Description of soils

Composite soil samples were collected from the upper layer (0–15 cm) of two fields located near Quebec City, QC, Canada (47°N, 71°W). The soils were air-dried and sieved to 2 mm before incubation. Each soil was characterized for biochemical properties and reported in Manirakiza et al. (2019). Briefly, the Kamouraska clay was an Orthic Humic Gleysol from a corn (Zea mays L.) field. It had a pH_water of 5.32, a total C of 30.2 g·kg⁻¹, a Mehlich-3 P of 36 mg·kg⁻¹, a P/Al ratio of 0.028, and a total P of 1236 mg·kg⁻¹. The St-Antoine sandy loam was an Orthic Humo-Ferric Podzol from a field under alfalfa (Medicago sativa L.)/timothy (Phleum pratense L.) pasture. It had a pH_water of 5.89, a total C of 16.3 g·kg⁻¹, a Mehlich-3 P of 33 mg·kg⁻¹, a P/Al ratio of 0.030, and a total P of 446 mg·kg⁻¹. Based on soil P saturation, each soil was considered low to medium in regards to available P content (0.025 < P/Al_M-III < 0.050; CRAAQ 2010) and had a high capacity to retain P (Pellerin et al. 2006; Wang et al. 2015).

Material characterization

Two PB and one biochar were used in this study. The two PB consisted of mixed primary and secondary wastewater-treated sludge from thermomechanical pulp (PB1; Kruger, Trois-Rivières, QC, Canada) or from acid-treated bleached Kraft pulp (PB2; Kruger Wayagamack, Trois-Rivières, QC, Canada). The biochar consisted of pine chips pyrolyzed at 700 °C in a BEC Beta base unit (Biochar Engineering Corporation, Golden, CO, USA).

The main chemical properties of each material were reported in Table 1. Chemical analysis was provided in detail in Gagnon and Ziadi (2012) for PB and in Lévesque et al. (2018) for the biochar. The selected PB differed significantly in their properties, notably pH, total N and P, and C/N ratio, with PB2 showing a higher potential nutrient value. For its part, the biochar should have high stability against degradation in soils due to its high fixed C content and low volatile matter and H/C and O/C ratio which in counterpart should reduce its role in supplying major nutrients for crops (Domingues et al. 2017).

Table 1.

Main properties of selected paper mill biosolids (PB) and pine biochar (dry matter basis except moisture).

Incubation study

The experiment consisted of eight treatments (unfertilized control, mineral NP fertilizer, and two PB types × three biochar rates) for each soil arranged in a completely randomized block design with three replicates for each sampling date. Biosolids were thoroughly mixed with 100 g of air-dried soil at a rate of 2.5% (w/w), equivalent to 30 Mg wet·ha⁻¹ considering a depth incorporation of 10 cm and a soil bulk density of 1.2 g·cm⁻³. This rate is typical of the mean applied in field in Québec (Hébert 2016). The biochar was added at three rates (0%, 2.5%, and 5% w/w) which gave a field application equivalent to 0, 30, and 60 Mg dry·ha⁻¹ in the 10 cm surface layer, respectively. The highest rate was used to amplify any impact of biochar on the P dynamic of PB even if this rate exceeded the agronomic and economic acceptance (Dai et al. 2017). The mineral NP fertilizer treatment consisted of 30 kg KH₂PO₄-P·ha⁻¹ along with 120 kg NH₄NO₃-N·ha⁻¹ added to evaluate the relative P contribution of each treatment combination.

Every amended soil mixture was incubated in 500 mL Mason™ glass jars during a 32 wk incubation period with periodic destructive samplings. In this study, samplings of weeks 2 and 16 were used for soil P fractionation. These two sampling periods were chosen to simulate short-term (early root growth) and typical duration for a growing season in eastern Canada. Distilled water was added to adjust water-filled pore space to 60%. The lids of the jars were inverted to allow aeration while limiting the loss of moisture. The jars were incubated in the dark in a controlled environment chamber at 25 °C. Water loss was monitored twice a week by weighing and corrected as necessary.

Soil P fractionation

At each sampling date, subsamples of the amended soil mixture were air-dried, sieved to pass a 2 mm screen, and then ground to 0.2 mm. The Hedley sequential extraction procedure was performed as described by Tiessen and Moir (2008), with modifications for soil digestion as proposed by Zheng et al. (2001) and performed in the same laboratory by Zhang et al. (2020). Briefly, 0.5 g ground soil subsample was weighed into 50 mL centrifuge tubes and sequentially extracted according to the following scheme:

The concentration of Pi in extracts and digests was determined using the ascorbic acid molybdenum blue method (Murphy and Riley 1962). The concentrations of Po in each of these two extractions (NaHCO₃ and NaOH) were calculated as the difference between total P determined after oxidation with potassium persulfate and Pi. Total P in soils was also determined in the same way as for residual P with 0.1 g ground soil. This made it possible to evaluate the P recovery from fractionation which varied from 81% to 97% (91% ± 3%).

Total Pi was the sum of resin-P, NaHCO₃-Pi, NaOH-Pi, and HCl-Pi, whereas total Po was the sum of NaHCO₃-Po, NaOH-Po, and residual-P. Usually but not exclusively, resin-P is considered as freely available Pi; NaHCO₃-P is assigned to Pi sorbed on crystalline Al and Fe oxides and to easily mineralized Po; NaOH-P is assigned to Pi sorbed to amorphous Al and Fe oxides and to stable Po; and HCl-P is considered to be Ca-P-compounds (Negassa and Leinweber 2009). Usually, the sum of resin-P and NaHCO₃-P was considered as labile P, whereas HCl-P and residual P were considered as P stable over time.

Material P fractionation

Each material P was also fractionated before application using the same method as reported in Zhang et al. (2020). Briefly, fresh PB and biochar (0.150 g dry basis for P fractionation and 0.030 g dry basis for total P) were used for the extraction, as suggested by Ajiboye et al. (2004). The supernatant of the resin step was kept to determine both Pi and Po (after digestion) in the water extract. No Pi was found in the water extract. Organic P in 1.0 mol·L⁻¹ HCl was also determined as recommended by He et al. (2010) to provide a more accurate P characterization of the materials. Usually, this fraction is recovered in the residual P pool and has been simply assumed to be negligible. Finally, extracts of residual P and total P were digested twice to completely release Pi from materials. This step did not give more Pi in the case of biochar.

Statistical analysis

All data for soil P incubation were checked for normality with the Shapiro–Wilk’s test, and square-root transformation was needed for residual P at 16 wk to improve the normality of distribution. Treatment effects were evaluated as a factorial of two soils × eight treatments replicated three times using the MIXED procedure (SAS Institute Inc. 2004). Due to interaction effects with soil type, analysis was performed by soil, with replicates and replicates × treatments as random effects, treatments as fixed effects, and incubation time as repeated effect. Main treatment effects and their interactions were tested using differences of least squares means. Differences were considered statistically significant at P < 0.05.

Material P fractionation

The P recovery from the fractionation of the three materials was high, ranging from 93% to 115% (Table 2). Compared with biochar, the total P content in the PB was very high; more specifically, total Po accounted for about 50% of total P in PB2. This situation was common for several PB (Zhang et al. 2020) and was related to the fact that this material had been subjected to a secondary biological treatment for further purification and stabilization before discharge to the water bodies, which stimulated microbial activities. By contrast, PB1 comprised most of its P in inorganic form (72%) and could be more closely compared with municipal biosolids in which Pi is dominant (Sui et al. 1999; Ajiboye et al. 2004).

Table 2.

Phosphorus (P) fractionation of paper mill biosolids^a and pine biochar.

The relative contribution of P in each fraction varied widely with the material used (Table 2). The resin-P (Pi + H₂O-Po), the most readily available form, and the labile P (resin + NaHCO₃), which is considered plant available, accounted, respectively, for 59% and 72% of total P in PB2. In the PB1, labile P constituted only 26% of total P, whereas stable P (HCl + residual) accounted for 35% of total P. This could be attributed to its high pH, which may contribute to the formation of recalcitrant Ca-phosphate minerals and to the high total Al content (Table 1), which reduced the P availability (Krogstad et al. 2005; Torri et al. 2017).

Most P in biochar was in stable form (75%, HCl +residual), whereas the labile fraction accounted for only 18% of total P, with 13% as resin-P (Table 2). The method of producing biochar alters the forms of P in the original feedstock. The total P and the proportion of Pi fraction at the expense of Po generally increase with pyrolysis treatment, but the solubility of P is reduced, forming orthophosphate and immobilizing P into minerals or complexes with Al, Ca, Mg, and Fe ions (Dai et al. 2016; Xu et al. 2016; Zornoza et al. 2016; Li et al. 2018; Adhikari et al. 2019). Schneider and Haderlein (2016) reported that the HCl fraction, representing Ca-bound P in minerals of low solubility, might act as a reservoir of intrinsic P to slowly replenish labile P in acidic soils.

Soil P fractionation

The soil resin and labile P, two fractions representing plant available P, were strongly influenced by treatments on both soils at the two sampling dates: after 2 wk and 16 wk of incubation (Tables 3 and 4). The PB2, which was the richest material in total P (Table 1), increased most of these fractions. Indeed, PB2 supplied 56 mg total P·kg⁻¹, of which 59% was in resin form and 72% was in labile form (Table 2). The PB1 supplied less total P (31 mg·kg⁻¹), which was in more recalcitrant forms and also contained higher Al. As a consequence, PB1 contributed less to increasing soil resin- and labile-P fractions. In terms of mineral P fertilizer, the P recovery in labile form corresponded to 29% and 59% of total P applied for PB1 and PB2, respectively, at week 2, and 53% and 98% for PB1 and PB2 at week 16 (Table 5). This indicated that PB had a good potential for P mineralization. This agrees with our previous finding (Zhang et al. 2020) where we reported a recovery of labile P that increased with time and reached 48% for PB1 and 87% for PB2 at the end of a 16 wk incubation.

Table 3.

Effects of co-application of pine biochar and paper mill biosolids^a on the soil phosphorus (P) fractions of the Kamouraska clay.

Table 4.

Effects of co-application of pine biochar and paper mill biosolids^a on the soil phosphorus (P) fractions of the St-Antoine sandy loam.

Table 5.

Recovery of total applied phosphorus (P) as labile P form from paper mill biosolids^a and pine biochar addition average over soil types.

In contrast to PB, the pine chip biochar had little impact on the resin and labile P fractions on both soils (Tables 3 and 4). Despite the fact that this biochar supplied 11 and 22 kg total P·ha⁻¹ with the two application rates, it did not increase the content of these fractions compared with the PB (Table 5). This was related to its low total P content (0.4 g·kg⁻¹; Table 1) and very recalcitrant P forms (75% HCl + residual; Table 2). This finding is in agreement with results obtained by Manirakiza et al. (2020), who reported that co-applying biochar and PB did not change soil Mehlich-3 extractable P concentration compared with the application of PB alone. Several studies reported increases in available P following biochar addition to soil (Glaser and Lehr 2019), but those applying biochars of softwood or pine chips did not show any effect (Gaskin et al. 2010; Tammeorg et al. 2014; Backer et al. 2016; Foster et al. 2016). In addition, increasing pyrolysis temperature to >600 °C significantly reduced the effect that biochar might have on P availability (Zornoza et al. 2016; Glaser and Lehr 2019, Li et al. 2020).

The pine biochar, however, affected the moderately (NaOH) and stable (HCl + residual) fractions, particularly in the low total-P sandy loam (Table 4). The highest rate decreased the NaOH-Pi and -Po content and caused a conversion to more stable forms (HCl and residual). The effect took place earlier for the alkaline PB1 than for the acidic PB2 (Fig. 1). The P release from biochar is characterized by two mechanisms: (1) an instantaneous direct release and (2) a long-term slow release by alteration of soil pH, microbial mineralization, and co-precipitation with cations present in the soil (Xu et al. 2013; Gao and DeLuca 2016; Li et al. 2020). Qian and Jiang (2014) reported that a more severe pyrolysis process promoted the migration of P to the long-term available HCl-P pool. The direct contribution of pine biochar or induced increase in soil pH is less plausible here because the material contained low amounts of labile P (Table 2) and caused only a small increase in soil pH after 16 wk (Fig. 2) and during the entire incubation study (0.1–0.2 units; Manirakiza et al. 2020). Hence, the pine biochar may have a limited impact on crop P nutrition in the year of application but may serve as a reservoir to improve soil P availability over the long term. As reported in the literature, the increase in soil pH following biochar addition may also be caused by the presence of ash in the biochar (Glaser et al. 2002). Smider and Singh (2014) reported that applying 1.5% tomato green waste biochar (ash content = 562 g·kg⁻¹) increased soil pH by between 0.76 and 1.93 units. The ash content of the biochar used in our study was low (48 g·kg⁻¹; Table 1) and may explain the weak increase in soil pH and consequently the soil P release (Gagnon and Ziadi 2020).

Fig. 1.

Effect of co-application of pine biochar and paper mill biosolids on the soil NaOH-Pi fraction of the St-Antoine sandy loam. PB1, paper mill biosolids with a C/N ratio of 25; PB2, paper mill biosolids with a C/N ratio of 12.

Fig. 2.

Effect of co-application of pine biochar and paper mill biosolids on the soil pH after 16 wk of incubation. PB1, paper mill biosolids with a C/N ratio of 25; PB2, paper mill biosolids with a C/N ratio of 12.

Biochar can be co-applied with an organic material to increase the P availability to crops (Liu et al. 2012; El-Naggar et al. 2015). It can also alleviate loss of dissolved P in runoff and reduce nonpoint source pollution by sorbing soluble P (Laird et al. 2010). Biochars produced at high pyrolysis temperatures show more promise in this regard (Mukherjee and Zimmerman 2013; Yuan et al. 2016) because these conditions induced a larger surface area (Gul et al. 2015; Adhikari et al. 2019) and increased the Ca-bound P (Xu et al. 2016; Li et al. 2018; Adhikari et al. 2019), contributing to soil P sorption (Xu et al. 2014). However, Soinne et al. (2014) reported that biochar made of a mixture of softwood chips had very low affinity to sorb phosphate but can retain some P in high phosphate solution (Zhang et al. 2016). In our study, pine biochar seemed to have a limited impact on soil P retention apart from that attributed to the conversion of a portion of P associated with Al and Fe oxides to recalcitrant P pools, which are more slowly available in time.

Relationships between P fractions in materials and soils

Several attempts have been made to establish relationships between material P properties including fractionation and soil P availability. In this study, soil resin-P was highly and positively related to the amount of resin-P added by the different treatments (Fig. 3). It was the same for the labile P (Fig. 4), indicating that it is important to characterize these pools when applying biochars to agricultural soils. This was caused by the very high proportion of recalcitrant P forms in biochar total P, which did not exist to the same degree with uncharred materials (Xu et al. 2016; Li et al. 2018; Adhikari et al. 2019). Total P concentration and water-extractable P have been reported as best predicting P release from manures and PB (Zvomuya et al. 2006; Zhang et al. 2020).

Fig. 3.

Relationship between the amount of resin phosphorus (P) applied and the net increase (treated soil minus control) in soil resin P average over soil types and sampling dates.

Fig. 4.

Relationship between the amount of labile phosphorus (P) applied and the net increase (treated soil minus control) in soil labile P average over soil types and sampling dates.

In contrast, the proportion of P in stable form (HCl + residual fractions) can be used as an index of material vulnerability to P loss by runoff water (Li et al. 2018). This pool was positively correlated to soil stable P (Fig. 5). This means that pine biochar contributed actively, through its stability, to minimize the environment P risk following land application. Recently, Li et al. (2020) concluded, from an incubation study, that biochar produced from corn stalks released P slowly and could increase P use efficiency if used as a P fertilizer.

Fig. 5.

Relationship between the amount of stable phosphorus (P) applied and the net increase (treated soil minus control) in soil stable P average over soil types and sampling dates. White dot from NP treatment was excluded of the regression.