Factors driving the carbon mineralization priming effect in a sandy loam soil amended with different types of biochar

The effect of biochar on the soil carbon mineralization priming effect depends on the characteristics of the raw materials, production method and pyrolysis conditions. The goal of the present study is to evaluate the impact of three different types of biochar on physicochemical properties and CO2 emissions of a sandy loam soil. For this purpose, soil was amended with three different biochars (BI, BII and BIII) at a rate of 8 wt % and soil CO 2 emissions were measured for 45 days. BI is produced from a mixed wood sieving from wood chip production, BII from a mixture of paper sludge and wheat husks and BIII from sewage sludge. Cumulative CO2 emissions of biochars, soil and amended soil were well fit to a simple first-order kinetic model with correlation coefficients (r2) greater than 0.97. Results show a negative priming effect in the soil after addition of BI and a positive priming effect in the case of soil amended with BII and BIII. These results can be related to different biochar properties such as carbon content, carbon aromaticity, volatile matter, fixed carbon, easily oxidized organic carbon or metal and phenolic substance content in addition to surface biochar properties. Three biochars increased the values of soil field capacity and wilting point, while effects over pH and cation exchange capacity were not observed.


Introduction
Biochar is a carbonaceous material obtained from biomass pyrolysis or gasification processes.Biochar production emits carbon dioxide and other greenhouse gases, but combined with proper waste disposal or biofuel production it offers a practical way of mitigating global warming (Barrow, 2012).For many years now, it has been researched as a significant means of improving soil productivity, carbon storage, and the filtration of soil percolating water (Lehmann and Joseph, 2009).In fact, land degradation is a worldwide phenomenon that affects soil quality, water resources, human societies and economic development (Zhao et al., 2013;Omutu et al., 2014).Biochar as a source of organic matter (Paz Ferreiro and Fu, 2014) can improve the quality of soils in crop and rangeland (Yan-Gui et al., 2013) and then the development of societies (Srinivasarao et al., 2014) and reduce the impact of climate change (Barbera et al., 2013).
Nowadays, biochar production is attracting more attention because it is a safer method of organic waste management.Many types of biomass can be transformed into biochar, including wood wastes, crop residues, switch grass, wastewater sludge or deinking sludges (Méndez et al., 2012;Paz-Ferreiro et al., 2014;Sohi et al., 2010).If enough farmers, larger agricultural enterprises, biofuel producers, and waste treatment plants established biochar production methods, it P. Cely et al.: Factors driving the carbon mineralization priming effect could reduce CO 2 emissions related to agriculture while improving soil productivity.
Biochar is a highly recalcitrant organic material, with a long-term stability in soil, which is on the scale of millennia or longer (Kuzyakov et al., 2014).The response that soil exhibits to biochar addition has global consequences for carbon cycling.Depending on the interaction between soil and biochar, the ecosystem could become a sink or source of carbon.
The term priming effect refers to increases or decreases in the mineralization of native soil organic matter due to the addition of substrates and has been observed in many studies, both in the field and under laboratory conditions (Paz-Ferreiro et al., 2012;Zavalloni et al., 2009;Zimmerman et al., 2011).While it is generally regarded that biochar addition results in a reduction in soil carbon emissions from the soil, the fact is that the results are biochar and soil specific.Indeed, previous works have shown that there is no clear trend in CO 2 emissions after biochar application.For example, Zimmerman et al. (2011) found that carbon mineralization was generally less than expected (negative priming) for soils combined with biochars produced at high temperatures (525 and 650 • C) and from hard woods, whereas carbon mineralization was greater than expected (positive priming) for soils combined with biochars produced at low temperatures (250 and 400 • C) and from grasses, particularly during the early incubation stage and in soils of lower organic carbon content.Luo et al. (2011) used biochar from plant residues and found during the first 13 days of incubation experiment that biochar obtained at 350 • C causes a large positive priming effect, while biochar prepared at higher temperatures (700 • C) caused a relatively small positive priming effect.These authors hypothesized that the priming effect was probably caused by labile organic matter remaining in the biochar after pyrolysis, which in turn activated the soil microorganism.Jones et al. (2011) hypothesized that the increment in soil respiration is due to a different mechanism than changes in soil physical properties (bulk density, porosity, moisture); biological breakdown of organic carbon released from the biochar; abiotic release of inorganic carbon contained in the biochar and a stimulation of decomposition of soil organic matter.Zavalloni et al. (2011) have shown that the amount of soil carbon respired was similar between the control and soil treated with biochar from coppiced woodland pyrolysis in a short-term incubation experiment.Also, Wardle et al. (2008) reported a priming effect from a boreal soil after biochar addition, although the results of this experiment have been disputed by others (Lehmann and Sohi, 2008).If a strong positive priming effect occurs after biochar addition to the soil, then the beneficial effects attained by biochar addition to the soil becomes mitigated.Furthermore, although the use of biochar measuring soil respiration has been evaluated (Méndez et al., 2012;Zimmerman et al., 2011), fewer studies have studied the role of biochar addition of native soil organic matter (Zimmermann et al., 2011;Cross and Sohi, 2011;Gascó et al., 2012).For example, Gascó et al. (2012) observed using thermal methods that there is a degradation of more complex structures after application of a sewage sludge biochar to a Haplic Cambisol.The final chemical composition and physical properties of biochar, and thus its potential for having a positive or negative priming effect, depends on the characteristics of the raw materials, production method and pyrolysis conditions.Different studies has been performed in order to study the influence of feedstock, production method and pyrolisis temperature on biochar properties and uses (Calvelo Pereira et al., 2011;Méndez et al., 2012;Zimmermann et al., 2011;Paz-Ferreiro et al., 2014).
In the present work, three different biochars were used in order to study their influence on soil properties and CO 2 emissions.Biochars were obtained from pyrolysis of different types of biomass: mixed wood sievings from wood chip production, paper sludge and wheat husks and sewage sludge at temperatures between 500 and 620 • C using slow pyrolysis processes.

Soil selection and characterization
The selected soil was taken from the northeast of Toledo (Spain) and the soil was air-dried, crushed and sieved through a 2 mm mesh prior to analysis.The initial pH and electrical conductivity (EC) were determined with a soil : water ratio of 1 : 2.5 (g m L −1 ) using a Crison micro-pH 2000 (Thomas et al., 1996) and a Crison 222 conductivimeter (Rhoades, 1996) respectively.CEC was determined by NH 4 OAc/HOAc at pH 7.0 (Sumner and Miller, 1996).Total organic matter (TOM) was determined using the dry combustion method at 540 • C (Nelson and Sommers, 1996).Soil metal content was determined using a Perkin Elmer 2280 atomic absorption spectrophotometer after sample extraction by digestion with concentrated HCl/HNO 3 following method 3051a (USEPA, 1997).Soil texture was determined following the methodology of Bouyoucos (1962).These analyses were performed in triplicate.

Biochar characterization
Three different biochar samples were selected and used for the present work: biochar I (BI) was produced by Swiss Biochar (Lausanne, Switzerland) from mixed wood sievings from wood chip production at 620 • C; biochar II (BII) was produced by Sonnenerde (Austria) from a mixture of paper sludge and wheat husks at 500 • C; and biochar III (BIII) was produced by Pyreg (Germany) from sewage sludge at 600 • C. The pyrolisis duration was 20 min in all cases.All biochar samples were produced using Pyreg500-III pyrolysis (Germany) units that can work until 650 • C in a continuous process.
The pH, EC, CEC and metal content in biochars were performed as in Sect.2.1.Proximate analysis was determined by thermogravimetry using Labsys Setaram equipment.The sample was heated to a temperature of 600 • C under N 2 atmosphere and 30 • C min −1 heating rate.Humidity was calculated as the weight loss from the initial temperature to 150 • C. The volatile matter (VM) was determined as the weight loss from 150 • C to 600 • C under N 2 atmosphere.At this temperature, air atmosphere was introduced and fixed carbon (FC) was calculated as the weight produced when the final sample was burned.The ashes were determined as the final weight of the samples.The content in C, H, N and S was analyzed by an elemental microanalyzer LECO CHNS-932 and the oxygen content was determined by difference.Biochar nitrogen adsorption analysis to determine BET surface area was carried out at 77 K in a Micromeritics Tristar 3000.Also, biochar CO 2 adsorption analysis to determine both CO 2 micropore surface area and monolayer capacity was performed at 273 K in a ASAP 2020 V3.01.

Treatments and soil respiration
The selected soil (S) was amended with the three biochar samples at 8 wt % (S+BI, S+BII, S+BIII) and mixtures were incubated at constant temperature (28 ± 2 • C) and humidity (60 % FC) for 45 days.Additionally, it was studied if the application of the different amendments had an additive or synergistic effect in the soil (priming effect); in this way each biochar (BI, BII, BIII) was incubated individually in the experimental conditions.
Each sample (100 g) was introduced into a 1 L airtight jar and the CO 2 produced during incubation was collected in 50 mL of a 0.3 N NaOH solution, which was then titrated using 0.3 N HCl after the BaCl 2 precipitation of the carbonates.All treatments were performed in triplicate.
Organic carbon oxidized with dichromate from initial and final biochars was determined by the Walkley-Black method (Nelson and Sommers, 1996).
After incubation time, the next soil properties were determined: pH, EC, CEC, field capacity (FC), wilting point (WP) and available water (AW).pH, EC and CEC were determined as in Sect.2.1.Field capacity (FC) and wilting point (WP) were determined as the soil moisture content at 33 kPa (FC) and 1500 kPa (WP) (Richards, 1954).Available water (AW) was calculated as the difference between FC and WP.All analyses were performed in triplicate.
In addition, thermal analysis (TG, dTG and DTA) of soil was performed in a thermogravimetric equipment Labsys Setaram.About 50 mg of each sample were heated at 15 • C min −1 until 850 • C in the air atmosphere using a flow rate of 40 mL min −1 .

Mineralization model
The cumulative mineralization data were fitted to a first-order kinetic model, which is widely used to model soil respiration data (Méndez et al., 2013).The kinetic model used to calculate the evolved CO 2 -C soil is described as follows: where Y is the cumulative CO 2 -C (mg CO 2 -C 100 g −1 soil), t is the cumulative time of incubation (d), and C and m are the mineralization constants, with C • m representing the initial mineralization rate.The convexity shape of Y in this model is defined mainly by m, with m ≤ 1 and C ≥ 0. This equation was fitted to describe the C mineralization in S, the biochars (BI, BII and BIII) and the amended soils (S+BI, S+BII and S+BIII).The mineralization rate parameters of Eq. ( 1) were estimated by a non-linear model method, minimizing RMSA.
To quantify the priming effect of the three raw materials, the model was fitted to the experimental data (Experiment) and to the respiration data with the addition of 92 g of soil with 8 g of biochars (Addition).Also, C 10 was calculated as the evolved CO 2 -C after 10 days according to the model.

Results and discussion
Table 1 shows the main properties of the soil and three biochars.The soil texture was sandy loam, it had a slightly alkaline pH, the EC value indicated that the soil has no risk of salinization and soil organic matter content was 6.30 %.
With respect to biochars, BI and BII showed basic pH, whereas BIII had a pH value near 7. Proximate analysis of three biochar samples showed differences in their composition.The ash content of biochars followed the next sequence BIII > BII > BI depending on the feedstock, i.e., BI is prepared from woodchip, BII from paper sludge and wheat husk and BIII from sewage sludge presenting a higher mineral content.Indeed, BIII had the highest EC and metal content.Biochar metal content did not exceed the limit values for concentrations of metals in soil set up by the European Union (European Community, 1986), with BIII presenting the highest content, which can be explained according to its origin.All biochars presented a similar CEC, which can be related to the comparable temperature of preparation.The volatile matter content of BI and BIII was similar and lower than that of BII.Fixed carbon of BI was significantly higher than that of BII and BIII.Combining VM and FC, the ratio FC/(FC + VM) could be indicative of the carbon stability.According to this, BI was a very recalcitrant carbon material, whereas BIII showed the lowest ratio.The molar H / C ratio was used as an indicator of the degree of aromatization.This ratio shows the sequence BI < BII < BIII.The O / C ratio was indicative of the degree of carbonization following the same trend as the H / C ratio, BI < BII < BIII.According to previous studies on biochars (Kuhbusch and Crutzen, 1995;Hammes et al., 2006) the H / C ratio of ≤ 0.3 (like BI) indicates a highly condensed aromatic ring system, whereas the H / C ratio of ≥ 0.7 (like BIII) represents a non-condensed structure.
Table 2 shows the changes in pH, EC and CEC after the 45 days of the incubation experiment.Instead, biochar pHs were different (Table 1); pH did not change after biochar application, though BI and BII presented pH 2 units higher than soil.Conversely, other studies have shown pH increments after biochar application.For example, Méndez et al. (2012) observed a pH increment on a Haplic Cambisol after the ad- Values in column followed by the same letter are not significantly different (P = 0.05) using the Duncan test.
The number of replicates was 3 for each determination.
dition of sewage sludge-derived biochar, Kloss et al. (2014) described a slight increment of soil pH (0.3 units) in an acid soil after application of woodchip-derived biochar, and Jien and Wang (2013) observed a significant increase in Ultisol pH from 3.9 to 5.1 after addition of biochar made from the waste wood of white lead trees, so both biochar and soil composition influence the pH changes.Biochar addition slightly increased soil EC (Table 1), but the risk of salinization was negligible at the applied dose (USDA, 1999).The increase in soil EC is very common in soils treated with biochar prepared from sludge, which is the case for BII and BIII, as reported in other studies (Hossain et al., 2010or Méndez et al., 2012).
With respect to CEC, biochars did not increase soil CEC, a result according to previous works (Méndez et al., 2012) and which can be related to the low CEC of biochar with respect to soil OM (Lehmann, 2007).
Biochars increased the values of soil FC and WP following the sequence S < S+BIII < S+BI ≈ S+BII for both properties.Also, there was an increment in the AW when the soil was treated by BI and BII.This improvement in water retention is in accordance with the results previously obtained by Méndez et al. (2012), which found the same trend in a soil with a similar sand content treated with biochar prepared for sewage sludge at 600 • C. The higher increment of FC, WP and AW in S+BI and S+BII treatments could be related to the higher values of FC and WP of these biochars according to their high surface area and porosity (Table 1).
In the last years, thermal analysis has been proposed as an interesting technique in the characterization of organic matter stabilization processes.Additionally, it has been applied to soil characterization to assess proportions of labile and recalcitrant organic matter (Plante et al., 2009) and to study the evolution of organic matter in amended soils (Barriga et al., 2010;Gascó et al., 2012).Thermal analysis has the advantage of providing information about the chemical characteristics of soil organic matter without any extraction step as all samples were analyzed.Figure 1 shows dTG (Fig. 1a) and DTA (Fig. 1b) of S, S+BI, S+BII and S+BIII samples after the incubation period.Different peaks were observed in Fig. 1; at temperatures lower than 150 • C, water releases were observed, then at temperatures from 200 to 650 • C, oxidation of organic matter takes place.Initially, weight loss corresponds to less humified matter (from 200 to 400 • C), whereas the peaks observed at temperatures higher than 400 • C correspond to more humified organic matter.At temperatures higher than 550 • C weight loss could be attributed to refractory carbon from biochars and clay decomposition (Gascó et al., 2012).
From the DTA curve, the first endothermic peak could be observed at temperatures lower than 150 • C due to moisture release from soil samples.Then, two small exothermic peaks could be observed between 200 and 650 • C due to combustion reactions of soil organic matter.It is established that the first peak was associated with combustion of less humified organic matter, whilst the second one was related to the more humified matter.Four samples are shown at 573 • C, the characteristic small endothermic peak due to the quartz α-β inversion.Comparison of four samples in Fig. 1a and  b shows the influence of different biochars in soil organic matter composition.Biochar addition increases the amount of more humified or thermally stable organic matter following the sequence S+BI > S+BII > S+BIII.It was interesting to note that S+BIII shows a thermal behavior similar to that of unamended soil (S), indicating a similar organic matter composition to original soil.With respect to biochar CO 2 emissions, these were higher in BI while significant differences between BII and BIII were not found.This fact can be attributed to the elevated TOM of BI (87.71%) with respect to BII (59.90 %) and BIII (25.15 %).In order to explain the similar CO 2 emissions of BII and BIII other factors need to be accounted for (Jones et al., 2011).Calvelo Pereira et al. (2011) found that dichromate oxidation reflects the degree of biochar carbonization and could therefore be used to estimate the labile fraction of  3.60a Values in column followed by the same letter are not significantly different (P = 0.05) using the Duncan test.
The number of replicates was 3 for each determination.
carbon in biochar.Figure 2 shows that BIII, with the highest ash content and the lowest C content and consequently, expected lowest CO 2 emissions, has the highest content of labile, so the H / C and O / C ratios have shown that BIII has non-condensed organic structures.After incubation, the labile carbon of BI decreases, whereas that of BII and BIII slightly increases, indicating that some of the more stable organic structures were transformed into labile carbon.This result was in accordance with that obtained previously by Gascó et al. (2012) using thermal analysis and biochar form sewage sludge.However, for BI the labile carbon slightly decreases after incubation.
Table 5. CO 2 −C evolved (mg CO 2 100 g −1 dry weight) during incubation experiment and parameters estimated according to a simple first-order kinetic model to describe the C mineralization in soil (S), biochars (BI, BII, BIII) and amended soils (S+ BI, S+ BII, S+ BIII).Root mean square deviation (RMSD), correlation coefficient (r 2 ) and coefficient of determination (R 2 ) of the fitted model are shown.1998), and a pH value exceeding 10 can have negative effects on soil pH, but it must noted that only the application of larger amounts of biochar will lead to changes in a soil's pH value (Schmidt et al., 2012).With respect to electrical conductivity, limits have been fixed according to the limits fixed by Richards (1954) where the high value (4 dS m −1 , 25 • C) is the limit between normal and saline soils.The organic carbon limits have been fixed according to the International Biochar Initiative (2012) and the recommendations of Schmidt et al. (2012), who described that the organic carbon content of pyrolysed chars fluctuates between 10 % and 95 % of the dry mass, depending on the feedstock and process temperature used.With respect to volatile matter (VM) and fixed carbon (FC), values over 20 % and 40 % of VM and FC can be considered high according to biochar prepared from different feedstocks as sewage sludge (Gascó et al., 2012;Méndez et al., 2012), rice husk (Kalderis et al., 2014), euca-lyptus wood or poultry litter (Paz-Ferreiro, 2012;Lu et al., 2014).Finally, BET surface area values should be preferably higher than 150 m 2 g −1 (Schmidt et al., 2012), values over 750 m 2 g −1 being very high and of the same order as montmorillonite.It must stand out that the negative effects are usually due to a combination of different factors and cannot be attributed to a unique factor.Table 5 and Fig. 3 show the parameters estimated according to simple first-order kinetic model to describe the C mineralization in soil (S), biochars (BI, BII, BIII) and amended soils (S+BI, S+BII, S+BIII).The kinetics of CO 2 evolved from biochars was well fit to the proposed model, presenting r 2 values higher to 0.97.With respect to the amended soils, the fit presented a root mean square deviation (RSMD) lower than 2 and r 2 values higher than 0.99.In fact, this model of a simple first-order kinetic model has been successfully used to estimate CO 2 emissions from biochar and biochar amended soil in a short-term incubation experiment (Méndez et al., 2013).Also, results show that the application of BI had a negative priming effect if data of the experiment (57.1 mg C-CO 2 /100 g) and addition (63.0 mg C-CO 2 /100 g) are compared (Table 4), according with the similar values of model parameters (m and C).This result was in accordance with that obtained by Zimmerman et al. (2011), who found that biochars produced at high temperatures and from hard woods like BI show negative priming.With respect to the application of BII and BIII to soil, results showed a positive priming effect.It is interesting to note that both biochars increase their labile carbon content during individual incubation (Fig. 2), whereas for BI, their content slightly decreases.The initial organic matter mineralization was very similar in all cases (C parameter ranged from 6.07 to 7.91) according to Méndez et al. (2012), who found an increment of CO 2 emissions after application at the same rate after application of biochar prepared from sludge to a similar sandy soil or results obtained by Smith (2010).Nevertheless, Paz-Ferreiro et al. (2012) found a negative priming effect after sewage sludge biochar application (prepared at 650 • C) to an Umbrisol.Indeed, Zimmerman et al. (2011) concluded that discrepancies in C mineralization of biochar-treated soils are likely due to the type of both soil and biochar, the duration of the experiment and the dose of used biochar.

Substrate
Finally, the C 10 parameter, i.e. evolved CO 2 -C after 10 days according to the model, is related to the labile fraction of biochar to be released by microbial activity.Results show that experimental data were very similar, and the difference between experiment and addition (Table 4) in the case of S+BI could suggest a toxic effect of biochar.

Conclusions
The effect of biochar on the soil carbon mineralization priming effect depends on the characteristics of the raw materials, production method and pyrolysis conditions.Indeed, results show a negative priming effect in the soil after addition of BI (prepared at 620 • C from a mixed wood sieving from wood chip production) and a positive priming effect in the case of soil amended with BII (prepared at 500 • C from a mixture of paper sludge and wheat husks) and BIII (prepared at 600 • C from sewage sludge).These facts can be related to different biochar properties such as carbon content, carbon aromaticity, volatile matter, fixed carbon, easily oxidized organic carbon, metal and phenolic substance content and surface biochar properties.In addition, experimental results show that cumulative CO 2 emissions were well fit to a simple first-order kinetic model for the different biochar and amended soils.Also, biochars additionally improved water soil retention.Finally, further research is required to determine the importance of the different biochar properties involved in soil CO 2 emissions.

Figure 2 .Figure 2 .
Figure 2. Evolution of organic carbon oxidised with dichromate.Values in column followed by the same letter are not significantly different (P = 0.05) using Duncan test

Figure 3 .
Figure 3. Exponential model of measured C mineralized (as CO2) and that calculated by addition of soil and BI, BII and BIII effects.

Figure 3 .
Figure 3. Exponential model of measured C mineralized (as CO 2 ) and that calculated by the addition of soil and BI, BII and BIII effects.

Table 1 .
Main properties of the soil (S) and biochars.
a VM: Volatile matter.b FC: Fixed carbon.

Table 2 .
pH, electrical conductivity (EC), cation exchange capacity of treated soils after the incubation experiment.

Table 4 .
Influence of different biochar properties on the increment of soil CO 2 emissions after biochar application.
b : pH