Abstract

Introduction

In January 2003 fire burnt approximately 1.5 million hectares of alpine and subalpine forest, grassland, bogs and farmland in southeast Australia. The fire was both extensive and intense and has impacted on fragile environments in a large water catchment area. This study focuses on the characterization of post-fire residues in order to assess possible impacts of these residues for soil and water.

The charred material, often referred to as black carbon (BC), consists of the organic constituents of plant biomass that are released to the surrounding environment during combustion (Goldberg 1985; Haumaier and Zech 1995; Schmidt and Noack 2000; Baldock and Smernik 2002; Mitra et al. 2002), as well as pyrogenic substances (substances produced under conditions involving intense heat) that are formed by combustion of the biomass. BC can be thought of as a continuum, from partly charred plant material, through char and charcoal to graphite and soot particles recondensed from the gas phase (Schmidt and Noack 2000; Mitra et al. 2002). The amount and composition of BC formed during biomass burning is controlled by a large number of variables related to the quantity, heterogeneity, and condition of combustible material present (Baldock and Smernik 2002). The nature of the fuel (amount, disposition, water content), the weather conditions during burning (wind, relative humidity, etc.), and the rate of rise and duration of the highest temperature all contribute to the type and amount of BC formed (Orioli and Curvetto 1978).

Chars are ubiquitous and persistent in atmospheric and aquatic particulate matter, soils, sediments and geological samples worldwide (Fernandes et al. 2003). High levels of charcoal carbon resulting from repeated historical burning of grasslands, open woodlands, and agricultural crop residues have been reported in Australian soils. For example, Skjemstad et al.(2002) reported that charcoal generated by fires can constitute up to 8 g C kg^-1 soil and represent up to 30% of the total organic carbon of Australian soils. Charred particles from biomass burning possess a relative lack of (bio)chemical reactivity and thereby strongly resist decomposition over a geological timescale (Goldberg 1985; Schmidt and Noack 2000).

In recent years, geochemical and biological studies of different forms of BC, such as plant chars, charcoal, and soots, have received increasing attention owing to their potential importance in a wide range of biogeochemical processes (Schmidt and Noack 2000). As examples, BC may represent a significant sink in the global carbon cycle. BC can affect the Earth’s radiative heat balance, be a useful tracer for Earth’s fire history, add significantly to the fraction of carbon buried in soils and sediments, and act as an important carrier of organic pollutants or heavy metals (Schmidt and Noack 2000).

The possible significance of carbonised materials, such as BC, for the formation of humic materials has been addressed by Kumada (1983), who suggested that the weathering of BC under natural conditions, or oxidative depolymerisation under the influence of oxygen and moisture resulted in the formation of humic and fulvic acids. These humic and fulvic acids could then be further transformed into lower molecular weight compounds and finally into carbon dioxide, methane and water. Shindo (1991) found that charred plant residues could produce large amounts of humic acid when treated with nitric acid, and proposed that similar humic acids form when residues are subjected to oxidative polymerisation under the influence of oxygen and moisture during a long period of time after burning.

Methods

Sampling

Samples of bushfire residues and soil from the northeastern Victorian high country were collected in late February 2003. All samples were collected using random stratified sampling. The sampling design comprised three separate 10 m x 10 m grids, each situated in a different location within a 100 m x 100 m sampling area. 10 sub-samples were collected from each grid and combined. Ash samples were collected by scraping residues from the ground using a small digging tool. Charcoal samples were collected by handpicking the residue from the ground. On collection, the ash and charcoal samples were air-dried at 35 °C and placed in air-tight plastic bags. The Global Positioning System (GPS) coordinates of the sample site were Australian map grid 84, zone 55, E516457 and N5933922, with elevation approximately 450 m above sea level.

Characterisation

The organic matter content of the bushfire residues was determined by the loss in weight after oven dried samples were ashed at 800^oC for 18 h. in a muffle furnace. BET surface areas and porosity were measured on six replicates with a Micromeritics Accelerated Surface Area and Porosity (ASAP 2000) system (Micromeritics Instrument Corp.). Scanning Electron Microscopy (SEM) was performed using a Cambridge Instruments S150 Scanning Electron Microscope, with image capture achieved using a Gatan Digiscan (Gatan Inc., USA). Diffuse reflectance infrared fourier transform (DRIFT) spectra were recorded on a Perkin Elmer FT-IR 1720X spectrometer with a PIKE Instruments diffuse reflectance attachment. Samples for DRIFT spectroscopy were ground up with KBr powder (1:20 mix) using an mortar and pestle. The instrument was set for 512 scans with a spectral resolution of 4 cm^-1.

Titrations

Potentiometric titrations were conducted on systems consisting of 0.50 g of charcoal, ash or the inorganic fraction of the ash (obtained after removal of carbon from the ash by treating in a muffle furnace at 800 °C) and made up to 300 mL with purified water in a borosilicate glass reaction vessel. The temperature of the reaction vessel was maintained at 25.0 ± 0.2°C by circulating water through an outer jacket from a constant temperature water bath. The electrolyte was 10 mM KNO₃ and the system was purged with nitrogen to eliminate CO₂ contamination. After pre-equilibration (24 h) at the natural pH the suspension was titrated with 0.100 M HNO₃ to about pH 3.5. The system was then titrated in steps to pH 9.5 with 0.100 M KOH using a PC-controlled titration system (Metrohm 691 pH meter and a Metrohm 665 Dosimat). After each addition of acid or base the pH was monitored until the drift was less than 0.1 mV per min, a criterion that was typically achieved within 20-30 min. Most titrations took about 48 h. Potentiometic titration data was modelled employing a constant capacitance surface complexation model and the computer program GRFIT (Ludwig 1992).

Leaching

Leaching experiments were conducted in two ways. In the first method 10.00 g of bushfire residue was placed with 300 mL of Milli-Q water in a borosilicate-glass reaction vessel and the suspension pH adjusted to 4.5, 7.5 or 8.5 using HNO₃ or NaOH. Samples (4 mL) were taken, using a glass syringe, at various times over a 15 day period. After each sample was taken it was filtered through 0.22 :m glass-fibre filters (Whatman, England), and the filtrate analysed using high performance liquid chromatography (HPLC). Adsorption of humic material to filters was negligible. In a second procedure, which we refer to as sequential extraction, 10.00 g of ash was leached in 300 mL of Milli-Q water for 24 h at the required pH (each sample was treated at pH 4.5, 7.5 and 8.5). After the 24 h leaching period the ash sample was filtered and then transferred into fresh water, adjusted to the desired pH and leached again for another 24 hours. This procedure was repeated three times. The leachate collected at each stage was either analysed for humic acid using HPLC or placed into an oven at 55 ^C to evaporate to dryness and the solid residue analysed using DRIFT spectroscopy.

Humic acid analysis

Humic acid standards were prepared from the sodium salt (Sigma Aldrich, Australia). This was further purified by mixing with Milli-Q water and adjusting to pH 10 with concentrated ammonium hydroxide. After centrifugation the supernatant was removed and then precipitated with concentrated hydrochloric acid. The precipitate was collected by vacuum filtration through Whatman #1 filter paper and then oven dried at 80 <C for 50 h. Humic acid concentration was measured according to a modification of the method of Susic and Boto (1989). The HPLC analysis used fluorescence detection with an excitation wavelength of 360 nm and monitoring emission at 425 nm. The mobile phase was 0.003% ammonia in Milli-Q water, with a flow rate of 1 mL/min. Separation was achieved on a Synergy Hydro-RP column (150 × 4.6 mm I.D., 4 μm, Phenomenex). A humic acid calibration curve (0-1000 mg/L) was used for the quantification of the bushfire residue leachate solutions with 50 μL samples injected for analysis. Validation of the analysis procedure was performed by precipitating humic material from leachate solution (as described above) and drying. Equivalent masses of purified humic acid and precipitated leachate gave equivalent assays by HPLC to within 2%.

Results

Characterisation

Scanning electron micrographs of charcoal and ash are shown in Fig. 1. The SEM images of the charcoal (A and B) show pores and channels in the charcoal particles, whilst those of the ash show randomly shaped particles. The surfaces of some of the ash particles (C) are smooth and flat, but most of the ash observed under the SEM consisted of irregular particles with little obvious information about their substructures. The ash was also found to contain small fragments of charcoal as can be seen in (D).

Figure 1: Scanning electron micrographs of bushfire residues: (A) a single fragment of charcoal, (B) higher magnification image of charcoal showing the pores and channels on the surface of the charcoal particle, (C) a collection of ash particles, and (D) a small fragment of charcoal found within the ash. (Scale bars shown below each image)

The charcoal contains a significantly higher fraction of organic material than the ash (Table 1). The specific surface areas of the bushfire residues used in this investigation are also listed in Table 1. The charcoal had the largest specific surface area, and the largest pore volume–consistent with the SEM images where the porous nature of the charcoal can be observed. The surface area of the ash, which contained approximately 21% organic matter, was more than double that of the inorganic ash fraction, which contained no organic matter. Pore volumes increased with increasing organic matter content.

DRIFT spectra of bushfire residues

DRIFT spectra of the charcoal, ash, inorganic ash fraction and purified humic acid are presented in Fig. 2. The main absorbance bands in the DRIFT spectra were: a broad band at 3300 to 3400 cm^-1 (OH stretch of phenolic and aliphatic OH, possibly NH), two peaks at 2930 and 2850 cm^-1 (asymmetric and symmetric CH stretch of -CH2-), a broad peak at 2600 cm^-1 (OH stretch of COOH), a peak or shoulder around 1700 to 1720 cm^-1 (C=O of COOH), a peak around 1585 to 1595 cm^-1 (COO- stretch), a peak at 1435 to 1450 cm-1 (C-H deformation of CH₂ or CH₃ groups), a small peak or shoulder at 1360 cm^-1 (symmetric COO- stretch or -CH bending of aliphatics), a broad peak around 1260 to 1280 cm^-1 (C-O stretch and OH deformation of COOH and phenolics), and a peak or shoulder around 1040 cm^-1 (aliphatic C-O-C or polysaccharides and/or Si-O of silicate minerals). The number of bands in the spectra around 1600 cm^-1 for the charcoal and humic acid is also typical of structures with high aromaticity (Baes and Bloom 1989; Stevenson 1994; Guo and Bustin 1998).

Table 1. Character of bushfire residues

Bushfire residue	Organic matter (% w/w) ± 2σ	Surface area (m2/g) ± 2σ	Total pore volume (cm3/g) ± 2σ
Charcoal	98 ± 2	403.8 ± 0.2	0.19 ± 0.01
Ash	21 ± 2	15.6 ± 0.2	0.04 ± 0.01
Inorganic ash fraction	0 ± 2	7.2 ± 0.2	0.03 ± 0.01

Figure 2: DRIFT spectra of (A) charcoal, (B) ash, (C) the inorganic fraction of the ash, and (D) humic acid.

Surface area of ash after leaching

The specific surface area of the ash was measured on dried samples that had been leached for 360 h at pH 4.5 and pH 8.5. At pH 4.5 the surface area was reduced to 14.5 ± 0.2 m²/g (from 15.6 ± 0.2 m²/g before leaching). The leached surface area was 12.2 ± 0.2 m²/g at pH 8.5.

Potentiometric titrations of bushfire residues

The acid-base properties for the bushfire residues were determined using potentiometric titrations. The charcoal exhibited two broad endpoints, one below pH 5 and the other above pH 8. For the ash, endpoints were less distinct, probably because of the presence of mineral and other inorganic components. Average pK_a values and concentrations of the organic functionalities were estimated from titration data and are given in Table 2.

Table 2. Surface acidity for ash and charcoal.

	Ash pKa, concentration	Charcoal pKa, concentration
carboxylate (COOH)	3.4, 0.32 mmol/g	5.5, 0.16 mmol/g
phenolic (Ar-OH)	8.5, 0.11 mmol/g	9.3, 0.15 mmol/g

Leaching

Fig. 3 shows the amount of humic acid that was leached from the bushfire residues as a function of time. A significant feature of these curves is that the leaching is a pH dependent process, with more humic material leached at higher pH values. For the ash residue there is an initial rapid leaching period in the first 24 h. At pH 7.5 and 8.5 leaching continues slowly for the time period of the experiments (360 h), while at pH 4.5, a decrease in humic concentration is observed after the initial desorption stage. Fig. 3 also shows that the charcoal was much less soluble than the ash residue.

Figure 3: Humic acid leaching curves for the bushfire residues (33.3 g/L, 25 °C) : ash at pH 8.5 (solid squares), pH 7.5 (open triangle) and pH 4.5 (solid circle), and charcoal all pH values (solid diamond). Experiments performed in duplicate.

DRIFT Spectra of Ash Leachates at pH 7.5

Sequential extractions were conducted to investigate if the character of the leaching material changed with successive extractions. DRIFT spectra of the leachates extracted on four consecutive days at pH 7.50 (the natural pH of the suspension) are shown in Fig. 4. The main absorbance bands of the DRIFT spectra of the ash leachates were: a broad band at 3300 to 3400 cm^-1 (OH stretch of phenolic and aliphatic OH), peak at 3090 (aromatic C-H stretch), two peaks at 2930 and 2850 cm^-1 (asymmetric and symmetric CH stretch of -CH₂-), a peak around 1600 cm^-1 (COO^- stretch and /or C=C aromatic), a peak or shoulder at 1430 cm^-1 (C-H deformation of CH₂ or CH₃ groups), a peak or shoulder at 1380 cm^-1 (symmetric COO^- stretch or -CH bending of aliphatics), a small shoulder at 1200 to 1220 cm^-1 (aromatic C , C-O stretch and/or OH deformation of COOH), and two peaks at 1095 and 1040 cm^-1 (C-O stretch of polysaccharides or polysaccharide-like structures). Fig. 4 shows that the leachate from day 1 is more aromatic in character (peaks at 3090, 1200 cm^-1, aromatic C=C at 1600 cm^-1, and C-H aromatic out-of-plane bending bands around 800 cm^-1). The relative intensity of the -OH hydrogen bond (3350 cm^-1) decreased upon successive days of leaching. Compared to the other bands in the spectrum, the relative intensity of the polysaccharide, or C-O-C aliphatic (1095 and 1040 cm^-1), increased with each extraction.

Figure 4. DRIFT spectra of humic acid leachate after sequential extraction from ash at pH 7.5.

Discussion

Characterisation of the fire residues

Analysis of the specific surface areas of the bushfire residues (Table 1) shows that the surface area of the charcoal was approximately 25 times greater than the ash. The difference results from the porosity of the charcoal, as is reflected by its larger pore volume and evident in the SEM images (Fig.1), where the porous nature of the charcoal can be observed. The surface area of the ash was more than twice that found for the inorganic ash fraction, suggesting that the organic fraction is responsible for a significant fraction of the surface area; this was also reflected by a decrease in the surface area after removal of organic material during leaching experiments.

DRIFT spectroscopy has been used by a number of authors to characterise various black carbons and humic substances (Goldberg 1985; Baes and Bloom 1989; Niemeyer et al. 1992; Stevenson 1994; Guo and Bustin 1998; Masserschmidt et al. 1999). Although band assignment gives general structural information, it is important to recognise that because of the complex structure of the fire residues such assignment is by no means definitive. The presence of a large number of bands in the DRIFT spectra of the charcoal (Fig. 2a) below 1600 cm^-1, along with the peaks evident around 800 cm^-1, suggests a surface with significant aromatic character. Similar features are observed in the spectra of the humic acid (Fig. 2d). Also present are strong bands that can be assigned to aromatic carboxyl groups and phenolics (1707 and 1281 cm^-1). Guo and Bustin (1998) reported similar infrared spectra for charcoal obtained under conditions of medium to high temperature (320^oC) and low charcoal formation time. By comparison, absorbance in the aromatic region for the ash sample was low, but significant bands were present in regions representing carboxyl (1587 cm^-1) and hydroxyl (3407 cm^-1) functional groups. Spectral comparisons between the two bushfire residues and the purified humic acid reveal that the charcoal had a similar functional group composition to the humic acid. There were few bands common between the ash and the purified humic sample, though some of the bands present in the DRIFT spectra of the ash sample represent absorbances of inorganic materials (the spectrum of the inorganic fraction shows that some of the peaks present in the ash spectrum result from absorbances of minerals and salts).

Modelling of the potentiometric titration data required the assumption that there were two different populations of acidic sites on the surfaces of the bushfire residues; carboxylate sites, with a lower pK_a and an end point below pH 7, and phenolic sites that have a higher pK_a and an end point at higher pH. Many authors have reported that the reactivity of humic materials results from the abundance of carboxyl and weakly acidic phenolic surface functional groups (Livens 1991; Stevenson 1994; Schnitzer 2000; Sanyal 2001). The presence of these sites in the bushfire residues was also confirmed in the DRIFT spectroscopy. The pK_a values obtained from the modelling do not represent a group of sites with the same pK_a value, but rather, a continuum of sites with similar acidity. The ash had a higher concentration of the more acidic functionalities, which collectively had a significantly lower pK_a compared to the charcoal. The charcoal, however, had a higher concentration of phenolic groups–reflected by the strong phenolic -OH stretching bands (1281 cm^!1) which were not evident in the spectrum of the ash. The pK_a values of the acid functional groups proposed from the modelling agree with the values generally accepted for carboxylate and phenolic functionalities for humic materials. For example, Stumm and Morgan (1996) report pK_a values ranging from 4!6 for carboxylate groups and 9!11 for phenolic groups.

Leaching

The leaching of humic acid from bushfire residues was pH dependent, with more humic material leached at higher pH values. It is not surprising that the solubility was higher at higher pH, as carboxylic acid and phenolic functional groups will become progressively more ionised as the pH increases, resulting in higher aqueous solubilities.

Leaching experiments also showed that the solubility of the charcoal was approximately 100 times lower than the ash, despite the higher organic composition of the charcoal. There are a number of factors that may contribute this. Titrations indicate that only a small fraction of the charcoal surface is occupied by polar phenolic or carboxylate functional groups, pointing to a surface that is less oxidised compared to the ash. This, along with the higher aromaticity of the charcoal, suggests that the charcoal surface is more hydrophobic. In addition, the SEM images indicate that the charcoal sample retains much of its original unburnt morphology, and as such, the material is more structured than the ash, with any acidic functional groups present at the surface likely to be part of an extended structural network. The increased surface hydrophobicity and extended structure of the charcoal contribute to lower water solubility.

An interesting feature of the humic acid leaching curves for the ash sample was the decrease in the amount of humic material leached at pH 4.5 after approximately 50 hours. This decrease probably results from re-adsorption of humic material back onto ash particles – probably onto the mineral phase which will carry a net positive charge at this pH.

Character of the leached material at pH 7.5

DRIFT spectroscopy of the leachates after sequential extraction indicates that the initial material leached into water is more aromatic in nature, and probably consists largely of aromatic acids and phenols. Upon repeated leaching the leachate becomes more aliphatic and contains more polysaccharide-like structure.

Conclusions

Leaching of humic acid from bushfire residues was a pH dependent process with more humic material leached at higher pH. Leaching was more significant for the ash residue compared to the charcoal, indicating that ash will have a greater short-term effect on water quality and the amount of humic material that may enter soil. The ash leachate initially contained compounds that were dominated by aromatic acid and phenolic functionalities, but became progressively more aliphatic in character as leaching continued.

References

Baes AU, Bloom PR (1989) Diffuse Reflectance and Transmission Fourier Transform Infrared (DRIFT) Spectroscopy of Humic and Fulvic Acids. Soil Science Society of America Journal 53, 695-700.

Baldock JA, Smernik RJ (2002) Chemical Composition and Bioavailability of Thermally Altered Pinus Resinosa (Red Pine) Wood. Organic Geochemistry 33, 1093-1109.

Fernandes MB, Skjemstad JO, Johnson BB, Wells JD, Brooks P (2003) Characterization of carbonaceous combustion residues. I. Morphological, elemental and spectroscopic features. Chemosphere 51, 785-795.

Goldberg ED (1985) 'Black Carbon in the Environment.' (John Wiley and Sons: New York).

Guo Y, Bustin RM (1998) FTIR Spectroscopy and Reflectance of Modern Charcoals and Fungal Decayed Woods: Implications for Studies of Inertinite in Coals. International Journal of Coal Geology 37, 29-53.

Haumaier L, Zech W (1995) Black Carbon - Possible Source of Highly Aromatic Components of Soil Humic Acids. Organic Geochemistry 23, 191-196.

Kumada K (1983) Carbonaceous Materials as a Possible Source of Soil Humus. Journal of Soil Science and Plant Nutrition 29, 383-386.

Livens FR (1991) Chemical Reactions of Metals with Humic Material. Environmental Pollution 70, 183-208.

Ludwig CD (1992). GRFIT, A Computer Program for Solving Speciation Problems: Evaluation of Equilibrium Constants, Concentrations and Other Physical Parameters. Berne, University of Berne.

Masserschmidt I, Cuelbas CJ, Poppi RJ, De Andrade JC, De Abreu CA, Davanzo CU (1999) Determination of Organic Matter in Soils by FTIR/Diffuse Reflectance and Multivariate Calibration. Journal of Chemometrics 13, 265-273.

Mitra S, Bianchi TS, McKee BA, Sutula M (2002) Black Carbon from the Mississippi River: Quantities, Sources, and Potential Implications for the Global Carbon Cycle. Environmental Science and Technology 36, 2296-2302.

Niemeyer J, Chen Y, Bollag JM (1992) Characterization of Humic Acids, Composts, and Peat by Diffuse Reflectance Fourier-Transform Infrared Spectroscopy. Soil Science Society of America Journal 56, 135-140.

Orioli GA, Curvetto NC (1978) The Effect of Fire on Soil Humic Substances. Plant and Soil 50, 91-98.

Sanyal SK (2001) Colloid Chemical Properties of Soil Humic Substances: A Relook. Journal of the Indian Society of Soil Science 49, 537-569.

Schmidt MWI, Noack AG (2000) Black Carbon in Soils and Sediments: Analysis, Distribution, Implications, and Current Challenges. Global Biogeochemical Cycles 14, 777-793.

Schnitzer M (2000) A Lifetime Perspective on the Chemistry of Soil Organic Matter. Advances in Agronomy 68, 1-58.

Shindo H (1991) Elementary Composition, Humus Composition, and Decomposition in Soil of Charred Grassland Plants. Journal of Soil Science and Plant Nutrition 37, 651-657.

Skjemstad JO, Reicosky DC, Wilts AR, McGowan JA (2002) Charcoal Carbon in U.S Agricultural Soils. Soil Science Society of America Journal 66, 1249-1255.

Stevenson FJ (1994) 'Humus Chemistry: Genesis, Composition, Reactions.' (John Wiley and Sons: New York).

Stumm W and Morgan JJ (1996) 'Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters.' (Wiley-Interscience: New York).

Susic M, Boto KG (1989) High-Performance Liquid Chromatographic Determination of Humic Acid In Environmental Samples at the Nanogram Level Using Fluorescence Detection. Journal of Chromatography 482, 175-187.