Science of the Total Environment 598 (2017) 604–618

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Comparing current chemical methods to assess biochar organic carbon in

a Mediterranean agricultural soil amended with two different biochars
Irene Raya-Moreno a,⁎,1, Rosa Cañizares a,b,1, Xavier Domene a,b, Vicenç Carabassa a, Josep M. Alcañiz a,b
a
CREAF, Cerdanyola del Vallès 08193, Spain
b
Universitat Autónoma de Barcelona, Cerdanyola del Vallès 08193, Spain

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Resistant organic carbon (ROC) was

quantified by different chemical
methods in a soil amended with two
biochars.
• Biochars significantly increased stable
organic C content of an amended soil
• The amount and type of biochars added
were stable enough not to vary along
two years of field experimentation.
• Mild oxidation with potassium dichro-
mate was a successful as indirect meth-
od to estimate recalcitrant C fraction in
soil.
• Hydrogen peroxide oxidation was not a
suitable method to determine biochar C
in soil.

a r t i c l e i n f o a b s t r a c t

Article history: Several methods have been proposed to quantify biochar C recalcitrance but their suitability is questionable. The
Received 9 November 2016 aims of this work are: i) to compare the suitability of thermal or chemical oxidation and acid hydrolysis methods
Received in revised form 17 March 2017 to quantify biochar C-pool in a biochar-amended soil,
Accepted 18 March 2017
and ii) to calculate the biochar content in the soil through a mass balance derived from the obtained data.
Available online 25 April 2017
Two contrasted biochars from pine wood and corn cob remains were incorporated at a rate of 5 Mg C ha−1 to a
Editor: D. Barcelo sandy loam vineyard soil with neutral pH and low organic carbon content, in field conditions. The analytical
methods used to determine the oxidability and hydrolyzation of soil and biochar-C were: i) weight loss-on-
Keywords: ignition (LOI) at three temperatures (375 °C, 550 °C and 950 °C) for the assessment of organic matter, and ii)
Biochar dry-combustion (TOC), strong (sO) and mild (mO) acid potassium dichromate oxidations, acid hydrolysis
Soil (AH) and peroxide oxidation (PO) for the assessment of organic C-pools. mO mainly estimated the easy
Resistant organic carbon oxidisable organic fraction of soil. Resistant organic carbon (ROC), estimated as non-hydrolysable organic carbon

Abbreviations: PB, Pine biochar; ZB, Corn cob biochar; S, Control soil; S + PB, Soil amended with pine biochar; S + ZB, Soil amended with corn cob biochar; AH, Acid hydrolysis; LOI,
Loss-on-ignition; LPO, Loss-on-peroxide-hydrogen oxidation; mO, Mild potassium dichromate oxidation; PO, Peroxide-hydrogen oxidation; sO, Strong potassium dichromate oxidation;
TOC, Total organic carbon (by dry combustion); BOCTOC, Biochar (of pine or corn cob) organic carbon content estimated by TOC; BOCAH, Biochar (of pine or corn cob) organic carbon con-
tent estimated by AH; BOCLPO, Biochar (of pine or corn cob) organic carbon content estimated by LPO; BOCmO, Biochar (of pine or corn cob) organic carbon content estimated by mO; CF,
Correction factor for mineral losses based on LOI 550 °C; ROCAH, Resistant organic carbon values calculated from AH; ROCmO, Resistant organic carbon values calculated from mO; ROCPO,
Resistant organic carbon values calculated from PO.
⁎ Corresponding author at: Center for Ecological Research and Forestry Applications, (CREAF), Edifici C, Campus de la UAB, 08193 Bellaterra, (Cerdanyola del Vallès), Spain.
E-mail addresses: [email protected] (I. Raya-Moreno), [email protected] (R. Cañizares), [email protected] (X. Domene), [email protected] (V. Carabassa),
[email protected] (J.M. Alcañiz).
1
The first two authors contributed equally to this work.

http://dx.doi.org/10.1016/j.scitotenv.2017.03.168
0048-9697/© 2017 Elsevier B.V. All rights reserved.
 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 605

Soil organic matter by AH and as non-oxidisable by mO, led to similar values in control soil (5 g C kg−1 soil), whereas different ROC
Chemical oxidation values were obtained in soils amended with biochar (6–12 g C kg−1 soil). The suitability of these different
Acid hydrolysis methods as proxies to quantify biochar C was verified through a mass balance observing differences between
them. PO removes well native soil organic matter, but also attacks partially biochar's fraction, so an underestima-
tion exists. However, mO leaves intact biochar in the amended soil. Summarising, LOI, TOC and mO were the best
proxies for biochar-C quantification, especially the last one, somewhat clarifying the debate on this topic.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction The main aims of this field study, carried out in a soil amended with
two different biochars, were: i) to evaluate which of the widely used
Biochar has received much attention in recent years due to its prop- chemical methods for the assessment of soil organic matter (SOM) or
erties as a potential soil amendment (Lehmann and Joseph, 2015). Bio- SOC are the most suitable for discriminating resistant soil and biochar
char is a solid carbonaceous material obtained from the thermochemical organic carbon pools, and ii) to calculate the biochar content in soils
conversion of biomass in an oxygen-limited environment and intended through a mass balance derived from the available analytical data and
for use as soil amendment (International Biochar Inititiative, 2012). confirm if these methods are good proxies to quantify biochar.
Furthermore, the application of biochar to soil provides additional envi-
ronmental benefits, such as long-term C sequestration in soil (Shackley 2. Materials and methods
et al., 2013), compensating CO2 emissions (Fang et al., 2014) and
playing a key role in a global C-negative strategy (Cheng et al., 2006; Table 1 summarise and define the acronyms used in this paper.
Qin et al., 2016).
Biochar stability could vary depending on biomass feedstock and py- 2.1. Biochar characterisation
rolysis procedure, which influences aromatic C condensation (Lehmann
et al., 2006; Wang et al., 2015). Most biochar C is very stable due to its Two biochars were tested in this study, one obtained as a residue of
chemical structure, which is rich in aromatic C structures (Baldock and the gasification of mixed pine wood splinters (Pinus radiata and
Smernik, 2002; Calvelo Pereira et al., 2011; Singh et al., 2012), highly re- P. pinaster) at 600–900 °C (PB), and the other by slow pyrolysis of corn
sistant to microbial decomposition (Shindo, 1991; Cheng et al., 2008). cobs (Zea mays) at 450–500 °C (ZB). Both biochars were grounded and
However, biochar contains a minor labile fraction that consist mostly sieved to 2 mm prior to analysis and field application.
of carbohydrates, proteins and fatty acids that have not been completely Biochar C and H content were determined using a Flash 2000 C.E. El-
charred (Lehmann and Joseph, 2009). This minor organic fraction alone emental Analyzer (Thermo Fisher Scientific) at 1020 °C, N content by a
might not contribute to efficient C sequestration in soil due to its fast Flash EA 1112 Elemental Analyzer (Thermo Fisher Scientific) at
turnover rates, although it may play a key role in soil structure owing
to physical protection processes that might increase its residence time Table 1
in soil (Plaza et al., 2016). List of acronyms used in this paper by alphabetical order.
While there can be no doubt that the application of biochar to agri- General
cultural soils increases the recalcitrant fraction of soil organic carbon acronyms Definition
(SOC), and hence C sequestration in soil, the most suitable methods Biochars
for its assessment are still under debate. Various attempts have been PB Pine biochar
made to correlate analytical determinations -chemical, biological and ZB Corn cob biochar
physical methods- with SOC pools (Strosser, 2010; von Lützow et al., Treatment
2007) but, in the specific case of chemical methods, their validity has S Control soil
been questioned (Naisse et al., 2013). A variety of current thermal and S + PB Soil amended with pine biochar
chemical soil analysis methods have been used for this purpose as dry S + ZB Soil amended with corn cob biochar

combustion, loss-on-ignition (LOI), and loss-on-hydrogen peroxide ox- Methods

idation (LPO) being the most widespread methods to determine soil or- AH Acid hydrolysis (HCl, 100 °C, 17 h)
ganic matter (Gustafsson et al., 1997; Mikutta et al., 2005). On the other LOI Loss-on-ignition (375 °C, 550 °C, 950 °C)
LPO Loss-on-peroxide-hydrogen oxidation
hand, dichromate oxidation has been widely used to assess soil organic
mO Mild potassium dichromate oxidation (K2Cr2O7, 60 °C, 8 h)
carbon (e.g. Walkley and Black, 1934) and acid hydrolysis (AH) to dis- PO Peroxide-hydrogen oxidation
criminate the recalcitrant C fraction (Rovira and Vallejo, 2007). Howev- sO Strong potassium dichromate oxidation (K2Cr2O7, 150 °C, 10 min)
er, the accuracy of these methods for the quantification of biochars is TOC Total organic carbon (by dry combustion)
unclear due to the diverse composition of carbon pools and their vulner-
Specific
ability to chemical attack. As an example of this, acid dichromate oxida- acronyms Definition
tion has been suggested by several authors as an alternative to test
BOCTOC Biochar (of pine or corn cob) organic carbon content estimated
biochar C reactivity in soil (Calvelo Pereira et al., 2011; Knicker et al., by TOC
2007). This procedure is a modification of the broadly used classic BOCAH Biochar (of pine or corn cob) organic carbon content estimated
methods to estimate organic carbon (e.g. Walkley and Black, 1934; by AH
Mebius, 1960). The K2Cr2O7/H2SO4 concentration, temperature and BOCLPO Biochar (of pine or corn cob) organic carbon content estimated
by LPO
time of digestion are the key parameters which determine the oxidation
BOCmO Biochar (of pine or corn cob) organic carbon content estimated
degree because not all biochar C is always oxidised by dichromate by mO
(Rumpel et al., 2006). Calvelo Pereira et al. (2011) proposed acid dichro- CF Correction factor for mineral losses based on LOI 550 °C
mate oxidation as a method to evaluate the most reactive fraction of ROCAH Resistant organic carbon values calculated using AH
biochar, suggesting that resistant C to that oxidation could reflect the ROCmO Resistant organic carbon values calculated using mO
ROCPO Resistant organic carbon values calculated using PO
most stable fraction and the degree of biochar aromatisation.
606 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

1020 °C, S by ICO-OES spectrometry using a Varian 725-ES Radial ICP acid pre-treatment to assess biochar organic content without the soot
Optical Emission Spectrometer (Varian Inc.). Organic O was estimated fraction (LOI 375 °C) (Gustafsson et al., 1997; Poot et al., 2009). In the
by subtraction of the others elements, ash and the mineral O loss from second, the calcined samples were further heated at 550 °C for 5 h to re-
carbonates. Inorganic C was measured by the titrimetric method de- move soot (Gustafsson et al., 1997), hence allowing the complete oxida-
scribed by Wang et al. (2014) and used to calculate the organic C tion of the organic carbon fraction (LOI 550 °C). Finally, the samples
(TOC) by subtraction from the total C (TC). Molar concentrations of O, were subjected to 950 °C for 5 h, in order to remove carbonates and
C and H were used to calculate O:C and H:C ratios (Baldock and other mineral components (LOI 950 °C) (Santisteban et al., 2004). LOI
Smernik, 2002; Knicker et al., 2005). The pH of biochar was measured 550 °C values were corrected subtracting mineral losses due to water
in a 1:20 w/v water suspension, and electrical conductivity (EC) was and mineral components using a linear regression between LOI 550 °C
measured after filtering the same extract. The ash content was deter- and TOC when organic C was 0 (Y intercept) in these soil samples (cor-
mined as the weight of the residue of combustion by a muffle furnace rection factor, CF = 28.42 g kg−1).
at 950 °C during four hours. Loss-on-ignition at three temperatures,
loss-on-hydrogen peroxide oxidation, total carbon, acid potassium di- 2.5. Loss-on-hydrogen peroxide oxidation (LPO)
chromate oxidation, hydrochloric acid hydrolysis and inorganic carbon
were also determined in biochar samples by the methods described Samples were subjected to a slow and progressive oxidation with an
later. excess of hydrogen peroxide (Mikutta et al., 2005). Briefly, 4 g of oven-
dried soil samples were placed in a 100 ml Erlenmeyer flask and hydro-
2.2. Site and soil description gen peroxide (33%) was added several times until the oxidation reaction
finished (one week was required in this study). Then the remaining res-
The study was conducted in a 20-year-old vineyard grown on a idue was dried and accurately weighted to estimate organic matter
Fluventic Haploxerept soil (Soil Survey Staff, 2014) located in Vimbodí- losses. To estimate resistant organic carbon to hydrogen peroxide oxi-
Poblet (Catalonia, NE Spain), on a hillside with a slope of 8%. The parent dation (PO), the residue was analysed using a Flash EA 1112 Elemental
material is a quaternary alluvial deposit formed by a mixture of slates, Analyzer at 1020 °C.
sandstone, granodiorite and limestone gravels (64% w/w) mixed in a
clay-loam matrix. Average annual rainfall and air temperature for the
2.6. Total carbon (TC), total organic carbon (TOC) and inorganic carbon
area are 550 mm and 14.6 °C, respectively. The soil can reach a depth
(IC)
of 120 cm and almost all the initial carbonates have been dissolved
(around 1% remaining). The field bulk density of topsoil is 1.43 Mg
Total carbon content in soil samples was determined by elemental
m−3. The fine fraction (b 2 mm) is 36% (w/w). The Ap horizon has a
analysis using a Flash EA 1112 Elemental Analyzer at 1020 °C. Inorganic
sandy loam texture with a clay content of 15%, a neutral pH (7.2), a
C was measured by the titrimetric method described by Wang et al.
low organic carbon content (0.97%), and low cation exchange capacity
(2014). TOC of soil samples was calculated as a difference between TC
(7.1 cmolc kg−1) mainly saturated by calcium.
and IC.
2.3. Experimental design and sampling
2.7. Oxidation by potassium dichromate
Two biochars were applied in a unique dose, to field plots at
5 Mg C ha−1, equivalent to 6.5 g kg−1 in the b2 mm soil fraction, in a Oxidisable organic carbon with potassium dichromate (K2Cr2O7) in
randomised block design with three treatments (in triplicate): control acid media was determined using a strong (sO) or a mild oxidation
(S), soil amended with pine biochar (S + PB) or amended with corn (mO). Strong oxidation was conducted according Nelson and Sommers
cob biochar (S + ZB). This application rate is representative of a realistic (1996) with some modifications, briefly 10 ml of 66.7 mM K2Cr2O7
application dosage for an agricultural soil (Jeffery et al., 2011). Three were dissolved in a concentrated mixture of H2SO4/H3PO4 (1:1 v/v)
plots (10 × 8.8 m2) were set up per treatment, each with four Vitis vinif- and added to a glass tube containing 200–300 mg of grounded soil. The
era rows including ten plants/row. The biochar was weighted and uni- tubes were vortexed (15 s) and heated to 150 °C in a digestion block
formly applied to the soil surface and then was incorporated to the for 10 min. After cooling, 90 ml of distilled water were added and the ex-
soil by tilling to a 15-cm plough depth. The vineyard in this study was cess of dichromate (Cr2O−2 7 ) was back titrated with 0.2 N ammonium

managed using ecological agriculture practices, and ploughed three or iron (II) sulphate (Mohr's salt) using diphenylamine as indicator. The
four times per year to control weeds. Two years before, the vineyard mild oxidation was conducted according to Rumpel et al. (2006) without
was fertilised with composted cow manure, though not fertilised during HF acid pre-treatment. Briefly, 200 mg of grounded soil was mixed with
the experimental period of this study. Agrochemicals had not been 5 ml of potassium dichromate solution (0.1 M K2Cr2O7/2 M H2SO4) and
added with the exception of the regular Bordeaux mixture treatments oxidised at 60 °C for 8 h. Oxidised C was quantified by titration as already
for fungal disease control. described.
Soil samples were collected in July 2013, July 2014 and July 2015
after 2, 14 and 26 months of biochar application, respectively. Each sam- 2.8. Acid hydrolysis (AH)
ple was the result of eight random soil cores of 4 dm3 pooling, taken at a
depth of 0–10 cm (Ap horizon) within each plot (approximately 45 kg/ Biochar recalcitrance has been assessed by its resistance to acid hy-
sample). Soil samples were sieved to 5 mm in the field to separate drolysis by some authors (Nocentini et al., 2010), as this method is wide-
gravels, and then air-dried, sieved to 2 mm in the laboratory, and stored ly used for organic carbon recalcitrance in soils (Plante et al., 2006;
at 4 °C. A representative portion of each sample was grounded and Rovira and Vallejo, 2002; Silveira et al., 2008). Acid hydrolysis was car-
sieved to 0.02 mm to carry out the analyses as required. Laboratory ried out according to Rovira and Vallejo (2007) with few modifications:
analyses were carried out in triplicate. 20 ml of 6 M HCl were applied to 500 mg of ground sample for 17 h at
105 °C; the resulting slurry was then vacuum filtered and washed with
2.4. Loss-on-ignition (LOI) distilled water using a weighted Duran® n°2 filter crucible. The filtrate
was discarded whereas the non-hydrolysed residue remaining in the fil-
Loss-on-ignition (LOI), both of soil and biochar, was measured in a ter crucible was dried at 105 °C for 3 h. The organic C resistant to acid hy-
muffle furnace using ground samples, in triplicate, at three tempera- drolysis was determined by elemental analysis of the residue using a
tures. In the first, samples were subjected to 375 °C for 18 h without Flash EA 1112 Elemental Analyzer at 1020 °C.
 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 607

2.9. Mass balance calculations acid hydrolysis was around 9% of TOC. Loss-on-peroxide oxidation or-
ganic matter (LPO) was slightly higher in corn cob than in pine biochar
For each of the different methods used in this work a mass balance even though values were minimal in both materials (b2.5% of TOC).
was conducted to estimate biochar organic carbon in biochar
amended-soils, subtracting the labile and resistant fraction of control 3.2. Soil organic matter content by loss-on-ignition (LOI) and loss-on-
soil. The respective equations used could be seen in the appendix section. hydrogen peroxide oxidation (LPO)

2.10. Data analysis Soil LOI values are shown in Table 3. As expected, significant differ-
ences were detected between control soil and that amended with bio-
Before statistical analysis, data were tested for homogeneity of vari- chars in LOI at 375 °C (p = 0.002), 550 °C (p = 0.002) and 950 °C
ance using the Shapiro-Wilk test. In each treatment, global significance (p = 0.002) (Table 3). However, no significant differences between
tests on the effects of each biochar type were performed using Two- sampling times were found, meaning that no relevant quantitative
Way RM ANOVA tests. Significant differences between treatments in a changes had occurred along two years after biochar application.
given sampling time, and between samplings in treatments, were Regarding LPO, no significant differences were detected between
assessed by using the Bonferroni test at a probability level of 0.05. All control soil and soil amended with biochars (p = 0.143). When LPO
tests were carried out using R software (R Core Team 2013). See p- was compared to LOI 375 °C (Table 3), it was clear that approximately
values of the analysis in the appendix section. half of SOM was oxidised by hydrogen peroxide in control soil, but
less in biochar amended soil.
3. Results
3.3. Total carbon (TC), total organic carbon (TOC) and inorganic carbon
3.1. Composition and chemical properties of pine and corn cob biochars
(IC) in soil

Table 2 shows the main chemical composition and properties of the

No significant variations in TC or TOC in the soil samples content
two biochars tested.
were detected over time in S, S + PB and S + ZB treatment (Table 4).
As expected, high pH and relatively low soluble salts values were ob-
The inorganic C content was very low and no significant differences
tained. The organic carbon content was high and similar, and the inor-
were found between treatments. Since both biochars contributes with
ganic C content (carbonates) was low in both biochars. Both met the
an even low carbonate content compared to soil (Tables 2 and 4) it
criteria of organic C content and H/C molar ratio b 0.7 (International
should mostly come from the latter. As expected, significant differences
Biochar Inititiative, 2012), although the H/C ratio was slightly lower in
in TOC were detected between control soil and soil amended with bio-
pine than in corn cob biochar while O/C ratio was the same. Corn cob
chars (p ≤0.001).
biochar showed higher N and H concentration, related to biochar raw
material and pyrolysis type. Slightly higher S values were observed in
3.4. Strong and mild dichromate oxidation of soil organic matter
pine than corn cob biochar. Organic matter estimated by LOI had a sim-
ilar thermochemical profile and ash content was approximately 9% in
Potassium dichromate strong oxidation (sO) values accounted for
both biochars.
approximately 90% of the total C concentration (Table 5). Therefore,
Regarding biochar C chemical stability estimated by different
this method was sensitive to the increase in C content after biochar ad-
methods, noticeable differences were observed between them. More
dition (p = 0.002). No significant differences were observed between 2,
specifically, when strong oxidation was conducted, 74% of TOC was
14, 26 months, either for soil amended with pine or corn cob biochars,
quantified in pine biochar whereas just 30% was measured in corn cob
once again showing the lack of variation in the whole organic carbon
biochar. Carbon losses by mild dichromate oxidation (mO) were ap-
content occurring over the two-year period.
proximately 6% of TOC in both biochars. Resistant C determined by
When soil samples were submitted to mild oxidisation (mO), a less
aggressive method for determining the labile C fraction (Table 5), no
Table 2
significant differences were observed between treatments at any time
Elemental analysis, molar ratios and chemical properties of biochar from pine wood (PB)
and corn cob (ZB). Total carbon (TC), Total nitrogen (TN), hydrogen (H), Total organic ox- (p = 0.798). In control soils oxidisable carbon estimated by mO repre-
ygen (O), O/C and H/C molar ratios, electrical conductivity (EC), ash, pH, sulphur (S), sented approximately 50% of TOC whereas in biochar-amended soils
weight loss-on-ignition (LOI) and weight loss-on-peroxide oxidation (LPO), inorganic car- only 30% of TOC.
bon (IC), organic carbon destroyed by strong potassium dichromate oxidation (sO), or
mild potassium dichromate oxidation (mO), and organic carbon resistant to acid hydroly-
sis (AH). 3.5. Organic carbon recalcitrance estimated by acid hydrolysis (AH)

Biochar feedstock Pine (BP) Corn cob (ZB)

The non-hydrolysable C content was dependent on the biochar addi-
pH (water, 1:20 w:v) 11.5 ± 0.04 10.3 ± 0.04 tion and significant differences were detected (p = 0.001) between
EC (dS m−1 25 °C) 0.69 ± 0.02 2.54 ± 0.5
control soil and soil amended with biochars at any time, but not be-
TC (g kg−1) 793.4 785.8
IC (g kg−1) 4.0 ± 0.07 2.7 ± 0.06 tween pine and corn cob biochar (Fig. 1).
TN (g kg−1) 0.20 6.80
H (g kg−1) 12.2 19.1 3.6. Comparative resistance of organic carbon (ROC) to chemical reagents
S (g kg−1) 1.48 0.64
O (g kg−1) 90.15 89.36
Ash (g kg−1) 91.9 91.1 The concentration of C in the residue of peroxide oxidation (S:
H/C 0.19 0.29 1.56 g kg−1; S + PH: 5.6 g kg−1; S + ZH: 7.36 g kg−1) and acid hydro-
O/C 0.11 0.11 lysis (Fig. 1), measured by elemental analysis, were used to determine
LOI (g kg−1) 375 °C 885.9 ± 0.2 891.7 ± 0.3
ROC. The ratio ROC/TOC estimated by three different methods was cal-
550 °C 892.1 ± 0.3 897.9 ± 0.2
950 °C 905.9 ± 0.2 917.7 ± 0.2 culated (data not shown). ROCPO was b20% of TOC in control soil, and
−1
LPO (g kg ) 0.95 ± 0.87 19.55 ± 3.84 b40% in biochar-amended one. On the other hand, ROCmO was lower
−1
sO (g kg ) 590.64 ± 32.57 235.27 ± 40.09 than ROCAH in control soil, while in biochar-amended soil ROC mea-
mO (g kg−1) 46.82 ± 1.59 43.69 ± 3.64 sured by mO and AH was approximately 70% of TOC. However, no sig-
AH (g kg−1) 84.05 ± 2.5 65.66 ± 8.46
nificant differences were observed over time (data not shown).
608 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

Table 3
Soil organic matter content estimated by loss-on-ignition (LOI) at different temperatures and loss-on-peroxide hydrogen oxidation (LPO) in a soil treated with pine (S + PB) and corn cob
(S + ZB) biochar compared to control (S) at three sampling times (2, 14 and 26 months) after biochar application. Mean values and standard deviation of three replicates are shown.
Values sharing capital letters indicate the lack of significant differences between sampling times within treatments (p b 0.05), while equal small letters imply the lack of differences be-
tween treatments within times (p b 0.05). If letters are not given, no significant differences were observed.

LOI 375 °C LOI 550 °C LOI 950 °C LPO

Biochar treatment (g kg−1) (g kg−1) (g kg−1) (g kg−1)

2 months
S 27.4 ± 1.34 a 41.67 ± 1.31 a 50.22 ± 1.39 a 11.67 ± 1.36 A
S + PB 34.99 ± 3.46 b 50.02 ± 3.79 b 58.84 ± 4.01 b 14.51 ± 1.30 A
S + ZB 36.80 ± 1.86 b 52.15 ± 1.55 b 61.60 ± 1.38 b 12.85 ± 1.30 A

14 months
S 28.18 ± 0.92 a 43.17 ± 1.32 a 51.96 ± 1.59 a 11.68 ± 0.53 A
S + PB 35.75 ± 1.05 b 51.47 ± 1.37 b 60.45 ± 1.63 b 13.67 ± 0.82 A
S + ZB 33.78 ± 2.47 b 50.72 ± 3.18 b 60.09 ± 3.37 b 13.01 ± 0.57 A

26 months
S 28.50 ± 0.80 a 42.70 ± 1.0 a 51.88 ± 1.35 a 15.24 ± 0.58 B
S + PB 34.43 ± 2.58 b 49.29 ± 2.56 b 58.85 ± 2.82 b 15.44 ± 1.98 B
S + ZB 33.20 ± 0.98 b 48.77 ± 1.36 b 56.30 ± 1.47 b 15.83 ± 0.86 B

If we compare methods, peroxide oxidation gives lower proportion terms of resistance to biological (Sun and Lu, 2014), chemical (Rovira
of resistant C than mO or AH (Fig. 2). In any case, soil amended with and Vallejo, 2002) and physical decomposition (Li et al., 2016).
both types of biochar showed higher ROC values compared to the con-
trol soil regardless of the method used (Fig. 2). 4.1. Suitable methods for total organic matter or carbon quantification

Elemental analysis was a reliable method to determine all sources of

3.7. Biochar organic carbon (BOC) quantification by different methods
carbon (TC) in soil, making it possible to estimate total organic carbon
(TOC) when inorganic carbon (IC) was subtracted.
As can be seen in Table 6, lower significant values of BOC were ob-
Weight loss-on-ignition at 375 °C corresponds mainly to organic
tained when estimation was performed using PO data (explained in ap-
matter although some crystalline water or mineral losses are included
pendix section) compared to the other methods. However, similar
(Santisteban et al., 2004); whereas weight loss at 550 °C overestimates
results were obtained using TOC, LOI and mO, agreeing with the nomi-
organic matter due to dehydration of clays, hydroxides and other min-
nal dose of biochar C applied (5 Mg C ha−1). A moderate underestima-
erals (Heiri and Lotter, 2001). Weight loss between 550 °C and 950 °C
tion is obtained with AH method.
estimated mainly CO2 released from carbonates, crystalline water and
volatile mineral elements losses (approximately 1% of weight losses in
4. Discussion this soil), showing no significant differences between treatments. At
each ignition temperature a correction factor (CF) for mineral losses is
In spite of a very different feedstock composition and pyrolysis proce- needed. In this work, the CF for LOI at 550 °C represents the half of the
dure, no great chemical composition differences were detected between weight losses at this temperature with a value of 28.42 g kg−1.
both biochars. In our study, pine biochar showed a slightly lower H/C Biochar loss-on-ignition depends on the feedstock and pyrolysis pro-
ratio than corn cob biochar. A lower H/C ratio suggested a more fused ar- cess used in biochar production. Brändli et al. (2009) observed that bio-
omatic structure (Baldock and Smernik, 2002; Knicker et al., 2005). This char is not completely thermo-oxidised at 375 °C, suggesting the higher
aromatic structure might confers biochars a high degree of stability in the temperature applied to obtain biochar, the higher the temperature
needed for its complete combustion. However, Gelinas et al. (2001)
found biochar to combust completely at 375 °C. In contrast, Brändli
Table 4
Soil total carbon (TC), inorganic carbon (IC), and total organic carbon (TOC), in a soil treat- et al. (2009) reported that higher temperatures were required. In our
ed with pine biochar (S + PB) and corn cob biochar (S + ZB) compared to control (S) at study differences in OM quantification at three different temperatures
three sampling times (2, 14 and 26 months) after biochar application. Mean values and (LOI 375 °C, 550 °C and 950 °C) were b2% in both pine and corn cob bio-
standard deviation of three replicates are shown. Values sharing capital letters indicate chars. Even though, we recommend the use of LOI 550 °C subtracting
the lack of significant differences between sampling times within treatments (p b 0.05),
while equal small letters imply the lack of differences between treatments within times
mineral losses (CF) to ensure that all biochar organic matter in amended
(p b 0.05). If letters are not given, no significant differences were observed. soils is completely thermo-oxidised. Moreover, Santisteban et al. (2004)
pointed out LOI 550 °C as a qualitatively reliable method to detect bio-
Total organic
Total carbon (TC) Inorganic carbon (IC) carbon (TOC)
char presence and possible losses in the field. Furthermore, Koide et al.
Biochar treatment (g kg−1) (g kg−1) (g kg−1) (2011) considered it as an appropriate method to quantify biochar in
the field.
2 months
S 10.72 ± 0.79 a 0.94 ± 0.25 9.77 ± 0.54 a
S + PB 17.98 ± 2.67 b 1.12 ± 0.52 16.87 ± 2.56 b 4.2. Suitable methods for the discrimination of resistant organic carbon
S + ZB 21.33 ± 1.50 b 1.12 ± 0.95 20.21 ± 2.37 b

14 months Hydrogen peroxide was a good oxidant agent for removing native OM
S 11.41 ± 0.86 a 1.43 ± 0.99 9.99 ± 1.03 a from soil samples, leaving the most resistant carbon, for example biochar
S + PB 18.97 ± 0.40 b 1.33 ± 1.18 17.65 ± 1.21 b (Liang et al., 2006; Mikutta et al., 2005). This method was easy to imple-
S + ZB 18.53 ± 2.98 b 1.35 ± 0.37 17.15 ± 3.31 b ment although long oxidation times are required and accurate weighting
26 months is needed to reduce gravimetric errors. Hydrogen peroxide oxidise a small
S 10.29 ± 0.73 a 0.49 ± 0.29 9.80 ± 0.85 a fraction of both biochars (Table 2) due to the high amount of resistant car-
S + PB 16.81 ± 0.89 b 0.84 ± 0.88 15.97 ± 1.23 b bon in these materials. The high hydrophobicity of pine biochar (data not
S + ZB 16.60 ± 1.03 b 0.96 ± 1.11 15.64 ± 2.03 b
shown), could reduce peroxide oxidation reactions. The hydrophobicity
 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 609

Table 5
Organic carbon resistant to a strong and mild K2Cr2O7 oxidation of a control soil (S) and soil treated with pine (S + PB) or corn cob (S + ZB) biochar at 2 months, 14 months and
26 months after biochar application. Mean values and standard deviation of three replicates. Values sharing capital letters indicate the lack of significant differences between sampling
times within treatments (p b 0.05), while equal small letters imply the lack of differences between treatments within times (p b 0.05). If letters are not given, no significant differences
were observed.

Strong oxidation (sO) Ratio Mild oxidation (mO) Ratio

Biochar treatment (g kg−1) sO/TOC (g kg−1) mO//TOC

2 months
S 9.32 ± 0.93 ABa 0.95 ± 0.07 5.97 ± 0.60 0.61 ± 0.03
S + PB 15.38 ± 1.85 b 0.91 ± 0.06 5.78 ± 0.14 0.35 ± 0.06
S + ZB 15.43 ± 0.41 b 0.77 ± 0.07 5.98 ± 0.05 0.30 ± 0.04

14 months
S 11.59 ± 1.68 Aa 0.95 ± 0.07 4.66 ± 1.05 0.46 ± 0.07
S + PB 16.07 ± 1.68 b 0.91 ± 0.05 5.14 ± 0.82 0.29 ± 0.03
S + ZB 14.16 ± 1.86 ab 0.83 ± 0.06 5.43 ± 0.42 0.33 ± 0.07

26 months
S 7.55 ± 1.18 Ba 0.77 ± 0.08 5.32 ± 0.38 0.54 ± 0.02
S + PB 13.70 ± 1.84 b 0.85 ± 0.05 5.72 ± 0.58 0.36 ± 0.04
S + ZB 14.43 ± 0.93 b 0.93 ± 0.09 4.75 ± 0.33 0.31 ± 0.04

can protect biochar from chemical attack (von Lützow et al., 2007), which Rumpel et al. (2006) used a similar method for biochar quantification in
might at least partly explain the lower oxidability of pine biochar com- soil because of dichromate's inability to oxidise some resistant mole-
pared to the corn cob type (Table 2). Moreover, biochar's physical charac- cules from biochar and black carbon. Furthermore, these authors also
teristics could change in the laboratory after exposition to hydrogen applied a hydrofluoric acid pre-treatment to liberate the organic matter
peroxide (Huff and Lee, 2016), and in the field by weathering, for exam- occluded in microagregates or in the interlaminar spaces of clays. How-
ple, by wetting and drying cycles in soil such as the reported increase in ever, the hydrofluoric acid was not used in our experiment so an under-
hydrophilicity of hydrophobic coal (Spokas et al., 2014; Velasco-Molina estimation of oxidisable organic matter could exist. Nevertheless, as
et al., 2016). Considering these observations, biochar organic carbon biochar was quantified by subtracting that of the control, mass balance
could be underestimated using this method (Table 6). is comparatively correct. Our results in biochar-amended soil also sup-
Total organic carbon quantification with dichromate strong oxida- port the view that the partial oxidation produced by mO only quantifies
tion method was lower than TOC estimated by elemental analysis, be- labile OC, for the most part from native soil organic matter in soil sam-
cause dichromate did not oxidise all organic carbon present in the ples. On the other hand, sO estimated 90% of TOC in soil amended
sample (Brändli et al., 2009). Biochar organic carbon (BOC) oxidation with pine biochar while 80% of TOC in that amended with corn cob bio-
depends on temperature, reaction time, concentration of reagents and char owing to the latter being less dichromate-oxidisable (see Table 2).
the degree of carbonisation of the biochar (Knicker et al., 2007). Several Such results indicate an underestimation of biochar carbon due to:
years ago, similar methods based on dichromate oxidation (Nelson and i) only partial oxidation of resistant organic carbon and, ii) reduced
Sommers, 1996) were widely used for organic C quantification in soils chemical attack capacity associated with biochar buoyancy when
due to the capacity of dichromate to oxidise resistant molecules such mixed with dichromate solution, since some particles remained on the
as lignins, pectins, tannic acids, and humic acids (Ball, 1964). However, Pyrex tube walls after agitation and digestion.

Fig. 1. Resistant organic carbon (ROCAH) to acid hydrolysis (AH) of a control soil (S) and soil treated with pine (S + PB) or corn cob (S + ZB) biochar, 2 months, 14 months and 26 months
after biochar application. Values sharing capital letters indicate the lack of significant differences between sampling times within treatments (p b 0.05), while equal small letters imply the
lack of differences between treatments within times (p b 0.05). If letters are not given, no significant differences were observed.
610 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

Fig. 2. Resistant organic carbon (ROC) content in a biochar-amended soil estimated by three methods: acid hydrolysis (AH); loss-on-peroxide hydrogen oxidation (LPO) determined as
differences between TOC and LPO; mild dichromate oxidation (mO) determined as differences between TOC and mO of control soil (S) and biochar-amended soil with pine (S + PB)
and corn cob (S + ZB) biochar. Mean values and standard deviation of three replicates at three sampling times (2, 14 and 26 months) after biochar application are shown. Values
sharing capital letters indicate the lack of significant differences between (p b 0.05) methods within treatments, while equal small letters imply the lack of differences between
treatments within the same methodology (p b 0.05). If letters are not given, no significant differences were observed.

Easily oxidisable C quantified by mO was approximately 50% of the ROCAH in residue of acid hydrolysis. Soil amended with biochar showed
TOC of the soil, as the ratio mO/TOC of control soil treatment indicated higher ROC values compared to the control soil, estimated by PO, mO
(Table 5). Yet it should be noted that this method is also suitable for indi- and AH, indicating that biochar increased the ROC pool in soil (Fig. 2).
rect estimation of the resistant carbon fraction of soil. All that said, mild Comparing ROCPO, ROCmO and ROCAH to TOC ratio in soil control, it
oxidation is a suitable proxy for estimating the amount of resistant organ- was observed that hydrogen peroxide was able to oxidise more soil or-
ic carbon added with biochar amendments by subtraction from the TOC. ganic carbon than mild dichromate oxidation or acid hydrolysis giving a
Acid hydrolysis had also been used to estimate the resistant organic lower ROC ratio value. Apparently, the ROCAH/TOC ratio was similar in
carbon fraction in the soil samples as it removes labile compounds such control soils and those treated with biochar, so one might reasonably
as carbohydrates and proteins by disruption of glycosidic and peptide think that biochar is partially sensitive to acid hydrolysis attack or that
bonding and leaves others such as alkyl and aryl chains (Leavitt et al., a portion of soil organic matter (such as lignin) was not hydrolysed, al-
1996; von Lützow et al., 2007). In our soil samples, ROC/TOC ratio esti- though in absolute values AH results (Fig. 1) were higher in biochar-
mated by acid hydrolysis showed higher values, mainly in control soil, treated soils. Nevertheless, it was found that biochar application nearly
compared to the rest of the methods. This behaviour could be explained doubled the amount of ROC in soil (Fig. 2). ROC content in soil obtained
due to acid hydrolysis is unable to break lignin molecules (Leavitt et al., by mO and AH was very similar, in contrast with ROC estimated by per-
1996), thus increasing the amount of resistant carbon in the soil resi- oxide oxidation which was lower. This finding supports the study of
dues. Moreover, according with Greenfield et al. (2013) carbohydrate Mikutta et al. (2005), in which different chemical agents were com-
destruction (by hydrolysis) derived components can be converted into pared, with hydrogen peroxide achieving the most efficient oxidation
new and more stable forms compounds (e.g. furfurals) which increase in removing labile soil organic matter and leaving the most resistant
its resistance to acid hydrolysis. ones. Several authors have described the efficacy of the PO method for
non-resistant soil organic matter oxidation (Huff and Lee, 2016; Liang
4.3. Estimation of ROC in amended soils et al., 2006; Mikutta et al., 2005), leaving the most resistant SOM pool
intact (Liang et al., 2006). However, when these methods were com-
Resistant organic C-fraction (ROC) was estimated and compared pared it was observed that PO was the most efficient method used in
using three different methods: i) ROCPO in the residue of peroxide oxi- this study to remove native soil organic matter, although a small frac-
dation; ii) ROCmO remaining after mild dichromate oxidation and iii) tion of biochar was partially oxidised. Also, mild dichromate oxidation

Table 6
Pine (PB) and corn cob (ZB) biochar organic carbon (BOC) content in amended soil estimated by different methods (TOC, LOI, LPO, mO and AH). Nominal biochar dose, 5 Mg C ha −1soil.
Mean values and standard deviation of three replicates of three sampling times are showed. Differences within a row, followed by the same small letter, are not significant between
methods (p b 0.05). No significant differences between biochar type (p b 0.05) were observed (within a column).

BOCTOC BOCLOI BOCPO BOCmO BOCAH

Biochar treatment (Mg C ha−1soil) (Mg C ha−1soil) (Mg C ha−1soil) (Mg C ha−1soil) (Mg C ha −1soil)

Pine (PB) 5.39 ± 0.58a 4.96 ± 1.64ab 2.67 ± 0.67b 5.21 ± 1.37a 4.37 ± 1.63ab
Corn cob (ZB) 6.03 ± 1.83a 5.01 ± 1.76ab 4.47 ± 2.43b 5.98 ± 2.13a 4.41 ± 2.13ab
 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 611

method (mO) was assumed to quantify mostly the native soil organic biochar organic carbon in soil were found, thus supporting the efficiency
carbon fraction, leaving biochar intact (Knicker et al., 2007; Naisse of biochar as a strategy to C storage in soil.
et al., 2013; Rumpel et al., 2006; Velasco-Molina et al., 2016), which
has been confirmed in our study. 5. Conclusions

4.4. Suitability of chemical methods for the quantification of biochar in The estimation of biochar using a mass balance in amended soils
amended soils needs the combination of (i) harsh methods to quantify the total organic
matter or carbon in amended and control soils, and (ii) milder, less ag-
The estimation of biochar content in amended soils based on chemi- gressive methods to quantify (labile) native soil organic matter com-
cal parameters that measure the concentration of organic C variables re- bined with total organic matter or carbon determination. As this
lies on a good knowledge of the applied biochars and its homogenous regard, loss-on-ignition and TOC were suitable methods to biochar's
application. In agricultural soils requires effective tilling operations and quantification when a control soil was available. However, mild dichro-
a moderate spatial variability of soil properties. When trying to apply mate oxidation combined with TOC was the most adequate procedure
this strategy to soils affected by wildfires, charcoal quantification could for biochar resistant organic carbon determination also comparing with
be a handicap. a control soil. Special attention will need to be paid to acid hydrolysis
In our work, the ideal method should (i) oxidise all soil native organic and loss-on-peroxide oxidation which might underestimate the biochar
matter, and (ii) leave biochar intact. No one of these methods used met amount. Therefore, as mild oxidation leaves practically intact biochar or-
completely these assumptions, but obtained estimations of biochar or- ganic carbon, it can be considered the best proxy to indirectly quantify
ganic carbon in soil reasonably agree with biochar nominal application biochar in amended soils.
(5 Mg C ha−1), with exception of PO method (Table 6). This is due as hy-
drogen peroxide oxidation removes well soil organic matter, but also par- Acknowledgements
tially attacks biochar, so an underestimation exists. However, mild
dichromate oxidation leaves quite intact biochar in amended soil. For This research was funded by MEDICHAR (ref. AGL2012-40037-
that reason, based on our results, mild oxidation can be considered the C02-01) and FERTICHAR (ref. AGL2015-70393-R) projects of the
best proxy to quantify indirectly biochar organic carbon in amended soils. Spanish Ministry of Economy and Competitiveness. The authors
Furthermore, our results provide evidence that, during two years wish to thank Stefania Mattana (CREAF) for providing field work
after biochar application in the field, no noticeable changes or losses of support.

Appendix A

A.1. Mass balance calculations

A mass balance of different organic fractions was performed to compare the different methods used to estimate resistant soil and biochar organic
matter (ROM and BOM), and organic carbon (ROC and BOC) as follows.

A.1.1. Resistant organic carbon (ROC) to chemical agents

The resistant organic carbon (ROC) was quantified by three methods corresponding to three different chemical attacks:
i) From the peroxide oxidation residue (ROCPO) by elemental analysis.
ii) From the carbon remaining after mild dichromate oxidation (ROCmO).

TOC−LSOCmO ¼ ROCmO ð1Þ

TOC: total organic carbon values estimated by TOC.

LSOCmO: labile organic carbon values estimated by mO.
ROCmO: resistant organic carbon values to mO.

iii) From the residue of acid hydrolysis (ROCAH) by elemental analysis.

A.1.2. Biochar quantification based on TOC

Differences in TOC between biochar treatments and control soil made it possible to determine the biochar C abundance in soil.

BOCTOC ¼ TOCSþB –TOCS ð2Þ

BOC TOC: biochar of pine or corn cob organic carbon values estimated by TOC.
TOCS + B: total organic carbon values of biochar-amended soil estimated by TOC.
TOCS: total organic carbon values of control soil estimated by TOC.

A.1.3. Biochar quantification based on LOI

LOI CFS + B minus LOI CFS were used to determine biochar organic matter (BOMLOI) in the samples. Then biochar organic carbon (BOCLOI) was cal-
culated as follows:

BOCLOI ¼ BOMLOI  F ð3Þ

BOC LOI: biochar of pine or corn cob organic carbon values estimated by LOI.
BOMLOI: biochar pine or corn cob organic matter values of biochar-amended soil estimated by LOI 550 °C.
612 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

F: Ratio OC/OM of biochar (F = 0.89 in pine biochar and F = 0.87 in corn cob biochar).

A.1.4. Biochar quantification based on PO method

Biochar organic carbon (BOCPO) was calculated subtracting resistant native soil organic C from peroxide oxidation of control soil (RSOCPO-S) of
those samples amended with biochars (RSOC PO − S + B), Eq. (4), as follows:

BOCPO ¼ RSOCPO−SþB −RSOCPO−S ð4Þ

BOCPO: Biochar of pine or corn cob organic carbon values estimated by PO.õ
RSOCPO − S + B: Resistant soil organic carbon values of biochar-amended to PO.
RSOCPO − S: Resistant soil organic carbon values of control soil to PO.

A.1.5. Biochar quantification based on mild dichromate oxidation method

Resistant soil organic carbon (RSOCmO − S) fraction was calculated as the difference between TOC and labile soil organic carbon estimated by mO
values (LSOCmO − S) of control soil samples, Eq. (5). Secondly, the pool of resistant organic carbon of control soil which was added to labile soil or-
ganic carbon of biochar-amended soil (LSOCmO – S + B) values, as this part of resistant organic carbon of soil was not quantified with mO analyses,
Eq. (6). Finally, soil organic carbon (SOCmO) estimation was subtracted from total organic carbon values (TOCS + B) to estimate biochar organic car-
bon (BOCmO) Eq. (7) as follows:

RSOCmO−S ¼ TOCS −LSOCmO−S ð5Þ

SOCmO ¼ LSOCmO−SþB þ RSOCmO−S ð6Þ

BOCmO ¼ TOCSþB − SOCmO ð7Þ

RSOCmO-S: resistant soil organic carbon values of control soil estimated by mO.
TOCS: total organic carbon values of control soils estimated by TOC.
LSOCmO-S: labile soil organic carbon values of control soil estimated by mO.
SOCmO: soil ORGANIC carbon values estimated by mO.
LSOCmO – S + B: Labile soil organic carbon values of biochar-amended samples estimated by mO.
BOCmO: Biochar pine or corn cob organic carbon values estimated by mO.
TOCS + B: Total organic carbon values of the biochar-amended soil by TOC.

A.1.6. Resistant biochar fraction estimation based on acid hydrolysis (AH)

Biochar organic carbon (BOCAH) was calculated subtracting resistant soil organic C from acid hydrolysis (RSOCAH-S) of those samples amended
with biochars (RSOC AH – S + B), Eq. (8), as follows:

BOCAH ¼ RSOCAH–SþB −RSOCAH−S ð8Þ

BOCAH: biochar of pine or corn cob organic carbon values estimated by AH.
RSOCAH – S + B: resistant soil organic carbon values of biochar-amended to AH.
RSOCAH – S: resistant soil organic carbon values of control soil to AH.

A.2. Statistics

A.2.1. Loss-on-ignition (LOI)

ANOVA RM: LOI 375

numDF denDF F-value p-value

(Intercept) 1 16 4679.525 0.000

Time 2 16 0.767 0.481
Treatment 2 6 22.748 0.002

Bonferroni test: LOI 375

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months 0.493 0.788 20.250 0.626 1.000

S,2 months - S,26 months 1.019 0.788 20.250 1.293 1.000
S,2 months - S + PB,2 months −7.030 1.289 13.500 −5.454 0.003
S,2 months - S + ZB,2 months −6.567 1.289 13.500 −5.095 0.007
S,14 months - S,26 months 0.526 0.788 20.250 0.668 1.000
S,14 months - S + PB,14 months −7.030 1.289 13.500 −5.454 0.003
S,14 months - S + ZB,14 months −6.567 1.289 13.500 −5.095 0.007
S,26 months - S + PB,26 months −7.030 1.289 13.500 −5.454 0.003
S,26 months - S + ZB,26 months −6.567 1.289 13.500 −5.095 0.007
S + PB,2 months - S + PB,14 months 0.493 0.788 20.250 0.626 1.000
S + PB,2 months - S + PB,26 months 1.019 0.788 20.250 1.293 1.000
S + PB,2 months - S + ZB,2 months 0.463 1.289 13.500 0.359 1.000
 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 613

(continued)

Bonferroni test: LOI 375

Contrast Estimate SE df t.ratio p-value

S + PB,14 months - S + PB,26 months 0.526 0.788 20.250 0.668 1.000

S + PB,14 months - S + ZB,2 months −0.030 1.511 25.996 −0.020 1.000
S + PB,14 months - S + ZB,14 months 0.463 1.289 13.500 0.359 1.000
S + PB,26 months - S + ZB,26 months 0.463 1.289 13.500 0.359 1.000
S + ZB,2 months - S + ZB,14 months 0.493 0.788 20.250 0.626 1.000
S + ZB,2 months - S + ZB,26 months 1.019 0.788 20.250 1.293 1.000
S + ZB,14 months - S + ZB,26 months 0.526 0.788 20.250 0.668 1.000

ANOVA RM: LOI 550

numDF denDF F-value p-value

(Intercept) 1 16 7863.007 0.000

Time 2 16 1.865 0.187
Treatment 2 6 23.835 0.001

Bonferroni test: LOI 550

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months −0.506 0.773 20.250 −0.655 1.000

S,2 months - S,26 months 1.023 0.773 20.250 1.325 1.000
S,2 months - S + PB,2 months −7.744 1.459 13.500 −5.308 0.004
S,2 months - S + ZB,2 months −8.029 1.459 13.500 −5.504 0.003
S,14 months - S,26 months 1.530 0.773 20.250 1.980 1.000
S,14 months - S + PB,14 months −7.744 1.459 13.500 −5.308 0.004
S,14 months - S + ZB,14 months −8.029 1.459 13.500 −5.504 0.003
S,26 months - S + PB,26 months −7.744 1.459 13.500 −5.308 0.004
S,26 months - S + ZB,26 months −8.029 1.459 13.500 −5.504 0.003
S + PB,2 months - S + PB,14 months −0.506 0.773 20.250 −0.655 1.000
S + PB,2 months - S + PB,26 months 1.023 0.773 20.250 1.325 1.000
S + PB,2 months - S + ZB,2 months −0.286 1.459 13.500 −0.196 1.000
S + PB,14 months - S + PB,26 months 1.530 0.773 20.250 1.980 1.000
S + PB,14 months - S + ZB,14 months −0.286 1.459 13.500 −0.196 1.000
S + PB,26 months - S + ZB,26 months −0.286 1.459 13.500 −0.196 1.000
S + ZB,2 months - S + ZB,14 months −0.506 0.773 20.250 −0.655 1.000
S + ZB,2 months - S + ZB,26 months 1.023 0.773 20.250 1.325 1.000
S + ZB,14 months - S + ZB,26 months 1.530 0.773 20.250 1.980 1.000

ANOVA RM: LOI 950

numDF denDF F-value p-value

(Intercept) 1 16 8864.376 0.000

Time 2 16 0.598 0.562
Treatment 2 6 22.092 0.002

Bonferroni test: LOI 950

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months −0.613 0.748 20.250 −0.820 1.000

S,2 months - S,26 months 0.208 0.748 20.250 0.279 1.000
S,2 months - S + PB,2 months −8.023 1.640 13.500 −4.892 0.010
S,2 months - S + ZB,2 months −8.976 1.640 13.500 −5.473 0.003
S,14 months - S,26 months 0.821 0.748 20.250 1.099 1.000
S,14 months - S + PB,14 months −8.023 1.640 13.500 −4.892 0.010
S,14 months - S + ZB,14 months −8.976 1.640 13.500 −5.473 0.003
S,26 months - S + PB,26 months −8.023 1.640 13.500 −4.892 0.010
S,26 months - S + ZB,26 months −8.976 1.640 13.500 −5.473 0.003
S + PB,2 months - S + PB,14 months −0.613 0.748 20.250 −0.820 1.000
S + PB,2 months - S + PB,26 months 0.208 0.748 20.250 0.279 1.000
S + PB,2 months - S + ZB,2 months −0.953 1.640 13.500 −0.581 1.000
S + PB,14 months - S + PB,26 months 0.821 0.748 20.250 1.099 1.000
S + PB,14 months - S + ZB,14 months −0.953 1.640 13.500 −0.581 1.000
S + PB,26 months - S + ZB,26 months −0.953 1.640 13.500 −0.581 1.000
S + ZB,2 months - S + ZB,14 months −0.613 0.748 20.250 −0.820 1.000
S + ZB,2 months - S + ZB,26 months 0.208 0.748 20.250 0.279 1.000
S + ZB,14 months - S + ZB,26 months 0.821 0.748 20.250 1.099 1.000
614 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

A.2.2. Loss-on-peroxide oxidation (LPO)

ANOVA RM: LPO

numDF denDF F-value p-value

(Intercept) 1 16 2191.744 0.000

Time 2 16 24.453 0.000
Treatment 2 6 2.738 0.143

Bonferroni test: LPO

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months 0.224 0.413 20.250 0.542 1.000

S,2 months - S,26 months −2.494 0.413 20.250 −6.037 0.000
S,2 months - S + PB,2 months −1.671 0.796 13.500 −2.098 1.000
S,2 months - S + ZB,2 months −1.029 0.796 13.500 −1.292 1.000
S,14 months - S,26 months −2.718 0.413 20.250 −6.579 0.000
S,14 months - S + PB,14 months −1.671 0.796 13.500 −2.098 1.000
S,14 months - S + ZB,14 months −1.029 0.796 13.500 −1.292 1.000
S,26 months - S + PB,26 months −1.671 0.796 13.500 −2.098 1.000
S,26 months - S + ZB,26 months −1.029 0.796 13.500 −1.292 1.000
S + PB,2 months - S + PB,14 months 0.224 0.413 20.250 0.542 1.000
S + PB,2 months - S + PB,26 months −2.494 0.413 20.250 −6.037 0.000
S + PB,2 months - S + ZB,2 months 0.641 0.796 13.500 0.805 1.000
S + PB,14 months - S + PB,26 months −2.718 0.413 20.250 −6.579 0.000
S + PB,14 months - S + ZB,14 months 0.641 0.796 13.500 0.805 1.000
S + PB,26 months - S + ZB,26 months 0.641 0.796 13.500 0.805 1.000
S + ZB,2 months - S + ZB,14 months 0.224 0.413 20.250 0.542 1.000
S + ZB,2 months - S + ZB,26 months −2.494 0.413 20.250 −6.037 0.000
S + ZB,14 months - S + ZB,26 months −2.718 0.413 20.250 −6.579 0.000

A.2.3. Total organic carbon (TOC)

ANOVA RM: TOC

numDF denDF F-value p-value

(Intercept) 1 16 2308.292 0.000

Time 2 16 3.877 0.042
Treatment 2 6 58.964 0.000

Bonferroni test: TOC

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months 0.379 0.799 23.014 0.475 1.000

S,2 months - S,26 months 2.110 0.799 23.014 2.642 0.524
S,2 months - S + PB,2 months −7.115 0.832 10.168 −8.550 0.000
S,2 months - S + ZB,2 months −8.005 0.832 10.168 −9.619 0.000
S,14 months - S,26 months 1.731 0.799 23.014 2.167 1.000
S,14 months - S + PB,14 months −7.115 0.832 10.168 −8.550 0.000
S,14 months - S + ZB,14 months −8.005 0.832 10.168 −9.619 0.000
S,26 months - S + PB,26 months −7.115 0.832 10.168 −8.550 0.000
S,26 months - S + ZB,26 months −8.005 0.832 10.168 −9.619 0.000
S + PB,2 months - S + PB,14 months 0.379 0.799 23.014 0.475 1.000
S + PB,2 months - S + PB,26 months 2.110 0.799 23.014 2.642 0.524
S + PB,2 months - S + ZB,2 months −0.889 0.832 10.168 −1.069 1.000
S + PB,14 months - S + PB,26 months 1.731 0.799 23.014 2.167 1.000
S + PB,14 months - S + ZB,14 months −0.889 0.832 10.168 −1.069 1.000
S + PB,26 months - S + ZB,26 months −0.889 0.832 10.168 −1.069 1.000
S + ZB,2 months - S + ZB,14 months 0.379 0.799 23.014 0.475 1.000
S + ZB,2 months - S + ZB,26 months 2.110 0.799 23.014 2.642 0.524
S + ZB,14 months - S + ZB,26 months 1.731 0.799 23.014 2.167 1.000

A.2.4. Strong dichromate oxidation (sO)

ANOVA RM: sO

numDF denDF F-value p-value

(Intercept) 1 12 1076.426 0.000

 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 615

(continued)

ANOVA RM: sO

numDF denDF F-value p-value

Time 2 12 9.497 0.003

Treatment 2 6 20.288 0.002
Time:Treatment 4 12 3.788 0.032

Bonferroni test: sO

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months −2.269 0.841 27.000 −2.699 0.427

S,2 months - S,26 months 1.771 0.841 27.000 2.107 1.000
S,2 months - S + PB,2 months −6.055 1.193 26.876 −5.074 0.001
S,2 months - S + ZB,2 months −6.107 1.193 26.876 −5.118 0.001
S,14 months - S,26 months 4.041 0.841 27.000 4.805 0.002
S,14 months - S + PB,14 months −4.483 1.193 26.876 −3.757 0.030
S,14 months - S + ZB,14 months −2.571 1.193 26.876 −2.155 1.000
S,26 months - S + PB,26 months −6.150 1.193 26.876 −5.153 0.001
S,26 months - S + ZB,26 months −6.880 1.193 26.876 −5.765 0.000
S + PB,2 months - S + PB,14 months −0.697 0.841 27.000 −0.829 1.000
S + PB,2 months - S + PB,26 months 1.676 0.841 27.000 1.994 1.000
S + PB,2 months - S + ZB,2 months −0.053 1.193 26.876 −0.044 1.000
S + PB,14 months - S + PB,26 months 2.374 0.841 27.000 2.823 0.318
S + PB,14 months - S + ZB,14 months 1.912 1.193 26.876 1.602 1.000
S + PB,26 months - S + ZB,26 months −0.730 1.193 26.876 −0.612 1.000
S + ZB,2 months - S + ZB,14 months 1.267 0.841 27.000 1.506 1.000
S + ZB,2 months - S + ZB,26 months 0.999 0.841 27.000 1.188 1.000
S + ZB,14 months - S + ZB,26 months −0.268 0.841 27.000 −0.319 1.000

A.2.5. Mild oxidation (mO)

ANOVA RM: mO

numDF denDF F-value p-value

(Intercept) 1 16 1442.364 0.000

Time 2 16 6.308 0.010
Treatment 2 6 0.234 0.798

Bonferroni test: mO

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months 0.833 0.235 20.25 3.539 0.073

S,2 months - S,26 months 0.643 0.235 20.25 2.733 0.458
S,2 months - S + PB,2 months −0.234 0.386 13.50 −0.606 1.000
S,2 months - S + ZB,2 months −0.074 0.386 13.50 −0.192 1.000
S,14 months - S,26 months −0.190 0.235 20.25 −0.806 1.000
S,14 months - S + PB,14 months −0.234 0.386 13.50 −0.606 1.000
S,14 months - S + ZB,14 months −0.074 0.386 13.50 −0.192 1.000
S,26 months - S + PB,26 months −0.234 0.386 13.50 −0.606 1.000
S,26 months - S + ZB,26 months −0.074 0.386 13.50 −0.192 1.000
S + PB,2 months - S + PB,14 months 0.833 0.235 20.25 3.539 0.073
S + PB,2 months - S + PB,26 months 0.643 0.235 20.25 2.733 0.458
S + PB,2 months - S + ZB,2 months 0.160 0.386 13.50 0.414 1.000
S + PB,14 months - S + PB,26 months −0.190 0.235 20.25 −0.806 1.000
S + PB,14 months - S + ZB,14 months 0.160 0.386 13.50 0.414 1.000
S + PB,26 months - S + ZB,26 months 0.160 0.386 13.50 0.414 1.000
S + ZB,2 months - S + ZB,14 months 0.833 0.235 20.25 3.539 0.073
S + ZB,2 months - S + ZB,26 months 0.643 0.235 20.25 2.733 0.458
S + ZB,14 months - S + ZB,26 months −0.190 0.235 20.25 −0.806 1.000

A.2.6. Acid hydrolysis (AH)

ANOVA RM: AH

numDF denDF F-value p-value

(Intercept) 1 16 855.852 0.000

Time 2 16 4.385 0.030
Treatment 2 6 28.929 0.001

Bonferroni test: AH

Contrast Estimate SE df t.ratio p-value

S,2 months - S,14 months 1.718 0.834 20.935 2.061 1.000

(continued on next page)

616 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

(continued)

Bonferroni test: AH

Contrast Estimate SE df t.ratio p-value

S,2 months - S,26 months 2.499 0.834 20.935 2.998 0.247

S,2 months - S + PB,2 months −5.661 0.939 12.609 −6.032 0.002
S,2 months - S + ZB,2 months −5.715 0.939 12.609 −6.089 0.002
S,14 months - S,26 months 0.781 0.834 20.935 0.937 1.000
S,14 months - S + PB,14 months −5.661 0.939 12.609 −6.032 0.002
S,14 months - S + ZB,14 months −5.715 0.939 12.609 −6.089 0.002
S,26 months - S + PB,26 months −5.661 0.939 12.609 −6.032 0.002
S,26 months - S + ZB,26 months −5.715 0.939 12.609 −6.089 0.002
S + PB,2 months - S + PB,14 months 1.718 0.834 20.935 2.061 1.000
S + PB,2 months - S + PB,26 months 2.499 0.834 20.935 2.998 0.247
S + PB,2 months - S + ZB,2 months −0.054 0.939 12.609 −0.057 1.000
S + PB,14 months - S + PB,26 months 0.781 0.834 20.935 0.937 1.000
S + PB,14 months - S + ZB,14 months −0.054 0.939 12.609 −0.057 1.000
S + PB,26 months - S + ZB,26 months −0.054 0.939 12.609 −0.057 1.000
S + ZB,2 months - S + ZB,14 months 1.718 0.834 20.935 2.061 1.000
S + ZB,2 months - S + ZB,26 months 2.499 0.834 20.935 2.998 0.247
S + ZB,14 months - S + ZB,26 months 0.781 0.834 20.935 0.937 1.000

A.2.7. Resistant organic carbon (ROC)

ANOVA RM: ROC

numDF denDF F-value p-value

(Intercept) 1 69 3202 0
Methods 2 69 169 0
Treatment 2 6 30 0

Bonferroni test: ROC

Contrast Estimate SE df t.ratio p-value

S,ROCmO- S,ROCAH −8.577 2.285 78.513 −3.753 0.012

S,ROCmO- S,ROCPO 31.729 2.309 78.754 13.741 0.000
S,ROCmO- S + PB,ROCmO −13.646 2.309 6.898 −5.910 0.023
S,ROCmO- S + ZB,ROCmO −16.762 2.285 6.654 −7.334 0.007
S,ROCAH- S,ROCPO 40.306 2.309 78.754 17.455 0.000
S,ROCAH- S + PB,ROCAH -13.646 2.309 6.898 −5.910 0.023
S,ROCAH- S + ZB,ROCAH -16.762 2.285 6.654 −7.334 0.007
S,ROCPO - S + PB,ROCPO -13.646 2.309 6.898 −5.910 0.023
S,ROCPO - S + ZB,ROCPO -16.762 2.285 6.654 −7.334 0.007
S + PB,ROCmO - S + PB,ROCAH -8.577 2.285 78.513 −3.753 0.012
S + PB,ROCmO - S + PB,ROCPO 31.729 2.309 78.754 13.741 0.000
S + PB,ROCmO - S + ZB,ROCmO −3.116 2.309 6.898 −1.349 1.000
S + PB,ROCAH - S + PB,ROCPO 40.306 2.309 78.754 17.455 0.000
S + PB,ROCAH - S + ZB,ROCAH -3.116 2.309 6.898 −1.349 1.000
S + PB,ROCPO - S + ZB,ROCPO -3.116 2.309 6.898 −1.349 1.000
S + ZB,ROCmO - S + ZB,ROCAH -8.577 2.285 78.513 −3.753 0.012
S + ZB,ROCmO - S + ZB,ROCPO 31.729 2.309 78.754 13.741 0.000
S + ZB,ROCAH - S + ZB,ROCPO 40.306 2.309 78.754 17.455 0.000

A.2.8. Biochar organic carbon (BOC)

ANOVA RM: BOC

numDF denDF F-value p-value

(Intercept) 1 81 125.188972 0.0000000

Methods 4 81 4.848911 0.0014787
Treatment 1 81 3.649027 0.0596411

Bonferroni test: BOC

Contrast Estimate SE df t.ratio p-value

S + PB,BOCmO- S + PB,BOCAH 1.200 0.557 91.303 2.154 1.000

S + PB,BOCmO- S + PB,BOCLOI 0.864 0.557 91.303 1.551 1.000
S + PB,BOCmO- S + PB,BOCPO 2.036 0.566 91.325 3.600 0.023
S + PB,BOCmO- S + PB,BOCTOC −0.119 0.557 91.303 −0.213 1.000
S + PB,BOCmO- S + ZB,BOCmO −0.681 0.355 91.312 −1.920 1.000
S + PB,BOCAH - S + PB,BOCLOI −0.336 0.557 91.303 −0.603 1.000
S + PB,BOCAH - S + PB,BOCPO 0.836 0.566 91.325 1.478 1.000
S + PB,BOCAH - S + PB,BOCTOC −1.319 0.557 91.303 −2.368 0.900
S + PB,BOCAH - S + ZB,BOCAH −0.681 0.355 91.312 −1.920 1.000
 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618 617

(continued)

Bonferroni test: BOC

Contrast Estimate SE df t.ratio p-value

S + PB,BOCLOI - S + PB,BOCPO 1.172 0.566 91.325 2.072 1.000

S + PB,BOCLOI - S + PB,BOCTOC −0.983 0.557 91.303 −1.764 1.000
S + PB,BOCLOI - S + ZB,BOCLOI −0.681 0.355 91.312 −1.920 1.000
S + PB,BOCPO - S + PB,BOCTOC −2.155 0.566 91.325 −3.810 0.011
S + PB,BOCPO - S + ZB,BOCPO −0.681 0.355 91.312 −1.920 1.000
S + PB,BOCTOC - S + ZB,BOCTOC −0.681 0.355 91.312 −1.920 1.000
S + ZB,BOCmO - S + ZB,BOCAH 1.200 0.557 91.303 2.154 1.000
S + ZB,BOCmO - S + ZB,BOCLOI 0.864 0.557 91.303 1.551 1.000
S + ZB,BOCmO - S + ZB,BOCPO 2.036 0.566 91.325 3.600 0.023
S + ZB,BOCmO - S + ZB,BOCTOC −0.119 0.557 91.303 −0.213 1.000
S + ZB,BOCAH - S + ZB,BOCLOI −0.336 0.557 91.303 −0.603 1.000
S + ZB,BOCAH - S + ZB,BOCPO 0.836 0.566 91.325 1.478 1.000
S + ZB,BOCAH - S + ZB,BOCTOC −1.319 0.557 91.303 −2.368 0.900
S + ZB,BOCLOI - S + ZB,BOCPO 1.172 0.566 91.325 2.072 1.000
S + ZB,BOCLOI - S + ZB,BOCTOC −0.983 0.557 91.303 −1.764 1.000
S + ZB,BOCPO - S + ZB,BOCTOC −2.155 0.566 91.325 −3.810 0.011

References Lehmann, J.J., Joseph, S., 2015. Biochar for Environmental Management: Science, Technol-
ogy and Implementation. Routledge, Routledge. ed.
Baldock, J.A., Smernik, R.J., 2002. Chemical composition and bioavailability of thermally al- Lehmann, J., Gaunt, J., Rondon, M., 2006. Bio-char sequestration in terrestrial ecosystems -
tered Pinus resinosa (Red pine) wood. Org. Geochem. 33:1093–1109. http://dx.doi. a review. Mitig. Adapt. Strateg. Glob. Chang. 11:403–427. http://dx.doi.org/10.1007/
org/10.1016/S0146-6380(02)00062-1. s11027-005-9006-5.
Ball, D.F., 1964. Loss-on-ignition as an estimate of organic matter and organic carbon in Li, S., Gu, X., Zhuang, J., An, T., Pei, J., Xie, H., Li, H., Fu, S., Wang, J., 2016. Distribution and
non-calcareous soils. J. Soil Sci. 15, 84–92. storage of crop residue carbon in aggregates and its contribution to organic carbon of
Brändli, R.C., Bergsli, A., Ghosh, U., Hartnik, T., Breedveld, G.D., Cornelissen, G., 2009. Quan- soil with low fertility. Soil Tillage Res. 155:199–206. http://dx.doi.org/10.1016/j.still.
tification of activated carbon contents in soils and sediments using chemothermal 2015.08.009.
and wet oxidation methods. Environ. Pollut. 157:3465–3470. http://dx.doi.org/10. Liang, B., Lehmann, J., Solomon, D., Kinyangi, J., Grossman, J., O'Neill, B., Skjemstad, J.O., Thies,
1016/j.envpol.2009.06.015. J., Luizão, F.J., Petersen, J., Neves, E.G., 2006. Black carbon increases cation exchange ca-
Calvelo Pereira, R., Kaal, J., Camps Arbestain, M., Pardo Lorenzo, R., Aitkenhead, W., pacity in soils. Soil Sci. Soc. Am. J. 70:1719. http://dx.doi.org/10.2136/sssaj2005.0383.
Hedley, M., Macías, F., Hindmarsh, J., Maciá-Agulló, J., 2011. Contribution to character- von Lützow, M., Kögel-Knabner, I., Ekschmitt, K., Flessa, H., Guggenberger, G., Matzner, E.,
isation of biochar to estimate the labile fraction of carbon. Org. Geochem. 42: Marschner, B., 2007. SOM fractionation methods: relevance to functional pools and to
1331–1342. http://dx.doi.org/10.1016/j.orggeochem.2011.09.002. stabilization mechanisms. Soil Biol. Biochem. 39:2183–2207. http://dx.doi.org/10.
Cheng, C.-H., Lehmann, J., Thies, J.E., Burton, S.D., Engelhard, M.H., 2006. Oxidation of black 1016/j.soilbio.2007.03.007.
carbon by biotic and abiotic processes. Org. Geochem. 37:1477–1488. http://dx.doi. Mebius, L., 1960. A rapid method for the determination of organic carbon in soil. Anal.
org/10.1016/j.orggeochem.2006.06.022. Chim. Acta 22, 120–124.
Cheng, C.-H., Lehmann, J., Engelhard, M.H., 2008. Natural oxidation of black carbon in Mikutta, R., Kleber, M., Kaiser, K., Jahn, R., 2005. Review: organic matter removal from
soils: changes in molecular form and surface charge along a climosequence. Geochim. soils using hydrogen peroxide. Soil Sci. Soc. Am. J. 69:120–135. http://dx.doi.org/10.
Cosmochim. Acta 72:1598–1610. http://dx.doi.org/10.1016/j.gca.2008.01.010. 2136/sssaj2005.0120.
Fang, Y., Singh, B.P., Singh, B., 2014. Temperature sensitivity of biochar and native carbon Naisse, C., Alexis, M., Plante, A., Wiedner, K., Glaser, B., Pozzi, A., Carcaillet, C., Criscuoli, I.,
mineralisation in biochar-amended soils. Agric. Ecosyst. Environ. 191:158–167. Rumpel, C., 2013. Can biochar and hydrochar stability be assessed with chemical
http://dx.doi.org/10.1016/j.agee.2014.02.018. methods? Org. Geochem. 60:40–44. http://dx.doi.org/10.1016/j.orggeochem.2013.
Gelinas, Y., Prentice, K.M., Baldock, J.A., Hedges, J.I., 2001. An improved thermal oxidation 04.011.
method for the quantification of soot/graphitic black carbon in sediments and soils. Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon, and organic matter. In:
Environ. Sci. Technol. 35, 3519–3525. Sparks, D.L. (Ed.), Methods of Soil Analysis. Part 3 - Chemical Methods. Soil Science
Greenfield, L.G., Gregorich, E.G., van Kessel, C., Baldock, J.A., Beare, M.H., Billings, S.A., Society of America, Madison, pp. 961–1010.
Clinton, P.W., Condron, L.M., Hill, S., Hopkins, D.W., Janzen, H.H., 2013. Acid Nocentini, C., Certini, G., Knicker, H., Francioso, O., Rumpel, C., 2010. Nature and
hydrolysis to define a biologically-resistant pool is compromised by carbon loss and reactivity of charcoal produced and added to soil during wildfire are particle-size
transformation. Soil Biol. Biochem. 64:122–126. http://dx.doi.org/10.1016/j.soilbio. dependent. Org. Geochem. 41:682–689. http://dx.doi.org/10.1016/j.orggeochem.
2013.04.009. 2010.03.010.
Gustafsson, Ö, Haghseta, F., Chan, C., Macfarlane, J., Gschwend, P.M., 1997. Quantification Plante, A.F., Conant, R.T., Paul, E.A., Paustian, K., Six, J., 2006. Acid hydrolysis of easily dis-
of the dilute sedimentary soot phase: implications for PAH speciation and bioavail- persed and microaggregate-derived silt- and clay-sized fractions to isolate resistant
ability. Environ. Sci. Technol. 31, 203–209. soil organic matter. Eur. J. Soil Sci. 57:456–467. http://dx.doi.org/10.1111/j.1365-
Heiri, O., Lotter, A., 2001. Loss on Ignition as a Method for Estimating Organic and Carbon- 2389.2006.00792.x.
ate Content in Sediments: Reproducibility and Comparability of Results. 25: Plaza, C., Giannetta, B., Fernandez, J.M., Lopez-de-Sa, E.G., Polo, A., Gasco, G., Mendez, A.,
pp. 101–110. http://dx.doi.org/10.1023/A. Zaccone, C., 2016. Response of different soil organic matter pools to biochar and or-
Huff, M.D., Lee, J.W., 2016. Biochar-surface oxygenation with hydrogen peroxide. ganic fertilizers. Agric. Ecosyst. Environ. 225:150–159. http://dx.doi.org/10.1016/j.
J. Environ. Manag. 165:17–21. http://dx.doi.org/10.1016/j.jenvman.2015.08.046. agee.2016.04.014.
International Biochar Inititiative, 2012. Standardized Product Definition and Product Test- Poot, A., Quik, J.T.K., Veld, H., Koelmans, A.A., 2009. Quantification methods of black car-
ing Guidelines for Biochar That is Used in Soil. IBI biochar Stand. Available: http:// bon: comparison of rock-Eval analysis with traditional methods. J. Chromatogr. A
www.biochar-international.org/characterizationstandard. Accessed 27 Sept 2016. 1216:613–622. http://dx.doi.org/10.1016/j.chroma.2008.08.011.
Jeffery, S., Verheijen, F.G.A., van der Velde, M., Bastos, A.C., 2011. A quantitative review Qin, X., Li, Y., Wang, H., Liu, C., Li, J., Wan, Y., Gao, Q., Fan, F., Liao, Y., 2016. Long-term effect
of the effects of biochar application to soils on crop productivity using meta- of biochar application on yield-scaled greenhouse gas emissions in a rice paddy
analysis. Agric. Ecosyst. Environ. 144:175–187. http://dx.doi.org/10.1016/j.agee. cropping system: a four-year case study in south China. Sci. Total Environ. 570:
2011.08.015. 1390–1401. http://dx.doi.org/10.1016/j.scitotenv.2016.06.222.
Knicker, H., Totsche, K.U., Almendros, G., González-Vila, F.J., 2005. Condensation degree of Rovira, P., Vallejo, V.R., 2002. Labile and recalcitrant pools of carbon and nitrogen in or-
burnt peat and plant residues and the reliability of solid-state VACP MAS 13C NMR ganic matter decomposing at different depths in soil: an acid hydrolysis approach.
spectra obtained from pyrogenic humic material. Org. Geochem. 36:1359–1377. Geoderma 107:109–141. http://dx.doi.org/10.1016/S0016-7061(01)00143-4.
http://dx.doi.org/10.1016/j.orggeochem.2005.06.006. Rovira, P., Vallejo, V.R., 2007. Labile, recalcitrant, and inert organic matter in Mediterra-
Knicker, H., Müller, P., Hilscher, A., 2007. How useful is chemical oxidation with dichro- nean forest soils. Soil Biol. Biochem. 39:202–215. http://dx.doi.org/10.1016/j.soilbio.
mate for the determination of “Black Carbon” in fire-affected soils? Geoderma 142: 2006.07.021.
178–196. http://dx.doi.org/10.1016/j.geoderma.2007.08.010. Rumpel, C., Alexis, M., Chabbi, A., Chaplot, V., Rasse, D.P., Valentin, C., Mariotti, A., 2006.
Koide, R.T., Petprakob, K., Peoples, M., 2011. Quantitative analysis of biochar in field soil. Black carbon contribution to soil organic matter composition in tropical sloping
Soil Biol. Biochem. 43:1563–1568. http://dx.doi.org/10.1016/j.soilbio.2011.04.006. land under slash and burn agriculture. Geoderma 130:35–46. http://dx.doi.org/10.
Leavitt, S.W., Follett, R.F., Paul, E.A., 1996. Estimation of slow- and fast-cycling soil organic 1016/j.geoderma.2005.01.007.
carbon pools from 6N HCl hydrolysis. Radiocarbon 38, 231–239. Santisteban, J.I., Mediavilla, R., Lopez-Pamo, E., Dabrio, C.J., Zapata, M.B.R., Garcia,
Lehmann, J., Joseph, S., 2009. Biochar for environmental management: science and tech- M.J.G., Castano, S., Martinez-Alfaro, P.E., 2004. Loss on ignition: a qualitative
nology. Earthscan, London http://dx.doi.org/10.4324/9781849770552. or quantitative method for organic matter and carbonate mineral content in
618 I. Raya-Moreno et al. / Science of the Total Environment 598 (2017) 604–618

sediments? J. Paleolimnol. 32:287–299. http://dx.doi.org/10.1023/B:JOPL. Strosser, E., 2010. Methods for determination of labile soil organic matter: an overview.
0000042999.30131.5b. J. Agrobiol. 27:49–60. http://dx.doi.org/10.2478/s10146-009-0008-x.
Shackley, S., Sohi, S., Ibarrola, R., Hammond, J., Mašek, O., Brownsort, P., Haszeldine, S., 2013. Sun, F., Lu, S., 2014. Biochars Improve Aggregate Stability, Water Retention, and
Biochar, tool for climate change mitigation and soil management. In: Lenton, T., Pore-space Properties of Clayey Soil. 26–33. http://dx.doi.org/10.1002/jpln.
Vaughan, N. (Eds.), In Geoengineering Responses to Climate Change. Geoengineering 201200639.
Responses to Climate Change, Springer, New York:pp. 73–140 http://dx.doi.org/10. Velasco-Molina, M., Berns, A.E., Macías, F., Knicker, H., 2016. Biochemically altered char-
1007/978-1-4419-0851-3. coal residues as an important source of soil organic matter in subsoils of fire-
Shindo, H., 1991. Elementary composition, humus composition, and decomposition in soil affected subtropical regions. Geoderma 262:62–70. http://dx.doi.org/10.1016/j.
of charred grassland plants. Soil Sci. Plant Nutr. 37:651–657. http://dx.doi.org/10. geoderma.2015.08.016.
1080/00380768.1991.10416933. Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for determining
Silveira, M.L., Comerford, N.B., Reddy, K.R., Cooper, W.T., El-Rifai, H., 2008. Characteriza- soil organic matter, and a proposed modification of the chromic acid titration meth-
tion of soil organic carbon pools by acid hydrolysis. Geoderma 144:405–414. http:// od. Soil Sci. 37, 26–38.
dx.doi.org/10.1016/j.geoderma.2008.01.002. Wang, T., Camps-Arbestain, M., Hedley, M., Singh, B.P., Calvelo-Pereira, R., Wang, C., 2014.
Singh, B.P., Cowie, A.L., Smernik, R.J., 2012. Biochar carbon stability in a clayey soil as a Determination of carbonate-C in biochars. Soil Res. 52:495–504. http://dx.doi.org/10.
function of feedstock and pyrolysis temperature. Environ. Sci. Technol. 46: 1071/SR13177.
11770–11778. http://dx.doi.org/10.1021/es302545b. Wang, X., Zhou, W., Liang, G., Song, D., Zhang, X., 2015. Characteristics of maize biochar
Soil Survey Staff, 2014. Keys to Soil Taxonomy. 12th ed. USDA-Natural Resources Conser- with different pyrolysis temperatures and its effects on organic carbon, nitrogen
vation Service, Washington, DC. and enzymatic activities after addition to fluvo-aquic soil. Sci. Total Environ. 538:
Spokas, K.A., Novak, J.M., Masiello, C.A., Johnson, M.G., Colosky, E.C., Ippolito, J.A., Trigo, C., 137–144. http://dx.doi.org/10.1016/j.scitotenv.2015.08.026.
2014. Physical disintegration of Biochar: an overlooked process. Environ. Sci. Technol.
Lett. 1:326–332. http://dx.doi.org/10.1021/ez500199t.