1 Introduction

Black carbon (BC) is a series of complex organic matter produced by the incomplete combustion of biomass and fossil fuels (Glaser and Birk 2012; Goldberg 1985). BC is believed to be highly stable and accumulates easily in the soil environment (Bond et al. 2013; Chen et al. 2022a; Czimczik and Masiello 2007; Masiello and Druffel 1998), which increases carbon sinks in ecosystems and mitigates climate change. Biochar is an artificially heated material that tends to become an increasingly important source of BC in the soil environment owing to its advantages in soil improvement, carbon sequestration, climate change mitigation, and pollution control (Dong et al. 2023; Guan et al. 2019; Shi et al. 2023; Wang et al. 2023a). In the past 10 years, the biochar industry has developed rapidly, with average biochar production rates of 30,000–50,000, ~ 50,000, > 20,000, and ~ 5000 t year−1 in China, the United States, Europe, and Australia, respectively (Garcia et al. 2022). Although biochar can theoretically increase soil BC content, a quantitative assessment of the increase in BC in soils with biochar addition in different environments is lacking. The lack of these data has become an important bottleneck limiting our ability to achieve soil carbon management through biochar application in the future.

One reason for the lack of relevant data is that BC is difficult to quantify and characterize. At present, BC analysis methods include pyrolysis (Meredith et al. 2012; Yang and Guo 2014), chemothermal oxidation (Wang et al. 2014), laevoglucose biomarker (Kuo et al. 2011), and benzene polycarboxylic acid (BPCA) molecular markers (Brodowski et al. 2005; Chen et al. 2022c; Fang et al. 2021; Glaser et al. 1998). Among them, the BPCA method destroys the fused ring aromatic structure through chemical oxidation to form a single small aromatic structural unit and then connects the carboxyl group at the bond-breaking position to form a series of BPCAs. This method is considered to be relatively accurate and suitable for the quantitative analysis of BC in the environment (Chang et al. 2020; Wiedemeier et al. 2015; Ziolkowski and Druffel 2010). In addition, the composition of BPCA monomers, including benzene tri- (B3CAs), tetra- (B4CAs), penta- (B5CAs), and hexa-carboxylic acids (B6CA), can provide information of the degree of aromatic condensation of BC (Brodowski et al. 2005; Dittmar 2008; Glaser et al. 1998). Application of the BPCA method allows the quantification of BC increases in soils with known amounts of biochar addition and can help infer the retention efficiencies and biogeochemical behavior of BC in soil.

Accompanying the formation of the condensed aromatic structure of BC in biochar manufacturing, polycyclic aromatic hydrocarbons—PAHs, such as naphthalene, phenanthrene, pyrene, benzo(a)pyrene, and benzo(g,h,i) pyrene, etc.) with 2 − 6 rings—are often yielded (Glaser et al. 1998; Goldberg 1985; Ziolkowski et al. 2011). Both BC and PAH are incomplete combustion products with high condensation and stability, and they often serve as molecular markers for thermal processes (Li et al. 2016; Lubecki et al. 2019). In addition, PAHs have received widespread attention owing to their carcinogenic, mutagenic, and teratogenic effects, which significantly threaten the health of humans and plants (Chen et al. 2022b; Mumtaz and George 1995; Wilcke 2000). Therefore, many studies have explored PAH content in biochar and biochar-amended soils, such as a continuous 5-year soil PAH investigation following 2 years of biochar addition in a vineyard (Rombola et al. 2019) and a 2.5 years of soil PAH investigation after a single biochar addition in barley- and oat-growing farmland (Kusmierz et al. 2016). These studies found that soil PAH contents were significantly increased after biochar application. However, once introduced into the environment, biochar undergoes aging over time, resulting in changes to its physical and chemical properties and, subsequently, alterations in soil properties (Liu et al. 2021). The changes in soil properties can be influenced by the biochar dosage, environmental factors, and treatment duration (Zhou et al. 2019). Therefore, it is important to quantitatively assess how the retention of BC and PAHs from biochar varies across different biochar dosages, soil environments, and treatment durations in long-term field experiments. In addition, although BC (which can be indicated by BPCAs) and PAHs are homologous substances from biochar, it remains unknown whether their occurrence and characteristics change synchronously over time and are in response to environmental change during biochar aging; it is of great significance to clarify the coupling or decoupling relationship between BPCAs and PAHs to evaluate the effectiveness of biochar application in increasing carbon sinks and regulating environmental risks in various environments.

To study the retention of BC and PAHs in soils with long-term biochar addition in different environments, we collected and studied the topsoil from yearly biochar addition experiments in two agro-ecological research stations in Ningxia Hui Autonomous Region (5 years of treatments; drier and cooler environment) and Shandong Province (7 and 11 years of treatments; wetter and hotter environment), China. We quantified the variations of BPCA and PAH contents and compositions with different cumulative biochar dosages under different environmental conditions, and further estimated the retention efficiencies of BC and PAHs in the topsoil. We hypothesized that: 1) the content of BC indicated by BPCAs (BCBPCA) and PAHs in soil would increase linearly with increasing cumulative biochar dosage, and the retention efficiencies of BCBPCA and PAHs would not change with the cumulative biochar dosage; 2) because the Shandong site has a higher mean annual temperature (MAT) and mean annual precipitation (MAP) and more sandy soil texture than the Ningxia site, the leaching loss of BC and the volatilization or leaching loss of PAHs would be greater at the Shandong site than at the Ningxia site, and thus the increases in BCBPCA and PAHs per unit biochar addition would be lower at the Shandong site than at the Ningxia site; and 3) because BC and PAHs have similar condensed aromatic structures, there would be significant correlations between the contents of BCBPCA and PAHs.

2 Materials and methods

2.1 Field experiments and sample collection

This study collected topsoil samples of 0−10 cm from two agro-ecological research stations in Ningxia and Shandong, China. The Qingtongxia Agro-ecological Research Station (106°11′35″E, 38°07′26″N) in Ningxia (hereafter, “Ningxia site”) was built in 2013. While the Huantai Agro-ecological Research Station (117°58′E, 36°05′N) in Shandong (hereafter, “Shandong site”) was built in 2007. The Ningxia site has a temperate continental semi-arid climate, with an annual average temperature of 8.9 °C, annual average precipitation of 193 mm (Sun et al. 2022), and annual sunshine duration of 3230 h. The soil type was Anthrosols based on the classification of the World Reference Base for Soil Resources (WRB 2022). Basic soil properties are shown in Table 1 (Sun et al. 2022). The nitrogen fertilizer addition amount was 0.25 t ha−1 year−1. The Ningxia site includes four treatments—biochar addition of 0, 4.5, 9.0, and 13.5 t ha−1 year−1—and each treatment has three parallel plots, each with an area of 65 m2 (13 × 5 m). Biochar was added to the soil surface and then plowed into topsoil. Rice was planted in May and harvested in September. The amount of irrigation water for farmland was approximately 1450 mm year−1 (i.e., 14,500 m3 ha−1 year−1) (Liu et al. 2021). Therefore, the total water input (precipitation plus irrigation) was approximately 1643 mm year−1. Farming and management practices were consistent among plots (Liu et al. 2021). A total of 12 composite soil samples (0−10 cm) from the 12 plots were collected in August 2018 (after 5 years of treatment). Each composite soil sample was a mixture of three random soil cores within a plot. Therefore, the cumulative biochar additions included 0 t ha−1 (no biochar addition for 5 years), 22.5 t ha−1 (4.5 t ha−1 year−1 for 5 years), 45 t ha−1 (9.0 t ha−1 year−1 for 5 years), and 67.5 t ha−1 (13.5 t ha−1 year−1 for 5 years); the corresponding sample types were noted as CKNX5, C22.5NX5, C45NX5, and C67.5NX5, respectively.

Table 1 Basic soil properties of Ningxia and Shandong sitesa
Full size table

The Shandong site has a temperate continental monsoon climate, with an annual average temperature of 12.4 °C, annual average precipitation of 600 mm, and annual sunshine hours of 2832.7 h (Zhang et al. 2013a, 2013b). The cropping mode was winter wheat-corn rotation. The soil type was also Anthrosols based on the classification of the World Reference Base for Soil Resources (WRB 2022). Basic soil properties are shown in Table 1 (Zhang et al. 2013a, 2013b). The nitrogen fertilizer addition amount in different treatment groups was 0.27 t ha−1 year−1. The Shandong site includes three treatments—biochar addition with 0, 4.5, and 9.0 t ha−1 year−1—and each treatment has three parallel plots, each with an area of 36 m2 (6 × 6 m). The amount of irrigation water in all plots was 225−400 mm year−1 (Xiao et al. 2019; Zhang et al. 2013a). Therefore, the precipitation plus irrigation was approximately 825−1000 mm year−1. Farming and management practices remained consistent among plots (Zou et al. 2023). Nine composite soil samples (0−10 cm) from nine plots were collected for both September 2014 (after 7 years of treatment) and September 2018 (after 11 years of treatment). The sample types collected in September 2014 included CKSD7 (no biochar addition for 7 years), C31.5SD7 (4.5 t ha−1 year−1 for 7 years), and C63SD7 (9.0 t ha−1 year−1 for 7 years), where the number after “C” represents the cumulative biochar dosage (in t ha−1). Similarly, sample types collected in September 2018 included CKSD11 (no addition for 11 years), C49.5SD11 (4.5 t ha−1 year−1 for 11 years), and C99SD11 (9.0 t ha−1 year−1 for 11 years). All composite soil samples from both Ningxia and Shandong sites were freeze-dried, gently crushed, passed through a 50-mesh sieve, and homogenized using a mortar and pestle. Subsequently, all soil samples were capable of passing through a 200-mesh sieve, confirming their thorough homogenization.

The added biochar (obtained from Dongxin Biochar Company, Shandong, with diameters of < 1 mm) for both sites was made from the incomplete combustion of crushed corncob at 360 °C for 24 h. The biochar was stored in cool and dry warehouses to minimize aging. The pH and specific surface area of the fresh biochar were 9.42 and 815 m2 g−1, respectively. The scanning electron microscopy (SEM) images indicated that the fresh biochar contained rich pore structures (Fig. S1). The carbon (C), nitrogen (N), and hydrogen (H) contents of biochar were 60.43 ± 0.58%, 1.28 ± 0.03%, and 2.40 ± 0.00%, respectively (Table S1). Similar to the soil samples, the fresh biochar samples were dried and homogenized using a mortar and pestle.

2.2 Elemental analysis

The contents of C, N, and H in the well-homogenized samples were analyzed using an Elementar Vario MICRO Cube (Germany). The atomic ratio of hydrogen to carbon (H/C) was calculated. The detailed data are shown in Table S1.

2.3 BPCA analysis

BPCAs in soil were analyzed according to the method of Vaezzadeh et al. (2021). We placed 50.0 mg of well-homogenized freeze-dried soil sample (or 25.0 mg of biochar) into glass ampoules, added 2 mL of 65% nitric acid (GR grade, Sinopharm), and sealed the ampoule. Then, we put the ampoule into a Teflon-lined reaction kettle. The reactor was sealed and heated in an oven at 170 °C for 8 h. After cooling to room temperature, the solution in the ampoule was dried under nitrogen. Dried samples were dissolved in ultrapure water and filtered through a 0.22 μm polytetrafluoroethylene membrane.

The BPCAs were detected by high-performance liquid chromatograph with photodiode-array detection (HPLC-DAD; U3000, ThermoFisher, USA) and separated on the Poroshell 120 SB-C18 column (100 × 4.6 mm, 2.7 μm; Agilent Technologies, Santa Clara, USA) under a mixed gradient of two mobile phases. Mobile phase A was phosphoric acid (Sigma Aldrich). Mobile phase B was HPLC-grade acetonitrile (Sigma Aldrich). The mixed standard included three benzene tricarboxylic acids (B3CAs, including hemimellitic acid, trimellitic acid, and trimesic acid), benzene tetracarboxylic acid (B4CA, pyromellitic acid), benzene pentacarboxylic acid (B5CA), and benzene hexacarboxylic acid (B6CA, mellitic acid), which were all purchased from Aladdin, Shanghai. Owing to the lack of the other two B4CA standards (prehnitic acid and mellophanic acid) on the market, their quantifications were performed according to the standard curve of their isomer (pyromellitic acid) following previous studies (Dittmar 2008; Vaezzadeh et al. 2021; Wiedemeier et al. 2016). The BCBPCA content was estimated as 2.27 times the carbon content of BPCAs (Glaser et al. 1998).

It has been reported that B3CA, B4CA, and B5CA are more likely to be generated after oxidation of aromatic structures at the edge of biochar, while B6CA can only be produced by the aromatic structure in the center of biochar macromolecules (Boot et al. 2015; Wiedemeier et al. 2015). Therefore, a higher proportion of B6CA in BPCAs indicates a higher aromatic condensation degree of BC (Boot et al. 2015; Wiedemeier et al. 2015). Furthermore, we calculated the average number of carboxylic groups for BPCAs (NCOOH, mol/mol) of soil based on Eq. 1 according to Sun et al. (2021). NCOOH was also used as an indicator of the aromatic condensation degree of BC; the larger the NCOOH value, the more condensed the aromatic structure contained in BC.

$$\it{N}_{\textrm{COOH}}=\frac{the\ amount\ of\ carboxylic\ groups\ of\ BPCAs\ \left( mol\ {g}^{-1}\right)}{the\ total\ amount\ of\ BPCAs\ \left( mol\ {g}^{-1}\right)\ }$$
(1)

2.4 PAH analysis

We added 10 g soil samples (or 2 g biochar samples) and a known amount of deuterated phenanthrene (PHE-d10) recovery standard (150 μL, 4 mg L−1) into fluorinated ethylene propylene tubes. Then, the PAHs in samples were extracted with n-hexane and acetone (volume ratio 1:1), and the combined extract was concentrated using a rotary evaporator. The concentrated sample was extracted with solid-phase extraction cartridge (Florisil PR, Anpel, China). Quantitative internal standards deuterated pyrene (PYR-d10) and deuterated perylene (PER-d12) were then added. Gas chromatograph-mass spectrometer (GC-MS, Agilent 7890B/5975) was used to detect the 18 PAHs, namely naphthalene (NAP), 1-methylnaphthalene (1-MeNAP), and 2-methylnaphthalene (2-MeNAP), acenaphthene (ACY), acenaphthylene (ACE), fluorene (FLU), anthracene (ANT), phenanthrene (PHE), fluoranthene (FLT), pyrene (PYR), benzo(a)anthracene (BaA), chrysene (CHR), benzo(b)fluoranthene (BbF), benzo(k)fluoranthene (BkF), benzo(a)pyrene (BaP), indene (1,2,3-c,d)pyrene (IcdP), benzo(a,h)anthracene (DBA), and benzo(g,h,i)pyrene (BghiP; Table S2). The standards used included the 18-PAH mixed standard, PYR-D10 and PER-D12 internal standard, and PHE-D10 recovery standard from Anpel Laboratory Technologies (Shanghai) Inc. Procedural and spiked blanks were analyzed every 10 samples for quality assurance and quality control. The average recovery of PAHs in spiked soil and biochar samples was 72%–89% (Table S5). The glassware was soaked in a 5% hydrochloric acid solution for more than 24 hours firstly and then ultrasonically cleaned with deionized water. At last, the glassware was baked in a muffle furnace at 450 °C for 4.5 hours to remove possible residual organic matter. For PAHs, we used the ratio between 5–6-ring and 2–4-ring PAHs [R(5–6/2–4)] as an indicator of the aromatic condensation degree of PAHs, where a larger R(5–6/2–4) indicated a greater aromatic condensation degree of PAHs (Sun et al. 2021).

2.5 Retention efficiencies of BCBPCA and PAHs in soils

The retention efficiencies of BCBPCA and PAHs in soil after biochar addition were estimated by quantifying the percentages of additional mass of BCBPCA or PAHs (i.e., detected mass difference between biochar-added and control plots) in the cumulative mass of BCBPCA or PAHs from the added biochar. Firstly, we calculated the mass of biochar-derived component i (mi_biochar; i = BCBPCA or PAH) entering 1 m2 of topsoil (0−10 cm) using Eq. 2:

$${m}_{i\_\textrm{biochar}}={c}_{i\_\textrm{biochar}}\times {m}_{\_\textrm{biochar}}$$
(2)

where ci_biochar is the content of i in biochar (in mg g-biochar−1 for BCBPCA and in ng g-biochar−1 for PAHs) and m_biochar is the mass of cumulative biochar (in g-biochar) added to 1 m2 of soil, which could be converted from the cumulative biochar dosage in t ha−1.

Based on the measured contents of BCBPCA and PAHs in the soil, the soil bulk density, and the soil depth, we also estimated the total masses of BCBPCA and PAHs (ng g−1) in soil of biochar-added and control plots based on the Eqs. 3−4:

$${m}_{i\_\textrm{treated}}={c}_{i\_\textrm{treated}}\times B{D}_{\textrm{treated}}\times {V}_{\textrm{treated}}$$
(3)
$${m}_{i\_\textrm{control}}={c}_{i\_\textrm{control}}\times B{D}_{\textrm{control}}\times {V}_{\textrm{control}}$$
(4)

where mi_treated and mi_control are the mass of i (i = BCBPCA or PAHs) in soil (1 m2 × 10 cm) of the biochar-added plot and control plot, respectively; ci_j is the concentration of i in soil (in mg g-soil−1 for BPCAs or in ng g-soil−1 for PAHs) of the j plot (j = treated or control); BDj is the soil bulk density (in g-soil cm-soil−3) in the j plot; and Vj is the volume of soil (in cm soil−3) with an area of 1 m2 and depth of 10 cm in the j plot.

Finally, the retention efficiencies of biochar-derived BCBPCA and PAHs in the topsoil (Ri; i = BCBPCA or PAHs) were estimated using Eq. 5 as follows:

$$Ri\approx \left({m}_{i_{\textrm{treated}}}-{m}_{i_{\textrm{control}}}\right)/{m}_{i\_\textrm{biochar}}\times 100\%$$
(5)

2.6 Health risk assessment

We used the incremental lifetime cancer risk (ILCR) model to estimate the cancer risk for adults and children exposed to PAHs in soils or pure biochar (Chen and Liao 2006; Peng et al. 2011; USEPA 1989). This model assumes human exposure to PAHs-contaminated soil through ingestion, dermal contact, and inhalation. The detailed estimation methods are described in Text S1 and Table S3. The United States Environmental Protection Agency has defined three risk levels based on ILCR values resulting from PAH exposure, including safe level (ILCR < 10−6), low-risk level (10−6 < ILCR < 10−4), and serious level (ILCR > 10−4) (USEPA 1989, 1996).

2.7 Data analyses

Experimental data were represented by the mean and standard error of three replicates. Spearman correlation was used to analyze the correlations between BCBPCA and PAH contents, NCOOH, and R(5–6/2–4). We used non-rotated principal components analysis (PCA) to study the relationships among parameters of soil BCBPCA and PAHs. Specifically, we normalized the raw data, established the correlation coefficient matrix between variables, calculated the eigenvalues and eigenvectors of the correlation coefficient matrix, and plotted out the principal components and comprehensive scores. Differences were considered statistically significant at the level of p < 0.05.

3 Results

3.1 Increased BC content with long-term biochar addition

The total content of BCBPCA in biochar was 211.11 ± 2.62 mg g−1, and it was dominated by B5CA and B6CA (Fig. S2, Table S4). The contents of BCBPCA and BPCA species linearly increased with cumulative biochar dosage (Fig. 1, Table S4). At the Ningxia site, the soil BCBPCA content was 0.73 ± 0.01 mg g−1 at CKNX5 (Figs. 1 and 2). After the application of biochar, the soil BCBPCA contents of C22.5NX5, C45NX5, and C67.5NX5 reached 1.21 ± 0.03, 1.14 ± 0.33, and 1.67 ± 0.21 mg g−1, respectively. Similar to the total BCBPCA, the contents of BPCA species all increased with increasing cumulative biochar dosage (Figs. 1 and 2). At the Shandong site, the soil BCBPCA contents in CKSD7 and CKSD11 were 0.48 ± 0.04 and 1.00 ± 0.34 mg g−1, respectively. After the application of biochar for 7 years, the contents of BCBPCA and BPCA species were 1.81 ± 1.09 mg g−1 in C31.5SD7 and 1.64 ± 0.88 mg g−1 in C63SD7. After the application of biochar for 11 years, the contents of BCBPCA and BPCA species were significantly higher in C49.5SD11 (3.02 ± 1.08 mg g−1) and C99SD11 (3.9 ± 1.03 mg g−1) than in CKSD11 (Fig. 1a). The fitting slopes of soil BCBPCA contents versus cumulative biochar dosage can reflect the increases of BC and BPCA species per mass unit biochar addition. The slopes of BCBPCA and BPCA monomers versus cumulative biochar dosage followed the order of SD11 > SD7 > NX5 (Fig. 1).

Fig. 1
figure 1

Contents of black carbon (BCBPCA; panel a) and benzene polycarboxylic acid (BPCA) species (panels b–e), and the average number of carboxylic groups for BPCAs (NCOOH; panel f) in soils (mean ± standard error; n = 3) from plots with different cumulative biochar dosages at the Ningxia site (5 years of treatments, NX5) and Shandong site (7 years of treatments, SD7; and 11 years of treatment, SD11). B3CA is the sum of hemimellitic acid, trimellitic acid, and trimesic acid; B4CA is the sum of prehnitic acid, mellophanic acid, and pyromellitic acid; B5CA is benzene pentacarboxylic acid; and B6CA is mellitic acid

Full size image
Fig. 2
figure 2

Relative abundances of benzene polycarboxylic acids (BPCAs) with B3CA, B4CA, B5CA, and B6CA in biochar and soils with different cumulative biochar dosages from the Ningxia site (5 years of treatments, NX5) and Shandong site (7 years of treatments, SD7; and 11 years of treatment, SD11). B3CA is the sum of hemimellitic acid, trimellitic acid, and trimesic acid; B4CA is the sum of prehnitic acid, mellophanic acid, and pyromellitic acid; B5CA is benzene pentacarboxylic acid; and B6CA is mellitic acid

Full size image

BPCAs were dominated by B6CA for the fresh biochar and all soil samples from both the Ningxia and Shandong sites. Interestingly, the proportion of B6CA was consistently lower in biochar-amended soils (i.e., mixtures of soils and biochar) than the two endmembers (i.e., non-biochar-amended control soils and fresh biochar). The average proportion of B6CA in BPCAs was higher in CKSD7 (46.0%) and CKSD11 (47.3%) than in CKNX5 (42.3%; Fig. 2). With increasing cumulative biochar dosage, the proportion of B6CA in BPCAs (proxy of aromatic condensation degree of BC) in soils continuously decreased for all sites (Fig. 2). This decrease was greater in SD7 and SD11 than in NX5, indicating that the increase in soil BC per unit amount of biochar addition was higher at the Shandong site than at the Ningxia site. Similarly, the slope of NCOOH versus cumulative biochar dosage followed the order of SD7 > SD11 > NX5, further indicating the lower aromatic condensation degree of BC at the Shandong site than at the Ningxia site.

3.2 Increased PAH content with long-term biochar addition

The PAH content of biochar was 3, 394.0 ng g−1 and was dominated by 2–3-ring species (Fig. S3, Table S5). The soil PAH content linearly increased with increasing cumulative biochar dosage (Fig. 3). Among PAH species, biochar addition induced increases in the order of 2-ring > 3-ring > 4-ring > 5–6-ring, where the content of 5–6-ring PAHs was not significantly increased for all sites (Fig. 3). This aligns with the fact that PAHs in fresh biochar was dominated by low-ring PAHs (2–3-ring) instead of high-ring PAHs (5–6-ring) (Fig. S3). At the Ningxia site, the soil PAH content in CKNX5 was 106.4 ± 2.9 ng g−1 (Fig. 3), with 4–6-ring species accounting for 62.3% (Fig. 4). With biochar addition, the soil PAH contents in C22.5NX5, C45NX5, and C67.5NX5 increased to 147.2 ± 16.1, 165.8 ± 14.6, and 193.6 ± 21.5 ng g−1, respectively. At the Shandong site, the soil PAH contents in CKSD1 and CKSD2 were 70.3 ± 14.4 and 95.3 ± 6.4 ng g−1, respectively. With 7 years of biochar addition, the soil PAH contents increased to 102.81 ± 9.28 and 127.31 ± 11.27 ng g−1 in C31.5SD7 and C63SD7, respectively (Fig. 3a). After 11 years of biochar addition, the soil PAH contents increased to 152.07 ± 2.20 and 184.50 ± 28.44 ng g−1 in C49.5SD11 and C99SD11, respectively (Fig. 3a).

Fig. 3
figure 3

Contents of 18 polycyclic aromatic hydrocarbons (Σ18PAHs; panel a) and polycyclic aromatic hydrocarbon (PAH) species with different ring numbers (panels b–e) and the ratio of 5–6-ring PAHs to 2–4-ring PAHs (R(5-–6/2-–4); panel f) in soils (mean ± standard error; n = 3) from plots with different cumulative biochar dosages at the Ningxia site (5 years of treatments, NX5) and Shandong site (7 years of treatments, SD7; and 11 years of treatment, SD11)

Full size image
Fig. 4
figure 4

Compositional profile of polycyclic aromatic hydrocarbons (PAHs) in biochar and soils from plots with different cumulative biochar dosages at the Ningxia site (5 years of treatments, NX5) and Shandong site (7 years of treatments, SD7; and 11 years of treatment, SD11)

Full size image

The contents of total PAHs and 2–4-ring PAHs in soils increased with increasing cumulative biochar dosage (Fig. 3a–e). The slope of soil PAH content versus cumulative biochar dosage followed the order of NX5 > SD7 > SD11, which was opposite to the ranking order for the regression slope of soil BPCAs content versus cumulative biochar dosage. This indicates that the increase in soil PAH per unit amount of biochar addition was lower at the Shandong site than at the Ningxia site. Similar to the indicator of aromatic condensation of BC, R(5–6/2–4), an indicator of the aromatic condensation of PAHs, decreased with increasing cumulative biochar dosage (Fig. 3f). The R(5–6/2–4) of PAHs at the Shandong site was higher than that at the Ningxia site (Fig. 3f), consistent with the lower content of 2–4-ring PAHs at the Shandong site than at the Ningxia site.

3.3 Retention efficiencies of BCBPCA and PAHs

With increasing cumulative biochar dosage, the retention efficiencies of BCBPCA and PAHs in soils decreased gradually in any given site (Table 2). That is, a small dosage of biochar addition increased soil BC and PAHs storage more effectively than a large amount of biochar addition. Comparing between sites, the retention efficiencies of BCBPCA were higher at the Shandong site than at the Ningxia site, whereas the retention efficiencies of PAHs were higher at the Ningxia site than at the Shandong site (Table 2).

Table 2 Estimated retention efficiencies of biochar-derived black carbon (BCBPCA) and polycyclic aromatic hydrocarbons (PAHs) in soils with different treatments at the Ningxia (NX) and Shandong (SD) sites
Full size table

3.4 Decoupling relationship between BCBPCA and PAHs

Principal component analysis (PCA) showed that the contents of BPCA species and the H/C ratio of soil had high loadings on the first principal component (PC1), while the contents of PAH species had high loadings on the second principal component (PC2; Fig. 5). Specifically, the contents of B3CA, B4CA, B5CA, and B6CA had positive loadings on PC1, and the H/C ratio of soil had negative loadings on PC1; the contents of 5- and 6-ring PAHs had positive loadings on PC2, and the contents of 2-, 3-, and 4-ring PAHs had negative loadings on PC2. Therefore, there was no significant correlation between the contents of BCBPCA and PAHs.

Fig. 5
figure 5

Principal component analysis of the contents and ratios of benzene polycarboxylic acid (BPCAs) and polycyclic aromatic hydrocarbon (PAHs) species for soils with different treatments from the Ningxia and Shandong sites. B3CA is the sum of hemimellitic acid, trimellitic acid, and trimesic acid; B4CA is the sum of prehnitic acid, mellophanic acid, and pyromellitic acid; B5CA is benzene pentacarboxylic acid; and B6CA is mellitic acid. The H/C is the atomic ratio of H to C in the sample

Full size image

Different samples were well separated on the score plot of the PCA (Fig. 5). The samples of the Shandong site mostly had positive scores on PC2, and those of the Ningxia site all had negative scores on PC2. This result aligns with the lower 2–4-ring PAHs at the Shandong site than at the Ningxia site. The sample scores on the PC1 generally increased with biochar addition, consistent with the increasing BCBPCA content with increasing cumulative biochar dosage for all sites. Furthermore, the sample scores on PC2 decreased with increasing cumulative biochar dosage for the Ningxia site but did not vary significantly with increasing accumulative biochar dosage for the Shandong site. This result visually indicates that biochar-derived 2–4-ring PAHs tend to have higher retention at the Ningxia site than at the Shandong site, which is consistent with the previous results (Figs. 3 and 4).

4 Discussion

4.1 Dosage-dependent retention efficiencies of BCBPCA and PAHs

Consistent with our first hypothesis, the contents of BCBPCA and PAHs both linearly increased with cumulative biochar dosage. This result is similar to previously reported findings that biochar addition increased soil PAH content (Cornelissen et al. 2005; Koelmans et al. 2006). However, the retention efficiencies of BPCAs and PAHs in soils decreased gradually in any given site (Table 2). This does not support our first hypothesis and indicates that a small dosage of biochar addition can increase soil BC and PAHs storage more effectively than a large amount of biochar addition. We noted that the decrease in aromatic condensation degree of BC (indicated by results of the proportion of B6CA in total BPCAs and NCOOH) was greater with increasing accumulative biochar dosage, which implies that BC might have experienced greater oxidation when the biochar dosage was higher. This may be because the application of more biochar will result in more macropores and better soil drainage and aeration (Zhou et al. 2019), which facilitates BC oxidation. However, such speculation awaits further verification.

4.2 Site-dependent retention efficiencies of BCBPCA and PAHs

Consistent with our second hypothesis, the retention efficiencies of PAHs from biochar were lower at the Shandong site than at the Ningxia site. Several factors may contribute to this difference. First, the 7 and 11 years of annual biochar addition at the Shandong site (average biochar ages of 3.5 years and 5.5 years, respectively) were longer than the 5 years of biochar addition at the Ningxia site (average biochar age of 2.5 years). The longer aging time of biochar should allow for greater PAH loss and lower PAH retention at the Shandong site compared with the Ningxia site. Second, the higher MAT at the Shandong site could lead to greater volatilization loss of PAHs, particularly low-ring PAHs with higher vapor pressures (Park et al. 1990). Third, compared to the Ningxia site, the Shandong site had much lower clay content (2.3% at Shandong vs. 18.2% at Ningxia) and organic carbon content (10.8 g kg−1 at Shandong vs. 16.1 g kg−1 at Ningxia) and thus much weaker adsorption of the hydrophobic PAHs (Krauss and Wilcke 2002; Kusmierz et al. 2016; Oleszczuk 2006; Reemtsma and Mehrtens 1997). Therefore, the differences in soil texture and organic carbon also support the lower retention efficiencies of PAHs at the Shandong site than at the Ningxia site. However, despite the Ningxia site receiving more precipitation plus irrigation (1643 mm year−1) than the Shandong site (825–1000 mm year−1), facilitating PAH leaching, this did not align with the observed higher PAH retention at the Ningxia site. It appears that lower MAT and the stronger adsorption capacity of the clay and organic carbon content in Ningxia overrode the effect of increased precipitation and irrigation.

In contrast to the PAHs, BCBPCA from biochar generally had higher retention efficiencies at the Shandong site than at the Ningxia site. This is contrary to our second hypothesis and cannot be explained by the longer treatment duration, higher MAT and MAP, or more sandy texture at the Shandong site, all of which should accelerate BC loss. However, our original hypothesis has omitted the potential impact of irrigation. The rice grown at the Ningxia site requires more irrigation water than the wheat and corn grown at the Shandong site, and the water from precipitation and irrigation was thus much higher at the Ningxia site than at the Shandong site. We speculated that a higher amount of precipitation plus irrigation was the main factor in the lower BC retention efficiencies at the Ningxia site than at the Shandong site. Therefore, a greater amount of BC is expected to be leached to subsoils and deep soils at the Ningxia site than at the Shandong site, resulting in lower BC retention in topsoil at the Ningxia site. Additionally, the Shandong site has much lower sunshine intensity than the Ningxia site (Feng et al. 2021; Tang et al. 2022). Despite that photolysis is generally considered as a relatively minor pathway for carbon loss in soil, the relatively weaker photolysis of soil BC might also facilitate the higher BC retention at the Shandong site.

4.3 Geochemical and environmental implications

Based on field experiments, we quantified the effects of long-term biochar addition on the storage of BCBPCA and PAHs in farmland soils. Based on the results of slope fitting for BCBPCA and PAHs versus cumulative biochar dosage, a ton of biochar added to soil was estimated to increase BCBPCA and PAHs in the topsoil by 47.40 ± 11.71 kg and 1.71 ± 0.11 g, respectively. With increasing cumulative biochar dosage, the contents of BCBPCA and PAHs increased but their retention efficiencies decreased. As such, low-dose biochar addition may be more efficient for increasing BC content than high-dose biochar addition in soil, and future biochar application can optimize soil carbon sequestration by adjusting the amount of biochar addition. We found that the average retention efficiencies of soil BC were consistently lower than 50%, which may be because a considerable part of biochar colloidal particles migrated to the subsoil or even deep soil (Wang et al. 2023b). Furthermore, the proportion of B6CA in the BPCAs was consistently lower in biochar-amended soils than in the two endmembers (i.e., non-biochar-amended soils and fresh biochar). This suggests that it was not a conservation mixing, and a significant amount of BC with high aromatic condensation degrees was degraded into BC with low aromatic condensation degrees. Based on the same sites, our previous study found that dissolved BC did not increase with 5 years of biochar addition but increased significantly with 11 years of biochar addition, which further supported the degradation of BC into soluble compounds with long-term biochar aging (Cai et al. 2020).

The large number of 2–4-ring PAHs carried by biochar significantly increased the contents of 2–4-ring PAHs in soil but had no significant effect on high-ring PAHs. Although BC derived from biochar has homology with PAHs, their contents in soils were decoupled and did not support our third hypothesis. Such results indicate that they have undergone different biogeochemical processes and different fates. Furthermore, the retention efficiencies of PAHs in all plots were higher than 60%, much higher than that of BCBPCA with a larger molecular structure (< 50%). This might be because during the aging process of biochar, non-extractable PAHs that were physically encapsulated or strongly adsorbed became extractable, resulting in higher retention efficiencies for PAHs than for BCBPCA. Fortunately, even after 11 years of biochar addition that resulted in a significant increase in soil PAH contents, the ILCR of soil PAH exposure still did not exceed 10−9 (Table S6); that is, the soil PAH content was not high enough to cause considerable health risks to humans through long-term soil exposure. However, PAHs also have ecological risks, and the potential impact of increased PAH content on soil animals, plants, and microorganisms might not be ignored. Notably, we found that the retention of BCBPCA and PAHs had distinct responses to environmental factors. That is, with higher MAT, more sandy soil, lower water inputs from precipitation plus irrigation, and lower sunlight intensity, the Shandong site had lower PAH retention but higher BCBPCA retention than the Ningxia site. In addition, the weak correlation between BPCA and PAHs further suggests that they have distinct environmental behaviors and fates; as such, extractable PAHs cannot be simply used as markers to track bulk BC.

5 Conclusions

Based on long-term field experiments, we investigated the variations of BC (evaluated using benzene polycarboxylic acids, BPCAs) and PAH contents and compositions with varying cumulative biochar dosages under different environmental conditions, and further estimated the retention efficiencies of BC and PAHs in the topsoil. The results showed that increasing cumulative biochar dosage causes elevated contents of BC and PAHs, accompanied by decreases in their retention efficiencies. The retention efficiencies were also highly site-dependent, with the Shandong site characterized by higher retention efficiencies of BPCAs and lower retention efficiencies of PAHs than the Ningxia site. The lower retention efficiencies of PAHs in the topsoil at the Shandong site than at the Ningxia site can be attributed to the higher temperature and more sandy soil texture in the Shandong site, which promote the leaching and volatilization losses of PAHs. However, the higher retention efficiencies of BC at the Shandong site could not be explained by these factors. Instead, the greater irrigation and subsequently greater BC leaching loss at the Shandong site, compared to the Ningxia site, are more likely responsible for the higher BC retention efficiencies in the topsoil. Despite both BC and PAHs originating from biochar and sharing similar condensed aromatic structures, their contents showed no significant correlation, indicating distinct behaviors and fates between BC and PAHs. Our findings emphasize the importance of optimizing biochar addition dosages and considering site-specific environmental factors for effective soil BC sequestration through biochar application. For example, given the same amount of biochar, applying a smaller dosage of biochar over a larger soil area might lead to a greater accumulation of BC in the topsoil. This study is the first to quantify the potential retention efficiencies of soil BC and PAHs under the influence of long-term biochar addition, providing valuable insights for the application of biochar in soil carbon sequestration in the future. Nevertheless, the possible impacts of different crop types and soil microbial communities and activities on BC and PAH retention remain unclear and await investigation in future research. Future research should also focus on the long-term carbon increase efficiencies of different types of biochar in various geographical regions or ecosystems, in situ biogeochemical transformation of BC, and the migration of BC into the subsoil and deep soil.