See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/260756237

Evaluation of flash and slow pyrolysis applied on heavy metal contaminated

Sorghum bicolor shoots resulting from phytoremediation

Article  in  Biomass and Bioenergy · April 2014

DOI: 10.1016/j.biombioe.2014.02.027

CITATIONS READS

61 2,208

7 authors, including:

Ziad Al Chami Jan Yperman

Mediterranean Agronomic Institute of Bari Hasselt University
37 PUBLICATIONS   656 CITATIONS    289 PUBLICATIONS   4,519 CITATIONS   

SEE PROFILE SEE PROFILE

Robert Carleer Stefano Dumontet

Hasselt University Parthenope University of Naples
311 PUBLICATIONS   7,260 CITATIONS    87 PUBLICATIONS   2,917 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Coal desulphurization View project

Difpolmine View project

All content following this page was uploaded by Stefano Dumontet on 19 October 2017.

The user has requested enhancement of the downloaded file.

 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

Available online at www.sciencedirect.com

ScienceDirect

http://www.elsevier.com/locate/biombioe

Evaluation of flash and slow pyrolysis applied on

heavy metal contaminated Sorghum bicolor shoots
resulting from phytoremediation

Ziad Al Chami a,*, Nasser Amer b,c, Koen Smets d, Jan Yperman d,
Robert Carleer d, Stefano Dumontet c, Jaco Vangronsveld d
a
Mediterranean Agronomic Institute of Bari e CIHEAM, Via Ceglie 9, 70010 Valenzano, Italy
b
Dipartimento di Scienze per l’Ambiente, Università degli Studi di Napoli “Parthenope”, Via F. Acton, 38, 80133
Napoli, Italy
c
Swaida Research Center, General Commission for Scientific Agricultural Research e GCSAR, Swaida, Syria
d
Centre for Environmental Sciences, Hasselt University, Agoralaan Building D, 3590 Diepenbeek, Belgium

article info abstract

Article history: Treatment and/or disposal of metal contaminated biomass are still an unsolved problem.
Received 7 May 2013 Knowledge of the metal distribution is of prime importance concerning the application of
Received in revised form pyrolysis product streams. Sorghum bicolor (L.) Moench was cultivated in a semi-hydroponic
12 February 2014 system to assess its potential use in phytoremediation and biomass production. Plants
Accepted 17 February 2014 were grown in a greenhouse using perlite as substrate, half-strength Hoagland’s solution
Available online xxx as control (CTR) and the same solution supplemented with a mixture of Ni and Zn (CTM;
10 g m
3 each). Shoot and root biomass were determined and analyzed for their metals
Keywords: content. Flash and slow pyrolysis were performed on S. bicolor shoots at 450  C. Biomass
Pyrolysis and pyrolysis products were analyzed focusing on metal distribution. Mass and energy
Zinc balances were determined. S. bicolor delivered good shoot biomass with relatively moderate
Nickel concentrations of Ni and Zn. Metal concentrations in the pyrolysis oils were below
Bioenergy detection limits and almost all metals accumulated in the char. In fact, 99% of Ni and 98%
Biomass of Zn were recovered in the char when a slow pyrolysis process was applied, while in flash
Char pyrolysis conditions both metals were found back in the char and in the heating transfer
medium. Furthermore, the percentages of char and oil were higher in slow pyrolysis
compared to flash pyrolysis. Energy recovery in the char from slow pyrolysis was higher
than flash pyrolysis. Flash and slow pyrolysis can likely offer a valuable processing method
for metal contaminated biomass, thus limiting the waste disposal problem associated with
phytoremediation of metals.
ª 2014 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: þ39 080 4606 223; fax: þ39 080 4606 206.
E-mail addresses: [email protected], [email protected] (Z. Al Chami).

http://dx.doi.org/10.1016/j.biombioe.2014.02.027
0961-9534/ª 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
2 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

be reasonably extracted per hectare per year, it becomes

1. Introduction evident that phytoremediation is only applicable to moder-
ately contaminated land, and cannot compete against con-
Growing global population together with increased demands ventional soil remediation in case heavily polluted sites are
for food, fiber and fuel are driving demand for clean renewable concerned [9].
alternatives to replace petroleum based fuels. In order to meet High yielding biomass crops offer good potential for the
this demand, economically feasible and environmentally sound phytoremediation of soils contaminated with heavy metals.
alternatives such as energy from biomass have been proposed Energy crops grown on agricultural, municipal or even in-
[1,2]. Using biomass feedstock from agricultural waste sources dustrial wastes, or grown on contaminated sites (e.g. urban
for bioenergy production might play a key role, since it will sites, mining spoils and contaminated industrial and ex-
reduce greenhouse gas emissions, will diversify energy sources industrial land), may provide additional revenue and
reducing the oil and gas dependence and, finally, will provide improve their appeal to growers. Energy crops grown on
new production scenarios and alternative sources of income to contaminated land also offer real opportunities for stabiliza-
farmers [3]. However, biomass feedstocks have relatively low tion, bioremediation and phytoremediation of contaminated
energy density compared to fossil fuels; vast areas would be soils. Biomass fuel crops can remove heavy metals from soils
required to grow enough biomass to meet even a fraction of and be used for bio-energy production.
today’s energy demand. This will even be more pronounced in Sorghum bicolor (L.) Moench (sweet sorghum), belonging to
the poorer countries and less developed agricultural regions, the family of Poaceae, is a hardy C4 grass widely used as a
but consequently will give them limited possibility to develop forage crop [10]. It is considered as a promising energy crop
sustainable production systems. In particular, this is especially due to its fast-growth and high biomass production. Addi-
common for the Mediterranean arid and semiarid regions, tionally, S. bicolor is relatively inexpensive to grow. In fact, it
where agricultural crop production is strongly limited by water can be processed into various high value added products, such
availability and soil quality. Furthermore, growing plants for as bio-ethanol, electricity/heat, charcoal, hydrogen, activated
biofuel production on the already limited agricultural land, will carbon, methanol and pulp for paper [11]. S. bicolor has a sig-
increase food and fodder prices [1]. nificant yield potential, even in conditions of limited water
Nowadays, soil contamination due to dispersal of indus- availability because of its capacity to overcome periods of
trial and urban emissions and wastes generated by human drought by temporarily slow growth and development [12]. S.
activities is of major environmental concern. Among the most bicolor can produce approximately 30 tons ha
1 year
1 of dry
important contaminants are the heavy metals. The necessity biomass on low quality soils with low inputs of fertilizer and
to decontaminate polluted soils is recognized, both socially limited water [13]. In addition, S. bicolor has the ability to
and politically, because of the increasing attention for envi- tolerate unfavorable soil and climate conditions in the Medi-
ronmental protection and human health. As the number of terranean arid and semi-arid area.
contaminated sites and levels of contamination increase, so In general, a metal phytoextraction protocol consists of the
does the need to develop effective and affordable approaches following steps: (a) cultivation of the appropriate plant species
for remediation. Most of the conventional remediation tech- on the contaminated site; (b) removal of harvestable metal-
nologies are expensive and have a negative impact on soil enriched biomass from the site; and (c) post-harvest treat-
structure and fertility; this subsequently causes negative im- ments (i.e., composting, compacting, thermal treatments like
pacts on the ecosystem. Phytoextraction focuses on metal burning or pyrolysis) to reduce volume and/or weight of
accumulation in above-ground plant parts and subsequent biomass for disposal as a hazardous waste or for its recycling
removal from the contaminated site. It is described as a cost to reclaim the metals that may have an economic value [7].
effective, environmental friendly, aesthetically pleasing Although this post-harvest treatment of contaminated
approach most suitable for developing countries [4]. Plants biomass following phytoextraction is of high relevance, it has
with high biomass production have been increasingly used as not been addressed as it should be [14], and only few papers
potential phytoremediator crops that make phytoremediation address this problem [15e21]. After a successful phytoex-
an environmentally acceptable source of biomass for bio- traction, it is important to reduce the crop volume [16,22,23]
energy and added value chemicals [5,6]. Phytoextraction is a and to remove the excess water in order to reduce the cost
long-term remediation effort, requiring many cropping cycles of transportation to the treatment or disposal site.
to decontaminate metal pollutants to acceptable levels [7]. Pyrolysis is a promising thermochemical conversion tech-
Performance of metal phytoextraction depends on the po- nique that can be applied for contaminated plant materials
tential biomass production and the concentration of metals in [24,25]. However, pyrolysis is an extremely complex process
harvestable parts. The fundamental traits of plants suitable that results in three fractions: solid (char), liquid (pyrolysis oil)
for phytoremediation, summarized by Chaney et al. [8], and gas. The yield and properties of these products are influ-
should be expressed in a wide range of environmental con- enced by many factors such as pyrolysis temperature, heating
ditions to transform phytoremediation from a natural phe- rate, vapor residence time and biomass input material [26,27].
nomenon to a sustainable technology for soil clean-up. Based on the heating rate, pyrolysis is categorized as slow, fast
Meers et al. [9] pointed out that improvement of the heavy or flash pyrolysis. Low process temperature and long vapor
metal extraction efficiency and the economic valorization of residence times favor the production of char. High temperature
the produced biomass are the major constraints that have and long vapor residence time increase the biomass conver-
to be overcome to make phytoextraction an appropriate sion to gas. Moderate temperature and short vapor residence
technique. Taking into account the amount of metals that can time are the optimum process conditions for pyrolysis oil

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2 3

production [28,29]. The pyrolysis oil can be easily stored and the American Environmental Protection Agency (EPA) for
transported, and/or being directly used as a fuel imitative or for plant uptake and translocation test [31]. The composition of
producing chemicals with high added value [30]. the nutrient solution was: 2.5 mol m
3 Ca(NO3)2; 2.5 mol m
3
The main objective of this study is to evaluate the potential KNO3; 1 mol m
3 MgSO4; 0.2 mmol m
3 KH2PO4; 50 mmol m
3
use of S. bicolor for phytoremediation and bioenergy produc- NaFeEDTA; 0.2 mmol m
3 Na2MoO4; 10 mmol m
3 H3BO3;
tion. This paper focuses on the fate of Ni and Zn, present in the 2 mmol m
3 MnCl2; 0.5 mmol m
3 CuSO4; 1.0 mmol m
3 ZnSO4
growth substrate, during uptake and translocation in S. bicolor. and 0.2 mmol m
3 NiSO4. The solution was buffered with
The behavior of Zn and Ni during pyrolysis of harvested 0.5 mol m
3 MES (2-[N-morpholino]ethanesulphonic acid).
contaminated S. bicolor biomass was of special interest. The pH of the nutrient solution (pH z 5.5e6.0) was controlled
all over the incubation period [32].
Five pots were irrigated using half strength Hoagland’s
2. Materials and methods nutrient solution (control treatment e CTR) and other five pots
were irrigated using half strength Hoagland’s nutrient solu-
2.1. Chemicals tion supplemented with Ni and Zn at concentrations of
10 g m
3 each (contaminated treatment e CTM). Nutrient so-
All chemicals used for nutrient solution [i.e., Ca(NO3)2, KNO3, lutions were stored in 100 L plastic containers equipped with a
MgSO4, KH2PO4, NaFeEDTA, Na2MoO4, H3BO3, MnCl2, CuSO4, water pump and wrapped with black plastic bags to avoid the
ZnSO4 and NiSO4] were purchased from Sigma Aldrich (Ger- development of algae. An automatic drip irrigation system
many). NiSO4.6H2O and ZnSO4.7H2O were supplied by Merck was used; six emitters (4 L h
1 flow rate) were spread in each
KGaA (Germany). Deionized water (Elix; Millipore Corporation) container. Irrigation schedule (frequency and quantity) was
was used to prepare the plant nutrient solution. For sample settled in order to prevent leaching from the pots and to
mineralization and chemical analysis, HNO3 and H2O2 Trace- maintain the moisture at 50% humidity.
Select from Sigma Aldrich (Germany) and ultrapure water
(18.2 MU cm
1) obtained with a Milli-Q purification system 2.3. Pyrolysis experiments setup
from Millipore (Bedford, MA, USA) were used.
2.3.1. Schematic overview
2.2. Plant materials and growth conditions Fig. 1 shows the schematic overview of the slow and flash
pyrolysis set-ups and the conducted analysis. Before the py-
In order to evaluate the potential use of Sorghum bicolor (L.) rolysis, shoots were oven dried at 105  C until constant weight
Moench for heavy metal phytoremediation and biomass pro- was reached. Flash and slow pyrolysis were performed on
duction, an open system hydroponic experiment was per- both contaminated and control samples to study the effect of
formed. Seeds of S. bicolor were purchased from Syngenta pyrolysis on the heavy metal distribution.
Seeds S.A.S (France). Seeds were sterilized with 5% solution of
NaOCl for 10 min, washed with distilled water and germinated 2.3.2. Slow pyrolysis using horizontal tubular reactor
on filter paper with 2 mL of distilled water in closed Petri The reactor is shown in Fig. 2 [25]. It consists of a char (1), a tar
dishes for 5e6 days at 23  C until the primary roots reached a (2) and an oil (3) collector. The char/ash and tar collectors are
length of 3e5 cm. Three uniform seedlings were transferred made of quartz, while the oil collector is made of borosilicate
into each plastic pot. Plastic pots (28 cm height and 33 cm glass. Dried biomass (about 7 g) was introduced in the quartz
diameter) were placed inside an open side’s greenhouse and reactor (1) and the system was assembled and placed in the
filled with perlite (4.5 kg). Modified half strength Hoagland Nabertherm furnace. A nitrogen flow (50 mL min
1) was
nutrient solution was used for irrigation as recommended by applied to ensure pyrolysis atmosphere. The furnace heating

Fig. 1 e Schematic overview of pyrolysis set-up and conducted analyses.

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
4 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

Fig. 2 e Horizontal tubular pyrolysis reactor from outside (A) and inside (B).

rate was 20  C min
1 till an end pyrolysis temperature of condensation units made of glass (the first water-cooled and
450  C was reached, then kept isothermal for 30 min. Produced the second ice-cooled) leading to two liquid fractions: (i) a very
pyrolysis oil (aqueous fraction) was collected in part 3, which aqueous liquid fraction named pyrolysis oil, and (ii) a water
was cooled with salty-ice. When the system was cooled down, poor liquid fraction named heavy oil. Non-condensable gases
char was collected from part 1. Tar (viscous fraction) was were leaving the condensation unit as permanent gases. At the
collected at the end of the reactor tube. At the end of the slow end of the pyrolysis process, it was necessary to remove the
pyrolysis experiment char, oil and tar were collected and sand from the char. Sand removal was performed in two steps:
weighed, while the gas yield was calculated by difference. In first the sample was sieved at 0.4 mm; subsequently the
this experiment, a low amount of tar was obtained and no remaining sand was electrostatically removed using a
further analyses could therefore be done. commercially available electrical handhold tool. Finally char
and oil were collected and weighed, while the gas yield was
2.3.3. Flash pyrolysis using semi-continuous reactor calculated by difference.
Flash pyrolysis experiments were performed in a semi contin-
uous lab-scale reactor (Fig. 3). The reactor setup was equipped 2.4. Chemical analysis
with an improved condensation unit in comparison with the
set-up described by Smets et al. [33]. Pyrolysis atmosphere was 2.4.1. Plant preparation
ensured by a double nitrogen inlet system (2  70 mL min
1). Plants were harvested at the beginning of the flowering stage
The first nitrogen inlet is located at the storage and the second is (four months). They were gently removed from the pots and
located at the top of the pyrolysis reactor as indicated in Fig. 3. separated into shoots and roots. Roots were washed with tap
The nitrogen is blown through the hollow shaft of the Archi- water to remove perlite and then rinsed with distilled water.
medes’ screw to the bottom of the reactor to make sure that the Shoots and roots were first air dried in the greenhouse and
reactor is completely free of oxygen. Before each experiment, subsequently transferred into an oven at 65  C till constant
the reactor was filled with 700 g of white sand (particle size: weight was reached. Residual humidity was determined at
125e500 mm) which served as a heat transfer medium. The sand 105  C.
was pre-heated at 600  C for 6 h to remove all organic Dried plant materials were first ground in a high-speed
contamination. The reactor and the sand (as heat transfer rotary cutting mill to pass through a 1 mm sieve to obtain a
medium) were heated by an electric heating jacket (Horst, homogenous representative sample. Before chemical anal-
GmbH) to reach the final pyrolysis temperature of 450  C. ysis, dried plant materials were homogenized using a Retsch
Temperature was PC monitored. A perforated Archimedes’ MM200 Mixer Mill.
screw ensured a homogenous temperature of the sand and
created a ‘fluidized bed effect’ which allowed maximal mixing 2.4.2. Heavy metal and alkaline earth metals determination
of the biomass and the sand at 450  C. Dried biomass (about 70 g) Total heavy metals and alkaline earth metals in shoots, roots,
was transferred from the storage vessel (at room temperature) char, tar and oil were extracted by wet digestion (1 mL H2O2
into the hot reactor by the screw injection system. The reaction and 5 mL HNO3 for 20 min at 190  C) using a microwave
and residence time were measured as the period of time be- digester (CEM model, MARS Xpress). Samples were then
tween the biomass sample entering the hot reactor and the first cooled, collected in volumetric flasks, and finally measured for
vapors observed in the water cooled condenser. Samples were their metal content by means of ICP-OES (Thermo Electron
heated from room temperature to pyrolysis temperature in a ICAP 6300 Series).
very short period of time (reaction and vapor residence time
less than 5 s) and flash pyrolysis conditions were accomplished. 2.4.3. Shoots thermogravimetric analysis
Resulting vapors were leaving the hot reactor (equipped with Dried samples of shoots were subjected to thermogravimetric
and without a hot gas filter system to prevent aerosol particles analysis (TGA) to mimic the pyrolysis behavior of the used
from entering the condensation vessels) and entered into the biomass and to define the best operational pyrolysis

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2 5

Fig. 3 e Flash pyrolysis reactor set-up.

temperature. The biomass was gradually heated at basis (mg kg
1 dw) and shoots or roots biomass (g plant
1) [36].
20  C min
1 to 600  C, under N2 atmosphere (60 mL min
1). The indexes were calculated according to the following
Then, temperature was kept isothermal at 600  C for 10 min. equations:
After switching from N2 to O2 (60 mL min
1), the sample was

1
heated to 900  C under oxygen atmosphere with heating rate mg metal kg shoot dw
TF ¼
1
of 20  C min
1. mg metal kg root dw

1
2.4.4. Energy content MU ¼ mg metal kg shoots or roots dw
Biomass and char energy contents were measured using a  kg shoots or roots dw per plant
bomb calorimeter (IKA C5003). The measurement was per-
formed according to the DIN 51900 standard. After calibrating 2.4.7. Mass and energy balance and heavy metal recovery
the bomb calorimeter with benzoic acid, the sample is intro- The mass balance was calculated from the weight of biomass
duced in a quartz crucible and combusted. The analysis was and the weight of the produced char and pyrolysis oil. The
done in duplicate. Oil energy content could not be determined gaseous fraction was calculated by difference. Since the HHV
by the IKA C5003 calorimeter due to its high water content of the pyrolysis oil could not be determined due to its high
even when paraffin strips were used as combustion aid. water content, the energy recovery (%) was calculated using
only the higher heating value (HHV) of the produced char ac-
2.4.5. Oil water content cording to the following equation [21]:
Oil water content was determined by volumetric Karl Fischer
(KF) titration. Titrations were performed with an apparatus of MassðcharÞ  HHVðcharÞ
Energy Recoveryð%Þ ¼  100
Schott Geräte (Titrator TR85). Component solution (Combiti- MassðbiomassÞ  HHVðbiomassÞ
trant 5, Merck) was used as titrating agent medium and
where HHV is the higher heating value.
methanol (Normapur AR1, VWR) as working medium. The
Heavy metal recovery was calculated using the following
endpoint of the titration was detected with a double platinum
equation:
electrode (type Pt1400). The KF apparatus was calibrated by
titrating six amounts of pure water in triplicate. The calibra- Metal ContentðPPÞ
Metal Recoveryð%Þ ¼  100
tion curve is made according to the method of the least Metal ContentðbiomassÞ
squares [34]. The pyrolysis oil samples were weighed on a
where PP is the pyrolysis product (char/oil/gas).
plastic container with an analytical balance (Mettler Toledo
AG245). After weighing the sample (17e30 mg) it was brought
into the titration vessel of the KF apparatus. When the sample 2.5. Statistical analysis
was completely dissolved, the titration was started. Each
sample was measured in triplicate. Two-way analysis of variance (ANOVA) was carried out and
separation of means was performed using LSD test at P ¼ 0.05
2.4.6. Translocation factor and total uptake index significance level. The means of the 5 measurements are re-
The translocation factor (TF) is the ratio of metal concentra- ported for each of the studied parameters. All statistics were
tion in shoots to that in roots [35]. Metal uptake (MU) is defined computed using SAS software version 9 (SAS Institute, Cary,
as the product of metal concentration in shoots or roots on dry NC).

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
6 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

hyperaccumulator. The high concentration of Ni found in

3. Results and discussions roots and relatively moderate concentrations Ni and Zn found
in shoots together with its high biomass production, suggest
3.1. Biomass production, metals uptake and that S. bicolor is suitable for phytostabilization applications
translocation when rapid immobilization is required, and for sustainable
use and economic valorization of metal contaminated soils
Shoots and roots dry weights, expressed in g plant
1 and in when the produced biomass is used as a feedstock for bio-
percentage of the control, are shown in Table 1. Shoots and energy or industrial processes.
roots dry weights in CTM treatment were significantly lower
than in the control treatment (17 and 33%, respectively). Ni 3.2. Biomass characterization
and Zn are essential elements for plant growth at low con-
centrations but become toxic at high concentrations. In fact, 3.2.1. Biomass chemical analysis
plants grown in CTM treatment, exhibit toxicity symptoms as Alkaline metals in biomass are considered as additional pa-
chlorosis and growth reduction. Our results are in accordance rameters controlling pyrolysis product characteristics. They
with Amer et al. [37] who studied the effect of Zn and Ni on catalyze cracking reactions, resulting in a lower average mo-
some Mediterranean species grown on hydroponics. They lecular weight and more water in pyrolysis oils [43]. The
reported that at low concentrations Ni had a growth concentrations of the alkaline (Na, K) and alkaline earth (Ca,
stimulating-effect, while at higher concentrations Ni became Mg) metals in S. bicolor shoots are presented in Table 2. Ca, K
toxic. Increased cellular concentrations of Ni and Zn indeed and Mg in the contaminated biomass were significantly higher
can lead to improper protein binding, which can alter the than in the control, while Na was significantly lower. Heavy
protein structure and to the formation of reactive oxygen metals in the nutrient solution may have antagonistic or
species [38]. synergistic interactions with other macronutrients, thus
Metal concentrations (mg kg
1 dw) in shoots and roots of S. confounding outcomes [44]. The interaction between heavy
bicolor are shown in Fig. 4. Ni and Zn are typically present in metals and macronutrients depends on many factors such as
plant tissues at concentrations of 15e22 and 15e50 mg kg
1, plant species. In our case, Zn and Ni nutrient solution
respectively [39]. In the CTR treatments, the Ni and Zn con- increased the concentration of Ca, K and Mg, and decreased
centrations in shoots were 11 and 58 mg kg
1, respectively. In the concentration of Na in the S. bicolor shoots. However, may
CTM treatments, considerable amounts of metals were found be the increase in Ca, Mg and K concentration is due to the
in shoots and roots. Zn was the dominant metal in the shoots decrease in biomass production in the metal exposed plants.
(179 mg kg
1), while Ni was the prevalent metal in the roots The decrease in Na concentration may be due to the antago-
(1217 mg kg
1). nism between K and Na.
Plant’s potential for phytoremediation can be estimated by
both MU and TF. Total metal uptakes are shown in Fig. 5 with 3.2.2. Biomass thermogravimetric analysis
translocation factors for Ni ¼ 0.06 and for Zn ¼ 0.74. Low Ni TGA was performed to gain a better understanding of the
translocation was observed to S. bicolor shoots in comparison pyrolytic behavior of S. bicolor control and contaminated
to Zn. Ni was accumulated in roots, while a significant amount shoots. Hence, the TG and DTG curves provide useful infor-
of Zn was translocated to shoots. Zn uptake was higher in mation on the yield of pyrolysis products (volatiles and non-
shoots than in roots. High shoot biomass produced compen- volatiles). Additionally, the content of the major biomass
sated the low Zn concentration and translocation. These re- structural components (cellulose, hemicellulose and lignin)
sults agree with Amer et al. [37] who found that roots are the can be estimated from the DTG curves since their decompo-
main accumulation site for Ni, while only a limited amount sition is believed to proceed in specific temperature zones.
was translocated to shoots. In contrast, Zn was very mobile Typically, hemicellulose decomposition is reported to occur
and a significant amount was translocated to shoots. In fact, between 220 and 315  C, while cellulose decomposes between
Al Chami et al. [40] and Terzano et al. [41] found that Zn was 315 and 400  C. Lignin decomposition, in contrast, usually
translocated to shoots with a translocation factor between 0.7 occurs in a wide temperature interval (200e900  C) and
and 1.1. therefore is much more difficult to observe separately [38].
Our conclusion agrees with Roosens et al. [42] who The TG- and DTG-curves of S. bicolor control and contami-
mentioned that S. bicolor cannot be considered as a Ni or Zn nated shoots are presented in Figs. 6 and 7, respectively. In
case of the control sample, a prominent DTG peak with two
shoulders was observed. The first (around 200  C) and second
Table 1 e Shoots and roots dry weight of S. bicolor shoulder peak (around 250  C) were assigned to the decom-
expressed as g per plant and as percentage of the control. position of hemicellulose, while the main peak (around 336  C)
Shoots Roots represented the degradation of cellulose. For the contami-
nated sample, the main DTG peak appeared around 329  C and
g % g %
had only one shoulder (around 250  C). In both cases, these
CTR 101  16a 100 68  11a 100
results indicate that S. bicolor had a higher content of cellulose
CTM 84  6b 83 46  8b 67
than of hemicellulose [45e47].
Values are means  standard deviation (n ¼ 5). DTG-curves showed that most of the decomposition re-
Means with different letters within the same column indicate sig-
actions were finished before about 600  C. Therefore, a pyrolysis
nificant difference between values; LSD test (P < 0.05).
temperature of 450  C was chosen because higher temperatures

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2 7

Fig. 4 e Total Ni and Zn concentrations in S. bicolor shoots and roots.

were expected to not significantly increase the yield of the py- 3.2.3. Biomass energy content
rolysis oil, but only to result in higher energy costs. The calorific value (Table 3), measured as Higher Heating
Biomass characteristics are shown in Table 3. Results Value (HHV), of the control sample was higher than that of the
pointed out that the control sample contains 2.3 wt% mois- contaminated sample (17.6 vs 15.2 MJ kg
1), because of its
ture, whereas the contaminated sample contains 3.7 wt%. The higher ash content as one of the reasons. Calorific value of the
ash content of the contaminated sample (12.5 wt%) was control was close to the value of 18.3 MJ kg
1, which has been
higher than that of the control sample (5.6 wt%). This differ- reported by Librenti et al. [3].
ence was expected since ash stands for the inorganic matter
in the biomass and this was higher in the contaminated 3.3. Flash and slow pyrolysis experiments
sample [48]. Raveendran et al. [49] mentioned that ash
strongly affects both the pyrolysis characteristics and the 3.3.1. Yields
product distribution. Product yields from flash and slow pyrolysis are presented in
Fig. 8. The char yield ranged from 22% for the control produced
by flash pyrolysis to 55% for the control produced by slow
pyrolysis. Pyrolysis oil yield ranged from 16% for the
contaminated sample with flash pyrolysis to 33% for the
control with slow pyrolysis. Slow pyrolysis produced higher
amounts of char and pyrolysis oil compared to the flash

Table 2 e Total Ca, K, Mg and Na content in S. bicolor

shoots.
Ca K Mg Na

1
g kg dw
CTR 5.29  0.12b 10.56  0.71b 4.84  0.31b 1.71  0.19a
CTM 8.83  0.24a 40.78  1.26a 6.55  0.42a 0.48  0.11b

Values are means  standard deviation (n ¼ 5); dw: dry weight.

Means with different letters within the same column indicate sig-
Fig. 5 e Total metal uptake (MU) in S. bicolor shoots and
nificant difference between values; LSD test (P < 0.05).
roots.

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
8 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

Fig. 6 e Thermogravimetric analysis of the control S. bicolor shoots.

pyrolysis. Char yield in all our treatments was higher than the of the biomass sample was composed of very small particles
typical value (20e35% of the original mass) which is in (almost dust) that were blown through the hot zone of the
accordance with Fahmi et al. [43]. pyrolysis reactor without complete conversion/pyrolysis.
Slow pyrolysis produced higher amounts of char and oil Therefore, we used a filter inside the reactor to prevent this as
compared to the flash pyrolysis. A possible explanation for the much as possible. However, a considerable amount was still
higher oil yield in case of the slow pyrolysis compared to the found in the condensation system when this filter was used.
flash pyrolysis is that the biomass sample was not completely Furthermore, no differences were found in the percentage
pyrolyzed during the flash pyrolysis experiment. Hence, part of different flash pyrolysis fractions of the contaminated

Fig. 7 e Thermogravimetric analysis of the contaminated S. bicolor shoots.

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2 9

Table 3 e Proximate analysis: mass fraction determined Table 4 e Water mass fraction in the oils from flash and
by TGA, and HHV calorific value of S. bicolor shoots slow pyrolysis.
determined by bomb calorimetry. CTR CTM
Proximate analysis CTR CTM
FP SP FP SP
Moisture (%) 2.3 3.7
Water content (%) 61.0  2.8 61  2.3 76  1.1 66  2.3
Volatile matter (%) 70.8 63.3
Fixed carbon (%) 21.3 20.5 Values are means  standard deviation (n ¼ 3).
Ash (%) 5.6 12.5
HHV calorific value (MJ kg
1) 17.6 15.2

of flash pyrolysis, the water content of the pyrolysis oil was 9%

sample with or without the hot gas filter. A similar result was (w/w) higher for the contaminated sample than for the refer-
obtained by Stals et al. [21] when they applied flash pyrolysis ence sample.
on heavy metal contaminated willow at 350  C. Water in the pyrolysis oil cannot be removed by conven-
Char production from the contaminated biomass was tional methods, such as distillation [24]. Consequently, the
higher than the control (31 vs 22%, respectively) in case of higher heating value of the pyrolysis oil could not be deter-
flash pyrolysis. This difference could be due to the higher ash mined using an IKA C5003 bomb calorimeter due to the
content (both alkali and heavy metals) in the contaminated elevated water content in the oil. The pyrolysis oil calorific
sample (Table 1). Ash content contributes directly to higher value is strongly correlated to its water content [21].
char production [50]. Piskorz et al. [51] found that char yield of
S. bicolor after fast pyrolysis at 450  C was 19.09%, while the 3.3.3. Char characterization
gaseous fraction was 15.39%. Higher gas yields were found in Char resulting from both flash and slow pyrolysis had higher
case of flash pyrolysis for both, control and contaminated calorific values than the initial raw materials (Table 5). Slow
samples, compared to the slow pyrolysis. A possible expla- pyrolysis of control sample produced char with a higher
nation is that part of the dusty biomass was stuck to the walls calorific value (26.9 MJ kg
1) compared to flash pyrolysis
of the connection pipe between the water cooled and ice (21.7 MJ kg
1). Conversely, in the contaminated sample flash
cooled condenser. The mass of this dusty biomass was not pyrolysis produced char with slightly higher calorific value
included in the mass of the pyrolysis oil, but rather included in (19.7 MJ kg
1) compared to slow pyrolysis (18.2 MJ kg
1). Stals
the mass of the gaseous fraction. et al. [21] found that the calorific value of the char produced
from heavy metal contaminated wood using flash pyrolysis at
3.3.2. Pyrolysis oil characterization 450  C was 24 MJ kg
1. Due to the higher mineral content of the
The pyrolysis oil had a light brown color, this could be due to contaminated sample, a lower HHV for obtained char could be
the high water content in the pyrolysis oil. Thus was expected compared to the control sample. The difference in
confirmed by KF volumetric titration. Oil derived from HHV values between contaminated and control samples were
contaminated sample had a higher water content than the great when slow pyrolysis was applied, a possible explanation
control sample (Table 4). This difference could be due to the could be that the high content of ash and mineral content in
higher content of alkali and earth alkali metals in the case of the contaminated sample enhances the volatilization/
contaminated biomass and the slightly higher moisture con- degradation of the biomass during slow pyrolysis in the lab-
tent of the contaminated sample compared to the control scale reactor (as is confirmed by the lower char yield). As a
[43,52]. Moreover, the water in the oil is mainly formed during result, more compounds with a high energy content could be
pyrolysis by dehydrogenation of hydroxyl groups [53]. In case removed.

Fig. 8 e Pyrolysis product yields as mass fraction of dry S. bicolor shoots by flash (FP) and slow pyrolysis (SP). * Gas yield was
calculated by difference

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
10 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

Table 5 e HHV calorific value and energy recovery of the Table 7 e Heavy metal recovery in the different pyrolysis
char from flash and slow pyrolysis. products obtained by flash (FP) and slow (SP) pyrolysis of
Parameter CTR CTM Sorghum bicolor shoots.
Treatments Product yields CTM
FP SP FP SP

1
HHV calorific value (MJ kg ) 21.7 26.9 19.7 18.2 Ni % Zn %
Energy recovery (%) 27.4 76.9 40.0 65.5 FP Char 51a 60a
Oil 0 0
Tar 4 2
SP Char 99 98
Energy recovery results (Table 5) pointed out that slow Oil 0 0
b b
pyrolysis was more efficient than flash pyrolysis, in case only Tar
a
the chars were taken into account. In fact, the energy recovery Rest of heavy metals is bound in the sand (as heat transfer
in the control sample was 76.9% for slow pyrolysis while it was medium).
b
only 27.4% for flash pyrolysis. A similar trend was observed for Trace amount of tar was produced during the slow pyrolysis and
therefore heavy metal concentration could not be determined.
the contaminated sample, energy recovery was 65.5% when
slow pyrolysis was applied and 40% when flash pyrolysis was
applied. These results are mainly caused by the higher char
production in case of slow pyrolysis. below 450  C. This result is also in agreement with Lievens
et al. [55] who reported that the oil was free of heavy metals
3.3.4. Fate and recovery of heavy metals when fast pyrolysis of heavy metal contaminated willow
The total content of heavy metals in the char, oil and tar containing high amount of heavy metals (e.g. Zn at 656 and
produced by flash and slow pyrolysis are shown in Table 6. To 4500 mg kg
1) was applied at 350  C. In addition, Kistler and
understand better the fate of heavy metals during the pyrol- Widmedz [56] found that for a pyrolysis temperature main-
ysis process, recoveries of Ni and Zn in each pyrolysis product tained between 500 and 600  C metals are retained within the
were calculated based on the total amount of heavy metals in char.
the raw materials (Fig. 4) and the total amount of heavy metals
obtained in the different pyrolysis fractions (Table 6). The re-
coveries results, presented in Table 7, demonstrate that 4. Conclusion
almost all Ni and Zn were recovered in the char when slow
pyrolysis was applied, while low percentages were recovered Phytoextraction is a long-term remediation effort, requiring
in the char fraction when flash pyrolysis was applied. In this many cropping cycles to decontaminate soils from metal
case, when the char fraction is collected, it is mixed with sand pollutants to acceptable levels. Metal phytoextraction per-
from the pyrolysis reactor used as heating transfer. Therefore, formance depends on the amount of biomass production and
sand is separated carefully from the char fraction [21]. the metal concentration in shoots and roots. Based on these
Consequently, fractions of the heavy metals are found back in criteria and on our results, S. bicolor could be a good candidate
the sand. However, many previous studies confirmed that for phytoremediation and biomass production in moderately
heavy metals are retained/bound in the heat transfer medium contaminated soils as it is able to concentrate metals in roots
[54]. as well to translocate low but substantial amounts to shoots,
For both types of pyrolysis, pyrolysis oil contained no yielding a considerable harvestable biomass.
detectable concentrations of Ni and Zn. This result was in As a result of the flash and slow pyrolysis, heavy metal
agreement with Stals et al. [21] who reported that only negli- contaminated biomass was reduced in weight and volume.
gible amounts of heavy metals were found in the oil when fast Furthermore, it is important to mention that heavy metals
pyrolysis of heavy metal contaminated hardwoods from were mainly recovered in the char fraction when the slow
phytoremediation containing high amount of heavy metals pyrolysis process was applied, while the metals were retained
(e.g. Zn at 349 and 822 mg kg
1) was applied at temperature in the char and the heating transfer medium sand in case of

Table 6 e Heavy metal recovery in the different pyrolysis products obtained by flash (FP) and slow (SP) pyrolysis of Sorghum
bicolor shoots.
Treatment Product CTM CTR
Ni (mg kg
1) Zn (mg kg
1) Ni (mg kg
1) Zn (mg kg
1)
FP Char 115 302 33 76
Pyrolysis oil <DL <DL <DL <DL
Heavy oil 36 55 14 14
SP Char 137 347 5 143
Pyrolysis oil <DL <DL <DL <DL
a a a a
Tar

<DL: below detection limit.

a
Trace amount of tar was produced during the slow pyrolysis and therefore heavy metal concentrations could not be determined.

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2 11

flash pyrolysis. In both pyrolysis processes produced oils were editors. Sustainable alternative fuel feedstock opportunities,
free of heavy metals. Char calorific value was higher than the challenges and roadmaps for six U.S. Regions; Proceedings of
calorific value of the raw materials. In addition, our results the Sustainable Feedstocks for Advance Biofuels Workshop;
28e30 September 2010; Atlanta, GA. IA, USA: Soil and Water
pointed out that slow pyrolysis was more efficient than flash
Conservation Society; 2010. pp. 149e59.
pyrolysis. In fact, the energy recovery in the slow pyrolysis [12] Monti A, Venturi G. Comparison of energy performance of
was strikingly higher than flash pyrolysis. However, when fibre sorghum, sweet sorghum and wheat monocultures in
using char for energy (combustion) purposes, additional gas northern Italy. Eur J Agron 2003;19(1):35e43.
filter requirements are necessary to trap the heavy metals. [13] Renewable Energy World. Bioethanol-Industrial world
Pyrolysis seems one of the promising ways for processing perspective www.jxj.com/magsandj/rew/200_03/bioethanol.
phytoremediation biomass, removing the bottleneck for html; 2000.
[14] Sas-Nowosielska A, Kucharski R, Ma-kowski E, Pogrzeba M,
further implementing phytoremediation cleanup procedures.
Kuperberg JM, Kryn  ski K. Phytoextraction crop disposal-an
It also allows to economically valorize contaminated soils for unsolved problem. Environ Pollut 2004;128(3):373e9.
production of biomass that can be converted to energy and/or [15] Kumar PBAN, Dushenkov V, Motto H, Raskin I.
other products with a higher added value. The possible eco- Phytoextraction e the use of plant to remove heavy metals
nomic revenue from any valorization opportunity can be from soils. Environ Sci Technol 1995;29(5):1232e8.
considered to contribute in the costs of remediation. However, [16] Salt DE, Blaylock M, Kumar NPBA, Dushenkov V, Ensley BD,
Chet I, et al. Phytoremediation: a novel strategy for removal
more in depth investigations are still required to valorize the
of toxic metals from the environment using plants. Curr
obtained pyrolysis products.
Opin Biotechnol 1995;13(5):468e74.
[17] Raskin I, Smith RD, Salt DE. Phytoremediation of metals:
using plants to remove pollutants from the environment.
references Curr Opin Biotechnol 1997;8(2):221e6.
[18] Garbisu C, Alkorta I. Phytoextraction: a cost-effective plant-
based technology for the removal of metals from the
[1] Weyens N, van der Lelie D, Taghavi S, Newman L, environment. Bioresour Technol 2001;77(33):229e36.
Vangronsveld J. Exploiting plantemicrobe partnerships for [19] Mulligan CN, Yong RN, Gibbs BF. Remediation technologies
improving biomass production and remediation. Trends for metal-contaminated soils and groundwater: an
Biotechnol 2009;27(10):591e8. evaluation. Eng Geol 2001;60(1e-4):193e207.
[2] Ruttens A, Boulet J, Weyens N, Smeets K, Adriaensen K, [20] Van Ginneken L, Meers E, Guisson R, Ruttens A, Tack FMG,
Meers E, et al. Short rotation coppice culture of willow and Vangronsveld J, et al. Phytoremediation for heavy metal
poplar as energy crops on metal contaminated agricultural contaminated soils and combined bio-energy production. J
soils. Int J Phytoremed 2011;13(1):194e207. Environ Eng Landsc Manag 2007;15(4):227e36.
[3] Librenti I, Ceotto E, Di Candilo M. Biomass characteristics and [21] Stals M, Carleer R, Reggers G, Schreurs S, Yperman J. Flash
energy contents of dedicated lignocellulosic crops. In: Third pyrolysis of heavy metal contaminated hardwoods from
International Symposium on Energy from Biomass and Waste. phytoremediation: characterization of biomass, pyrolysis oil
8e11 November 2010; Venice, Italy. Italy: CISA Publisher, and char/ash fraction. J Anal Appl Pyrol 2010;89(1):22e9.
Environmental Sanitary Engineering Centre; 2010. pp. 8e15. [22] McGrath SP. Phytoextraction for soil remediation. In:
[4] Sheoran V, Sheoran S, Poonia P. Role of hyperaccumulators Brooks RR, editor. Plants that hyperaccumulate heavy
in phytoextraction of metals from contaminated mining metals, their role in phytoremdiation, microbiology,
sites: a review. Crit Rev Environ Sci Tec 2011;41(2):168e214. archaeology, mineral exploration and phytomining.
[5] Ghosh M, Singh SP. A review on phytoremediation of heavy Wallingford, UK: CAB International; 1998. pp. 261e87.
metals and utilization of its by products. Appl Ecol Environ [23] Blaylock MJ, Huang JW. Phytoextraction of metals. In:
Res 2005;3(41):14e8. Raskin I, Ensley BD, editors. Phytoremediation of toxic
[6] Schröder P, Herzig R, Bojnov B, Ruttens A, Nehnevajova E, metals: using plants to clean up the environment. New York:
Stamatiadis S, et al. Bioenergy to save the world. Producing John Wiley and Sons; 2000. pp. 53e70.
novel energy plants for growth on abandoned land. Environ [24] Bridgwater AV, Meier D, Radlein D. An overview of fast
Sci Pollut Res 2008;15(3):196e204. pyrolysis of biomass. Org Geochem 1999;30(12):1479e93.
[7] Vangronsveld J, Herzig R, Weyens N, Boulet J, Adriaensen K, [25] Lievens C, Yperman J, Vangronsveld J, Carleer R. Study of
Ruttens A, et al. Phytoremediation of contaminated soils and the potential valorization of heavy metal contaminated
groundwater: lessons from the field. Environ Sci Pollut Res biomass via phytoremediation by fast pyrolysis: part I.
2009;16(7):765e94. Influence of temperature, biomass species and solid heat
[8] Chaney RL, Reeves PG, Ryan JA, Simmons RW, Welch RM, carrier on the behavior of heavy metals. Fuel 2008;87(10e-
Angle JS. An improved understanding of soil Cd risk to 11):1894e905.
humans and low cost methods to phytoextract Cd from [26] Demirbas A. Biomass resource facilities and biomass
contaminated soils to prevent soil Cd risks. Biometals conversion processing for fuels and chemicals. Energ
2004;17(5):549e53. Convers Manag 2001;42(11):1357e78.
[9] Meers E, Ruttens A, Hopgood M, Lesage E, Tack FMG. [27] Raveendran K, Ganesh A, Khilar KC. Influence of mineral
Potential of Brassica rapa, Cannabis sativa, Helianthus matter on biomass pyrolysis characteristics. Fuel
annuus and Zea mays for phytoextraction of heavy metals 1995;74(12):1812e22.
from calcareous dredged sediment derived soils. [28] Demirbas A, Arin G. An overview of biomass pyrolysis. Energ
Chemosphere 2005;61(4):561e72. Sources 2002;24(5):471e82.
[10] Unger PW. Alternative and opportunity dry land crops and [29] Demirbas A. Pyrolysis of ground beech wood in irregular
related soil conditions in the Southern Great Plains. Agron J heating rate conditions. J Anal Appl Pyrol 2005;73(1):39e43.
2001;93(1):216e26. [30] Demirbas A. Effect of initial moisture content on the yields of
[11] Shoemaker CE, Bransby DI. The role of sorghum as a oily products from pyrolysis of biomass. J Anal Appl Pyrol
bioenergy feedstock. In: Braun R, Karlen D, Johnson D, 2004;71(2):803e15.

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
12 b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e1 2

[31] Environmental Protection Agency (EPA). Ecological effects, populations of Thlaspi caerulescens from Western Europe.
test guidelines; OPPTS 850.4800 plant uptake and Plant Cell Environ 2003;26(10):1657e72.
translocation test; 1996. EPA 712-C-96e159. [43] Fahmi R, Bridgwater AV, Donnison I, Yates N, Jones JM. The
[32] Millner PD, Kitt DG. The Beltsville method of soilless effect of lignin and inorganic species in biomass on pyrolysis
production of vesicular-arbuscular mycorrhizal fungi. oil yields, quality and stability. Fuel 2008;87(7):1230e40.
Mycorrhiza 1992;2(1):9e15. [44] Reichman SM. The responses of plants to metal toxicity: a
[33] Smets K, Adriaensens P, Vandewijngaarden J, Stals M, review focusing on copper, manganese and zinc. Melbourne;
Cornelissen T, Schreurs S, et al. Water content of pyrolysis Australia: Australian Minerals & Energy Environment
oil: comparison between Karl Fischer titration, GC/MS- Foundation; 2002.
corrected azeotropic distillation and 1H NMR spectroscopy. J [45] Meszaros E, Jakab E, Varhegyi G, Szepesvary P, Marosvolgyi B.
Anal Appl Pyrol 2011;90(2):100e5. Comparative study of the thermal behavior of wood and bark
[34] Skoog DA, West DM, Holler FJ. Fundamentals of analytical of young shoots obtained from an energy plantation. Anal
chemistry: an introduction. 7th ed. New York, USA: Saunders Appl Pyrol 2004;72(2):317e28.
College Publishing; 1996. [46] Prins MJ, Ptasinski KJ, Janssen F. Torrefaction of wood part 1.
[35] McGrath SP, Zhao FG. Phytoextraction of metals and Weight loss kinetics. Anal Appl Pyrol 2006;77(1):28e34.
metalloids from contaminated soils. Curr Opin Biotechnol [47] Tsamba AJ, Yang WH, Blasiak W. Pyrolysis characteristics
2003;14(3):277e82. and global kinetics of coconut and cashew nut shells. Fuel
[36] Utmazian MNDS, Wieshammer G, Vega R, Wenzel WW. Process Technol 2006;87(6):523e30.
Hydroponic screening for metal resistance and accumulation [48] Mohan D, Pittman CU, Steele PH. Pyrolysis of wood/biomass
of cadmium and zinc in twenty clones of willows and for bio-oil: a critical review. Energ Fuel 2006;20(3):848e89.
poplars. Environ Pollut 2007;148(1):155e65. [49] Raveendran K, Ganesh A, Khilar KC. Pyrolysis characteristics
[37] Amer N, Al Chami Z, Al Bitar L, Mondelli D, Dumontet S. of biomass and biomass components. Fuel 1996;75(8):987e98.
Evaluation of Atriplex halimus, Medicago lupulina and [50] Antal MJ, Grønli M. The art, science and technology of
Portulaca oleracea for phytoremediation of Ni, Pb, and Zn. Int charcoal production. Ind Eng Chem Res 2003;42(8):1619e40.
J Phytoremed 2013;15(5):498e512. [51] Piskorz J, Majerski P, Radlein D, Scott DS, Bridgwater AV. Fast
[38] Yang M, Cobine PA, Molik S, Naranuntarat A, Lill R, pyrolysis of sweet sorghum and sweet sorghum bagasse. J
Winge DR, et al. The effects of mitochondrial iron Anal Appl Pyrol 1998;46(1):15e29.
homeostasis on cofactor specifity of superoxide dismutase 2. [52] Demirbas A. Progress and recent trends in biofuels. Prog
EMBO J 2006;25(8):1775e83. Energy Combust 2009;33(1):1e18.
[39] Hansch R, Mendel R. Physiological functions of mineral [53] Sánchez ME, Menéndez JA, Domı́nguez A, Pis JJ, Martı́nez O,
micronutrients (Cu, Zn, Mn, Fe,Ni, Mo, B, Cl). Curr Opin Plant Calvo LF, et al. Effect of pyrolysis temperature on the
Biol 2009;12(3):259e66. composition of the oils obtained from sewage sludge.
[40] Al Chami Z, Cavoski I, Mondelli D, Miano T. Effect of compost Biomass Bioenergy 2009;33(6e7):933e40.
and manure amendments on zinc speciation, plant content, [54] Stals M, Thijssen E, Vangronsveld J, Carleer R, Schreurs S,
and translocation in an artificially contaminated soil. Yperman J. Flash pyrolysis of heavy metal contaminated
Environ Sci Pollut Res 2013;20(7):4766e76. biomass from phytoremediation: influence of temperature,
[41] Terzano R, Al Chami Z, Vekemans B, Janssens K, Miano T, entrained flow and wood/leaves blended pyrolysis on the
Ruggiero P. Zinc distribution and speciation within rocket behaviour of heavy metals. J Anal Appl Pyrol 2010;87(1):1e7.
plants (Eruca vesicaria L. Cavalieri) grown on a polluted soil [55] Lievens C, Carleer R, Cornelissen T, Yperman J. Fast pyrolysis
amended with compost as determined by XRF of heavy metal contaminated willow: influence of the plant
microtomography and micro-XANES. J Agric Food Chem part. Fuel 2009;88(8):1417e25.
2008;56(9):3222e31. [56] Kistler RC, Widmer F. Behavior of chromium, nickel, copper,
[42] Roosens N, Verbruggen N, Meerts P, Ximenez-Embun P, zinc, cadmium, mercury and lead during the pyrolysis of
Smith J. Natural variation in cadmium tolerance and its sewage sludge. Environ Sci Technol 1987;21(7):704e8.
relationship to metal hyperaccumulation for seven

Please cite this article in press as: Al Chami Z, et al., Evaluation of flash and slow pyrolysis applied on heavy metal contaminated
Sorghum bicolor shoots resulting from phytoremediation, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/
j.biombioe.2014.02.027
View publication stats