b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5

Available at www.sciencedirect.com

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

Surface texture, chemistry and adsorption properties of acid

blue 9 of hemp (Cannabis sativa L.) bast-based activated carbon
fibers prepared by phosphoric acid activation

Ru Yang a,*, Guoqiang Liu a, Xinhua Xu a, Min Li a, Jianchun Zhang b, Xinmin Hao b
a
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, 15 Beisanhuan East Road, P.O. Box 82,
Beijing, 100029, China
b
Quartermaster Institute of General Logistics Department of CPLA, Beijing, 100088, China

article info abstract

Article history:
Received 15 November 2009
Received in revised form
21 August 2010
Accepted 25 August 2010
Available online 16 September 2010
Keywords:
Hemp (Cannabis sativa L.) bast
Activated carbon fibers
Textural properties
Acid blue 9 (CAS: 2650e18e2)
Adsorption isotherm
Kinetics
Hemp (Cannabis sativa L.) bast was used to prepare activated carbon fibers by phosphoric
acid activation at 400e600
C. The pyrolysis process, textural and chemical properties for
the samples were investigated by means of TG/DTA, SEM, cryogenic N2 adsorption, FTIR
and XPS. Dye adsorption on the resultant sample was also measured. The textural prop-
erties of the activated carbon fibers were found to be strongly dependent on the activation
temperature. Activated carbon fibers exhibited narrow pore size distributions with maxima
in the micropore and small mesopore regions. BET surface area, total pore volume,
micropore volume and mesopore volume increased with the increase of activation
temperature up to 450
C and then decreased with further heating, and a sample with
maximum surface area of 1142 m2 g1 and total pore volume of 0.67 cm3 g1 was obtained.
Phosphoric acid facilitated the conservation of porous structure, led to the creation of
tremendous porosity, and resulted in various P-containing functional structures on the
surface and in the bulk phase of the resultant samples. The adsorption of acid blue 9 on the
sample could be favorably described by Langmuir isotherm, and the adsorption kinetics
was found to be well fitted by the intraparticle diffusion model.
ª 2010 Elsevier Ltd. All rights reserved.

1. Introduction Nowadays, hemp is regarded to be of important industrial and

With the fast development of society and economy, the
demand of renewable and environmental friendly energy and
raw materials sources becomes urgent due to the pollution of
environment, global warming and depletion of fossil oil
reserves. Hemp (Cannabis sativa L.) is one of the most productive
and useful plants known, and widely planted in China. It grows
quickly without any great need of pesticides in most locations
and climates with only moderate water and fertilizer require-
ments, becoming a valuable and environmentally friendly crop.
economic value as a source of paper, textiles, building mate-
rials, food, medicine, paint, detergent, varnish, oil, ink, and fuel
[1], especially in the new research field such as biodiesel
production from hempseed oil and textiles production from
hemp stems. As a result, the use of hemp generates an exten-
sive amount of solid residues, which were set on fire under
most situation and presents an environmental problem in
terms of adequate disposal.
One of the effective alternative uses of agricultural waste
biomass that has gained much attention over the last decade

* Corresponding author. Tel./fax: þ86 10 64436736.

E-mail address: yangru@bbn.cn (R. Yang).
0961-9534/$ e see front matter ª 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biombioe.2010.08.061
438 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5
is the production of activated carbons by thermo-chemical
conversion [2], and many agricultural wastes have been
proved to be promising raw materials for the production of
activated carbons [3]. Activated carbon fiber (ACFs) constitute
a particular type of activated carbon with certain specific
advantages such as extended specific surface area, high
adsorption rates and capacities, microporous structure,
special surface reactivity and ease of synthesis in form of
cloths or felts [4]. Thus, ACFs exhibit a wide applicability in
separation, purification and catalytic processes in comparison
with powdered or granular activated carbons [5].
It is widely agreed that the pore structure and pore size
distribution of an activated carbon is largely determined by the
nature of the starting material [6]. Studies also have shown
that it is easier to activate carbonaceous materials that
contains a significant proportion of fibrous structure like
cellulose and hemicellulose rather than those that has less
fibrous structure but higher percentage of lignin in nature [7],
because of a less robust and more labile nature of cellulose and
hemicellulose as compared to lignin [8]. It is known that hemp
stems consist of about 65% woody core fibers and 35% bast
fibers, and the core fibers consist of 40e48% cellulose, 18e24%
hemicellulose and 21e24% lignin, and the bast fibers are
57e77% cellulose, 9e14% hemicellulose and 5e9% lignin [9].
Obviously, the property of rich in cellulose and hemicellulose
and low content of lignin for hemp bast together with its low
price, abundant source and benefit for environment protection
make it an attractive precursor for ACFs compared with core
fibers.
Recently, Rosas et al. [10] has prepared ACFs by chemical
activation of hemp fibers with phosphoric acid, and analyzed
the influence of the activation temperature and the impreg-
nation ratio on the porous structure and surface chemistry of
ACFs, which presents a high oxidation resistance attributing
to the existence of CePO3 and CeP, acting as a physical barrier,
blocking the active carbon sites for the oxidation reaction.
However, there are still few works on the preparation of ACFs
derived from hemp bast. The main objectives of the present
work were to prepare hemp bast-based ACFs by phosphoric
acid activation under different activation temperatures, and
to examine the surface texture, chemistry and dye adsorption
properties of the final samples.
2. Materials and methods
2.1. Preparation of samples
The hemp bast was taken from the eight months of growth
hemp with a stage of maturity at harvest of 90e120 days in
Xishuangbanna Dai Autonomous Prefecture (E100
500 ,
N21
950 ), Yunnan Province in China. An anti-pulling decorti-
cator was used to isolate the bast from hemp stem, and the
hemp bast was stored in a special chamber with good venti-
lation, temperature of 18e23
C and relative humidity of
40e65%. Clean hemp bast of about 1.5e2.0 m long was folded
into about 10 cm long bundles, and each bundle was about
40 g in weight. Five bundles (ca. 200  5 g) were impregnated
with 4 L phosphoric acid aqueous solution of 2.5 mol L1
prepared by phosphoric acid (85 wt.%, A.R.). The mixture was
boiled at 110
C for 6 h and then filtrated. Then the precursor
was dried with vacuum dryer at 60
C for 24 h. The impreg-
nated hemp bast was placed in a tubular furnace and acti-
vated under continuous N2 flow (50 mL min1). The activation
temperature was reached at a heating rate of 5
C min1 and
maintained for 2 h, and then the system was cooled to room
temperature in N2 atmosphere. The resultant ACFs were
washed with hot distilled water until the pH value of the
washing solution reached 6e7, and then they were dried in an
oven at 110
C and weighted to determine the yield of the
activation process (weight of activated carbon related to
weight of raw material, in dry basis). Different activation
temperatures within 400e600
C in step of 50
C were studied,
and the ACFs were labeled as P,  represents the
activation temperature in degree Celsius.
2.2. Characterization of samples
Thermogravimetric experiments were carried out with a TG/
DTA thermal analyzer to investigate the pyrolysis process and
activation mechanism. Two samples, raw hemp bast dried at
110
C and hemp bast impregnated with H3PO4 were
measured, and the weight loss was recorded in the range
0e1100
C. About 10 mg of sample was heated from 0 to 1100
C
at a ramping rate of 10
C min1 under flowing nitrogen at
a rate of 50 mL min1 throughout the measurements.
The surface morphologies of the resultant samples were
imaged by using a scanning electron microscope Model
Cambridge S250Mk3.
Textural characteristics were determined by N2 adsorption
at 196
C on automatic apparatus (ASAP 2020, Micromeritics),
and the samples were degassed at 250
C for 12 h under vacuum
before the measurements. The specific surface areas (SBET)
were estimated by the BrunauereEmmetteTeller (BET) equa-
tion [11], and the pore size distribution (PSDs) was estimated by
the density functional theory (DFT) [12]. The total pore volume
(Vtot) was calculated by the adsorption data at P/P0 ¼ 0.995. The
micropore volume (Vmic), mesopore volume (Vmes) and macro-
pore volume (Vmac) were obtained by applying DFT method.
The chemical characterization of functional groups in
bulk phase and on surface of the samples was studied by
a Fourier transform infrared spectrometer (FTIR-2000, Per-
kineElmer) in the range 4000e400 cm1, using pellets with
samples dispersed in KBr.
The elemental contents and functional groups types on the
surface of the samples were measured below 106 Pa using
a Physical Electronics VG ESCALAB MK II XPS spectropho-
tometer with an achromatic MgKa X-ray source. The wide
scan, C1 s, O1 s and P2p3 core level spectra of the samples were
recorded at pass energy of 20 eV. The peak positions were
corrected based on the most intense graphitic peak, which
was taken as 284.6 eV. The spectra obtained were curve-fitted
with the nonlinear least-square iterative technique based on
the Gaussian function after baseline subtraction using Shir-
ley’s method [13].
2.3. Adsorption of acid blue 9
Acid blue 9 (CAS: 2650e18e2) was chosen as adsorbate, and
the sample prepared at 450
C was chosen as adsorbent. The
 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5
chemical formula of acid blue 9 is C37H34O9N2S3Na3 with
a molar weight 815 g mol1, and the molecular size is
1.88 nm  1.47 nm  1.07 nm [14]. The adsorption isotherms
were fitted by Langmuir [15] and Freundlich [16] equations,
and three kinetic models including intraparticle diffusion,
pseudo-first-order, and pseudo-second-order [17] were used
to fit the data of kinetic experiments.
3. Results and discussion
3.1. Pyrogenation process and activation mechanism
Fig. 1 shows the TG/DTG curves of raw hemp bast and H3PO4-
impregnated hemp bast. A weight loss of approximately
88.2 wt.% in 0e1100
C is observed for the raw hemp bast, and
the process can be divided into four steps: a little weight loss of
w2.8 wt.% below 200
C, due to the decrease of physically
adsorbed water [18]; a significant weight loss of 2.8e66.6 wt.%
from 200 to 520
C, corresponding to the decomposition of
hemicellulose, cellulose and lignin, for which the ranges of
initial decomposition temperature are 150e350, 275e350 and
250e500
C, respectively [19]; a very slow weight loss of
66.6e69.9 wt.% at the third step in 520e880
C, confirming the
resultant char in this temperature range is thermally more
stable; a rapid weight loss of 69.9e88.2 wt.% at the last step in
880e1100
C, being assigned to the excessive burn-off of carbon
constituents.
The TG/DTG curves of H3PO4-impregnated hemp bast pres-
ents a great weight loss of w30 wt.% at lower temperatures
<260
C, and then a continuous loss of 30e73.9 wt.% with the
temperature increasing up to 1100
C. It is noticeable that the
mass lose of H3PO4-impregnated hemp bast is higher than that
of raw hemp bast at temperature lower than 300
C, while
significantly lower at temperature beyond 300
C, which
suggests that H3PO4 not only decreases the pyrolysis tempera-
ture of hemp bast, but also leads to a higher carbon yield.
Generally, H3PO4 is believed to function as a dehydrating
catalyst, which can promote decomposition of the cellulosic
0
-20
-40
TG (%)
-60
-80
Raw hemp bast
Impregnated hemp bast
-100
0 200 400 600 800 1000
Temperature (oC)
Fig. 1 e TGeDTG curves of raw hemp bast and hemp bast
impregnated with phosphoric acid.
439
precursors at a lower heat treatment temperature [20]. More-
over, the presence of H3PO4 in the interior of the precursor
restricts the formation of tar as well as other liquids such as
acetic acid and methanol by formation of cross-links, and
inhibits the shrinkage of the precursor particle by occupying
certain substantial volumes [10], resulting in the lower weight
loss and higher yield for the H3PO4-impregnated hemp bast at
high heat treatment temperature. H3PO4 may also produce
activation through formation of phosphate and polyphosphate
bridges which connect crosslink biopolymer fragments, avoid-
ing the contraction of the material by effect of temperature, and
removal of the activating agent during the washing step will lead
to a matrix in an expanded state with an accessible pore
structure.
3.2. Surface morphology
The SEM micrographs of hemp bast-based ACFs prepared by
direct carbonization and by phosphoric acid activation at
450
C are shown in Fig. 2. It is observed that some granular
matters disperse on or between the fairly ordered fiber arrays,
and the granular matters may be attributed to the deposition
of tar, a predominant product of devolatilization in pyroge-
nation process [21]. Note that the surface of the ACF treated
with phosphoric acid is relatively planar and clean with little
tar deposition compared to that of the sample obtained by
direct carbonization, indicating that the presence of H3PO4 in
the interior of the hemp bast can restrict the formation of tar.
3.3. Surface textural characterization
Fig. 3 shows the N2 adsorptionedesorption isotherms at
196
C and the DFT pore size distributions (PSDs) of hemp
bast-based ACFs prepared by phosphoric acid activation at
different temperatures. All isotherms belong to type I according
to the IUPAC classification [11], showing a significant increase
in the adsorption at low P/P0 values (<0.1), a broad knee and
a long plateau which extends to P/P0 z 1.0, indicating the ACFs
are essentially microporous solids. The N2 uptake reaches
maximum when the activated temperature increases up to
450
C, and increasing the temperature to 500
C results in
a significant decrease of N2 uptake over the entire pressure
14 range. A further rise in the carbonization temperature to 550
C
DTG does not affect the N2 uptake obviously. Further heating to
12
DTG 600
C provokes a strong increase in the N2 uptake, and a small
10 tail is observed, indicating the existence of macropores. H2
DTG (mg/min)
hystereses were observed for as prepared ACFs, indicating the
8
existence of ink-bottle type mesopores [11]. The desorption
6 branches of ACFs prepared at 450, 500 and 550
C present
a relatively larger and more remarkable hysteresis loops at
4 intermediate pressures, suggesting that there are much more
mesopores with wider PSDs in these samples.
TG 2
Fig. 3b shows the PSDs obtained by applying DFT method to
TG 0 N2 adsorption data. The PSDs in micropore region for the ACFs
exhibit a broad similarity that each curve displays four
-2
1200 distinct peaks at ca. 0.5, 0.7, 1.3 and 1.5 nm, while the PSDs in
mesopore region show some difference that the mesopores
mainly accumulate in a narrow range around 2e3 nm for the
sample prepared at 400
C whereas in a relative wider range
around 2e6 nm for the samples prepared beyond 400
C. The
440 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5

Fig. 2 e SEM photographs of the ACFs prepared from hemp bast by direct carbonization (a, b) and phosphoric acid activation
(c, d) at 450
C.

variation expresses that, to some extent, an increase in the 400
C to 23.19% at 600
C with increasing the temperature, as
heat treatment temperature can enlarge the pore width and a consequence of a deeper dehydration of the carbonaceous
improve the proportion of mesopores. structure of the precursor. There are two sharp decreases in the
Table 1 reports the yield and various textural parameters yields, one from 400 to 500
C and the other from 550 to 600
C,
calculated from the N2 adsorption isotherms at 196
C, and whereas the slope is lower between 500 and 550
C. The first
the variation of these parameters with the activation temper- sharp weight loss corresponds to the loss of most of the volatile
ature are shown in Fig. 4. The yield decreases from 30.97% at matter as a result of intensifying dehydration and elimination

450 0.10
400 a b
Differential pore volume (cm g )
3 -1

350 0.08
Quantity adsorbed (cm g )
3 -1

300
0.06
250

200
0.04 P400
150 P400 P450
P450 P500
100 P500 0.02
P550
50 P550 P600
P600 0.00
0
0.0 0.2 0.4 0.6 0.8 1.0 1 10 100
Relative pressure (P/P0) Pore width (nm)

Fig. 3 e Nitrogen adsorption isotherms (a) and DFT pore size distributions (b) of hemp bast activated carbon fibers activated
by phosphoric acid at different activation temperatures (400e600
C). Full symbols, adsorption; open symbols, desorption.
 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5 441

Table 1 e Yield, specific surface area and pore volumes of hemp bast activated carbon fibers activated by phosphoric acid at
different activation temperatures.a
Sample code Yield (%) SBET (m2 g1) Vmic (cm3 g1) Vmes (cm3 g1) Vmac (cm3 g1) Vtot (cm3 g1) Vmic/Vtot (%)

P400 30.97 986.3 0.4240 0.1316 0.0011 0.5567 76.16

P450 28.90 1142.4 0.4755 0.1884 0.0024 0.6663 71.36
P500 26.52 1069.6 0.4555 0.1563 0.0044 0.6162 73.92
P550 25.95 1058.3 0.4521 0.1466 0.0030 0.6017 75.14
P600 23.19 1038.6 0.4350 0.1321 0.0133 0.5804 74.95

a SBET: specific surface area obtained by BET equation; Vmic, Vmes, Vmac: micropore volume, mesopore volume and macropore volume calculated
by DFT method; and Vtot: total pore volume.

reactions of cellulose, hemicellulose or lignin. The second one micropore volumes (Vmic), mesopore volumes (Vmes) and total
corresponds to the volatilization of the P2O5 coming from the pore volumes (Vtot) with activation temperature follow the
H3PO4 in excess and/or to the resultant sample combustion same trend which is observed for BET surface area, indicating
once the formerly protecting P2O5 is lost [22]. that the increase of activation temperature produces
The effect of activation temperature on the development of a progressive porosity development of the ACFs, and 450
C
surface area and pore volumes is shown in Fig. 4b. The BET would be the best temperature in the range studied. Fierro et al.
surface area increases with the increase of temperature and [22] reported a similar trend for surface development with the
reaches maximum value of 1142.4 m2 g1 at 450
C, and then activation temperature (400e650
C) in the preparation of acti-
decreases with further heating. The noticeable increase of vated carbons derived from kraft lignin by phosphoric acid
surface area with temperature up to 450
C may correspond to activation, with maximum BET surface area of 1305 m2 g1
the evolution of compounds produced from the cross-linking obtained at 600
C. It is worthy to note that the present results
reactions [23], while the decrease beyond 450
C would be are quite comparable with that of Rosas et al. [10] who reported
attributed to the degradation of the phosphate and poly- that the highest surface area for hemp bast-based ACFs was
phosphate bridges or the collapse of the resultant carbon obtained by phosphoric acid activation at 550
C, but presented
samples. Moreover, the loss of the protecting P2O5 and the higher mesopore volume and wider PSDs, suggesting that the
combustion of the carbon could also explain this decrease in the hemp bast-based ACFs in the current work are more promising
surface area. Fig. 4b also shows that the variation manners of for the application of separation and purification.

32 3.4. Chemical structure characteristics

a
30 The 4000e400 cm1 infrared spectral region of raw hemp bast
and a series of hemp bast-based ACFs prepared by phosphoric
acid activation at 400e600
C are shown in Fig. 5. All spectra
28
exhibit a wide band at 3200e3600 cm1 with a maximum at
Yield (%)

3420 cm1, which can be assigned to the OeH stretching mode

26 of hydroxyl groups and adsorbed water [24]. A weak band at
604 cm1 also appears in the spectra of each sample, may be
24 attributed to the out-of-plane deformation mode of OeH [19],
and the intensity of the OeH peaks suggests that the content
22
of OeH in raw hemp bast is much higher.
0.7 1200 The FTIR spectra of raw hemp bast shows a remarkable
b band at around 2920 cm1 due to the CeH stretching in methyl
0.6 Vtot and methylene groups [25], and a group of bands between 1460
BET surface area (m g )
2 -1

and 1240 cm1 with three main shoulder peaks at 1375, 1329
Pore volume (cm g )

1100
3 -1

0.5
Vmic and 1244 cm1, respectively. The peak at 1375 cm1 could be
ascribed to the in-plane symmetric deformation vibration of
0.4
1000 eCH3 in lignin, and the band at 1329 cm1 is related to the in-
0.3
SBET plane bending vibrations of OeH or stretching of C]O in
cellulose, and the shoulder at 1244 cm1 corresponds to the
0.2 900
Vmes asymmetric stretching of ]CeOeC connected with aryl
groups in lignin [19]. A very weak peak at 897 cm1 is also
0.1
observed, associated with the in-plane bending vibrations of
800
400 450 500 550 600 CeH or out-of-plane deformation mode of CeH and OeH in
o pyranoid rings involved in cellulose [26]. Note that all of the
Activation temperature ( C)
spectrum bands as mentioned above are mainly concerned
Fig. 4 e Influence of activation temperature on yield (a), with the structures of cellulose and lignin in hemp bast,
pore volumes and surface area (b) of hemp bast activated whereas these bands do not appear in the FTIR spectra of
carbon fibers prepared by H3PO4 activation. hemp bast after phosphoric acid activation, which reveals
442 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5

600 Table 2 e Chemical environments of surface elements for

activated carbon fiber prepared by phosphoric acid
550 activation at 450
C.
C1 s O1 s
500
Binding Chemical Binding Chemical
450 energy (ev) environment energy (ev) environment

284.4 CeC, CeH 532.3 eOH, O]CeO

400
286.3 C]O, OeCeO 533.3 O]CeO
HMP 291 eCOOH 535.4 eCOOH

4000 3500 3000 2500 2000 1500
Wavenumbers(cm )
-1
Fig. 5 e FTIR spectrum of raw hemp bast (HMP) and hemp
bast activated carbon fibers by phosphoric acid activation
at different temperatures.
that the presence of phosphoric acid can result in thermal
decomposition and destruction of organic compounds in
hemp bast as explained in thermogravimetric analysis.
Fig. 5 also shows that a band at 1648 cm1 due to the C]O
stretching in quinones [27] exists in all spectra, and the inten-
sity of the band for raw hemp bast is stronger than that of the
ACFs, indicating that the samples whether treated with phos-
phoric acid or not all contain some amount of quinones, and
the content of quinones in the ACFs is much lower when the
temperature is increased, which is attributed to the destruction
of C]O bonds by conjugating with the aromatic rings [28].
These results suggest that the crosslink reactions together with
aromatization will take place during the phosphoric acid acti-
vation process of hemp bast and easily perform at a lower
temperature. The band at 1735 cm1 could be assigned to the
stretching vibrations of C]O in ketones, carboxylic acids or
isolated carbonyl groups that are not involved in conjugation
with a double bond or aryl group [27], and the band at 1594 cm1
may be ascribed to the aromatic ring stretching vibrations (C]
C) enhanced by polar functional groups [29]. It must be noticed
that the former one only appears in the spectrum of raw hemp
bast, while the latter one only appears in the spectrum of ACFs
and its intensity decreases as the temperature is increased,
confirming that phosphoric acid can favor the crosslink reac-
tions and aromatization of carbonized matter at lower
1000 500 temperature. A similar effect of phosphoric acid on the acti-
vation of Nomex has been reported by Suárez-Garcı́a et al. [4],
who consider the reason for the decrease of aromatization at
higher temperature may be attributable to the collapse of
aromatic rings (skeletal carbon). The FTIR spectrum of the
sample prepared at 400
C exhibits three quite weak shoulder
peaks at 871, 804 and 756 cm1, respectively, all of which are
related to the out-of-plane bending vibrations of CeH in
aromatic rings with a large degree of substitution [27]. This
aromatic CeH band becomes so weak that it is even not readily
observable with increasing the activation temperature,
demonstrating a progressive substitution of CeH bonds in the
aromatic system and the formation of new CeR bonds, which
was proved by Puziy et al. [25], who prepared synthetic acti-
vated carbons by phosphoric acid activation of a styr-
eneedivinylbenzene copolymer.
All spectra show a broad band between 1160 and 1020 cm1,
which may be concerned with the CeO stretching in acids,
phenols, ethers or esters [30] for the raw hemp bast, while the
bond stretching of PþeO, PeOeP, and PeOeC resulted from the
formation of P-containing carbonaceous structures like acid
phosphates and polyphosphates [4,25] should also be taken into
account for the ACFs besides the aforesaid reason. As a conclu-
sion, phosphoric acid has a strong effect on the chemical
structures of hemp bast-based ACFs, which not only owns the
groups of OeH and aryl but also contains the groups of C]O
conjugated with a double bond or aromatic rings, CeO situated
in acids, phenols, ethers or esters, and the P-containing struc-
tures like acid phosphates and polyphosphates.

30000
100000 a Survey C1s 100000 b C1s c O1s
25000

80000 80000
20000
Counts/s

1
Counts/s
Counts/s

60000
60000 O1s 15000
CAuger
40000 2 1
40000 10000
2
20000 5000
20000
P2p3
3
0
3
0
0
1200 1000 800 600 400 200 0 298 296 294 292 290 288 286 284 282 280 544 542 540 538 536 534 532 530 528 526
Binding Energy (eV) Binding Energy (eV) Binding Energy (eV)

Fig. 6 e XPS spectra of hemp bast activated carbon fibers activated by phosphoric acid at 450
C: (a) a wide survey scan, (b)
C1 s spectrum, (c) O1 s spectrum.
 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5 443

spectrum exhibits a significantly asymmetric C1 s peak drag-

Table 3 e Langmuir and Freundlich adsorption isotherm
ging tail towards the high binding energy, demonstrating that
constants for acid blue 9 adsorbed onto the hemp bast
activated carbon fibers prepared by phosphoric acid one portion of oxygen-containing groups situated in different
activation at 450
C.a chemical environment are exposed to the sample surface
Model Constants [33]. Table 2 reveals that a series of groups such as graphitic
carbons (CeC), CeH, C]O, OeCeO, eCOOH, eOH or O]CeO
Langmuir equation may exist on the surface of the ACFs, which is partly
Qm (mg g1) 28.75 compatible with the conclusion obtained from FTIR analysis.
b (L mg1) 0.0855 Some researchers consider that the P-containing compounds
R2 0.9905 arising from treatment with phosphoric acid are tightly
RL 0.04e0.2
bound to the carbon lattice [25], indicating that the P content
Freundlich equation
is higher in the bulk than that on the surface, may be a reason
1/n 0.1664 that contribute to the difficulty to be detected by XPS on the
KF (mg g1) 11.60
sample surface.
R2 0.8346

a Adsorption conditions: adsorbent dosage, 40 mg/20 mL; 3.5. Equilibrium and kinetics adsorption of acid blue 9
temperature, 20
C.

3.5.1. Adsorption isotherms

In order to make further understanding of the chemical
characteristics of the hemp bast-based ACFs prepared by
phosphoric acid activation, XPS technique which only
analyzes the first 5e10 nm of the surface of the fibers [31] was
employed to detect the surface chemistry of the sample acti-
vated at 450
C. The scanning spectra are shown in Fig. 6, in
which part (a) shows a wide survey scan of the sample, part (b)
shows the C1 s spectrum and part (c) shows the O1 s spectrum.
Fig. 6a shows that two strong symbol peaks for C1 s and
O1 s, and a quite weak one for P2p3 can be observed at 284.6,
532 and 132.9 eV, respectively [32], suggesting the presence of
abundant C, O and minor P on the surface of the sample. The
mass concentrations of C, O and P calculated from the cor-
responding peak areas are 85.21%, 13.4% and 1.39%, respec-
tively. To reveal the local chemical state of functional groups
on the surface, the curve fittings of the raw data of the C1 s
and O1 s spectra are performed in Fig. 6b and c, respectively,
and the deconvolution results with respect to each peak
position [13] are given in Table 2. Fig. 6b shows that the
Adsorption isotherm is basically helpful to describe how
adsorbates interact with adsorbents and how the adsorption
molecules distribute between the liquid phase and the solid
phase when the adsorption process reaches an equilibrium
state, so it is critical in optimizing the use of adsorbents. In the
current work, both of the Langmuir and the Freundlich
equations were applied to isotherm data fitting for the
adsorption of acid blue 9 onto the hemp bast-based ACF
prepared at 450
C, and the related correlation coefficients (R2)
and calculated parameters are recorded in Table 3.
Table 3 shows that the adsorption isotherms of acid blue 9
onto the hemp bast-based ACF could be best fitted by the
Langmuir model (R2 ¼ 0.9905) with an adsorption capacity of
about 28.75 mg g1 other than the Freundlich model
(R2 ¼ 0.8346). As also summarized in Table 3, the value of RL is
found to be 0.04e0.2, falling in the favorable range of 0e1,
which further indicates more perfect data fitting by the
Langmuir equation, implying that the adsorption of acid blue
9 on the as prepared carbon fibers is monolayer adsorption
due to its small pore size and narrow PSD.

Table 4 e Comparison of the kinetic parameters for the adsorption of acid blue 9 onto the hemp bast activated carbon fiber
prepared by phosphoric acid activation at 450
C.a
Model Constants
1
C0 (mg L ) 50 90 120 150 200 270
qe,exp (mg g1) 16.1  0.57 23.8  0.94 25.7  1.07 26.5  0.79 26.3  1.11 27.7  0.93

Pseudoefirsteorder k1 (1 h1) 0.0182 0.0154 0.0147 0.0136 0.0091 0.0124

qe,cal (mg g1) 14.5 19.2 28.8 23.9 23.9 23.0
R2 0.966 0.964 0.969 0.955 0.959 0.972
SSE 2.42 3.99 2.43 2.98 2.29 4.78

Pseudoesecondeorder k2 (g mg1 h1) 0.0022 0.0020 0.0010 0.0006 0.0003 0.0011

qe,cal(mg g1) 18.3 22.3 30.8 31.9 32.2 30.3
R2 0.968 0.954 0.946 0.948 0.834 0.953
SSE 1.35 1.60 4.70 5.03 6.87 2.39

Intraparticle diffusion kp (mg g1 h1/2) 1.2263 1.3333 2.2709 2.2504 1.8559 1.9304
qe,cal (mg g1) 17.0 24.3 26.2 25.6 25.0 28.6
R2 0.976 0.975 0.988 0.985 0.977 0.981
SSE 0.95 1.01 0.58 0.70 1.31 0.86

a Adsorption conditions: adsorbent dosage, 40 mg/20 mL; temperature, 20
C.

444 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5
3.5.2. Kinetic models
Having examined the adsorption equilibrium state of acid blue
9 onto the hemp bast-based ACF prepared at 450
C, a further
investigation of adsorption kinetics is desirable as it provides
information about the mechanism of adsorption and potential
rate controlling steps such as mass transfer and chemical
reaction, which is essential for designing a fast and effective
model adsorption system in practice [34]. Applying the afore-
said three kinetic equations to experiment data, the calculated
parameters are listed in Table 4. It is shown that all of the
correlation coefficients (R2) are greater, above 0.975, and the
sum of error squares (SSE) are smaller, below 1.31, for the
intraparticle diffusion kinetic model compared to those for the
other two kinetic models. Besides, all calculated equilibrium
adsorption amount (qe) also agree very well with the experi-
mental data in the case of intraparticle diffusion kinetics.
Therefore, the kinetic adsorption behavior of acid blue 9 onto
the hemp bast-based ACF can be favorably described by the
intraparticle diffusion model based on the prediction that the
rate-limiting step may be physical adsorption and dye diffu-
sion inside the ACF.
4. Conclusions
Hemp bast-based ACFs prepared by phosphoric acid activation
exhibit narrow multimodal PSDs with maxima in the micro-
pore and mesopore regions. The surface area and total pore
volume reach maxima at 450
C with respective values of
1142 m2 g1 and 0.67 cm3 g1. Phosphoric acid not only facili-
tated the conservation of porous structures in the precursor
fibers, but also lead to the creation of tremendous porosity, and
resulted in various functional structures on the surface and in
the bulk phases of the resultant samples. The adsorption of
acid blue 9 on the sample could be favorably described by
Langmuir isotherm, and the adsorption kinetics was found to
be well fitted by the intraparticle diffusion model. Hemp bast is
a suitable candidate for use in the production of adsorbents for
removal of dyes, thus contributing for the implementation of
sustainable development in both the hemp production and
environmental protection.
references
[1] Batra SK. In: Lewin M, Pearce EM, editors. Handbook of fibre
chemistry. New York: Marcel Dekker Inc.; 1998.
[2] Nunes AA, Franca AS, Oliveira LS. Activated carbons from
waste biomass: an alternative use for biodiesel production
solid residues. Bioresour Technol 2009;100:1786e92.
[3] Ioannidou O, Zabaniotou A. Agricultural residues as
precursors for activated carbon productiondA review.
Renew Sustain Energy Rev 2007;11:1966e2005.
[4] Suárez-Garcı́a F, Martı́nez-Alonso A, Tascón JMD. Activated
carbon fibers from Nomex by chemical activation with
phosphoric acid. Carbon 2004;42:1419e26.
[5] Jiménez V, Sánchez P, Lucas AD, Valverde JL, Romero A.
Influence of the nature of the metal hydroxide in the porosity
development of carbon nanofibers. J Colloid Interface Sci
2009;336:226e34.
[6] Bansode RR, Losso JN, Marshall WE, Rao RM, Portier RJ.
Adsorption of volatile organic compounds by pecan shell-
and almond shell-based granular activated carbons.
Bioresour Technol 2003;90:175e84.
[7] Daud WMAW, Ali WSW. Comparison on pore development
of activated carbon produced from palm shell and coconut
shell. Bioresour Technol 2004;93:63e9.
[8] Toles CA, Marshall WE, Johns MM. Granular activated
carbons from nutshells for the uptake of metals and organic
compounds. Carbon 1997;35(9):1407e14.
[9] Gümüs‚kaya E, Usta M, Balaban M. Carbohydrate components
and crystalline structure of organosolv hemp (Cannabis
sativa L.) bast fibers pulp. Bioresour Technol 2007;98:491e7.
[10] Rosas JM, Bedia J, Rodrı́guez-Mirasol J, Cordero T. HEMP-
derived activated carbon fibers by chemical activation with
phosphoric acid. Fuel 2009;88:19e26.
[11] Sing KSW, Everett DH, Haul RAW. Reporting physisorption
data for gas/solid systems with special reference to the
determination of surface area and porosity. Pure Appl Chem
1985;57:603e19.
[12] Ryu ZY, Zheng JT, Wang MZ, Zhang BJ. Characterization of
pore size distributions on carbonaceous adsorbents by DFT.
Carbon 1999;37:1257e64.
[13] Su FB, Zhao XS, Lv L, Zhou ZC. Synthesis and
characterization of microporous carbons templated by
ammonium-form zeolite Y. Carbon 2004;42:2821e31.
[14] Tamai H, Yoshida T, Sasaki M, Yasuda H. Dye adsorption on
mesoporous activated carbon fiber obtained from pitch
containing yttrium complex. Carbon 1999;37:983e9.
[15] Deng H, Yang L, Tao GH, Dai JL. Preparation and
characterization of activated carbon from cotton stalk by
microwave assisted chemical activationdapplication in
methylene blue adsorption from aqueous solution. J Hazard
Mater 2009;166:151421.
[16] McKay G, Blair HS, Gardner JR. Adsorption of dyes on
chitin. I. Equilibrium studies. J Appl Polym Sci 1982;8:
3043e57.
[17] Gücek A, S‚ener S, Bilgen S, Mazmancı MA. Adsorption and
kinetic studies of cationic and anionic dyes on pyrophyllite
from aqueous solutions. J Colloid Interface Sci 2005;286:
53e60.
[18] Zhang ZD, Yan XX, Tian BZ, Yu CZ, Tu B, Zhu ZS, et al.
Synthesis of ordered small pore mesoporous silicates with
tailorable pore structures and sizes by polyoxyethylene
alkyl amine surfactant. Micropor Mesopor Mater 2006;90:
23e31.
[19] Shafizadeh F. Introduction to pyrolysis of biomass. J Anal
Appl Pyrolysis 1982;3(4):283e305.
[20] Jagtoyen M, Derbyshire F. Some considerations of the origins
of porosity in carbons from chemically activated wood.
Carbon 1993;31:1185e92.
[21] Hu ZH, Srinivasan MP, Ni YM. Novel activation process for
preparing highly microporous and mesoporous activated
carbons. Carbon 2001;39:877e86.
[22] Fierro V, Torné-Fernández V, Celzard A. Kraft lignin as
a precursor for microporous activated carbons prepared by
impregnation with ortho-phosphoric acid: synthesis and
textural characterization. Micropor Mesopor Mater 2006;92:
243e50.
[23] Ruthven DM. Principles of adsorption and adsorption
processes. New York: Wiley; 1984. p. 55e58.
[24] Guo YP, Rockstraw DA. Physical and chemical properties of
carbons synthesized from xylan, cellulose, and Kraft lignin
by H3PO4 activation. Carbon 2006;44:1464e75.
[25] Puziy AM, Poddubnaya OI, Martı́nez-Alonso A, Suárez-
Garcı́a F, Tascón JMD. Synthetic carbons activated with
phosphoric acid I. Surface chemistry and ion binding
properties. Carbon 2002;40:1493e505.
 b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 4 3 7 e4 4 5
[26] del Rı́o JC, Gutiérrez A, Rodrı́guez IM, Ibarra D, Martı́nez Á T.
Composition of non-woody plant lignins and cinnamic acids
by Py-GC/MS, Py/TMAH and FT-IR. J Anal Appl Pyrolysis 2007;
79:39e46.
[27] Guo J, Lua AC. Characterization of adsorbent prepared from
oil-palm shell by CO2 activation for removal of gaseous
pollutants. Mater Lett 2002;55:334e9.
[28] Olivares-Marı́n M, Fernández-González C, Macı́as-Garcı́a A,
Gómez-Serrano V. Thermal behaviour of lignocellulosic
material in the presence of phosphoric acid. Influence of the
acid content in the initial solution. Letters to the Editor.
Carbon 2006;44:2330e56.
[29] Wang ZM, Kanoh H, Kaneko K, Lu GQ, Do D. Structural and
surface property changes of macadamia nut-shell char upon
activation and high temperature treatment. Carbon 2002;40:
1231e9.
[30] Olivares-Marı́n M, Fernández-González C, Macı́as-Garcı́a A,
Gómez-Serrano V. Preparation of activated carbons from
445
cherry stones by activation with potassium hydroxide. Appl
Surf Sci 2006;252:5980e3.
[31] Mangun CL, Benak KR, Economy J, Foster KL. Surface
chemistry, pore sizes and adsorption properties of activated
carbon fibers and precursors treated with ammonia. Carbon
2001;39:1809e20.
[32] Mirji SA, Halligudi SB, Mathew N, Jacob NE, Patil KR,
Gaikwad AB. Adsorption of methanol on mesoporous SBA-
15. Mater Lett 2006;61:88e92.
[33] Al-Khatib MF, Iyuke SE, Mohamad AB, Daud WRW,
Kadhum AAH, Shariff AM, et al. The effect of impregnation
of activated carbon with SnCl2.2H2O on its porosity, surface
composition and CO gas adsorption. Carbon 2002;40:
1929e36.
[34] Senthilkumaar S, Varadarajan PR, Porkodi K,
Subbhuraam CV. Adsorption of methylene blue onto jute
fiber carbon: kinetics and equilibrium studies. J Colloid
Interface Sci 2005;284:78e82.