GCB Bioenergy (2013) 5, 104–115, doi: 10.1111/gcbb.

12018

Production and characterization of slow pyrolysis

biochar: influence of feedstock type and pyrolysis
conditions
F R E D E R I K R O N S S E * , S V E N V A N H E C K E * , D A N E D I C K I N S O N * and W O L T E R P R I N S *
*Biosystems Engineering, Ghent University, Coupure Links 653, Gent, B-9000, Belgium

Abstract
Biochar was produced by fixed-bed slow pyrolysis from various feedstock biomasses under a range of process
conditions. Feedstocks used were pine wood, wheat straw, green waste and dried algae. Process conditions var-
ied were the highest treatment temperature (HTT) and residence time. The produced chars were characterized
by proximate analysis, CHN-elemental analysis, pH in solution, bomb calorimetry for higher heating value, N2
adsorption for BET surface area and two biological degradation assays (oxygen demand, carbon mineralization
in soil). In proximate analysis, it was found that the fixed carbon content (expressed in wt% of dry and ash-free
biochar) in the biochar samples strongly depended on the intensity of the thermal treatment (i.e. higher tempera-
tures and longer residence times in the pyrolysis process). The actual yield in fixed carbon (i.e. the biochar fixed
carbon content expressed as wt% of the dry and ash-free original feedstock biomass weight) was practically
insensitive to the highest treatment temperature or residence time. The pH in solution, higher heating value and
BET surface positively correlated with pyrolysis temperature. Finally, soil incubation tests showed that the addi-
tion of biochar to the soil initially marginally reduced the C-mineralization rate compared against the control
soil samples, for which a possible explanation could be that the soil microbial community needs to adapt to the
new conditions. This effect was more pronounced when adding chars with high fixed carbon content (resulting
from more severe thermal treatment), as chars with low fixed carbon content (produced through mild thermal
treatment) had a larger amount of volatile, more easily biodegradable, carbon compounds.
Keywords: biochar, biological degradation, characterization, feedstock, packed bed reactor, pyrolysis, pyrolysis conditions,
slow pyrolysis

Received 29 June 2012; revised version received 6 September 2012 and accepted 7 September 2012

has focused on selecting process conditions to maximize

Introduction
the yield of the bio-oil fraction, as it is a crude liquid
Rising energy costs, concerns over greenhouse gas emis- biofuel that can be used without modification in station-
sions and the nonrenewable nature of fossil fuels have ary heat and power applications, or can be further
prompted significant research into the conversion of upgraded to a drop-in biofuel (Czernik & Bridgwater,
biomass into biofuels and other value-added renewable 2004; Balat et al., 2009; Bridgwater, 2012). Conversely,
products (Hossain et al., 2011). Among the array of bio- the solid or char fraction obtained from biomass pyroly-
mass conversion technologies, pyrolysis is a relatively sis is often considered a waste product and conse-
simple technique in which organic material is heated in quently combusted to provide the necessary heat for the
the absence of oxygen. During the pyrolysis process, the pyrolysis process. However, recent research suggests
natural polymeric constituents (i.e. lignin, cellulose, fats that char from pyrolysis can be used as a soil amend-
and starches) are thermally broken down into three dif- ment, hence termed biochar, to substantially increase
ferent fractions: bio-oil (condensed vapours), char (solid soil fertility (Chan et al., 2007; Laird et al., 2009; Novak
fraction) and noncondensable gases (Mohan et al., 2006). et al., 2009; Major et al., 2010; Jeffery et al., 2011).
Depending on the heating rate and temperature, differ- At present, the mechanisms by which biochar
ent distributions and yields of the aforementioned frac- increases soil fertility are not fully understood. Research
tions can be obtained (Bridgwater & Peacocke, 2000; Lu has demonstrated that biochar application to soil
et al., 2009). To date, most biomass pyrolysis research increases the soil organic carbon, improves water hold-
ing capacity and soil aeration, increases the cation
Correspondence: Frederik Ronsse, tel. + 32 9 264 62 00, exchange capacity, neutralizes the pH of acidic soils
fax + 32 9 264 62 35, e-mail: [email protected] and improves the soil microbial ecology (Sohi et al.,

104 © 2012 Blackwell Publishing Ltd

 S L O W P Y R O L Y S I S B I O C H A R C H A R A C T E R I Z A T I O N 105

2010). In addition to these purported benefits, biochar overall influence on the final product characteristics
largely consists of a recalcitrant carbon fraction which (Antal & Gronli, 2003; Lua et al., 2004; Özçimen &
has been demonstrated to be very stable, with a half-life Ersoy-Meriçboyu, 2008).
of over 1000 years in the soil (Lehmann et al., 2006; To address the variability in biochar production
Kuzyakov et al., 2009; Lehmann & Joseph, 2009; Zimm- techniques (other than slow pyrolysis), the wide selec-
erman, 2010). Consequently, biochar production by tion of process parameters within each production
pyrolysis of biomass effectively removes carbon from technique, and the variability in feedstock selection – all
the atmospheric carbon cycle, transferring it to long- of which may affect biochar performance in soils – a
term storage in soils. Biochar therefore could help in the large number of biochar quality parameters have been
global challenge of carbon dioxide (CO2) mitigation, as described (Schimmelpfennig & Glaser, 2012; Song &
it results in a net removal of carbon from the atmo- Guo, 2012; Sun et al., 2012). More recently, attempts
sphere (Lehmann et al., 2006; Lehmann & Joseph, 2009; have been made to define a set of assays and relevant
Roberts et al., 2010; Matovic, 2011). characteristics for assessing and comparing the quality
Biochar can be produced from various types of of biochars (International Biochar Initiative (IBI), 2012;
processes including slow and fast pyrolysis, and gasifi- Schmidt et al., 2012). These parameters include (apart
cation. Each type of process is distinguished by differ- from toxicity testing): biochar particle size distribution,
ent ranges of temperatures, heating rates, biomass and elemental and proximate analysis, pH and liming effect,
vapour residence times. Given this variability in as well as more analytically advanced properties includ-
pyrolysis processes and their accompanying process ing BET surface area, porosity and plant available nutri-
conditions, in combination with a wide range of avail- ent content (IBI, 2012). However, it should be noted that
able biomass feedstocks for biochar production (includ- the identification of the ranges or values of these param-
ing wood, energy crops, agricultural waste residues, eters which classify biochars according to their quality
sewage sludge, anaerobic digestate, municipal wastes, is still ongoing. Furthermore, these quality parameters
etc.), large variability is to be expected in the physico- alone do not warrant the agronomic benefits of biochar
chemical properties of the biochars, and ultimately, in by themselves, as soil type and climate properties also
their performance as a soil amendment and/or in their determine the net effect of biochar use.
ability to store carbon permanently in the soil (Antal & This study is an experimental investigation of the
Gronli, 2003). Consequently, the challenge for biochar production of biochar by slow pyrolysis and aims to
science is to predict and assure product quality, agro- relate the various feedstock and process parameters to
nomic benefits and environmental effects of any given the yield and physicochemical characteristics of biochar.
biochar produced from any given feedstock by any Once the relationships between feedstock, production
given pyrolysis technology and process conditions process and biochar properties are known, this knowl-
(Masek et al., 2011). edge will permit the production of tailor-made biochar.
To produce high-quality biochars intended for More specifically, an appropriate selection of biomass
agricultural use with reliable and consistent product feedstock and pyrolysis parameters could be made that
qualities, slow pyrolysis is often considered as the most yields a biochar optimized for a specific behaviour
feasible production process (Song & Guo, 2012). In within the soil both in terms of soil fertility improvement
slow pyrolysis, the biomass is heated in an oxygen lim- and enlarging the pool of stable carbon within the soil.
ited or oxygen free environment, with typical heating
rates between 1 and 30 °C min
1 (Lua et al., 2004).
Materials and methods
Slow pyrolysis is usually carried out at atmospheric
pressure, and the process heat is typically supplied
Feedstock materials
from an external energy source – generally from com-
bustion of the produced gases or by partial combustion Four biomass input materials were selected for this study:
of the biomass feedstock (Laird et al., 2009). Under wood (pine), green waste, wheat straw, and spray-dried algae.
these circumstances biochar yields are usually up to 30 To ensure more comparable conditions (i.e. biomass heating
rate) in the pyrolysis reactor for the used different feedstocks,
wt%, on dry feedstock weight basis. However, yield
the materials were all pelletized.
and the properties of the resulting biochar are depen-
Both the pine wood (Stelmet, Poland) and straw (Strovan,
dent on several factors including the biomass
Belgium) were acquired commercially in a pelletized form
feedstock, heating rate, pyrolysis temperature and (6 mm pellet diameter). Prior to the pyrolysis experiments,
vapour residence time. Of these parameters, it has been pine wood and straw pellets were stored at ambient tempera-
established that the highest treatment temperature ture and ambient air humidity.
(HTT; the maximum temperature which the biomass is Green waste was obtained from a local garden contractor
subjected to in the pyrolysis reactor) has the greatest and consisted of shredded leaves, twigs and branches of

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

106 F . R O N S S E et al.

mainly coniferous trees and shrubs. Green waste feedstock Biochar characterization
material was ground in a cutting mill passing over a 2 mm
screen and then cold pelletized in a laboratory pellet press Biochar yield. The yields (g) of the recovered biochar were
(6 mm diameter). The green waste pellets were then air dried expressed as weight percentages of dry ash-free biochar recov-
at 105 °C for 1 h, before finally being stored at
18 °C due to ered to dry ash-free initial biomass. Dry ash-free (daf) basis for
their sensitivity to microbial decay. Spray-dried algae were yield expression was chosen to avoid positive bias in yield in
acquired commercially (SBAE Industries, Sleidinge, Belgium), case of using biomass samples with a high mineral (ash) con-
and manually pelletized to a diameter of 15 mm. tent. The yields were calculated by Eqn (1):
Mc
Mash;c
g¼  100% ð1Þ
Mdry;b
Mash;b
Slow pyrolysis set-up
The slow pyrolysis reactions were carried out in a vertical, where Mc is the weight of the char recovered from the
tubular, stainless steel reactor (d 9 L = 3.8 9 30 cm) which pyrolysis reactor (kg) which after the pyrolysis process was
was heated by an electric furnace (schematic in Fig. 1). considered to be oven dry (see below), Mdry,b is the oven-dry
The maximum temperature ramp rate for the reactor was ca. mass of raw biomass material, and Mash,c and Mash,b represent
17 °C min
1 and the reactor was continuously swept with the respective ash contents (i.e. weight in kg) in the biochar
nitrogen (800 ml min
1) to remove the produced gases and tars and raw biomass samples. More details regarding ash content
produced during pyrolysis. determination are given in the section below.
For each slow pyrolysis experiment, biomass pellets were
loosely packed in the reactor to form a bed height of 25 cm. Proximate analysis. Moisture, volatile matter and ash con-
Because of the differences in bulk density between feedstocks, tent were determined according to D1762-84 (ASTM, 2007). In
the actual mass of biomass in the reactor differed: ca. 135, 100, brief, biochar samples of ca. 1 g – in duplicate – were heated in
70, and 70 g of wood, straw, green waste and algae were porcelain crucibles and the sample weight differences before
respectively used in each pyrolysis experiment. Each pyrolysis and after heating were determined. For moisture content, sam-
experiment consisted of heating the reactor at the maximum ples were dried at 105 °C for 2 h (i.e. oven dry); for volatile
heating rate (17 °C min
1) until the HTT was reached, which matter samples were heated to 950 °C for 11 min (covered cru-
ranged between 300 and 750 °C depending on the experiment. cible) and for ash content 750 °C for a minimum of 2 h (uncov-
The reactor was then kept at the nominated HTT for a specific ered crucible). The weight of the original sample, subtracted by
duration (residence time), before the furnace was shut off and its moisture content, ash content and volatile matter content (as
the reactor ambiently cooled. The nitrogen flow was continued determined by the aforementioned proximate analysis) corre-
during cooling to purge the reactor of any remaining pyrolysis sponds to the stable carbon fraction of that sample and hence,
gases and to prevent any oxygen exposure to the char while this fraction is termed ‘fixed carbon or fixed-C fraction’. In this
still above ignition temperature. In total 32 experiments were study, the fixed carbon fraction is expressed on dry ash-free
performed, testing combinations of four highest treatment basis, or:
temperatures (HTT = 300, 450, 600, and 750 °C) and two
Mdry
Mvm
Mash
different residence times (10 and 60 min). An overview of all %Mfc ¼ 100  ð2Þ
Mdry
Mash
experiments, including used feedstock and pyrolysis condi-
tions, is given in Table 1.
where %Mfc is the fraction of fixed carbon (in wt%), Mvm is the
weight of volatile matter in the sample (kg), Mdry is the oven
12
dry weight of the sample (kg) and Mash is the weight of the ash
8
13 residue of the sample (kg).
9 11
2
In this study, ratio (in wt%) of weight of fixed carbon in a
biomass sample to the weight of the original biomass feedstock
5 10 on a dry and ash-free basis is defined as the fixed carbon yield,
4 or


7 %Mfc;c  Mdry;c
Mash;c
3 14 %Yfc ¼ ð3Þ
1 Mdry;b
Mash;b

6
where %Yfc is the fixed carbon yield (in wt%), Mdry,c and Mdry,b
are the dry weights (kg) of the biochar and the biomass feed-
Fig. 1 Slow pyrolysis set-up for the batch production of bio- stock respectively. Furthermore, Mash,c and Mash,b and are the
char: (1) nitrogen gas supply, (2) flow control, (3) gas preheater, weights (kg) of the ash in the biochar and in the original
(4) electric tube furnace, (5) pyrolysis reactor, (6) sintered base biomass feedstock out of which the biochar was produced,
plate, (7) packed biomass bed, (8) biomass lock hopper, (9) con- respectively.
denser, (10) condensate/gas separator, (11) cotton filter,
(12) diaphragm gas flow meter, (13) gas vent and (14) bio-oil Elemental analysis. The elemental (CHN) analysis was per-
recovery. formed in duplicate using a Flash 2000 Elemental Analyser

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

 Table 1 Overview of the tested properties of the biochars from different feedstocks, and using different HTTs and residence times in the pyrolysis process

Process conditions Proximate analysis Elemental analysis Biological tests

Biochar Volatile Fixed Ash Fixed Total C Total H BET Soil

HTT Residence yield matter carbon content carbon content content HVV pH in surface BOD14 incubation
(°C)* time (min) (wt%)† (wt%)† (wt%)† (wt%)‡ yield (wt%)§ (wt%)† (wt%)† H/C-ratio (MJ kg
1 db) solution area (m² g
1) (mg O2 l
1)* performed?

Wood
300 10 89.8 78.0 22.0 0.3 19.8 54.1 5.9 1.30 n.d.¶ 4.5 n.d. n.d.
60 43.7 42.6 57.4 0.5 25.1 71.3 4.7 0.79 n.d. 5.7 6 n.d. xk
450 10 29.2 21.4 78.6 1.0 23.0 82.5 3.8 0.56 32.5 6.6 4 79.3
60 27.0 16.8 83.2 1.2 22.5 86.3 3.5 0.49 32.9 6.7 23 54.8 x
600 10 24.4 8.2 91.8 1.2 22.4 90.0 2.6 0.35 34.4 6.7 196 40.5
60 23.3 6.4 93.6 1.3 21.8 92.3 2.3 0.30 34.4 9.1 127 25.9 x
750 10 23.0 2.6 97.4 1.1 22.4 92.5 1.4 0.19 n.d. 10.2 128 n.d.
60 22.7 2.6 97.4 1.1 22.1 92.5 1.1 0.15 n.d. 10.4 n.d. n.d. x
Straw
300 10 94.8 76.3 23.7 8.0 22.5 50.3 6.2 1.47 n.d. 6.1 n.d. n.d.
60 36.8 33.5 66.5 19.1 24.5 76.2 5.0 0.79 n.d. 9.4 n.d. n.d.
450 10 28.5 19.4 80.6 22.4 23.0 84.1 3.6 0.51 25.1 9.8 n.d. 41.6
60 27.5 15.9 84.1 22.9 23.1 86.4 3.5 0.49 25.5 10.1 16 42.0 x
600 10 25.4 8.8 91.2 24.5 23.2 90.1 2.4 0.32 25.6 10.9 n.d. 32.4

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

60 25.2 7.4 92.6 24.5 23.3 90.3 2.1 0.28 25.5 11.3 22 65.1
750 10 23.7 4.2 95.8 26.2 22.7 92.2 1.6 0.20 n.d. 12.1 n.d. n.d.
60 24.4 4.1 95.9 25.8 23.4 93.7 1.2 0.16 n.d. 11.9 n.d. n.d.
Green waste
300 10 98.4 74.3 25.7 3.6 25.3 53.2 6.2 1.41 n.d.¶ 7.4 n.d. n.d.
60 48.6 48.6 51.4 6.8 25.0 69.3 5.4 0.94 n.d. 8.1 n.d. n.d.
450 10 31.3 25.3 74.7 11.1 23.4 78.8 4.2 0.63 27.5 9.6 n.d. 65.1
60 27.8 18.5 81.5 12.0 22.7 82.9 3.5 0.51 27.9 10.0 17 60.8 xk
600 10 24.9 11.5 88.5 13.2 22.0 87.7 2.3 0.32 27.9 10.4 n.d. 57.4
60 24.4 8.8 91.2 13.4 22.3 88.4 2.0 0.27 28.0 11.3 46 31.4
750 10 26.4 3.5 96.5 13.9 25.5 87.5 1.5 0.21 n.d. 11.4 n.d. n.d.
60 23.7 1.9 98.1 13.4 23.2 93.2 1.3 0.16 n.d. 11.6 n.d. n.d.
Dry algae
300 10 72.8 70.0 30.0 46.3 21.8 62.7 7.2 1.38 n.d. 4.9 n.d. n.d.
60 50.1 55.2 44.8 55.8 22.4 69.5 6.9 1.19 n.d. 7.7 n.d. n.d.
450 10 28.4 27.5 72.5 68.6 20.6 74.5 4.5 0.72 9.22 9.1 n.d. 97.0
60 25.0 19.1 80.9 71.8 20.2 78.7 4.0 0.61 8.68 9.3 14 70.8 x
600 10 24.1 18.9 81.1 72.2 19.5 80.1 2.7 0.41 8.29 11.1 n.d. 71.5
60 22.9 15.7 84.3 73.0 19.3 83.4 2.0 0.29 8.17 11.9 19 65.2
750 10 21.0 10.1 89.9 74.8 18.9 86.4 1.5 0.21 n.d. 12.4 n.d. n.d.
60 19.3 3.9 96.1 76.4 18.5 90.6 1.4 0.19 n.d. 12.5 n.d. n.d.

*HTT, Highest treatment temperature; BOD, Biological oxygen demand.

†
Biochar on dry and ash-free basis.
‡
Biochar on dry basis.
§
Expressed on dry and ash-free feedstock weight basis.
¶
For HVV (higher heating value), BET surface area and biological tests, only selected samples were subjected to analysis (n.d. = not determined).
k
Indicates that for these samples a soil incubation test was performed.
S L O W P Y R O L Y S I S B I O C H A R C H A R A C T E R I Z A T I O N 107
108 F . R O N S S E et al.

(Thermo Fisher Scientific, Waltham, MA, USA). Oxygen con- for 42 days at 25 °C within sealed glass jars. The CO2, evolved
tent was not measured or calculated because of the interference from each incubation, was trapped by a 15 ml 0.5 N NaOH
of inorganic oxides in the ash, especially as the ash contents in solution. Periodically the lye solutions were replaced and
the biochar varied from less than 1 to more than 40 wt% titrated with 0.5 N HCl to a pH of 8.24 (702 SM Titrino,
depending on the feedstock material used (see also Table 1). Metrohm, Herisau, Switzerland). Evolved CO2 was therefore
determined by subtracting the amount of titrant used in a solu-
Higher heating value. The HHV (higher heating value) of tion from a given biochar treatment from the average amount
chars and input materials were determined by bomb calorimetry, required to neutralize the solution of the blank control; the dif-
in duplicate (Parr model 6200 calorimeter with a model 1108 ference being attributable to soil respired CO2 reacting with
oxygen bomb, Parr Instrument Company, Moline, IL) according OH
in solution to form CO32
. Although it has to be
to the instructions of Parr sheet no. 205M, 207M, and 442M. remarked that a part of the evolved CO2 could be attributed to
abiotic processes. The time elapsed between titrations (and
replacement of the NaOH solutions) was varied during the
pH in solution. Biochar samples were suspended in a 0.1 N
incubation period according to the expected respiration rate.
KCl solution in a 1 : 10 (wt/wt) ratio. After 10 min of stirring,
Measurements were taken on days 2, 4, 6, 10, 14, 18, 25, 36, and
the pH of the biochar suspension was measured using a Model
42.
420 Thermo Orion (Thermo Fisher Scientific). The analyses of
pH were performed in duplicate.
Results
BET surface area. The BET specific surface area of the
biochars were calculated by measuring nitrogen gas adsorption A summary of all tests performed on the different types
at -196 °C using a Ströhlein Areameter II (CIS Ingenieurbüro of biochar produced along with the results is given in
Seifert, Dresden, Germany) according to the single-point method Table 1. A more detailed analysis of each physicochemi-
DIN66132 (Deutsches Institut für Normung, 1975). Samples cal or biological test is given below.
were degassed at 100 °C under continuous nitrogen flow for
24 h prior to analysis. Analyses were performed in duplicate.
Biochar yield
Biological degradation experiments. Two short-term biolog- The effects of pyrolysis residence time and HTT on
ical degradation tests were performed: a biological oxygen
biochar yield are illustrated in Fig. 2. The char yields
demand (BOD) test and a soil incubation test. In the first, the
(daf basis) are negatively correlated with increasing
BOD of each biochar sample was measured over 14 days in a
2.6 g l
1 dispersion containing a microbial inoculum obtained pyrolysis severity (i.e. increased HTT and longer
from a soil sample. The soil sample from which the microbial residence times). Of the various feedstocks tested, algae
inoculum was prepared was the same as the one used in the had the lowest biochar yield (daf basis), with the excep-
soil incubation test (see below). The dispersion was incubated tion for the treatment at 300 °C and with a residence
at 20 °C and buffered at a pH of 7.2. The amount of oxygen time of 60 min. This could be due to a different compo-
consumed by biological oxidation of each biochar sample was sition in terms of extractives and cell wall components
measured using an Oxitop Respirometer (WTW Gmbh, Weil- in comparison with terrestrial biomass. Also, the algae
heim, Germany). The oxygen consumption metering principle sample used had a high ash content (38.4 wt% daf)
is based on absorbing evolved CO2 by soda lime pellets – the which may act catalytically during the pyrolysis process
consumption of oxygen (biological demand) is then calculated
and alter the product distribution in terms of yield of
from measuring the headspace pressure drop in the container
gas, char and bio-oil (Patwardhan et al., 2010).
of the biochar dispersion (Veeken et al., 2003). Apart from the
14-day measuring period and the use of a soil-derived
Generally, the yields in this study were lower than
microbial inoculum, the procedure was similar to ISO (1999). results reported in literature (e.g. Özçimen &
The second biological degradation test consisted of measur- Ersoy-Meriçboyu, 2008; Masek et al., 2011), where
ing the respired CO2 from incubated biochar–soil mixtures biochar yield above 30 wt% are reported. The low biochar
using a method adapted from Anderson & Ineson (1982). yields in this study can be attributed to the high nitro-
A sandy loam soil, retrieved from the area of Lendelede (Bel- gen sweeping rate used in our production experiments
gium) was used in the incubations. The soil sample contained (800 ml min
1). The sweep gas reduces the vapour resi-
7.1 g organic carbon per kg of soil and had a pH-H2O of 5.33. dence time and therefore partially inhibits secondary
The soil was sampled from an arable farming plot and its char forming reactions (Özçimen & Ersoy-Meriçboyu,
average composition was 50 wt% sand, 43 wt% silt and 7 wt%
2008).
clay. To prepare the incubations, the soil was thoroughly
mixed, passed through a 2 mm sieve, and for each individual
experiment 3.5 g of biochar was mixed with 250 g of soil (dry Proximate analysis
basis) and placed in a 6.9 cm diameter PVC tube. Next, water
was added to achieve a moisture content of 50% water filled Proximate analysis was performed to measure the
pore space. The biochar–soil preparations were then incubated moisture, volatile matter, fixed carbon and ash contained

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

 S L O W P Y R O L Y S I S B I O C H A R C H A R A C T E R I Z A T I O N 109

(a) 100 (a) 100

Biochar yield (wt%, daf basis)

80 80

fixed C content

wt%, daf basis

60 60

40 40

fixed C yield

20 20

0 0
250 350 450 550 650 750 250 350 450 550 650 750

HTT (°C) HTT (°C)

(b) 100
(b) 100
Biochar yield (wt%, daf basis)

80
80
fixed C content

wt%, daf basis

60
60

40
40
fixed C yield
20
20

0
0
250 350 450 550 650 750 250 350 450 550 650 750
HTT (°C)
HTT (°C)
Fig. 3 The effect of HTT on fixed carbon content and on the
Fig. 2 The effect of HTT on biochar yield, expressed in wt% of
fixed carbon yield, both expressed in wt% of biomass feedstock
biomass feedstock, on dry and ash-free basis (daf), for two dif-
on dry and ash-free basis (daf), for two different residence
ferent residence times: (a) 10 min and (b) 60 min. Biochar sam-
times: (a) 10 min and (b) 60 min. Biochar samples prepared
ples prepared from wood (♦), straw (□), green waste (▲) and
from wood (♦), straw (□), green waste (▲) and algae (○).
algae (○).

within the produced biochars and raw biomasses. The content is plotted against HTT for two different resi-
moisture contents of the input materials were 5.84, dence times (10 and 60 min). To allow comparison and
7.99, 31.64, and 5.32 wt% for wood, straw, green waste to exclude bias from the ash content figures, the fixed
and algae respectively. The relative ash content of the carbon results are expressed on a dry ash-free basis.
biochar samples shows an important increase with As the severity of pyrolysis increases, the fixed carbon
increasing pyrolysis severity (i.e. residence time and content in the produced biochars similarly increases,
HTT), which is expected as ash remains in the solid with HTT having a more notable impact than residence
fraction whereas the organic matter undergoes thermal time in these experiments. However, the overall fixed
decomposition, resulting in weight loss in the C-con- carbon yield – the total fixed carbon content expressed
taining fraction. The calculated ash yields – i.e. the as wt% of the original feedstock biomass used to pro-
ratio (in per cent) of the weight of ash in the biochar duce the biochar (daf basis) – was seen to be independent
to the weight of ash in the original biomass feedstock of both process variations as well as of the feedstocks
sample – were >98% for wood, green waste and algae used. The increase in the fixed carbon content is thus a
and >95% for straw biochar. The minor losses in ash result of the reduction in the overall biochar mass rather
content after the pyrolysis process are likely due to lost than additional ‘carbon-fixing’ reactions. Conceptually,
potassium, which can volatilize at pyrolysis tempera- this is unsurprising as the method used to measure fixed
tures (Long et al., 2012). In Fig. 3, the fixed carbon carbon is itself also a pyrolytic process (devolatilization

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

110 F . R O N S S E et al.

1.6 1.6

1.4 1.4

1.2 1.2

1.0 1.0

H/C ratio
H/C ratio

0.8
0.8

0.6
0.6
y = –0.0169 x + 1.8459
0.4
0.4 R2 = 0.9714

0.2
0.2
0.0
0.0 20 30 40 50 60 70 80 90 100
250 350 450 550 650 750
Fixed carbon content, wt% (on daf basis)
HTT (°C)
Fig. 5 The measured H/C ratio of the biochar samples plotted
Fig. 4 The relationship between HTT and measured H/C
against fixed carbon content (in wt% and expressed on dry ash-
ratio of the biochar samples. Biochar samples prepared from
free basis). Biochar samples prepared from wood (♦), straw (□),
wood (♦), straw (□), green waste (▲) and algae (○), samples
green waste (▲) and algae (○).
produced with a residence time of 10 min connected with solid
black lines, and grey dotted lines for residence times of 60 min.

at 950 °C). Considering the consistency of the fixed used to make an estimate for the average size of the
carbon yields and the fact that all the biochar produc- polyaromatic graphene clusters in the biochars –
tion experiments in this study were subject to HTTs which is likely to be an indicative measure of the overall
much lower than 950 °C, we conclude that the proxi- biochar stability in the soil. For instance, for a biochar
mate analysis had the overall net effect of subjecting all with an H/C ratio of 0.37, in its simplest construction,
the biochars to the same pyrolysis process – 950 °C could be composed of carbon graphene clusters having
treatment for 6 min – in essence overriding the lower an average of 22 aryl rings per cluster. In reality, the
temperature pyrolysis that previously occurred in the cluster formation and aryl ring consistency is unlikely
actual biochar production process. This observation fur- to be well defined – with an extensive inhomogeneous
ther supports the notion that HTT is the fundamental network more probable – but such estimations may be
process parameter in slow pyrolysis. Slow pyrolysis can useful in developing a better understanding of the
therefore be understood as a series of devolatilization molecular structure of biochar. The linear correlation
reactions that progressively leave behind an increas- between H/C ratio and fixed carbon content indicates
ingly condensed carbonaceous matrix. that fixed carbon has a very low H content and also that
volatile matter released during pyrolysis is composed of
compounds with higher H/C ratios than the remaining
Elemental analysis
biochar (fixed carbon). Hence the devolatilization
The elemental composition of the produced biochars removes most of the H from the biomass as the pyrolysis
shift from that of the feedstock biomass to high-carbon, conversion reaction takes place.
low-hydrogen compounds as a function of pyrolysis
process intensity (HTT, residence time). These results
Higher heating value
are summarized in Fig. 4, where the H/C ratio is plotted
against the HTT of production. Although large differ- The calorific data are plotted in Fig. 6. The higher heat-
ences in elemental composition were observed between ing values (HHV) of the produced biochars increased
the biochars produced from the different feedstocks at with increasing pyrolysis residence time and HTT for
lower HTTs, there was a convergence of the H/C ratio all feedstocks except algae. This anomalous observation
to 0.18 for all biomass types at a HTT of 750 °C. is explained by the fact that energy densification from
The H/C ratios measured in this research, as demon- pyrolysis only occurs in the organic fraction of the feed-
strated in Fig. 5, are similar to the values published by stock. For algae, the reduction of the HHV in pyrolysis
Brown et al. (2005), Keiluweit et al. (2010), Schimmelp- is caused by the high ash content of the algae, which
fennig & Glaser (2012) and Sun et al. (2012). The H/C was 38.2 wt% (dry basis), whereas the ash contents of
ratio can be considered a basic proxy for the average wood, straw and green waste were 0.2, 7.9 and 3.5 wt%
number of H–C bonds per carbon atom, which can be respectively. Similarly, the differences in HHV of the

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

 S L O W P Y R O L Y S I S B I O C H A R C H A R A C T E R I Z A T I O N 111

40 (a)
Wood
13
Straw

Higher heating value (MJ kg–1 db)

35
Green waste
Algae
30 11

25

pH in 0.1 N KCl
9
20

15 7

10
5
5

0 3
Raw 10 min, 60 min, 10 min, 60 min, 200 400 600 800
450 °C 450 °C 600 °C 600 °C
HTT (°C)
Fig. 6 The biochar higher heating value (MJ kg
1 – expressed (b)
on dry, i.e. ash-containing, biomass feedstock basis) of the dif- 13
ferent feedstocks tested, with varying intensity of the slow
pyrolysis process (residence time, HTT).
11

biochars formed from other feedstocks are also attribut-

pH in 0.1 N KCl
9
able to relative ash contents. As pyrolysis occurs, the
char slowly trends towards pure carbon (graphitic
structure) which carries a HHV of 32.8 kJ kg
1 (Soares 7
et al., 1997). Ash contained within char will therefore
‘dilute’ the energy content, meaning the limited value of
5
the HHV for completely carbonized materials will be
lower than that of pure C (as seen in the HHVs of
biochars from straw and green waste). However, the 3
200 400 600 800
HHV of the biochars derived from wood is higher than HTT (°C)
the HHV of pure carbon (up to almost 35 MJ kg
1).
This is due to the presence of C–H, C–O, and perhaps O Fig. 7 The pH of biochar in solution vs. HTT and using a resi-
dence time of 10′ (a) and 60′ (b) of the different biomass feed-
–H bonds remaining in the carbonaceous mass.
stocks tested: wood (♦), straw (□), green waste (▲) and algae
Considering biochars with carbon contents from 90% to
(○).
100%, one expects the HHV to peak and slowly decline
as the heterogeneous bonding is sequentially eliminated,
and the carbon molecular structures become increasingly labile – and more oxygenated – carbon is retained.
energetically stabilized by aromatic resonances and p-p Consequently, at higher pyrolysis severity, the amount
stacking of graphitic sheets. of carboxyl groups in the resulting biochar has been
reduced and/or the acidic groups have become deprot-
onated to the conjugate bases resulting in more alkaline
Biochar pH in solution
pH of the biochar in suspension. Another contributing
The solution pH of biochar in suspension (Fig. 7) factor to the rise in pH at more severe pyrolysis condi-
increases with intensification of the thermochemical tions is the relative increase of ash content in the
treatment (i.e. longer pyrolysis processes with higher biochar.
HTT produce more alkaline biochars). Differences in
pH can be observed for the biomass types: biochar
Specific surface area
produced from wood has an average pH in solution
that is in general 2 pH units lower than the values for The results of the BET surface analysis are given in
the three other feedstocks produced at similar pyrolysis Table 2. Two series of analyses were performed: one
conditions. The pH of the biochar, next to its ash that included most of the pine wood-derived biochars
content, is likely to be correlated with the presence of (to evaluate the effect of HTT and residence time) and
oxygen functionalities in the biochar: during thermo- one where different feedstocks were compared at two
chemical conversion with lower process intensity, more given HTT/residence time combinations. The highest

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

112 F . R O N S S E et al.

Table 2 The BET surface area of (a) woody biochar pro- biochar or due to the onset of ash melting at higher tem-
duced at varying highest treatment temperatures (HTTs) and peratures.
residence times and (b) of different feedstock materials pyrol- When comparing the different biomass feedstocks in
ysed with a HTT of 450 and 600 °C and a residence time of Table 2b, woody biochar offers the highest potential of
60 min.
surface area as all other biochar types had a BET spe-
(a) Wood biochar cific surface below 50 m² g
1. Also, wood had the low-
Highest treatment temperature (°C) est ash content of all feedstocks used (0.2, 7.9, 3.5, and
38.4 wt% for wood, straw, green waste and algae,
Residence time (min) 300 450 600 750
respectively). From these observations, higher amount
10 – 4 196 128 of inorganics (i.e. ash content) in the biomass feedstock
60 6 23 127 – negatively correlate with specific surface area in the
produced biochar. This is possibly explained by fusion
(b) Residence time = 60 min
Biomass input material of molten ash filling up pores in the biochar, thereby
decreasing accessible surface area.
HTT (°C) Wood Straw Green waste Algae
450 23 16 17 14
600 127 22 46 19 Biological degradation

–, No data recorded. Figure 8 shows the results of the cumulative BOD for
biochars produced from wood under different pyrolysis
BET specific surface (196 m2 g
1) was observed for intensities (residence time, HTT). These data show that
biochar produced from wood at an HTT of 600 °C and the rate of oxygen demand during the 2-week incuba-
a shorter residence time of 10 min. At lower pyrolysis tion period decreased exponentially, with a logarithmic
temperatures (300 and 450 °C), the biochar surface area relationship existing between incubation time and
was generally low, but gradually increased in biochars cumulative BOD. In case of wood, the R2 in the logarith-
produced with longer residence times. Pyrolysis at mic regression analysis varied between 0.899 and 0.956.
600 °C produced the most accessible surface in the Similar logarithmic trends were seen with biochars
chars for all feedstock biomasses tested, but an inverse produced from feedstocks other than wood.
relationship between specific surface and residence time Total cumulated BOD was found to be directly
was also observed. When the pyrolysis temperature was proportional to the volatile content of biochar with a
increased further, the BET surface area reduced again, correlation of R2 = 0.728 (Fig. 9). Higher carbon
which is likely due to restructuring taking place in the mineralization rates in soils amended with biochars

90
450 °C, 10 min
450 °C, 60 min
80
Biological oxygen demand (mg O2 L–1)

600 °C, 10 min

600 °C, 60 min
70 Log. (450 °C, 10 min)
Log. (450 °C, 60 min)
60 Log. (600 °C, 10 min)
Log. (600 °C, 60 min)
50

40

30

20

10

0
0 2 4 6 8 10 12 14 16 18
Oxidation time (days)

Fig. 8 Cumulative biological oxygen demand (BOD), measured over a 14-day period with regular sampling and for wood-based
biochar, treated at different HTTs and different residence times.

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

 S L O W P Y R O L Y S I S B I O C H A R C H A R A C T E R I Z A T I O N 113

35 0.2

Cumulative respiration (mmol CO2-C)

Wood 300 °C Wood 450 °C
Wood 600 °C Wood 750 °C
30 0.0

BOD14 (mg O2 g–1 biochar)

25 –0.2 zero biochar
control sample

20 –0.4

15 –0.6

10 y = 0.8783 x + 4.2894 –0.8

R² = 0.7283

5 1.0
5 10 15 20 25 30 0 10 20 30 40

Volatile matter (wt%) Time (days)

Fig. 9 Cumulative biological oxygen demand (BOD), mea- Fig. 10 Cumulative respiration, relative to the untreated con-
sured over a 14-day period plotted against volatile matter con- trol soil sample and expressed as mol CO2-C produced per
tent, for wood (♦), straw (□), green waste (▲) and algae (○) and day, recorded during a 42-day incubation period in a biochar-
under varying HTTs and residence times. treated soil sample. Biochar samples were all produced with a
residence time of 60 min.

produced from low-temperature pyrolysis processes the relative carbon respiration curves in Fig. 10. After
have also been previously observed by Zimmerman their initial reduced values, respiration rates were seen
(2010) and Cross & Sohi (2011). The high-volatile matter to restore to a level above that of the control (untreated)
content of these chars was identified as the main cause soil sample – as can be seen by the positive slope of the
of the higher carbon mineralization. relative respiration curve near the end of the 42-day
Given the results in Table 1, assuming exponential measurement period of most of the biochar-amended
decay of the volatile matter in the biochar and by taken soil samples. Also, the depression of initial carbon min-
into account the pool of volatile matter in the biochar, eralization is proportional to the intensity of thermal
then extrapolation indicates that all ‘volatile carbon’ treatment during pyrolysis – although the sample pro-
should be consumed after 7 months (algae, HTT = 600 °C duced at 600 °C does not follow trend. Chars produced
and tr = 60 min) to 2.3 years (wood, HTT = 750 °C and at higher HTT tend to have lower, more suppressed, ini-
tr = 60 min). However, these crude estimates are based tial carbon mineralization rates. The initial depression
on the assumption that the volatile matter and stable of the respiration rate could be attributed to the time
carbon fraction as determined by the proximate analysis needed for the soil microculture to adapt to the new
method correspond respectively to the labile and recal- conditions. The changes brought upon the soil sample
citrant carbon fractions in the soil, which not necessarily by the addition of biochar occur on different levels: for
holds true. instance, the pH and soil texture will be altered; biochar
In the second biological degradation test, biochar was could have the ability to adsorb certain compounds –
mixed with an actual soil sample and the CO2 produc- i.e. nutrients or phytotoxic compounds (Beesley et al.,
tion was measured during a 42-day incubation period. 2010; Borchard et al., 2012) – or release additional nutri-
The tests were only limited to the woody biochar ents; there may be a mild toxicity associated with ‘fresh’
samples produced using a residence time of 60 min. biochar; perhaps the added biochars alters the soil
In Fig. 10, the cumulative respiration, relative to the structure and/or water-holding capacity in a way that
untreated control soil sample (i.e. cumulative respira- disrupts the soil biota. Regarding the effect of biochar
tion biochar-amended sample minus cumulative addition to the initial CO2 respiration in soils, some
respiration control soil), for the different-tested woody researchers have found an opposite effect, i.e. an
biochars is plotted for a 42-day incubation period. It is increase in evolved CO2 in biochar-amended soil
important to stress that in this test, no distinction can be samples – an effect also know as positive ‘priming
made between the mineralization of preexisting C effect’ (Novak et al., 2010; Jones et al., 2011; Zimmerman
already present in the soil sample and the C from et al., 2011). However, this positive priming effect has
biochar itself. The results show that the initial carbon not always been observed to significant extent (Cross &
mineralization rates are suppressed by the addition of Sohi, 2011; Case et al., 2012) or a negative priming
biochar, which is clearly seen by the negative slopes of effect, a reduction in initial CO2 soil respiration in

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

114 F . R O N S S E et al.

biochar-amended soil samples could be observed content. Thus, the energy densification in pyrolysis is
(Zimmerman et al., 2011). Clearly, the outcome of confined to the C fraction and was consistent across the
changes in CO2 respiration in biochar supplemented soil tested feedstocks.
samples depends on a large number of variables, The pH in dispersion of the produced biochars was
including biochar type (feedstock, pyrolysis conditions), governed by feedstock type, ash content and pyrolysis
soil type, biochar loading in the soil (i.e. wt% biochar in intensity, with more increased pyrolysis temperature
soil) and physical conditions during the soil incubation and duration generally increasing the solution pH.
tests (temperature, air humidity, soil moisture content, Specific BET surface areas of the biochars produced in
etc.). Consequently, it is difficult to interpret and com- this study were generally low, with the exception of
pare the C mineralization observations in this study. those produced from wood and high HTT which had
Longer soil incubation tests, with different biochar load- comparatively moderate specific surfaces. Higher HTT
ings, are needed to clarify the overall effects biochar has increased the BET specific surface across all feed-
on overall mineralization in the soil. stocks, whereas longer residence times reduced the
BET specific surfaces in biochars produced at high
temperatures, possibly due to ash fusion filling mi-
Discussion
cropores.
The fixed-bed pyrolysis experiments in this study In the biological degradation experiments, the volatile
tested a range of biomass feedstocks and variable pro- matter content of biochar correlated positively with
cess parameters. Analysis of the produced biochars initial BOD. Whereas in the soil incubation carbon
provided significant data and increased the overall mineralization tests, the addition of biochar to the soil
understanding of slow pyrolysis. Raw biochar yield initially reduced the C-mineralization rate. Hence indi-
was seen to decrease with both the intensity of the cating that the soil microculture needs to adapt to the
pyrolysis process (i.e. residence time and highest new conditions. This effect was more pronounced when
treatment temperature) and by the ash content of the adding chars with high fixed carbon content, as chars
feedstock biomass. The volatile matter content of the with low C-content had a larger amount of volatile, pre-
produced biochars also declined with pyrolysis sever- sumably, more easily biodegradable, C-compounds. The
ity, whereas ash and fixed carbon contents increased. extent to which labile and stable C-fraction is present is
If, however, the fixed carbon content of the different greatly determined by the intensity of the pyrolysis
biochars is corrected for ash content (i.e. reported on process, both in terms of residence time and highest
daf basis), it was found that whereas the fixed carbon treatment temperature (HTT).
content of the biochars strongly depended on the
intensity of the thermal treatment, the actual overall Acknowledgements
yield in fixed carbon was practically insensitive to the
treatment temperature and residence time. This obser- The authors gratefully acknowledge the financial support of
Evonik Industries AG and the Multidisciplinary Research Part-
vation could result in the faulty conclusion that the
nership Ghent Bio-Economy in conducting this study.
thermochemical conversion that occurs in the organic
fraction of biomasses during slow pyrolysis is largely
invariant across the tested range of HTTs and resi- References
dence times. However, the observed constant fixed Anderson JM, Ineson P (1982) A soil microcosm system and its application to mea-
surements of respiration and nutrient leaching. Soil Biology and Biochemistry, 14,
carbon yields were merely the result of the method
415–416.
employed for measuring ‘fixed carbon’, as this Antal MJJ, Gronli M (2003) The art, science, and technology of charcoal production.
method itself is a pyrolysis process (up to 950 °C) Industrial and Engineering Chemistry Research, 42, 1619–1640.
ASTM (2007) D1762–84: Standard Method for Chemical Analysis of Wood Charcoal.
and in effect, supersedes any previous slow pyrolysis
American Society for Testing and Materials international, West Conshohocken,
production. This conclusion is further supported by PA, USA.
the consistent linear relationship between fixed carbon Balat M, Balat M, Kirtay E, Balat H (2009) Main routes for the thermo-conversion of
and H/C atomic ratio found across the produced biomass into fuels and chemicals. Part 1: pyrolysis systems. Energy Conversion and
Management, 50, 3147–3157.
biochars. Beesley L, Moreno-Jimenez E, Gomez-Eyles JL (2010) Effects of biochar and green
The HHV of the produced chars was largely consis- waste compost amendments on mobility, bioavailability and toxicity of inorganic
and organic contaminants in a multi-element polluted soil. Environmental Pollu-
tent for any given feedstock material beyond a moder-
tion, 158, 2282–2287.
ate pyrolysis treatment. The HHV of the biochars Borchard N, Prost K, Kautz T, Moeller A, Siemens J (2012) Sorption of copper (II)
trended towards a consistent limit for each feedstock and sulphate to different biochars before ans after composting with farmyard
manure. European Journal of Soil Science, 63, 399–409.
with increasing pyrolysis intensity. The actual HHVs of
Bridgwater AV (2012) Review of fast pyrolysis of biomass and product upgrading.
the produced biochars were largely determined by the Biomass and Bioenergy, 38, 68–94.
amount of ‘dilution’ of the combustible fraction by ash

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115

 S L O W P Y R O L Y S I S B I O C H A R C H A R A C T E R I Z A T I O N 115

Bridgwater AV, Peacocke GVC (2000) Fast pyrolysis processes for biomass. Renew- Lua AC, Yang T, Guo J (2004) Effects of pyrolysis conditions on the properties of
able Sustainable Energy Reviews, 4, 1–73. activated carbons prepared from pistachio-nut shells. Journal of Analytical and
Brown RA, Kercher AK, Nguyen TH, Nagle DC, Ball WP (2005) Production and Applied Pyrolysis, 72, 279–287.
characterisation of synthetic wood chars for use as surrogates for natural sor- Major J, Rondon M, Molina D, Riha SJ, Lehmann J (2010) Maize yield and nutrition
bents. Organic Geochemistry, 37, 321–333. during 4 years after biochar application to a Colombian savanna oxisol. Plant and
Case DCC, McNamara NP, Reay DS, Whitaker J (2012) The effect of biochar addition Soil, 333, 117–128.
on N2O and CO2 emissions from a sandy soil – the role of soil aeration. Soil Biol- Masek O, Brownsort P, Cross A, Sohi S (2011) Influence of production conditions on
ogy and Biochemistry, 51, 125–134. the yield and environmental stability of biochar. Fuel. doi: 10.1016/j.fuel.2011.
Chan KY, Van Zwieten L, Meszaros I, Downie A, Joseph S (2007) Agronomic values 08.044 (in press).
of greenwaste biochar as a soil amendment. Australian Journal of Soil Research, 45, Matovic D (2011) Biochar as a viable carbon sequestration option: global and Cana-
629–634. dian perspective. Energy, 36, 2011–2016.
Cross A, Sohi SP (2011) The priming potential of biochar products in relation to Mohan D, Pittman CU Jr., Steele PH (2006) Pyrolysis of wood/biomass for bio-oil: a
labile carbon contents and soil organic matter status. Soil Biology and Biochemistry, critical review. Energy & Fuels, 20, 848–889.
43, 2127–2134. Novak JM, Busscher WJ, Laird DA, Ahmedna M, Watts DM, Niandou M (2009)
Czernik S, Bridgwater AV (2004) Overview of applications of biomass fast pyrolysis Impact of biochar amendment on fertility of a southeastern Coastal Plain soil. Soil
oil. Energy & Fuels, 18, 590–598. Science, 174, 105–112.
Deutsches Institut für Normung (1975) DIN66132: Bestimmung der spezifischen Obe- Novak JM, Busscher WJ, Watts DW, Laird DA, Ahmedna MA, Niangou MAS (2010)
rfläche von Feststoffen durch Stickstoffadsorption (Einpunkt-Differenzverfahren nach Short-term CO2 mineralization after additions of biochar and switchgrass to
Haul und Dümbgen). Beuth Verlag GmbH, Berlin, Germany. Typic Kandiudult. Geoderma, 154, 281–288.
Hossain MK, Strezov V, Chan KY, Ziolkowski A, Nelson PF (2011) Influence of Özçimen D, Ersoy-Meriçboyu A (2008) A study on the carbonisation of grapeseed
pyrolysis temperature on production and nutrient properties of wastewater and chestnut shell. Fuel Processing Technology, 89, 1041–1046.
sludge biochar. Journal of Environmental Management, 92, 223–228. Patwardhan PR, Satrio JA, Brown RC, Shanks BH (2010) Influence of inorganic salts on
International Biochar Initiative (IBI) (2012) Guidelines for Specifications of Biochars for the primary pyrolysis products of cellulose. Bioresource Technology, 101, 4646–4655.
Use in Soils, Final Version (May 2012), IBI, Westerville, OH, USA. Roberts KG, Gloy BA, Joseph S, Scott NR, Lehmann J (2010) Life cycle assessment of
ISO (1999) ISO 14851:1999, Determination of the Ultimate Aerobic Biodegradability of biochar systems: estimating the energetic, economic, and climate change poten-
Plastic Materials in an Aqueous Medium – Method by Measuring the Oxygen Demand tial. Environmental Science and Technology, 44, 827–833.
in a Closed Respirometer, International Organization for Standardisation, Geneva, Schimmelpfennig S, Glaser B (2012) One step forward toward characterization: some
Switzerland. important material properties to distinguish biochars. Journal of Environmental
Jeffery S, Verheijen FGA, van der Velde M, Bastos AC (2011) A quantitative review Quality, 41, 1–13.
of the effects of biochar application to soils on crop productivity using meta-anal- Schmidt HP, Abiven S, Kammann C, Glaser B, Bucheli T, Leifeld J (2012) Guide-
ysis. Agriculture, Ecosystems and Environment, 144, 175–187. lines on the Production of Biochar – Euorpäisches Pflanzenkohle Zertifikat/European
Jones DL, Murphy DV, Khalid M, Ahmad W, Edwards-Jones G, DeLuca TH (2011) Biochar Certificate. Delinat Institute and Biochar Science Network, Ayent, Swit-
Short-term biochar-induced increase in soil CO2 release is both biotically and abi- zerland.
otically mediated. Soil Biology and Biochemistry, 43, 1723–1731. Soares RW, Menzes VJ, Fonsec MVA, Dweck J (1997) Characterization of carbona-
Keiluweit M, Nico PS, Johnson MG, Kleber M (2010) Dynamic molecular structure ceous products by TG and DTA. Journal of Thermal Analysis, 49, 657–661.
of plant biomass-derived black carbon (biochar). Environmental Science and Tech- Sohi SP, Krull E, Lopez-Capel E, Bol R (2010) A review of biochar and its use and
nology, 44, 1247–1253. function in soil. Advances in Agronomy, 105, 47–82.
Kuzyakov Y, Subbotina I, Chen H, Bogomolova I, Xu X (2009) Black carbon decom- Song W, Guo M (2012) Quality variations of poultry litter biochar generated at differ-
position and incorporation into soil microbial biomass estimated by 14C labeling. ent pyrolysis temperatures. Journal of Analytical and Applied Pyrolysis, 94, 138–145.
Soil Biology and Biochemistry, 41, 210–219. Sun H, Hockaday WC, Masiello CA, Zygourakis K (2012) Multiple controls on the
Laird DA, Brown RC, Amonette JE, Lehmann J (2009) Review of the pyrolysis chemical and physical structure of biochars. Industrial and Engineering Chemistry
platform for co-producing bio-oil and biochar. Biofuels, Bioproducts and Biorefineries, 3, Research, 51, 3587–3597.
547–562. Veeken AHM, de Wilde V, Hamelers HVM (2003) OxiTop® Measuring System for
Lehmann J, Joseph S (2009) Biochar for Environmental Management – Science and Tech- Standardised Determination of the Respiration Rate and N-Mineralisation Rate of
nology. Earthscan, London. Organic Matter in Waste Material, Compost and Soil. Nutrient Management Institute
Lehmann J, Gaunt J, Rondom M (2006) Bio-char sequestration in terrestrial (NMI), Wageningen, the Netherlands.
ecosystems – a review. Mitigation and Adaptation Strategies for Global Change, 11, Zimmerman AR (2010) Abiotic and microbial oxidation of laboratory-produced
403–427. black carbon (biochar). Environmental Science and Technology, 44, 1295–1301.
Long J, Song H, Jun X, Sheng S, Lun-shi S, Kai X, Yao Y (2012) Release characteris- Zimmerman AR, Gao B, Ahn MY (2011) Positive and negative carbon mineralization
tics of alkali and alkaline earth metallic species during biomass pyrolysis and priming effects among a variety of biochar-amended soils. Soil Biology and Bio-
steam gasification process. Bioresource Technology, 116, 278–284. chemistry, 43, 1169–1179.
Lu Q, Li W-Z, Zhu X-F (2009) Overview of fuel properties of biomass fast pyrolysis
oils. Energy Conversion and Management, 50, 1376–1383.

© 2012 Blackwell Publishing Ltd, GCB Bioenergy, 5, 104–115