i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8

Available online at www.sciencedirect.com

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

Hydrogen production from glycerol by supercritical water

gasification in a continuous flow tubular reactor

Simao Guo, Liejin Guo*, Changqing Cao, Jiarong Yin, Youjun Lu, Ximin Zhang
State Key Laboratory of Multiphase Flow in Power Engineering (SKLMF), Xi’an Jiaotong University, Xi’an 710049, Shaanxi, PR China

article info abstract

Article history: In this work, glycerol was used for hydrogen production by supercritical water gasification.
Received 17 November 2011 Experiments were conducted in a continuous flow tubular reactor at 445w600
C, 25 MPa,
Received in revised form with a short residence time of 3.9w9.0 s. The effects of reaction temperature, residence time,
22 December 2011 glycerol concentration and alkali catalysts on gasification were systematically studied. The
Accepted 25 December 2011 results showed that the gasification efficiency increased sharply with increasing tempera-
Available online 20 January 2012 ture above 487
C. A short residence time of 7.0 s was enough for 10 wt% glycerol gasification
at 567
C. With the increase of glycerol concentration from 10 to 50 wt%, the gasification
Keywords: efficiency decreased from 88% to 71% at 567
C. The alkali catalysts greatly enhanced water-
Hydrogen production gas shift reaction and the hydrogen yield in relation to catalysts was in the following order:
Glycerol NaOH > Na2CO3>KOH > K2CO3. The hydrogen yield of 4.93 mol/mol was achieved at 526
C
Supercritical water gasification with 0.1 wt% NaOH. No char or tar was observed in all experiments. The apparent activation
(SCWG) energy and apparent pre-exponential factor for glycerol carbon gasification were obtained
Kinetics by assuming pseudo first-order kinetics.
Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.

1. Introduction for future sustainable utilization and environmental advan-

tages. The biodiesel glycerol produced from biomass
With the rapid increase of the world biodiesel production in was considered to be renewable and carbon neutral, which
recent years, a glut of glycerol is produced as a byproduct of can provide the opportunity for producing renewable
the biodiesel industry, and nearly 1 ton of glycerol was hydrogen. Many methods have been applied to convert glyc-
produced for 9 tons of biodiesel. A large surplus leads to erol to hydrogen, including fermentation [6,7], steam
inexpensive glycerol, which was listed in top 10 potential reforming [8e10], autothermal reforming [11], aqueous-phase
candidates by U.S.DOE [1] for producing bio-based chemical reforming [12] and supercritical water gasification
products. Many researchers focused on using this cheap (SCWG) [13,14]. All these methods were reviewed by Adhikari
glycerol as a raw material to produce high value-added et al. [15].
chemicals like 1,3-propanediol [2], acrolein [3] and hydroxyl Among these methods, SCWG of hydrocarbons and
acetone [4]. In particular, using glycerol to produce hydrogen biomass are supposed to be a promising method for hydrogen
has also received much attention. production [16e18]. Supercritical water (SCW) is defined as the
Hydrogen has attracted extensive attention as a clean fuel. water beyond the critical point (374
C, 22.1 MPa) and has
However, at present, 96% of hydrogen is produced from fossil many unique properties, such as low viscosity, high diffusivity
fuels [5]. Bio-based or other renewable hydrogen is expected and low dielectric constant. These properties make SCW

* Corresponding author. Tel.: þ86 29 8266 3895; fax: þ86 29 8266 9033.
E-mail address: [email protected] (L. Guo).
0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2011.12.135
5560 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8

a good solvent for both organics and gases, providing a bench-scale continuous flow tubular reactor at a tempera-
a homogenous condition for gasification with short residence ture range of 445w600
C and 25 MPa, with a short residence
time and high-efficiency [19,20]. In recent decades, much time of 3.9w9.0 s. The effects of reaction temperature, resi-
research into SCWG has been done and a large variety of dence time and glycerol concentration on gasification were
feedstocks and model compounds were gasified in SCW, such systematically studied. Considering the fact that the crude
as wood [21], sludge [22], coal [23], glucose [24], cellulose [25], glycerol obtained from biodiesel industry always contains
phenol [26] and C-1 compounds [27]. alkalis [31,32], the influences of four types of alkali catalysts
However, for glycerol gasification in SCW, very few studies (NaOH, Na2CO3, KOH and K2CO3) on gasification were inves-
have been reported(less than 10), especially with respect to tigated. Based on the pseudo first-order kinetics assumption,
hydrogen production. Antal et al. [28] first studied the glycerol the overall reaction rate of glycerol carbon gasification was
decomposition in SCW at 500
C and 34.5 MPa, they found the analyzed. The activation energy and pre-exponential factor
gaseous products contained H2, CO2, CO, CH4, C2H4 and C2H6. were also obtained for chemical reaction engineering to
Kersten et al. [21] reported that only dilute glycerol solution design and scale up the SCWG reactor.
can be completely converted to gas phase without catalysts.
Byrd et al. [13] reported that near-theoretical hydrogen yield of
7 mol/mol glycerol was achieved by reforming 5 wt% glycerol 2. Experimental
solution in SCW over Ru/Al2O3 as a catalyst at 800
C. May et al.
[29] used Ru/ZrO2 to catalyze glycerol gasification in SCW, but 2.1. Materials
the hydrogen yield was very low, which was less than 0.6 mol/
mol glycerol at 550
C and 35 MPa. Chakinala et al. [14] gasified The glycerol(>99.0 wt%) used in this study was produced by
glycerol as a model compound of microalgae in SCW and Tianjin Fu Chen Chemical Reagent Factory. The KOH and
studied the influence of additives (L-alanine, glycine, L-proline NaOH were obtained from Tianjin HonYan Chemical Reagent
and K2CO3) on gasification, they found the addition of amino Factory. Anhydrous K2CO3 and Na2CO3 were provided by
acid caused coke formation, while the addition of K2CO3 Tianjin TianLi Chemical Reagents Ltd. All these reagents were
enhanced the gasification. Onwudili et al. [30] studied bio- analytical pure.
diesel plant waste (20.8 wt% methanol, 42.3 wt% glycerol and
33.1 wt% fatty acid methyl esters) reforming under subcritical 2.2. Reactor system and experimental procedures
and supercritical water conditions, a gaseous effluent con-
taining up to 90% by molar fraction of hydrogen was obtained The experiments were done in a continuous SCWG system
during alkaline gasification. developed in our laboratory. Fig. 1 shows a schematic diagram
To date, the number of experimental runs on SCWG of of this system [23]. The maximum throughput capacity of the
glycerol is still limited. In addition, most of the previous works system is 16 kg/h. The reactor was fabricated from Hastelloy-
were done with low glycerol concentration and in miniature C276 tubing with 17.15 mm o.d.  10.85 mm i.d.  1.24 m
scale (both batch and continuous flow reactors). In order to length. The design temperature and pressure were 800
C and
obtain more information about the hydrogen production 30 MPa, respectively. During the experiments, the feedstock
behavior of the glycerol gasification in SCW and further (glycerol solution) was first mixed with preheated water to
determine the feasibility of SCWG of glycerol, in this work, realize fast heating for avoiding side reactions [33]. The flow
high glycerol concentration (10e50 wt%) was gasified in ratio of feedstock and preheated water was 1:4 in all

Fig. 1 e Scheme of supercritical water gasification system [23]. 1-Water tank 2-flush water tank 3-main flow pump
4- auxiliary pump 5-flush pump 6-cooler 7-preheater 8-mixer 9-reactor 10-charger 11- silo 12-filter 13-back-pressure
regulator s 14-gas-liquid separators 15-gas flow meter 16-manometer 17-main flow meter 18-auxiliary flow meter.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8 5561

experiments. Then, the mixture flow entered the reactor

which was located in an oven and heated by electric stove 3. Result and discussion
wires. A PID controller was used to maintain the reactor at the
desired temperature. The reaction temperature mentioned in 3.1. Effect of temperature
this paper was measured by a K type thermocouple inserted in
the reactor. The pressure of the system was controlled by Reaction temperature is the most important parameter that
a back pressure regulator. The products flow after the reaction influences the SCWG [19]. Higher temperatures promote the
were cooled down in the cooler, and a gas-liquid separator freeeradical reactions [35] which lead to higher gas yield. Fig. 2
after the back pressure regulator was used to separate the shows results of 10 wt% glycerol gasification in SCW in the
products. The flow rate of gaseous product was measured by
a wet type gas meter. The liquid was collected for further a 120
analysis. More detailed information about this system can be
found in our previous work [23,34].
100
The uncertainty of experimental measurements was

GE,CE and HE/ %

analyzed. The uncertainty of temperature and pressure are
80
less than 2.06% and 2.00% respectively; the uncertainty of gas
yield is less than 2.17%.
60

2.3. Analytical methods 40 GE

CE
The composition of the gaseous products was measured by 20 HE
Agilent 7890A gas chromatograph(GC), which was equipped
with thermal conductivity detector (TCD) and capillary
0
column C-2000 purchased from LanZhou Institute of Chemical 450 475 500 525 550 575 600
Physics in China. High purity helium with a flow rate of 30 ml/ Temperature / °C
min was used as the carrier gas. b 4.0
The liquid effluent was pretreated by solid phase extraction
Gas yield / mol⋅mol glycerol

3.5 H2 CO
(SPE), using Agilent SampliQ C8 SPE columns. Then, the organic
CO2 CH4
extract sample was injected into the GC(Agilent 6890) with HP- 3.0
INNOWax capillary column. A mass selective detector C2H4 C2H6
-1

(MSD5973) was used to identify the organics in the sample. 2.5

2.0
2.4. Data interpretation 1.5

The residence time is defined as the reactor volume (V) 1.0

divided by volumetric flow rate of reaction flow (Vo) at reaction
0.5
temperature and pressure:
0.0
V LAr 450 475 500 525 550 575 600
s¼ ¼ (1)
Vo Q Temperature / °C
Where L is the length of reactor (m), A is the inner cross- c
60
section area (m2), r is the density of the reaction flow which
was assumed to be the pure water at the reaction condition
(kg/m3), Q is the mass flow rate of reaction flow (kg/s).
Molar fraction / %

The gasification efficiency (GE), carbon gasification effi- H2 CO

40 CO2 CH4
ciency (CE) and hydrogen gasification efficiency (HE) were
defined as the following equations: C2H4 C2H6

20
GE ¼ (the total mass of the product gas)/(the total mass of the
feedstock) (2)

0
CE ¼ (the total carbon in the product gas)/(the total carbon in 480 500 520 540 560 580 600
the feedstock) (3) Temperature / °C
Fig. 2 e Effect of temperature on glycerol gasification (a) GE,
HE ¼ (total hydrogen in the product gas)/(the total hydrogen in CE and HE; (b) Gas yield; (c)Molar fraction(10 wt% glycerol,
the feedstock) (4) 5 kg/h, 25 MPa).
5562 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8

temperature range of 445w600
C at 25 MPa and a flow rate of

a
5 kg/h. The main gaseous products were H2, CO, CO2 and CH4,
40
with a small amount of C2H4 and C2H6. As shown in Fig. 2,
when the temperature was 445
C, the gasification efficiency

GE,CE and HE/ %

was only 4.3%, indicating that at 445
C or below, the CeC
30
bond of glycerol may not be broken, and little glycerol was
gasified. However, even in this low temperature, no tar or char
was observed. As the temperature increased from 487 to 20
600
C, the gasification efficiency was greatly promoted,
GE
increasing from 38% to 102%, carbon gasification efficiency
CE
increased from 31% to 89%, and hydrogen gasification effi- 10 HE
ciency increased from 35% to 105%, meaning that the SCW
also contributed some hydrogen in product gas. The hydrogen
yield of 3.47 mol/mol was obtained at 600
C, which was half of 0
5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0
theoretical limit of equation (5). At different temperatures, the
Residence time/ s
yields of CO and CO2 were about the same value. The methane b
yield slightly increased with increasing temperature, while it 1.4

Gas yield / mol⋅mol glycerol

was still low and only 0.24 mol/mol at 600
C. The gas
composition was almost unchanged when the temperature 1.2
increased, the molar faction of H2, CO/CO2, CH4 and C2Hx were
1.0 H2 CO
about 57%, 18%w20%, 3%w4% and <1%, respectively. The
-1
fast-type water-gas shift reaction [24] was not observed in this CO2 CH4
0.8
work, perhaps because our experiments did not reach the C2H 4 C2H6
needed temperature for that reaction, which must exceed 0.6
660
C according to Lee’s work [24].
0.4

0.2
C3H8O3 þ 3H2O / 3CO2 þ 7H2 (5)
0.0
It is believed that the gases and organics can dissolve in 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0
SCW, making the mixture a single-phase [20], but when the Residence time / s
effluent flow was cooled to ambient temperature, since the
Fig. 3 e Effect of residence time on glycerol gasification in
smaller solubility at ambient temperature than that in SCW,
SCW at 487
C (a) GE, CE and HE; (b) Gas yield(10 wt%
the gaseous products like hydrogen and carbon monoxide
glycerol, 25 MPa).
would escape from the water. So, the gradually cooled effluent
would change from a single-phase flow to a gas-liquid two-
phase flow. Moreover, when the total gas yield reached
a certain value, for instance, 2.33L/min (at 600
C,10 wt%,5 kg/
a 4-fold increase. However, the increasing trend of gasification
h), which meant that the volume fraction of total gas yield was
efficiency decreased with increasing residence time, indi-
as high as 12% in the whole effluent flow at 25 MPa. We
cating that further extending the residence time cannot
observed that the liquid and the gas passed through the back
significantly increase gasification efficiency if the temperature
pressure regulator alternately, causing slight intermittent
remained at 487
C. While at a higher temperature of 567
C, as
impact on the back pressure regulator and pipeline system
shown in Fig. 4, the gasification efficiency, carbon gasification
due to the quite different compressibility factors between
efficiency and hydrogen gasification efficiency only showed
liquid effluent and gas. This impact phenomenon was similar
a slightly increasing tendency with increasing residence time.
to the impact which occurred in slug flow and might cause
When the residence time increased from 4.2 to 7.3 s, the
damages to the experimental system [36]. So, this phenom-
gasification efficiency increased from 84% to 94% and the
enon should be considered and avoided in the further SCWG
hydrogen yield only increased from 3.13 to 3.37 mol/mol, such
system developments, especially for scaling-up.
results indicated it was not necessary to take longer residence
time at higher temperatures. A shorter residence time of 7.0 s
3.2. Effect of residence time
was sufficient to obtain relatively high gasification efficiency
and could increase the treating capacity of the system.
The effect of residence time on the gasification of 10 wt%
glycerol in SCW was investigated at 487
C and 567
C,
respectively. The different residence times were obtained by 3.3. Effect of concentration
changing the flow rate. As shown in Fig. 3, at 487
C, the
increase of residence time can significantly enhance glycerol Fig. 5 shows the effect of glycerol concentration on gasifica-
gasification. When the residence time increased from 5.2 to tion at 567
C, 25 MPa. As the glycerol concentration increased
9.0 s, the gasification efficiency increased from 8% to 42% and from 10 to 25 wt%, the gasification efficiency, hydrogen gasi-
the hydrogen yield increased from 0.35 to 1.35 mol/mol, nearly fication efficiency and carbon gasification efficiency decreased
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8 5563

a a 100

100
90 GE

CE,GE and HE/ %

CE
CE,GE and HE / %

80 HE
80

70
GE
60 CE
60
HE

50
40 10 20 30 40 50
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 Mass concentration / wt%
Residence time/ s
b
b 3.5
60

50
Gas yield / mol⋅mol -1 glycerol

3.0 H2 CO

Molar fraction / %
H2 CO CO2 CH4
40
2.5 C2H4 C2H6
CO2 CH4
2.0 C2H4 C2H6 30

1.5 20

1.0
10
0.5
0
10 20 30 40 50
0.0
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 Mass concentration / wt%
Residence time / s
Fig. 5 e Effect of concentration on glycerol gasification in
Fig. 4 e Effect of residence time on glycerol gasification in SCW (a) GE, CE and HE; (b) Gas molar fraction(567
C, 5.8 s
SCW at 567
C (a) GE, CE and HE; (b)Gas yield(10 wt% residence time, 25 MPa).
glycerol, 25 MPa).

and higher glycerol concentration may lead water to become

from 88%, 93% and 75% to 72%, 72% and 69%, respectively. the limiting reactant. While the pyrolysis Reaction (7) may be
The hydrogen gasification efficiency decreased more quickly insensitive to water concentration. When the glycerol
than carbon gasification efficiency. The molar fractions of H2 concentration initially increased from 10 to 25 wt%, the yields
and CO2 decreased from 59.3% and 18.3% to 53.6% and 10.7%, of H2 and CO2 from the steam reforming Reaction (6)
respectively. The molar fraction of CO increased greatly from decreased sharply because of the decreased water concen-
17.9% to 29.7%. The molar fractions of CH4 and C2Hx changed tration, so the gasification efficiency and hydrogen gasifica-
little with increasing concentration. However, further tion efficiency decreased, and the molar fractions of H2 and
increase in concentration did not affect the gas yield and gas CO2 also decreased. Further increase in glycerol concentra-
composition too much, the gasification efficiency and the tion, the gas yield from the steam reforming reaction became
molar fractions of different gases remained stable when the very low, while the pyrolysis reaction which was insensitive to
concentration increased from 25 to 50 wt%. In addition, it is water concentration became the main source of gas yield,
worth noting that even the glycerol concentration was as high resulting in that the gas yield and gas composition remained
as 50 wt%, the gasification efficiency and hydrogen gasifica- stable with increasing glycerol concentration. So, the increase
tion efficiency still maintained a relatively high level (were of glycerol concentration may have an unfavorable effect on
about 70%), and the hydrogen molar fraction reached 52%, no the steam reforming reaction, while it has little effect on
tar or char was observed. pyrolysis reaction.
To explain the tendency shown in Fig. 5, we inferred that
the gases were mainly from two types of reactions, one was
the pyrolysis of glycerol and intermediate products, and the CHxOy þ (2y) H2O / CO2 þ (2 þ x/2  y)H2 (6)
other was steam reforming of glycerol and intermediate
products. The steam reforming Reaction (6) may be very
sensitive to water concentration, because water is a reactant CHxOy / CO þ CO2 þ H2 þ intermediate products (7)
5564 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8

3.4. Effect of alkali catalysts addition observed some black intermediates in preheat process at
200
C), the Na2CO3 may promoted these side reactions in
3.4.1. The effect of different alkali catalysts subcritical condition. While in our experiments, the glycerol
The crude glycerol obtained from biodiesel industry always was quickly heated to higher temperature, so the side reac-
contains some alkalis due to base catalytic transesterification tions may be avoided. Although the hydrogen gasification
process [31,32]. Many smaller biodiesel plants simply discard efficiency was largely increased after adding alkali catalysts, It
the glycerol byproduct as a waste, because it was expensive to can be seen from Table 1 that the improvement of carbon
remove soluble alkalis in crude glycerol [37]. While in SCWG gasification efficiency was not obvious, which was similar to
process, it is reported [38e40] that alkali catalysts like KOH, the results of Kersten et al. [21].
NaOH, K2CO3 and Na2CO3 can significantly enhance the water-
gas shift Reaction (8). We can take advantage of alkalis in
crude glycerol in SCWG. In this study, the effect of various CO þ H2O / CO2 þ H2 (8)
alkali catalysts (K2CO3, KOH, NaOH and Na2CO3) on 10 wt%
Meanwhile, the composition of gas products before and
glycerol gasification was investigated at the lower tempera-
after adding catalysts also changed significantly. Table 2
ture of 526
C, 6.5 s residence time. The amount of catalysts
displays the molar fractions of gases with and without
addition was 0.5 wt%. As shown in Fig. 6, the addition of alkali
various catalysts. The theoretical values of molar fractions of
catalysts significantly increased the H2 and CO2 yields and
gases at the thermodynamic equilibrium state were also given
decreased CO yield. When 0.5 wt% NaOH was used, the H2 gas
by using the method from Lu and Yan [42,43]. From Table 2, it
yield was up to 4.55 mol/mol, the CO gas yield was only
can be seen that the addition of alkali catalysts made the
0.09 mol/mol, while these values without catalyst were
molar fractions of gases draw close to the thermodynamic
2.57 mol/mol and 0.86 mol/mol, respectively. The hydrogen
equilibrium values. The molar fractions of hydrogen and CO2
gasification efficiency was up to 118.2%, which confirmed that
without catalysts were 57.5% and 19.3%, respectively, while
the water-gas shift reaction was greatly enhanced, and some
these values were increased to 68.9% and 28.4% with 0.5 wt%
extra gaseous hydrogen was from water. Moreover, although
NaOH, which were very close to the thermodynamic equilib-
we try to control the reaction temperature at 526
C using the
rium value 69.3% and 30.1%, respectively. The molar fraction
same thermal boundary conditions in all experiments, we
of CO decreased from 19.2% to 1.4% before and after adding
found the reaction temperatures with alkali catalysts were
0.5 wt% NaOH.
about 10
C higher than those without alkali catalysts, indi-
cating that the exothermic reaction which was supposed to be
3.4.2. The effect of catalysts amount
water-gas shift Reaction (8) occurred intensely. Fig. 6 also
The effect of the amount of alkali catalysts on glycerol gasi-
showed that the activity of catalysts with respect to the
fication was investigated by adding different amount of (0.05,
hydrogen yields decreased in the order of NaOH > Na2CO3>
0.1, 0.3 and 0.5 wt%) NaOH into 10 wt% glycerol solution at
KOH > K2CO3. The different catalytic effects may be the
526
C. As shown in Fig. 7, the gasification efficiency, carbon
integrated result of different solubility, different dissociation
gasification efficiency and hydrogen gasification efficiency
constants, different charge density and different molecular
firstly increased with increasing amount of NaOH, the
weights among these four alkali catalysts in SCW. Xu et al. [41]
maximum of hydrogen yield of 4.93 mol/mol was obtained
reported that Na2CO3 had a negative effect on glycerol gasifi-
when the amount of NaOH was 0.1 wt%, the hydrogen gasifi-
cation, which was in contrast to our results. This may be
cation efficiency was up to 128%. As the amount of NaOH
because in his experiments, the side reactions like poly-
increased further, the hydrogen yield and gasification effi-
glycerine formation occurred in preheater(the author
ciency slightly decreased. We assumed the solubility limita-
tion may be one of the reasons causing the tendency shown in
5.0 Fig. 7. That is, the catalytic effect increased with increasing
H2
concentration of NaOH below the solubility value. However,
Gas yield / mol⋅mol glycerol

CO
when the concentration of NaOH exceeded the solubility
4.0 CH4
CO2 value, further increase in NaOH addition may not improve the
C2H4 gasification because the surplus NaOH could not be dissolved.
-1

3.0 C2H6 Therefore, it did not mean that the more amount of catalyst

2.0
Table 1 e The GE, CE and HE with different alkali
catalysts(526
C, 6.5 s residence time, 10 wt%
1.0 glycerol D 0.5 wt% catalyst, 25 MPa).
CE/% GE/% HE/%
0.0
NONE K2CO3 Na2CO3 NaOH NONE 64.2 76.5 73.4
KOH
K2CO3 67.5 96.7 105.6
KOH 66.2 96.1 112.7
Fig. 6 e The gas yields of glycerol gasification with different
Na2CO3 71.7 108.2 112.9
alkali catalysts (526
C, 6.5 s residence time, 10 wt%
NaOH 68.6 103.8 118.2
glycerol D 0.5 wt% catalyst, 25 MPa).
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8 5565

Table 2 e The molar fractions of gases in different 4. Kinetics

catalytic conditions and theoretical values of
thermodynamic equilibrium state (526
C, 6.5 s residence
The liquid collected in our experiments was analyzed by
time, 10 wt% glycerol D 0.5 wt% catalyst, 25 MPa).
GCeMS and the liquid products, such as acrolein, allyl alcohol,
H2 CO CH4 CO2 C2H6 C2H4
hydroxyl acetone, acetaldehyde and propionic acid were
NONE 57.5 19.2 3.4 19.3 0.3 0.24 identified. Moreover, small molecular products like methanol,
K2CO3 65.2 0.31 3.7 30.4 0.17 0.24 formaldehyde and formic acid were also detected in near- and
KOH 67.4 0.29 3.2 28.7 0.15 0.21 SCWG of glycerol by other researchers [29,35]. According to
Na2CO3 67.0 1.5 1.1 30.3 0.034 0.042
these liquid products and the discussion of 3.3, we inferred
NaOH 68.9 1.4 1.2 28.4 0.033 0.067
Thermodynamic 69.3 0.0014 0.617 30.1 0 0
that when the glycerol mixed with hot preheated water, the
equilibrium value dehydration of glycerol first occurred, forming acrolein and
hydroxyl acetone. As the temperature increased and reached
a certain value, the pyrolysis of glycerol happened and the
CeC bond of glycerol was broken, resulting in the formation of
C-1 and C-2 compounds like formaldehyde and acetaldehyde
the better glycerol gasification result in SCW. Moreover, it is [28]. Meanwhile, some CO, CO2, H2 and a small amount of CH4
reported that the alkalis were always precipitated in the may be also released in the pyrolysis process [13]. The inter-
reactor system due to their low solubility [34], and too much mediate products produced from the reactions mentioned
alkali salts may accelerate the plugging process. So, on the above were subsequently converted to CO, CO2 and H2 by
premise of guaranteeing the catalytic effect for gasification, steam reforming and further pyrolysis.
the amount of alkali catalysts used should be as small as In order to obtain the basic kinetics information of glycerol
possible. gasification in SCW, we did a series of experiments at 25 MPa,
using 10%wt glycerol solution, with various temperatures and
residence times. Generally, the overall reaction rate of glycerol
carbon gasification r may be expressed as follows:
a 140  
d Cgc  a  b
r¼ ¼ k0 Cgc CH2 O (9)
ds
120
Where Cgc, CH2 O are the molar fraction of total carbon (mol/L)
CE,GE and HE/ %

100 and the molar fraction of water (mol/L) in glycerol solution

respectively; s is residence time (s); a, b are the empirical
orders of reaction and k’ is the “true” reaction rate constant.
80
Considering that the initial glycerol concentration was fixed
and the molar fraction of water was much larger than that of
60
GE glycerol, we assumed ½CH2 O b was a constant. Meanwhile, we
CE supposed a ¼ 1, so, the equation (9) can be expressed as:
40
HE  
d Cgc  
r¼ ¼ k Cgc (10)
20 ds
0.0 0.1 0.2 0.3 0.4 0.5 0
Mass concentration of NaOH / wt% where k ¼ k ½CH2 O b , represents an apparent reaction rate
b constant for glycerol carbon gasification. In fact, the equation
(10) is a pseudo first-order kinetics for glycerol carbon gasifi-
5.0
Gas yield / mol⋅mol glycerol

cation in SCW. For pseudo first-order kinetics, the relationship

between  ln(1 X ) and s can be shown as below by integra-
4.0 H2 CO tion of equation (10) [19]:
CO2 CH4
-1

C2H4 C2H6 lnð1  XÞ ¼ ks (11)

3.0
where X is the carbon gasification efficiency (CE).
Fig. 8 shows the plots of  ln(1 X ) versus s at different
2.0
reaction temperatures. According to equation (11), k value at
each reaction temperature can be determined by the slope of
1.0 the straight line in Fig. 8. The k values were in the range of
0.042w0.379 s1 at the temperature of 487w600
C. Although
0.0 the non-Arrhenius behavior of the overall degradation of
0.0 0.1 0.2 0.3 0.4 0.5
glycerol was founded by Buhler et al. [35], the non-Arrhenius
The amount of NaOH catalyst / wt% behavior mainly occurred at sub- and near-critical condi-
Fig. 7 e The effect of amount of NaOH catalyst on glycerol tions, considered to be the overlay of ionic and freeeradical
gasification (a) GE, CE and HE; (b) Gas yield (526
C, 6.5 s reactions [35]. While in this work, the temperature range was
residence time, 10 wt% glycerol, 25 MPa). 487w600
C and the low temperature range was avoided.
5566 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8

 
3.0 Ea
k ¼ Aexp  (12)
y = (0.379 ± 0.009)x RT
487°C
2.5
526°C It is worth noting that the kinetics obtained above is not
567°C a universal equation which could predict all experimental
y = (0.263 ± 0.009)x
2.0 results due to the assumptions made above. For example, it may
600°C
-Ln(1-X)

not be suitable for SCWG of other concentration glycerol solu-

0
1.5 y = (0.165 ± 0.004)x tions which largely deviate from 10 wt%, because k ¼ k ½CH2 O b
may be greatly changed due to various molar fractions of water.
1.0 Nevertheless, the main purpose of giving our simplified pseudo
first-order kinetics was to provide a basic information of glyc-
0.5 y = (0.042 ± 0.006)x erol gasification in SCW for those who are using similar reactor
systems and experimental conditions, and give them an initial
0.0 concept of the magnitude about reaction rate constants and
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Arrhenius parameters of SCWG of glycerol.
Residence time τ / s
Fig. 8 e First-order plot of L ln(1 L X ) against residence 5. Conclusions
time s for glycerol carbon gasification in SCW at 25 MPa.
The straight lines are the results of linear least-squares fits Glycerol was successfully gasified in SCW and hydrogen-rich
to the data. gas products (hydrogen molar fraction was 52.0e68.9%) were
obtained. No char or tar was observed in all experiments. The
Therefore, in our experiments, the ionic pathway which was experimental results showed that using biodiesel glycerol as
supposed to be dominated at sub- and near-critical tempera- a renewable source to produce hydrogen by SCWG had good
prospects.
ture could be neglected compared to the free-radical pathway
The glycerol gasification efficiency and the hydrogen yield
which prefers higher temperatures. Fig. 9 shows the Arrhe-
increased sharply with increasing temperature above 487
C,
nius plot of the k values obtained in the range of 487w600
C.
102% and 3.47 mol/mol was achieved at 600
C, respectively.
The apparent activation energy Ea and apparent pre-
While at the temperature below 445
C, the glycerol was little
exponential factor A in Arrhenius equation (12) were ob-
gasified. The increasing of residence time can significantly
tained through linear least-squares regression, which were
enhance glycerol gasification at 487
C. While at a higher
104.5  20.3 kJ/mol and 105.901.30 s1 respectively
(The  indicates 95% confidence interval). Chakinala et al. [14] temperature of 567
C, a short residence time of 7.0 s was
reported the apparent activation energy of carbon gasification sufficient to obtain relatively high gasification efficiency, and
of 10 wt% glycerol was 110 kJ/mol, which was quite close to further extending of residence time was not necessary. When
our result. Byrd et al. [13] reported the activation energy Ea of the glycerol concentration increased from 10 to 50 wt%, the
gasification efficiency and hydrogen yield decreased rapidly at
glycerol gasification was 55.9 kJ/mol, which was smaller than
first and then remained steady. Due to the gas-water two-
our result, because in their work, they used Ru/Al2O3 as
phase flow of cooled reactor effluent, the intermittent impact
a catalyst, resulting in a reduction in the activation energy.
on the back pressure regulator and pipeline system was
o observed when the gas yield was high.
Temprature / C
625 600 575 550 525 500 475 The alkali catalysts can greatly increase the gasification
0.0 efficiency and hydrogen yield by promoting water-gas shift
reaction. The hydrogen yield with respect to catalysts was in
-0.5
the following order: NaOH > Na2CO3>KOH > K2CO3. The
-1.0 hydrogen yield of 4.93 mol/mol was achieved at 526
C with
0.1 wt% NaOH.
-1.5
A simple kinetics study was done by assuming pseudo first-
Ln k

-2.0 order kinetics. The overall rate constant of glycerol carbon

gasification in SCW was in the range of 0.042w0.379 s1 at
-2.5
a temperature of 487w600
C. The apparent activation energy
-3.0 This work and apparent pre-exponential factor for glycerol carbon gasifi-
Chakinala et al. cation were 104.5 kJ/mol and 105.90 s1, respectively.
-3.5

-4.0
1.10 1.15 1.20 1.25 1.30 1.35
Reciprocal temprature / 1000K
-1 Acknowledgments

Fig. 9 e Arrhenius plot of the global rate constants k for This work was financially supported by the National Basic
glycerol carbon gasification in SCW at 25 MPa. This Research Program of China (Contracted No. 2009CB220000)
work(-), Chakinala et al.(C). The line was the results of and the National Natural Science Foundation of China
linear least-squares fits to present data. (Contracted No. 50821064).
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8 5567

references [20] Peterson AA, Vogel F, Lachance RP, Froling M, Antal MJ,
Tester JW. Thermochemical biofuel production in
hydrothermal media: a review of sub- and supercritical
water technologies. Energy & Environmental Science 2008;1:
[1] Werpy T, Petersen G, Aden A, Bozell J, Holladay J, White J,
32e65.
et al. Top value added chemicals from biomass. Available at:
[21] Kersten SRA, Potic B, Prins W, Van Swaaij WPM. Gasification
In: Results of screening for potential candidates from sugars
of model compounds and wood in hot compressed water.
and synthesis gas, vol. I. United States Department of Energy
Industrial & Engineering Chemistry Research 2006;45:
http://wwwostigov/bridge/purlcoverjsp?purl¼/15008859-
4169e77.
s6ri0N/native/; 2004.
[22] Gasafi E, Reinecke MY, Kruse A, Schebek L. Economic
[2] Azbar N, Gungormusler M, Gonen C. Continuous production
analysis of sewage sludge gasification in supercritical water
of 1,3-propanediol using raw glycerol with immobilized
for hydrogen production. Biomass and Bioenergy 2008;32:
Clostridium beijerinckii NRRL B-593 in comparison to
1085e96.
suspended culture. Bioprocess and Biosystems Engineering
[23] Li YL, Guo LJ, Zhang XM, Jin H, Lu YJ. Hydrogen production
2011;34:727e33.
from coal gasification in supercritical water with
[3] Corma A, Huber GW, Sauvanauda L, O’Connor P. Biomass to
a continuous flowing system. International Journal of
chemicals: catalytic conversion of glycerol/water mixtures into
Hydrogen Energy 2010;35:3036e45.
acrolein, reaction network. Journal of Catalysts 2008;257:163e71.
[24] Lee IG, Kim MS, Ihm SK. Gasification of glucose in
[4] Suppes GJ, Chin CW, Dasari MA, Sutterlin WR. Dehydration
supercritical water. Industrial & Engineering Chemistry
of glycerol to acetol via catalytic reactive distillation. AIChE
Research 2002;41:1182e8.
Journal 2006;52:3543e8.
[25] Sasaki M, Kabyemela B, Malaluan R, Hirose S, Takeda N,
[5] Ewan BCR, Allen RWK. A figure of merit assessment of the
Adschiri T, et al. Cellulose hydrolysis in subcritical and
routes to hydrogen. International Journal of Hydrogen Energy
supercritical water. Journal of Supercritical Fluids 1998;13:
2005;30:809e19.
261e8.
[6] Sabourin-Provost G, Hallenbeck PC. High yield conversion of
[26] Dileo GJ, Neff ME, Kim S, Savage PE. Supercritical water
a crude glycerol fraction from biodiesel production to
gasification of phenol and glycine as models for plant and
hydrogen by photofermentation. Bioresource Technology
protein biomass. Energ Fuel 2008;22:871e7.
2009;100:3513e7.
[27] Watanabe M, Sato T, Inomata H, Smith RL, Arai K, Kruse A,
[7] Seifert K, Waligorska M, Wojtowski M, Laniecki M. Hydrogen
et al. Chemical reactions of C-1 compounds in near-critical
generation from glycerol in batch fermentation process.
and supercritical water. Chemical Reviews 2004;104:5803e21.
International Journal of Hydrogen Energy 2009;34:3671e8.
[28] Antal Jr MJ, Mok WSL, Roy JC, Raissi AT, Anderson DGM.
[8] Douette AMD, Turn SQ, Wang W, Keffer VI. Experimental
Pyrolytic sources of hydrocarbons from biomass. Journal of
investigation of hydrogen production from glycerin
Analytical and Applied Pyrolysis 1985;8:291e303.
reforming. Energ Fuel 2007;21:3499e504.
[29] May A, Salvado J, Torras C, Montane D. Catalytic gasification
[9] Adhikari S, Fernando SD, Haryanto A. Hydrogen production
of glycerol in supercritical water. Chemical Engineering
from glycerin by steam reforming over nickel catalysts.
Journal 2010;160:751e9.
Renewable Energy 2008;33:1097e100.
[30] Onwudili JA, Williams PT. Hydrothermal reforming of bio-
[10] Dou BL, Rickett GL, Dupont V, Williams PT, Chen HS, Ding YL,
diesel plant waste: products distribution and
et al. Steam reforming of crude glycerol with in situ CO2
characterization. Fuel 2010;89:501e9.
sorption. Bioresource Technology 2010;101:2436e42.
[31] Vyas AP, Verma JL, Subrahmanyam N. A review on FAME
[11] Schmidt LD, Dauenhauer PJ, Salge JR. Renewable hydrogen
production processes. Fuel 2010;89:1e9.
by autothermal steam reforming of volatile carbohydrates.
[32] Thompson JC, He BB. Characterization of crude glycerol from
Journal of Catalysts 2006;244:238e47.
biodiesel production from multiple feedstocks. Applied
[12] Dumesic JA, Cortright RD, Davda RR. Hydrogen from catalytic
Engineering in Agriculture 2006;22:261e5.
reforming of biomass-derived hydrocarbons in liquid water.
[33] Sinag A, Kruse A, Rathert J. Influence of the heating rate and
Nature 2002;418:964e7.
the type of catalyst on the formation of key intermediates
[13] Byrd AJ, Pant KK, Gupta RB. Hydrogen production from
and on the generation of gases during hydropyrolysis of
glycerol by reforming in supercritical water over Ru/Al2O3
glucose in supercritical water in a batch reactor. Industrial &
catalyst. Fuel 2008;87:2956e60.
Engineering Chemistry Research 2004;43:502e8.
[14] Chakinala AG, Brilman DWF, van Swaaij WPM, Kersten SRA.
[34] Cao CQ, Guo LJ, Chen YA, Guo SM, Lu YJ. Hydrogen
Catalytic and Non-catalytic supercritical water gasification
production from supercritical water gasification of alkaline
of microalgae and glycerol. Industrial & Engineering
wheat straw pulping black liquor in continuous flow system.
Chemistry Research 2010;49:1113e22.
International Journal of Hydrogen Energy 2011;36:13528e35.
[15] Adhikari S, Fernando SD, Haryanto A. Hydrogen production
[35] Buhler W, Dinjus E, Ederer HJ, Kruse A, Mas C. Ionic reactions
from glycerol: an update. Energy Conversion and
and pyrolysis of glycerol as competing reaction pathways in
Management 2009;50:2600e4.
near- and supercritical water. Journal of Supercritical Fluids
[16] Matsumura Y, Minowa T, Potic B, Kersten SRA, Prins W, van
2002;22:37e53.
Swaaij WPM, et al. Biomass gasification in near- and super-
[36] Sakaguchi T, Ozawa M, Hamaguchi H, Nishiwaki F, Fujii E.
critical water: status and prospects. Biomass and Bioenergy
Analysis of the impact force by a transient liquid slug flowing
2005;29:269e92.
out of a horizontal pipe. Nuclear Engineering and Design
[17] Guo LJ, Lu YJ, Zhang XM, Ji CM, Guan Y, Pei AX. Hydrogen
1987;99:63e71.
production by biomass gasification in supercritical water:
[37] Chiu CW, Dasari MA, Sutterlin WR, Suppes GJ. Removal of
a systematic experimental and analytical study. Catalysts
residual catalyst from simulated biodiesel’s crude glycerol
Today 2007;129:275e86.
for glycerol hydrogenolysis to propylene glycol. Industrial &
[18] Kruse A. Supercritical water gasification. Biofuels
Engineering Chemistry Research 2006;45:791e5.
Bioproducts & Biorefining-Biofpr 2008;2:415e37.
[38] Minowa T, Zhen F, Ogi T. Cellulose decomposition in hot-
[19] Basu P, Mettanant V. Biomass gasification in supercritical
compressed water with alkali or nickel catalyst. Journal of
water e a review. International Journal Chemical Reactor
Supercritical Fluids 1998;13:253e9.
Engineering 2009;7:R3.
5568 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 5 5 5 9 e5 5 6 8

[39] Sinag A, Kruse A, Schwarzkopf V. Key compounds of the water. International Journal of Hydrogen Energy 2009;34:
hydropyrolysis of glucose in supercritical water in the 5357e64.
presence of K2CO3. Industrial & Engineering Chemistry [42] Lu YJ, Guo LJ, Zhang XM. Thermodynamic modeling and
Research 2003;42:3516e21. analysis of hydrogen production process by biomass
[40] Kruse A, Meier D, Rimbrecht P, Schacht M. Gasification of gasification in supercritical water. Abstracts of Papers of the
pyrocatechol in supercritical water in the presence of American Chemical Society 2006;231. 9-PETR.
potassium hydroxide. Industrial & Engineering Chemistry [43] Yan Q, Guo L, Lu Y. Thermodynamic analysis of hydrogen
Research 2000;39:4842e8. production from biomass gasification in supercritical
[41] Xu DH, Wang SZ, Hu X, Chen CM, Zhang QM, Gong YM. water. Energy Conversion and Management 2006;47:
Catalytic gasification of glycine and glycerol in supercritical 1515e28.