than that of the conclusion of Yan et al. (2020). Compared with two immobilization methods, on the contrary, the biodegradation performance of free Ochrobactrum sp. (FB group) at initial stage was very low (only 6.87% NH₄⁺-N was transformed in 24 h) and NH₄⁺-N removal efficiency reached 64.82% after 168 h. However, the value was 41.52 and 43.75% in SA-B and SA/PVA-B group, respectively, and they did not beat the FB group. Although the removal efficiency decreased, it could also be concluded that SA/PVA as immobilized material was more appropriate.

The fact that more nitrogen source and oxygen were obtained by free bacteria than immobilized bacteria prolonged the removal time by gel particle–immobilized bacteria (Yan et al., 2020). However, gel particles could provide stable micropores and protect cells from environmental changes and toxic substances (Hsieh et al., 2008; Hou et al., 2013). Zhang et al. (2021) prepared magnetic PVA–SA–diatomite composite carriers for immobilized microorganism and the highest NH₄⁺-N removal rate reached 72.5% at 12 h. It was found that the adsorptions of NH₄⁺-N by non-magnetic ingredients and Fe₃O₄ contributed 21.2 and 25.5%, respectively, and microorganism metabolism contributed 53.2%. Immobilized degrading bacteria in PVA–SA hydrogel bead was also reported to remove polycyclic aromatic hydrocarbons (PAHs) and the removal efficiency was around 77% in 96 h (Chen et al., 2021). Liu et al. (2019) investigated the effect of PVA–SA–cell cryogel bead–immobilized Bacillus sp. on the degradation of phenanthrene. The results indicated that the use of gel beads increased the number of adsorption sites to accelerate phenanthrene degradation.

On the basis of optimizing the immobilization method, BCCP was proposed as a carrier for Ochrobactrum sp. immobilization and the NH₄⁺-N removal efficiency is demonstrated in Figure 4B. The strong adsorption of BCCP and gel made NH₄⁺-N removal rate reach a high value on the first day. The efficiency of BC group (BCCP alone for adsorption as control group, 30.95%) was higher than that of SA/PVA-C group (29.94%), which reflected that the adsorption performance of gel particle was lower than BCCP and gel hindered the adsorption process of BCCP. Until 168 h, the removal efficiency was 34.35 and 31.55%, respectively, and little change was discovered during the process. On the contrary, the degradation efficiency of free Ochrobactrum sp. was dilatory and only 6.87% was achieved at the first 24 h, and increased to 84.82% at 168 h. This phenomenon reflected that biodegradation played its advantages. For SA/PVA-BC-B group, the degradation efficiency of NH₄⁺-N kept higher than free Ochrobactrum sp. during the reaction process and it was up to 79.39% at 168 h, which exceeded 14.57% than free Ochrobactrum sp. group.

Compared with SA/PVA for microbial immobilization without adding BCCP, the results showed that BCCP as carrier to immobilize Ochrobactrum sp. could improve its degradation efficiency for NH₄⁺-N. Because the porous structure of BCCP provided a larger surface area and a greater number of holes, it could store more substrate and promote microbial growth. The porous structure of BCCPs provides larger specific surface area and more pores, and can store more matrix and promote microbial growth (Chen et al., 2016).

Based on the aforementioned research results, the NH₄⁺-N degradation mechanism of immobilizing Ochrobactrum sp. with BCCP is demonstrated in Figure 5. The potential mechanisms were summarized as follows: the physical adsorption of gel and BCCP promoted NH₄⁺-N accumulation rapidly on the surface of BCCP, and provided more appropriate condition for microbial degradation. However, the gel covering on BCCP had a certain resistance for BCCP adsorption. Besides, the Ochrobactrum sp. embedded in gel could degrade the high concentration of NH₄⁺-N adsorbed on BCCP surface. The porosity of BCCP provides a necessary place for Ochrobactrum sp. growth and reproduction, and the adsorption driving force from BCCP promoted the biotransformation of Ochrobactrum sp.

Microbial immobilization can resist the adverse environment. Salinity, as an important factor, affected the growth of microorganisms and osmotic pressure of cell membrane. In high-salinity environment, the growth of microorganisms was inhibited (Wang et al., 2017). The effect results of salinity on NH₄⁺-N degradation are demonstrated in Figure 6A. For low salinity (lower than 5‰), free Ochrobactrum sp. group kept high NH₄⁺-N removal efficiency (89.37–90.43%) and the degradation performance decreased to 36.24% when the salinity was up to 35‰. The phenomenon reflected that the nitrification process of Ochrobactrum sp. was inhabited under high salinity condition. However, the NH₄⁺-N degradation efficiency was 69.32∼72.31% in 0 and 5‰ experiment groups, and the immobilization with BCCP produced a marked enhancement performance that displayed 12.47% higher than free Ochrobactrum sp. when the salinity increased to 35‰. Gao et al. (2020) found that immobilized materials owned a protective effect on bacteria in environments with high salinity and bacterial growth was inhibited when the salinity was higher than 15‰. Bacteria needed to obtain additional energy from the substrate to maintain cell activity in a high-salinity environment, and they could gradually adapt to high salinity environments over time (Moussa et al., 2005; Ge et al., 2019).

Temperature is another key factor affecting microbial growth and enzyme activity, and the low temperature resistance for immobilized microorganisms is shown in Figure 6B. The NH₄⁺-N removal efficiency of free or immobilized Ochrobactrum sp. groups increased with the temperature increasing from 15 to 30°C, while the efficiency decreased rapidly at 35°C. Compared with free bacteria group, microbial immobilization did not show its advantages, and its degradation efficiency was inferior to free Ochrobactrum sp. under different temperature conditions. For free bacteria group, 81% degradation efficiency was achieved at 30°C, which was 20% more than immobilized Ochrobactrum sp. group. For AOB, the optimal temperature is 30°C and bacteria grow perfectly at this temperature (Huang et al., 2017). When the temperature was lower than the optimal temperature, it affected the enzymatic reaction of cells and limited the growth rate of bacteria (Serra-Maia et al., 2016; Binnal and Babu, 2017; Huang et al., 2017; Manhaeghe et al., 2019). On the contrary, higher temperature could inactivate certain proteins in cell, reduced the activity of the microorganism, and even led to cell death (Ras et al., 2013; Serra-Maia et al., 2016; Nwoba et al., 2019). In this study, immobilized Ochrobactrum sp. did not play an effective role in resisting low temperature, but protected the Ochrobactrum sp. from the changes in salinity. The reason might be that as SA and PVA are the embedding materials of immobilized Ochrobactrum sp., the dense protective layer formed by them could buffer the salt concentration of microbial layer on the surface of BCCP, but could not block the impact of low temperature.