Article pubs.acs.org/est
Discrimination in Degradability of Soil Pyrogenic Organic Matter Follows a Return-On-Energy-Investment Principle Omar R. Harvey,*,† Allison N. Myers-Pigg,‡ Li-Jung Kuo,§ Bhupinder Pal Singh,∥ Kevin A. Kuehn,⊥ and Patrick Louchouarn‡,# †
School of Geology, Energy and the Environment, Texas Christian University, Fort Worth, Texas 76129, United States Department of Oceanography, Texas A&M University, College Station, Texas 77840, United States § Marine Science Laboratory, Pacific Northwest National Laboratory, Sequim, Washington 98382, United States ∥ NSW Department of Primary Industries, Elizabeth Macarthur Agricultural Institute, Menangle, New South Wales 2568, Australia ⊥ Department of Biological Sciences, The University of Southern Mississippi, Hattiesburg, Mississippi 39406, United States # Department of Marine Sciences, Texas A&M University at Galveston, Galveston, Texas 77553 United States ‡
S Supporting Information *
ABSTRACT: A fundamental understanding of biodegradability is central to elucidating the role(s) of pyrogenic organic matter (PyOM) in biogeochemical cycles. Since microbial community and ecosystem dynamics are driven by net energy flows, then a quantitative assessment of energy value versus energy requirement for oxidation of PyOM should yield important insights into their biodegradability. We used bomb calorimetry, stepwise isothermal thermogravimetric analysis (isoTGA), and 5-year in situ bidegradation data to develop energy-biodegradability relationships for a suite of plant- and manure-derived PyOM (n = 10). The net energy value (ΔE) for PyOM was between 4.0 and 175 kJ mol−1; with manure-derived PyOM having the highest ΔE. Thermal-oxidation activation energy (Ea) requirements ranged from 51 to 125 kJ mol−1, with wood-derived PyOM having the highest Ea requirements. We propose a return-on-investment (ROI) parameter (ΔE/Ea) for differentiating short-to-medium term biodegradability of PyOM and deciphering if biodegradation will most likely proceed via cometabolism (ROI < 1) or direct metabolism (ROI ≥ 1). The ROI-biodegradability relationship was sigmoidal with higher biodegradability associated with PyOM of higher ROI; indicating that microbes exhibit a higher preference for “high investment value” PyOM.
■
INTRODUCTION Pyrogenic organic matter (PyOM) refers to the physicochemically diverse group of combustion-derived organic residue produced during biomass burning.1 Current estimates of annual global carbon input from the production of PyOM in vegetation fires are on the order of 118−385 megatons;2 with fire data indicating that the average size of vegetation fires in the United States have tripled to 120 acres per fire (and continue to increase) since 1990 (Figure S1 of the Supporting Information). With the increasing recognition of fire as an important driver of current and future landscape dynamics, the fate of PyOM and its environmental significance have been the subject of numerous research efforts. The overall consensus from these efforts indicates that PyOM represents a key component in biogeochemical cycles controlling Earth’s climate, as well as nutrient and contaminant dynamics in soils and aquatic systems. Central to elucidating the fate and environmental significance of PyOM materials is an understanding of their soil degradability and degradation characteristics. Many studies © 2016 American Chemical Society
have examined the soil degradability and degradation characteristics of organic matter in general (and PyOM in particular) from a mass change and chemical structure perspective. Results from these studies point to a confluence of molecular structure, substrate−subtrate interactions, microbial community, and abiotic conditions as the determinants of degradability and degradation characteristics. For example, there is a significant amount of literature indicating that, ceteris paribus, degradability/degradation (defined as resistance to abiotic or biotic loss) decreases with increasing condensation of the PyOM molecular structure.3−7 Both Zimmerman4 and Singh et al.3 showed that increasing PyOM condensation, via increased heat treatment temperature (HTT) or using hard tissue (e.g., wood) versus soft tissue (e.g., leaves and grass) feedstocks, resulted in increased PyOM resistance to degradation and lower Received: Revised: Accepted: Published: 8578
February 29, 2016 June 19, 2016 July 11, 2016 July 11, 2016 DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology Table 1. Selected Properties and Associated Arrhenius Parameters for PyOMs Used in the Study Arrhenius parametersc a
PyOM
R50 (SE)
EuW400Ab EuW550A EuW400 EuW550 EuL400A EuL550A PL400 PL550A CM400 CM550A
0.532 0.577 0.527 0.584 0.504 0.508 0.503(0.012) 0.507(0.020) 0.460(0.007) 0.468(0.010)
b
−1
nonaromatic C (%)
CHNO (mol g )
14.7 1.3 17.0 1.2 18.4 8.0 26.4 7.6 29.1 5.5
0.105 0.100 0.106 0.099 0.106 0.099 0.080 0.062 0.039 0.029
ln (A) 10.7 17.0 10.5 18.0 9.98 11.3 6.76 10.1 5.35 8.52
± ± ± ± ± ± ± ± ± ±
1.80 1.32 1.95 1.72 0.858 2.45 1.49 1.60 1.07 1.39
Ea (kJ mol−1)
R2
± ± ± ± ± ± ± ± ± ±
0.935 0.989 0.909 0.969 0.977 0.870 0.886 0.919 0.911 0.914
81.3 120.0 80.3 125.4 76.0 82.9 59.5 77.5 51.4 67.9
9.62 7.01 10.4 9.19 4.37 13.1 8.08 8.69 5.37 6.96
a
Recalcitrance index determined according to Harvey et al.5 SE is standard error for triplicate measurements. Other values were average of duplicate measurements. bData from Singh et al.2 cFitted parameters for Arrhenius equation; ln(k) = ln (A) − Ea/RT, where k is the first-order rate constant (in units of min−1), A is the frequency factor (in min−1), Ea is activation energy (in kJ mol−1), R is the gas constant (in kJ mol−1K−1), and T is temperature (in K). R2 is the goodness of fit when experimentally determined k values were fitted at T = 473.15−773.15K (200−500 °C). dPyOM with “A” were steam activated.3
different types of PyOM and whether they are comparable to those for nonpyrolyzed OM are yet to be determined. Rovira et al.15 used differential scanning calorimetry to show that litter of “higher quality” had the highest energy content and susceptibility to microbial degradation. Their description of “quality” took into consideration energy benefits and the energy inputs (needed to obtain such benefit) but a quantitative value/ index for “quality” was not developed. The “recalcitrance” values from Harvey et al.6 and the Ea values from Leifeld et al.14 provide useful quantitative values; but both assessments only consider the relationship between energy input and biodegradability. In the current study, we combine the underlying principles of “quality” and “recalcitrance” into a single unifying hypothesis to describe the biodegradability of PyOM. This unifying hypothesis proposes that, “irrespective of feedstock or pyrolysis conditions, PyOM degradability will vary predictably along a common quality-recalcitrance gradient; whereby PyOM with the highest ratios of energy output-to-energy input requirement will exhibit the highest susceptibility to degradation within a given environment”. To test this hypothesis, we combined stepwise-isothermal thermogravimetic analysis (isoTGA), bomb calorimetry, a return-on-energy-investment (ROI) parameter and 5 years of PyOM degradation data to study the relationship between net energy gains and biodegradability in a range of PyOM residues produced from different biomass feedstocks and at different HTT. Specific objectives were to (1) determine the activation energy (Ea) needs for the oxidation of different PyOM; (2) determine the calorific value of different PyOM; (3) estimate the net energy gains and ROI associated with the oxidation of PyOM and; (4) quantify the energy-PyOM biodegradability relationship and use this relationship to contribute new knowledge to current understanding of biodegradation preferences and pathways for PyOM biodegradation in aerobic soils.
degradation rates. Work by other researchers (albeit primarily on nonpyrogenic organic matter) however, suggests that molecular structure is not always a good indicator of persistence.8−12 As evidence of this, Kleber et al.12 noted that compared to 107 year old carbon extracted from a Oxisol, older carbon (680 years old) from an inceptisol had a higher proportion of alkyl/aromatic C, thermally labile materials and fatty-acid-to-lignin phenol ratios; thereby, lacking the highly condensed molecular structure that would be expected for the “recalcitrant” carbon remaining after six centuries. Instead, they suggested that organic matter degradability be seen not as a function of molecular structure but, of microbial ecology and resources availability within a given physical soil environment. Despite significant progress, there are still noticeable knowledge gaps in current understanding of PyOM degradability. For example, while there is overwhelming evidence pointing to both molecular structure and environmental conditions as key determinants of PyOM degradability in the short-to-medium term,13 understanding on a basic quantitative and thermodynamic level is still largely unresolved. Understanding PyOM degradability from a first principle perspective is especially important for discerning the role (and predicting the behavior) of PyOM in environmental systemswhere net energy flow drives biogeochemical cycling of nutrients and contaminants. There are in fact several lines of evidence to suggest that an energy-based approach may indeed be pivotal to understanding PyOM degradability at the most fundamental level. First, the degree of microbial activity and soil temperature noted by a number of researchers as having controlling effects on both the quantity and rate of mineralization of a given PyOM are both energy driven environmental factors.8−11 Second, the degradation of native soil organic matter is known to have an activation energy (Ea) requirement on the order of 67−120 kJ per mol;14 with indications of at least a qualitative link between Ea and biodegradability.15 Harvey et al.6 developed a recalcitrance index (R50) as a tool for classifying PyOM into biodegradability classes by combining thermal analysis with the concept of a link between energy input requirements and biodegradability. Strong inverse relationships between R50 and microbial mineralization of PyOM confirmed the link between energy input requirements and recalcitrance. However, what the values of the activation energy look like for
■
EXPERIMENTAL SECTION Pyrogenic Organic Matter Materials and Their Degradability. The materials (n = 10) used in this study were a suite of well-characterized plant- and manure-derived PyOM produced at heat treatment temperatures (HTT) 400 or 550 °C.3,16,17 Detailed description of the production, chemical analysis, and chemical properties for the fresh PyOM can be 8579
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology
Figure 1. Stepped and derivative thermogravimetic thermograms for (A) EuW400A, (B) EuW550A, CM400, and CM550 PyOM. Thermograms were obtained under air flow (10 mL min−1) and ramp-to-isothermal conditions at 200, 300, 350, 400, 450, and 500 °C. Isothermal conditions were maintained for 60 min at each temperature and ramp rate between isothermal equilibration was 10 °C min−1.
found elsewhere,3,16,17 with selected properties listed in Table 1. Briefly, PyOM materials were produced at 400 or 550 °C with or without steam activation. Steam activated PyOM are denoted with an “A” at the end of the sample designation (e.g., EuW400A). Plant-derived PyOM materials were produced from a C3 vegetation feedstock, Eucalyptus saligna wood and leaves (denoted as EuW400, EuW550 for wood, and EuL400 and EuL550 for leaves). Manure-derived PyOM are produced from poultry litter (PL400, PL550) and cow manure (CM400, CM550). The susceptibility of the PyOM materials to microbial degradation was evaluated both indirectly and directly. Indirect assessment of microbial degradability was based on the indexbased approach by Harvey et al.6 In this approach, the thermal degradation of 14 to 16 mg of each PyOM in air (flow rate = 10 mL min−1) was compared to that of graphite and the T50 (temperature corresponding to 50% weight loss) values used to calculate the recalcitrance parameter, R50. Further details on the calculation and applicability of R50 to PyOM degradation is outlined in Harvey et al.6 Direct evaluation of microbial degradation of PyOM materials was through a 5-year long laboratory incubation experiment.3 In this experiment, PyOM materials were added to a vertisol vegetated with the C4 (and hence δ13C-enriched) grass, Astrebla spp., for over 100 years. The PyOM materials were added at a proportion of 8.17 g per kg soil and incubated under optimal soil conditions (dark, 67% water holding capacity, 22 °C); with respired C being captured throughout the 5 year period as CO2 in NaOH, precipitated as SrCO3 and analyzed for δ13C. To quantify the level of microbial
degradation, the authors took advantage of the different isotopic signatures in CO2 released from the δ13C-enriched native soil C compared to δ13C-depleted PyOM. The reader is directed to Singh et al.3 for details on calculations. We combined this 5-year data set of direct PyOM degradation with data collected from thermal analysis (including R50, activation energy and net energy value) of the fresh unincubated PyOM to achieve research objectives of the current study. Activation Energy, Net Energy Value, and Return-onEnergy Investment (ROI) Analyses. Stepwise-isothermal thermogravimetic analysis (isoTGA; TA Q500, TA Instruments) was used to assess the thermal degradation of each PyOM in air (10 mL min−1) at isothermal temperatures (isoTemp) of 200°, 300°, 350°, 400°, 450°, and 500 °C. Hold time at each isothermal temperature was 60 min with a ramp rate of 10 °C min−1 between each isothermal step. For each isoTGA run, the resulting data for each PyOM was discretized by isoTemp (removing overtemperature effects to within 1 °C of the target isoTemp), plotted as a function time and used to determine the thermal oxidation rate constant (k) at respective isoTemp. The activation energy, Ea, was then derived from the linear form of Arrhenius equation: ln(k) = ln(A) − Ea /RT
(1)
where k denotes rate constant; A is the frequency factor; T is the isoTemp (in K); R is the gas constant (0.008314 kJ mol−1K−1). A plot of ln (k) as a function of −1/RT, produces a straight line with a slope equal to the activation energy, Ea. A bomb calorimeter equipped with a semimicro oxygen bomb (Parr 6200 model 1109A, Parr Instruments) was used to 8580
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology obtain net energy values (ΔE; kJ g−1) by completely oxidizing between 45 and 50 mg of each PyOM sample (weighed to ±0.01 mg). Oxidation was performed under a high purity oxygen headspace (>99.5% O2; pressure = 30 bar) with NiCr fuse wire (10 cm and helically coiled) providing the ignition charge. Values for ΔE were corrected for ash content, converted to kJ mol−1 (using the molar CHNO content) and weighted by degree of nonaromaticity for each PyOM. While ΔE in units of kJ g−1 of C may be of ecological significance; units of kJ mol−1 of CHNO were considered more appropriate/ complete for the objectives of this study and facilitated the calculation of the return-on-energy-investment (ROI) parameter, which is defined as follows: ROI = ΔE /Ea
(2)
where, Ea (obtained from isoTGA analysis) is in units of kJ mol−1 and hence requiring that ΔE also be in kJ mol−1. Ash content, molar CHN content, and nonaromaticity used were obtained from Wang et al.17 and Singh et al.3 Oxygen content for each PyOM was calculated by difference, using ash and CHN content. Weighting by nonaromaticity was needed to account for the effect of molecular structure on energy distribution18 such as that observed in preliminary data assessment showing distinct structure-induced biases in data for HTT 450 versus 500 °C PyOM (Figure S1).
Figure 2. van-Krevelen plot showing atomic H/C, atomic O/C, and calculated CHO base formula of PyOM. Delineation of different classes of organic matter were after Kuhnert et al.24 and Hertkorn et al.25
■
RESULTS AND DISCUSSION Thermal Oxidation, Its Kinetics and Activation Energy of PyOM. Thermograms (and their respective derivatives) from isoTGA analysis of PyOM materials are shown in Figures 1 and S2. The thermograms reflected distinctive trends in thermal degradability/degradation, across both feedstock and HTT. Irrespective of feedstock, increasing HTT (from 400 to 550 °C) resulted in a decline in the thermally labile PyOM fraction at isoTemp ≤300 °C. For example, EuW/A, EuL, PL, and CM PyOM produced at HTT550 had 1, 8, 10, and 20% of this readily oxidizable fraction, respectivelycompared to 15, 17, 25, and 33% in respective HTT400 PyOM. Differences in the trends of thermal degradation characteristics across feedstock types were most apparent at isoTemps ≥300 °C. For the CM feedstock, 70−75% of the PyOM was thermally labile at isoTemps of ≤350 °C with peak thermal-lability (∼40% weight loss) occurring at isoTemp350 (Figure 1C,D). By comparison, only 20−35% of PyOM from EuW feedstock (activated and nonactivated) was thermally labile at isoTemps ≤350 °C; with peak thermal-lability (30−45% weight loss) occurring at isoTemp450 (Figure 1A,B). With the exception of a higher thermally labile fraction at isoTemp ≤300 °C in the PL400 samples, EuL and PL PyOM showed thermal degradation characteristics that were similar to each other but intermediate between that of CM (on the lower end) and EuW PyOM. For example, peak thermal-lability in all EuL and PL PyOM occurred at isoTemp400 (compared to isoTemp350 and 450 for CM and EuW PyOM, respectively). Observed differences in the thermal degradation character of the PyOM were directly attributable to variability in their molecular composition/structure.3,5,19 The consistently lower thermally labile fraction at isoTemp300 in PyOM produced at HTT550 versus HTT400 was attributable to increasingly condensed structures as HTT increased (Figure 2).19 Nuclear magnetic resonance data from Singh et al.3 indicated that the PyOM produced at HTT550 had higher aromatic and lower nonaromatic carbon contents than those produced at HTT400.
Comparison of CHO base formulas (calculated from atomic CHO ratios) pointed to a combination of dehydrogenation with dehydration (PL), dehydroxylation (EuL), or decarboxylation (EuW) as the primary mechanisms driving differences in the degree of PyOM condensation with HTT.5 For example, the CHO base formula for EuW400/A, EuW550/A, CM400, and CM550 is C12H6O3, C11H4O, C8H8O4, and C8H6O4, respectivelya difference of CH2O2 (i.e., CO2 + H2; indicating decarboxylation-dehydrogenation) and H2 (indicating dehydrogenation) for EuW and CM PyOMs, respectively. The increasing degree of condensation reflected in Figure 2 (EuW ≥ EuL = PL > CM) also provided a plausible explanation for the isoTemps at which peak thermal-lability of the PyOM (EuW = 450 °C; EuL = PL = 400 °C; CM = 350 °C) were observed. Discretization of isoTGA data for PyOMs into respective isoTemps (200, 300, 350, 400, 450, and 500 °C) indicated that thermal oxidation kinetics was first order and varies systematically with feedstock and/or HTT (Figure 3). In all instances, thermal oxidation kinetics could be adequately defined with the model: W = Woe−kt + β
(3)
where W is the weight (%) at time t, W0 is the initial weight (100%), k is the first-order rate constant, and β is the residual weight after a given isoTemp stage. Values for k ranged from 6.20 × 10−5 to 0.137 min−1 across the PyOM materials and increased with increasing isoTemp for a given PyOM (Figure S3). Plots of ln (k) versus 1/RT are provided in Figure S4. Fitted slopes of these plots, and hence activation energy of thermal oxidation (Ea), ranged from 51.4 (± 5.37) kJ mol−1 in CM400 to 125 (± 9.19) kJ mol−1 in EuW550 (Table 1). For a given feedstock, activation energy was higher in PyOM produced at HTT550 than HTT400 and at a given HTT, followed the general trend EuW > EuL > PL> CMconsistent with the degree of condensation reflected in Figure 2. It is also 8581
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology
Figure 3. Discretized weight-loss data and associated first-order kinetic modeling of thermal degradation for (A) EuW400A, (B) EuW550A, (C) CM400, and (D) CM550A PyOMs. Discretization was based on isothermal temperatures of 200−500 °C.
worth mentioning that our values for Ea were within the ranges reported by Leifeld et al.,14 who used differential scanning calorimetry to estimate chemical/thermal and biological activation energy for soil organic matter decomposition. Links between Energetics and Biodegradability of PyOM. The relationship between the relative recalcitrance index (R50) of the PyOM materials, their activation energy (Ea) and their net energy value (ΔE) is shown in Figures S6 and 4. Harvey et al.6 found that the susceptibility of PyOM to biodegradation decreases exponentially with increasing R50. This decrease in biodegradability (with increasing R50) was linked to an increase in the degree of conjugation/condensation in the molecular structure of the PyOM.5,6 A strong positive linear correlation (r2 = 0.88) between Ea and R50, as well as the relative positions of respective PyOM in the relationship was consistent with a recalcitrance-molecular structure link (Figure S6). That is, PyOM with higher degrees of conjugation/ condensation also had higher relative recalcitrance and require a higher energy input to initiate degradation.6 By comparison, the strong negative exponential nature of Figure 4 (which was also reflected in a plot of ΔE versus R50; Figure S4) indicated that PyOM with the lowest degree of conjugation/condensation (or lowest Ea and R50) had the highest ΔE (Figure 4B). For example, the ΔE of CM400 (Ea = 51.4 kJ mol−1; R50 = 0.46) was 175 kJ mol−1 of CHNO compared to 4 kJ mol−1 of CHNO for EuW550 (Ea = 125 kJ mol−1; R50 = 0.58). From a biodegradability perspective, this suggests that PyOM requiring the lowest energy investment had the potential for providing the greatest energy returns to support microbial growth.
Figure 4. Relationship between activation energy and net energy value (ΔE) of plant- and manure-derived pyrogenic organic matter produced at heat treatment temperatures of 400 or 500 °C.
Comparing the energetics of oxidation for the CC (which dominate the molecular structure of non- or low conjugated/ condensed organic structures)5,20 and CC bonds (whose proportion and significance increases with conjugation/ condensation)3,5,19 provide valuable insights into observed trends in both Ea and net energy value of the PyOM. The average bond energy (or Ea) needed to initiate oxidation of a 8582
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology
Figure 5. Return-on-investment (ROI)-degradability relationships for plant- and manure-derived pyrogenic organic materials produced at heat treatment temperatures of 400 or 500 °C. Measures of degradability were (A) directly measured microbial carbon mineralization and (B) relative recalcitrant index (R50)an indirect measure of microbial degradability. The fraction of total organic (TOC) existing as water-extractable organic carbon (WEOC) is also shown in (B; unfilled squares).
CC bond is on the order of 348 kJ mol−1, while the enthalpy of reaction (ΔH) for its oxidation, CC + 2O2 → 2CO2 is approximately −1858 kJ mol−1.21,22 By comparison, the average bond energy for the CC double bond is much higher at 614 kJ mol−1 and its ΔH is lower at −1592 kJ mol−1 (effectively −796 kJ bond−1).21,22 The trends we observed in Ea and ΔE can be plausibly explained by the fact that the molecular structures of non- or low conjugated/condensed PyOMs such as CM400, CM550, and PL400 are dominated by a CC backbone (and other single-bonded atoms) with low bond energy and high energy yields upon oxidationhence there lower Ea and higher ΔE.3,5 Comparatively, the progressively increasing conjugation/condensation reflected in Figure 2 for EuL400, EW400, PL550, EuL550, and EuW550 was consistent with molecular structures increasingly dominated by CC bonds and other high bond energy (high Ea), low energy yielding (low ΔE) bonds.3,5,19 The combined effect of a linearly increasing Ea and an exponentially decreasing ΔE on the biodegradation and biodegradability of the PyOM is reflected in the relationship between the return on investment parameter (ROI), measured microbial mineralization of PyOM, R50, and fraction of TOC existing in water-extractable (WEOC) form (Figures 5). Values of ROI for the PyOM were between 0.03 and 3.41; with the highest ROIs observed in CM400 (3.41) and PL400 (1.38) and becoming progressively lowerin CM550 (0.82), EuL400A (0.63), EuW400 (0.52), EuW400A (0.50), PL550A (0.43), EuL550A (0.29), and EuW550/A (0.03)as the degree of molecular conjugation/condensation increased and the fraction of TOC as WEOC decreased. The clustering of the lower ROI PyOM toward the base (and the higher ROI PyOM toward the crest) of the sigmoidal PyOM mineralized-ROI relationship (Figure 5A) was supportive of an energy-driven discrimination by microorganisms toward preferential degradation of PyOM with increasingly higher ROIs. The paralleling of the ROImineralization relationship in Figure 5A by that of the ROI− WEOC relationship in Figure 5B point to the water-extractable organic matter as the source of the high energy-yielding, low Ea components comprising PyOM with increasingly higher ROIs. The fact that mineralization (up to 2.5%) was observed in low
ROI PyOM materials (those with the least potential to supply energy for microbial growth) pointed to the likely involvement of indirect or cometabolism in the biodegradation of these PyOM. Although an identification of the metabolic pathway(s) involved is beyond the realms of possibility for this study; the fact that a material with no (to low) chemical energy value was biodegraded dictates the involvement of an energy-yielding auxiliary process. However, as ROI increases, indicative of the net energy output increasing relative to the required energy investment, the probability of PyOM degrading via direct metabolism increases. Biodegradation via direct metabolism would, therefore, be most likely for the CM400 and CM550A while cometabolism would be most likely in CM550, EuL400A, EuW400, EuW400A, PL550A, EuL550A, and EuW550/A. As expected, R50 of the PyOM showed a general decrease with increasing ROI (Figure 5B). That is, PyOM with the highest relative recalcitrance and conjugation/condensation also had the lowest ROI values. There were however clear indications of very different ROI-R50 relationships existing for PyOM with “low investment” value compared to those with “high investment” value. Initial segmental regression analysis of the ROI-R50 data (R2 = 0.81) showed an inflection point at ROI = 0.75 ± 0.22 and R50 = 0.49. However, there was no evidence that the ROI at this inflection point was significantly different than 1 (p-value = 0.2179). With the inflection point set to ROI= 1, the ROI-R50 relationship for PyOM with ROIs < 1 and ROIs > 1 were described by the equations; R 50 = 0.57 − 0.097(ROI)
(4)
and R 50 = 0.473 − 0.003(ROI)
(5)
respectively. A shifting of the inflection point from 0.75 toward 1 would also be expected in natural environments; since actual biological Ea is often significantly lower than the chemical Ea determined by thermal analysis techniques.14 Leifeld et al.14 found that biological Ea was, on average, 64% of respective chemical Ea determined by thermal analysis. Assuming the same relationship between chemical and biological Ea (Ea,bio), values of ROI would be 56% higher and the inflection point of the R508583
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology ROIbio relationship would be at 1.18 ± 0.35, where ROIbio = ΔE/Ea,bio, Ea,bio = 0.64Ea, and Ea is the activation energy determined by isoTGA. Interestingly, a similar inflection to that in the R50-ROIbio relationship was also observed in first derivative plots for the PyOM mineralized versus ROIbio and the water-extractable organic carbon (WEOC) versus ROIbio sigmoidal curves (Figure S7)pointing to a close link between variations in PyOM recalcitrance, water-extractable organic carbon, PyOM degradability, and net energy returns. Differences in slope between eqs 4 (slope = −0.097) and 5 (slope = −0.003), was supportive of a different biodegradation scenario being dominant in PyOM with ROIs < 1 versus those with ROIs > 1. At 32 times higher than in that eq 5, the slope in eq 4 reflected a structure-dominated, energy-deficient degradation scenario in PyOM with ROI < 1. That is, changes in molecular structure differences (as reflected in decreasing R50) at ROI < 1 is accompanied by relatively small gains in energy returns. This was very much in line with expectations for indirect or cometabolismsuch as that observed in positive priming of PyOMwhere energy needs for microbial growth are satisfied via metabolism of alternate energy-rich substrate with PyOM biodegradation occurring largely as a byproduct. In contrast, the much smaller slope of eq 5 was indicative of PyOM with ROI > 1 degrading in a structure-deficient, energyrich scenario where small variations in molecular structure favors comparatively large changes in net energy returns. Such observations are consistent with direct metabolismas would occur in the degradation of readily oxidizable water-extractable organic carbonwhere the energy return from PyOM degradation is sufficient to satisfy requirements of microbial growth without influx of an external “primer”. The inflection point at ROI = 1 (R50 = 0.47) in Figure 5B therefore represents the point where PyOM biodegradation transitions from being dominated by indirect/cometabolism to being dominated by direct metabolism and vice versa. Interestingly, the R50 (0.47) at the transition/inflection point between the two PyOM groups, and metabolic scenarios, was also comparable to that delineating Class B and Class C biochars.5 Class C biochars have R50 < 0.50, retains some characteristics of its uncharred parent material and are more susceptible to biodegradation than Class B biochars (0.50 ≤ R50 < 0.70) whose properties and behavior are very distinct from their uncharred parent materialdue to the removal of labile components and increased conjugation/condensation.5 The new data in the present study thus provide additional lines of evidence to show that PyOM classification5 can also be explained with thermodynamics parameters. Environmental Significance. We present, to our knowledge, the first study to quantitatively relate short- to mediumterm PyOM biodegradation and biodegradability directly to energy and energy returns dynamics. Our data indicated that microorganisms would have increasing preference to degrading PyOM with increasingly higher return on energy investment (higher ROI). This ROI concept also quantitatively verifies the OM quality concept proposed by Rovira et al.;15 whereby OM with higher relative energy gain is of higher quality and more favored by microbes. The significant links between ROI and biodegradability makes ROI a very informative parameter when assessing the short- to medium-term biodegradability of PyOM in the environment. We thus propose to include ROI in the characterization of PyOM for environmental applications. Such applications include carbon sequestration, soil amendments for agronomic purposes, and sediment amelioration for contami-
nant immobilization. It is also important to note that the ROI estimated using activation energies from isoTGA is a conservative estimate of return on energy investment to that expected in natural systems. This is due to the fact that microbial degradation is an enzyme-mediated/catalyzed oxidative process and generally require lower Ea than the noncatalyzed chemical oxidation measured by isoTGA.14 As we illustrated the “biological ROI” of PyOM will therefore be higher than the chemical ROI we determined. In this regard, PyOM with low chemical ROI may still be biodegradable via direct metabolism. Additionally, the mineralization data we used in the present study is from a well-controlled laboratory soil incubation experiment,3 and represented a more “optimal” degradation environment than would be expected in a natural temperate environment.23 We also only considered complete aerobic biodegradation of the PyOM and did not consider environmentally induced shifts in microbial communities, incomplete oxidation, anaerobic oxidation, or differences in enzymatic pathways. For information on microbial community and microbial activity in our experiments, the reader is directed to Singh and Cowie.26 Even with all these considerations, we believe that stability assessment of PyOM by ROI (or ROIbio) is a valuable metric and, like R50, provides an easily determined approach for selecting/differentiating PyOM samples for environmental applications.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b01010. Additional data on the elemental composition, thermal degradation, and thermal degradation kinetics of the various PyOM (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 817-257-4272; fax: 817-257-7789; e-mail: omar.
[email protected] (O.R.H.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We acknowledge Adriana Downie for arranging production of PyOM by Pacific Pyrolysis Australia (previously Best Energies). Financial support from the National Science Foundation (DBI 0923063) to K.A.K. is also gratefully acknowledged. The authors gratefully acknowledge the presubmission review and insightful comments from Dr. Pere Rovira (Centre Tecnològic Forestal de Catalunya, Spain). Comments from the Associate Editor and three anonymous reviewers improved the manuscript.
■
REFERENCES
(1) Masiello, C. A. New directions in black carbon organic geochemistry. Mar. Chem. 2004, 92 (1−4), 201−213. (2) Santin, C.; Doerr, S. H.; Preston, C. M.; Gonzalez-Rodriguez, G. Pyrogenic organic matter production from wildfires: a missing sink in the global carbon cycle. Glob. Change Biol. 2015, 21 (4), 1621−1633. (3) Singh, B. P.; Cowie, A. L.; Smernik, R. J. Biochar Carbon Stability in a Clayey Soil As a Function of Feedstock and Pyrolysis Temperature. Environ. Sci. Technol. 2012, 46 (21), 11770−11778. 8584
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
Article
Environmental Science & Technology (4) Zimmerman, A. R. Abiotic and Microbial Oxidation of Laboratory-Produced Black Carbon (Biochar). Environ. Sci. Technol. 2010, 44 (4), 1295−1301. (5) Harvey, O. R.; Herbert, B. E.; Kuo, L.-J.; Louchouarn, P. Generalized Two-Dimensional Perturbation Correlation Infrared Spectroscopy Reveals Mechanisms for the Development of Surface Charge and Recalcitrance in Plant-Derived Biochars. Environ. Sci. Technol. 2012, 46 (19), 10641−10650. (6) Harvey, O. R.; Kuo, L.-J.; Zimmerman, A. R.; Louchouarn, P.; Amonette, J. E.; Herbert, B. E. An Index-Based Approach to Assessing Recalcitrance and Soil Carbon Sequestration Potential of Engineered Black Carbons (Biochars). Environ. Sci. Technol. 2012, 46 (3), 1415− 1421. (7) Bruun, S.; Jensen, E. S.; Jensen, L. S. Microbial mineralization and assimilation of black carbon: Dependency on degree of thermal alteration. Org. Geochem. 2008, 39 (7), 839−845. (8) Steinbeiss, S.; Gleixner, G.; Antonietti, M. Effect of biochar amendment on soil carbon balance and soil microbial activity. Soil Biol. Biochem. 2009, 41 (6), 1301−1310. (9) Davidson, E. A.; Janssens, I. A. Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 2006, 440 (7081), 165−173. (10) Cheng, C. H.; Lehmann, J.; Thies, J. E.; Burton, S. D. Stability of black carbon in soils across a climatic gradient. J. Geophys. Res. Biogeosci. 2008, 113, (G2).10.1029/2007JG000642 (11) Fang, Y.; Singh, B.; Singh, B. P. Effect of temperature on biochar priming effects and its stability in soils. Soil Biol. Biochem. 2015, 80, 136−145. (12) Kleber, M.; Nico, P. S.; Plante, A. F.; Filley, T.; Kramer, M.; Swanston, C.; Sollins, P. Old and stable soil organic matter is not necessarily chemically recalcitrant: implications for modeling concepts and temperature sensitivity. Glob. Change Biol. 2011, 17 (2), 1097− 1107. (13) Lehmann, J.; Abiven, S.; Kleber, M.; Pan, G.; Singh, B. P.; Sohi, S. P.; Zimmerman, A. R.; Lehmann, J.; Joseph, S. Persistence of biochar in soil. Biochar for Environmental Management: Science, Technology and Implementation 2015, 233−280. (14) Leifeld, J.; Lützow, M. Chemical and microbial activation energies of soil organic matter decomposition. Biol. Fertil. Soils 2014, 50 (1), 147−153. (15) Rovira, P.; Kurz-Besson, C.; Coûteaux, M.-M.; Ramón Vallejo, V. Changes in litter properties during decomposition: A study by differential thermogravimetry and scanning calorimetry. Soil Biol. Biochem. 2008, 40 (1), 172−185. (16) Singh, B.; Singh, B. P.; Cowie, A. L. Characterisation and evaluation of biochars for their application as a soil amendment. Aust. J. Soil Res. 2010, 48 (7), 516−525. (17) Wang, T.; Arbestain, M. C.; Hedley, M.; Singh, B. P.; CalveloPereira, R.; Wang, C. Determination of carbonate-C in biochars. Soil Res. 2014, 52 (5), 495−504. (18) Seredych, M.; Hulicova-Jurcakova, D.; Lu, G. Q.; Bandosz, T. J. Surface functional groups of carbons and the effects of their chemical character, density and accessibility to ions on electrochemical performance. Carbon 2008, 46 (11), 1475−1488. (19) Wiedemeier, D. B.; Abiven, S.; Hockaday, W. C.; Keiluweit, M.; Kleber, M.; Masiello, C. A.; McBeath, A. V.; Nico, P. S.; Pyle, L. A.; Schneider, M. P. W.; Smernik, R. J.; Wiesenberg, G. L. B.; Schmidt, M. W. I. Aromaticity and degree of aromatic condensation of char. Org. Geochem. 2015, 78, 135−143. (20) Harvey, O. R.; Herbert, B. E.; Rhue, R. D.; Kuo, L.-J. Metal Interactions at the Biochar-Water Interface: Energetics and StructureSorption Relationships Elucidated by Flow Adsorption Microcalorimetry. Environ. Sci. Technol. 2011, 45 (13), 5550−5556. (21) Benson, S. W., III Bond energies. J. Chem. Educ. 1965, 42 (9), 502. (22) Petrucci, R. H.; Herring, F. G.; Bissonnette, C.; Madura, J. D. General Chemistry: Principles and Modern Applications; Pearson College Division: Boston, 2011.
(23) Kuzyakov, Y.; Subbotina, I.; Chen, H.; Bogomolova, I.; Xu, X. Black carbon decomposition and incorporation into soil microbial biomass estimated by 14C labeling. Soil Biol. Biochem. 2009, 41 (2), 210−219. (24) Kuhnert, N.; Dairpoosh, F.; Yassin, G.; Golon, A.; Jaiswal, R. What is under the hump? Mass spectrometry based analysis of complex mixtures in processed food - lessons from the characterisation of black tea thearubigins, coffee melanoidines and caramel. Food Funct. 2013, 4 (8), 1130−1147. (25) Hertkorn, N.; Frommberger, M.; Witt, M.; Koch, B. P.; SchmittKopplin, P.; Perdue, E. M. Natural Organic Matter and the Event Horizon of Mass Spectrometry. Anal. Chem. 2008, 80 (23), 8908− 8919. (26) Singh, B. P.; Cowie, A. L. Long-term influence of biochar on native organic carbon mineralisation in a low-carbon clayey soil. Sci. Rep. 2014, 4, 3687.
8585
DOI: 10.1021/acs.est.6b01010 Environ. Sci. Technol. 2016, 50, 8578−8585
