This research was carried out in April 2013 at Shenyang Agricultural University in Liaoning Province, China, as part of the Long-term stationary biochar experiment (40°48’N and 123°33’E). According to the FAO classification, the soil in the region belongs to the Alfisols profile (23). The site climate has been given in detail by Li et al. (24). Each treatment was repeated three times, the plot area was 25.5 m² (3.6 m×7.0 m), and all treatments were randomly arranged. The following were the basic physical and chemical properties of the initial soil (CK0): 6.0 soil pH, 9.9 g kg^–1 organic C, 0.9 g kg^–1 total nitrogen (N), 112.65 mg kg^–1 alkali-hydrolyzable N, 16.30 mg kg^–1 Olsen-P, 109.90 mg kg^–1available potassium (K), 1.25 g cm^-3 There were four treatments: CK (no fertilizers), NPK: urea (46% N) +superphosphate (5% P) +potassium chloride (42% K) fertilizers; CNPK: biochar +NPK; and SNPK: corn stover +NPK. As shown in all figures, CK was the control treatment from 2018, and CK0 was the initial soil from 2013. At a surface density of 3000 kg ha^–1 yr^–1, biochar was applied. At a surface density of 9000 kg ha^–1 yr^–1, corn stover was cut into 2–3 cm and applied, as well as corn stover from the harvested crop. The annual surface densities of N, P, and K fertilizers were 195 kg ha^–1, 39 kg ha^–1, and 62 kg ha^–1, respectively. Corn stover, biochar, and chemical fertilizer were applied annually (April 2013 to 2018) before sowing and spread on the soil surface before being tilled into a 0–20 cm soil layer by a rotavator. Corn was planted in April–May and harvested in September–October. The experimental field cropping system included one annual spring crop of corn in continuous cultivation using the cultivar mongolia6531, with a planting density of 60000 plants hm^–2. The biochar was produced from corn stover and heated at 450–600°C with a pH of 10.4, 490. 0 g kg^–1 total C, 14.4 g kg^–1 total N, 8.5 g kg^–1 total P, 14.08 g kg^–1 total K, 22.34 m² g^–1 specific surface area, 0.043 cm³ g^–1 pore volume, 7.12 nm pore size, and 33.5% ash content. The total N, P, K, and C contents of the stover were 0.96%, 0.32%, 0.72%, and 42.08%, respectively.

For all treatments, soil samples were taken from 0–20 cm depth after harvesting in October. Each plot had five randomly selected points where the soil was mixed as a soil sample, air-dried naturally, and sieved through 0.85 mm and 0.15 mm sieves.

The alkali diffusion method was used to determine the available N in the soil using 1.0 mol L^–1 of NaOH (25). Olsen-P was measured using the molybdenum-antimony anti-colorimetric method in 0.5 mol L^–1 of NaHCO₃ (26). The available K was determined with 1.0 mol L^–1 of NH₄OAc (Shanghai, China) using the flame photometric method (27). The pH of the water:soil mixture was 2.5:1, according to a Lei Magnetic PHS-3C type pH meter from China. The total P content was determined using the sodium hydroxide melting-molybdenum-barium colorimetric method (28). The C and N contents of the soil were determined using an elemental analyzer (Elemental III, Germany).

In the current study, the Hedley-P fractions were classified into seven soil P fractions based on previous research (29). (1) In 50 mL screw-cap centrifuge tubes, soil samples weighing about 0.5 g (0.15 mm) were collected. These samples were shaken for 16 h with 30 mL of deionized water and two saturated HC O 3 − anion resin membranes. The resin membranes were removed and shaken in a 0.5 mol L^–1 HCl solution for 16 h. The inorganic phosphorus, labeled Resin-P, was determined using the molybdenum blue method. (2) The soil residue was centrifuged in a centrifuge tube at 0°C for 10 min at 25000 g before being filtered and rinsed with 30 mL NaHCO₃. After shaking for 16 h, the solution was centrifuged for 10 min. The supernatant was separated into two parts: one labeled as NaHCO₃-Pi (NaHCO₃ extracted inorganic P, primarily adsorbed on the soil’s surface, and this part of the P was effective), and the other was frozen and centrifuged for further analysis. Ammonium sulfate was added to the other part and heated in an autoclave for 1 h to 121°C, yielding NaHCO₃-Po (NaHCO₃ extracted organic P, which was easy to mineralize and could be used by plants within a short period). (3) The soil was washed with 30 mL NaOH on the filter membrane in a centrifuge tube. After shaking for 16 h, the sample was centrifuged for 10 min and filtered. The supernatant was divided into two parts; to the first, 0.9 mol L^–1 of H₂SO₄ was added, which was then frozen and centrifuged for analysis and labeled as NaOH-Pi (inorganic P extracted with NaOH, chemically adsorbed to the soil Fe, Al compounds, and on the surface of clay particles). Ammonium sulfate was added to the other part, which was autoclaved for 1.5 h at 121°C and labeled as NaOH-Pt (NaOH extracted total P); NaOH-Po (organic P) = NaOH-Pt minus NaOH-Pi. (4) The soil was washed through a filter membrane in a centrifuge tube with 30 mL of 1 mol L^–1 HCl. After shaking for 16 h, the sample was centrifuged for 10 min and filtered. As HCl-Pi, the supernatant was labeled and analyzed. (5) Approximately 10 mL of concentrated HCl was added to the soil residual sample and heated in a water bath for 10 min at 80°C. After treating the sample with 5 mL of concentrated HCl, it was centrifuged for 10 min. The supernatant was tested and labeled with HCl-Pt (HCl extracted total P). HCl-P = HCl-Pt -HCl-Pi (6) Of concentrated H₂SO₄, 5 mL was added to the soil residue. The sample was treated with H₂O₂ and digested several times at 360°C. The sample was shaken, filtered, and set aside for a short period. The supernatant was referred to as Residual-P. Table S1 shows the recovery rate of the Hedley-P fraction.

The Swiss Bruker AVANCE III Bruker-500MHz nuclear magnetic resonance instrument installed at Jilin University and Shenyang Agriculture University was used to perform liquid-state ³¹P-NMR spectroscopy of soil samples (extracts) (30, 31). The soil sample was pretreated by passing approximately 3.00 g through a 2 mm sieve into a 100 mL centrifuge tube and treating it with a 60 mL mixture of 0.25 mol L^–1 NaOH and 0.05 mol L^–1 Na₂EDTA extractant at a water-soil ratio of 20:1. After mixing, the sample was shaken for 16 h at 20°C before being centrifuged (20°C, 10000 g, 20 min) and the supernatant filtered through a 0.45-micron filter membrane. A small amount of the extract, about 15 mL, was frozen. After dissolving the lyophilized sample in 1 mL of 0.25 mol L^–1 NaOH, it was centrifuged for 5 min (4°C, 10000 g). The supernatant was removed. Before being transferred to a 5 mm NMR tube for analysis, 0.6 mL of supernatant was treated with 0.05 mL of D₂O. After digesting 5 mL of the filtered extract with the H₂SO₄-HClO₄ mixture, the total extracted P concentration in the soils was determined using ICP-OES. The standard was orthophosphoric acid (85%) with a chemical shift of 6 ppm (32), and the other peaks were assigned to P forms based on the spiked experiment results and references (30, 33, 34). Although the spin-lattice relaxation times for these samples were not measured using ³¹P-NMR, the delay time was estimated using the P/(Fe + Mn) ratio for these extracts (6). Using the MestReC software, the nuclear magnetic resonance spectra were plotted, and the integrals and peak areas were calculated. Through integration, we also used MestReNova software (Mestrelab Research, S.L., Spain) to determine the relative proportion of each compound in the spectrum. We calculated the concentration of all compounds in each group from the total extracted P concentration and obtained relative proportion.

SPSS 21.0 (IBM Corp., Armonk, NY, USA) was used for the statistical analyses. The data are presented as arithmetic means with standard deviations. Before analysis, data were checked for variance homogeneity (Levene’s test) and residual normality (Shapiro-Wilk test) and transformed to log or square root if necessary. A one-way analysis of variance procedure was used to test the effect of different fertilization treatments on soil properties and NaOH-EDTA P extraction. To determine whether the differences in P forms between treatments were significant, the least significant difference (LSD) test was used (P <0.05). The correlations between the P fractions and P forms were investigated using Pearson’s correlation coefficients.

Several studies have found that some biochar contains a high concentration of base cations, which can cause a significant increase in soil pH (35, 36). In our soils, chemical fertilizers significantly decreased soil pH. The combination of corn stover (and its biochar) and chemical fertilizers application did not increase soil pH compared to the initial soil pH ( Figure 1 ). This meant that the soil pH remained unchanged, and the combination of corn stover (and its biochar) and chemical fertilizer application reduced the impact of chemical fertilizer to some extent. Soil total C increased significantly with the combination of biochar and chemical fertilizers application, followed by the combination of corn stover and chemical fertilizers application when compared to the control treatment ( Figure 1 ). Indeed, biochar was found to be more beneficial to soil C accumulation than corn stover application (37). The aging process influences biochar decomposition and mineralization, increasing soil organic C. (38). The study found that total N was lowest in control soils. Highest in soils treated with NPK ( Figure 1 ). The alkali-hydrolyzable N was reduced compared to the initial soil, whether fertilized or not. The minimum value was obtained in the control soils ( Figure 1 ). In comparison to the control soils, the combination of biochar and chemical fertilizers application slightly increased alkali-hydrolyzable N, indicating that N availability increases when biochar and chemical fertilizers are combined. These findings were consistent with those of Fiorentino et al. (39), who discovered that biochar could reduce inorganic N immobilization in natural soil and fertilizers by increasing microbial activity. Our research found that the highest C/N ratio was 11.5 in biochar application, ranging from 9.8 to 11.5 in all treatments. Other studies suggested that higher C/N ratios occurred in applied amounts of biochar as more stable C input to soils (40).

Higher total P levels were found after the soil was treated with a combination of biochar and chemical fertilizers, which coincided with high organic matter additions ( Figure 2 ), indicating that adding the combination of biochar and chemical fertilizers increased total P. Both biochar (41, 42) and corn stover (43, 44) are likely to be good sources of P. The plots with the highest increase in Olsen-P had a combination of corn stover, and chemical fertilizers applied. The second-largest had a combination of biochar and chemical fertilizers applied ( Figure 2 ). The Olsen-P level in the chemical fertilizer-added soils then increased to the level in the initial soils when compared to the control soil. These findings indicated that adding P to mineral and organic sources positively affected Olsen-P, which could be attributed to Olsen-P being directly transformed by labile organic P produced by additional mineral fertilizer or stover application (45, 46). Furthermore, biochar’s surface is rich in oxygen-containing functional groups. Its negative charge and complex pore structure endow it with a high cation exchange capacity and adsorption capacity (47). When biochar combined with chemical fertilizer, biochar could be used as a slow-release fertilizer carrier to delay the release of fertilizer nutrients. 5 years of Olsen-P changes that could be attributed to P were added at high rates (39 kg ha^–1 yr^–1).

Corn’s increased P uptake after the addition of corn stover (and its biochar) and chemical fertilizers was most likely due to an increase in soil organic matter and pH value ( Figures 1 , 2 ). In contrast to Barrow et al. (48), a pH of 5 may have promoted plant P uptake. This deviation from our results could be attributed to differences in soil parent material, soil P content, and crop variety. The highest yield was obtained in the combination of biochar and chemical fertilizers, and there was no significant difference between chemical fertilizers and the combination of corn stover and chemical fertilizers. Therefore, we propose that adding the combination of biochar and chemical fertilizers may be a more effective strategy for increasing soil total P, P uptake by corn, and yield; however, the effects of biochar on P uptake from soil P and plants remain unknown.

The combination of biochar and chemical fertilizers resulted in a significant increase in total Hedley-P fraction concentrations ( Figure 3 ). Resin-P increased more when biochar and chemical fertilizers were used together than when corn stover and chemical fertilizers were used together. It increased slightly in chemical fertilizer soils, as it did in corn stover and chemical fertilizer soils. This could be attributed to changes in soil properties and higher availability of P in soils. The current study found no significant differences between initial soil and control soils at NaHCO₃-Pi, indicating that five years of planting had no effect on NaHCO₃-Pi, and there were no significant differences between fertilized treatments at NaHCO₃-Pi ( Figure 3 ). NaOH-Pi followed a similar pattern to NaHCO₃-Pi, with the lowest levels found in control soils. Our findings revealed that the combination of corn stover (and its biochar) and chemical fertilizer application increased the content of inorganic P fractions. The organic P fractions (NaHCO₃-Po and NaOH-Po) of biochar with chemical fertilizers were significantly higher than that of the combination of corn stover and chemical fertilizers. This could be due to the distinct physical properties of biochar, which is also a rich source of organic P fractions than corn stover. On the other hand, it could be due to the interaction of chemical fertilizers and biochar/corn stover. NaOH-Po is commonly referred to as sorbed P, and biochar with a larger surface area has a greater adsorption capacity for sorbed P. Furthermore, biochar influences soil P adsorption capacity by influencing various factors, including soil pH, P content, cation content, and microbial activity. In all treatments, Resin-P followed the same pattern as NaHCO₃-Po. In this study, we found no significant effects of chemical fertilizers or the combination of corn stover and chemical fertilizers soils on Residual-P, but the opposite was true for HCl-P, which was lowest in the combination of biochar and chemical fertilizers soils and highest in chemical fertilizers soils ( Figure 3 ). In comparison to chemical fertilizers, the application of corn stover (biochar) and chemical fertilizers can increase soil available P content by changing soil pH. The findings of our study were also consistent with previous research (49, 50).

The solution ³¹P-NMR of NaOH-EDTA soil extracts collected from all treatments is depicted in Figure 4 . Adenosine monophosphate (AMP) and inositol hexakisphosphate were identified as the orthophosphate monoester peaks (IHP; Figure 5 ). Neo-IHP and D-chiro-IHP were found in the initial soil; additionally, neo-IHP was found in the combination of corn stover and chemical fertilizers soils, and D-chiro-IHP was found in the combination of biochar and chemical fertilizers soils. This could be related to their origins, as D-chiro-IHP can be derived from plants and microbes, whereas neo-IHP can be derived from microbes (24, 51).

Based on ³¹P-NMR, the combination of biochar and chemical fertilizer application significantly increased the contents of total inorganic P and orthophosphate compared to the other treatments ( Figure 6 ). This could be related to soil pH, organic matter, and total P. Furthermore, when compared to control soils, the use of chemical fertilizers significantly increased orthophosphate content. Our data showed that orthophosphate had a different trend than Olsen-P, which could be attributed to differences in soil properties. On the other hand, pyrophosphate exhibited the opposite trend as orthophosphate, with pyrophosphate in biochar soils being the lowest of all treatments. Pyrophosphate is produced by microbial activity in soils (52).

Total organic P based on ³¹P-NMR was higher in the SNPK and CNPK treatments than in the CK treatment ( Figure 6 ), indicating that the use of a combination of corn stover (and its biochar) and chemical fertilizers improved soil P retention when compared to the CK and NPK treatments. This increase in orthophosphate diester and total IHP corresponds to an increase in the Po/Pi ratio in the combination of corn stover and chemical fertilizer soils ( Figures 6 , 7 ). Organic P levels were highest in soils containing a combination of corn stover and chemical fertilizers; this component of IHP may be derived from plants (53). This indicated that organic P in soils containing corn stover and chemical fertilizers was mostly composed of orthophosphate diester and IHP. The majority of organic P was orthophosphate monoester (after diester degradation correction, the monoesters and diesters were corrected by using -glycerophosphate and -glycerophosphate as the diesters, as shown below; Figure 7 ). Furthermore, orthophosphate monoester and total IHP followed a similar pattern, with orthophosphate monoester being highest in soils containing corn stover (and biochar) and chemical fertilizers and lowest in control soils. Organic P increases due to the application of chemical fertilizers, and corn stover (and its biochar) are primarily composed of P microbial products such as the easily mineralizable orthophosphate monoester form AMP, stable total IHP (24) and orthophosphate diester (34, 54). The combination of corn stover (and its biochar) and chemical fertilizers, as well as the porous structure of the combination of corn stover (and its biochar), provided a habitable environment for soil microorganisms, particularly biochar. After applying the combination of corn stover (and its biochar) and chemical fertilizers, the initial soil had the highest value of orthophosphate diester (after diester degradation correction), followed by the combination of corn stover (and its biochar) and chemical fertilizers. Similarly, the highest level of DNA was found in the initial soil, followed by the combination of corn stover (and its biochar) and chemical fertilizers soils, but not in the control soil. This is because high orthophosphate diester or DNA levels indicate high microbial biomass, whereas control soils had the lowest soil organic C levels ( Figure 1 ).

Orthophosphate was the inorganic P fraction most positively correlated with Resin-P and P uptake by corn (P <0.05, Table 1 ). This could be due to the fact that orthophosphate and Resin-P were available P that plants could absorb in the soil (55, 56). The majority of the Hedley-P fractions (Resin-P, NaHCO₃-Pi, NaHCO₃-Po, NaOH-Po, and Residual-P) and corn P uptake were inversely related to pyrophosphate. Only AMP showed significantly positive correlations with NaHCO₃-Pi, NaOH-Pi, and P uptake by corn in all organic P forms (P <0.01). Resin-P, NaHCO₃-Pi, and NaOH-Pi were positively related to orthophosphate and AMP. In this study, P uptake by corn was found to be strongly related to AMP, NaHCO₃-Pi, and NaOH-Pi. This result suggested that AMP, NaHCO₃-Pi, and NaOH-Pi could be long-term P sources.