The biochar obtained from Takino Filter Inc. contains bacteria. The biochar was made from used wood pallet carbonized at 700˚C. The major pore distribution was from 1.6 to 1.7 nm. Microorganisms and their nutrient sources were impregnated in the charcoal. The charcoal was compression-molded into a cylinder with a diameter of 5 cm, height of 2 cm and specific surface area of 350 m²/g (Figure 1) [13]. The biochar was covered with a commercially available draining net and placed near the middle of a 1 L beaker. One liter of solution was stirred with a hot stirrer over 8 days at a water temperature of 35˚C and 400 rpm (Figure 1). Ten milliliters of treated solution were filtered through a membrane filter (pore diameter: 0.45 µm).

In this experiment, Gram staining was performed with reference to the modified Hucker method [19]. Cells were fixed in methanol for 1 to 2 min, completely dried and then 1 mL of crystal violet was added and stained for 1 min. Then, the crystal violet on the slide glass other than the fixed part was washed with distilled water, 1 mL of Lugol’s solution was added, and the stain solution was fixed to the cells. Then, after staining with 1 mL of safranin solution for 1 min, the mixture was washed with distilled water and the excess water was removed. A Nikon ECLIPSE E600 microscope was used for observation.

The preserved strain of Cronobacter was spread onto fresh bouillon medium and incubated at 35˚C. A single colony from the plate culture was inoculated into 5 mL of bouillon liquid medium and incubated in a shaker at a constant temperature of 35˚C while shaken at 120 rpm. One milliliter of this primary culture was inoculated into a 500 mL Erlenmeyer flask containing 250 mL of liquid bouillon medium and incubated in a shaker at a constant temperature of 35˚C while shaken at 120 rpm.

Aliquots of the culture solution from the Erlenmeyer flask were collected every 3 h to monitor culture growth. Absorption was measured on an ultraviolet-visible spectrophotometer at 660 nm and serially diluted with sterile 0.8% sodium chloride solution. One hundred microliter aliquots of the diluted solutions were spread plated onto bouillon agar plates and incubated at 35˚C. The number of colonies formed was used to determine the colony-forming units (CFU) of viable bacteria (CFU/ml).

Cronobacter strains were pre-streaked onto bouillon agar plates and then inoculated into test tubes containing 5 mL of bouillon liquid medium. The cells were incubated for 24 h in a shaker while shaken at 120 rpm at a constant temperature of 35˚C. One hundred microliters of this pre-culture was inoculated into ammonium sulfate liquid medium (Table 2) and incubated for 24 h in a shaker while shaken at 120 rpm at a constant temperature of 35˚C. The carbon sources used

were sugars D-glucose, fructose, galactose, xylose, maltose, and sucrose; and organic acids citric acid, succinic acid, acetic acid, lactic acid, malic acid, and sodium acetate. The culture solution was filtered with a 0.45 µm membrane filter prior to analysis.

The raw wastewater contained mean values of TOC and TN of 43.6 and 115.9 mg/L, respectively, with color and pH of 493 and 6.39, respectively (n = 3). Of the total TN, 98% was present as ammonia nitrogen. This result was based on the release of wastewater from the livestock farm into the waterway and dilution by the natural water flow.

The effect of charcoal was determined by monitoring the changes in TN and color when charcoal was added to the livestock wastewater and incubated for 10 days (Figure 3). The TN decreased from an initial concentration of 115.9 mg/L to 26.2 mg/L and the color decreased from 493 to 324 after 10 days. The charcoal treatment removed approximately 77.4% of TN from raw effluent. In this experiment, it was suspected that a biofilm was formed by microorganisms adhering to the surface of biological charcoal, and an aerobic layer was formed on the surface of the biofilm (Figure 4). Ammonium was mainly removed by assimilation into cells [20].

When biological charcoal was added to the synthetic wastewater and incubated for 20 days, the diurnal changes in NH 4 + , NO 2 − + NO 3 − , ORP and TOC were monitored (Figure 5).

The NH 4 + concentration decreased from 112 ± 4 mg/L to 46.2 mg/L by day 8, showing that approximately 60% of NH 4 + could be removed by biological charcoal treatment over 8 days. Because the initial pH of 8.02 showed little change to pH 8.34 ± 0.26 during the treatment period, it was considered that there was almost no effect of NH 4 + stripping that occurs under basic conditions of pH 9 or higher in synthetic wastewater [14]. In addition, low or non-detectable concentrations of NO 2 − (0 mg/L) and NO 3 − (0.13 ± 0.05 mg/L) were detected in the wastewater.

The ammonium concentration during ammonium utilization regulates ammonium assimilation, which results in improved aerobic denitrification efficiency [21].

The TOC hardly decreased and stayed between 243 and 298 mg/L after the second day, and ORP maintained anaerobic conditions from −53.5 mV to −70.5 mV.

A biofilm was formed on the surface of biological charcoal and it was considered that the bacteria within the biofilm secreted extracellular polysaccharides [21].

Because the detected NO 3 − concentration was relatively lower than the decrease in NH 4 + , it was considered that the decrease in TN was taken up by anabolic action. Kim et al. reported that a decrease in NH 4 + concentration and accompanying cell production by bacterial isolates occurred under conditions of 30% dissolved oxygen and carbon: nitrogen ratio of 8 [3]. In this study, the carbon: nitrogen ratio = 0.37 for actual wastewater and 2.67 for artificial wastewater. In this study, carbon sources were considered to be supplied from charcoal.

Table 4 shows the tentative identification of isolated microorganisms based on the color and shape of the isolated colonies. Table 5 shows the results of bacterial identification. Fragments of the 16S rRNA genes of YW_1-8, YW_9, and Y_9, which had different colony shapes and colors (Table 4) were amplified with sizes of 474 bp, 552 bp, and 516 bp for nucleotide sequence analysis. A BLAST search of these nucleotide sequences against the DNA Data Bank of Japan (DDBJ) showed 99% homology of YW_1-8 with Lysinibacillus fusiformis strain P103, 99% homology of YW_9 with Brevibacillus agri strain ST11, and 99% homology of YC_9 with Cronobacter sakazakii strain BDCSS041 (Table 5). Lysinibacillus and Brevibacillus are aerobic Gram-positive rods and are closely related to Bacillus subtilis, suggesting that these bacteria were supported on biological charcoal. In particular, the Y_9 strain, which showed high homology with Cronobacter, had the highest NH 4 + removal rate Therefore, it is likely that amino acids and proteins were synthesized in the Cronobacter sp. by assimilation of NH 4 + . In previous reported examples, NH 4 + treatment using nitrogen assimilation targeted NH 4 + concentrations in compost of 0.001 mg/L to 50 mg/L, and NH 4 + concentrations in agricultural wastewater of 100 mg/L or more. There are few reports of this process and no studies on nitrogen assimilation by Cronobacter [22]. Cronobacter is a facultative anaerobe Gram-negative bacillus that is widely distributed in the intestinal tract of animals and in the natural environment; it is known to form biofilms.

The Y_9 strain was confirmed as Gram-negative (Table 5) and showed high homology with the facultative anaerobe Cronobacter that was preferentially removed NH 4 + in the synthetic wastewater. The nitrogen metabolism of microorganisms includes nitrate denitrification and nitrogen assimilation. During nitrogen assimilation, cells synthesize material from NH 4 + and NO 3 − ; NO 2 − and NO 3 − are not generated and NH 4 + is removed from the environment. Therefore, it is likely that YW_9 and Y_9 absorbed NH 4 + from the medium and used it for cellular growth and activity [23] [24].

The ability of Cronobacter to remove NH 4 + was tested in media using different carbon sources at 1 g/L (D-glucose, fructose, galactose, xylose, maltose and sucrose; Figure 6). In ammonium sulfate liquid medium containing 103 ± 12 mg NH 4 + /L, growth of Cronobacter on D-glucose or galactose resulted in decreases in the NH 4 + concentration by 25.1 and 23.1 mg/L, respectively. Growth of the strain increased through to stationary phase concentrations of 43 × 10⁷ CFU/mL or more.

The strain was also tested in liquid medium culture containing different organic acids (citric acid, succinic acid, acetic acid, lactic acid, and malic acid; Figure 7). The decrease in NH 4 + concentration was higher during growth on organic acids than when compared with that during growth on sugars. The maximum NH 4 + removal and cell growth were obtained during growth on succinic acid, but even during growth on acetic acid, the NH 4 + removal was approximately 1.5 times that obtained during growth on glucose.