- Email: [email protected]
Sustainable porous carbons from lignocellulosic wastes obtained from the extraction of tannins
Microporous and Mesoporous Materials xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Sustainable porous carbons from lignocellulosic wastes obtained from the extraction of tannins B. Ruiz a, E. Ruisánchez a, R.R. Gil a, N. Ferrera-Lorenzo a, M.S. Lozano b, E. Fuente a,⇑ a b
Instituto Nacional del Carbón (INCAR), CSIC, Francisco Pintado Fe, 26, 33011 Oviedo, Spain Asociación de Investigación de las Industrias del Curtido y Anexas (AIICA), Igualada, Barcelona, Spain
a r t i c l e
i n f o
Article history: Received 28 May 2014 Accepted 1 September 2014 Available online xxxx Keywords: Lignocellulosic wastes Tannins Activated carbon Pyrolysis Chemical activation
a b s t r a c t The present research study explores the possibility of obtaining high surface area activated carbons (ACs) from lignocellulosic wastes from the extraction of tannins. The use of vegetable tannins in the leather industry has the serious drawback that it involves the mass destruction of trees. Currently, studies are being conducted to obtain tannins from different lignocellulosic wastes. Two lignocellulosic wastes from the extraction of tannins, defatted grape seeds and acacia seed shells, with high carbon and nitrogen contents and a low ash content were obtained and investigated as a potential precursor for the preparation of activated carbons. KOH chemical activation, with a previous pyrolysis step, was performed in a conventional electric furnace varying the experimental conditions of KOH/precursor weight ratio, final activation temperature, and inert flow gas. After activation the samples were washed with a solution of HCl and water or just with hot water. The ACs obtained were essentially microporous with a specific surface area up to 2000 m2 g and presented low ash content, less than to 0.10% in the case of adsorbent materials from the defatted grape seeds and up to 3.60% for the materials from the acacia seed shells. The best results were obtained with the largest KOH/precursor weight ratio or the highest activation temperature (900 °C). Their moderate nitrogen content (up to 1.5%) makes them especially suitable materials for CO2 capture. Some of them were more effective for CO2 adsorption than the commercial activated carbon F400 and therefore represent an attractive alternative to more expensive adsorbent materials. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The tanning industry uses animal hide, a putrescible material, and converts it into a stable material, leather. If the tanning industry did not exist animal skins, a by-product of the meat and dairy industries, would have to be disposed of by other means such as by land filling or incineration. The tanning industry mainly uses one of two methods for transforming skin into leather: mineral or vegetable tanning. Mineral tanning which uses chromium salts and acids is most modern and the most used method but it is also the most polluting. Vegetable tanning uses vegetable tanning substances, called tannins. Tannins are very numerous and are distributed abundantly in the nature. The disadvantage of this type of tanning is deforestation. Currently, studies are being conducted to obtain tannins from different lignocellulosic wastes. Thus, AIICA has developed a methodology for extracting them from defatted grape seeds under solvent extraction in basic conditions [1] with very good results. Grape seeds, which are a solid waste from the wine industry, are ⇑ Corresponding author. Tel.: +34 985 119090; fax: +34 985 297662. E-mail address: [email protected] (E. Fuente).
mostly burnt as fuel and to some extent used for cattle feed, despite the fact that they are also a source of oil for human consumption [2,3] or tannin vegetables which could be used in the tanning industry [1]. The possibility of obtaining tannins from acacia seed shells by means of a grinding and sieving operation has also been investigated. The fruits of Acacia (Faidherbia albida) are composed of two parts: the seeds and the seed shell. The seed serves as food for animals and the shell seeds after undergoing a process of grinding and sieving provide the vegetable tannin concentrates in the fine powder fraction. This fraction is used as tanning agent by the tanners from northern Africa. After the process of obtaining tannins from defatted grape seeds and acacia seed shells a lignocellulosic residue with a high carbon and low ash content are obtained. The leather industry has often been associated with a high pollution of the water and air and to generating a large amount of solid wastes. Wastewater is loaded of sulfides, chlorides, sulfates, phosphates, settleable solids, color. . . and chromium in the case of the mineral tanning. The main parameters affecting the air quality in the tanneries are volatile organic compounds (VOCs), H2S, NH3 and dust. Furthermore, pollutants such as nitrogen oxides
http://dx.doi.org/10.1016/j.micromeso.2014.09.004 1387-1811/Ó 2014 Elsevier Inc. All rights reserved.
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
2
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
(NOx), sulfur oxides (SOx), and carbon dioxide (CO2) are generated in the thermal systems to generate heat. The adsorption is an effective abatement technology for contaminant compounds. The most important consideration is the selection of an appropriate adsorbent. The adsorbents most commonly used for contaminant compounds removal are the activated carbons because they have a large surface area, high porosity and great adsorption capabilities. Activated carbons (ACs) can be obtained from almost any material with high carbon content. The precursors generally used for the production of ACs are coal, lignite and wood, among others. In order to reduce the production costs of these adsorbents, many research studies have been undertaken to find alternatives to the raw materials used in activated carbon production, such as agricultural and industrial wastes [47]. Furthermore, an evaluation of agricultural residues or by-products as precursors of activated carbon seems to be very promising from a sustainable viewpoint. In this work, the precursors of the ACs will be two agricultural wastes obtained from defatted grape seeds and acacia seed shells after that these residues have been subjected to a process of solvent extraction and grinding–sieving, respectively, for the separation of the tannins. The use of grape seeds in the preparation of ACs has been reported in only a few instances in the scientific literature [8–10] and almost none from acacia seed shells [11]. However, there is no background literature information related to the use of the lignocellulosic wastes reported in this work. The aim of the present research work is to evaluate the possibility of obtaining sustainable porous carbons from lignocellulosic wastes from the extraction of tannins by means of KOH chemical activation using a previous pyrolysis step. The influence of the activating agent/precursor weight ratio, activation temperature, inert flow gas and different washing processes of the final materials is studied in order to achieve a suitable development of porosity in the activated carbons. 2. Materials and methods 2.1. Raw materials This study was carried out in order to valorise the lignocellulosic wastes from the extraction of tannins. Two representative samples of these wastes, residual materials left behind after the extraction of tannins from defatted grape seeds and acacia seed shells, were supplied by ‘‘Asociación de Investigación de las Industrias del Curtido y Anexas’’, AIICA. The extraction of tannins from defatted grape seeds was performed in an aqueous medium using sodium metabisulfite as a solubilizer of tannins, and applying pressure and temperature [1]. After the extraction process a solid lignocellulosic residue, GS, is left behind. This will serve as one of the precursor materials of the activated carbons. The method of extracting tannin from Acacia Albida fruits was as follows: the acacia fruits (seeds and seed shells) were crushed and then passed through an air flow where the seeds and the seed shells were separated due to their different densities. After this, the seed shells were crushed by a cutter mill and were separated into two fractions by sieving: a coarse fraction and a fine powder. The fine powder contained the highest concentration of tannins (47.9%) while the other fraction contained less than half (21.4%). The coarse fraction, AP, was selected as precursor material for the activated carbons. In Appendix A, Supplementary Data, the Fig. S1 shows the leitmotiv of this research study and the Figs. S2 and S3 show an image illustrating the equipment for the extraction of tannins from defatted grape seeds and the scheme of the tannins separation from the acacia seed shells, respectively.
2.2. Activated carbons The activated carbons from these lignocellulosic wastes from the extraction of tannins were obtained in a two-stage thermochemical process which involved carbonizing the raw material followed by chemical activation of the char with an alkali agent (KOH). The thermochemical processes, pyrolysis and thermochemical activation, were carried out in an alumina crucible placed in a conventional tubular furnace Carbolite CTF 12/65/550. The pyrolysis step of the raw material (GS, AP) was performed under a N2 flow gas of 150 ml/min 1 at a heating rate of 5 °C/min up to a temperature of 750 °C for 60 min. In order to establish the most suitable pyrolysis conditions, several parameters were studied in a thermo-balance, Q5000IR: heating rate (5, 10 and 15 °C/min), final temperature (900 and 1000 °C), maintenance time of final temperature (60 and 120 min), under a flow nitrogen rate of 100 ml min 1 in all the experiments. In the chemical activation, the activating agent (KOH) and the pyrolyzed sample (GSP or APP) were mixed in solid state (physical mixture) in different weight ratios (1:1. and 2:1). The experimental conditions of the chemical activation were: a heating rate of 5 °C/min, a final activation temperature of from 700 to 900 °C, the final temperature being held for 60 min, and different N2 gas flows of 150 and 500 ml min 1. After chemical activation, in order to remove the activation products and the mineral matter blocking the porosity, the adsorbent materials were washed with a 5M hydrochloric acid solution and subjected to a series of deionised water (Milli-Q) rinses. Some activated materials were washed only with hot water at 100 °C and without using hydrochloric acid. Finally the samples were dried at 105 °C. The nomenclature for the activated carbons (ACs) prepared is summarized in Table 1. In order to conduct a comparative study of the CO2 adsorption capacity of the obtained activated carbons a commercial activated carbon (Filtrasorb F400) was selected since it had been used in previous CO2-sorption studies [12–15]. The commercial activated carbon ‘‘Maxsorb3000’’ was also used as reference since it is a material that is widely used in gas storage [16–19] and has very good textural properties [20].
Table 1 Nomenclature of the activated carbons. Sample
KOH/precursor ratio
Activation temperature (°C)
N2 flow gas (ml min 1)
GSP-1-700-f GSP-1-750-f GSP-1-750-F GSP-2-750-f GSP-1-800-f GSP-1-900-f GSP-1-900-F GSP-1-900-fW
1:1 1:1 1:1 2:1 1:1 1:1 1:1 1:1
700 750 750 750 800 900 900 900
150 150 500 150 150 150 500 150
APP-1-700-f APP-1-750-f APP-1-750-F APP-2-750-f APP-1-800-f APP-1-900-f APP-1-900-fW
1:1 1:1 1:1 2:1 1:1 1:1 1:1
700 750 750 750 800 900 900
150 150 500 150 150 150 150
GSP (defatted grape seeds carbonized) and APP (acacia seed shells carbonized). 1 or 2: activating agent/precursor weight ratio, 1:1 or 2:1, respectively. 700–750–800–900: activation temperature (°C). f or F: inert gas flow in the chemical activation process, 150 ml min 1 or 500 ml min 1, respectively. W: samples washed with hot water; samples washed with HCl and water have not annotation.
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
3
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
2.3. Characterization of the materials The moisture and ash contents of the sample were determined following the UNE 32002 and 32004 norms, respectively. The ultimate analysis (carbon, hydrogen, nitrogen and sulfur content) was carried out on automatic instruments (LECO CHN-2000 and S-144DR). Atomic absorption spectrophotometry was used to determine the inorganic composition. The ashes were dissolved with inorganic acids and analysed on Atomic Absorption Spectrophotometer Shimazu AA-6300 equipment. The morphology of the samples was examined using a Scanning Electron Microscope equipped with an Energy-Dispersive X-ray analysing system (SEM/EDX). Textural characterization was performed by measuring the N2 adsorption isotherms at 196 °C in a Micromeritics ASAP 2420 automatic apparatus. The isotherms were used to calculate the BET specific surface area, SBET, and total pore volume, VTOT, at a relative pressure of 0.95. The pore size distributions, microporosity and mesoporosity, were obtained by applying the density functional theory (DFT) model to the N2 adsorption data, assuming slit-shaped pore geometry [21]. Before these adsorption experiments, the samples were out gassed under vacuum at 120 °C overnight to remove any adsorbed moisture and/or gases. To measure the capacity of the sorbents to capture CO2, H2 and CH4, high pressure adsorption isotherms were obtained using a Rubotherm-VTI magnetic suspension balance at ambient temperature and under static conditions. CO2, H2 and CH4 adsorption isotherms were determined up to 3 MPa. Prior to the measurements the samples were outgassed at 120 °C under a high vacuum for 240 min. 3. Results and discussion
Table 2 Chemical characterization of the raw materials, carbonized materials and activated carbons. Sample
Mass fraction (%), db Ash
C
H
N
S
2.07 5.53
53.32 50.57
5.68 4.77
1.05 1.45
0.23 0.26
5.04 12.60
91.14 82.86
1.2 1.14
1.8 2.08
0.1 0.07
GSP-1-700-f GSP-1-750-f GSP-1-750-F GSP-2-750-f GSP-1-800-f GSP-1-900-f
0.33 0.14 0.27 0.05 0.35 0.92
93.26 92.60 95.76 92.61 96.24 92.16
0.69 0.40 0.35 0.50 0.26 0.15
1.11 1.24 0.86 1.02 0.96 0.77
0.07 0.07 0.06 0.09 0.09 0.09
APP-1-700-f APP-1-750-f APP-1-750-F APP-2-750-f APP-1-800-f APP-1-900-f
2.81 3.60 3.22 0.86 2.26 3.43
92.91 91.90 92.14 94.93 93.63 91.84
0.59 0.50 0.46 – 0.26 0.24
1.51 1.25 1.23 0.92 1.17 0.99
– – 0.10 – 0.11 0.12
91.00 93.13
0.30 0.14
1.00 0.25
0.70 –
GS AP GSP APP
F400 MAXSORB
the Fig. 1. It shows that in the case of grape seeds CaO is the principal oxide with a 36.7% presence while in the case of the acacia seed shells SiO2 (38.0%) is the main oxide, due to the parts of the plant chosen for this study, since acacia is the seed shell and the Si is part of its cell structure. However, in the grape seed there is more Ca which contributes to the growth of seeds. The second most abundant compound in both cases is K2O which is the nutrient that most plants absorb after nitrogen in the form of cation K+. Finally, defatted grape seeds contains SiO2 (13%) and the acacia seed shells CaO (15.8%). The rest of the oxides are minor components in both wastes.
3.1. Chemical characterization of the materials The results of the proximate and ultimate analyses of the raw materials (GS and AP), their corresponding carbonized materials (GSP, APP), the adsorbents material obtained and the commercial activated carbons are shown in Table 2. The chemical analysis of both raw materials (GS, AP) showed a carbon content higher than 50% but a low ash and sulfur content. This high carbon and low ash content make the materials good precursors for obtaining activated carbons. In this work, a previous pyrolysis step was used before KOH chemical activation in order to study the possible energetic use of the different fractions obtained (char, fuel, gas) using a procedure similar to that used for other types of industrial waste of leather and macroalgae [22–24]. Once these raw materials were pyrolyzed the carbon content of the chars (GSP, APP) increased significantly, especially in the case of grape seeds, up to 91.14%, and for the acacia seed shells up to 82.86%. Although the ash content doubled in both cases, from 5.53% to 12.60% in the case of acacia seed shells and from 2.07% to 5.34% in the case of the grape seeds, the ash content remained low in these pyrolyzed materials. The chemical composition values of the GSP are similar to those obtained by Jimenez-Cordero et al. [25] in his study of granular chars from the flash and conventional pyrolysis of grape seeds subjected to n-hexane extraction for 24 h in an apparatus to remove oil. As can be seen from Table 2, the adsorbent materials obtained from GSP and APP show a high carbon content (91.84–96.24%), a low ash content (<1% and up to 3.60% for the activated carbons from defatted grape seeds and acacia seed shells, respectively) and a low sulfur content (up to 0.12%). The ash composition of the raw materials (GS, AP) which was studied by atomic absorption spectrophotometry is presented in
3.2. Thermal degradation of the raw materials (GS, AP) The thermal degradation of the lignocellulosic wastes (GS, AP) was studied in a thermo-balance, Q5000IR, in a nitrogen atmosphere. The behavior of these materials is shown in the form of thermograms in Fig. 2. Three heating rates at 5, 10 and 15 °C/min were used, to simulate the thermal shock undergone by the wastes when they were subjected to the working temperature (750 °C), The pyrolysis of the raw materials (GS, AP) revealed an initial slight weight loss at about 100 °C, due to the elimination of water. In both cases the weight loss started at about 200–250 °C and ended at about 550 °C. 40 35 30
GS
AP
25 % 20 15 10 5 0 CaO
SiO2
Al2O3 Fe2O3
MgO
Na2O
K2O
Fig. 1. The ash composition of the lignocellulosic wastes from the extraction of tannins (GS, AP).
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
4
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
This loss resulted from the decomposition of the major constituents of the lignocellulosic materials; hemicellulose decomposing mainly between 150 and 350 °C, cellulose between 275 and 350 °C, and lignin gradually decomposing between 250 and 500 °C [26,27]. As the pyrolysis curves of the wastes show, the percentage of final weight loss depends only minimally on the heating rate, the differences in the final residue being of the order of 3% in the case of acacia and 1.5% for the grape. In light of these data and decomposition temperatures, as mentioned previously, the pyrolysis temperature and heating rate were adjusted to 750 °C (to allow minimum weight deviation and ramp weight loss stabilized) and 5 °C/min. (fast enough to cause degradation processes), respectively. Once the optimal conditions of pyrolysis were established, as a function of the results obtained in the thermo-balance, the raw materials (GS, AP) were pyrolyzed at a heating rate of 5 °C/min, a pyrolysis temperature of 750 °C for 60 min. After the pyrolysis step, the carbonized materials or chars obtained (GSP, APP) were chemically activated with KOH in different experimental conditions to yield several activated carbons, Table 1. 3.3. SEM microscopy SEM images of the raw materials (GS, AP) and the adsorbent materials obtained in this work are shown in Fig. 3. It can be seen that the raw materials have a characteristic lignocellulosic structure. In the pyrolysis process the raw structure was modified with the loss of its cellular structure. The external surfaces of the activated carbons show a lot cavities and are 0,6 Defatted Grape seeds 5ºC/min Defatted Grape Seeds 10ºC/min Defatted Grape Seeds 15ºC/min
Weight (%)
100
0,5
80
0,4
60
0,3
40
0,2
20
0,1
0
Deriv.Weight (%/ºC)
120
0 50
150
250
350 450 550 650 Temperature (ºC)
750
850
120
0,4 Acacia seed shells 5ºC/min Acacia seed shells 10ºC/min Acacia seed shells 15ºC/min
100
Weight (%)
60
0,2
40
Deriv. Weight (%)
0,3 80
0,1 20 0
0,0 50
150
250
350
450
550
650
750
850
950
Temperature (ºC) Fig. 2. Pyrolysis of defatted grape seeds (GS) and acacia seed shells (AP) at different heating rates (5, 10, 15 °C/min).
very irregular, indicating generation of porosity in the adsorbent materials produced after the attack of the alkaline reagent (KOH) during activation. 3.4. Textural characterization of the materials In the KOH chemical activation process different parameters were considered for obtaining activated carbons: the activating agent/precursor weight ratio, the nitrogen flow, the activation temperature and finally, when the materials were obtained, the different solutions to use in the washing process. The results obtained in the textural characterization of the activated carbons are shown in Fig. 4 (nitrogen adsorption isotherms at 196 °C) and in Table 3 where, the data corresponding to the equivalent specific surface area -SBET-, total pore volume at p/p0 = 0.95 -VTOTand the distribution of porosity obtained by applying the density functional theory (DFT) model to the N2 adsorption data, are presented. Fig. 4 shows all the N2 adsorption isotherms at 196 °C for the materials derived from the defatted grape seeds. As can be seen the raw material does not show any nitrogen adsorption. Therefore the equivalent specific surface areas (BET) are negligible for this sample, <1 m2 g 1 (Table 3). The adsorption of N2 by the activated carbons obtained from this residue takes place, fundamentally, at low relative pressures which is typical of microporous solids [28,29]. Only the samples obtained in more severe conditions of activation show a small hysteresis loop, which is associated to capillary condensation inside the mesopores, but the hysteresis is very small so the materials are mainly microporous. In general the isotherms are of type I, according to the BDDT classification and type I–IV hybrid for the materials obtained in several activation conditions [30]. The capacity of the nitrogen adsorption of the activated carbon increases considerably with the increase in the KOH/precursor weight ratio which was found to be the most important parameter in the chemical activation process (Fig. 4) [31,32]. As can be seen in this figure an increase in the activating agent produces a widening of the knee of the isotherms, indicating a change in the pore size distribution. The quantity of N2 adsorbed was found to be greater in the adsorbent which had been subjected to the greatest amount of activating agent (KOH:sample ratio = 2:1). The activated carbon also has the highest equivalent specific surface area (SBET = 1860 m2 g 1) and total pore volume (0.772 cm3 g 1), Table 3. Okman et al. produced activated carbons from grape seeds by chemical activation with potassium hydroxide at different activation temperature (600 and 800 °C) using activating agent/precursor ratios of 0.25:1, 0.50:1 and 1:1; the mixture of chemical and precursor was in a state of dissolution; in these conditions an increase in the weight ratio did not appear to promote an increase in the BET specific surface [10]. The other experimental variable that has a significant influence on the adsorption capacity of the activated carbons obtained is the activation temperature. An increase in this variable led to an increase in the nitrogen adsorption capacity of the adsorbents obtained (Fig. 4), in the specific surface area (1038 and 1604 m2 g 1 for the activated carbons obtained at 700 °C and 900 °C, respectively) and also in the total pore volume (0.415 and 0.672 cm3 g 1 for the adsorbents obtained at 700 °C and 900 °C, respectively), Table 3. It has also been reported that increases in activation temperatures produce adsorbent materials with higher surface areas [10,33]. The effect of the flow inert gas (150 and 500 ml min 1) used on the KOH chemical activation was studied at 750 °C and 900 °C. This variable caused few changes to the materials obtained. There was a slight increase in the textural development of the activated carbons with the increase in nitrogen flow (Fig. 4 and Table 3). It
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
5
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
(a)
(b)
(a*)
(c)
(d)
(d*)
Fig. 3. SEM photographs: (a) an acacia seed shell, (a⁄) detail of acacia seed shell, (b) activated carbon from acacia seed shells (900 °C, 150 ml min (d) activated carbon from defatted grape seeds (900 °C, 150 ml min 1), (d⁄) detail of activated carbon.
1200 MAXSORB GSP-2-750-f GSP-1-900-F GSP-1-900-f F400 GSP-1-800-f GSP-1-750-F GSP-1-750-f GSP-1-700-f GS
Vads (cm3g-1, STP)
1000 800 600 400 200 0 0
0,2
0,4
p/po
0,6
0,8
1
Fig. 4. N2 adsorption isotherms at 196 °C for the raw material and the activated carbons obtained from the defatted grape seeds and the commercial activated carbons.
has been reported that the flow gas has an important effect on the porosity development of activated carbons obtained from coal by KOH chemical activation. A higher nitrogen flow rate during the carbonization process produces activated carbons with a much higher micropore volume [31]. The results for the activated carbons obtained from the two residues (defatted grape seeds and acacia seed shells) are very similar, so conclusions can be applied to both residues. It should be noted
1
), (c) defatted grape seeds,
that the activated carbons obtained from the acacia seed shells exhibit a nitrogen adsorption capacity slightly higher than the corresponding materials obtained from defatted grapes seeds. Also the BET specific surface area and total pore volume values are slightly higher. The adsorbent material with the higher SBET (1988 m2 g 1) and total pore volume (0.809 cm3 g 1) was obtained by KOH chemical activation from acacia seeds shell at a KOH/precursor weight ratio of 2:1 at 750 °C and under an inert gas flow rate of 150 ml min 1. Another experimental variable object studied was the effect of the washing stage. The washing stage is the last step in the preparation of a chemically activated carbon. This washing serves to remove activation products and mineral matter blocking the porosity of the adsorbent materials obtained. In order to see the effect of the washing step, activated carbons were prepared from both residues (defatted grape seeds and acacia seed shells) by KOH chemical activation at a weight ratio of 1:1, at 900 °C and under a nitrogen flow of 150 ml min 1. The materials thus obtained were washed with 5M hydrochloric acid solution and subjected to a series of deionised water rinses (GSP-1-900-f and APP-1-900-f) or with hot water (GSP-1-900-fW and APP-1-900-fW). In Appendix A, Supplementary Data, the Fig. S4 shows the N2 adsorption isotherms corresponding to the four activated carbons. The samples washed with hydrochloric acid and water (GSP-1900-f and APP-1-900-f) show a somewhat higher N2 adsorption than the adsorbent materials washed only with hot water (GSP-1-900-fW and APP-1-900-fW) which is indicative of a more efficient removal of potassium compounds by washing with HCl. Moreover, they also present a higher specific surface area (1604 and 1684 m2 g 1 for GSP-1-900-f and APP-1-900-f, respectively) and total pore volume (0.672 and 0.716 cm3 g 1 for GSP-1-900-f
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
– –
15.2 24.1 25.5 54.2 29.1 48.3 48.1 47.6
54.7 52.3 45.2
45.3 49.7
– –
82.7 75.1 72.9 42.4 69.0 44.4 42.5 45.4
43.6 42.4 49.6
32.2 5.0
22.5 45.3
1.7 5.3 5.2
Vmmi (%) (0.7–2 nm)
– –
and APP-1-900-f, respectively), Table 3. Similar results have been reported for KOH chemical activated carbons from coal by Lozano-Castelló [31]. Table 3 and Fig. 5 show the pore-size distributions by DFT of the adsorbent materials obtained from the defatted grape seeds and the commercial activated carbons (F400, MAXSORB). It can be observed from Table 3 that all of the adsorbents materials obtained in this work are microporous materials and in all the cases the micropore volume is higher than 90%. It is noteworthy than the adsorbent materials obtained at an activation temperature between 700 and 800 °C are ultramicroporous materials with micropore volume higher than 98% and the contribution of ultramicropore volume being between from 82.7% to 69.0%. It can be seen that a higher activation temperature (900 °C) or high activating agent/precursor weight ratio (2:1) favors the widening of pores with the activated carbons obtained showing a greater contribution of wider micropores (up to 54.2%) and mesopores (up to 9.3%). The commercial activated carbon MAXSORB has an important mesopore volume (45%).
3.5. CO2/CH4/H2 adsorption capacity of the adsorbent materials
) (2–50 nm) – –
0.007 0.003 0.006 0.021 0.008 0.041 0.054 0.036
0.011 0.030 0.028
0.114 0.645
– –
0.052 0.092 0.098 0.337 0.122 0.272 0.275 0.244
0.359 0.298 0.242
0.229 0.708 0.163 0.071
0.286 0.241 0.266
0.282 0.287 0.280 0.264 0.289 0.250 0.243 0.233
– –
c
0.641 1.681
0.809 0.716 0.641
0.415 0.458 0.469 0.772 0.510 0.672 0.706 0.632
– –
1271 3420 F400 MAXSORB
1988 1684 1547 APP-2-750-f APP-1-900-f APP-1-900-fW
1038 1169 1176 1860 1275 1604 1613 1496 GSP-1-700-f GSP-1-750-f GSP-1-750-F GSP-2-750-f GSP-1-800-f GSP-1-900-f GSP-1-900-F GSP-1-900-fW
<1 <1 GS AP
Vumi: Volume ultramicroporous (<0.7 nm). Vmmi: Volume medium-microporous (0.7–2 nm). Vme: Volume mesoporous (2–50 nm). a
b
Vumi (cm3 g a
) p/p0 = 0.95
1
)
VTOT (cm3 g
0,14
MAXSORB GSP-2-750-f GSP-1-900-F GSP-1-900-f F400 GSP-1-800-f GSP-1-750-F GSP-1-750-f GSP-1-700-f
0,12
1
SBET (m2 g Samples
Table 3 Textural characterization of the raw materials and activated carbons.
1
) (<0.7 nm)
b
Vmmi (cm3 g
1
) (0.7–2 nm)
c
Vme (cm3 g
1
Fig. 6 shows the CO2, H2 and CH4 adsorption isotherms at high pressure for two selected activated carbons obtained in this work from the lignocellulosic waste (GSP-1-900-f and APP-1-750-f) together with the commercial activated carbons, F400 and MAXSORB; Table 4 shows the adsorption capacities of CO2, H2 and CH4 in terms of the mass fraction (%) of these materials at air pressure (P = 0.1 MPa) and the maximum pressure studied (P = 3 MPa). The materials tested exhibit a high capacity for retaining CO2, moderate CH4 adsorption capacity and a very low H2 adsorption capacity, Table 4 and Fig. 6. This clearly evidences the selectivity of the adsorbents to CO2 so they could be used to separate gases in order to obtain pure hydrogen or to separate CO2/H2 and CH4/H2. It is well known that the adsorption capacity of an activated carbon is mainly governed by its texture and its surface chemistry. Maxsorb presents the highest capacity for adsorbing CO2 due to the high textural development of this commercial activated carbon (>3000 m2 g 1) followed by the adsorbent material obtained from defatted grape seeds at high temperature (900 °C), GSP-1-900-f. This adsorbent obtained from the lignocellulosic waste displays a CO2 adsorption capacity higher than the commercial activated carbon F400 which was selected as reference since it
0,1 Inv Vol (cm3g-1)
a
b
Vumi (%) (<0.7 nm)
c
2.1 0.8 1.6 3.4 1.9 7.3 9.4 7.0
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
Vme (%) (2–50 nm)
6
0,08 0,06 0,04 0,02 0 0,1
1
10
Pore diameter (nm) Fig. 5. Pore-size distributions by DFT of the adsorbent materials obtained from the defatted grape seeds and the commercial activated carbons (F400, MAXSORB).
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
CO2, CH4, H2 uptake (mass fraction %)
120 100 MAXSORB
80
CO2
to be good precursors for the production of activated carbon as the activated carbons obtained had no heavy metals and a low ash content of between 0.1% and 3% with high specific surface areas, SBET. BET specific surface areas of up to 1988 m2 g 1 were obtained. The adsorbent materials are fundamentally microporous with a micropore volume higher than 90%. The activated carbons obtained in this work appear excellent candidates for CO2 capture.
60 GSP-1-900-f
Acknowledgement F400
40
The authors would like to thank AIICA for supply the lignocellulosic wastes studied in the present manuscript.
APP-1-750-f
20 CH4 H2
0 0
1
2
3
Pressure (MPa) Fig. 6. High pressure CO2, CH4 and H2 adsorption isotherms for two adsorbent materials (GSP-1-900-f and APP-1-750-f) obtained from lignocellulosic wastes from the extraction of tannins and the commercial activated carbons (F400, MAXSORB).
Table 4 Gas adsorption capacities at high pressure of the adsorbent carbons obtained and commercial activated carbons. Samples
P (MPa) ⁄
GSP-1-900-f APP-1-750-f F400 MAXSORB *
7
%CO2 adsorbed
⁄
%CH4 adsorbed
⁄
%H2 adsorbed
0.1
3
0.1
3
0.1
3
13.91 16.49 13.43 12.00
50.53 37.26 40.18 110.00
2.6 2.9 1.47 1.8
11.1 9.1 8.29 15.5
0.0 0.0 0.01 0.01
0.23 0.19 0.16 0.30
% mass fraction.
has been used in previous CO2-sorption studies [12–15]. The other activated carbon obtained in this work (APP-1-750-f) and selected for studying gas adsorption capacity at high pressure also presents up to a pressure of 1 MPa a CO2 adsorption capacity higher than the F400 but a lower capacity than that of the other adsorbent material obtained at high temperature, GSP-1-900-f. A high CO2 adsorption capacity may be due also to the presence of high contents of nitrogen in the precursor since it has been suggested in the scientific literature that the presence of nitrogen-containing functional groups on the activated carbons may contribute to a higher selectivity towards CO2 [34,35]. The activated carbons obtained in this work seem excellent candidates for CO2 capture. Both the presence of basic surface groups (nitrogenous groups) and the development of porosity give rise to carbon materials with excellent properties in terms of adsorption capacity and selectivity to CO2. 4. Conclusions Activated carbons were prepared from lignocellulosic wastes obtained from the extraction of tannins (defatted grape seeds and acacia seed shells) by the KOH activated method at different KOH to lignocellulosic waste weight ratios, activation temperatures and flow nitrogen rates. The results show high BET surface areas, and porosity development directly related to the ratio used during chemical activation. Another property that has a positive effect on the textural development of the adsorbent materials is the activation temperature. The higher the activation temperature the higher the textural development. The flow rate in the chemical activation does not appears to influence in the textural development. The defatted grape seeds and acacia seed shells are shown
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.micromeso.2014. 09.004. References [1] Grape Tannins, Programa LIFE (LIFE-04/ENV/ES/000237). Union Europea (2004–2007). [2] J.M. Luque-Rodríguez, M.D. Luque de Castro, P. Pérez-Juan, Talanta 68 (2005) 126–130. [3] A. Molero, C. Pereyra, E. Martínez de la Ossa, Chem. Eng. J. 61 (1996) 227–231. [4] N.M. Nor, L.C. Lau, K.T. Lee, A.R. Mohamed, J. Environ. Chem. Eng. 1 (2013) 658– 666. [5] O. Ioannidou, A. Zabaniotou, Renew. Sust. Energy Rev. 11 (2007) 1966–2005. [6] B.S. Girgis, A.M. Soliman, N.A. Fathy, Microporous Mesoporous Mater. 142 (2011) 518–525. [7] J.M. Dias, M.C.M. Alvim-Ferraz, M.F. Almeida, J. Rivera-Utrilla, M. Sánchez-Polo, J. Environ. Manage. 85 (2007) 833–846. [8] M. Al Bahri, L. Calvo, M.A. Gilarranz, J.J. Rodriguez, Chem. Eng. J. 203 (2012) 348–356. [9] D. Jimenez-Cordero, F. Heras, N. Alonso-Morales, M.A. Gilarranz, J.J. Rodriguez, Chem. Eng. J. 231 (2013) 172–181. [10] I. Okman, S. Karagöz, T. Tay, M. Erdem, Appl. Surf. Sci. 293 (2014) 138–142. [11] M. Kumar, R. Tamilarasan, J. Environ. Chem. Eng. 1 (4) (2013) 1108–1116. [12] Y. Gensterblum, P. van Hemert, P. Billemont, A. Busch, D. Charriére, D. Li, B.M. Krooss, G. de Weireld, D. Prinz, K.H.A.A. Wolf, Carbon 47 (2009) 2958–2969. [13] J.E. Fitzgerald, R.L. Robinson, K.A.M. Gasem, Langmuir 22 (2006) 9610–9618. [14] R. Humayun, D.L. Tomasko, AIChE J. 46 (2000) 2065–2075. [15] M. Sudibandriyo, Z. Pan, J.E. Fitzgerald, R.L. Robinson, K.A.M. Gasem, Langmuir 19 (2003) 5323–5331. [16] J.P. Marco-Lozar, J. Juan-Juan, F. Suárez-García, D. Cazorla-Amorós, A. LinaresSolano, Int. J. Hydrogen Energy 37 (2012) 2370–2381. [17] J.P. Marco-Lozar, M. Kunowsky, F. Suarez-Garcia, J.D. Carruthers, A. LinaresSolano, Energy Environ. Sci. 5 (2012) 9833–9842. [18] M.A. Sheikh, M.M. Hassan, K.F. Loughlin, Gas Sep. Purif. 10 (1996) 161–168. [19] S. Jribi, B.B. Saha, S. Koyama, H. Bentaher, Energy Convers. Manage. 78 (2014) 985–991. [20] T. Otowa, R. Tanibata, M. Itoh, Gas Sep. Purif. 7 (1993) 241–245. [21] J.P. Olivier, W.B. Conklin, M.V. Szombathely, Stud. Surf. Sci. Catal. 87 (1994) 81–89. [22] R.R. Gil, R.P. Girón, M.S. Lozano, B. Ruiz, E. Fuente, J. Anal. Appl. Pyrol. 98 (2012) 129–136. [23] N. Ferrera-Lorenzo, E. Fuente, I. Suárez-Ruiz, R.R. Gil, B. Ruiz, J. Anal. Appl. Pyrol. 105 (2014) 209–216. [24] N. Ferrera-Lorenzo, E. Fuente, J.M. Bermúdez, I. Suárez-Ruiz, B. Ruiz, Bioresour. Technol. 151 (2014) 199–206. [25] D. Jimenez-Cordero, F. Heras, N. Alonso-Morales, M.A. Gilarranz, J.J. Rodriguez, Biomass Bioenergy 54 (2013) 123–132. [26] K.G. Mansaray, A.E. Ghaly, Bioresour. Technol. 65 (1998) 13–20. [27] H. Yang, R. Yan, H. Chen, D.H. Lee, C. Zheng, Fuel 86 (2007) 1781–1788. [28] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscor, R.A. Pierotti, J. Rouquérol, T. Siemieniewska, Pure Appl. Phys. Chem. 57 (1985) 603–619. [29] S. Román, J.M. Valente, Microporous Mesoporous Mater. 165 (2013) 127–133. [30] S. Brunauer, L.S. Deming, W.E. Deming, E. Teller, J. Am. Chem. Soc. 62 (1940) 1723–1732. [31] D. Lozano-Castelló, M.A. Lillo-Ródenas, D. Cazorla-Amorós, A. Linares-Solano, Carbon 39 (2001) 741–749. [32] R.R. Gil, B. Ruiz, M.S. Lozano, M.J. Martín, E. Fuente, Chem. Eng. J. 245 (2014) 80–88. [33] N. Ferrera-Lorenzo, E. Fuente, I. Suárez-Ruiz, B. Ruiz, Fuel Process. Technol. 121 (2014) 25–31. [34] M.G. Plaza, C. Pevida, A. Arenillas, F. Rubiera, J.J. Pis, Fuel 86 (2007) 2204–2212. [35] M.S. Shafeeyan, W.M.A.W. Daud, A. Houshmand, A. Shamiri, J. Anal. Appl. Pyrol. 89 (2010) 143–151.
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Sustainable porous carbons from lignocellulosic wastes obtained from the extraction of tannins B. Ruiz a, E. Ruisánchez a, R.R. Gil a, N. Ferrera-Lorenzo a, M.S. Lozano b, E. Fuente a,⇑ a b
Instituto Nacional del Carbón (INCAR), CSIC, Francisco Pintado Fe, 26, 33011 Oviedo, Spain Asociación de Investigación de las Industrias del Curtido y Anexas (AIICA), Igualada, Barcelona, Spain
a r t i c l e
i n f o
Article history: Received 28 May 2014 Accepted 1 September 2014 Available online xxxx Keywords: Lignocellulosic wastes Tannins Activated carbon Pyrolysis Chemical activation
a b s t r a c t The present research study explores the possibility of obtaining high surface area activated carbons (ACs) from lignocellulosic wastes from the extraction of tannins. The use of vegetable tannins in the leather industry has the serious drawback that it involves the mass destruction of trees. Currently, studies are being conducted to obtain tannins from different lignocellulosic wastes. Two lignocellulosic wastes from the extraction of tannins, defatted grape seeds and acacia seed shells, with high carbon and nitrogen contents and a low ash content were obtained and investigated as a potential precursor for the preparation of activated carbons. KOH chemical activation, with a previous pyrolysis step, was performed in a conventional electric furnace varying the experimental conditions of KOH/precursor weight ratio, final activation temperature, and inert flow gas. After activation the samples were washed with a solution of HCl and water or just with hot water. The ACs obtained were essentially microporous with a specific surface area up to 2000 m2 g and presented low ash content, less than to 0.10% in the case of adsorbent materials from the defatted grape seeds and up to 3.60% for the materials from the acacia seed shells. The best results were obtained with the largest KOH/precursor weight ratio or the highest activation temperature (900 °C). Their moderate nitrogen content (up to 1.5%) makes them especially suitable materials for CO2 capture. Some of them were more effective for CO2 adsorption than the commercial activated carbon F400 and therefore represent an attractive alternative to more expensive adsorbent materials. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The tanning industry uses animal hide, a putrescible material, and converts it into a stable material, leather. If the tanning industry did not exist animal skins, a by-product of the meat and dairy industries, would have to be disposed of by other means such as by land filling or incineration. The tanning industry mainly uses one of two methods for transforming skin into leather: mineral or vegetable tanning. Mineral tanning which uses chromium salts and acids is most modern and the most used method but it is also the most polluting. Vegetable tanning uses vegetable tanning substances, called tannins. Tannins are very numerous and are distributed abundantly in the nature. The disadvantage of this type of tanning is deforestation. Currently, studies are being conducted to obtain tannins from different lignocellulosic wastes. Thus, AIICA has developed a methodology for extracting them from defatted grape seeds under solvent extraction in basic conditions [1] with very good results. Grape seeds, which are a solid waste from the wine industry, are ⇑ Corresponding author. Tel.: +34 985 119090; fax: +34 985 297662. E-mail address: [email protected] (E. Fuente).
mostly burnt as fuel and to some extent used for cattle feed, despite the fact that they are also a source of oil for human consumption [2,3] or tannin vegetables which could be used in the tanning industry [1]. The possibility of obtaining tannins from acacia seed shells by means of a grinding and sieving operation has also been investigated. The fruits of Acacia (Faidherbia albida) are composed of two parts: the seeds and the seed shell. The seed serves as food for animals and the shell seeds after undergoing a process of grinding and sieving provide the vegetable tannin concentrates in the fine powder fraction. This fraction is used as tanning agent by the tanners from northern Africa. After the process of obtaining tannins from defatted grape seeds and acacia seed shells a lignocellulosic residue with a high carbon and low ash content are obtained. The leather industry has often been associated with a high pollution of the water and air and to generating a large amount of solid wastes. Wastewater is loaded of sulfides, chlorides, sulfates, phosphates, settleable solids, color. . . and chromium in the case of the mineral tanning. The main parameters affecting the air quality in the tanneries are volatile organic compounds (VOCs), H2S, NH3 and dust. Furthermore, pollutants such as nitrogen oxides
http://dx.doi.org/10.1016/j.micromeso.2014.09.004 1387-1811/Ó 2014 Elsevier Inc. All rights reserved.
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
2
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
(NOx), sulfur oxides (SOx), and carbon dioxide (CO2) are generated in the thermal systems to generate heat. The adsorption is an effective abatement technology for contaminant compounds. The most important consideration is the selection of an appropriate adsorbent. The adsorbents most commonly used for contaminant compounds removal are the activated carbons because they have a large surface area, high porosity and great adsorption capabilities. Activated carbons (ACs) can be obtained from almost any material with high carbon content. The precursors generally used for the production of ACs are coal, lignite and wood, among others. In order to reduce the production costs of these adsorbents, many research studies have been undertaken to find alternatives to the raw materials used in activated carbon production, such as agricultural and industrial wastes [47]. Furthermore, an evaluation of agricultural residues or by-products as precursors of activated carbon seems to be very promising from a sustainable viewpoint. In this work, the precursors of the ACs will be two agricultural wastes obtained from defatted grape seeds and acacia seed shells after that these residues have been subjected to a process of solvent extraction and grinding–sieving, respectively, for the separation of the tannins. The use of grape seeds in the preparation of ACs has been reported in only a few instances in the scientific literature [8–10] and almost none from acacia seed shells [11]. However, there is no background literature information related to the use of the lignocellulosic wastes reported in this work. The aim of the present research work is to evaluate the possibility of obtaining sustainable porous carbons from lignocellulosic wastes from the extraction of tannins by means of KOH chemical activation using a previous pyrolysis step. The influence of the activating agent/precursor weight ratio, activation temperature, inert flow gas and different washing processes of the final materials is studied in order to achieve a suitable development of porosity in the activated carbons. 2. Materials and methods 2.1. Raw materials This study was carried out in order to valorise the lignocellulosic wastes from the extraction of tannins. Two representative samples of these wastes, residual materials left behind after the extraction of tannins from defatted grape seeds and acacia seed shells, were supplied by ‘‘Asociación de Investigación de las Industrias del Curtido y Anexas’’, AIICA. The extraction of tannins from defatted grape seeds was performed in an aqueous medium using sodium metabisulfite as a solubilizer of tannins, and applying pressure and temperature [1]. After the extraction process a solid lignocellulosic residue, GS, is left behind. This will serve as one of the precursor materials of the activated carbons. The method of extracting tannin from Acacia Albida fruits was as follows: the acacia fruits (seeds and seed shells) were crushed and then passed through an air flow where the seeds and the seed shells were separated due to their different densities. After this, the seed shells were crushed by a cutter mill and were separated into two fractions by sieving: a coarse fraction and a fine powder. The fine powder contained the highest concentration of tannins (47.9%) while the other fraction contained less than half (21.4%). The coarse fraction, AP, was selected as precursor material for the activated carbons. In Appendix A, Supplementary Data, the Fig. S1 shows the leitmotiv of this research study and the Figs. S2 and S3 show an image illustrating the equipment for the extraction of tannins from defatted grape seeds and the scheme of the tannins separation from the acacia seed shells, respectively.
2.2. Activated carbons The activated carbons from these lignocellulosic wastes from the extraction of tannins were obtained in a two-stage thermochemical process which involved carbonizing the raw material followed by chemical activation of the char with an alkali agent (KOH). The thermochemical processes, pyrolysis and thermochemical activation, were carried out in an alumina crucible placed in a conventional tubular furnace Carbolite CTF 12/65/550. The pyrolysis step of the raw material (GS, AP) was performed under a N2 flow gas of 150 ml/min 1 at a heating rate of 5 °C/min up to a temperature of 750 °C for 60 min. In order to establish the most suitable pyrolysis conditions, several parameters were studied in a thermo-balance, Q5000IR: heating rate (5, 10 and 15 °C/min), final temperature (900 and 1000 °C), maintenance time of final temperature (60 and 120 min), under a flow nitrogen rate of 100 ml min 1 in all the experiments. In the chemical activation, the activating agent (KOH) and the pyrolyzed sample (GSP or APP) were mixed in solid state (physical mixture) in different weight ratios (1:1. and 2:1). The experimental conditions of the chemical activation were: a heating rate of 5 °C/min, a final activation temperature of from 700 to 900 °C, the final temperature being held for 60 min, and different N2 gas flows of 150 and 500 ml min 1. After chemical activation, in order to remove the activation products and the mineral matter blocking the porosity, the adsorbent materials were washed with a 5M hydrochloric acid solution and subjected to a series of deionised water (Milli-Q) rinses. Some activated materials were washed only with hot water at 100 °C and without using hydrochloric acid. Finally the samples were dried at 105 °C. The nomenclature for the activated carbons (ACs) prepared is summarized in Table 1. In order to conduct a comparative study of the CO2 adsorption capacity of the obtained activated carbons a commercial activated carbon (Filtrasorb F400) was selected since it had been used in previous CO2-sorption studies [12–15]. The commercial activated carbon ‘‘Maxsorb3000’’ was also used as reference since it is a material that is widely used in gas storage [16–19] and has very good textural properties [20].
Table 1 Nomenclature of the activated carbons. Sample
KOH/precursor ratio
Activation temperature (°C)
N2 flow gas (ml min 1)
GSP-1-700-f GSP-1-750-f GSP-1-750-F GSP-2-750-f GSP-1-800-f GSP-1-900-f GSP-1-900-F GSP-1-900-fW
1:1 1:1 1:1 2:1 1:1 1:1 1:1 1:1
700 750 750 750 800 900 900 900
150 150 500 150 150 150 500 150
APP-1-700-f APP-1-750-f APP-1-750-F APP-2-750-f APP-1-800-f APP-1-900-f APP-1-900-fW
1:1 1:1 1:1 2:1 1:1 1:1 1:1
700 750 750 750 800 900 900
150 150 500 150 150 150 150
GSP (defatted grape seeds carbonized) and APP (acacia seed shells carbonized). 1 or 2: activating agent/precursor weight ratio, 1:1 or 2:1, respectively. 700–750–800–900: activation temperature (°C). f or F: inert gas flow in the chemical activation process, 150 ml min 1 or 500 ml min 1, respectively. W: samples washed with hot water; samples washed with HCl and water have not annotation.
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
3
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
2.3. Characterization of the materials The moisture and ash contents of the sample were determined following the UNE 32002 and 32004 norms, respectively. The ultimate analysis (carbon, hydrogen, nitrogen and sulfur content) was carried out on automatic instruments (LECO CHN-2000 and S-144DR). Atomic absorption spectrophotometry was used to determine the inorganic composition. The ashes were dissolved with inorganic acids and analysed on Atomic Absorption Spectrophotometer Shimazu AA-6300 equipment. The morphology of the samples was examined using a Scanning Electron Microscope equipped with an Energy-Dispersive X-ray analysing system (SEM/EDX). Textural characterization was performed by measuring the N2 adsorption isotherms at 196 °C in a Micromeritics ASAP 2420 automatic apparatus. The isotherms were used to calculate the BET specific surface area, SBET, and total pore volume, VTOT, at a relative pressure of 0.95. The pore size distributions, microporosity and mesoporosity, were obtained by applying the density functional theory (DFT) model to the N2 adsorption data, assuming slit-shaped pore geometry [21]. Before these adsorption experiments, the samples were out gassed under vacuum at 120 °C overnight to remove any adsorbed moisture and/or gases. To measure the capacity of the sorbents to capture CO2, H2 and CH4, high pressure adsorption isotherms were obtained using a Rubotherm-VTI magnetic suspension balance at ambient temperature and under static conditions. CO2, H2 and CH4 adsorption isotherms were determined up to 3 MPa. Prior to the measurements the samples were outgassed at 120 °C under a high vacuum for 240 min. 3. Results and discussion
Table 2 Chemical characterization of the raw materials, carbonized materials and activated carbons. Sample
Mass fraction (%), db Ash
C
H
N
S
2.07 5.53
53.32 50.57
5.68 4.77
1.05 1.45
0.23 0.26
5.04 12.60
91.14 82.86
1.2 1.14
1.8 2.08
0.1 0.07
GSP-1-700-f GSP-1-750-f GSP-1-750-F GSP-2-750-f GSP-1-800-f GSP-1-900-f
0.33 0.14 0.27 0.05 0.35 0.92
93.26 92.60 95.76 92.61 96.24 92.16
0.69 0.40 0.35 0.50 0.26 0.15
1.11 1.24 0.86 1.02 0.96 0.77
0.07 0.07 0.06 0.09 0.09 0.09
APP-1-700-f APP-1-750-f APP-1-750-F APP-2-750-f APP-1-800-f APP-1-900-f
2.81 3.60 3.22 0.86 2.26 3.43
92.91 91.90 92.14 94.93 93.63 91.84
0.59 0.50 0.46 – 0.26 0.24
1.51 1.25 1.23 0.92 1.17 0.99
– – 0.10 – 0.11 0.12
91.00 93.13
0.30 0.14
1.00 0.25
0.70 –
GS AP GSP APP
F400 MAXSORB
the Fig. 1. It shows that in the case of grape seeds CaO is the principal oxide with a 36.7% presence while in the case of the acacia seed shells SiO2 (38.0%) is the main oxide, due to the parts of the plant chosen for this study, since acacia is the seed shell and the Si is part of its cell structure. However, in the grape seed there is more Ca which contributes to the growth of seeds. The second most abundant compound in both cases is K2O which is the nutrient that most plants absorb after nitrogen in the form of cation K+. Finally, defatted grape seeds contains SiO2 (13%) and the acacia seed shells CaO (15.8%). The rest of the oxides are minor components in both wastes.
3.1. Chemical characterization of the materials The results of the proximate and ultimate analyses of the raw materials (GS and AP), their corresponding carbonized materials (GSP, APP), the adsorbents material obtained and the commercial activated carbons are shown in Table 2. The chemical analysis of both raw materials (GS, AP) showed a carbon content higher than 50% but a low ash and sulfur content. This high carbon and low ash content make the materials good precursors for obtaining activated carbons. In this work, a previous pyrolysis step was used before KOH chemical activation in order to study the possible energetic use of the different fractions obtained (char, fuel, gas) using a procedure similar to that used for other types of industrial waste of leather and macroalgae [22–24]. Once these raw materials were pyrolyzed the carbon content of the chars (GSP, APP) increased significantly, especially in the case of grape seeds, up to 91.14%, and for the acacia seed shells up to 82.86%. Although the ash content doubled in both cases, from 5.53% to 12.60% in the case of acacia seed shells and from 2.07% to 5.34% in the case of the grape seeds, the ash content remained low in these pyrolyzed materials. The chemical composition values of the GSP are similar to those obtained by Jimenez-Cordero et al. [25] in his study of granular chars from the flash and conventional pyrolysis of grape seeds subjected to n-hexane extraction for 24 h in an apparatus to remove oil. As can be seen from Table 2, the adsorbent materials obtained from GSP and APP show a high carbon content (91.84–96.24%), a low ash content (<1% and up to 3.60% for the activated carbons from defatted grape seeds and acacia seed shells, respectively) and a low sulfur content (up to 0.12%). The ash composition of the raw materials (GS, AP) which was studied by atomic absorption spectrophotometry is presented in
3.2. Thermal degradation of the raw materials (GS, AP) The thermal degradation of the lignocellulosic wastes (GS, AP) was studied in a thermo-balance, Q5000IR, in a nitrogen atmosphere. The behavior of these materials is shown in the form of thermograms in Fig. 2. Three heating rates at 5, 10 and 15 °C/min were used, to simulate the thermal shock undergone by the wastes when they were subjected to the working temperature (750 °C), The pyrolysis of the raw materials (GS, AP) revealed an initial slight weight loss at about 100 °C, due to the elimination of water. In both cases the weight loss started at about 200–250 °C and ended at about 550 °C. 40 35 30
GS
AP
25 % 20 15 10 5 0 CaO
SiO2
Al2O3 Fe2O3
MgO
Na2O
K2O
Fig. 1. The ash composition of the lignocellulosic wastes from the extraction of tannins (GS, AP).
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
4
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
This loss resulted from the decomposition of the major constituents of the lignocellulosic materials; hemicellulose decomposing mainly between 150 and 350 °C, cellulose between 275 and 350 °C, and lignin gradually decomposing between 250 and 500 °C [26,27]. As the pyrolysis curves of the wastes show, the percentage of final weight loss depends only minimally on the heating rate, the differences in the final residue being of the order of 3% in the case of acacia and 1.5% for the grape. In light of these data and decomposition temperatures, as mentioned previously, the pyrolysis temperature and heating rate were adjusted to 750 °C (to allow minimum weight deviation and ramp weight loss stabilized) and 5 °C/min. (fast enough to cause degradation processes), respectively. Once the optimal conditions of pyrolysis were established, as a function of the results obtained in the thermo-balance, the raw materials (GS, AP) were pyrolyzed at a heating rate of 5 °C/min, a pyrolysis temperature of 750 °C for 60 min. After the pyrolysis step, the carbonized materials or chars obtained (GSP, APP) were chemically activated with KOH in different experimental conditions to yield several activated carbons, Table 1. 3.3. SEM microscopy SEM images of the raw materials (GS, AP) and the adsorbent materials obtained in this work are shown in Fig. 3. It can be seen that the raw materials have a characteristic lignocellulosic structure. In the pyrolysis process the raw structure was modified with the loss of its cellular structure. The external surfaces of the activated carbons show a lot cavities and are 0,6 Defatted Grape seeds 5ºC/min Defatted Grape Seeds 10ºC/min Defatted Grape Seeds 15ºC/min
Weight (%)
100
0,5
80
0,4
60
0,3
40
0,2
20
0,1
0
Deriv.Weight (%/ºC)
120
0 50
150
250
350 450 550 650 Temperature (ºC)
750
850
120
0,4 Acacia seed shells 5ºC/min Acacia seed shells 10ºC/min Acacia seed shells 15ºC/min
100
Weight (%)
60
0,2
40
Deriv. Weight (%)
0,3 80
0,1 20 0
0,0 50
150
250
350
450
550
650
750
850
950
Temperature (ºC) Fig. 2. Pyrolysis of defatted grape seeds (GS) and acacia seed shells (AP) at different heating rates (5, 10, 15 °C/min).
very irregular, indicating generation of porosity in the adsorbent materials produced after the attack of the alkaline reagent (KOH) during activation. 3.4. Textural characterization of the materials In the KOH chemical activation process different parameters were considered for obtaining activated carbons: the activating agent/precursor weight ratio, the nitrogen flow, the activation temperature and finally, when the materials were obtained, the different solutions to use in the washing process. The results obtained in the textural characterization of the activated carbons are shown in Fig. 4 (nitrogen adsorption isotherms at 196 °C) and in Table 3 where, the data corresponding to the equivalent specific surface area -SBET-, total pore volume at p/p0 = 0.95 -VTOTand the distribution of porosity obtained by applying the density functional theory (DFT) model to the N2 adsorption data, are presented. Fig. 4 shows all the N2 adsorption isotherms at 196 °C for the materials derived from the defatted grape seeds. As can be seen the raw material does not show any nitrogen adsorption. Therefore the equivalent specific surface areas (BET) are negligible for this sample, <1 m2 g 1 (Table 3). The adsorption of N2 by the activated carbons obtained from this residue takes place, fundamentally, at low relative pressures which is typical of microporous solids [28,29]. Only the samples obtained in more severe conditions of activation show a small hysteresis loop, which is associated to capillary condensation inside the mesopores, but the hysteresis is very small so the materials are mainly microporous. In general the isotherms are of type I, according to the BDDT classification and type I–IV hybrid for the materials obtained in several activation conditions [30]. The capacity of the nitrogen adsorption of the activated carbon increases considerably with the increase in the KOH/precursor weight ratio which was found to be the most important parameter in the chemical activation process (Fig. 4) [31,32]. As can be seen in this figure an increase in the activating agent produces a widening of the knee of the isotherms, indicating a change in the pore size distribution. The quantity of N2 adsorbed was found to be greater in the adsorbent which had been subjected to the greatest amount of activating agent (KOH:sample ratio = 2:1). The activated carbon also has the highest equivalent specific surface area (SBET = 1860 m2 g 1) and total pore volume (0.772 cm3 g 1), Table 3. Okman et al. produced activated carbons from grape seeds by chemical activation with potassium hydroxide at different activation temperature (600 and 800 °C) using activating agent/precursor ratios of 0.25:1, 0.50:1 and 1:1; the mixture of chemical and precursor was in a state of dissolution; in these conditions an increase in the weight ratio did not appear to promote an increase in the BET specific surface [10]. The other experimental variable that has a significant influence on the adsorption capacity of the activated carbons obtained is the activation temperature. An increase in this variable led to an increase in the nitrogen adsorption capacity of the adsorbents obtained (Fig. 4), in the specific surface area (1038 and 1604 m2 g 1 for the activated carbons obtained at 700 °C and 900 °C, respectively) and also in the total pore volume (0.415 and 0.672 cm3 g 1 for the adsorbents obtained at 700 °C and 900 °C, respectively), Table 3. It has also been reported that increases in activation temperatures produce adsorbent materials with higher surface areas [10,33]. The effect of the flow inert gas (150 and 500 ml min 1) used on the KOH chemical activation was studied at 750 °C and 900 °C. This variable caused few changes to the materials obtained. There was a slight increase in the textural development of the activated carbons with the increase in nitrogen flow (Fig. 4 and Table 3). It
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
5
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
(a)
(b)
(a*)
(c)
(d)
(d*)
Fig. 3. SEM photographs: (a) an acacia seed shell, (a⁄) detail of acacia seed shell, (b) activated carbon from acacia seed shells (900 °C, 150 ml min (d) activated carbon from defatted grape seeds (900 °C, 150 ml min 1), (d⁄) detail of activated carbon.
1200 MAXSORB GSP-2-750-f GSP-1-900-F GSP-1-900-f F400 GSP-1-800-f GSP-1-750-F GSP-1-750-f GSP-1-700-f GS
Vads (cm3g-1, STP)
1000 800 600 400 200 0 0
0,2
0,4
p/po
0,6
0,8
1
Fig. 4. N2 adsorption isotherms at 196 °C for the raw material and the activated carbons obtained from the defatted grape seeds and the commercial activated carbons.
has been reported that the flow gas has an important effect on the porosity development of activated carbons obtained from coal by KOH chemical activation. A higher nitrogen flow rate during the carbonization process produces activated carbons with a much higher micropore volume [31]. The results for the activated carbons obtained from the two residues (defatted grape seeds and acacia seed shells) are very similar, so conclusions can be applied to both residues. It should be noted
1
), (c) defatted grape seeds,
that the activated carbons obtained from the acacia seed shells exhibit a nitrogen adsorption capacity slightly higher than the corresponding materials obtained from defatted grapes seeds. Also the BET specific surface area and total pore volume values are slightly higher. The adsorbent material with the higher SBET (1988 m2 g 1) and total pore volume (0.809 cm3 g 1) was obtained by KOH chemical activation from acacia seeds shell at a KOH/precursor weight ratio of 2:1 at 750 °C and under an inert gas flow rate of 150 ml min 1. Another experimental variable object studied was the effect of the washing stage. The washing stage is the last step in the preparation of a chemically activated carbon. This washing serves to remove activation products and mineral matter blocking the porosity of the adsorbent materials obtained. In order to see the effect of the washing step, activated carbons were prepared from both residues (defatted grape seeds and acacia seed shells) by KOH chemical activation at a weight ratio of 1:1, at 900 °C and under a nitrogen flow of 150 ml min 1. The materials thus obtained were washed with 5M hydrochloric acid solution and subjected to a series of deionised water rinses (GSP-1-900-f and APP-1-900-f) or with hot water (GSP-1-900-fW and APP-1-900-fW). In Appendix A, Supplementary Data, the Fig. S4 shows the N2 adsorption isotherms corresponding to the four activated carbons. The samples washed with hydrochloric acid and water (GSP-1900-f and APP-1-900-f) show a somewhat higher N2 adsorption than the adsorbent materials washed only with hot water (GSP-1-900-fW and APP-1-900-fW) which is indicative of a more efficient removal of potassium compounds by washing with HCl. Moreover, they also present a higher specific surface area (1604 and 1684 m2 g 1 for GSP-1-900-f and APP-1-900-f, respectively) and total pore volume (0.672 and 0.716 cm3 g 1 for GSP-1-900-f
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
– –
15.2 24.1 25.5 54.2 29.1 48.3 48.1 47.6
54.7 52.3 45.2
45.3 49.7
– –
82.7 75.1 72.9 42.4 69.0 44.4 42.5 45.4
43.6 42.4 49.6
32.2 5.0
22.5 45.3
1.7 5.3 5.2
Vmmi (%) (0.7–2 nm)
– –
and APP-1-900-f, respectively), Table 3. Similar results have been reported for KOH chemical activated carbons from coal by Lozano-Castelló [31]. Table 3 and Fig. 5 show the pore-size distributions by DFT of the adsorbent materials obtained from the defatted grape seeds and the commercial activated carbons (F400, MAXSORB). It can be observed from Table 3 that all of the adsorbents materials obtained in this work are microporous materials and in all the cases the micropore volume is higher than 90%. It is noteworthy than the adsorbent materials obtained at an activation temperature between 700 and 800 °C are ultramicroporous materials with micropore volume higher than 98% and the contribution of ultramicropore volume being between from 82.7% to 69.0%. It can be seen that a higher activation temperature (900 °C) or high activating agent/precursor weight ratio (2:1) favors the widening of pores with the activated carbons obtained showing a greater contribution of wider micropores (up to 54.2%) and mesopores (up to 9.3%). The commercial activated carbon MAXSORB has an important mesopore volume (45%).
3.5. CO2/CH4/H2 adsorption capacity of the adsorbent materials
) (2–50 nm) – –
0.007 0.003 0.006 0.021 0.008 0.041 0.054 0.036
0.011 0.030 0.028
0.114 0.645
– –
0.052 0.092 0.098 0.337 0.122 0.272 0.275 0.244
0.359 0.298 0.242
0.229 0.708 0.163 0.071
0.286 0.241 0.266
0.282 0.287 0.280 0.264 0.289 0.250 0.243 0.233
– –
c
0.641 1.681
0.809 0.716 0.641
0.415 0.458 0.469 0.772 0.510 0.672 0.706 0.632
– –
1271 3420 F400 MAXSORB
1988 1684 1547 APP-2-750-f APP-1-900-f APP-1-900-fW
1038 1169 1176 1860 1275 1604 1613 1496 GSP-1-700-f GSP-1-750-f GSP-1-750-F GSP-2-750-f GSP-1-800-f GSP-1-900-f GSP-1-900-F GSP-1-900-fW
<1 <1 GS AP
Vumi: Volume ultramicroporous (<0.7 nm). Vmmi: Volume medium-microporous (0.7–2 nm). Vme: Volume mesoporous (2–50 nm). a
b
Vumi (cm3 g a
) p/p0 = 0.95
1
)
VTOT (cm3 g
0,14
MAXSORB GSP-2-750-f GSP-1-900-F GSP-1-900-f F400 GSP-1-800-f GSP-1-750-F GSP-1-750-f GSP-1-700-f
0,12
1
SBET (m2 g Samples
Table 3 Textural characterization of the raw materials and activated carbons.
1
) (<0.7 nm)
b
Vmmi (cm3 g
1
) (0.7–2 nm)
c
Vme (cm3 g
1
Fig. 6 shows the CO2, H2 and CH4 adsorption isotherms at high pressure for two selected activated carbons obtained in this work from the lignocellulosic waste (GSP-1-900-f and APP-1-750-f) together with the commercial activated carbons, F400 and MAXSORB; Table 4 shows the adsorption capacities of CO2, H2 and CH4 in terms of the mass fraction (%) of these materials at air pressure (P = 0.1 MPa) and the maximum pressure studied (P = 3 MPa). The materials tested exhibit a high capacity for retaining CO2, moderate CH4 adsorption capacity and a very low H2 adsorption capacity, Table 4 and Fig. 6. This clearly evidences the selectivity of the adsorbents to CO2 so they could be used to separate gases in order to obtain pure hydrogen or to separate CO2/H2 and CH4/H2. It is well known that the adsorption capacity of an activated carbon is mainly governed by its texture and its surface chemistry. Maxsorb presents the highest capacity for adsorbing CO2 due to the high textural development of this commercial activated carbon (>3000 m2 g 1) followed by the adsorbent material obtained from defatted grape seeds at high temperature (900 °C), GSP-1-900-f. This adsorbent obtained from the lignocellulosic waste displays a CO2 adsorption capacity higher than the commercial activated carbon F400 which was selected as reference since it
0,1 Inv Vol (cm3g-1)
a
b
Vumi (%) (<0.7 nm)
c
2.1 0.8 1.6 3.4 1.9 7.3 9.4 7.0
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
Vme (%) (2–50 nm)
6
0,08 0,06 0,04 0,02 0 0,1
1
10
Pore diameter (nm) Fig. 5. Pore-size distributions by DFT of the adsorbent materials obtained from the defatted grape seeds and the commercial activated carbons (F400, MAXSORB).
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004
B. Ruiz et al. / Microporous and Mesoporous Materials xxx (2014) xxx–xxx
CO2, CH4, H2 uptake (mass fraction %)
120 100 MAXSORB
80
CO2
to be good precursors for the production of activated carbon as the activated carbons obtained had no heavy metals and a low ash content of between 0.1% and 3% with high specific surface areas, SBET. BET specific surface areas of up to 1988 m2 g 1 were obtained. The adsorbent materials are fundamentally microporous with a micropore volume higher than 90%. The activated carbons obtained in this work appear excellent candidates for CO2 capture.
60 GSP-1-900-f
Acknowledgement F400
40
The authors would like to thank AIICA for supply the lignocellulosic wastes studied in the present manuscript.
APP-1-750-f
20 CH4 H2
0 0
1
2
3
Pressure (MPa) Fig. 6. High pressure CO2, CH4 and H2 adsorption isotherms for two adsorbent materials (GSP-1-900-f and APP-1-750-f) obtained from lignocellulosic wastes from the extraction of tannins and the commercial activated carbons (F400, MAXSORB).
Table 4 Gas adsorption capacities at high pressure of the adsorbent carbons obtained and commercial activated carbons. Samples
P (MPa) ⁄
GSP-1-900-f APP-1-750-f F400 MAXSORB *
7
%CO2 adsorbed
⁄
%CH4 adsorbed
⁄
%H2 adsorbed
0.1
3
0.1
3
0.1
3
13.91 16.49 13.43 12.00
50.53 37.26 40.18 110.00
2.6 2.9 1.47 1.8
11.1 9.1 8.29 15.5
0.0 0.0 0.01 0.01
0.23 0.19 0.16 0.30
% mass fraction.
has been used in previous CO2-sorption studies [12–15]. The other activated carbon obtained in this work (APP-1-750-f) and selected for studying gas adsorption capacity at high pressure also presents up to a pressure of 1 MPa a CO2 adsorption capacity higher than the F400 but a lower capacity than that of the other adsorbent material obtained at high temperature, GSP-1-900-f. A high CO2 adsorption capacity may be due also to the presence of high contents of nitrogen in the precursor since it has been suggested in the scientific literature that the presence of nitrogen-containing functional groups on the activated carbons may contribute to a higher selectivity towards CO2 [34,35]. The activated carbons obtained in this work seem excellent candidates for CO2 capture. Both the presence of basic surface groups (nitrogenous groups) and the development of porosity give rise to carbon materials with excellent properties in terms of adsorption capacity and selectivity to CO2. 4. Conclusions Activated carbons were prepared from lignocellulosic wastes obtained from the extraction of tannins (defatted grape seeds and acacia seed shells) by the KOH activated method at different KOH to lignocellulosic waste weight ratios, activation temperatures and flow nitrogen rates. The results show high BET surface areas, and porosity development directly related to the ratio used during chemical activation. Another property that has a positive effect on the textural development of the adsorbent materials is the activation temperature. The higher the activation temperature the higher the textural development. The flow rate in the chemical activation does not appears to influence in the textural development. The defatted grape seeds and acacia seed shells are shown
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.micromeso.2014. 09.004. References [1] Grape Tannins, Programa LIFE (LIFE-04/ENV/ES/000237). Union Europea (2004–2007). [2] J.M. Luque-Rodríguez, M.D. Luque de Castro, P. Pérez-Juan, Talanta 68 (2005) 126–130. [3] A. Molero, C. Pereyra, E. Martínez de la Ossa, Chem. Eng. J. 61 (1996) 227–231. [4] N.M. Nor, L.C. Lau, K.T. Lee, A.R. Mohamed, J. Environ. Chem. Eng. 1 (2013) 658– 666. [5] O. Ioannidou, A. Zabaniotou, Renew. Sust. Energy Rev. 11 (2007) 1966–2005. [6] B.S. Girgis, A.M. Soliman, N.A. Fathy, Microporous Mesoporous Mater. 142 (2011) 518–525. [7] J.M. Dias, M.C.M. Alvim-Ferraz, M.F. Almeida, J. Rivera-Utrilla, M. Sánchez-Polo, J. Environ. Manage. 85 (2007) 833–846. [8] M. Al Bahri, L. Calvo, M.A. Gilarranz, J.J. Rodriguez, Chem. Eng. J. 203 (2012) 348–356. [9] D. Jimenez-Cordero, F. Heras, N. Alonso-Morales, M.A. Gilarranz, J.J. Rodriguez, Chem. Eng. J. 231 (2013) 172–181. [10] I. Okman, S. Karagöz, T. Tay, M. Erdem, Appl. Surf. Sci. 293 (2014) 138–142. [11] M. Kumar, R. Tamilarasan, J. Environ. Chem. Eng. 1 (4) (2013) 1108–1116. [12] Y. Gensterblum, P. van Hemert, P. Billemont, A. Busch, D. Charriére, D. Li, B.M. Krooss, G. de Weireld, D. Prinz, K.H.A.A. Wolf, Carbon 47 (2009) 2958–2969. [13] J.E. Fitzgerald, R.L. Robinson, K.A.M. Gasem, Langmuir 22 (2006) 9610–9618. [14] R. Humayun, D.L. Tomasko, AIChE J. 46 (2000) 2065–2075. [15] M. Sudibandriyo, Z. Pan, J.E. Fitzgerald, R.L. Robinson, K.A.M. Gasem, Langmuir 19 (2003) 5323–5331. [16] J.P. Marco-Lozar, J. Juan-Juan, F. Suárez-García, D. Cazorla-Amorós, A. LinaresSolano, Int. J. Hydrogen Energy 37 (2012) 2370–2381. [17] J.P. Marco-Lozar, M. Kunowsky, F. Suarez-Garcia, J.D. Carruthers, A. LinaresSolano, Energy Environ. Sci. 5 (2012) 9833–9842. [18] M.A. Sheikh, M.M. Hassan, K.F. Loughlin, Gas Sep. Purif. 10 (1996) 161–168. [19] S. Jribi, B.B. Saha, S. Koyama, H. Bentaher, Energy Convers. Manage. 78 (2014) 985–991. [20] T. Otowa, R. Tanibata, M. Itoh, Gas Sep. Purif. 7 (1993) 241–245. [21] J.P. Olivier, W.B. Conklin, M.V. Szombathely, Stud. Surf. Sci. Catal. 87 (1994) 81–89. [22] R.R. Gil, R.P. Girón, M.S. Lozano, B. Ruiz, E. Fuente, J. Anal. Appl. Pyrol. 98 (2012) 129–136. [23] N. Ferrera-Lorenzo, E. Fuente, I. Suárez-Ruiz, R.R. Gil, B. Ruiz, J. Anal. Appl. Pyrol. 105 (2014) 209–216. [24] N. Ferrera-Lorenzo, E. Fuente, J.M. Bermúdez, I. Suárez-Ruiz, B. Ruiz, Bioresour. Technol. 151 (2014) 199–206. [25] D. Jimenez-Cordero, F. Heras, N. Alonso-Morales, M.A. Gilarranz, J.J. Rodriguez, Biomass Bioenergy 54 (2013) 123–132. [26] K.G. Mansaray, A.E. Ghaly, Bioresour. Technol. 65 (1998) 13–20. [27] H. Yang, R. Yan, H. Chen, D.H. Lee, C. Zheng, Fuel 86 (2007) 1781–1788. [28] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscor, R.A. Pierotti, J. Rouquérol, T. Siemieniewska, Pure Appl. Phys. Chem. 57 (1985) 603–619. [29] S. Román, J.M. Valente, Microporous Mesoporous Mater. 165 (2013) 127–133. [30] S. Brunauer, L.S. Deming, W.E. Deming, E. Teller, J. Am. Chem. Soc. 62 (1940) 1723–1732. [31] D. Lozano-Castelló, M.A. Lillo-Ródenas, D. Cazorla-Amorós, A. Linares-Solano, Carbon 39 (2001) 741–749. [32] R.R. Gil, B. Ruiz, M.S. Lozano, M.J. Martín, E. Fuente, Chem. Eng. J. 245 (2014) 80–88. [33] N. Ferrera-Lorenzo, E. Fuente, I. Suárez-Ruiz, B. Ruiz, Fuel Process. Technol. 121 (2014) 25–31. [34] M.G. Plaza, C. Pevida, A. Arenillas, F. Rubiera, J.J. Pis, Fuel 86 (2007) 2204–2212. [35] M.S. Shafeeyan, W.M.A.W. Daud, A. Houshmand, A. Shamiri, J. Anal. Appl. Pyrol. 89 (2010) 143–151.
Please cite this article in press as: B. Ruiz et al., Micropor. Mesopor. Mater. (2014), http://dx.doi.org/10.1016/j.micromeso.2014.09.004