Contribution to characterisation of biochar to estimate the labile fraction of carbon

Abstract

Different analytical techniques were used to find the most reliable and economic method for determining the labile fraction of C in biochar. Biochar was produced from pine, poplar and willow (PI, PO and WI, respectively) at two temperatures (400 and 550 °C) and characterised using spectroscopic techniques [solid state ¹³C nuclear magnetic resonance spectroscopy (NMR)], molecular markers [pyrolysis–gas chromatography–mass spectrometry (Py–GC–MS)], thermogravimetry (TG), elemental composition and wet oxidation (potassium permanganate and potassium dichromate). Short term incubation (110 h) of an A horizon from an Umbrisol amended with the biochar samples at two doses (7.5 and 15 t ha⁻¹) was also carried out to provide supplementary information on the influence of biochar–soil interaction on CO₂ evolution. Spectroscopic analysis demonstrated that the degree of biochar carbonisation was influenced by the type of feedstock and heating conditions and followed the order WI-400 < PI-400 ∼ WI-550 ∼ PO-400 < PO-550 < PI-550. The thermo-labile fraction of the biochar samples, estimated from TG, ranged between 21% and 49%. The fraction of total C oxidised with potassium permanganate (C_per/C_total) was <50 g kg⁻¹ in all cases, whereas potassium dichromate (C_dichro/C_total) oxidation efficiency ranged between 180 and 545 g kg⁻¹. For each type of feedstock, the highest values of either chemically or thermally degradable C corresponded to the biochar produced at low temperature. Results indicate that low cost methodologies, such as dichromate oxidation and TG, reflected the degree of biochar carbonisation, and could therefore be used to estimate the labile fraction of C in biochar.

Introduction

Biochar is pyrolysed organic material intended for use as a soil amendment to sustainably sequester C and concurrently improve soil function, while avoiding any adverse effects, on both the short and long terms (Lehmann and Joseph, 2009, Verheijen et al., 2009). Because of its condensed aromatic nature – especially when pyrolysed at high temperature – biochar is difficult for microorganisms to degrade and is therefore more stable than non-charred biomass (Paris et al., 2005, Lehmann et al., 2009, Shackley and Sohi, 2010), even though it is, over long periods, thermodynamically unstable under the oxidative conditions in most surface soils (Macías and Camps Arbestain, 2010). The relative stability of biochar determines the length of its contribution to the mitigation of greenhouse gas (GHG) emissions. However, its properties, including stability, depend not only on feedstock type and processing conditions (referred to as intrinsic recalcitrance) (Labbe et al., 2006, Nguyen and Lehmann, 2009), but also on the pedoclimatic conditions of the soil to which it was applied (Czimczik and Masiello, 2007).

Obviously, if biochar is to become accountable in trading C offset, robust protocols are needed to demonstrate and monitor its stability over time in the environment where it is deployed (Lehmann et al., 2009). In the context of biochar as C storage tool for mitigating GHG emissions and, based on the 100 yr time horizon used to calculate the global warming potential, as in the Kyoto Protocol (Forster et al., 2007), any amount of C stored from the atmosphere for at least 100 yr could be computed as a GHG mitigation benefit. Unfortunately, direct measurements in a timeframe of a century or longer make the approach unattainable. The stability of biochar should thus be evaluated in terms of the labile and stable fraction of C in the charred material (Lehmann et al., 2006, Lehmann et al., 2009, Hammes and Schmidt, 2009), the first being the fraction decomposed or lost with a turnover time ranging between weeks and decades, the second being the amount remaining in the soil C pool for centuries and millennia. While the mean residence time of the stable fraction might be difficult to determine, knowledge of the labile fraction under different pedoclimatic conditions should provide valuable information for future assessments of the proportion of stable C in biochar needed for trading C offset.

Recent studies that have focussed on characterisation of charred material – from which the labile fraction of C can be estimated – typically rely on the use of one of the following: (i) spectroscopic approaches, such as solid state nuclear magnetic resonance spectroscopy (solid state ¹³C NMR) (McBeath and Smernik, 2009) and Fourier transform infrared spectroscopy (FTIR) (Michel et al., 2009); (ii) thermal analysis (thermogravimetry, TG; de la Rosa et al., 2008); (iii) molecular markers by means of pyrolysis–gas chromatography–mass spectrometry (Py–GC–MS; Kaal et al., 2008a, Kaal et al., 2009) or the benzene polycarboxylic acid method (Brodowski et al., 2005); (iv) an array of chemo-thermal (Hammes et al., 2007) and chemical (e.g. potassium dichromate: Knicker et al., 2007; nitric acid: Marques Trompowsky et al., 2005) oxidation. Studies to determine the labile fraction of C in biochar under specific pedoclimatic conditions include incubation experiments based on CO₂ efflux (Kimetu and Lehmann, 2010, Bruun et al., 2011) and C isotope evolution (Bruun et al., 2008, Kuzyakov et al., 2009, Zimmerman et al., 2011). To our knowledge, however, no specific attempts have been made to look at relationships among the different estimates of the labile fraction of C in biochar.

The objectives of this study were (i) to characterise the C fraction in biochar produced from different feedstock types (pine, poplar and willow) at two temperatures (400 and 550 °C) using Py–GC–MS and solid state ¹³C NMR, and (ii) to compare the results with those from less sophisticated but more affordable techniques (TG, oxidation with different reagents) to find the most reliable and economic method for determining the intrinsic labile fraction of C in biochar. In addition, short term incubation using an acid soil with and without the addition of these biochar samples was carried out. These were not designed to address specifically the short term lability of C in biochar, as isotopes were not used, but rather to provide supplementary information on the influence of biochar–soil interaction on CO₂ evolution.